« Back to Results

AEA Poster Session

Poster Session

Friday, Jan. 3, 2020 7:00 AM - 6:00 PM (PDT)

Marriott Marquis, North Tower - Grand Ballroom Foyer & West Lobby Lounge

Saturday, Jan. 4, 2020 7:00 AM - 6:00 PM (PDT)

Marriott Marquis, North Tower - Grand Ballroom Foyer & West Lobby Lounge

Sunday, Jan. 5, 2020 7:00 AM - 3:00 PM (PDT)

Marriott Marquis, North Tower - Grand Ballroom Foyer & West Lobby Lounge
Hosted By: American Economic Association

Safe Asset Carry Trade (G1, E4)

Benedikt Fabian Ballensiefen
,
University of St. Gallen and World Bank Group
Angelo Ranaldo
,
University of St. Gallen and Swiss Finance Institute

Abstract

Investors pay a "convenience premium" for the safety and liquidity benefits provided by near-money or safe assets. This convenience premium varies across time and securities. For instance, U.S. Treasuries have a lower yield than otherwise equivalent safe assets. While the temporal and cross-sectional variation in the prices of safe assets is well documented in the literature, our understanding of their asset pricing implications is still very limited. We provide the first systematic asset pricing analysis of one important category of safe assets, the repurchase agreement (repo).

We find that heterogeneity in repo rates allows for a remunerative carry trade. This carry trade involves a long position in "less expensive" safe assets and a simultaneous short position in their "more expensive" counterparts. The return on this investment strategy, our carry factor, exhibits time-variation, which points toward time-dependent cross-sectional differences among safe assets. A standard, no-arbitrage model with two risk factors, a market factor and a carry factor, is able to explain the price of safe assets: While the market factor determines the level of short-term interest rates, the carry factor accounts for their cross-sectional dispersion. From the perspective of the safe asset literature, our carry factor reflects a portfolio-based convenience premium differential and is explained by three main factors: the safety premium, the liquidity premium, and the opportunity cost of holding money.

Our analysis contributes to three strands of the literature. First, we contribute to the asset pricing literature by introducing a new focus on short-term interest rates. Second, our analysis contributes to the literature on the dispersion in short-term interest rates by highlighting that heterogeneity in repo rates allows for a remunerative carry trade. Third, we add to the growing literature on safe assets shortages and their macroeconomic and financial stability implications by explaining the existence of our carry factor from a safe asset perspective.

Attention Triggers and Retail Investors' Risk Taking (G4, G0)

Marc Arnold
,
University of St. Gallen
Matthias Pelster
,
Paderborn University
Marti G. Subrahmanyam
,
New York University

Abstract

This paper investigates how individual attention triggers affect financial risk taking. We analyze a large sample of trading records from a brokerage service that sends standardized push messages on stocks to retail investors. This micro-level data allows us to isolate the push messages as individual stock-attention triggers. By exploiting this data in a difference-in-differences setting, we find that attention triggers increase investors' willingness to take risk. We also provide cross-sectional analyses to identify for which investors and stocks this effect is more pronounced.

Childlessness and Inter-Temporal Fertility Choice (J1, J0)

Fabian Siuda
,
Ifo Institute and University of Munich

Abstract

This paper develops and estimates a dynamic structural model of fertility with endogenous marriage formation, linking the timing of fertility to its intensive (number of children) and extensive (having children) margin. The model features rational, forward-looking agents who make decisions on marriage and fertility, and are exposed to declining fecundity rates over time. In every period, agents face a trade-off between work and child-rearing, and across time there is a trade-off between having children early or late in life. The model parameters are identified using four distinct facts of the 2008 and 2012 German Microcensus: (i) fertility until age 30 decreases with education for married and single women, (ii) fertility after age 30 increases with education for married and single women, (iii) childlessness increases with women’s education, (iv) marriage rates decrease with education for women and increase with education for men. I obtain three main insights. First, postponement of childbirth combined with the natural decline of fecundity over time can explain up to 15% of childlessness, depending on education. Second, by estimating the model separately for East and West Germany, I find that institutions and economic conditions matter: the two major factors for childlessness in West Germany are postponement of childbirth and high opportunity costs of children due to lack of public childcare. By contrast, in East Germany, social sterility plays a larger role. Finally, using the estimated model parameters for counterfactual analysis, I evaluate consequences of reoccurring labor market interruptions and policies aimed at reconciling work and family life.

Comparative Studies of M&A and Greenfield Investments in Response to Changes in Tax Regulations and Administration of Host Economies (H3, F3)

Gohar S. Sedrakyan
,
Georgia State University

Abstract

We analyze the impact of the tax indicators described by Paying Taxes scores of Doing Business report on two modes of foreign direct investments in equity capital- greenfield investments and cross-border mergers and acquisitions. Then, we assume 25 percent improvement in each of studied tax determinants and evaluate this effect on the inbound flow of both types of FDI. The study compares four econometric methods-the ordinary least squares, random-effects, fixed-effects, and Hausman-Taylor models- applied to the panel dataset for one hundred sixty countries for the period from 2009 to 2017. The comprehensive nature of Hausman-Taylor tools, which allows for analysis of panel datasets with time-varying, time-invariant and endogenous parameters of tax administration, reveals divergence among the factors that impact transactions of cross-border mergers and acquisitions and greenfield investments. This study estimates higher sensitivity of cross-border mergers and acquisitions to the factors of economic development, overall business friendliness and location of potential host countries. Greenfield investments are more likely to be directed to the countries with effective tax administration rules and they are less sensitive to the size of a host economy; and thus, any country, that implements rigorous tax reforms, has potential for attracting greenfield investments.

Do Investors Exploit Bank Earnings Management Information to Earn Profits in Stock Markets? (G1)

Nguyen Ngoc Thao Vo
,
RMIT University-Vietnam
Thai Vu Hong Nguyen
,
RMIT University-Vietnam
Duc Thi Hong Phan
,
RMIT University-Australia

Abstract

The issue of asymmetric information remains the concern in investment decisions and has regained attention in relation to financial crisis. This study investigates whether investors exploit bank earnings management information to earn profits by trading on the basis of this informational advantage in stock markets. We argue that investors are divergent in capability to detect bank earnings management, which will be subsequently reflected in their trading positions. We answer the questions if bank accounting regulations, enforcement regimes, institutional quality and IFRS adoption play the roles in shaping investors’ behaviours in exploiting trading profits on the basis of detecting bank earnings management. Our research, on the sample of 198 banks across 13 Asia-Pacific countries in the period of 2002-2017, finds that there is a significant positive relationship between bank income-decreasing earnings management and bid-ask spreads. Our study also sheds light on the signalling effects of bank accounting regulations/ enforcement/ institutions/ IFRS adoption in guiding investors’ behaviours in response to bank earnings management.

Government Credit and Trade War (F0, G2)

Ning Cai
,
China Development Bank
Jinlu Feng
,
China Development Bank
Yong Liu
,
China Development Bank
Hong Ru
,
Nanyang Technological University
Endong Yang
,
University of Macau

Abstract

Using transaction-level trade data from China Customs and loan data from the China Development Bank (CDB), we analyze how government credit affects trade activities. CDB credit to strategic industries at the top of supply chains leads to lower prices, higher volumes, and more product varieties and destinations of exports for firms in downstream industries. This increased import competition crowds out U.S. firms in the same industry but crowds in downstream firms. In the recent trade war, the U.S. government strategically raised tariffs on China’s imports that compete with domestic manufacturers rather than the imports that are used as inputs for U.S. firms.

Household Heterogeneity and the Value of Government Spending Multiplier: An Analytical Characterization (E2, E6)

Pawel Kopiec
,
National Bank of Poland

Abstract

This paper provides an analytical decomposition of the fiscal multiplier
in economy populated with heterogeneous households, uninsured idiosyncratic
income risk and frictional product market. Similarly to Auclert (2019),
the derived expression consists of interpretable, model-based channels that
describe the transmission of government spending shocks by private consumption.
Calibrated model is used to estimate the magnitude of multipliers
and their structure under alternative fiscal and monetary rules. Analytical
and quantitative comparison to the multiplier’s formula in economy with
identical agents indicates that household heterogeneity plays a crucial role
in the propagation of fiscal expenditures shocks.

Information Sharing Is Not Always the Right Option in CPR Extraction Management: Experimental Findings (D9, Q5)

Dimitri Dubois
,
University of Montpellier
Stefano Farolfi
,
UMR G-eau and CIRAD
Phu Nguyen-Van
,
BETA, CNRS, INRA, University of Strasbourg and Thang Long University
Julliette Rouchier
,
LAMSADE, CNRS, and University of Dauphine

Abstract

This paper experimentally investigates the impact of different information sharing mechanisms in a common pool resource (CPR) game, with the aim to find a mechanism that is at the same time efficient and not too expensive for the managing agency. More precisely, we compare the extraction level observed with three mechanisms: a mandatory information sharing mechanism and two voluntary information sharing mechanisms that differ by the degree of freedom given to the players. Our main result is that the mechanism of voluntary information sharing can help achieve a lower average extraction level than that observed with the mandatory one but also that giving actors a lot of freedom introduces strategic considerations that can lessen this positive impact.

Optimal Taxation of Assets (H2, D3)

Nicolaus Tideman
,
Virginia Tech
Thomas Mecherikunnel
,
Virginia Tech

Abstract

The optimal taxation of assets requires attention to two concerns: 1) the elasticity of the supply of assets and 2) the impact of taxing assets on distributional objectives. The most efficient way to attend to these two concerns is to tax assets of different types separately, rather than having one tax on all assets. When assets are created by specialized effort rather than by saving, as with innovations, discoveries of mineral deposits and development of unregulated natural monopolies, it is interesting to consider a regime in which the government awards a prize for the creation of the asset and then collects the remaining value of the asset in taxes. Analytically, the prize is like a wage after taxes. In this perspective, prizes are awarded based on a variation on optimal taxation theory, while assets of different types are taxed in divergent ways, depending on their characteristics. Some categories of assets are abolished.

Partitioned Pricing and Consumer Welfare (L0, D9)

Kevin Ducbao Tran
,
DIW Berlin

Abstract

In online commerce, obfuscation strategies by sellers are hypothesized to mislead consumers to their detriment and to the profit of sellers. One such obfuscation strategy is partitioned pricing in which the price is split into a base price and add-on fees. While empirical evidence suggests that partitioned pricing impacts consumer decisions through salience effects, its consumer welfare consequences are largely unexplored. Therefore, I provide a quantification of the welfare impact of the behavioral response to partitioned pricing. To do so, I derive a discrete choice model that jointly allows for differences in the reaction to marginal changes in add-on fees and the base price as well as a discontinuous effect of a zero fee. The model is based on a framework on limited attention and I estimate it using web scraped data of posted price transactions on eBay Germany. My results suggest under-reaction to marginal changes in the shipping fee, consistent with previous results in the literature. However, I also document a discontinuous positive effect of free shipping on consumer demand, which is novel to the literature. The combined impact of these effects on consumer welfare is less than six percent of consumer surplus. The welfare impact is attenuated because the maximum shipping fee on eBay is capped and the free shipping effect partly counteracts the under-reaction to shipping fees in expectation.

Sovereign Bond Premium and Global Macroeconomic Conditions (G0)

Sandro Andrade
,
University of Miami
Adelphe Ekponon
,
Cambridge University
Alexandre Jeanneret
,
HEC Montreal

Abstract

This paper studies how global macroeconomic conditions affect sovereign bond prices. Weak and volatile economic performance during recessions increases a country’s default probability more than strong and stable performance during expansions reduces it, leading to countercyclical and unconditionally high sovereign credit spreads. We identify the sovereign bond premium arising from this exposure to severe but low-frequency changes in global macroeconomic conditions. Our model predicts that this bond premium is higher for countries that are more exposed to the global business cycle, particularly around recessions. We find support for this prediction using emerging market sovereign bond data over the 1994Q1-2018Q2 period.

The Invisible Costs of Promoting Competition in the Airline Industry (R4, H2)

Hoang Dao
,
State University of New York-Buffalo

Abstract

Ever since the Airline Deregulation Act in 1978, the airline industry drastically transitioned from the most condensed and regulated industry to one of the most competitive one. Severe competition makes airline industry's profit margin among the lowest at 8.2% in 2018, only slightly more than half of U.S. average (15.2%). The introduction of electronic booking and budget airlines increased the competition in this market to an unparalleled degree. How would such competition affect airline's behavior, market structure, and ultimately, the future of aviation? This paper explores the effects of airline industry competition on the firms' costs and operations behavior. Specifically, the effects of competition on airline's safety expense, product-differentiation expense, route choices and fleets.

This article begins by deriving an estimate for the degree of competition, employing the discrete choice techniques with differentiated product to estimate the demand for air travel in the U.S. domestic markets. A substitution matrix and a vector of preferences for observable characteristics are obtained for each quarter from 1993Q1 to 2018Q4. A measure for competition in the industry is formulated for each airline during the period. This competition index is then used to evaluate the effect of competition on the multiple costs and characteristics of airlines, using the instrumented difference in differences method. My result shows that, the expense for safety does not change significantly, but the expenses for product differentiation decreases as the markets become more competitive. Airlines fly longer routes on average, and air fleets gravitates towards homogeneously narrow-body, long-distance airplanes as airlines face more competition.

Welfare Multiplier of Public Investment (E6, F4)

Giovanni Ganelli
,
International Monetary Fund
Juha Tervala
,
University of Helsinki

Abstract

We analyze the welfare multipliers of public spending—the consumption equivalent change in welfare for a one dollar change in public spending—in a DSGE model. The welfare multiplier of public investment depends crucially not only on the productivity (output elasticity) of public capital, as shown by earlier studies, but also on the depreciation rate of public capital and the efficiency of public investment defined as a fraction of public investment spending that translates into the public capital stock. When the key parameter values are set based on the empirical estimates for advanced economies and the output multipliers are consistent with the empirical estimates, the welfare multiplier is positive and sizable. Our welfare multiplier of choice is 1.4 for advanced economies, meaning that households are willing to pay 1.4 dollars for a one dollar increase in public investment spending. The welfare multiplier is roughly zero when the key parameter values are set to match the features of developing economies. A public infrastructure push in advanced economies makes sense, but developing economies should enhance the efficiency and productivity of public investment.

A Time to Print, a Time to Reform (N3, O3)

Lars Boerner
,
Martin Luther University of Halle-Wittenberg and King's College London
Jared Rubin
,
Chapman University
Battista Severgnini
,
Copenhagen Business School

Abstract

The public mechanical clock and movable type printing press were arguably the most important and complex technologies of the late medieval period. We document two of their most important, yet unforeseeable, consequences. First, towns that were early adopters of clocks were more likely to also be early adopters of presses. We posit that towns with clocks became upper-tail human capital hubs|both technologies required extensive technical know-how that had many points of overlap. Second, a three-stage instrumental variables analysis indicates that the press in
uenced the adoption of Protestantism, while the clock's effect on the Reformation was mostly indirect.

Airbnb and Private Investment in Chicago Neighborhoods (R0, O1)

Minhong Xu
,
Nanjing Audit University
Yilan Xu
,
University of Illinois-Urbana-Champaign

Abstract

The Airbnb-based home-sharing platform reduces the market frictions of short-term rentals, which raises the potential economic returns to a property. The conversion of residential units into tourist accommodation and the associated new revenue flows create incentives for capital investment. This study examines how the expansion of the Airbnb market has stimulated capital investment in Chicago neighborhoods. The instrumental variable estimates show that, given a 1% increase in the Airbnb listings, the number building permits issued in a quarter increased by 0.84% while capital investment increased by 3.19% or $81,000 equivalently. Besides direct investment in residential properties, we find spillover capital flows to retail and commercial zones where amenities and businesses arise to meet the demand of the shifting population. We show that the effects were primarily driven by commercial hosts rather than casual hosts. Moreover, Airbnb disproportionally enhanced capital investment in declining and stable communities.

Automation in Developing Countries: Plant-Level Evidence and Inter-Industry Spillovers (O3, O1)

Giorgio Presidente
,
World Bank

Abstract

This paper provides novel evidence on the impact of industrial automation in a developing country. Matching industry-level data on shipments of industrial robots to a large sample of manufacturing plants in Indonesia, we find that robotisation, both at the plant-level and through interindustry linkages, boosts plant performance. We interpret the evidence as suggesting that adoption of robots allows producers to upgrade their position in the global value chain.

Bank Relationship, Covenant Enforcement, and Creditor Control (G2, G3)

Yong Kyu Gam
,
Southwestern University of Finance and Economics
Chunbo Liu
,
Southwestern University of Finance and Economics

Abstract

This paper investigates the effect of banking relationship on the likelihood of lenders’ enforcement of loan covenant violations. We find that if lenders have a long-run relationship with borrowers, these lenders enforce material covenant violations at a substantially lower rate when borrowers breach financial covenants. Moreover, borrowers with such relationships are less likely to experience raises in loan interest rates and a deterioration of subsequent financing and investment activities when they fail to fulfill their financial covenants. Further evidence shows that the mitigation of information asymmetry along the lending relationship is the driving force of the empirical findings.

Birth Weight and Cognitive Development during Childhood: Evidence from India (I1, O1)

Santosh Kumar
,
Sam Houston State University
Kaushalendra Kumar
,
International Institute for Population Sciences-Mumbai
Ramanan Laxminarayan
,
Center for Disease Dynamics, Economics & Policy-New Delhi and Princeton University
Arindam Nandi
,
Center for Disease Dynamics, Economics & Policy-Washington

Abstract

Health at birth is an important indicator of human capital development over the life course. This paper uses longitudinal data from the Young Lives survey and employs instrumental variable regression models to estimate the effect of birth weight on cognitive development during childhood in India. We find that a 10 percent increase in birth weight increases cognitive test score by 8.1 percent or 0.11 standard deviations at ages 5-8 years. Low birth weight infants experienced a lower test score compared with normal birth weight infants. The positive effect of birth weight on a cognitive test score is larger for boys, children from rural or poor households, and those with less-educated mothers. Our findings suggest that health policies designed to improve birth weight could improve human capital in resource-poor settings.

Climate and Tropical Cyclone Effects on Economic Activity: Evidence at the Firm Level from Mexico (O1, Q0)

Miriam Juarez-Torres
,
Bank of Mexico
Puigvert Jonathan Sr.
,
Bank of Mexico

Abstract

How tropical cyclones affect the economic activity of firms of manufacturing and services sectors in Mexico? For answering this question, we constructed a unique database of monthly observations of firms’ economic activity indicators and at the municipality level tropical cyclone exposure, measured as maximum sustained winds, and weather data to estimate their impact on production and remunerations per worker in manufacturing for the period 1994-2014 and revenues and remunerations per worker at the services sector for 2005-2015. Controlling by climate and other variables, we find that for an average firm located in a municipality with high frequency of tropical cyclones exposure, the maximum sustained winds show a negative impact on production and remunerations per worker in the manufacturing sector and a low and persistent negative effect in revenues of firms of the services sector. An important contribution of this paper is based on the more disaggregated temporal and spatial character of its results, which allow identifying how heterogeneity plays a role on the magnitude of the effect, which is especially useful for policy design.

Comparative Analysis of Russian Citizens’ Well-Being Before and After the Collapse of the Soviet Union (P2, I3)

N. N. Tarun Chakravorty
,
Siberian Federal University
Vladislav N. Rutskiy
,
Siberian Federal University
Elena P. Sevastyanova
,
Siberian Federal University

Abstract

This study compares the happiness and well-being of Russian citizens between the Soviet period and the present capitalist Russia using quantitative and qualitative data collected by authors’ students from the residents of some Russian regions while the respondents are 60 years’ old or above, who have experienced family lives in both of the periods. It makes inferences from the content analysis of the qualitative information provided by the respondents on their perception of happiness in two systems with the aid of NVivo. It analyses the quantitative data and compares the levels of well-being that the citizens had in the past and have now. We gather the following findings: a larger number of people suggest that they were happier in Soviet period than in present capitalist Russia; stability in their life, stability of the state, certainty of life they wanted to live, guarantee of livelihood and confidence in the ability to live a better life in the days to come, were the main reasons for their happiness in Soviet Russia; those who think they were happier in Soviet period, do not want to go back to that system as they fear that such a shift might involve uncertainty while certainty was the main reason for their happiness in that period; those who suggest that they are happier in the present capitalist Russia, put forward freedom, expansion of choices, expansion of opportunities as their arguments; although freedom is considered a main element of happiness in present capitalist Russia, it can contribute to increasing happiness only after an individual is able to fulfil his or her necessities and has some excess money for investment or entertainment.

Conflicts in Private Family Firms (G3, D1)

Janis Berzins
,
BI Norwegian Business School
Alminas Zaldokas
,
Hong Kong University of Science and Technology

Abstract

We use Norwegian household-level data and full structures of family relationships to understand how succession decisions are made when the family has multiple potential heirs. We argue that the decisions on ownership distribution in family firms are related to the potential of future family conflicts. We first theoretically show that both joint and concentrated control can be optimal, depending on how likely the heirs are to disagree about the corporate policy and such effect is non-monotonic. We document that the patterns of divided bequest are more likely when the potential disagreement is lower in private Norwegian family firms as captured by heirs being of different gender, levels of education, ages, and born to different parents. The firms controlled by the families that are expected to have fewer potential conflicts are also more likely to remain within the family control. We identify the potential of family conflicts based on the history of the founder's divorces instrumented by the divorces in the extended family relationships, outside of the nuclear family.

Consequences of Immigrating During a Recession: Evidence from the United States Refugee Resettlement Program (J6, J3)

Joshua Mask
,
University of Illinois-Chicago

Abstract

Are there long-term labor consequences from migrating to the United States during a recession? For most immigrants, credibly estimating this effect is difficult because of selective migration. Some immigrants may not move if economic conditions become unfavorable. However, identification is possible for refugees as their arrival dates are exogenously determined through the US Refugee Resettlement program. A one percentage point increase in the arrival national unemployment rate reduces refugee wages by 1.66 percent and employment probability by 1.39 percentage points after five years.

Consumption, Government Failure, and Asset Prices (G1, E0)

Thiago de Oliveira Souza
,
University of Southern Denmark

Abstract

The equity premium–risk-free rate level and predictability puzzles in standard power utility consumption-based asset pricing models disappear once we remove the government-imposed component from the consumption expenditure series. I calibrate this component based on the growth rates of two proxies for government intervention, which I also show to forecast the short- and long-term equity premiums between 1974 (or 1981) and 2017. In summary, investors require large premiums to hold stocks because stocks deliver poor returns when government intervention (failure) increases, systematically reducing individual utility levels. Government failure is likely the key macro-finance variable linking asset prices and economic fluctuations.

Contagion and Return Predictability in Asset Markets: An Experiment with Two Trees (G4, G0)

Andreea Popescu
,
Tilburg University
Charles Noussair
,
University of Arizona

Abstract

Using a laboratory experiment, we investigate if contagion can emerge between two risky assets even though their fundamentals are not correlated. To guide our experimental design, we use the ‘Two trees’ asset pricing model developed by Cochrane et al. (2007). The model makes time-series and cross-section return predictions following a shock to one of the assets’ dividend share. Consistent with the predictions of the model, we observe positive autocorrelation in the shocked asset, a positive contemporaneous correlation between the two risky assets, and time-series and cross-sectional return predictability using the dividend-price ratio. In line with the rational foundation of the model, the model’s predictions have higher support in markets with more sophisticated agents.

Costly Information Processing and Consumption Dynamics (E7, E2)

Jeremy Boccanfuso
,
Paris School of Economics

Abstract

This paper studies the consumption-saving problem of a consumer who faces a fixed cost for paying attention to noisy information and whose attention strategy, i.e., whether or not she pays attention, can be a function of the underlying information. At the optimum, consumers chose to be attentive when evidence accumulates far from their prior beliefs. The model provides an explanation to four puzzling empirical findings on consumption and expectations. First, consumers’ attention depends on the information content. Second, aggregate information rigidities vary over the business cycle. Third, consumers only react to large anticipated shocks and neglect the impact of small ones. Fourth, aggregate consumption dynamics vary over the business cycle.

Crowdfunding Dynamics (G3, L1)

Paul Belleflamme
,
Catholic University of Louvain
Thomas Lambert
,
Erasmus University
Armin Schwienbacher
,
University of Côte d’Azur

Abstract

Various forms of social learning and network effects are at work on crowdfunding platforms, giving rise to informational and payoff externalities. We use novel entrepreneur-backer data to study how these externalities shape funding dynamics, within and across projects. We find that backers decide to back a particular project based on past contributions not only to that project — as documented by prior work — but also to other contemporaneous projects — a novel result. Our difference-in-differences estimates indicate that such ‘cross-project funding dynamics’ account for 4-5% in the increase of contributions that projects generate on a daily basis. We show that recurrent backers are the main transmission channel of cross-project funding dynamics: by initiating social learning about project existence and quality, recurrent backers encourage future funding by other backers. Our results demonstrate that even though contemporaneous projects compete for funding, they jointly benefit from their common presence on the platform. We finally show that these crowdfunding dynamics stir platform growth, with important consequences for competition among platforms.

Data Network Effects: The Example of Internet Search (L4, L8)

Maximilian Schaefer
,
DIW Berlin

Abstract

The rise of dominant firms in data driven industries is often credited to their alleged data advantage. Empirical evidence lending support to this conjecture is lacking. In this paper, we show that data as an input into machine learning tasks displays features that favor the hypothesis that data is a source of market power. We study the search result quality for search keywords on Yahoo!. Search result quality improves when more users search a keyword. In addition to this direct network effect caused by more users, we observe an additional externality that is caused by the amount of data that the search engine collects on the users. More data on the users reinforces the direct network effect. We propose to view this reinforcement effect due to additional user-specific data as a data network effect. Our findings are
consistent with the consensus that data display diminishing returns to scale for a given prediction task. This feature of data is often regarded as incompatible with the hypothesis that data is a source of market power. Our results rationalize the market power hypothesis through a different mechanism by suggesting that data, in addition to being an input, is also a technology shifter.

Default and Liquidity: A Continuous Time Approach (G0, E5)

Theofanis Papamichalis
,
University of Oxford
Dimitrios Tsomocos
,
University of Oxford

Abstract

Classical General Equilibrium models are discrete-time and assume away financial frictions, especially endogenous default and liquidity. We develop a continuous time stochastic macroeconomic model that incorporates both features. We identify the default channel that accelerates the rate of decrease of leverage and consequently investment. We also assess the equilibrium trade-off relationship between money supply and default penalty. We show that there exists an optimal monetary and regulatory mix, which achieves optimal levels of equilibrium welfare.

Differential Treatment in the Bond Market: Sovereign Risk and Mutual Fund Portfolios (F3, G1)

Nathan Converse
,
Federal Reserve Board
Enrico Mallucci
,
Federal Reserve Board

Abstract

How does sovereign risk affect investors' behavior? We answer this question using a novel database that combines sovereign default probabilities for 27 developed and emerging markets with monthly data on the portfolios of individual bond mutual funds. We first show that changes in yields do not fully compensate investors for additional sovereign risk, so that bond funds reduce their exposure to a country's assets when its sovereign default risk increases. However, the magnitude of the response varies widely across countries. Fund managers aggressively reduce their exposure to high-debt countries and high-risk countries. By contrast, they are more lenient toward core developed markets. In this sense, these economies appear to receive preferential treatment. Second, we document what determines the destination of reallocation flows. When fund managers reduce their exposure to a country in response to its sovereign risk, they shift their assets to countries outside the immediate geographic region while at the same time avoiding countries with high debt-to-GDP ratios and markets to which they are already heavily exposed. These results are supportive of models of sovereign default that assign a nontrivial role to the preferences of international creditors.

Discount Rates and Cash Flows: A Local Projection Approach (G1, C3)

Matthijs Lof
,
Aalto University
Henri Nyberg
,
University of Turku

Abstract

We develop a volatility decomposition derived from flexible and robust local projections (Jordà, 2005) to quantify the relative contributions of expected discount rates and cash flows to the variation of dividend yields. Compared to conventional vector autoregressions (VARs), local projections are more robust to misspecification, e.g. in the presence of dividend smoothing. In addition, local projections enable the incorporation of large information sets, the use of monthly data in addition to annual data, and the consideration of time variation of the volatility decomposition. We apply the LASSO and Model Averaging methods to accommodate large sets of predictive variables.
In contrast to conventional VAR-based volatility decompositions (e.g. Campbell and Shiller, 1989; Cochrane, 2008), we find that the contribution of expected cash flows is by no means non-negligible, in particular when extending the information set beyond the lagged dividends yield or when allowing for time-variation of the volatility decomposition. The variation of expected discount rates remains nevertheless the dominant contributor to market volatility.

Diversification in Lottery-Like Features and Portfolio Pricing Discounts (G4, G1)

Xin Liu
,
University of Bath

Abstract

I study the asset pricing implications of cumulative prospect theory on portfolio discounts. I extend Barberis and Huang (2008) and show that a portfolio consisting of lottery-like stocks should trade at a discount due to diversification. This discount can be partially mitigated if lottery-like stocks tend to produce extreme payoffs at the same time. I utilize three empirical settings to support this theoretical prediction: the closed-end fund puzzle, the announcement returns of mergers and acquisitions, and conglomerate discounts. My findings support cumulative prospect theory from an alternative perspective and provide a novel and unifying explanation for three seemingly unrelated phenomena.

Do Supply-Side Incentives Improve the Use of Healthcare Services? New Evidence from a Field Experiment (D9, I1)

Ahmad Reshad Osmani
,
University of Memphis
Albert Okunade
,
University of Memphis

Abstract

Studies have recently kindled light on the effects of incentive modalities in the healthcare sector. However, there is insufficient evidence on underlying causes of partial effectiveness of these strategies in the health systems of developing countries. This study presents results from a large-scale randomized experiment across 6,848 households in Afghanistan that evaluates the impact of a conditional incentive pay scheme to the health facilities. Supported by the target income hypothesis framework and relaxing the compliance assumption in the empirical modeling, the estimated coefficients yield causal effects of supply-side conditional incentive on the demand for healthcare services. After two years, conditional incentive led to induce the demand for the pre- targeted maternal and children healthcare services among the households at lower levels and contracted-out health facilities. Further, the incentive scheme is associated with sizable efficiency gains at facility level. These gains are realized at the expense of deterring service users’ satisfaction with the physician’s communication qualities. This study establishes that margins of improvement do exist on the supply-side performance conditioning on organizational structure and the service contractual arrangements of health facilities. The current work provides a framework for plausible implementation of incentive policies in the healthcare sector.

Domestic Migration: For Money or For Love? (J6, H2)

Ben Klemens
,
U.S. Treasury Department

Abstract

Economic models of migration, domestic and international, typically begin with the assumption that a moving household's primary goal is to attain higher income than it would earn by staying. This article uses administrative records for almost all people earning formal market income in the U.S., 2001-2015, totaling about 1.7 billion household observations with 82 million long-distance moves, to develop a detailed match between movers and comparable stayers and thus a comparison of movers' income changes relative to stayers. In aggregate, movers see about a median 1% gain in income after moving relative to the counterfactual of staying, with wide variance. Even a decade later, about two out of five households have lower income relative to staying, with an overall median relative income gain of about 6%. Pecuniary benefits are not evenly distributed: movers leaving school and younger single households without children are likely to see higher income relative to staying, but other movers, most notably single parents, are roughly half as likely to see a relative income gain. The overall story is a bifurcated population of movers. Roughly half move to higher income relative to staying, and the rest do not, indicating for whom the hypothesis of income maximization is difficult to support, and where future research about the many motives for moving may focus.

Don't Expect Too Much: The Effect of Biased Expectations on (Over)-Indebtedness (D1, D8)

Melanie Koch
,
German Institute for Economic Research
Theres Kluehs
,
Leibniz University Hannover
Wiebke Stein
,
Leibniz University Hannover

Abstract

Household indebtedness is rising worldwide. This study investigates one possible driver of this increase that is rooted in the theory of permanent income: high income expectations. We collect data from an emerging country, Thailand, as (over-) indebtedness in markets with incomplete financial infrastructure and social security can be devastating. Furthermore, our sample of rural households is exposed to a high degree of uncertainty, which makes expectation formation prone to behavioral biases. We implement a new measure for high income expectations and show that it is strongly and robustly related to both objective and subjectively felt over-indebtedness. Controlling for various household characteristics, unexpected shocks, and other possible confounding factors reduces the concern about reverse causality. In an additional lab-in-the-field experiment, we explicitly find that overconfidence, a specific form of biased expectation, is related to overborrowing.

Employment Effects of Three Rounds of Federal Minimum Wage Hikes (J2)

Kenneth J. McLaughlin
,
City University of New York-Hunter College and Graduate Center

Abstract

This paper presents event-study estimates of the effects of the 1990–1991, 1996– 1997, and 2007–2009 rounds of federal minimum wage hikes on the employment of teens and high school dropouts in states without super-federal minimum wages. In state-year panel data from the Current Population Survey, a control group of people ages 25–59 with at least a high school education generates counterfactual series that track teen and dropout employment rates quite well (outside the periods of minimum wage hikes). Deviations from the counterfactual series in the post-hike period identify the employment effects of the minimum wage hikes.

For the 1990–1991 and 2007–2009 rounds, the employment effects for teens and dropouts are negative, statistically significant, economically large, and robust to the treatment of trends and year effects. Differences by sex and race are small compared to the difference by age: disemployment effects for younger teens (ages 15–17) are twice the size of the effects for older teens (ages 18–19). Welfare reform contaminates analysis of the 1996–1997 round, but monthly estimates of the employment effects in that round resemble monthly estimates in the 1990–1991 round until welfare reform rolled out in the second half of 1997.

Employment Shocks and Demand for Pain Medication (I1)

Isabel Musse
,
University of Illinois-Urbana-Champaign

Abstract

Declining economic opportunity is often portrayed as one of the drivers of the opioid epidemic. Better employment conditions can, however, affect opioid use through two channels: increasing physical pain from working or reducing mental distress that can contribute to substance abuse. I use a large dataset of opioid and over-the-counter (OTC) painkiller sales to measure the effect of employment shocks on demand for pain medication. To separate the channels, I contrast the effect of labor demand shocks on the use of opioids with the effect on the use of OTC painkillers---which address pain but not mental health---allowing for the effects to depend on the injury rate of local industries. I find that a 1 percent increase in the employment-to-population ratio decreases the per-capita demand for opioids by 0.20 percent, while it increases the per-capita demand for OTC painkillers by 0.14 percent. To decompose the effect of employment on opioid use in the two channels, I calculate the substitution between these pain medications, exploring the introduction of a policy that increased requirements to prescribe opioids. My findings show that during local economic expansions, the decline in opioid abuse is 40 percent larger than the total effect on use while, at the same time, the demand for pain relief medication increases and is related to jobs in high injury industries.

Employment-maximising Minimum Wages (R1, J3)

Duncan H.W. Roth
,
Institute for Employment Research-Nuremberg (IAB)
Gabriel Ahlfeldt
,
London School of Economics and Political Science
Tobias Seidel
,
University of Duisburg-Essen

Abstract

Motivated by a reduced-form evaluation of the impacts of the German na-tionally uniform minimum wage on labour, goods and housing markets, wedevelop a simple model that nests a monopsonistic labour market with hetero-geneous firms. The model predicts that the employment effect of a minimumwage is a bell-shaped function of the minimum wage level. Consistent withthe model prediction, we find the largest positive employment effects in re-gions where the minimum wage corresponds to 48% of the pre-policy medianwage and negative employment effects in regions where the minimum exceeds80% of the pre-policy median wage. Our estimates provide first bounds for anevidence-based minimum wages that avoid detrimental employment effects.

Estimating Hysteresis Effects (E3)

Pål Bergset Ulvedal
,
Norges Bank
Francesco Furlanetto
,
Norges Bank
Ørjan Robstad
,
Norges Bank

Abstract

Standard Blanchard-Quah decompositions disentangle shocks with a permanent effect on output from shocks with only a temporary effect in VAR models. Shocks with permanent effect are normally interpreted as supply shocks (driving potential output) whereas shocks with temporary are interpreted as demand shocks. This interpretation of shocks rule out the possibility that demand shocks may have permanent effect on output through hysteresis effects.

Our paper extends the standard Blanchard-Quah decomposition, using a combination of sign and long-run restrictions, to disentangle supply and demand shocks with permanent effects from supply and demand shocks with temporary effects, in a VAR model estimated on US data.

We identify a demand shock that has permanent effects on the levels of output and employment. The shock explains a large fraction of the variance of these variables, indicating that hysteresis effects are important and that demand shocks may have permanent effects on output and employment.

Family Time Allocations over the Last Half Century (D1, E2)

Alexandros Theloudis
,
LISER-Luxembourg and University College London

Abstract

Over the last half century, married American women doubled their labor supply and halved the time they spend on household chores. By contrast, married American men reduced their labor supply and increased their involvement in chores. Both increased the time they devote to children but men, especially the most educated, did so by far more. What explains these dramatic changes in family time allocations? I develop a dynamic collective model that incorporates family time use and features financial and human capital, rich earnings dynamics, and endogenous divorce. The model quantifies the role of (i) the wage and education gender gaps; (ii) fertility and the cost to raise children; (iii) labor market experience; (iv) technical change in the household; and (v) features of the marriage market such as the expansion of divorce in the early 1970s. Preliminary results suggest that -at least- the gender wage gap and the divorce option are important for the observed time allocation dynamics. The latter implied an improvement in intra-household bargaining power among women born in the 1950s and a transfer of non-market work to their husbands.

Financial inclusion and patience: Longitudinal Evidence from Indonesia (D9, R0)

Jiaying Chen
,
Hong Kong University of Science and Technology

Abstract

Financial development plays a key role in economic growth, but there is little evidence on how greater access to financial services affects households’ patience when making intertemporal choices. Theory suggests that such choices are guided by interest rates when individuals freely access banking services and reflect underlying time preference parameters when individuals are credit constrained. This may lead to a reduction in the patience of individuals when the financial system expands, especially when saving in banks increases. This study is the first to empirically examine the effect of greater financial inclusion on patience using longitudinal data. Analyzing two waves of the Indonesian Family Life Survey that spans a period of rapid expansion of bank branches to rural villages, I find that a commercial bank branch in the village reduces measured patience, and the result remains robust using matching estimators. I also find evidence that financial access affects patience mainly by increasing access to deposit services rather than loans.

Firms’ Wage Structures, Workers’ Fairness Perceptions, Job Satisfaction and Turnover Intentions: Evidence from Linked Employer-Employee Data (J3, M5)

Christian Pfeifer
,
Leuphana University Lueneburg
Jens Mohrenweiser
,
Bournemouth University

Abstract

Relatively few studies analyze the effects of wage structures of firms, because administrative linked employer-employee data are necessary that include wage information about all workers and their characteristics to compute conditional wage structure variables. Even fewer studies look at the nexus between wage structures, worker attitudes (e.g., fairness perceptions), job satisfaction, and behavioral intentions (e.g., turnover intentions), because linked employer-employee data need to be supplemented with worker surveys. In this paper, we use novel data that allow us to link worker and establishment surveys with administrative social security data for all workers in the surveyed establishments. From these data, we can generate four wage structure variables which also account for interpersonal comparisons: workers’ own absolute wages, workers’ conditional internal reference wages within firms, the conditional wage dispersion within each firm, and workers’ conditional external reference wages across firms. We make three empirical contributions, which we combine in a mediation-style regression analysis. First, we are interested in the impact of firms’ wage structures on workers’ perceived wage fairness as an important organizational justice variable. Second, we are interested in the impact of firms’ wage structures on workers’ job satisfaction and turnover intentions, which are costly for firms because turnover intentions are positively correlated with actual quit behavior (turnover costs) and negatively correlated with work effort (productivity). Third, we are interested in the contribution of the fairness considerations on the overall effect of firms’ wage structures on workers’ job satisfaction and turnover intentions. The findings suggest that equity and social status considerations as well as altruistic preferences towards co-workers and inequality aversion are important, whereas the evidence for signal considerations is limited.

Frugality and Firms' Financial Flexibility: Evidence from Natural Disasters (G4, F3)

Matthew Wynter
,
University of Illinois-Chicago

Abstract

Firms headquartered in frugal countries tend to use shorter maturity debt. We hypothesize that when financing conditions worsen, having a larger share of debt due sooner increases firms’ need to readjust. We use natural disasters to identify periods when financing conditions worsen. We account for the economic impact of the disaster on firms’ growth opportunities by estimating disaster-specific cumulative abnormal returns (DisasterCARs). Controlling for DisasterCARs, we find that firms in frugal and less frugal countries issue similarly the year before the disasters, but afterwards the affected firms in frugal countries raise debt of shorter maturity, smaller stock proceeds, attempt to tap non-local capital markets, and invest less, suggesting that frugality can reduce firms’ financial flexibility.

Gender Biases in Education Investment: The Unexpected Effect of Reduction in Years of Compulsory Schooling in a Developing Country (J1, I2)

Ahmed Elsayed
,
Institute of Labor Economics (IZA)
Olivier Marie
,
Erasmus University Rotterdam

Abstract

We study the impact of reducing the cost of education on households’ human capital investment decisions for their children and how this differs by the child’s gender in a context of strong preference for sons. We exploit a unique policy reform in the education system of Egypt at the end of 1980’s which reduced compulsory schooling from nine to eight years, and consequently decreased the cost of education for families. Using rich dataset and comparing treated and non-treated cohorts within schools, we show that the policy not only had positive impact on the probability to finish basic education, but also unexpectedly improved the probability to finish post-compulsory schooling, and raised the average age at which students leave education. Most of this impact stems from educational improvements for girls from poorer/lower socio-economic background. We attribute this to changes in the behavior of credit constrained households who start investing more in their daughters’ education when costs of education are reduced. The results are robust to different specifications. We finally further show evidence that the policy had significant long-term positive impact on labor market outcomes, age and quality of marriage, and social empowerment for the treated women.

Gender Grading Bias at The University: Quasi-Experimental Evidence from an Anonymous Grading Reform (I2, J7)

Bjorn Tyrefors
,
Research Institute of Industrial Economics
Joakim Jansson
,
Linnaeus University

Abstract

In this paper, we first present novel evidence of female university students benefiting from being graded fairly, i.e., anonymously. This finding conflicts with previous results at the secondary education level. In contrast to the teacher gender composition at lower levels of education, teachers at the university level are predominantly male. Thus, an in-group bias mechanism could consistently explain the evidence at both the university and secondary education levels. We find that the treatment effect is driven by departments in which male teachers constitute the majority and estimate a significant in-group bias effect using randomly assigned graders. However, unexpectedly, further analysis shows that the treatment effect is essentially independent of the grader/student gender match. Hence, we document evidence of both grading bias stemming from in-group bias and other match independent factors, for example, shared culture and institutional factors.

Happy 18th Birthday, Now Leave: The Hardships of Aging Out of Foster Care (J1, Y4)

Alexa Prettyman
,
Georgia State University

Abstract

Over 20,000 youth age out of foster care each year in the United States without achieving permanency and unprepared to be adults: 20% will become homeless, less than 3% will receive a college degree, and 50% will still be unemployed by 24 years old. Under the Fostering Connections Act and state legislation, 22 states extended their age-out age beyond 18 years old between January 2012 and December 2016. I exploit the temporal and geographic variation of extended foster care to provide some of the first national evidence of the effects of federally-funded extended foster care on the transition to adulthood. Data comes from the National Youth in Transition Database, a longitudinal survey that collects information from foster care youth at 17, 19, and 21 years old, and is linked to the Adoption and Foster Care Analysis and Reporting System, which contains information about individuals’ foster care history. Using a linear probability model that controls for individual and state characteristics and a sample of 10,000 19-year-old foster youth nationwide, I find evidence that federally-funded extended foster care increases the likelihood that youth remain in foster care by 39%, reduces homelessness by 36%, and increases employment by 15%. Overall, these estimates are modest compared to comparison-of-means tests. The effects on college enrollment, disconnectedness (neither working nor enrolled in school), and incarceration are inconclusive. I also interact federally-funded extended foster care with individual experiences at 17 years old, such as homelessness and substance abuse, to better understand who benefits from this program. Federally-funded extended foster care appears to mitigate the negative effects of homelessness and substance abuse on college enrollment and disconnectedness. Foster care may not be a panacea for all vulnerable youth, but extended foster care appears to provide resources that aid in the transition to adulthood for some.

Inertia of Institutional Investors: Rational or Behavioral? (G2, G4)

Hugh Hoikwang Kim
,
University of South Carolina
Mohammad (Vahid) Irani
,
University of South Carolina

Abstract

We examine institutional investors’ tendency not to change their portfolio for an extended period, called portfolio inertia. Studying over 39 million investor-stock-quarter observations, we document that institutional investors do not trade a single share in one of five stocks in their portfolio for at least a quarter of the year. Trading costs do not fully explain this inertia behavior. We find that the inertia is associated with the inferior future performance of institutional investors. The results suggest that the inertia is driven by a potential behavioral bias, rather than a rational attention allocation strategy aimed at improving overall performance.

Information Design in Simultaneous All-Pay Auction Contests (D8, C7)

Zhonghong Kuang
,
Tsinghua University
Hangcheng Zhao
,
University of Chicago
Jie Zheng
,
Tsinghua University

Abstract

We study the information design problem of the contest organizer in a simultaneous 2-player 2-type all-pay auction contest environment, where players have limited information about own/others valuations of the prize. The contest organizer can send a public message to the contestants about their type distribution, in order to maximize expected total effort. We allow the players' ex-ante symmetric type distributions to be correlated, and the information disclosure policy to take the stochastic approach of Bayesian persuasion. The optimal design, the structure of which depends on the degree of the correlation of players' types, is completely characterized and shown to achieve higher effort than the traditional discrete type-dependent information disclosure policy. Given players' types are private information, if there is a strong positive correlation, an optimal design consists of two posteriors with one representing a perfect positive correlation and the other representing a positive correlation identified by a cutoff condition; if there is a weak positive correlation or negative correlation between types, the optimal design consists of two posteriors with one where both being high types is impossible and the other where a positive correlation is identified by the same cutoff condition. We also consider the case in which types are unknown to both players and the case in which the type information is asymmetric between the two players. Welfare comparisons are also conducted across different informational setups. Our work is the first study on full characterization of information design for games with two-sided asymmetric information and infinite action space.

Information Rigidity and Economic Uncertainty: A New Theory and Stylized Facts (D8, E3)

Romina Kazandjian
,
American University

Abstract

I propose a structural micro-founded sticky-noisy information model with high- and low-uncertainty regimes. Agents first appraise the state of uncertainty and only spend resources to update their inflation expectations if they perceive uncertainty as sufficiently high. Time-varying uncertainty affects expectation formation through two direct channels: 1) the "wake-up call" effect, which causes agents to pay more attention, increasing their quantity of information; and 2) the "wait-and-see" effect, which decreases their quality of information and prompts them to put less weight on new noisier information. Using structural estimation of alternative models with information frictions, I find that accounting for the indirect state-dependence channel, the proposed innovation of the model, better explains the observed information rigidity, since it considers the interaction between the two direct effects. A substantial amount of information rigidity is due to inattention, leaving ample room for policymakers to employ frequent, direct, and simple forward guidance to pierce the veil of inattention.

Institutional Design of Pension Systems and Individual Behavior: How do Households Respond? (G2, H7)

Renata Herrerias
,
Technological Autonomous University of Mexico (ITAM)
Guillermo Zamarripa
,
FUNDEF

Abstract

Mexico adopted a Defined Contribution (DC) Pension System with private individual accounts in 1997. We use the pension reform as a natural experiment to compare the expectations of policy makers at that time with the actual behavior of workers participating in the system. Pension rules were intended to encourage workers to assume full ownership of their pension accounts, to participate in the formal labor market (tax paying firms), and to induce additional pension savings. Using a panel data of Mexican workers, we find that individuals participate much less than required, the alternative of voluntary savings have almost been ignored, and 30 percent of affiliated workers do not claim ownership of their pension accounts. We analyze the dynamic of workers’ behaviors finding that men, higher income workers, workers with housing credit and individuals living in richer regions are the ones most likely to comply with the rules. Worker characteristics and local labor market rigidities prevailed over pension incentives. The consequence is that only one third of participants are expected to receive lifetime pension income under the reformed system. We conclude that pension policy design should consider as a restriction the structure of local labor markets and not the other way.

Is the RMB a Safe-haven Currency? Evidence from Conditional Coskewness and Cokurtosis (F3, G1)

Xin Cheng
,
Xiamen University
Yinggang Zhou
,
Xiamen University

Abstract

We examine whether RMB is a safe haven currency in terms of its hedging effectiveness of financial stress for global or Asian equity investors. RMB coskewness (covariance between RMB premium and equity volatility) is mostly negative, implying that RMB is not a good hedge in the volatile market. Moreover, positive RMB cokurtosis (covariance between RMB premium and equity skewness) implies that RMB is unable to hedge against stock market crash. Neither RMB coskewness nor cokurtosis are priced, suggesting equity investors with skewness and kurtosis preferences would not use RMB to hedge against financial stress. Therefore, RMB is not a safe haven currency yet.

Is Work a Burden? The Role of the Living Standard (J0, I3)

Jianbo Luo
,
State University of New York-Buffalo

Abstract

Many mainstream schools of economics argue that work is a burden, while nonmainstream schools argue that this might not be entirely true. This paper aims to reconcile the differences by suggesting that individuals will balance income and leisure only after the necessary expense for their current living standard is met. Thus, whether work is a burden depends upon two criteria. (1) In terms of marginal utility, before (after) such expense is met, the marginal utility per labor hour is positive (negative), i.e., work is not (is) a burden. (2) In terms of total utility, the total utility provided by labor is positive, i.e., work is not a burden. Three applications show that the above explanation can reconcile different historical perspectives, explain various discrepancies about labor supply between neoclassical theory predictions and empirical findings, and reconcile the different interpretations about lottery winners’ labor supply.

Liquidity Spillover in the Foreign Exchange Market (C3, G1)

Ya-Ting Chang
,
National Central University
Yin-feng Gau
,
National Central University
Chih-Chiang Hsu
,
National Central University

Abstract

We use an index of spillover based on the generalized variance decomposition developed by Diebold and Yilmaz (2009, 2012, 2014) to measure the connectedness in liquidity in nine currency markets during 2008 and 2013. The results of spillover tables and indices demonstrate how the liquidity of one market spillovers to and from other currency markets. We find that the market of Euro against the US dollar is the main net transmitter in liquidity to the other foreign exchange markets.

Local Economic Impacts of Legislative Malapportionment (E6, H3)

Masami Imai
,
Wesleyan University

Abstract

Malapportionment, or unequal legislative representation, is a highly contested, and yet a common and persistent feature of electoral systems in many countries where more delegates per capita are granted to rural, sparsely populated, and economically struggling regions. Does legislative representation matter to local economies, and if so, how much? In Japan, the Lower House seats were severely malapportioned until an electoral reform substantially equalized the geographical distribution of representation for the 1996 election. We use this episode as a quasi-experimental setting to investigate the causal effect of malapportionment on the relative performance of local economies. There are two main results. First, an additional seat in the Lower House significantly expands local governments’ fiscal space. An extra delegate is associated with more fiscal transfers, confirming Horiuchi and Saito (2003), and it also leads to more borrowing and more spending (largely on public capital), suggesting strong flypaper effects. Second and perhaps more interestingly, over-represented communities ultimately do not seem to benefit from this political and fiscal gift. We detect no discernible effects of legislative representation on establishment or employment. Our null results do not seem to be driven by omitted variable bias or general noise in the data. Rather, it is due to crowding-out effects in local labor markets. An additional representation and (the resulting additional transfers) produce more construction and public sector jobs, and yet these positive effects are entirely offset by comparable losses of jobs in other sectors.

Long-term Effects of Childhood Exposure to Persecution: Human Capital, Marriage Market, and Intergenerational Outcomes (J1, N4)

Xuechao Qian
,
Ohio State University

Abstract

There were substantial class struggles in mid-20th Century China for more than two decades (1950-1976) in the form of persecution of certain groups of people, such as landlords, capitalists, and intellectuals. This paper investigates the impacts of early life exposure to persecutions during the class struggle period on human capital development and marriage market outcomes, and it analyses how such impacts depend on exposure happened in the life course. Using a difference-in-differences strategy, I show that individuals with longer exposure to persecutions in early childhood (0-6 years old) completed less formal schooling, have worse verbal and math skills, and earn lower incomes in the long-term. Furthermore, they are more likely to marry people from classes favored by the regime but with poorer human capital outcomes. While males are affected by the exposure in early childhood more through direct impact on human capital development, females are impacted more indirectly through marriage.

Macroeconomic Effects of Political Risk Shocks (E3, C3)

Sinem Hacioglu Hoke
,
Bank of England

Abstract

We investigate the macroeconomic effects of political risk in an information-rich SVAR. Using an external instrument based on an index of US partisan conflict for identification, we find that reduced political risk has an expansionary impact: it is immediately priced into stock prices; increases firms' credit availability, employment and investments while households invest and consume more - ultimately output rises. As an important driver of economic dynamics in medium to long term, the shock creates an aggregate supply effect where output and inflation move in opposite directions and generates a trade-off between inflation stabilization and output growth during turbulent periods.

Managing GDP Tail Risk (E5, G1)

Thibaut Duprey
,
Bank of Canada
Alexander Ueberfeldt
,
Bank of Canada

Abstract

We propose a novel framework to analyze how policy-makers can manage risks to the median projection and risks specific to the tail of GDP growth. By combining a quantile regression of GDP growth with a VAR, we show that monetary and macroprudential policy shocks can reduce credit growth and thus GDP tail risk. So, policy-makers concerned about GDP tail risk would choose a tighter policy stance at the expense of macroeconomic stability. Using Canadian data, we show how our framework can add tail event information to projection models which ignore them and give policy-makers a tool to communicate the tradeoffs they face.

Online Platform for In-Home Healthcare: Customer Defection and Platform Exploitation (L1, L8)

Kris Zhou
,
University of Texas-San Antonio
Bradley J. Allen
,
University of Arkansas
Richard T. Gretz
,
University of Texas-San Antonio
Mark B. Houston
,
Texas Christian University

Abstract

A major concern for managers and policy makers in online service platform industries (e.g. Zeel, Amazon Home Services, Freelancer.com) is platform exploitation, the opportunistic behavior of service agents defecting with customers to avoid platform fees. Existing theoretical models of platforms overlook the possibility of exploitation. We examine this phenomenon using data from an in-home healthcare platform that connects nurses and patients in China. Results are consistent with high-quality and long-tenure nurses engaging in exploitation. A standard deviation increase in either nurse quality or nurse platform tenure decreases patient retention on the platform by either 0.033 (~6.73%) or 0.036 percentage points (~7.33%), respectively. Retention increases then decreases as nurse quality improves; this relationship is more pronounced at higher prices. The negative effect of nurse tenure on retention is greater at higher prices. Finally, characteristics that decrease patient retention also reduces nurse return rate to the platform, suggesting that high-quality and long-tenure nurses are less likely to use the platform as they serve more customers privately. Our results contrast with prior research that emphasizes the importance of service agent quality and tenure for platform success.

Pension Incentives and Labor Supply: Evidence from the Introduction of Universal Old-Age Assistance in the UK (J2, H2)

Matthias Giesecke
,
RWI-Leibniz Institute for Economic Research
Philipp Jäger
,
RWI-Leibniz Institute for Economic Research

Abstract

We estimate the labor supply response to the introduction of means-tested minimum pensions in the UK through the Old-Age Pension Act (OPA) of 1908. The OPA was a major social policy intervention and the first one to universally target older workers in a time of very limited social protection. Using full-count census data covering the entire population, we identify the labor supply effects of the program based on variation at the age-based eligibility threshold between 69 and 70. Our results show a considerable and abrupt decline in labor force participation of 5.3 to 6.0 percentage points when older workers turn 70. This sudden drop only occurs at the age cutoff and only after the OPA was implemented. The unique historical setting permits us to study unintended labor supply responses when labor earnings are taxed implicitly through government transfers.

Private Equity and Taxes (G3, H2)

Peter Heinz Severin
,
University of Mannheim
Marcel Olbert
,
University of Mannheim

Abstract

We study companies' tax avoidance behavior after being acquired in a private equity transaction. Exploiting rich European firm-level data in a matched-sample difference-in-differences setting, we find that target companies' effective tax rates decrease by 13 percent. This finding is in line with the hypothesis that private equity investors create shareholder value by extracting money from the government. While our evidence suggests that target firms engage more heavily in profit shifting, we do not find strong evidence for a tax-motivated leverage channel. Target firms experience lower asset and productivity growth when they engage in significant tax avoidance after the deal.

Prospect Theory and Currency Returns (G4, F3)

Qi Xu
,
Zhejiang University
Roman Kozhan
,
University of Warwick
Mark Taylor
,
Washington University-St. Louis

Abstract

We empirically investigate the role of prospect theory in the FX market. Using the historical distribution of exchange rate changes, we construct a currency-level measure of prospect theory value and show that it forecasts future currency excess returns. High prospect theory value currencies significantly underperform low prospect theory value currencies. The predictability is higher during periods of excessive speculative demand of irrational traders and when arbitrage is limited. These findings are consistent with the hypothesis that investors mentally represent the portfolio of currencies by their historical distributions or charts and evaluate this distribution in the way described by prospect theory.

Redesigning Unemployment Insurance: Who to Target and How to Trigger (J6, I3)

Rachel Sederberg
,
Bowdoin College

Abstract

The United States Unemployment Insurance system is a federally funded and state administered program meant to assist in smoothing consumption during spells of involuntary joblessness. In the absence of extenuating circumstances, UI benefits are available for 26 weeks on average, with some deviations in a few states. In times of economic distress, additional weeks of UI benefits may become available if certain criteria are met. These benefits, known as Extended Benefits or Emergency Unemployment Compensation may become available to qualifying individuals in a given state dependent on the state’s economic situation meeting certain criteria. These one-size-fits-all mechanisms largely do not account for across state variation in labor market conditions and characteristics, which may have large impacts on the distress felt by unemployed workers. This paper seeks to ascertain the efficacy of previously employed triggering mechanisms for extended benefits, as well as investigate the appropriateness of a one-size-fits-all base UI program. When applicable, suggestions regarding efficiency improving changes to UI and extended/emergency benefit programs are made.

Sophistication and Cautiousness in College Applications (C7, D4)

Kentaro Tomoeda
,
University of Technology Sydney
Yan Song
,
Jinan University
Xiaoyu Xia
,
Chinese University of Hong Kong

Abstract

In many school choice and college admissions markets, matching mechanisms have been reformed from the Immediate Acceptance (IA) mechanism to some variants of the Deferred Acceptance (DA) mechanism. In order to evaluate these policy reforms, it is essential to investigate what (behavioral) models could explain the major impact of the reforms. In this article, we provide an answer to this question in the context of Chinese college admissions reforms, where the IA mechanism has been replaced by the Chinese parallel mechanisms (Chen and Kesten, 2017). First, we show that the reforms would not affect the matching outcome if students played the equilibrium, but our data exhibit a clear pattern that the matching became more assortative under the parallel mechanisms. Motivated by this observation, we extend the model in two dimensions: (i) heterogeneous beliefs and (ii) strategic sophistication. We identify and estimate the fractions of the student behavioral types, and show that both dimensions are important in explaining the patterns of the data. In our counterfactual simulations, we show that the distributional effects of the reforms on the student welfare differ from what they could have obtained from introducing the DA mechanism.

Stock Liquidity and Corporate Diversification: Evidence from Index Reconstitution (G1, G3)

Seungjoon Oh
,
Peking University
Hursit Celil
,
Peking University
Srinivasan Selvam
,
Peking University
Xueying Shangi
,
Peking University

Abstract

Using the Russell index reconstitution as an exogenous shock to stock liquidity, we examine how stock liquidity affects corporate diversification in a regression discontinuity framework. We find that a positive liquidity shock to firms included in the Russell 2000 index reduces diversification. Furthermore, post index-inclusion, these firms also rely less on internal capital markets. The causal inference can be linked to two mechanisms: improvements in the information environment and firm's governance quality, which help to lower financing frictions, manager's wealth diversification needs, and empire-building tendencies. Overall, financial markets by affecting demand for firm-specific information and governance quality can shape a firm's boundaries and its scope.

The Effect of Property Rights on Capital Structure: Evidence from a Chinese Natural Experiment (K1, G3)

Yixin Liu
,
University of New Hampshire
Yu Liu
,
University of Texas-Rio Grande Valley
William L. Megginson
,
University of Oklahoma
Zuobao Wei
,
University of Texas-El Paso

Abstract

We examine how changes in property rights security impact firm capital structure decisions by exploiting a natural experiment, the enactment of China’s Property Rights Law in 2007 (the Law). Using a large dataset of non-listed firms, we document a significant overall decrease in leverage after the Law’s passage, consistent with the reinvestment hypothesis’ prediction that owners are willing to reinvest more of their profits as property rights protection strengthens, substituting for external debt. Difference in difference analyses document that relatively more financially constrained firms experience an increase in leverage following the Law’s enactment, consistent with the financial constraint hypothesis’ prediction that lenders become more willing to extend credits to constrained firms as creditor rights are strengthened. For listed firms, we verify that leverage also declines after the Law’s passage, but financial constraints do not significantly impact the Law–leverage relation.

The Effect of the Child Care Tax Credit on Maternal Labor Supply (J2, H2)

Haibin Jiang
,
Clemson University

Abstract

The Child Care Tax credit (CCTC) is a child care subsidy program that allows working parents to claim a tax credit for their child care expenses. In the study, I document a comprehensive legislative history of the CCTC enactments, amendments, and repeals at both federal and state levels. Using the CCTC variation generated by exogenous law changes and focusing on mothers between the ages of 20 to 55 in the Panel Study of Income Dynamics, I estimate the effect of the CCTC on maternal labor supply. Using differences-in-differences estimation and policy shocks measured at the state level, I found that a \$1000 increase in the CCTC increases the maternal labor force participation rate by three percentage points. To incorporate the policy variation within a state, I construct an individual CCTC treatment variable for each woman using observable individual characteristics and state-specific regulation. I use the state-level CCTC treatment variable as the instrumental variable for the individual CCTC treatment variable and find that a \$1000 increase in CCTC increases the maternal labor force participation rate by six percentage points. The effects are more pronounced in married mothers than single mothers.

The Long-Term Effects of Labor Market Entry in a Recession: Evidence from the Asian Financial Crisis (J2, E3)

Eleanor Jawon Choi
,
Hanyang University
Jaewoo Choi
,
Korea Development Institute
Hyelim Son
,
University of Seoul

Abstract

This paper investigates the long-term effects of initial labor market conditions by comparing cohorts who graduated from college before, during, and after the 1997–1998 Asian financial crisis. We measure the overall welfare impact by examining not only labor market activities but also family formation and wealth accumulation. Using data from 20 waves of the Korean Labor and Income Panel Study, we find a substantial and persistent reduction in employment, earnings, marriage, fertility, and financial assets among men who graduated in a bad economy. For women, limited job opportunities at graduation result in an increase in childbearing.

The Unintended Effects of California’s Paid Family Leave Program on Children's Birth Outcomes (I1, O0)

Feng Chen
,
Tulane University

Abstract

Over the past decades, the United States has witnessed increasing changes in family structure, the labor force market, and demography. The federal government and some states have taken different avenues in the way to help working mothers balance work and family in the form of family leave policy. California is the first state to provide paid family leave program, which offers 6 weeks postnatal paid leave to employed mothers. The precautionary saving theory states that people may delay consumption and save in the current period due to the lack of completeness of insurance markets. This theory can also be applied to leave-taking. Pregnant women may delay leave-taking and save their paid leave before childbirth due to limited paid leave after the childbirth. Baum and Ruhm (2016) using data from survey NLSY97 finds that mothers in California with the availability of paid family leave take around one extra week of leave in their quarter before birth. Evidence from the Survey of Income and Program Participation also found that mothers in California take 3.7% more prenatal paid leave after the program. This paper uses natality data from the National Vital Statistics System and a Difference-in-Difference method to evaluate the Unintended Effects of the paid family leave program of California on children’s birth outcomes. The results show that California’s paid family leave decreases the premature birth rate by 0.29 percentage points or 2.8 percent. One possible mechanism for these effects is that mothers who take more prenatal leave are more likely to begin prenatal care earlier, increase their doctor visits, and hence reduce potential risks that adversely affect birth outcomes. Heterogeneous analysis indicates that the effects are particularly large on children of African American mothers and on mothers without a high school degree, who are more likely cannot afford to the unpaid prenatal leave and substitute their postnatal paid leave for prenatal leave.

Threats and Opportunities in the Digital Era: Automation Spikes and Employment Dynamics (O3, F1)

Giacomo Domini
,
Erasmus University Rotterdam
Marco Grazzi
,
Catholic University of the Sacred Heart
Daniele Moschella
,
Sant'Anna School of Advanced Studies
Tania Treibich
,
University of Maastricht

Abstract

This paper investigates how investment in automation-intensive goods impacts on worker flows at the firm level and, within firms, across occupational categories. Resorting to an integrated dataset encompassing detailed information on firms, their imports, and employer-employee data for French manufacturing employers over 2002-2015, we identify `automation spikes' using imports of intermediates embedding automation technologies and then test their impact on employment dynamics. We find that automation spikes are positively correlated with preceding and contemporaneous growth in employment, mainly due to lower separation rates of investing firms. These differential patterns of net and gross worker flows do not appear to change significantly across different types of workers (occupational categories, `techies', routine-intensive vs. non routine-intensive jobs).

Trade War and Social Welfare: A Structural Model of the United States Solar Industry (Q5, L1)

Wenjun Wang
,
Agricultural Bank of China

Abstract

As a green industry, the solar power sector has grown rapidly over the past 10 years, with capacity soaring from 6,660 MW in 2006 to 229,300 MW in 2015 worldwide. In October 2012, a trade war broke up between US and China, and the US imposed anti-dumping and countervailing duties on imported solar cells from China. The trade war in the solar industry has been widely discussed, but little is know about its total economic cost. In this paper, I fill this gap and estimate the incidence of US anti-dumping policy among different market participants.

My data come from the Lawrence Berkeley National Laboratory (LBNL)’s Tracking the Sun report series. This data set provides household-level information on almost all of the solar PV installations in the US market for the period 2010 - 2015. On the demand side estimation, I use BLP instrument as the IV for the solar price; on the supply side, I model the vertical structure of the industry and explicitly account for the strategic behaviors of manufacturers and installers. The demand and supply parameters are estimated jointly by using Generalized Methods of Moments (GMM).

Based on the estimated model, I conduct two counterfactual simulations regarding different hypothetical changes on anti-dumping duties and subsidy rates. The results of my simulations show that the installation capacity in the US solar market would expand by 24.5% if there were no anti-dumping policies. Compared with the big losses in Chinese solar manufacturers, the US manufacturers have only gained small profits from the anti-dumping policy. The US-China trade war has also brought non-trivial damage to the environment, in the form of increased greenhouse gas emission and air pollution.

Understanding Bank and Nonbank Credit Cycles: A Structural Exploration (E3, G2)

Bora Durdu
,
Federal Reserve Board
Molin Zhong
,
Federal Reserve Board

Abstract

We explore the structural drivers of bank and nonbank credit cycles using a medium-scale DSGE model with two types of financial intermediation. We posit economy-wide and sectoral disturbances in both macro and financial sectors. We estimate that sectoral shocks to the balance sheets of entrepreneurs are important for fluctuations in bank and nonbank credit growth at the business cycle frequency. Economy-wide entrepreneurial risk shocks gain predominance for explaining the lower frequency co-movement between the two series. Macro shocks play very little role in explaining financial cycles.

Uninsurance and Purchases of Prescription Drugs with High Abuse Potential: Evidence from the Federal Dependent Coverage Mandate (I1)

Michael DiNardi
,
University of Rhode Island

Abstract

In recent years, use of prescription drugs with high abuse potential such as central nervous system depressants, opioids, and stimulants rose in the United States. Because health insurance lowers the cost of purchasing prescription drugs, losing coverage may cause individuals to forgo treatment and decrease prescription drug consumption which could reduce health and increase the likelihood of overdose and death if individuals substitute to risky drugs sold on the black market. Using a regression discontinuity design, I estimate the effect of aging out of health insurance from the Affordable Care Act’s dependent coverage mandate at age 26 on legal purchases of prescription central nervous system depressants, opioids, and stimulants. Individuals are 0.8-1 percentage points purchase a prescription central nervous system depressant and 1 percentage point less likely to purchase a prescription opioid after turning 26. These changes are driven by women. A back of the envelope calculation suggests that at age 26, up to 44,000 women may suffer adverse health effects from a reduction in the use of central depressants and up to 35,000 may suffer worse health from a reduction in the use of opioids. Estimated effects for men are generally negative but imprecise.

Who Disseminates Foreign Information Into Stock Prices? (F3, G4)

Wei Jiao
,
University of Wisconsin-Green Bay

Abstract

As the economy is increasingly globalized, firms are operating and competing on a global scale. Inevitably, foreign information is of growing importance on firm value and understanding how foreign information disseminates into stock prices becomes critical. This paper finds that portfolio managers of active international mutual funds can disseminate foreign information across countries into U.S. multinationals’ stock prices. Investing in non-U.S. markets helps these managers gather foreign information that is relevant to U.S. multinationals’ foreign operations. When these managers use their foreign information to trade U.S. multinationals’ stocks, they move stock prices and hence disseminate foreign information. In doing so, they generate abnormal returns of 0.5% monthly from investing in U.S. multinationals. Through disseminating foreign information, these managers increase U.S. multinationals’ stock price informativeness on foreign information, which in turn improves U.S. multinationals’ real economic efficiency.

“Law of One Price” in the Internet Era ---- Search Cost, Platform Competition and Customer Lock-in (L1, D4)

Zhen Sun
,
Tsinghua University
Jianping Liu
,
Tsinghua University
Taoxiong Liu
,
Tsinghua University

Abstract

In this study, we use a unique dataset from Tsinghua University's iCPI database to analyze the price dispersion in the online market. The results show that the cross-platform price dispersion of the same product is a pervasive and stable phenomenon. The dispersion on weekdays, during which consumers incur higher search cost, is higher than that during the weekends. Moreover, we find a positive relationship between price dispersion and the number of platforms on which a product is listed, contrary to the classic industrial organization theory that competition leads to price convergence. We conjecture that “platform lock-in”, the phenomenon that consumers are “locked” to particular platform due to its closed-circle advertising and other related (e.g., financial support) services. These findings indicate that search costs and “platform lock-in” are two important factors for the price dispersion in online market, while platform competition cannot reduce the dispersion. The dynamics of price dispersion following a change in the number of platforms suggest that platforms do not strategically adjust prices in response to the change in market structure, which provides support for the platform lock-in hypothesis. Overall, our findings suggest that in China's highly developed online market, “platform lock-in” is a phenomenon that prevents market integration in the internet era. These findings have important implications for platform regulation and governance.

A Macroeconomic Model with Interbank Markets and Regulated Banks (G2, G0)

Chao Huang
,
University of Edinburgh
Fernando Moreira
,
University of Edinburgh
Thomas Archibald
,
University of Edinburgh

Abstract

We propose an enhanced dynamic model in which counterparty banks suffer idiosyncratic liquidity shocks and are connected by an interbank market for liquidity risk management. The aim of this paper is to shed new light on the impact of the interbank market on the bank lending activity to entrepreneurs, under different requirement regimes. Banks are subject to loan investment risk, consisting of both permanent and transitory components, and idiosyncratic liquidity shocks leave them in liquidity surplus or liquidity deficit. We find that liquidity-deficit banks will be more likely to be affected by the requirement regime constraints, which restrict their interbank borrowing volume; consequently, under stricter (looser) regimes, like a liquidity (capital) requirement, there is a supply (demand) surplus in the interbank market. The increase in the transitory risk volatility, keeping the total volatility constant, will noticeably deactivate the interbank market; however, the effect of a permanent risk volatility increase is insignificant. This finding indicates that the interbank market is more sensitive to the transitory effects of financial shocks. We also find that once the liquidity supply or demand is not satisfied by the interbank market, banks will be prone to curtail lending, especially in the case of the liquidity-deficit banks; therefore, we argue that central bank intervention in the distressed banks will help to sustain the loan investment scale and contribute to social welfare growth. In normal times, the liquidity requirement (represented by the Liquidity Coverage Ratio) lowers social welfare by limiting bank lending and interbank participation, while in extreme scenarios such as financial crises, the liquidity requirement is more effective in creating social welfare than the capital regulation by stabilizing the bank lending activity.

A Model of Investment-and-Marriage with Differential Fecundity (D1, J1)

Hanzhe Zhang
,
Michigan State University

Abstract

This paper provides a model that can explain why women go to college at higher rates but have lower average incomes than men in the US and other countries. Differential fecundity and a general-equilibrium marriage-market effect form the basis of our explanation. The model can also explain the relationships between age at marriage and personal midlife income for men and for women as well as the relationship between age at marriage and spousal midlife income for women. Our explanations for these facts are supported by empirical evidence and calibration results.

A Natural Experiment Test for Asset Pricing Models (G1)

Tongbin Zhang
,
Shanghai University of Finance and Economics
Renbin Zhang
,
Autonomous University of Barcelona

Abstract

The high and volatile AH premium in the connected Chinese stock markets provides a natural experiment to test asset pricing models. We show that various present-value asset pricing models, in which stock prices are determined only by fundamentals, are rejected because of their difficulties in explaining AH premium. We find that an internal rationality learning model can be squared with the AH premium and show the importance of subjective stock price expectations in equity pricing. Convergence traders are highly likely to suffer a big loss in this situation.

A Race to the Top? Staggered Electoral Cycles and Strategic Interactions in Business Taxes (H7, H2)

Sebastian Garmann
,
Ruhr-University Bochum

Abstract

Many theories predict strategic interactions in setting taxes across neighboring jurisdictions. The few papers that use quasi-experimental settings for causal empirical evidence come to very diverse results, and seldom study the source of interactions. I add to this literature by exploiting staggered election dates, which create variation in neighbor’s tax rates that is driven by election-year manipulations rather than reverse causality or spatially correlated omitted variables. I find that municipalities only mimic the tax rate increases that occur in their neighbor’s post-election years, while they do not mimic the tax rate decreases that occur in election years. Such a behavior would lead to a race to the top in tax rates. Consistent with yardstick competition, tax increases in neighboring municipalities create a window of opportunity for politicians to implement fiscally necessary tax increases in their own municipalities.

A Spatial Panel Quantile Model with Unobserved Heterogeneity (C3, E4)

Lina Lu
,
Federal Reserve Bank of Boston
Tomohiro Ando
,
University of Melbourne

Abstract

This paper introduces a spatial panel quantile model with unobserved heterogeneity. The proposed model is capable of capturing high-dimensional cross-sectional dependence and allows heterogeneous regression coefficients. For estimating model parameters, a new estimation procedure is proposed. When both the time and cross-sectional dimensions of the panel go to infinity, the uniform consistency and the asymptotic normality of the estimated parameters are established. In order to determine the dimension of the interactive fixed effects, we propose a new information criterion. It is shown that the criterion asymptotically selects the true dimension. Monte Carlo simulations document the satisfactory performance of the proposed method. Finally, the method is applied to study the quantile co-movement structure of the U.S. stock market by taking into account the input-output linkages as firms are connected through the input-output production network.

Adding Fuel to the Fire Sales: Banks, Capital Regulation, and Systemic Risk (G1, G2)

Samuel Rosen
,
Temple University

Abstract

In a model with heterogeneous banks and endogenous fire sales, the tightening of bank capital regulation can aggravate fire sales, leading to larger bank losses and higher systemic risk. When calibrated to the data, the least costly policies to mitigate systemic risk raise both ex ante capital requirements and ex post shortfall penalties. These policies also assign relatively higher capital requirements to banks that can better offset price declines during a fire sale, consistent with the recently implemented capital surcharge for global systemically important banks (G-SIBs). My findings provide further support for leading-edge macroprudential tools, including stress tests and countercyclical capital buffers.

Adverse Selection and Switching Costs in Health Insurance Marketplaces: Using Nudges to Fight the Death Spiral (L5, I1)

Krisztina Horvath
,
Boston College

Abstract

In health insurance markets where regulators limit insurers' ability to price on the health status of individuals, the traditional regulatory intervention to protect the market from adverse selection and expand coverage among young and healthy people is the implementation of an insurance mandate. In this paper I analyze an alternative, behavioral solution in the context of the Affordable Care Act Marketplaces: the automatic enrollment of the uninsured. I build a theoretical model which shows that this nudging policy is welfare improving under inefficiently low coverage rates, and the size of its benefit depends on the strength of consumer inertia. Using an individual-level panel dataset on health insurance plan choice and claims from the Colorado All Payer Claims Database, I estimate a structural model of health insurance demand and supply in the presence of switching costs. Simulating the effects of the policy, I find that auto-enrollment increases coverage rates and reduces premiums, generating a gain in consumer surplus. Moreover, I show that it is essential to take into account the heterogeneity of preferences when designing default plans to auto-enrolled individuals. Defaulting everyone into the same contract type leads to more quitting due to inefficient matching and it may also indirectly increase adverse selection on the intensive margin through the price adjustment mechanism. The results of this paper suggest that in order to avoid these problems and maximize the welfare gains of automatic enrollment in health insurance markets, it is important to design smart default policies.

Affirmative Action Policies and Interracial Marriage (J1, J1)

Shiyi Chen
,
University of Connecticut

Abstract

During the 1970s and 1980s many states had passed their own affirmative action law, which created a unique opportunity to study how state affirmative action laws affect personal relationships between people of different races. This paper examines the impact of these policies on white/black interracial marriage for public-sector employees using the 2008-2017 American Community Survey. The empirical analysis exploits time and state variations in initiating state affirmative action law, to estimate triple-difference regression of the implications of interracial marriage. I find that the probability of a white male working in the public sector has black spouse increases by 0.12 percentage points if he married in a year when there was an affirmative action policy in place in his state of residence. However, the policies do not appear to affect the likelihood that a white female marrying a black husband. Besides, affirmative action laws decrease the possibility that black males or females marry a white spouse. These results suggest that state affirmative action laws improve race relations, but with gender and racial differences.

Agricultural Productivity Shocks and Poverty in India: The Short and Long-Term Effects of Monsoon Rainfall (C3, O1)

Matthias Sebastian Hertweck
,
Deutsche Bundesbank

Abstract

This paper examines the dynamic effects of monsoon rainfall shocks on yield, wages, and prices in the Indian agricultural sector. We distinguish between positive and negative rainfall shocks and explicitly consider their spatial dimension (local/regional). We find that particularly negative regional shocks exert adverse effects. The enormous drop in agricultural yield is short-lived, but elicits a persistent decline (increase) in wages (food prices). Negative local shocks affect only wages, but not prices. This indicates that, in the food market, intra-regional trading mitigates the impact of local shocks. However, in the labour market, the arbitrage mechanism through migration appears substantially weaker.

Antitrust Policy in a Globalized Economy (E1, F6)

Linyi Cao
,
Washington University-St. Louis
Lijun Zhu
,
Peking University

Abstract

The antitrust policy has been relaxed and the number of Mergers and acquisitions
(M&A) has risen rapidly since the 1980s in the United States. This paper provides a
framework to evaluate cost and benefits of antitrust policy in a global context. M&A
reallocate resources from small to large and typically more productivity firms, while
also increase monopoly power of the latter. Optimal antitrust policy seeks a balance
between the positive productivity effect and the negative markup effect. In a globalized
economy, increasing productivity fully accrues to domestic firms/workers while
a higher markup only partially hurts domestic consumers. A weakening antitrust policy
since the 1980s is thus an optimal response to the increasing globalization in the
same period. We present a dynamic general equilibrium model of M&A, and show
that welfare, measured as aggregate consumption/production in stationary equilibrium,
is a hump-shape function of the antitrust policy parameter in model. In a work
in progress, we are extending the model to an open economy, and aim to formalize
the intuition that open to trade demands a more lenient optimal antitrust policy and
explore its quantitative implications for aggregate markup and welfare.

Are Firms with Female CEOs More Environmentally Friendly? (Q5, G3)

Zigan Wang
,
University of Hong Kong
Luping Yu
,
University of Hong Kong

Abstract

In this paper, we document a previously unknown benefit of women’s role in firm management: the enhancement of environmental protection. Through a panel data analysis, we find that firms with female CEOs produce less air and water pollution and greenhouse gas emissions, and receive fewer environmental penalties, compared to firms with male CEOs. Our difference-in-differences analysis shows that firms also reduce air and water toxic releases, greenhouse gas emissions, and receive fewer environmental penalties after experiencing a male-to-female CEO transition. Moreover, firms demonstrate higher awareness of environmental protection, reflected in their 10-K filings, when being led by female CEOs.

Are Private Health Care Providers Politically Discriminated Against? (I1, H4)

Adam Pilny
,
RWI-Leibniz Institute for Economic Research
Felix Roesel
,
Ifo Institute

Abstract

We identify causal effects of political distortions on the allocation of public capital funds in the German hospital market. Particularly, we investigate whether private hospital providers are politically discriminated against in the granting procedure. In Germany, federal states are responsible for funding capital costs of hospitals. State governments can exclusively decide on the amount of funds to be allocated and which hospitals to fund. They are legally obligated not only to fund public but also private providers, i.e. equal treatment of all ownership types is required by law.
We assess the effects of electoral cycles and the governments’ ideology (i.e. left-wing versus right-wing) on the allocation of public capital funds. We exploit self-compiled historical data of the German hospital market between 1955 and 2017 and link this information with political data, budget accounts, and socioeconomics of German states. We estimate regression discontinuity design (RDD) models to identify ideology-induced effects on (i) the volume and (ii) the budget forecast errors of allocated capital funds. Thus, we investigate whether a left-wing government with a tight majority of seats in the state parliament exhibits a different pattern in capital funding than a right-wing government with a tight majority of seats in parliament. For examining the case of political discrimination against certain types of hospital owners, we differentiate the effects by public versus private providers.
Preliminary results show a sufficient variation in budget forecast errors between public and private providers, i.e. that the actual funding in many states does deviate from the scheduled amounts to be funded according to the state budget plans. First results reveal a significant negative effect of left-wing governments on budget forecast errors, indicating that these governments seem to cut the volume of actual funds. By January further results will be shown.

Asset Prices, Corporate Actions, and Bank of Japan Equity Purchases (E5, E2)

Ben Charoenwong
,
National University of Singapore
Randall Morck
,
University of Alberta
Yupana Wiwattanakantang
,
National University of Singapore

Abstract

Since 2010, the Bank of Japan (BOJ) has purchased stocks to boost domestic firms’ valuations to increase GDP growth. The stock return elasticity with respect to BOJ purchases relative to the previous month’s market cap is around 2 and increases across longer horizons. Over one quarter, BOJ share purchases worth 1% of assets correspond to an increase of 1% in share valuation and a .27% increase in assets. Consistent with elevated valuations letting firms “cash out,” BOJ share purchases predict equity issuances and fewer stock buybacks. However, less than 9% of increased assets reflect net tangible capital investments, whereas cash and short-term investments account for over 50%. This unconventional monetary stimulus thus boosts share prices but is largely not transmitted into real investment growth.

Asset Safety Versus Asset Liquidity (E4, G1)

Athanasios Geromichalos
,
University of California-Davis
Lucas Herrenbrueck
,
Simon Fraser University
Sukjoon Lee
,
University of California-Davis

Abstract

Recently, a lot of attention has been paid to the role “safe and liquid assets” play in the macroeconomy. Many economists take as given that safer assets will also be more liquid, and some go a step further by practically using the two terms as synonyms. However, they are not synonyms: safety refers to the probability that the (issuer of the) asset will pay the promised cash flow, and liquidity refers to the ease with which an asset can be sold if needed. Mixing up these terms can lead to confusion and wrong policy recommendations. In this paper, we build a multi-asset model in which an asset’s safety and liquidity are well-defined and distinct from one another. Treating safety as a primitive, we examine the relationship between an asset’s safety and liquidity in general equilibrium. We show that the commonly held belief that “safety implies liquidity” is generally justified, but there may be exceptions. We then describe the conditions under which a relatively riskier asset can be more liquid than its safe(r) counterparts. Finally, we use our model to rationalize the puzzling observation that AAA corporate bonds are considered less liquid than (the riskier) AA corporate bonds.

Balancing Work and Family: The Impact of Paid Family Leave on United States Labor Supply (J2, J3)

Kelly Jones
,
American University
Britni Wilcher
,
American University

Abstract

In 2018, less than 20 percent of U.S. workers had access to paid family leave (PFL) during periods spent caring for a new child resulting in $20.8 billion in lost wages. A growing body of literature explores whether PFL mitigates costs associated with caregiving. Existing research exploits state policy changes to identify the causal impact of access to PFL on leave taking behavior (i.e. use and duration) and its subsequent impact on labor market outcomes (e.g. labor force participation, likelihood of returning to previously held job, hours worked, and earnings). Evidence from early adopters of PFL, California and New Jersey, suggests that the policy enhances labor market attachment of parents. It is unclear what existing findings imply for design and efficacy of PFL in other states that may differ on political, social, and economic dimensions. This is the first study to estimate the impact of non-job protected PFL (California and New Jersey) and job-protected PFL (Rhode Island) on labor short- and long-term labor supply decisions.

Using Current Population Survey (CPS) Basic Monthly samples, we exploit the introduction of family leave insurance programs in California (2004), New Jersey (2009), and Rhode Island (2014) to identify the causal impact of PFL on short- and long-term labor decisions of male and female U.S. workers. Five labor market outcomes are examined: labor force participation, employment, part-time employment, unemployment, and underemployment rate. We combine an event study and difference-in-differences model to estimate the effect of PFL on short- term labor market decisions. Long-term effects of PFL are estimated using synthetic control methods on a state panel containing annual state averages of labor force characteristics aggregated from CPS Basic monthly samples.

Bank Capital, Financial Stability and Basel Regulation in a Low Interest-Rate Environment (E3)

Margarita Rubio
,
University of Nottingham
Fang Yao
,
Reserve Bank of New Zealand

Abstract

The current low interest-rate environment poses new challenges to bank regulation policies. This paper analyzes the role of the Basel regulation in this new context. We study this issue by using a DSGE model with housing and collateral constraints, both on borrowers and banks. Results show that, on the one hand, low interest rates lead to an increase in bank capital, calling for less strict regulatory regimes. On the other hand, this environment encourages more indebtedness of borrowers, calling for a stricter regulation for macroprudential purposes. An optimal analysis of the countercyclical buffer in Basel III regulation reveals that risks to financial stability outweigh the extra accumulation of bank capital. Thus, the countercyclical buffer should be implemented more aggressively in a low interest-rate world.

Behavioural Spatial Economics: Utility Misprediction in Locational Choices and Spatial Sorting (D9, R2)

Christian Krekel
,
London School of Economics and Political Science
Reto Odermatt
,
University of Basel

Abstract

The formation of accurate expectations about the costs and benefits of different locational choices is a central element in moving decisions and spatial sorting. It involves the prediction of changes in locational characteristics as well as utility consequences of these changes. We assess whether prospective movers make systematic errors when forming their expectations regarding their future wellbeing in the new location. Using geo-referenced, longitudinal household data from Germany that span two decades and exploiting the German reunification as an exogenous shock to moving costs, we document that, while all movers expect to be better off, people who move to deprived cities make systematic and large prediction errors. Reasons include failure to anticipate adaptation to compensating differentials and hedonic assimilation. We discuss the relevance of these findings for spatial equilibrium theory, and show that behavioural reasons can explain some of the persistent wellbeing differences across regions.

Behind the Flames: Unintended Impacts of Straw Burning on Human Capital (Q5, I0)

Joshua Graff Zivin
,
University of California-San Diego
Tong Liu
,
Hong Kong University of Science and Technology
Yingquan Song
,
Peking University
Qu Tang
,
Jinan University
Peng Zhang
,
Hong Kong Polytechnic University

Abstract

Air pollution potentially impairs human capital, but the effects remain largely unobserved in rural and developing contexts. Agricultural straw burning is prevalent in agrarian economies worldwide with a long history, yet the external costs of pollution from burning are underexplored. Using data from the National College Entrance Examination (NCEE, or gaokao) in China and straw burnings detected by high-resolution satellites during 2005 to 2011, this paper investigates the impacts of straw burning, a traditional rural source of air pollution on high-stakes cognitive performance of young people. To address the endogeneity of straw burning, we differentiate upwind fires from non-upwind fires to isolate socioeconomic confounders. We find that a one-standard-deviation increase in the difference between upwind and non-upwind fires during the exam decreases the total exam score by 1.45% of a standard deviation (or 0.6 points), and further decreases the probability of getting into first-tier universities by 1.24% of a standard deviation. This paper contributes to understanding the effects of air pollution on cognitive performance in a high-stakes setting for young people in a developing context, and highlights the unintended and overlooked social costs of straw burning on human capital.

Beyond Tuition: Cost of Living and College Affordability (I2, R0)

Tatiana Mocanu
,
University of Illinois-Urbana-Champaign

Abstract

Recent public debate about college affordability has focused on rising tuition values, leaving a substantial component of cost of college neglected. Cost of living (COL) comprises 50% of cost of attendance, with shares even higher in cities with high housing costs. Universities are in charge of estimating and reporting COL figures, which raises the question of whether these estimates are correctly measured. Accurate living expenses are crucial, since the cost of attendance limits the amount of federal loans and aid students are eligible to acquire, and thus finance college. In this paper, I document that inaccurate COL estimates provided by universities are pervasive and show how misreporting affects student outcomes. There are two parts to my analysis. First, I derive local cost of living indices that reflect the categories considered by universities when publishing living expenses estimates. These indices are new and measure local housing, food, and transportation costs based on a national representative student, with locally adjusted multipliers. I use these indices to benchmark how the accuracy of living expenses reported by U.S. universities behaves over time and space. Inaccurate COL reporting is widespread across all institution types. Almost half of all colleges report housing and food costs with at least 20% error, with for-profit universities under-reporting the most. In the second part, I show that under-reporting induces higher dropout rates, increase in student working hours, and higher loan default rates, and propose mechanisms to explain these effects. By having all universities following standardized COL estimates and making the average under-reporting university accurate, dropout rates would decrease by almost 1 percentage point.

Birds of a Feather Flock Together (More): Homophily in the Age of Trump (J1, D6)

Deserina Sulaeman
,
Singapore Management University

Abstract

Homophily, the tendency for humans to associate more with others who are similar to themselves, is pervasive in social and economic interactions. In particular, racial and ethnic homophily has been quite persistent over time despite profound changes in social, political, demographic, and economic conditions (Smith et al., 2014). This study examines ethnic homophily on a charitable crowdfunding platform, focusing on campaigns associated with 2017 Atlantic hurricane season – the costliest season on record with three Category 4 hurricanes: Harvey, Irma, and Maria. These hurricanes heavily affected regions that are among those with the highest percentage of Hispanic population in the U.S: Texas, Florida, and Puerto Rico.

This study documents three related results. First, we identify ethnic homophily in charitable crowdfunding. Compared to non-Hispanic fundraisers, Hispanic fundraisers receive a larger proportion of donations from Hispanic donors. As Hispanics are under-represented in the donors’ pool (relative to the fundraisers’ pool), this ethnic homophily is associated with lower average donations for Hispanic fundraisers.

Second, we document a time-series variation in ethnic homophily associated with statements issued by a divisive figure: the President of the U.S. During this hurricane season, President Trump often tweeted regarding the relief efforts provided by government entities. Ethnic homophily is stronger following such tweets: Hispanic donors’ proportion of donations to Hispanic fundraisers more than doubled following the tweets.

Third, despite increasing ethnic homophily, President Trump’s tweets do not appear to put Hispanic fundraisers at a disadvantage. Following the tweets, Hispanic fundraisers receive increased donations from non-Hispanic donors as well as Hispanic donors. As these tweets are likely to raise the awareness regarding the issues faced by the various relief efforts, the tweets’ positive effect on Hispanic fundraisers highlights the importance of awareness in dampening homophily’s role in amplifying divisive policies enacted by political figures.

Black-Cat Markets and the Value of Superstition: Evidence from Housing Prices in China (Z1, R3)

Guoying Deng
,
Sichuan University
Manuel Hernandez
,
IFPRI

Abstract

The potential role and influence of superstitious beliefs on agents’ behavior and the economy is still relatively new in the literature, in part because there are few settings where the effect of these beliefs can be formally detected and quantified. This paper examines the association between superstitious beliefs and housing prices in China. We focus on the numbers “4” and “8” as negative and positive beliefs associated with these numbers, respectively, are the strongest among the Chinese culture. We use an extensive daily transaction dataset for new apartment units in a major city for the period 2004-2015 and evaluate price differentials based on the floor level and unit number. We find that units located in the fourth floor consistently show a discount of around 80 Yuan (or 2%) per square meter compared to units in other floors, while units located in the eighth floor show a premium of 20 Yuan (or 0.5%) and the latter are sold faster. These price differentials persist when alternatively considering unit numbers ending in four or eight, although the price differences are less salient (between 0.4-0.6%). The results are robust to alternative buyer and seller characteristics. Similarly, the superstitious pricing behavior is positively correlated with the economic cycle and seems more prevalent in higher income neighborhoods. The beliefs further appear temporally exacerbated after the occurrence of an extreme event such as the Wenchuan earthquake of 2008. Overall, the results show that superstitious beliefs can have sizeable and continuous effects on the Chinese housing market and total household wealth portfolio.

Can Banks Use Their Liquid Asset Buffers? (D8, C7)

Guillaume Arnould
,
Bank of England
Antoine Lallour
,
Bank of England

Abstract

We assess how disclosure affects the usability of the liquid asset buffer that banks hold due to liquidity regulation (the Liquidity Coverage Ratio, or LCR). Using recent data, we show that banks tend to disclose quite detailed information around their LCR buffer holdings, sometimes more frequent and detailed than required by regulation. This has consequences on buffer usability. Using a model, we show that such disclosures can sometimes reduce incentives for banks to use their liquid asset buffer in stress, making the buffer partially unusable.

Can Consumers Avoid Cost Sharing by Reclassifying Injuries? (I1, D4)

Olesya Fomenko
,
Workers Compensation Research Institute (WCRI)
Jonathan Gruber
,
Massachusetts Institute of Technology

Abstract

We examine how consumers respond to being effectively double insured under two systems: group health and workers compensation. Many GH plans have substantial consumer cost-sharing burden, while WC coverage has no cost-sharing for medical services for work-related injuries. In particular, the percentage of workers enrolled in high deductible (more than $1000) plans quadrupled from 12% in 2007 to 51% in 2017. As a result, a consumer facing a large deductible under their GH plan will have a strong financial incentive to make a claim under WC instead. We use a unique data set of claims under both GH and WC to study how WC filing responds to GH deductibles for the most common injuries covered under both programs. We identify the impact of case shifting by using interactions of deductible levels and previous spending. We incorporate this in two ways: through spot prices and end of year prices. For spot prices, we measure the remaining deductible at the time when the enrollee files their claim, while for end of year prices, we rely on the remaining deductible at the end of year. The latter is of course difficult since the actual spending during the year depends on whether the episode in question is filed as GH or WC. We address this potential endogeneity by using last year’s total spending as an instrument for this year’s total spending. We find that a typical claim is about 1.4 percentage points (5.3%) more likely to be filed as a WC claim when facing an average deductible (about $630) compared to a plan with no deductible. Also, we find that consumers do not appear to be forward looking, focusing on the “spot price” rather than the full “end of year price” in deciding to file for WC coverage.

Can E-Commerce Reduce Traffic Congestion? Evidence from Alibaba Single Day Shopping Event (R4, F6)

Cong Peng
,
UC San Diego

Abstract

Traditional retail involves traffic both from warehouses to stores and from consumers to stores. E-commerce cuts these intermediate traffic by delivering goods directly from the warehouses to the consumers. Although plenty of evidence has shown that vans that are servicing the e-commerce are a growing contributor to traffic and congestion, consumers are also found making less shopping trips in a vehicle. This poses the question of whether e-commerce can reduce traffic congestion. The paper exploits the exogenous shock of an influential online shopping retail discount event in China (similar to Black Friday), to investigate how rapid growth of e-commerce affects urban traffic congestion. Perceiving e-commerce as trade across cities, I specify a CES demand system with Fréchet distributed taste shocks on online shopping to model the change of online consumption and offline consumption due to the price shock in the sale event. Mapping consumption to traffic delivers a mechanism that reveals the relation between traffic efficiency and the change of traffic during the event. I tracked hourly traffic congestion data in one hundred Chinese cities in one week before and two weeks after the event. In the week after the event, intracity traffic congestion dropped by 1.7% during peaks and 1% during non-peak hours. Using Baidu Index (similar to Google Trends) as a proxy for online shopping, I find that online shopping increased by about 2.5 times during the event. Guided by the model, I find that a 100% increase in online shopping causes a 1.3%-1.8% reduction in traffic congestion. The effect is the most salient from 9 am to 11 am and from 7 pm to midnight. This reduction in congestion has a monetary value of around 6 billion dollars a year.

Can Protectionism Improve the Trade Balance (F1, F4)

Serge Shikher
,
United States International Trade Commission
Sunghyun Henry Kim
,
Sungkyunkwan University

Abstract

The paper uses a dynamic structural gravity model of trade to analyze changes in the trade balance in response to changes in trade policy. Production uses labor and capital that are subject to adjustment costs. The model has incomplete asset markets with bond trading so that countries have both current account and financial account transactions. This feature allows us to analyze dynamic effects on trade balance and international asset positions of trade policies, such as changes in tariff rates. We use the model to analyze the effects of an increase in US tariff on (a) Chinese good and (b) all imported goods with and without retaliation. We find an increase in US tariffs leads to a temporary improvement in US trade trade balance and increase in US foreign asset position. Both savings and investment fall in the US, but the initial fall in investment is greater than the fall in savings.

Can Public Recognition Reward Backfire? Field Experimental Evidence on the Retention and Performance of Volunteers with Social-Image Concerns (O1, I2)

Asad Islam
,
Monash University
Abdul Malek
,
Brac University
Sakiba Tasneem
,
Monash University
Liang Choon Wang
,
Monash University

Abstract

We embed a large-scale randomized controlled experiment within an existing volunteer tutor program of BRAC in Bangladesh to examine the effects of offering non-financial incentives on volunteers’ dropout rates and performance. Consistent with the hypothesis that volunteers are motivated by social-image concerns, we find that dropout rates increase when volunteers, especially those with high career and other-regarding motives for volunteering, are offered a performance-contingent public-recognition certificate. Despite nearly doubling the dropout rate for the most desirable volunteers, the treatment improves overall student performance because it motivates performance among volunteers with low other-regarding motives for volunteering and low past achievement.

CEO Optimism and Net Interest Margin (G4, G3)

Jiashan Wu
,
University of Illinois-Chicago

Abstract

we study the link between CEO optimism and commercial banks’ net interest margin (NIM). We find that NIM among banks with optimistic CEOs is higher. We specifically test four of the channels where optimism could have affected NIM. Our results show that the high NIM could be due to higher risk taking and higher market power among optimistic banks.

Challenges of Credible Forward Guidance (E5, E0)

Akatsuki Sukeda
,
University of California-Santa Cruz

Abstract

This paper analyzes the effectiveness of optimal incentive-compatible forward guidance at the effective lower bound (ELB) on nominal interest rates when the economy faces occasional financial disruptions. Financial disruptions make households respond less to future monetary shocks because of precautionary-saving motives and make firms respond less to future monetary shocks because the value of current profit increases. As a result, the following three results emerge. First, forward guidance affects the economy at the ELB much less strongly without changing its effect when the economy exits the ELB. Second, the effectiveness of forward guidance becomes weaker even if the central bank has perfect credibility and can commit to forward guidance with a longer horizon than the case without financial disruption. Third, since the output gap and inflation experience more significant deviation from their optimal levels at exit periods, credibility concerns make the effectiveness of optimal sustainable forward guidance even weaker.

Complementarities in Public and Private Health Service Delivery: Results from a Randomized Controlled Trial in Nigeria (I1, H4)

Pedro Carneiro
,
University College London
Marcus Holmlund
,
World Bank
Michell Yoonjei Dong
,
World Bank
Sanghmitra Gautam
,
Washington University-St. Louis

Abstract

This paper presents experimental evidence from a cluster-randomized controlled trial (RCT) evaluating the effectiveness of involving non-traditional health services providers in malaria prevention and treatment in Nigeria. Community health workers were trained in malaria prevention, diagnosis, and treatment and provided with fully-subsidized malaria drugs to be used in treating members of their community; and private drug retailers were trained in diagnosis and treatment of malaria and provided with access to partially-subsidized drugs. To test the effectiveness of both interventions individually and jointly in comparison to a control condition, we conducted a four-arm cluster RCT in southern Nigeria.

We generate three main empirical results. First, we find evidence that both interventions improve household knowledge on malaria. Better household knowledge doesn’t necessarily lead to higher preventive or better care-seeking practices, however. Still, we observe lower malaria prevalence among children in areas where drug retailers were trained and subsidized. We further examine the causal chain between improved knowledge and reduction in malaria by conducting heterogeneity and mediation analyses. We find that primary health facilities in Nigeria differ in terms of their quality and find evidence of greater impact in areas with better quality primary health facilities at baseline. This suggests that non-traditional health providers should be regarded as complementary with rather than substituting for the existing public health system. Furthermore, through mediation analysis we confirm that impacts on health outcomes resulted from improvements in the quality of health providers rather than better knowledge or care-seeking behavior by households.

Credit Surface of Mortgage Loans: Lenders' Belief of Housing Markets (G1, R3)

Qizhou Xiong
,
Halle Institute for Economic Research (IWH)

Abstract

This paper investigates the relationship between lenders’ beliefs about future house price changes and their mortgage pricing choices. The mortgage lending market prices the risk of default, which is correlated with both borrower characteristics, leverage level, and regional housing market expectation. From a theoretical perspective, conditional on borrower risk, the default probability positively correlates with the loan-to-value ratio (LTV), which in turn positive correlates with the mortgage interest rate. At the collateral equilibrium (Fostel and Geanakoplos, 2015; Bailey et al., 2018), all mortgages are priced at the credit surface of LTV and mortgage rates, which is bounded by two important thresholds: the threshold below which no one would default and the threshold above which no one would lend. Existing Theories show that the first threshold indicates the lenders’ belief in the worst case scenario of the local housing market; and that the second threshold reflects the expected housing market growth. Using the FHFA mortgage loan performance data from Freddie Mac, Fannie Mae and Ginnie Mae in the past two decades, I estimate the thresholds in the top 20 housing markets by population in the United States after controlling the borrower risk by FICO score and lender heterogeneity. This paper exploits the cross region variation in mortgage contracts to the same type of households within the same lender to reveal the relative housing market assessment difference across regions by the lenders. Finally, I compare the estimated ex-ante housing market belief with the ex-post housing market movement using Zillow house price index to evaluate the lenders' performance in assessing and pricing housing market risks.

Cultural Biases in Equity Analysis (G4, G1)

Vesa Pursiainen
,
Imperial College London and University of Hong Kong

Abstract

I study the role of cultural biases in decision-making among equity analysts. I construct a Eurobarometer-based measure of cultural trust bias between European countries and find that a more positive trust bias by the analyst's country of origin toward the firm's headquarter country is associated with significantly more positive stock recommendations, controlling for analyst-month and firm-month fixed effects. The cultural bias effect is stronger for eponymous firms whose names mention their home country, suggesting firm names can have a priming effect for cultural biases. A more positive trust bias toward the CEO's home country is also associated with more positive recommendations. The bias effect varies over time, increasing with the aggregate level of pessimism in Europe and decreasing with consumer confidence. In addition, I find evidence of a negative North-South bias emerging during the European debt crisis, a UK-Europe divergence amid Brexit, and a Franco-British bias during the Iraq war. The share price reaction to buy-recommendations by more positively biased analysts is weaker, suggesting that the market may be aware of these biases. The effect of cultural biases is stronger for analysts from countries with a more negative attitude toward globalization and increases with analyst tenure, while analysts working at larger brokers are less affected by cultural bias.

Currency Returns and Interest Rate Slopes (F3, G1)

Pedro Castro
,
PUC-Rio
Ruy Ribeiro
,
PUC-Rio

Abstract

We construct measures of term structure slope changes for the US and other G10 countries using short-term interest futures at different horizons. These changes in term structure slopes have immediate impact on currency returns but also a strong delayed effect over the following weeks, implying that currencies are predictable. We find strong evidence of out-of-sample short-term predictability for both individual currencies and a portfolio that is long G10 currencies and short US dollar. For instance, investors that condition on international slopes to dynamically trade this long G10 portfolio improve Sharpe ratios to 0.6, relative to 0.12 for a buy-and-hold strategy. We also construct a novel dollar-neutral currency portfolio by sorting G10 country currencies based on the cross-section of the slope measures, that deliver higher Sharpe ratios than other currency strategies, such as the carry trade. This new strategy is also uncorrelated to other standard currency strategies. These findings are compatible with delayed currency market reaction to information in interest rates. Consistent with this interpretation, we find evidence that this momentum persists longer in non-FOMC weeks.

Dead Men Tell No Tales: How the Homo Sapiens Became Homo Economicus (J1, Z1)

Roman L. Zakharenko
,
National Research University Higher School of Economics

Abstract

The paper explains long-term changes in birth, death rates and attitude to personal consumption
by changing patterns of cultural transmission. When communities are culturally isolated, they are focused on population growth, resulting in large fertility and welfare transfers to children, limited adult consumption and lack of old-age support. With increasing cultural contact across communities, successful cultural traits induce their hosts to attempt becoming celebrities by limiting fertility and increasing longevity via higher consumption and old-age arrangements. Empirical analysis confirms that celebrities have fewer children
and live longer; their presence precedes reduced aggregate birth and death rates.

Deadweight Loss of the Value of Information with Investors’ Temporal Beliefs (D8, D6)

Zhi Dong
,
University of Auckland

Abstract

This paper introduces and investigates the deadweight loss of the value of information when investors have contingent beliefs on updated market signals. It introduces a framework to theoretically examine the loss of social welfare when there is ‘distorted’ or ‘hidden’ information in markets with less transparent assets, for e.g., physical assets and collaterals in banking sectors. It proposes empirical framework to test the impact of information on the deadweight loss of social welfare when ‘distorted’ information prevails or ‘real’ information is hidden in markets. The paper highlights the inter-connection between the value of information, investors’ interpretation of signals and their adjustable beliefs. The findings provide additional insight into the understanding of asymmetric information, corporate finance and behavioural finance and economics for market agents. The study disentangles the influence of information and psychological elements in market agents’ decision making process in a dynamic setting.

Dealer Leverage and Exchange Rates: Heterogeneity Across Intermediaries (G2, F3)

Laurie Pounder DeMarco
,
Federal Reserve Board
Ricardo Correa
,
Federal Reserve Board

Abstract

In line with a growing literature on intermediary asset pricing, we find that changes in the leverage of primary dealers have predictive power in forecasting exchange rates. Unlike previous studies, though, we find that broker-dealer heterogeneity matters for their role in asset pricing. Foreign-headquartered dealers entirely drive the predictive power while the leverage of domestic U.S.-headquartered dealers is insignificant. We argue that this is due to foreign broker-dealers having more balance sheet capacity relative to domestic dealers during the 2000s. This result conflicts with an assumption of homogeneity among intermediaries which is implicit in most modern intermediary asset pricing models. In addition, we find that currency market positions, including derivatives positions, are likely stronger than cross-border lending as the main channel through which leverage manifests itself in exchange rate changes.

Deciphering China’s Female Employment Paradox (J1, J2)

Jin Cao
,
Norges Bank
Haiyue Yu
,
Dongbei University of Finance and Economics
Jin Cao
,
Norges Bank
Shulong Kang
,
Dongbei University of Finance and Economics

Abstract

China is featured by extraordinarily high female post-childbirth labor market participation rate and labor intensity, given that the public subsidy to childcare is poor and policy support for childbearing female employees is largely absent. Establishing a panel dataset that tracks females’ childbirths and employment, we find that such paradox is largely explained by the intra-family grandparental childcare. Correcting the selection bias that stems from females’ fertility decisions using PSM-DID model, we find that females without grandparental support suffer a substantial drop in post-childbirth employment, while the employment of females with grandparental support even rises after childbirth. It takes females without grandparental support twice as long to recover their employment after childbirth.

Dinosaur Judges: Conservative Experts in a Changing Society (D8, O3)

Kim Sau Chung
,
Hong Kong Baptist University
Yujing Xu
,
University of Hong Kong

Abstract

Modern societies thrive on the advices of experts in a garden variety of areas. How do we identify these experts? In circumstances where an expert's track record cannot be easily assessed by the general public, our society relies on peer reviews from "known'' experts to identify new experts. This gives rise to an aristocratic expert class that is inevitably conservative. Young scholars, in order to earn the approval of old "known'' experts, have incentives to study old subjects or follow old schools of thought at the expenses of new subjects and new schools of thought that better serve a changing society. Our society tradeoffs conservatism against competence in its endeavor to identify experts, but the optimal tradeoff may not be achieved due to time-inconsistency. We formalize this problem with a model described in terms of legal experts such as lawyers and judges, and use it to shed light on noise voters and anti-intellectualism in the Trumpian era.

Direct and Spillover Effects of Free Compulsory Education on Schooling and Migration (I2, J1)

Naijia Guo
,
Chinese University of Hong Kong
Zhao Rong
,
Nanjing Audit University
Shuangxin Wang
,
Chinese University of Hong Kong

Abstract

In 2006, China started the Two Exemptions One Subsidy (TEOS) program, providing free compulsory education which includes six years of primary education and three years of lower secondary education, in rural China. This paper exploits TEOS to study its direct effect on the education and labor migration decisions of the eligible children and its intra-household spillover effects on their ineligible siblings who had already finished their compulsory education as well as their parents. By using the National Fixed Point Survey data and constructing a difference-in-differences framework, we first find that more eligible children attend lower secondary schools and fewer migrate to work after TEOS. Second, ineligible brothers and sisters are affected differently by TEOS: more ineligible brothers attend upper secondary schools, while more ineligible sisters drop out of upper secondary school to participate in labor migration. In the long run, we find TEOS improved completed years of schooling of males, but had no impact on the schooling of females since, for them, the positive impact on lower secondary education is offset by the negative impact on upper secondary education, leading to greater gender disparity in education. Third, parents with eligible minor children are less likely to migrate since more school input after TEOS is complemented by more parental input, while parents with eligible adult children migrate more to compensate for the loss of family laborers.

Directed Search, Nominal Rigidities, and Mark-up Cyclicality (E3)

Zhesheng Qiu
,
City University of Hong Kong
José-Víctor Ríos-Rull
,
University of Pennsylvania

Abstract

New Keynesian models with sticky prices built on the Dixit-Stiglitz framework must have countercyclical mark-ups conditional on monetary shocks, which is inconsistent with empirical evidence based on labor share data. We pose a directed search style shopping friction in goods market to model firms' price setting, on top of Dixit-Stiglitz. Our theory allows for procyclical mark-ups conditional on monetary shocks, without sacrificing the performance of the model along other dimensions. We prove this in a static model, and test it in an estimated medium scale DSGE model. Unlike the literature that criticizes the use of inverse labor shares to get procyclical mark-ups, we provide a complementary view that New Keynesian models can in fact be compatible with procyclical mark-ups conditional on monetary shocks, when there is goods market shopping friction. These results come from the assumption that firms compete via prices for higher production capacity realizations which is bounded above by 100%.

Discouragement Traps (E2, J6)

Tristan Potter
,
Drexel University

Abstract

I construct a simple measure of the rate of discouragement among potential labor market participants and show that this measure rose sharply during the Great Recession and never recovered to its pre-crisis level. Accordingly, I propose a theory in which fears of prolonged joblessness can become self-fulfilling, drawing the economy into a permanent high-discouragement, low-participation state: a discouragement trap. When job losers fear it will be difficult to find work if they remain jobless for too long, they search aggressively, thus crowding out those at the back of the queue, inducing labor force withdrawal and discouragement, and rationalizing fears of prolonged joblessness. This mechanism emerges naturally when firms rank workers based on jobless duration and workers make search decisions. The interaction between ranking and endogenous search effort generates strong strategic complementarities that give rise to multiple steady state equilibria characterized by significantly different discouragement rates despite a unique natural rate of unemployment. Rich global dynamics emerge: I show that for any initial conditions, spontaneous fears of prolonged joblessness can thrust the economy into a discouragement trap that is unlikely to be detected from unemployment data alone.

Do Corporate Taxes Affect Workplace Safety? (K2, J4)

Daniel Bradley
,
University of South Florida
Connie Mao
,
Temple University
Chi Zhang
,
University of Massachusetts-Lowell

Abstract

We examine how workplace injury rates change when firms are subject to a corporate tax shock. We find that tax increases lead to a significance increase in reported injuries, but tax decreases have no similar effect. Our difference-in-differences empirical strategy relies on staggered state-level corporate tax changes that exploits spatial discontinuity in treatment and control establishments located in contiguous border counties within the same firm. The results are strongest in industries with low union bargaining power, for firms with high marginal tax rates, poor safety culture scores, firms that barely meet or beat analysts’ earnings forecasts and firms that hire seasonal workers. Our results suggest that tax increases lead to real effects at the expense of employees, with no similar benefit accruing for tax cuts.

Does a High School Diploma Matter? Evidence Using Regression Discontinuity Design (I2, J3)

Deni Mazrekaj
,
KU Leuven
Sofie Cabus
,
KU Leuven

Abstract

It is well documented that more educated people tend to earn higher wages. The reason for this education premium, however, has been a source of considerable debate. The current study explores whether education raises productivity (human capital theory) or just reflects it (signaling theory) by estimating earnings returns to a high school diploma. Most early studies that aimed to distinguish between the human capital and the signaling theory based on the years of schooling suffered from omitted-variables bias and therefore produced only correlational evidence. As opposed to years of schooling, a high school diploma is unlikely to affect the productivity, as it is essentially only a piece of paper. To approximate the random assignment of diplomas, we exploit the standardized exit exams which the students must take at the end of the final year of secondary education to obtain a high school diploma in the Netherlands. These exit exams are the last obstacle before graduation. Using a regression discontinuity design on administrative population data, we compare the earnings of students who barely passed and barely failed the standardized exit exams. These students have similar levels of human capital but different diploma status. Our results indicate that a high school diploma raises hourly net earnings by about 4 percent. Although we find that a diploma increases earnings regardless of gender, ethnicity, or educational track, we find a larger positive effect for girls and students who have completed a program in the pre-university track. Moreover, we also find that students with a diploma are 2 to 4 percentage points more likely to find a job. Thereby, the results indicate that a high school diploma serves as an important signal on the labor market.

Does Drug Enforcement Cause Violence? Evidence from Colombia (K4, C3)

Hernan Botero Degiovanni
,
Scotland's Rural College
Santiago Pinto
,
Federal Reserve Bank of Richmond

Abstract

Drug enforcement may have unintended consequences. This is especially true in source countries where military manoeuvres are utilized to support the activities of the anti-narcotics police. We use Colombian drug war data for 2005 to test whether the anti-narcotics military operations carried out by the Colombian government had any influence on the country’s homicide rate at municipality level. As the government’s military actions are found to be endogenous to the homicide rate, we run IV regressions to determine a measure of military attacks which only represents those actions performed to support coca eradication. In order to accomplish that, geographical characteristics of the Colombian municipalities are used in the first stage regressions as instruments to proxy for the municipalities’ productivity to cultivate coca crops. The underlying idea is that the military attacks that took place in municipalities with a large productivity to produce coca were perpetrated with the aim of reducing that activity. One issue that arises from these IV regressions is that their residuals are spatially correlated. To tackle this problem, we estimated Spatial Autocorrelation (SAR) models. Another issue that arises is that the fitted values that result from the first stage regressions are not necessarily non-negative. We run Tobit models to address this problem. According to our estimations, every additional anti-narcotics military operation in 2005 increased the average homicide rate of the intervened municipalities in nearly 12 homicides per 100,000 inhabitants. In addition, these operations incentivized drug dealers’ counterattacks, which raised the average homicide rate in the affected municipalities in nearly 14 homicides per 100,000 inhabitants per additional attack executed.

Does Publication Lead to Publication? The Effect of Author Reputation on Publication Success (A2, L2)

Oliver Rehbein
,
University of Bonn

Abstract

I test whether non-quality-related author reputation affects publication success. I exploit the fact that post-publication journal reputation changes affect an author's reputation exogenously. This quality-independent reputation change significantly increases the publication success of the next publication, without influencing citations. The effect of reputation increases with journal prestige and is particularly pronounced in general interest journals.

Does the Daylight Savings Time Causes People to Change More Than Their Clock? (D1, J2)

Sanjukta Basu
,
Tulane University

Abstract

A government policy, passed decades ago, forces every citizen in some countries, around the world, to change their clocks twice a year; once in spring and the other time in fall. In the spring, the clock moves one hour forward. This is a practice of advancing clocks during summer months so that evening daylight lasts longer, while sacrificing normal sunrise times. The policy was introduced in the US during the World War I to conserve energy during wartime and give longer daylight. Existing literature shows that contrary to the intent of the policy, DST has not been successful in reducing energy consumption in the modern world.

My research attempts to study how the change in time caused by DST impacts individuals' day-to-day schedules and moods. I use various outcome variables such as sleep pattern, labor productivity, exercise pattern, well-being and mood indicators such as happiness, sadness, tiredness and stress scale. I also classify activities into groups based on their MET values. I use the American Time Use Survey to study the DST change in USA. I find that respondents reduce light intensity activities after the fall DST implementation and increases light and moderate intensity activities while reducing sedentary tasks after the spring DST implementation.

Retired individuals have a larger impact than the whole population or working age group from a DST change. In the fall, they reduce sedentary and moderate to vigorous activities by 4.5 and 3.6 hours, respectively. Light intensity activity increases. In the spring, these individuals only reduce high intensity activities. Similarly, disabled individuals reduce high intensity activities like sleeping and relaxing by 2.23 hours in spring. I intent to further study if the DST effect crime rates and traffic accidents.

Does the WTO Balance Biosecurity with International Trade? (F1, Q1)

LINDA FERNANDEZ
,
Virginia Commonwealth University
Monica Das
,
Skidmore College

Abstract

International trade poses potential risks to worldwide biosecurity. Invasive species that hitchhike on traded goods represent a biosecurity threat. Invasive species are non-native species that spread in a new area with deleterious impacts. We explore the trade and biosecurity relationship to answer the following: Does the World Trade Organization’s (WTO) Sanitary and Phytosanitary Program (SPS) enable simultaneous biosecurity and enhanced trade for importers and exporters? The SPS has not been quantitatively analyzed. Fernandez and Sheriff (2013) theoretically examine importing and exporting countries contending with biosecurity risk and show potential gains from SPS for importers and exporters in trade and biosecurity through Article 9. Article 9 allows for technical assistance to reduce biosecurity risk. Article 10 focuses on how special needs of developing countries are taken into account. Our panel data covers 1993 to 2014 for 33 countries. Our data are grouped and vary by: (1) size of agricultural sector, (ii) size of manufacturing, (iii) GNI/capita and (iv) geographical region. We estimate a panel model with varying slopes and intercepts to explain how volume of aggregate agricultural exports responds to changes in technical assistance through the WTO’s SPS program (Articles 9 and 10). We control for exogenous factors such as share of manufactures exports and real gross domestic product. We explore three measures of technical assistance in terms of number of commodities, duration and value. We find the impact of technical assistance on agricultural exports depends on the definition of technical assistance applied, the country group and a country’s GNI/capita as well as its geographical location. Additionally, we explore how the interaction between WTO policy and an Enabling Trade Index influences the relationship between biosecurity and trade.

Economic Growth and the Debt Tax Shield (H2, E2)

Marcel Fischer
,
Copenhagen Business School and University of Konstanz
Bjarne Astrup Jensen
,
Copenhagen Business School

Abstract

We study the general equilibrium implications of the corporate debt tax shield in a growth economy that taxes household income and firm profits and redistributes tax revenues in an attempt to harmonize the lifetime consumption opportunities among households that differ in their endowments. Our model predicts that in general equilibrium the tax shield's reduction in the corporate after-tax borrowing rate is counteracted by an increase in the pre-tax rate. In our model the debt tax shield leads to a simultaneous increase in the growth rate of the economy and in the degree of disparity in households' lifetime consumption opportunities.

Economics of The Fintech Industry in Small Open Transition Economies: A New Competitive Advantage? (G2, O3)

Aleksandr V. Gevorkyan
,
St. John's University

Abstract

Following applied industry analysis this paper derives three immediate constraints to the recent rise in financial technology (fintech) in the small open economies of Eastern Europe and former Soviet Union. First, the paper finds that much of the continued growth and local implementation of fintech-driven innovation is constrained by the structure and conduct of the domestic financial sectors. Second, central banks, currently enjoying broader macroeconomic policy mandates than peers elsewhere, may be caught off-guard balancing further financial deepening, as domestic financial credit swells, and sustaining challenges to foreign exchange markets balance. Finally, the third constraint to fintech’s growth is technological, implied in the limited ability of the firms in the local high-tech sectors to source from domestic innovation and talent pool, further limiting sector’s solidification domestically and minimizing sustained global presence. As such, despite recent attention from the global business community, the scalability of fintech’s innovative potential for transition economies remains uncertain with lessons for smaller emerging markets prioritizing financial technology growth.

Effect of Caring for Pets on the Well-being of Older Americans (D1, I3)

Charlene Marie Kalenkoski
,
Texas Tech University
Thomas Korankye
,
Texas Tech University

Abstract

Standard economic household production models (e.g. Becker, 1965) state that a person produces household commodities (such as a meal, a clean home, and quality children) with market-purchased goods and services and the household member’s time. The person derives utility (i.e., happiness, satisfaction, or well-being) from these commodities. One such commodity is human-animal interaction. This paper uses data from the American Time Use Survey and its associated Well-being Module to examine the relationship between engaging with pets/household animals and well-being, examining several measures of well-being. Probit and ordered probit models are estimated to examine how time spent engaging with a household pet/animal is associated with several self-reported health outcome measures that cover general health, hypertension, the taking of pain medication, and feelings of being well-rested. Additional analyses at the respondent level examine, using continuous regression models, how time spent engaging with a household pet/animal is associated with daily average scores for meaningfulness, happiness, sadness, stress, tiredness, and pain. All models control for standard demographic and economic variables available in the ATUS as well as a baseline measure of well-being to isolate the effect of the pet/animal engagement activity on well-being.

Effect of Parental Preference for the Sex of a Child on Child Outcomes: Evidence from Korea (J1)

Won Fy Lee
,
University of Minnesota

Abstract

This study examines the effect of parental sex preference for children on parental inputs and child outcomes, using a nationally representative survey data of birth cohorts born in 2008. I utilize the survey question that directly ask parents' underlying gender preferences towards a newborn child and used the responses as a measure of gender preference to examine causal effect of gender preferences on parental inputs and child outcomes at the age of 4. I found that parents’ sex preferences had a statistically significant effect on the parents’ inputs to their child and the child’s subsequent cognitive outcome. This is the first study to use parent-level preference measures to estimate the effect of such preferences on child well being and outcomes.

Effects of Test Translation on Kindergarten English Learners’ Performance on Literacy, Math and Executive Function (J1, I2)

Irma A. Arteaga
,
University of Missouri

Abstract

This study examines the effects of translating literacy, mathematics and executive function tests from English into the student’s native language, an understudied accommodation. Given that English language learning (ELL) students represent the fastest growing segment of the U.S. student population, it is important to understand how tests’ translations affect performance. In 2015, 4.8 million children were ELL students, and most ELL children were enrolled in elementary school grades. More than 77 percent of ELL students spoke Spanish at home (NCES, 2018). This study uses data from the Early Childhood Longitudinal Study, Kindergarten (ECLS-K: 2010), a nationally representative dataset of more than 15,000 children who attended kindergarten in 2010. In kindergarten, all children took a language screener test, the preLAS. Children living in Spanish speaking households and who were below the cutoff point in the preLAS test were administered all tests in Spanish, while children who were at or above the cutoff point, or whose households did not speak a foreign language, were administered the test in English. I used a regression discontinuity design, which resembles a randomized experiment because children who are just below the cutoff point and who were administered the tests in Spanish would be similar to children who are just above the cutoff point and who were administered the tests in English. Around the vicinity of the cutoff point, this method allows to answer the causal question of whether translation accommodations for young English language learners have an effect on math, executive function and reading test scores. Our findings show that linguistic accommodations improve math and executive functions scores in the spring, but not in the fall of kindergarten. This suggests that accommodations are effective when children have to answer linguistically challenging and difficult questions, and that occurs in the spring, after children were taught something new.

Endogenous Prosperity on Competition: Capital Allocation on Efficient Frontiers (O3, D6)

Zhi Dong
,
University of Auckland

Abstract

This paper explores and analyses forces driving efficient frontiers for investment equity portfolios. It is perhaps the first study to explicitly incorporate dynamic competition in production and R&D prosperity in evaluating equity performance for efficient frontiers, with regard to multiple industries. A factor of prosperity is introduced into the optimization of the system. This factor is theoretically determined through endogenous competition and empirically correlated with equity growth. It is found that technological shocks leading to prosperity of equity growth will push out efficient frontiers for investment equity portfolios. In a dynamic setting for competitive capital markets, the results imply potential over-flow of capital attracted into high growth equity sectors. The over-flow of capital may result in systematic error and pricing failure in equity markets. Loss of efficiency in capital allocation is correlated with market players’ beliefs on prosperity of technology shocks; while market players’ beliefs are adaptive to new signals that diminish past shocks on temporal pace. The study recommends regulations on competition to encourage releasing new technological developments under smoothed patterns and to reduce impact of technological surprise to capital markets.

Estimating the Collateral Eligibility Premium: An Euro Area Study (E5, G2)

Dawid Zochowski
,
European Central Bank
Ioana Alexopoulou
,
European Central Bank
Ulrich Bindseil
,
European Central Bank
Agueda Solis Alonso
,
European Central Bank
Giacomo Beltrame
,
European Central Bank

Abstract

The impact of central bank collateral frameworks has recently found renewed interest in the economic literature. A key measure of the relevance of central bank collateral frameworks for securities markets and funding costs is the eligibility premium. Brunnermeier et al. (2016) and Nyborg (2017) argue that the impact of collateral eligibility on security prices is significant and has profound monetary policy and economic implications. The empirical literature focusing on the impact of concrete episodes of eligibility changes on security yields provides the estimate of the eligibility premium in the magnitude of 5 to 13 basis points.

This paper adds to the literature by consistently exploiting unique data set on all major changes of the eligibility status of securities in the Eurosystem collateral framework. Using diff-in-diff and distinguishing between the announcement and the implementation effects, we find that the eligibility premium is influenced by two channels operating in two different directions. First, increased usefulness of the security as a funding vehicle related to the ability of posting is as collateral with the ECB when it becomes eligible is associated with the negative impact of the yields. Second, lower market liquidity of the security that becomes eligible as it is hoarded on balance sheets of financial institutions or locked in operations with central banks increases the yields.

We document that while before the global financial crisis the first channel dominated leading to a negative eligibility premium on yields in the range of 5 basis points, once the crisis erupted the impact of lower market liquidity more than outweighed the effects of the first channel, which led to a positive eligibility premium on yields in the range of 5-10 basis points.

Exercising Real Options Sooner or Later? New Insights from Quantile-Preserving Spreads on How to Fasten or Delay Exercise (G0, H3)

Pascal Letourneau
,
University of Wisconsin-Whitewater
Sang Baum Kang
,
Illinois Institute of Technology

Abstract

Real options address many different real-world problems including investment, merger and acquisition, debt valuation, managerial compensation, and government policy. A critical difference between real options and financial options lies in that real option holders, writers, and even external stakeholders can modify the characteristics of real options to increase or decrease its exercise probability. The stakeholders encompass a firm, a creditor, a manager, and a government. For example, government policies may have for objectives to promote or discourage investments in various industries.
Using a general quantile-preserving spread and stochastic dominance, this paper studies how modifications of real option characteristics affect the optimal real option exercise decision. We study under which conditions an increase in risk may increase the exercise probability and hasten real option exercise. We find that the exercise timing depends on the preserved quantile. Contrary to the conventional wisdom, an increase in risk can decrease the holding value if the risk is added through asymmetric spread. We significantly generalize previously obtained results by developing propositions for an unspecified underlying process and a general call-like payoff function.
Our paper enriches a strand of literature in financial economics where agents make a decision to maximize the expected utility of a payoff function. Our paper also adds to literature on the effect of risk on option value. To the best of our knowledge, very few articles investigate the intentional actions of stakeholders to modify a real option's exercise probability in a theoretical perspective. Our results are useful to determine an optimally parsimonious modification of a real option to increase or decrease its exercise probability and have policy implications. Specifically, we find that the modification of strike price tends to be more cost effective than that of the underlying process.

Exploring the Roots of Sicilian Technological Underdevelopment: The Role of Mafia Since the Unification of Italian Kingdom Up to the Present Day (N9, O3)

Giuseppe Di Vita
,
University of Catania
Fabio Di Vita
,
University of Catania

Abstract

This paper investigates, from an economic point of view, the impact of Mafia on the technological development of Sicily. We attempt to demonstrate that the presence of the Mafia in Sicily have hampered the process of technological development since the unification of the Italian Kingdom (1861). Moreover, we demonstrate that the negative impact of such criminal association on this region of Southern Italy are still persistent until contemporary times. To these aims are considered the data of patents in Sicily from 1874 to 1913, never be used in previous analyses, because fruit of archives manual consultation. The main result of the analysis is that Sicilian provinces with high presence of Mafia show, since 1874 to the first World War, low levels of technological development, and that this negative impact on technological progress persists up to now. This observed phenomenon seems to be useful to explain from a new point of view the persistent underdevelopment of Sicily.

Falling Behind: Has Rising Inequality Fueled the American Debt Boom 1980–2007? (E5, R3)

Fabian Greimel
,
University of Mannheim
Moritz Drechsel-Grau
,
University of Mannheim

Abstract

The household debt boom since 1980 is considered one of the main drivers of
the Great Recession of 2007–9. In lockstep with household debt, income inequality
has risen to new extremes. We build a model that links rising inequality to the
mortgage debt boom. It builds on the old idea that people care about their social
status. In an attempt to keep up with ever richer Joneses, the middle class substitutes
status-enhancing houses for status-neutral consumption. These houses are mortgage-
financed, creating a debt boom across the income distribution.
Our mechanism is consistent with the following stylized facts: (i) Real mortgage
debt, (ii) debt-service-to-income ratios and (iii) house sizes (in sqft) have increased
since the 1980 across all income quintiles. This happened despite (iv) stagnating real
incomes for the bottom 50% since the 1980s.
We build a tractable dynamic network model with housing to illustrate how our
mechanism generates these facts. We extend it to a quantitative general equilibrium
life-cycle model to show how status concerns and rising inequality amplify previously
studied origins of the debt boom: the saving glut, the banking glut and financial
innovation. Preliminary results suggest that social comparisons boost the debt boom
and the house price boom by about 25%.

Feel the Burn: Mental and Behavioral Responses to Agricultural Air Pollution in China (Q5, I0)

Joshua Graff Zivin
,
University of California-San Diego
Tong Liu
,
Hong Kong University of Science and Technology
Guojun He
,
Hong Kong University of Science and Technology

Abstract

Farmers in many countries burn agricultural crop residues after harvest to facilitate farming. This paper estimates the impacts of straw burning on air pollution and mental health measured by Center for Epidemiologic Studies Depression Scale (CES-D) in China during 2013-2015. Specifically, a 1-point increase in summer straw burning in July and August in a Chinese county will lower the mental health status of urban residents by 1.22% of a standard deviation, and the total monetary losses in urban China exceed 2.6 billion USD. Since rural straw burning affects urban mental health only through air pollution, the estimates approximate the short-run pollution effect. In contrast, rural farmers feel happier when there are more burnings, suggesting that agricultural gains from straw burning may outweigh farmers’ perceived costs of burning. In response to straw burning, Chinese citizens adopt avoidance behaviors, such as searching and purchasing anti-PM2.5 masks and air filters in autumn when ambient air quality is much worse but not in summer. The monthly defensive expenditure on anti-PM2.5 masks would exceed 1.3 million USD if autumn straw burning increases by 10 points in all cities. There is also a decline in public interest in outdoor activities but a growing interest in air quality information, travel and entertainment, signaling the perceived risks and mitigation efforts by the citizens. This study is among the first to document a causal impact of air pollution on mental health using a traditional but underexplored pollution source, and it highlights the urgency of restraining straw burning in China. The evidence further helps explain the failure of command-and-control regulation banning straw burning which neglects the potential losses to the farmers, and calls for incentive-based policies such as subsidizing straw recycling to achieve Pareto efficiency. The findings can also provide reference for other countries with similar issues.

Financial Literacy and Student Debt (A2, I2)

Nikolaos Artavanis
,
Virginia Tech
Soumya Karra
,
University of Massachusetts-Amherst

Abstract

Using a large sample of over 1,000 students from a major, land-grant, public university in Massachusetts, we examine the financial literacy level of college students, and its implications on the repayment of student debt. We find low levels of financial literacy (39.5%), particularly among female (26%), minority (24%) and first-generation (33%) students. Based on survey responses, we show students with a deficit in financial literacy are more likely to underestimate their future student loan payments; 38.2% of low-literacy students underestimate future payments by more than $1,000 annually. Furthermore, we document a financial literacy wage gap as lower literacy students expect significantly lower starting salaries than their literate peers. As a result, low-literacy students are more vulnerable to unexpected, adverse shocks on their payment-to-income ratios that can impair their future creditworthiness and undermine their ability to service debt post-graduation.

Forecasting GDP Growth from Outer Space (C8, E0)

Jaqueson Kingeski Galimberti
,
ETH Zurich

Abstract

We evaluate the usefulness of satellite-based data on nighttime lights for forecasting GDP growth across a global sample of countries, proposing innovative location-based indicators to extract new predictive information from the lights data. Our findings are generally favorable to the use of the night lights data to improve the accuracy of model-based forecasts. We also find a substantial degree of heterogeneity across countries on the relationship between lights and economic activity: individually estimated models tend to outperform panel specifications. Key factors underlying the night lights performance include the country’s income level, logistics infrastructure, and quality of national statistics.

Fueling the Engines of Liberation with Cleaner Cooking Fuel: Evidence From Indonesia (J2, I1)

Tushar Bharati
,
University of Western Australia
Yiwei Qian
,
University of Southern California
Jeonghwan Yun
,
University of Southern California

Abstract

In its attempt to reduce the subsidy burden of kerosene, the cooking fuel for 48 out of the 52 million Indonesian households in 2004, the Indonesian government launched the “Conversion to Liquefied Petroleum Gas (LPG) Program” in 2007. Cooking with LPG not only reduces indoor pollution due to cooking but also saves cooking time. Using the staggered rollout of the program across provinces and variation in the pre-program kerosene adoption across communities, we find that the program improved the health and increased the labor force participation of women. The program also had benefits for the rest of the family - improved levels of health and consumption and an increase in subjective well-being. Importantly, the program was associated with increased financial decision-making power with women. In light of the findings that intra-household externalities are one of the main reasons for the lack of adoption of cleaner cooking technology, increased decision-making power with women has important implications for the sustained use of this technology, even in the absence of the subsidy.

Fundamental and Speculative Demands for Housing (E3, E2)

Weicheng Lian
,
International Monetary Fund

Abstract

Using an approach resembling the production-based asset pricing, I infer over-time changes in the price of housing services based on the adjustment cost of changing housing stock through residential investment, and changes in the price of residential structures. It also decomposes the deviation of house price from its trend into a fundamental component--the contribution from changes in demand for housing services of current households, and the rest, which can be caused by speculation in the housing market. I show that (i) the price of housing services increased by around 35% between 2000-06, whereas the CPI rent, an imputed measure of the price of housing services, increased by only 8% in real term during this period, and (ii) three-quarters of the house price surge from 2000 to 2006 was attributed to the fundamental component and the trend of the house price.

Gender Gap in Higher Education and Gender Norms (I2, D1)

Stefanie Jeanette Huber
,
University of Amsterdam
Hannah Paule-Paludkiewicz
,
Goethe University Frankfurt

Abstract

Cross-country differences in the gender gap of higher education attainment are large. The gender gap in tertiary education attainment (measured as the female minus the male rate), ranged between rates of negative 13% in Switzerland to positive 11% in New Zealand in 2010. In this paper, we test the hypothesis that part of these cross-country differences is driven by culture. To isolate the effect of culture from the effects of institutions and purely economic factors, we investigate the decision of second-generation immigrants in the United States in attaining at least a Bachelor degree. We robustly find that cross-country differences in cultural values and believes are an important explanatory factor for the observed differences in the gender gap of higher education attainment across countries.

Geopolitical Risk and Corporate Investment (G3, D8)

Ruchith Dissanayake
,
Queensland University of Technology
Vikas Mehrotra
,
University of Alberta
Yanhui Wu
,
Queensland University of Technology

Abstract

Shocks to geopolitical risk are known to adversely affect real activity, as well as a flight to safety by invested capital. In this study we explore the channels via which this occurs. We find that firms respond to geopolitical risk by cutting back on capital investments. This effect is stronger for firms with more irreversible investments and foreign operations. Geopolitical threats appear to influence investments more than geopolitical acts do, perhaps because acts are perceived as resolving uncertainty. Dividends, another use of cash by firms, are not adversely affected by changes in geopolitical risk, indicating finite half-lives for geopolitical shocks.

Growth of Illicit Drug Markets and Its Effects on Crime Rates (I1, K4)

Sujeong Park
,
RAND Corporation
David Powell
,
RAND Corporation

Abstract

After years of reductions in the rate of murder and violent crime in the United States, the national murder rate increased in both 2015 and 2016. The causes of this trend are generally unknown, though there is some evidence that it is related to narcotic drugs. Arrests related to heroin-cocaine had been stable between 2010 and 2014 before a sudden increase in 2015. Likewise, the number of murders related to narcotic drugs has increased since 2013 with an especially large jump in 2015. These types of crime increases parallel recent dramatic growth in overdoses involving heroin and synthetic opioids, indicators of advancing illicit drug markets. However, the causal relationship between the recent opioid crisis and the rise in murder rates is missing from the literature. The study aims at ascertaining the presence of a causal link.
We show that local areas hit hardest by the rise in heroin and fentanyl overdoses are also those that have been disproportionately affected by the increase in crime. Furthermore, we use the reformulation of OxyContin as an exogenous shock to illicit markets. OxyContin reformulation led people misusing OxyContin to switch to illicit opioids, and previous work has shown that areas with more people misusing OxyContin experienced faster growth in heroin and fentanyl overdoses. We test whether this growth in the illicit drug markets caused an increase in crime.
We find that after reformulation, there are larger increases in murder rates in states with high pre-reformulation rates of OxyContin misuse. This relationship is strongest for white homicide victim rates. The results provide a causal link between the opioid epidemic and crime and show that the rise in murder rates was due to increased exposure of victims to risky environments, such as access to illicit drug markets.

Has WTO Membership Promoted Trade Since the WTO Foundation? (F1, F6)

Lourenco S. Paz
,
Baylor University
Andre Filipe de Azevedo
,
Unisinos University
Magnus dos Reis
,
Federal University of Rio Grande do Sul (UFRGS)

Abstract

This article proposes a new methodology based on the gravity model to estimate the effects of the WTO membership on trade flows after the creation of the WTO in 1995. We employ the Poisson Pseudo-Maximum Likelihood estimator with several different specifications of fixed effects. The sample includes bilateral imports from 133 countries in the primary, textile and manufacturing sectors covering the period from 1995 till 2014. Estimates suggest that the WTO has succeeded in expanding international trade, mainly on primary goods and textiles. WTO membership had no effect on trade flows between developed nations, though it substantially fostered trade flows involving developing nations, including that of manufacturing goods. These important heterogeneous affects are overlooked by the extant literature that investigates the effects of WTO membership only on the total trade flows.

High-Speed Rail, Structural Transformation and Inclusive Growth:Quantitative Case Study of the Experimental High-Speed Rail Project in China (O1, R1)

Xiao Ke
,
Peking University
Justin Yifu Lin
,
Peking University

Abstract

Modern transportation infrastructure is vital to an economy and is a prerequisite for future growth in cities in developing countries. Hence, high-speed rail networks in China have expanded drastically over the past decade. The study investigates the economic impacts of High Speed Rail projects for targeted city locations with heavy-industry structure, taking the first HSR project in China as a case study. Using the synthetic control method by Abadie et al. (2003, 2010, 2015) to construct appropriate counterfactuals, we find that within policy evaluation period 1999-2013, most HSR cities experience sustained increase in real GDP per capita. However, the magnitudes of such HSR impacts are very heterogeneous. As for some locations, the local income level increase between 4.6% and 28.1% on average when compared with the counterfactual non-HSR cities. The rest receive negligible impacts. By further examining the local labor market, we find that the targeted locations experience structural transformation toward the service sector and industry diversification. This in turn mitigates policy burden in the industry sector. First, as HSR projects greatly reduce travel time cost and increase punctuality, employment increase in tourism and business related sectors. Second, HSR connection breaks through bottlenecks for freight transport capacity on the existing railway lines and brings gains to bulk commodity industry. Besides, increase in employment and income in tradable sectors generates demand for non-tradable local good and services. Finally, for heavy industry cities in Northeast China with large amount of redundant workers, expansion of the service sector helps reallocate the low skilled labor from heavy industry to labor intensive service sector that used to be suppressed, hence further facilitate structural transformation. Results suggest that the HSR infrastructure project in China can help heavy-industry cities along the line take comparative advantage-following approach to facilitate structural transformation and diversification thus contribute to inclusive growth.

Home Visitation or Group Meeting? Effects of Early Stimulation on Child Well-Being: A Cluster Randomized Control Trial in Guatemala (I2, J1)

Irma A. Arteaga
,
University of Missouri
Julieta Trias
,
World Bank

Abstract

This paper examines the effects of parental training on early childhood stimulation on parent-child interactions and child’s nutritional and cognitive outcomes in Guatemala. This study assesses the cost effectiveness of using home-visits or group-sessions. More than 40 percent of Guatemala’s population is indigenous, making the country an ideal setting for new comparative research of an at-risk population that has not been studied comprehensively. Additionally, the literature on cost-effectiveness and cost-benefit analysis of early childhood stimulation programs is limited (Gowani, Yousafzai, Armstrong, & Bhutta, 2014). While more common in the developed world (Barnett & Masse, 2007; Belfield, Nores, Barnett, & Schweinhart, 2006), this type of analysis is most needed in the developing world given the limited resources and the need to assess programs’ feasibility,sustainability, and scalability. To our knowledge, this study is the first that used a clustered randomized controlled trial to assign communities (n=111) to three treatment arms: home visitations, group sessions, and no intervention in an indigenous setting. As such, it will advance the literature by providing evidence as to whether one early childhood stimulation modality produces better outcomes than the other does and, if so, by how much and at what cost. 2,022 children between 6 and 18 months of age were assessed at baseline. The interventions were also complemented with training, supervision, and mentoring of mother educators. Children received 10 months of intervention and we assessed program effects on children’s nutritional status using anthropometric measures, receptive and expressive language, and fine motor, using the MDAT scales of child development, and home stimulation environment, using the HOME and FCI instruments. Intent-to-treat preliminary findings suggest that children who receive home visits are less likely to be overweight, and more likely to have a positive home environment than children who participate in group meetings or children in the control group.

Hong Kong Can Escape the Trilemma in the Short Run (E5, E4)

Thomas Willett
,
Claremont McKenna College and Claremont Graduate University
Shan Xue
,
Claremont Graduate University
Alice Ouyang
,
Central University of Finance and Economics

Abstract

The monetary trilemma, that countries cannot have all three of a fixed exchange rate, no controls, and monetary autonomy at the same time, has been at the heart of a great deal of international monetary analysis. What is often not recognized is that this trilemma is only a long-term constraint unless capital mobility is close to perfect. With imperfect capital mobility, central banks can sterilize international reserve flows in the short run and hence maintain a degree of monetary autonomy. Hong Kong is a natural case to test this possibility since it has a credibly fixed exchange rate against the dollar and no major capital controls. We investigate this issue using two approaches. One is to estimate the effects of changes in US interest rates on those in Hong Kong. The second is to estimate whether Hong Hong has been able to engage in some degree of sterilization. Appropriate estimation of sterilization coefficients requires also estimation of offset coefficients which show how much of a change in the domestic monetary base is offset by capital flows, a useful measure of capital mobility. We use both monthly and daily data between January 1999 and December 2018 from the HKMA. We find that the extent of arbitrage between the US and HK varies across the different asset classes and maturities. The interest rate pass-through from the US to HK is less than unity in both the interbank and bond markets, indicating some monetary autonomy. We find evidence of effective sterilization and offset coefficients below 1, again indicating imperfect capital mobility. The specific estimates are sensitive to both the measurement of HK’s monetary base and the specification of the monetary authority’s reaction function, but the general conclusion of imperfect capital mobility and ability to sterilize holds.

How Do Enhanced Derivative Disclosures Affect Information Asymmetry Between Informed and Uninformed Investors? (G3, M4)

Mengbing Ren
,
University of Warwick

Abstract

Firms use derivatives both for hedging purposes and non-hedging purposes. The inadequate derivative disclosures about the purposes of firms’ derivative usage prior to the implementation of FASB Statement No.161 (SFAS 161) made it difficult for investors to appraise the effects of derivatives on firm performance. Prior research finds that the accounting designation of derivatives required by SFAS 161 is informative as to how firms use derivatives. Given that the transparency of firms’ derivative disclosure improves after SFAS 161, I examine whether the enhanced derivative disclosures reduce the information asymmetry between informed and uninformed investors. Tension exists as the derivative information may not be comprehensible to relatively uninformed investors. Using a hand-collected sample of U.S. firms, I find that derivative users compliant with SFAS 161 experience significantly greater reduction in stock illiquidity and probability of informed trade in the post-SFAS 161 period, and such impact is more pronounced for firms with greater investor attention.

How Forecast Accuracy Depends on Conditioning Assumptions (C5, E3)

Katja Heinisch
,
Halle Institute for Economic Research (IWH)
Christoph Schult
,
Halle Institute for Economic Research (IWH)
Carola Engelke
,
University of Cologne

Abstract

This paper examines the extent to which forecast errors on the economic development are driven by external or technical assumptions that prove to be incorrect ex post. Therefore, we use a new data set comprising an unbalanced panel of annual forecasts from different institutions forecasting German GDP, exports and imports and the underlying assumptions. We test whether forecast errors decrease significantly with smaller horizons. A high fraction of the variation in forecast errors for GDP, exports and imports growth can be accounted by the variation in the underlying assumption errors. We find that forecast errors in world trade drive most of the variation in GDP forecast errors.

How Words Matter: Machine Learning & Movie Success (M3, Z1)

Louis R. Nemzer
,
Nova Southeastern University
Florence Neymotin
,
Nova Southeastern University

Abstract

We employ a recurrent neural network (RNN) machine learning approach to decipher the relationship between word choice in movie descriptions and box office ticket sales. The baseline analysis was repeated with the additional outcome of movie ratings as an alternative measure of box office success. Data were collected from the Internet Movie Database (IMDB) website, and movies were restricted to full-length comedies released from 2014 to 2018. A Long short-term memory (LSTM) architecture, which is widely used in machine learning for parsing text, was implemented for the RNN. Additional controls employ runtime and other characteristics that may potentially impact ticket sales. Our work has implications for directors and movie houses considering the importance of descriptions in generating a box office hit via enticing movie descriptions.

Ignorance Is Bliss (D8)

Yufeng Sun
,
Shanghai University of Finance and Economics

Abstract

A principal can select a better project when she has more prior information, she will also hold a strong belief that the selected project is promising. However, given this strong belief, when an agent implements the project and submits a report with negative feedback, the principal will doubt the agent’s competence rather than the project’s quality and even dismiss the agent. The agent can anticipate this, so he has an incentive to manipulate his disclosure which includes less negative news to avoid being dismissed by the principal. The principal will then suffer a loss from the agent’s information distortion, as she cannot immediately adjust the project. This paper shows that if the principal is initially ignorant about the project, the agent has less concern about being judged as incompetent and a greater incentive to provide more information about project progress, including negative news, then the wrong project is more likely to be adjusted immediately. In contrast to the classical principal-agent literature, where the principal can be better off having more information because less information rent is paid to the agent, this paper contends that in a bounded rationality environment, it is possible that for the principal, “ignorance is bliss.” She may be better off intentionally choosing “not to know,” as the agent then has an incentive to reveal more information which makes the project better-off.

Immigration and Electoral Outcomes: Evidence from the 2015 Refugee Inflow to Germany (J6, H0)

Julia Bredtmann
,
RWI-Leibniz Institute for Economic Research

Abstract

During the second half of 2015, Germany witnessed a dramatic increase in the number of refugees. In the first elections after this mass arrival of refugees, right-wing populist parties, most notably the Alternative for Germany (AfD), won significant shares of the popular vote. Using unique data on refugee populations and their type of accommodation at the municipality level, this paper investigates the effects of the 2015 refugee inflow on the outcomes of the state elections held in March 2016 in the German state of Rhineland-Palatinate. For identification, we make use of the fact that the “German refugee crisis” served as a natural experiment, in which the sheer necessity to accommodate the large number of refugees led to exogeneous variation in the allocation of refugees across municipalities. Results based on fixed effects estimations show that an increase in the population share of refugees increases the vote share of right-wing parties and decreases the vote share of the incumbent federal government parties. However, exploring variation in the type of accommodation provided for refugees, our results reveal that the estimated electoral effects of the refugee inflow are solely driven by municipalities that host large reception centers for refugees. This finding has important implications for the design of asylum procedures, as it suggests that an early transfer of refugees to decentralized follow-up accommodation could reduce anti-refugee sentiments among the native population and thus the support for right-wing populist parties.

Immigration and Inequality: New Macroeconomic Evidence (E2, J2)

Ørjan Robstad
,
Norges Bank
Francesco Furlanetto
,
Norges Bank
Samad Sarferaz
,
ETH Zurich

Abstract

In this paper, we reconsider the link between immigration and labor income inequality using detailed micro and macro data for Norway. Immigration has increased substantially in Norway during the last 15 years in response to several European Union enlargements to Eastern European countries. At the same time, several measures of income inequality have started rising, although not as abruptly as in other developed economies. The link between immigration and inequality is not obvious. On one hand, some recent research based on a micro analysis of local labor markets finds that immigration from poor countries has steepened the social gradient in Norway (thus increasing labor income inequality) while immigration from rich countries has leveled the social gradient (cf. Hoen, Markussen and Røed, 2018). On the other hand, a study by Basso, Peri and Rahman (2018) for the US finds that immigration has reduced the job polarization effects induced by technological progress in the form of computerization. In contrast with the previous studies, we study immigration at the macro-level to capture general equilibrium effects and to disentangle the exogenous component of immigration. Our goal is to estimate the effects of various kinds of immigration shocks on labor income inequality but also on the level of labor earnings. Our analysis is feasible since Norway is one of the few countries for which a quarterly detailed measure of net immigration at the macro level and labor earnings based on administrative data covering the population of Norwegian workers at monthly frequency are available. We estimate a Structural Vector Autoregression model to disentangle immigration shocks from other shocks driving the business cycle. We trace the effects of immigration shocks on the labor earnings distribution and we are thus able to investigate whether different kinds of immigration shocks have increased or decreased inequality.

Immigration, Product Quality, and Intensive and Extensive Margins of Trade (F1, F2)

Syed Al-Helal Uddin
,
College of Saint Benedict & Saint John's University
Ahmed Tariq Aziz
,
Iowa State University

Abstract

This paper studies the impact of immigrants on the extensive and intensive margins of US exports. We contribute to the literature by exploring variation in the impacts of the ethnic networks on the intensive and extensive margin of trade across different skill groups of immigrants and quality levels of products. Since a large part of the immigrant population suffers from poor skill transferability in the destination country, occupation in the destination country may contain more information about their role in enhancing trade. This paper proposes a more carefully defined measure of migration business network based on the types of occupations they are employed in and quantifies its impact on the US export. Since the direction of causality in immigration-trade literature is less clearly established from earlier studies, our analysis takes advantage of the fact that political refugees to the US are exogenously allocated across locations to establish the causal relationship. For the empirical estimation, we have utilized the US Census Bureau data provided by Schott (2008) for the USA inter and intra-state trade with foreign countries and IPUMS-America for the immigration from a specific country to a particular state. Data on the number of refugees per US state come from the Office of Refugee Resettlement (ORR). Our finding shows that immigrant stock in the US has a significant effect on both the extensive and intensive margins of the US exports and the impact is higher for high-quality products and through high skill immigrants.

Implications of the Occupational Wage-Skill Mismatch in the US (J3, J2)

Orhun Sevinc
,
Central Bank of the Republic of Turkey

Abstract

The US labor market has been polarizing since the 1980s along occupational wage percentiles while at the same time exhibiting monotonic growth by non-wage skill measures such as education, training, and cognitive skills. These seemingly contrasting trends are driven by the wage-skill mismatch at the lower part of the wage distribution. Using ILO-based definitions of working conditions and occupational measures from ONET this paper develops an index of task difficulty that can correct the occupational wage-skill mismatch and estimates significant labor market returns to more difficult working conditions focusing on job switchers in the NLSY. Evidence suggests a more multi-faceted approach including both cognitive and non-cognitive aspects of human capital to interpret and explain inequality trends.

In Art We Trust (G1, G0)

Yuexin Li
,
Tilburg University
Marshall Ma
,
Erasmus University Rotterdam
Luc Renneboog
,
Tilburg University

Abstract

As trust is a cornerstone in financial markets, especially in illiquid, opaque, and unregulated markets such as art markets, we investigate the economic effects of provenance information on the sales probability of paintings, their hammer prices, and repeat sales returns. We collect provenance data from auction catalogues by applying textual analysis and partition the data according to four provenance dimensions related to pedigree (ownership chains relating buyers to artists), exhibition history (museums, art fairs, cultural cities), art (history) literature coverage, and certification (physical and non-physical proof of authenticity by artists and experts). We find that these four categories of provenance information increase the sales probability by 2% to 4%, lead to a price premium by 14% to 54%, and increase the annualized return by 5% to 16% (percentage points) after controlling for artwork characteristics (such as topic, signature), artist, time, and auction house branch fixed effects. We perform a variety of robustness tests, e.g., by means of LASSO estimations. In order to address the problem of endogeneity in the sales decisions, we examine repeat sales samples with longer holding periods and investigate the estate sales which we expect to be less affected by past prices. In order to address the potential Past Price–Provenance Change reverse causality issue, we apply a two-stage regressions on repeat sales, exploit Christie’s provenance policy change as a quasi-natural experiment in a Difference-in-Differences (DiD) setting, and also study the provenance effects in a DiD setting for artists affected by fakes and forgeries after discovery. All attempts to address endogeneities do not invalidate our results.

In Search of Information: Use of Google Trends' Data to Narrow Information Gaps for Low-income Developing Countries (E3, O4)

Futoshi Narita
,
International Monetary Fund
Rujun Yin
,
International Monetary Fund

Abstract

Timely data availability is a long-standing challenge in policy-making and analysis for low-income developing countries. This paper explores the use of Google Trends' data to narrow such information gaps and finds that online search frequencies about a country significantly correlate with macroeconomic variables (e.g., real GDP, inflation, capital flows), conditional on other covariates. The correlation with real GDP is stronger than that of nighttime lights, whereas the opposite is found for emerging market economies. The search frequencies also improve out-of-sample forecasting performance albeit slightly, demonstrating their potential to facilitate timely assessments of economic conditions in low-income developing countries.

Income Inequality and Capital Reallocation in the Presence of Financial Frictions (G2, E2)

Matias Ossandon Busch
,
Halle Institute for Economic Research (IWH)

Abstract

Does income inequality exacerbate or constrain the reallocation of credit risk in periods of financial distress? Using novel matched bank-branch data on Colombian banks I find that after the collapse of Lehman Brothers, ex-ante exposed banks operating in municipalities with a more unequal income distribution shift credit towards safer borrowers to a larger extent. For identification I compare branches' reaction to this shock within banks and across municipalities by simultaneously absorbing local demand. Collateral constraints play a key role: While the overall effect is stronger in regions where banks demand higher collateral, credit backed by better collateral remains consistently shielded. Regional-level estimates suggest that this inequality risk-taking channel can account for a significant share of consumption growth during crisis periods.

Inflation Target Uncertainty and Monetary Policy (E5)

Yevgeniy Teryoshin
,
University of Nevada-Las Vegas

Abstract

I develop an extension of the standard New Keynesian model to monetary policy regime switching to study the impact of uncertainty around the future inflation target. First, I fully characterize how the responses of inflation and output to inflation target uncertainty depend on the monetary policy rule. If monetary policy is passive, inflation may increase far beyond the anticipated increase in the inflation target, while a strong monetary response to expected inflation results in an immediate drop in the inflation rate. Next, I derive the optimal response of the central bank, which can be achieved by adjusting the current inflation target. A central bank unwilling to adjust the inflation target can optimally adjust other policy rule parameters and can often obtain comparatively similar welfare benefits. Finally, I examine the implications of a perfectly anticipated change in the inflation target and find it is likely to generate cyclical dynamics for inflation and output under a constant policy rule. An optimal time varying policy rule or uncertainty in the period of the inflation target change eliminates cyclical fluctuations and improves welfare.

Information Monopolies and Monetary Policy Pass-through (G2, E5)

Charles O'Donnell
,
European Central Bank
Fergal McCann
,
Central Bank of Ireland

Abstract

We empirically investigate the role of frictions in bank-borrower relationships on the transmission channel of monetary policy. We argue that banks' incentives to pass on reductions in their funding costs depend on the ease at which their borrowers can solicit outside competition for their financing. We test this hypothesis by comparing the loan spreads of small bank-dependent firms with those of group-affiliated and large firms. To limit the impact of the endogeneity of monetary policy to macro conditions, we restrict our analysis to firms which have an investment-grade credit rating. Using a large sample of French firms, we show that following the ECB's monetary policy stimulus in the winter of 2008, the loan spreads (banks' mark-up) of small bank-dependent firms increased by 42 basis points more than those of large and group-affiliated firms. This effect is robust, but at a lower magnitude, when using alternative measures of bank-dependency based on firms' debt concentration with a main lender, and controlling for firm size and group-affiliation. We also show that pass-through is stronger in counties with lower levels of local bank market concentration (HHI), but that this effect only holds for larger (or more diversified) borrowers. We perform several further tests to rule out a risk premium story. First, we find no evidence for a flight-to-quality effect for stand-alone SMEs at this time. Second, we find no similar pricing impact when including in our sample speculative-grade firms and assigning treatment by firms' credit rating. Our evidence is in line with theories which show that banks' information monopoly allows them to charge higher rates to their `locked-in' borrowers.

Is Gift Card Really A Gift? An Empirical Study of Income Elasticity for Special Labeling of Dairy Product (D9, L1)

Jialiang Zhu
,
Xiamen University

Abstract

Products with labels representing different special characteristics, such as non-GMO or dolphin-friendly, can never be assessed by consumption of consumers, yet many consumers are still attached to those labels. Literature treats products with those characteristics that consumers cannot judge by themselves as credence goods. Therefore, consumers highly rely on product labels to help make purchasing decisions. Clearly, there is a price premium associated with those labels, but the empirical research of income elasticity for those credence-good labels is still scant.

This paper covers this research gap by cooperating with a large local grocery chain store in China and collecting individual-level transaction data with different payment information. Specifically, this paper uses grocery-store gift cards as a proxy for consumers’ income twist to estimate the income elasticity for the goods with credence features, such as organic labeled and geographic indication labeled dairy products. Using the random coefficient structural model, this paper first finds that consumers doing grocery with gift cards are more likely to purchase organic milk or other high-quality milk. The income elasticity for the organic label or foreign imported label is as high as 1.4. Second, the structural model incorporates protein, calcium, carbohydrate, and calorie as four other dimensions of dairy products, and finds in a fast-growing society such as China, the demand price elasticity for protein is much smaller than calcium or carbohydrate. Lastly, after tracking the VIP members’ four-year purchasing history, the paper finds that gift cards change consumer’s purchasing behavior and the changed behavior remains even without gift cards. The paper contributes to the literature as it empirically estimates the income elasticity for organic labels and the demand elasticity for different types of nutrition. It also quantitatively demonstrates that gift cards make consumers spend more and buy high-quality products that they typically not do if using cash.

Is the FOMC Overly-Optimistic? (E3, E5)

Sam Bullard
,
Wells Fargo Securities, LLC
Azhar Iqbal
,
Wells Fargo Securities, LLC

Abstract

This study develops a new framework to analyze whether the FOMC’s forecasts are overly optimistic. We find that the FOMC tended to over-forecast GDP growth and the unemployment rate and under-forecast inflation during the 1992-2017 period. We also find that the FOMC’s forecasts displayed different behavior under different chairs. Our analysis suggests that the FOMC is overly-optimistic, as it tends to forecast a more optimistic GDP growth outlook with “controlled” inflation expectations, while its optimism has also increased over time.
We also find that the Blue Chip consensus forecasts differ from the FOMC’s forecasts, as the Blue Chip consensus under-forecasted GDP growth and over-forecasted inflation for the 1992-2017 period. The Blue Chip consensus’ forecasting behavior changes in the post-2001 period, as the consensus over-forecasted GDP growth and inflation. From a theoretical perspective, the Blue Chip consensus forecasts are “consistent” in the post-2001 period, while the FOMC’s forecasts are “inconsistent,” as over-forecasting GDP growth is consistent with over-forecasting inflation.
In sum, our analysis suggests that the FOMC has an incentive to forecast an economic outlook that is consistent with its mandate, where the economy is growing close to its potential, yet inflation remains in check. The FOMC’s forecasts paint a bright economic outlook, and we find that the FOMC’s forecasts are more optimistic than the Blue Chip consensus forecasts.

Leadership in Scholarship: Editors' Influence on the Profession's Narrative (A1, O3)

Sergey V. Popov
,
Cardiff University
Ali Sina Önder
,
University of Portsmouth
Sascha Schweitzer
,
University of Bayreuth

Abstract

Academic journals disseminate new knowledge, and therefore can influence the direction and composition of ongoing research by choosing what to publish. We study the influence of editors and coeditors of the American Economic Review (AER) on the topic structure of papers published in the AER between 1976 and 2013 using a textual analysis of manuscripts. We compare AER's topic structure to that of the other top general interest journals. The appointment of new AER editors, while accompanied by a minor comovement of AER topics towards topics of editor's post-appointment publications, serves more to premediate trends in the other Top 5 journals.

Learning, Equilibrium Trend, Cycle, and Spread in Bond Yields (G1, G4)

Guihai Zhao
,
Bank of Canada

Abstract

While the empirical literature has shown the importance of macro trends in modeling the term structure of interest rates, the standard stationarity assumption makes it hard for equilibrium models to capture the trend in bond yields. This paper presents an equilibrium model to explain the trend, cycle, and spread in historical U.S. Treasury bond yields. The trend in yields is generated by learning from the stable components in GDP growth and inflation, which share similar patterns to the neutral rate of interest (r Star) and trend inflation (pi Star) estimates in the literature. Cyclical movements in yields and spread, are mainly driven by learning from the transitory components in GDP growth and inflation. The upward trend in the Treasury yield spread found in the data and the recent secular stagnation are tightly coupled due to persistently negative short-run beliefs. The upward-sloping yield curve is mainly driven by the fact that the amount of Knightian uncertainty that investors face is different for the long run versus the short run.

Lighting up the Dark: Liquidity in the German Corporate Bond Market (G1, G2)

Yalin Gunduz
,
Deutsche Bundesbank
Giorgio Ottonello
,
Vienna Graduate School of Finance
Loriana Pelizzon
,
Goethe University Frankfurt, SAFE and Ca' Foscari University
Michael Schneider
,
Deutsche Bundesbank
Marti G. Subrahmanyam
,
New York University

Abstract

We study the impact of transparency on liquidity in OTC markets. We do so by providing an analysis of liquidity in a corporate bond market without trade transparency (Germany), and comparing our findings to a market with full post-trade disclosure (the U.S.). We employ a unique regulatory dataset of transactions of German financial institutions from 2008 until 2014 to find that: First, overall trading activity is much lower in the German market than in the U.S. Second, similar to the U.S., the determinants of German corporate bond liquidity are in line with search theories of OTC markets. Third, surprisingly, frequently traded German bonds have transaction costs that are 39-61 bps lower than a matched sample of bonds in the U.S. Our results support the notion that, while market liquidity is generally higher in transparent markets, a sub-set of bonds could be more liquid in more opaque markets because of investors "crowding" their demand into a small number of more actively traded securities.

Local Labor Demand Shocks and Earnings Differentials: Evidence from Shale Oil and Gas Booms. (J3, Q3)

Gregory B. Upton Jr.
,
Louisiana State University
Han Yu
,
Texas A&M University

Abstract

In this research, we show that labor demand shocks to specific workers (male workers with high school education) in a specific industry that makes up a relatively small share of employment (oil and gas sector) can have significant impacts on earnings differentials both economy wide and within unrelated sectors. We estimate that a million dollars of new oil and gas production per person is associated with a 6.4% decrease in the college/high school earnings differential, and a 6.5% increase in the male/female earnings differential.

Identification comes from the plausibly exogenous technological advancement that allowed for resources to be produced from areas with specific geological formations thousands of feet below the earth's surface. The timing of these advancements and areas are impacted are well understood.

We decompose the possible channels through which a productivity shock to the oil and gas sector can impact regional earnings differentials. While we find evidence that male workers with less than a college education substitute into the high paying oil and gas sector, the lion's share of the change in earnings differentials can be explained by changes within the non-oil and gas sectors. This research highlights the importance of understanding labor demand shocks in seemingly unrelated sectors that make up a relatively small percentage of aggregate employment in describing economy wide earnings differentials. Implications for understanding changing trends in income inequality both across the income spectrum and between groups are significant.

Long-Term Effects of Child Care Assistance Policies (D1, I3)

Juliana Chen Peraza
,
RAND Corporation

Abstract

This paper aims to identify the effects of child care assistance policies on children’s long-run outcomes. Specifically, I focus on whether certain eligibility criteria for child care assistance determined by the state children are born in (i.e. having minimum hours of work requirements and the number of activities that classify as work activities) have an effect on years of education completed by 2013. To answer this question, I use a reduced form quasi-experimental approach that takes advantage of the natural variation in state child care assistance policies’ eligibility criteria after the welfare reform of 1996. The data used is a combination of the Panel Study of Income Dynamics (PSID) and the Child Care and Development Fund (CCDF) Policies Database. Preliminary results using OLS regression with controls and cohort fixed effects seem to indicate that children born in states without minimum hours of work requirements have more years of education by 2013, and that increasing the number of activities that qualify under the work requirement also has small positive effects on schooling. Similarly, ordered logistic regression of educational attainment (i.e. HS drop out, HS graduate, college enrolled) indicates these two eligibility criteria have effects on higher educational attainment. However, I do not see any effects of not having minimum hours of work requirements on years of schooling using a difference-in-difference approach.

Long-Term Health Effect of Earned Income Tax Credit (H2, I1)

Ze Song
,
Rutgers University

Abstract

Using decades of variation in the federal and state Earned Income Tax Credit (EITC) and the Panel Study of Income Dynamics (PSID) dataset, I examine the impact of exposure to EITC expansions in utero and during childhood on health outcomes in adulthood. In order to overcome the confounding relationship between family income and health outcomes, this study uses the maximum EITC benefit as the key variable. Reduced-form estimates show that EITC expansions had a positive impact on self-reported health status. Specifically, a $1000 increase in the maximum EITC exposure from ages 13 to 18 corresponds with a 0.01 point increase in the reported health status during adulthood. In addition, being exposed to EITC expansions in utero increases reported health status by 0.05 point. Relative to the range of reported health of 1 to 5 and the standard deviation of 0.94, these are very small effects. Nonetheless, these health effects are consequential, associating with increases in both family income and maternal labor supply.

Love Thy Neighbor? Ethnic Favoritism and Trust in Indonesia (I3, Q5)

Almedina Music
,
World Bank

Abstract

Ethnic favoritism refers to a situation where co-ethnics receive a disproportionate share of resources when members of their ethnic group control the local government - a phenomenon that is prevalent in many countries. This paper addresses two empirical questions: At household level, is there evidence of ethnic favoritism in the allocation of post-disaster aid transfers? And if so, how does this sort of ethnic favoritism affect trust levels in communities? Using household panel data, the analysis draws on a quasi-
natural experiment with an unexpected inflow of post-disaster aid to Indonesian villages that have been affected by a natural disaster. Preliminary results show that households that share the same ethnicity as their community/village head are more likely to receive post-disaster aid transfers than households that were equally affected by a natural disaster, but do not share the same ethnicity as the community/village leader. Results also indicate that trust levels towards other ethnicities are lower in areas where ethnic favoritism prevails while general trust levels are less affected.

Macroeconomic Effects of Discretionary Tax Changes in Canada: Evidence from a New Narrative Measure of Tax Shocks (E6, H0)

Syed Muhammad Hussain
,
James Madison University
Lin Liu
,
University of Liverpool

Abstract

In this paper we study the macroeconomic effects of changes in federal taxes for the Canadian economy for the time period 1961:1 - 2014:4. We employ the narrative methodology of Romer and Romer (2010) and Cloyne (2013) to identify exogenous changes in federal taxes. In particular, we study, in detail, all the budget documents for the Canadian economy and document all legislated tax changes and the motivations behind them. We then isolate those tax changes that were not motivated by contemporaneous movements in the economy and classify them as exogenous tax changes. Our main empirical result shows that an exogenous tax cut of 1 percent of GDP leads to a significant but short-run increase in output. Our analysis of disaggregated measures exogenous tax changes shows that 1) tax hikes associated with deficit consolidation have the biggest (contractionary) effects on output, 2) changes in personal income taxes have larger effects than changes in other types of taxes, 3) anticipated tax changes have strong expansionary announcement effects, 4) tax increases tend to have bigger (contractionary) effects on output than the (expansionary) effects of tax decreases, and 5) the effectiveness of tax policy has drastically decreased over time.

Make or Buy, and/or Cooperate? The Property Rights Approach to Auto Parts Procurement in Japan (L1, L6)

Yosuke Takeda
,
Sophia University
Ichihiro Uchida
,
Aichi University

Abstract

We conducted an empirical analysis on the hypothesis on auto parts procurement in Japan, raised by Asanuma
(1989; 1992). The Asanuma hypothesis of Japanese subcontractors claims that there is a new classification of
auto parts and their producers according to the degree of initiative for product and process designs. The
initiative results in “relation-specific skills” acquired by the suppliers in relation to the auto manufacturers in
the first tier. Among the responses to the hypothesis, Milgrom and Roberts (1992) and Holmstrom and Roberts
(1998) focused upon a role of the supplier association in the Japanese hierarchy system, where communication
among the suppliers alleviates opportunistic misbehavior of the automakers. This paper, instead of the
reputational role of the association, takes an alternative stand on the technology cooperation association, from
the property rights theory, especially a general setup of Whinston (2003). Participation in the associations
should be considered as non-contractible investments for the relation-specific skill. The empirical implications
of some specified models concern the effects on a vertical integration likelihood of both the importance of
buyers’ or sellers’ non-contractible investments and specificity in the acquired relation-specific skills. We
estimate an equation of vertical integration wherein the determinants are dummy variables of the parent firm
and the subsidiary’s participation in the cooperation associations and variables representing the degree of their
relation specificity. The significance and the signs of these variables suggest that, other than a model of
exogenous acquisition of relation-specific skills, a model can be also applicable to the Japanese auto parts
suppliers-manufacturers, where it is not the manufacturers’ but instead are the suppliers’ investments which
create their own relation-specific skills through the association activities. The Asanuma hypothesis turns out to
be alive.

Measuring (In)Attention to Mutual Fund Fees: Evidence from Experiments (D1, G4)

Hugh Hoikwang Kim
,
University of South Carolina
Wenhao Yang
,
University of South Carolina

Abstract

We estimate investors’ attention level to mutual fund fees based on a parsimonious asset allocation model with limited attention in an experimental setting. We find that, on average, investors allocate 62.5% of their full attention to mutual fund fees. We also find that cognitive ability is an important factor in determining investors’ attention to the fees. The estimated attention level implies investors in the U.S. mutual funds market pay $16 billion more in fees per year than the level they recognize. We evaluate policy options to increase attention to fees and find that the policy effectiveness may depend on one’s cognitive ability.

Medicaid Cuts and Consumer Financial Outcomes: Evidence from Missouri (I1, D1)

James Bailey
,
Providence College
Nathan Blascak
,
Federal Reserve Bank of Philadelphia
Slava Mikhed
,
Federal Reserve Bank of Philadelphia

Abstract

In July 2005, a set of cuts to Medicaid eligibility and coverage went into effect in Missouri. Eligibility requirements became more stringent, requiring lower incomes for some workers, more spenddown of assets for the elderly, and entirely cutting the Medical Assistance for Workers with Disabilities program. Medicaid coverage also become less generous even for those who retained eligibility, cutting dental and optometric coverage and requiring co-pays and premiums from some relatively high-income recipients. Overall these cuts removed about 100,000 Missourians from the program, and reduced the value of the insurance for the remaining 375,000 enrollees. Using data from the Consumer Expenditure Survey, we show how these cuts led to increased spending on insurance premiums and medical services by the affected groups. Using data from the Federal Reserve Bank of New York/Equifax Consumer Credit Panel (CCP), we show how the Medicaid cuts affected a series of consumer financial outcomes including credit card balances and delinquent debt. We employ a variety of difference-in-difference empirical strategies, including comparisons of Missouri to nearby states that did not make changes to their Medicaid programs; comparisons of border cities that span Missouri and non-cutting states; and comparisons with Missouri based on Medicaid eligibility. We also take advantage of a partial reversal of the Missouri cuts in 2007. We compare our results to a broader literature on Medicaid and consumer finance that has generally measured the effects of Medicaid expansions rather than cuts, aiming to determine which outcomes are asymmetric.

Migration, Education and Urban Divergence: Evidence from United States Patent Counts (J2, R2)

Oudom Hean
,
Ohio State University
Nattanicha Chairassamee
,
Ohio State University
Mark D. Partridge
,
Ohio State University

Abstract

The skill-biased technical change empirical literature has been silent regarding two key channels of labor supply, migration and demand for education, through which technology increases urban divergence. Using patents as a measure of skill-biased technical change, this paper sheds new light on technology-induced spatial inequality. Instrumental variables estimations show that between 2005 and 2015, which corresponds to the decade of relatively low U.S. migration rate, the educational channel played a greater role in explaining skill divergence between U.S. urban counties. We also find positive relationships between urban income/wage divergence and patent counts in computer and data processing, telecommunications, and automation.

Mind the Basel Gap (G2, E5)

Petri Jylhä
,
Aalto University
Matthijs Lof
,
Aalto University

Abstract

The so-called Basel gap, the difference between a country’s credit-to-GDP ratio and its estimated long-term trend, is used as a basis for setting the countercyclical regulatory capital buffers under the Basel III regulatory framework and, hence, has important real implications. In this paper, we show that the Basel gap, estimated by a seemingly sophisticated recursive filtration method, is nearly equivalent to a naive 16 to 18 quarter change in the credit-to-GDP ratio.

The Basel gap was developed by Drehman et al. (2010) as an early warning signal of financial crises. It is calculated as a country’s credit-to-GDP ratio minus the long-term trend of the ratio. As with any such actual-minus-trend gap measure, a crucial step in constructing the Basel gap is defining the long-term trend of the credit-to-GDP ratio. Drehman et al. (2010) employ a recursive Hodrick-Prescott (1981) filter with a very large smoothing parameter (400,000) to calculate the trend. In this paper we show that the recursive trend estimation mechanically results in a trend that simply lags the original time series. Since the estimated trend closely approximates the lagged credit-to-GDP ratio, the estimated gap closely approximates the change in the ratio.

The equivalence between credit gaps estimated recursively by the HP filter and by simple changes in the credit-to-GDP ratio holds for both smaller and larger values of the smoothing parameter, with the difference that a higher smoothing parameter generates a gap that approximates a longer difference in the credit-to-GDP ratio. Using data from 44 countries as well as simulated data, we find that the gap estimated with a smoothing parameter of 400,000 closely approximates the 16-18 quarter change in the credit-to-GDP ratio.

Monetary Policy and Heterogeneity of Bank Risk Taking: Evidence from China (G2, E5)

Jiaming Zhang
,
China Bohai Bank and Nankai University
Xiangrong Deng
,
Nankai University
Xueliang Feng
,
Nankai University

Abstract

In this paper, we construct a model incorporating the bank liability-side risk taking into the bank risk taking channel of monetary policy, investigating the impact of monetary policy on heterogeneous bank risk taking, and conduct an empirical test with panel data of 155 banks in China. The results of our model show that monetary policy not only affects bank asset-side risk taking, but also bank liability-side risk taking. In addition, monetary policy transmits through bank liability-side risk taking to bank asset-side risk taking. The empirical results show that, monetary policy has significant and asymmetric impact on bank asset-side risk taking and liability-side risk taking. Specifically, loose monetary policy leads to an increase in bank asset-side risk taking but a decrease in bank liability-side risk taking, while tight monetary policy results in an increase in bank liability-side risk taking but a decrease in bank asset-side risk taking. Further, bank liability-side risk taking play a mediating role in the transmission of monetary policy to bank asset-side risk taking, and the more the bank liability-side risk taking, the less the bank asset-side risk taking.

Monetary Transmission and Inflation Expectations in the Euro Area (E5, C5)

Catalina Martinez Hernandez
,
Free University of Berlin and DIW Berlin

Abstract

In this paper, I empirically investigate the role of inflation expectations for explaining heterogeneous monetary transmission in the four largest countries of the Euro Area. I analyze if imperfect information in the expectations-formation process explain potential differences in the responses of inflation to monetary policy shocks. Since the number of variables that the ECB monitors is large, I concentrate on a data-rich environment framework and propose a Large Bayesian Proxy Vector Autoregression. I estimate this model by combining two strands of the literature on Bayesian Macroeconometrics: The Large Bayesian VAR and the Bayesian Proxy VAR. I find evidence of incomplete information in the formation of expectations and distinguish that for the case of Spain, these rigidities can explain the heterogeneous response of inflation compared to the aggregate response of the Euro Area. The policy implications of this paper are crucial for understanding the role of inflation expectations for monetary transmission especially in periods of unconventional tools.

Multi-Dimensional Social Networks and Employment Opportunities: Evidence from Rural India (O1, J2)

Youjin Hahn
,
Yonsei University
Jun Sung Kim
,
Monash University
Hee-Seung Yang
,
KDI School of Public Policy and Management

Abstract

We study the impact of multi-dimensional social networks on employment outcomes. Using data on social networks in rural India, we find that females’ labor supply is influenced by peers, such as friends, relatives and risk sharing partners, while males’ labor market outcomes seem to be unaffected by their peers. Peer effects of females are mostly driven by the aggregate outcome of peers (the social multiplier effect) instead of the average outcome of peers (social norm effect). The same-gender friends and relatives have stronger social multiplier effects on females than the risk sharing networks. Our analysis indicates that social network information can be utilized to improve women’s employment outcomes in rural communities, in particular by targeting central individuals in their social network.

Music Consumption Decisions with Non-Durable Streaming Options (L2, D1)

Jason Walter
,
University of Wisconsin-Stout
R. Scott Hiller
,
Fairfield University

Abstract

Consumers are increasingly purchasing non-durable music products, consumed through a streaming bundle delivered via a subscription model. In this paper we examine how individual preferences influence a consumer's music format decision. We analyze consumption differences between durable retail music products and non-durable streaming music subscription bundles. A user's preferred format depends on the intensity of their music interests, scope of interests, and how quickly a song's utility depreciates. Our empirical analysis shows that streaming consumers have greater depreciation rates than the traditional distribution of terrestrial radio, and that digital sales decline at a slower rate than does the usage of the streaming version. Our theory model and empirical evidence suggest that consumers prefer a non-durable subscription over a durable purchase of information goods when they have higher depreciation rates or a greater scope of music interests. Using simulation, we identify the ideal consumption format for consumers based on their individual listening preferences.

New Digital Technologies and Heterogeneous Employment and Wage Dynamics in the United States: Evidence from Individual-Level Data (J2, O3)

Frank M. Fossen
,
University of Nevada-Reno
Alina Sorgner
,
John Cabot University-Rome

Abstract

We investigate heterogeneous effects of new digital technologies on the individual-level employment- and wage dynamics in the U.S. labor market in the period from 2011-2018. We employ three measures that reflect different aspects of impacts of new digital technologies on occupations. The first measure, as developed by Frey and Osborne (2017), assesses the computerization risk of occupations, the second measure, developed by Felten et al. (2018), provides an estimate of recent advances in artificial intelligence (AI), and the third measure assesses the suitability of occupations for machine learning (Brynjolfsson et al., 2018), which is a subfield of AI. Our empirical analysis is based on large representative panel data, the matched monthly Current Population Survey (CPS) and its Annual Social and Economic Supplement (ASEC). The results suggest that the effects of new digital technologies on employment stability and wage growth are already observable at the individual level. High computerization risk is associated with a high likelihood of switching one’s occupation or becoming non-employed, as well as a decrease in wage growth. However, advances in AI are likely to improve an individual’s job stability and wage growth. We further document that the effects are heterogeneous. In particular, individuals with high levels of formal education and older workers are most affected by new digital technologies.

Non-Linear Pricing with Reneging (D8, C7)

Menghan Xu
,
Xiamen University
Yujing Xu
,
University of Hong Kong

Abstract

This paper studies a dynamic non-linear pricing problem, adding the possibility that the seller can costly renege on the initial contracts, which is common in reality in the forms of false advertising, add-on pricing and bait-and-switch. While reneging allows the seller to earn more surplus by offering a new full-extraction contract after learning the buyer’s preference, a forward looking buyer will hide information. We fully characterize the equilibrium direct mechanism with the presence of this strategic interaction and show that the quality distortion may be mitigated and participation can be higher when the market moves from full-commitment to one with modest reneging cost. Moreover, the seller may strategically choose not to screen by providing a pooling mechanism and we show that such mechanism is welfare improving compared with a (partial-)separating one. We establish the precise condition under which the welfare improvement happens and further relate it to whether the market is niche or mass. In addition, we show that in the sequential equilibrium, there always exists an implementable contract throughout the game even if the seller has already incurred the reneging cost and is free to modify it. By explicitly modeling seller’s information extraction problem without full commitment, our results have policy implications on protecting consumers from deceptive business tactics.

Nonprofits as Innovators: Profit Status and Competition in the Mixed Microfinance Industry (D2, M1)

Sarah Wolfolds
,
Cornell University

Abstract

The competitive dynamics between organizations with distinct business models has only recently become an area for theoretical and empirical analysis. Using the setting of microfinance where non-profits and for-profits directly compete, this paper develops a formal analytical model investigating the objectives that organizations pursue in order to understand the resulting differences in business models. I find support, using a firm-level panel dataset from 2003 through 2012 in Latin America, for the model’s predictions of differentiation in the wealth of borrowers served by the firm’s profit status, as well as differentiation by its sources of financing. In addition, allowing for the endogenous choice of sources of financing leads to the hypothesis
that non-profits may focus on even lower income borrowers when they face more for-profit competitors. These results suggest a continued role for the non-profit organization even in the face of better-capitalized for-profit competitors, and I conclude by suggesting this role is to provide innovative products and create and legitimize new markets.

Optimal Allocation of Sample Sizes in Group-Randomized Trials (C9, C1)

Zuchao Shen
,
University of Florida
Ben Kelcey
,
University of Cincinnati
Gang Wang
,
Winthrop University

Abstract

Group-randomized trials have been widely used in education and social sciences to evaluate causal effects of interventions. Two key considerations in designing experimental studies are statistical power and the minimal use of resources. Optimal design frameworks simultaneously address both of these considerations. In this paper, we developed a more flexible framework that expands the scope of optimal design considerations to include, for example, varied sampling costs across treatment conditions and levels of hierarchy. The results suggest that the proposed framework can identify more efficient designs and that it is robust to misspecifications of concomitant design parameters (e.g., intraclass correlation coefficient). The resulting framework applies to a wide range of group-randomized trials and is implemented in the R package odr. The presentation is designed to be accessible to a wide range of scholars and emphasizes conceptual understanding and practical application (e.g., R package odr).

Optimal Patent Policy for the Pharmaceutical Industry (O3, I1)

Olena Izhak
,
Düsseldorf Institute for Competition Economics
Tanja Saxell
,
VATT Institute for Economic Research
Tuomas Takalo
,
Bank of Finland

Abstract

We derive formulas for the design of the optimal patent policy that are functions of competitors’ responses to changes in the effective duration and breadth of new drug patents prior to their expiration. To estimate these responses, we use data on Paragraph IV patent challenges by generic firms in the US, and two quasi-experimental approaches: one based on changes in patent laws and another on the allocation of patent applications to examiners. We find that extending effective patent duration increases Paragraph IV entry whereas broadening protection reduces it. The formulas combined with these results imply that pharmaceutical patents should be shorter but broader.

Optimal Stratification of Platforms: Two-Sided Markets Design with Mechanism Design Approach (D4, L1)

Gaoyang Cai
,
Tsinghua University
Yong Wang
,
Tsinghua University
Danxia Xie
,
Tsinghua University
Yitao Lv
,
Tsinghua University

Abstract

We study the optimal stratification of platforms (two-sided markets) with mechanism design approach. Market designer faces two types (high and low) of buyers and sellers whose type information is private to themselves. Market designer can stratify one centralized platform into several sub-platforms to maximize the total profit. To implement platform stratification, a set of incentive compatible conditions must be satisfied for each sub-platform. The efficiency of platform stratification is compared to the benchmark case of non-stratification, i.e. one centralized platform that accommodates all types of buyers and sellers together without stratification. We derive and characterize the conditions under which platform stratification dominates the benchmark of non-stratification in terms of social welfare and platform profits respectively. Finally, we generalize the discussion to the case of buyers and sellers with many or continuously-distributed types, and investigate the market design problem of optimal platform stratification, including choosing the optimal number of sub-platforms while corresponding sets of incentive compatible constraints being satisfied.

Political Network, Work Experience, and Economic Performance: Evidence from China (H1, P3)

Yukun Sun
,
Zhongnan University of Economics and Law

Abstract

In this paper, I study the impact of networks and experience of top local leaders on economic growth. Employing the detailed data of top provincial and prefecture leaders in China between 1991 and 2014, I find that political networks and work experience are the important determinants on the promotion of local leaders. Using a difference-in-difference estimation method, I find that local leaders that having connections with central leaders are more likely to get promoted. While politicians with work experience as top local leaders are more likely to have better economic performance during their tenure. These findings are robust to various sensitivity tests. My study adds evidence to a growing literature emphasizing the role of political incentives of government officials in promoting local economic growth.

Pollution on the Growth Path: The Calibration, Simulation and Prediction of Environmental Kuznets Curve in China (O1, Q5)

Zhe Fu
,
University of International Business and Economics
Yanmei Sun
,
University of International Business and Economics
Jianye Yan
,
Peking University

Abstract

The fast and persistent (till now) growth of China gives rise to the unavoidable concern on environmental pressure. First we add micro-foundation into the Green Solow Model (Brock and Taylor, 2010), by which we put one step further to the theoretical aspect of the research on canonical Environmental Kuznets Curve (EKC). Then we calibrate based on Chinese economic data. From theoretical and simulated perspective, our research provides a deeper essence of support to the majority of empirical conclusions by Chinese scholars on the topic of judging where should China be positioned on EKC and some related questions. We confirm that China has not achieved the turning point of EKC, i.e., in the stage that pollution and emission positively relate to the economic growth. Furthermore, based on current empirical evidence, our quantitative simulation from our theoretical model predicts that the turning point of inverted-U in EKC for China would happen in 2036 or so. Besides, we also expect that our paper supplements the EKC literature from the perspectives of methodology, viewpoints and reasoning.

Predicting Consumer Spending Using Expenditure Micro-Data (C5, D0)

Bibaswan Chatterjee
,
Xi’an Jiaotong University
Dooti Roy
,
Boehringer Ingelheim

Abstract

Predictive models using machine learning are increasingly being used in economics to leverage large datasets in testing theories and discovering effects, along with traditional econometric techniques. To that extent we use data on spending to predict the purchasing behavior of consumers. By using data on prior spending patterns, we see if we can link demographic and income-education information of groups of people to their spending on different consumption bundles. As a measure of consumption patterns, we use deviation of spending from prior means for a consumption bundle. Controlling for the year and month of spending expenditures, we expect to be able to see how much of the spending is predicted by non-business cycle facts. In the course of our exploration, we would like to highlight the importance of education in affecting spending patterns.


If spending datasets can be used to predict if people with certain characteristics on average are likely to spend more or less than the existing average levels of spending,then we could potentially design policies targeted towards such demographics.

Price Dispersion Across United States Cities: The Role of Walmart (L1, F0)

Md Rafayet Alam
,
University of Tennessee-Chattanooga
Akm Mahbub Morshed
,
Southern Illinois University-Carbondale

Abstract

This paper examines the effects of Walmart on price dispersion across US cities. Although US retail architecture has been changed over the past two decades by the expansion of big-box chain-stores such as Wal-Mart, the impact of such market consolidation on price dispersion is not examined rigorously in the literature. Using opening/conversion dates of Walmart and US city level individual good-specific quarterly retail price data for 101 cities over 25 years, we examine the role of Walmart on price dispersion. Our analysis shows that price dispersion in a city-pair increases when Walmart is opened in one of the cities of the pair. When Walmart is opened in the other city of the city-pair, price dispersion falls from that increased level but the fall is not enough to bring the price dispersion back to the level where there is no Walmart in the city-pair. This finding implies that Walmart drives down prices in ‘early’ cities more than in ‘late’ cities and that Walmart wields its market power gradually. Our analysis also shows that in the context of price dispersion, the presence of Walmart in one city in a city-pair is equivalent to adding between 500 and 1000 miles distance between the two cities while the presence of Walmart in both the cities in a city-pair is equivalent to adding distance between 200 and 700 miles depending on the commodity under consideration.

Price Sensitivity and Substitution Among Prescription Medications: Evidence from the Medicare Part D Donut Hole Closure (I1, D4)

Cameron Kaplan
,
University of Southern California

Abstract

Prescription drugs are an increasingly large contributor to healthcare costs, and access to these medications critically depends on out-of-pocket costs. Policies to lower the cost of drugs should theoretically increase utilization, but it is unclear whether there are heterogeneous effects across different types of drugs based on their characteristics (e.g. addictive vs. non-addictive, number of medications in the class, tolerability, use for mental health conditions, etc.). The ACA sought to lower prescription drug costs in Medicare by closing the so-called prescription drug donut hole. Drugs have been covered under Medicare Part D since 2006, and from 2006-2010, patients who were not eligible for low-income subsidies faced a large coverage gap (donut hole) where they were responsible for the full cost of medications after their spending on medications exceeded approximately $2500 per year. Following the passage of the ACA, the coverage gap began to partially close in 2011, starting with a 50% discount for brand-name drugs in the initial year. This led to a change in the relative price between brand and generic drugs for prescriptions filled during the coverage gap. Using a 100% sample of Medicare claims data from 2006-2016, we examine the impact of the coverage gap closure on utilization of drugs in each of 464 therapeutic classes using a difference-in-difference model comparing individuals who are enrolled in the low-income subsidy program (and do not face a coverage gap) to all others. We then use the results from these separate regressions to perform a meta-regression to examine factors associated drugs that experience larger vs. smaller price effects. Finally, we examine cross price elasticities by comparing brand vs. generic utilization within class.

Product Hopping and Innovation Incentives (O3, L4)

Jorge Lemus
,
University of Illinois-Urbana-Champaign
Olgu Ozkul
,
University of Illinois-Urbana-Champaign

Abstract

We study innovation incentives in the presence of “product hopping,” whereby the incumbent patents a minor modification of a drug (e.g., a new delivery method) and invests in marketing to switch demand towards the minor modification. In our setting firms compete sequentially to discover two innovative drugs. The winner of the first R&D race (the incumbent) can alter the market structure that follows the second R&D race through product hopping. This can increase investments during the second R&D race when product hopping softens competition or when the incumbent benefits from becoming a multi-product monopolist. The change in expected continuation values can increase or decrease investments during the first R&D race. Thus, the welfare effect of product hopping is ambiguous. We discuss our results in the context of the current policy debate on product hopping, welfare, and antitrust.

Public Annuities: Buyers’ Behavior and Policy Design (H5, E6)

Qilin Zhang
,
University of Hong Kong
Sau-Him Paul Lau
,
University of Hong Kong

Abstract

Motivated by Diamond (2004) and the experiences in various economies adopting the defined contribution pension system, we study the number, form and offering time of public annuities when individuals’ survival probabilities and strength of bequest motive are heterogeneous. After obtaining a useful measure of the severity of adverse selection, we examine public annuities with a guarantee clause that a fraction of the annuity payout is still received by the buyer’s beneficiary after she dies. We show that an increase in the guarantee proportion is generally able to mitigate adverse selection. However, when the heterogeneities in survival probability and bequest strength are limited, the guarantee clause is irrelevant. Regarding the offering time, we find that a deferred public annuity contract offered at a young age eliminates adverse selection and crowds out the transaction of the immediate annuity contract offered at an older age. Our results are useful to countries planning to introduce public annuities.

Public Listing Choice with Persistent Hidden Information (G3, G2)

Francesco Celentano
,
University of Wisconsin-Madison
Mark Rempel
,
University of Wisconsin-Madison

Abstract

The listing propensity of US firms on top stock exchanges has substantially
declined since the late 1990s. While increasing transparency has been imposed on
public firms since the Sarbones-Oxley Act, the increased reliance of US firms on
private funding sources predates these reforms suggesting more fundamental, technological
changes in the environment. Over this same period, firms have become increasingly
intangible as well as compensation to public executives has
increased but become more volatile. To rationalize these trends, we build a model of
firm contracting with competition between private and public funding financiers. We
empirically validate the model by testing its ancillary predictions on Capital IQ data
of both private and public firms.
Firm managers are risk averse and have private information over the productivity
of the firm. Due to the persistence of this private information, to preclude
the manager from diverting cash-flows, the optimal contract of an uninformed financier
will induce excessive volatility and higher drift in the compensation of managers in excess
of the underlying cash-flow process. With the subject-matter expertise and the
non-disclosure agreements available by private financiers, private financiers are able to
observe the realized productivity path of the project and save on the compensation to
the manager paid to align incentives. Private funding is however limited, and must
be acquired ahead of time as opposed to the deep-pockets of public markets. This
rationing of private financiers leads to a selection of higher volatility, more intangible
firms being funded by private financiers and the residual funded by the public markets.
Using compensation and balance sheet data on a large sample of public and
private firms, we empirically test and find support
for the model predictions. Our results suggest that the changing composition of firms
has been a driver of both the decline in listed firms and the growth of
compensation of public executives.

Racial/Ethnic Disparities in Rural-Urban Mortality Gap in the United States: A 24 Years Longitudinal Study (I1, R1)

Nasim B. Ferdows
,
University of Southern California
Soroosh Baghban Ferdows
,
Istanbul Technical University
Amit Kumar
,
Northern Arizona University

Abstract

This research aims to study racial and geographical disparities in the US by comparing the race and sex-specific mortality trends of the US rural and urban populations. We created a longitudinal county level analytic file of the US population 65 years and older, over the period of 1968 to 2015 obtained from Compressed Mortality Files (CMF) from CDC-WONDER and Area Health Resources Files (AHRF). First, we used an OLS regression of age-adjusted mortality rate onto year indicators interaction with race and gender to depict the race and sex-specific trend in age-adjusted mortality rates. We also estimated the change in in mortality rate over time, for each race and gender, relative to values in 1968. Finally, we estimated race and sex specific trend in rural-urban mortality gap using state fixed effects regression. Our results indicate that overall mortality rate in the US older population has declined steadily over the last five decades, and the racial gap in mortality rate have been declining since early 2000s, with a more considerable decline for females. However, decomposing the trends into the rural and urban populations, the racial gap in mortality rates has only declined in urban areas. Moreover, mortality rates of the whites in rural areas declined more rapidly than their Black counterparts, resulting in a gap that has been widening in the last three decades. The racial gap has increased considerably for the male population residing in rural counties not adjacent to an urban county. Thus, racial disparity in mortality has increased in rural areas, with a considerable widening between white and black male population living in the more remote rural areas.

Reassessing Trade Barriers with Global Value Chains (F4, F1)

Yuko Imura
,
Bank of Canada

Abstract

This paper provides a systematic, quantitative analysis of the short-run and long-run effects of various trade-restricting policies in the presence of global value chains and multinational production. Using a two-country DSGE model with endogenous firm entry and exit in exporting and multinational production, I compare the effects of (i) tariffs on final-good imports, (ii) tariffs on intermediate-input imports, and (iii) barriers to access a foreign market.
I show that, in the long run, all three policies lead to a recession in both countries, but the relative effects on the two countries’ GDP vary across policies. At the firm level, less productive exporters exit from the destination market, while the most productive few find it profitable to locate production in the foreign country as multinationals, thereby partially recovering the loss from exporting. In the short run, the dynamics differ across policies and also from their long-run outcomes. Final-good tariffs and market-access barriers lead to a temporary production boom in the policy-imposing country, while intermediate-input tariffs result in an immediate recession in both countries. The latter also discourages multinational operation over the short run when the input tariffs dominate the declining costs of labor and capital.

Reference-Dependent Preferences for Information (H0, D9)

Tabaré Capitán
,
University of Wyoming
Linda Thunstrom
,
University of Wyoming
Klaas Van 't Veld
,
University of Wyoming
Jonas Nordstrom
,
Lund University and University of Copenhagen

Abstract

If people’s preferences for information are reference-dependent, welfare analyses of information policies – such as Cost-Benefit Analyses – would also depend on the referent. For example, a group of people used to see calorie information might value that information more than a group that is not used to see calorie information, this difference might translate into differences between the willingness to pay (WTP) to receive the information and the willingness to accept (WTA) not to receive the information across groups – both are measures of their preferences for information. We conducted a laboratory experiment to test if preferences for information are reference-dependent and found that more people chose to see calorie information in the endowed treatment than in the not endowed treatment – i.e., there is an endowment effect. First, the data is incompatible with the sharp null hypothesis of no treatment effect for all participants – i.e., the treatment has an effect. Second, the data is compatible with the bounded null hypothesis of a weakly positive treatment effect for all participants – i.e., the effect of the treatment is weakly positive –. These results are based on randomization inference and we claim internal validity. As a complement, we show that the data is incompatible with the weak null hypothesis of zero average treatment effect. Like previous studies, we find that experience reduces the endowment effect. All results are robust to alternative model specifications. This study contributes knowledge to those evaluating welfare effects of information policies: Benefits to information will depend on the welfare measure used (WTP or WTA).

Regulating Financial Networks Under Uncertainty (G1, E0)

Carlos A. Ramirez
,
Federal Reserve Board

Abstract

I study the problem of regulating a network of interdependent financial institutions when there is uncertainty regarding its precise structure. I show that such uncertainty reduces the scope for welfare improving interventions. Although acquiring institution-level information potentially reduces this uncertainty, it does not always lead to welfare improvements. Under certain conditions, regulation that reduces the risk-taking incentives of a small set of institutions can improve welfare. The size and composition of such a set crucially depends on the cost of acquiring institution-level information, the cost of regulating institutions, and investors’ preferences.

Regulatory Effects on Short-Term Interest Rates (G2)

Patrick Schaffner
,
University of St. Gallen
Angelo Ranaldo
,
University of St. Gallen
Michalis Vasios
,
Bank of England

Abstract

We analyse the effects of EMIR and Basel III regulations on short-term interest rates.
EMIR requires central clearing houses (CCP) to continually acquire safe assets, thus expanding the lending supply.
Basel III, in contrast, disincentivises the borrowing demand of repurchase agreements (repo) by tightening banks' balance sheet constraints.
Using unique datasets of repo transactions and CCP activity, we find compelling evidence for both supply and demand channels. The overall effects are decreasing short-term rates and increasing market imbalances in various forms, all of which entail unintended consequences originated from the new regulatory framework.

Regulatory Races Revisited (C3, L5)

Sanchari Choudhury
,
Saint Xavier University
Jayjit Roy
,
Appalachian State University

Abstract

Regulatory competition across U.S. states has been a debatable topic among economists for the last few decades as the extant literature is inconclusive and contradictory about the direction of the race. Empirical studies on this issue have majorly concentrated on regulatory restrictions in four particular areas, namely labor-, environmental-, banking- and corporate governance-regulations. Though these studies are consequential, they restrict the scope of an analysis to a very specific regulatory context. To overcome this notable shortcoming, we use the first panel data set on federal regulation of all industries - published recently in 2015 - to re-address the question of regulatory races between jurisdictions.

Return of the Bond-Price Support Regime: Bank of Japan’s Dual Bond-Purchase Program (E5, G1)

Takahiro Hattori
,
Ministry of Finance of Japan
Jiro Yoshida
,
Pennsylvania State University

Abstract

This is the first study that analyzes the Bank of Japan’s dual bond-purchase program since 2016 consisting of fixed-amount operations through auctions and fixed-rate (i.e., unlimited-amount) operations. This Yield Curve Control (YCC) regime as part of the Quantitative and Qualitative Easing (QQE) exhibits similarities to the Fed’s bond-price support regime during WWII. We show that bond yields across the entire yield curve became stationary processes suggesting credible monetary policy. Our intra-day analysis additionally demonstrates that the largest fixed-rate operation on July 30, 2018, was effective in controlling government bond yields but not interest-rate swap rates.

Risk Aversion, Credit and Banking (G2, E4)

Jonathan Benchimol
,
Bank of Israel
Caroline Bozou
,
Panthéon-Assas University

Abstract

The degree of risk aversion between households, firms, and bankers is differentiated to assess its influence in the business cycle. A Constant Relative Risk Aversion (CRRA) utility function is assumed for all agents to compare three different transmission channels: standard, financial and risk aversion shocks. Our non-linear framework allows a reinterpretation of economic and financial dynamics under several risk aversion levels and fluctuations. Agents' risk aversion is found to be an essential indicator for policymakers. We find that an increased risk aversion level generally attenuates the response of output to economic and financial shocks. A positive risk aversion shock substantially influences the real economy. Differentiating the degree of risk aversion between agents matches better the business cycle, while a risk aversion shock is found to substantially influences central and retail bank interest rates through consumption smoothing and precautionary saving behaviors.

Risk Pooling, Leverage, and the Business Cycle (E3, G2)

Pietro Dindo
,
Ca' Foscari University of Venice
Andrea Modena
,
Ca' Foscari University of Venice
Loriana Pelizzon
,
Goethe University Frankfurt

Abstract

This paper investigates the interdependence between the risk pooling activity
of the financial sector and the macroeconomic and financial dynamics in a general
equilibrium model of a productive economy. Due to their exposure to idiosyncratic
shocks, heterogeneous entrepreneurs are willing to mitigate their risk through a financial sector.
After the payment of an intermediation cost, the financial sector
pools risky claims issued by different firms, and uses leverage to profi t from
this opportunity. Exogenous systematic shocks change the relative size of the financial
sector, and thus the equilibrium amount of pooled risk, making financial
leverage state-dependent and counter-cyclical. We study how this mechanism endogenously
channels ampli cation and mitigation of both output and consumption fluctuations.
In equilibrium, financial leverage also determines Sharpe ratios and
risk-free interest rates, as a function of intermediation costs as well as idiosyncratic
and systematic volatility. Last, we investigate the effect of leverage constraints.
When they are binding, the share of pooled idiosyncratic risk is sub-optimal, although
associated with milder fluctuations of aggregate consumption, and phases
of low capitalization become more persistent. Overall, we find leverage constraints
to be welfare-improving for the entrepreneurs. This suggests that there exists an
optimal size of the financial sector as well as of leverage constraints.

Robots and Labor Market Adjustments (O3, J2)

Wolfgang Dauth
,
University of Wuerzburg and IAB
Sebastian Findeisen
,
University of Mannheim
Jens Suedekum
,
University of Dusseldorf
Nicole Woessner
,
University of Dusseldorf

Abstract

We estimate the effect of industrial robots on employment, wages, and the composition of jobs in German labor markets between 1994 and 2014. We find that the adoption of industrial robots had no effect on total employment in local labor markets specializing in industries with high robot usage. Robot adoption led to job losses in manufacturing that were offset by gains in the business service sector. We analyze the impact on individual workers and find that robot adoption has not increased the risk of displacement for incumbent manufacturing workers. They stay with their original employer, and many workers adjust by switching occupations at their original workplace. The loss of manufacturing jobs is solely driven by fewer new jobs for young labor market entrants. Moreover, we find that, in regions with higher exposure to automation, labor productivity increases while the labor share in total income declines.

Rural-Urban Migration, Family Arrangement and Children’s Well-Being: Evidence from China’s Rural Areas (R2, I0)

Lili Wei
,
Lanzhou University
Jing Zhang
,
Lanzhou University
Ying Yang
,
Lanzhou University
Hao-yu Cao
,
Lanzhou University
Yu-qi Hou
,
Lanzhou University

Abstract

China's rapid urbanization process and large-scale rural-to-urban population migration have changed the family structure in China's rural areas. Parents migration and family arrangements have profoundly affected children's well-being. Hukou system and the economic situation in rural areas have made migrant parents face whether they take their children to the city or leave them behind. The changes in family arrangements and the gap between urban and rural development in China have made a huge difference in the education, health, cognition and happiness of children. Based on the China Household Survey data of rural hukou children aged 10-15, we examine the impact of family arrangements on the comprehensive well-being of migrant children and left-behind children. We find that parents migration is a double-edged sword: parents migration reduces the happiness and family relationships of left-behind children, but it is conducive to children’s education; for migrant children, parents migration improves children's happiness and education level, social cognition and family relationships, but it is not conducive to the children's health and life schedule. Parents urban-rural migration has significant negative impacts on children's language expression skills. From the perspective of internal structure of family arrangements, our findings demonstrate left-behind children living with one parent only, which has negative performance on language expression, health, initiative of communication with parents; left-behind children with no parents, living with grandparents and others is not conducive to the happiness, social cognition, and that improves children’s depression tendency. Migrant children living with father only, happiness and health are significantly reduced, but their academic performance tend to do better, especially in mathematics.

School Outcomes, Information Campaigns and Parental Involvement: Evidence from a Sequence of Two RCTs in the Dominican Republic (I0, O0)

Daniel R. Morales
,
Instituto Dominicano de Evaluación e Investigación de la Calidad Educativa & Pontificia Universidad Católica Madre y Maestra
Claudia Curiel
,
Dominican Institute for Evaluation and Research of Educational Quality (IDEICE)
Carlos Schmidt-Padilla
,
University of California-Berkeley
Patricia Mones
,
Dominican Institute for Evaluation and Research of Educational Quality (IDEICE)

Abstract

In 2017, the Dominican Republic's Ministry of Education began implementing triannual diagnostic evaluations in third and sixth grade evaluating students in language, math and science. Exams are followed by a "school report card" with an analysis of the main results tailored to the school's leadership, teachers and parents to use as inputs to improve education quality. Despite this effort, little is known regarding to what extent these reports are used in schools and at home to guide policy and set benchmarks for improvement.

To assess whether reports are used and if they have an impact on school quality, we worked with the Ministry of Education to design an information campaign that provides specific instructions to school principals on how to get the reports, how to analyze them and design and improvement plan, and how to organize three different workshops among the school’s leadership team, teachers, and parents, respectively. As part of this information campaign, we ran two randomized control trials to raise workshops compliance:

RCT 1 (Morales, Rodriguez and Schmidt-Padilla, 2019a): Improving School Outcomes through Information Campaigns. https://doi.org/10.1257/rct.4954-1.0.

RCT 2 (Morales, Rodríguez and Schmidt-Padilla, 2019b): Improving School Outcomes through Parental Involvement. https://doi.org/10.1257/rct.5048-1.0.

This is an ongoing evaluation that shows the importance of verifying that the information reaches the agents, and that reminding them about their tasks has an impact on the adoption of policies for school principals.

The preliminary results of the RCT1 show that reminders increase compliance levels by more than 400% for school principals. Also, that encouraging parents to ask principals about the school results (and the workshop that should be organized for them) increases the adoption of the workshops and the dissemination of the results of the diagnostic evaluations by more than 100%.

With these exogenous variations in the adoption of workshops, the impact of the use of reports and information diffusion through workshops on student learning levels will be estimated. The 3rd grade diagnostic evaluation will be in April 2020 and 6th grade in May 2021.

School’s Out: How Summer Youth Employment Programs Impact Academic Outcomes (I2, J1)

Alicia Modestino
,
Northeastern University

Abstract

Over the past several decades, many urban high schools have experienced little or no improvement in closing academic achievement gaps along socioeconomic and racial lines, highlighting chronic absenteeism among low-income and at-risk youth as a serious challenge for policies aimed at improving academic performance among these groups. In high poverty areas, as many as one third of all high-school students are chronically absent resulting in poor outcomes such as inability to read at grade level, increased risk of drop-out, and reduced rates of post-secondary enrollment. This paper provides experimental evidence regarding the impact of a large-scale Summer Youth Employment Program (SYEP) on high school students’ school attendance and academic performance during the school year after participation. Using an embedded randomized controlled trial (RCT), I find that the Boston SYEP has a significant impact on improving attendance but little evidence of improvements in academic performance. Youth who were randomly selected into the SYEP treatment group experienced significant improvements in reducing unexcused absences (-1.9 days) and achieving an attendance rate of greater than 85 percent (+2.9 percentage points) relative to the control group that provide a meaningful improvement in reducing chronic absenteeism relative to baseline (27 percent). I also find modest impacts on reducing course failures and increasing standardized test-taking, but no impact or disciplinary incidents. Moreover, these medium-term academic outcomes appear to be linked to improvements in social skills and academic aspirations among participants that occur during the summer, as measured by a pre-/post-program survey, and are greater in magnitude for older and “at-risk” youth. These results give policymakers some insight into the broader set of short-term program effects while also providing a look inside the “black box” as to how SYEPs affect youth in the long-run and which youth benefit the most.

Sleeplessness, Distraction and Stock Market Performance: Evidence from the World Cup (G4, G1)

Ko Chiu Yu
,
National University of Singapore
Manyi Yu Fan
,
National University of Singapore
Jinghan Cai
,
University of Scranton
Marco Richione
,
University of Scranton
Natalie Russo
,
National University of Singapore

Abstract

We study how sleeplessness and distraction impact global financial markets. Using the world's most widely viewed sporting event, the FIFA World Cup, we uncover significantly negative stock returns due to the sleeplessness from watching the games overnight and the distraction from the games during trading hours. The effects are stronger for countries with better soccer game records, and are independent of the sentiment effect due to game results documented in the literature.

Solve the Riddle on Infrastructure-Debt Nexus from Outer Space (E2, O1)

Xuehui Han
,
Asian Infrastructure Investment Bank
Yuan Cheng
,
Fudan University

Abstract

With the debt concerns amounting worldwide, large infrastructure investments are commonly blamed as the cause, especially for the less-development economies. Since satellite image on nighttime lights are unbiased and found being more reliable to reflect economic activities in countries with less-developed statistic system, in this analysis, we propose an approach to measure the effectiveness of infrastructure investment by using satellite image on nighttime light and examine whether the infrastructure investments have brought about the expected economic returns that can support debt sustainability. We use the Version 4 DMSP-OLS Nighttime Lights Time Series from 1992 to 2013 provided by NOAA to examine, following the infrastructure investment, (1) whether nighttime lights in the region are brighter; (2) how long it took to be brighter; (3) the effectiveness measure as the difference between the degree of light brightened and changes in infrastructure investment; and (4) whether the debt-infrastructure investment combination would harm the effectiveness of the infrastructure investments. A new matching method for causal inference with time-series cross-sectional data by Imai, Kim, and Wang (2018)1 is used to assess the causality.

With respect to the three types of infrastructures, such as paved roads, electricity generating capacity, and rails, we found: (1) the investment increase must be persistent to reach 8 years for rails, 6 years for electricity, and 3 years for paved roads to have the most significant economic return; (2)The best performing country-episodes in terms of generating economic return usually observe positive relationship with the government debt until certain level; and (3) sharp reductions in debt during investment periods would harm the economic returns for the following periods, especially for electricity generating capacity.

State Ownership, Intellectual Property Protection and M&A Manipulation: Evidence from China’s Mixed-Ownership Reform (G3)

Xiaoqian Zhang
,
Zhejiang University

Abstract

Intangible assets play an important role in Chinese M&A. Since they are difficult to measure, it may be manipulated in M&A. We investigate the effect by proxying them with R&D expenses and an earnings-based measure, adjROTA. Our evidence shows there is a U-shape between intangibles and M&A deal value. But for SOEs, this U-shape may have a trend of negative lineararity after China’s Mixed-ownership Reform. Tests on structural change and state-ownership implies that SOEs are sold at a lower price than POEs even it has higher intangibles, especially in those regions where local government defaulted in realizing their GDP targets. Our 2SLS with IVs from political turnover, i.e. CCP secretary leave age and tenure still, shows a robust significant effect. But the effect from CCP secretary becomes significant for POEs while it disappears for SOEs.

State Variations in Nurse Practitioners Scope of Practice and Rural-Urban Mortality Gap in the United States (I1, R0)

Nasim B. Ferdows
,
University of Southern California
Amit Kumar
,
Northern Arizona University

Abstract

Although nurse practitioners (NPs) might be the main providers of primary care to some communities, different states pursue different regulations for NP’s practice authority. This study compared the trends in age-adjusted mortality rates and physician supply in states with different prescriptive authority on NPs, using AHRF and CMF data. The outcome was age-adjusted mortality rate per 100k population for adults older than 65. Physician supply was defined as the number of physicians per 1k population. Linear regression with county fixed effects was used to evaluate the association between physician supply and mortality among rural and urban counties in states with different NP prescription authority. Age-adjusted death rate in rural counties in states with current restrictions on NPs declined from 5,500 to 4,800 deaths per 100k from 1992 to 2014. However, the death rate in rural counties in other states declined from 5,000 to 4,300 deaths per 100k population from 1992 to 2014. The supply of physicians in rural counties in states with current restriction on NPs increased from 0.9 to 1.1 physicians per 1k population from 1992 to 2014. While in states with no restrictions on NPs, the supply of physicians increased from 1.2 to 1.8 in the most remote rural counties and increased from 1.1 to 1.5 physicians per 1k population in states with prior restrictions on NPs. Our results indicate that rural-urban mortality gap is most considerable in states where NPs are not authorized to prescribe medication, compared to other states. Counterintuitively, the states with the restriction on NPs prescription authority have the lowest supply of physicians in rural areas, compared to other states’ rural counties. Therefore, people living in rural areas in states where NPs are not authorized to prescribe medication suffer from not receiving care due to both the lower physician supply and restrictions on NPs.

Subjective Expectations, Educational Choice Heterogeneity and Gender: Evidence from a Sample of Swedish High School Students (I2, J2)

Nikolay Angelov
,
Uppsala Center for Labor Studies
Per Johansson
,
Uppsala University
Mikael Lindahl
,
University of Gothenburg
Ariel Pihl
,
University of Gothenburg

Abstract

In this paper we elicit the subjective expectations of a sample of Swedish high school students, just before they officially make their field of study choice for university. This carefully collected data provides us with a window into the students’ counterfactual educational choices by summarizing their expectations about their top choice field of study, as well as about fields which they rank lower in their choice set. To take full advantage of this education choice data, we estimate mixed logit models, allowing for random coefficients, something which is rarely done in this literature. We show that stated preferences over educational amenities (including earnings, college experiences and work-life experiences) are heterogenous. This heterogeneity is significant both in terms of observables like gender, and unobservables. We find that both pecuniary and non-pecuniary expectations of outcomes of an educational choice strongly predict which field students chose, but non-pecuniary expectations are relatively more important for female students. We also apply our model to the gender gap in enrollment in STEM majors. The STEM gender gap in stated preferences can be fully explained by our elicited subjective expectations variables. Most important are the two taste variables capturing enjoyment of course work and of expected occupation, whereas expected earnings explains very little.

Surrender Contagion in Life Insurance: Modeling and Valuation (G2, G4)

Chunli Cheng
,
Sun Yat-Sen University
Christian Martin Hilpert
,
Sun Yat-Sen University
Aidin Miri Lavasani
,
University of Hamburg
Mick Schaefer
,
University of Hamburg

Abstract

This paper incorporates contagious surrender behavior into the pricing and regulation of participating life insurance contracts, allowing for structural default. The model features a financially sophisticated representative professional and several retail policyholders. In a model of surrender contagion combining a stochastic surrender intensity with an endogenous self-exciting effect, contract values and surrender behaviors of both policyholder types depend on each other. While professional’s surrender exploit retail policyholders, the latter benefit from a reasonable regulatory intervention. A riskier investment strategy caters to the insurer’s equity holders, professionals prefer an intermediate volatility level to balance bankruptcy costs against participation benefits.

Tariffs and Chinese Steel Futures (F1, G1)

Benjamin H. Liebman
,
St. Joseph's University
Yochanan Shachmurove
,
City University of New York-City College and Graduate Center

Abstract

This is the first paper to investigate the impact of trade protection on Chinese steel futures prices. The paper sheds light on the relatively new Chinese futures market for steel reinforcing bar, which has quickly become the most highly traded metal future in the world and yet has received almost no attention in the academic literature. Moreover, the paper contributes to the existing literature on the consequences of trade barriers by observing the behavior of both spot and futures markets in response to tariffs.

We construct a Vector Error Correction model of Chinese steel prices that includes weekly observations of rebar prices from 2013-2018 and incorporate all tariffs applied to Chinese rebar exports, including President Trump’s recent "section 232" tariffs on foreign steel. While both spot and future prices fall in response to tariffs, the magnitude of this drop is smaller for spot prices and is followed by an even larger increase in the subsequent week. In contrast, rebar futures display an initial drop that is larger than the fall in spot prices, and then experience a rebound over the next week that fails to match the magnitude of the initial decline. These results suggest that Chinese steel futures are more sensitive to foreign protection than spot prices. Moreover, the subsequent rebound in both the future and spot markets suggests that investors initially respond fearfully to pending protection but then reverse their behavior to a lesser degree (for the future) or greater degree (for the spot) when greater clarity emerges regarding the actual impact that these measures will have. In the final part of the paper, we gauge the impact of tariffs on the volatility of share prices of individual Chinese steel companies, taking advantage of a variety of volatility series provided by the NYU’s V-LAB.

Tax Literacy and Personal Investments for Post-Retirement Years (H3, J3)

Keiko Iwasaki
,
NLI Research Institute
Kunio Nakashima
,
NLI Research Institute
Tomoki Kitamura
,
Tohoku Gakuin University

Abstract

Facing the decreasing benefit of the public pension system in an aging society, governments offer various tax advantages to encourage individuals to have personal investments for their post-retirement years. However, the effects of these tax breaks on individual decisions depend on the level of tax literacy as well as behavioral characteristics such as time and risk preference.

While previous studies demonstrate that financial literacy plays an important role in increasing the probability of having investments for post-retirement years, tax literacy has not been included in measurements of financial literacy, and thus the role of tax literacy has not been well investigated either.

Using our unique data of Japanese residents aged 40–64 years (n=3,685), we measure tax literacy separately from financial literacy and investigate their role in various personal investment decisions. Employing parental educational attainment and experience of having different types of corporate welfare plans for retirement as instruments for endogenous variables of tax and financial literacy, we conduct instrumental variables estimations.

We find that tax literacy plays a significant role in increasing the probability of having investments for post-retirement years such as the individual-type Defined Contribution pension plan, Nippon Individual Saving Account, and personal pension insurance, while we find no significant effect of tax literacy on increasing the probability of having individual brokerage accounts without any tax benefits. Furthermore, on contrary to previous studies, we find no significant impact of financial literacy on increasing the probability of having investments for post-retirement years after controlling for its endogeneity, tax literacy and time and risk preference.

Our results suggest that it is crucial to invest in tax literacy education as well as getting tax benefits across people to achieve the intended goal of the tax advantages to encourage individuals to have investments for post-retirement years.

Teacher Quality, Family Inputs, and Student Outcomes: Evidence from Random Assignment of Students to Classrooms (I2, I0)

Chuanyi Guo
,
University of Illinois-Chicago

Abstract

Understanding the importance of schools and families in determining student outcomes has been a central concern among education researchers. However, there is very little understanding of how school and family inputs interact with each other and jointly affect student outcomes. In this study, I pay close attention to whether teacher quality affects family investment in children. For example, it is possible that parents devote less time to their children when they are assigned to experienced teachers. In addition, I am interested in how teacher quality interacts with family inputs to affect student achievement. Answering this question will further contribute to the understanding of the complementarity between schools and families. For example, analyzing whether teachers are differentially effective depending on how much families invest in children, might help me explain the mechanism through which the complementarity would determine student outcomes. This type of school-family interaction is neglected in the current literature which solely focuses on whether one single input is effective or not.

To study these questions, I utilize a nationally representative survey of middle school teachers, students, and their parents in China and focus on schools in which the assignment of students to classrooms is random. This allows me to mitigate potential selection problems—for example, high-achieving students or students with better-motivated parents are more likely to be assigned to high-quality teachers. By comparing outcomes for students taught by high-quality versus low-quality teachers, I document that parents in fact spend more time on their children when they are randomly assigned to high-quality teachers. The interaction between teacher quality and family inputs statistically significantly increases students' test scores and their happiness at school. This paper is the first to directly examine the complementarity between school and family inputs and its impact on student academic and noncognitive achievement in a causal sense.

The Achilles Tendon of Dynamic Pricing ---The Effect of Consumers’ Fairness Preferences on Platform Dynamic Pricing Strategies (L1, D4)

Yong Wang
,
Tsinghua University
Tianze Tang
,
Tsinghua University
Qiaoqin Xiong
,
Tsinghua University
Zhen Sun
,
Tsinghua University
Weiyi Zhang
,
Tsinghua University

Abstract

A longstanding challenge in the platform economy is how to reconcile the economic efficiency improvement of the dynamic pricing strategy and the incurring users’ perception of unfairness associated with the strategy. In this paper, we study the effect of consumers’ fairness preferences on dynamic pricing strategies adopted by platforms in a non-cooperative game. Our study reveals that, in a one-shot game, if consumers have fairness preferences, dynamic prices will slightly decline. In a repeated game, dynamic prices will be reduced even when consumers do not have fairness preferences. When fairness preferences and repeated game are considered simultaneously, dynamic prices are most likely to be set at fair prices. We also discuss the effect of platforms’ discounting factors, the consumers’ income and alternative choices of consumption on the dynamic prices. Our findings illustrate the importance of incorporating behavioral elements in understanding and designing the dynamic pricing strategies for platforms and the implications on social welfare in general.

The Common Currency Effect on International Trade: Evidence from an Accidental Monetary Union (F1, F3)

Roger Vicquery
,
London School of Economics

Abstract

This paper provides for the first time quasi-experimental evidence on the treatment effect of common currencies on international trade. Following a seminal study by Rose (2000), who found currency unions to have strikingly high trade effects, a vast literature has developed disputing the size and significance of the “Rose Effect”. While a literature meta-analysis still confirms a large pro-trade effect, results inherently suffer from endogeneity and self-selection bias and should not be interpreted as causal (Head and Mayer, 2014). This is an important point as currency union membership is currently a key policy choice for a number of countries around the world.

My paper fills a key gap in the literature by providing causal identification of the Rose Effect. I exploit as a natural experiment a rare event involving what is probably the closest historical precedent to the Euro Area, a French Franc zone that existed throughout the 19th century.

Around 1861, a major exogenous variation in the membership of this currency area occurred. This was driven by random diplomatic and military events as Piedmont-Sardinia, a long-standing French monetary client, unexpectedly annexed most of modern-day Italy. As part of the Italian unification, the annexed territories “accidentally” became part of a currency area with France and its satellites. Importantly, I show that key annexed Italian states had little endogenous reasons to join the French Franc area in a non-unification counter-factual.

I estimate the trade effect of this “accidental” currency union relying on new data and theory-consistent gravity equations. The estimated causal Rose Effect is large and significant, with a preferred estimate at 35%. Opposite to recent work looking at the endogeneity issue (Campbell, 2013), my findings broadly corroborate the original policy implications of the Rose literature. However, they also show a significantly smaller effect than the average one found in the literature.

The Competition of PMNCs During the Internationalization: Ecosystem, Informal Institution and Network Effect (L1, M2)

Ke Rong
,
Tsinghua University
Huiyi Litan
,
Tsinghua University
Di Zhou
,
Tsinghua University

Abstract

Platform companies that connect buyers with sellers of goods and service are increasingly venturing abroad. However, it is harder for these platform multinational companies (PMNCs) to grab market share from the local competitors in the foreign market than traditional MNCs. In this study, we employ a duopoly two-side platform model based on a platform ecosystem perspective. We suggest that PMNCs’ internationalization performance depends on the ecosystems of both foreign and local platforms, and informal institution distance (IID) between countries, instead of being solely determined by the platform. Also, the network effect is introduced into the model as a moderator. Based on our model, we hypothesize that (1) a foreign platform with a richer ecosystem or a higher penetration rate is more likely to capture larger market share in the local market, (2) if the local platform has a richer ecosystem or a higher internationalization rate, the foreign platform will capture smaller market share in the local market, (3) if the local market is large enough, the larger the IID is, the smaller is the market share of the foreign platform, (4) the network effect would enhance the influence of both foreign and local ecosystem. For this study, we acquire a longitudinal, cross-country dataset on the market share of search engine platforms. Our sample comprises more than 10 search engines of 50 countries from 2009 to 2017, which are tracked on a daily basis. This study makes several contributions. This study introduces PMNCs into the international business context and proposes an extended two-side platform model based on a platform ecosystem perspective. Also, we provide an effective strategy for PMNCs’ internationalization progress.

The Consistency of Knowledge Context, Human Capital and Absorptive Capability of Firm (O3, M1)

Shin-Ren Pan
,
Taiwan High Prosecutors Office-Taichung Branch

Abstract

This paper proposes that the efficiency of an absorptive capability of the firm can be explained by the context in which human capital accumulated knowledge on a specific trajectory related to external knowledge. I hypothesize that the consistency of knowledge context, conceptualized as the degree of similarity of the knowledge production among actors formed by the different process of knowledge accumulation, will moderate the cost of the R&D activity involving heterogeneous logics from distinct field knowledge and thus lead to higher innovation performance of firms. A panel data collected from the public firms who are qualified applying a specific program of academic-industrial collaboration in the Taiwanese science parks is used to test the hypothesis. The result, with the quasi-experiment design combining matching, difference-in-differences, and third-differences, shows that the firms owning more doctoral talents, who bridged the distinct paths of knowledge stock and presented higher consistency of knowledge context, raise innovative performance under an academic-industrial program. Some factors which might influence the estimation, such as firm size, quality of the cooperative academic institutes, the configuration of the human capital, and R&D inputs of the firms, were considered. The findings expand the understanding of knowledge spillover for innovation.

The Cost Effects of a Private Health Insurance Shared Savings Model (I1, L2)

Brigham Walker
,
Tulane University

Abstract

Shared savings models, where physician practices are paid a portion of the total savings achieved for their patients, have found increased popularity to create cost reduction incentives. These models have yielded modest effects in the public payer setting, but few studies have assessed the effects of private payer variations to the model. This paper assesses the cost effects of a large private payer-based model.


In this specific model, Accountable Care Organizations (ACOs) agree on a projected cost target with the payer. The practices continued to be paid under a fee-for-service arrangement, but if total costs are under the target, the savings are shared between the insurer and the ACOs. The launch was staggered and virtually all non-participating practices eventually participated the program. Using differences-in-differences and event study designs with several years of privately held healthcare claims data, I assess the effects of the program by comparing participating practices against non-participating practices while controlling for patient risk scores, various fixed effects, and including group-specific time trends.

Cost outcomes include overall, pharmacy, medical, inpatient, emergency room (ER), and preventative care costs. Efficiency measures include use of imaging, potentially preventable ER visits, and generic drug utilization rates. Utilization outcomes include inpatient, ER, and length of stay averages. Preliminary results suggest that the program changed utilization and cost patterns among the treatment practices.

The Cyclical Behavior of Factor Shares (E0, F6)

Lijun Zhu
,
Peking University
Michele Boldrin
,
Washington University-St. Louis
Yong Wang
,
Peking University

Abstract

We review the empirical evidence about factor shares and show that, apart from a varying trend, they are characterized by a strong and persistent cyclical pattern. Next, we provide a theory of why this may be due to the pattern of technological innovation under competition. Central to our theory are endogenous movements in relative factor prices creating incentives for replacing old technologies with new ones. The endogenous interaction between labor-saving innovations and changes in the relative price of labor is the source of both growth and cycles.

The Distributional Consequences of Macroeconomic Stabilization Policies (E5, E2)

Christian Friedrich
,
Bank of Canada
Alexandra Effenberger
,
German Federal Ministry for Economics and Energy
Jeromin Zettelmeyer
,
Peterson Institute for International Economics and CEPR

Abstract

Using harmonized household-level data from 26 European countries over the period 2005 to 2013, we examine the distributional impacts of unexpected changes in monetary and fiscal policy on households. In response to an unexpected tightening of fiscal policy, the growth of household disposable income decreases on average, while it increases in response to an unexpected tightening of monetary policy. This is the case even when we control for business cycle fluctuations, the financial cycle and the real effective exchange rate, which may themselves be responding to policy shocks. Repeating the regressions for the five quintiles of the disposable household income distribution indicates that the positive effects of tighter monetary policy on average household income growth operate through the richest 20 percent of households, while the impact on the bottom quintile is negative. Tighter fiscal policy, on the other hand, affects all households negatively. Exploring the reasons for the varying impact of monetary policy on different parts of the income distribution, we find that in response to an unexpected monetary tightening, households at the bottom of the distribution mainly suffer from adverse changes in their employment status, while households at the top benefit from higher capital incomes.

The Economic Impact of Internet Censorship (L8, O2)

Ayesha Ali
,
Lahore University of Management Sciences
Ihsan Ayyub Qazi
,
Lahore University of Management Sciences

Abstract

Internet censorship has become increasingly pervasive with nearly 70 countries restricting Internet access to their citizens. The blocking of YouTube in Pakistan, Google and Facebook in China and Iran, and Twitter in Turkey are some recent examples of censorship, which have affected large number of users. Internet censorship can have a substantial economic impact on various stakeholders in the Internet ecosystem including end-users, content providers, Internet service providers, and advertisers. We assemble a unique data set that combines information about major long term censorship events with country level indicators of economic and political freedoms to provide evidence on which web services are censored, trends in censorship over time, and the correlates of Internet censorship practices across countries. We employ an event study design to understand the impact of censorship of Internet services in China on the local service providers. Using data on stock price returns of censored firms and their local counterparts, we study the immediate and long term impact of censorship events on the growth of firms. We discuss our results in the light of existing models in International trade theory that predict intra-industry reallocation towards more productive firms in an environment of free trade. The results of our research can inform Internet censorship policy by providing a structured framework for understanding the economic implications of such measures.

The Effect of Anticorruption on the Structure of Public Revenue and Expenditure (H5, D7)

Guangjun Qu
,
Birmingham Southern College
Kevin Sylwester
,
Southern Illinois University-Carbondale
Feng Wang
,
Chongqing University

Abstract

This study investigates the effectiveness of anticorruption campaigns run by the Communist Party of China (the Party, hereafter) in recent years. Some believe that the Party’s efforts did reduce corruption, but others argued that these campaigns were initiated primarily to attack political rivals and, hence, had limited impacts on corruption. Unfortunately, relevant empirical evidence is scarce. This study aims at filling the void in the literature by examining the effect of the anticorruption campaigns on the structure of public revenue and expenditure of Chinese governments. The literature suggests that corruption distorts the ways a government collects its revenue and allocates its expenditure. If true, the anticorruption efforts made by the Party are supposed to mitigate the distortion as long as the campaigns are effective. Focusing on the Party’s official newspapers, we use the newspaper-based measures of anticorruption for three levels of Chinese government, from national to provincial to municipal, and investigate how anticorruption is associated with various components of public revenue and expenditure. We find that the lower the level of the government, the stronger the influence of anticorruption on the ways it collects revenue and allocates spending. The evidence also indicates that the anticorruption efforts reduce land-leasing revenue and infrastructure spending, but they are not associated with the spending on education and public services. The staggered implementation of the anticorruption campaigns that can be taken as a quasi-natural experiment helps address the issues of endogeneity and reverse causation.

The Effect of Electronic Monitoring on Offenders and Their Families (K4)

Hans Gronqvist
,
Uppsala University
Julien Grenet
,
Paris School of Economics
Susan Niknami
,
Stockholm University

Abstract

Electronic monitoring (EM) is one of the most popular tools to reduce high rates of imprisonment. Theory suggests that EM may increase the risk of relapsing into crime by making low punishment salient (e.g. Chalfin and McCrary 2015) or reduce recidivism by preventing accumulation of criminal capital in prison, lower depreciation of human capital, or reduce stigma. EM may also have social and economic implications beyond that of recidivism by potentially influencing future employment prospects or the well-being of family members. Yet, the few existing design based have only estimated the effects of EM on reoffending (e.g. Di Tella and Schargrodsky 2013). Evidence on the magnitude and sign of the effect of EM on key outcomes such as own labor supply, marital stability and child development is lacking.

We study a major reform in Sweden where EM was scaled up from a small pilot scheme to the entire country. Individuals sentenced to at most three months in prison (50 percent of the prison sentences) were made eligible to apply for EM, thereby potentially avoiding entering prison (i.e. “front-door” EM). The rules of eligibility were general so almost everyone applying was also approved. While the number of incarcerated individuals sentenced to up to three months fell by close to 50 percent post-reform, incarceration among those sentenced to between 3 and 6 months was unchanged. We leverage population based register data and compare the outcomes of eligible offenders to ineligible offenders using DD and RD designs.

The reform lowered recidivism by about 15 percent and increased the likelihood of future employment by 10 percent. The effects are stronger for younger offenders and those with a criminal history. For the children of the offenders, the reform increased compulsory school GPA (about 5 percentile ranks); although there are no significant effects on crime.

The Exports of Higher Education Services from OECD Countries to Asian Countries. A Gravity Approach (F1, I2)

John Beghin
,
North Carolina State University
Byung Yul Parks
,
North Carolina State University

Abstract

We analyze bilateral exports of higher education services between OECD countries and Asia, using a gravity equation approach, panel data from 1998 to 2016, and PPML regression. The approach treats higher education consumption by Asian countries as a consumable durable good. Asian Students come to OECD countries to obtain degrees from their universities. Structurally, the flow of students from Asian country j to OECD country i depends on bilateral transaction costs between i and j, the income per capita in j, school-age demographics in j, the higher-education capacity of i, the perceived quality of universities in i, and the usual multilateral trade resistance terms We find that bilateral flows of students are strongly influenced by bilateral distance, importers’ income, real exchange rate, demographics, common language, common border, commonality in religions, the visa regime prevailing in bilateral country pairs, the size of the higher education sector in OECD countries and to a lesser extent, the perceived university reputation. The evolution over time of the size of the higher education sector, visa regimes, strong income growth and changes in demographics in nearby export markets and to a lesser extent, the change in university reputation, explain the emergence of Australia, Canada, Korea, and New Zealand and the loss of market share by the US which still strongly dominates international trade in higher education services.

The Government Spending Multiplier at the Zero Lower Bound: International Evidence from Historical Data (E3)

Mathias Klein
,
Sveriges Riksbank
Roland Winkler
,
University of Antwerp

Abstract

Based on a large historical panel dataset, this paper provides robust evidence that
the government spending multiplier is significantly higher when interest rates are
at, or near, the zero lower bound. We estimate fiscal multipliers that are around
1.5 during zero lower bound episodes and significantly below unity outside of it.
We show that the difference in multipliers is not driven by multipliers being higher
during periods of economic slack.

The Highs and Lows of Medical Marijuana Legalization (K4, I1)

Sanjukta Basu
,
Tulane University
Siobhan S. Innes-Gawn
,
Tulane University
Mary H. Penn
,
Tulane University

Abstract

Starting with California in 1996, thirty-three states and the District of Columbia have passed medical marijuana laws (MMLs). Although MMLs are aimed at providing patients with the therapeutic benefits of marijuana, they may also increase access to and use of marijuana for recreational purposes, and this may affect other risky behaviors, such as substance consumption or criminal activity. For example, if marijuana is a gateway drug, MMLs could also increase the use of hard drugs, such as cocaine or heroin. Alternatively, if MMLs increase access to marijuana through friends and family members, marijuana users may be less likely to purchase marijuana from drug dealers providing pathways to hard drugs and the initiation of hard drug use by marijuana users may decrease. We investigate whether MMLs affect legal and illegal substance use and criminal behavior by exploiting the variation among states in the presence, timing of adoption, and characteristics of these laws.
We use data from the National Longitudinal Survey of Youth 1997 Cohort, which is a panel data set that follows the lives of a sample of American youth over time (from 1997 to 2015). This data set allows us to apply difference-in-difference and event study models to identify causal effects on individual-level, self-reported criminal activity and use of marijuana, alcohol, tobacco, and hard drugs. State MMLs have considerable variation in their characteristics, which may have important consequences on substance use and crime. Due to the nature of the data and state laws, we are able to examine the effects of MMLs on illegal behavior that may go undetected by law enforcement and investigate how possession limits, cultivation limits, and the availability of dispensaries affect these risky behaviors. Preliminary results indicate that MMLs may increase the probability and frequency of smoking, suggesting there may be some unintended negative effects.

The Impact of Derivative Disclosures on Managerial Opportunism: Evidence from SFAS 161 (G1, G3)

Guanming He
,
Durham University
Mengbing Ren
,
University of Warwick

Abstract

Derivatives are increasingly used by managers not only to hedge risks but also to pursue non-hedging activities for fulfilling opportunistic incentives. Using the mandatory adoption of a financial reporting standard (SFAS 161) on derivative disclosures, we examine whether and how derivative disclosures influence managerial opportunistic behavior. We employ insider trades and stock price crash risk to capture managerial opportunism. Applying a difference-in-differences research design with hand-collected data on derivative designations, we find that, after the implementation of SFAS 161, derivative users that comply with SFAS 161 experience a significantly greater decrease in both insider trades and stock price crash risk, compared with a matched control sample of non-derivative-users. Our cross-sectional analyses reveal that SFAS 161 has greater impact on firms in the case of high information opacity, high financial risk, or high business risk. We find no evidence that, compared to the non-derivative-users, derivative users not compliant with SFAS 161 have a greater reduction in either insider trades or stock price crash risk in the post-SFAS 161 period, implying the importance of enhancing the enforcement of the regulation. Overall, our results suggest that the enhanced disclosures on firms’ objectives and strategies of using derivatives curb managerial opportunism.

The Impact of Fake News: Evidence from the Anti-Vaccination Movement (D8, I1)

Meradee Tangvatcharapong
,
Texas A&M University

Abstract

The increasing amount of fake news has generated significant debate about the proper role of government and social media platforms in combating it. However, little is known about whether fake news can actually change behavior. This paper addresses this question by examining how vaccination rates responded to the unexpected surge in media coverage in 2007 of the verifiably false claim that the MMR vaccine caused autism. Specifically, I use a difference-in-difference approach to compare the MMR vaccination rates of children whose parents were most and least likely to be affected by the news over time. I determine parents’ susceptibility using three predetermined characteristics: whether their child is a firstborn, the child’s gender, and the parents’ age. Results show that susceptible parents were 3.3 percentage points less likely to vaccinate their children with an MMR shot by the recommended age of 15 months and 4.1 percentage points less likely to do so by 29 months. This indicates that at a minimum, fake news caused parents to delay vaccinating their children by over a year, and at most prevented them from ever immunizing their children.

The Impact of Indoor Climate on Human Cognition: Evidence from Chess Tournaments (Q5, J2)

Juan Palacios
,
Maastricht University and IZA
Steffen Kuenn
,
Maastricht University
Nico Pestel
,
Institute of Labor Economics (IZA)

Abstract

While there is extensive evidence on the harming effects of temperature and pollution on human health, the current knowledge about how these stressors affect human performance is still rather limited.We link measures of indoor environmental conditions to the performance of chess players at official tournaments where players face incentives to exert high effort. We construct an objective outcome measure for cognitive performance by comparing the quality of a player’s actual moves with “optimal” moves as predicted by a computer. The results indicate that air pollution (PM2.5) is the main driver of the probability of making mistakes and the magnitude of such errors. We find that a increase of 10 μg leads to 1% higher probability of making a mistake, and 6.67 % larger errors (relative to the average error). The impact of pollution is exacerbated by time pressure. When players approach the time control of games, an increase of 10 μg (similar to one standard deviation in our sample) leads to a increase of 2% in the probability of making a mistake, and those errors are 14.71 % larger.Given the type of the cognitive task chess players have to perform (and which we actually measure with our outcome variable), it is very likely that our results have strong implications for the productivity of high-skilled office workers, in particular for those executing non-routine cognitive tasks requiring problem-solving skills.

The Impact of Temperature Shocks on Credit Market (G2, G3)

Mandeep Singh
,
University of New South Wales
Emdad Islam
,
University of New South Wales

Abstract

A study by NASA (2005) suggests that changes in climate affect not only average temperatures, but also extreme temperatures, and that such changes increase the likelihood and intensity of weather-related disasters. Given changes in surface temperature are asymmetric across regions, variations in the number and severity of natural disasters are evident across geographical regions. We conjecture that such cross regional differences in the likelihood of weather related disasters would induce heterogeneity in the very way firms are organized, operate, and survive. Specifically, we ask whether banks, as important financial intermediaries, do consider risks associated with climate change while making lending decisions. Relatedly, we also examine how firm level strategic decisions such as investments and financing are affected due to such exposure to extreme weather related disasters. To answer these important questions, we assemble a novel dataset combining data from NASA's Goddard Institute of Space Studies, SHELDUS, and LPC Dealscan dataset. Exploiting the granularity of such rich dataset, we utilize identification strategy of Khwaja and Mian (2008), in the spirit of Acharya et al.(2018), who exploit multiple bank-firm relationships to control for loan demand, and other observed and unobserved borrower characteristics. We show that lending volume, loan spread and other important terms of a loan contract are adversely impacted by the likelihood of extreme weather related disasters. We also show the direct and trickle down effect of constrained financing on firm level outcome and firm survival likelihood. In addition to the other direct costs of climate change like food and water security, migration and political security, our study reveals that the severity of temperature change may translate into lower credit availability and higher cost of credit with stringent collateral requirements. Thus, we contribute to the nascent literature that links climate change and finance with broader implications for macro-economy.

The Long-term Causal Effect of U.S. Bombing Missions on Economic Development: Evidence from Ho Chi Minh Trail and Xieng Khouang Province in Lao P.D.R. (O1, P5)

Takahiro Yamada
,
Japan Ministry of Finance
Hiroyuki Yamada
,
Keio University

Abstract

This paper investigates the long-term causal effect of massive U.S. bombing missions on later economic development in the case of Laos during the Vietnam War, 1955-1975. The empirical strategy relies on the instrumental variable approach based on the heavily bombed targets Ho Chi Minh Trail and Xieng Khouang province as a proxy of U.S. bombing missions by measuring the distance between them and the centroid of the village-level administrative boundary. We use the three rounds of nightlight data (1992, 2005 and 2013), and two rounds of population data (1990 and 2005) to estimate the effect of bombs on later economic development. The estimation results show the no robust effect of U.S. bombing missions on economic development in the long term. In addition, the results do not necessarily support the convergence hypothesis in the neoclassical growth theory arguing the regional economic convergence within a country.

The Mexican Drug War: Elections and Homicides (K4, D7)

Aixa Garcia-Ramos
,
University of Passau

Abstract

Mexico has experienced a dramatic increase in homicides during the last decade. This increase has been associated with the turf wars among Drug Trafficking Organisations (DTOs) for the control of strategic territories. This paper examines whether these territorial disputes are higher during the lame duck period, when incumbent DTOs might be relatively weaker. Using homicides as a proxy for turf wars, my results show support for this hypothesis. The increase in homicides is concentrated on municipalities in which the PRI wins the election. In contrast, those in which the PAN wins experience a decrease.

The Price of Capital Goods: A Driver of Investment Under Threat (E2, F6)

Natalija Novta
,
International Monetary Fund
Weicheng Lian
,
International Monetary Fund
Evgenia Pugacheva
,
International Monetary Fund
Yannick Timmer
,
International Monetary Fund
Petia Topalova
,
International Monetary Fund

Abstract

Over the past three decades, the price of machinery and equipment fell dramatically relative to other prices in advanced and emerging market and developing economies. Using cross-country and sectoral data, we show that the decline in the relative price of tangible tradable capital goods provided a significant impetus to the capital deepening that took place during the same time period. The broad-based decline in the relative price of machinery and equipment, in turn, was driven by the faster productivity growth in the capital goods producing sector relative to the rest of the economy, and its deeper trade integration, which induced domestic producer to lower prices and increase their efficiency. Our findings suggest an additional channel through which rising trade tensions and sluggish productivity could threaten real investment growth going forward.

The Pricing Implications of the Oligopolistic Securities Lending Market: A Beneficial Owner Perspective (G1, G2)

Zorka Simon
,
Goethe University Frankfurt
Zsuzsa R. Huszar
,
S. P. Jain School of Global Management

Abstract

In this study, we examine the welfare implications of market inefficiencies in the European securities lending market for contractual savings institutions (CSIs) from July 2006 to June 2015. Following the global financial crisis, increased regulatory efforts, quantitative easing, and flights-to-safety gave rise to a shortage of high-quality liquid assets (HQLAs) and depressed yields on safe European treasuries. The resulting low interest rates affect CSIs’ balance sheet. On the liability side, liability values increased that led to a decline in solvency. On the asset side, as a large part of CSI portfolios is invested in long-term treasuries due to regulatory mandate, falling asset prices and low HQLA yields forced these institutions to seek alternative revenue opportunities, where fully indemnified securities lending transactions could offer a solution.
In this setting, we study the functioning of the OTC securities lending market for the prime European benchmark securities, German Treasuries. We find that demand pressure is priced in and captured in lending fees with a significant delay despite the exceptionally high utilization rates. This suggests that brokers and agents underrepresent less connected lenders’ interests, such as that of smaller pension funds and insurers with limited bargaining power. We also document high relative price spreads across lending contracts on the same security at the same time, especially when fees are already high, indicating that some lenders are unable to extract the “real” rents due to prime brokers’ discriminatory pricing behavior and information advantage. These inefficiencies are most evident in the long maturity segment, where most lenders are contractual savings institutions, whose inability to capitalize on lending income has non-negligible negative welfare consequences for the average European citizen. As such, we urge regulators to consider introducing measures to protect the interest of CSI market participants, allowing them to access the income channel from securities lending.

The Puzzling Politics of R&D: Signaling Competence Through Risky Projects (O3, H5)

Natalia Lamberova
,
University of California-Los Angeles

Abstract

Why do some incumbents devote significant funds to the support of Research and Development
(R&D), even though the fruits of such investment typically appear after the incumbent
has left the office? Unlike many other types of long-term government investment (such
as education or infrastructure), government investment in R&D is less visible to the average
voter and highly risky. This paper investigates whether the mere fact of incumbent pursuing
pro-R&D policy can result in higher expectations of future economic growth rewarded
by the voters, or greater perception of incumbent’s competence. It further asks whether
pursuing R&D policy helps the incumbent to secure reelection. The paper proposes a theoretical
model of signaling, provides evidence from survey experiments conducted in USA
and Russia, and corroborates findings with cross-country evidence. The paper concludes
that investment in R&D can generate higher perceptions of incumbent’s competence.

The Real Effects of Fear and Fundamental Uncertainty Shocks (E3, D8)

Gene Paul Gerard Ambrocio
,
Bank of Finland

Abstract

This study exploits biases in survey-based estimates of forecast uncertainty to provide evidence on the macroeconomic effects of macro-uncertainty as decomposed into fundamental and overconfidence bias components. Crucially, estimated effects using overconfidence biases mitigate endogeneiety concerns that beset those from fundamental uncertainty. Using a sign and zero restrictions identification scheme in a vector autoregression, I find that increases in fundamental uncertainty and declines in overconfidence may lower real activity.

The Real Estate Consequences of Immigration Shocks: Evidence from the United States’ Mexican Repatriation (R3, J6)

Vinicios Sant'Anna
,
University of Illinois-Urbana-Champaign
Gustavo Cortes
,
University of Florida

Abstract

How do extreme migration shocks affect local economies? We analyze the United States' Mexican Repatriation of 1930--36 and its impact on housing and construction activity in US cities. Using Census full-count and hand-collected building permits data, we show that repatriating Mexicans during the Great Depression slowed down city growth. Employing an instrumental variable approach, we find that cities with greater Mexican outflow experienced lower growth in the number and values of building permits, as well as median house values. These results suggest that repatriation policies negatively impact real estate conditions and local economic growth.

The Social Costs of Financial Crisis: Worker Level Evidence (J2, E2)

Hans Gronqvist
,
Uppsala University
Julien Grenet
,
Paris School of Economics
Daniel Jahnson
,
Uppsala University

Abstract

While a growing literature investigates the effects of financial crisis on firm-level employment we know almost nothing about the long-run effects on the workers and their families. The main reason is limited access to data. Although some firm-level datasets allows researchers to measure exposure to crisis using debts (e.g., Chodorow-Reich 2014, Giroud and Mueller 2017, Huber 2018) they do not contain individual workers.

This paper examines the long-run effects on the workers who were exposed to the sudden Swedish banking crisis in 1990/91, one of the most severe financial crises in history (Reinhart and Rogoff 2008). Our data, starting in 1985, contain financial statements for a large sample of firms linked to workers. We calculate each firm’s position in the pre-crisis debt ratio distribution and show how the year-by-debt-quartile coefficients evolve before, during, and after the crisis, holding constant firm effects and local labor market shocks. Our DD estimates show no significant changes in the outcomes prior to the crisis. There is no effect on total firm sales, conditional on LM-shocks, which make consumer demand a less likely transmission mechanism.

Firms more dependent on external capital before the crisis experience 20% larger employment losses than less dependent firms. Employment has recuperated after about 7 years. The employment contraction is explained both increased worker separations and decreased hirings. Consistent with Mueller (2017), employment falls relatively more for high skilled workers. While we find no significant effects on investments, firms under financial pressure temporarily experienced lower productivity.

At the worker level, individuals more exposed to the crisis experience suffer in the long-run. A decade after the crisis, these workers are still less likely to work and have 10 percent lower wages. The effects are concentrated among less-skilled males. Post-crisis these workers work in less productive firms.

The Spillover Effects of a Medicare Payment Reform (I1, L2)

Brigham Walker
,
Tulane University

Abstract

By altering the incentives faced by healthcare providers, Medicare can indirectly affect care for the privately insured. Healthcare research on these spillover effects has focused on Medicare Advantage and traditional fee-for-service (FFS) Medicare or used payer mix variation to identify the effects of public insurance schemes, but scarce research assesses the indirect effects of Medicare on the privately insured population under age 65. This paper does just that using the Oncology Care Model (OCM) payment shock – which included a monthly payment to encourage care coordination and a shared-savings payment to encourage lower patient spending – to directly measure spillover effects.

Anecdotal evidence suggests that the OCM program prompted the hiring of non-physician clinicians and led to an increased emphasis on evidence-based treatment. I utilize a unique dataset of medical claims from a large provider network where, despite having unified care pathways, input prices, and information technology, only half of the practices participated in the program. I use a differences-in-differences methodology to measure the effects of the OCM program on utilization and cost among patients with private insurance. I couple this analysis with an event study to assess the onset effects of the policy. Primary outcomes include changes in expenditures, such as drug-spend and total-spend, and secondary outcomes include changes in utilization of specific services such as regular visits to mitigate unnecessary hospitalizations or emergency room visits.

Preliminary results suggest that the Medicare change affected the utilization and cost outcomes for the privately insured population under age 65. Shared-savings models have elsewhere shown modest first year effects among those directly managed under those payment systems; this analysis will clarify whether, where, and to what magnitude these effects spillover to privately-insured patients uninvolved with the program.

The Supply Side of Mortgage Lending (G2, M3)

Vardges Levonyan
,
University of Zurich

Abstract

We find strong supplier effects in mortgage currency choice using credit register data of all loans issued in Armenia 2005-2017. Controlling for the relative price of dollar- and local-currency denominated mortgages, households are more likely to choose dollar-denominated mortgages from banks with higher dollar deposit inflows, and from banks on the margin to their dollar capital reserve requirements. The supply-side effects are reduced with competition, are larger for less sophisticated customers, and for customers with relationship banking. The effect increases for foreign banks and is reduced after periods of exchange rate shocks. We provide a model of bank lending, where prices and bank steering predict these results. We explore our interpretation with alternative non-price channels.

The Value of Firm Reputation in the Labor Market: Evidence from Corporate Scandals (M5, J0)

Salil Gadgil
,
University of California-Los Angeles
Jason Sockin
,
University of Pennsylvania

Abstract

Treating corporate scandals as unanticipated shocks to firm reputation, we study how reputational losses affect a firm's relationships with job seekers and current employees. Using proprietary data from the website Glassdoor, we find that conditional on expressing interest in similar firms, potential applicants are 16% less likely to click on job postings for a firm in the wake of a scandal. The magnitude of this effect increases to 29% for job seekers with an advanced degree, which may reflect their stronger outside options. Online reviews reveal that firms are also perceived more negatively by employees for an extended period after a scandal. In the three months following a reputation shock, overall ratings drop by 4.7%, with ratings for work-life balance and culture dropping by 4.3% and 6.9%, respectively. Workers are also 8 percentage points less likely to recommend their employer to someone in their social network after a scandal, suggesting that referral networks are damaged when a firm loses reputation. Taken together, our results indicate that a firm's ability to hire and retain workers is compromised after a reputational shock. Consistent with the theory on labor search and matching, we find that firms respond to this diminished capacity by paying 3% higher wages relative to their peers who have not faced a scandal. This effect is several times larger for workers with advanced degrees, perhaps due to their bargaining power. While a sizeable literature explores the impact of signals sent by job seekers to employers, comparatively little has been done to understand the importance of signals sent in the other direction. Our findings provide strong evidence that despite its intangibility, reputation is a valuable asset for firms in the labor market.

Too Many Dealerships? The Impact of State Franchise Laws in Automobile Dealerships Distribution (L5, K2)

Kaida Zhang
,
Pennsylvania State University

Abstract

This paper quantifies the effect of automobile franchise laws on US automobile dealership networks. The franchise laws prohibit automobile manufacturers to terminate a franchise on its own discretion. Due to historical reasons, US brand manufacturers have more dealerships than their foreign brand competitors. It has been long debated on whether this large number of dealerships have harmed their profitability. Based on the moment inequality approach, this paper estimates a static entry model with flexible network effects and unobservable firm and market heterogeneities. Automobile manufacturers simultaneously decide the number of dealerships to put in each market. In the equilibrium, US manufacturers are assumed to have no profitable upward deviations but may have profitable downward deviations, while foreign manufacturers have profitable deviations in neither side. The results show that the cannibalization effect between dealerships overwhelms scale of economy. In the counterfactual scenario where US manufacturers can freely terminate franchises, this paper finds that they would like to close about 28% of existing dealerships.

Total Labor Requirements, Value Relative Prices and Market Prices: Determination of Relative Prices by Input Output Analysis of Japan for 1951,1960,1970,1980,1990,2000 and 2011 (L1, D4)

Akiko Nakajima
,
Fukuoka University

Abstract

This paper constructs natural prices as prices that will prevail when all labor participating in production are given equal value added generation. Natural prices are obtained by total labor requirements of each sector multiplied by average income per participating labor. In the calculations of total labor requirements, capital costs consider depreciation cost proportioned to sectors supplying capital goods, and labor embodied in imports are replaced by the amount of labor embodied in the same monetary value of exports.
Market prices are theoretically equal to 1. They are calculated as sum of labor content from each sector multiplied by actual value added per labor of each sector, and this becomes 1 for all sectors for all periods empirically.
Empirical findings (Number of sectors are 32) show that the gap between natural prices and market prices decrease over time from 1951 to 2011 according to RIETI (Research Institute affiliated to Ministry of Economics and International Trade) data. Weighted (weighted by each sector’s total output) Standard Deviation of the Difference between Natural prices and Market prices (WSDDNM) decrease from 0.48(1951), 0.417(1960), 0.406(1970), 0.341(1980), 0.301(1990), 0.2833(2000) to 0.2627(2011).
Gap between natural prices and market prices widens when the economy is in boom(1990) , however there are gradual decrease over time according to Supplementary-Employment-Table of Input-Output-Tables, published by Ministry of Internal Affairs and Communications. WSDDNM are, 0.4807(1951), 0.4371(1960), 0.4125(1970), 0.3442(1980), 0.4607(1990), 0.3575(2000) and 0.3403(2011).
Both international and domestic competition among sectors decrease the gap. Wage negotiation process of the 1970s must have contributed to decrease the gap within manufacturing industries.
Study concludes that the gap between market prices and natural prices has very slow tendency to decrease, while profitable industry at profitable period increase the gap.

TV in Times of Political Crisis : New Evidence from the 2017 Election in Kenya (D7, D8)

Elisa Mougin
,
Sciences Po

Abstract

This paper provides new evidence on the influence of exposure to TV broadcasts on
voters in times of political turmoil. I investigate the impact of the staggered introduction
of digital television between 2013 and 2017 in Kenya on electoral mobilization using the
nullification by the Supreme Court of the results of the presidential election of 2017,
followed by the boycott of the second round by the opposition. I construct a novel
geo-coded dataset that includes granular data on terrain features and electoral results.
The main predictors of signal reception are identified using a logistic model; a series of
exogeneity checks confirms the empirical strategy. On average, the influence of TV is
negative: turnout is lowered by 2.4 percentage points in polling stations covered by the
signal. Yet, there is a large heterogeneity across groups: the depressing effect is more
pronounced in places where the race was close in the first stage, and when voters have
already been exposed to violence and protests. The preference for the "exit" strategy is
offset in pro-incumbent political bastions, where broadcasts turn into a mobilization
amplifier.

Uncertainty in Procurement Contracting with Time Incentives (L9, C5)

Daiqiang Zhang
,
State University of New York-Albany

Abstract

This paper studies A+B (cost-plus-time) procurement contracting with time
incentives in the highway construction industry. In the presence of construction
uncertainty, the contractor's actual completion time may deviate from the bidding
completion time, and the A+B contract design is not ex post ecient. Using data
from highway procurement contracts in California, we show in a counterfactual
analysis that the ex post ecient lane rental contract would reduce the social cost by
$44.73 million (43.66%) on average. In particular, the average commuter cost would
reduce by $60.27 million (85.47%), suggesting a substantial reduction of construction
externality to commuters from lane rental contracts.

Understanding the Effect of Digital Literacy on the Spread of Misinformation – Evidence from Pakistan (D8, L8)

Ayesha Ali
,
Lahore University of Management Sciences
Ihsan Ayyub Qazi
,
Lahore University of Management Sciences

Abstract

The increasing availability of low-cost mobile phones and mobile Internet access in emerging markets has led to widespread use of social media platforms, making them an important source of news and place for social and political activity. This trend has brought many new users online including those with limited exposure to technology. Concurrently, we have also observed an increasing trend in the spread of misinformation on such platforms. The spread of misinformation can create and exacerbate polarization, while also affecting the beliefs and actions of users without a prior ideological leaning. We carry out a household level survey which provides unique first-hand evidence on the trends in social media use, sharing of news, and spread of misinformation among low and middle-income users in Pakistan. Using a list of actual news stories circulated on social media we create a test to measure the extent to which users are likely to believe and share misinformation. We evaluate the effectiveness of two interventions for countering misinformation in a randomized control setting. Our first intervention educates users about common features of misinformation through a video, while our second intervention in addition to the video provides specific feedback to users about their own past behaviour in engaging with misinformation. We measure the effectiveness of our interventions on future behaviour of participants in believing and sharing misinformation.

United States and Euro Area External Adjustment: The Role of Commodity Prices and Emerging Market Shocks (F4, F3)

Massimo Giovannini
,
European Commission-Joint Research Centre
Stefan Hohberger
,
European Commission-Joint Research Centre
Robert Kollmann
,
European Centre for Advanced Research in Economics and Statistics (ECARES), Free University of Brussels & CEPR
Marco Ratto
,
European Commission-Joint Research Centre
Werner Roeger
,
European Commission, DG-ECFIN (Directorate‑General for Economic and Financial Affairs)

Abstract

The trade balances of the US and of the Euro Area (EA) have improved markedly after the
Global Financial Crisis. This paper quantifies the drivers of EA and US economic fluctuations and external adjustment, using an estimated (1999-2017) three-region (US, EA, rest of world) DSGE model with trade in manufactured goods and in commodities. In the model, commodity prices reflect global demand and supply conditions. The paper highlights the key contribution of the post-crisis collapse in commodity prices for the EA and US trade balance reversal. Aggregate demand shocks originating in Emerging Markets too had a significant impact on EA and US trade balances. The broader lesson of this paper is that Emerging Markets and commodity shocks are major drivers of advanced countries’ trade balances and terms of trade.

United States Monetary Policy Spillovers to Emerging Markets: The Effect of Stress Tests (G2, F3)

Friederike Niepmann
,
Federal Reserve Board
Tim Schmidt-Eisenlohr
,
Federal Reserve Board
Emily Liu
,
Federal Reserve Board

Abstract

This paper shows that monetary policy and prudential policies interact. U.S. banks
issue fewer commercial and industrial loans to emerging market borrowers when the
Federal Reserve tightens monetary policy. The effect is less pronounced for banks that
are more constrained through the U.S. bank stress tests, reflected in a lower minimum
capital ratio in the severely adverse scenario. This suggests that monetary policy
spillovers depend on banks’ capital positions. In particular during a period of quantitative
easing when liquidity is abundant, banks are more flexible, and the scope for
adjusting lending is larger when they have a bigger capital buffer. We conjecture that
bank lending to emerging markets during the zero-lower bound period would have been
even higher had the United States not introduced stress tests for their banks.

Universality, Equality and Equity in The Italian National Health Service: Highlighting Discrimination Phenomena in Accessing Health Services Using Institutional and Administrative Data (I1, K3)

Silvia Gatti
,
University of Bologna

Abstract

The Italian NHS must guarantee to all citizens, in conditions of equality, universal access to the equitable provision of health services.
In the current institutional framework of healthcare federalism, the central government has the responsibility to ensure these rights through a strong system of guarantees summarized in the Essential Levels of Care (LEA), and at the same time the Regional Authorities have direct responsibility through their Local Health Authorities (AUSLs) for the implementation of government and expenditure for achieving the country’s health objectives.
The monitoring of the LEA by the Ministry of Health (data available since 2001) it would be the tool for verification of the maintenance of the NHS principles, avoiding discrimination in accessing health services. It is an ex post instrument with indicators that do not specifically address each policy of reorganization of services or spending containment. But it is the only complete tool available to everyone and it is possible to read the regional indicators, highlighting the phenomena of non-access to NHS services, the groups that are affected, and the changes over time.
More in depth concerning the progressive changes in the offer of services for the early detection of breast cancer in the area of Bologna AUSL (poster at AEA Meeting 2017), the data (2002-2016) from the Regional Health Service of Emilia-Romagna on the access to the mammography services in the AUSLs of the region allow to present the results of a longitudinal analysis on the different paths for the early detection of breast cancer undertaken by the women in Bologna and in the other AUSLs in Emilia-Romagna after the solutions adopted after 2010 to deal with the problems of waiting lists and the control of spending for the services of early detection of breast cancer, redirecting the services toward the screening of public health.

Upward Income Mobility and Legislator Support for Education Policies (I2, D7)

Luna Bellani
,
University of Konstanz

Abstract

This paper investigates how upward mobility affects legislator voting behavior towards education policies. We develop an electoral competition model where voters are altruistic parents and politicians are office seeking. In this setting the future economic status of the children is affected both by current public education spending and by the level of upward mobility. This paper is among the first to provide evidence for the association between upward mobility and policy formation. Using a newly compiled dataset of roll call voting on California education legislation matched with electoral district-level upward mobility we find that a standard deviation decrease in upward mobility in a legislator’s district is associated with a decrease in his likelihood of voting “no” on redistributive education bills of almost 6 percentage points.

Using Machine Learning and Big Data to Identify the Food Security Problem in Afghanistan (I3, R2)

Chen Gao
,
Texas A&M University
Chengcheng J. Fei
,
Texas A&M University
Reid Stevens
,
Texas A&M University
David J. Leatham
,
Texas A&M University

Abstract

After suffering from decades of civil war and severe environmental issues in recent years, poverty and food security issues persist for households in Afghanistan. Policymakers in Afghanistan are trying to eliminate poverty and food insecurity problem using global food aid and other subsidies. These policies require identifying food insecure households precisely and accurately.
We use big data from Afghanistan’s National Risk and Vulnerability Assessment (NRVA) survey to measure food security in Afghanistan using household food consumption data. We then use supervised machine learning models to find variables in the NRVA survey that predict food insecure households. Research on food security tends to use causal models to capture contemporaneous relationships or to predict future food security. We believe food security is a persistent problem and there are bidirectional effects between the food security and the socioeconomic factors collected in the NRVA. Thus, we do not focus on causality and focus instead on the association between NRVA variables and food security.
We find that many variables are useful to identify food-insecure households, many of which are not typically included in food security models. Surprisingly, we find that some factors that are traditionally used to model food security, like disaster shocks and availability of clean water, are not useful predictors of food security. Other variables, like the material of the exterior wall of the dwelling, the sources from which the households borrow money, and expenditures on charity are important variables in classification models. Of the 28% of total households that are food insecure, we are correctly classify 87% of that group using about 10 variable out of the 581 variables collected by the NRVA. In additional to offering accurate identification of the food insecure households, our model also suggests that the relationship between food security and socioeconomic factors is bidirectional.

Violence and Growth in the Mexican Drug War (F1, O1)

Jesus Andres Gorrin
,
University of Warwick
Jose Morales
,
Harvard University
Bernardo Ricca
,
London School of Economics

Abstract

This paper studies the trade effects of increases in violence following the Mexican Drug War. A focus on exports allows us to control for demand shocks. We compare exports of the same product to the same country of destination, but produced in municipalities with different exposure to violence after a close electoral outcome. Building on the close elections identification strategy proposed by Dell (2015), we show that municipalities that are exogenously exposed to the Drug War experience a 40% decrease in export growth on the intensive margin. We show similar results using micro-data on single plant exporters located in differently treated municipalities, but exporting the same product to the same destination. We show also evidence of the mechanism through which violence affects trade. Large exporters suffer larger effects, along with exports of more complex, capital intensive, and skill intensive products. Finally, we provide evidence consistent with violence increasing marginal exporting costs. The reduction in exports only occurs in the intensive margin; however, we find no changes in exit.

Water Innovation and Water Governance: Adaptive Responses to Regulatory Change and Extreme Weather Events (Q2, O3)

Horatiu A. Rus
,
University of Waterloo
Hongxiu Li
,
University of Waterloo

Abstract

During the past several decades, many countries have achieved significant progress on environmental issues such as water pollution through the increased stringency of their water policies. While a few studies have examined the stringency of overall environmental policy and its impacts on innovation, the direct link between water policies and water-related technological innovation has not yet been investigated. This paper focuses on three aspects of water policy (drinking water quality, water pollution, and water quantity) in the United States and examines the effects of federal and state level regulatory changes on the level of relevant technological innovation. We construct a unique panel dataset covering major amendments and regulated contaminants lists as stipulated in the legislative acts most relevant to each water policy area, along with a set of technological patents pertaining to drinking water quality, wastewater treatment, and water quantity, over a period of more than 30 years. We find that, in general, the impact of water regulation on innovation is both statistically and economically significant. In the case of drinking water, new regulations on national drinking water standards are found to have a stimulating effect on related patents. On water pollution, amendments to the Clean Water Act spur more water pollution-reduction technologies. Last, water quantity related technological patents respond positively to both state-level water policy and past economic damages due to water scarcity, and the latter has a more substantial impact on water quantity related technologies.

Welfare Analysis under Probabilistic Choices in a Rational Expectations Equilibrium Model (D8, G1)

Xuezhong He
,
University of Technology Sydney
Lei Shi
,
Macquarie University
Marco Tolotti
,
Ca' Foscari University of Venice

Abstract

In a rational expectations equilibrium model, when traders can make optimally probabilistic choices to determine the level of uncertainty in information acquisition, traders can be made better off in the asymmetric information equilibrium compared to a symmetric-information equilibrium with no informed traders. Although informed trading turns the risky asset into a low risk and low return investment (risk-return effect) and distorts risk sharing (Hirshleifer effect), it also provides those traders who become informed with higher trading profitability (asymmetric information effect). We show that the asymmetric information effect is not completely washed out by information acquisition cost and provide necessary and sufficient conditions under which it dominates the risk-return and Hirshleifer effects.

What Information Does Risk Neutral Skewness Contain? Evidence From Price Rebounds and Momentum Crashes (G1)

Paul Borochin
,
University of Miami
Yanhui Zhao
,
University of Wisconsin-Whitewater

Abstract

Positive option-implied risk-neutral skewness (RNS) predicts next-month
abnormal underlying stock returns driven by upward rebounds of previously
undervalued stocks. The RNS anomaly is strongest in periods of post-recession
rebounds when momentum crashes occur. Furthermore, the momentum
anomaly is strongest (weakest) in stocks with the most negative (positive)
RNS. We generalize our findings to non-optionable stocks by constructing an
RNS factor-mimicking portfolio, finding that a momentum strategy that avoids
performance reversals has meaningfully superior performance. Our results
hold after controlling for trading frictions, firm characteristics, and common
risk factors.

When Does Being Ethical Confer a Profit Advantage Over Rivals? A Theoretical Model of Ecological Awareness (L1, Q5)

Gherardo Girardi
,
University of St. Mary and Queen Mary University of London

Abstract

Being ecologically aware need not imply lower profits than rivals, even if it implies a higher unit cost of production. We consider a Stackelberg leader who wishes to reduce the harmful impact of its industry on the environment and who competes with a conventional profit-maximizing follower. We assume that the leader’s output is less polluting than that of its rival, this being achieved at the expense of a higher unit cost of production. Being ecologically aware, the leader expands its output so as to displace its rival’s dirtier output. This increases its market share and can make it more profitable than its rival. Specifically, we show that an ecologically aware market leader can be more profitable than its profit-maximizing rival in situations where, had it not been ecologically aware, it would have been less profitable.

When Does the Platform Tell You the Truth? Optimal Design of Persuasion Policy in the Two-Sided Market (L8, M3)

Andy Tao Li
,
Tsinghua University
Jie Zheng
,
Tsinghua University

Abstract

We study the information announcement strategy for the C2C Service platform, whereas the platform provides not only a less profitable standard service with stochastic stock-out risk, but also a more profitable guaranteed service with no stock-out risk. While the expected stock-out rate of the standard service is commonly known, customers do not observe the availability of the standard service when they choose between these two services, and thus their decision can be influenced by the service availability information announced by the platform. By leveraging the stock-out forecasting technique, we show that the platform has incentive to manipulate its disclosure policy via a Bayesian persuasion approach. This implies that the platform will not always truthfully disclose the stock-out rate of the standard service for profit maximization concern. Thus, designing a state-contingent probabilistic disclosure policy, which allows the platform to announce a higher stock-out rate of the standard service to persuade customers to adopt the guaranteed service and hence involves lying under some conditions, will be optimal for the platform to boost its profit. We solve for the optimal Bayesian persuasion strategy for the platform, and fully characterize the conditions under which when the standard service is not available the platform tells the truth, the platform tells the partial truth and the partial lie, and the platform tells all lies (which is information-wise equivalent to revealing no information), respectively. We also investigate the optimal truth telling probability of the equilibrium outcomes in the short-run and in the long-run, and conduct analyses on the consequences for the platform's profit, suppliers' profit and consumers' surplus. Our study has important policy implications and provide a better understanding about the information manipulation incentive of the platform in a two-sided market of differentiated service with information asymmetry between the platform and customers.

Why Companies Are So Different? Alternative View on The Firms' Financial Design (G3, M2)

Anastasia Stepanova
,
Higher School of Economics
Maria Kokoreva
,
Higher School of Economics

Abstract

For years we believed in sectoral structure of the economy. We used to believe in the similar business models and, therefore, decision-making process, risk preferences and, finally, quite similar performance indicators, for different companies in one sector. We use sectoral benchmarks for corporate decisions. We had a good reason for that; oil companies, the largest companies in 90s, have several business models and similar decision-making process. However, in recent years we observe the structural changes of the global economy. Today Apple, Google, Microsoft, Amazon and Facebook are the five largest companies in the world, and they have very different business models with different stakeholders and their risk preferences. So, we believe it’s time to think deeper and look for other special features that make companies comparable to each other. We believe that today it is more about people. Decision-makers (founders, CEOs, stakeholders) develop the firm financial design, or architecture, depending on their risk preferences, goals and behavioral biases.
The prior work reports conflicting evidence on the relationship between owners, corporate governance, and capital structure of public companies as well as on the industry specific drivers of firm performance. We reexamine the links between the financial decisions in the companies and governance mechanisms by applying the inductive approach to the pool of public companies in Russell 3000. We apply the cluster analysis for determination of typical patterns of firm financial architecture. We demonstrate that there are 9 sustainable patterns of firm architecture in US market nowadays and describe the portraits of typical firms. Interestingly, we show that performance and firm risk differ significantly for companies with different architecture patterns. Finally, we show that the majority of patterns are not strongly correlated the industry, and industry factors do not play anymore the crucial role for the firm’s risk preferences and performance.

Why So Negative? Belief Formation in Boom and Bust Markets (G4, E3)

Jan Mueller-Dethard
,
University of Mannheim
Pascal Kieren
,
University of Mannheim
Martin Weber
,
University of Mannheim and CEPR

Abstract

What determines investors’ risk-taking across macroeconomic cycles? Recently, Cohn et al. (2015) find evidence that investors’ risk preferences change across boom and bust markets (countercyclical risk aversion). In this study, we test whether the belief channel can cause the same observed investment pattern. More precisely, we examine whether investors’ belief formation is inherently different in boom and bust markets, which can ultimately lead to the observed changes in risk-taking. Findings from neuroscience suggest that this may be the case, as brain processes differ depending on whether people learn from negative or positive outcomes (Knutson and Bossaerts, 2007).
We conduct an experiment in which participants first form beliefs in either a boom or a bust market and then invest in one of two lotteries, a risky (with fixed probabilities) or an ambiguous lottery (with unknown probabilities). The market phase itself (boom/bust) and the strength of the market phase (strong/weak) define our two treatment variables. We purposely give participants the freedom to form subjective beliefs about the ambiguous lottery by sampling from it. Our main results can be summarized as follows. Subjects who learn to form beliefs in the negative domain (bust) are significantly more pessimistic in their return expectations both compared to the Bayesian forecast and to those who learn in the positive domain (boom), consistent with Kuhnen (2015). We find that the induced pessimism translates to lower risk-taking. In particular, those subjects who learn to form beliefs during a severe bust market invest on average 17 % less in the ambiguous lottery than the average participant. For the risky lottery, however, we do not find any significant difference, indicating that the belief channel indeed drives our findings. By demonstrating that changes in beliefs affect investors’ risk appetite in boom and bust markets, our results provide new policy implications.

Willingness to Pay for Brand Reputation: Lessons from the Volkswagen Emissions Scandal (L5, M2)

Xiaogang Che
,
City University of London
Hajime Katayama
,
Waseda University
Peter Lee
,
JP Morgan Chase Institute

Abstract

Brand reputation, which reflects consumers' view of a firm and affects how much they are willing to pay for its products, is a valuable asset for the firm. A large body of theoretical research offers discussion on the relationship between brand reputation and revenue. Despite this recognition, it is difficult, if not impossible, to measure the value of brand reputation, in particular, whether and how much a consumer is willing to pay for reputation, as reputation is not tangible like the quality or design of a good.

In September 2015, the United States Environmental Protection Agency (EPA) publicly announced that Volkswagen (VW) had installed emissions-compliance ``defeat device'' software in their diesel models produced from 2009 to 2015 to help them pass the emissions tests, which as a result were identified as emissions violators. As there was no prior warning for the scandal, the EPA's announcement was a surprise to the U.S. car market. Exploiting this exogenous shock as a natural experiment, we estimate the impacts of the scandal on the prices of VW cars, which reflect whether and to what extent consumers' willingness to pay (WTP) for VW's brand reputation is influenced. The data is from eBay, where the auction mechanism is second-price; it is a weakly dominant strategy for buyers to bid their true values and hence transaction prices in the auctions give us a direct measure for WTP.

Our difference-in-differences estimates show that non-violators, namely diesel cars produced in 2000-2008 and gasoline cars, experienced significant price drops by 14 percent and 9 percent, which purely reflect a decline in consumers' WTP for VW's brand reputation. Additionally, the magnitude of price-drop is smaller for violators than for non-violators of VW diesel cars, suggesting that consumers rationally adjust their WTP by considering compensation provided by VW for violating models.