Attitudes towards democracy and fiscal policy
Many analysts have stressed the important role of the middle sectors in the functioning of the democratic system and social cohesion. Latin America has been steadily becoming more democratic since the mid-1980s, according to the "Polity IV" ranking, a widely used data series in political science research (Figure 4.1).
2 Out of 23 Latin American and Caribbean countries included in this database, 18 were ranked as democracies in 2008, with only Cuba left as an autocracy – whereas in 1980 there were eight autocracies and only seven democracies. From the early to the mid-1990s this expansion was accompanied by a decline in the average quality of democracy, a reflection of the relatively imperfect nature of the new regimes. Since then there has been a fairly steady democratic consolidation in the region.3 There are of course considerable differences across countries – from consolidated democracies such as Costa Rica, Chile and Uruguay (with a Polity score of 10, the same as most OECD countries), to countries such as Ecuador and Venezuela where democratic consolidation is considerably weaker.
Figure 4.1. Democratic consolidation in Latin America and the Caribbean
Democratic consolidation is often associated with increased demand for social expenditure, as sections of the population that were previously excluded from the decision-making process begin to exert their civil rights. Brazil’s transition towards democracy is emblematic, being accompanied by a substantial increase in government expenditure to meet the state’s new obligations under the country’s 1988 constitution (Figure 4.2). There are potentially important development challenges here: if the state does not gather sufficient financial resources to meet voters’ legitimate demands, then its choice is between satisfying them at the cost of unsustainable macroeconomic policies, or leaving them unfulfilled and undermining the democratic system.4
How Latin America is navigating this dilemma can be tested by looking at two key indicators of public perceptions: support for the proposition that democracy is the best system; and satisfaction with the actual way democracy functions in their country (Figure 4.3). The picture that emerges is one of preference for democracy in principle, but low satisfaction with how democracy is working. With the sole exception of Uruguay (where over 70% of the population is satisfied), the majority of people in every country in the region are not satisfied with the way democracy is currently working.
This does not reflect disillusion with democracy itself, support for which is much higher in most countries. In Venezuela, Dominican Republic, Uruguay, Paraguay and Guatemala more than 70% of the population support democracy. In a second group, though levels are lower, democracy still clearly enjoys the support of the majority. This group includes Nicaragua, Chile, Honduras, Argentina and Peru. In the rear, Bolivia, Colombia, Mexico, Panama, Costa Rica, Ecuador Brazil and El Salvador see support from around just 50% of the population – among this group are the two most populous countries in the region, Brazil and Mexico. Democracy is far from having consolidated either support or satisfaction across the region.
Figure 4.2. Democratic transition in Brazil and government consumption (percentage of GDP)
Figure 4.3. Satisfaction with and support for democracy by country (percentage of respondents, 2008)
What part do the Latin American middle sectors play in this? The data available allow analysis across self-perceived income quintiles (Figure 4.4)5. Satisfaction with democracy increases steadily with perceived economic status. A person who puts him or herself in the highest quintile is almost twice as likely to be satisfied with the way the democratic system works than a person in the first quintile (57% satisfaction against 31%).6 Support for democracy is more nuanced. It is the self-declared middle sectors that value democracy most.
Figure 4.4. Attitudes towards democracy by perceived income quintiles in Latin America (percentage of respondents)
Political stance can also be analysed by where people place themselves on a left-right scale (Figure 4.5). These positions are often used as an approximate measure of the demand for redistribution, with the left being associated with more redistribution and the right with more economically liberal views.7 Two interesting results emerge. First, people who perceive themselves as part of the middle sectors (those in the second to fourth quintiles) tend also to put themselves in the centre of the distribution of political preference. For example, over 54% of these middle sectors put themselves between 4 and 6 (the political centre). The equivalent figure for the disadvantaged is around 41% and for the affluent 28%. Second, the proportion of the middle sectors that place themselves at the extremes (of either left or right) is lower than the disadvantaged or the affluent. This is reflected also by a lower dispersion in political preferences within the middle sectors against the other groups.8
Figure 4.5. Distribution of political preferences by perceived income quintiles (percentage of respondents)
The evidence, then, shows the middle sectors in Latin America are in principle supporters of democracy and have rather moderate views on politics, yet remain dissatisfied with how democracy actually functions. Is this dissatisfaction evident in their views on taxation and public services? Figure 4.6 synthesises the main findings. Clearly, the middle sectors display greater "tax morale": members of the middle sectors are more likely than other members of society to consider that citizens should pay their taxes, are less likely to consider that taxes are too high, and less likely to justify tax evasion. However, they are also less satisfied with the provision of public services, compared to the affluent. In short, members of the middle sectors have a "dissatisfied customer" relationship with the state: while relatively supportive of taxation, they are not satisfied with the services they receive.9