Why Do People Give Interest-Free Loans to the Government? An Experimental Study of Interim Tax Payments
Benjamin C. Ayers
University of Georgia
Steven J. Kachelmeier
University of Texas at Austin - Department of Accounting
John R. Robinson
University of Texas at Austin
Journal of the American Taxation Association, Fall 1999
Taxpayers remit through withholding and estimated tax payments substantially more than the amounts assessed upon filing of returns. Some excess payments may reflect involuntary institutional features (e.g., default withholding rules), but there may also be voluntary excess remittances reflecting taxpayer preferences. Our study uses a case-based experiment administered to 162 work-experienced MBA student volunteers to investigate such preferences in an environment stripped of transaction costs and sources of confusion regarding minimum payment provisions. In addition, we use a factorial design to explore the sensitivity of taxpayer interim payment preferences to tax policy considerations involving uncertainty, payment form, and experience.
We present three related arguments for why taxpayers might prefer interim payment patterns that violate strict adherence to the time value of money. First, if taxpayers adopt a reference point of covering the end-of-year tax bill with their interim payments, the alternative of delaying payment in order to receive a larger sum now may be perceived as an unattractive prospect (Shelley 1993; Loewenstein 1988). Second, the dread phenomenon may explain why people often express preferences to get a negative outcome over with rather than anticipate its future occurrence (Loewenstein and Thaler 1992; Loewenstein 1987). Finally, taxpayers may view excess interim payments as a form of forced savings, anticipating their limited self-control (Shefrin and Thaler 1992; Thaler and Shefrin 1981; Schelling 1984). All three arguments offer consistent predictions in our setting, offering theoretical reasons why policy makers should consider taxpayers' preferences when formulating policies involving interim tax remittances.
The experiment involved asking participants to indicate desired quarterly tax remittances for a hypothetical case in which a taxpayer expected $16,000 of tax on an annual income of $80,000. We manipulated the uncertainty associated with the end-of-year tax liability by adding a description of the annual tax liability in a worst-case scenario. The form of payment was either withholding from wages or estimated tax installments. We also assessed the effects of self-reported taxpaying experience on taxpayer preferences. In all treatment conditions, case information indicated a minimum quarterly payment that would avoid any underpayment penalty (based on the prior-year tax liability). Participants basing decisions strictly on the time value of money would prefer to remit only this minimum amount in each of our treatment conditions.
We find that 43 percent of the participants indicate a preference to pay more tax on an interim basis than the minimum necessary to avoid an underpayment penalty. Because participants are screened for case comprehension, we interpret this frequency as reflecting genuine preferences, not ignorance or confusion. We also find that participants' overpayment preferences are significantly more pronounced in cases where the current-year tax liability is more uncertain. This effect is different from risk aversion, because even the most risk averse taxpayer would be better off economically by paying the minimum possible to the government and investing any desired provision for taxes in a risk-free account. In contrast to the effect of uncertainty, we find that preferences for overpayment are robust to whether the payment form is withholding from wages or estimated tax installments. Finally, we find that taxpayer experience mitigates the general propensity to overpay and also mutes the influence of uncertainty.
JEL Classification: H24, M40Accepted Paper Series
Date posted: September 12, 1999
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo4 in 0.422 seconds