Fernando Eguren Martin
Dollar shortages in funding markets outside the United States have been a recurrent feature of the last three major crises, including the turmoil associated with the ongoing Covid-19 pandemic. The Federal Reserve has responded by improving conditions and extending the reach of its network of central bank swap lines, with the aim of channelling US dollars to non-US financial systems. Despite the recurrence of this phenomena, little is known about the macroeconomic consequences of both dollar shortage shocks and central bank swap lines. In this post (and in an underlying Staff Working Paper) I provide some tentative answers.
Dollar shortages and policy response during the Covid-19 pandemic
The financial markets turmoil resulting from the Covid-19 pandemic included a marked increase in “offshore” US dollar funding costs, affecting non-US financial systems most directly. A widely followed measure of the cost of dollars “offshore” relative to US markets – so-called covered interest rate parity (CIP) deviations – rose sharply across currencies in mid-March (Figure 1). This dollar shortage in global banking is not a new phenomenon though: similar instances were observed during the global financial crisis and euro area crisis.
Figure 1: CIP deviations across currencies
Note: the figure displays CIP deviations between the US dollar and a range of currencies, computed using 3-month interest rates.
The policy response to this spike in US dollar funding costs outside the US included an expansion in the network of currency swap lines between the Federal Reserve and a range of central banks from advanced and emerging market economies. As part of this swap operation the Fed lends US dollars to a foreign central bank, receiving the other country’s currency as collateral. This swap is agreed to be reversed after a set period, and the Fed earns an interest on its loan.
In response to the Covid-19 crisis, the cost of swap lines was reduced for a set of countries with standing facilities with the Fed, and new facilities were set up with nine additional central banks. This policy response is also in line with those observed during the global financial crisis and the euro area crisis. Figure 2 shows the magnitude of swap lines drawings since March this year, and makes it clear that quantities are comparable to those observed during 2008-2009.
Figure 2: Drawings from Federal Reserve Swap Lines
Despite the recurrence of dollar shortage shocks and resulting expansions in Fed currency swap lines, relatively little is known about the effect of these phenomena on the macroeconomy (Akinci et al offer related insights to mine but do not consider the effect of swap lines quantitatively). In forthcoming work, I shed some light on that dimension.
The macroeconomic consequences of dollar shortage shocks
In a new research paper, I present a two-country medium-scale macroeconomic model featuring dollar shortage shocks and central bank swap lines. The two countries can be thought of as an emerging market economy (EM), in which banks (or borrowing corporates) display exchange rate mismatches in their balance sheets, and the United States, which central bank has the capacity to offer US dollar swap lines. The model is calibrated accordingly.
In the model, EM banks fund their (local currency) loans by borrowing in both local and foreign currency (US dollars). A dollar shortage shock is modelled as a sudden increase in the collateral banks need to post in order to borrow US dollars, unrelated to EM fundamentals (it could capture, for example, a deterioration in global sentiment). This shock leads to an increase in a measure of expected funding rate differentials – uncovered interest rate parity (UIP) deviations, which can be linked to an increase in CIP deviations, as shown in Figure 1 – and an exchange rate depreciation. The combination of the increase in the collateral required to borrow and the deterioration in mismatched balance sheets resulting from the exchange rate depreciation lead to a sharp contraction in bank lending. This in turn depresses investment, consumption and ultimately output.
Figure 3 shows the model dynamics described above in response to a shock calibrated to match the magnitude of observed CIP deviations between the US dollar and the Korean won before the announcement of a swap line between the Bank of Korea and the Federal Reserve. The model features UIP deviations but not related CIP deviations, absent an explicit market for FX derivatives needed to measure CIP. Korea is selected as the calibration target of the exercise as an EM that both suffered from dollar shortages and was granted access to a swap line with the Fed. It can be seen that the contractions in investment, consumption and output are quantitatively meaningful.
Figure 3: Model-implied macro effects of dollar shortage shock in Korea
Note: the figure shows impulse response functions of a series of endogenous variables of the main model in response to a “dollar shortage” shock calibrated to match the increase in CIP deviations in Korea during the Covid-19 pandemic.
Central bank swap lines to the rescue
The dynamics observed in Figure 3 do not incorporate the existence of a dollar swap line between Korea and the US. When allowing for such possibility, the negative consequences of dollar shortage shocks are reduced. In the model, the domestic central bank channels the US dollars obtained from its swap line with the Fed to the domestic banking system, which therefore demands less dollars in the market. This in turn reduces the pressure on the domestic exchange rate. The combination of a reduction in the type of borrowing subject to increased collateral constraints and a smaller hit to net worth resulting from a milder exchange rate depreciation lead to a smaller contraction in bank lending. The smaller contraction in lending leads to a smaller fall in investment, consumption and output.
Figure 4 compares the dynamics of macroeconomic variables under two scenarios: one in which swap lines are not available (as shown in Figure 3) and one in which they become available. Swap lines are calibrated to match the size of the line established between the Bank of Korea and the Federal Reserve. It can be seen that the presence of swap lines has a significant impact on macroeconomic variables: the negative impact of the dollar shortage shock is reduced by about 20%.
Figure 4: Macro effects of dollar shortage shock, with swap lines
Note: the figure shows impulse response functions of a series of endogenous variables of two versions of the main model in response to a “dollar shortage” shock calibrated to match the increase in CIP deviations in Korea during the Covid-19 pandemic. A baseline version does not consider the existence of swap lines (black solid lines), and an alternative version allows for swap lines, calibrated to match the size of the Fed-BoK arrangement (green dotted lines).
It is worth noting that the modelling of swap lines is highly stylised, and does not incorporate potential adverse effects such as the creation of moral hazard in EM banking systems. Also, the shocks discussed and modelled are restricted to the access of the financial sector to foreign currency funding, are unrelated to EM fundamentals and do not represent more general balance of payments stresses, which central bank swap lines are not designed to address.
In sum, the model presents a framework to think about the mechanisms at play and magnitudes involved in the interplay between dollar shortage shocks, central bank swap lines and the macroeconomy. It concludes that dollar shortage shocks can have sizeable negative effects on investment, consumption and output, and that these can be significantly reduced by the presence of central bank swap lines.
If you want to get in touch, please email us at firstname.lastname@example.org or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.