Our politicians are anxious. And not just because no-one has a decent poll lead or idea where Brexit will end up. No, some are finding the time to get anxious about other things too, including the state of capitalism in the UK. The angst isn’t limited to the left either, with Michael Gove becoming a prominent worrier about the “failure of our current model of capitalism to deliver the progress we all aspire to”.
These concerns are no surprise given we’re still emerging from the deepest financial crisis and squeeze on living standards in any of our lifetimes. But what exactly should we be worrying about? One prominent concern, prompted by the fast rise of huge tech firms and controversial takeovers, is that our economy has become ever more concentrated in the hands of a few big firms. Worrying about that is fair enough – greater concentration could lead to weaker competition, with other companies, consumers and workers losing out. But does that anxiety fit with what has actually happened?
While there is evidence that big firms have become more powerful in the US, to date there has been very little UK focused work. The last detailed whole-economy study of market concentration was published in the mid-2000s. But in new Resolution Foundation analysis, we find that concentration has increased quite substantially on some measures. Britain’s 100 biggest firms now account for nearly a quarter (23 per cent) of total revenue across the economy, up by a quarter since 2003-04.
We also analyse concentration at a lower level, looking across around 600 sub-sectors of the UK economy. We find that on average bigger firms account for a larger share of sub-sector revenue now than in the early 2000s – although the increase at the sub-sector level has been smaller. The five biggest firms within each sub-sector account for an average of 43 per cent of revenue, up from 39 per cent in 2003-04.
There are distinct phases behind these headline results of higher concentration: it rose gently in the 2000s, increased sharply during the financial crisis and fell back more recently. Whether more recent falls are a one off post-crisis effect or a trend remains to be seen.
As well as variation over time, there has been very significant variation across sectors. Amongst the biggest increases are general retail (i.e. our supermarkets) and the fast growing gambling sector, in which the five largest businesses accounted for 68 per cent of sub-sector revenue in 2003-04, but a notably larger 85 per cent in 2015-16.
We should care not just about product market concentration, but also about what has happened in the labour market. This matters because workers being more reliant on working for a few big firms might mean less ability to negotiate higher wages. Some have argued that this could be part of the explanation for recent weak pay growth. It might be expected that as product market concentration has increased, so too will labour market concentration as the biggest firms employ a larger proportion of workers. However the opposite has happened, with employees slightly less concentrated among the biggest employers now than in 2003-04. The average share of employees working for the top five firms in each sub-sector has fallen from 31 per cent then to 28.5 per cent in 2015-16.
This fall in labour market concentration undermines speculation that growing labour market power of a few firms could be part of the cause of recent weak wage growth in the UK. But it doesn’t mean we should be totally relaxed on the workers behalf. Firstly, that’s because labour market concentration is much higher for low-paid workers and in low-paying sectors such as retail. Secondly, it’s because of what might be driving this paradox of shifts in labour market concentration, particularly in the 2000s, being different to what we see in product markets.
Our work suggests that part of the reason for this is that as big firms have gained market share they have done so by making use of a relatively small amount of labour. More specifically, we find that big firms in sub-sectors that have become more concentrated have pulled away from their smaller rivals with relatively high growth in ‘turnover per employee’. It does not follow automatically that these firms have higher profit levels than other firms, although that would be consistent with other recent research from the IMF showing that margins have risen for the biggest firms in the UK by 20 per cent since 2003 and on so called ‘superstar firms’ in the US.
These findings raise the prospect that as a sector becomes more concentrated, either the labour share within that sub-sector may fall as revenue growth occurs in firms with relatively few employers or (if workers in the leading firms are able to secure higher wages than in smaller firms) wage inequality between firms may rise.
So what does this all mean for policy makers? Is their anxiety about our capitalism justified and, if so, what should we do about it?
First, yes a bit of anxiety is reasonable; product market concentration has risen across all measures.
Second, they should focus not just on the big tech firms that make the headlines. Google and Facebook matter, but so does the rest of the economy – the majority of which has become more concentrated since 2003-04.
Third, we should not put recent weak wage growth down to increased labour market concentration. But more work on the level of labour market concentration, and whether it is particular bad in some local areas, is still needed. In particular the lack of employment options for those in low-paying sectors is a concern.
Fourth, the paradox of falling labour market concentration could mean that it matters more than we thought not just what job you do, but what firm you do it for.
Overall, our research puts the claims about capitalism’s failings into context. Concentration has risen. Big firms matter more than they used to, but not in the sense of employing more people. The UK economy is complex, perhaps more than ever. So if you’re looking for an easy answer to the question of whether it’s working, beware of simple solutions.
This work contains statistical data from ONS which is Crown Copyright. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates.