Page 60 - TaxAdviser_2022

business-component level. However, if test requirements that have histori- credit computations is included in Rev.
the overall business component does cally received less scrutiny, such as the Proc. 2011-42, which provides general
not meet all four tests, certain steps are technological-information test. guidance for taxpayers that use statisti-
followed. In accordance with Regs. Sec. Two recent cases that reflect this cal sampling and statistical estimates
1.41-4(b)(2), the shrink-back rule pro- trend and emphasize that the burden of for tax filing purposes. Provided that
vides that the four-part test is first ap- proof is on the taxpayer to satisfy and taxpayers follow the IRS guidelines, sta-
plied at the level of the discrete business substantiate the four-part test at the tistical sampling can be a cost-effective,
component. If the requirements are not business-component level are Siemer efficient approach to calculating and
met at this level, then the four-part test Milling Co., T.C. Memo. 2019-37, and substantiating the research credit. In ad-
is applied at the most significant subset Max, T.C. Memo. 2021-37. In Siemer dition, the IRS often encourages the use
of elements of the business component. Milling, the court analyzed and evalu- of statistical sampling for the research
For example, if the new engine product ated the four-part test for each of the credit due to the efficiencies provided
developed by the motorcycle engine taxpayer’s seven business components during examination.
builder referenced earlier does not meet independently. The court disallowed the Statistical sampling often minimizes
the four-part test, the requirements may taxpayer’s research credits because none the level of involvement required from
be applied at the next most significant of the business components met all four the taxpayer to collect the quantitative
subset of elements of the engine (e.g., parts of the four-part test requirement. data required to compute QREs and cal-
the carburetor). Similarly, in Max, the Tax Court con- culate the research credit. Additionally,
cluded that the taxpayer could not claim statistical sampling may significantly
Substantiation requirements a research credit because the taxpayer’s reduce the level of qualitative analysis
Regs. Sec. 1.41-4(d) provides that activities did not meet three of the required to meet the substantiation
taxpayers claiming a research credit are four-part test requirements, including requirements of Regs. Sec. 1.41-4(d).
required to keep records “in sufficiently the process-of-experimentation test. In Taxpayers with a considerable number
usable form and detail to substantiate particular, the court agreed with the IRS of business components may find it
that expenditures claimed are eligible that the taxpayer did not fundamentally burdensome to properly document the
for the credit.” Sec. 6001 and the regu- rely on principles of the hard sciences four-part test requirements for each
lations thereunder provide taxpayers and did not employ a process similar business component. By implementing a
specific guidance with respect to record to the scientific method during its pre- statistical sampling approach, the analy-
retention, including the requirement to production process. sis is limited to the taxpayer’s sample
keep permanent books of account or The FAA concludes that, while sta- of records.
records. As highlighted in the FAA, the tistical methods can be used for estima- Further, the use of statistical sam-
substantiation requirements must also tion purposes, the general requirement pling can make the IRS examination
be satisfied at the business-component that the four-part test be applied sepa- process more efficient for the taxpayer
level. Taxpayers can look to case law for rately to each business component and and the IRS. Provided the taxpayer
examples of how the courts interpret the substantiated at such level is not altered uses valid statistical methods to de-
substantiation requirement with respect by using statistical methods. Therefore, termine its sample selections, the IRS
to business components and qualified careful consideration of statistical ap- will generally limit its purview to the
research activities. proaches that capture and allow proper taxpayer’s sampling selections during
There are a variety of cases that substantiation of the four-part test by exam. As a result, the process of is-
examine the substantiation requirement each business component is key to suc- suing and responding to information
and demonstrate that the four-part test cessfully defending a research credit documentation requests (IDRs) during
is applied at the business-component claim during audit. an IRS exam is often more streamlined
level. With respect to substantiation, the because taxpayers generally find it less
FAA cites several notable cases includ- Benefits of statistical sampling arduous to sufficiently document and
ing Union Carbide Corp., T.C. Memo. The IRS has a history of accepting substantiate their research credit claims
2009-50, aff’d, 697 F.3d 104 (2d Cir. taxpayer estimates based on statistical as part of the initial filing rather than in
2012), and Suder, T.C. Memo. 2014-201. analysis and has provided guidance response to an IDR.
Based on the recent audit environment, on sampling through field directives,
IRS examinations appear to be plac- revenue procedures, and audit technique Sampling process overview
ing more focus on taxpayers’ business procedures. Guidance that best supports In a typical research credit study in-
components as well as certain four-part the use of statistical analysis in research volving sampling, the taxpayer’s list of

www.thetaxadviser.com February 2022 13