Scott Shane is an academic based in the economics department at Case Western Reserve University who specializes in entrepreneurship. I got to know Scott when I was heading up the Center for Regional Issues at Case Western Reserve.
A friend recently sent me Scott's article -- Everything you know about entrepreneurship is wrong -- and asked me to comment.
My response:
Shane's views are grounded in basic Economics 101. The only problem is that few markets function this way. The hard categories he is suggesting are not that clear in the real world. Markets are not mechanisms. They function as open networks where various actors make investments based on risk/return calculations.
These calculations are based on imperfect information and made by people who are not the "rational actors" with whom economists populate their models.
Government investment in start-ups does not work well because governments do not do a good job structuring their investments. (Solyndra or the more recent bomb in Rhode Island are good examples.)
The challenge is to structure these investments with more rigor. This is very hard to do, and that's why I don't really recommend it. Few government employees have the skill to structure these investments.
Generally, I'm skeptical of any government program that does not structure investments in individual firms as co-investments. Formula loan guarantees are generally foolish. They push the risk off to government investors without any compensating upside. The risk profiles of individual investments are ignored. (In contrast, the auto bail-out, structured by some administration experts with deep backgrounds in corporate finance, seems to have worked out well.)
Shane is correct in saying that government programs become captured. Agency capture is an old problem that is very difficult to solve. Just consider the problems of writing sensible banking regulations, even in the wake of the JP Morgan-Chase bet gone bad.
To avoid this problem in regional economic development, local and state governments can forget about funding individual companies and instead focus on funding the civic infrastructure -- the networks -- needed by entrepreneurs. That's the economic gardening idea.
With economic gardening, local and state governments invest in a public good that does not generate sufficient private returns to induce private investment.
Government programs can be structured to respond progressively to market signals, even though they may be weak. The best example, perhaps, is the SBIR program, originally designed by Roland Tibbets at the National Science Foundation. In the absence of a structure to guide investments along the risk/return continuum, sunsetting these investments is a good idea as a default.
My guess is that Shane's contrarian views are grounded in a misinterpretation of government data. Shane's problem with interpreting small business data comes in his inability to distinguish among different types of small business start-ups. To be fair, government data is not organized this way. Yet, understanding the strategic dimension of small business development is critical to crafting sensible public policy. Not all small businesses are the same.
Strategically, a company that manufactures vinegar in Cody, NE is different from a family farm producing strawberries. (These two examples come from my trip to Nebraska this week.) The vinegar start-up serves a national market (a lot of good money); while the family farm serves only a local market (increasing the velocity of neutral money, mostly).
Shane doesn't understand these strategic differences, but they exist and they are important. Here's a paper I wrote over 25 years ago on the issue. More recently, here's a paper from the UK on these high growth companies.
What's missing from Shane's argument is that government investments in start-ups are peanuts, compared to the government incentives to move companies from here to there. A recent book estimates these incentives total $50B to $70B. These investments are wasteful, and they need to be restructured.
The Pew Foundation recently produced a report on the lack of accountability for these incentives. If you want a target for stupid investment, this is it. We would be far better off with a broader tax base, lower rates and a simpler code. (In contrast, I would be surprised if government investments in start-ups total more than $3 billion nationally.)
Finally, I find his argument about multipliers as just odd, and his point about Eaton is, well, nonsensical. To my mind, Shane lacks a strategic perspective on small business development, so he comes up with some pretty goofy hypotheticals.
The key policy issue is how do we identify, ex ante, the start-up firms likely to grow quickly. They are a small number, but they have a powerful impact. That's the challenge that economic gardening has set out to solve. It has nothing to do with supporting small firms as small firms.
