Public And Private Provision Of Public And Private Goods
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Public And Private Provision Of Public And Private Goods as PDF for free.
More details
- Words: 927
- Pages: 19
24th Congress of the European Economic Association Barcelona, August 23-27, 2009
Public and private provision of public and private goods* * Paper Title: Consumer Welfare and Market Structure in a Model of Competition between Open Source and Proprietary Software
Alexia Gaudeul GSBC, Jena
Introduction
A number of industries see public provision of public goods along with private provision of private goods:
Health care (Sloan 2000), Education (Epple and Romano 1998), Art (Frey 1999).
Differences in incentives and objectives (Brhlikova 2004 and 2006, Sloan 2000), disciplining effect (Glaeser 2002), crowding out (Frey 1999).
Introduction
In the software industry, however, cohabitation between private provision of public goods and private provision of private goods.
software developed under proprietary license terms and
software developed under open source (OS) license terms.
Introduction
Strand of literature on private provision of public good (Bergstrom et al. 1986).
One public good, many private goods.
Application to Open Source:
Schmidtke (2006): OS software (OSS) as complement to proprietary software is worth investing into. Polborn (2008): Developing OSS is like joining a club, competition between clubs may lead to over-provision of OSS.
Contribution
Model inspired by “The number and size of nations”, Alesina and Spolaore 1997.
Advantage:
Many OS projects and proprietary firms all using the same technology. Competition in attracting users/developers.
Drawback: Free riding is not modeled.
Consider entry by proprietary software in OS industry and conversely. Consider equilibrium of a mixed industry.
Motivation
Historical notes: OS was the original system of software development (50s to 70s), with proprietary software coming later (70s, 80s).
GNU (1983) was a reaction, led by Stallmann, to a trend towards proprietarization of software (cf. Unix and AT&T).
Share of OS: survey of a number of development areas to determine the prevalence of the use of OSS vs proprietary software (cf. “Why OSS does not succeed”, Gaudeul 2008).
Among other findings, OSS almost systematically has lower shares than proprietary software, except in some specific areas, where freeware and OSS cohabit.
Motivation
LATEX case study (Gaudeul 2007): focus on the dynamics and patterns of cohabitation of proprietary and OS software.
Proprietary software may either dominate an area or
Install itself in niches not served by OSS o r
Attach itself to OSS.
Questions
The above notes raise several questions:
Are each production models robust to entry by the competing alternative production model?
Do mixed industry models provide higher or lower welfare than industries with a single production model?
What is the share of users of OS vs. proprietary software in a mixed industry model?
Results
The OS development model on its own may be more efficient from the point of view of welfare than proprietary development. OS industry vulnerable to entry by proprietary products. Proprietary industry robust to entry by OS products. A mixed industry where OS and proprietary development methods co-exist may exhibit large OS projects cohabiting with more specialized proprietary projects.
This co-existence improves welfare vs. proprietary model, and may do so also vs. OS.
Model
A location model: Each product costs K to produce, and has value g to the consumer that is closest to its location. A consumer that is located at distance li from the location of the product derives utility g(1-ali) from the product. There are two modes of production.
Either users contribute to collective production, with costs shared equally. or an entrepreneur produces the good and sells it for profit at price p.
Free entry.
Equilibria
Optimally, size s*=(4k/ag)1/2 (few large projects). Proprietary: size sp=(k/ag)1/2 (many small projects). Open source: size sos>(2k/ag)1/2 (large projects, but may grow too large!).
However, size limited by risk of forking (setting up independent project).
Size further limited by threat of proprietary entry.
Mixed industry
What would consumers prefer?
OS system, with possibly less specialized projects but requiring only low contribution? PS system, with possibly more specialized projects at a higher price?
Mixed industry
Suppose open-source and proprietary projects alternate in the consumer's preference space.
Proprietary Open-source
Mixed industry
First equilibrium:
open-source and proprietary projects are of the same size, s p .
Half of consumers use OSS.
Second equilibrium:
Open-source projects of size s * .
Proprietary projects of size s p .
2/3 of consumers use OSS.
Welfare 5 Optimal size
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 Proprietary size, inefficient
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 OS size, may be too large
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 Mixed industry, if second equilibrium
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Extensions
Proportion of consumers who cannot or will not use OSS. Proportion of consumers who “free ride” on OSS (use it but do not or cannot develop it). Proportion of developers who will never buy PS. Dynamics in choice of location and growth of projects. Complementarities OS and PS:
PS built on top of OSS (e.g. to facilitate access).
PS using OS code.
OSS reverse-engineering PS.
Public and private provision of public and private goods* * Paper Title: Consumer Welfare and Market Structure in a Model of Competition between Open Source and Proprietary Software
Alexia Gaudeul GSBC, Jena
Introduction
A number of industries see public provision of public goods along with private provision of private goods:
Health care (Sloan 2000), Education (Epple and Romano 1998), Art (Frey 1999).
Differences in incentives and objectives (Brhlikova 2004 and 2006, Sloan 2000), disciplining effect (Glaeser 2002), crowding out (Frey 1999).
Introduction
In the software industry, however, cohabitation between private provision of public goods and private provision of private goods.
software developed under proprietary license terms and
software developed under open source (OS) license terms.
Introduction
Strand of literature on private provision of public good (Bergstrom et al. 1986).
One public good, many private goods.
Application to Open Source:
Schmidtke (2006): OS software (OSS) as complement to proprietary software is worth investing into. Polborn (2008): Developing OSS is like joining a club, competition between clubs may lead to over-provision of OSS.
Contribution
Model inspired by “The number and size of nations”, Alesina and Spolaore 1997.
Advantage:
Many OS projects and proprietary firms all using the same technology. Competition in attracting users/developers.
Drawback: Free riding is not modeled.
Consider entry by proprietary software in OS industry and conversely. Consider equilibrium of a mixed industry.
Motivation
Historical notes: OS was the original system of software development (50s to 70s), with proprietary software coming later (70s, 80s).
GNU (1983) was a reaction, led by Stallmann, to a trend towards proprietarization of software (cf. Unix and AT&T).
Share of OS: survey of a number of development areas to determine the prevalence of the use of OSS vs proprietary software (cf. “Why OSS does not succeed”, Gaudeul 2008).
Among other findings, OSS almost systematically has lower shares than proprietary software, except in some specific areas, where freeware and OSS cohabit.
Motivation
LATEX case study (Gaudeul 2007): focus on the dynamics and patterns of cohabitation of proprietary and OS software.
Proprietary software may either dominate an area or
Install itself in niches not served by OSS o r
Attach itself to OSS.
Questions
The above notes raise several questions:
Are each production models robust to entry by the competing alternative production model?
Do mixed industry models provide higher or lower welfare than industries with a single production model?
What is the share of users of OS vs. proprietary software in a mixed industry model?
Results
The OS development model on its own may be more efficient from the point of view of welfare than proprietary development. OS industry vulnerable to entry by proprietary products. Proprietary industry robust to entry by OS products. A mixed industry where OS and proprietary development methods co-exist may exhibit large OS projects cohabiting with more specialized proprietary projects.
This co-existence improves welfare vs. proprietary model, and may do so also vs. OS.
Model
A location model: Each product costs K to produce, and has value g to the consumer that is closest to its location. A consumer that is located at distance li from the location of the product derives utility g(1-ali) from the product. There are two modes of production.
Either users contribute to collective production, with costs shared equally. or an entrepreneur produces the good and sells it for profit at price p.
Free entry.
Equilibria
Optimally, size s*=(4k/ag)1/2 (few large projects). Proprietary: size sp=(k/ag)1/2 (many small projects). Open source: size sos>(2k/ag)1/2 (large projects, but may grow too large!).
However, size limited by risk of forking (setting up independent project).
Size further limited by threat of proprietary entry.
Mixed industry
What would consumers prefer?
OS system, with possibly less specialized projects but requiring only low contribution? PS system, with possibly more specialized projects at a higher price?
Mixed industry
Suppose open-source and proprietary projects alternate in the consumer's preference space.
Proprietary Open-source
Mixed industry
First equilibrium:
open-source and proprietary projects are of the same size, s p .
Half of consumers use OSS.
Second equilibrium:
Open-source projects of size s * .
Proprietary projects of size s p .
2/3 of consumers use OSS.
Welfare 5 Optimal size
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 Proprietary size, inefficient
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 OS size, may be too large
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Welfare 5 Mixed industry, if second equilibrium
SW* 4 SWmixed SWp 3
SW=g-K/s-ags/4
2 1 0 -1
Somin Sp S*
Soentry0.2 Sofork
0.3
0.4
s
0.5
Extensions
Proportion of consumers who cannot or will not use OSS. Proportion of consumers who “free ride” on OSS (use it but do not or cannot develop it). Proportion of developers who will never buy PS. Dynamics in choice of location and growth of projects. Complementarities OS and PS:
PS built on top of OSS (e.g. to facilitate access).
PS using OS code.
OSS reverse-engineering PS.
Related Documents
Public And Private
November 2019 33
Public Infrastructure And Private Productivity
November 2019 21
Analysis Public And Private Sectors.docx
June 2020 22
Public- Private Partnership
June 2020 28
Public Art, Private Places
June 2020 11More Documents from "Lisa Temple-Cox"
