The words of the U.S. Treasury Secretary John Connally “It is our currency but it is your problem” originally used in a different context, can be paraphrased to summarise a subtle and perhaps a relatively small problem, but a problem nevertheless. In a new paper we investigate what, if any, is the impact of monetary policy on the tail risk of exchange rates. The paper focuses on nine advanced economies whose currencies represent around 85% of global foreign exchange turnover and covers the last 20 years.
Author:
Evarist Stoja
Professor of Finance
University of Bristol Business School
The Impact of Unconventional Monetary Policy
Anecdotal evidence suggests that unconventional monetary policy has a considerable impact on exchange rate tail risk. While at face value, this may be of little or esoteric interest, given the paramount importance and pervasiveness of the FX market in the global economy, it can have serious consequences on investors, financial institutions and economies. For example, one of the largest one-day depreciations of the Japanese Yen in recent years ensued the Bank of Japan’s announcement of an expansion of its asset purchase program which led to substantial turbulence in the market. Similarly, the de-pegging of the Swiss Franc from Euro by the Swiss National Bank in January 2015 gave rise to a tail event in the Franc/Euro exchange rate which in turn led to the bankruptcy of several financial firms with serious repercussions for financial stability. Yet another example is the sharp depreciation of the ‘Fragile Five’ (Brazil, India, Indonesia, Turkey, and South Africa) currencies in response to the U.S. Fed’s announcement on 22 May 2013 that it intended to start tapering asset purchases at some future date. The capital outflows that ensued deteriorated further the large current account deficits of these countries with serious repercussions for their economies.
The reasons underlying this relationship are complex. Extraordinary events since the beginning of the new millennium have resulted in a reduction of safe haven assets, elusive liquidity, sustained volatility in markets, and jumps in risk aversion, which in turn have led to increased turbulence and exacerbated fluctuations in asset prices. Moreover, with policy rates at the effective lower bound or, in some cases negative over sustained periods as Figure 1 below shows, the available headroom for central banks to respond was severely constrained.
In response, many central banks resorted to non-standard or unconventional policies to restore price stability when the bank base rate proved ineffective as Figure 2 below shows. These unconventional policies include Large Scale Asset Purchases also known as QE – the purchase of large quantities of financial assets, typically Government or other highly-rated bonds, Forward Guidance – announcements about the future path of short-term interest rates or Swap Lines – readiness to increase the supply of domestic currency to other central banks. These policies have been pursued at very large scales by most central banks in advanced economies.
Figure 1: The movement in the base rate across time and currencies. Note that the difference in rates across jurisdictions has become smaller since the Global Financial Crisis.
Figure 2: The frequency of policy measure implementation across time and currencies. The figure is a structured scatter plot so the intensity in colour represents the frequency a central bank has intervened with an unconventional monetary policy tool over a particular period.
Yet, precisely because central bank rates were at or close to zero for more than 10 years, currencies have been bearing an increasingly larger share of the adjustment burden from such policies. Indeed, there is increasing evidence to suggest that the impact of the monetary policy on the exchange rates has increased substantially over time.
Measuring Tail Risk
Studying the impact of monetary policy on exchange rate risk introduces additional levels of complexity because of the two-sided nature of the FX market and the multi-layered nature of risk. The impact of an intervention by a central bank on its domestic currency may be offset by similar interventions undertaken by other central banks on their own currencies. For example, the Bank of England and the Federal Reserve both have undertaken large and frequent conventional and unconventional monetary policy actions but it is not clear what net impact these actions have had on the Dollar/Pound exchange rate. Additionally, it may also be that these activities, over prolonged periods, give rise to complex dynamics, which in turn may lead to sustained volatility, turbulence and increase tail risk in the exchange rate.
The nature of risk is similarly complex. Classical finance argues that the disentangling of systematic from idiosyncratic risk is paramount as the latter can be diversified away but the former cannot, so systematic risk must be treated with care. This reasoning can be directly applied to tail risk. Intuition suggests that a monetary policy action taken by a central bank in isolation, may mainly contribute to the risk of its domestic currency, which would count as idiosyncratic risk since it would not affect the fundamentals of other currencies. If so, this would matter little to investors, institutions and economies with exposure to this currency because idiosyncratic risk can be diversified away. However, because these actions are often taken simultaneously or in close synchronisation (as illustrated in Figures 1 and 2) – whether implicit or otherwise – it may result in common variation across currencies and so impact their systematic component. Because systematic risk cannot be diversified away, it would increase the overall risk exposure of investors, institutions and economies relying on the FX market for investments and trade. In our paper, we empirically evaluate the impact of monetary policy decisions on the systematic component of the tail risk of currencies.
As a measure of tail risk we use the Value-at-Risk (VaR) which shows how much the investor is likely to lose with a given probability over a given horizon. VaR has been extensively embraced by regulators and practitioners in financial markets under the Basel II and III frameworks as the basis of risk measurement for the purpose of ensuring regulatory capital adequacy as well as risk management and strategic planning at industry level.
To conduct the empirical analysis, we use a number of econometric techniques ranging from simple linear regressions to the more technically-sophisticated such as instrumental variables in Panel Regressions and Bayesian Global VAR.
Findings
Our findings confirm the existence of a cross-border transmission channel of monetary policy through the tail risk of the FX market. This may have implications for portfolio allocations and capital flows, risk management and financial stability. Though arguably short-lived – the effect fades off after 1 month – the tail impact is particularly pronounced for some monetary policy instruments such as QE, Swap Lines and Forward Guidance. This cross-border source of tail risk is equally present for central banks that do not have an explicit exchange rate target, even after controlling for the U.S. dollar dominance and the effects of their own monetary policy stance.
More specifically, we find that both conventional and unconventional monetary policy tools have a substantial impact on the tail risk of currencies, particularly in its systematic component. This transmission is larger for measures such as Asset Purchase Programs and Swaps, and in particular since the Euro Area debt crisis. Moreover, the effects are stronger for countries that have heavily engaged in unconventional monetary policy, shedding new light on the (unintended) consequences of UMP on financial markets. The effects last for up to 1 month and are proportionally higher for joint QE actions. This suggests a reinforcement of monetary policy effects. Lastly, our empirical evidence confirms the central role played by Fed’s monetary policy.
Based on this analysis, an extrapolation may be ventured. Quantitative tightening – the reversal of QE currently high on the agenda of many central banks – will likely impact the tails of exchange rates but the aggregate effect will depend on its implementation. In particular, if QT is undertaken at similar times across central banks this may reduce the idiosyncratic component of the tail risk but will increase the systematic component that cannot be diversified by investors which in turn may result in market turbulence. The magnitude of any impact would increase as central banks whose currencies represent a large share of the FX market adopt such policy. However central banks can implement additional measures to tame the impact on the tail risk, for example, maintaining or increasing swap lines will help to control the tail risk. An alternative could be the spreading of the implementation of QT policies by central banks across time so they are asynchronous which should contribute to the reduction of the common variation and any ensuing turbulence in the markets.
Find out more about financial markets research at the University of Bristol Business School.