Behind Chile’s political crisis
More than one million people marched in Santiago on October 26 to protest the Government’s security response to Chile’s current political crisis and to demand structural economic reforms to reduce inequality and increase social services. In this post I analyze these grievances from a quantitative perspective and explore what it would take to translate them into policy.
This is my fourth inequality-related post. I use the same sources of data and framework of analysis as in my initial analysis focused on Canada, its update to include inequality/redistribution “models” and my analysis of inequality in Venezuela and encourage interested readers to refer to these for further conceptual and technical background.
Chile has long been an economically unequal society.
Figure 1 shows that Chile’s market income inequality, as measured by the Gini coefficient, has been very high for as long as the data has been available, including for the last 40 years, as shown in Figure 1. For comparative purposes Figure 1 also includes two group averages (“LatAm-7” group: Argentina, Brazil, Colombia, Costa Rica, Mexico, Panama and Peru; “OECD-11” group: Australia, Canada, Finland, France, Germany, Italy, Japan, Norway, Sweden, UK and USA) and three other countries.
Chile joined the OECD in 2010 and is currently the Latin America country with the highest GDP per capita, currently (2018) at USD $25,000. The data for the OECD-11 and Chile are from the OECD’s Income Distribution Database, supplemented with the Standardized World Income Inequality Database (SWIID), which is the main source of data for the LatAm-7.
The Gini coefficient varies from 0 to 1.00, with higher values representing higher inequality. The Gini is one of a number of ways in which inequality may be measured. Other measures include the proportion of income flowing to the 1% or 10% of the population with the highest income. I use the Gini because it is more conducive to redistribution-related policy analysis and is more broadly available than other measures.
Market income Gini coefficients measure the distribution of “market” income from the labour and capital markets and differs from “disposable” (sometimes referred to as “net”) income, which measures income after the payment of direct taxes and the receipt of public cash transfers/benefits. The difference between “market” and “disposable” income inequality reflects the extent to which Governments reduce market income inequality by direct taxes and cash benefits.
Starting at group averages, Figure 1 shows that market inequality was significantly higher (almost 0.10 Gini points) in Latin America than in the OECD in 1980. Inequality in both groups started to climb in the mid-1980’s until the early-2000’s, at which point inequality in Latin America began to decrease.
Chile’s market inequality has remained very high and is now well above the LatAm-7 average. For comparative purposes I have included Canada, France and the USA, which are representative of three types of inequality/redistribution country “models”. Market inequality in Canada, France and the USA has increased by 0.07, 0.05 and 0.10 Gini points, respectively, with France and the USA now reached levels similar to Chile. Figure 1 also presents the year in which each country reached Chile’s current GDP per capita and shows that they generally had lower or similar inequality than currently, indicating that Chile’s high market inequality is not related to its relative or absolute GDP per capita, but rather to other factors, including the distribution of human and financial capital and other structural elements, including the extent to which markets have been designed and/or regulated to benefit the elite.
Chile has historically had very low economic redistribution.
The difference between the market income Gini and the disposable income Gini is referred to as “fiscal redistribution” or just “redistribution”. Figure 2 shows that Chile’s redistribution has been similar to that LatAm-7 but is well below that of the OECD-11 and Canada, France and USA. Readers will note that these countries have chosen very different levels of redistribution to counteract their underlying market inequality. France and the USA are high-inequality countries, but each chooses a different level of redistribution, with France having about twice that of the USA. Canada is a low-inequality country with the same low level of redistribution as the USA. Figure 2 shows that the level of redistribution is not related to GDP per capita: redistribution was similar to current levels in all countries when they had GDP/capita comparable to that of Chile today. Redistribution is fundamentally a political choice with real societal and welfare consequences.
To be clear, redistribution is not designed to, nor does it generally have an immediate or large impact on the underlying determinants of market inequality, including the distribution of human or financial capital. Rather, it is principally designed to moderate market inequality to politically-acceptable levels – it is a form of “social contract”. As added economic bonus, as I noted previously, the recent economic evidence is that redistribution tends to (slightly) increase economic growth. That is, there is no “efficiency-equity trade-off” that proponents of neoliberal policies used to justify a minimalist State. And that a minimalist State is certainly what Chile has had.
Figure 3 shows Government (all levels) taxation revenues as a percentage of GDP (data from the OECD Global Revenue Statistics database), which is my preferred widely-available indicator to use as a proxy for the comparative economic weight of the State. Figure 3 shows that since comparable data has been, available starting in 1990, Chile’s taxation revenues have been only about half of those in the OECD-11. Figure 3 shows that the level of taxation revenues (as a proxy) is not related to GDP per capita. Comparing Figures 2 and 3, readers will note the direct relationship between tax revenues and redistribution – in modern democracies, citizens choose, generally indirectly, to have higher/lower levels of redistribution, including by having higher/lower taxes.
The composition of taxation revenues in the OECD-11, with a relatively heavy weight on personal and corporate income taxes, are relatively progressive and account for about one-third of redistribution (measured in Gini points). In contrast, Chile mostly relies on value-added taxes (VAT) and hence taxes have little or no redistributive impact (Chile also receives non-tax revenues – royalties – from copper and other mining).
The remaining two-thirds of the distributive impact in the OECD-11 is the result of cash benefits, which will depend on their quantum and degree of targeting at low-income households (“efficiency”). Based on the most recent data (2015) from the OECD’s Social Expenditure database, cash benefits account for 5% of GDP in Chile, compared to 9% for Canada and USA, 13% for OECD-11 and 19% for France.
Disposable Income Inequality
Figure 4 shows disposable income inequality. Because Chile has chosen very low redistribution, Chile’s disposable income inequality remains very high, with a Gini coefficient of about 0.44. In contrast, the two countries that had similar market inequality as Chile, France and USA, both moderate their disposable income inequality by different degrees, USA to a Gini coefficient of about 0.39 and France to such an extent that it ends up having a Gini coefficient similar to Canada’s (and the OECD-11) of about 0.30.
To see how these Gini coefficients relate to other income inequality measures, the OECD estimates that for 2015 a very large 36% of disposable income goes to the highest 10% in Chile; that figure is 29% in the USA, and 24% in Canada, France and OECD-11.
Public financing and privatization of Social Services and Education
A key demand of the Chileans marching in the streets is increased public financing of social services, including for education, health and retirement. In keeping with the neoliberal model of a minimalist state, in each of these sectors Chile has established parallel public/private systems.
The publicly-financed systems are designed to be universal and have relatively modest or no user fees/contribution charges, depending on the level of service selected and income level of the user. In practice, however, many middle-income and almost all high-income Chileans “opt out” of the perceived lower-quality public systems by subscribing into a health or retirement “plan” offered by a user-fee-based for-profit private provider that make up the private systems. In fact, many of the grievances of the marchers are directed at these health plan providers (“ISAPRES”) or the retirement plan financial entities (“AFPs”), or the for-profit universities.
This system perpetuates Chile’s high market inequality because it relegates lower-income households to generally lower-quality public systems that are less conducive to building (education) and maintaining (health and retirement) human capital to participate in labour markets. At the same time, the privatized systems allows for the creation of financial capital, which is predominantly held by higher-income households to which the investment income flows.
Figure 5 shows how comparatively little public financing Chile provides for social expenditures (e.g. health, retirement, etc.) and education, reaching about 15% of GDP, compared to 22-23% for Canada and USA, 28% for the OECD-11 and 37% for France.
This ranking of public financing generally corresponds to the ranking of taxation revenues in Figure 3, which puts back the question into political economy terms of the choice between low/high public provision of services and low/high taxes. Chile is a prime but not the only example that the demand for “public” services, whether it be education, health or retirement security, is independent of their public provision and how private systems can be designed to complement or even substitute them.
Financing of Health and Education Sectors
Figures 6 and 7 show the relationship between the public and private systems in the health and education sectors.
The OECD collects detailed health statistics, including on public and private expenditures, including out-of-pocket medical costs. Figure 6 presents the relative size of privately-financed/funded expenditures and shows that Chile’s system is highly privatized. Chile is similar to the USA, at around 50%. The slight decrease in the USA in 2014 and 2015 are the result from the coming into force of the ACA (“Obamacare”). In contrast, Canada, France and OECD-11 are around 25-30% private.
Total (public and private) health expenditures are highest in the USA (16% of GDP) and lowest in Chile (8%) with Canada, France and the OECd-11 at between 10-11%.
Similarly, based on detailed OECD education financing statistics, Figure 7 presents the relative size of privately-financed/funded education expenditures. It shows that Chile has long been a highly privatized system, even higher than that of the USA, and much higher than Canada, OECD-11 and France.
The reason is that while a number of OECD-11 countries have a high private participation in tertiary education comparable to Chile (60-70%) none of the comparator countries also have Chile’s relatively high private participation in primary and secondary education (20%), which makes up three-quarters of overall expenditures (which are between 5-6% of GDP for the comparator countries).
The structural economic reforms that are being demanded by the millions marching in Chile are simply those that most advanced democracies have long implemented: policies to substantially moderate income inequality and to finance high-quality universal social services.
These policies are relatively simple to administer and have been implemented by dozens of countries with national incomes well below that of Chile. There is no administrative or economic constraint. But that has never really been the issue in Chile (or elsewhere). Rather, until the last two weeks, the political equilibrium in Chile appeared to be one of incremental reforms to the neoliberal policies imposed during the military dictatorship of 1973-1990.
So assuming the political equilibrium in Chile is more responsive to demands for structural reforms, what would it take to have such demands translated into policy?
For example, Chileans could decide to increase redistribution. Based on the analysis above, I estimate that matching the levels of redistribution in Canada and USA would require about 5% of GDP to finance, while matching those of the OECD-11 about 8% of GDP. Similarly, Chileans could decide to increase public financing of high-quality and universal social services and education, which to match Canada and USA levels would require about 5% of GDP, and to match OECD-11 levels about 7% of GDP.