IRS Must Rethink Foreign Corp Nonfiler Campaign, TIGTA Says

The Internal Revenue Service Should Revise the Approach Used to Identify Foreign Corporations That Do Not File Required U.S. Income Tax Returns

September 11, 2019

This report has cleared the Treasury Inspector General for Tax Administration disclosure review process and information determined to be restricted from public release has been redacted from this document.

Redaction Legend:

1 = Tax Return/Return Information

2 = Law Enforcement Techniques/Procedures and Guidelines for Law Enforcement Investigations or Prosecutions.

Highlights

Final Report issued on September 11, 2019

Highlights of Reference Number: 2019-30-067 to the Commissioner of Internal Revenue.

IMPACT ON TAXPAYERS

Form 1120-F, U.S. Income Tax Return of a Foreign Corporation, nonfilers are foreign entities that are legally required to file but fail to do so. The IRS considers nonfiling an egregious problem because it contributes to the Tax Gap and undermines tax compliance. Compliant taxpayers who timely file and pay their taxes want to be confident that other taxpayers are doing the same.

WHY TIGTA DID THE AUDIT

The Form 1120-F Non-Filer Campaign was one of the first campaigns rolled out by the Large Business and International (LB&I) Division. This audit was initiated to assess how effectively the Form 1120-F Non-Filer Campaign is being conducted.

WHAT TIGTA FOUND

The objectives of the Form 1120-F Non-Filer Campaign are to identify nonfilers using external and internal sources to identify and address Form 1120-F nonfiling compliance risk. Although the Form 1120-F Non-Filer Campaign is ongoing, this review focused on the initial results which were based on the first three data sources (two external, one internal) used by the campaign to build cases.

The campaign was initiated without establishing evidence of a significant compliance risk pertaining to Form 1120-F nonfilers. Prior to launch, several campaign documents noted difficulties in estimating the potential nonfiler population. Additionally, the compliance initiative that preceded this campaign did not provide evidence of a potentially large Form 1120-F nonfiler population. However, it did identify a general misunderstanding of filing requirements, forms, instructions, etc. through correspondence with affected taxpayers and/or their representatives.

Also, a coordinated strategy was not established to ensure that critical campaign case actions were performed timely. Specifically, TIGTA identified untimely case actions and follow-up actions, as well as undocumented procedures and determinations. There was no mechanism in place to ensure that personnel responsible for approving or mailing the compliance letters to potential nonfilers had actually done so in a timely manner, resulting in significant delays. Furthermore, the LB&I Division did not properly plan for nonresponsive taxpayers and, consequently, it was unable to follow up appropriately with those who failed to respond.

Overall, initial campaign case results show low examination referral and proposed assessment rates. Because of this, TIGTA believes that the campaign results do not support an effective nonfiler identification process or the underlying premise of a large Form 1120-F nonfiler population.

WHAT TIGTA RECOMMENDED

TIGTA recommended that the Commissioner, LB&I Division: 1) consider (in light of the IRS's limited examination resources) whether to continue the campaign based on the initial 22 examination referrals from the first 50 soft letters issued or whether a significant change in the methods to identify nonfilers is warranted, and 2) establish controls over case building, case selection, and soft letter issuance processes to ensure consistency, documentation, and timely case actions if the IRS plans to continue with this campaign.

The IRS agreed with both recommendations and will assess the success of the Form 1120-F Non-Filer Campaign based on the 22 examinations referrals and consider whether to continue with the campaign or change methods to identify nonfilers, and the IRS has already established controls over case building, case selection, and soft letter issuance processes.

---

September 11, 2019

MEMORANDUM FOR
COMMISSIONER OF INTERNAL REVENUE

FROM:
Michael E. McKenney
Deputy Inspector General for Audit

SUBJECT:
Final Audit Report — The Internal Revenue Service Should Revise the Approach Used to Identify Foreign Corporations That Do Not File Required U.S. Income Tax Returns (Audit # 201830020)

This report presents the results of our review to assess how effectively the Large Business and International Division is conducting the Form 1120-F Non-Filer Campaign. This audit is included in our Fiscal Year 2019 Annual Audit Plan and addresses the major management challenge of the Impact of Global Economy on Tax Administration.

Management's complete response to the draft report is included as Appendix VI.

Copies of this report are also being sent to the Internal Revenue Service managers affected by the report recommendations. If you have any questions, please contact me or Matthew A. Weir, Assistant Inspector General for Audit (Compliance and Enforcement Operations).

---

Table of Contents

Background

Results of Review

Appendices

Abbreviations

BMF	Business Master File
CCNIP	Case Creation Nonfiler Identification Process
CIP	Compliance Initiative Project
CY	Calendar Year
ECI	Effectively Connected Income
FY	Fiscal Year
I.R.C.	Internal Revenue Code
IRS	Internal Revenue Service
LB&I	Large Business and International
NFA	No Further Action
TIGTA	Treasury Inspector General for Tax Administration
TIN	Taxpayer Identification Number
USTB	United States Trade or Business

---

Background

The Large Business and International (LB&I) Division believes that its mission could be improved with a strategic focus upon issues, transactions, and processes.¹ In furtherance of this belief, the LB&I Division has implemented a new approach described as “campaigns” to address taxpayer noncompliance and audit selection. The Internal Revenue Service (IRS) developed the campaign concept as a part of its Future State initiative and designed it to focus on specific compliance-related issues, instead of whole tax return characteristics. Campaigns can include audits as well as other taxpayer interactions, such as reaching out to taxpayers and tax professionals, issuing guidance, and participating in industry events for outreach.

Using campaigns to address compliance risks represents a fundamental change in the way the LB&I Division selects and performs work. This approach has focused on using a variety of treatments to achieve compliance across the LB&I Division's filing population instead of just auditing more tax returns. On January 31, 2017, the LB&I Division announced the rollout of its 13 initial campaigns. One campaign was the Form 1120-F Non-Filer Campaign.²

The Form 1120-F Non-Filer Campaign was approved on November 16, 2016, with the purposes of identifying nonfilers, encouraging filing of Forms 1120-F by foreign taxpayers, and determining the effectiveness of various data sources in uncovering nonfilers. The LB&I Division used both external and internal sources in conjunction with “soft letter” correspondence to achieve these goals.³ More specifically, it began with U.S. Customs data (external) to select 15 cases; Form 1042-S, Foreign Person's U.S. Source Income Subject to Withholding, data (internal) to select 15 cases; and State Sales Tax/Business Registrations data (external) to select 20 cases.⁴ As part of the campaign, the LB&I Division's Compliance Integration Council authorized nine full-time equivalents to complete 50 total field examinations of campaign cases for a two-year period.⁵ This particular campaign originated from a LB&I Division Compliance Initiative Project (CIP) that focused on Form 1120-F filers and nonfilers using data from the two external sources previously mentioned. In Calendar Year (CY) 2016, the LB&I Division converted this CIP into the Form 1120-F Non-Filer Campaign. The treatment stream for this campaign started with soft letter outreach and proceeded to examination when appropriate. See Appendix IV for a flowchart of the campaign process.

Internal Revenue Code (I.R.C.) provisions and related regulations subject income earned by foreign corporations to taxation in the United States under either of the two following circumstances:

The term “trade or business within the United States” is not specifically defined in either the Code or regulations. Under case law, a foreign person's business activities within the United States give rise to a USTB if those activities are “considerable, continuous and regular.” This gives rise to a facts and circumstances determination. Except under limited circumstances, the performance of personal services within the United States by a foreign person is treated as a USTB.⁹

Generally, when an entity engages in a trade or business within the United States, all income from domestic sources, other than Fixed, Determinable, Annual, or Periodical income that is not connected with the conduct of a USTB, is treated as ECI.¹⁰ Under limited circumstances, certain types of a foreign corporation's foreign source income can also be ECI if attributable to the corporation's U.S. office or other fixed place of business. In some cases, treaties can reduce the amount of tax that a foreign corporation would normally pay under the U.S. tax code. If a treaty exists, a foreign entity's foreign source income would only be subject to full taxation if it conducts business through a U.S.-based permanent establishment with attributable profits.¹¹

As a result of the foregoing rules, many foreign companies are required to file Form 1120-F. These include:

• Those that engaged in a USTB during the taxable year, even if they have no gross income, or that experienced net losses attributable to the USTB during that year.

• Those having income that is exempt by treaty and must claim such exemption by filing Form 8833, Treaty-Based Return Position Disclosure Under Section 6114 or 7701(b), with the IRS.

In general, if the foreign corporation fails to file Form 1120-F before the earlier of the date of notification by the IRS that a return is due or 18 months after the due date, it is not entitled to deductions and credits. The filing deadlines may be waived if the foreign corporation establishes to the satisfaction of the IRS Commissioner that the corporation, based on the facts and circumstances, acted reasonably and in good faith in failing to file a Form 1120-F income tax return.¹²

Foreign entities may file protective returns for any taxable year if they conducted limited activities within the United States during that year which they do not believe rise to the level of a trade or business. Foreign corporations that file protective returns are not required to report income. This will be discussed later in the report.

Form 1120-F nonfilers are foreign entities that are legally required to file but fail to do so. The IRS considers nonfiling an egregious problem because it contributes to the Tax Gap and can cause compliant taxpayers to lose faith in the fairness of the tax system.¹³ Compliant taxpayers that timely file and pay their taxes want to be confident that other taxpayers are doing the same.

This review was performed with information obtained from the LB&I Division Cross Border Activities Practice Area in Washington, D.C. We also performed this review at the following LB&I Division posts of duty: Plantation, Florida; Rochester, New York; and Farmers Branch, Texas, during the period September 2018 through May 2019. We conducted this performance audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objective. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective. Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.

Results of Review

The objectives of the Form 1120-F Non-Filer Campaign are to identify nonfilers using external, e.g., U.S. Customs Import Data, State Department registrations, and State sales and use tax filings, and internal, e.g., Form 1042-S, the Integrated Data Retrieval System, sources to identify and address Form 1120-F nonfiling compliance risk.¹⁴ LB&I Division management also want to determine whether the campaign has led to an increase in Form 1120-F return filing.¹⁵ However, the Form 1120-F Non-Filer Campaign continues to face two major challenges.

• How to identify foreign corporations doing business in the United States that did not file a Form 1120-F. The LB&I Division initially chose to use U.S. State Sales Tax/Business Registrations data, U.S. Customs data, and Form 1042-S data to identify entities doing business in the United States.¹⁶ It compared these data against information in the IRS's Business Master File (BMF) to identify the population of foreign businesses that potentially failed to file a Form 1120-F.¹⁷

• How to determine whether these foreign corporations were required to file a Form 1120-F. Using information from the foreign companies (via soft letter responses), the LB&I Division would attempt to determine whether they were required to file. Campaign management acknowledge that it is difficult to obtain information with correspondence because there is no incentive for taxpayers to comply. However, not responding may lead to an examination.

Although the Form 1120-F Non-Filer Campaign is ongoing, we focused on the initial results based on the first three data sources analyzed: U.S. State Sales Tax/Business Registrations data, U.S. Customs data, and IRS Form 1042-S data. As of February 13, 2019, the campaign had built 220 cases using these datasets. Figure 1 shows a breakdown of the cases built by data source, including the number of soft letters issued by the campaign and the responses received as of April 19, 2019.

Figure 1: Campaign Activity for Each Data Source

Campaign Action	Customs Data	Business Registrations and Sales Tax Data	Form 1042-S Data	Total
Number of Cases Built	60	80	80	220
Number of Soft Letters Issued	15	20	15	50
Number of Responses	10	12	9	31
Number of Nonresponsive Entities	5	8	6	19
Source: Treasury Inspector General for Tax Administration (TIGTA) analysis of campaign status reports.

Overall, we found that:

• The campaign was initiated without establishing evidence of a significant compliance risk pertaining to Form 1120-F nonfilers.

• A coordinated strategy was not established to ensure that critical campaign case actions were performed timely.

The Form 1120-F Non-Filer Campaign Was Initiated Without Establishing Evidence of a Significant Compliance Risk

In their presentation to the Compliance Integration Council seeking approval for this campaign, the Practice Area cited data sources identified in a CIP, indicating a potentially large Form 1120-F nonfiler population. The CIP resulted in the issuance of five soft letters built out of U.S. Customs data and ultimately transitioned into the campaign under review. * * * 1 * * * entities responded to the soft letters and were all referred to the Examination function. However, CIP results did not provide specific evidence that U.S. Customs data contained a potentially large Form 1120-F nonfiler population. CIP results did identify a general misunderstanding of filing requirements, forms, instructions, etc. through their correspondence with affected taxpayers and/or their representatives. The CIP also found that many foreign corporations filed protective returns year after year reflecting no taxable income earned in the United States. By using protective returns, these companies are considered to have timely filed their returns in the event that the IRS at some later point determines that the taxpayer did earn taxable income in the United States. The CIP did not determine the extent of any noncompliance or recommend the initiation of a campaign.

The Form 1120-F Non-Filer Campaign was one of the first 13 campaigns rolled out because it was considered among the most “shovel-ready.”¹⁸ More specifically, the CIP was thought to be easily convertible into a campaign. In June 2018, the IRS developed a tool for scoring campaigns, and this campaign ranked *2* out of 37 campaigns based on the scoring tool's factor.¹⁹

• Business value — Overall compliance benefit and performance against LB&I Division strategic compliance objectives (ranked *2* out of 37).

• Feasibility — Minimal external obstacles or limitations to campaign implementation or can be mitigated by the LB&I Division (ranked *2* out of 37).

• Ease of implementation — Resources available to implement critical campaign elements (ranked *2* out of 37).²⁰

While discussing our results with campaign management, management stated that they do not use the scoring tool to prioritize the order of how campaigns are presented for approval or the order in which they are implemented. The intent was to provide leadership with insight into how one campaign idea being presented for approval compared against other previously approved and rejected campaign ideas. These factors consider limited compliance data. In a prior report, TIGTA noted that of the approved campaigns, nine of the 10 lowest priority campaigns were among the first announced.²¹

LB&I Division officials did not assess whether existing compliance results supported a Form 1120-F campaign

In our prior audit, TIGTA reported that the IRS did not select issues or prioritize campaigns based on past or potential impact on compliance results. The LB&I Division had planned to use the new campaign approach to create specific objectives focused on improving compliance across the population of taxpayers for whom it is responsible. Accordingly, campaigns should address the greatest risks to compliance and choose work based on them. However, in reviewing the Form 1120-F Non-Filer Campaign, we found that compliance risk was not fully evaluated based on existing compliance data prior to campaign initiation. Furthermore, the underlying campaign documents discussed the difficulty in identifying Form 1120-F nonfilers and quantifying the Form 1120-F nonfiler population.

As previously noted, the precursor to this campaign was the Form 1120-F CIP which used information from U.S. Customs data to issue soft letters to five foreign entities.²² However, soft letter responses and subsequent examination results associated with the CIP ultimately contradicted claims of compliance risk made to support the subsequent campaign. * * * 1 * * * cases referred to the Examination function were closed no change. Even though CIP results did not support a significant compliance risk with nonfiling Forms 1120-F, this campaign was rolled out to specifically address what campaign documents stated was a “potentially large” Form 1120-F nonfiler population.

Overall, in our review of CIP and campaign documents, we found several indications that the compliance risk for Form 1120-F is unknown. Campaign staff could not provide quantifiable evidence or documentation based on past compliance results to corroborate an elevated compliance risk due to 1) examination experience with delinquent voluntary filings and field examinations of domestic entities that uncovered nonfiling parent corporations and 2) data sources (U.S. Customs data, Business Registrations data, Sales Tax information, etc.) indicating a “potentially large nonfiler population.” While discussing our results with the campaign staff, they agreed that only limited documentation of compliance results was considered to support a large Form 1120-F nonfiler population.

However, existing compliance results did show low audit coverage rates of Form 1120-F examinations despite large additional assessments from these examinations. Figure 2 shows the audit coverage rate of Forms 1120-F filed during CYs 2014 through 2016.²³ The audit coverage rate for these years never rose above 2 percent, but the average assessment was more than $2 million for CY 2015 and nearly $690,000 for CY 2014.²⁴

Figure 2: Form 1120-F Examination Data for Fiscal Years (FY) 2015 Through 2017

FY	Forms 1120-F Filed in the Previous Calendar Year	Forms 1120-F Examined in the Fiscal Year	Audit Coverage Rate in the Fiscal Year	Total Additional Assessments in the Fiscal Year	Average Assessment Per Form 1120-F Examined in the Fiscal Year
2015	42,514	857	2.0%	$590,926,000	$689,529
2016	44,601	503	1.1%	$1,024,043,000	$2,035,871
2017	47,992	656	1.4%	$80,228,000	$122,299
Source: IRS Data Book.25

The IRS also calculates the no-change rate that represents the percentage of returns examined resulting in no additional assessment. The no-change rate was 52 percent for FFY) 2016 and 46 percent for FY 2015.²⁶ Although the data could indicate a possible underreporting problem, the data do not indicate a Form 1120-F nonfiling issue. In fact, Figure 3 shows the filing trend in the number of Forms 1120-F has been increasing, although there may be factors other than increased tax compliance that are driving the increased filings, such as increased economic activity.

The number of Forms 1120-F filed has increased 13 percent in CY 2017 from CY 2014, which is an increase of more than 5,500 foreign entities. We believe that the LB&I Division did not consider some indicators that did not support a significant Form 1120-F nonfiler compliance risk such as 1) the CIP results' failure to support a significant compliance risk regarding Form 1120-F nonfiling and 2) the increasing trend in the number of Forms 1120-F filed.

Furthermore, as expanded upon later in this report, we determined that the external databases used by the campaign (U.S. Customs, Sales Tax/Business Registrations, etc.) have not yielded evidence supporting the LB&I Division's assertion of a potentially large Form 1120-F nonfiler population.

Because campaigns were meant to be compliance risk based, we believe that the quantitative results of prior compliance efforts should have been considered prior to campaign launch. Additionally, LB&I Division management should have continuously evaluated compliance risk and productivity associated with Form 1120-F work being delivered through this campaign. A preliminary issue-based assessment comparing potential enforcement results to the costs associated with the development of a campaign inventory would have allowed the LB&I Division to determine if there was actually an issue worth pursuing and/or focus its efforts on specific areas of Form 1120-F noncompliance. Because the campaign did not develop a profile of the Form 1120-F nonfiling population, it did not identify: 1) groups of foreign entities that may have compliance issues, 2) issues specific to industry type, and 3) filing and nonfiling trends to inform where outreach is most needed to target nonfiling taxpayers.

The IRS's examination resources have been reduced over the last five years.²⁷ IRS Examination function personnel have decreased 38 percent from 13,138 examiners in FY 2010 to 8,205 examiners FY 2017. The number of audits has also decreased by 31 percent from 1.6 million in FY 2013 to 1.1 million in FY 2017. Proposed assessments have declined over the last 10 years, from $44 billion in FY 2007 to $29 billion in FY 2017. Given the diminished examination resources, the IRS should be even more focused on emphasizing areas that have the highest compliance risk.

After we presented our results to the campaign staff, they collaborated with Campaign Development and Administration and developed alternative tests to estimate a possible Form 1120-F nonfiler population. This was accomplished by testing internal data in terms of nonfilers and stop filers.

• They analyzed Form SS-4, Application for Employer Identification Number, data to identify foreign entities that applied for a Taxpayer Identification Number (TIN).²⁸ Typically, the IRS assigns 98-Series TINs to foreign companies. They matched the 98-Series TINs for a three-year period to verify compliance. In this analysis, they identified 19 percent to be potential nonfilers.

• They also analyzed Form 1120-F filers on an entity-by-entity basis from Processing Year 2012 to Processing Year 2015 to identify those that stopped filing. In this analysis, they identified a potential non-filer compliance risk of six percent (per year).

Using these methods, the campaign staff believes that the test results warrant further investigation into potential nonfilers. Campaign staff agreed that this type of analysis should have been completed, considered, and addressed earlier in the campaign.

The campaign did not consider existing IRS nonfiler projects

The IRS uses third-party information reporting to identify nonfilers with earned income. Specifically, the IRS identifies nonfilers who receive income from an employer or business as reflected on information returns such as Form W-2, Wage and Tax Statement, and Form 1099-Misc, Miscellaneous Income. The income reported on such forms indicates whether the taxpayer should have filed a tax return. In addition to third-party reporting, the IRS identifies nonfilers by analyzing taxpayers' prior year filing history with current year data. For instance, the IRS attempts to identify nonfilers who filed in a prior year but failed to file a tax return for the current tax year.

The Nonfiler Inventory and Analysis group performs this analysis twice a year to identify and address nonfilers.²⁹ The analysis performed by the Nonfiler Inventory and Analysis group for individuals is referred to as the Individual Master File Case Creation Nonfiler Identification Process (CCNIP). Once the nonfilers are identified, the cases are categorized based on a number of characteristics, such as the amount of third-party reported income, withholding data, and estimated net tax due. After cases are categorized, they are assigned a selection code that is used to prioritize Individual Master File CCNIP inventory, i.e., the population of cases that show a potential for securing a tax return.

The BMF CCNIP does not work the same way as Individual Master File CCNIP. BMF nonfiler identification occurs weekly and is based on the scored results of the following factors:

• 16 weeks has lapsed beyond the return due date.

• The business has an open filing requirement for the return.

• There is no satisfying condition on the tax module.

• The tax period is available for selection (business was operational at the time the return was due)

The Form 1120-F Non-Filer Campaign did not consider using the BMF CCNIP during the planning of the campaign. Though the campaign was approved in November 2016, it was not until April 2017 that the LB&I Division contacted the IRS's Nonfiler Inventory and Analysis group to advise them of the Form 1120-F Non-Filer Campaign and the process to be used to identify Form 1120-F nonfilers. According to IRS officials, to have been able to use the BMF CCNIP nonfiler case creation process, the LB&I Division would need to develop a business case and make a formal request, and the IRS Information Technology organization would need to make certain changes to the BMF CCNIP program in order to identify Form 1120-F nonfilers. Changes are made periodically to the BMF CCNIP program, and while no formal analysis was performed to cost out such a change, IRS officials did not consider the change to identify Form 1120-F nonfilers to be significant.

The campaign reduced or eliminated some data from consideration

LB&I Division campaign staff used various filters to narrow the scope of the campaign. U.S. Customs data, in particular, reduced a large body of available data into smaller subsets without properly justifying the reduction criteria. Specifically, the campaign's data strategy did not include foreign entities with:

• * * * 2 * * *

• * * * 2 * * *

In addition, staff eliminated from consideration entities that filed Form 1120-F protective returns in all three initial databases. They also faced obstacles in processing the State Sales Tax/Business Registrations data, which we describe in more detail further in the report. Ultimately, 220 cases were built from the remaining data, and 50 soft letters were issued to the selected entities. Taxpayers who responded to the soft letter inquiries did so several ways. Figure 4 details how entities responded.

Figure 4: Number of Campaign Cases, Soft Letters, and Responses

Case Item/Action	U.S. Customs Data	Sales Tax/Business Registrations Data	Form 1042-S Data	Totals
Number of cases built	60	80	80	220
Soft Letters Sent	15	20	15	50
No response received	5	8	6	19
Response received	10	12	9	31
Response rate	67%	60%	60%	62%
Type of Response
Letter from foreign entity	5	5	4	14
Telephone call from foreign entity	0	2	3	5
Protective return from foreign entity	4	4	1	9
Foreign entity filed a Form 1120-F	1	1	1	3
Source: TIGTA analysis of campaign status reports.

Of the 31 responses received, 15 resulted from the initial soft letter and were received within 112 calendar days of the soft letter mail date. The remaining 16 responses came after a second and in some cases a third letter. As of April 19, 2019, the campaign referred 10 of the19 no-response cases and 12 of the 31 response cases to the Examination function. Figure 5 shows the source for the 22 cases referred to the Examination function.

Figure 5: Number of Soft Letters and Referrals to the Examination Function

Case Item	Examination Referrals — Customs Data	Examination Referrals — Sales Tax Data	Examination Referrals — Form 1042-S Data	Examination Referral Totals
No response received	5	2	3	10
Response received	4	3	5	12
Source of Examination Referral
No response received	5	2	3	10
Letter from foreign entity	2	1	0	3
Telephone call from foreign entity	0	0	3	3
Foreign entity filed a protective return	2	2	1	5
Foreign entity filed a Form 1120-F	0	0	1	1
Total Examination Referrals	9	5	8	22
Source: TIGTA analysis of campaign status reports.

U.S. Customs database: The U.S. Customs database initially contained records of 169,172 importers with a total import value exceeding $1.6 trillion in CY 2015.³⁰ However, the campaign * * * 2 * * *.

* * * 2 * * *.³¹ * * * 2 * * * When we inquired about this issue, LB&I Division officials stated that this information was excluded due to limited resources and the complexity involved to verify such a large number of entities. However, they stated that they may expand into this area later.

U.S. State Sales Tax database: There are 28 States that maintain formal sales tax information sharing agreements with the IRS. The IRS obtained business registrations data from 14 of the 28 States during the CIP and supplemented that information with the sales tax information that was already available through the sharing agreements. Using the State business registrations data, CIP personnel developed a list of foreign entities registered to conduct business in various States as of CY 2013.³² Formulating this list posed many challenges as each State provided different types of information. In some instances, CIP personnel had to manually format the business registrations lists provided to isolate the potential foreign businesses that had registered.

For the campaign, the IRS matched newly acquired sales tax data in CY 2015 against the CIP's CY 2013 business registrations data. In doing so, the IRS eliminated any foreign entities that registered in a State subsequent to CY 2013 (CYs 2014 and 2015). The campaign used the resulting data to create a list of potential nonfilers (composed of active, foreign corporations). This list was provided to the Large Business International Planning, Analysis, Inventory Research function to perform the following:

• Conduct * * * 2 * * *

• * * * 2 * * *

LB&I Division employees working on the campaign believed this information represented a list of potential nonfiling foreign corporations that were registered to do business in the United States. They also believed that these entities would be more likely to have ECI and a potential filing requirement. However, to date, the database has failed to yield effective results, with 40 percent of foreign corporations failing to respond to IRS notices. * * * 1 * * * cases in which the foreign businesses were nonresponsive were eventually classified as “no further action” (NFA) due to failure to respond, low examination potential, company dissolution, deliverability, or low materiality. IRS campaign management has since determined that researching and building cases from the State Sales Tax/Business Registrations database was too long and arduous, concluding that it was a resource intensive endeavor with low returns.

Consequently, campaign management has stated that, based on their initial analysis, they will no longer be using the State Sales Tax/Business Registrations database to identify potential Form 1120-F nonfilers in the future.³³

Form 1042-S Database: There were a total of 5,451,475 unique Form 1042-S filings in the Form 1042-S database for Tax Year 2015. The campaign team applied the following criteria for the Form 1042-S data to identify a population of potential Form 1120-F nonfilers:

• * * * 2 * * *

• * * * 2 * * *

• * * * 2 * * *

Using these criteria, the campaign isolated and sampled 282 taxpayers. When asked about the * * * 2 * * *

Form 1120-F Protective Returns: Foreign entities that file protective returns mitigate their risk of penalties and losing various credits and deductions if the IRS later deems their income effectively connected and therefore taxable. The protective return does not need to report any gross income or provide support for its tax calculation. Instead, filers must include a statement that the return is being filed to preserve the foreign corporation's right to deductions and credits and check the box on page 1 of the return indicating that they are filing a protective return.

* * * 2 * * * Given that campaign personnel suggested that the campaign is exploratory in nature, we believe that protective returns filers should also have been reviewed for the purposes of evaluating their compliance risk during the case building stage. If underreporting had been identified within the protective return filing population in this campaign, more work could have been initiated to address the elevated compliance risk.

Campaign cases are not yielding anticipated results

Campaign results from the three databases we reviewed have yet to show that the LB&I Division has effectively identified Form 1120-F nonfilers. Overall, we determined that the campaign has been unable to support the premise of a large Form 1120-F nonfiler population. Initial campaign case results show low examination referral and proposed assessment rates thus far. More specifically, LB&I Division campaign personnel issued:

• 15 soft letters on 60 U.S. Customs data cases and received 10 responses. Of the 10 responses, four were referred for examination. The five that did not respond were all referred for examination as well. This resulted in nine total examination referrals out of the database (15 percent examination rate from the database).

• 20 soft letters on 80 Sales Tax/Business Registrations data cases and received12 responses. Of the 12 responses, three were referred for examination. * * * 1 * * * the eight no-response entities were also referred for examination. This resulted in five total examination referrals (6 percent examination rate from the database).

• 15 soft letters on 80 Form 1042-S data cases and received nine responses. Of the nine responses, five were referred for examination. Three of the six no-response entities were also referred for examination. This resulted in eight total exam referrals (10 percent examination rate from of the database).

As previously stated, 22 examination referrals were made. Of these referrals, four examinations have been completed without any assessments as of March 2019. Figure 6 captures some of the examination referral details.

Figure 6: Examination Referrals and Assessments

Campaign Action	U.S. Customs Data	Sales Tax/Business Registrations Data	Form 1042-S Data	Total
Referral to the Examination Function	9	5	8	22
Unassigned	3	2	2	7
Assigned — Open	5	0	6	11
Assigned — Closed	1	3	0	4
Closed With Additional Assessment	0	0	0	0
Source: TIGTA analysis of campaign status reports.

The campaign's current strategy does not ensure that the campaign is appropriately reviewing campaign results to determine whether this is an area that requires the specific focus of an IRS initiative.

Recommendation

Recommendation 1: The Commissioner, Large Business and International Division, should assess the success of the campaign based on the 22 examinations referrals from the first 50 soft letters issued and consider whether limited examination resources warrant campaign continuation or whether a significant change in the methods to identify nonfilers is warranted.

A Coordinated Strategy Was Not Established to Ensure That Critical Campaign Case Actions Were Performed Timely

We reviewed the first 50 campaign cases that resulted in soft letters. These cases were developed using data from U.S. State Sales Tax/Business Registrations, U.S. Customs, and IRS Forms 1042-S. Our review showed:

• Delays in mailing correspondence.

• Delays initiating follow-up actions.

• Lack of documentation.

Delays in mailing correspondence

When IRS correspondence requests a taxpayer response within a set period of time, it is essential that the IRS provide the taxpayer the entire time allotted. To ensure that taxpayers are provided the full window of response, the IRS generally dates correspondence in advance of the anticipated mail date or mails the correspondence immediately after it is printed. If the letters are not mailed out in a timely manner, the recipient must respond to the notice within a truncated window.

For the purposes of this campaign, when taxpayers responded to soft letters, the campaign staff would evaluate the response and make a decision about whether to refer the case to Field Examination or not. If taxpayers failed to respond, the campaign staff would reissue the first letter. Taxpayers that remained nonresponsive after the second letter were sent a third letter containing language that was more aggressive. When taxpayers failed to respond to the more aggressive letter, the campaign staff would make a decision about whether to pass with a NFA or refer the nonresponsive taxpayer to Field Examination. A response does not necessarily yield an examination referral — instead, it is evaluated to determine whether an examination should be pursued or not based on how the taxpayer explained their tax position.

In our case review, we noted untimely case actions and follow-ups, as well as case procedures and determinations that were not documented. In many instances, entities received a soft letter stating they must respond within 45 calendar days after much of that time had already elapsed based on the letter date. Most likely this scenario would cause confusion about the expected response due date. Figure 7 is an example of an actual letter that was mailed on November 28, 2017, despite showing a letter date of October 30, 2017.

Nine case files pertaining to the first 50 soft letters sent did not have sufficient documentation of the correspondence between campaign personnel and the foreign entities. Of the remaining 41 cases, 31 were mailed six or more calendar days after the printed letter date, including * * * 1 * * *. Figure 8 breaks down our review of the time between the printed date on the letterhead and when it was actually sent (the fourth letter in Figure 8 is the NFA letter sent to notify the foreign entities that the IRS's case review was complete).

Figure 8: Time Between Letter Date and Mail Date

Mail Period (Calendar Days)	1st Letter	2nd Letter	3rd Letter	4th Letter (NFA)
Total letters sent	50	25	21	20
Mailed after the letter date timeframes
Mailed timely (less than five calendar days)	10	1	2	2
Mailed six through 10 calendar days late	17	1	15	0
11 through 20 calendar days	9	0	2	0
21 through 30 calendar days	1	0	0	0
31 through 40 calendar days	2	3	0	0
More than 60 calendar days	2	20	2	14
Unable to determine due to lack of mail or letter date	9	0	0	4
Source: IRS Nonfiler Classification logs circa 11/7/2017-11/7/2018.

Delays initiating follow-up actions

Timely follow up to taxpayer correspondence is critical to prevent taxpayers from delaying potential enforcement action and keeps a case moving toward resolution. This campaign's initial soft letter required foreign entities to respond within 45 calendar days and warned the recipients of a potential examination if they failed to comply:

However, campaign personnel did not follow up with 35 (70 percent) of the 50 entities that failed to respond within the 45-calendar day window for at least three weeks after the LB&I Division-imposed deadline. In the early stages of the campaign, team members followed up with those who did not respond to the initial soft letter by simply resending them a copy of the initial soft letter. Once a dedicated follow-up letter with more aggressive language was developed and approved, campaign team members began using it to pursue nonresponsive taxpayers (going forward this follow-up letter was used instead of resending a copy of the initial soft letter).³⁴ If a recipient remained nonresponsive after being sent follow-up letters, the campaign team would make a decision about whether to make an examination referral.

Of the entities that received the initial soft letters, 15 entities responded within 46 calendar days. Of those responses:

• Four responded with a telephone call.

• Four responded by sending a protective tax return.

• Six responded with a letter.

• One responded with a copy of a previously filed protective return.

Of the remaining 35 entities that did not respond within the allotted time, 22 received a follow-up soft letter after 67 or more calendar days had passed from the date of the first soft letter issuance.³⁵ For many cases, the gaps in correspondence between the initial soft letters and the final resolution reached the 400-through 500-calendar day range.

• 10 no-response cases that were referred for examination averaged 458 calendar days before such a decision was made.

• 9 no-response cases that resulted in NFAs averaged 457 calendar days before the IRS issued a NFA letter conveying the IRS's case review completion.

TIGTA noted that a potential contributing factor for delays in letter issuance is the requirement that all correspondence must pass through a minimum of three different levels of review prior to being mailed. Furthermore, there was no mechanism in place to ensure that whomever was responsible for approving or mailing the letter had actually done so in a timely manner. LB&I Division management stated that it generally should take two weeks to mail a letter out once a determination to do so has been made. We believe that this seems extensive, and the campaign should look for ways to streamline this process. Below are some of the more egregious examples of cases in which we determined critical follow-up actions were not timely completed:

• * * * 1 * * *

• * * * 1 * * *

• * * * 1 * * *

• * * * 1 * * *

These delays were driven primarily by the fact that the IRS did not consider nonresponsive taxpayers and therefore did not develop a contingency plan for them. Furthermore, classification duties appeared to have changed during the course of the campaign initially beginning with Campaign Development and Administration and then transitioning through three separate people inside the Practice Area. The campaign team also failed to develop a comprehensive timeline covering significant aspects of the campaign.

Due to the delays, taxpayers may have been uncertain about whether they needed to respond to the letters, especially within the context of the ongoing IRS impersonation scams. The requests also increased taxpayer burden on entities that were appropriately complying with Form 1120-F filing requirements. Most importantly, the IRS risks losing credibility with taxpayers and their representatives when it does not act as it advised in its correspondence after deadlines are missed.

Management Actions: In response to our review, the campaign plans to implement the following changes:³⁶

• Stop reissuing a duplicate of the first soft letter as a follow-up when entities failed to respond initially.

• Develop a separate NFA letter without the “thank you for responding” section.

• Reduce the number of personnel involved in approving letters to streamline the process and reduce the time it takes to mail the letter.

• Implement a tracking function within the associated classification logs that documents and monitors approval for mail and required follow-up dates.

The LB&I Division did not establish plans to address nonresponsive taxpayers

In planning, campaign management did not consider the possibility of a high no-response rate. As a result, they did not develop a follow-up letter or establish procedures to address cases in which the foreign entities did not respond to the initial soft letters. As previously noted, the campaign initially addressed this issue by simply sending the taxpayer a photocopy of the first soft letter with a “Second Request” stamp near the letterhead. Regardless, between November 2017 and November 2018, the no-response rate for the initial soft letter was 49 percent as one-half of the foreign entities did not respond.

It took the campaign nearly 20 months from the date of the launch to develop and send the first follow-up letters created specifically to address nonresponsive taxpayers. These follow-up letters contained stronger language to prompt a response from nonresponsive entities. Because the campaign did not establish plans to address nonresponsive cases, many nonresponsive taxpayers remained unaddressed by the campaign for well over a year. One factor contributing to this delay was a program manager's failure to review and approve the follow-up letter's May 2018 release. It took the campaign nearly three months to discover this error before eventually mailing these letters in late August 2018. The campaign has stated that competing priorities contributed to the delay. If the IRS had properly planned for nonresponsive taxpayers, it would have been able to timely address nonresponsive cases.

Another part of the campaign's notification process required the development and deployment of a NFA letter to inform entities that they had satisfied the IRS inquiry into their filing requirements and no enforcement actions would be pursued. The letter read in part:

Thank you for your response to our previous letter about Form 1120-F, U.S. Income Tax Return of a Foreign Corporation. Your response satisfied the questions we had about filing Form 1120-F for tax year 20xx. You don't need to take any action at this time.

Because the NFA letter used standardized language, in some instances, the campaign essentially thanked nonresponsive entities for their cooperation despite the fact that there was no correspondence.

Management Actions: In response to our review, the campaign plans to implement the following changes:

• The campaign analyst must also document the reason for soft letter determination and sign off on the cases once the work is completed.

• The manager must sign off on the case once the review is completed.

• The campaign should also standardize what resources must be used in researching an entity to ensure consistency between analysts when case building.

Recommendation

Recommendation 2: If the IRS is going to continue with the Form 1120-F Non-Filer Campaign, the Commissioner, Large Business and International Division, should establish controls over case building, case selection, and soft letter issuance processes to ensure consistency, documentation, and timely case actions.

---

Appendix I
Detailed Objective, Scope, and Methodology

Our overall objective was to assess how effectively the LB&I Division is conducting the Form 1120-F Non-Filer Campaign. To accomplish our objective, we:

I. Determined whether the IRS took the necessary actions to properly implement the Form 1120-F Non-Filer Campaign.

II. Assessed the effectiveness of the case development process.

III. Assessed how the IRS evaluates the success of the campaign.

Internal controls methodology

Internal controls relate to management's plans, methods, and procedures used to meet their mission, goals, and objectives. Internal controls include the processes and procedures for planning, organizing, directing, and controlling program operations. They include the systems for measuring, reporting, and monitoring program performance. We determined that the following internal controls were relevant to our audit objective: IRS procedures for implementing campaigns, processing soft letter responses, and taking case actions. We accomplished this by interviewing IRS personnel, reviewing key documentation related to the campaign, and reviewing campaign cases.

---

Appendix II
Major Contributors to This Report

Matthew A. Weir, Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Christina M. Dreyer, Director

Timothy F. Greiner, Audit Manager

Shalin Basnayake, Lead Auditor

Charles S. Nall, Senior Auditor

---

Appendix III
Report Distribution List

Deputy Commissioner for Services and Enforcement

Chief Counsel

Chief Financial Officer

Chief Information Officer

Commissioner, Wage and Investment Division

Deputy Commissioner, Large Business and International Division

Director, Enterprise Audit Management

---

Appendix IV

---

Appendix V

---

Appendix VI
Management's Response to the Draft Report

August 23, 2019

MEMORANDUM FOR
MICHAEL E. MCKENNEY
DEPUTY INSPECTOR GENERAL FOR AUDIT

FROM:
Douglas W. O'Donnell
Commissioner, Large Business and International Division

SUBJECT:
Draft Audit Report # 2018-30-020, The Internal Revenue Service Should Revise the Approach Used to Identify Foreign Corporations That Do Not File Required U.S. Income Tax Returns

Thank you for the opportunity to respond to the above referenced draft report which reviewed our campaign and strategy for ensuring filing compliance by foreign entities involved in a U.S. Trade or Business (USTB) or Permanent Establishment. It is the position of the IRS that nonfiling is a problem that contributes to the tax gap and undermines tax compliance. This campaign was approved using a phased approach to evaluate the efficacy of government data, both internal and external to the IRS, in identifying potential Form 1120-F non-filers.

The campaign performed analysis of transactional data and classification by experienced examiners to identify potential Form 1120-F non-filers with substantial U.S. business activities. Filters were developed based on a set of risk factors, and revenue agents and specialists with substantial inbound tax experience were involved in the classification process. These procedures resulted in the classification of 220 cases and the selection of the initial 50 potential nonfilers to receive soft letters.

As the report highlights, the initial results show that 62 percent of the foreign based companies sampled understood the content of the soft letter. In other words, the response rate provides adequate support that soft letters are an effective means to initiate contact with a foreign entity and attempt to obtain information regarding their potential USTB activities. This effort aligns with the IRS goal of enforcing the law and helping taxpayers fully understand and meet their tax obligations. The campaign is also conducting further analysis of entities that did not respond to determine the cause.

The IRS is committed to improving voluntary compliance, reducing the international tax gap and helping nonresident businesses understand their obligations under the law to include the timely filing of U.S. tax returns. Your report notes that we had not established evidence of a significant compliance risk with respect to 1120-F nonfilers at the time the campaign was initiated (and acknowledges the data challenges in accurately quantifying this universe of nonfilers). As discussed during the audit, the IRS believes there is evidence of compliance risk, though it is insufficient to statistically prove the magnitude of the risk. It is also important to note that this campaign allocated a relatively small amount of resources and the majority of our resources and most of our existing compliance programs are focused on addressing the areas of highest compliance risk. We do not intend to limit the purview of Campaigns to focusing solely on the highest compliance risk because it is important for us to broadly allocate our compliance resources to have a positive impact across all areas of noncompliance.

While we disagree with some of the factual assertions made in the report with respect to the campaign process and other material facts, we agree with TIGTA on some of the challenges identified and agree with your recommendations. The Attachment provide details on our corrective actions responding to TIGTA's specific recommendations.

If you have any questions, please contact me or members of your staff may contact John E. Hinding, Director, Cross Border Activities Compliance Practice Area at (202) 317-8686.

Attachment

---

Attachment I

RECOMMENDATION 1:

The Commissioner, Large Business and International Division, should assess the success of the campaign based on the 22 examinations referrals from the first 50 soft letters issued and consider whether limited examination resources warrant campaign continuation or whether a significant change in the methods to identify non-filers is warranted.

CORRECTIVE ACTIONS:

We agree with the recommendation. The IRS will assess the success of the Form 1120-F Non-Filer Campaign based on the 22 examinations referrals from the first 50 soft letters issued and consider whether limited examination resources warrant campaign continuation or whether a significant change in the methods to identify non-filers is warranted.

IMPLEMENTATION DATE:

December 15, 2019

RESPONSIBLE OFFICIAL(S):

Director, Cross Border Activities Compliance Practice Area, Large Business and International Division

CORRECTIVE ACTION(S) MONITORING PLAN:

We will monitor this corrective action as part of our internal management system of controls.

RECOMMENDATION 2:

If the IRS is going to continue with the Form 1120-F Non-Filer Campaign, the Commissioner, Large Business and International Division, should establish controls over case building, case selection, and soft letter issuance processes to ensure consistency, documentation, and timely case actions.

CORRECTIVE ACTIONS:

We agree with the recommendation and as a result of this audit we have done the following:

• Stopped reissuing a duplicate of the first soft letter as a follow-up when entities failed to respond initially.

• Developed a separate No Further Action letter.

• Reduced the number of personnel involved in approving letters to streamline the process and reduced the time it takes to mail the letter.

• Implemented a tracking function within the associated classification logs that documents and monitors approval for mail and required follow-up dates.

• Established a process in the classification log for the classifier to document the reason for soft letter determination and sign off on the cases once the work is completed.

• Established a process in the classification log for the manager to sign off on the case once the review is completed.

• Established a process in the classification log to standardize what resources must be used in researching an entity to ensure consistency between analysts when case building.

IMPLEMENTATION DATE:

Implemented

RESPONSIBLE OFFICIAL(S):

Director, Cross Border Activities Compliance Practice Area, Large Business and International Division

CORRECTIVE ACTION(S) MONITORING PLAN:

We will monitor this corrective action as part of our internal management system of controls.