Abstract

http://ssrn.com/abstract=519783
 
 

Footnotes (35)



 


 



Who Should Bear Tax Compliance Costs?


Joseph Bankman


Stanford Law School

March 10, 2004

Stanford Law and Economics Olin Working Paper No. 279

Abstract:     
Regulation is expensive. The federal income tax comprises one of the most extensive forms of government regulation, and one of the most expensive. Much of this expense is recognized in the form of reduced work effort or saving. Economic models that evaluate fundamental tax reform proposals often focus exclusively on these two forms of tax-induced changes in behavior, ignoring compliance costs. These costs, however, are quite significant. They include the time spent filing one's tax return and maintaining records related to that filing; the time spent learning and negotiating the rules when engaged in various forms of tax planning; and the amounts paid to third parties, such as accountants, lawyers, financial planners or software providers, to that same end. They also include the costs the government incurs to promulgate and enforce the law.

Estimates of compliance costs imposed on both individuals and business entities range between 10% to 25% of revenue raised, or from about $100 billion to $250 billion a year. Compliance costs substantially reduce the social gains from taxation; in some areas, these costs may outweigh those gains altogether.

This paper examines the question: Who should bear tax compliance costs? The question is important for a number of reasons. First, many voters believe that those responsible for the tax law do not accurately reflect constituent needs, and that as a result the government does not adequately take taxpayer compliance costs into account when designing tax rules. Second, some compliance costs fall disproportionately upon a small group of taxpayers. Third, many compliance costs are variable, and it may be efficient to require those costs be borne by one party or another. Finally, compliance costs reduce the social value of regulation in fields other than tax, and it may be possible to extend the analysis here to those other fields.

Part I introduces the subject and analysis with an example that provides probably the strongest case for government reimbursement of compliance costs: taxpayer compliance management program (TCMP) or TCMP-like audits. Part II sets forth a preliminary analytical framework from which to view compliance costs. Part II.A. examines the special case in which voter preferences are flawlessly translated into law and procedure. Part II.B. assumes that government officials maximize values other than (or possibly in addition to) those they impute to voters and do not adequately weight compliance costs; and that those costs would receive greater weight if they were treated as a separate budget item. Part II.C. assumes that the IRS is well intentioned but impolitic; that the legislature and/or electorate wrongly believes that the agency does not adequately weigh compliance costs when setting policy and so sets undesirable constraints on agency behavior. Part III uses this analytical framework to analyze two significant forms of taxpayer compliance costs: the costs of garden-variety audits and the costs of filing individual tax returns.

Number of Pages in PDF File: 28

working papers series


Download This Paper

Date posted: March 19, 2004  

Suggested Citation

Bankman, Joseph, Who Should Bear Tax Compliance Costs? (March 10, 2004). Stanford Law and Economics Olin Working Paper No. 279. Available at SSRN: http://ssrn.com/abstract=519783 or http://dx.doi.org/10.2139/ssrn.519783

Contact Information

Joseph Bankman (Contact Author)
Stanford Law School ( email )
559 Nathan Abbott Way
Stanford, CA 94305-8610
United States
650-725-3825 (Phone)
650-725-7663 (Fax)
Feedback to SSRN


Paper statistics
Abstract Views: 2,569
Downloads: 416
Download Rank: 38,396
Footnotes:  35

© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright   Contact Us
This page was processed by apollo5 in 0.250 seconds