Episodes of high financial stress have occurred both globally and in specific countries over the past six decades. This column introduces a new quarterly financial stress index for 110 countries, constructed using country reports from the Economist Intelligence Unit. This index has wider coverage than existing measures and can also capture the severity of financial stress. Increases in financial stress are associated with sizeable and persistent reductions in the level of output. These effects are stronger in emerging markets compared with advanced economies and in cases where the initial level of financial stress is already high.
In this column, we present a new quarterly index of financial stress – the Financial Stress Index (FSI) – for 110 countries (Ahir et al. 2023). Using the FSI, we examine the impact of financial stress on the economy and how it varies across countries, firms, and depending on the initial level of financial stress.
Constructing the FSI
We follow the work of Bernanke (1983) and Romer and Romer (2017) and define financial stress as an increase in the cost of credit intermediation or disruptions to the credit supply. We then use country reports of the Economist Intelligence Unit (EIU) to identify episodes of financial stress and measure their intensity. We follow a three-step process to construct the index. First, we search the EIU reports for words associated with our definition of financial stress and identify paragraphs in which two sets of keywords appear: (i) “credit”, “financial”, “bank”, “lending”, and “fund”; and (ii) “crisis”, “crunch”, “squeeze”, “bailout”, “rescue”, “tight”, “contract”, and “reluctant”. In the second step, we read these paragraphs to confirm that the text is indeed describing developments associated with contemporaneous financial stress. In the third step, we sum the words associated with financial stress and scale this raw count by the total number of words in each report.
New index versus existing measures
This is the first effort to create a continuous financial stress index for a large set of advanced and developing economies. Our work builds on the pioneering work of Romer and Romer (2017), whi develop a semi-annual continuous measure of financial stress for 30 countries for the period of 1967 to 2017 using the OECD Economic Outlook. We make two important contributions to their work. First, instead of the OECD reports, we rely on the EIU country reports which allows us to extend the country coverage to 110 countries and the frequency from semi-annual to quarterly over the period of 1967-2023. Two factors help improve the comparability of the FSI across countries: the FSI is based on a single source (the EIU reports), and these reports follow a standardised process and structure (Ahir et al. 2018).
Our second contribution is to provide an exogenous measure of financial stress. To this purpose, following the same approach of Peek and Rosengren (2000), we carefully examine the narrative in the EIU reports and identify episodes of financial stress stemming from financial stress in other countries.
Our work complements existing chronologies of financial crises (e.g. Reinhart and Rogoff 2009, Laeven and Valencia 2020) – we show that increases in our measure of financial stress typically precedes the occurrence of a financial crisis – as well as studies using model-based approaches to identify financial stress episodes (Klaus et al. 2016).
A seismograph for financial stress
Our global financial stress index (Figure 1) rose during the Latin America debt crisis in the 1980s (often known as The Lost Decade), the Mexican Peso Crisis in the mid-1990s, the financial crisis in Asia, Russia, and Latin America – also coinciding with the long-term capital management episode in the late 1990s – and then rose sharply during the Global Crisis and Europe’s sovereign debt crisis between 2008-2013. The index then remained broadly flat except for the short-lived increase during the Covid-19 crisis (Mamayski 2020).
Figure 1 Financial Stress Index (FSI)
Notes: The Financial Stress Index (FSI) is summing the number of keywords identified with financial stress in EIU country reports. The index is then normalised by total number of words. A higher number means higher financial stress and vice versa. The figure above is an average across 110 countries and covers 1967 to 2018 at a quarterly frequency.
The aftermath of financial stress
Our research suggests that increases in financial stress are associated with sizeable and persistent reductions in the level of output, and transitory ones in the growth rate of the economy. We find that a one standard deviation increase in financial stress index is associated with a reduction in the level of output by 0.35% one year after the increase in financial stress and by 0.2% five years after (Figure 2).
Figure 2 Average impact of a change in the Financial Stress Index
(Evolution of log output following a one-standard deviation increase in the Financial Stress Index)
Notes: The graph shows the dynamic response and shaded areas denote 90% confidence bands. Time is indicated on the x-axis. Estimates are obtained using a sample of 49 countries over the period 1996q1-2018q4. For details, see notes to Figure 6 in Ahir et al. (2023).
To put this in perspective, the results suggest that the peak increase in financial stress seen in the US during the Global Crisis would have been associated with a reduction in US GDP by about 6% in 2009.
Heterogeneity impacts
The effect of financial stress varies markedly across countries and time. First, the decline in GDP following an equal increase in financial stress is more than twice as large in emerging markets and developing economies (EMDEs) as in advanced economies (AEs) (shown in Figure 3.) This reflects differences in economic resilience, including the ability and space of fiscal and monetary policies to respond to financial stress.
Second, the economic effect of financial stress increases with the level of financial stress. In other words, the same increase in financial stress has larger effects when the initial level of financial stress is already high, while small increases in financial stress have typically negligible macroeconomic effects.
Figure 3 Variation in the impact of a change in the Financial Stress Index
(Evolution of log output following a one-standard deviation increase in the Financial Stress Index)
Notes: The graph shows the response and shaded areas denote 90% confidence bands. Time is indicated on the x-axis. Estimates are obtained using a sample of 49 countries over the period 1996q1-2018q4. For details, see notes to Figure 7 in Ahir et al. (2023).
Besides aggregate impacts, we find that financial stress triggers persistent declines in firm investment, with the effect being larger on those firms that are characterised by lower profits, lower return on assets, and more financial and debt burdens.
Source : VOXeu
