Search the archive.

The fair use economy has continued to expand since the end of the Great Recession, reflecting the growth of the Internet and the increased use of it by businesses and consumers. The fair use economy is changing. Output is increasingly driven by the core Internet-based industries that depend most on fair use and other limitations and exceptions to copyright.

In 2014, the fair use economy accounted for $5.6 trillion in revenues and $2.8 trillion in value added, roughly 16 percent of U.S. GDP. It employed approximately 18 million people and supported an annual payroll of more than $1.4 trillion. Additionally, the fair use economy generated $368 billion in exports and fueled rapid productivity growth.

When comparing copyrighted works to those in the public domain, the monopoly power created by copyright results in higher prices and lower availability (Buccafusco & Heald, 2013; Flynn et al., 2019; Heald, 2014, 2018; Li et al., 2018). Evidence also shows that the reverse is true: once copyright expires, works are more often performed (Cuntz, 2023; Watson et al., 2023) and lower prices of accessing copyrighted works encourage follow-on innovation (Biasi & Moser, 2021). Frictions and inefficiencies in the licensing market might explain the conflicting results. Contacting and negotiating with rightsholders can be time-consuming, leading to high transaction costs for license acquisition (Akmon, 2010; Vuopala, 2010). In some cases, the copyright owner cannot be identified, leading to so-called orphan works that cannot be licensed for use or re-use. The longer the copyright term, the more difficult it may be for potential licensees to locate the copyright owner. Because of a lack of empirical research, it is unclear, however, to what extent frictions in the licensing market and orphan works as an extreme case represent an economically important problem. In the context of entertainment formats, such as music and video, platforms implementing automated N&T systems switched focus from pure enforcement to enabling a digital licensing market. For example, around the time YouTube launched the affiliate program in 2009, in which the platform agreed to share advertising revenue with contributors, YouTube introduced its “Content ID” system, where rightsholders are notified of infringing content and can choose to block the allegedly infringing video or monetize the content by sharing advertising revenue. YouTube reports that over 90% of rightsholders choose monetizing, which has accumulated to $7.5 bn in advertising revenue paid to rightsholders from claims through the system since 2016 (YouTube, 2021). Over 9000 partners use the Content ID system, including large network broadcasters, movie studios, and record companies. In the second half of 2021, the Content ID system made close to 760 mn claims, of which only 0.5% were disputed, suggesting that it has become feasible to license millions of derivative works with negligible cost. The Content ID system is just one example of digital technology that has spurred the introduction of market mechanisms and allowed to circumvent the institutions that would traditionally be involved in the settling of copyright infringement cases (see Towse, 2017 for an overview of the history of music publishing and licensing), thereby potentially enhancing the efficiency of licensing markets. However, to date, no empirical studies estimate the welfare effects of algorithmic licensing.

This report provides updated information on U.S. industries that depend on and/or benefit from limitations and exceptions to copyrights, including the fair use of copyrighted materials. This “fair use” economy has grown in importance since the publication of the original CapTrade study, which was published in 2007 and covered data through 2006.3 Subsequent updates found that the fair use economy continued to grow, even during the Great Recession.4

Fair use in the strict sense is an important statutory restriction on the rights conferred on original works by the U.S. Copyright Act of 1976:5 According to Sec. 107, “[t]he fair use of a copyrighted work for . . . purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research is not an infringement of copyright.”6 The fair use doctrine, and other limitations and exceptions to copyright,7 have grown in importance with the rise of the digital economy, as fair use permits a range of activities that are critical to many high technology businesses, including search portals and web hosting.8 In the discussion that follows, the term “fair use” sometimes will be used as a shorthand expression referring to the full range of limitations and exceptions in U.S. copyright law.

It is also the case that in many AI contexts, parties simply prefer trade secrecy over patent protection. 395In such scenarios, it would thus seem that strengthening AI patents would have little to no effect on the parties' intellectual property choices. That preference for trade secrecy, in turn, is likely to continue to push the industry towards consolidation in many cases, regardless of the relative strength of AI patents. This is so for the reasons discussed above: trade secrecy tends to increase transaction costs relative to patent protection, meaning that vertical integration becomes the more economically efficient means of producing the AI solution, even if it falls short of being the most innovative one. 396

Indeed, this point is buttressed by the fact that important parts of AI systems are either not protectable under patent law (e.g., the training data) or only barely so because of the existing AI prior art. 397And as discussed above, some of those complementary assets are more feasible to access when the industry is consolidated. Hence, strengthening AI patents by expanding patent eligibility is not a panacea for the AI industry. Indeed, doing so may actively harm AI innovation by reviving an AI patent troll market. Instead, the real solution may lie outside of intellectual property rights altogether.

A proliferation of patents in the life sciences and biotechnology fields could lead to more patent infringement lawsuits and costly legal battles over the boundaries of intellectual property rights. This situation was directly addressed by the Supreme Court in the wake of the litigation explosion in the late 1990s and early 2000s. Should the PERA’s new patent eligibility standards restore those circumstances, increased litigation costs may be passed on to consumers in the form of higher drug prices and reduced choices, further exacerbating affordability issues.

Any proposed expansion of patent eligibility for inventions involving natural products or biological molecules and processes could have significant implications for drug prices and affordability. By allowing broader patenting of naturally occurring compounds, genes, proteins, and biological processes, pharmaceutical companies would gain stronger exclusive rights over these fundamental building blocks of drug development. This may be particularly problematic in rapidly evolving fields like AI, precision medicine, and genetic testing, where the closer a patent is to an abstract idea or natural phenomenon, the more stifling its impact will be on the development of new inventions and applications.

In addition, Section 101 is unique because district courts can decide patent eligibility as a question of law without resort to expert testimony or evidence beyond the patent. As a result, meritless cases can be resolved at an early stage before huge sums of money are wasted on attorney’s fees and other litigation costs. By contrast, Section 103 requires specific factual determinations, and therefore typically requires expert evidence and can only be resolved at trial. In some cases, courts can and have decided Section 103 issues at summary judgment, but we are not aware of any rulings on the merits at the motion to dismiss stage, when many dispositive Section 101 motions are decided.40 The difference for wrongly accused defendants is enormous: motions to dismiss are decided before discovery or trial, and thus before hundreds of thousands, and more often, millions, of dollars in litigation costs are accrued.41

Patents on genetic information and uses thereof will not only limit the availability of current medical diagnostics and treatments, but also harm the availability of future health innovations. Because patents allow their owners to block competition, they would empower their owners to block competition from those who would otherwise offer more effective or affordable tests and treatments. This exclusionary power would, in turn, deter researchers from developing new tests and treatments for life-threatening and debilitating conditions, like cancer and Alzheimer’s disease. The result will be medical care that is more expensive, less innovative, and less accessible than it is today—just as it was in the days before the Myriad decision.

Further, critics against gene patents argue that such patents, when given to a single company or multiple companies, for instance, could prevent competitors from offering substitute tests and/or testing services.208 Under this view, exclusive rights obtained through patents can lead to increased “cost[s] of testing, hampered verification of test results,” and fewer substitute tests commercially available to consumers.209 At such a crossroads, it appears that competing entities attempting to enter the testing market must conduct a cost-benefit analysis of whether they wish to negotiate (i.e., pay) for patent licenses, which could be costly, or stay out of the market altogether. Because “diagnostics produce relatively low profits,”210 the latter is more likely to occur. Here, the lack of competition, or even the existence of a monopoly, could increase pricing for tests due to the absence of substitute options on the market. In other words, companies could charge more for each test simply because they can; there is no incentive for them to make their products more affordable in the absence of competition. Then, the “Iron Triangle” predicts that an increase in price will decrease overall patient access to tests.211 This could further impact the quality of the available test either positively or negatively: quality could improve because companies with patent protection are given such a monopoly of the market that they can further invest funds into the research and development of better diagnostics; conversely, quality could decrease if the companies become complacent due to their exclusive market position and decide to cut corners in producing the tests. Overall, under this view, the consequence is that a lack of active competition due to excessive protection granted by isolated DNA patents could increase costs, which would likely reduce the number of patients able to afford genetic testing, and thereby decrease access to tests.

Patents on genetic information and uses thereof will not only limit the availability of current medical diagnostics and treatments, they will also restrict the availability of future health innovations. These patents would empower their owners to block competition from those who would otherwise offer more effective or affordable tests and treatments. This exclusionary power would, in turn, deter researchers from developing new tests and treatments for life-threatening and debilitating conditions, like cancer and Alzheimer’s disease. The result will be medical care that is more expensive, less innovative, and less accessible than it is today.

If PERA were enacted, it would create a patent environment similar to the time before the Myriad and Mayo decisions and as such, biomarkers (including human genes) and their association with health status would once again be patent eligible. The answer to your question is not hypothetical in that we have a real world test case of what this would mean for consumers and patients, as we saw the consequences pre-2013 and the benefits since then. Specifically, prior to 2013, patient access to genetic testing was greatly limited due to testing monopolies stemming from patents on human genes. I refer you back to my written testimony beginning on page 6 in which I describe at length multiple examples of these consequences for patients.

To summarize my written testimony, consumers and the broader healthcare system should expect dramatic increases in the cost of testing and concomitant reduced access due to reduced competition. PERA would return us to a pre-Myriad environment with genetic testing monopolists charging high prices for partial results and no provider able to offer comprehensive testing as is offered by multiple genomic testing companies today. These consequences will be particularly felt by lower income patients if they live in states with Medicaid programs that do not cover the only test available due to a monopoly. Further, patients will not have access to tests reflecting the latest scientific advances because just as was the case during the pre-Myriad period, providers having a patent on a specific necessary test will lack an adequate business incentive for them to update their tests. Most concerning of all, patients in the United States may find themselves with no access at all to critical, lifesaving genetic information if a patent holder opts not to commercialize a test nor license it, which was the case of long QT syndrome pre-2013.

When innovation scholars think about the means that governments use to encourage R&D, patents are probably the first tool that come to mind. The standard account of patents as innovation incentives is well known.78 Inventions are public goods that are both expensive to develop and difficult to prevent others from copying.79 Like other public goods, these features suggest that the market will not optimally supply desired inventions. If an innovator would have to spend a million dollars to develop an invention that could easily be copied by its rivals, the innovator simply won’t bother to invest its money in the first place. Patents solve the public goods problem by giving inventors exclusive rights to their inventions for a limited period of time.80 This period of exclusivity may enable inventors to charge supra-competitive prices for products that embody their inventions, thus allowing them to recoup their R&D costs and make a profit.81 Unlike some of the incentives described above, the government only issues patent rights to people who actually produce an invention. In order to obtain patents, inventors must demonstrate that they have generated something that is both novel and nonobvious to other skilled inventors.82 And, in theory at least, their inventions are required to be “useful.”83

Patents differ from prizes, grants, and tax incentives in another way. While the latter are paid directly by the government (and, thus, taxpayers) through spending, issuing patents doesn’t cost the government much.84 But patents definitely have a price. Patent incentives are financed by consumers of products that incorporate patents who must pay higher prices for those goods.85 By enabling patent owners to charge supracompetitive prices, patents act as a tax paid by consumers for the benefit of innovators.86 Thus, unlike other innovation expenditures, when governments promote innovation with the exclusive rights patents confer, their expenditures are “off-budget.”87 This means that it is enormously difficult to estimate the size of society’s contribution to innovation promotion through higher prices for patented goods.88

Finally, some have also argued that concerns about PERA should be ignored because the expanded availability of patents, in and of itself, would enhance U.S. competitiveness. While providing a comprehensive response to this claim is beyond the scope of my testimony, it should suffice to point out that the majority of patents granted by the USPTO claim foreign inventions and were issued to foreign applicants. The principal effects of enacting PERA would thus be to incentivize foreign innovation and enrich foreign patent owners. This would come at the expense of U.S. businesses and consumers, because substantially all the harms with respect to increased liability, greater business uncertainty, and higher litigation costs would fall on those doing business in the United States, with most of these costs ultimately being passed on to American consumers in the form of higher prices. There can be no serious argument that subsidizing foreign innovation or enriching foreign entities at the expense of American consumers and businesses somehow benefits the United States. This alone should cast serious doubt on claims that an unprecedented expansion of patent eligibility is justified simply because it would allow for increased patenting.

In sum, HTIA and its member companies are deeply concerned about the broad expansion of patentable subject matter that would result from the enactment of PERA as currently drafted. Both empirical evidence and historical experience indicate that expanding patent eligibility to include nontechnological subject matter will undermine incentives to invest in technological innovation and impose substantial dead-weight losses on the U.S. economy, thereby harming U.S. competitiveness at a time when retaining our technological and economic advantages relative to adversarial nations is critical to our national security.

Regarding the impact on consumers, the whole purpose of the patent system is to provide inventors with exclusive rights that allow them to constrain competition and charge higher prices. The increased profit from these higher prices is the incentive to innovate that is provided by the patent system. In a well-functioning patent system, the hope is that consumers will benefit more from the increased pace of innovation than they will be harmed by increased prices. Unfortunately, PERA would substantially increase prices to consumers without producing any significant countervailing benefits from innovation. This is because PERA would expand patent eligibility to non-technological fields, such as business methods, where patent protection has been shown to reduce rather than increase R&D investment.2 Thus, the predictable impact of PERA on consumers would be to increase prices and reduce selection, while at the same time depriving them of the benefits they otherwise would have had from increased innovation.

There is an old saying that in Washington, no bad idea stays dead for long. The Patent Eligibility Restoration Act (PERA), proposed by Senators Thom Tillis (R-NC) and Chris Coons (D-DE) seems designed to prove that proposition.

The basic idea behind the act is to increase the types of things for which people and corporations can get government-granted patent monopolies. These could raise the price of a wide range of items from prescription drugs and medical procedures to computer software and smartphones. It is also likely to mean billions of dollars more being wasted every year in patent suits and related litigation.

This paper examines empirically the effect of patent protection on the price of an array of consumer products. We collect original data on patent-product associations and study the effect of an exogenous loss of patent protection on product prices. We have matched 2778 patents to 825 products available on the Amazon.com e-commerce website and have tracked the prices of these products for a period of up to eight years. We study the change in price around the time of patent expiry. Because patent protection is limited in time by law, patent expiry is exogenous to the quality of the underlying invention or to its commercial value. Furthermore, we are able to isolate the effect of patent expiry from the effect of product depreciation by controlling for product model displacement and product age. The empirical analysis then explores the heterogeneous effects of patent expiry across patent type and importance. It also portrays the price evolution around the time when patent terms expire. Finally, it considers how prices react to the intensity of the competition in the product market.

We find that patent expiry is associated with a 7-8 percent drop in product prices, and that the effect is larger for more important patents (e.g., patents that claim a broader slice of technology). We observe that the price starts dropping about one year before patent expiry, possibly suggesting strategic entry deterrence from the incumbent (Milgrom and Roberts,1982;Goolsbee and Syverson,2008). We also observe that the decline in price is more pronounced in more competitive markets, with some evidence of a U- shape relationship between the price drop and the level of the competitive pressure. Finally, placebo tests on samples of fake patent expiry events confirm the validity of our identification strategy.

As it stands, the Sahm rule is not pointing to a recession. The U.S. added nearly 200,000 jobs in November, which helped to stop an upward climb in the unemployment rate, which fell to 3.7 per cent from 3.9 per cent.

Still, the recent increase in the jobless rate – 0.3 percentage points, based on the Sahm rule’s formula – is within spitting distance of signalling a recession.

The inverted yield curve is Wall Street’s most obsessed-over recession indicator, with a track record of foreshadowing every downturn since 1960 with no false signals. But since last year it’s been screaming downturn and no U.S. recession has come, leading some analysts to wonder if the metric is broken.

Not so fast, according to Duke University finance professor Campbell Harvey, the Canadian-born economist who pioneered it.

Weaker consumer spending is all according to the Fed’s plan: cool economic activity and thereby inflation, hopefully without inducing a recession. We believe Q2 and Q3 real GDP growth will be anemic (between 0% and 1% q/q annualized). Still a recession is unlikely because businesses are still generally willing to hold on to their best talent.

Against this backdrop, we anticipate that slower real GDP growth and inflation, plus a relatively healthy labor market will enable the Fed to cut interest rates twice this year – potentially at the November and December meetings. Indeed, only a slight majority of FOMC participants desired one cut at the June meeting, with others mostly anticipating two cuts.

The recession, predicted by business executives, economists, and investors, refuses to show up.

Steady hiring continues to fuel consumer spending and, in turn, an economic expansion unlike any the U.S. has seen. Employers added 2.75 million jobs over the last 12 months, including 272,000 in May, the Labor Department said Friday.

US economic data points to a soft landing as the labor market and inflation gradually cool.

Until recently those two forces were holding back a soft-landing narrative.