Positive encouragements can be non-pecuniaire (intellectual curiosity; to social incentive or financial such as natural market forces or push-pull mechanisms such as public funding through grants, IP and prizes. While they all incentivise innovation to an extent, they also all have disadvantages. Public funding and prizes suffer from lack of information; IP solves the information problem by introducing deadweight loss. A mix of incentives is therefore likely to be the most appropriate solution across the board.
Incentives can be conceptualised in two ways, as positive encouragements, and as positive encouragements minus negative hindrances. Encouragements are things like financial rewards and respect. Hindrances are things like bureaucracy, legal restrictions, and closed access to information. While conceptualising incentives as encouragements minus hindrances is more comprehensive, it is also a less standard approach. Here, incentives will be considered as encouragements.
Some incentives are non-pecuniary. These range from personal incentives, like intellectual curiosity; to social incentives, like respect; right up to altruistic incentives, like the benefit of others. Then there are pecuniary or financial incentives. Of all the kinds of incentive, financial ones have been most written about in economic literature, are the most accessible to policy makers, and are plausibly the most important incentives. Without the money to cover the costs of innovation, it seems likely that personal, social and altruistic incentives would not be sufficient to generate the level of innovation required by society.
Focusing in on financial incentives, natural market forces provide some stimulation to innovation. People will pay for things they value, and some people have money. Moreover, because imitation is not costless and does not happen immediately, the first innovator in an area will have a market advantage for a time.1 There are also particular market conditions which can protect innovators’ profits. For instance, encryption and trade secrecy prevent competitors from accessing an innovator’s work, while network effects can create a de facto monopoly.2
However, market forces alone usually offer insufficient financial incentives to innovation. On the one hand, there is the problem of competition. The costs of the initial innovation as an idea or piece of research are born by the innovator alone. Their competitors only have to bear the costs of replication, which can be very low. This puts innovators at a disadvantage and makes investment in innovation less attractive.3 On the other hand, there is the problem of efficient supply. The price of a product excludes some would-be consumers, who cannot afford it. With physical goods, this can be efficient if it saves resources or frees up the good for another consumer. With information goods which are nonrival, depriving a consumer of access is never efficient.4 The efficient price of a pure information good, where the marginal cost is zero, would be nothing. This is the only way to ensure efficient access. ‘However, the efficient competitive price, zero, will not cover the costs of developing the software, and therefore the market will not work.’5 The problem of efficient supply is more fundamental than that of competition: the former erodes profits, while the latter sets profits to zero. Where information goods are concerned, these problems mean that market forces alone do not provide sufficient incentives to innovation.
A number of incentive mechanisms have been devised to solve this problem and ensure that there are sufficient incentives to innovate. Often such mechanisms are characterised as ‘push’ or ‘pull’ mechanisms, in that they provide direct or indirect incentives to innovate.6 All mechanisms which are currently in use incentivise innovation to some extent, in that they do pay out financial rewards to innovators. However, they stimulate different kinds of innovation to differing extents. Besides, there is the problem of the counterfactual: we do not know how much more or less innovation would have occurred under different incentive mechanisms. Here we shall consider the main incentive mechanisms in use today: public funding via grants, intellectual property (IP), and prizes.
Public funding through grants is a push incentive which stimulates innovation by selecting and funding projects upfront. In a world of perfect information, this would be the most efficient mechanism for incentivising innovation. The fundamental problem for public funding through grants is that in reality we do not know which innovations will be valuable in advance. This lack of information makes public funding less efficient, though it is still considered one of the best ways of incentivising basic research.
Intellectual property is ‘[t]he only fundamentally new incentive scheme of the past 400 years is intellectual property.’7 IP, including copyright and patents, is ‘an exclusive right to market an invention for a fixed time period.’8 In effect IP creates a temporary monopoly, and so creates a deadweight loss. ‘Deadweight loss occurs when people are excluded from using the good even though their willingnesses to pay are higher than the marginal cost.’9 This is the main inefficiency of IP as an incentive to innovation. One solution to the problem of deadweight loss is price discrimination, whereby a producer sets different prices according to the willingnesses to pay of different groups. However, because of arbitrage and pressure from those paying higher prices, this is an imperfect solution.10 A system which could provide the same incentives as IP without deadweight loss would be superior. Moreover, IP incentivises profitable innovation, innovation which has a market. This means that it offers a poor incentive to innovate in areas where the market is predominately poor - including neglected diseases.
The other main incentive to innovation besides public funding and IP is prizes. ‘Prizes were eclipsed by patents during the Industrial Revolution, but they have never vanished as an incentive mechanism.’11 Prizes have the same advantages as IP without deadweight loss, as they delink the price of the product from the profit gained. However, the efficient functioning of prizes as incentives depends on the observability and verifiability of either social value or cost.12 Otherwise, the prize may be set higher or lower than is efficient. Unfortunately, both social value and cost are difficult to measure, so as with public funding there is an information problem. In economic literature, ‘social value’ is sometimes used to mean consumer surplus.13 However, in the case of healthcare, a metric which includes health outcomes as well as consumer surplus would be more appropriate. These two factors would only coincide if all individuals had equal resources and were perfectly rational. Given that this is not the case, metrics of social impact like QALYs may be more appropriate for setting prizes for healthcare innovation. These metrics are also hard to evaluate, but arguably less so than consumer surplus or costs.
Public funding through grants, IP and prizes all incentivise innovation to an extent. They also all have disadvantages. Public funding and prizes suffer from lack of information; IP solves the information problem by introducing deadweight loss. A mix of incentives is therefore likely to be the most appropriate solution across the board. However, the suitability of an incentive mechanism in a specific context depends on numerous factors.
Firstly, the nature of innovation in question influences which incentive will be most powerful. For instance, it may be that when innovation is both sequential and complementary, strong IP actually inhibits innovation.14
Secondly, the environment in which innovation takes place can determine the appropriate incentive. An example of this is the relationship between IP, innovation and GDP. IP’s influence on innovation is nonlinear: increased IP in countries with relatively low or high IP in the first place increases innovation, but in other cases it decreases innovation. Moreover, GDP’s influence on the relationship between innovation and IP is also nonlinear. For countries with high levels of IP, increased IP stimulates innovation - but only provided that GDP is sufficiently high. This means that blanket IP regulations like TRIPS are unlikely to be optimal either globally or for individual countries.15
Finally, the social impact of the innovations in question influences the appropriate incentive mechanism. One of the advantages of IP is that users pay for innovation, rather than the general public.16 While this may be appropriate for luxury goods, it is highly undesirable in healthcare, where making the sick pay is generally regarded as immoral. As Scotchmer puts it, ‘Whereas wealthy benefactors and governments can indulge in basic science and curiosity-driven research, a research agenda driven by patents is hostage to the market and to consumer sovereignty. The consumers who are sovereign are those with resources.’17 In the case of healthcare, this creates huge inequality and leads to preventable death.
Bessen, James, and Eric Maskin. “Sequential Innovation, Patents, and Imitation.” The RAND Journal of Economics 40, no. 4 (2009): 611–635.
Dalton, Max. “Should You Fund Research into Tropical Diseases?,” n.d. https://docs.google.com/document/d/1Jmp1LQEem9E29ynSrqYrkiCsFFMrvyZLPqPl3Itdaak/edit?usp=sharing&usp=embed_facebook.
Gallini, Nancy, and Suzanne Scotchmer. “Intellectual Property: When Is It the Best Incentive System?” Innovation Policy and the Economy 2 (2002): 51–77.
Hecht, Robert, Paul Wilson, and Amrita Palriwala. “Improving Health R&d Financing for Developing Countries: A Menu of Innovative Policy Options.” Health Affairs (Project Hope) 28, no. 4 (2009): 974–85. doi:10.1377/hlthaff.28.4.974.
Hudson, John, and Alexandru Minea. “Innovation, Intellectual Property Rights, and Economic Development: A Unified Empirical Investigation.” World Development 46 (2013): 66.
Moser, Petra, and Alessandra Voena. “Compulsory Licensing: Evidence from the Trading with the Enemy Act.” American Economic Review 102, no. 1 (2012): 396–427. doi:10.1257/aer.102.1.396.
Scotchmer, Suzanne. Innovation and Incentives. Cambridge, Mass. ; London: MIT Press, 2004.
