Gompers Lerner Sharfstein (2003) - Entrepreneurial Spawning

Article
Has bibtex key
Has article title
Has year
In journal
In volume
In number
Has pages
Has publisher
© edegan.com, 2016

• This page is referenced in BPP Field Exam Papers

Reference

• Gompers, P., J. Lerner, and D. Sharfstein (2003), "Entrepreneurial Spawning: Public Corporations and the Genesis of New Ventures, 1986-1999," NBER working paper #9816. pdf

Abstract

This paper examines the factors that lead to the creation of venture capital backed start-ups, a process we term "entrepreneurial spawning." We contrast two alternative views of the spawning process. In one view, employees of established firms are trained and conditioned to be entrepreneurs by being exposed to the entrepreneurial process and by working in a network of entrepreneurs and venture capitalists. Alternatively, individuals become entrepreneurs because the large bureaucratic companies for which they work are reluctant to fund their entrepreneurial ideas. Controlling for a firm's size, patent portfolio and industry, we find that the most prolific spawning firms were public companies located in Silicon Valley and Massachusetts that were themselves once venture capital backed. Less diversified firms are also more likely to spawn new firms. Spawning levels for these firms rise as their sales growth declines. Firms based in Silicon Valley and Massachusetts and originally backed by venture capitalists are more likely to spawn firms only peripherally related to their core businesses. Overall, these findings appear to be more consistent with the view that entrepreneurial learning and networks are important factors in the creation of venture capital backed firms.

Entrepreneurial Spawning

In the context of this paper entrepreneurial spawning refers to the creation of startup that recieves venture capital by a former employee of a large and established firm. The paper considers two views of spawning: The Fairchild View and the Xerox View.

The Fairchild View

The Fairchild view is loosesly that VC back firms will spawn more VC backed firms.

The Fairchild view supposes that:

• Employees are trained and conditioned to be entrepreneurs by working in a VC backed firm
  • Exposure to the entrepreneurial process
  • Access to network of entrepreneurs and VCs
• Located in a hotbed of VC
  • Easy access to network of suppliers, labor, goods and capital, as well as customers
  • Greater likelihood of relational-specific investments
• Selection effect - individuals with taste for higher risk, entrepreneurial firms may self select into VC backed firms

The Xerox View

The Xerox view is loosely that large bureaucratic firms may be unwilling to fund new entrepreneurial ideas, and so innovators leave.

The Xerox view supposes that:

• Large firms are incapable of responding to radical changes that upset their business
  • Thus the invention of a disruptive technology leads the inventor to leave
• High-level managers may be incapable of evaluating technologies outside of the large firm's core
  • Internal capital markets favour established lines of business
• These firms may focus on their core competencies
  • Supported by findings that diversified firms trade at a discount, and are more likely to be broken up
  • Also diversification leads to declines in productivity in existing business units
• Possibly employees get frustrated - though how do you measure this?

Data and Findings

The data is on VC backed firms from 1986 to 1999. The founders of the firms were identified, as was their previous employer.

Basic Results:

• Previous work in silicon valley increases spawning by 38%
• Previous work in MA increases spawning by 23%
• Previous work at a VC backed firm increases spawning by 23%
  • This supports the Fairchild view over the Xerox view
• Previous work at single line of business firm increases spawning by 19%
  • Therefore diversified firms don't spawn more
• Older firms spawn less (6%), and spawning declines with age.
• Larger firms spawn more
• Firms with more patents in category 2 or 3 spawn more, patents in other categories lead to less spawning
• Prior VC affects spawning but not how much.
• Spawning is unrelated to tobin's Q and EBITDA on an annual basis
• Spawned firms from VC backed firms are less related (in terms of patent overlap) than from non-VC backed firms
  • But less focused firms do less spawning
  • CA/MA firms (i.e. VC backed) appear to focus more on their core
• Entrepreneurial activity in a region appears to have increasing returns

Overall, a non-VC backed firm has a 33% chance of spawning, and a VC backed firm a 53% chance of spawning. Ultimate success may be bounded - when growth slows, spawning increases. As a firm's growth declines employees are more likely to leave and spawn. Also, entrepreneurs coming from public companies may have increased over time, or at least been stable.

Results are robust to choice of industry, technology, etc. So VC backed spawns aren't more likely because they feature VC relevant technology. Furthermore VC backed spawns are in a different line of business from their parent. Repeat entrepreneurs are also examined, and as a robustness check excluded - this has little effect on the results. Spawning doesn't appear to be caused by layoffs.

Problems

A few of the problems include:

• Limiting the patent control variable search to firms with two or more patents in heavily VC dominated industries.
• The industry segment(s) of operation measure is extremely weak.