Bank of England: The banks that said no: banking relationships, credit supply and productivity in the United Kingdom

09 October 2015

This paper provides new evidence on the impact of the credit supply shock caused by the financial crisis of 2008 on corporate outcomes in the UK.

Authors find that firms facing a reduction in credit supply experienced greater falls in labour productivity and capital per worker. Authors argue that this is due to an increase in the shadow price of capital causing firms to substitute towards more labour-intensive technologies in production. Their results suggest that a 10% contraction in credit supply led, on average, to a 5-6% fall in capital per worker and a 5-8% in labour productivity. The estimated impact on labour productivity is large, suggesting lower capital investment may have been associated with lower levels of innovation and technological development. However further research is needed in this area.

Their results also suggest average pay fell further in firms more exposed to the credit shock, and in similar proportion to labour productivity, even though these firms were hiring labour in the same markets as less exposed firms. Authors find that a 10% fall in credit supply led, on average, to a 7-9% fall in average pay for the firms affected. This suggests firms were able to share the impact of the credit shock with workers, through lower wage growth.

Authors find that firms facing adverse credit supply shocks were more likely to fail. Their results predict that a 10% decrease in credit supply would increase the probability of bankruptcy by 60%.

These parameter estimates are both statistically significant and economically large. They suggest that the credit supply shock caused by the recent financial crisis may explain around 5-8 percentage points of the 17% shortfall in labour productivity relative to its pre-crisis trend by 2013, half of the shortfall in wages, and nearly half of the pickup in company liquidations between 2007 and 2009.

Their identification strategy relies on pre-crisis banking relationships that decay non-randomly over time. A key limitation of this empirical design is that authors are unable to say how persistent these effects might be. Although they are only able to identify the impact of the contraction in credit supply in 2008 and 2009, it may be that either the shock was more persistent or the effects on the corporate sector longer lasting.

On the intensive margin, it could be that firms are somehow permanently scarred by temporary credit shocks, or it could be that they are able to catch up to the counterfactual pre-crisis trend path once the credit shock abates. And on the extensive margin, this study does not tell us how firm entry is affected by credit supply, so authors cannot say what happens to the factors that become unemployed when firms fail, either during or after the period of crisis. Barnett et al. (2014) suggests impaired capital allocation between firms may have been an important driver of the weakness in productivity. There are also likely to have been factors, not covered in this study, helping to keep unproductive firms alive after the crisis, including forbearance by banks and the tax authorities, and low levels of interest rates (Barnett et al. (2014)).

Overall, the durability of the productivity slowdown since the crisis makes this an important avenue for future research.

Full working paper


© Bank of England