Two weeks ago, we published new research on Britain’s productivity performance that has created some debate.

It showed that over 90 per cent of British productivity growth in the decade prior to the recession came from existing businesses, and that many high productivity businesses were going out business. It also found that the average new firm, despite the hype that surrounds startups, made a negligible contribution to British productivity growth.

This result surprised us. Prior studies have found a more positive contribution of entry (although they typically looked at new plants/establishments rather than only at new firms, and thus included new store openings by the likes of Tesco and Starbucks).

Moreover, while it is true that new firms go through a learning phase and cannot take advantage of economies of scale (and official statistics may also underestimate their productivity), the research also found that even young firms that were between 5 and 10 years old were not more productive on average than existing incumbents.

On the positive side, new firms (or some of them) not only create a disproportionate share of new jobs, but also introduce new ideas and increase competition in the marketplace, forcing incumbents to be more productive (an indirect channel that was not captured in the research).

And reassuringly, the analysis found that there is a minority of entrepreneurs which make a significant contribution to productivity growth, a useful reminder that quality matters more than quantity when promoting entrepreneurship.

The puzzle

But the result that really puzzled us was another one. In a nutshell, the UK economy could have been 100 billion larger if allocative efficiency had not fallen as it did (or to be more precise, 96 billion, which corresponds to 5 per cent of UK’s GDP or 7.4 per cent higher private sector productivity)*. This is a substantial figure, so it is useful to try to understand what it means and what may explain it.

In the UK, as in other economies, there is a large dispersion in the productivity of firms within narrowly defined sectors. The higher the correlation between firm size and firm productivity, the better for aggregate productivity.

In other words, given that different firms have different productivity levels, it is better for UK productivity when the largest firms are the most productive ones, rather than the other way around.

Figure 1: Allocative efficiency in Britain, 1998-2007 (Source: Nesta/NIESR/ONS)*

As the figure above shows, what we saw during the decade prior to the recession was a large decline in how efficiently resources were allocated in the British economy.

In other words, there continued to be large differences in the productivity of firms within each sector, but more people were working in lower productivity firms within the sector at end than at the beginning of the period (more people were also working in lower productivity sectors, but that’s another story).

Figure 2: Allocative efficiency and productivity (level and growth) by sector in Britain, 1998-2007 (Source: Nesta/NIESR/ONS)*

The research looked at 111 sectors in detail and, as the figure above shows, there were large differences across sectors, with allocative efficiency falling more in services than manufacturing on average (you can navigate this figure and see which sector is which here).

There was also a strong correlation between the change in allocative efficiency and productivity growth. Sectors with the largest decline in allocative efficiency were the ones with the lowest productivity growth during the period.

Our data pre-dates the recession, but more recent estimates suggest that the reallocation of resources has slowed down significantly in recent years, so the allocative efficiency of the British economy is likely to have worsen even more.

Not yet the full story

The obvious question that follows is: Why is it that more people were working in lower productivity firms within the sector? Was it because low productivity firms were growing at the expense of high productivity firms?

The answer is no. The figure below breaks down the contribution that 'continuing' firms (that is, excluding entering and exiting firms) made to Britain's productivity growth over the decade.

Specifically, it shows that resources were reallocating from low to high productivity firms, increasing British productivity by 15 per cent (productivity improvements within firms accounted for another 20 per cent, but overall productivity growth was smaller since firms with the fastest productivity growth were downsizing, while firms growing fast in employment saw their productivity worsen).

Figure 3: Decomposition of the productivity performance of continuing firms in Britain, 1998-2007 (Source: Nesta/NIESR/ONS)

Putting the pieces together

So here is the overall picture: more people ended up working in lower productivity firms within each sector, despite the fact that during the period people had been reallocating towards higher productivity firms in the sector.

What could explain this? Here is one hypothesis. Due to technology and globalisation, the speed of change in the UK economy may have accelerated so much in recent years that, even if the resources are continuously reallocating from low productivity firms to high productivity firms in the sector, this process may not be fast enough to take advantage of new opportunities.

In other words, high productivity firms may not be able to scale up sufficiently rapidly before their competitive advantage is lost. If this is correct, it means that, on average, high productivity firms in the sector can only sustain their advantage for a shorter period of time, before other firms successfully challenge them. The tale of Palm, Blackberry or Nokia would be consistent with that, but counterexamples also abound.

This is only one of several possible hypotheses, so if you have ideas, views or thoughts on what may be going on please add them below in the comments section.

What does this mean in practice?

Regardless of what the ultimate cause is, the evidence shows that there would be substantial productivity benefits of unlocking the growth of high productivity firms. Even if no firm were to improve its productivity, UK productivity would be much higher if the UK created the conditions for the most productive firms to scale up.

The challenge is how to do that. Sherry Coutu’s new report on scale-ups tackles precisely this question, including many recommendations that I hope the UK government will take on board to help British businesses to thrive.

Ultimately, my own view is that it requires action at 3 levels:

1. Developing more impactful support schemes, which requires more experimentation with new approaches and better learning. This is why Nesta has partnered with several organisations around the world to launch the Innovation Growth Lab, a new global laboratory for innovation and growth policy that develops and tests different approaches to accelerate innovation, entrepreneurship and growth.
2. A better functioning eco-system, since businesses don’t operate in a vacuum. Without good access to talent, knowledge, infrastructure and finance, it is difficult for businesses to grow. A new approach to industrial policy, with more active participation from businesses themselves, may help to address some of these gaps.
3. Improving the regulatory framework, making it easier for businesses to scale up across Europe (for instance, creating a new European single market for entrepreneurs) as well as enabling faster resource reallocation across firms. If we are moving towards a more dynamic business environment in which people switch jobs regularly, adopting the Scandinavian flexisecurity model (protecting individuals rather than jobs) or the Austrian labour regulation system (which allows portability of severance packages across companies) are alternatives worth considering.

These are all subjects that I’ll be writing more about in future posts.

@ABravoBiosca

*The allocative efficiency figure is based on the original research published in October 2014, which has been revised since then. To access the updated measures of allocative efficiency and their impact on productivity and GDP, see the revised working paper and research brief reissued in October 2015.