Did Core Deposits Hedge Loan-Commitment Risk during the Financial Crisis?
A financial crisis is generally a time of great stress for banks, firms, and individuals all at once. Firms and banks might be cut off from funding options that just before the crisis were considered stable. At the same time, individuals and firms might be forced to draw down their credit lines to deal with unemployment, slower sales, and other expenses. Under such conditions, banks can find it challenging to provide funding to stressed borrowers, because they are stressed themselves.
Besides providing liquidity through loans, banks provide loan commitments. These represent a promise to fund future credit demand by borrowers. A familiar example of a loan commitment is a credit card. Your credit limit is the amount the bank promises to fund when you make purchases. If you have a $1,000 limit and you have spent $250, then $250 will show up on a bank’s balance sheet as a loan and $750 will show up off the balance sheet as an unused loan commitment.
During a crisis, banks might experience an unusually large drawdown on these unused commitments. The exposure to the demand for liquidity can leave banks scrounging for cash to cover their commitments. A number of research papers claim that banks should be able to meet these demands because funds from depositors should simultaneously be flowing in, as investors, scared by the market turmoil, seek the safe haven of deposits. If this relationship does in fact exist, we would expect to see core deposits and unused commitments moving in opposite directions during a crisis. Core deposits include total transaction accounts, savings deposits, and time deposits of less than $100,000, and they are generally considered a stable source of funds for a bank’s lending base.
The relationship seems to hold in the aggregate during the most recent crisis. The aggregate amount of unused commitments declined, while total core deposits increased. But that might be misleading because the increase in deposits might be happening at banks with no decline in loan commitments. At the outset of the crisis, the level of unused commitments was significantly higher than the deposit base supporting these promises. As the crisis unfolded, the gap closed, and deposits overtook the level of unused commitments by the end of the crisis.
The decrease in unused commitments could have occurred for two main reasons: one, individuals and businesses made purchases and drew down their lines of credit; two, banks withdrew or reduced their previously extended lines of credit. Probably both factors were at play during the crisis, but we think the action we see is more due to drawdowns. It is generally hard for banks to back out of their promises, because they fear the consequences of loss of reputation. Firms would not buy these commitments in the future, unless they think it is very likely the bank would fulfill them in times of need.
The US banking industry is dominated by a few big bank holding companies, which usually behave differently than the rest of the pack, and it is no different here. The top ten bank holding companies hold the vast majority of unused loan commitments. In June of 2007, right as the crisis hit, the top ten held nearly 80 percent of unused loan commitments, while the top two had 43 percent.
The top ten bank holding companies’ total unused commitments are significantly greater than their total core deposits. At the beginning of 2007, the top ten banks had in aggregate 2.6 times as many dollars promised in unused commitments as they had sitting in core deposits. By the end of 2009, this ratio had dropped to 1.5.
Core deposits increased during the crisis period by 35 percent, or $445 billion. Unused loan commitments, on the other hand, fell by $1.29 trillion between 2007:Q4 and 2009:Q2.This decline seems to have significantly accelerated after the Lehman collapse in September 2008. If the decrease in unused commitments was due to drawdowns, then the increase in deposits would have cushioned the blow. On the other hand, it was not nearly enough to cover the entire change in commitments, leaving a whopping shortfall of $845 billion.
Banks below the top ten, conversely, live in an entirely different universe. Core deposits are much higher than their total unused commitments throughout. The increase in core deposits during the crisis seems to be significantly higher than the slight decline in unused commitments. At the beginning of 2007, banks below the top ten had in aggregate 1.8 times as many dollars sitting in core deposits as they had promised in unused commitments. By the end of 2009, this ratio was up to 2.5.
All in all, it seems that unused commitments experienced a significant decline in the top ten banks, with the increase in core deposits only slightly mitigating the demand for liquidity accompanying this decline. Banks below the top ten seem to have experienced little change in their unused commitments, with a significant increase in deposits. We conclude that some evidence in the crisis trends suggests that the increase in core deposits hedges to some degree the decline in unused commitments. But the aggregate picture is different from one that considers banks of different sizes, suggesting the actual situation is more complex.