Changes in bank capital and lending: a microeconomic approach

During the period 1996-2007, in an environment characterised by relative macroeconomic stability, significant accumulation of savings at the global level and exceptionally loose financial conditions, bank lending to the non-financial private sector grew significantly in a large number of advanced economies, with the highest rates recorded in those experiencing property booms. The outbreak of the financial crisis and, in some cases, the house price correction triggered a deterioration in private-sector balance sheets and cut off flows of bank financing to households and firms. Against this background, there has been an intense debate about the relative contribution to changes in bank lending of supply-side factors (deriving from the weakness of banks’ solvency position, linked to the rise in default rates) and demand-side factors (associated with the sharp fall in activity and, in some cases, the need to reduce the level of private-sector debt). As regards the supply of credit, various studies have attempted to confirm the existence of a link between changes in a bank’s capital and its ability to lend. In the presence of perfect markets the level of capital of a financial institution would have no effect on the amount of financing granted to projects with an appropriate return-to-risk ratio. However, the existence of various financial frictions (such as those arising from information asymmetries between banks and their customers) means that the supply of credit may be limited when the level of capital of an institution is below certain levels. Specifically, in a recessionary environment, the existence of minimum capital requirements, rising default rates that eventually lead to a fall in capital and the greater difficulty of obtaining additional capital may, in principle, have a negative impact on the supply of credit. To be able to distinguish between the contributions of supply and demand-side factors to changes in lending, it is necessary to solve what in economics terminology is known as an identification problem. The impact of changes in banks’ capital on the amount of credit available cannot be inferred simply from changes over time in aggregate lending, owing to the existence of factors that simultaneously affect the level of banks’ capital and firms’ demand for credit. To resolve this problem, the literature has frequently resorted to microeconomic data, in order to compare the lending of banks with different amounts of capital. However, these comparisons only enable the effect of supply-side factors to be isolated on the assumption that all banks face the same demand behaviour, which is unlikely when financial institutions often specialise in different types of agents, whose solvency may change in different ways over the business cycle. Other studies have combined information on banks and firms or have analysed the impact of one-off changes in capital, linked to factors or events that are barely correlated with changes in the demand for credit. Following this latter approach, this article presents the results of a study that analyses the impact of changes in bank capital on lending to firms, by comparing financial institutions with different degrees of exposure to the real estate sector and facing different house price changes (depending on the provinces in which they operate). Following a brief review of the empirical literature that has analysed, using diverse methodological approaches, the impact of changes in banks’ solvency position on their supply of credit, the methodological approach used to isolate the effect of factors associated with the supply of credit is described in greater detail. Finally, the results of this approach, based on disaggregated data obtained from the balance sheets of Spanish banks between 1995 and 2009, are summarised.