The Local Stress Index: A New Tool to Summarize Local Market Conditions, and an Application to the South African Local Market

Tobias Adrian Financial Counsellor and Director, Monetary and Capital Markets Department International Monetary Fund

October 27, 2020

Thank you very much for your invitation to join you at this discussion of the fourth edition of the Absa “Africa Financial Markets Index.”

Since its launch in 2017, the AFMI has become an important reference point for policymakers, investors and analysts as they consider the development of financial markets across the continent. This year’s edition takes a deeper look than ever before into Africa’s financial markets, assessing conditions in key geographies across six key areas: market depth; access to foreign exchange; market transparency; taxation and the regulatory environment; macroeconomic opportunity; and the legality and enforceability of standard financial markets master agreements.

This is a turbulent time for the global economy — and especially for Africa, as the COVID-19 pandemic causes painful disruptions for many Emerging Markets. They were already under intense pressure before the pandemic, with currencies depreciating, growth slowing and interest rates rising. Many of them — particularly Low-Income Countries — now confront dire public-health challenges and stark social tensions, and some face the prospect of debt distress.

Financial markets in these economies have been continuously becoming more complex across different products. Since the Global Financial Crisis of a decade ago, the currency and interest-rate derivative markets have increased in volume and sophistication, while non-residents have become dominant players in select local-currency bond markets. This has created the need for policymakers to monitor the signals derived from different markets.

In Chapter 2 of the October 2020 edition of the IMF’s Global Financial Stability Report www.imf.org/GFSR — entitled, “Emerging and Frontier Market Economies: A Greater Set of Policy Options to Restore Stability” — we introduce our “Local Stress Index” (LSI) as a tool to address this need. The LSI is constructed from local currency market liquidity and stress indicators — such as bid-offer spreads, realized volatility, and other risk-premium measures for local currency bonds and exchange rates.

The usefulness of the LSI is most evident when we use it to look at the impact of the COVID-19 shock. It brings into sharp focus the rapid deterioration of financial conditions that was propagated by the shock, across risk assets. Policymakers in wealthier economies were able to deliver an unprecedented policy response, with far-reaching fiscal and monetary measures to restore market functioning — but policymakers in poorer economies had much less latitude for action.

The LSI confirms that the overall stress in EM local markets has been comparable to the impact of the Global Financial Crisis of 2008 and 2009. However, the pace of the deterioration of market conditions, as well as their subsequent normalization, was unprecedented. There is also a notable difference in terms of the behavior of currency and bond markets.

The stress in currency markets was less severe and conditions normalized more quickly, thanks to the rapid policy response by major central banks, including the swift establishment of dollar liquidity lines by the Federal Reserve. Structural changes in Emerging Market currency markets, such as decentralization and the shift to more electronic trading, may also have played an important role.

The stress in local currency bond markets, however, was larger than in any of recent crisis, including the shock of 2008 and 2009. Three significant factors may have contributed to that stress. First, the COVID-19 response was significantly more skewed toward fiscal policy, which naturally increased pressure through bond supply risks. Second, Emerging Market local bond markets have become much more integrated with the global financial system, and non-residents are often the largest group of investors in local-currency bond securities. Third, local market depth has not matched higher non-resident participation, with the domestic investor base often remaining underdeveloped. By contract to Advanced Economies, which have many insurance and pension funds, domestic banks are often the sole alternative liquidity providers in Emerging Market economies.

Developments in South Africa provide a useful case study. South African financial markets are among the most developed across Emerging Markets. Non-resident holdings of local government bonds stood around 35 percent of the total debt stock just before the pandemic. While a high degree of integration with global financial markets is often a positive factor, it can lead to increased volatility spillovers from other developed economies as well as from Emerging Markets. The COVID-19 shock has been particularly challenging for South Africa, because it entered the pandemic with weak economic growth as well as limited fiscal-policy space.

South Africa’s currency market experienced significant pandemic-related stress that reached a crescendo in early April. Realized and options-implied volatility jumped to levels not seen since the Global Financial Crisis. But the extent of the dollar liquidity shortage, as measured by the currency basis, was much less pronounced compared to the crisis of a decade ago, and the widening of bid-ask spreads was also limited.

South Africa’s domestic bond market, by contrast, experienced an unprecedented shock, when analyzed through the LSI measure. Even now, it has not yet fully normalized. Prior to the pandemic, investors were already concerned about the country’s fiscal developments and about the prospect of further credit downgrades. This seems evident, judging by elevated asset swap spreads as well as the steep bond yield curve. The global FCI shock coincided with a major credit downgrade, which, in turn, led to the exclusion of local bonds from the World Government Bond Index. As a result, the level of outflows, volatility and deterioration of liquidity (as was evident from the widening of bid-offer spreads) signaled severe distress in the bond market. While volatility has subsided and bid-offer spread have normalized, the risk premium in the bond market remains elevated, reflecting bond supply risk as well as absence of non-resident inflows.

The developments followed a similar pattern in other major Emerging Markets, including Brazil, India, Indonesia, Mexico, Poland and Turkey. The COVID-19 shock led to a “sudden stop” in local bond markets, which had been a reliable source of financing over the past decade. This factor coincided with a rapid growth in budget financing needs due to governments’ fiscal stimulus response. Moreover, a number of Emerging Markets saw the exhaustion of conventional monetary policy space, with policy rates reduced to all-time lows.

Given this unique set of circumstances, many Emerging Market economies introduced asset purchase programs (APPs) for the very first time. The decision to introduce APPs is likely to have been supported by the increased credibility of Emerging Market central banks’ inflation-targeting regimes, as well as by the positive experience of Advanced Economies with their APPs. The programs introduced in Emerging Markets varied in scope, scale and modality, but they were united by their primary objective of restoring market functioning at the height of pandemic-induced turbulence.

The analysis carried out by the October 2020 edition of the GFSR concludes that APPs had a positive impact on bond markets by lowering yields, improving market liquidity, and reducing the eventual, overall level of stress, as measured by the LSI. At the same time, APPs did not lead to currency depreciation or increased stress in currency markets, as is evident from the LSI. In the aggregate, countries that introduced APPs saw a swifter reduction of stress in local markets by comparison to countries without such programs. That said, further research needs to be carried out as the APPs progress and diverge on cross-country basis.

So, as I mentioned, the LSI has served as a useful indicator in identifying stressed conditions in the COVID-19 episode. We hope that it can become a yardstick for policymakers when they measure financial-market stress in the future. The LSI aims to provide a coherent summary of market condition across multiple liquidity, activity and risk-premia indicators — while also capturing contagion effects by considering the increase in correlations during crises.

This methodology was pioneered by the European Central Bank for the SovCiss index, but the IMF is the first to apply it to a broad set of Emerging Markets. The Fund will continue developing the LSI and will share the results in an upcoming IMF Working Paper, as well as through other avenues.

We believe that the “Local Stress Index” (LSI) will be a useful tool for policymakers as they consider various countries’ economic conditions. I’m pleased to share these ideas about the LSI with you today, and I invite you to continue exploring the reasoning within our latest edition of the GFSR — which offers useful ideas for policymakers worldwide, in Advanced Economies and Emerging Markets alike.

Thank you very much. Now I’d be pleased to discuss any questions you may have.