software patents, economic evidence and competition

Software patents, economic evidence and competition

Open Forum on the Substantive Patent Law TreatyWorld Intellectual Property OrganisationGeneva, March 3, 2006

Rishab Aiyer Ghosh - [email protected] Nations University / MERITMaastrichtDisclaimer: this presentation represents the author's personal views

Overview Software patents: definitions and evidence

What are software patents Who owns them Why do firms patent software

Software, networks and competition Economic effects of patents on competition Economic effects of types of standards

What are software patents? Computer Implemented Inventions =

“inventions whose implementation involves the use of a computer, computer network or other programmable apparatus, the invention having one or more features which are realized wholly or partly by means of a computer program”

If “realized wholly”, then =software Such definitions cannot be used for large-

scale empirical analysis

What are software patents? Large scale searches possible using:

Specific patent classes (“IPC method”) Keywords (“Bessen & Hunt method”) All classes for largest software firms (“Hall

& MacGarvie method”

Who holds software patents?Industry

Bessen-Hunt

Graham-Mowery

Hall-MacGarvie

All definitions combined

Telecommunications 2 1 744 222Electrical machinery 3 2 566 137Transportation equipment 1 415 128 54Machinery & engines 3 576 149 47Autos 2 386 117 38Oil 1 475 259 28Auto parts 497 197 64 25Chemicals 1 160 56 23Printing 119 96 69 12Wholesale trade 115 250 169 8Fabricated metals 315 93 50 7Misc 210 82 70 7Computing equipment 21 18 11 6Paper 415 78 33 5Furniture 138 45 18 4Pharmaceuticals 1 153 87 4Food & tobacco 253 146 94 2Primary metals 200 74 41 2Business services NEC 118 145 102 2Instruments & Comm. eq. 12 9 4 2Medical instruments 19 28 21 1Computing systems & software 2 2 2 1Textiles & apparel 54 105 81 1Rubber & plastics 103 48 29 1Stone, clay, & glass 116 50 28 1Lumber & wood 17 27 22 0Soap 205 24 16 0Total 72.657 63.838 41.361 10.455

Source: Torrisi & Thoma, 2006 CESPRI

Evolution of software patentsCountry of assignee for EPO software patents


0

20 0 0

40 0 0

60 0 0

80 0 0

1 00 0 0

1 20 0 0

79 8 0 8 1 82 8 3 84 8 5 8 6 87 8 8 8 9 9 0 9 1 92 9 3 9 4 95 9 6 97 9 8 9 9 00 0 1 0 2 03

DE FR G B IT US JP o the rs

Evolution of software patents

0% 2% 4% 6% 8%

10% 12% 14% 16%

1981 1984 1987 1990 1993 1996 1999 2002 Year

% o

f all

pate

nts

EPO - keyword method

EPO - IPC method

USPTO – keyword method

USPTO - IPC method

USPTO - HM method

Software patents as a share of all patents, US vs EPO


Evolution of software patentsConcentration of EPO software patents by assignee sector


0%

1 0%

2 0%

3 0%

4 0%

5 0%

6 0%

7 0%

8 0%

9 0%

1 0 0%

1% 5% 9%13% 17%

21%25% 29%

33% 37% 41% 45%49%

53%57%

61% 65%69% 73% 77%

81%85%

89% 93%

97%

cumulative perc entage of as s ignee s

cum

ulat

ive

perc

enta

ge o

f pat

ents

s ic3 6 E lectrica l,ele ctro nic an d te lecom m . e qu ip

s ic3 5 C o m p ute r e q uip . a n d ind u strial m a chin ery an d equip .

s ic3 8 In strum e nts an d re late d p rod ucts

s ic7 3 So ftw are an d o the r bu s in e ss se rvice s

s ic4 8 Te le com m u nica tio n s e rvices

The top 10 assignees account for between 57 and 78 per cent of EPO software patents

Evolution of software patentsEPO software patents by publication year and size* of assignee


*Size is EU definition. Large >250; Medium >50, Small >10 employees

0

1000

2000

3000

4000

5000

6000

1982

1983

1984198

519

86198

719

881989

1990

1991

199219

931994

1995

1996

199719

981999

2000

2001

2002200

32004

pu b lica tio n ye ar s

soft

war

e p

aten

ts

Micro Small Med ium Large

“Quality” indicatorsTechnical quality can only be measured by inspection of patents themselves; citations etc are possible proxies


0.00

0.50

1.00

1.50

2.00

2.50

1991 1992 1993 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

publication ye ar

inde

x num

bers

(bas

e=19

91)

IPCs codes claims family backw ard citations forw ard c itations w ithin 5 years

Software patents and R&DEPO software patent stocks per million euros of R&D stocks – US software firms


0 .0 0

0 .0 1

0 .0 1

0 .0 2

0 .0 2

0 .0 3

0 .0 3

0 .0 4

1 9 9 1 1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9

y e a r

pate

nts/

inta

ngib

les

Software patents and jobsEmployment average growth rate for European SSPs, holders vs non-holders of software patents


Sector Software patenting

1997-99 (base 1994-96)

2000-02 (base 1997-99)

2003-04 (base 2000-02)

IT services no 81.89% 88.37% 27.96%

IT services yes 61.47% 59.42% -6.84% Pre-packaged software no 61.66% 66.13% 30.47%

Pre-packaged software yes 107.40% 131.87% -11.66%

Why firms patents softwareReasons for filing for a software / CII patent

8

11

15

23

43

49

0 10 20 30 40 50 60

Generate publicity or attract venture capital

Use for cross-licensing or patent exchanges

Earn licensing revenue

Block competitors from entering yourmarket

Defend your firm against infringement suits

Prevent competitors from copying yourideas

% respondents "very important" or "moderately important"Source: Ghosh, Glott et al, 2006 MERIT

Software patents: inhibition

Source: Ghosh, Glott et al, 2006 MERIT

In the last three years, has concern that your software might infringe a patent led your firm to:

Percentage (n = 498)

Change or abandon a software development project to avoid infringement

10.2

Avoid markets where your software might infringe patents

11.0

Obtain a patent license in order to develop or use software

12.2

multiple response

Software patents: disclosure


In the last 3 years, were any of the following important sources of ideas for your firm's new or improved software?

5.2

12.4

17.1

21.9

29.9

42.8

Patent databases (published patents)

Proprietary software source code in other products

Free / Open Source software source code in other products

Scientific publications or trade journals

Internet forums or other informal exchanges

Products on the market

multiple response

n = 498

Software patents: importance


Which of the following are important methods for helping your firm to profit from its new or improved software?

8.6

10.0

11.2

14.5

17.5

21.5

23.9

30.3

patenting the hardware linked to the software

patenting the software

first to publish

copyright

trade secrecy

technical protection measures

trademarks / marketing

first to market

multiple response

n = 498

Patents and competition Patents do not only “protect ideas from theft

or imitation” - they also prevent independent creation of ideas

“Patents amount to temporary monopolies on useful new inventions” (Economist)

These monopolies are justified on the basis of increasing knowledge sharing through disclosure, thus increasing innovation

Needs to be demonstrated through evidence

Patents and competition Software has three unusual properties

Mainly incremental innovation Several problems have only one solution Network effects are very strong

Patents and competition Network effects can form entry barriers for

new technologies Path dependence, QWERTY... Natural monopolies to maximise welfare from

network effects Monopolies can lead to rent-seeking and

capture of network externalities

Economics of standards Alternative approach: separate technology

from producer Truly open standards allow natural

monopolies of technologies (standards) while providing for full competition among vendors

Types of standards Proprietary (“standard”?) technologies

Natural monopoly in technology leads to natural monopoly in market for products and services based on that technology

Results when access to the technology is available only to the rights holders

Types of standards (“Semi-open”?) Standard technologies

Natural monopoly in technology arises (de facto) or is defined (de jure) but some competition provided for in market for products and services

Results when access to the technology is available to players other than the rights holders, perhaps retaining advantages for the rights holders

Types of standards Open standard technologies

Natural monopoly in technology arises (de facto) or is defined (de jure) but full competition provided for in market for products and services

Results when access to the technology is available to all (potential) players on equal terms providing no a priori advantages based on ownership of rights

Open source software Provides main competing product in:

Web servers (#1 in market share) Server operating systems (#2) Network file systems (#2) Office productivity software (#2) Web browsers (#2)

Assumes authors have full rights to created software – compatible with copyright (independent creation), not with patents

software patents, economic evidence and competition

Documents