TREASURY INSPECTOR GENERAL FOR TAX ADMINISTRATION

Office of Audit

Highlights

The Large Case Examination Selection Method
Consistently Results in High No-Change Rates

Final Report issued on June 22, 2020

Highlights of Reference Number:† 2020-30-031 to the Commissioner of Internal Revenue.

IMPACT ON TAXPAYERS

The IRSís primary objective in selecting returns for examination is to promote the highest degree of voluntary compliance.† The LB&I Division has a variety of examination programs and uses a multitude of methods to select returns.† However, it consistently spent most of its examination resources on large business returns.

WHY TIGTA DID THE AUDIT

The IRS compiles Tax Gap data to periodically update appraisals of the nature and extent of tax payment noncompliance for use in formulating tax administration strategies.† The IRS estimates the average annual gross Tax Gap for Tax Years 2011 through 2013 to be $411 billion.† The largest component, $352 billion, is attributable to underreporting of taxes. †Large corporation (assets of $10 million or more) tax noncompliance contributes an estimated $26 billion to the average annual underreporting Tax Gap.

This audit was initiated to evaluate the selection process, use of resources, and examination productivity for corporate returns examined as part of the Large Business and International (LB&I) Divisionís Discriminant Analysis System (DAS) workstream.† Approximately 44 percent of Form 1120, U.S. Corporation Income Tax Return, examinations during Fiscal Years 2015 through 2018 were closed from the DAS workstream.

WHAT TIGTA FOUND

TIGTA analyzed the 10,755 returns closed in the DAS workstream during Fiscal Years 2015 through 2018 and found that 47 percent were closed with no change to the tax return.† TIGTA analyzed the potential cost for excessive time charged to no-change returns, i.e., time in excess of 200 hours, and estimated that potentially $22.7 million was spent examining no‑change returns in excess of 200 hours.

Of the 10,755 returns, 7,831 returns (73 percent) were systemically selected, i.e., were selected as the primary tax return to be examined.† The overall no-change rate for these returns was about 55 percent (4,327 of the 7,831), and the no‑change rate was generally high across all activity codes for businesses with assets of $10 million or more (ranging from 44 percent to 61 percent).

The LB&I Division is updating the DAS model to improve the no‑change rates.† However, TIGTA found that the LB&I Division is not leveraging all available information to improve the model, such as the examination scope and which tax issues are the most productive to examine.† LB&I also plans to test the new formulas only on returns that are nearly a decade old.

TIGTA reviewed the examination results for the 10,755 DAS returns and found that the LB&I Division is not adequately monitoring DAS examination results to assess whether the model is effectively ranking returns based on the likelihood of potential tax adjustment.† When assessing the productivity of its models, the LB&I Division does not use the actual examination amount when an examination results in a refund.† Instead, it treats examinations that result in a refund as no change in tax.† By not using the actual examination amount for refunds, the LB&I Divisionís productivity is skewed to the positive and does not accurately reflect the true compliance impact.

WHAT TIGTA RECOMMENDED

TIGTA recommended that the LB&I Division develop an action plan to reduce the examination no-change rates; avoid working pickup returns unless issues are established on primary tax returns that may affect prior or subsequent years; minimize hours expended on no‑change closures; test newly developed formulas on current examined returns and consider breadth of scope and noncompliance issues found in past examinations; and analyze DAS return actual examination results on a regular basis.†

The LB&I Division agreed with two recommendations and will formulate a plan to reduce the no‑change rate and hours incurred.† The LB&I Division also plans to analyze examination results on a regular basis.† The LB&I Division disagreed with three recommendations pertaining to the DAS model and noted that it released a new DAS model in April 2020.† The LB&I Division also disagreed with using actual examination dollar results for evaluating the effectiveness of the DAS model.

 

READ THE FULL REPORT

To view the report, including the scope, methodology, and full IRS response, go to:

https://www.treasury.gov/tigta/auditreports/2020reports/202030031fr.pdf.

 

Phone Number ††/† 202-622-6500

E-mail Address †/TIGTACommunications@tigta.treas.gov

Website†††††† ††††††/https://www.treasury.gov/tigta