US dollar funding: an international perspective

Report prepared by a Working Group chaired by Sally Davies (Board of Governors of the Federal Reserve System) and Christopher Kent (Reserve Bank of Australia)

The US dollar dominates international finance as a funding and investment currency. Although the United States accounts for one quarter of global economic activity, around half of all cross-border bank loans and international debt securities are denominated in US dollars. Deep and liquid US dollar markets are attractive to non-US entities because they provide borrowers and lenders access to a large set of counterparties. The pre-eminence of the US dollar as the global reserve currency and in trade invoicing further motivates its international use.

The widespread use of a dominant currency for funding gives rise to a complex and geographically dispersed network of relationships. This has important implications for the resilience of the global financial system. Specifically, the central role of the US dollar in international finance means that global economic and financial activity is highly dependent on the ability of US dollar funding to flow smoothly and efficiently between users. This broad international use of the US dollar generates significant benefits to the global financial system. These benefits arise from economies of scale and network effects, which reduce the costs of transferring capital and risks around the financial system. But it can also lead to vulnerabilities, as the resulting interconnectedness can transmit and amplify shocks that emanate from the United States or elsewhere in US dollar funding markets, across the globe.

This report seeks to understand better the role of US dollar funding in the global financial system by: (i) taking stock of its structure and evolution since the global financial crisis (GFC); (ii) assessing its resilience and highlighting its potential vulnerabilities; and (iii) identifying implications for policy. The scope of the report is limited to US dollar borrowing, lending and intermediation by non-US entities with each other and with US entities. The bulk of the work reported here was concluded prior to the outbreak of Covid-19. The ensuing crisis validated many of the messages from the analysis, but at the same time had an important impact on US dollar funding activity. The final section of the report provides some pertinent, albeit necessarily preliminary, observations in this regard.

US dollar funding remains below its peak a decade ago relative to the size of the global economy, despite having grown in nominal terms. By contrast, the US dollar's share in international borrowing has reversed its pre-GFC downward trend to again reach levels seen in 2000. It is clearly the dominant international funding currency.

There have been major changes in the structure of the US dollar funding landscape since the GFC. For instance, there is less activity in Europe but more elsewhere, including in emerging market economies (EMEs). Less intermediation is now conducted by banks, and there has been an increase in market-based finance. As a result, non-banks have become more important providers and users of US dollar funding. These trends have been influenced by a number of factors, including (but not limited to) new regulatory reforms, the recovery and recapitalisation of weak banks, and shifting business models of intermediaries in many jurisdictions. Moreover, the robust performance of the United States and some EMEs in recent years and the resulting higher interest rates compared with many advanced economies (AEs) contributed to a shift of global portfolios towards US securities and cross-border lending into EMEs, much of which is in US dollars.

In some respects, vulnerabilities to the global financial system stemming from US dollar funding and the lending activities of non-US banks have declined. Banks have become more resilient to shocks, as they hold more capital, have larger liquidity buffers, manage risks more carefully and have reduced bilateral credit exposures. In addition, more borrowing is undertaken on a collateralised basis and more transactions are cleared through central counterparties. Bilateral liquidity swap lines between the Federal Reserve and other central banks provide a prudent liquidity backstop. As a result, key markets and institutions are better positioned to withstand shocks. Nevertheless, this does not imply that shocks cannot materialise or that volatility is a thing of the past, as illustrated by the Covid-19 crisis.

Indeed, some of the improvements since the GFC may have been offset, at least in part, by the rise in the US dollar activity of non-banks. On the one hand, institutions such as pension funds and insurers tend to have more stable regular funding sources and operate with less leverage than banks, which by itself would improve the resilience of international US dollar activity. However, certain vulnerabilities may have arisen because these institutions are playing a larger role in US dollar markets while having less recourse to a range of US dollar funding sources (including central bank facilities). The large footprint of non-bank financial institutions in some markets also suggests that should they experience distress, this could trigger fire sales that could amplify any market volatility. Finally, unhedged US dollar borrowers could face the simultaneous adverse realisations of exchange rate risk, interest rate risk and refinancing risk. The systemic-risk implications of these developments are hard to evaluate owing to the limited visibility of activities conducted by these players.

International US dollar funding activity remains large, and economies and sectors remain interconnected. As a result, there is the potential for the transmission of shocks with large effects on the global financial system and the global economy. The cost and availability of US dollar financing can shift as a result of changes in US interest rates, shifts in global risk sentiment or periods of market stress (eg the Covid-19-related stress). Cross-border and cross-sector linkages also often transcend regulatory jurisdictions, which complicates participants' and regulators' efforts to monitor and manage the risk of a retrenchment in cross-border liquidity.

Significant data gaps, in combination with the complexity and interconnectedness of the system, make assessing the evolving risks and vulnerabilities especially challenging. Reflecting this, a key policy message of this report is that authorities should seek to improve the transparency of global US dollar funding activities - for example, through additional data collection, greater data-sharing and improved disclosure. Most prominently, significant gaps remain around the increasingly important role played by some non-bank financial institutions and non-financial corporations, and around key markets such as the repo and FX swap markets. Moreover, the global nature of these markets necessitates taking a global approach to monitoring - no single jurisdiction has the whole picture.

The report identifies regulatory and structural policy options that could further reduce certain vulnerabilities. For example, in time, regulators of non-bank financial institutions could provide guidance on the inclusion of a currency dimension in their liquidity risk management. Some jurisdictions could consider policies aimed at deepening domestic capital markets. Further thought might be given to improving safety nets that can cushion the negative impact when US dollar-related risks crystallise - for example, through increased self-insurance or increased bilateral, regional or global liquidity support mechanisms, although all these options present governance and policy challenges.