Cockburn MacGarvie (2009) - Patents Thickets And The Financing Of Early Stage Firms

From edegan.com
Revision as of 19:18, 25 March 2013 by imported>John
Jump to navigation Jump to search

Reference

  • Cockburn, I.M. and MacGarvie, M.J. (2009), "Patents, Thickets and the Financing of Early-Stage Firms: Evidence from the Software Industry", Journal of Economics & Management Strategy, Vol.18, No.3, pp.729--773
@article{cockburn2009patents,
  title={Patents, Thickets and the Financing of Early-Stage Firms: Evidence from the Software Industry},
  author={Cockburn, I.M. and MacGarvie, M.J.},
  journal={Journal of Economics \& Management Strategy},
  volume={18},
  number={3},
  pages={729--773},
  year={2009},
  abstract={The impact of stronger intellectual property rights in the software industry is controversial. One means by which patents can affect technical change, industry dynamics, and ultimately welfare, is through their role in stimulating or stifling entry by new ventures. Patents can block entry, or raise entrants’ costs in variety of ways, while at the same time they may stimulate entry by improving the bargaining position of entrants vis-à-vis incumbents, and supporting a “market for technology” which enables new ventures to license their way into the market, or realize value through trade in their intangible assets. One important impact of patents may be their influence on capital markets, and here we find evidence that the extraordinary growth in patenting of software during the 1990s is associated with significant effects on the financing of software companies. Start-up software companies operating in markets characterized by denser patent thickets see their initial acquisition of VC funding delayed relative to firms in markets less affected by patents. The relationship between patents and the probability of IPO or acquisition is more complex, but there is some evidence that firms without patents are less likely to go public if they operate in a market characterized by patent thickets.},
  discipline={Econ},
  research_type={Empirical},
  industry={Software},
  thicket_stance={},
  thicket_stance_extract={},
  thicket_def={},
  thicket_def_extract={},  
  tags={},
  filename={Cockburn MacGarvie (2009) - Patents Thickets And The Financing Of Early Stage Firms.pdf}
}

File(s)

Abstract

The impact of stronger intellectual property rights in the software industry is controversial. One means by which patents can affect technical change, industry dynamics, and ultimately welfare, is through their role in stimulating or stifling entry by new ventures. Patents can block entry, or raise entrants’ costs in variety of ways, while at the same time they may stimulate entry by improving the bargaining position of entrants vis-à-vis incumbents, and supporting a “market for technology” which enables new ventures to license their way into the market, or realize value through trade in their intangible assets. One important impact of patents may be their influence on capital markets, and here we find evidence that the extraordinary growth in patenting of software during the 1990s is associated with significant effects on the financing of software companies. Start-up software companies operating in markets characterized by denser patent thickets see their initial acquisition of VC funding delayed relative to firms in markets less affected by patents. The relationship between patents and the probability of IPO or acquisition is more complex, but there is some evidence that firms without patents are less likely to go public if they operate in a market characterized by patent thickets.

Review

Measures of thicket

The key measures of patent thickets identified in the paper are:

  • The cumulative number of patents in a market, which the authors characterize as the:" 'main effect' of patent thickets".
    • Patent classes in NBER Patent Database are mapped to product market considered through text search and review of the Manual of Patent Classification to identify patents "relevant" to each market.
  • Indicators of patent regimes periods for each market (with regime changes in 1996 or 1998 depending on product market).

Other measures of patent thickets in a market include:

  • A Herfindahl index for the concentration of citations among patent assignees in a market;
  • The number of assignees cited in patents in a market;
  • The average number of claims per patent in a market;
  • The average number of non-patent references to prior art;
  • The ratio of the number of patent claims to the citations (more indicating greater uncertainty as to patent validity);

Sample

Size, data source, industries, geography:

  • 19,717 public and private firms from 1990-2004 identified in the CorpTech directory as software development companies (and further classificied into 27 distinct product markets). For these companies, external venture or corporate funding is drawn from the SDC VentureXpert database and sales and other information is drawn from Compustat.

Results

"[There is]...a significant positive effect of the number of patents in the market on the probability of exiting via IPO, although the interaction effect with the legal regime change is negative...[so] patent thickets tend to protect incumbents, but only up to a point. ...[However, for] the subset of firms that did not obtain any patents of their own...are much less likely to exit via IPO, relative either to liquidation/censoring or to acquisition, in markets with a large patent thicket. This finding is striking in light of the fact that only a minority of the firms in our sample ( 22%) hold patents by the time of exit."

Most important patent thicket measures:

  • In general, the number of patents in the market, and the indicators of regime changes that strengthened patent protection were found to affect the dependent variables analyzed.
    • The Herfindahl index of assignee concentration was also found significant for one dependent variable.
    • Other patent thicket measures related to uncertainty over patent validity were not found to be significantly related to dependent variables analyzed.

The number of market entrants:

  • The number of market entrants is significantly reduced by patent thicket measures:
    • The number of patents in the market (with an elasticity of -0.438);
    • The average claimes per cite in the market (with an elasticity of -0.371);
    • There is a net reduction in entry after regime changes that strengthened patents (based on median impact between regime and number of patents interaction term).

Initial Funding:

"Patent-intensive markets saw a reduction in number of firms receiving initial investment by external parties in early-stage firms relative to low-patent markets following the expansion of the software patentability in the 1990s"
  • Funding levels are significantly increased by the number of patents in a market (with an elasticity near 1), which authors attribute to selection effect reflecting higher entry costs in markets with patent thickets.

The number of IPOs:

"We also find a statistically significant association between our measures of patent-related uncertainty and the number of IPOs in a market in a given year. Investors in public securities appear to be more willing to invest in software companies operating in markets in which there are fewer “problem patents” and in which patents cite more non-patent prior art."
  • The number of IPOs is significantly positively increased by number of pattents, but also has a large offsetting negative effect associated with the strengthening of patenting rights, which authors view as having made such markets less attractive.

Obtaining initial funding:

"Start-up software companies operating in markets characterized by denser patent thickets see their initial acquisition of VC or corporate funding delayed relative to firms in markets less affected by patents."
  • Firm exits from the entreprenuerial phase by going public or being acquired are positively related to the number of patents in the market, but the reverse effect is found for IPOs among the 78% of firms that hold no patents.
  • Software ventures in the most thicketed markets have a very small probability of obtaining outside funding:
    • The hazard of funding is higher for those firms with their own patents;
    • After 1996 strengthening of patents, chances of funding fall by :"0.4 for a one standard deviation increase in the patent stock in the market. Given that the mean probability of obtaining funding in a given year is 0.085 overall and 0.102 after 1996, this effect appears to be of considerable importance."
    • The Herfindahl index of patent citations over assignees in each market has a positive effect on reciept of funding, which authors attribute to higher bargaining costs.

Social Welfare Consequences:

  • While the the number of patents in a market is positively related to initial funding, legal regimes strengthening patent rights reduce this relationship.
"...patents confer private benefits to software companies. These benefits appear to have been substantial, and are reflected in the extraordinary surge in patenting in this industry. However, these incentives to obtain patents may ultimately become collectively self-destructive. Our differences-in-differences estimates of the relative impact of strengthening patent rights show a generally negative effect on entry and financing of software firms in the most heavily thicketed markets. Continued accumulation of patents may therefore result in the “stifling” effects identified here swamping the offsetting “stimulating” effect on innovation."

Dependent Variables and Models analyzed:

  • The number of "new" market entrants (firms founded within two years) using Poisson regressions;
  • Measures of investment (the number of firms receiving an initial round of funding from external investors, median amount invested, number of IPO) using Poisson and OLS regressions;
  • Time to receipt of external funding from VC or corporate investors using hazards models;
  • Odds of exiting the entrepreneurial phase as a result of acquisition or IPO using hazards multinomial logit model.

Models control for the patent thicket measures listed above and a range of market and patenting conditions:

  • The mean number of non-patent references (patent quality);
  • The modal citation lag between citing and cited years in a market (shorter lags indicating an earlier stage in product cycle);
  • Number of other firms in the market;
  • Revenue growth in the market;
  • Year and market fixed effects;
  • A difference-in-difference approach is sometimes used to evaluate impacts of entry and financing before and after the 1990s when a change in legal regime led to an increase in the issuance of patents overall.