Chart of the day: Do IPOs create jobs?
In the wake of its fabulous report about how investors in VC funds are stupid, the Kauffman foundation has released another report, this time about IPOs. This one comes with a very bad press release, which says in breathless fashion that “nearly 1.9 million new jobs forfeited in the past decade as fewer entrepreneurial firms join ranks of public companies”. In fact, the report itself is much less alarmist, and a single chart does a very good job of debunking the idea that if we had more IPOs, we’d automagically have much more employment.
What this chart shows is that during the dot-com bubble, companies like Amazon and eBay would go public and promptly reinvest the proceeds, using them to grow as fast as they could. Both of them were just three years old at IPO, and used their equity capital to carve out dominant positions in their respective corners of the internet. As the report says, the market’s mantra during the dot-com bubble was “grow rapidly or fail”, and so all companies which went to market adopted pretty much that strategy.
With hindsight, many of those companies would have been better off conserving their capital, preserving a bunch of liquidity for a rainy day, and going for sustainability over growth. But that didn’t happen, and what you can see in the chart is a spectacular failure of public companies to create jobs after the dot-com bubble burst. The older companies certainly didn’t: total employment in companies which went public in 1996, for instance, actually fell significantly between 2000 and 2003. And even newly-public companies, if they went public in 2001 onwards, basically gained no jobs at all; the only exception was 2004, the year Google and Salesforce went public.
So yes, the number of companies going public after the dot-com bubble burst was lower than the number of companies which went public during the bubble, when equity capital was dirt cheap. That doesn’t mean that jobs were forfeited as a result: if more companies had gone public, there’s no way they would have grown their payrolls at the rate that the cohorts of 1996 and 1997 did.
Indeed, the report itself explains very clearly that the 1.9 million number is not remotely something to be taken literally:
Since the number of years in which to grow would have been shorter than for the firms that went public in the late 1990s, the jobs created through 2010 probably would be lower. Second, there is an assumption that the average quality of firms going public would remain the same as those that actually did go public. In other words, that there would have been additional eBays, Amazon.coms, and Googles if there had just been more IPOs. Third, that the people that would have been hired would not have been doing something else. In other words, there is an implicit assumption that a mass army of would-be engineers, scientists, and marketing experts is sitting at home watching television. And fourth, that the capital invested when a company raises funds in an IPO would not otherwise have been invested in job-creating activities. The average company that conducted an IPO during our sample period raised $162 million in inflation-adjusted dollars, and if there were 2,288 more IPOs of the same average size, $370 billion of capital would have been pulled from other uses.
Instead, the point of the 1.9 million jobs number is that it’s low, not high: it’s being presented in order to contrast with insanely overinflated figures elsewhere, such as Grant Thornton report which says (slide 15) that the decrease in IPOs “may have cost the United States 22 million jobs over the last decade”.
In fact, what has happened over the past decade or so is that companies have been getting older and older at IPO, and have been able to raise, in some cases, billions of dollars in venture capital before going public. As such, IPOs have not been a way of raising growth capital, so much as a way of creating an exit for VC funders. Or, to put it another way, there are still lots of hot 3-year-old technology companies raising huge amounts of equity and using it to hire loads of people. They’re just doing it in the private markets rather than the public markets.
What’s more, the fast-growing technology companies which are going public now, or which have gone public in recent years, are hiring precisely the one group of people where there’s no unemployment problem at all: computer engineers in general, and Silicon Valley computer engineers in particular.
Once upon a time, when IPOs were primarily ways for young, fast-growing companies to raise the capital they needed to continue to grow, there was a strong case to be made that they helped create jobs. Today, however, IPOs are something else entirely. If there were more IPOs, that might be a good thing, but it’s silly to believe that we’d have more jobs that way. IPOs, like leveraged buyouts, are financial tools used by financial professionals to make money. Those financial professionals surely like to think of themselves as job creators. But their real job is to make money, not jobs. And so while there are reasons to bemoan the lack of IPOs in recent years, this idea — that we’d have many more jobs right now if there had only been more IPOs — isn’t one of them.