When arm’s length is too far

Thorsten Beck, Hans Degryse, Ralph De Haas, Neeltje van Horen 25 July 2014

a

A

In the wake of the global financial crisis, policymakers’ attention has focused on lending to small and medium-sized enterprises (SMEs) as these were among the most affected firms when the credit cycle turned. In the US, president Obama signed the Small Business Jobs Act in 2010, which authorised the creation of the Small Business Lending Fund Programme to increase the availability of credit for small businesses. In the UK, policymakers have put a lot of pressure on banks to increase, or at least not reduce, lending to SMEs – often seen as the backbone of the economy.

But what is the best way for banks to reach out to SMEs? While the traditional literature has focused on ‘relationship lending’ as the prime lending tool for SMEs, recent evidence shows that transaction-based or arms-length lending – using hard information to screen firms and hard assets as collateral – can be more cost-effective (Berger and Udell 2006). It also allows larger and non-local banks to lend to SMEs (Beck et al. 2011).

In recent research (Beck et al. 2014), we gauge how these different lending techniques co-vary with firms’ financing constraints in good and bad times. Are certain banks better able to continue lending to small firms during a downturn? A recent paper by Bolton et al. (2013) suggests that the importance of different lending techniques might indeed vary over the business cycle. During normal times, relationship lenders tend to incur higher costs and therefore charge higher lending rates than transaction-based lenders. However, as relationship banks learn more about their borrowers over time, they can continue to lend at more favourable terms to profitable firms when a crisis hits. Central and eastern Europe is an ideal region to test this hypothesis as it experienced a pronounced credit cycle, with credit growing around 35% in 2005, and then dropping to negative growth in 2009 (Figure 1).

Figure 1. The credit cycle across emerging Europe

Note: This figure shows annual nominal credit growth (%) across emerging Europe over the period 2005-13. The bars and line indicate total and corporate credit growth, respectively. Growth rates are based on the difference in end-year credit stocks.
Source: CEIC.

New micro data on local banking markets

To identify the relationship between banks’ lending techniques and SMEs’ access to credit, we create a new and detailed cross-country dataset with locality level (towns and cities) information. Specifically, we combine:

  • Information on firms’ access to bank credit as taken from the Business Environment and Enterprise Performance Survey (BEEPS), conducted by the EBRD and the World Bank in 2005 and again in 2008/9.

We have data on over 7,000 firms in both 2005 and 2008/9 across 21 transition economies.

  • Unique information on the extent to which relationship lending is a dominating element for banks when dealing with SMEs.

We take this bank-level information from the Banking Environment and Performance Survey (BEPS II), undertaken by the EBRD and Tilburg University in 2011. BEPS II contains information based on detailed face-to-face interviews with bank CEOs. We have data available for almost 400 banks, representing 80% of all bank assets in the 21 sample countries.

  • Banks’ balance sheet information and ownership structure as taken from Bankscope and the Claessens and van Horen (2014) database on bank ownership.
  • Newly hand-collected information about the exact geographical location of the bank branches in our sample countries.

We combine these different datasets to construct:

  • An indicator of financing constraints, based on information on whether a firm has a loan (demand and unconstrained), has no loan and no need for one (no demand and thus unconstrained), or has no loan because its application was rejected or it was discouraged from applying (demand and constrained). This three-way classification allows us to control for demand-side effects when studying the effect of bank lending techniques on individual firms’ financing constraints.
  • The share of branches in each firm’s geographic proximity that belongs to banks which consider themselves relationship as opposed to transaction lenders.
  • A series of firm-level control variables, including size, ownership, age, and dummy variables capturing whether the firm is an exporter and whether its financial statements are audited. We also create locality-level variables that control for other characteristics of the local banking landscape, such as bank ownership, capitalisation, funding structure, and average distance to headquarters in the case of foreign banks.

It is important to control for demand-side effects as the share of firms indicating a need for external finance decreased from 70% in 2005 to 62% in 2008/9. At the same time, the share of constrained firms (rejected or discouraged) increased from 34 to 42%.

We document a high variation of relationship and transaction lenders across but also within countries (Figure 2). On average, half of the branches belong to banks that describe themselves as relationship lenders. Interestingly, there is no clear correlation between lending technique and bank ownership – the share of relationship lenders among foreign banks is even somewhat higher than among domestic banks.

Figure 2. Regional variation in relationship banking

Note: This heat map plots the geographical localities in our dataset. Each dot indicates a locality that contains at least one surveyed firm. Darker colors indicate a higher proportion of bank branches owned by relationship banks as opposed to transaction banks. Relationship banks are defined as banks whose CEO mentioned that relationship lending was a "Very important" technique when lending to SMEs.

Our findings

Relating firms’ financing constraints in 2005 and 2008/9 to the share of relationship lenders in their proximity yields the following findings:

  • While there is no relationship between firms’ access to bank finance and the dominance of relationship or transaction lenders in 2005 (during the height of the credit boom in many of the sample countries), firms’ access to credit suffered less in 2008/9 in localities dominated by relationship lenders. The economic effect is also large -- moving from a locality with 20% relationship lenders to one with 80% relationship lenders reduces the probability of being credit constrained in 2008/9 by 31 percentage points.
  • The mitigating effect of relationship lending on firms financing constraints is stronger for smaller, younger, and more opaque firms with less collateral to pledge in 2008/9. There is no such difference in 2005.
  • This credit constraint easing effect of relationship lenders is especially prominent in adverse macroeconomic environments, as gauged by regional or country-level GDP drops in 2008/9.

All the results are confirmed when we control for possible selection bias by explicitly controlling for the lower credit demand in 2008/9. In robustness tests, we also control for several possible sources of endogeneity. Specifically, we control for the possibility that banks might have located their branches in specific localities with more credit-worthy enterprises by limiting our regressions to the pre-2005 branch network. Similarly, we control for enterprise creation being endogenous to the location of relationship lenders by focusing on older companies. In both cases we confirm our findings.

Policy implications

Our results have important policy implications. While the recent literature has clearly pointed to the benefits of having diverse lending techniques within a banking system, relationship lending seems to have a more prominent role to play during economic downturns. During such periods SME lending tends to be particularly subdued, potentially delaying and weakening the subsequent phase of economic recovery (Chodorow-Reich 2014). The effect of a financial crisis on the real economy would therefore likely be smaller if more firms could be induced to seek a long-term banking relationship with financial institutions. Supporting the collection of the necessary ‘hard’ information about SMEs through credit registries and thus incentivising banks to invest more in generating ‘soft’ information themselves is another important policy message supported by our findings. Relatedly, our results also warn against an excessive short-term focus of banks and their shareholders on reducing costs by laying-off loan officers and other frontline staff. In the medium-term, and especially when an economic boom turns to bust, such cuts may negatively affect banks’ ability to continue to distinguish between firms with and without adequate growth prospects.

References

Beck, T, H Degryse, R De Haas and N Van Horen (2014), “When Arm’s Length is Too Far: Relationship Banking over the Business Cycle.” EBRD Working Paper 169.

Beck, T, A Demirgüc-Kunt, and M S Martinez Peria (2011), “Banking Financing for SME's: Evidence Across Countries and Bank Ownership Types”, Journal of Financial Services Research 39, 35-54.

Beck, T, V Ioannidou, and L Schäfer (2012), “Foreigners vs. Natives: Bank Lending Technologies and Loan Pricing”, CentER Discussion Paper No. 55, Tilburg University.

Berger, A and G Udell (2006), “A More Complete Conceptual Framework for SME Finance”, Journal of Banking & Finance 30(11), 2945-2966.

Bolton, P, X Freixas, L Gambacorta, and P E Mistrulli (2013), “Relationship and Transaction Lending in a Crisis”, Temi di Discussione 917, Bank of Italy.

Chodorow-Reich, G (2014), “The Employment Effects of Credit Market Disruptions: Firm-Level Evidence from the 2008-09 Financial Crisis”, Quarterly Journal of Economics 129(1), 1-59.

Claessens, S and N Van Horen (2014), “Foreign banks: Trends and Impact”, Journal of Money, Credit and Banking 46, 295-326.

a

A

Topics:  Financial markets

Tags:  bank lending, SMEs

Thorsten Beck

Professor of Banking and Finance, Cass Business School; Professor of Economics, Tilburg University; Research Fellow, CEPR

Ralph De Haas

Deputy Director of Research, EBRD

Professor of Finance at the Universities of Leuven and Tilburg, CEPR research fellow.

Neeltje van Horen

Senior Economist at the Research Department of the Dutch Central Bank