Rethinking Pharmaceutical Innovation Policy

The pharmaceutical industry relies more heavily on patent protection than any other sector of the economy. Empirical surveys across a wide range of industries consistently show that pharmaceutical firms are exceptional in how strongly they rate patents compared to other means of appropriating the benefits of innovative activity such as secrecy and first-mover advantages (e.g., lead time, moving down the learning curve, and establishing production, sales, and services facilities). (Mansfield et. al 1981; Levin et. al 1987; Cohen et. al 2000) Legal and economic analysts, for their part, agree that, whatever other concerns may be raised about the overall costs and benefits of patent protection in different sectors or across the economy as a whole, the case for such protection remains strong for pharmaceuticals. (Jaffe & Lerner 2004; Bessen & Meurer 2008) This includes those who are most skeptical of strong patent protection, such as leading legal economist and former seventh circuit judge, Richard Posner, who observed that “pharmaceuticals are the poster child for the patent system. But few industries resemble pharmaceuticals” (Posner 2009).

But what makes pharma so special? Most analysts have focused on particular ways that patents operate with respect to innovation in drug compounds, so as to provide especially strong, needed, or comparatively low-cost barriers to imitation with respect to such compounds. In a new INET-supported article published in the Yale Law Journal, I show that this focus is fundamentally misplaced. What the prevailing consensus on drug patents misses is that innovation in this sector centrally consists of not one, but two separate—and very distinct—types of information goods: knowledge of a new chemical or biological compound (the “compound information good”) and knowledge of the safety and efficacy of that compound for use in humans, as validated by clinical trial data (the “data information good”). And it is the second, not the first, information good that is the key driver of the economics in this sector and, consequently, the apt focus of its innovation-policy rules.

Yet both patent law and innovation policy analysis have focused almost exclusively on the first information good, leading to a severe mismatch between existing innovation-policy rules and the relevant features of the underlying information goods they seek to incentivize. This system of misaligned incentives accounts for many of the most troubling features of the industry’s present practices and performance, including “evergreening” practices, “reverse-settlement agreements” (RSAs), and “me-too” drugs. Curbing the high costs associated with these practices—of unduly high prices, restricted access to drugs, duplicative research, and waste from gaming—requires realigning the system’s incentives, away from patents and toward a revised system of regulatory exclusivity.

Innovation as Information Goods

That innovations are properly conceived, from a policy point of view, as “information goods,” is an insight that goes back to at least Kenneth Arrow’s foundational work on the topic. (Arrow 1962, stating at 609: “Invention is here broadly interpreted as the production of knowledge.”) The point has since been given systematic treatment by (Benkler 1999 [2002] and Varian 1999), who emphasize the two central economic features of such goods, which pull in distinct—indeed, opposing—policy directions: (1) their high of degree of nonexcludability, meaning difficulty of providing access to some while preventing it for others, which may lead to their undersupply by for-profit actors and point to the need for some innovation-policy intervention in competitive markets; and (2) their pure degree of nonrivalry, meaning that use by one does not subtract from like use by another, which points to a significant cost of using exclusionary rights as our innovation-policy solution.

Yet despite the centrality of this analytic to the field, its lessons have not always been fully absorbed, and legal and policy analysis of patents has often been hobbled by recurrent “physicalist” errors that misconceive patents as obtaining in some tangible product or process rather than, what is always the case, intangible knowledge of some product or process (i.e., knowledge of the structure or property of some “thing,” or knowledge of a way of doing, using, or making something). In the specific case of drugs, such “physicalism” has gotten in the way of clearly discerning the separate information goods involved in pharmaceutical innovation and analyzing their distinct technological and economic features as relevant to innovation policy analysis. This applies not only to the two “downstream” information goods discussed presently (namely, compound and data information) but also more “upstream” information goods (such as knowledge of molecular targets and mechanisms of action) that are often generated by public-sector researchers or public/private university-industry partnerships and that are properly seen to be ineligible for patent protection for sound underlying innovation policy reasons. (See Syed 2018)

Pharma’s Two Distinct Information Goods

Pharmaceutical firms’ innovative activity centers on generating two separate information goods: knowledge of a promising new drug product (the compound information good) and knowledge of the drug’s safety and efficacy for humans, as shown by clinical trial data (data information good). These information goods are not only separate but also sharply distinct in their risk/cost profiles as relevant to innovation policy analysis.

Generating the compound information good involves the exploration of a highly uncertain possibility frontier: each step involves many risks—only about one in a thousand candidate compounds make it through the drug-discovery phases of “search, synthesis, and screening” to enter clinical trials—so as to warrant comparatively low expenditures per step. By contrast, generating the clinical information good is a comparatively low-risk, high-cost endeavor: roughly one out of five to ten drugs that enter clinical trials successfully navigate the process of testing and refinement to receive Food and Drug Administration (FDA) approval, while the costs of phase 1, 2, and 3 trials dwarf those of each step of preclinical drug discovery.[1]

The divergence in risk/cost profiles of these information goods is highly significant for two distinct reasons. First, from a purely economic point of view, it is the data, not the compound, information good that is the driver of the industry’s innovation costs. While the overall cost of drug development remains a topic of fierce controversy, what is not controversial is that clinical-trial expenditures comprise the lion’s share of the costs, running to around seventy percent according to industry-sponsored studies (DiMasi et. al 2003, DiMasi et. al 2016),[2] and even higher for others. Indeed, a 2021 metareview of twenty-two studies of drug-development costs conducted over the past four decades found that over half (thirteen) of the studies reviewed did not even consider preclinical drug-discovery expenditures as part of total costs. (Schlander et. al 2021)

Second, in addition to their varying economic significance, these goods also differ in the technological features of the respective processes that generate them. Preclinical drug discovery, with its high risks and lower costs, is well-suited for a decentralized search, where “many minds” are given free rein to explore different avenues, even at the risk of a fair bit of overlapping, duplicative activity. (Merges & Nelson 1990). Clinical trials, on the other hand, with their lower risks and high costs, are better-suited for coordinated development, to curb duplicative efforts that would be highly wasteful at this stage. (Kitch 1977) In other words, preclinical research should be a nonexclusionary zone, to enable many-minded exploration unencumbered by proprietary barriers. Meanwhile, for clinical trials, some mechanism is needed to call off the innovation race at its outset.

Policy Implications of the Two Distinct Information Goods

Two sets of conclusions follow from integrating the distinct economic and technological features of the two innovations. First, the compound information good poses no special incentive case for patent protection. Its share of overall industry innovation costs is relatively modest and—what is really the relevant focus for innovation-policy analysis—the differential between its average innovation costs and risks and its average imitation costs and speed is unlikely to be much greater than in many other sectors where a combination of first-mover advantages and secrecy suffice to enable a relatively robust level of innovative activity. In addition, patents also serve no useful “coordinating” function during the research phase leading to their generation: their comparatively high risks and low costs make that phase suitable for a competitive, decentralized search.

Second, the data information good does present a strong case for an innovation-policy intervention, but it is one for which patents are a highly unsuitable instrument. That strong case stems not only from its large share of overall industry innovation costs, but also—what is again the relevant focus—from the large difference between its average costs and risks of generation and its average costs and speed of replication (with the latter massively reduced by regulatory permission of imitator piggybacking on innovator data). Yet the patent system provides little to no direct protection over this information good, as its doctrines center on the results of preclinical research, not clinical testing. Further, given the technological features of this innovation, it would be untenable to try to reform the patent system to protect it; inquiries into its desirability and validity are simply not ones that the patent system is well-suited to carry out.

In sum, patents serve their two primary functions in pharmaceutical innovation—incentivizing innovative activity and coordinating innovation races—only indirectly, with respect to an information good, clinical data, that they do not directly protect. Whereas for the information good that patents do directly cover—knowledge of the compound—they play little to no coordinating role and only a secondary incentive role. A sounder innovation policy would be to replace the primary, yet indirect, role played by patents over data information with a form of regulatory exclusivity that specifically attends to the distinctive features of this innovation, and simultaneously to phase out the direct but secondary role patents play over compound information.

Explaining Pharma’s Industry-Specific Patent Distortions

Meanwhile, the present system of indirect, and hence misaligned, incentives is the main culprit for many of the most concerning aspects of the industry’s current practices and performance. A system of innovation-policy rules that does not directly attend to the relevant features of the information goods that it governs is ill-equipped to address the tradeoffs facing any innovation policy.

Specifically, our current innovation policy for drugs fares very badly in handling the two key tradeoffs facing any incentive system that uses exclusionary rights (be they patents or data exclusivity). The first is to weigh any incentive benefits from stronger exclusionary rights—in terms of increased levels of innovation—against unduly high prices and curbed access for those innovations that would have been generated at lower levels of protection. The second is to weigh any such incentive benefits against undue rent dissipation—that is, wastefully duplicative innovative activity—that may occur for those innovations that would have been incentivized by lower levels of protection.

Concerns regarding each of these have been prominently voiced in the literature on pharma over the past two decades, the first under the heading of “evergreening” practices and the second that of “me-too” drugs. Missing from this literature, however, has been an explanation for why such practices are especially rife in—indeed, in some cases, specific to—pharma. And lacking a proper diagnosis of the problems, the solutions on offer have been similarly wanting. By shifting our focus to the central innovation, the data information good, we can supply both the missing explanation and the tools for more effective reforms.

“Evergreening” is the umbrella term used—usually by critics—to refer to efforts by drug companies to prolong the effective exclusivity enjoyed by a drug product beyond the formal expiration of the core patents on the compound. While encompassing a bewildering variety of practices, such efforts come down to two principal forms: (1) efforts to obtain and defend “secondary patents” on a drug that expire at a later date than the primary or core patents that originally covered it (e.g., Novartis obtaining patents on the beta crystalline form taken by its leukemia treatment Gleevic, that expired years after the parent patents on the basic compound itself; or Schering-Plough, for its allergy drug Claritin, taking out a patent on a “metabolite,” or chemical byproduct automatically produced upon its ingestion, when the patent on the active ingredient was set to expire); and (2) efforts to obtain and defend patents on new “secondary products” that can effectively compete with generics of the original (e.g., AstraZeneca pulling its heartburn treatment Prilosec off the market before it went off patent, to swap in its “new and improved” single-enantiomer variant Nexium; or Forest swapping out its twice-a-day Alzheimer’s drug Nameda IR for once-a-day Namenda XR). These practices have spawned a huge scholarly literature, with debate raging about the extent to which any given practice should count as “evergreening,” whether such practices are even able to successfully extend protection, and if successful, why they should be considered troubling. (For representative contrasting assessments, compare Feldman 2018 and HaCohen 2020 with Lietzan 2019 and List 2021.)

Once we set aside various possible confused criticisms and evasive defenses, the core concern with evergreening is that, when successful (as it often can be, for reasons detailed in my article), such practices may provide an unduly long buffer for parent- or improvement-drug products against effective generic price competition, to result in some consumers or patients having to pay much higher prices, and others simply priced out of access altogether. And to the reply that such higher prices may redound to greater incentives for innovation, the response is that such practices obtain any such incentives very indirectly, and as such incur high extra administrative and related costs (of obtaining and defending secondary patents, promoting secondary products, etc.) and distortions to innovative activity (by skewing incentives toward developing secondary products that derive much of their value parasitically, from the primary product). Such administrative and distortionary costs may be referred to as the costs of “gaming” the system.

Clarifying how evergreening works, however, raises its own puzzle: why is the practice so heavily concentrated in, indeed specific to, pharma? A patentee’s basic incentive to try to extend the effective protection from competition that its product enjoys, beyond the formal life of its core patents, would seem to be present more generally, in all sectors that enjoy robust patent protection. Yet the literature on evergreening has focused its attention exclusively on pharma, but without adducing a satisfactory explanation for why it is this sector, more than others, that so intensively engages in the practice.

That explanation lies in the specific industry structure of pharma, namely a sharp bifurcation into innovator/imitator profiles of its firms and products, with patented products made and sold as “brand name” ones by firms in one sector of the industry, and fully imitative ones made and sold as “generics” by firms in another sector. With this sharp bifurcation comes a sharp—indeed massive—differential in the prices of the competing brand-name and generic products, with the latter being 75-85% cheaper than the former on average. (Congressional Budget Office 2010; Leedan 2020) It is this steep drop in price—operating upon a base of product sales in the millions to hundreds of millions per year—that the generic form of competition threatens in pharma that provides its firms the massive extra fuel, on top of the basic incentive shared by all patentees, to extend effective patent life on their products. And what explains this steep price differential? The gap between innovator and imitator costs with respect to clinical data: ever since the passage in 1984 of the Hatch-Waxman Act, generic firms have been allowed to regulatorily “piggyback” on the data originally generated by the brand-name firm. Indeed, the effect of Hatch Waxman has not just been to lower entry costs for specific generic firms but to have made such entry widely feasible enough as to create a generic industry.

In other words, it is the centrality of the data information good to the industry’s economics and its regulatory treatment that explains the pharma-specific character of intensive evergreening. The gap between the generation and replication costs of this data explains not only the steep price differential between particular brand-name and generic products that provides individual firms the special fuel to engage in evergreening, but also the generalized industry structure that has made the practice pervasive in its wake.

The same applies to another widespread form of pharma-specific gaming, which has generated a large literature of its own over the past two decades: “reverse settlement agreements” (RSAs) between brand-name plaintiffs and generic defendants involved in patent litigation. (See, e.g., Hemphill 2006; FTC 2010; Elhauge & Krueger 2012; Hovenkamp & Lemus 2022.) In typical litigation settlements, it is the defendant who pays the plaintiff some amount to drop the lawsuit, so as to avoid higher prospective damages should they be found liable for infringement. In RSAs, by contrast, it is the reverse: the plaintiff patentee pays the defendant(s) to drop the suit. This raises the specter that plaintiffs are “buying off” a challenge to their (potentially weak) patents—that is, “paying for delay” of generic entry.

As with evergreening more generally, there remains debate on the extent to which RSAs should trouble us. But missing in the literature, again, is a satisfactory explanation for the one feature of such agreements that all are agreed on: that they are specific to patent litigation in the pharmaceutical context. And here too, the missing explanation is supplied by the fact that the key distinguishing innovation in pharma is the data information good and its regulatory treatment: this is what explains the price gap fueling patentees in this sector (alone) to seek to ward off (generic) competition with such intensity.

A First Set of Reforms: Abolishing Orange-Book Linkage and RSAs

Understanding why intensive gaming practices such as evergreening and RSAs are pharma-specific provides, in turn, the basis for more fundamental reforms to curb them than those typically proposed. Thus, with respect to evergreening in general, the Federal Trade Commission (FTC) has sought to “delist” up to 100 drug patents from the “Orange Book”—which links patents on brand-name products with delays in FDA approval of generics—so as to curb abusive delays owing to frivolous secondary patents. (FTC 2023) And regarding RSAs, the FTC has pushed for treating them as presumptively anticompetitive under antitrust law. (FTC v. Actavis 2018)

Both these proposals are sensible, but neither goes far enough. Delisting some patents from the Orange Book falls far short of what is merited, namely, delinking all patents from the FDA regulatory process. Why? Because there is simply no plausible rationale for entangling the FDA’s regulatory decision-making, pertaining to the data information good, with the patent system’s resolution of disputes pertaining to the compound information good. Similarly, RSAs should be deemed per se, not merely presumptively, anticompetitive. Why? Because, once we understand what fuels and structures pharma-specific reverse settlements, we can see that, unlike settlements of patent litigation in other contexts, RSAs are a form temporal market division that, like all market divisions, falls into the category of horizontal agreements that are impermissible absent a narrow “productive joint venture” exception, one that is in applicable to RSAs. (See Syed 2025 at 2107-09)

And yet even these reforms, although far-reaching, are not thoroughgoing enough. Why? Because by working within the patent system, to improve its operation in the application to drugs, they fail to get at the underlying, generative source of the problem. While Orange-Book linkage and the legality of RSAs provide extra opportunities to pursue evergreening gaming practices, removing them would still leave untouched the underlying motive—namely, the extra fuel given to holders of drug patents to protect themselves from the especially fierce drop in price associated with the loss of the patent and entry of generic competition. This competition enjoys much lower costs owing to regulatory piggybacking on the innovator’s data information goods. To tackle the problem at its root requires attending to the specific innovation policy needs of that information good. And that requires ignoring the distraction of patents and their gaming in terms of linkages, litigation, settlements, etc., all of which center on the wrong (compound) information good.

“Me-too” Drugs and Rent Dissipation

Similar considerations apply when we turn from the high prices and gaming costs of evergreening to the distinct problem of duplication wastes from “me too” drugs. Me-too drugs refer to highly similar brand-name drugs clustered together in the same therapeutic class, meaning they use the same mechanism of action (e.g., selective serotonin reuptake inhibition) to treat the same condition (e.g., depression), but each with its own distinctly patented compound. The dynamic that drives such clustering is familiar from the general literature on innovation races: the “rents” held out by strong patent protection may lure multiple participants to engage in overlapping, duplicative activity that, while privately rational, may become socially wasteful. (Barzel 1968; Kitch 1977; McFetridge & Smith 1980; Grady & Alexander 1992; Carleton & Perloff 2005)

To distinguish between socially beneficial versus wasteful overlapping innovative activity requires distinguishing between the risk/cost profiles of different phases of activity—precisely what an analytical focus on the distinct character of compound and data information goods allows us to do. Such an analysis can guide reforms aimed at retaining decentralized, potentially overlapping, exploration at the higher risk, lower cost upstream—i.e., compound information—phase, while curbing it via coordination at the lower risk, higher cost downstream—i.e., data information—phase.

A Second Set of Reforms: Phasing out Patents with Revised Regulatory Exclusivity

In sum, the access and gaming costs stemming from evergreening and related practices require an analysis of the distinctive economic significance of data information goods to guide effective reforms. Meanwhile, the duplication costs associated with me-too drugs require an analysis of their distinctive technological features to effectively redress.

For both, the best metric of their costs is to step back from particulars of specific cases or examples and take a comprehensive view of the industry’s output and the types and extent of innovation it represents. Such a review, carried out in the article, reveals that, of the 2,872 new drugs approved in the years 1990 to 2023,[3] almost 70% were secondary products, and 86% of these were rated by the FDA not to hold out a significant advance (see Table 1). In other words, 60% of the industry’s output consists of secondary products securing patent protection that is likely incommensurate with the modest innovation they hold out. Moreover, of the roughly 30% of output that consisted of primary products, over half (51%) were similarly rated as standard—i.e., held to be somewhat to highly duplicative of already-available treatments. These figures represent an inordinate amount of wasteful, duplicative R&D and administrative costs incurred by the industry under existing innovation policy rules, wastes that not only drive up underlying drug costs and hence prices, but also reduce net overall innovation.

Table 1: Breakdown of New Drug Approvals, 1990-2004, 2008-2023

	Drug Product Type
Drug Rating	NMEs	IMPs	Totals
Priority	433 (49% of NMEs)	287 (14% of IMPs)	720 (25% of all approvals)
Standard	452 (51% of NMEs)	1700 (86% of IMPs)	2152 (75% of all approvals)
Totals	885 (31% of all new approvals)	1987 (69% of all new approvals)	2872

Source: Syed 2025

The access, gaming, and duplication costs incurred by evergreening practices and me-too drugs stem from the misaligned incentives of the present system of innovation policy rules in place for pharmaceuticals. In each case, the cause lies in different aspects of how the central innovation in pharmaceuticals, the data information good, is handled by the present system of regulatory requirements, permissions, and data exclusivity. And for both, the solution lies in the same domain: to replace patent protection with a tailored system of regulatory exclusivity, one that retains strong incentives for truly socially valuable forms of drug innovations while curtailing them for others. The overall contours of such a revised system of regulatory exclusivity are set out in my longer article.

Rethinking Patent Policy: The Special Case of Pharma as Regulatory Artifact

Turning from pharmaceutical innovation policy to broader debates in patent theory, the conventional view that pharma presents an especially strong case for patent protection turns out to be triply wrong. First, the innovation taking pride of place in judicial and scholarly attention—the compound information good—presents no special case for patents. Second, the innovation that does present a strong case for innovation-policy support—the data information good—is both sidelined by the patent system and, in any case, ill-suited for patent protection. The special case posed by pharma, then, is not for patents but for an alternative innovation-policy intervention. Finally, the basis of that special case from some policy intervention lies in a specific regulatory regime—namely, the way that regulatory requirements and permissions massively drive up innovator costs and drive down imitator ones. None of this is to query this regime of regulatory requirements and permissions. Far from it. Rather, it is simply to underline that it is this regime, rather than any generalizable economic or technological features of drugs, that makes pharma special, putting it in need of special innovation-policy support.

This last point is crucial for general debates in patent theory. In those debates, pharma has long cast a shadow over the standard conclusion that the overall case for patents—across the economy as a whole—is uneasy, and likely at its best for modest protection for small inventors at the margins. (Machlup 1958; Scherer 2007; Bessen & Meurer 2008) Pharma has long operated as the key exception to that general rule, serving both to gnaw away at confidence in the rule in theory and to skew the operation of the system in practice. Showing that this exception can be not only explained, but explained away, reinforces the broader conclusion—that for most sectors, strong patents are likely not needed for robust innovation—which may now be retained in its original force, without qualification, and as such perhaps serve as the basis for instituting a more modest system of patent rights in practice.

References

Arrow, Kenneth J., Economic Welfare and the Allocation of Resources for Invention, in The Rate and Direction of Inventive Activity: Economic and Social Factors 609 (Richard R. Nelson ed., 1962)

Barzel, Yoram, Optimal Timing of Innovations, 50 Review of Economics & Statistics 348 (1968)

Benkler, Yochai, Intellectual Property and the Organization of Information Production, (Oct. 1999) (draft), http://www. benkler.org/Ipec99.pdf (final version published in 22 International Review of Law & Economics 81 (2002)

Carlton, Dennis W. & Jeffrey W. Perloff, Modern Industrial Organization (4th ed. 2005)

Cohen, Wesley M., Richard R. Nelson & John P. Walsh, Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or Not) (Nat’l Bureau of Econ. Rsch., Working Paper No. 7552, 2000)

Congressional Budget Office, Effects of Using Generic Drugs on Medicare’s Prescription Drug Spending (September 2010)

DiMasi, Joseph A., Ronald W. Hansen & Henry G. Grabowski, The Price of Innovation: New Estimates of Drug Development Costs 22 Journal of Health Economics 151 (2003)

DiMasi, Joseph A., Henry G. Grabowski & Ronald W. Hansen, Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs, 47 Journal of Health Economics 20 (2016)

Elhauge, Einer & Alex Krueger, Solving the Patent Settlement Puzzle, 91 Texax Law Review 283 (2012)

Federal Trade Commission, Pay-for-Delay: How Drug Company Pay-offs Cost Consumers Billions (Jan. 2010)

Federal Trade Commission, FTC Challenges More Than 100 Patents as Improperly Listed in the FDA’s Orange Book (2023) available at https://www.ftc.gov/news-events/news/press-releases/2023/11/ftc-challenges-more-100-patents-improperly-listed-fdas-orange-book

Feldman, Robin, May Your Drug Price Be Evergreen, 5 Journal of Law & Biosciences 590 (2018)

FTC v. Actavis, Inc., 570 U.S. 136 (2013)

Grady, Mark F. & Jay I. Alexander, Patent Law and Rent Dissipation, 78 Virginia Law Review 305 (1992)

Hacohen, Uri Y., Evergreening at Risk, 33 Harvard Journal of Law & Technology 479 (2020)

Hemphill, C. Scott, Paying for Delay: Pharmaceutical Patent Settlement as a Regulatory Design Problem, 81 New York University Law Review 1553 (2006)

Hovenkamp, Erik & Jorge Lemus, Antitrust Limits on Patent Settlements: A New Approach, 70 Journal of Industrial Economic 258 (2022)

Kitch, Edmund W., The Nature and Function of the Patent System, 20 Journal of Law & Economics 265 (1977)

Ledan, Seema Discussing Brand Versus Generic Medications, U.S. Pharmacist (June 2020)

Levin, Richard C., Alvin K. Klevorick, Richard R. Nelson & Sidney G. Winter, Appropriating the Returns from Industrial Research and Development, 3 Brookings Papers on Econ. Activity 783 (1987)

Lietzan, Erika, The “Evergreening” Metaphor in Intellectual Property Scholarship, 53 Akron Law Review. 805 (2019)

List, McKenzie E., The Hollow Rhetoric of Evergreening, 61 Jurimetrics J. 495 (2021)

Machlup, Fritz, An Economic Review of the Patent System, Staff of Subcomm. on Patents, Trademarks & Copyrights of the S. Comm. on the Judiciary, 85th Cong., Study No. 15 (1958)

Mansfield, Edwin, Mark Schwartz & Samuel Wagner, Imitation Costs and Patents: An Empirical Study, 91 Economics Journal 907 (1981)

McFetridge, Donald G. & Douglas A. Smith, Patents, Prospects and Economic Surplus: A Comment, 23 Journal of Law & Economics 197 (1980)

Merges, Robert P., and Richard R. Nelson, On the Complex Economics of Patent Scope, 90 Columbia Law Review, 839 (1990)

Posner, Richard A., “Why There Are Too Many Patents in America,” The Atlantic (July 12, 2012), https://www.theatlantic.com/business/archive/2012/07/why-there-are-too-many-patents-in-america/259725

Scherer, F.M., The Political Economy of Patent Policy Reform in the United States (John F. Kennedy School of Government Working Paper RWP07-042, 2007)

Schlander, Michael, Karla Hernandez-Villafuerte, Chih-Yuan Cheng, Jorge Mestre-Ferrandiz & Michael Baunmann, How Much Does It Cost to Research and Develop a New Drug? A Systematic Review and Assessment, 39 PharmacoEcon. 1243 (2021)

Syed, Talha, Reconstructing Patent Eligibility, 70 American University Law Review 1937 (2021)

Syed, Talha, Does Pharma Need Patents?, 134 Yale Law Journal 2038 (2025)

Varian, Hal R., Markets for Information Goods (Bank of Japan Institute for Monetary & Economic Studies, Discussion Paper No. 99-E-9, 1999), https://www.imes.boj.or.jp/res…

Notes

[1] A detailed review of the literature on the risks and costs of these respective phases of innovation is provided in (Syed 2025 at 2068-69 and 2078-80).

[2] It should be noted that even these estimates significantly understate the contribution of clinical trials to industry R&D costs: since the figures they provide are for capitalized costs rather than out-of-pocket cash outlays, the estimated shares for preclinical R&D costs include the time such expenditures are tied up without seeing a return, and that time is significantly extended by the length of clinical trials.

[3] With the exception of years 2004-2007, for which refined data were not available.