Symbol of the Government of Canada

Common menu bar links

Innovation Canada: A Call to Action

6. Program Mix and Design (continued)

Net Public Benefit

In 2007, the Department of Finance conducted a thorough benefit–cost analysis of the SR&ED program (Parsons and Phillips 2007). It concluded that the public benefit of the program — the part due to the incremental R&D investment stimulated by the SR&ED subsidy and the estimated social return on that incremental investment (i.e., extra output generated in the economy as a whole) — exceeded its full costs. These costs include not only administrative and compliance costs, but also the cost to the economy of having to raise additional tax revenues to finance the credit. This type of benefit–cost analysis relies on key parameter estimates — particularly of the extra R&D performed per dollar of tax credit provided and of the social rate of return on business R&D expenditure — which are subject to considerable estimation error. These measurement problems, together with the difficulty of applying the methodology to other innovation support programs — for example, to business network and internship programs — have led the Panel to conclude that the calculation of net public benefit is not sufficiently precise at this time to permit a benefit–cost ranking of the government's business R&D support programs, nor to fine-tune the mix between SR&ED and the portfolio of direct expenditures.

The estimation of net public benefit of innovation programs is nevertheless an excellent goal and should be included, as it is developed and refined, within the suite of program evaluation tools, as referred to in Chapter 5 on program effectiveness. What can be said, in any event, is that the net public benefit of a support program declines as administration and compliance costs increase, and as the subsidy rate for the program increases beyond a certain threshold. Some level of subsidy is of course needed to stimulate the behaviour that a program is trying to encourage. But an increasing subsidy obviously raises the "cost" side of the equation directly and also steadily increases the likelihood that projects with poorer expected returns will be undertaken, or that businesses with poor prospects will be sustained longer than they otherwise would be. Either of these outcomes reduces the added benefit contribution of the program. At some point, the additional cost of a higher subsidy will exceed the additional social benefit that more generously subsidized projects induce. Were this to occur, the subsidy rate should be reduced. When subsidies associated with several different programs are combined to support a given project, the risk of excessive subsidization obviously increases. This is why it is important to take an overall portfolio approach to a suite of R&D programs and to establish close cooperation between federal and provincial governments in this regard.

As noted in Chapter 5, the SR&ED program is both lauded and criticized. In the latter regard, its complexity results in excessive compliance costs, unpredictability about qualification and firms having to resort to retaining consultants, the cost of which diminishes the benefit. Compliance costs are currently estimated to be approximately 14 percent of the value of credits earned by small firms and 5 percent for large firms.3 At the same time, the program's enhanced benefit for small firms may be richer than necessary, particularly given the compounded effect of provincial credits and other government support programs. Figure 6.4 shows the impact of combining the R&D tax credits provided by individual provinces with federal SR&ED credits. Box 6.3 provides information on how firms are able to "stack" benefits from tax credits and direct assistance. It also illustrates the present limits on stacking.

Figure 6.4 Federal and Provincial Tax Credit Rates (%)

a The federal credit is 35 percent for small CCPCs and 20 percent for other firms. The base for the federal credit is reduced by the amount of provincial credits. (Return to reference a)

b The Quebec credit is paid on wages and salaries plus 50 percent of contracts. The federal–provincial rate is expressed as a percentage of R&D costs eligible for the SR&ED credit. (Return to reference b)

Source: PricewaterhouseCoopers (2011) and calculations undertaken for this review.

Provinces Provincial Tax Credit Federal Plus Provinciala
CCPCs Other Firms
Alberta and British Columbia 10 42 28
Manitoba 20 48 36
New Brunswick, Newfoundland and Labrador, Nova Scotia, Saskatchewan and Yukon 15 45 32
Northwest Territories and Prince Edward Island 0 35 20
Ontario (small/large firms) 10/4.5 42 24
Quebec (small/large firms)b 37.5/17.5 48 27

Box 6.3 Stacking of R&D Support

Firms undertaking R&D in Canada have access to a wide range of federal and provincial support programs, and can frequently obtain financial support for the same project from more than one program. For example, in 2007, some 86 percent of small firms claiming the SR&ED credit made use of at least one other (federal and/or provincial) R&D support program. In that year, approximately 70 percent of all small firms received financial assistance amounting to 40–50 percent of their spending on R&D and almost 10 percent of small firms (about 1600) received assistance amounting to more than 50 percent of their R&D spending.

Within the federal government, the key constraint on "stacking" — that is, the use of two or more programs to support a given initiative — is a Treasury Board requirement that funding from all government sources (federal, provincial and municipal), including tax credits, cannot exceed 100 percent of the eligible costs of a project. However, lower limits are typically established for individual programs. For example, the stacking limit for IRAP projects is 75 percent of eligible costs, which consist of wages, contract expenses and overhead. Programs also limit the proportional amount of funding that can be provided by the program itself to support any one project. In the case of IRAP, this "stand-alone" limit is also 75 percent, but the base is smaller because of the exclusion of overhead expenses.

It is important to bear in mind that IRAP does not automatically provide the maximum subsidy to firms — some firms may receive the maximum, but this would be the exception, given that the average subsidy rate is actually closer to 20 percent.a But as stated above, a substantial number of firms are able to obtain government assistance that exceeds half of their spending on R&D.

The Panel believes that the government should undertake a comprehensive study of the appropriate level of public support for a project, including the maximum allowable stacking rate. Over-subsidization is a potential concern because, beyond a certain point, subsidization no longer generates a net benefit for the economy as a whole.

a This number is based on a web-based survey undertaken in May and June 2011 on behalf of the Panel. (Return to reference a)

The research and consultation undertaken by the Panel point to both the need and opportunity to improve several aspects of the SR&ED program in ways that would increase the net public benefit of the program. It is likely that fiscal savings could thus be obtained while increasing the net benefit that is currently being delivered by the program. The savings would be available to finance the incremental costs of implementing the Panel's recommendations, which are directed primarily toward SMEs, as outlined in other chapters. The following recommendation in respect of the SR&ED program constitutes the Panel's advice with regard to a more appropriate mix and design of tax incentives and direct support.

Recommendation 2

Simplify the SR&ED program by basing the tax credit for SMEs on labour-related costs. Redeploy funds from the tax credit to a more complete set of direct support initiatives to help SMEs grow into larger, competitive firms.

The Vision of the Panel

The SR&ED tax credit is the flagship of federal support for innovation, allowing almost 24 000 firms across all economic sectors and regions of the country to make individual, market-driven decisions about the industrial R&D they need in order to compete and succeed. It is essential for this highly valued program to be made simpler, more predictable and more cost effective in promoting innovation through business R&D spending. Much more information on the costs and benefits of the SR&ED program needs to be provided on a regular basis to enable future program adjustments and to ensure that the program continues to contribute cost effectively to business innovation in Canada.

Getting There

To realize this vision, the Panel makes the following recommendations.

2.1 Simpler compliance and administration — The tax credit benefiting small and medium-sized Canadian-controlled private corporations (CCPCs) should be based on labour-related costs in order to reduce compliance and administration costs. Because the credit would be calculated on a smaller cost base than at present, its rate would be increased. Over time, the government should also consider extending this new labour-based approach to all firms, provided it is able to concurrently provide compensatory assistance to offset the negative impacts of this approach on large firms with high non-labour R&D costs.

2.2 More predictable qualification — Improve the Canada Revenue Agency's preclaim project review service to provide firms with pre-approval of their eligibility for the credit.

2.3 More cost effective — Reduce the amount of SR&ED tax credit assistance by introducing incentives that encourage the growth and profitability of small and medium-sized enterprises (SMEs) while decreasing the refundable portion of the credit over time. Redeploy the savings to fund new and/or enhanced support for innovation by SMEs, as proposed in the Panel's other recommendations.

2.4 More accountable — Provide data on the performance of the SR&ED tax credit on a regular basis to permit evaluation of its cost-effectiveness in stimulating R&D, innovation and productivity growth.

2.5 Phased implementation and consultation — Adopt the proposed changes through a phased-in approach to give the business sector time to plan and adjust smoothly. There should be early consultations with the provinces on the proposed changes, given that they may want to consider adopting the same base as the federal government.

Simpler Compliance and Administration

The base for the SR&ED tax credit currently comprises — in addition to the direct labour cost of researchers — overhead expenses, materials, capital equipment expenses, contracts and payments to third parties such as post-secondary institutions. This Canadian definition of the tax credit base is broader than that in most other countries. Four developed countries (the US, Japan, Australia and Singapore) exclude capital costs from the base. The US also excludes overhead expenses, while the credit in the Netherlands is based on wage costs only.

The calculation of non-labour costs can be complex, particularly for expenditures on capital and materials. This complexity imposes costs on R&D performers by increasing their compliance effort and, particularly for SMEs, leads to increased reliance on third-party consultants to prepare R&D claims. Recall that compliance costs for the SR&ED program are currently estimated to average approximately 14 percent of the value of credits earned by small firms and 5 percent for large firms.

The program could be greatly simplified by moving to a program cost base comprised of (i) direct labour costs, (ii) the labour component of contracts (assumed to be 50 percent, the same as for in-house R&D) and (iii) payments to third parties (such as post-secondary institutions), much of which are labour based. Moving to a labour cost base will reduce compliance costs by eliminating all of the calculations related to capital, materials and overheads.

Since high compliance costs mainly affect SMEs and since switching to a labour cost base would adversely affect large firms in industries where material inputs and equipment are a large fraction of the cost of doing research, it is recommended that this new labour-based approach immediately apply only to the tax credits for the CCPCs. Thereafter, the government should consult with the business community and the provinces on the possibility of applying the approach to all firms. Such universal application should be pursued on the condition that initiatives are developed to address the potentially negative impacts of a labour-based approach on large firms with capital-intensive R&D activity.

More Predictable Qualification

The CRA does not review all filed claims, but instead uses a risk management approach, reviewing some claims for both their financial content and scientific aspects (that is, whether the R&D expenditures are eligible for the tax credit). Businesses must be able to demonstrate that their projects meet the SR&ED eligibility criteria, and must identify the particular costs associated with the eligible project. Businesses therefore face "eligibility risk" should the project not qualify as SR&ED in the view of CRA, or should the costs allocated to the eligible SR&ED project be determined by CRA to be less than originally expected.

Box 6.4 Claiming Overhead Expenses through the SR&ED Program

Firms claiming the SR&ED tax credit currently have the option of claiming overhead expenses (administrative and financial support, utilities, etc.) either directly or through the proxy method. Under the proxy method, firms forgo claiming their actual overhead costs, but instead gross up their wage costs by 65 percent. Almost all firms choose the proxy method, but the reasons for this are unclear to the Panel. For example, for some firms, the reduced compliance costs in using the proxy method may justify its use, even when the actual overhead is greater than 65 percent of wage costs. In other cases, firms may find their overhead costs are less than 65 percent and so the proxy method provides a net benefit. The Panel believes that the proxy method is justified, but it is unclear what the appropriate rate should be. Therefore, the Panel advises the government to undertake a study of the true overhead costs borne by firms for their eligible R&D activities and to establish the proxy rate that best reflects these costs.

Budget 2008 announced $10 million in annual funding to enable the CRA to implement an action plan to improve the administration of the SR&ED program by increasing the agency's scientific capacity and by improving its services to claimants (Department of Finance 2008, p. 87). Since then, CRA has undertaken several initiatives, some of which are still under way. For example, it has increased the number of technical reviewers who determine program eligibility and provide claimant services. The CRA also provides First-Time Claimant, Preclaim Project Review and Account Executive services. These free services are demand-driven and are intended to improve claimants' access to the program.4 The Panel heard evidence from stakeholders that CRA's existing "pre-approval" service is not achieving its objective of improving the speed and predictability of the claims process. One factor contributing to this outcome is that third-party preparers — who are involved in more than half of all claims — may be reluctant to communicate to their clients the availability of CRA's free services.

With its existing services, clients can obtain an informal ruling on eligibility from CRA, which can be useful to firms seeking loans and using tax credits as security. A formal ruling, particularly one that is available at any stage of the R&D process, would evidently be more valuable to R&D performers and would reduce the incentive to make use of contingency fee arrangements with third parties. A more predictable eligibility determination process would also reduce the current tension between CRA and taxpayers when projects are determined not to qualify ex post, which sometimes imposes substantial costs on the company performing the R&D. In addition, the proposed labour base for the credit would remove a great deal of uncertainty for the CCPCs regarding qualifying expenditures.

More Cost Effective

Analysis by the Department of Finance of start-ups created over the 2000–2004 period indicates that, within five years following incorporation, approximately 2 percent of innovative start-ups grow into large firms that continue to undertake R&D. This outcome suggests that the enhanced SR&ED tax credit is a blunt instrument for supporting the "gazelles" among the many start-ups that are eligible for the enhanced SR&ED credit and that a shift in assistance for SMEs to other more targeted programs could be more supportive and cost effective. Furthermore, calculation of the net public benefit undertaken for this review indicates that the SR&ED credit for small companies likely does not generate a positive net benefit because of the high subsidy rate and compliance costs relative to the estimated spillover benefit.

The effectiveness of the SR&ED program would be improved if its parameters could be adjusted to target more of the benefit toward those firms that have a greater probability of achieving high R&D-based growth and profitability. If the program were to adopt the proposed labour cost base set out in Recommendation 2.1, the prevailing 35-percent rate for small companies would have to be increased to compensate for the new narrower cost base of the tax credit. The Panel is not in a position to recommend new rate parameters to be applied to the labour cost base. That will require considerable technical analysis, which is the purview of the Department of Finance. But it is evident that, in readjusting the parameters, there is an opportunity to retarget the benefit. Specifically, the refundable portion of the credit for CCPCs should be reduced over time so that a portion of the benefit would be claimed only if the company is profitable and therefore has tax owing against which the non-refundable portion of the credit could be applied. In other words, all relevant parameters of the SR&ED tax credit regime applied to CCPCs should be changed to enable implementation of the Panel's vision of a new approach to CCPC refundability that better "targets success" and thus contributes to the objective of increasing the number of small, innovative Canadian businesses that become larger, globally competitive enterprises.

The resulting savings will depend on the extent of the shift to only partial refundability for small businesses. As discussed above, the Panel is recommending that these savings be used to finance increases in direct programs that will deliver more "bang for the buck" and serve other needs of SMEs, including providing support in certain areas that are not currently covered under the SR&ED program. The enhanced direct spending programs being recommended are discussed in more detail in Chapters 5 and 7.

The Panel wishes to emphasize that these recommendations — savings from the SR&ED tax credit regime, on the one hand, and increases in direct program spending to benefit SMEs, on the other — are intended to be directly linked. That is, any savings through the SR&ED program for CCPCs must necessarily be contingent upon a commensurate increase in direct program spending benefiting SMEs, particularly to increase the likelihood that a small innovative company will succeed in the transition to becoming a large Canadian enterprise.

More Accountable

To facilitate broader assessments of the SR&ED program, the government should publish data on use of the tax credit, subject to respecting taxpayer confidentiality. More detailed information could also be submitted by taxpayers on their costs of compliance, including in particular the use of contingency fee arrangements and consulting costs in order to assess the effectiveness of Panel recommendations related to reducing complexity and improving effectiveness.

A main stumbling block to date has been the lack of data from taxpayers and the reluctance of CRA to insist that taxpayers provide quality data on the program apart from what is directly relevant to the audit function. Taxpayers would understandably be reticent to provide more detailed information on their projects if there were any concerns about confidentiality and response burden. Confidentiality must be assured, and firms must understand that the additional efforts requested are in their interest. To this end, a dialogue should be undertaken with major industry associations on the rationale for the information collected, including a clear indication of how it would be used and protected.

Phased Implementation and Consultation

The Panel is aware that changing the base for the federal credit has implications for provincial governments with tax credits that (except for Quebec) use the federal base with only minor modifications. Consultation will be required with these provinces before changing the federal base pursuant to Recommendation 2. During its consultations, many stakeholders also advised the Panel that it should take account of the potentially disruptive effect of making changes to programs, especially the larger ones. The government should therefore adopt a phased approach to the implementation of the proposed changes, so businesses will have ample time for transition planning and adjustment. In particular, given that the move to a more performance-based approach for CCPCs would represent a major change in philosophy for the SR&ED regime, the reduction of refundability for existing small businesses should be phased in gradually, in tandem with the gradual redeployment of funds for enhanced direct support measures benefiting SMEs, as recommended in other chapters of this report.

It is also important to note that about 650 newly-created start-ups apply for SR&ED credits for the first time each year. Lack of adequate financing is a constraint on the growth of most start-ups, and the current SR&ED tax credit helps to address this problem through the 35-percent refundable tax credit. Our recommendations will provide start-ups, as well as established SMEs, with a wider range of support to help them bring their innovations to market through an expanded IRAP program, a new commercialization vouchers pilot program, a new concierge service, greater access to a highly skilled and entrepreneurial workforce, enhanced procurement programming, and improved access to risk capital at the start-up and later stages of growth. While the enhanced direct support measures that we are recommending would provide alternative sources of finance for start-ups and a portion of the SR&ED credit would still be refundable, some start-ups might receive a lower level of support through these measures. In view of this, the government may wish to investigate the administrative feasibility and budgetary consequences of providing a fully refundable labour-based tax credit for a fixed number of years for start-ups to ensure continued support for Canadian entrepreneurs who take on the risks of establishing new firms.

3 These estimates are based on a web-based survey undertaken in May and June 2011 on behalf of the Panel. (Return to reference 3)

4 More information on the services the CRA provides to SR&ED claimants is available from the CRA website at: (Return to reference 4)