HOW INCREASED COMPETITION FROM
`GENERIC DRUGS HAS AFFECTED PRICES
`AND RETURNS IN THE PHARMACEUTICAL INDUSTRY
`
`JULY 1998
`
`The Congress of the United States
`Congressional Budget Office
`
`Exhibit 1088
`IPR2017-00807
`ARGENTUM
`
`000001
`
`

`

`NOTES
`
`The numbers in the text and tables of this study may not add up to totals because of rounding.
`
`Cover photo ©The Stock Market/Dennis Meyler.
`
`000002
`
`

`

`Preface
`In 1984, the Drug Price Competition and Patent Term Restoration Act (also known as
`
`the Hatch-Waxman Act) created an abbreviated approval process for generic prescrip-
`tion drugs and at the same time extended patent terms for innovator drugs. This
`Congressional Budget Office (CBO) study examines the extent to which competition from
`generic drugs has increased since the act. It also analyzes how that competition has af-
`fected the returns from developing a drug. The analysis was conducted at the request of
`the Chairman of the Senate Committee on the Budget.
`
`Anna Cook of CBO's Natural Resources and Commerce Division wrote the study
`under the supervision of Jan Paul Acton and Elliot Schwartz. The analysis would not have
`been possible without data and information provided by the Food and Drug Administration
`(FDA), the Patent and Trademark Office (PTO), the Health Care Financing Administra-
`tion, and Henry Grabowski of Duke University. A variety of industry experts provided
`information and insights, including Philip Chao and Donald Hare of the FDA, Karin Tyson
`of the PTO, Joel Hamilton of the General Accounting Office, David Reiffen of the Federal
`Trade Commission, Paul Wilson of IMS America, and Gary Persinger of the Pharmaceuti-
`cal Research and Manufacturers of America (now of the National Pharmaceutical Coun-
`cil). Other outside reviewers included the following economics professors: Ernst Berndt
`and Scott Stern of MIT, Fiona Scott Morton of Stanford, David Salkever of Johns
`Hopkins, and F.M. Scherer of Harvard. Within CBO, John Peterson, Linda Bilheimer,
`Judith Wagner, Patrice Gordon, and Anne Cappabianca (now at Hoffman-La Roche) made
`extensive and valuable comments. Aaron Zeisler and Carl Muehlmann provided research
`assistance.
`
`Christian Spoor edited the manuscript, and Melissa Burman proofread it. Angela
`McCollough typed the many drafts. Kathryn Quattrone prepared the study for publication,
`and Laurie Brown prepared the electronic version for CBO's World Wide Web site.
`
`July 1998
`
`June E. O'Neill
`Director
`
`This study and other CBO publications
`are available at CBO's Web site:
`http://www.cbo.gov/
`
`000003
`
`

`

`
`
`000004
`
`000004
`
`

`

`Contents
`
`SUMMARY
`
`ONE
`
`INTRODUCTION
`
`The Basis for Competition Among Drug Companies 2
`Changes Made by the Hatch-Waxman Act 3
`Data Used in This Analysis 4
`
`TWO
`
`THE EFFECT OF MANAGED CARE ON
`THE PHARMACEUTICAL MARKET
`
`The Rise of Managed Care 5
`How PBMs Help Hold Down Drug Expenditures 6
`How Managed Care Affects the Demand
`for Prescription Drugs 10
`Conclusions 11
`
`THREE
`
`PRICING AND COMPETITION IN THE
`PHARMACEUTICAL MARKET
`
`Competition Among Brand-Name Drugs 14
`Factors That Determine Discounts on Brand-
`Name Drugs 23
`Competition Between Brand-Name and
`Generic Drugs 27
`Competition Among Generic Drugs 32
`Conclusions 34
`
`FOUR
`
`THE EFFECTS OF THE HATCH-WAXMAN ACT
`ON THE RETURNS FROM INNOVATION
`
`Changes to the Length of Patents for Brand-
`Name Drugs 38
`Changes to the Approval Process for Generic
`Drugs 43
`Effects on the Returns from Marketing a Drug 45
`Effects of Proposed Changes to the Hatch-
`Waxman Act 49
`Conclusions 50
`
`ix
`
`1
`
`5
`
`13
`
`37
`
`000005
`
`

`

`vi HOW INCREASED COMPETITION FROM GENERIC DRUGS
`
`July 1998
`
`APPENDIXES
`
`A
`
`B
`
`C
`
`D
`
`Data Used for the Empirical Estimates 53
`
`Regression Results on Discounting 59
`
`Assumptions Behind the Calculation of Returns
`from Marketing a New Drug 65
`
`The Replacement Effect 73
`
`000006
`
`

`

`CONTENTS
`
`TABLES
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`
`10.
`
`A-1.
`
`B-1.
`
`B-2.
`
`B-3.
`
`C-1.
`
`C-2.
`
`C-3.
`
`Market Share and Average Retail Prescription Price,
`by Type of Drug, 1994
`
`Average Time from Clinical Testing to Final Approval
`for an Innovator Drug
`
`Percentage of New Drugs Acquired Rather Than Self-
`Originated by U.S.-Owned Drug Companies
`
`Average Price Differences for Various Types of
`Purchasers in the Pharmaceutical Market
`
`Price Comparison of Generic and Innovator Drugs,
`by Number of Manufacturers, 1994
`
`Generic Subsidiaries or Divisions of Brand-Name Manufacturers
`
`Changes in Patent Protection for U.S. Pharmaceuticals
`
`Average Length of Hatch-Waxman Extensions for
`Drugs Approved Between 1992 and 1995
`
`Limits on Hatch-Waxman Extensions for Drugs
`Approved Between 1992 and 1995
`
`Reasons That Some Drugs Approved Between 1992
`and 1995 Did Not Receive a Hatch-Waxman Extension
`
`Data and Methods Behind CBO’s Estimates
`
`Variables Used in the Regression Analysis of Discounting
`
`Regression Results on Price Dispersion in 1994
`
`The Effects of Generic and Brand-Name Competition
`on Price Dispersion
`
`Assumptions Used to Calculate the Change in Returns
`from Marketing a Drug
`
`Formulas for Calculating Generic Market Share
`
`How Sensitive Is the Calculation of Returns to Changes
`in the Base-Case Assumptions?
`
`vii
`
`15
`
`17
`
`22
`
`25
`
`33
`
`34
`
`39
`
`40
`
`41
`
`41
`
`54
`
`61
`
`63
`
`64
`
`67
`
`69
`
`71
`
`000007
`
`

`

`viii HOW INCREASED COMPETITION FROM GENERIC DRUGS
`
`July 1998
`
`FIGURES
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`
`BOXES
`
`1.
`
`2.
`
`3.
`
`4.
`
`C-1.
`
`D-1.
`
`How PBMs Fit Into the Payment System for
`Prescription Drugs
`
`Channels of Distribution for Prescription Drugs
`
`Change in the Profit Stream for a Typical
`Innovator Drug
`
`Choosing a Profit-Maximizing Price for a Drug
`
`Market Share of the Top Three Innovator Drugs
`in 66 Therapeutic Classes, 1994
`
`Growth in the Market Share of Generic Drugs
`Since 1984
`
`The Average Profit Stream for a Brand-Name Drug
`Before and After the Hatch-Waxman Act
`
`Types of Prescription Drugs
`
`The Role of Changes in State Drug-Product
`Substitution Laws
`
`Defining Therapeutic Classes of Drugs
`
`Studies of How Generic Entry Affects Brand-Name Prices
`
`Calculating the Probability of Generic Entry
`
`Calculating the Impact of the Replacement Effect and
`Generic Competition on the Returns from Innovation
`
`8
`
`14
`
`16
`
`19
`
`23
`
`27
`
`46
`
`2
`
`7
`
`23
`
`30
`
`68
`
`74
`
`000008
`
`

`

`Summary
`
`The pharmaceutical market has become increas-
`
`ingly competitive since the early 1980s, in part
`because of the dramatic growth of the generic
`drug industry. In 1996, 43 percent of the prescription
`drugs sold in the United States (as measured in total
`countable units, such as tablets and capsules) were
`generic. Twelve years earlier, the figure was just 19
`percent. Generic drugs cost less than their brand-
`name, or "innovator," counterparts. Thus, they have
`played an important role in holding down national
`spending on prescription drugs from what it would
`otherwise have been. Considering only sales through
`pharmacies, the Congressional Budget Office (CBO)
`estimates that by substituting generic for brand-name
`drugs, purchasers saved roughly $8 billion to $10 bil-
`lion in 1994 (at retail prices).
`
`Three factors are behind the dramatic rise in
`sales of generic drugs that has made those savings
`possible. First, the Drug Price Competition and Pat-
`ent Term Restoration Act of 1984—commonly known
`as the Hatch-Waxman Act—made it easier and less
`costly for manufacturers to enter the market for ge-
`neric, nonantibiotic drugs. Second, by 1980, most
`states had passed drug-product substitution laws that
`allowed pharmacists to dispense a generic drug even
`when the prescription called for a brand-name drug.
`And third, some government health programs, such as
`Medicaid, and many private health insurance plans
`have actively promoted such generic substitution.
`
`Greater sales of generic drugs reduce the returns
`that pharmaceutical companies earn from developing
`brand-name drugs. The Hatch-Waxman Act aimed to
`
`limit that effect by extending the length of time that a
`new drug is under patent—and thus protected from
`generic competitors. Those extensions compensate for
`the fact that part of the time a drug is under patent it is
`being reviewed by the Food and Drug Administration
`(FDA) rather than being sold. The act tried to balance
`two competing objectives: encouraging competition
`from generic drugs while maintaining the incentive to
`invest in developing innovative drugs. It fell some-
`what short of achieving that balance, however, in part
`because the act shortened the average time between the
`expiration of a brand-name drug's patent and the ar-
`rival of generic copies on the market (so-called generic
`entry) from more than three years to less than three
`months. More important, it also greatly increased the
`number of drugs that experience generic competition
`and, thus, contributed to an increase in the supply of
`generic drugs. In the end, the cost to producers of
`brand-name drugs from faster generic entry has
`roughly offset the benefit they receive from extended
`patent terms. Meanwhile, the greater competition
`from generic drugs has somewhat eroded their ex-
`pected returns from research and development.
`
`CBO estimates that those factors have lowered
`the average returns from marketing a new drug by
`roughly 12 percent (or $27 million in 1990 dollars).
`In this study, "returns from marketing a new drug"
`refers to the present discounted value of the total
`stream of future profits expected from an average
`brand-name drug. Previous studies estimate that those
`profits had an average present discounted value of
`$210 million to $230 million (in 1990 dollars) for
`drugs introduced in the early 1980s. Those returns are
`
`000009
`
`

`

`x HOW INCREASED COMPETITION FROM GENERIC DRUGS HAS AFFECTED PRICES AND RETURNS
`
`July 1998
`
`valued at the date of market introduction, after sub-
`tracting production costs but not the costs of research
`and development. Also, because the drugs in those
`studies were not eligible for the patent-term extensions
`provided by the Hatch-Waxman Act, those estimates
`do not account for the benefits of the extensions now
`available under the act. Thus, those figures can be
`considered a minimum estimate of the returns from
`marketing. Only part of the estimated decline in re-
`turns can be attributed to the Hatch-Waxman Act; the
`other factors that have boosted sales of generic drugs
`have played a role as well.
`
`This study relies on a variety of data to produce
`its estimates, including a data set that represents about
`70 percent of prescription drug sales through retail
`pharmacies in the United States. The various sets of
`data all have strengths and weaknesses, which are dis-
`cussed along with the estimates they generate. In gen-
`eral, the empirical estimates in this study are rough
`rather than precise measures. They help characterize
`the increase in competition in the pharmaceutical mar-
`ket and its effects on the profits of drug manufacturers
`and the prices paid for prescription drugs.
`
`The Effects of Managed Care
`on the Pharmaceutical Market
`
`At the same time that the Hatch-Waxman Act has
`helped increase the supply of generic drugs, changes in
`the demand for pharmaceuticals have affected the fre-
`quency with which generic and brand-name drugs are
`prescribed and the prices paid for them. Those
`changes in demand were brought on by newer forms of
`health care delivery and financing. In particular, be-
`cause of competitive pressure in the health insurance
`market, more private-sector health plans have adopted
`managed care techniques in an effort to hold down
`overall health spending. The net effect of those tech-
`niques on spending for prescription drugs, however, is
`unclear.
`
`On the one hand, many health plans (including
`traditional fee-for-service plans) hold down drug costs
`by "managing" their outpatient prescription drug
`benefits—either themselves or through organizations
`called pharmaceutical benefit management companies
`
`(PBMs). Those plans and PBMs use computer net-
`works at pharmacies and electronic card systems for
`enrollees that allow pharmacists, before filling an
`enrollee's prescription, to consult a list (or formulary)
`of the plan's suggested drugs. Formularies typically
`encourage substituting brand-name drugs with generic
`versions, or sometimes with other, less expensive
`brand-name drugs. Savings result not only because of
`that substitution but also because many manufacturers
`of brand-name drugs offer discounts to health plans or
`PBMs in exchange for being included on their formu-
`lary. In addition, because they represent a large pool
`of customers, PBMs can negotiate with pharmacies
`over the retail prices charged for prescriptions. Since
`the late 1980s, those various techniques have been
`putting downward pressure on the prices that PBMs
`and health plans pay for prescription drugs sold
`through pharmacies.
`
`On the other hand, health maintenance organiza-
`tions (HMOs) and some other managed care plans fre-
`quently charge lower copayments for health care ser-
`vices—including physician visits and prescription
`drugs—than traditional fee-for-service plans do.
`Those lower copayments may lead to greater use of
`prescription drugs by beneficiaries. The treatment
`practices of HMOs may also favor more intensive use
`of prescription drugs, perhaps as an alternative to
`costlier forms of treatment. As a result, the increasing
`prevalence of managed care plans may have helped
`boost the quantity of prescription drugs sold in the
`United States.
`
`For brand-name drugs still under patent (which
`do not yet have generic competitors), managed care
`techniques may have only a small effect on profits,
`assuming that greater use offsets the downward pres-
`sure on prices. For brand-name drugs whose patents
`have expired, however, profits are probably lower than
`they would have been without the generic substitution
`promoted in part by managed care plans and PBMs;
`that substitution has cut dramatically into the market
`share of those drugs. (CBO's calculation of the
`change in returns accounts for the full increase in ge-
`neric market share since 1984, part of which is attrib-
`utable to the rise in managed care techniques, but it
`does not measure managed care's effect on profitabil-
`ity through other variables, such as increases in pre-
`scription drug use and changes in pricing.)
`
`000010
`
`

`

`SUMMARY
`
` xi
`
`Pricing and Competition in the
`Pharmaceutical Market
`
`Competition in the pharmaceutical market takes three
`forms: among brand-name drugs that are therapeuti-
`cally similar, between brand-name drugs and generic
`substitutes, and among generic versions of the same
`drug. Manufacturers of brand-name drugs compete
`for market share primarily through advertising and the
`quality of their products (including efficacy and side
`effects), as well as through pricing. Manufacturers of
`generic drugs increase their market share mainly by
`lowering prices. (In general, companies produce either
`generic or brand-name drugs, not both, although some
`generic manufacturers are subsidiaries of brand-name
`manufacturers.)
`
`Competition Among Brand-Name
`Drugs
`
`Patents do not grant complete monopoly power in the
`pharmaceutical industry. The reason is that compa-
`nies can frequently discover and patent several differ-
`ent drugs that use the same basic mechanism to treat
`an illness. The first drug using the new mechanism to
`treat that illness—the breakthrough drug—usually has
`between one and six years on the market before a ther-
`apeutically similar patented drug (sometimes called a
`"me-too" drug) is introduced. Economic theory and
`various studies suggest that the presence of several
`therapeutically similar drugs limits manufacturers'
`ability to raise prices as much as would otherwise be
`the case. In addition, brand-name manufacturers are
`more likely to agree to give purchasers a discount if
`those purchasers have the option of switching to a ge-
`neric or me-too competitor.
`
`The factors that limit the number of similar but
`slightly differentiated brand-name drugs on the market
`are unclear. In some cases, perhaps, only a limited
`number of slightly different chemicals that target a
`given enzyme can be developed into drugs. Or, as one
`economist has suggested, the high cost of developing a
`drug may limit the number of similar brand-name
`drugs that are eventually brought to market. Compa-
`nies will undertake such investment only if they be-
`
`lieve the market is not already saturated or their drug
`has some quality advantage that could enable it to
`compete effectively and earn an adequate return. For
`that reason, competition among patented brand-name
`drugs probably results in companies' earning roughly a
`normal rate of return on their investment in research
`and development (R&D), on average.
`
`Overall, the pharmaceutical market is not highly
`concentrated, but when that market is divided into nar-
`rowly defined therapeutic classes, it becomes quite
`concentrated. The top manufacturers of brand-name
`drugs, ranked by pharmaceutical sales, each account
`for no more than 7 percent of the entire market for
`prescription drugs (which totaled $60.7 billion in 1995
`at manufacturer prices). Within each therapeutic
`class, however, higher levels of concentration appear.
`In 35 of the 66 therapeutic classes that CBO examined
`in this study, the top three innovator drugs together
`constituted at least 80 percent of retail pharmacy sales
`in their class.
`
`Studies of the average prices paid by pharmacies
`and hospitals have shown that manufacturers of
`brand-name drugs do compete with each other through
`pricing. The markups they charge over the marginal
`cost of producing a drug are consistent with economic
`models of price competition in which entry by manu-
`facturers is limited (such as by patents). Offering dis-
`counts to some buyers may also be an important di-
`mension of price competition for brand-name drugs.
`But its extent is difficult to measure because of lack of
`data.
`
`Discounts on Brand-Name Drugs
`
`Different buyers pay different prices for brand-name
`prescription drugs. In theory, when companies are
`permitted to charge different types of purchasers dif-
`ferent prices, those purchasers least sensitive to price
`will pay the most. In today's market for outpatient
`drugs, purchasers that have no insurance coverage for
`drugs, or third-party payers that do not use a formu-
`lary to manage their outpatient drug benefits, pay the
`highest prices for brand-name drugs.
`
`Manufacturers offer discounts on brand-name
`drugs based not only on the volume purchased but also
`
`000011
`
`

`

`xii HOW INCREASED COMPETITION FROM GENERIC DRUGS HAS AFFECTED PRICES AND RETURNS
`
`July 1998
`
`on the buyer's ability to affect the drug's market share
`by using a formulary to systematically favor one
`brand-name drug over another for a large number of
`patients. Pharmacies themselves do not generally pro-
`mote substitution between brand-name drugs, so they
`do not generally receive large discounts or rebates
`from manufacturers. Rather, it is the PBMs and in-
`surers who manage benefits for drugs sold through
`pharmacies that promote brand-name substitution and
`receive discounts.
`
`Such price discrimination, or discounting, may
`be an important mechanism for facilitating price com-
`petition in the pharmaceutical market. It rewards in-
`stitutional purchasers that organize their patient base
`through formularies so as to encourage the use of less
`costly drugs. Prohibiting discounts, as some policy-
`makers have called for, could decrease price competi-
`tion.
`
`Drug companies usually do not make their dis-
`counts public, but CBO was able to obtain limited
`information on the prices paid by different types of
`purchasers for prescription drugs. The prices that
`pharmacies pay can be seen as a proxy for the final
`price paid by customers who do not have a managed
`drug benefit or PBM to negotiate rebates from manu-
`facturers. Based on the average invoice prices for top-
`selling drugs sold primarily to retail pharmacies, hos-
`pitals and clinics pay 9 percent less than retail phar-
`macies, on average, and HMOs pay 18 percent less.
`Federal facilities, such as veterans' hospitals, get an
`even more substantial discount—over 40 percent, on
`average, compared with the price paid by retail phar-
`macies. (Those comparisons are based only on in-
`voice prices, so they do not account for rebates and
`other types of discounts that do not appear on in-
`voices.)
`
`Statistical analysis shows that manufacturers'
`discounts on brand-name drugs tend to be higher when
`more generic and me-too drugs are available. That
`analysis is based on the difference between the average
`price paid by pharmacies and the lowest price paid by
`any private purchaser in the United States (the best-
`price discount), as reported under the Medicaid drug
`rebate program. CBO found that the best-price dis-
`count for a brand-name drug was 10 to 14 percentage
`points greater when a generic version was available
`from four or more manufacturers. That analysis also
`
`showed that as the number of brand-name manufactur-
`ers in a therapeutic class increases from one to five,
`the best-price discount grows by 10 percentage points.
`Those statistical results imply that discounts are at
`least partly a response to competitive market condi-
`tions and may be a sign of greater price competition in
`some segments of the pharmaceutical market.
`
`Competition Between Brand-Name
`and Generic Drugs
`
`The Hatch-Waxman Act eliminated the duplicative
`tests that had been required for a generic drug to ob-
`tain approval from the FDA. (That change applied
`only to nonantibiotic drugs, since antibiotics already
`had an abbreviated approval process.) Before 1984,
`manufacturers of generic drugs were required to inde-
`pendently prove the safety and efficacy of their prod-
`ucts. They were prohibited from using the unpub-
`lished test results of the original innovator drug, which
`were considered trade secrets of its manufacturer.1
`The Hatch-Waxman Act streamlined the process for
`approving generic drugs by requiring only that manu-
`facturers demonstrate "bioequivalence" to an already-
`approved innovator drug. (Bioequivalence means that
`the active ingredient is absorbed at the same rate and
`to the same extent for the generic drug as for the inno-
`vator drug.) The tests necessary to prove bioequiva-
`lence are much less costly than those required to prove
`safety and efficacy.
`
`By accelerating the approval process for a ge-
`neric drug and also allowing its producer to begin clin-
`ical tests before the patent on the innovator drug had
`expired, the Hatch-Waxman Act reduced the average
`delay between patent expiration and generic entry
`from more than three years to less than three months
`for top-selling drugs. Even more important, the act
`increased the proportion of brand-name drugs that
`face generic competition once their patents expire. In
`1983, only 35 percent of the top-selling drugs with
`expired patents (excluding antibiotics and drugs ap-
`proved before 1962) had generic versions available.
`Today, nearly all do.
`
`1.
`
`This study uses the terms "brand-name" and "innovator" inter-
`changeably.
`
`000012
`
`

`

`SUMMARY
`
` xiii
`
`After a drug's patent expires, generic copies
`quickly gain a large share of its market. CBO exam-
`ined 21 brand-name prescription drugs in its retail
`pharmacy data set that first saw generic competition
`between 1991 and 1993. Within their first full calen-
`dar year after patent expiration, those drugs lost an
`average of 44 percent of their market (as measured by
`the quantity of prescriptions sold through pharmacies)
`to generic drugs. And the generic versions cost an
`average of 25 percent less than the original brand-
`name drugs at retail prices. That rapid growth in ge-
`neric market share after patent expiration is a substan-
`tial change from the situation before the 1984 Hatch-
`Waxman Act. In 1983, for example, generic market
`share averaged just 13 percent for nonantibiotic drugs.
`
`Various studies have found that generic entry has
`little effect on the prices of brand-name drugs, which
`continue to increase faster than inflation. CBO's anal-
`ysis of the average prices that manufacturers charge
`for drugs distributed to retail pharmacies is consistent
`with that result. However, CBO's analysis of dis-
`counting shows that certain purchasers other than re-
`tail pharmacies receive steeper discounts on brand-
`name drugs once generic alternatives are available.
`Taken together, those results imply that the impact of
`generic entry on brand-name prices may vary consid-
`erably among different types of purchasers.
`
`Even if brand-name prices frequently do not re-
`spond to generic competition, such competition can
`effectively save money because price-sensitive buyers
`may switch to lower-priced generic drugs. CBO esti-
`mates that in 1994, purchasers saved a total of $8 bil-
`lion to $10 billion on prescriptions at retail pharmacies
`by substituting generic drugs for their brand-name
`counterparts. (That estimate assumes that all of the
`generic prescriptions dispensed in 1994 would have
`been filled with a higher-priced brand-name drug if a
`generic drug was not available.)
`
`Competition Among Generic Drugs
`
`By making generic entry easier and less costly, the
`Hatch-Waxman Act helped increase the number of
`generic manufacturers producing the same drug. As
`the number of manufacturers rises, the average pre-
`scription price of a generic drug falls. CBO's analysis
`
`shows that when one to 10 firms are manufacturing
`and distributing generic forms of a particular drug, the
`generic retail price of that drug averages about 60 per-
`cent of the brand-name price. When more than 10
`manufacturers have entered the market, the average
`generic prescription price falls to less than half of the
`brand-name price.
`
`The Effects of the Hatch-
`Waxman Act on the Returns
`from Innovation
`
`The patent provisions in the Hatch-Waxman Act have
`not completely protected drug companies' profits from
`the dramatic rise in generic competition since 1984.
`Manufacturers of brand-name drugs invest an average
`of about $200 million (in 1990 dollars) to bring a new
`drug to market, when the cost of capital and the cost
`of failures (investment in drugs that never make it to
`market) are included. Patent protection enables manu-
`facturers to earn an adequate return on that invest-
`ment. By itself, generic entry increases the rate at
`which sales erode after patent expiration, thus reduc-
`ing the returns from marketing a new drug. Two stud-
`ies have estimated that drugs introduced in the early
`1980s earned returns that exceeded their capitalized
`costs of development by $22 million to $36 million, on
`average. (Those figures represent the present dis-
`counted value in 1990 dollars.) CBO concludes that
`since 1984, the expected returns from marketing a new
`drug have declined by about 12 percent, or $27 million
`in 1990 dollars. That decline has probably not made
`drug development unprofitable on average, but it may
`have made some specific projects unprofitable.
`
`Changes to the Length of Patents
`for Brand-Name Drugs
`
`Under the Hatch-Waxman Act, drugs that contain a
`new chemical entity never before approved by the
`FDA can qualify for an extension of their patent term.
`Those extensions, granted after the drug is approved,
`equal half of the time the drug spent in clinical testing
`(usually a total of six to eight years) plus all of the
`
`000013
`
`

`

`xiv HOW INCREASED COMPETITION FROM GENERIC DRUGS HAS AFFECTED PRICES AND RETURNS
`
`July 1998
`
`time it spent having the FDA review its new drug ap-
`plication (usually about two years). Two key limita-
`tions apply. First, the extension cannot be longer than
`five years, and second, it cannot grant a total period of
`patent protection that exceeds 14 years after the drug
`is approved.
`
`June 8, 1995 (most of which have yet to be introduced
`on the market). However, many products that were
`already under patent by that date have benefited from
`the URAA, since their manufacturers can choose be-
`tween the 17-year and 20-year patent terms and still
`be eligible for a Hatch-Waxman extension.
`
`The 14-year limit is the main reason that Hatch-
`Waxman extensions now average about three years in
`length. Fifty-one drugs approved between 1992 and
`1995 received an extension. Excluding the eight drugs
`that were subject to a transitional two-year cap (which
`applied to products already in testing when the act
`took effect), half of the drugs had their extensions lim-
`ited by the 14-year cap.
`
`Not all of the new drugs that are approved obtain
`an extension. Out of 101 drugs approved between
`1992 and 1995, 38 did not apply for a Hatch-Waxman
`extension. Nineteen of those drugs had no patent to
`extend, and 15 others already had 14 years of patent
`protection left after obtaining FDA approval.
`
`the Hatch-
`Besides patent-term extensions,
`Waxman Act contains other provisions that postpone
`generic competition. One key provision is the require-
`ment that manufacturers wait five years after an inno-
`vator drug is approved before filing an application to
`sell a generic copy. That requirement benefits drugs
`that have no patent, or that have very little time left
`under patent, when they are approved. That exclusiv-
`ity provision, together with the patent-term extensions,
`postpones generic entry by an average of 2.8 years for
`all drugs approved that contain a new chemical entity.
`Another exclusivity provision delays generic entry for
`three years when a new application is approved that
`requires clinical tests (such as for a new dosage form
`or over-the-counter version of an already-approved
`drug).
`
`Ten years after the Hatch-Waxman Act, another
`piece of federal legislation—the Uruguay Round
`Agreements Act of 1994 (URAA)—further changed
`the patent terms of prescription drugs. That act al-
`tered the length of a patent for all types of inventions
`to 20 years from the date the application is filed rather
`than 17 years from the date the patent is granted.
`That change should have little effect on the average
`amount of time between market introduction and pat-
`ent expiration for brand-name drugs patented after
`
`The Change in Returns from Innovation
`
`As noted earlier, the Hatch-Waxman Act greatly in-
`creased the probability that a generic copy would be-
`come available once the patent on a brand-name drug
`expired. It also contributed to a dramatic rise in ge-
`neric market share. In addition, the act reduced the
`delay between patent expiration and generic entry, but
`that acceleration was roughly offset by patent-term
`extensions and exclusivity provisions that postpone
`generic entry.
`
`CBO estimates that the increase in the size of the
`generic market since 1984—part of which is attribut-
`able to the act—has reduced the expected level of re-
`turns from marketing a brand-name drug by an aver-
`age of $27 million in 1990 dollars. That amount is
`roughly 12 percent of the total discounted returns from
`selling a brand-name drug, which previous studies
`have estimated at $210 million to $230 million in 1990
`dollars for drugs introduced in the early 1980s.
`(Those figures represent the present discounted value
`of the total stream of profits from those drugs dis-
`counted to the date of market introduction, deducting
`manufacturing costs but not R&D costs.) That 12
`percent decline does not change significantly under
`reasonable variations in CBO's underlying assump-
`tions.
`
`Other factors besides the Hatch-Waxman Act
`have played a role in increasing the frequency of ge-
`neric competition and the average size of generic mar-
`ket share. For example, changes in state laws have
`given pharmacists more leeway to substitute generic
`drugs for brand-name ones. And for reasons of cost,
`many purchasers have put increasing emphasis on ge-
`neric substitution.
`
`Total returns from selling a brand-name pre-
`scription drug vary significantly among different
`drugs. As noted above, the average cost of developing
`
`000014
`
`

`

`SUMMARY
`
` xv
`
`such drugs, including failures, is around $200 million
`in 1990 dollars. But on average only three in 10 drugs
`earn that much in discounted returns (after deducting
`manufacturing, advertising, distribution, and other
`non-R&D-related costs). For most drugs, the returns
`from marketing do not exceed the average capitalized
`costs of development. As a result, for a company's
`average returns to exceed its average development
`costs, the company must discover and market a highly
`profitable drug from time to time.
`
`For all drugs, on average, the increase in generic
`sales since 1984 has probably not reduced expected
`returns below the average capitalized costs of R&D.
`On the margin, however, it is possible that a few drugs
`that were barely profitable to develop before may no
`longer be so now.
`
`CBO's calculation of the change