The economics of claim preparation

1Summary

This paper looks at the preparation of a federal research credit claim as an economic transaction. A business buys a service: the identification, calculation and documentation of a credit under Section 41. The adviser who supplies it can be paid in different ways. The paper asks how each way of paying could affect the work done and the claim filed.

The paper is analytical. It contains no statistics except those read in an official source, and it states plainly where no data exists. The most important gap is simple to state: no IRS form records how a research credit claim's preparer was paid, so no official data shows whether claims prepared under different fee arrangements differ in size, quality or outcome.12

What can be said rests on three things. The first is the structure of the claim, which is built from many judgments.34 The second is the structure of each fee, which determines what the adviser is paid more for. The third is the allocation of risk, which the tax law places on the taxpayer.567 Treasury and the IRS have drawn the same connection between fee structure and claim behavior in their own published reasoning, and so have tax authorities in the United Kingdom.8910

The paper sets out the economic case for success fees in its strongest form before answering it. It does not suggest that advisers who charge such fees act improperly.

2What a claim preparer produces

2.1A calculation built on judgments

The credit is generally 20 percent of qualified research expenses above a base amount, or 14 percent above a three year average under the alternative simplified credit.3 The test of qualified research is applied separately to each business component of the taxpayer.3 Among other things, substantially all of the activities must be elements of a process of experimentation, which the regulations set at 80 percent or more of the research activities, measured on a cost or other consistently applied reasonable basis.4

Each of these rules calls for a judgment. Which projects are business components; which activities within them were experimentation; how much of each employee's time went to qualified services; and which supplies and contract costs count. The regulations require the taxpayer to keep records in sufficiently usable form and detail to substantiate that the expenditures claimed are eligible.4 The product the adviser delivers is therefore not only a number. It is a set of judgments, the number they produce, and the records that support them.

2.2The work behind a claim

The IRS's own estimates give a sense of the work involved. For taxpayers whose burden is shown separately in the Form 6765 instructions, the IRS estimates 10 hours 31 minutes for recordkeeping, 2 hours 25 minutes for learning about the law or the form, and 5 hours 6 minutes for preparing and sending it.1 For tax years beginning after 2025, most claimants must also report at least 80 percent of their qualified research expenses by business component, for up to 50 components.1 For refund claims, the IRS requires the business components, the research activities for each, and the totals of wage, supply and contract research expenses.11

These estimates describe a form, not a full engagement, and they say nothing about how long any particular claim takes. They show that the work has identifiable parts that can be priced.

3Three ways to pay for the work

Advisers are paid in three broad ways, and each rewards something different.

A fixed fee is agreed in advance for a defined scope of work. It does not change with the size of the credit.

An hourly fee is calculated on the time spent. It also does not change with the size of the credit, though it rises with the work done.

A contingent fee depends on the result. Treasury's rules for practice before the IRS define it as any fee based, in whole or in part, on whether a position avoids challenge or is sustained, and say it includes a fee based on a percentage of the refund reported on a return, a percentage of the taxes saved, or one that otherwise depends on the specific result attained.12 The AICPA Code defines it as a fee under an arrangement in which no fee is charged unless a specified finding or result is attained, or in which the amount otherwise depends on the result.13

3.1Figure 1: what each fee structure pays for

Feature	Fixed fee	Hourly fee	Fee calculated on the credit
Changes with the size of the credit claimed	No	No	Yes
Changes with the work done	Only if the scope changes	Yes	Not directly
Paid if no credit is found	Yes	Yes	Usually not
Adviser's income from a judgment that enlarges the claim	Unchanged	Unchanged	Higher
Adviser's income from a judgment that reduces the claim	Unchanged	Unchanged	Lower
Within Treasury's definition of a contingent fee	No	No	Yes12

3.2Mixed arrangements

Many engagements combine elements. A fee may have a fixed part and a percentage part, or a percentage subject to a cap. Because Treasury's definition covers a fee based "in whole or in part" on the result, any percentage element keeps the arrangement within it.12 A cap limits how much the fee can rise, but below the cap each additional dollar of credit still raises the fee. A fixed element reduces the share of the adviser's income that depends on the claim, but does not remove it.

3.3What the buyer can compare

The structure of the fee also affects what a business can compare before it signs. Two fixed quotes can be compared directly. A fixed quote and a percentage quote cannot, because the percentage fee is unknown until the credit is calculated, and the credit is calculated by the adviser who will be paid on it. As an illustration only: a quote of 20 percent costs less than a fixed fee of $25,000 if the credit turns out to be below $125,000, and more if it turns out to be above. The business cannot know which until the work is done.

The last two rows of Figure 1 carry the argument of this paper. Under a fixed or hourly fee, the adviser's income does not depend on which way a judgment falls. Under a fee calculated on the credit, it does.

4How the fee structure meets the work

4.1Search and scope

Every engagement starts by deciding what to look at. A fee calculated on the credit gives the adviser a reason to search widely, because each additional qualifying cost raises the fee. That is a real benefit when a business has under-claimed. The same fee gives no reward for the time spent deciding that a project does not qualify, because that work produces no credit. A fixed or hourly fee pays for both kinds of work equally.

4.2Judgment calls

At each point where the law calls for judgment, a reasonable view could go either way. A fee calculated on the credit pays more each time a judgment favors a larger claim and pays nothing extra for a judgment that favors a smaller one. As an illustration only, with round numbers: on a claim prepared at $100,000, a 20 percent fee is $20,000; if each close judgment is resolved upward and the claim is prepared at $150,000, the same percentage produces a fee of $30,000. Under a fixed fee the adviser's income is the same at either figure.

This is a property of the arrangement. It does not mean that advisers act on it. Many advisers paid on a contingent basis may resolve close questions carefully and against their own financial interest.

4.3Documentation

The regulations require records that substantiate the claim.4 Documentation is costly to produce and does not itself increase the credit. Under a fee calculated on the credit, effort spent on records is paid for out of a fee that is fixed by the size of the claim, not by the quality of its support. The IRS's 2008 guide for examiners, which states that it is not an official pronouncement of the law or of IRS position, described "a growing trend" of prepackaged submissions in support of research credit claims, which "frequently fail to substantiate" the expenses claimed.14 The guide does not link that observation to any fee arrangement, and this paper does not either.

4.4Timing: original returns and refund claims

A credit can be claimed on the original return or later on a refund claim. Treasury has explained why timing matters to fee rules. When it adopted its current contingent fee rule in 2007, it said that restricting contingent fees for preparing returns "supports voluntary compliance with the Federal tax laws by discouraging return positions that exploit the audit selection process", and that it was particularly concerned with contingent fees for refund claims or amended returns "filed late in the examination process".8 It allowed contingent fees for interest and penalty reviews "because there is no exploitation of the audit lottery in these situations".8

In economic terms, Treasury's concern is that a position's value to a fee earner depends on the chance that it is never examined. That chance is not a measure of whether the position is right.

4.5Examination and control

The 2008 examiner guide also said: "If the fee is based on the ultimate determination of the claim amount there may be a limitation on to ability of the taxpayer to resolve the issue. Some fee agreements shift ultimate control over the resolution of the claim to the preparer."14 Where the fee depends on the final amount allowed, the adviser has a financial stake in how an examination is settled, and that stake may differ from the taxpayer's interest in settling quickly or conceding a weak item.

5Who bears which risk

A contingent fee is often described as sharing risk. The law shows which risks are shared and which are not.

If the work finds no credit, an adviser on a contingent fee usually earns nothing, so that risk is shared. If a credit is claimed and later reduced, the consequences fall on the taxpayer:

the taxpayer repays the credit or loses the refund;
interest runs from the date the tax was due, at the underpayment rate;7
a refund claim for an excessive amount carries a penalty of 20 percent of the excess on the person making the claim, unless there is reasonable cause;5 and
an underpayment from negligence or a substantial understatement carries a 20 percent accuracy-related penalty.6

The preparer faces its own penalty only where an understatement is due to a position it knew or should have known lacked the required support, or to willful or reckless conduct. That penalty is the greater of a fixed sum or a percentage of the income the preparer derived from the return or claim.15 No federal rule requires a contingent fee to be returned when a claim is reduced. Under Treasury's definition, an arrangement to refund the fee if a position is not sustained is itself a contingent fee.12

5.1Figure 2: an illustration of the allocation (round numbers, illustrative only)

Item	Claim as filed	After examination
Credit	$150,000	$100,000
Fee at 20 percent of the credit claimed	$30,000	$30,000 unless the contract provides otherwise
Repaid by the taxpayer	None	$50,000 plus interest7
Penalty on a refund claim for an excessive amount, if no reasonable cause	None	Up to $10,000 (20 percent of the excess)5

The figure assumes a refund claim and no refund of the fee. Actual outcomes depend on the facts, the contract and the IRS's findings. It shows that the part of the risk a contingent fee shares is the risk that no credit is found, while the risk that a credit fails on examination stays with the taxpayer.

6What tax authorities have said about the incentive

Several official sources describe the same incentive in their own words.

United States, 2006. The Commissioner of Internal Revenue told the Senate Committee on Finance that taxpayers were filing refund claims, "often marketed to them on a contingency fee basis", to claim additional research credit, and that these claims were "frequently based on unsupportable amounts, nonqualified expenditures, or estimates for which the taxpayers do not have contemporaneous documentation".16
United States, 2024. Proposing to treat contingent fees for returns and pre-examination refund claims as disreputable conduct, Treasury and the IRS wrote that "A practitioner with a direct, financial interest in the tax benefits of a client may be incentivized to increase such tax benefits."9 The proposal has not been finalized.9
United Kingdom, 2024. Explaining why it restricted the payment of R&D tax credits to agents, HMRC said the change would help reduce non-compliance "by reducing the financial incentive for agents to overstate claims".10
Canada, 2012. The federal budget announced a study "to better understand the practice of businesses hiring consultants on a contingency-fee basis" to prepare SR&ED claims, "and to determine whether any action is required".17

These are statements of reasoning, not measurements. None of them reports how often, or by how much, claims prepared on a contingent fee differ from others. Their weight lies in the fact that tax administrators in four countries have reached the same view of the incentive independently.

7The economic case for success fees

The arguments for contingent fees deserve to be stated in their strongest form. When Treasury last revised its rule, commenters said the restrictions were overly broad, interfered with the practitioner and client relationship, and "prohibited some small and middle market taxpayers from appropriately requesting refunds".8

Liquidity. A business short of cash may be unable to pay for a study before it knows whether a credit exists. A fee paid from the credit matches cost to benefit and makes the claim possible.

Risk transfer. The business does not pay for work that finds nothing. For a business unsure whether its activities qualify, that is valuable insurance.

Information. A business cannot easily judge an adviser's quality in advance. An adviser willing to be paid only from results is signaling confidence in its own work, and a client can rely on that signal.

Effort. An adviser paid on results has a reason to look hard for every credit the law allows. Supporters argue that fixed fees can reward doing the minimum.

Access. Many businesses that do qualifying research do not know it. Contingent pricing lets advisers offer to look at no cost, which may bring the credit to businesses Congress meant it to reach.

These arguments describe real advantages to real businesses.

8Answering the case

Liquidity is an argument about when a fee is paid, not how it is calculated. A fixed or capped fee paid in stages, including after the credit is received, meets the same need. Treasury's definition turns on how a fee is calculated.12

Risk transfer covers the risk that no credit is found. It does not cover the risk that a credit claimed fails on examination, which the law leaves with the taxpayer.567

Information is a fair point about the difficulty of judging advice in advance. But a fee calculated on the credit signals the adviser's confidence that a credit will be claimed and paid, not that it will survive examination. The IRS sees the claim long after the fee is earned.

Effort cuts both ways. A fee calculated on the credit rewards effort that enlarges the claim. It does not reward effort spent on records or on deciding that a cost does not qualify, which protects the taxpayer and produces no credit.

Access is the strongest argument. It is an argument for low or deferred upfront cost and for advisers who will look at a business's activities without charge, both of which are possible under fees that do not depend on the size of the credit.

Treasury and the AICPA have each reached the same conclusion for returns and refund claims. Treasury considered and rejected leaving pre-examination refund claims unregulated, because that "would not have met the objective to discourage unduly aggressive tax positions".9 The AICPA Code bars members in public practice from preparing an original or amended return or a refund claim for a contingent fee.13

Our position

Congress should prohibit contingent fees, commonly called success fees, in federal research and development tax credit engagements under Section 41, so that advisers are paid for the work they perform rather than for the size of the claim they produce. Swanson Reed, which publishes this paper, is a fixed-fee R&D tax adviser. It has a direct commercial interest in the outcome: it does not charge success fees, and a ban would remove a pricing model it competes against.

9What the data does and does not show

9.1What exists

The IRS has said it receives thousands of research credit refund claims each year involving hundreds of millions of dollars, and that these claims are examined in a substantial number of cases.11
The IRS Statistics of Income program publishes tables on corporations claiming the credit, but its corporation research credit tables listed on the IRS website run only to tax year 2014.2
Canada collects each SR&ED claim preparer's billing arrangement on the claim form, including a code for a contingency fee calculated as a percentage of the credit.1819 No official publication analyzing that data was found for this paper.
The United Kingdom measures non-compliance through a Mandatory Random Enquiry Programme. HMRC reported that in 2021 to 2022 around half of claims from small and medium enterprises were non-compliant in part or in full.10 It also reported that around half of companies used a specialist R&D agent and that non-compliance, as a value of claims, was slightly higher in claims submitted with one.10 That analysis is by use of an agent, not by how the agent was paid.10

9.2What does not exist

Any official count of US research credit claims prepared on a contingent fee. Form 6765 does not ask, and its instructions make no reference to a preparer.1
Any official comparison of claim size, adjustment rates or penalties by fee arrangement, in the United States or in any of the other countries reviewed for this paper.
Any current IRS statistics on research credit claims by preparer type.

The absence of data is not evidence for or against the position in this paper. It means the question cannot be answered from official records, and that the argument rests on the structure of the fee, the allocation of risk in the law, and the published reasoning of tax authorities. Readers who disagree with the position can test it against the same sources.

References

Internal Revenue Service, Instructions for Form 6765 (Rev. December 2025), "Section G" and "Paperwork Reduction Act Notice", read in full. https://www.irs.gov/instructions/i6765 (accessed 11 October 2026). Back
Internal Revenue Service, SOI Tax Stats: Corporation research credit, table list. https://www.irs.gov/statistics/soi-tax-stats-corporation-research-credit (accessed 11 October 2026). Back
Office of the Law Revision Counsel, 26 U.S.C. 41, Credit for increasing research activities, subsections (a)(1), (c)(4)(A) and (d)(2)(A), United States Code 2024 edition published by the Government Publishing Office. https://www.govinfo.gov/content/pkg/USCODE-2024-title26/html/USCODE-2024-title26-subtitleA-chap1-subchapA-partIV-subpartD-sec41.htm (accessed 11 October 2026). Back
Treasury Department, 26 CFR 1.41-4, Qualified research for expenditures paid or incurred in taxable years ending on or after December 31, 2003, paragraphs (a)(6) and (d). https://www.ecfr.gov/current/title-26/section-1.41-4 (accessed 11 October 2026). Back
Office of the Law Revision Counsel, 26 U.S.C. 6676, Erroneous claim for refund or credit, subsections (a) and (b), United States Code 2024 edition published by the Government Publishing Office. https://www.govinfo.gov/content/pkg/USCODE-2024-title26/html/USCODE-2024-title26-subtitleF-chap68-subchapB-partI-sec6676.htm (accessed 11 October 2026). Back
Office of the Law Revision Counsel, 26 U.S.C. 6662, Imposition of accuracy-related penalty on underpayments, subsections (a) and (b)(1) and (2), United States Code 2024 edition published by the Government Publishing Office. https://www.govinfo.gov/content/pkg/USCODE-2024-title26/html/USCODE-2024-title26-subtitleF-chap68-subchapA-partII-sec6662.htm (accessed 11 October 2026). Back
Office of the Law Revision Counsel, 26 U.S.C. 6601, Interest on underpayment, nonpayment, or extensions of time for payment, of tax, subsection (a), United States Code 2024 edition published by the Government Publishing Office. https://www.govinfo.gov/content/pkg/USCODE-2024-title26/html/USCODE-2024-title26-subtitleF-chap67-subchapA-sec6601.htm (accessed 11 October 2026). Back
Treasury Department and Internal Revenue Service, Regulations Governing Practice Before the Internal Revenue Service, final regulations, T.D. 9359, 72 FR 54540 (26 September 2007), preamble, "Contingent Fees", pages 54541 to 54542. https://www.federalregister.gov/documents/2007/09/26/E7-18918/regulations-governing-practice-before-the-internal-revenue-service (accessed 11 October 2026). Back
Treasury Department and Internal Revenue Service, Regulations Governing Practice Before the Internal Revenue Service, notice of proposed rulemaking, REG-116610-20, 89 FR 104915 (26 December 2024), Explanation of Provisions part B and Regulatory Flexibility Act analysis, page 104924. https://www.federalregister.gov/documents/2024/12/26/2024-29371/regulations-governing-practice-before-the-internal-revenue-service (accessed 11 October 2026; Federal Register docket search on that date showed no final rule). Back
HM Revenue and Customs, Approach to Research and Development tax reliefs 2023 to 2024 (published 30 October 2024), sections 1.2 and 2.1 and Annex B, Theme 5. https://www.gov.uk/government/publications/hmrcs-approach-to-research-and-development-tax-reliefs-2023-to-2024/approach-to-research-and-development-tax-reliefs-2023-to-2024 (accessed 11 October 2026). Back
Internal Revenue Service, IRS sets forth required information for a valid research credit claim for refund, News release IR-2021-203, 15 October 2021, with update of 18 June 2024. https://www.irs.gov/newsroom/irs-sets-forth-required-information-for-a-valid-research-credit-claim-for-refund (accessed 11 October 2026). Back
Treasury Department, Circular No. 230, 31 CFR 10.27, Fees, paragraphs (b) and (c). https://www.ecfr.gov/current/title-31/subtitle-A/part-10/subpart-B/section-10.27 (accessed 11 October 2026). Back
American Institute of CPAs, AICPA Code of Professional Conduct (updated for all official releases through September 2026), ET 1.510.001 Contingent Fees Rule, paragraphs .01 and .03. https://pub.aicpa.org/codeofconduct/ethicsresources/et-cod.pdf (accessed 11 October 2026). Back
Internal Revenue Service, Large and Mid Size Business Division, Research Credit Claims Audit Techniques Guide (RCCATG): Credit for Increasing Research Activities Section 41, LMSB-04-0508-030 (May 2008), introductory note, Chapter I "Background" and Chapter II. https://www.irs.gov/businesses/research-credit-claims-audit-techniques-guide-rccatg-credit-for-increasing-research-activities-section-41 (accessed 11 October 2026). Back
Office of the Law Revision Counsel, 26 U.S.C. 6694, Understatement of taxpayer's liability by tax return preparer, subsections (a), (b) and (e), United States Code 2024 edition published by the Government Publishing Office. https://www.govinfo.gov/content/pkg/USCODE-2024-title26/html/USCODE-2024-title26-subtitleF-chap68-subchapB-partI-sec6694.htm (accessed 11 October 2026). Back
Internal Revenue Service, IRS Commissioner Testifies before Senate Committee on Finance on Compliance Concerns Relative to Large and Mid-Size Businesses, IR-2006-94 (13 June 2006), written testimony, section "Research & Experimentation (R&E) Credit Claims". https://www.irs.gov/pub/irs-news/ir-06-094.pdf (accessed 11 October 2026). Back
Canada Revenue Agency, Archived: Budget 2012, Scientific Research and Experimental Development Program, question 9. https://www.canada.ca/en/revenue-agency/programs/about-canada-revenue-agency-cra/federal-government-budgets/archived-budget-2012/archived-budget-2012-scientific-research-experimental-development-program.html (accessed 11 October 2026). Back
Government of Canada, Income Tax Act, RSC 1985, c 1 (5th Supp), subsection 162(5.3). https://laws-lois.justice.gc.ca/eng/acts/I-3.3/section-162.html (accessed 11 October 2026). Back
Canada Revenue Agency, T4088, Guide to Form T661 (T4088(E) Rev. 20), Part 9, Claim preparer information, lines 950 to 965. https://www.canada.ca/en/revenue-agency/services/forms-publications/publications/t4088/guide-form-t661-scientific-research-experimental-development-expenditures-claim-guide-form-t661.html (accessed 11 October 2026). Back

Cite this paper

Summary

1Summary

2What a claim preparer produces

2.1A calculation built on judgments

2.2The work behind a claim

3Three ways to pay for the work

3.1Figure 1: what each fee structure pays for

3.2Mixed arrangements

3.3What the buyer can compare

4How the fee structure meets the work

4.1Search and scope

4.2Judgment calls

4.3Documentation

4.4Timing: original returns and refund claims

4.5Examination and control

5Who bears which risk

5.1Figure 2: an illustration of the allocation (round numbers, illustrative only)

6What tax authorities have said about the incentive

7The economic case for success fees

8Answering the case

9What the data does and does not show

9.1What exists

9.2What does not exist

References

About this paper

Cite this paper

Summary

1Summary

2What a claim preparer produces

2.1A calculation built on judgments

2.2The work behind a claim

3Three ways to pay for the work

3.1Figure 1: what each fee structure pays for

3.2Mixed arrangements

3.3What the buyer can compare

4How the fee structure meets the work

4.1Search and scope

4.2Judgment calls

4.3Documentation

4.4Timing: original returns and refund claims

4.5Examination and control

5Who bears which risk

5.1Figure 2: an illustration of the allocation (round numbers, illustrative only)

6What tax authorities have said about the incentive

7The economic case for success fees

8Answering the case

9What the data does and does not show

9.1What exists

9.2What does not exist

References

About this paper

Related papers