A recent report by the Bank of Italy looks at why the various banks of Italy use derivatives. Specifically, the central bank of that country wanted to know: is it a matter of hedging? Or is it a matter of keeping a proprietary book?
Hedge fund managers and other pursuers of alpha will be interested in this paper to the extent that they find themselves on the opposite side of transactions with banks. It is often important to understand why your counterparty is in the transaction.
The short answer: banks chiefly use derivatives for hedging.
The authors of the paper are: Luigi Infante; Stefano Piermattei; Raffaele Santioni; and Bianca Sorvillo. They use the B of I’s quarterly supervisory data to look at that country’s banks’ use of derivatives through the period 2003-2017. The supervisory data includes information on: (a) the types of derivatives used; (b) notional amounts and relative fair values; (c) classifications in the banking or trading book. The “banking book” consists of all assets being held to maturity, while the “trading book” consists of assets and liabilities for which a trading intention exists.
Banks unsurprisingly use derivatives to hedge against possible moves in interest rates, and against counterparty default. At the end of 2017, interest rate derivatives accounted for a little less than 90% of the outstanding amounts of such instruments held by Italy’s banks. Within that family of derivatives, the interest rates swaps dominate.
The report also finds that banks holding more capital and with greater liquidity in their asset base rely less on derivatives than do those with smaller numbers in either respect. This indicates that capital and liquidity are substitutes for derivatives, which in turn answers the original question. Capital and liquidity are substitutes precisely for the hedging purpose of derivatives.
The report reviews some basics about the function and significance of banks: they are in the business of transforming short-term liabilities into long-term assets. When interest rates move, the mismatch in the time horizon between liabilities and assets becomes a matter of risk. This is precisely why banks need to hedge against interest rate moves.
The difference at any given time between the assets and the liabilities maturing within 12 months may be called the “short-term duration gap” or somewhat more concisely the maturity gap. In general, the maturity gap is positively associated with long positions, negatively associated with short.
Private Loans and Banking Groups
Another key point in understanding the role of derivatives in banks: some banks lend more to the private sector than others. Private sector loans are generally higher risk (a public sector default is a very rate, sometimes a world-historical, event). So, the more such loans a bank make, one might infer, the more hedging it must do. The Bank of Italy report says that banks belonging to banking groups make fewer private sector loans than do stand-alone banks. Despite this, membership in a banking group is positively correlated with the use of derivatives.
The reason: membership in a group allows a bank, even one that is comparatively small, to make use of economies of scale in banking. The exploitation of group-level resources—or the adequate level of activity and the specialized resources necessary to manage the contracts—frees them to participate more actively in the derivatives markets than stand-alone banks their size could. This effect overwhelms the impact of their relatively low reliance on private sector loans. Indeed, in Italy 80% of the market for derivatives is accounting for by the five largest banking groups.
But, to return to the question that set this study in motion: banks generally use derivatives to hedge much more than to trade. This conclusion is, the authors say, robust to different specifications. This remains their conclusion, for example, when they control for the presence of censored observations in the sample and for the cross-sectional dependence of bank activity.
Much of the previous academic work on this subject has been US-centered, and the authors note that they are widening the discussion by looking at the Italian interface of banks and derivatives. It is an appropriate widening, because the regulatory and industry structure in the US is in some important respects sui generis.