Effectiveness of RampD tax incentives in small a

Authors: Rufin Baghana Pierre Mohnen

Publish Date: 2009/04/08

Volume: 33, Issue: 1, Pages: 91-107

PDF Link

Abstract

In this paper we evaluate the effectiveness of RD tax incentives in Quebec using manufacturing firm data from 1997 to 2003 originating from RD surveys annual surveys of manufactures and administrative data The estimated price elasticity of RD is –010 in the short run and –014 in the long run with slightly higher elasticities for small firms than for large firms We show that there is a deadweight loss associated with levelbased RD tax incentives that is particularly acute for large firms For small firms it is not sizeable enough to suppress the RD additionality at least not for quite a number of years after the initial tax change Incremental RD tax credits do not suffer from this deadweight loss and are from that perspective preferable to levelbased tax incentivesMany countries rely on a policy of RD tax incentives to spur RD in the private sector In some countries the budget allocated to this policy is substantive Canada is one of these countries In 2006 the Canadian federal government spent about 3 billion a year on its RD tax incentives program Finance Canada 2007 Among the Canadian provinces Québec has been the most generous in its RD tax incentives for many years In 2005 provincial payments for fiscal assistance amounted to 538 million dollars Tax incentives have the virtue of being more neutral in the type of firms and projects funded than measures of direct RD support in the form of grants and subsidies letting private business determine the projects and the amount of RD This virtue could also be seen as a weakness in so far as it might be socially preferable to steer RD towards projects with high spillovers Many countries try to focus their support on small and mediumsize enterprises which are more than big firms plagued by the market failure of financing intangible investment in the presence of information asymmetries As firms rely more on internal funds than on external debt and equity for financing their investments especially RD investments small firms might be more financially constrained than large firms1A number of studies have been done to evaluate the effectiveness of tax incentives in making firms spend more on RD The usual evaluation consists in checking whether there is RD additionality in the sense that private firms increase their RD expenditures by more than it costs the government to support the tax incentives program If per dollar of government support less than one dollar of additional private RD gets spent by business then public support partially crowds out private funding for RD The evaluation of the socalled “bang for the buck” is a rough cost–benefit analysis that generally does not take into account factors like spillovers indirect tax returns administration costs and the opportunity cost of spending taxable income on RD supportIn the literature there are two approaches to the evaluation of the effectiveness of RD tax incentives The first approach estimates the treatment effect by constructing counterfactuals instrumenting the treatment comparing experimental and treated firms before and after the introduction of a policy change or by comparing firms that are close to a discontinuity in the treatment design Matching estimators compare the average RD effort of firms that receive RD tax credits with the average RD of firms that do not but that are otherwise similar in particular in having the same likelihood of receiving RD tax credits Czarnitzki et al 2004 Duguet 2008 Instrumental variable estimators instrument the treatment by regressing it on a certain number of explanatory variables Cappelen et al 2008 or by controlling for the selected firms that know about RD tax incentives and apply for it Corchuelo MartínezAzúa and MartínezRos 2008 Differenceindifference estimators compare the RD of firms in the reference and treated group before and after a policy change in this case a new feature in RD tax incentives Cornet and Vroomen 2005 Regression discontinuity design compares the RD of firms that are affected or unaffected by an exogenous discontinuity in the treatment function for example just below and just above a ceiling in the conditions for being eligible to receive RD tax credits Haegeland and Moen 2007The second approach is based on a structural model deriving from some kind of optimization objective a demand for an RD equation that depends inter alia on a user cost of RD that is itself a function of RD tax credit parameters Bernstein 1986 Hall 1993 Bloom et al 2002 Mairesse and Mulkay 2004 The structural modeling approach permits simulations of the effects of future tax changes and allows distinguishing between shortrun and longrun effects Given the endogeneity of the amount of RD tax credits received and a possible distributed lag specification instrumental variable estimations are usedThis study evaluates the effectiveness of the RD tax incentives program in the province of Québec We rely on a modeling of factor demand similar to the ones adopted in studies that have opted for an estimation of a factor demand model for RD with RD tax parameters included in the user cost of RD as in Bloom et al 2002 Mairesse and Mulkay 2004 Dagenais et al 2004 and Haegeland and Moen 2007 We construct an observational specific Bindex that enters the user cost of capital and that reflects the various changes over time and across firms in the RD tax credit scheme in Quebec and in Canada As opposed to most other studies we have access to administrative data capturing the real RD tax support received by firms operating in the province of Quebec For a survey of the empirical literature on RD tax incentives see Hall and van Reenen 2000 and Mohnen 2000 For previous work on Canadian firms see Bernstein 1986 Lebeau 1996 Department of Finance Canada and Revenue 1997 Dagenais et al 2004 Czarnitzki et al 2004 and Parsons and Phillips 2007The originality of this study is that it exploits the actual RD tax credits received by every firm instead of just relying on the statutory tax rates and eligibility conditions and attributing tax incentives to every eligible firm Indeed many firms may either not know of the existence of tax incentives especially small firms or decide not to apply for RD tax credits because of administration costs inexperience or apprehension about dealing with the tax authorities The second advantage of using the observed payments is that it does not oblige us to collect data on certain types of RD like cooperative RD that would be difficult to get The third advantage has to do with the fact that in the RD surveys firms operating in more than one province may not always split their RD by province whereas from the administrative data the provincial figures are declared automatically Few studies in the literature have been able to make use of the actual payments of tax incentives Brouwer et al 2002 Haegeland and Moen 2007 de Jong and Verhoeven 2007 Duguet 2008Since the tax support is more generous towards small and mediumsized enterprises it is of interest to compare the effectiveness of this policy for SMEs and large enterprises Lokshin and Mohnen 2007 report that small firms with less than 200 employees are more sensitive to RD tax incentives than large firms in The Netherlands Small firms are likely to be more reactive to changes in RD tax incentives at least if they actually apply for them as they have more difficulty in financing their RD They have little collateral they may be young firms with little to show in terms of success and they may not even have patents to signal their capability to innovate The estimation of the model and the evaluation of the program in terms of RD additionality will be done separately for both types of firms We want to investigate whether the same holds for firms in Quebec

Keywords: