Ottawa’s Research Subsidy Reforms Underwhelming

Published in the Financial Post.

Though they were largely lost in the political storms that broke over Ottawa just before Christmas, in mid-December the federal government announced long-awaited reforms to the scientific research and experimental development (SR&ED) program. The government’s largest R&D program, SR&ED provides a refundable 35 per cent tax credit to privately owned small and medium-sized enterprises (SMEs), as well as a non-refundable 15 per cent tax credit to larger firms. It pays out almost $4 billion in benefits a year, with slightly more than half going to SMEs.

In the past round of public consultations on SR&ED, Ottawa raised expectations for fundamental change by asking if all firms should receive the same subsidy rate and if the credit for larger firms should also be refundable. But the changes announced last month were underwhelming. Ottawa opted for marginal measures — that are nevertheless costly.

A small but positive change was to extend the 35 per cent refundable credit to small Canadian businesses that have gone public. Right now, small firms going public see their cash flow fall sharply when the 15 per cent, non-refundable credit kicks in. That prompts some firms to forgo the benefits of going public, which harms their growth prospects. To complement this initiative, the government raised both the R&D spending eligible for the refundable credit and the size of firm able to access it.

The biggest change was to reverse a 2014 decision and make capital expenditures an eligible expense, a measure I estimate will cost $1.6 billion over six years. That will help boost R&D, but in my view the money would have been better spent financing refundability for large firms.

Why? Two main reasons. First, in any year big firms claim less than half the 15 per cent credit they’ve earned simply because they don’t have enough taxable income. Even if they do eventually claim it all, the delay reduces its value to them. And, second, the U.S. taxes its multinationals on their worldwide income, so some of the SR&ED credit effectively ends up being a gift to the U.S. Treasury. The tax credit increases the net income of U.S.-controlled firms, which raises their U.S. tax liabilities. Refundable credits don’t have this effect, however. They are considered grants, which don’t trigger a “Treasury transfer.”

Phasing-in refundability over 10 years would have cost about the same as making capital spending an eligible expense. But because it would make the value of the credit more predictable, its impact on R&D would be bigger. And for U.S.-controlled firms operating in Canada preventing the Treasury transfer would raise the value of the credit by as much as 40 per cent.

Governments subsidize R&D because its benefits are not confined to the firm that performs it. Some of the knowledge created inevitably spills over to other firms, allowing them to benefit without performing the R&D themselves. My research shows that these knowledge spillovers are greater for larger firms than smaller ones, so rebalancing support in favour of large firms would provide more R&D bang for the buck.

Rate rebalancing would also make it more likely that subsidized R&D projects will be commercialized. Subsidies lower the “hurdle rate” for investment, giving the go-ahead for some lower-quality projects with less potential for commercialization. Reducing the subsidy rate for smaller firms would therefore raise both the average quality of projects and the share commercialized. These gains could be realized at no additional fiscal cost by setting a single subsidy rate of 21 per cent (versus the current 35 per cent for smaller firms).

The most common justification for favouring small firms is to compensate for the difficulties they face accessing risk capital. But trying to fix this problem with an extra subsidy for all firms doing R&D is wasteful and unnecessary given the support available to small firms from the Business Development Bank of Canada.

The government is still thinking about a potentially productive simplification of SR&ED. The best approach here would be to break the link between receiving refundable tax credits and filing a tax return. The credits are effectively grants, so there is no real need to tie them to a tax return. Untying them would shorten the delay between performing R&D and receiving the subsidy, which would increase the subsidy’s value to firms and strengthen their response to it.

The government is also still thinking about making more kinds of innovation spending eligible for SR&ED credits. It should drop that idea. It’s unlikely the SR&ED subsidy rates would provide the right amount of support for other innovation activities. If more activities deserve subsidies, they should get separate programs with separate subsidy rates.

On balance, Ottawa’s latest reform of SR&ED is timid. Fortunately, this timidity did not extend to implementing a special low tax rate on income from intellectual property. Designed properly, this could be a cost-effective way to promote R&D and address Canada’s poor performance commercializing and scaling up inventions developed from it. Details are to come — after the political storms pass, presumably.

John Lester, fellow-in-residence at the C.D. Howe Institute, was director of research for the Expert Panel Review of Federal Support to Research and Development (2011).