This Web page has been archived on the Web.

2000 April Report of the Auditor General of Canada

Main Points

6.1 In 1994 the government restricted the period for filing tax credit claims for scientific research and experimental development (SR&ED) to 18 months after the year in which taxpayers incurred them. The Canada Customs and Revenue Agency then received 16,000 additional claims, which represented $2.8 billion in credits for taxation years as far back as 1985.

6.2 Our audit revealed significant inconsistencies in the handling of these claims. These inconsistencies compromised the fairness of the review process. In addition, serious professional differences of opinion arose within the Agency on how to resolve a dispute with a claimant over a significant claim. The Agency did not demonstrate that it had effective mechanisms to resolve this type of difference among its officers on individual cases.

6.3 The lack of clarity about the eligibility of SR&ED projects has resulted in unresolved claims dating back to 1985. The claimed tax credits amount to hundreds of millions of dollars. Since the creation of the SR&ED program, there have been disagreements between claimants and administrators and among administrators on what constitutes an eligible activity or project, how thoroughly a science officer should review a claim and how much documentation is necessary to support a claim. Clearer rules are urgently needed to deliver and administer the program effectively and efficiently.

6.4 The Agency has recognized the need to improve the administration of the program. To this end, after consulting stakeholders, the Minister of National Revenue released an action plan in November 1998. The Agency created a new Directorate and appointed a new head of the program.

6.5 According to our audit, the Agency needs to improve the assessment of the risk that taxpayers' claims are ineligible so that science and audit staff can focus on claims with the highest risk. The Agency manages the risk of ineligibility by conducting completeness checks, risk assessments, audits and science reviews.

6.6 An evaluation of the SR&ED program, done by the Department of Finance and the Agency, may have overstated the impact of the program in generating additional spending on SR&ED. The evaluation also indicated that the program generated only a net increase in Canada's real income of between $20 million and $55 million per year.

Background and other observations

6.7 The federal government encourages research and development in the private sector through tax incentives. Taxpayers receive tax credits once they have incurred expenses on their projects.

6.8 The Agency's information systems show that in 1997 about 11,000 corporations and individual taxpayers claimed over $10 billion in SR&ED expenses and $2 billion in tax credits. Large corporations claimed 85 percent of total expenses.

6.9 A staff of 144 science advisors and 268 financial auditors review claims in 37 tax services offices across the country. Advisors determine whether the work is eligible under the program and auditors examine whether the related costs qualify.

6.10 A key objective of the Agency's 1998 action plan was to reduce uncertainty about the eligibility of science projects. As a result, the Agency and industry proposed a definition of an SR&ED project that, in our view, modifies the level of eligible SR&ED work and expenses and that could result in additional costs in tax credits. If any significant changes were made to the definition of an SR&ED project, they would need to be reviewed by the Department of Finance, which is responsible for the program's policy objectives, costs and related legislation.

The responses of the Canada Customs and Revenue Agency and the Department of Finance are included in the chapter. Their responses describe the action they have taken and will continue to take to address our recommendations.

Introduction

6.11 Research and development is an important factor in improving the productivity of a country's economy and the standard of living of its citizens. Canada's federal government has encouraged research and development in the private sector for many years through tax incentives.

6.12 Incentives currently available for work in scientific research and experimental development (SR&ED) include:

  • accelerated deductions for qualifying capital expenses of scientific research and experimental development incurred in Canada directly by, or on behalf of, a taxpayer's business;
  • investment tax credits for qualifying current and capital expenses incurred in Canada. These can be claimed at rates of 20 percent or 35 percent and are refundable in some cases. Unused investment tax credits can be carried back three years or carried forward 10 years.

6.13 One of the key controls built into the design of these incentives is that the funds must be spent first before any tax credits are received. This also helps to guard against government funding of research and development that the marketplace would not support.

6.14 Several provinces have additional tax incentives for research and development. The federal and provincial tax deductions and credits can reduce the cost of SR&ED by about half.

6.15 The Income Tax Act defines eligible SR&ED as a "systematic investigation or search carried out in a field of science or technology by means of experiment or analysis". Work undertaken to advance scientific knowledge or to achieve technological advancement for the purpose of creating new (or improving existing) materials, devices, products or processes will qualify. Certain work, such as market research or sales promotion, is excluded from the definition. Certain work that directly supports SR&ED and is commensurate with the needs of SR&ED is also eligible.

6.16 The Canada Customs and Revenue Agency - in consultation with the Department of Finance, other federal organizations and industry representatives - developed three criteria to determine the activities that fall within the definition of SR&ED and qualify for the tax incentives. As set out in the Agency's administrative guidelines of the SR&ED program, these criteria are:

  • scientific or technical advancement - the activity must generate information that advances our understanding of scientific relations or technologies;
  • scientific or technological uncertainty - whether a given result or objective can be achieved and how to achieve it are not yet known or able to be determined on the basis of generally available scientific or technological knowledge or experience;
  • scientific and technological content - the activity must incorporate a systematic investigation and be carried out by qualified personnel.

6.17 Exhibit 6.1 lists the objectives of the SR&ED program.

6.18 Administration of the SR&ED program. The Canada Customs and Revenue Agency administers the SR&ED program through 37 tax services offices. Science advisors review the technical aspects of the claims and financial auditors audit the related costs. Taxpayers' activities must qualify as scientific research and experimental development under the law before related costs can be eligible for a tax credit.

6.19 The Agency administers the program with the help of 144 scientists (11 science managers, 104 science advisors and 29 science officers) and 268 financial auditors. It also hires science experts on contract, when necessary, to supplement its in-house expertise. Agency auditors of large files audit SR&ED claims by large corporations. Staff at the Agency's head office develop administrative policy and provide advice to science advisors and auditors in the field.

6.20 Size of the SR&ED program and trends. The Agency's information systems show that in 1997 about 11,000 corporations and individual taxpayers claimed over $10 billion in SR&ED expenses and over $2 billion in tax credits. The number of claimants increased from 6,500 in 1992 to 12,000 in 1994 and has remained at around 11,000 since. Eight percent of the claims accounted for 85 percent of total expenses (see Exhibit 6.2).

6.21 SR&ED action plan. In November 1998, the Minister of National Revenue announced a series of initiatives, known as the SR&ED action plan, to improve the delivery of the program. These initiatives included creating a new Directorate and reorganizing the administration of the program, clarifying roles and responsibilities of the Agency and taxpayers, reviewing the program's rules of eligibility and finding new ways to resolve disputes during the processing of a claim.

Focus of the audit

6.22 The purpose of the audit was to assess how well the government has administered the tax incentive program for scientific research and experimental development since our last audit in 1994. We followed up on the main findings and recommendations that we presented in our 1994 Report, Chapter 32, Income Tax Incentives for Research and Development. More details are provided in About the Audit at the end of the chapter.

Observations and Recommendations

Adjustments Requested by Taxpayers in 1994

The Agency was faced with an administrative nightmare

6.23 Before 1994, taxpayers could claim credits on SR&ED expenses anytime after the year in which they incurred them by requesting an adjustment to the income tax return they had filed for that year. The government decided that the program should encourage new research rather than provide an incentive to reopen tax returns already filed or to recalculate tax credits for expenses of prior years. The February 1994 Budget proposed to restrict expenses qualifying for tax credits to those that taxpayers identify no later than 18 months after the year in which they incurred them. This change became law in June 1994. The amended legislation contained a provision that gave taxpayers until 13 September 1994 to file claims relating to previous taxation years.

6.24 As a result of the amendment, the Agency received 16,000 taxpayer- requested adjustments (TRAs) representing $2.8 billion in tax credits (see Exhibit 6.3). Some of these TRAs involved adding expenses to previous claims, but most of them were from 4,000 new claimants for several taxation years, as far back as 1985. The 16,000 TRAs created a huge backlog in the work of science advisors and financial auditors. These claims are referred to as the "bulge claims".

6.25 The Agency had to delay the processing of the bulge claims. It was unable to quickly increase the number of science advisors needed to process the claims. It also gave priority to current-year claims and strived to process all complete refundable claims within 120 days after receiving them.

6.26 The bulge claims were a challenge for the taxpayer and the tax administrator. The claims were often poorly prepared because of their age, missing documentation and the inexperience of the new claimants and those who assisted them. Files or personnel who could confirm that the activities had been experimental were often no longer available. Even after science consultants interviewed claimants and received more information on the claims, often they could not determine precisely what activities were eligible for a tax credit. Their decisions were often based on judgment and taxpayers' assertions. The review of related expenses also proved difficult because appropriate financial documentation was often missing. The Department of Finance and the Agency reached the same conclusion in their 1997 document entitled The Federal System of Income Tax Incentives for SR&ED: Evaluation Report (see Exhibit 6.4).

6.27 By 31 March 1999, the Agency had processed most of the bulge claims it had received in 1994. Science advisors and auditors disallowed about $1.3 billion in tax credits for the processed claims, which was close to half of the amount claimed.

The processing of the "bulge" claims raised questions of fairness

6.28 In our 1994 Report we had expressed concerns about the impact of the legislative change on the administration of the program. During this audit, we followed up these concerns by examining the retroactive and current claims of 100 claimants. We also interviewed tax lawyers and accountants, financial auditors and science reviewers and looked at representations that various taxpayer associations had made to the Agency.

6.29 In files that we reviewed and in our discussions with financial auditors and science reviewers, we noted that, at times, poor documentation left the tax administrator with a choice - either reject the claim if the taxpayer could not support it or accept the claim in whole or in part, based on the taxpayer's assertions. This trade-off between accuracy and efficiency created inconsistencies in the way files were handled.

6.30 In some cases, the tax services office rejected claims if projects or expenses were poorly documented. In others, financial auditors and science reviewers helped taxpayers to prepare eligible claims.

6.31 Sometimes the first science review found activities ineligible. Some claimants asked for and received a second review, which concluded on occasion that they were eligible. In other cases, the Agency did not offer a second review. By granting a request for a second science review or even a third one, the Agency gave some claimants an opportunity to, in effect, benefit from an additional redress mechanism during the audit of the claim. The Agency did not publicize that it was willing to undertake a second science review. Consequently, many taxpayers might have been deprived of the opportunity to ask for one.

6.32 We found a case where three science opinions had rejected a claim. The unsatisfied claimant asked for a fourth opinion that, this time, resulted in the approval of the claim. The Agency allowed the tax credit.

6.33 The Agency's databases could not provide us with a listing of claims where the claimant had asked for and received a second science review. However, the Agency informed us that there had been a limited number (20) of such cases. We were also unable to determine whether some claimants had asked for but had not received a second science review.

6.34 In January 2000, the Agency released guidelines on how to provide second opinions to claimants who dispute the science review. It is also considering an alternative mechanism to resolve disputes, which would supplement the current appeal process.

6.35 We noted other inconsistent approaches to resolving disputes over the eligibility of science activities. In one case, two consultants reviewed unrelated science areas of a claim. The claimant disagreed with one consultant whose conclusions would have disallowed most of the area of the claim he had reviewed. Rather than obtain a second science review, the Agency resolved the claim by applying the outcome of the other, more favourable review to both areas of the claim. The claimant received about $500,000 more than the science report had indicated.

6.36 In another case, a science advisor found various projects of a claimant ineligible. The audit report noted that only a portion of the allowed credit was based on reviewed work. According to the file, the Agency allowed an additional tens of millions of dollars in SR&ED expenses to settle the claim. As a result, the claimant received millions more in tax credits.

6.37 In our view, there were significant inconsistencies in the handling of these claims, which compromised the fairness of the review process.

There were serious internal differences of opinion on a significant claim

6.38 Serious differences of opinion arose within the Agency on how to resolve a large claim.

6.39 At the time this claim was filed, the taxpayer had poorly documented the science work and expenses. The staff at a tax services office spent over 10,000 hours on the audit and $300,000 on a science report. It was the auditors' professional opinion that they had to deny most of the claim because of the lack of substantiation by the taxpayer. The Agency's head office spent $300,000 more on a second science report. The Agency informed us that, in keeping with its functional authority, the head office took over responsibility for settling the file. The head office then approved tens of millions of dollars more in tax credits than the audit from the tax services office and the region had supported.

6.40 This file raises a number of serious concerns about the process that was followed to resolve the dispute. The head office made first and second offers and then a final one to settle the claim; each offer significantly exceeded the amount that auditors had supported. The file did not document the reasons for the first and second offers, which the head office had made before the completion of the science report and the financial audit. As for the final offer, we have concerns about how it was reached.

6.41 Included in the final offer were tens of millions of dollars in tax credits for costs that the taxpayer had paid to another taxpayer and then claimed. The other taxpayer invoiced the taxpayer for the costs of SR&ED work and other work. The other taxpayer was not required to indicate what was eligible SR&ED work.

6.42 The taxpayer claimed the costs of what it thought was the other taxpayer's SR&ED work as well as an additional amount. This claimed amount was substantially higher than the amount the Agency had allowed in the other taxpayer's SR&ED claim for the same work. (At that time, the legislation permitted both taxpayers to claim for the same SR&ED credits.)

6.43 Auditors at the tax services office determined, and the head office agreed, that a significant portion of the costs claimed by the taxpayer for the work of the other taxpayer did not qualify for a tax credit. Based on information in the Agency's files, including the audit of the other taxpayer's claim, the auditors at the tax services office thought that it would be reasonable to allow the taxpayer a credit of tens of millions of dollars for the other taxpayer's work done in Canada. The head office finally allowed twice that amount.

6.44 According to the files at the tax services office, the head office told the auditors that, owing to policy, they could not examine the other taxpayer's SR&ED claim because the taxpayer and the other taxpayer were dealing at arm's length. The auditors had to accept the amount determined by the taxpayer for work done in Canada.

6.45 In our view, since the taxpayer had estimated the SR&ED work of the other taxpayer, the Agency had an obligation to examine and verify how this was done. It was inappropriate for the head office to disregard the results of its previous audit and assessments of the other taxpayer's SR&ED claim as well as other documentation on file in determining the credit allowed to the taxpayer.

6.46 The head office also accepted, without adequate verification, other expenses that auditors had rejected for lack of substantiation. This resulted in an additional tax credit of tens of millions of dollars.

6.47 The head office has the authority and the expertise to review decisions of tax services offices, regional offices and consultants, and to override incorrect ones. However, we believe that the resolution of any major internal dispute over the application of administrative policies needs to reconcile the opposing positions.

6.48 In this case, we would have expected the Agency to analyze in depth the history of the disputed claim. The lessons it would have drawn from this experience could have been used to better plan, manage and monitor the administration of the SR&ED program. We have learned that the Agency began an internal review following our audit.

6.49 The Canada Customs and Revenue Agency should revise its procedures to ensure that SR&ED claims are processed in a fair and consistent manner. The Agency should establish a process to reconcile internal differences of opinion on the application of administrative and audit policies related to verifying SR&ED claims.

Agency's response: The Canada Customs and Revenue Agency (CCRA) strives to ensure that all of its services, including the SR&ED program, are administered in a fair and consistent manner. This objective is articulated in our Fairness Pledge, published in 1998, which includes a specific commitment to apply laws consistently and equitably. Furthermore, the CCRA's objective for the SR&ED program, as set out in Exhibit 6.1 of this chapter, reinforces our commitment to consistency and fairness.

We acknowledge that there is room for improvement and this is reflected in the SR&ED Action Plan announced in 1998, which is captured in Exhibit 6.5. Of particular note is the action relating to a code of rights, obligations and responsibilities, the action to establish committees to develop standards and guidelines and the action to clarify documentation requirements. All of these actions are intended to bring about greater fairness and consistency.

The CCRA is making significant progress in implementing the SR&ED Action Plan and will take additional measures as necessary to reinforce fairness and consistency in our actions and to ensure we are able to effectively demonstrate to the Auditor General and others our record in being fair and consistent.

The CCRA's accountability framework, internal communications practices and decision-making fora provide effective means to address differences of opinion. We do, however, acknowledge that a significant difference of opinion did occur within the Agency respecting one file. In this one instance, existing mechanisms were not fully used to resolve the difference of opinion. The CCRA has already taken action to address this specific occurrence and will reinforce within the Agency the importance of using available mechanisms to resolve differences of opinion.

Challenges of the SR&ED Program

6.50 Although the pressures associated with reviewing and resolving the bulge claims have subsided, the program continues to face major challenges.

Clearer eligibility rules are needed to improve the administration of the program

6.51 Since the creation of the program, there have been disagreements between claimants and administrators and among administrators on what is an eligible activity or project, how thorough the science review should be, and how much documentation is necessary to support the claim. In our view, uncertainty about what makes a claim eligible can adversely affect relations with claimants. Resolving disputes over eligibility drains resources, jeopardizes the program's incentive objective and puts at risk large amounts of tax revenue.

6.52 Observers of the program have suggested that better training of science reviewers would reduce uncertainty. We believe that this would only lessen inconsistencies and that clearer eligibility rules are needed to reduce uncertainty.

6.53 In November 1998, the Agency released an action plan to improve the delivery of the SR&ED program (see Exhibit 6.5). A steering committee, with the Agency as a member, is overseeing its implementation. Several actions of the plan were aimed at reducing the number of disputes between claimants and the Agency. Other actions were aimed at reducing uncertainty by developing new interpretation guidelines and standards and clearer documentation requirements. We agree that clearer rules on eligibility and documentation are needed to deliver the program effectively and efficiently. Clear rules are a key concern in a tax system that relies on self-assessment and voluntary compliance.

6.54 The Agency had tried to clarify the administrative rules on the eligibility of science work by revising the program's information circular (IC86-4R) in 1997. It stopped this exercise to study further the many suggestions and complaints of stakeholders.

6.55 Under the action plan, several industry subcommittees looked at SR&ED eligibility in their respective sectors. A draft paper entitled "SR&ED Project Definition - Principles" reflected their views. In January 2000, it was presented to industry and representatives of industry associations as a collective opinion of industry and the Agency. The paper indicated that these principles would help to apply the definition of an SR&ED project as set out in the information circular.

6.56 This definition deviates from the current practice of defining, documenting and reviewing claims. In our view, the paper's definition modifies the level of eligible SR&ED work and expenses and could result in additional costs in tax credits.

6.57 The Department of Finance is responsible for the policy objectives and legislation of the program. Therefore, if any significant changes were made to the definition of an SR&ED project, the Department would need to determine their intent and potential cost.

6.58 The Canada Customs and Revenue Agency should clarify the administrative rules on eligible work under the SR&ED program. The Department of Finance should examine the legislative rules to ensure that they adequately reflect government tax policy.

Agency's response: As indicated in the preceding paragraphs, the CCRA has already launched a process to clarify the administrative rules governing the work eligible under the SR&ED program. At the January 2000 Minister's Conference on SR&ED, the need for clearer eligibility rules was identified as a priority for the future. The CCRA believes that the meaningful consultations under way with industry representatives will result in administrative rules that are clearer and more relevant to specific sectors, which will facilitate voluntary compliance.

The CCRA is working closely with the Department of Finance in this undertaking to ensure that new administrative rules for eligible work are consistent with government tax policy. This will also expedite the development of any legislative amendments that may be required to support the ongoing effectiveness and integrity of the SR&ED program.

Department of Finance's response: The Department of Finance fully supports this recommendation. The Department and the CCRA work closely together on the SR&ED tax incentive program. For example, a number of formal and informal mechanisms are in place to ensure that the SR&ED tax incentives are monitored and managed appropriately. These mechanisms include liaisons between officials responsible for policy or administration of the SR&ED tax incentive program, the SR&ED Interdepartmental Working Group and the CCRA Advisory Committee on SR&ED.

As indicated in the 2000 budget, the federal government will consult with industry representatives to ensure that the guidelines on software development, in particular internal use software, both reflect government policy and provide clarity and certainty of application for compliance purposes as well as administration. Once consultations are completed, the government will determine whether amendments to the Income Tax Act are required.

Unresolved disputes over eligibility of science work have reached hundreds of millions of dollars in tax credits

6.59 On 31 March 1999, the Agency had an inventory of 7,680 outstanding claims dating back to 1985 that involved SR&ED tax credits of $5 billion. Hundreds of millions of dollars were included in claims designated as "unworkable". The unworkable claims from the financial and telecommunication sectors were outstanding because of unresolved disputes between claimants and the Agency. We examined these claims.

6.60 Financial institutions are challenging the decision to disallow hundreds of millions of dollars in tax credits. Financial institutions filed SR&ED claims in 1990 dating back to 1985. By 31 March 1996, some institutions had claimed SR&ED expenses of billions of dollars for information technology and software development. This represented tax credits of hundreds of millions of dollars.

6.61 In 1992, science advisors reviewed these claims and reached different conclusions about their eligibility. Some in one tax services office (TSO) approved the claims of a financial institution, whereas some in another TSO rejected similar claims of a different financial institution. The Agency formed a panel of experts to review claims by financial institutions for software development expenses and to ensure that all TSOs would give them consistent treatment. In 1996 the panel concluded that only 10 percent of the claimed expenses were eligible.

6.62 In the 1995 Budget, the Minister of Finance noted that the government was reviewing the income tax rules and administrative guidelines on the eligibility of SR&ED in information technology. He announced that during the review, no bank or financial institution could claim a tax credit for its activities in information technology. On 12 December 1995, the Minister tabled a Notice of Ways and Means Motion to ensure this.

6.63 Following the Agency's assessment that much of the work claimed by financial institutions would not qualify for SR&ED incentives, the Department of Finance concluded that there was no need to change the law as had been proposed in the Notice of Ways and Means Motion. The Department indicated that revising guidelines and enhancing audit coverage would resolve any difficulties.

6.64 In February 1997, the Agency revised its administrative guidelines on the eligibility of software development.

6.65 In January 2000, however, some financial institutions were still challenging the Agency on their information technology claims for billions of dollars in SR&ED expenses, which represented a possible tax credit of hundreds of millions of dollars. The Agency had rejected amounts claimed by many other institutions that had not challenged its assessment.

6.66 Telecommunication companies are challenging the decision to disallow hundreds of millions of dollars in tax credits. Telecommunication companies have submitted SR&ED claims for the development of management information systems dating back to 1991. The Agency has estimated that outstanding claims amount to billions of dollars in expenses, which represents hundreds of millions of dollars in tax revenue that is at risk.

6.67 The Agency carried out science reviews of the 1990-91 claims of two telecommunication companies. These reviews reached inconsistent conclusions about the eligibility of the science work. In 1997, the Agency completed a national review of the 1993-94 claims of some telecommunication companies and rejected a large percentage of them. The claimants have challenged the outcome of this review.

6.68 There have been unsuccessful attempts to resolve these claims. In 1997 some companies drafted an agreement on filing and auditing SR&ED claims. The Agency did not accept this agreement and prepared an explanation of its position on some general issues concerning the eligibility of claims by telecommunication companies. The companies rejected this explanation. Since 1994 they have continued to submit annual SR&ED claims for similar types of activities. At the time of our audit, an independent third party had begun a second national review of these claims.

6.69 The Canada Customs and Revenue Agency, with the departments of Justice and Finance, should develop and implement a strategy to resolve outstanding claims by the financial and telecommunication sectors.

Agency's response: The CCRA shares the Auditor General's concerns about the inventory of outstanding claims in the sectors identified. Given the amounts in question, the CCRA and the departments of Justice and Finance are proceeding in a prudent manner by undertaking additional reviews and exploring possible options for resolution. In addition, a number of formal challenges have been launched by claimants in these sectors and the normal resolution processes are being followed by the CCRA and the Department of Justice. We are seized with the importance of resolving these outstanding claims and are committed to a fair and correct resolution as early as practical. However, we recognize that the outstanding claims must meet the three well-established criteria - scientific or technological uncertainty, advancement and content - in order to qualify for the SR&ED tax incentives.

It should be noted that elements of the Action Plan focus on preventing disputes through better and earlier communication, establishing a standard mechanism for dealing with disputes and resolving outstanding disputes about past claims.

Department of Finance's response: The Department of Finance will be assisting the CCRA in order to achieve this result. As indicated in the 2000 budget, the government is committed to rigorously applying the three well-established criteria to address the backlog of SR&ED claims related to information technology. These criteria are scientific or technological uncertainty, advancement and content.

Management of Compliance Risk

6.70 In 1997, eight percent of claims represented 85 percent of all SR&ED expenses claimed that year (see paragraph 6.20). The rest of the claims were for small amounts. This and the fact that the Agency strives to process all refundable tax credit claims within 120 days after receiving a completed claim present challenges to managing compliance risk (the risk that claims may not comply with the eligibility criteria of the program).

6.71 There are no standard criteria for assessing risk. We found that the head office had not developed standard criteria to assess risk and to identify high-risk claims for review and audit. As a result, tax services offices developed their own criteria. Some offices relied mainly on financial criteria; others used a mix of financial and science criteria.

6.72 Our review of program data showed how the Agency had managed risk in recent claims. By May 1999, the Agency had received 29,000 claims for taxation years from 1996 to 1999, which involved $5 billion in tax credits. It accepted 10,000 claims ($460 million in credits) based on a screening without further verification and it audited 13,000 claims ($1.3 billion in credits). The Agency had yet to process the remaining 6,000 claims ($3.2 billion in credits) in its inventory.

6.73 According to program data, 29 percent of completed audits, or 100,000 audit hours, did not lead to claims adjustments. This suggests that the risk assessment criteria may not be successful in identifying risky files for review and audit.

6.74 Several tax services offices indicated to us that during peak periods, they accepted more refundable claims as filed if they could not process them within 120 days after receiving them.

6.75 Most claims from large corporations are audited several years after they are filed. Agency auditors of large files audit SR&ED claims by large corporations. Science advisors review these claims when auditors start their work.

6.76 We looked at the claims of 25 of the 100 largest claimants and learned that most science reviews and audits were not current. They covered only taxation years between 1988 and 1993.

6.77 Our audit found that science advisors had reviewed the claims differently. We found that some had reviewed a large corporation's claim year after year, even when past reviews had identified no problems with eligibility. Others reviewed an unrepresentative sample of projects from the claim; unreviewed projects were accepted as filed. In some tax services offices, advisors and auditors were proposing to adopt project-sampling methods that would allow them to apply the results of their review and audit of one claim to other claims over several years.

6.78 The Canada Customs and Revenue Agency should adopt standard criteria to assess the compliance risk of claims. The Agency should also ensure that it processes large claims in a timely and consistent manner.

Agency's response: The CCRA, in all its programs, is committed to continuously improving its management of risk. With regard to the SR&ED program, the Auditor General recognized in his 1994 Report that all claims are subject to a desk review by a science advisor and an auditor when they are filed. This review is designed to determine whether the activities being claimed meet basic eligibility requirements and whether the costs are reasonable. Since then, we have included standard risk criteria in the annual workplans and have taken other measures, such as the provision of detailed compliance rates by strata to the field, inclusion of risk assessment practices in training materials and other guidelines, and sharing and discussion of best practices.

The general framework for assessing risk is included in the recently released Stakeholders Report for the program, Action Plan - Consultation to Implementation. The cornerstone of risk assessment in the SR&ED program is "Knowing Your Client". With the implementation of the client-centered approaches called for in the Action Plan, our ability to assess risk will be further enhanced.

The CCRA is committed to processing large claims in a timely, consistent manner. The objective is to process all large claims within 365 days of receiving a complete claim. We recognize that the timely delivery of these claims represents a significant challenge. There are several measures in the Action Plan, including "process audit" and "Account Executive Service", that address this issue.

In addition, the resources for processing large claims were recently consolidated in the SR&ED program. This will allow the CCRA to better focus its resources on the science, technology and incentive aspects of the program and improve timeliness, while maintaining the program's fiscal integrity. Furthermore, these measures will establish greater uniformity and consistency in the way the program is delivered across the country. For instance, one of the key functions of the National Technology Sector Specialist initiative, which was recently implemented, is to ensure consistency in program delivery.

Impacts of the SR&ED Program

A joint evaluation assessed the impacts of the program

6.79 In 1997 the Department of Finance and the Agency completed a joint evaluation of the program. Here are two main findings:

  • Federal SR&ED tax incentives encouraged participants to incur more expenses in research and development than they would have without the program. Participants in the program had spent 38 cents over and above every dollar in federal incentives that they received.
  • The program generated a net increase in Canada's real income of between $20 million and $55 million per year.

6.80 In a survey of claimants, the evaluation determined the additional spending generated by the program. The Department indicated that there were advantages and disadvantages to using this method to obtain information. The main advantage of a survey is that it explains a subject in more detail. Among its disadvantages, respondents have a natural tendency to overestimate the impacts of policies that are beneficial to them. Claimants were asked if investments would have been different without federal SR&ED tax incentives. It is unclear whether taxpayers attributed all additional spending on SR&ED to federal tax incentives or to both federal and provincial incentives.

6.81 Given the uncertainty about the extent of additional spending generated by federal tax incentives, sensitivity analyses would have been appropriate to estimate the ratio of additional spending on SR&ED to tax revenue forgone. For example, assuming that survey respondents attributed all additional spending to federal and provincial incentives, the tax revenue forgone would have been higher and the cost-spending ratio would have been lower than 1.38. Provincial tax incentives represent between 20 percent and 30 percent of total tax incentives.

6.82 Additional spending on SR&ED directly reflects the impact of the program on the economy. An overstatement of additional spending would overstate the impact. The evaluation revealed that the economic benefits of the program exceeded the costs by a relatively small amount, between $20 million and $55 million. In other words, the SR&ED program was marginally cost-effective.

6.83 To estimate the impact of the program on the economy, the evaluation used conservative estimates of benefits to persons or businesses other than those paying for research and development (spillover benefits).

6.84 In our November 1999 Report, Chapter 19, Investing in Innovation, we commented that spending on research and development is not the only factor that determines the rate of growth of productivity in the economy; and it may not be the most important one. We called on Industry Canada to thoroughly examine the government's main direct funding programs for research and development. In our view, it is important for the government to look at how federal tax incentives promote growth and innovation.

6.85 In future evaluations of the SR&ED program, the Department of Finance needs to consider the following questions.

  • Why, among the G-7 countries, does Canada have the second lowest ratio of total spending on research and development to gross domestic product? This is despite the fact that Canada has the most generous tax incentive for research and development.
  • How much should government support research and development? Should the federal government invest some of its SR&ED funds in other activities that promote innovation?
  • Why is the number of claimants not growing?

6.86 The Department of Finance, with Industry Canada, should look at the respective roles of tax incentives and program incentives within the overall federal strategy of providing assistance for research and development.

Department of Finance's response: SR&ED investment tax credits are the most important element of the federal strategy of providing assistance for research and development (R&D). Tax credits are provided in recognition of the fact that the benefits of SR&ED accrue not only to the SR&ED performer but also to other participants in the economy. These credits are designed to encourage firms to perform SR&ED in Canada given the key role of SR&ED in the emerging new economy.

The Department will continue to review the program to ensure its effectiveness in the context of the overall federal strategy of providing assistance for R&D, including any expenditure programs administered by other departments, such as Industry Canada. The Department of Finance completed an evaluation of the SR&ED tax incentive program, which was published in 1997.

Ongoing review is important to ensure that the SR&ED tax incentive program is effective in meeting its policy objectives. For example, as indicated in the 2000 budget, the federal government will consult with industry representatives to ensure that the guidelines on software development, in particular internal use software, both reflect government policy and provide clarity and certainty of application for compliance purposes as well as administration. The federal government must ensure that tax credits are only provided to those activities to which the SR&ED program was intended to apply.

Conclusion

6.87 Our audit revealed a number of significant problems in the administration of the program.

6.88 The handling of bulge claims compromised the fairness of the review process. Serious professional differences of opinion on how to handle a significant claim were not resolved. The Agency requires a mechanism to reconcile internal differences of opinion.

6.89 There is uncertainty about what constitutes an eligible activity or project, how thoroughly a science advisor should review a claim and how much documentation is necessary to support a claim. This uncertainty needs to be eliminated. In a tax incentive program such as the one for SR&ED, the legislative and administrative rules must define as completely as possible who and what qualify for the tax credit. This helps to ensure that only eligible SR&ED work is claimed and allowed. Clearer rules are urgently needed to deliver the SR&ED program effectively and efficiently.

6.90 Amounts in disputes over the eligibility of science work in the financial and telecommunications sectors have reached hundreds of millions of dollars in tax credits. The Agency, with the departments of Justice and Finance, needs to develop and implement a strategy to resolve these disputed claims.

6.91 The Agency and industry proposed a definition of an SR&ED project that would modify the level of eligible SR&ED work and expenses and could result in additional costs in tax credits. The Department of Finance is responsible for the program's policy objectives, costs and related legislation. Therefore, if any significant changes were made to the definition of an eligible SR&ED project, the Department would need to determine their intent and potential cost.

6.92 There is a need to strengthen the methods of assessing the risk that claims may not comply with the eligibility criteria of the program.

6.93 An evaluation by the Department of Finance and the Agency revealed that the program generated a net increase in Canada's real income of between $20 million and $55 million per year. In our November 1999 Report, Chapter 19, Investing in Innovation, we commented that spending on research and development is not the only factor that determines the rate of growth of productivity in the economy; and it may not be the most important one.

About the Audit

Objective

The objective of the audit was to assess how well the government has administered the tax incentive program for scientific research and experimental development.

Scope and Approach

In our examination, we determined:

  • how the Agency handled 16,000 retroactive claims that it had received in 1994;
  • whether there were problems that impeded the administration of the program;
  • how the Agency managed the risk that claims for tax credits may not have been eligible;
  • what is known about the program's impacts.

We conducted our audit work in the Canada Customs and Revenue Agency, its head office and seven of its tax services offices and in the Department of Finance.

We examined various Agency documents and representations that various taxpayers associations made to the Agency. We interviewed science advisors, financial auditors, and private-sector tax advisors.

We reviewed the evaluation of the SR&ED program that the Department and the Agency carried out in 1997.

We analysed SR&ED program data.

We examined the files of 100 (25 large and 75 small) claimants, many of whom had several multiyear claims. The total value of these claims exceeded $3 billion in tax credits. Our examination covered bulge claims and recent claims. We selected these claims based on our judgment. Therefore, the results cannot be generalized statistically to the broader population of similar claims. We focussed only on the actions of government officials as they administered these claims. We did not audit the claimants and we made no comment on their actions.

Criteria

To effectively administer the program, the Agency was expected to:

  • clearly articulate the objectives of the program and balance them with the requirement that it prudently disburse program funds;
  • have clear rules and guidelines on eligibility to guide claimants and their reviewers;
  • support and document conclusions of reviewers;
  • resolve claims efficiently and effectively and ensure that taxpayers receive consistent treatment; and
  • have appropriate procedures in place to manage the risk that claims may not be eligible.

Audit Team

Assistant Auditor General: Shahid Minto
Principal: Barry Elkin
Directors: Richard Gaudreau, Brenda Siegel and Abid Raza

For information, please contact Barry Elkin.