Parameter Estimation Using On-Line Household Surveys

Paper Session

Saturday, Jan. 7, 2017 2:30 PM – 4:30 PM

Hyatt Regency Chicago, Regency D
Hosted By: American Economic Association
  • Chair: Wilbert Van der Klaauw, Federal Reserve Bank of New York

The Strategic Survey Question (SSQ) Approach to Estimating Preference Parameters

John Ameriks
,
Vanguard Group
Joseph Briggs
,
New York University
Andrew Caplin
,
New York University
Matthew D. Shapiro
,
University of Michigan
Christopher Tonetti
,
Stanford University

Abstract

Strategic Survey Questions (SSQs) elicit preference parameters by posing large scale, hypothetical stated preference questions framed by life-cycle choices confronted by individuals. SSQ question are designed so that the responses can be mapped exactly into optimum problems that economists use to model decision-making. Building on work by Barsky, Juster, Kimball, and Shapiro on eliciting risk tolerance and on Ameriks, Caplin, Laufer, and Van Nieuwerburgh on eliciting public care aversion and bequest motives, the Vanguard Research Initiative (VRI) has designed, tested, and implemented SSQs to elicit and analyze parameters essential for understanding asset accumulation and decumulation in the face of multiple late-in-life risks. The VRI embeds these SSQs in a large panel of older Americans who have significant financial assets and who thus face decisions about self-financing late-in-life risks. The VRI combines a rich set of SSQs, survey data on income, wealth, labor, and expectations, and administrative account data on asset holdings. This paper demonstrates the internal validity of SSQ responses using a test-retest approach. The paper then shows how using the behavioral data and SSQ response jointly can provide much sharper understanding of decision-making than by using behavioral data alone, specifically by examining the interaction between financial risk and health risk.

Preference for the Workplace, Investment in Human Capital, and Gender

Matthew Wiswall
,
Arizona State University
Basit Zafar
,
Federal Reserve Bank of New York

Abstract

We use a hypothetical choice methodology to estimate preferences for workplace attributes and quantify how much these preferences influence pre-labor market human capital investments. This method robustly identifies preferences for various job attributes, free from omitted variable bias and considering the equilibrium job match. Women on average have a higher willingness-to-pay (WTP) for jobs with greater work flexibility and job stability, and men have a higher WTP for jobs with higher earnings growth. These job preferences relate to college major choices and actual job choices, and explain as much as 25 percent of the gender wage gap.

Analyzing Consumer Decision Making Under Uncertainty Using Strategic Survey Questions (SSQs)

John Sabelhaus
,
Federal Reserve Board and University of Maryland

Abstract

Standard expected utility maximization models make strong predictions about how changing the payoff to a risky decision affects the desirability of that decision, given specifications for the risk environment and underlying utility function parameters. In this Strategic Survey Question (SSQ) experimental module run with Survey of Consumer Expectations (SCE) on-line panel members, respondents were asked to score several risky decisions. Each decision was presented and scored under four different payoff assumptions. The idiosyncratic risk components can be subtracted out for each respondent, because the questions were written such that underlying risks do not vary with the expected payoffs. The key baseline expectations need to compute expected values for each respondent are taken from the SCE core questions. The risky decisions involved choices about changing jobs, expanding a business, buying a home, investing in human capital, and changing portfolio composition. As expected, respondents scored any given decision more favorably when the expected payoff is higher. Respondents who self-reported higher risk aversion require a higher payoff in order for a given decision to score above the break-even point, meaning the point where taking the action just becomes preferred to not taking the action. This new multiple-payoff approach makes it possible to go beyond point estimates for parameters of interest, and test assumptions about curvature across the payoff possibilities, and the role of ambiguity in underlying expectations.

Estimating Discount Functions with Consumption Choices Over the Lifecycle

David Laibson
,
Harvard University
Peter Maxted
,
Harvard University
Andrea Repetto
,
Adolfo Ibanez University
Jeremy Tobacman
,
University of Pennsylvania

Abstract

Intertemporal preferences are difficult to measure because financial payments (e.g., checks in the mail) do not coincide with utility flows. We estimate time preferences using a structural buffer stock consumption model and the Method of Simulated Moments. The model includes stochastic labor income, liquidity constraints, child and adult dependents, liquid and illiquid assets, revolving credit, retirement, and discount functions that allow short-run and long-run discount rates to differ. Data on wealth accumulation and credit card borrowing over the lifecycle identify the parameters in the model. In almost all specifications we reject the restriction to a constant discount factor (i.e., exponential discounting). Our benchmark estimates imply a short-term discount factor of β = 0.50 and a long-term annualized discount factor of δ = 0.97.
Discussant(s)
Christopher Carroll
,
Johns Hopkins University
Eleanor Dillon
,
Arizona State University
Claudia Sahm
,
Federal Reserve Board
Mariacristina De Nardi
,
NBER
JEL Classifications
  • C4 - Econometric and Statistical Methods: Special Topics
  • C8 - Data Collection and Data Estimation Methodology; Computer Programs