Monthly Budgeting Heuristics: Evidence from ‘Extra’ Paychecks†

This paper provides empirical evidence that households follow budgeting heuristics which can generate consumption patterns that are not easily explained by standard models. Specifically, I show that households follow a heuristic in which budgets are constructed based on typical monthly income while ignoring the fact that true income may vary from month to month. To do so, I first formalize and explore the implications of a general model of budgeting heuristics in which individuals mistakenly extrapolate their current monthly income into the future. When the frequency of consumption and the frequency of income are sufficiently misaligned, the model predicts excess sensitivity of consumption in months with atypical income. To empirically test the model’s predictions, I exploit a unique feature of bi-weekly pay schedules. Because bi-weekly workers are paid on a regular two-week schedule, they typically receive two paychecks per month with the exception of two months out of the year, during which they receive three. Importantly, these third paychecks can only be viewed as ‘extra’ if households are evaluating their income on a monthly basis, a decision that is not a priori assumed. Using household-level micro data from the Consumer Expenditure Survey, I find evidence of excess sensitivity of consumption to these third paychecks and that this response is entirely due to the purchase of durable goods, and specifically vehicle purchases. Household spending increases by approximately $262 on average in months following third paychecks, or roughly 15.7 percent of the average paycheck. I also find no corresponding response in non-durable goods, suggesting that liquidity constraints alone cannot explain these results. These findings are consistent with the predictions of the budgeting heuristic model. †I am deeply indebted to my advisors Daniel Gottlieb, Judd Kessler, Nikolai Roussanov, Todd Sinai, and Jeremy Tobacman for their encouragement and support. I am also grateful to Santosh Anagol, John Beshears, Alexander Gelber, Mark Jenkins, Jonathan Kolstad, Katherine Milkman, Devin Pope, Nicholas Souleles, and numerous participants in the 2013 Spring School in Behavioral Economics, the 2013 LBS Trans-Atlantic Doctoral Conference, and the 2013 Consumer Expenditure Survey Microdata Workshop for helpful suggestions and comments. I gratefully acknowledge support from the Carlos de la Cruz Fellowship. All errors are my own. The most recent version of this paper can always be found on my website, https: //bepp.wharton.upenn.edu/profile/756/ ‡The Wharton School, University of Pennsylvania, 3000 Steinberg Hall-Dietrich Hall, 3620 Locust Walk, Philadelphia, PA 19104, [email protected].