And I would have gotten away with it too, if it weren’t for you meddling civil servants..

Rohan Silva, who worked in the Number 10 Policy Unit until recently, and is now “Entrepreneur in Residence for London VC Index Ventures” is intelligent, engaging and passionate about the potential for digitals startups to boost the UK economy. All of these are good things. However, I was very amused by this passage in an adulatory blog profile of him in the Wall Street Journal (pointed out by Stian Westlake of NESTA, who’s commissioned much of the key research in this area):

I remember the first economic speech the Prime Minister gave … I wrote in the final third about startups and disruptive innovation. I put in all these statistics which were all true, like 6% of high growth firms generate 60% of jobs. All that stuff I was arguing should skew government policy towards those companies. Whitehall kept taking the stats out, even though they were true. It didn’t feel right to them because these civil servants would never, ever encounter startups.

So what did the Prime Minister say in the end? In the speech Rohan is referring to, he said:

And today, many more of our jobs are dependent on new, young and dynamic businesses.  It’s astonishing to think that between 1980-2005, nearly all net job creation in the United States occurred in firms less than five years old. And here in Britain, just six per cent of UK businesses are high-growth but they generated over half of the net employment growth between 2005 and 2008.

Subsequently, he went even further:

Because if you look at where growth has come from in recent years, you see that it is the small, innovative companies that hold an immense amount of the potential.  Over and over again studies show that about one in 20 companies – the small, high-growth firms – are responsible for half of new job creation.  One in 20 firms, half of the new jobs.  So far from ignoring the start-ups – the insurgents if you like – we should be laying out the red carpet for them.

Rohan says these statistics are “all true”.  Really? The source for the UK statistics is this report published by NESTA.  It does indeed say that 6 percent of businesses are “high growth” (that is, firms where employment grew sharply in the relevant period), and that they generated over half of the net increase in employment between 2005 and 2008. But there are at least three separate issues with this figure.

1.  As the report states, “it’s not just about startups”:

It’s not just about start-ups Young firms are more likely to be high-growth, but the majority of high-growth firms (70 per cent) are at least five years old. Still, young high-growth firms are responsible for a fifth of the increase in employment in all high-growth firms.

One fifth of half is one tenth.  So, in other words, young high growth firms – successful startups – account for about a tenth of the net increase in employment. Not quite so dramatic.

2.  And it’s definitely not all about small firms. In fact, even if you look at firms which started the period with less than 50 employees (so not that small), the Prime Minister’s “small, high growth firms” accounted for about 15% of net new job creation, not the half he claimed in his second speech.  Big firms – those with more than 250 employees – accounted for about 25%.   The report has a rather different – much more accurate and hence less hyperbolic – take than Rohan and the Prime Minister, concluding:

small high-growth firms employing fewer than 50 employees are not unimportant.

3. And finally, of course, net employment growth is a fairly meaningless statistic, as I’ve frequently written before in the context of immigration. Companies don’t create “net” jobs; they create (and destroy) jobs all the time. The NESTA report shows that “high growth” companies account for half of the 2.4 million net increase in employment in existing businesses in the three year period 2005-08. But, as NIESR’s Bob Butcher recently showed, the UK private sector created about 4 million new jobs every single year during that period (while a somewhat smaller number were destroyed). It is of course gross job creation we should be thinking about at the company level and that should be the object of government policy towards high growth companies.  In this light, the Prime Minister’s statement that:

It’s astonishing to think that between 1980-2005, nearly all net job creation in the United States occurred in firms less than five years old.

is risible (although, to be fair, the original source is equally misleading). It’s a bit like saying all net job creation in that period was accounted for by people who were under 30 when they got their first job. The surprise would be if this were not the case. For  a reductio ad absurdum on this, see Declan Gaffney’s explanation of how disabled people “took all the jobs”.

To conclude, the statistic Rohan cited is indeed true, but, contrary to his claims, has rather little to do with startups or small firms.  All it really shows is that high (employment) growth firms (young or old, small or big) account for a large share of employment growth.  This is perhaps not exactly earth-shattering, nor is it likely to have come as news to civil servants. The idea that you would build a strategy for growth or job creation around this is absurd. 

As for what the Prime Minister, at Rohan’s prompting, actually said, his first speech was merely highly misleading, while his subsequent one was flat wrong.   The only justified criticism of civil servants here is that Number 10’s normally rigorous fact-checking procedures failed to stop this.