Latin America’s growing middle class gives cause for optimism about the region’s economic and democratic prospects, according to a new analysis. In some countries, the growth is large enough to generate the virtuous circle of more middle class societies reinforcing the inclusive politics that sustain broadly shared growth, writes Nancy Birdsall. There is no evidence that a politically powerful middle class will discourage policies that help the poor, she notes. Indeed, in countries with center-left as opposed to populist governments, changes in social policies have benefited the poorest and most vulnerable in society.
The Latin American and Caribbean region is becoming middle-class. Thirty percent of the population is middle class (up from 20 percent two decades ago) and its income share has increased from 40 to nearly 50 percent. Is that because of global economic expansion? Will the middle class continue to grow even as overall growth rates slow?
An analysis of household data provides a cause for optimism about the economic and democratic prospects in Latin America and the Caribbean.
First, it is encouraging that whatever is captured in self-responses about being “middle class” seems to accord reasonably well with the objective criterion of relative invulnerability to falling into poverty. It would surely make Alexis de Tocqueville happy that the proportion of people that feel more economically secure has grown from 20 to at least 30 percent in many countries of the region.
Second, it is encouraging that across time and countries people in households with per capita income between $10 and $50 share such key characteristics as education and work status. This suggests that it is reasonable to have a construct called the Latin American middle class – with potential spillovers across countries in policy norms. A secure middle class is likely to both object to corruption in government as well as be more effective in fighting it. Perhaps the exposure of high-level corruption in Brazil will reinforce intolerance for such corruption in Peru and Bolivia.
The same might be said about respect for human rights: While the middle class would certainly not view the issue differently, it would be better able to advance its views the larger it is.
Third, that the middle class is not exceptional in its characteristics after controlling for their income is another cause for optimism. It suggests, as Charles Kenny recently argued, that the poor are no different from you and me. Most of the world’s poor are not poor because they are deficient in ambition or enterprise, but because they lack the assets – schooling, land, access to finance– that provide a middle class income. A larger middle class suggests some positive trends at the societal level; more people have one or several of those assets that are sufficient to insure reasonable economic security.
Moreover, the fact that the middle class is not exceptional after controlling for current income vindicates this measure as a reasonably sensible proxy for education, permanent income, occupation, or whatever other characteristics analysts might prefer to identify the middle class – at least in the case of Latin America. An income level that is sufficient to be part of the middle class does not guarantee happiness, but it does go along with characteristics and capabilities that are associated with well-being and, in the wider sense, agency and human freedom.
Fourth, the finding that the middle class is not exceptional in behavior or values does not imply that societies are not somehow different when the middle class represents a larger share of population and income. The measures of middle class “society” above indicate that Chile, Brazil and Costa Rica are the most middle class countries in our sample of eight LAC countries. This can largely but not entirely be explained by their higher average income.
The countries’ income distributions also matter, at least in a statistical sense. Costa Rica has a lower average income than Brazil, but a larger middle class population share with a larger command of total income. In Costa Rica and Chile (but not in Brazil) the middle class captured an estimated 53 percent of total income in 2009.
Will the LAC “rich” middle class align themselves politically with their richer counterparts or the much larger poor and vulnerable populations? There is indirect evidence that this well-off middle class is open to redistribution. Birdsall, Lustig and McLeod (2011) find that in countries with center-left as opposed to populist governments, changes in social policies have benefited the bottom four quintiles more than the top quintiles in which the middle class is heavily concentrated. Brazil and Chile had such center-left governments throughout most of the 2000s.
This is another possible reason for optimism. Perhaps a growing middle class, with its growing political influence, supports political regimes that promise a combination of sound macroeconomic policy with a heavy emphasis on social programs that reach the majority of the population. A growing middle class perhaps even supports such policies if their own group does not directly benefit through public expenditures. This would suggest that the middle class in Latin America sees its future aligned with governments that deliver a combination of property rights protection, stability, and greater overall access to public goods. The middle class would then be a consequence of good economic policy and, in a virtuous circle, a supporter of such policy.
It is often surmised that a politically powerful middle class will discourage policies that help the poor. There is no obvious evidence that this has been the case in Latin America over the past decade. Though the LAC middle class is rich relative terms, the profiles above show that in absolute income and other characteristics it is closer to the large group of poor and vulnerable households than to the rich. This is true with respect to income itself and with respect to education once the high returns to university education – the norm only among the rich – are taken into account.