This is a summary of "Who Bears Interest Rate Risk?" by Peter Hoffmann, Sam Langfield, Federico Pierobon, and Guillaume Vuillemey, published in the Review of Financial Studies.
The authors find that although European banks’ collective exposure to interest rate risk is smaller than conventionally believed, it is also mainly differentiated by country and (mortgage) loan contractual preferences. Their findings of sharp heterogeneity in the balance sheet risks of banks, however, are relevant to euro-area institutions and have direct implications for the conduct of monetary policy across the currency union.
What Is the Investment Issue?
The authors measure the impact of interest rate risk on banks to test the conventional wisdom that the banking sector suffers from rising interest rates. They study this impact in terms of net worth and income by first defining the determinants of interest rate risk exposure and their variations—that is, by country, asset type, or market power, for example—and then also factoring in the proportion of risk hedged by interest rate swaps. Their analysis suggests that the structural distribution of interest rate risk is more diverse than generally assumed.
How Did the Authors Conduct This Research?
The authors use a dataset of balance sheet exposures of most banks directly supervised by the European Central Bank (ECB) as of 31 December 2015. Repricing cash flows are obtained across 14 maturity buckets. For fixed-rate instruments, they use the difference between assets and liabilities maturing in each time bucket. For variable-rate instruments, they use the notional amount allotted to each maturity bucket corresponding to the nearest repricing date.
The authors also use transaction-level data restricted to interest rate swaps referenced to Eonia and Euribor, obtained from two trade repositories— DTCC Derivatives Repository Ltd. and Regis-TR—that collect and maintain records of transactions conducted in derivatives markets by European financial institutions. For each trade, the authors document the identity of both counterparties, the residual maturity, the underlying benchmark rate, and the fixed rate agreed on at trade execution.
Their data cover 104 of the 129 banks directly supervised by the ECB at the end of 2015 (i.e., 97% of the total assets of banks under ECB).
The authors also obtain annual data on banks’ interest income and expenses since the euro’s inception in 1999 from Orbis (formerly Bankscope), as well as data on loan composition and bank business model classification (i.e., distinguishing nine categories of business models). The official euro-area yield curve is used for computing the present values of cash flows from on-balance-sheet items.
Interest rate risk is assessed using three different measures based on a given change in interest rates:
- The change in the bank’s net interest margin (measuring up to one year of impact)
- The change in the bank’s net worth, defined as the present value of net repricing cash flows (measured across the entire maturity spectrum)
- A time-series regression model on the change in net interest margin
What Are the Findings and Implications for Investors and Investment Professionals?
The authors determine the impact of rising interest rates on banks’ profitability and net worth. Their findings are in contrast to conventional wisdom, which assumes that banks’ incomes and net worth are necessarily hurt by an increase in interest rates and vice versa.
The authors articulate their findings: “Banks’ exposure to interest rate risk is small on aggregate, but heterogeneous in the cross-section.” Half of the sample benefits from rising interest rates, and the authors call this finding the “reverse maturity transformation” for banks that “hold variable-rate assets funded by sticky sight deposits.” The authors reference several research papers showing that banks have considerable pricing power over retail deposits, converting them, essentially, into much longer liabilities to match their asset re-pricing profiles. Hedging interest rate risk through swaps, however, makes up only around one-fourth of the exposure, according to the authors. Interestingly, the resulting cross-sectional differences in interest rate risk are not determined by business model but rather are principally a function of regional variations in loan rate fixation preferences for mortgages.
The authors suggest their research has important implications for policymakers. For example, their simulations estimate that the sample banks would lose a total of €4.6 billion in the case of a 25-bp increase in interest rates. Cross-country disparities and varying industry and household residual exposures, however, imply that interest rate changes can have meaningful redistributive effects within the banking sector. Understanding these largely heterogeneous effects is essential for the successful transmission of monetary policy throughout the currency union.
In recent years, banks have been quite successful at hedging balance sheet risks through matching variable rate assets and liabilities and by using derivatives under strict supervisory oversight. Such may not be the case, however, for smaller or niche banks without significant market power to dictate deposit rates or effectively manage high intermediation costs. Therefore, it is imperative to understand the market position and business niche of the institution that may be targeted for investment.