In last week’s post, we looked at balance sheets of broker-dealers and how the notion of ‘marking-to-market’ encouraged a procyclicality of leverage. A paper released by the IMF last month has advanced this work and looked at the link between balance sheet strength and lending during the global financial crisis. This follows the recent theme of this blog, where we have been looking at financial market imperfections and frictions that have an impact on the real economy. Kapan and Minoiu (hereafter K&M) have used an empirical approach to show that both a bank’s capital and its exposure to market funding via its structural liquidity are important conduits for the transmissions of shocks.
In trying to understand banks’ ability to sustain lending during a financial crisis, K&M found three important results:
- Banks that relied more on market based funding reduced their supply of lending more than other banks. By relying more on the market for funding, these banks were more vulnerable to liquidity shocks during the crisis. This result supports the findings of last week’s post.
- This effect was influenced by the quantity and quality of bank capital. Better capitalised banks reduced their lending less than other banks with the same exposure to shocks.
- Importantly, there were complementarities between bank capital and its structural liquidity. In other words, higher structural liquidity helped to maintain lending only for those banks that were well-capitalised.
How they did it
K&M analysed the lending behaviour of more than 800 financial institutions. Their data set covered 55 countries from 2006 to 2010, allowing the distinction of ‘before’, ‘during’ and ‘after’ the global financial crisis. This distinction allows the authors to capture the variation in balance sheet quality before the crisis and document its effect on lending. Specifically, K&M regress the change in loan volume between the ‘before’ and ‘after’ periods on a number of balance sheet indicators.
An important empirical challenge in this research was the ability to separate the supply of credit from the demand for credit, as both can be affected by an aggregate demand shock. Cleverly, K&M exploit a notion referred to as ‘within-borrower’ variation. To do this, the authors kept only those borrowers in the sample that had borrowed from the same lender in both the before and after periods for at least two lenders. If we can assume that each borrower reduced their demand proportionately vis-à-vis its lenders, this technique ensured that demand effects were removed from the analysis.
The empirical model employed by K&M was as follows:
Where delta C is the log-change in syndicated bank credit extended by bank i in country j to bank k, L is a proxy for the bank-specific liquidity shock, captured by the wholesale funding and liquidity measures, eta is the fixed effects of the borrower and delta refers to bank nationality effects, such as exchange rate movements.
Definitions and clarifications of key concepts
K&M have made a highly useful contribution to this area by examining different aspects of bank balance sheets and their relationship with the bank lending channel. As this is a complicated area, it is worth going through these balance sheet indicators to put the results of the study into perspective.
Firstly, to measure banks funding structure, K&M look at the Net Stable Funding Ratio (NSFR). Defined in Basel III, the NFSR is a long term liquidity requirement that considers both the asset and liability side of the balance sheet. This ratio measures the stability of a bank’s funding sources relative to the liquidity profile of its assets. K&M show that the NSFR is negatively correlated with measures of reliance on wholesale funding.
Secondly, reliance on wholesale funding is captured by two variables: non-deposit liabilities as a share of total liabilities, and non-deposit funding as a share of total funding.
Finally, to assess another aspect of bank health, K&M focused on a bank’s capital base. K&M found that better capitalised banks were better able to maintain lending during the crisis. Possible explanations for this effect include the relative ease that well-capitalised banks could raise debt under stressful conditions compared to poorly capitalised banks. The key finding of this paper was the complementarities found between capital and structural liquidity – that is, a stable funding structure relative to a bank’s asset liquidity profile is only effective if the bank has a strong capital base.
Important results in the paper also related to the measurement of a bank’s capitalisation via three capital ratios. ‘Tangible common equity’ is the ratio of total shareholder equity – minus preferred stock, goodwill and intangible assets – to total assets minus intangible assets. This ratio is more in line with Basel III regulations, whereas the other two are calculated on Basel II: Regulatory Tier 1 and Total regulatory capital (Tier 1 + Tier 2), divided by risk weighted assets.
Implications for policy and theory
For policy, this important study suggests quite strongly that both funding measures and capital measures are crucial for bank lending sustainability. On the funding side, the NSFR is a valuable measure, while on the capital side, common tangible equity ratio has proved the most useful here.
For economic theory, particularly for bringing banking and financial frictions into macro models, these findings suggest that the supply of credit needs to be influenced by the interaction of both capital adequacy and leverage with funding liquidity concepts. One could possibly use heterogeneous banks along the lines of their NSFR and have endogenous leverage leading to the increased vulnerability to shocks.
These are very interesting areas for macro theory, the difficulty is, however, they are easier said than done.
 Kapan, T & Minoiu, C, (2013). Balance sheet strength and lending during the global financial crisis. IMF Working Paper.