The collapse of Lehman Brothers in September 2008 was an unprecedented shock to banks’ funding opportunities. Banks transmitted this funding shock to their borrowers, but not necessarily in a homogeneous way. For example, several papers indicate that there was significant heterogeneity in the geographical reallocation decisions of banks (e.g. De Haas and Van Horen 2012, Giannetti and Laeven 2012). In this column, we investigate a different type of reallocation, by providing a comprehensive and detailed analysis of the reallocation that Belgian banks pursued across sectors and firms.
Banks operating in Belgium relied significantly on the interbank market for their funding. As can be seen in Figure 1, the total volume of interbank funding of banks active in Belgium was more than €500 billion in August 2008; a year later, more than half of that funding had dried up. This drying-up was significant for the banks and firms active in Belgium. The average firm in Belgium borrowed from a bank for which this interbank funding outflow represented 10% of its total assets.
Figure 1 Aggregate interbank funding of banks active in Belgium
To identify the reallocation in the supply of credit following these funding problems, we use 160,223 fully documented bank-firm combinations. We combine monthly bank-firm level data from a comprehensive credit register that contains all credit granted in Belgium by all financial institutions, monthly balance sheets of these financial institutions, and annual balance sheets of all registered firms. The richness of our data allows us to study various measures of credit growth and makes it possible to disentangle credit supply from demand. The latter is done by saturating the corresponding loan growth specifications with a comprehensive set of fixed effects in order to control for credit demand (Khwaja and Mian 2008, Jimenez et al. 2017).
Figure 2 summarises the main results of our study (De Jonghe et al. 2020). It illustrates the lending decisions made by a bank experiencing an interbank funding outflow equal to 10% of its assets. Our analysis shows that banks that face such a funding shock significantly reduce their credit supply to firms. Specifically, a 10% funding outflow leads on average to a reduction in credit growth of 4.14 percentage points. Given that the total amount of granted credit prior to the shock to all firms in the sample was €100 billion, and that the average firm in our sample was borrowing from a bank that faced a 10% interbank funding outflow, we estimate that the credit supply-induced ‘missing credit’ in Belgium in the year after the Lehman Brothers collapse was around €4 billion.
Figure 2 Heterogeneous impact of a 10% interbank funding outflow on credit growth
Besides the average effect, Figure 2 also plots the heterogeneous effect of a 10% interbank funding outflow. Specifically, the figure shows the effect for each combination of high/low bank sector market share and high/low bank sector specialisation, as well as one addition with high/low firm leverage. High (low) is defined as the mean plus (minus) one standard deviation. As the funding shock increases the marginal cost of lending, banks start to prefer their inframarginal borrowers over their marginal borrowers. We identify the marginal borrowers along three different lines. First, we show that banks are, on average, able to charge relatively higher interest rates to firms when they have a high market share in the sector in which the firm operates. Second, we show that relatively fewer firms go bankrupt in a banks’ sectoral portfolio if the bank is specialised in lending to that sector. Finally, we also consider several measures of firm risk.
Our analysis indicates that, compared to the average effect, the impact of the funding shock is 24% smaller for firms borrowing from banks that have a high market share and high specialisation in the firms' sector (i.e. an effect on credit growth of -3.13 percentage points). The opposite holds for firms borrowing from banks with a low market share and low specialisation; they experience a funding shock that is larger than average, with an effect on credit growth of -5.15 percentage points. Hence, banks direct their attention to sectors where they can more easily extract rents (higher sector market share) or where they have built up superior knowledge (higher sector specialisation).
The difference becomes even more pronounced when taking into account firm riskiness (in this example, proxied by leverage). The impact of the funding shock is more than twice as large for risky firms borrowing from banks that have a low market share and low specialization in the firms' sector (-5.62 percentage points) compared to safe firms borrowing from banks that have a high market share and high specialization in the firms' sector (-2.66 percentage points). This suggests that there is a flight-to-quality effect that co-exists with the two aforementioned reallocation effects. This is in line with findings of, for example, Liberti and Sturgess (2018), who show that borrower composition shifts toward larger and less risky firms during a credit supply shock.
On the real side, we find a moderate reduction in investments for firms borrowing from banks that were hit harder by the funding shock, but the reduction is less pronounced for firms when their bank has a high sector market share.
Overall, our results provide useful information for policymakers seeking to ensure access to finance for firms during crisis times, as we show that riskier firms and firms borrowing from banks that have low sector market shares and specialisation are more vulnerable to shocks in the banking sector. Related to this, firms may prefer matching with banks with a larger sector market share. While this implies a higher cost of borrowing, it also acts as an insurance premium that guarantees access to finance when the bank faces a funding shock. Our findings are also of interest to bank regulators, as they reveal a bright sight of lending concentration during crisis times.
De Haas, R and N Van Horen (2013), “Running for the exit? International bank lending during a financial crisis”, Review of Financial Studies 26(1): 244-285.
De Jonghe, O, H Dewachter, K Mulier, S Ongena, and G Schepens (2020), "Some borrowers are more equal than others: Bank funding shocks and credit reallocation", Review of Finance 24(1): 1-43.
Giannetti, M and L Laeven (2012), “The flight home effect: Evidence from the syndicated loan market during financial crises”, Journal of Financial Economics 104(1): 23-43.
Jimenez, G, S Ongena, J L Peydro, and J Saurina (2017), “Do demand or supply factors drive bank credit, in good and crisis times?”, Barcelona GSE Working Paper 966.
Khwaja, A and A Mian (2008), “Tracing the impact of bank liquidity shocks: Evidence from an emerging market”, American Economic Review 98(4): 1413-1442.
Liberti, J M and J Sturgess (2018), “The anatomy of a credit supply shock: evidence from an internal credit market”, Journal of Financial and Quantitative Analysis 53(2): 547-579.