Pay-for-Delay:
`How Drug Company Pay-Offs
`Cost Consumers Billions
`
`An FTC Staff Study
`January 2010
`
`Federal Trade Commission | ftc.gov
`
`CFAD VI 1051 - 0001
`CFAD VI v. CELGENE
`IPR2015-01103
`
`

`
`CFAD VI 1051 - 0002
`
`CFAD VI 1051 - 0002
`
`

`
`Summary
`
`Brand-name pharmaceutical companies can delay generic competition that lowers
`prices by agreeing to pay a generic competitor to hold its competing product off the
`market for a certain period of time. These so-called “pay-for-delay” agreements have
`arisen as part of patent litigation settlement agreements between brand-name and
`generic pharmaceutical companies.
`
`“Pay-for-delay” agreements are “win-win” for the companies: brand-name
`pharmaceutical prices stay high, and the brand and generic share the benefits of the
`brand’s monopoly profits. Consumers lose, however: they miss out on generic prices
`that can be as much as 90 percent less than brand prices. For example, brand-name
`medication that costs $300 per month might be sold as a generic for as little as $30 per
`month.
`
`The Federal Trade Commission’s (FTC) investigations and enforcement actions against
`pay-for-delay agreements deterred their use from April 1999 through 2004.1 In 2003,
`an appellate court held that such agreements were automatically (or per se) illegal.2
`
`Since 2005, however, a few appellate courts have misapplied the antitrust law to uphold
`these agreements.3 Following those court decisions, patent settlements that combine
`restrictions on generic entry with compensation from the brand to the generic have re-
`emerged.
`
`Agreements with Delay and Compensation4
`
`19
`
`16
`
`14
`
`14
`
`3
`
`0
`
`2004
`
`2005
`
`2006
`
`2007
`
`2008
`
`2009
`
`5
`
`1
`
` ●
`
` ●
`
` ●
`
` ●
`
`20
`
`18
`
`16
`
`14
`
`12
`
`10
`
`02468
`
`CFAD VI 1051 - 0003
`
`

`
` ●
`
`Agreements with compensation from the brand to the generic on average prohibit
`generic entry for nearly 17 months longer than agreements without payments, where
`the average is calculated using a weighted average based on sales of the drugs.6 Most of
`these agreements are still in effect. They currently protect at least $20 billion in sales of
`brand-name pharmaceuticals from generic competition.7
`
` ●
`
`Pay-for-delay agreements are estimated to cost American consumers $3.5 billion per
`year – $35 billion over the next 10 years.8
`
`
`Recommendation
`
`Pay-for-delay agreements have significantly postponed substantial consumer savings from
`lower generic drug prices. The Commission has recommended that Congress should pass
`legislation to protect consumers from such anticompetitive agreements.
`
`2
`
`CFAD VI 1051 - 0004
`
`

`
`Background
`
`Pay-for-delay agreements appear in some settlements of patent litigation between brand-name
`and generic pharmaceutical companies. That patent litigation usually takes place within the
`framework for generic entry established by the Hatch-Waxman Act.9 Under that Act, a generic
`competitor may seek entry prior to expiration of the patents on a brand-name drug. Generic
`drug entry before patent expiration can save consumers billions of dollars. Generics have an
`incentive to challenge brand patents because the first generic to file its application can obtain
`180 days of marketing exclusivity during which it is the only generic on the market. To seek
`FDA approval for entry before patent expiration, a generic must declare that its product does
`not infringe the relevant patents or that the relevant patents are invalid.
`
`Typically, brand-name pharmaceutical companies challenge the generic’s declaration, and
`litigation ensues between the brand-name and generic pharmaceutical manufacturers to
`determine whether the relevant patents are valid and infringed. For the brand to prevail and
`block entry, it must successfully defend the validity of its patents and demonstrate that the
`generic’s product would infringe those patents. In 2002, the FTC issued a study showing
`that generics prevailed in 73% of the patent litigation ultimately resolved by a court decision
`between 1992 and June 2002.10
`
`Given the costs and potential uncertainty of patent litigation, brand-name and generic
`pharmaceutical companies sometimes settle their patent litigation before a final court
`decision. For example, the parties may agree that the generic can enter at some time before
`the patent’s expiration date, but not as soon as the generic seeks through its litigation. Absent
`compensation to the generic for the delay in its entry, such settlement agreements are unlikely
`to raise antitrust issues.
`
`The FTC’s 2002 study determined, however, that some brand-name and generic
`pharmaceutical companies had settled their patent litigation through agreements that
`compensated generics for substantial delays in generic entry. The FTC recommended that
`Congress pass legislation to require pharmaceutical companies to file certain agreements
`with the FTC. The intent of the legislation was “to put an end to this exploitation of the
`provision in Hatch-Waxman that grants a short-term protection from competition to the first
`manufacturer to bring a generic version of a brand name drug to market.”11
`
`Congress acted on the FTC’s recommendation. Under the Medicare Prescription Drug,
`Improvement, and Modernization Act of 2003 (the “MMA”), pharmaceutical companies must
`file certain agreements with the FTC and the Department of Justice within ten days of their
`execution.12
`
`3
`
`CFAD VI 1051 - 0005
`
`

`
`Findings from Pharmaceutical
`Agreement Filings from FY2004
`through FY2009
`
` h
`
`How Many Final Agreements Have Involved Compensation from the Brand to the
`Generic Combined with Restrictions on Generic Entry?
`
`From FY2004-FY2009, 66 final agreements involved some form of compensation
`from the brand to the generic combined with a delay in generic entry.
`
` h
`
`Can Pharmaceutical Companies Settle Patent Litigation without Pay-for-Delay
`Agreements?
`
`Yes. From FY2004-FY2009, pharmaceutical companies filed a total of 218 final
`settlement agreements involving brand and generic companies. Seventy percent of
`those patent settlements – 152 – did not involve compensation from the brand
`to the generic combined with a delay in generic entry. This large number of
`settlements not involving compensation from the brand to the generic undermines
`brand and generic firms’ arguments that compensation is the only way to settle
`patent litigation. In fact, there are a variety of ways to settle litigation that do not
`involve these payments.
`
` h
`
`Do Agreements with Compensation from the Brand to the Generic Postpone
`Generic Entry Significantly Longer than Other Patent Settlement Agreements?
`
`Yes. Staff analysis of patent settlements restricting generic entry finds that
`agreements with compensation on average prohibit generic entry for nearly 17
`months longer than agreements without payments, where the average is calculated
`using a weighted average based on sales of the drugs.13 This difference in time to
`entry is very unlikely to be caused by random variation in the agreements. In fact,
`there is less than a 1% chance that this large a difference in average time to entry
`would be observed if the amount of delay from the two types of agreements were
`drawn from the same statistical distribution.
`
`A hypothetical consumer paying $300 per month for a brand-name drug, instead of
`a generic price as low as $30 per month, could pay as much as $270 per month more
`for prescription drugs. Over a 17-month period, this could total additional expenses
`of $4,590 resulting from the extra delay that occurs, on average and weighted for
`sales.
`
`4
`
`CFAD VI 1051 - 0006
`
`

`
` h
`
`Is the First Generic to Seek Entry Prior to Patent Expiration Involved in Most of
`the Potential Pay-for-Delay Settlements?
`
`Yes. Out of the 66 agreements that combined compensation from the brand to the
`generic with deferred generic entry, 51 agreements (77%) were between the brand
`pharmaceutical company and the generic company that was the first to seek entry
`prior to patent expiration for the relevant brand-name drug.
`
`Settlements with first-filer generics can prevent all generic entry. Those agreements
`place a “cork in the bottle” that typically ensures the brand-name drug’s lock on the
`market. This cork-in-the-bottle effect occurs because every subsequent generic
`entrant has to wait until the first generic has been marketed for 180 days. 14
`
` h
`
`Do All Pay-for-Delay Agreements Involve Dollar Payments from the Brand to the
`Generic?
`
`No. Brand-name pharmaceutical companies have found a wide variety of techniques
`through which to compensate generic companies for delaying their entry.
`
`Recently, brand-name pharmaceutical companies have sometimes compensated
`generics by agreeing not to compete through a so-called “authorized generic.” Under
`the Hatch-Waxman Act, the generic that is first to file its approval application can
`be entitled to market its generic product for 180 days with no competition from
`other generics.15 This rule, however, does not protect the first-filer generic from
`competition from an “authorized generic” or “AG” during those 180 days.
`
`AGs are brand-name pharmaceutical products marketed as generics. AG
`competition can substantially reduce the revenues a first-filer generic earns during its
`180 days of marketing exclusivity.16
`
`About 25% of patent settlement agreements from FY2004-FY2008 that were with
`first-filer generics involved an explicit agreement by the brand not to launch an
`AG to compete against the first filer, combined with an agreement by the first-filer
`generic to defer entry past the date of the agreement.17 In effect, by agreeing not to
`launch an AG, the brand agrees not to subtract from the generic’s profits during the
`180-day period.
`
` h
`
`Has the FTC Given Up Litigating Pay-for-Delay Cases under the Antitrust Laws?
`
`No. The FTC has multiple investigations underway and currently is litigating two
`cases in the trial courts.18 Over the past nine years, the FTC has invested substantial
`resources in investigating and, when necessary, litigating cases involving patent
`settlements in which brand-name pharmaceutical companies allegedly paid generic
`companies to stay off the market, thus depriving consumers of millions of dollars in
`cost savings that would otherwise have been available.19
`
`5
`
`CFAD VI 1051 - 0007
`
`

`
`Given the magnitude of consumer harm from pay-for-delay settlements – an
`estimated $35 billion over the next ten years – a legislative solution offers the
`quickest and clearest way to deter these agreements and obtain the benefits of
`generic competition for consumers.
`
`
`
`6
`
`CFAD VI 1051 - 0008
`
`

`
`Study Methodology
`
`This study was prepared by staff from the FTC’s Bureau of Competition, Bureau of
`Economics, and Office of Policy Planning.
`
`This study is based on patent settlement agreements filed with the FTC between January 1,
`2004 and September 30, 2009 pursuant to the Medicare Prescription Drug, Improvement, and
`Modernization Act of 2003, Pub. L. No. 108-173, 117 Stat. 2066, codified in relevant part at
`42 U.S.C. § 1395w-101 note (section 110), 21 U.S.C. § 355 note (sections 1111-1118), 21
`U.S.C. § 355(j)(5) (section 1102).
`
`Staff identified agreements in which restrictions on generic entry were combined with
`compensation from the brand to the generic. The FTC has challenged some of these
`agreements as violating the antitrust laws, but the agency lacks sufficient resources to
`investigate and litigate the legality of all of these agreements.
`
`How staff calculated the additional delay in generic entry
`associated with agreements that involved compensation
`from the brand to the generic.
`
`To calculate how long (on average and weighted for sales) generic entry was delayed as
`a result of compensation from brand-name pharmaceutical companies to generic drug
`companies, staff compared agreements with and without compensation to the generic in
`terms of the sales-weighted average time between the date of the agreement’s execution and
`the date of generic entry.
`
`To avoid double counting multiple settlements on the same drug, only the settlement that
`establishes the earliest date for generic entry was used in this calculation.
`
`To better reflect the amount of consumer savings held up by the delay, staff used weighted
`averages of sales.
`
`This calculation established that, on average and weighted for sales, agreements with
`compensation from the brand to the generic delayed generic entry for nearly 17 months
`longer than agreements without compensation. Staff determined that the 17 month
`difference in time until generic entry was statistically significant at the 99% confidence level.
`Thus, this difference in time to entry is very unlikely to be caused by random variation in
`the agreements. In fact, there is less than a 1% chance that this large a difference in average
`time to entry would be observed if the amount of delay from the two types of agreements
`were drawn from the same statistical distribution.
`
`7
`
`CFAD VI 1051 - 0009
`
`

`
`How staff calculated the estimate of $3.5 billion annually
`that consumers lose due to pay-for-delay agreements.
`
`The calculation below is a method of estimating the likely harm to consumers from the loss
`of competition when patent settlements delay generic entry.20 The analysis estimates that
`under relatively conservative assumptions, the annual savings to purchasers of drugs that
`would result from eliminating “reverse-payment” settlements would be approximately
`$3.5 billion.
`
`This calculation requires four factors:
`
`1.
`
`the consumer savings that result from generic competition in any given month,
`
`2.
`
`the likelihood that a generic manufacturer and brand-name manufacturer will
`reach a settlement that delays entry in return for compensation,
`
`3.
`
`the length of entry delay resulting from such settlement, and
`
`4.
`
`the combined sales volume of drugs for which settlements are likely.
`
`(1) Consumer savings from generic competition.
`
`When generic entry occurs, purchasers immediately begin to benefit from the savings
`associated with lower generic drug prices. Following an initial entry period, the generic
`market matures and consumers receive the full savings from generic competition. Thus,
`any delay in entry results in a longer period of purchases at the full brand price and
`correspondingly fewer purchases at the mature competitive prices.21 This means that
`the costs to consumers (or what they would have saved but for the entry delay) are
`equal to the monthly savings from the mature generic market multiplied by the number
`of months of delay.
`
`Publicly available information about recent generic launches suggests that a generic
`market typically matures about one year after the first entrant comes on the market.
`The generic penetration rate at that point is about 90% on average, i.e. pharmacists fill
`90 of every 100 prescriptions for the molecule with an AB-rated (or bioequivalent)
`generic. Recent information also shows that in a mature generic market, generic prices
`are, on average, 85% lower than the pre-entry branded drug price.22
`
`Using the above figures and assumptions, the average consumer savings from a mature
`generic market relative to pre-generic levels are approximately 77% (85% savings
`multiplied by 90% of market demand). If purchasers discount future savings at the
`same rate as they expect drug prices and quantities to increase, then all future savings
`can be expressed in terms of today’s dollars without complicated net present value
`calculations. Thus, the costs of delay are the average discount (77%) times the length
`of the delay times the pre-generic entry revenues of the branded drugs that will reach a
`settlement with delay.
`
`8
`
`CFAD VI 1051 - 0010
`
`

`
`(2) Likelihood of Settlements with Payment to Delay, and the Length of Delay
`
`It is more difficult accurately to estimate how much delay is likely to result from
`settlements that have not yet been reached, especially because future legislative or
`judicial actions could alter the types of settlements that are likely. Therefore, the
`calculation assumes that recent settlements provide the best information about what
`may happen in the future. Data on settlements reported to the FTC from FY2004 to
`FY2008 show that of all patent settlements resulting from a Paragraph IV (invalidity
`or non-infringement) challenge, approximately 24% included both restrictions on
`timing of generic entry and a payment to the generic firm.
`
`The additional length of the delay that is attributed to the payments in these
`settlements can be calculated by taking the universe of Paragraph IV settlements that
`have restrictions on entry, then comparing the average number of months between
`the execution of the agreement and the date of generic entry in agreements with and
`without payments to the generic entrant. Agreements with payments on average allow
`entry nearly 17 months (1.42 years) later than agreements without payments.
`
`This does not mean that we are assuming that all settlements with payments would
`“become” settlements without payments if the former were banned. Some would;
`others might involve litigation of the patent. But since settlements without payments
`will tend to reflect patent strength, they can provide a benchmark for the consumer
`impact of either alternative.
`
`(3) Sales Volume of Drugs for which Settlements are Likely
`
`Staff relied on recent history as a guide to the settlements likely to be seen in the future.
`The analysis starts with the FDA’s list of all drugs that have received a Paragraph IV
`filing.23 It then uses information from the FDA’s Orange Book, IMS NPA retail sales
`data, and the settlement filings to determine whether there had been a generic version
`of a challenged drug launched before 2004. If a generic had entered, it was removed
`from the list of drugs that could have settled between FY2004 and FY2008. The
`analysis next uses the IMS data to determine the total dollar sales associated with
`those drugs remaining in the sample for each year. It adjusts these annual totals by
`removing drugs that reached a settlement or experienced generic entry due to a non-
`settlement event such as a court victory or patent expiration.
`
`By the end of FY2008, the above method estimates that there were $90 billion of
`branded drug sales still facing a Paragraph IV challenge. Since the IMS data used
`does not cover all purchasing channels and excludes injectable drugs, $90 billion is a
`conservative estimate of the total branded dollars affected by possible settlements.
`
`The next step is to look at the number of settlements per year as a percentage of all
`Paragraph IV-challenged drugs that could possibly settle. Over the FY2004 to FY2008
`time period, the percentage of drugs that settled per year (not including injectables)
`increased from 7 percent to 18 percent, with most of the increase following the
`Eleventh Circuit’s Schering decision. Since this post-Schering era is probably a better
`
`9
`
`CFAD VI 1051 - 0011
`
`

`
`reflection of likely future settlement patterns, it seems appropriate and conservative to
`use the 15 percent per year average from this period in the estimate calculations.
`
`Multiplying $90 billion by 15 percent yields $13.5 billion in drug purchases that are
`predicted to be affected by settlements each year. Multiplying this $13.5 billion total by
`24 percent (an assumption based on the percentage of past settlements with payment
`and delayed entry), leads to a prediction of $3.2 billion in drug sales that will be
`affected by reverse payment settlements in a given year.
`
`(4) Final Estimate Calculation
`
`The final steps in calculating the savings to be gained by eliminating pay-for-delay
`settlements are to factor in the discount consumers would receive from matured generic
`entry and the length of delay. From the 77 percent savings and 1.42 year delay figures
`above, the calculation is therefore:
`
` 77% savings
`x $3.2 billion (15% per year settling)
`x 1.42 years (median delay)
`
`$3.5 billion of annual purchaser savings
`
`In sum, the calculation yields a conservative estimate of $3.5 billion per year of
`potential savings from eliminating pay-for-delay settlements.
`
`Results with Varied Assumptions
`
`The $3.5 billion figure represents staff ’s best estimate of the effect based on what staff
`believes to be the most reasonable assumptions. Nonetheless, this estimate is sensitive
`to changes in the assumptions.24 Reasonable estimates about the length of delay and
`the sales of drugs likely to be affected by the legislation can vary. The calculations
`below present high and low estimates of savings derived from the data ranges.
`
` 77% savings
`x $3.9 billion (18% per year settling)
`x 2.5 years (high of interquartile distribution of delay)
`
`$7.5 billion of annual purchaser savings
`
` 77% savings
`x $1.5 billion (7% per year settling)
`x 0.5 years (low of interquartile distribution of delay)
`
`$0.6 billion of annual purchaser savings
`
`10
`
`CFAD VI 1051 - 0012
`
`

`
`Endnotes
`
`1 See Generic Drug Entry Prior to Patent Expiration: An FTC Study, Exec. Summary at viii ( July
`2002), available at http://www.ftc.gov/os/2002/07/genericdrugstudy.pdf. This study covered
`the period through June 2002. The FTC began receiving patent settlement agreements in January
`2004 pursuant to the Medicare Prescription Drug, Improvement, and Modernization Act of 2003.
`Although there is a gap between July 2002 and December 2003, we are unaware that brand and
`generic firms entered into any pay-for-delay settlement agreements during this time period.
`
`2 See In re Cardizem CD Antitrust Litigation, 332 F.3d 896 (6th Cir. 2003).
`
`3 See Schering-Plough Corp. v. Fed. Trade Comm’n, 402 F.3d 1056 (11th Cir. 2005); see also In re
`Tamoxifen Citrate Antitrust Litigation, 466 F.3d 187 (2d Cir. 2006); In re Ciprofloxacin Hydrochloride
`Antitrust Litigation, 544 F.3d 1323 (Fed. Cir. 2008). But see Brief For the United States In
`Response To the Court’s Invitation, In re Ciprofloxacin Hydrochloride Antitrust Litigation,
`No. 05-cv-2851(L) (2d Cir. July 6, 2009), available at
`http://www.justice.gov/atr/cases/f247700/247708.htm.
`
`4 These agreements were filed with the FTC pursuant to the Medicare Prescription Drug,
`Improvement, and Modernization Act of 2003, Pub. L. No. 108-173, 117 Stat. 2066 (codified
`in relevant part 42 U.S.C. § 1395w-101 (2009) note (section 110), 21 U.S.C. § 355 (2009) note
`(sections 1111-1118), 21 U.S.C. § 355(j)(5) (2009) (section 1102)). All of these agreements
`involved patent settlements that combined restrictions on generic entry with compensation from
`the brand to the generic. The FTC has challenged some of these agreements as violating the
`antitrust laws, but the agency lacks sufficient resources to investigate and litigate the legality of all of
`the agreements represented in this chart.
`
`5 These years represent fiscal years.
`
`6 The 17-month delay attributed to payments was calculated by comparing the sales-weighted
`average time between the date of the agreement’s execution and the date of generic entry for
`agreements with and without compensation to the generic.
`
`7 This dollar amount represents the prior-year total sales of the brand-name pharmaceuticals that
`are currently covered by agreements with delay and compensation and thus indicates the order of
`magnitude of brand-name pharmaceutical sales for which generic competition (with lower prices)
`has likely been delayed.
`
`8 See Jon Leibowitz, Chairman, Fed. Trade Comm’n, “Pay-for-Delay” Settlements in the Pharmaceutical
`Industry: How Congress Can Stop Anticompetitive Conduct, Protect Consumers’ Wallets, and Help Pay
`for Health Care Reform (The $35 Billion Solution) at 8 ( June 23, 2009), available at
`http://www.ftc.gov/speeches/leibowitz/090623payfordelayspeech.pdf.
`
`9 The Drug Price Competition and Patent Term Restoration Act of 1984, Pub. L. No. 98-417, 98
`Stat. 1585 (1984) (codified as amended 21 U.S.C. § 355 (2009)) governs how generics may enter
`the marketplace to compete with brand-name pharmaceuticals.
`
`10 See supra note 1.
`
`11 See S. Rep. No. 107-147 at 4 (2002).
`
`12 See Pharmaceutical Agreement Filing Requirements, available at
`http://www.ftc.gov/os/2004/01/04106pharmrules.pdf.
`
`11
`
`CFAD VI 1051 - 0013
`
`

`
`13 The delay attributed to payments was calculated by comparing the sales-weighted average time
`between the date of the agreement’s execution and the date of generic entry for agreements with
`and without compensation to the generic. The distribution of annual sales figures for drugs covered
`by these pay-for-delay agreements is not discernibly different from the distribution of annual sales
`figures for drugs covered by agreements that restrict generic entry with no payment to the generic.
`
`14 Later-filing generics cannot enter the market until they win their own patent litigation at the court
`of appeals level and the first filer generic either markets its product for 180 days or forfeits its right
`to do so. 21 U.S.C. § 355(j)(5)(D) (2009) (forfeiture provisions).
`
`15 There may be more than one “first-filer” if more than one generic firm files its application on the
`same, “first” day.
`
`16 Authorized Generics: An Interim Report, Fed. Trade Comm’n at 3 ( June 2009); available at
`http://www.ftc.gov/os/2009/06/P062105authorizedgenericsreport.pdf.
`
`17 Id.
`
`18 See Fed. Trade Comm’n v. Cephalon, No. 08-cv-2141-RBS (E.D. Pa. May 8, 2008) (transfer order);
`Fed. Trade Comm’n v. Watson, No. 09-cv-00598 (N.D. GA Feb. 9, 2009) (transfer order).
`
`19 See In re Hoechst Marion Roussel Inc., Carderm Capital L.P. and Andrx Corp.; 131 F.T.C. 927 (2001)
`(consent order); In re Abbott Laboratories and Geneva Pharmaceuticals, Inc., C-3945, C-3946
`(consent orders issued May 22, 2000); In re Schering-Plough Corp., et al, D. 9297, Initial Decision
`issued June 27, 2003; rev’d by Commission Decision and Order, December 8, 2003(136. F.T.C. 956
`(2003)); rev’d 402 F.3d 1056 (11th Cir. 2005); In re Bristol-Myers Squibb, 135 F.T.C. 444 (2003)
`(consent order); Fed. Trade Comm’n v. Cephalon, No. 08-cv-2141-RBS (E.D. Pa. May 8, 2008)
`(transfer order); Fed. Trade Comm’n v. Watson, No. 09-cv-00598 (N.D. GA Feb. 9 2009) (transfer
`order).
`
`20 This calculation first appeared as an Appendix to Chairman Leibowitz’s speech. See supra note 8.
`
`21 If one assumes some future end-point in the drug’s life on the market, delayed entry means that, by
`that end-point, consumers will have had less time buying in the mature competitive market.
`
`22 The calculation assumes that the total demand for the drug/molecule (market size in unit sales)
`remains the same after generic entry occurs. It also assumes that the brand’s price stays the same
`after generic entry occurs. Data show that branded prices often rise following generic entry, but
`there are also instances when brand price declines. Assuming the price stays the same simplifies the
`analysis.
`
`23 This is based on a version downloaded from the FDA’s website on May 19, 2009.
`
`24 In addition, a possible effect in the other direction could arise if a future legislative or judicial action
`made pay-for-delay agreements illegal. To the extent that such an action would reduce generic
`firms’ incentives to file Paragraph IV challenges, it could reduce the sales volume of drugs facing
`such challenges. Any such deterrent effect would likely be very low, however. As noted above, only
`24% of all cases settled with both payment and delay, so presumably generic drug firms do not
`assume that they will be able to settle their patent litigation through compensation for deferred
`generic entry. Moreover, a generic would still have a strong incentive to challenge a weak patent in a
`large market.
`
`12
`
`CFAD VI 1051 - 0014
`
`

`
`CFAD VI 1051 - 0015
`
`CFAD VI 1051 - 0015
`
`

`
`Federal Trade Commission | ftc.gov
`
`CFAD VI 1051 - 0016