|
Activity Number:
|
567
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #303739 |
|
Title:
|
Propensity Scoring: An IRS Case Study in Determining the Relationship Between Bank Products and Noncompliance
|
|
Author(s):
|
Karen C. Masken*+ and Mark Mazur and Joanne Meikle and Roy Nord
|
|
Companies:
|
IRS and IRS and IRS and IRS
|
|
Address:
|
1111 Constitution Ave NW, Washington, DC, 20224,
|
|
Keywords:
|
propensity scoring ; scoring ; matching ; estimation ; sensitivity analysis
|
|
Abstract:
|
An estimated 20 million individual income taxpayers utilize bank products (Refund Anticipation Loans (RALs) and Refund Anticipation Checks/Cards (RACs)) to obtain their income tax refund. Analysts in IRS were asked to determine if the use of RALs and RACs may also lead to less compliant behavior on the part of taxpayers and perhaps the preparers who facilitate access to these bank products. The primary issue we faced in conducting this research was that the only available data addressing noncompliance was from operational audits. Returns are not randomly selected for audit and therefore this data has selection bias. To address this bias, we chose to use propensity scoring techniques. This paper discusses the methodology we used in scoring, matching, estimation and sensitivity analysis. We found that the use of RALs was correlated with noncompliance, however the use of RACs was not.
|
