The Industrial Research Assistance Program (IRAP): Are Taxpayers Getting Value for Money?
Managers of federal programs are required to evaluate their programs once every five years. Evaluation results are made public and include an action plan for dealing with any problems identified. Programs found to be inconsistent with government priorities or ineffective in achieving program goals would become candidates for cutbacks as part of the federal government’s Strategic Reviews of spending.
Based on the latest internal evaluation, IRAP would not soon be on the chopping block. The IRAP evaluation reports benefits of $1,568 million and program costs of $137 million, for a net benefit of $1,431 million. Too good to be true? Unfortunately, yes. The benefits identified consist of all of the additional activity induced by the subsidy, but the output and job losses that occur when the government raises taxes or cuts spending, as it must, to finance IRAP are not included. To a close approximation, the net impact of IRAP on overall employment in the economy will be zero, which means that the output and employment induced by the subsidy is irrelevant in a true benefit-cost analysis.
A standard benefit-cost analysis of a business subsidy starts off with the assumption that overall real income should be higher as a result of the policy – taxpayers should be better off as a result of their decision to provide a subsidy. Society has an interest in subsidizing industrial research because the knowledge created benefits society at large, not just the firms undertaking the research. Despite their best efforts, firms cannot prevent the knowledge created from leaking out or spilling over to other firms, which allows them to reduce costs for the general benefit of society.
There is abundant evidence that these knowledge spillovers are important, but there are substantial costs associated with providing assistance to firms, so the existence of spillovers is not sufficient to conclude that the subsidy is good public policy. The most straightforward costs are the expenses incurred by governments to administer the program and the costs incurred by firms in applying for the subsidy and reporting on how it was used. Taxes have a more subtle cost: in addition to affecting the demand for goods and services, the higher taxes required to finance IRAP harm economic performance through their effects on incentives to work, save and invest. Finally, subsidies cause labour and capital to be shifted from their market-determined uses, which also harms economic performance.
My analysis using the standard framework shows that IRAP fails a benefit-cost test – the net loss is about 15% of program spending. The ability to choose recipients of the subsidy likely means that the spillover benefit from IRAP projects is larger than for the average R&D project, but this advantage is offset by very high administration expenses, which amount to about a quarter of the assistance provided to firms.
The requirement to evaluate programs sounds like a great system for bringing evidence to bear on expenditure management decisions, but its usefulness depends on the quality of the evaluations undertaken. In the case of IRAP, the benefit-cost analysis undertaken in the latest evaluation gives a misleading assessment of the program’s contribution to society; a standard benefit-cost analysis shows that taxpayers are losing out by funding IRAP.
 See “Benefit-Cost Analysis of R&D Support Programs” Canadian Tax Journal 60:4 pp. 793-836.