Lean Entrepreneurship is Undermining the U.S. Economy – The 60% Challenge
According to Mike Maples of Floodgate, “$500,000 is the new $5,000,000.” The idea is that the growing opportunity to rent rather than build and own various assets, from servers to payment systems to inventory management, means that a new company can get a lot accomplished on much less money. Startup engineering methods, like customer discovery, take advantage of (and to some extent are predicated on) this opportunity.
Last Spring, a lot of bloggers were worrying about what this means for the traditional venture capital model, where successful VCs manage too much money to break up into $500,000 bits. In response, “super angels” like Maples, Chris Dixon, and Ron Conway have raised small funds and are investing relatively small amounts of money. Big traditional funds have reacted in different ways, setting up captive seed funds (which are turning out to be problematic) and/or putting more money to work in expensive, later-stage opportunities, like Kleiner Perkins’ greentech fund.
But forget the VCs, the real problem is the U.S. economy.
There’s an old saw that small businesses are the engine of job creation. This isn’t true in general, but it is true if you focus on young small businesses – startups. For example, a 2010 NBER study concluded that “firm births contribute substantially to both gross and net job creation.”
Other studies, like Tom Kane’s piece for the Kaufman Foundation, show that we’re talking about several million new jobs annually. The U.S. needs to create about 3.5 million jobs per year to get unemployment down to a 5% level in 5 years, and startups have historically contributed the lion’s share.
But look at this:
That engine has stalled in a big way. We’ve gone from startups generating 4.5 million jobs per year a decade ago, down to 2.5 million today, a drop of about 45%. Why?
Last March, the New York Times published an article titled “A Decline in American Entrepreneurship,” asserting that it was due to fewer new businesses. That’s not so. The chart below shows the number of companies established and how many are in business after 24 and 36 months
There’s a decline of about 20% in new business creation linked to the recession, but if you take the series back another twenty years, you can see that these numbers are within the historical range – highs of about 550,000 during boom times, lows of above 400,000 in recessions. How does a 20% cyclical decline in the number of startups translate into a 45% decline in new startup-related jobs?
Ask Mike “half a million dollars is the new five million” Maples. Startups these days can get to work for a lot less money – and, a lot fewer jobs. BLS data (calculated in another Kaufman foundation study) shows that from 1998 to 2010, the average number of employees at a startup has fallen from 7.7 to 4.9. And that looks like a long term, secular, not cyclical, decline. The problem for employment isn’t the cyclical 20% decline in startups, it’s the 36% decline in the number of people at each new startup. Lean startups, and the technology and service scaffolding that make them possible, are undermining the U.S. economy. Put another way, greater efficiency in startup creation is costing a lot of jobs. To make up for the loss, new startups can’t just be more efficient – they have to be more successful. To make up for the jobs they never create because they’re more efficient, startups need to avoid bankruptcy and growing faster. They need to be, roughly, 60% more successful.
This is the promise, and the challenge, for startup engineering. There’s not a lot of objective data, yet, on whether startup engineering really makes startups more successful. But it better. The stakes couldn’t be much higher.