Menu
Tax Notes logo

Marijuana Industry Needs More Tax Compliance, TIGTA Finds

MAR. 30, 2020

2020-30-017

DATED MAR. 30, 2020
DOCUMENT ATTRIBUTES
Citations: 2020-30-017

The Growth of the Marijuana Industry Warrants Increased Tax Compliance Efforts and Additional Guidance

March 30, 2020

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:

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

4 = Attorney Client/ Attorney Work Product.

Highlights

Final Report issued on March 30, 2020

Highlights of Reference Number: 2020-30-017 to the Commissioner of Internal Revenue.

IMPACT ON TAXPAYERS

Marijuana is classified as a Schedule I controlled substance under the Controlled Substances Act. Businesses in this industry have limited banking access and are subject to Internal Revenue Code (I.R.C.) Section (§) 280E, which prohibits the deduction of expenses incurred in trafficking Schedule I controlled substances. The IRS risks diminished taxpayer compliance when marijuana businesses fail to report all income as required under I.R.C § 61, regardless of source, and deduct expenses not allowed under I.R.C. § 280E.

WHY TIGTA DID THE AUDIT

This audit was initiated to evaluate the IRS's examination and education approach to certain cash-based industries with an emphasis on legal marijuana operations.

WHAT TIGTA FOUND

TIGTA reviewed statistical random samples of marijuana businesses in three States and determined that 59 percent (140 out of 237) of the tax return filings for Tax Year 2016 had likely I.R.C. § 280E adjustments, which when projected over the population totaled $48.5 million in unassessed taxes for Tax Year 2016 or $242.6 million when the results are forecasted over five years.

TIGTA also estimated the tax impact to comply with I.R.C. § 280E for the same sampled marijuana business taxpayers. When projected to the population, TIGTA estimated a $95 million Federal income tax impact to these taxpayers from the application of I.R.C. § 280E on their Tax Year 2016, or $475.1 million when forecasted over five years.

In addition, TIGTA selected a statistically random sample of 90 marijuana businesses that filed State returns for Tax Year 2016 in the State of Washington to determine whether these taxpayers were reporting all of their income in compliance with I.R.C. § 61. TIGTA found that 23 (26 percent) of 90 returns likely have I.R.C. §61 adjustments involving either underreported income or nonfiling of tax returns. When projected over the population for Washington, the IRS missed the opportunity to address $3.9 million of potential assessments for Tax Year 2016, or $19.3 million when forecasted over five years.

Also, the IRS lacks guidance to taxpayers and tax professionals in the marijuana industry. Such guidance would improve awareness of tax filing requirements for taxpayers in this industry, such as the correct application of I.R.C. §§ 280E and 471(c), which would reduce the burden of tracking inventory for certain small businesses.

WHAT TIGTA RECOMMENDED

TIGTA recommended that the IRS develop a comprehensive compliance approach for the marijuana industry, including a method to identify businesses in this industry and track examination results; develop and publicize guidance specific to the marijuana industry, such as guidance on the application of I.R.C. § 471(c) in conjunction with I.R.C. § 280E; leverage publically available information at the State level and expand the use of existing Fed/State agreements to identify nonfilers and unreported income in the marijuana industry; and increase educational outreach towards unbanked taxpayers making cash deposits regarding the unbanked relief policies available.

The IRS agreed with five of the six recommendations. The IRS did not agree with the recommendation to develop and provide guidance on I.R.C. § 471(c) citing other priorities. However, the IRS added that once the 2019-2020 Priority Guidance Plan is resolved, developing guidance to ensure coordination between I.R.C. §§ 280E and 471(c) will be considered.


March 30, 2020

MEMORANDUM FOR
COMMISSIONER OF INTERNAL REVENUE

FROM:
Michael E. McKenney
Deputy Inspector General for Audit

SUBJECT:
Final Audit Report — The Growth of the Marijuana Industry Warrants Increased Tax Compliance Efforts and Additional Guidance (Audit # 201830022)

This report presents the results of our review to evaluate the Internal Revenue Service's (IRS) examination and education approach to certain cash-based industries with an emphasis on legal marijuana operations. This audit is included in our Fiscal Year 2020 Annual Audit Plan and addresses the major management challenge of Improving Tax Reporting and Payment Compliance.

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

Copies of this report are also being sent to the IRS 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

High-Risk Marijuana Business Tax Returns With Millions of Dollars in Potential Tax Adjustments Are Not Worked

Recommendations 1 and 2:

The IRS Lacks Guidance Regarding Internal Revenue Code Section 280E and the Uncertain Impact From Internal Revenue Code Section 471(c) of the Tax Cuts and Jobs Act

Recommendations 3 and 4:

The IRS Is Not Using Available Marijuana State Tax Return Information to Identify Nonfilers and Underreported Income

Recommendation 5:

Some Unbanked Marijuana Businesses Are Assessed Federal Tax Deposit Penalties Based on Non-Electronic Funds Transfers

Recommendation 6:

Appendices

Appendix I — Detailed Objective, Scope, and Methodology

Appendix II — Major Contributors to This Report

Appendix III — Report Distribution List

Appendix IV — Outcome Measures

Appendix V — Tax Year 2016 Graduated Corporate Tax Rate Schedule

Appendix VI — Glossary of Terms

Appendix VII — Management's Response to the Draft Report

Abbreviations

ATG

Audit Technique Guide

BSA

Bank Secrecy Act

CIP

Compliance Initiative Project

DIF

Discriminant Function

FTD

Failure to Deposit

FY

Fiscal Year

IDRS

Integrated Data Retrieval System

I.R.C.

Internal Revenue Code

IRS

Internal Revenue Service

SAR

Suspicious Activity Report

SB/SE

Small Business/Self-Employed

TIGTA

Treasury Inspector General for Tax Administration


Background

The United States legal marijuana industry has been growing in recent years. Industry sources projected that the U.S. legal marijuana industry took in nearly $11 billion in sales in 2018 and are expected to rise to $13 billion in 2019 and $25 billion by 2025.1 The growth is a result of several States in the United States passing laws to legalize the use of marijuana for either medical and/or recreational purposes. Figure 1 summarizes the States in the United States that have legalized the use of marijuana for either medical and/or recreational purposes as of July 2019.2

State view of marijuana legal status

Figure 1 reflects that 10 States (Alaska, California, Colorado, Maine, Massachusetts, Michigan, Nevada, Oregon, Vermont, and Washington) and the District of Columbia allow for both medical and recreational use of marijuana. Another 23 States allow for medical marijuana use only. In summary, 33 (66 percent) of 50 States and the District of Columbia allow for either medical and/or recreational use of marijuana.3

While marijuana businesses may legally operate under State laws, they remain illegal under Federal law

Marijuana businesses may operate under one or more industry classifications such as medical marijuana, recreational marijuana, head shop, social lounge/smoking club, producer,4 cooperative, and distributor. While marijuana businesses may be allowed to legally operate under State law, they remain illegal under Federal law. Marijuana remains classified as a Schedule I controlled substance under the Controlled Substances Act.5

Under the Controlled Substances Act, it is illegal to manufacture or distribute marijuana. Further, Internal Revenue Code (I.R.C.) Section (§) 280E prohibits the deduction of expenses incurred in trafficking controlled substances.6 I.R.C. § 280E does not apply to the cost of goods sold. Consequently, businesses that sell marijuana can reduce gross receipts by the cost of goods sold but cannot deduct other business expenses.7 Figure 2 illustrates the Federal tax effect for a hypothetical marijuana business applying I.R.C. § 280E to the calculation of taxable income for Tax Year 2016.

Figure 2: Hypothetical Example of I.R.C. § 280E Tax Impact to a Marijuana Business for Tax Year 2016

Form 1120 Return*

Example

Gross Income

$500,000

Cost of Goods Sold

($250,000)

Business Expenses

($150,000)

Net Income

$100,000

I.R.C. § 280E Adjustment

$150,000

Taxable Income

$250,000

Graduated Corporate Tax8

$80,750

Source: TIGTA hypothetical example. *Form 1120, U.S. Corporation Income Tax Return.

As detailed in Figure 2, business expenses are not deductible under I.R.C. § 280E. In this example, the “I.R.C. § 280E Adjustment” disallows business expenses of $150,000 thereby increasing taxable income to $250,000, which results in a net corporate tax of $80,750.

However, the Internal Revenue Service (IRS) has provided little guidance associated with I.R.C. §280E other than Office of Chief Counsel Advice 201504011. No references to marijuana businesses can be found in IRS publications. Based on Office of Chief Counsel guidance, taxpayers subject to I.R.C. § 280E determine costs of goods sold pursuant to I.R.C. § 471. The Small Business/Self-Employed (SB/SE) Division has developed an internal document “Participant Guide” that provides revenue agents with guidelines on how to audit marijuana businesses. According to the IRS internal Participant Guide, pursuant to Treas. Reg. 1.471-11, cost of goods sold for producers includes direct material cost, e.g., marijuana seeds or plants; direct labor costs, e.g., planting, cultivating, harvesting, sorting; and indirect costs. Indirect costs may include repair expenses, maintenance, utilities, rent, indirect labor and production supervisory wages, indirect materials, tools, and cost of quality control. The IRS internal Participant Guide noted the cost of goods sold for retailers includes the costs of the marijuana purchased, less trade or other discounts, plus transportation in or other necessary charges in acquiring possession of the marijuana. According to marijuana industry training conducted by the IRS Large Business and International Division inventory issue practice area, cost of goods sold does not include:

  • Marketing, advertising, selling expenses, distribution, interest expense, and other costs as listed in Treas. Reg. 1.471-11(c)(2)(ii). If these costs are included as part of costs of goods sold, they would be disallowed as an “I.R.C. § 280E Adjustment” also increasing the taxable income and tax obligation of a business.

Three additional potential adjustments to cost of goods sold based on I.R.C. § 280E include I.R.C. § 263A (uniform capitalization rules), accelerated depreciation under I.R.C. §168(k), and depreciable assets expensed under I.R.C. § 179 as described below.

1) I.R.C. § 263A amounts reported as part of cost of goods sold for taxpayers subject to I.R.C. § 280E. I.R.C. § 263A was enacted four years after I.R.C. § 280E and expanded upon I.R.C. § 471 by providing guidance on the costs associated with producing products including rules for capitalizing indirect expenses. I.R.C. § 263A(a)(2) states: “Any cost which (but for this subsection) could not be taken into account in computing taxable income for any tax year shall not be treated as a cost described in this paragraph.” I.R.C. § 263A(a)(2) prevents marijuana businesses from obtaining a tax benefit by capitalizing disallowed deductions under I.R.C. § 280E.

2) Accelerated depreciation amounts for taxpayers subject to I.R.C. § 280E. According to Treas. Reg. § 1.471-11(c)(2)(iii)(b), the only depreciation that can be included in cost of goods sold for a taxpayer subject to I.R.C. § 280E would be depreciation deducted on a taxpayer's financial statements, or book depreciation, on assets “incident to and necessary for” marijuana production or manufacturing. Based on this limitation, depreciation for Federal tax purposes in excess of book depreciation would not typically be allowed based on I.R.C. § 280E, e.g., depreciation determined under I.R.C. § 168 in excess of book depreciation.

3) Depreciable assets expensed under I.R.C. § 179. I.R.C. § 179 allows an expense deduction for taxpayers that elect to treat the cost of qualifying property, referred to as I.R.C. § 179 property, as an expense rather than a capital expenditure. I.R.C. § 179 deductions result in capital purchases getting expensed. These I.R.C § 179 deductions may be subject to limitations under I.R.C § 280E.

States that have legalized marijuana typically apply a higher excise or sales tax

We reviewed State taxes for those States allowing legal recreational marijuana retail sales.9 We generally found that the States charged either a higher excise or sales tax than those applicable for sales made by non-marijuana industry businesses. For instance, the State of Washington appears to have one of the highest tax rates for marijuana recreational sales with a base tax rate of 43.5 percent (37 percent excise tax plus the State standard retail sales tax of 6.5 percent), which is significantly higher than its 6.5 percent State standard sales tax for non-marijuana industry businesses.10 Conversely, the State of Oregon has one of the lowest recreational marijuana tax rates for marijuana sales starting at 17 percent (retail sales tax), which is significant when considering that Oregon does not have a retail sales tax on non-marijuana industry sales.11 For these two States, the consumer pays the taxes on the purchase and the retailer remits the tax to the State; whereas in the State of California, the distributor collects 15 percent excise tax from the retailer for recreational sales.12

Limited access to banking services increases risk associated with cash only businesses and creates barriers with meeting Federal tax obligations

Marijuana businesses have limited access to banking because marijuana is classified as a Schedule I controlled substance, and banks and credit unions who service marijuana businesses can potentially be charged with money laundering. Many financial institutions are not willing to risk potential civil or criminal liability associated with their obligations under the Bank Secrecy Act (BSA).13

One of the main barriers for banks and credit unions is the information reporting requirements when providing banking services to marijuana businesses. For example, BSA regulations require the filing of a Suspicious Activity Report (SAR) when a financial institution knows, suspects, or has reason to suspect that a transaction of $5,000 or more involves funds derived from an illegal activity or is an attempt to disguise funds derived from an illegal activity.14 The SAR filing requirement is both costly and risky as the reporting of all transactions the financial institution has with the respective marijuana business can be extensive, and if the activity is incorrectly reported, fines to the financial institution could result.

Banks and credit unions that service marijuana businesses may charge large fees to compensate for the extensive reporting requirements and risk for providing services to these businesses. One credit union in California stated it was charging banking fees to marijuana businesses of up to $10,000 as an upfront fee and $5,000 a month for producers and $7,500 a month for dispensaries.15 Another small credit union in Oregon that serves marijuana businesses stated the credit union filed more than 13,500 individual reports over the past two years (2017 and 2018) for approximately 500 cannabis clients.16

We have also identified recent trends with banks and credit unions providing banking services to marijuana businesses. According to the U.S. Treasury Financial Crimes Enforcement Network, the number of financial institutions actively banking marijuana-related businesses increased from 401 in October 2017 to 715 in June 2019.17

However, the lack of banking access continues to be an issue in the marijuana industry with most banks or credit unions across the United States not willing to accept marijuana business customers. Marijuana businesses without bank account access are also unable to set up merchant accounts for accepting credit or debit cards. This results in most marijuana businesses conducting business transactions in cash only. Marijuana businesses may have automated teller machines on the premises for customers to facilitate cash only transactions.

The main tax-related concern about cash intensive businesses is that cash transactions are more difficult to track and are therefore more likely to go unreported to the IRS. Unlike checks and credit card receipts, cash transactions do not generally result in third-party information capable of being reported to the IRS. To the extent that Government laws and regulations discourage banking for marijuana businesses (and to the extent they encourage cash only transactions), they also may be indirectly and unintentionally encouraging tax noncompliance.

The lack of banking services also creates a barrier for paying Federal taxes. The IRS allows for cash payments at customer service locations. However, the availability and access to make cash payments is limited to specific locations, times, and is subject to IRS procedures that include scheduling an appointment in advance.

This lack of banking access also can be a barrier to collection for the IRS as the traditional methods of levying a bank account are unavailable. The IRS awarded a contract to MITRE starting in August, 2018 to review this issue and propose methods and locations for the cannabis industry to pay Federal taxes safely and conveniently.18

Federal attempts to mitigate conflict with State law

On August 29, 2013, the U.S. Deputy Attorney General issued a memorandum indicating that the U.S. Government would rely on State and local governments to address marijuana activity and would not prosecute individuals and companies complying with State laws (the Cole Memorandum).19 Following the issuance of the Cole Memorandum, the Financial Crimes Enforcement Network issued guidance intended to enhance the availability of financial services for marijuana businesses and mitigate risks of financial institutions violating the BSA by servicing such businesses.20 The guidance provides financial institutions with three standards upon which to issue a SAR.21

The Cole Memorandum was subsequently rescinded on January 4, 2018, by the then U.S. Attorney General. The current Federal impact of the rescission is unclear; however, the succeeding Attorney General has stated the Government would not pursue prosecutions of marijuana businesses who relied on the Cole Memorandum.22 However, the Department of Justice has not issued a formal memorandum addressing this issue since the January 2018 rescission.

There are also a number of proposed legislative initiatives in Congress, including the:

  • Secure and Fair Enforcement Banking Act of 2019.23

  • Strengthening the Tenth Amendment Through Entrusting States Act.24

This review was performed at the SB/SE Division Western Area (hereafter referred to as the Western Area) Headquarters in Denver, Colorado, and with information obtained from SB/SE Division Examination function executives during the period September 2018 through December 2019. There was an impairment to the scope of our audit work pertaining to the identification of underreporting of tax and failure to file tax returns. The IRS would not use its existing information sharing agreements with the respective State agencies of California and Oregon to request and to obtain State gross receipts and income information for determining whether this information assisted in the identification of cash based businesses that underreport or fail to file tax returns. The IRS said it had “no need or use” for the information. Further, officials from both States told us that their State law did not permit them to provide the data to us directly. As such, our analysis pertaining to underreporting and nonfiling is limited to Washington State, where we were able to obtain gross receipts information on licensed marijuana businesses directly from public information available on the State's website.

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

High-Risk Marijuana Business Tax Returns With Millions of Dollars in Potential Tax Adjustments Are Not Worked

There are unique tax compliance risks in the marijuana industry due to I.R.C. § 280E and cash intensive sales. The SB/SE Division identified the marijuana industry as a high-impact compliance area in examination program letters for both Fiscal Years (FY) 2018 and 2019.

The current IRS compliance approach for marijuana businesses is primarily concentrated in the Western Area.25 On January 4, 2016, the Western Area approved a Compliance Initiative Project (CIP) in the marijuana industry. The Part 1 CIP was limited to 50 marijuana businesses in Colorado and was terminated on June 30, 2017, without expansion into Part 2.26 According to Western Area management, the marijuana CIP was not continued as a Part 2 CIP based on resource constraints.27 The 50 marijuana business examinations of the terminated Western Area Part 1 CIP are still in progress with closures being reported in FYs 2018 and 2019.

The Western Area has been involved in preparing internal guidance materials and conducting training for revenue agents for both CIP and non-CIP examinations of marijuana businesses. The training has been conducted for the following locations: Denver, Seattle, Portland, Northern California, Nevada, Phoenix, and Detroit.

Each CIP has its own project and tracking codes; however, non-CIP marijuana examinations are only consistently tracked in the Western Area. The Western Area tracks non-CIP marijuana examinations by using Aging Reason Code 097 to monitor non-CIP marijuana examinations. TIGTA noted that all other SB/SE Division Examination function areas outside of the Western Area do not consistently use Aging Reason Codes to track non-CIP marijuana examinations.

Figures 3 and 4 compare the productivity of Western Area marijuana CIP return closures with Western Area marijuana non-CIP return closures and include the Discriminant Function (DIF) examination results for comparison.28

Figure 3: Western Area Marijuana Examination Results for FY 2018

Returns

Returns Closed

Dollars Per Return

Dollars Per Hour

No Change Rate

CIP-Marijuana

188

$82,831

$2,497

2.1%

Non-CIP Marijuana

122

$172,737

$3,375

0.8%

DIF Returns

5,715

$36,658

$967

19.3%

Source: Closed Case Monitoring Reports from the Western Area.

Figure 4: Western Area Marijuana Examination Results for FY 2019 Through June 30, 2019

Returns

Returns Closed

Dollars Per Return

Dollars Per Hour

No Change Rate

CIP-Marijuana

102

$148,084

$2,752

0.0%

Non-CIP Marijuana

128

$144,698

$3,598

6.3%

DIF Returns

3,559

$43,096

$1,065

17.3%

Source: Closed Case Monitoring Reports from the Western Area.

As reflected in Figures 3 and 4, marijuana examination results for both CIP and non-CIP marijuana returns generate a significantly higher dollars per return and dollars per hour rate of return than DIF examinations.29 Also, the no change rate for both CIP marijuana and non-CIP marijuana examinations is significantly lower than DIF examinations.

Other IRS marijuana compliance efforts include a current CIP specific to the city of Detroit in the Midwest Area, approved on December 31, 2017. Past IRS compliance efforts in the marijuana industry have included one CIP in the Southwest Area (California) and another CIP in the Western area (separate from the ones noted in Figures 3 and 4).

The SB/SE Division has designated an overall manager to provide direction and oversight for issues unique to the marijuana industry. This manager is identified as the National Champion and is located in Baltimore, Maryland. In addition, there are eight Area-level Champions who are Area-level managers assigned to provide oversight for examination issues in the marijuana industry for their respective examination area.30 The Western Area-level Champion is located in Denver, Colorado. The National Champion and Area-level Champions conduct meetings on a regular basis to discuss issues and share knowledge related to the marijuana industry.

Western Area management noted the following two primary compliance issues for marijuana returns: 1) business expenses incorrectly deducted based on I.R.C. § 280E and 2) unreported income. IRS management explained that most adjustments on marijuana examinations are based on the inaccurate application of I.R.C. § 280E.

According to Western Area management, the marijuana examination cycle time can be lengthy due to significant related returns picked up and delays based on representative tactics such as quashing summons. SB/SE Division management also noted that marijuana cases are often closed to Appeals as unagreed.

According to the IRS, there is no easy method to identify marijuana businesses based on tax return filing information. The IRS should explore alternatives, such as leveraging State information on marijuana-based businesses through existing information sharing agreements or available public information to develop and initiate a national CIP. The objective of the national CIP should focus on creating a comprehensive compliance approach to identify noncompliant taxpayers in this industry.

Marijuana businesses in California, Oregon, and Washington have a high rate of noncompliance with I.R.C. § 280E

Our review of three statistical random samples identified 140 (59 percent) of 237 marijuana businesses with a Tax Year 2016 filing requirement with likely I.R.C. § 280E adjustments.31 Specifically, 38 (78 percent) of the 49 returns for California, 49 (50 percent) of the 98 returns for Oregon, and 53 (59 percent) of the 90 returns for Washington likely had I.R.C § 280E adjustments. We calculated the likely tax assessment impact to these respective marijuana businesses from the exceptions in these States as follows: $5,060,011 in California, $1,147,124 in Oregon, and $1,091,551 in Washington.32 We also calculated the average I.R.C. § 280E assessment per return from these overall likely tax assessments to be $103,266 in California, $11,705 in Oregon, and $12,128 in Washington for Tax Year 2016.33 When we project the results of the exceptions to each State's population, 260 returns for California, 261 for Oregon, and 380 for Washington would have likely assessments for Tax Year 2016. The corresponding estimated total of likely assessments for the populations would be $34.6 million for California, $6.1 million for Oregon, and $7.8 million for Washington. Based on the results of our samples and adjusted populations, we estimate that the IRS could have issued $48.5 million in tax assessments for Tax Year 2016 in these three States or $242.6 million in these three States when the results are forecasted over five years.34 This forecast represents only a portion of the tax noncompliance related to I.R.C. § 280E in that it only includes three out of 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana and does not consider the growth in the industry since Tax Year 2016.

In order to analyze compliance risk with I.R.C. § 280E, TIGTA identified marijuana business lists from State websites for California, Oregon, and Washington. For each respective State's population of marijuana businesses, we selected a statistically valid random sample. Figure 5 lists active retail and producer business licenses listed on Oregon and Washington websites. For California, the business licenses represent active retail and distributor licenses listed on its website. The public business license lists TIGTA used for each State were obtained in 2018.

We conducted additional research using the online State license registration data for each State to determine whether the business license selected was active during Calendar Year 2016. If the business license was active during Calendar Year 2016, we used the Integrated Data Retrieval System (IDRS) to determine if the taxpayer filed a tax return for Tax Year 2016. We continued this process until we obtained the minimum required sample size for each State. Because some of the business licenses were not active or did not have a Federal return filing requirement for Tax Year 2016, we adjusted the population accordingly as reflected in Figure 5 for projection purposes.

Figure 5: TIGTA Sample Plan

Stratum

Information Date

Active Licenses

Minimum Sample Size

Adjusted Population35

California

9/10/2018

835

49

335

Oregon

8/20/2018

1,668

98

521

Washington

9/5/2018

1,533

90

645

Source: TIGTA-developed sample plan based on State license lists from California, Oregon, and Washington.

Potential adjustments were calculated by reviewing Tax Year 2016 returns and identifying deductions reported on the face of the return and cost of goods sold expenses such as I.R.C. § 263A expense and accelerated depreciation not allowed under I.R.C. § 280E.36 TIGTA's testing also included disallowing I.R.C § 179 expenses that may not be allowed under I.R.C. § 280E. The review did not include a detailed analysis of additional potential adjustments, such as those that could be identified only through a detailed review of supporting schedules and taxpayer records such as “other cost” included in cost of goods sold or issues with taxpayers potentially understating ending inventory.37

The average marijuana business size, based on gross revenues for our sample of active licenses reviewed, was $2,685,774 in California, $694,765 in Oregon, and $1,261,941 in Washington. The actual compliance risk could potentially be much larger, as this analysis does not include any other potential examination adjustments that may be found during an examination-level review. Further, as noted previously, 33 out of 50 States and the District of Columbia have passed legislation to legalize medical and/or recreational use of marijuana.

The IRS's current compliance approach in the marijuana industry does not include leveraging State data available to identify marijuana businesses for possible audit selection. In addition, there is no national CIP at this time for the marijuana industry. TIGTA noted the termination report dated July 2, 2018, for the Western Area CIP included a recommendation for a national marijuana CIP and an increase in taxpayer and preparer education. However, the IRS has not moved forward with a national CIP at this time. Without a national CIP to address this high-risk industry, the IRS will be unable to accurately determine the overall noncompliance risk associated with taxpayers within this industry let alone the accuracy of their reporting compliance with the limitations of I.R.C. § 280E.

Recommendations

The Commissioner, SB/SE Division, should:

Recommendation 1: Develop a comprehensive compliance approach, i.e., national CIP, for this industry and leverage State marijuana business lists to identify noncompliant taxpayers.

Management's Response: The IRS agreed with this recommendation but disagreed with the related outcome measure (see Appendix IV for more detailed information on the reported outcome measures). In its response, the IRS noted that whether it pursues taxpayers in the marijuana industry depends on priorities and available resources. The IRS stated it will use data analytics to identify the size and scope of noncompliant taxpayers and prioritize the compliance activities based on resources available.

Office of Audit Comment: As previously stated, the SB/SE Division identified the marijuana industry as a high-impact compliance area in examination program letters for both FYs 2018 and 2019. However, IRS management's response does not directly address the compliance risk in the marijuana industry. Therefore, as the IRS evaluates its resource allocation, it should take a comprehensive approach and prioritize high-impact compliance areas such as the marijuana industry.

Recommendation 2: Direct all examination areas to use Aging Reason Codes to track non-CIP marijuana business examination results.

Management's Response: The IRS agreed with this recommendation and will provide guidance to examiners to apply the appropriate code(s) to cases to track marijuana examinations.

The IRS Lacks Guidance Regarding Internal Revenue Code Section 280E and the Uncertain Impact From Internal Revenue Code Section 471(c) of the Tax Cuts and Jobs Act

While the IRS had conducted outreach at a variety of industry and tax professional venues, the IRS has not published nationwide guidance to taxpayers and tax professionals in the marijuana industry. IRS outreach has not included the use of soft letters to inform marijuana businesses of tax issues in this industry. Additional guidance in the industry is critical to improve the compliance rate with I.R.C. § 280E. As previously stated, the SB/SE Division has developed an internal document “Participant Guide” that provides revenue agents with guidelines on how to audit marijuana businesses. However, the Participant Guide has not been made public like other Audit Technique Guides (ATG) that provide insight and guidance for issues unique to a particular industry. One of the existing ATGs is for cash intensive businesses and was last updated in April 2010. However, the cash intensive businesses ATG does not address issues unique to the marijuana business such as I.R.C. § 280E and banking limitations.

IRS management stated that they do not plan to make the marijuana Participant Guide public, stating that there is too much difficulty in addressing the differences in State and local law and there is too much uncertainty given the potential for changes in Federal and State laws.

However, differences in State and local law do not impact the IRS's examination of Federal tax returns of businesses in the marijuana industry. The IRS should develop guidance specific to the marijuana industry and publish it on its IRS.gov website. The information should incorporate such topics as Frequently Asked Questions and other information that would improve awareness of tax filing requirements for taxpayers in this industry, such as the correct application of I.R.C. § 280E and 471(c).

As previously stated, the IRS has identified the marijuana industry as a high-impact compliance area. TIGTA also found a significant noncompliance rate for marijuana businesses with I.R.C. § 280E for California, Oregon, and Washington. As previously shown:

  • Figure 2 provides a basic example of the application of I.R.C. § 280E which illustrates that business expenses of marijuana businesses are not allowed under I.R.C. § 280E. Taxpayers subject to I.R.C. § 280E are also subject to I.R.C. § 471.

  • Figures 3 and 4 noted significantly higher assessments for marijuana examinations than DIF examinations. TIGTA's testing also noted an overall 59 percent noncompliance rate for marijuana businesses with I.R.C. § 280E for the three States tested.

It is important that the IRS have a multipronged strategy to increase taxpayers' and tax professionals' understanding of I.R.C. § 280E to improve compliance in this industry. This strategy should include both outreach and written guidance. The IRS should develop guidance specific to the marijuana industry such as a Frequently Asked Questions document, publicize existing guidance, i.e., Participant Guide, and conduct outreach in this industry to improve awareness of the application of I.R.C. § 280E. Absent a strategy to educate these taxpayers, compliance-related issues will continue to grow and negatively affect limited IRS resources.

The impact of I.R.C. § 471(c) of the Tax Cuts and Jobs Act38 is uncertain

The Tax Cuts and Jobs Act contained a new provision, I.R.C. § 471(c), which is applicable to taxable years after December 31, 2017, and will have the effect of reducing the burden for tracking inventory for small businesses with less than $25 million in gross receipts.39 These qualified businesses would not be subject to the general rule for determining inventory. Instead, they may elect to use internal financial statements or accounting procedures to account for costs in lieu of keeping inventories in the manner otherwise required by I.R.C. § 471(a).40

Under this new provision, marijuana businesses could argue they are entitled to use a method of accounting that includes all expenses in cost of goods sold to potentially avoid the impact of I.R.C. § 280E. According to IRS Chief Counsel, at least two practitioners have identified this issue and have questioned IRS personnel on how the IRS plans to handle I.R.C. § 471(c) as applied to marijuana industry taxpayers.

These practitioners have identified the potential unintended consequence of I.R.C. § 471(c) that appears to allow small marijuana businesses to include non-cost of goods sold expenses in their cost of goods sold and potentially avoid the application of I.R.C. § 280E. IRS Chief Counsel noted that practitioners assert that the new law may provide small business taxpayers wide latitude to characterize all expenditures as cost of goods sold. The effect of the law is still uncertain.

See the following hypothetical examples in Figure 6 that illustrate the potential impact of I.R.C. § 471(c) and requested IRS comment.

Figure 6: Hypothetical Examples Based on I.R.C. § 471(c)

Income/Expense

Case A
Tax Year 2017

Tax Cuts and Jobs Act Impact

Case B
Tax Year 201841
(change in accounting method)

Case C
Tax Year 2018
(no change)

Gross Sales

$1,000,000

$1,000,000

$1,000,000

Cost of Goods Sold

($500,000)

($750,000)

($500,000)

Gross Income

$500,000

$250,000

$500,000

Advertising

$100,000

$0

$100,000

Repairs and Maintenance

$50,000

$0

$50,000

Salaries and Wages42

$100,000

$0

$100,000

Total Business Expenses

$250,000

$0

$250,000

I.R.C. § 280E Adjustment

($250,000)

$0

($250,000)

Net Business Expenses

$0

$0

$0

Taxable Income

$500,000

$250,000

$500,000

Source: TIGTA-developed examples based on IRS discussions and IRS written responses.

SB/SE Division Field Counsel noted that the hypothetical examples represent practitioner arguments and further clarified that cost of goods sold are not expenses or deductions, but rather

adjustments to income made in arriving at gross income. IRS Chief Counsel noted in its May 2019 response to TIGTA that taxpayers may take the position illustrated in Case B to avoid the application of I.R.C. § 280E * * * 4 * * *

  • Case A represents a Tax Year 2017 retail marijuana return with cost of goods sold based on I.R.C. § 471. The Case A return reported advertising, repairs and maintenance, and salaries/wages as business expenses, which were disallowed based on the application of I.R.C. § 280E. This has been the typical application of I.R.C. § 280E for many years.

  • Case B represents a Tax Year 2018 retail marijuana tax return based on the application of I.R.C. § 471(c) (allowing the taxpayer to include non-cost of goods sold expenses in the cost of goods sold thereby avoiding the application of I.R.C. § 280E) and includes the same expenses as Case A. However, in this example, advertising, repairs and maintenance, and salaries/wages totaling $250,000 have been moved to cost of goods sold, increasing its value to $750,000. Based on the application of I.R.C. § 471(c), the taxpayer would not be subject to I.R.C. § 280E.

  • Case C represents a Tax Year 2018 retail marijuana return and is based on the taxpayer separately reporting advertising, repairs and maintenance, and salaries/ wages as business expenses. The business expenses were disallowed based on application of I.R.C. § 280E.

Based on examples B and C, the same taxpayer could potentially avoid I.R.C. § 280E by changing the accounting method under the newly enacted I.R.C. § 471(c). Based on these examples, there may be an unintended consequence of this provision of Tax Cuts and Jobs Act that may significantly reduce the application of I.R.C. § 280E.

The potential variance in treatment of taxpayers and lack of guidance provided by the IRS could lead to unfair treatment of similar taxpayers in the marijuana industry. It is important that the IRS have a strategy to inform this industry of the impact that I.R.C. § 471(c) may have on I.R.C. § 280E and take a consistent approach to taxpayer treatment based on the application of I.R.C. § 471(c).

The burden for marijuana businesses may change based on the tax impact of I.R.C. § 471(c)

TIGTA's analysis also included a review of the overall tax impact to marijuana businesses complying with I.R.C. § 280E. The purpose of this analysis was to provide a complete picture of the issues that may impact marijuana businesses. This industry is in a very unique position, where many States have legalized the purchase of marijuana while it remains illegal for Federal purposes. Therefore, it is important to keep in mind the potential impact of enacted and pending legislation to ensure that taxpayers remain in compliance with their tax obligations and that stakeholders are also aware of the tax impact in this industry. Our analysis considered the cost for marijuana businesses complying with I.R.C. § 280E and includes an estimate of the tax impact. The estimate is based on both affirmative adjustments identified on returns made by taxpayers as part of their filing and potential examination adjustments identified by TIGTA. As previously stated, because some of the business licenses were not active or did not have a Federal return filing requirement for Tax Year 2016, the populations were adjusted accordingly as reflected in Figure 5.

Our review of three statistically valid samples previously mentioned identified 169 (71 percent) of 237 marijuana businesses with Tax Year 2016 filing requirements with a tax impact due to I.R.C. § 280E. Specifically, 42 (86 percent) of the 49 returns for California, 67 (68 percent) of the 98 returns for Oregon, and 60 (67 percent) of the 90 returns for Washington likely had I.R.C § 280E adjustments.

We determined the tax impact from the exceptions was $7,429,602 in California, $2,868,562 in Oregon, and $4,043,993 in Washington.43 We found the average likely I.R.C. § 280E tax impact per return to be $151,625 in California, $29,271 in Oregon, and $44,933 in Washington for Tax Year 2016.44 When we project the results of the exceptions to each State's population, 287 returns for California, 356 for Oregon, and 430 for Washington would be affected by I.R.C. § 280E. The total likely tax impact would then be $50.8 million for California, $15.3 million for Oregon, and $29 million for Washington. The total potential tax impact for 1,073 taxpayers would be $95 million.45 When the results are forecasted over five years, we estimate that the tax impact is $475.1 million.46 This forecast represents only a portion of the tax noncompliance related to I.R.C. § 280E in that it includes only three out of 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana and does not consider the growth in the industry since Tax Year 2016.

The potential tax impact of I.R.C. § 471(c) for marijuana businesses will be based on how the IRS applies and administers this new statute as well as, potentially, how the courts interpret it.

Recommendations

Recommendation 3: The Commissioner, SB/SE Division, should develop guidance specific to the marijuana industry, such as a Frequently Asked Questions, and document and publicize it on its IRS.gov website to improve awareness of the tax filing requirements for taxpayers in this industry, such as the application of I.R.C. § 280E.

Management's Response: The IRS agreed with this recommendation but disagreed with the related outcome measure (see Appendix IV for more detailed information on the reported outcome measures). In its response, the IRS stated it has a comprehensive audit technique guide, available on IRS.gov, for cash intensive businesses that would assist taxpayers in this industry and plans to provide additional information for cash intensive businesses on IRS.gov including Frequently Asked Questions as needed.

Office of Audit Comment: The audit technique guide for cash intensive businesses does not address compliance issues unique to marijuana businesses. As such, the IRS should focus on developing new guidance information on IRS.gov specific to the marijuana industry.

Recommendation 4: IRS Chief Counsel should coordinate with the Commissioner, SB/SE Division, to develop and distribute, internally and externally, specific guidance on the application of I.R.C. § 471(c) in conjunction with I.R.C. § 280E for taxpayers that report Schedule I related activities on Federal tax returns.

Management's Response: IRS Chief Counsel disagreed with this recommendation because the Department of the Treasury and Chief Counsel resources at present are focused on priority guidance in response to the Tax Cuts and Jobs Act and identifying and reducing regulatory burdens in response to Executive Order 13789. IRS Chief Counsel added that the 2019-2020 Priority Guidance Plan includes an item under I.R.C. §§ 263A, 448, 460, and 471 to reflect Tax Cuts and Jobs Act changes affecting small businesses. The IRS Chief Counsel further noted they will consider developing and issuing guidance to ensure proper coordination between I.R.C. §§ 280E and 471(c) after the Priority Guidance Plan has been resolved.

Office of Audit Comment: While IRS management disagreed with this recommendation, they are in agreement with the need for developing guidance on the application of I.R.C. § 471(c) in conjunction with I.R.C. § 280E. If the IRS Chief Counsel ultimately develops and issues this guidance, the recommendation will be considered addressed. Prioritizing this project would help to avoid potential noncompliance issues with I.R.C. § 471(c) and I.R.C. § 280E, as the marijuana industry's growth and tax-related complexities will likely continue.

The IRS Is Not Using Available Marijuana State Tax Return Information to Identify Nonfilers and Underreported Income

I.R.C. § 61 requires that gross income must be reported whether it is from legal or illegal sources. The State of Washington publically posts on its website a list of marijuana businesses and sales. TIGTA analyzed public information available from the State of Washington website to identify potential underreported income and nonfilers.

Based on our analysis, 23 (26 percent) of the 90 returns for Washington likely have I.R.C § 61 adjustments.47 Specifically, eight returns involved underreported income and 15 were nonfilers.48 We determined the total likely tax assessments from the exceptions was $439,119. Using a weighted average per return for our sample, we determined the likely average I.R.C. § 61 assessment per return to be $4,879.

As part of our testing, we identified the State return filing rate for Tax Year 2016 returns reporting of sales for our sample population. Our prior testing under I.R.C. § 280E required a Federal return while this test required State sales reporting. We looked at each taxpayer and determined if Tax Year 2016 State sales were reported. We continued this process until we obtained the minimum required sample size of 90. Because some of the business licenses were not active or did not have a State return filing requirement for Tax Year 2016, we adjusted the population accordingly. We looked at 174 taxpayers to find our sample of 90 taxpayers that reported Tax Year 2016 sales. We determined the State return filing rate to be 52 percent.49

Based on the filing rates, we then adjusted the population for Washington to 793.50 Using the weighted average per return and the Tax Year 2016 filing rate, we projected the potential assessments for Tax Year 2016 based on I.R.C. § 61 to be $3.9 million.51 When the results are forecasted over five years, we estimate that the potential tax assessments from unreported income for Washington to be $19.3 million.52

The IRS agreed with our sample methodology and agreed with all cases reviewed using no materiality threshold for exception cases. TIGTA noted that Washington State tax information for marijuana businesses is available to the public which may result in higher compliance than other States. The IRS is currently not using this publically available information.

TIGTA planned to use State income tax and total sales, i.e., gross receipts, data for three States: California, Oregon, and Washington. Washington State sales data with respect to marijuana businesses is publicly available. As discussed on page 7 of this report, we were unable to obtain this information for California and Oregon. Accordingly, our analysis pertaining to underreporting and nonfiling is limited to Washington State.

Recommendation

Recommendation 5: The Commissioner, SB/SE Division, should leverage publically available State tax information and expand use of Fed/State agreements to identify nonfilers and unreported income in the marijuana industry.

Management's Response: The IRS agreed with this recommendation but disagreed with the related outcome measure (see Appendix IV for more detailed information on the reported outcome measures). In its response, the IRS noted that whether it pursues taxpayers in the marijuana industry depends on priorities and available resources. The IRS stated it will review the publically available State tax information and Fed/State agreements to determine whether and how they could be legally, systemically, effectively, and efficiently used in compliance activities.

Some Unbanked Marijuana Businesses Are Assessed Federal Tax Deposit Penalties Based on Non-Electronic Funds Transfers

All businesses are required to make their Federal tax deposits via the Electronic Funds Transfer system,53 and the failure to use the Electronic Funds Transfer system subjects businesses to a 10 percent penalty.54 Electronic Funds Transfer system payments are made through financial institutions, and as was previously described, marijuana businesses are sometimes unable to establish bank accounts. Consequently, these businesses may resort to cash payment of tax deposits.

Marijuana businesses that make cash payment deposits for employment tax returns may be subjected to Failure to Deposit (FTD) penalties.55 Penalty relief is available on a case-by-case basis for unbanked taxpayers based on reasonable cause.56 A taxpayer can establish reasonable cause by documenting at least one attempt to obtain a bank account every 24 months. The taxpayer should attach documentation of the attempt to obtain a bank account to their tax return to qualify for reasonable cause.57

* * * 2 * * *

* * * 2 * * *

Redacted "Law Enforcement Techniques/ Procedures and Guidelines for Law Enforcement Investigations or Prosecutions"table

[Editor’s Note: The reference mark for footnotes 5859 and 60 appear in the image above.]

To illustrate this potential impact, TIGTA conducted a limited review of the sampled marijuana businesses in Oregon for Tax Year 2016. TIGTA reviewed 98 marijuana businesses and found 60 who filed employment tax returns.61 We found that 13 (22 percent) of the 60 taxpayers were potentially unbanked and six (46 percent) of the 13 had been assessed FTD penalties.62 * * * 2 * * * will likely cause an increase in FTD * * * 2 * * *

The overall upward trend in * * * 2 * * * is most likely due to the growth in the legalization and retail sales of recreational use in the marijuana industry. The increased growth is causing a unique taxpayer burden in the marijuana industry due to FTD penalties assessed on unbanked marijuana businesses that are unable to access banking services.

The responsibility to request an abatement rests with the taxpayer. Some taxpayers may be unaware of the process to manually request an abatement as the procedures are not easily found on the IRS.gov website. The procedures are technical and are contained in various sections of the Internal Revenue Manual. This lack of understanding and administrative burden results in taxpayers repeatedly having to go through the penalty abatement process and may result in some taxpayers unnecessarily paying the FTD penalty.

Taxpayers including marijuana businesses should not be penalized because they cannot satisfy their respective employment tax obligations via the required electronic transmission process. The current conflict between Federal and State law regarding marijuana business activity is well established regarding banking access. The IRS needs to increase awareness of the current FTD penalty relief policies for unbanked taxpayers such as marijuana businesses.

Recommendation

Recommendation 6: The Commissioner, SB/SE Division, should increase educational outreach towards unbanked taxpayers making cash deposits regarding the unbanked relief policies available.

Management's Response: The IRS agreed with this recommendation. In its response, the IRS stated it will expand the penalty relief information currently available on IRS.gov.


Appendix I
Detailed Objective, Scope, and Methodology

The overall objective of this review was to evaluate the IRS's examination and education approach to certain cash-based industries with an emphasis on legal marijuana operations. To accomplish our objective, we:

I. Researched and analyzed Federal and State laws regulating the marijuana industry.

A. Researched and analyzed Federal law, including I.R.C. Sections (§§) 61, 168(k), 179, 263A, 280E, 448(c)(1), 471(c), 6302(h), and 6656. We also reviewed and analyzed the Control Substance Act, the Tax Cuts and Jobs Act, and key tax cases specific to marijuana businesses.

B. Researched and analyzed State laws for legalized medical and/or recreational use of marijuana as of July 2019.

II. Evaluated what the IRS is doing to inform, educate, and assist taxpayers in the marijuana industry.

A. Reviewed IRS guidance on IRS.gov and in publications to determine whether guidance was adequate to educate and inform marijuana businesses on compliance regulations including I.R.C. § 280E.

1. Determined if the marijuana Participant Guide that includes examples of how to calculate I.R.C. § 280E cost of goods sold was available to practitioners and marijuana businesses.

B. Determined if the IRS has participated in practitioner discussions regarding marijuana compliance with tax professionals including attorneys, certified public accountants, and enrolled agents.

C. Determined if the IRS has conducted any outreach to marijuana businesses including presentations, soft letters, or any other type of communication.

III. Evaluated the IRS's efforts to develop a compliance strategy for the marijuana industry.

A. Discussed past and current marijuana examination results for the CIP with the IRS.

1. Visited SB/SE Division Examination function Western Area Headquarters and conducted interviews with management and employees responsible for conducting marijuana compliance activities.

IV. Tested the level of compliance for income reporting and I.R.C. § 280E for marijuana businesses for Calendar Year 2016.

A. Identified marijuana businesses from California, Oregon, and Washington from State websites.

1. Analyzed active retail and producer business licenses listed on Oregon website (1,668 active licenses) as of August 20, 2018, and Washington website (1,533 active licenses) as of September 5, 2018. For California, analyzed active retail and distributor licenses (835 active licenses) listed on its website as of September 10, 2018.

2. Conducted additional research using the online State license registration data for each State to determine whether the business license selected was active during Calendar Year 2016 and used the IDRS to determine if the taxpayer filed a tax return for Tax Year 2016. Because some of the business licenses were not active or did not have a Federal return filing requirement for Tax Year 2016, we adjusted the population based on the Tax Year 2016 filing rate.

3. Computed the Tax Year 2016 filing rate by dividing the sample size by the number of taxpayers reviewed to find the sample. We identified a 40.16 percent filing rate in California by reviewing 122 taxpayers to select our sample of 49 businesses. We identified a 31.21 percent filing rate in Oregon by reviewing 314 taxpayers to select our sample of 98 businesses. We identified a 42.06 percent filing rate for Washington by reviewing 214 taxpayers to select our sample of 90 businesses.1

4. Reviewed the selected statistical samples totaling 237 marijuana businesses from three States; i.e., 49 from California, 98 from Oregon, and 90 from Washington. We analyzed each selected taxpayer's Tax Year 2016 returns from the IDRS or the Employee User Portal and determined whether the taxpayer was in compliance with I.R.C. § 280E. We selected the samples using a confidence level of 90 percent, a precision rate of ± 5 percent, and an error rate of 35.5 percent. A statistical sample was taken because we wanted to estimate and project the potential tax assessments that may be owed from marijuana businesses in these States that did not properly comply with I.R.C. § 280E. We shared our sampling methodology with our contracted statistician, who confirmed the accuracy of our methodology. We also shared our sampling methodology with the IRS, and they provided their concurrence. The contracted statistician assisted with reviewing our sampling plans and developing our projections.

5. Summarized the results and projected over the population.

6. Using the same methodology and sample selections as previously noted, we reviewed the total tax impact to marijuana businesses subject to I.R.C. § 280E. This analysis included both affirmative adjustments identified on returns and adjustments identified in Step IV.A.4.

7. Summarized the tax impact results and projected over the population.

B. Tested the level of compliance of income reporting for marijuana businesses using available State tax data for Washington. This test was limited to Washington only.

1. Analyzed active retail and producer business licenses listed on the Washington website (1,533 active licenses) as of September 5, 2018.

2. Conducted additional research using public data reported to the State to determine whether the business reported sales to the State of Washington for Tax Year 2016. Because some of the business licenses were not active or did not report sales to the State of Washington for Tax Year 2016, we adjusted the population based on the State return filing rate.

3. Computed the State of Washington Tax Year 2016 filing rate by dividing the sample size by the number of taxpayers reviewed to find the sample. We identified a 52 percent filing rate for Washington by reviewing 174 taxpayers to select our sample of 90 businesses.2

4. Reviewed the selected 90 taxpayers from Washington. We analyzed each selected taxpayer's Tax Year 2016 Federal returns from the IDRS or the Employee User Portal against the sales reported to the State and determined whether the taxpayer was in compliance with I.R.C. § 61. We selected the samples using a confidence level of 90 percent, a precision rate of ± 5 percent, and an error rate of 35.5 percent. A statistical sample was taken because we wanted to estimate and project the potential tax assessments that may be owed from marijuana businesses in the State from unreported income. We shared our sampling methodology with our contracted statistician, who confirmed the accuracy of our methodology. We also shared our sampling methodology with the IRS, and they provided their concurrence. The contracted statistician assisted with reviewing our sampling plans and developing our projections.

5. Summarized the results and projected over population.

V. Reviewed FTD penalty policies for unbanked taxpayers and reviewed 98 Oregon marijuana businesses for FTD penalties assessed to determine if FTD penalties were assessed on unbanked businesses.

Data validation methodology

We assessed the reliability of the returns reviewed for I.R.C. § 280E testing by: verifying that the tax per return matched the tax posted on the IDRS for all exceptions found and as part of the identification process of the Federal filings for our sample returns, we validated that the Federal return name and address matched State records. We found no variances between the tax reported on the returns and the tax reported on the IDRS or discrepancies in comparing addresses between the State and Federal records. We assessed the reliability for our I.R.C. § 61, i.e., gross income, testing through reviewing IDRS documents, State records, and background checks. We also requested agreement with the I.R.C. § 61 exceptions from the IRS. The IRS conducted its own review of the exception cases and agreed with all exception cases found. We concluded the data reviewed were sufficiently reliable for the purpose of this report for both the I.R.C § 280E and I.R.C. § 61 testing.

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: the IRS's policies, procedures, and practices related to compliance and education of marijuana businesses regarding

I.R.C. §§ 280E, 61, and 6302(h). We evaluated these controls by conducting interviews with management, reviewing examination results specific to the marijuana industry, reviewing written guidance documents, and conducting independent testing of marijuana returns for compliance.


Appendix II
Major Contributors to This Report

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

Christina M. Dreyer, Director

Javier L. Fernandez, Audit Manager

Lee S. Hoyt, Lead Auditor

Jesse Fenton, Auditor


Appendix III
Report Distribution List

Deputy Commissioner for Services and Enforcement

Commissioner, Small Business/Self-Employed Division

Deputy Commissioner, Small Business/Self-Employed Division

Director, Examination, Small Business/Self-Employed Division

Director, Enterprise Audit Management


Appendix IV
Outcome Measures

This appendix presents detailed information on the measurable impact that our recommended corrective actions will have on tax administration. These benefits will be incorporated into our Semiannual Report to Congress.

Type and Value of Outcome Measure:

Increased Revenue — Potential; $48.5 million in Federal income taxes from 901 taxpayers reporting marijuana businesses activity that did not comply with I.R.C. Section (§) 280E for Tax Year1 2016 or $242.6 million for 4,505 taxpayers when the sample is forecast over five years (see page 8).2

Methodology Used to Measure the Reported Benefit:

We obtained lists with a total of 4,036 taxpayers with marijuana business activity reported for Tax Year 2016 for the States of California, Oregon, and Washington. From the 4,036, we selected a statistically random sample of 237 taxpayers with Tax Year 2016 tax returns using three stratified subpopulations for each State to determine compliance with I.R.C. § 280E.3

We conducted additional research using the online State license registration data for each State to determine whether the business license for the business selected was active during Calendar Year 2016 and filed a tax return for Tax Year 2016. For those with an active license during Calendar Year 2016, we used the IDRS to determine if the taxpayer filed a tax return for Tax Year 2016 and continued the process until we obtained the minimum required sample size for each State. Because some of the business licenses were not active or did not have a Federal return filing requirement for Tax Year 2016, we adjusted the population based on the Tax

Year 2016 filing rate. The Tax Year 2016 filing rate was based on the sample size divided by number of taxpayers reviewed to find the sample. The filing rates computed were as follows:

40.16 percent (49/122) for stratum 1 (California), 31.21 percent (98/314) for stratum 2 (Oregon), and 42.06 percent (90/214) for stratum 3 (Washington). We then applied the filing rate to the original population. We adjusted the population to 335 returns (835 x 40.16 percent) for stratum 1; 521 returns (1,668 x 31.21 percent) for stratum 2; and 645 returns (1,533 x 42.06 percent) for stratum 3. Figure 1 illustrates TIGTA's methodology for our stratified populations and adjusted populations of returns reviewed for compliance with I.R.C. § 280E. 

Figure 1: Population for I.R.C. § 280E Compliance Testing for Tax Year 2016

I.R.C. § 280E

California
(Stratum 1)

Oregon
(Stratum 2)

Washington
(Stratum 3)

Totals

Population

835

1,668

1,533

4,036

Sample size

49

98

90

237

Taxpayers Reviewed to Identify Sample

122

314

214

650

Tax Year 2016 Filing Rate

40.16%

31.21%

42.06%

 

Adjusted Population

335

521

645

1,501

Source: TIGTA analysis of Tax Year 2016 Federal filing rate based on randomly selected businesses for each State.

We found 140 of the 237 taxpayers are not in compliance with I.R.C. § 280E. The total potential expense adjustments from the 140 exceptions totaled $26,257,602. For these 140 exceptions, $7,298,686 in tax may be due based on noncompliance with I.R.C. § 280E.4 For these 140 exceptions, we calculated the error rate for each stratum by dividing the stratum number of errors by the stratum sample cases. This resulted in computed error rates of 77.55 percent (38/49) for stratum 1, 50.00 percent (49/98) for stratum 2, and 58.88 percent (53/90) for stratum 3. We applied the error rates to the adjusted populations of Tax Year 2016 marijuana businesses tax returns. Based on these parameters, 260 returns (335 x 77.55 percent) for stratum 1, 261 returns (521 x 50.00 percent) for stratum 2, and 380 returns (645 x 58.88 percent) for stratum 3 had potential I.R.C. § 280E adjustments. The total overall returns with potential I.R.C. § 280E adjustments were 901 (260 + 261 +380).

To compute the dollar amount for the exceptions, we calculated the error dollars for each stratum by dividing the error dollars for each stratum by the stratum sample cases. This computed dollar rates of $103,266 ($5,060,011/49) for stratum 1, $11,705 ($1,147,124/98) for stratum 2, and $12,128 ($1,091,551/90) for stratum 3. We applied this sample weighted average to the population of marijuana tax returns by stratum. Based on these parameters, $34,594,110 (335 x $103,266) for stratum 1, $6,098,305 (521 x $11,705) for stratum 2, and $7,822,560 (645 x $12,128) for stratum 3 of Federal income tax may be owed for Tax Year 2016.5 The total Federal income tax owed for Tax Year 2016 for the three stratums was $48,514,975 ($34,594,110 + $6,098,305 + $7,822,560)6 based on noncompliance with I.R.C. § 280E or $242,574,875 in Federal income tax may be due when forecast over the next five years.7 This forecast is considered a conservative estimate because it includes only three out of 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana and does not consider the growth in the industry since Tax Year 2016. Figure 2 identifies and illustrates our sample results and projections.

Figure 2: I.R.C. § 280E Compliance Testing Results for Tax Year 2016 for Taxpayers That Did Not Comply With I.R.C. § 280E

I.R.C. § 280E

California

Oregon

Washington

Totals

Adjusted Population

335

521

645

1,501

Sample Size

49

98

90

237

Number With Potential Adjustments

38

49

53

140

Percentage With Potential Adjustments

77.55%

50.00%

58.88%

 

Projected Exception Cases

260

261

380

901

Total Potential Expense Adjustments From Exceptions8

$16,773,489

$5,187,389

$4,296,724

$26,257,602

Total Potential Tax From Exceptions9

$5,060,011

$1,147,124

$1,091,551

$7,298,686

Weighted Average for Sample Returns

$103,266

$11,705

$12,128

 

Tax Year 2016 Potential Tax Assessments Lost10 11 12

$34,594,110

$6,098,305

$7,822,560

$48,514,975

Five-Year Forecast

$172,970,550

$30,491,525

$39,112,800

$242,574,875

Source: Results and calculated outcomes are based on TIGTA's analysis of the IDRS, the IRS Employee User Portal data for sampled taxpayers, and tax calculations using the Tax Year 2016 Graduated Corporate Tax Rate Schedule.

Management's Response: The IRS did not agree with the 90 percent confidence interval used to compute the outcome measure, stating that it is not scientifically rigorous enough to use for projections and yields imprecision and uncertainty about the outcome measure.

Office of Audit Comment: We believe that the methodology used to identify these taxpayers and quantify the outcome was appropriate and provided a reasonable estimate of the forecasted $242.6 million potential tax impact. During fieldwork, the IRS found no issues with our sampling methodology and agreed with all identified exceptions from our sample. We believe the forecasted potential tax assessment amount is both reasonable and conservative, providing relevant and substantive data regarding the potential risk in this industry. Further, our forecast is limited to only the three States we reviewed. The overall risk across all 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana would be significantly greater.

Type and Value of Outcome Measure:

  • Taxpayer Burden (tax impact)13 — Potential; $95 million in Federal income taxes for 1,073 taxpayers based on the application of I.R.C. § 280E on Tax Year 2016 returns or $475.1 million for 5,365 taxpayers when the sample is forecast over five years (see page 14).14

Methodology Used to Measure the Reported Benefit:

We obtained lists with a total of 4,036 taxpayers with potential marijuana business activity reported for Tax Year 2016 for the States of California, Oregon, and Washington. From the 4,036, we selected a statistically random sample of 237 taxpayers with Tax Year 2016 tax returns using three stratified subpopulations representing each State to determine the potential burden for taxpayers complying with I.R.C. § 280E. The initial population was adjusted based on using the same methodology previously stated and illustrated in Figure 1 of Appendix IV.

We found that 169 of the 237 taxpayers are subject to tax based on I.R.C. § 280E.15 The total expense adjustments from the 169 exception returns reviewed totaled $53,650,703. This estimate includes both taxpayer-reported adjustments and potential adjustments based on TIGTA's review. TIGTA's estimate of potential tax impact is based on return information, and we recognize the actual tax impact may be higher when considering additional information that may not be reflected on the returns reviewed. For these 169 exceptions, $14,342,157 in tax is required to comply with I.R.C. § 280E. For these 169 exceptions, we calculated the error rate for each stratum by dividing the stratum number of errors by the stratum sample cases. This resulted in computed error rates of 85.71 percent (42/49) for stratum 1, 68.37 percent (67/98) for stratum 2, and 66.67 percent (60/90) for stratum 3. We applied the error rates to the adjusted populations of Tax Year 2016 marijuana business tax returns. Based on these parameters, 287 returns (335 x 85.71 percent) for stratum 1, 356 returns (521 x 68.37 percent) for stratum 2, and 430 returns (645 x 66.67 percent) for stratum 3 have tax impact based on I.R.C. § 280E issues. The total overall returns with tax impact based on I.R.C. § 280E were 1,073 (287 + 356 +430).

To compute the dollar amount for the exceptions, we calculated the error dollars for each stratum by dividing the error dollars for each stratum by the stratum sample cases. This computed dollar rates of $151,625 ($7,429,602/49) for stratum 1, $29,271 (2,868,562/98) for stratum 2, and $44,933 ($4,043,993/90) for stratum 3. We applied this sample weighted average to the population of marijuana tax returns by stratum. Based on these parameters, $50,794,375 (335 x $151,625) for stratum 1, $15,250,191 (521 x $29,271) for stratum 2, and $28,981,785 (645 x $44,933) for stratum 3 in Federal income tax for Tax Year 2016 would be incurred by taxpayers based on compliance with I.R.C. § 280E. The total Federal income tax owed for Tax Year 2016 for the three stratums would then be $95,026,351 ($50,794,375 + $15,250,191 + $28,981,785) or $475,131,755 in Federal income tax when forecast over the next five years.16 This forecast is for only three out of 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana and does not consider the growth in the industry since Tax Year 2016. Figure 3 identifies and illustrates our sample results and projections.

Figure 3: I.R.C. § 280E Tax Impact Testing Results for Tax Year 2016

Taxpayer Impact Summary

California

Oregon

Washington

Totals

Adjusted Population

335

521

645

1,501

Number of Businesses in Sample

49

98

90

237

Number of Businesses With Tax Impact

42

67

60

169

Percentage With Tax Impact

85.71%

68.37%

66.67%

 

Projected Exception Cases

287

356

430

1,073

I.R.C. § 280E Adjustments Reported by Taxpayer on Return

$7,696,459

$9,476,815

$10,089,373

$27,262,647

Potential I.R.C. § 280E Examination Adjustments Based on TIGTA Review

$16,773,489

$5,187,239

$4,427,328

$26,388,056

Total Adjustments Per I.R.C. § 280E

$24,469,948

$14,664,054

$14,516,701

$53,650,703

Total Potential Tax Impact

$7,429,602

$2,868,562

$4,043,993

$14,342,157

Average Tax Impact Per Business

$151,625

$29,271

$44,933

 

Tax Year 2016 Tax Impact17

$50,794,375

$15,250,191

$28,981,785

$95,026,351

Five-Year Forecast18

$253,971,875

$76,250,955

$144,908,925

$475,131,755

Source: Results and calculated outcomes are based on TIGTA's analysis of the IDRS, the IRS Employee User Portal data for sampled taxpayers, and tax calculations using the Tax Year 2016 Graduated Corporate Tax Rate Schedule.

Management's Response: The IRS disagreed with the taxpayer burden outcome measure on the basis of its definition and noted concerns regarding “double counting” by including the estimated tax effect of noncompliant taxpayers from the first outcome measure in this outcome measure.

Office of Audit Comment: We believe that the methodology used for the outcome measure was appropriate and provided a reasonable estimate of the forecasted $475.1 million in potential tax impact. With respect to the IRS's concerns as to whether the outcome measure includes double counting, while marijuana businesses may be allowed to legally operate under State law, they remain illegal under Federal law and certain expenses are not deductible under I.R.C. § 280E. To determine the overall tax impact of I.R.C. § 280E, we included both compliant and noncompliant marijuana businesses in this calculation. We do not agree that this is double counting. Instead our analysis shows the tax impact of I.R.C. § 280E should marijuana remain classified as a Schedule I controlled substance under the Controlled Substances Act for these marijuana businesses. Additionally, the IRS asserted that this outcome measure did not meet the official definition of “taxpayer burden.” However, we use the term “taxpayer burden” consistent with the IRS's own definition, i.e., the associated tax impact for marijuana businesses complying with the law.19

Type and Value of Outcome Measure:

  • Increased Revenue — Potential; $3.9 million in Federal income taxes from 203 taxpayers reporting marijuana business activity who did not comply with I.R.C. § 61 for Tax Year 2016 or $19.3 million for 1,015 taxpayers when the sample is forecast over five years(see page 20).20

Methodology Used to Measure the Reported Benefit:

We obtained lists with a total of 1,533 taxpayers with potential marijuana business activity reported for Tax Year 2016 for the State of Washington. From the 1,533, we selected a statistically random sample of 90 taxpayers with Tax Year 2016 State sales to determine the taxpayer compliance with I.R.C. § 61. The initial population was adjusted based on the Tax Year 2016 State return filing rate.

Our prior testing under I.R.C. § 280E required a Federal return, while this test required State sales reporting. We looked at each taxpayer and determined if Tax Year 2016 State sales were reported. We continued this process until we obtained the minimum required sample size of 90. We looked at 174 taxpayers to find our sample of 90 taxpayers that reported Tax Year 2016 State sales. We determined the State return filing rate to be 51.72 percent. We calculated the adjusted population to be 793 (1,533 x 51.72 percent).21

We found 23 of the 90 taxpayers were not in compliance with I.R.C. § 61. The total potential income adjustments from the 23 exceptions totaled $3,259,179. TIGTA's calculations included an estimate of 56 percent for cost of goods sold for nonfilers.22 For these 23 exceptions, $439,119 in tax may be due based on noncompliance with I.R.C. § 61. For these 23 exceptions, we calculated the error rate by dividing the stratum number of errors by the stratum sample cases, which resulted in an error rate of 25.56 percent (23/90). We then applied the error rate to the adjusted population of Tax Year 2016 marijuana business tax returns. Based on these parameters, 203 returns (793 x 25.56 percent) have potential I.R.C. § 61 adjustments.

To compute the dollar amount for the exceptions, we calculated the error dollars by dividing the error dollars by the stratum sample cases. This computed a dollar rate of $4,879 ($439,119/90). We applied this sample weighted average to the population of marijuana tax returns. Based on these parameters, $3,869,047 (793 x $4,879) of Federal income tax may be owed for Tax Year 2016 from noncompliance with I.R.C. § 61 or $19,345,235 when forecast over the next five years. Figure 4 identifies and illustrates our sample results and projections.23

Figure 4: I.R.C. § 61 Compliance Testing Results for Tax Year 2016

Income Test

Washington

Total Population

1,533

Projected Tax Year 2016 Filings Rate24

51.72%

Adjusted Population

793

Number of Businesses in Sample

90

Number With Potential Adjustments

23

Percentage With Potential Adjustments

25.56%

Projected Exception Cases

203

Total Potential Income Adjustments From Exceptions

$3,259,179

Total Potential Tax From Exceptions

$439,119

Weighted Average for Sample Returns25

$4,879

Tax Year 2016 Potential Tax Assessments Lost26

$3,869,047

Five-Year Forecast27

$19,345,235

Source: Results and calculated outcomes are based on TIGTA's analysis of the IDRS, the IRS Employee User Portal data for sampled taxpayers, and tax calculations using the Tax Year 2016 Graduated Corporate Tax Rate Schedule.

Management's Response: The IRS did not agree with the 90 percent confidence interval used to compute the outcome measure, stating that it is not scientifically rigorous enough to use for projections and yields imprecision and uncertainty about the outcome measures.

Office of Audit Comment: The methodology used to identify these taxpayers and quantify the outcome was appropriate and provided a reasonable estimate of the forecasted $19.3 million potential tax impact. As previously stated, during fieldwork, the IRS found no issues with our sampling methodology and agreed with all exceptions found in our sample. We believe the forecasted potential tax assessment amount is both reasonable and conservative, because the estimate is for only one of 33 States providing relevant and substantive data regarding the potential risk in this industry. The overall risk across all 33 States and the District of Columbia that allow for either medical and/or recreational use of marijuana is significantly greater.


Appendix V
Tax Year 2016 Graduated Corporate Tax Rate Schedule

Tax Rate Schedule

If taxable income (line 3D, Form 1120) on page 1 is:

Over —

But not over —

Tax is:

Of the amount over —

$0

$50,000

15%

$0

50.000

75,000

$7,500 + 25%

50,000

75.000

100,000

13,750 + 34%

75,000

100.000

335,000

22,250 + 39%

100,000

335,000

10,000,000

113,900 + 34%

335.000

10,000,000

15,000,000

3,400,000 + 35%

10,000,000

15,000,000

18,333,333

5,150,000 + 38%

15,000,000

18,333,333

. . . . .

35%

0

Source: Tax rate schedule from Tax Year12016 Instructions for Form 1120, U.S. Corporation Income Tax Return, Schedule J, Tax Computation and Payment section on page 17.


Appendix VI
Glossary of Terms

Term

Definition

Audit Technique Guide

Helps IRS examiners during audits by providing insight into issues and accounting methods unique to specific industries. While the ATGs are designed to provide guidance for IRS employees, they are also useful to small business owners and tax professionals who prepare returns.

Calendar Year

The 12-consecutive-month period ending on December 31.

Compliance Initiative Project

Authorized activity outside of planned strategies involving taxpayer contact to identify potential areas of noncompliance for correcting the noncompliance.

Discriminant Function

A mathematical technique used to computer score income tax returns as to examination potential. Examination potential is indicated by a numeric score which is assigned to each return by examination class; the greater the score, the greater the examination potential within each examination class.

Electronic Funds Transfer

The electronic transfer of money over an online network.

Fiscal Year

Any yearly accounting period, regardless of its relationship to a calendar year. The Federal Government's fiscal year begins on October 1 and ends on September 30.

Integrated Data Retrieval System

IRS computer system capable of retrieving or updating stored information. It works in conjunction with a taxpayer's account records.

Internal Revenue Code

The Federal tax law enacted by Congress in Title 26 of the United States Code. It is organized by topics such as income, estate and gift, employment, and miscellaneous excise taxes.

Internal Revenue Manual

The official source of IRS policies, procedures, and guidelines.

Revenue Agent

Employees in the Examination function who conduct face-to-face examinations of more complex tax returns such as businesses, partnerships, corporations, and specialty taxes.

Tax Year

A 12-month accounting period for keeping records on income and expenses used as the basis for calculating the annual taxes due. For most individual taxpayers, the tax year is synonymous with the calendar year.


Appendix VII
Management's Response to the Draft Report

March 11, 2020

MEMORANDUM FOR
MICHAEL E. McKENNEY

FROM:
DEPUTY INSPECTOR GENERAL FOR AUDIT
Eric C. Hylton C .
Commissioner, Small Business/Self-Employed Division

SUBJECT: Draft Report — The Growth of the Marijuana Industry Warrants Increased Tax Compliance Efforts and Additional Guidance
(Audit #201830022)

Thank you for the opportunity to review and comment on the above subject draft audit report. While marijuana businesses are legalized for state purposes, marijuana remains classified as a Schedule I controlled substance under the Control Substance Act presenting challenges for what is allowed as deductible business expenses for federal tax purposes.

Cash intensive businesses, particularly those that are illegal under federal law, create additional challenges for the IRS and the taxpayer. The marijuana industry is largely unbanked due to the illegal status of marijuana at the federal level. As TIGTA notes, it is often difficult for a taxpayer engaged in the marijuana business to open a bank account as many banks are hesitant to service this industry due to the increased reporting requirements (e.g., suspicious activity reports) and risk of being charged with violating anti-money laundering and illicit finance laws. The taxpayer’s inability to utilize a bank for these businesses creates concerns of public safety, transparency, and accountability. It also hinders our enforcement efforts as it could minimize the types of third-party information documents available (e.g., Form 1099-K). The lack of available banking services creates an additional burden to the taxpayer, subjecting them to failure to deposit penalties. Legislative changes could address the challenges encountered with Internal Revenue Code (IRC) §280E, banking policies, and public safety.

We began addressing the marijuana industry in 2015 by providing guidance on the application of IRC §280E. Subsequently, we initiated multiple compliance initiative projects (CIPs) and delivered training to our examiners. We continue to evaluate the results of the CIPs to determine the overall compliance in the industry and we are devoting the appropriate amount of resources based on our priorities. In April 2017, we established a cross-functional marijuana team that has since developed strategic

objectives, ensured training and job aids are available to examiners, ensured examiners are aware of proper procedures with respect to contacting city and state agencies, evaluated methods for identifying potential noncompliance, and collaborated with external stakeholders, including practitioners, to increase an awareness of tax responsibilities and improve compliance.

Our budget has generally suffered cuts since Fiscal Year (FY) 2010 with our full-time equivalents decreasing over 15 percent from FY 2013 to FY 20181. We also experienced a hiring freeze from FY 2011 to FY 2018 with very few hiring exceptions made during this time. During this decrease in both funding and personnel we faced increasing challenges, including implementation of new legislation such as the Foreign Account Tax Compliance Act (FATCA), Affordable Care Act (ACA), and Tax Cuts and Jobs Act (TCJA). It will take considerable time for us to restore our enforcement levels. Our funding and staffing limitations required us to make difficult decisions regarding priorities and the types of enforcement actions we pursued. We must balance these limited resources with priorities that include, but are not limited to, customer service and enforcement efforts. We must achieve balanced coverage across the various types of returns as well as geographic considerations. In FY 2020, we have a renewed focus on high-income taxpayers and combating fraud. To address this trend of declining resources, our FY 2021 budget request included funds for investments in expanding and improving the effectiveness and efficiency of our overall tax enforcement program.

Your report states that the scope of your audit work was impaired because “[t]he IRS would not use existing information sharing agreements with the respective State agencies [. . .] for the purpose of determining whether this information assisted in the identification of cash-based businesses that underreport or failure [sic] to file tax returns." During the audit, we made it clear that requesting such information from the states on behalf of TIGTA solely for use by TIGTA in conducting this audit was not within the scope of our Memorandums of Understanding (MOUs) with the states, which solely govern requests for which the IRS has a compliance need.

During the audit, we informed TIGTA that we did not concur with the outcome measures used. Although the sampling method was shared with the IRS, the 90% confidence interval is not scientifically rigorous enough to use for projections and yields imprecision and uncertainty about the outcome measures. The use of this confidence level increases the risk of a false positive, which could ultimately mislead the reader to believe there is a problem when in fact no problem exists. In addition, the five-year projection amplifies the potential revenue without considering the high probability of subsequent compliance after a taxpayer has been subjected to a compliance activity. Lastly, pursuing the cases suggested in this audit report comes with an opportunity cost that was not accounted for in the outcome measures, and potentially causes a negative impact to voluntary compliance due to the redirection of resources from other
compliance activities.

With respect to the taxpayer burden measure, the official definition of “taxpayer burden” is the time spent and out-of-pocket costs associated with tax compliance activities (e.g. tax-related record keeping or return preparation) performed by the taxpayer or their representative. The compliance burden that IRC §280E places on taxpayers is the need to know that they do not qualify for IRC §162 deductions. Your outcome measure relating to taxpayer burden reflects non-compliance with IRC §280E addressed in the first outcome measure as well as your estimate of the tax effect of compliant taxpayers voluntarily forgoing deductions for operating expenses. Including noncompliance from the first outcome measure within the second outcome measure on taxpayer burden arising from complying with the law results in double counting.

Attached is our response to your recommendations. If you have any questions, please contact me or Scott Irick, Director, Examination Operations, Small Business/Self-Employed (SB/SE) Division.

Attachment


Attachment

Recommendation 1:

The Commissioner, SB/SE Division, should develop a comprehensive compliance approach, i.e., national CIP, for this industry and leverage State marijuana business lists to identify noncompliant taxpayers.

Planned Corrective Action:

Whether we pursue taxpayers in the marijuana industry, or any other industry, depends upon IRS priorities and availability of resources. We will use data analytics to identify the size and scope of non-compliant taxpayers. From this analysis, we will determine whether additional compliance activities are necessary and, if so, identify and prioritize them based on resources available.

Implementation Date:

April 15, 2021

Responsible Official:

Director, Exam Case Selection

Corrective Action Monitoring Plan:

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

Recommendation 2:

The Commissioner, SB/SE Division, should direct all examination areas to use Aging Reason Codes to track non-CIP marijuana business examination results.

Planned Corrective Action:

We will provide guidance to examiners to apply the appropriate code(s) to cases to track marijuana examinations.

Implementation Date:

October 15, 2020

Responsible Official:

Director, Exam Case Selection

Corrective Action Monitoring Plan:

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

Recommendation 3:

The Commissioner, SB/SE Division, should develop guidance specific to the marijuana industry such as a Frequently Asked Questions document and publicize it on its IRS.gov website to improve awareness of the tax filing requirements for taxpayers in this industry, such as the application of I.R.C. §280E.

Planned Corrective Action:

We have a comprehensive audit technique guide (ATG) for cash intensive businesses that would assist taxpayers in this industry, and which is already available on IRS.gov. We will provide additional information for cash intensive businesses on IRS.gov, including in the form of frequently asked questions, as needed.

Implementation Date:

January 15, 2021

Responsible Official:

Director, Exam Quality & Technical Support

Corrective Action Monitoring Plan:

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

Recommendation 4:

IRS Chief Counsel should coordinate with the Commissioner, SB/SE Division, to develop and distribute, internally and externally, specific guidance on the application of I.R.C. §471 (c) in conjunction with I.R.C. §280E for taxpayers that report Schedule I related activities on Federal tax returns.

Planned Corrective Action:

We shared your recommendation with Chief Counsel. Counsel provided the following response:

We disagree. Treasury and Counsel resources at present are focused on priority guidance in response to the TCJA and identifying and reducing regulatory burdens in response to Executive Order 13789. Currently, the 2019-2020 Priority Guidance Plan (PGP) for regulations and other guidance lists an item under §§263A, 448, 460, and 471 to reflect TCJA changes affecting small businesses. Chief Counsel in consultation with the Commissioner, Treasury, and the SB/SE Division will consider developing and issuing guidance to ensure proper coordination between I.R.C. §§280E and 471 (c) and general tax principles once the PGP item has been resolved.

Implementation Date:

N/A

Responsible Official:

N/A

Corrective Action Monitoring Plan:

N/A

Recommendation 5:

The Commissioner, SB/SE Division, should leverage publicly available State tax information and expand use of Fed/State agreements to identify nonfilers and unreported income in the marijuana industry.

Planned Corrective Action:

Whether we pursue taxpayers in the marijuana industry, or any other industry, depends upon IRS priorities and availability of resources. We will review the publicly available State tax information and Fed/State agreements to determine whether and how they could be legally, systemically, effectively, and efficiently used in compliance activities.

Implementation Date:

October 15, 2021

Responsible Official:

Director, Exam Case Selection

Corrective Action Monitoring Plan:

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

Recommendation 6:

The Commissioner, SB/SE Division, should increase educational outreach towards unbanked taxpayers making cash deposits regarding the unbanked relief policies available.

Planned Corrective Action:

We will expand the penalty relief information currently available on IRS.gov to educate unbanked taxpayers about the penalty relief policies applicable to them.

Implementation Date:

January 15, 2021

Responsible Official:

Director, Business Support Organization, SB/SE Operations Support

Corrective Action Monitoring Plan:

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

FOOTNOTES

1Sources for marijuana industry sales include: bdsanalytics.com in partnership with Arcview Market Research and newfrontierdata.com including reports posted at globenewswire.com.

2The Figure 1 analysis of States that have legalized recreational use is limited to effective dates as of July 2019 and does not include States that have passed laws with effective dates after July 2019.

3None of the 50 States allow for recreational use only.

4Marijuana producers may produce, harvest, trim, dry, cure, and package marijuana into lots for sale at wholesale to marijuana processors. They may also produce and sell marijuana plants, seed, and plant tissue to other marijuana producers.

5Pub. L. No. 91-513, title II, § 202, 84 Stat. 1247 (1970).

6See Appendix VI for a glossary of terms.

7Other business expenses include all deductions that reduce income not including cost of goods sold.

8See Appendix V for the calculations based on the Tax Year 2016 Graduated Corporate Tax Rate Schedule.

9TIGTA's review of State-level retail taxes is summary level only and does not attempt to analyze the differences in tax treatment for medical and recreational sales.

10Washington's State sales tax of 6.5 percent does not include local sales tax rates.

11Oregon's State marijuana tax of 17 percent does not include local marijuana taxes. In Oregon, cities and counties can add up to an additional 3 percent marijuana retail tax.

12The State of California has a 7.25 percent State standard sales and use tax, and local sales taxes may also apply.

13Currency and Foreign Transactions Reporting Act, Pub. L. No. 91-508, Title II, 84 Stat.1118 (1970) (codified as amended at 31 §§ 321, 5311–5314, and 5316–5322). Part II of the BSA is cited as the Currency and Foreign Transactions Reporting Act. It is codified now at 31 U.S.C., Money and Finance, Chapter 53, Monetary Transactions, Part II, Records and Reports on Monetary Instruments Transactions.

1412 C.F.R. § 208.62.

15James R. Koren, Why Some Pot Businesses Hide Their Cash-and Others Truck It Straight to a Federal Vault, Los Angeles Times, July 7, 2017, Business, https://www.latimes.com/business/la-fi-cannabis-banking-20170707-story.html.

16Testimony of Rachel Pross, Chief Risk Officer, Maps Credit Union, dated Feb. 13, 2019, to the House Financial Services Committee. The 13,500 individual reports includes both SAR and Cash Transaction Reports.

17Marijuana Banking Update report from the Financial Crimes Enforcement Network website (June 2019).

18MITRE is a Government contractor commonly used by the IRS for special projects.

19This memorandum was referred to as the “Cole Memorandum” which was issued on August 29, 2013, by James Cole, U.S. Deputy Attorney General. The Cole Memorandum was sent to all U.S. attorneys providing updated guidance on marijuana enforcement under the Controlled Substance Act. The memorandum established eight priorities to: 1) prevent the distribution of marijuana to minors, 2) prevent revenue from the sale of marijuana from going to criminal enterprises, 3) prevent diversion of marijuana from States where it is legal to States where it is not legal, 4) prevent marijuana activity from covering trafficking of other illegal substances, 5) prevent drugged driving and other public health consequences, 6) prevent marijuana use or possession on Federal property, 7) prevent violence associated with the distribution of marijuana, and 8) prevent growth of marijuana on public lands.

20The Financial Crimes Enforcement Network Guidance, BSA Expectations Regarding Marijuana-Related Businesses (Feb. 14, 2014).

21The three SAR filing standards are: “Marijuana Limited” (the business does not run afoul of the eight priorities listed by the Cole Memorandum); the “Marijuana Priority” SAR must be made when the financial institution reasonably believes the business violated one of the Cole Memorandum priorities; and the “Marijuana Termination” SARs are required when the financial institution deems it necessary to terminate a relationship with a marijuana-related business.

22U.S. Attorney General William Barr Confirmation Hearing (January 19, 2019) who stated, “My approach to this would be not to upset settled expectations and the reliance interest that have arisen as a result of the Cole Memorandum and investments have been made and so there has been reliance on it, so I don't think it's appropriate to upset those interests.”

23H.R. 1595, 116th Cong. § 1 (2019). The Secure and Fair Enforcement Banking Act bill would prevent Federal banking regulators from taking the following actions in States which have legalized cannabis: i) terminating deposit insurance for a bank solely because it has provided financial services to a cannabis-related business, ii) imposing restrictions or penalties on depository institutions for providing financial services to cannabis-related businesses, iii) discouraging institutions from offering financial services to individuals or entities because they are involved in the cannabis industry, and iv) otherwise taking adverse supervisory action with respect to cannabis-related lending activity.

24H.R. 2093, 116th Cong. § 1 (2019).

25The Western Area is one of seven geographic SB/SE Division Field Examination Areas. The others are North Atlantic, South Atlantic, Central, Midwest, Gulf States, and Southwest.

26A Part 1 CIP authorization is requested when fewer than 50 taxpayer contacts are anticipated for a field office compliance initiative.

27A Part 2 CIP authorization is requested to expand a Part 1 CIP when the Part 1 reveals documented noncompliance or when there is an otherwise established compliance risk.

28DIF examinations are the largest category of SB/SE Division examinations.

29The comparisons in Figures 3 and 4 are general in nature and based on IRS summary reports. TIGTA did not attempt to compare productivity results by the type of return, size of business, or any other classification.

30The eight areas include seven geographical examination areas and one area for technical services.

31To select each of the three statistically valid samples, we used an expected error rate of 35.5 percent, a precision rate of ± 5 percent, and a confidence interval of 90 percent.

32The tax impact amount does not account for some returns reviewed for which the likely I.R.C. § 280E adjustment was reduced or eliminated based on current year net operating losses.

33This is the total weighted average per case based on the stratified sample calculations as shown in Appendix IV.

34See Appendix IV. When projecting the results of our statistical sample, we are 95 percent confident that the actual total amount is between $30.1 million and $66.9 million. The five-year forecast for potential tax assessments is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. The forecast does not include changes to tax rates and possible tax law changes based on the Tax Cuts and Jobs Act. In addition, it assumes that additional taxpayers are impacted each year. The actual amount of revenue tax assessed is contingent upon the IRS implementing an effective process to identify questionable I.R.C. § 280E-related tax returns and on the extent that taxpayers audited by the IRS cannot adequately support their expenses.

35Based on our analysis, the population was adjusted to 40.16 percent, 31.21 percent, and 42.06 percent for California, Oregon, and Washington, respectively. Rounded for ease of presentation. See Appendix IV for detailed calculations.

36The likely tax from exceptions was calculated based on actual tax for Form 1120 corporations using the Tax Year 2016 Graduated Corporate Tax Rate Schedule; Tax Year 2016 individual tax rates for flow-through entities with one, two, or three shareholders or partners; and an estimated tax rate of 10 percent for flow-through entities with more than three partners or shareholders or entities with multiple tiers. TIGTA noted that some I.R.C. § 280E adjustments were absorbed by current year net operating losses and resulted in no current year tax effect.

37TIGTA requested IRS input and agreement on our analysis. The IRS agreed with all cases reviewed.

38Pub. L. No. 115-97. Officially known as “An act to provide for reconciliation pursuant to titles II and V of the concurrent resolution on the budget for Fiscal Year 2018.”

39I.R.C. § 448(c)(1) states that a corporation or partnership meets the gross receipts test of this subsection for any taxable year if the average annual gross receipts of such entity for the three-taxable-year period ending with the taxable year which precedes such taxable year does not exceed $25,000,000.

40I.R.C. § 471(c)(1)(B)(ii).

41Case B is based on including all business expenses in cost of goods sold.

42Salaries and Wages are not related to cost of goods sold.

43The tax impact amount does not account for some returns reviewed for which the likely I.R.C. § 280E adjustment was reduced or eliminated based on current year net operating losses.

44This is the total weighted average per case based on the stratified sample calculations as shown in Appendix IV.

45The confidence interval projections for each State were based on a two-sided 95 percent confidence interval. We are 95 percent confident the actual total amount is between $70.5 million and $119.5 million. Total rounded for ease of presentation.

46The five-year forecast for potential tax assessments is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. The forecast does not include changes to tax rates, tax law changes based on the Tax Cuts and Jobs Act. It assumes that additional taxpayers are impacted each year. See Appendix IV for a detailed analysis of taxpayer burden (tax impact).

47TIGTA requested IRS input and agreement on analysis. The IRS agreed with all cases reviewed.

48TIGTA counted all nonfilers and unreported income regardless of materiality in order to accurately project risk.

49The State return filing rate is based on 90/174 = 51.72 percent.

50Rounded for ease of presentation (1,533 x 51.72 percent = 793). See Appendix IV for detailed calculations.

51The confidence interval projections were based on a two-sided 95 percent confidence interval. We are 95 percent confident that the actual total amount for Washington is between $954,341 and $12,868,924. This projection is based on using empirical likelihood. Numbers have been adjusted to be consistent with previously rounded numbers.

52The five-year forecast for potential tax assessments is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. The forecast does not include changes to tax rates, tax law changes based on the Tax Cuts and Jobs Act. In addition, it assumes that additional taxpayers are impacted each year.

53I.R.C. § 6302(h); Treas. Reg. § 31.6302-1(h); Internal Revenue Manual 20.1.4.2.2 (Feb. 9, 2018).

54I.R.C. § 6656.

55Internal Revenue Manual 20.1.4.2.2 (Feb. 9, 2018).

56Internal Revenue Manual 20.1.4.26.1.1 (Jul. 17, 2015).

57Internal Revenue Manual 20.1.4.26.1.1 (3) and (7) (Jul. 17, 2015).

58In some States, there is a period of time between the date marijuana was legalized for recreational use and when retail sales began.

59* * * 2 * * * Totals rounded for ease of presentation.

60IRS report of cash receipts at Taxpayer Assistance Centers by State was prepared and provided by IRS and could not be independently validated by TIGTA.

61TIGTA did not review the 28 marijuana businesses with no employment tax returns. These businesses may also be unbanked.

62FTD penalties may include a late deposit or be based on payments not made by Electronic Funds Transfer or both.

1See Appendix IV for detailed calculations.

2See Appendix IV for detailed calculations.

1See Appendix VI for a glossary of terms.

2The five-year forecast is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

3To select each of the three statistically valid samples, we used an expected error rate of 35.5 percent, a precision rate of ± 5 percent, and a confidence interval of 90 percent.

4The tax impact amount does not account for some returns reviewed for which the potential I.R.C. § 280E adjustment was reduced or eliminated based on current year net operating losses.

5The totals for each stratum are based on multiplying the unrounded exception dollar rates for each stratum.

6Numbers have been adjusted to be consistent with previously rounded numbers and are supported by statistician calculations.

7The five-year forecast for potential tax assessments is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. The forecast does not include changes to tax rates, possible tax law changes based on the Tax Cuts and Jobs Act. In addition, it assumes that additional taxpayers are impacted each year.

8California potential adjustments were $15,639,249 business expenses, $585,564 I.R.C. § 263A adjustments, and $548,676 accelerated depreciation (includes potential adjustments from I.R.C. § 179). Oregon potential adjustments were $3,700,377 business expenses, $1,017,561 I.R.C. § 263A adjustments, and $469,451 accelerated depreciation (includes potential adjustments from I.R.C. § 179). Washington potential adjustments were $3,497,063 business expenses, $414,051 I.R.C. § 263A adjustments, and $385,610 accelerated depreciation (includes potential adjustments from I.R.C. § 179).

9The potential tax from exceptions was calculated based on actual tax for Form 1120, U.S. Corporation Income Tax Return, corporations; flow-through entities with one, two, or three shareholders or partners; and an estimated tax rate of 10 percent for flow-through entities with more than three partners or shareholders. TIGTA noted that some I.R.C. § 280E adjustments were absorbed by current year net operating losses and resulted in no current year tax effect.

10Adjusted population times weighted average for sample returns.

11As part of our compliance testing, we reviewed shareholder and partner returns for tax calculations and noted a few potential nonfilers and underreported income based on the Schedules K-1 issued and reported. Our I.R.C. § 280E testing results do include the tax impact of unreported income on cases with potential I.R.C. § 280E adjustments based on the actual Schedules K-1 that were issued and reported by the shareholders and partners.

12The confidence interval projections (Tax Year 2016 Potential Tax Assessments Lost) for each State were based on a two-sided 95 percent confidence interval. We are 95 percent confident that the actual total amount for California is between $16,528,021 and $52,659,883; Oregon is between $3,233,439 and $8,963,530; and Washington is between $3,235,950 and $12,409,621.

13The tax impact is the cost of marijuana businesses complying with I.R.C. § 280E.

14The five-year forecast is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

15TIGTA's analysis included a review of 210 returns and 27 IDRS transcripts when returns were unavailable.

16Numbers have been adjusted to be consistent with previously rounded numbers and are supported by statistician calculations using weighted average numbers from the projection. Forecast does not consider changes based on the Tax Cuts and Jobs Act for which guidance regarding I.R.C. § 471(c) is still pending and tax rate changes. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

17The point estimate projections for each State were based on a two-sided 95 percent confidence interval. We are 95 percent confident the point estimate (estimated Tax Year 2016 tax burden) for California is between $29,229,071 and $72,359,357; Oregon is between $9,511,974 and $20,988,449; and Washington is between $17,618,851 and $40,345,045.

18Forecast does not consider changes based on the Tax Cuts and Jobs Act for which guidance regarding I.R.C. § 471(c) is still pending and tax rate changes. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

19Internal Revenue Manual 22.24.1.2 (Jan. 8, 2016), defines how the IRS measures taxpayer burden as follows: “Taxpayer burden is defined as the time or money expended by taxpayers to fulfill their tax compliance responsibilities. The IRS employs Taxpayer Burden Models to estimate taxpayer compliance burden. The models are used to estimate the burden impact of passed and proposed legislative, policy and administrative changes . . . ”

20The five-year forecast is based on multiplying the base year by five and assumes, among other considerations, that economic conditions and tax laws do not change. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

21Rounded for ease of presentation (1,533 x 51.72 percent = 793).

22The 56 percent is based on IRS input attributed to third-party information.

23The final sample results and projections have been rounded for ease or presentation and are based on previously rounded numbers and supported by statistical calculations using weighted average numbers from the projections.

24The 1,533 population for Washington is based on a September 5, 2018, State business list that includes businesses started after December 31, 2016. For Washington, TIGTA reviewed the first 174 sample selections to locate 90 Tax Year 2016 businesses with gross receipts reported to the State of Washington. As such, the State return filing percentage is 51.72 percent (90/174). The filing percentage rate was applied to the overall population of 1,533 to project Tax Year 2016 State filings.

25The tax change calculations are based on allowing 56 percent for cost of goods sold for nonfilers. The tax calculations are based on using actual tax if available and a 10 percent tax rate if undetermined.

26The confidence interval projections were based on a two-sided 95 percent confidence interval. We are 95 percent confident that the actual total amount for Washington is between $954,341 and $12,868,924. This projection is based on using empirical likelihood.

27Forecast does not consider tax rate changes based on the Tax Cuts and Jobs Act. With respect to the number of impacted taxpayers, the projection assumes that additional taxpayers are impacted each year.

See Appendix VI for a glossary of terms.

1Source: Internal Revenue Service Data Book, 2018, available at https://www.irs.gov/pub/irs-pdf/p55b.pdf.

END FOOTNOTES

DOCUMENT ATTRIBUTES
Copy RID