Commodities | Economics & Growth
How Commodity Prices Can Predict Banking Crises
- By Sam van de Schootbrugge
- Sign-up to discuss on Slack
- 5 min read
Summary
- A new Journal of International Economics paper explores the link between commodity prices and financial stability in low-income countries.
- It finds commodity price volatility is a significant predictor of banking crises, especially in fixed exchange rate economies and ones with a high share of primary goods in production.
- Financial instability is created through a reduction in government revenues and a shortening of sovereign debt maturity, which is likely to weaken banks’ balance sheets.
Introduction
Commodity prices are a key factor driving economic aggregates in developing countries, accounting for about one quarter of output fluctuations since the 1960s. A new Journal of International Economics paper investigates how commodity prices can also influence financial stability and so predict banking crises within low-income countries (LICs). They find:
- A one standard deviation increase in commodity price volatility is associated with a 2.5ppt increase in the probability of a banking crisis.
- Credit growth and net capital inflows, which past research identifies as leading indicators of financial instability, are found not to be predictors of banking crises.
- The effect of commodity price volatility on the probability of banking crises is concentrated in countries with a dominant primary sector of production and those with a fixed exchange rate.
Data and Methodology
The authors study 60 LICs over the period 1963-2015, amounting to 2,120 observations with an average time series of 35+ years per country. The majority are African countries, but the sample also includes Bangladesh, Cambodia and Myanmar.
The paper classifies a banking crisis as when there are either significant signs of financial distress (e.g., bank runs and/or bank liquidation) or significant banking policy intervention in response to substantial losses in the banking system. In total, 45 crises occurred in the sample – most in the 1980s and 1990s.
On commodity prices, the authors gather monthly data for 44 commodities from the IMF Primary Commodity Price Database. They combine this with annual information on the country-specific share of net exports of that commodity deflated by aggregate GDP for each primary commodity. The authors then construct a country-specific aggregate commodity price (ACP) index that adopts fixed country-specific commodity weights.
Lastly, the authors control for (i) macroeconomic fundamentals; (ii) external sector variables; (iii) monetary indicators; (iv) measures of banking system structure; (v) measures of global economic activity; and (vi) indicator variables for armed conflict, deposit guarantee schemes, debt crises, and currency crises.
In terms of methodology, the authors implement a random effects and fixed effects logit estimator to estimate the likelihood of banking crises across countries. It allows for the inclusion of countries that have never experienced a crisis (around half in the sample).
Why Commodity Prices Influence Financial Stability
In the literature, commodity price fluctuations can have real effects on a country’s output through a financial channel. Intuitively, a fall in commodity prices reduces a bank’s lending capacity, especially for commodity-exporting developing countries, because of higher non-performing loans (NPLs) and lower bank profitability. This leads to liquidity and solvency problems for exposed banks, which translates into banking crises.
A drop in commodity prices can also pressure the public sector, which could start running arrears to suppliers and contractors. This triggers second round effects on banks’ balance sheets. Also, a reduction in government revenues could incentivise the issuance of public debt and increase banks’ exposure to sovereign risk through government interference and moral suasion.
Commodity Price Volatility and Banking Crises
To begin, the authors show how episodes of declining commodity prices and increasing volatility overlap with LIC banking crises. Banking crises are clustered between the late 1980s and the early 1990s but are almost absent in the 2000s. In total, two thirds of banking crises happened when commodity price volatility was higher than its short-run average (Chart 1).
Source: Paper, page 14 and 15
They also find that commodity price growth increases significantly two years before the crisis year but does not persist. Commodity price volatility, however, starts increasing in the lead-up to the crisis and remains above the value even in later tranquil periods (Chart 2). This highlights the importance of volatility over commodity price growth in predicting banking crises.
Empirically, the authors find that a 1 SD increase in the annual volatility of commodity prices is associated with a 2.5ppt higher probability of a banking crisis. The share of short-term external public debt over total external debt is also a good predictor – a 1 SD increase comes with a 2.2ppt increase in the likelihood of a crisis. Banking crises are also more likely after periods of high inflation and high foreign aid inflows.
In terms of country characteristics, the authors find that commodity price volatility most affects countries in which primary goods production dominates. Countries with fixed exchange rates and hard pegs are also more susceptible to banking crises following episodes of commodity price volatility. And countries less open to trade are simply more likely to have banking crises, regardless of volatility.
Public Sectors Drive Boom-and-Bust Dynamics in LICs
The empirical results give no indication that credit growth or capital flows (change in net foreign assets) matter for the prediction of banking crises. Other research finds that excessive credit growth is a leading indicator of financial crises.
The authors rationalise the results by suggesting ‘the wave of de jure financial liberalisation that occurred in the 1980s and 1990s has not been accompanied by a de facto financial liberalisation’. And so, instead of the private sector borrowers and lenders driving boom-and-bust dynamics (as is the case in DM and EM countries), government-owned institutions and central actors are key features of systemic crises in several LICs.
On capital flows, a lack of private foreign investment means flows comprise mostly of official debt, which is generally directed at the government. The banking systems are less likely to intermediate these flows and so they do not fuel the boom-and-bust cycles generally seen in emerging markets. For these reasons, the typical drivers of EM banking crises are less prevalent in LICs.
Commodity Price Volatility and the Government’s Role
The paper finds higher commodity price volatility is associated with lower government revenues. Specifically, a 1 SD increase in commodity price volatility comes with about a 1ppt decline in government revenues. Lower revenues result in governments needing to take on more external public debt to meet large external financing needs. This shifts the composition of external public debt towards short-term debt.
The results align with past research showing that emerging and developing countries shift towards short-term borrowing during crises when uncertainty and informational asymmetries are larger. This hurts the banking system because banks then increase their holdings of government debt, further exposing their balance sheets to sovereign risk and raising the probability of a banking crisis. Indeed, the authors find some evidence that banking crises follow episodes of sovereign default (Chart 3).
Source: Paper, Page 30
Bottom Line
Commodity price volatility has increased to late-1980s levels and, at least for copper and oil, is at its highest in a decade (Chart 4). The paper shows that this elevated volatility increases the likelihood of financial distress among LICs. Given that the paper also shows banking crises are more prevalent after periods of increasing public debt, decreasing real GDP growth, decreasing reserves, and high inflation, LICs are set for a difficult post-pandemic recovery. VoxEU and IMFBlog recently covered this topic.
Citation
Eberhardt, M., Presbitero, A. F., (021), Commodity prices and banking crises, Journal of International Economics, (103474), https://www.sciencedirect.com/science/article/abs/pii/S0022199621000519#s0100
Sam van de Schootbrugge is a macro research economist taking a one year industrial break from his Ph.D. in Economics. He has 2 years of experience working in government and has an MPhil degree in Economic Research from the University of Cambridge. His research expertise are in international finance, macroeconomics and fiscal policy.
(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)