VoxEU Column Competition Policy Productivity and Innovation

Competition and R&D: Evidence from biopharma

R&D-intensive firms such as biopharmaceutical companies operate in a competitive and risky environment. This column presents new evidence on how competition affects the investment decision of R&D-intensive firms. An increase in competition will make the firm increase the R&D investment, and as a response the firm will carry more cash and reduce its debt. Also, more competition will increase the idiosyncratic risk of R&D-intensive firms.

Investment and finance decisions of biopharmaceutical companies

The development of new drugs to treat diseases is an important social good and a significant driver of economic growth and welfare. As a result, most governments have an interest in promoting such investments. However, R&D-intensive firms such as biopharmaceutical companies operate in an increasingly competitive and risky environment, and investments in R&D are often very large and with highly uncertain payoffs.

What induces biopharmaceutical firms to make these investments, and what are the factors that affect them and the willingness of investors to support them? To answer these questions, we need to understand the economics underlying the investment and financing decisions of these firms, the evolution of their risk characteristics, and the impact of competition.

While current theories of the corporation make numerous predictions about many industries, they may not apply to R&D-intensive firms because of these firms’ special characteristics. For example, R&D expenditures by such firms often dwarf spending on property, plant, and equipment, and capital budgeting and financing for R&D can differ significantly in nature from that of other capital projects due to the high-risk and staged nature of R&D investment and the absence of observable post-investment cash flows over many periods (see Myers and Howe 1997, who lay out these issues for the pharmaceutical industry). Moreover, competition in R&D-intensive industries affects the likelihood of winning a patent race associated with a given R&D investment as well as the profits following patent expiration. Nowhere has the effect of competition been more evident than in the biopharmaceutical (biopharma) industry, where competitive pressures have increased due to changes in regulation, lower costs of entry due to improvements in technology, and the expiration of patents combined with high development costs of new therapeutics (e.g. Caves et al. 1991, Grabowski and Vernon 1992, among others). These factors have squeezed margins from existing products associated with assets-in-place, with significant implications for R&D investments.

New evidence

In our study, we examine how the competitive environment affects an R&D-intensive firm’s investment and capital structure decisions, as well as its financial risk characteristics (Thakor and Lo 2015). The reason for focusing on competition is that this has been one of the most salient changes in the industry, and also the subject of much policy debate in terms of the pros and cons of allowing the industry to become more competitive, e.g., would higher competition impair or facilitate R&D investments?  To address this, we develop a theory that relates how competition interacts with and affects the firm's decisions and present supporting empirical evidence. While the focus is on biotechnology and pharmaceutical firms, our framework and analysis are applicable to other R&D-intensive firms as well.

Analytical predictions

We begin by considering a biopharma firm that faces a decision to undertake a staged R&D investment and must fund this investment through the capital market. We develop predictions for how changes in competition affect the firm’s decisions.

  • Our first prediction is that an increase in competition will make the firm increase R&D investment and reduce investment in assets-in-place that support existing products. 

The economic driver of this effect is that higher competition—due either to technological advances or structural changes introduced by laws—increases the marginal benefit of investing in R&D compared to existing products. For example, one of the reasons why firms in the biopharma industry engage in R&D is to replace old drugs (many of which may be off-patent and thus face competitive pressures) with new drugs (which are patent-protected and insulated from competition for some time).

  • Our second prediction is that, in response to an increase in competition in the industry, the biopharma firm will carry more cash and reduce its debt.

The reason is that R&D-intensive firms must release information to investors to get funding. However, this information also (inadvertently) leaks to the firm’s competitors, potentially increasing competitive entry and thus damaging the future viability of the R&D given the winner-takes-it-all nature of the patent process. A firm that raises money from capital markets after it learns that its R&D is viable will unavoidably signal some of this information to competitors. The firm can avoid this leak by raising financing when it knows less about the success of the R&D and build up cash reserves from this ‘early’ financing to avoid subsequent fundraising from capital markets. The reason for reducing financial leverage is that, as the firm shifts its investments to R&D and away from assets-in-place, it has less collateral to support debt.

  • Our third prediction is that an increase in competition will affect the risk characteristics of R&D-intensive firms by lowering their betas but also increasing their idiosyncratic risk.

The intuition is that existing products/assets of the firm carry only systematic risk, whereas R&D has largely idiosyncratic risk (by definition, R&D involves new projects that are one-off stand-alone investments by individual firms, and hence uncorrelated with the economy before they yield commercial products). Existing products involve similar investments by many other firms, and hence contain systematic risk. For example, in the biopharma industry, new drugs have patent protection and are thus less affected by investments by other firms due to the monopoly that patents confer on the specific drug. In contrast, existing products include generic drugs that no longer enjoy patent protection and are sensitive to the investments of other firms.

Empirical evidence

To determine whether competition affects R&D-intensive firms in the ways we predict, we examine historical data from the biopharma industry. We first document competition over time for the biopharma industry in Figure 1, through two measures: The concentration ratio and the number of competitors. As Figure 1 shows, these measures indicate that competition has increased for the biopharma industry over time. 

Figure 1. Competition in the biopharma industry

Notes: These figures present estimates of competition over time for the biopharma industry. The top figure gives the 4-rm Concentration Ratio, and a higher number indicates increased concentration. The bottom figure gives the number of competitors in the biopharma industry over time.

We then examine trends for the industry over time in terms of its investments and financial characteristics. Figure 2 depicts the average level of R&D investments, assets-in-place (proxied by property, plant, and equipment), cash holdings, debt, and net debt (debt minus cash). Consistent with the predictions of our framework, as competition has increased, R&D investments have increased while investment in assets-in-place has decreased. In addition, cash holdings have increased, while debt and net debt have generally decreased.1 For all of the variables, the trends for the biopharma industry are more striking than those of other industries, suggesting that the biopharma trends are not driven by aggregate trends affecting all industries.

Figure 2. Financial characteristics over time

Notes: These graphs show the mean (solid blue line) and median (dashed red line) values of financial characteristics for the biopharma industry in each year. The green dotted lines represent the mean values of financial characteristics for all other industries.

For the risk characteristics, Figure 3 shows beta and volatility estimates for the biopharma industry over time. Again, consistent with our predictions, the Fama-French 3-factor betas have generally declined over time, while idiosyncratic volatility has increased over time.

Figure 3. Biopharma betas and idiosyncratic volatility

Notes: The top figure shows the betas of a value-weighted portfolio of biopharma stocks, calculated via the Fama-French 3-factor model. The blue line represents the market factor, the green line represents the market-to-book factor (HML), and the red line represents the size factor (SMB). The dotted lines are trend lines for the factors. The bottom figure gives estimates of idiosyncratic volatility.

Further evidence from new legislation

Additional support for our model’s predictions comes from a natural experiment involving a legislative change in 1984. The Drug Price Competition and Patent Term Restoration Act of 1984 (informally known as the Hatch-Waxman Act) was introduced for the express purpose of increasing competition in the biopharma industry by facilitating the entry of generic drugs after the expiration of a patent. Prior to the passage of the Hatch-Waxman Act, onerous Food and Drug Administration requirements made it necessary for generic drugs to replicate many of the original drug's clinical trials to gain regulatory approval. However, once the law was passed, generic drugs only needed to demonstrate bioequivalence to the original drug, greatly decreasing the barriers to competitive entry. A number of papers have provided evidence that the Hatch-Waxman Act did indeed increase competition and facilitate the entry of generic drugs (see, for example, analysis and evidence by Grabowski and Vernon 1986, 1992, who look at entry and price data for a sample of drugs after the enactment of the law).

We, therefore, examine the effect of this new law on the biopharma industry compared to other R&D-intensive firms with similar characteristics and to which the law did not apply. Figure 4 displays the average values of various financial variables of the biopharma industry (solid blue) and other R&D-intensive industries (dashed red lines) for the 20 years surrounding the passage of the Hatch-Waxman Act (the vertical red lines represent the year that the Hatch-Waxman Act was implemented). The levels of R&D expenditures, cash holdings, debt, net debt, and assets-in-place are all similar for both the biopharma and other R&D-intensive firms in the pre-period, confirming that the two industries are similar in terms of these financial characteristics. After the Act was implemented, the values for the two groups diverge in ways consistent with our model’s predictions. Specifically, in the period following the enactment of the law, R&D expenditures and cash holdings for biopharma firms increased while debt, net debt, and assets-in-place decreased relative to the trends for the other firms.

Figure 4. Financial characteristic trends for biopharma and other firms

Notes: Trends for financial characteristic variables for R&D expenditures, cash holdings, debt, net debt, and assets-in-place, all scaled by total assets. All variables are averages for each group. The solid blue lines give averages for the biopharma industry, while the red dashed lines give averages for a matched sample of R&D-intensive rms. A vertical red line is included in each graph, representing the year that the Hatch-Waxman Act was implemented.

Concluding remarks

Our research highlights the importance of some unique features of R&D-intensive firms, and explains some interesting secular trends in their financial characteristics and risk over time. This is by no means a complete characterisation of the financial features of such firms, but rather a starting point for understanding their unique features. Our research also provides a framework for understanding how policy changes that may affect competition in R&D-intensive industries can affect R&D and other investment decisions, as well as the risk of such industries and, therefore, capital flows into and out of these industries.

References

Caves, R E, M D Whinston, and M A Hurwitz (1991), “Patent Expiration, Entry, and Competition in the U.S. Pharmaceutical Industry”, Brookings Papers on Economic Activity: Microeconomics Vol. 1991, 1-66.

Grabowski, H, and J Vernon (1986), "Longer Patents for Lower Imitation Barriers: The 1984 Drug Act", The American Economic Review 76, no. 2: 195-98.

Grabowski, H, and J M Vernon (1992), "Brand Loyalty, Entry, and Price Competition in Pharmaceuticals after the 1984 Drug Act", Journal of Law and Economics 35, no. 2: 331-50.

Myers, S C, and C D Howe (1997), "A Life-Cycle Financial Model of Pharmaceutical R&D", Program on the Pharmaceutical Industry, Sloan School of Management, Massachusetts Institute of Technology.

Thakor R T and A W Lo (2015), “Competition and R&D Financing Decisions: Theory and Evidence from the Biopharmaceutical Industry”, NBER Working Paper No. 20903.

Footnote

1 While the mean level of debt has increased over time (mostly in the 1970s and the 2000s), the median level of debt has declined consistently from the mid-1970s. The debt levels are cross-sectionally skewed across firms, with some firms holding very large amounts of debt—this drives the mean values upwards. But the median debt levels indicate that the majority of firms have decreased their debt levels in the industry. Net debt shows a similar trend, although the decline in both mean and median values are more marked until the late-1990s. While the mean level increases after that point (concurrent with the increase in debt), the median level of net debt stays relatively flat, consistent with how the competition measures behaved over this period.

3,150 Reads