Exposure to Interest Rate Risk and the Transmission of Monetary Policy

Banks ... with a smaller disparity between the interest rate sensitivity of their assets and liabilities will not curtail their lending as much as those with larger disparities.

The income streams of most commercial banks are sensitive to interest rate risk: commercial banks fund their long-term, fixed-rate lending with short-term loans, so any hike in rates by the Federal Reserve System raises their cost of securing deposits, reduces their cash flow, and increases their leverage. Because banks usually try to keep their leverage constant, higher interest rates mean they have to issue more stock or reduce the growth of their lending -- but previous research has shown that banks have trouble raising equity in the short term, so they tend to reduce their lending instead. This reduction in lending represents an important channel by which monetary policy changes are transmitted to the real economy.

The extent to which a particular bank needs to curtail its lending to achieve a leverage target in the aftermath of an interest rate increase depends on the difference between the value of its short-term assets, which generate interest-rate sensitive income streams, and the value of its liabilities, which are similarly interest-rate dependent. Stronger banks -- those with a smaller disparity between the interest rate sensitivity of their assets and liabilities -- will not curtail their lending as much as those with larger disparities. For example, in the aftermath of a 100-basis-point increase in the Fed funds rate, a bank with a "gap" in the 25th percentile will lend about 1.6 percentage points more than a bank in the 75th percentile, according to Augustin Landier, David Sraer, and David Thesmar. In Banks' Exposure to Interest Rate Risk and the Transmission of Monetary Policy (NBER Working Paper No. 18857), they conclude that the income gap -- that is, the disparity between the interest-sensitive assets and liabilities -- significantly affects the lending channel.

Examining quarterly bank holding data from 1986 to 2011, the authors find that the income gap of U.S. institutions with more than $1 billion in assets has varied dramatically over time. In 1993, it averaged 22 percent; in 2009, the average gap was 5 percent. The income gap also varies among institutions: at the 25th percentile of large commercial banks, the income gap is zero; at the 75th percentile, it's 25 percent of total assets.

Banks that have an income gap could use interest-rate derivatives to hedge against the risks of a rate hike by the Fed, but they appear not to fully hedge their interest-rate exposure, according to this study. The difference in the effect of a 100-basis-point rise in interest rates on the quarterly earnings of a bank in the 75th percentile of the income gap distribution and the earnings of a bank in the 25th percentile is about 0.02 percent of total assets. Given that the average quarterly return on assets is 0.2 percent, that 0.02 percent difference is significant.

This study also documents that the income gap strongly predicts not only a bank's lending but also its earnings. A rate rise will have a smaller effect on trimming the growth of lending at a stronger institution than at a weaker one. The authors calculate that in the face of an increase of 100 basis points, a bank at the 25th percentile will reduce its lending by about 0.4 percentage points more than a bank at the 75th percentile.

--Laurent Belsie