« Back to Results

High-frequency Data and Real Economic Activity

Paper Session

Saturday, Jan. 5, 2019 2:30 PM - 4:30 PM

Atlanta Marriott Marquis, International 3
Hosted By: American Economic Association
  • Chair: Andrew H. McCallum, Federal Reserve Board

High-frequency Spending Responses to the Earned Income Tax Credit

Aditya Aladangady
,
Federal Reserve Board
Shifrah Aron-Dine
,
Stanford University
David Cashin
,
Federal Reserve Board
Wendy Dunn
,
Federal Reserve Board
Laura Feiveson
,
Federal Reserve Board
Paul Lengermann
,
Federal Reserve Board
Katherine Richard
,
University of Michigan
Claudia Sahm
,
Federal Reserve Board

Abstract

Many lower-income households face substantial, high-frequency income volatility and have limited financial buffers. Thus, even temporary shifts in the timing of income may lead to large welfare costs due to an inability to smooth consumption. Few studies have quantified the spending effects of short-lived income shocks due to a lack of sufficiently high-frequency spending data. Using a novel dataset of anonymized daily, geographic spending, we study the spending effects of a two-week delay in 2017 for tax refunds claiming the earned income tax credit (EITC). The regression results show that spending out of the EITC is highly sensitive to the timing of receipt, suggesting that these households are unable to smooth their spending through an income delay of only a few weeks. Moreover, EITC receipt affects daily spending on nondurable necessities, such as groceries, in addition to durable goods. These results point to substantial, negative welfare impacts on lower-income households from temporary income shocks.

The Self-Constrained Hand-to-Mouth

Michael Gelman
,
Claremont McKenna College

Abstract

This paper examines the response of food expenditures to the receipt of paychecks using financial account data from a personal finance app. Similar to previous studies, this paper finds that food expenditures increase during the week the paycheck is received. While the standard explanation for this result is temporary liquidity constraints, this paper argues otherwise. Intuitively, it's unlikely that individuals will be liquidity constrained during the weeks they receive their paycheck. Therefore, their decision to spend more during weeks in which they have more liquidity likely reflects preferences and not constraints. The intuition is formalized through specifying a buffer stock model of consumption. Model simulations show that indeed consumption behavior is not affected by liquidity during the week the paycheck is received. The empirical results match the theoretical predictions and confirm that temporary liquidity constraints cannot explain excess sensitivity to regular paychecks.

Weekly Payroll Employment Data for the United States

Tomaz Cajner
,
Federal Reserve Board
Leland Crane
,
Federal Reserve Board
Ryan Decker
,
Federal Reserve Board
Adrian Hamins-Puertolas
,
Federal Reserve Board
Christopher Kurz
,
Federal Reserve Board

Abstract

This paper examines new weekly employment series based on payroll processor microdata covering one fifth of U.S. private employment. These series provide very high frequency information about a large segment of the labor market, potentially complementing more well-known labor market indicators. We document several interesting features of the data. First, the seasonal fluctuations are generally similar to comparable BLS series, though with some important deviations that may be related to end-of-year bonuses. We also document within-month cycles, where employment grows more quickly at the start of most months and weakens later in the month. Second, we show a strong relationship between our weekly series and weekly unemployment insurance claims, which suggests some forecasting power. Third, we study the impact of time aggregation on the measurement of employment inaction, which is relevant for fitting models with non-convex adjustment costs. Finally, we explore whether the weekly data have any additional forecasting power for official employment releases once monthly data have been accounted for.

Fast Adjustment to Exchange Rate Shocks

Andrew H. McCallum
,
Federal Reserve Board
Nikolas Zolas
,
U.S. Census Bureau

Abstract

We use the universe of U.S. international trade observed at the daily frequency to document how quickly exchange rate shocks pass-through to trade quantities and trade prices. By focusing on large, unanticipated exchange rate shocks and daily trade flows, our specifications do not face the same challenges as typical exchange rate pass-through regressions. Our results provide new evidence about the degree and speed of pass-through. We close with a discussion of the implications these results have for macroeconomic models.
Discussant(s)
Dan Silverman
,
Arizona State University
Lorenz Kueng
,
Northwestern University
Matthew Notowidigdo
,
Northwestern University
JEL Classifications
  • E2 - Consumption, Saving, Production, Investment, Labor Markets, and Informal Economy
  • F4 - Macroeconomic Aspects of International Trade and Finance