Patent Pools, Litigation and Innovation Jay Pil Choi (UNSW, Michigan - - PowerPoint PPT Presentation

Patent Pools, Litigation and Innovation

Jay Pil Choi (UNSW, Michigan State) Heiko Gerlach (U of Queensland) 2nd ATE Symposium at UNSW Sydney December 15, 2014

Patent Pools, Litigation and Innovation 1

Patent Pools and Antitrust

patent pools=agreements of two or more IP owners to jointly license to third parties single license contract for a bundle of patents with multiple owners many patent pools in different industries until end of WWII Supreme Court decision in 1945 impeded further pool formation 1995 DOJ/FTC Guidelines for the Licensing and Acquisition of Intellectual Property

Patent Pools, Litigation and Innovation

• 1. Introduction

2

competitive effects of patent pools depend on relationship among constituent patents (e.g. Lerner & Tirole 2004) complementary patents: pro-competitive substitute patents: anti-competitive 2004/2007/2014: European Commission also reviews patent pools favourably, introduces safeguards

Golden ¡Rice ¡ ¡Pool ¡(2000) ¡ Gene0c ¡Tes0ng ¡ ¡(2010) ¡ HIV ¡drugs ¡(2010) ¡ Wireless ¡Lan ¡ ¡ (2008) ¡ One ¡Blue ¡ (2009) ¡

Patent Pools, Litigation and Innovation

• 1. Introduction

3

Patent Pools and Probabilistic Patents

“probabilistic patents” (Lemley & Shapiro, 2004): half of litigated patents found to be invalid patents right to “try to” exclude licensing in the shadow of litigation: royalty rates reflect the strength

• f patents

pools with probabilistic, complementary patents have two effects

1 litigating individual patents not profitable: patent pools are

stronger litigation entity than individual firms

2 prevent individual patent holders from engaging in Bertrand type

competition to avoid litigation against their own patent

Patent Pools, Litigation and Innovation

• 1. Introduction

4

patent pools can be used as a mechanism to discourage patent litigation and charge higher licensing fees this in turn reduces future downstream innovation incentives if patents sufficiently weak, complementary patent pools can be anti-competitive formalize the court’s idea in Duplan vs Deering Milliken:

“the ... patents in suit were known ... to be weak and, ..., they [the parties] were confident that these patents could be invalidated. The main purpose of the patent pool in the case was to protect the parties from challenges to the validity of their patents in order to gain the power to fix and maintain prices in the form of royalties which they... exercised thereafter.”

Patent Pools, Litigation and Innovation

• 1. Introduction

5

Theoretical Literature on Patent Pools

Patent Pools and Static Welfare

◮ Shapiro (2003) ◮ Lerner and Tirole (2004) ◮ Lerner, Strojwas and Tirole (2007)

Patent Pools and Innovation Incentives with Ironclad Patents

◮ Llanes and Trento (2013) ◮ Dequiedt and Veraevel (2006) ◮ Kwon et al. (2008)

Patent Pools and Probabilistic Patents

◮ Choi (2010) ◮ Gilbert and Katz (2006) Patent Pools, Litigation and Innovation

• 1. Introduction

6

Empirical Literature

Lampe and Moser (2010, 2013)

◮ Sewing Machine Combination (1856-1877), the first patent pool in

U.S. history

◮ pool discouraged patenting and innovation, more likely to sue outsiders ◮ attributed to fact that patent pools create more formidable entities in

court and thus increase the threat of litigation for outside firms

Joshi and Nerkar (2011)

◮ global optical disc industry: MPEG-2, DVD3C, DVD6C patent pools ◮ patent pool inhibits, rather than enhances, innovation by firms Patent Pools, Litigation and Innovation

• 1. Introduction

7

Lampe and Moser (2010, 2013): “19th century data suggest that the sewing machine patent pool discouraged invention by outside firms..”

Patent Pools, Litigation and Innovation

• 1. Introduction

6

Lampe and Moser (2010, 2013): “19th century data suggest that the sewing machine patent pool discouraged invention by outside firms..”

Patent Pools, Litigation and Innovation

• 1. Introduction

6

• 2. Complementary Patent Pools and Future Development

Incentives

two complementary patents, A and B, which are owned by two separate firms patents essential for development new technology by third firm C potential innovation can create a total value of v implementation cost of c ≥ 0 distributed according to cdf G(c) reversed hazard rate of G(.), r(.) = g()/G(), monotonically decreasing in its argument how does patent pool affect future incentives to develop new innovations? consider ex ante licensing: commitment mechanism not to hold up firm C’s investment incentives

Patent Pools, Litigation and Innovation

• 2. Complementary Patent Pools and Future Development Incentives 7

timing of licensing game in the absence of litigation:

1

firms decide whether or not to form a patent pool

2

set licensing fees

⋆ If they do not form a patent pool, they set the licensing fees (fA, fB)

independently

⋆ If they form a patent pool, they offer package licensing at F 3

firm C decides whether to buy licenses and develop innovation or not

firm C develops the new innovation if development cost satisfies c < v − fA − fB probability new innovation to be developed is given by G(v − fA − fB )

Patent Pools, Litigation and Innovation

• 2. Complementary Patent Pools and Future Development Incentives 8

Licensing with Independent Firms firm A solves the following problem given firm B’s royalty rate fB max

fA

fAG(v − fA − fB ) first order condition for firm A’s optimal royalty rate fA is given by G(v − fA − fB ) − fAg(v − fA − fB ) = 0 (1) which implicitly defines reaction function fi = Θ(fj) fA = G(v − fA − fB ) g(v − fA − fB ) . MRHR condition ensures unique Nash equilibrium at intersection of reaction functions such that f ∗

A = f ∗ B = f ∗ and

F ∗ = f ∗

A + f ∗ B

Patent Pools, Litigation and Innovation

• 2. Complementary Patent Pools and Future Development Incentives 9

Licensing with Patent Pool the optimal package license is derived by solving max

F

FG(v − F)

• ptimal ex ante licensing fee F ∗∗ satisfies the following first order

condition: G(v − F ∗∗) − F ∗∗g(v − F ∗∗) = 0, (2) which can be rewritten as F ∗∗ = G(v − F ∗∗) g(v − F ∗∗) . patentees do not internalize increase in other patentee’s profits when demand for package is increased by price reduction patent pool decreases overall royalty rates and simultaneously increases both patentees’ profits and induces more future innovations

Patent Pools, Litigation and Innovation

• 2. Complementary Patent Pools and Future Development Incentives 10

Proposition

When firms form a patent pool, total licensing fees are lower and there are more subsequent innovations, F ∗ = f ∗

A + f ∗ B > F ∗∗

Social and private incentives to form a patent pool are perfectly aligned. variation of well-known result that dates back to Cournot’s (1927) analysis of the complementary monopoly problem patent pool increases social welfare and is pro-competitive in the presence of multiple blocking patents dynamic efficiency interpretation for allowing patent pools of complementary innovations

Patent Pools, Litigation and Innovation

• 2. Complementary Patent Pools and Future Development Incentives 11

• 3. Probabilistic Patent Rights and Litigation with Patent

Pools

probabilistic patents: pA = α ≥ 0 and pB = β ≤ 1 where α ≤ β probabilities that court will uphold validity of patents if challenged symmetric information structure: α and β common knowledge two additional stages after firm C’s decision whether to buy licenses

◮ if firm C does not buy, upstream firm can decide whether to sue for

infringement; in this case, firm C will contest the validity of the patent

◮ court decides on validity of patent and litigation outcome revealed

fee setting has two potential margins: demand or litigation margin we first consider game where only litigation margin is binding, then we consider game where both margins may be binding

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

12

timing of the game with litigation and demand margin

1 firms A and B decide whether or not to form a patent pool 2 firms A, B or pool set licensing fees 3 firm C draws development cost c from G(.) and decides whether to engage in development

⋆ if firm C does not develop, the game ends

4 firm C decides for each technology whether to buy license or not 5 upstream firms (or pool) decide(s) whether to sue firm C for infringement; if sued, firm C contests the validity of the patent in court 6 litigation outcomes revealed; if patent has been challenged and upheld, its holder proposes new license fee for firm C 7 if firm C has license for all non-invalidated patents, it receives v.

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

13

timing of the game with litigation margin only

1 firms A and B decide whether or not to form a patent pool 2 firms A, B or pool set licensing fees 3L firm C develops the subsequent innovation 4 firm C decides for each technology whether to buy license or not 5 upstream firms (or pool) decide(s) whether to sue firm C for infringement; if sued, firm C contests the validity of the patent in court 6 litigation outcomes revealed; if patent has been challenged and upheld, its holder proposes new license fee for firm C 7 if firm C has license for all non-invalidated patents, it receives v.

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

13

both parties incur litigation cost per patent L ≥ 0 independent of patent validity parameters if patent invalidated: downstream firm can use technology at no cost if patent validated: patent holder sets new license fee and firm C decides whether to purchase license or not

• ther existing licensing contract remains unaffected by court decision

if firm C does not acquire license of validated or unchallenged patent, it is unable to produce and receives zero profit parameter values such that threat of infringement litigation is credible L v ≤ α 2α + 1(1 − α)(1 − β). (A)

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

14

• 1. Licensing and Litigation with Independent Firms

firms A and B propose license fees, fA and fB

• ption 1 for firm C: don’t buy and infringe on both patents; validity
• f both patents tested; both need to be invalidated for C to receive

positive payoff VAB = (1 − α)(1 − β)v − 2L

• ption 2: infringe on patent i and buy license for patent j yields

Vi = (1 − α)(v − fj) − L

• ption 3: buy both licenses yields

V0 = v − fA − fB if firm C is indifferent between litigating against A or B it randomises with equal probabilities

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

15

Optimal Litigation and Licensing for firm C

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ (4) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡ Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

16

Nash equilibrium in License Fees firms’ best response is to set limit licensing fees to avoid litigation

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

17

Nash equilibrium in License Fees firms’ best response is to set limit licensing fees to avoid litigation

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡

reac*on ¡func*on ¡firm ¡B ¡

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

17

Nash equilibrium in License Fees firms’ best response is to set limit licensing fees to avoid litigation

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡

reac*on ¡func*on ¡firm ¡B ¡ reac*on ¡func*on ¡firm ¡A ¡

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

17

Nash equilibrium in License Fees firms’ best response is to set limit licensing fees to avoid litigation

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡

reac*on ¡func*on ¡firm ¡B ¡ reac*on ¡func*on ¡firm ¡A ¡

Nash equilibrium license fees with independent firms: (¯ f ∗

A, ¯

f ∗

B)

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

17

• 2. Licensing and Litigation with Patent Pool and Package License

firms A and B form patent pool and sell package licenses for a fee ¯ F patent pool maximizes the joint profit of patent holders downstream firm will either (i) buy the package license or (ii) enter into litigation against both patents by assumption (A) challenging both is superior to remaining inactive challenging exactly one patent not enough to break up monopoly power; dominated by remaining inactive downstream firm buys the package license if v − F ≥ (1 − α)(1 − β)v − 2L (3)

• r

F ≤ [1 − (1 − α)(1 − β)]v + 2L ≡ ¯ F ∗∗.

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

18

patent pool can either: (i) limit license at highest fee ¯ F ∗∗ that avoids litigation or (ii) enter litigation against both patents litigation yields monopoly profits if at least one patent is deemed valid by the court expected profits from litigation are dominated by limit licensing revenues since [1 − (1 − α)(1 − β)]v − 2L < ¯ F ∗∗ = 1 − (1 − α)(1 − β)]v + 2L ⇒ limit licensing at ¯ F ∗∗ at is always optimal for the patent pool how does limit license fee of pool compare with total license fees with independent firms?

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

19

Aggregate licensing fees with independent firms and patent pool

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡

it holds that VA|fB=f

∗ B = VB|fA=f ∗ A = V0|F=F ∗ > VAB = V0|F=F ∗∗ Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

20

Proposition

Suppose only the litigation margin is binding. A patent pool with a package license charges a higher aggregate limit license fee compared to independent patent holders, that is, F

∗ = f ∗ A + f ∗ B < F ∗∗.

with weak patents and litigation: reverse result of Proposition 1 first, package licensing makes it unprofitable to selectively challenge

• ne patent: only way to reduce C’s payment to the pool is to

successfully challenge all its patents all-or-nothing litigation proposition allows pool to charge higher aggregate limit license fees second, independent patent holders unable to sustain such high license fees because they are engaged in Bertrand-type competition with respect to litigation incentives

Patent Pools, Litigation and Innovation

• 3. Probabilistic Patent Rights and Litigation with Patent Pools

21

• 4. The Interplay of the Demand Margin and Litigation

Incentives

• 1. Litigation and Licensing with Independent Firms

three different cases can arise as function of the equilibrium values with demand and litigation margin

1

litigation margins are not binding f ∗ < ¯ fA ≤ ¯ fB (9)

2

both litigation margins are binding: limit licensing fee less than best responses to rival’s limit licensing fee ¯ fB ≤ Θ(¯ fA) (10)

3

litigation margin only binds for firm A: conditions (9) and (10) not satisfied; requires asymmetric patent positions

Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

22

Case 1: Litigation margins not binding and f ∗ < ¯ fA fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡

Θ(fA) ¡ Θ(fB) ¡

f ¡ * ¡ f* ¡ Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

23

Case 2: Both litigation margins binding and ¯ fB ≤ Θ(¯ fA) fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡

Θ(fA) ¡ Θ(fB) ¡

f ¡ * ¡ f ¡ * ¡

Nash equilibrium in which both set limit licensing fees (¯ fA, ¯ fB) exists if L/v sufficiently small

Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

24

Case 3: Only firm A’s litigation margin binding fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ (5) ¡ (6) ¡ fA ¡* ¡ fB ¡ * ¡

Θ(fA) ¡ Θ(fB) ¡

f ¡ * ¡ f ¡ * ¡ Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

25

• 2. Litigation and Licensing with Patent Pool and Package License

if optimal license fee with demand margin only F ∗∗ is below ¯ F ∗∗, litigation margin is not binding

• therwise, litigation margin is binding and optimal price is ¯

F ∗∗ hence, the optimal license fee satisfies min{F ∗∗, ¯ F ∗∗} ¯ F ∗∗ is decreasing in L/v while F ∗∗ is independent limit license fee ¯ F ∗∗ is optimal when L/v sufficiently small ⇒ there exist values such that litigation margin is binding with and without patent pool

Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

26

Proposition

Consider the full game with demand and litigation margin. If patents are sufficiently weak and litigation cost low relative to the value of the innovation, then patent pools hinder subsequent innovations and reduce welfare.

Linear Example: symmetric patent positions α = β v = 1 and c uniformly distributed along [0, 1]

• ptimal demand margin fees are

2f ∗ = F ∗ = 2 3 and F ∗∗ = 1 2

• ptimal litigation margin fees are

¯ f ∗

A = ¯

f ∗

B = L + α

1 + α and F ∗ = α(2 − α) + 2L

Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

27

Optimal License Fees in Linear Example

L ¡ F ¡

(1-­‑2α)/3 ¡ (1-­‑3α)/4 ¡ (1-­‑2α(2-­‑α))/4 ¡

F** ¡ F** ¡ F* ¡ F* ¡ Patent ¡Pool ¡ Independent ¡firms ¡

Patent Pools, Litigation and Innovation

• 4. The Interplay of the Demand Margin and Litigation Incentives

28

• 5. Patent Pool with Individual Licenses

patent pool issues individual licenses at fees fA and fB to maximise joint profit of holders downstream firm’s litigation and licensing behaviour like with independent patent holders patent pool can sustain fees that induce exactly one litigation (no more Bertrand competition) litigation against j allows to limit license patent i at Vj = VAB or fi = piv + L 1 − pj patent pool makes same expected profits independent of which patent litigated pjv + (1 − pj)[piv + L 1 − pj ] − L = [1 − (1 − α)(1 − β)]v

Patent Pools, Litigation and Innovation

• 5. Patent Pool with Individual Licenses

29

Optimal Fees with Individual Licenses

fA ¡ fB ¡

0 ¡(no ¡li'ga'on, ¡ ¡buy ¡both ¡licenses) ¡ A ¡(li'gate ¡patent ¡A, ¡ ¡buy ¡license ¡from ¡B) ¡ B ¡(li'gate ¡patent ¡B, ¡ ¡buy ¡license ¡from ¡A) ¡ AB ¡(li'gate ¡ ¡ both ¡patents) ¡ v ¡ v ¡ ( 4 ) ¡ αv+L/(1-­‑β) ¡ βv+L/ (1-­‑α) ¡ αv+L ¡ βv+L ¡ (5) ¡ (6) ¡ fA ¡ * ¡ fB ¡ * ¡

pool ¡with ¡ ¡ individual ¡licenses ¡ pool ¡with ¡ ¡ package ¡license ¡

• with individual licenses pool might find it optimal to raise ex ante fees and induce

litigation against one patent

• however, package licensing dominates as it avoids cost of litigation

Patent Pools, Litigation and Innovation

• 5. Patent Pool with Individual Licenses

30

Optimal Pool Organization and Welfare

Proposition

Package licensing yields (weakly) higher profits for the patent pool compared to selling individual licenses. If L ≤ L′′, package licensing is welfare superior to individual licenses. For higher values of L, individual licensing yields weakly higher welfare.

Corollary

For intermediate litigation costs L′′ < L ≤ L′, mandatory individual licensing for patent pools increases welfare and implements the same

• utcome as with independent patent holders.

Patent Pools, Litigation and Innovation

• 5. Patent Pool with Individual Licenses

31

• 6. Extension 1: Licensing and litigation with more than

two patents.

• ur qualitative results carry over to the case with n > 2 symmetric

patents with independent patent holders: Firm C buys cheapest licenses and litigates against more expensive ones limit licensing equilibrium with total license fees n¯ f ∗(n) which is increasing in n with pool and package license limit license increases exponentially ¯ F ∗∗(n) = (1 − (1 − α)n)v + nL when only litigation margin bind, result is reinforced as the number of firms increases

Patent Pools, Litigation and Innovation

• 6. Extensions

32

when the demand margin is also active, the pool’s optimal charge is bounded as n increases since min{ ¯ F ∗∗(n), F ∗∗} < v with independent patent holders the fee is min{nf ∗(n), n¯ f ∗} → v as n → ∞

Proposition

If nf

∗(n) ≤ F ∗∗, then patent pools with n 2 patents charge higher total

license fees and reduce total welfare relative to independent patent

• holders. This condition is harder to satisfy, the higher the number of

complementary patents. There exists a finite upper bound n on the number of patents, above which patent pools increase total welfare.

Patent Pools, Litigation and Innovation

• 6. Extensions

33

• 6. Extension 2: Sequential Litigation

in some situations there might be sufficient time for firm C to consider sequential litigation rather than litigating all patents simultaneously sequential litigation increases the value of litigating all patents as it might save the cost of second litigation if first court case is lost it is optimal to litigate against stronger patent first this affects licensing and litigation with patent pool and independent patent holder alike reduces equilibrium licensing fees relative simultaneous litigation however, same qualitative results as with simultaneous litigation

Patent Pools, Litigation and Innovation

• 6. Extensions

34

• 7. Independent Licensing Requirement

Independent Licensing Requirement (ILR): individual patents, which are part of pool, also licensed from original patent owner Lerner and Tirole (2004): ILR can screen in welfare-enhancing patent pools and prevent formation of welfare-reducing pools ILR does not prevent patent pool from charging higher aggregate licensing fees in this context

Proposition

Consider a patent pool with independent licensing. If n > 2, then there exist equilibria in which a patent pool sells its package at a total license fee exceeding the total price charged in the absence of a pool. best deviation for individual license: lower fee to sell this license and induce litigation against other holders rather than buy package license profitability of deviation is limited by the cost of litigation it induces

Patent Pools, Litigation and Innovation

• 7. Independent Licensing Requirement

35

ILR and Unbundling

additional constraint: pool is not allowed to give package discount package license fee not exceeding sum of individual license fees: F = nf

Proposition

Consider independent licensing together with an unbundling requirement. Then, there exists no equilibrium in which the pool is able to sell its package license at a fee exceeding total license fees in the absence of a patent pool. ILR and strict unbundling requirement can screen out welfare-reducing patent pools

Patent Pools, Litigation and Innovation

• 7. Independent Licensing Requirement

36

Conclusions

effect of patent pools on innovation/licensing/litigation when patents are probabilistic and weak patent pools including complementary patents can increase total licensing fees and reduce welfare relative to independent patent holders patent pools with package licenses are stronger litigation entity as litigating individual patents is not profitable (“all-or-nothing proposal”) with independent patent holders selective litigation profitable and patent holders compete in Bertrand fashion to avoid litigation blanket approval of pools with complementary patents is not warranted cautious policy should take into account patent strength and litigation incentives

Patent Pools, Litigation and Innovation Conclusions 37