Startups, incumbents and the UK's productivity puzzle
Productivity didn’t feature very prominently in Osborne’s budget speech just two weeks ago, but this doesn’t mean that UK’s productivity challenge has faded away. As the latest data released by the Office for National Statistics shows, UK productivity growth continues to disappoint.
UK labour productivity has stagnated since the financial crisis in 2007, and with it so has wage growth. Labour productivity levels only returned to their pre-recession peak at the end of 2014. This contrasts to average growth of just under 2 per cent per annum in the 10 years leading up to the financial crisis. Many people are asking: what explains this productivity puzzle?
New research by Nesta and the National Institute for Economic and Social Research sheds some light on this. The study analyses the sources of productivity growth in the UK over the period 1998-2013, using firm-level data from the Office for National Statistics (ONS) to understand where productivity came from. Specifically, it assesses the importance of productivity improvements within individual firms in different sectors and the distribution and re-distribution of labour across more and less productive firms in different sectors.
Six findings emerge from this research:
1) New firms created after the recession have dragged UK productivity down
The UK is not creating sufficient high-productivity startups to compensate for the larger number of new firms with low levels of productivity. The result is that, overall, new firms created after the recession made a negative contribution to productivity growth in the short term, reducing UK productivity levels by around 2 per cent between 2010 and 2013.
The low level of productivity displayed by recent new firms is potentially worrying. While it is normal for new firms to improve their productivity over time as they learn about the market and how to best serve it, the productivity performance of the latest cohort of new firms was below historical levels (particularly, but not only, in the services sector).
Historically, new firms had on average similar levels of productivity to that of the average firm in their sector in their first years of life, but in the post-recession period, the productivity of new firms was significantly lower.
This does not mean that every new firm is a drain on productivity nor that we should stop supporting new firms altogether, since averages hide the large dispersion in the productivity performance of new firms (as other research published by Nesta has shown). However, it is yet another reminder that, when it comes to new firms, quality matters more than quantity.
2) In the years since the recession the closure of underperforming firms has substantially increased UK productivity
The exit of low productivity firms has made a large positive contribution to UK productivity growth in the years after the recession, pushing UK productivity upwards by more than 3 per cent between 2010 and 2013.
This was not the case in the years of the recession, in which more ‘good companies’ went out of business and therefore the contribution of exit to productivity growth was lower (close to nothing between 2007 and 2010). This result highlights the importance of keeping barriers to exit low, but also the need to better understand the reasons that force highly productive companies to go out of business.
3) The growth of highly productive companies and the contraction of underperforming ones has slowed down since the recession
The contribution of resource reallocation among existing firms to UK productivity growth weakened after the recession (falling by around 2 percentage points between 2007-2010 and 2010-2013).
There are at least two potential explanations that could explain this finding (the research does not allow us to say which is the main one). One explanation is that good UK firms are growing and bad ones contracting more slowly than they did in the past. Another explanation is that the wrong type of companies are growing and contracting (e.g. low productivity firms scaling up and/or high productivity firms shrinking). Regardless of the explanation, the results suggest that barriers to the growth of high-productivity firms may be hampering UK productivity growth.
4) Overall there is little evidence that since 2007 the economy has become materially worse at reallocating resources across firms
One of the prominent and popular explanations for UK productivity weakness is the idea that the allocation of resources has become less efficient since the financial crisis. But the evidence for this being a major explanation for the productivity puzzle is scarce.
Taking together the combined effects on productivity growth of expanding and shrinking firms and firm births and deaths, these do appear a little weaker post 2010 than is usually the case. But, on average since 2007 are not very different to the pre-crisis period. In some sectors (such as Manufacturing and Retail) the contribution to productivity growth from the reallocation of resources across firms was less after 2007 than before, but in other sectors (such as Information & Communication and Construction) it was little different.
These findings do not categorically prove that the financial crisis did not affect the efficiency of resource reallocation. This is because we do not have a clear counterfactual against which to compare outcomes. For example, it is unclear whether we should have expected an increase in the contributions of resource reallocation associated with the economic downturn, or whether we should have seen a reduction in these contributions due to heightened uncertainty.
5) Slow productivity growth within firms is the main factor responsible for the UK productivity puzzle
UK firms experienced a decline in their productivity levels during the recession, something that is not unusual comparing to other recessions. However, what is unusual is that their productivity levels have not recovered more quickly in the post-recession years.
Indeed, the stagnant productivity growth within incumbent firms accounts for most of the difference between actual and pre-crisis trend productivity levels.
While differences in firm-level productivity growth explain most of the differences across sectors in sector-level productivity growth, firm-level productivity has stagnated across many different sectors. This suggests that the productivity puzzle is not driven by a particular sector and therefore that any explanation needs to consider economy-wide market factors.
6) A dynamic firm growth distribution is an important driver of UK productivity growth
The reallocation of resources among firms (via entry, expansion, contraction and exit of firms) adds around 3 to 4 per cent to productivity levels in the UK market sector every three years.
In fact, this may understate the importance of a dynamic firm growth distribution for productivity growth, since it is competition from growing high-productivity firms that forces other existing firms to improve their productivity levels.
Creating and maintaining the conditions that enable the continuous reallocation of labour and capital towards the most productive firms is therefore crucial to sustaining long term productivity growth.
Three policy recommendations
These findings suggest a number of implications for policymakers. At Nesta we would like to draw attention to three.
First of all, the potential waste of large amounts of taxpayers’ money when Government support is applied indiscriminately to all new firms or SMEs. Such support is likely to exacerbate the problem of low productivity firms entering the market or remaining in the market, dragging down average levels of productivity. A topical example of such a policy is the measure, announced in the 2016 Budget, for Small Business Rates Relief: a £7 billion tax cut applied to small businesses, regardless of their productivity or growth capacity.
Secondly, every year the Government spends £10 billion in different forms of business support. Not only should this money be devoted to productivity-enhancing measures, but it is important to make sure that it is actually having the intended effect. Despite recent initiatives such as the What Works Centre for Local Economic Growth or the Nesta-led Innovation Growth Lab, we still know very little about what works and what doesn’t. More experimentation with new support schemes and better evidence on their effectiveness should be an important priority, and there are several actions that the Government could take to achieve that (from setting up an experimentation fund for innovation and growth to making much better use of the data sitting in their own computers as well as new forms of online-based big data that have become available in recent years).
Thirdly, the Government should consider how to make it easier for high-productivity smaller firms to grow at the expense of less productive incumbents. One way to accomplish this would be by what might be called a Dynamic Enterprises and Markets Initiative (DEMI): a series of reviews, each led by a minister, of business regulation in an area where technology offers the potential for new entrants to disrupt established ways of doing business. Each review would work with new entrants to identify obstacles to new technologies and then work swiftly to remove them when appropriate.
A similar approach has already benefited high-growth disruptive sectors like the Sharing Economy and Alternative Finance (peer-to-peer lending and crowdfunding): in both these fields, careful work by the UK government, together with new companies, has resulted in the UK being one of the best places in the world for disruptive businesses to operate. This is good for productivity. The UK should make the most of this, using DEMI to expand the approach to new areas (examples include health data, unmanned aerial vehicles, or data-driven insurance).
Figure: Contribution to UK productivity growth
 The amount of output produced per person or per hour worked.
 Output per worker: Whole Economy, ONS mnemonic A4YM, 23 December 2015. The UK is not the only country to have experienced a weakening of labour productivity growth in recent years, see e.g. International Comparisons of Productivity – First Estimates: 2014, ONS, 18 September 2015.
 For the full results, including tables of the productivity performance of different British sectors 1998-2013, see Nesta Working Paper No. 16/01 and the tables released with it.
 UK productivity levels here refers to productivity levels in the UK market sector excluding agriculture, mining, utilities, real estate and finance.
 Productivity for new firms measured relative to the productivity of incumbent firms.
 Fixing the foundations: Creating a more prosperous nation. HMT. July 2015.
Photo: Will Heidelbach, CC by 2.0 www.flickr.com/photos/wilhei/109403306