Do derivatives make the world safer?

Guillaume Vuillemey

12 February 2015

a

A

The interest rate derivatives market is the largest market in the world, with an aggregate notional exposure of 563 trillion USD as of June 2014. Its fast growth over the past 15 years (shown in Figure 1) has raised concerns from policymakers. Currently, no theory provides guidance regarding the effect of the use of derivatives on other decisions by financial intermediaries.

In a recent paper, I develop a framework to show how hedging using interest rate derivatives affects:

  • Risk management in banking,
  • The response of bank lending (both to interest rate and real shocks), and
  • The occurrence of bank defaults.

Figure 1. The growth of derivatives markets.

What are interest rate derivatives?

Interest rate derivatives are contracts by which two parties commit to exchange future interest rate cash flows, computed as percentages of a given amount – the notional amount. The most popular of these contracts is the interest rate swap, which makes it possible to exchange a fixed rate against a floating rate until the maturity of the contract is reached. Derivative contracts have hedging properties: they make it possible to insure against some future realizations of the short rate, which would otherwise induce losses.

One reason why banks are active in the interest rate derivatives market is because most of the cash flows they receive (e.g. loans) or pay (e.g. interbank borrowing) are interest rates whose maturities do not match: they tend to ‘borrow short’ and ‘lend long’. As a consequence of maturity mismatch, changes in interest rates either increase or decrease a bank’s profitability and possibly induce default.

Derivatives and risk management

In my framework, hedging is motivated by the existence of financial constraints (as in Froot et al. 1993). Banks aim to manage internal funds so that they have sufficient resources at times profitable lending opportunities arise. A shortage of funds would imply turning to costly external financing sources. Banks optimally engage in risk management either by

  • Preserving debt capacity – i.e. by not borrowing up to their collateral constraint and instead keeping cash – or
  • Using derivatives to transfer resources to future states where large lending outlays will be optimal.

My framework features two risks faced by a bank, which give rise to two opposite motives for risk management.

  • On the liability side, the risk is that the cost of debt financing will be high precisely in states where lending opportunities will be large. This risk gives an incentive to transfer resources from future states where the short rate is low to states where it is high.
  • On the asset side, the risk is that for a given cost of debt financing, the bank will be unable to seize lending opportunities arising from a low short rate, as such states are typically associated with greater optimal lending. This risk gives an incentive to transfer resources from future states where the short rate is high to states where it is low.

From the existence of these two opposite forces – which I call respectively the ‘financing’ and the ‘investment’ motives for risk management, – it follows that both pay-fixed and pay-float swaps may be used for hedging. In previous discussions, the fact that banks use pay-float positions – i.e. they get exposed to interest rate spikes – was usually considered a puzzle or as evidence of speculation. I show it is also consistent with genuine hedging.

Derivatives and bank lending

Does lending react differently to shocks? Hedging using derivatives induces a ‘procyclical but asymmetric’ lending policy. Derivatives users are able to secure more funds for states where lending opportunities arise. They are thus better able to seize them, inducing some procyclicality.

As bad shocks hit, derivative users tend to cut lending to a lower extent. In such states, all banks aim to restore their lending capacity for future periods. Non-users of derivatives do so simply by cutting lending and keeping cash. Users of derivatives do so to a lesser extent, as they have an extra instrument to ensure that their lending capacity is sufficient in later periods. Using swaps, they can transfer resources precisely to future states where high lending will be optimal. They cut less lending, as empirical evidence also suggests (Purnanandam 2007).

Derivatives and default

Absent speculative motives, does hedging using derivatives reduce the occurrence of bank defaults? It can, but not necessarily.

When default is considered, swap contracts may either transfer funds from non-default states to default states, or the reverse. The fact that both pay-fixed and pay-float swaps can be used for hedging implies that both effects can occur. Because default states are associated with high realizations of the short rate, the use of pay-fixed swaps (which are valuable when the short-rate is high) provides additional resources precisely in such states, and consequently reduces the occurrence of bank defaults at the margin. If, on the contrary, pay-float swaps are used, defaults will occur more often.

Which effect dominates? This depends on the structural frictions that provide an incentive to hedge. If default costs are high, the incentive to avoid default is a first-order concern, and derivatives will be used to do so.

If instead, the costs associated with accessing external finance when high lending is optimal are high relative to default costs, banks use pay-float swaps more often. Doing so, they transfer resources to states where the short rate is low, as these states are associated with greater lending opportunities. While they will be better able to exploit profitable lending opportunities as they arise, they will also default more often when the realized short rate is high. It can thus be that – absent speculating motives – derivatives hedging increases the occurrence of defaults.

References

Froot, K A, D S Scharfstein, and J C Stein (1993), “Risk management: Coordinating corporate investments and financing policies”, Journal of Finance 5, 1629–1658.

Purnanandam, A (2007), “Interest rate derivatives at commercial banks: An empirical investigation”, Journal of Monetary Economics 54 (6), 1769–1808.

Vuillemey, G (2014), “Derivatives and Risk Management by Commercial Banks”, Job Market Paper. 

a

A

Topics:  Financial markets

Tags:  derivatives

PhD candidate in Economics, Sciences Po Paris

Events