A Win-Win-Win? Motivating innovation in a knowledge economy with tax incentives

https://doi.org/10.1016/j.techfore.2017.05.030Get rights and content

Highlights

  • We propose a novel tax incentive directed on profit sharing schemes (PSS).

  • This is compared to the standard ‘IP box’ incentive in a knowledge economy context.

  • We show that the PSS incentive benefits workers, firms and the economy as a whole.

  • Its effectiveness rises with the role of labor relative to capital investments.

  • Its relative efficacy is also moderated by labor mobility and knowledge spillovers.

Abstract

In this study we explore effects of two distinct tax policies on innovation in a pure knowledge economy: an ‘IP box’ incentive and a (hypothetical) tax incentive on compensation earned by agents from profit sharing schemes (PSS). In contrast to the conventional assumption that firms decide on whether to innovate or not, we focus on a bottom-up innovation process (sometimes also called ‘bootleg innovation’), where firms set incentives to fulfill different tasks, but the final decision on whether to make the more innovative task is taken by an employee. We compare the two tax incentives under several distinct specifications demonstrating that the tax incentive on PSS can be a powerful mechanism fostering innovative activity and benefiting at the same time workers, firms and the economy as a whole. This study shows that the more critical for firms is attracting and motivating highly skilled workers, the larger the expected gain from employing the tax incentive on agents' compensation. We also find that the relative efficacy of this tax incentive is moderated by labor mobility and the extent of knowledge spillovers.

Introduction

Innovation plays an increasingly prominent role in shaping competitive advantages of firms and countries: skilled employees generate new knowledge, which is later incorporated into better products or processes. Numerous policies have been proposed to foster innovation, ranging from regulatory reforms of the intellectual property (IP) rights (Boldrin and Levine, 2008) to tax incentives designed to obtain the maximum boost in innovation with the least loss in tax revenues. Previous tax research has paid considerable attention to tax incentives for innovation in the form of credits or deductions related to corporate income taxation (good examples are Hall and Van Reenen, 2000, Myles, 2009, Griffith et al., 2014, Gemmell et al., 2014), as these are the types of incentives most commonly used in practice. The limits of such policies have been as well highlighted: volume-based tax incentives may induce a reclassification of accounting items and lack additionality (Dimos and Pugh, 2016), while incremental incentives may entail large administrative costs and cause distortions to the inter-temporal allocation of the investments. ‘IP box’ incentives have gained momentum and have been adopted in several EU countries (Evers et al., 2015), however researchers have documented that IP boxes mostly cause a relocation of existing R&D activities rather than fostering domestic innovative efforts (Alstadsæter et al., 2015). The latter point also makes IP boxes unsustainable in the long run as they might trigger a form of tax competition and a race to the bottom of IP-related tax rates.

Given the aforementioned issues, the challenge is to find alternative policy designs able to stimulate innovation (while entailing small losses in tax revenues), and to compensate the behavioral distortions caused by other existing policies. The research on the link between labor income taxation and innovation, though, is not well developed. Moreover, tax incentives to innovation on the side of labor income are adopted in few countries and are also limited in extent. The tax wedge on labor income in more advanced countries is often high and, because of the progressivity of the personal income tax, is even higher for any increase in pay that might stem from a valuable innovation. If one considers that about 60–70% of R&D costs are used for labor (Harhoff et al., 2003, OECD, 2015), the extent to which the existing personal tax systems can be hampering innovation by making labor more costly for firms becomes clear. Regardless of the source of financing (internal cash flow, or external financing) used to support R&D outlays, the additional costs due to the combined effects of the personal and corporate taxes might restrain innovation and lead to socially sub-optimal levels of investment. Furthermore, the decision-making process in innovative firms is often complex and combines top-down with bottom-up approaches, centralized with decentralized decisions, and relevant degrees of information asymmetry between management and employees involved in technical functions. A consequence of the latter observations is that the way personal taxation affects the overall outcome of innovative efforts within companies is not immediate to foresee.

In a recent contribution d’Andria (2016a) proposed a novel (hypothetical) form of tax incentive on profit sharing schemes (PSS) based on the grounds of theoretical arguments. By PSS is meant any form of compensation to employees (e.g., direct participation to profits, bonus pay, stock options and stock grants) that links compensation to measures of company's success and, specifically, of innovation. The idea of a tax incentive on PSS stems from the empirical observations that R&D intensive firms offer PSS compensations to employees who are capable to innovate (Balkin and Gomez-Mejia, 1984, Ittner et al., 2003, Lerner and Wulf, 2007), and that inventors' pay rises on average around the time of a patent application (Depalo and Di Addario, 2014) or grant (Toivanen and Väänänen, 2012). d’Andria (2016a) demonstrated that a policy mix that includes some level of tax reductions on PSS income could better foster innovation in a setting where information asymmetry between managers and employees is a relevant issue. Another appealing property of this tax incentive is that it does require neither any a priori definition of what an R&D expenditure is (as it is for standard R&D tax credits and deductions), nor the innovation to be subject to intellectual property rights (as it is the case for IP box incentives). All innovations that increase the profitability of the firm can be captured by PSS monetary incentives, and as a consequence, a tax incentive on PSS is capable to affect a wider set of potential innovative behaviors. Our contribution is to study the performance of this PSS incentive and compare it with an IP box incentive under a rich set of scenarios. The flexibility of our simulation environment allows to enable a large number of market features (of special relevancy, features related to dynamic characteristics of firms' behavior) providing a large set of possible markets where the two incentives may be introduced. The aim is then to better understand under what observable conditions a PSS incentive may be a viable policy to support innovation.

The political attention to the topic has increased recently. A good example is the US presidential candidate Hillary Clinton's proposal for a Profit Sharing Tax Credit announced in July 2015, whose supposed beneficial effects are ascribed to its ability to foster innovation.1 The proposal is motivated, among others, as PSS “makes businesses more productive and innovative. Studies find that [PSS] on the whole result in increased business productivity and innovation. This makes sense: when employees share in profits, they have a stronger stake in the company's success” advocating “a win-win situation [..] good for workers and good for business”. Given that innovation and subsequent economic growth are a major concern, this study explores the possibility of a ‘win-win-win’ situation (for workers, business and economy as whole) modeling a design of the incentive that provides a tax reduction to employees directly rather than a tax credit to the employing company. To this end, we develop an Agent-Based Model (ABM) representing, in a stylized way, a market for product innovations made by inventors who are employed in private firms, where the latter compete to attract the best workers and offer them profit-maximizing compensation contracts. Under the assumptions of the model, a PSS-based incentive is shown to outperform an IP box incentive as a means to induce more innovations. The interactions between the PSS-based incentive and different labor mobility regimes are studied as well. Anticipating here some of the conclusions, our main result is that a tax incentive on the PSS part of employees' compensation can be a useful addition to the arsenal of policies stimulating innovative activity, particularly for those countries where the transition to a knowledge-based economy is at an advanced stage.

The paper is organized as follows. Section 2 reviews the links with existing literature. Section 3 describes the ‘baseline’ model, while in Section 4 we present its simulation results and gradually add complexity by allowing firms to accumulate their technological capabilities and additionally introducing R&D investments. In Section 5 we provide a quantitative assessment of the tax incentives' efficacy and Section 6 concludes.

Section snippets

Previous literature

In knowledge-intensive firms PSS compensations are offered to employees for multiple reasons. A survey done by Ittner et al. (2003) on US firms from the ‘New Economy’2 reports that one of the major reasons they offer stock options and stock grants is to attract and retain skilled staff. A

The model

Conventional theoretical models in economics begin with a set of definitions and assumptions which lead to proofs of theorems. Such models are often characterized by tractability and parsimony, which are in turn obtained by means of (over)simplifying assumptions (Judd, 2005). One relevant difference found in Agent-Based Models is the attempt to circumvent such simplifications and to attain insight in cases where complexities are not dismissed. Such complexities often arise from local

Simulation results for different scenarios

We run our numerical experiment for 2000 periods reporting averaged values (over the last 500 periods) for the variables of interest, such as the number of innovations successfully introduced, contracts being selected and tax revenues being collected.17 As our ABM provides a simplified representation of the real economy

Regression analysis

To draw general conclusions based on the extensive number of parameter combinations tested in our model, we use regression analysis of the data generated by our model. This is particularly helpful as one can distinguish between the different effects of each specific variable on our output measures and test their significance. In the following, we focus on three variables that are interesting policy-wise (thus, concentrating on the outcome for the economy): the aggregate value of innovations

Conclusions

Empirical and anecdotal evidence point to organizational issues being as relevant obstacles to innovation within private companies as are external constraints. Moreover, the competition for the best minds puts pressure on wages and, hence, on the cost of labor in R&D functions. As about 70% of total R&D expenditures are labor costs, understanding how the labor market for innovative employees works and how policy may shape it is of primary importance, particularly in light of the changes toward

Diego d’Andria studied business and economics at the University of Naples “Federico II”, Italy. He completed a PhD program in economics at the DFG Research Training Group 1411 “The Economics of Innovative Change” at the University of Jena in 2016. Currently he is Economic Analyst at the European Commission Joint Research Center (JRC) - Fiscal Policy Analysis unit. His main research interests are in the fields of public finance, taxation and human capital. More at http://www.diegodandria.info/.

References (84)

  • LensbergT.

    Investment behavior under Knightian uncertainty-an evolutionary approach

    J. Econ. Dyn. Control.

    (1999)
  • MalerbaF. et al.

    Public policies and changing boundaries of firms in a ‘history-friendly’ model of the co-evolution of the computer and semiconductor industries

    J. Econ. Behav. Organ.

    (2008)
  • ReinganumJ.F.

    The timing of innovation: research, development, and diffusion

    Handb. Ind. Organ.

    (1989)
  • TesfatsionL.

    Agent-based computational economics: modeling economies as complex adaptive systems

    Inf. Sci.

    (2003)
  • AcharyaV.l.V. et al.

    Seeking alpha: excess risk taking and competition for managerial talent

  • AlmeidaP. et al.

    Localization of knowledge and the mobility of engineers in regional networks

    Manag. Sci.

    (1999)
  • AlstadsæterA. et al.

    Patent boxes design, patents location and local R&D

  • AnderssonF. et al.

    Reaching for the stars: who pays for talent in innovative industries?

    Econ. J.

    (2009)
  • AntunesL. et al.

    Tactical exploration of tax compliance decisions in multi-agent based simulation

  • ArrowK.

    Economic welfare and the allocation of resources for invention

  • BalkinD.B. et al.

    Determinants of R&D compensation strategies in the high tech industry

    Pers. Psychol.

    (1984)
  • BaranchukN. et al.

    The economics of super managers

    Rev. Financ. Stud.

    (2011)
  • BeatonA.E. et al.

    The fitting of power series, meaning polynomials, illustrated on band-spectroscopic data

    Technometrics

    (1974)
  • BeetsmaR.M.W.J. et al.

    Measuring risk attitudes in a natural experiment: data from the television game show LINGO

    Econ. J.

    (2001)
  • BenabouR. et al.

    Bonus culture: competitive pay, screening, and multitasking

  • BlackS.E. et al.

    What's driving the new economy?: the benefits of workplace innovation

    Econ. J.

    (2004)
  • BloomquistK.

    Tax compliance as an evolutionary coordination game: an agent-based approach

    Public Financ. Rev.

    (2011)
  • BoldrinM. et al.

    Against Intellectual Monopoly

    (2008)
  • BoltonP. et al.

    Contract Theory

    (2005)
  • BrownD.E.

    Inventing Modern America: From the Microwave to the Mouse

    (2002)
  • CloggC.C. et al.

    Statistical methods for comparing regression coefficients between models

    Am. J. Sociol.

    (1995)
  • CohenW. et al.

    Schumpeter's prophecy and individual incentives as a driver of innovation

  • d’AndriaD.

    Taxation and incentives to innovate: a principal-agent approach

    FinanzArchiv: Public Financ. Anal.

    (2016)
  • d’AndriaD. et al.

    Motivating innovation in a knowledge economy with tax incentives

  • d’AndriaD. et al.

    The relationship between R&D intensity and profit-sharing pay: evidence from Europe

  • DavisJ.S. et al.

    Social behaviors, enforcement, and tax compliance dynamics

    Account. Rev.

    (2003)
  • DawidH. et al.

    How do social networks contribute to wage inequality? Insights from an agent-based analysis

    Ind. Corp. Chang.

    (2014)
  • De GrauweP.

    Animal spirits and monetary policy

    Economic Theory

    (2011)
  • DemaryM.

    Transaction taxes and traders with heterogeneous investment horizons in an agent-based financial market model

    Economics

    (2010)
  • DepaloD. et al.

    Shedding light on inventors' returns to patents

  • EdererF.P.

    Incentives for parallel innovation

    SSRN Scholarly Paper 2309664

    (2013)
  • EgbetokunA. et al.

    Absorptive capacity and innovation: when is it better to cooperate?

    J. Evol. Econ.

    (2014)
  • Cited by (0)

    Diego d’Andria studied business and economics at the University of Naples “Federico II”, Italy. He completed a PhD program in economics at the DFG Research Training Group 1411 “The Economics of Innovative Change” at the University of Jena in 2016. Currently he is Economic Analyst at the European Commission Joint Research Center (JRC) - Fiscal Policy Analysis unit. His main research interests are in the fields of public finance, taxation and human capital. More at http://www.diegodandria.info/.

    Ivan Savin studied International Economics at the Ural State University in Yekaterinburg (Russia) and at the University of Passau. In 2011 he obtained a PhD from the Department of Statistics and Econometrics at the University of Giessen. He has been a postdoctoral researcher at the DFG Research Training Group 1411 “The Economics of Innovative Change” at the University of Jena between 2011 and 2015. Since May 2015 he is researcher within the KIT-BETA research project devoted to general purpose technologies, creativity and sustainability. His research interests include economics of innovation and economics of complexity. More at http://ivansavin.wix.com/researcher-webpage.

    The first version of this work was written when both authors were employed at the Graduate College ‘Economics of Innovative Change’ (DFG RTG 1411) at the Friedrich Schiller University led by Uwe Cantner, to whom we are thankful for support and guidance. We would also like to thank Holger Graf, Silke Übelmesser, participants of the GENED Workshop at the Kiel Institute for the World Economy and the 15th International Schumpeter Society Conference and two anonymous reviewers for helpful comments and suggestions. IS also acknowledges support from the Helmholtz Association (HIRG-0069) and Projex CSES, Initiative d’Excellence, Université de Strasbourg. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They should not be attributed to the European Commission.

    View full text