How will the debt crisis impact countries’ political stability?
By Sam Wilkin, Director of Political Risk Analytics, WTW
In early October 2023, Oxfam put out a press release with the headline, “world’s poorest countries to slash public spending by more than $220 billion in face of crushing debt.” This headline has alarming implications for political stability, considering that austerity and unrest have been shown to have a strong correlation. In general, despite dramatic examples in Ecuador and Sri Lanka, austerity appeared to play a relatively small role in global protest in 2022 and 2023, when nearly 80% of countries either held government spending constant or increased expenditures. 2024 appears likely to be a different story, as governments struggle to manage unsustainable debts. The current emerging market debt crisis could be compounded by social unrest.
Looking at the mechanisms by which austerity links to unrest, we might expect the current emerging markets debt crisis to result in more countries opting for less transparent short-term funding; significant risks that debt sustainability initiatives will prove self-defeating; and an elevated chance of ‘contagious protests’ spreading to multiple countries. An Index of people power and pressure for spending cuts suggests that countries already in default will face the highest risk, although Brazil appears to have a surprising potential for austerity-linked unrest.
This essay begins with a quiz. What do the following political events have in common?
The 2011 collapse of the Egyptian government during the Arab Spring
The 2018 gilets jaunes protests in France
The 2019-20 riots in Chile
The 2022 flight into exile of the president of Sri Lanka
One obvious answer might be that each incident involves mass political action. A less obvious answer would be that each incident involves policies of austerity (that is, net reductions in government spending, either through tax increases or spending cuts).
One trigger for the collapse of Sri Lanka’s government – following decades of rapid economic growth – was that the government-subsidized fuel price was allowed to appreciate. In Chile, a trigger for months of protests was the effective reduction in government subsidies for public transport (and thus, famously, higher subway fares in Santiago). In France, one spark for protests was provided by a proposed increase in fuel taxes. In Egypt, Arab Spring unrest was augmented by public opposition, expressed on the streets, over a government decision to allow the subsidized price of bread to rise.
The relationship between austerity and public protest is a well-known one. (Had we added explicit anti-austerity protests to the above list, such as those in Argentina, Greece or, more recently, Ecuador and Lebanon, we would surely have given the game away.) And yet, on a statistical basis, the relationship between austerity and unrest has been relatively little researched (although excellent case studies are abundant).
That omission is surprising when one considers that efforts to create models to explain the timing of political instability have thus far had limited success. Over decades of research, many factors have been shown to correlate with state fragility. For instance, anocracies (states in which power is contested but not via free and fair elections) appear to be more likely than either democracies or dictatorships to experience violent turmoil; countries with large natural resource exports appear to have longer, and perhaps more frequent, civil wars. Predicting the timing of instability appears to be more of a challenge, though, with efforts focusing on artificial intelligence early warning systems.
Austerity, however, has already been indicated to correlate with the timing of unrest, using relatively simple statistical approaches. In the next section, we will review this research and offer an update based on ACLED data.* We will then consider why austerity correlates with unrest, and look at some of the links between austerity and unrest provided by the Oxford Analytica contributors who authored profiles in this edition of the Willis Political Risk Index.
In early October 2023, Oxfam put out a press release with the headline, “world’s poorest countries to slash public spending by more than $220 billion in face of crushing debt.”Now seems an appropriate time to dig into the topic of austerity and unrest. This essay will conclude with an index of countries most at risk in 2024.
*ACLED provides day-by-day, geocoded political event data drawing on media sources, and has compiled records of more than a million events already.
Austerity and unrest in history: austerity vs. incidents of instability, 1919-2008
Source: “Austerity and Anarchy: Budget Cuts and Social Unrest in Europe, 1919-2008,” by Jacopo Ponticelli and Hans-Joachim Voth (2011); see also “Tightening Tensions: Fiscal Policy and Civil Unrest in Eleven South American Countries, 1937-1995” by Hans-Joachim Voth (2012)
One of the most comprehensive statistical studies of the relationship between austerity and unrest was published about a decade ago by Hans-Joachim Voth and Jacopo Ponticelli. In that study on Europe, and then in a subsequent piece on Latin America, the authors looked at a long time horizon and found a strong statistical relationship between spending cuts and the incidence of social unrest. In Latin America, there was a further relationship between austerity and military coups (the relationships for Europe are shown in the graph above).
In more recent years, there have been numerous studies of a special case of this phenomenon – the link between removal of subsidies on food or energy and social unrest. The link between these two areas of research is not perfect; in countries that are food or energy producers, price controls can have a similar political effect as explicit fiscal transfers. (Indeed, in many protests in the ACLED database demonstrators will demand that the government “fix prices” without specifying a mechanism.) This special case was addressed in the last edition of the Political Risk Index, which looked at cost of living crises.
Not the years of austerity: proportion of countries cutting federal budgets in 2022 and 2023
Source: IMF
Voth and Ponticelli’s research relied on a long-term database of New York Times stories. Using the ACLED database, which is more comprehensive, but only available for a shorter time period, we decided to see if this relationship between austerity and unrest still held in the post-pandemic years.
There were a number of complications to this research. Most importantly, 2022 and 2023 were not, by and large, years of austerity. Nearly 80 percent of countries either held spending stable or increased spending in 2022 and 2023 (see graph). As a result, our statistics on the link between austerity and unrest rely on a small number of austerity cases.
Moreover, some of the most well-known examples of austerity and unrest would not be picked up by the government spending measures we use (following Voth and Ponticelli). A tremendous wave of protests in France was sparked by a proposed increase in the retirement age, but in 2023 the French government was on balance holding spending roughly constant. Lebanon, meanwhile, is clawing its way back from sovereign default; the ratio of government spending to economic output has risen to an estimated 15 percent in 2023. That is still vastly below the pre-crisis ratio of about 30 percent, but much better than the shocking 2021 figure of 9 percent, and so both 2022 and 2023 count as fiscal expansion for Lebanon.
Austerity and unrest today: austerity vs. person-days expended on protests, 2022-23
Source: IMF, WTW analysis of ACLED database
Nonetheless, our simple graph of austerity and unrest was at least somewhat reminiscent of that produced by Voth and Ponticelli. Countries that did not impose cuts (such as France) had the largest number of estimated person-days expended on protest annually.* But in countries that made major spending cuts, including Azerbaijan, Chile, and Myanmar in 2022, and Burkina Faso, Sierra Leone, Ghana, Sri Lanka and Sudan in 2023, had nearly as high levels of protest intensity.
Of course, 2024 is likely to look quite different in terms of levels of austerity. The number of countries in debt distress has soared; high global interest rates are encouraging even solvent countries to cut spending (for more, see the accompanying essays, “Which countries are at high risk of debt distress?” and “Review of key risk rating changes in this edition of the Index”).
Are there warning signals that trouble is coming? So far, the indications are not too alarming. According to the ACLED database, large anti-IMF protests appear to have been limited to Argentina and Pakistan, and peaked in March of 2023 (with less frequent anti-IMF protests reported in Lebanon and Sri Lanka).
*Note that in this context, a “person-day” indicates not a 24-hour period but the participation of a person at a protest or riot event on a given calendar day.
Trends in Tensions: Fiscal Policy Protest and Riot Incidents by Month, 5/2022-8/2023
Source: WTW analysis of ACLED database; Credendo
We constructed a simple search for anti-austerity protests, including terms relating to fiscal policy such as “subsidies,” “taxes,” and “budget cuts.” A graph of the month-by-month frequency of such protest incidents in the ACLED database appears below. There is some indication that protest frequency is rising in frontier market countries and countries with a weak fiscal position (defined as countries assigned a long-term credit rating from Credendo of 5 or worse), while declining in the advanced economies.
Why is austerity associated with political unrest? Many of the studies that approach this question on a statistical basis have not delved into the reasons, referring instead to case examples. But considering the mechanism by which spending cuts translate into unrest may shed some light on the challenges that may arise during the current emerging markets debt crisis.
We can think of the composition of a country’s national budget as being determined by the outcome of contests amongst groups of citizens. In theory, the groups with the most political power will tend over time, via these contests for influence, to gain the largest share of government spending. Of course, things are not that simple in reality. Ideas matter; the political system mediates this contest; and better policy can make a larger pie for everyone in the future (to name a few objections). And yet pork barrel politics – for instance, subsidies for farm lobbies – will be familiar to most readers. So will the need of dictators to keep the army well-paid.
Which groups will have the most power within society? In part, that depends on the ability to solve so-called “collective action problems” – the ability to organize for effective political action. Groups that have clear common interests will tend to be better at acting together, as will groups that have a clear common identity. Other things being equal, smaller groups tend to be more effective at political action (which is one reason why, as countries industrialize and the agricultural sector shrinks, per capita subsidies for farmers tend, counterintuitively, to grow).*
In the previous edition of the Index, we compared measures of “people power” across countries. We found that countries where people had higher incomes, greater access to the internet, and greater literacy rates saw more person-days dedicated to mass unrest. The same idea very likely holds within countries. We might hope that government spending would go to those who need it most; but perhaps more often it goes to those who are best able to press their case.
For this reason, austerity can be politically risky. (Indeed, Voth and Ponticelli contend that governments implementing austerity are more afraid of triggering unrest than of punishment at the ballot box.) If a government is spending a lot of money on something frivolous, it may well be for good reason. Lavish expenditures probably reflect the political clout of the groups benefiting from them – and so cutting these programs could bring people on the streets. The inverse may be true of the tax system. Those who are not currently taxed may well be enjoying that status based on their effective lobbying, or their potential to stage an uprising.
If at least some of these claims about the links between austerity and unrest are correct, the current emerging markets debt crisis could pose a particular challenge to political stability, for several reasons.
First, in emerging markets, the most obvious targets of spending cuts may have the most power to respond. Over the years, the IMF has become increasingly sensitive to the dangers of triggering social unrest via the “conditions” it imposes on recipients of concessional lending. One of the IMF’s responses has been to shift its focus from cuts to reform – instead of cutting social spending entirely, the IMF attempts to replace expensive programs that largely benefit the rich with cheaper programs that largely benefit the poor. That approach certainly gives the IMF the moral high ground, and it is hard to imagine any multilateral institution taking the opposite tack.
On the other hand, this approach does not necessarily reduce the political risks associated with austerity. A recent example is the cuts to fuel subsidies in Sri Lanka. These subsidies did tend to benefit the relatively well-off (owners of cars and other motorized vehicles) – but as subsequent events demonstrated, these people also had a great ability to organize protests in response to cuts, and indeed to set off a chain reaction of protest that led to the collapse of the government.
One implication is that governments seeking to avoid the political risks associated with IMF conditions may be tempted to seek other, less transparent sources of financing that offer only a temporary solution and may make subsequent defaults messier (see the accompanying essay on “The Politics of Sovereign Debt Restructuring”).
*Olson, M., 1965. The Logic of Collective Action. Cambridge: Harvard University Press.
The Costs of Unrest: The Impact of Social Unrest on Economic Growth and Equity Returns
Source: The economics of social unrest
Second, in emerging markets, austerity may defeat its intended purpose. The IMF’s own research on unrest indicates that social unrest tends to undermine both economic growth and equity market performance, and that the impact on equity markets is larger in autocratic regimes (see graph). In contrast to the cost of living crisis reviewed in the last edition of the Index, the debt crisis looks most severe in frontier markets, where autocracies are more prevalent. Hence efforts to impose austerity and regain investor confidence may cause unrest that inadvertently undermines the objectives of the program.
Third, in the current debt crisis, there is a particularly high risk of contagion. One finding of research on social unrest tends to be that the best predictor of unrest is prior unrest. That finding may in some cases reflect omitted variables, but it also accords with theory. In general, one of the greatest challenges in organizing a protest is to convince people that their action can be effective. Prior protests can be a good demonstration of effectiveness, and prior protests can help convince people that if they take to the streets, others will join
them.* The same general idea can apply across countries, which may be why protests tend to be contagious (for instance, the Arab Spring). The current emerging market debt crisis is systemic in nature, in the sense that a strong U.S. dollar, high interest rates, and high commodity prices are imposing a burden on many countries simultaneously. This simultaneous burden could translate into simultaneous protests.
Bearing these ideas in mind, we now turn to an index of countries at risk in 2024.
If the above comments have convinced the reader that catastrophe is imminent, here is a silver lining: for better or worse, many of the countries impacted by the debt crisis have relatively little people power. In contrast to the cost of living crisis, which triggered large protests in some of the world’s richest countries, the debt crisis appears likely to impact emerging markets, and particularly frontier markets, where protests tend, on balance, to be more difficult to organize (that said, to assess the risk accurately, one would need to know the people power of the specific groups impacted by spending cuts).
In the Index that follows, we have attempted, at a country level at least, to balance the unstoppable force of spending cuts against the immovable object of people power (look to the profiles in each country for a more nuanced analysis). The countries at the top of the Index have the most alarming combination of educated and well-off people together with an urgent need to cut the budget.
In the profiles, contributors point to a few trends to watch. One is cost of living crises. While inflation rates appear to be slowing in advanced economies, prices of food and fuel continue to be a political issue in the emerging world. That is particularly true in countries in debt distress, where governments have little leeway to subsidize these staples. Examples include Zambia, Laos, Iran, Kenya, and Pakistan. (For a full Index of countries at risk from cost of living crises, see the previous edition of the Index.)
Another point is currency devaluation. Many countries have attempted to prop up the value of their currencies, in part to control inflation in the prices of imported goods. Some countries in IMF programs are now being forced to devalue, which may spark inflation and protests, including Egypt, Bangladesh, and Ethiopia. Other countries, such as Nigeria and Laos, have devalued voluntarily, but at some risk to stability.
Sanctions are also playing an outsized, although contradictory, role in this debt crisis. Sanctions have forced Russia and Mali into default. At the same time, sanctions have prevented Myanmar and Belarus from building up foreign debt, but saddled Iraq with large debts that are a legacy of the 1990s.
A final point is conflict with lenders. In a populist age, governments from Egypt to Argentina to Tunisia are failing to meet IMF conditions, and in some case attempting to justify their failures by presenting themselves as popular champions against neo-colonialism. This trend is one to be watched. As the contributor covering the Republic of Congo (Congo-Brazzaville) reminds us, that country has yet to successfully exit an IMF program, and this failure has been detrimental to the country’s development.
Below, we present the top 15 countries scoring worst on our measures of austerity and unrest (which is applied only to countries covered in this Index). Most of those countries appearing have already suffered anti-austerity protests; the top countries have in most cases defaulted on their debts and are grappling with difficult reforms, a sudden stop in foreign lending, or in many cases popular discontent.
The high position of Brazil is a surprise; unlike most of the other top countries, Brazil is unlikely to be forced into austerity by an immediate debt crisis. On the other hand, the Oxford Analytica contributor notes, in August 2023 Brazil passed into law a new fiscal framework, and “there is a growing expectation among market players that spending cuts are unavoidable for the administration to comply with the new framework.”
*van Stekelenburg, J. & Klandermans, B., 2010. Individuals in movements: A social psychology of contention. In: Handbook of Social Movements across Disciplines. s.l.:Springer, pp. 103-139.
A Risk Index for Austerity and Social Unrest in 2024
Source: WTW calculations based on IMF, Credendo, World Bank, and Freedom House
