Today we’ve published new research looking at business dynamism and productivity growth before the crisis in Britain. Some may wonder what we can learn from looking at the past, when the problem that many are worrying about is Britain’s productivity growth slowdown after the recession. But the answer is simple. The analysis shows that not all was well and good prior to the recession. And that unless we address some of the deep-seated problems that it has identified, Britain’s productivity performance will suffer.

Four key findings emerge from this research:

1. Most startups were not particularly special from an economic point of view.

The average new British business was no more productive than the average existing business, either at its foundation or after five years of existence. While some new firms made a positive contribution to productivity growth, this was offset by the negative contribution of other new firms.

It is true that some entrepreneurs do contribute to British economic growth even if they don’t end up being successful. The ideas resulting from the experimentation that they undertake and the competition pressure that they create for incumbents can create productivity gains that prevail.

But this finding is a useful reminder of a known but yet often forgotten fact. Quality matters more than quantity when looking at entrepreneurship. Therefore, policy makers should focus on supporting high-impact entrepreneurship rather than on increasing the overall number of entrepreneurs, which are often low productivity businesses.

2. Many good companies went out of business.

A surprisingly high number of businesses that left the market were ones that showed above-average productivity. For every ten low productivity firms that left the market, six high-productivity ones did too. Had the high productivity exiting firms stayed in business, British productivity growth between 1998-2007 could have been up to 6 percentage points higher than it was.

Of course, it is to be expected that some high-productivity businesses will go bust even in the best of circumstances. But the sheer number of high-productivity firms leaving the market gives some cause for concern. We need to understand why this is happening and whether anything can be done to mitigate this trend.

3. Existing businesses were by far the main contributors to productivity growth.

Productivity improvements in the existing population of firms contributed an estimated 15 percentage points to growth in labour productivity, accounting for over 90 per cent of Britain’s productivity growth between 1998-2007.

The main driver was firms’ success at improving their own productivity performance, followed by the reallocation of resources from low to high productivity companies. But as we will discuss in a future blog, it appears this process was not fast enough. Moreover, on average firms growing fast in employment typically saw their productivity worsen, while firms with the fastest productivity growth were downsizing.

4. The British economy became significantly worse at allocating resources to the best businesses over the period.

This is complex but important. The research measured the so called ‘allocative efficiency’ of the economy over the period – the extent to which the best (most productive) companies were larger than the worst (least productive) in each sector of the economy. This gives an idea of whether the best businesses are scaling up.

The results show that allocative efficiency fell sharply in Britain from 1998 to 2007 among firms with ten or more employees. If Britain’s productive resources were as efficiently allocated at the end of the period as they had been at the beginning, productivity would had been 7.6 percentage points higher among firms of this size. Today this would be equivalent to around £79 billion of lost GDP.

In other words, in many sectors of the British economy the most productive firms struggled to grow, while large firms that had lost their productivity advantage managed to maintain their market share. Evidence from other sources suggests that this pattern has not improved since the financial crisis.

We'll discuss some potential explanations for this troubling finding in a future blog, but whatever the reason, this research suggests that economic policymakers should, as a matter of priority, seek to make it easier for high-productivity businesses to scale up. The impact on Britain’s productivity performance would be substantial.

To read more, see Nesta’s new publication “The Other Productivity Puzzle" and the underlying working paper “Sources of labour productivity at sector level in Britain, 1998-2007: a firm-level analysis” by Geoff Mason, Catherine Robinson and Chiara Rosazza Bondibene, Nesta Working Paper No. 14/09.

This blog post was originally published in October 2014 and has been ammended to incorporate revisions made to the underlying research in October 2015.