Middle-Class Heroes: The Best Guarantee of Good Governance

February 16, 2016

This article first appeared in Foreign Affairs.[1]

The two economic developments that have garnered the most attention in recent years are the concentration of massive wealth in the richest one percent of the world’s population and the tremendous, growth-driven decline in extreme poverty in the developing world, especially in China. But just as important has been the emergence of large middle classes in developing countries around the planet. This phenomenon—the result of more than two decades of nearly continuous fast-paced global economic growth—has been good not only for economies but also for governance. After all, history suggests that a large and secure middle class is a solid foundation on which to build and sustain an effective, democratic state. Middle classes not only have the wherewithal to finance vital services such as roads and public education through taxes; they also demand regulations, the fair enforcement of contracts, and the rule of law more generally—public goods that create a level social and economic playing field on which all can prosper.

The birth of new middle classes all over the world therefore qualifies as a triumph of capitalism and globalization. But it is a fragile victory. For the world now faces a period of prolonged slow growth. That is bad news, not only because it could halt the impressive declines in poverty but also because it could set back hopes for better governance and fair-minded economic policy across the developing world, harming both middle classes and the far larger populations of poorer people in the developing world who are the chief victims of weak or abusive governments. The rich world could lose out, too, since improvements in governance allow poor countries to collaborate with the international community in managing the risks posed by pandemics, terrorist groups, climate change, waves of political refugees, and other regional and global problems. Governments in the developing world and in rich countries alike would do well to nurture and protect the legitimate interests of the new middle classes.

Who is Middle Class?

In today’s high-income countries, “middle class” is a relative measure: most households earn enough money to place them more or less in the middle of the national income distribution. But “middle class” has always been an absolute concept as well: to be middle class means enjoying sufficient material security to be able to credibly plan for the future. That definition is particularly important in the developing world, where economists increasingly identify a middle-class household as one with enough income to survive such shocks as a spell of unemployment, a health emergency, or even the bankruptcy of a small business without a major or permanent decline in its living standard. Middle-class citizens deal with plenty of economic anxiety and stress, but they don’t worry about being able to pay next month’s rent, car loan installment, or credit card bill.

Evidence from Latin America suggests that reaching middle-class economic security in that region requires a daily income of somewhere around $10 per person, or the equivalent of around $10,000 a year for a family of three[2]. That family is likely to include at least one adult who has completed secondary school and works in an office, a factory, or a retail job with a steady paycheck, as opposed to working in agriculture or the informal economy[3].

That $10-per-day threshold, adjusted for differences in prices across countries, can be applied elsewhere as a rough proxy for middle-class status. Of course, it’s not a perfect measure: for example, it puts middle-class households in developing countries such as India and Kenya nowhere near the actual middle of those countries’ income distributions. They’re much closer to the top: of India’s 1.25 billion people, at most 100 million enjoy that level of income[4]. Indeed, for India, where the median daily income per person is less than $5[5], the $10-per-day threshold is probably too high; for Chile, the richest country in Latin America, with a median daily income of $14, it is probably too low. Imprecise as it may be, however, the absolute $10 figure is nonetheless useful, since it allows economists to compare the sizes of middle classes in various developing countries and to track their growth over time.

Bourgeois Benefits

The size of a country’s middle class has significant economic and political implications. A large middle class increases the demand for domestic goods and services and helps fuel consumption-led growth. Middle-class parents have the resources to save and invest in their children’s education, building human capital for the country as a whole. And in the developing world, people living on $10 a day or more are able to take reasonable business risks, becoming investors as well as consumers and workers. In all these ways, the emergence of a middle class drives economic growth.

Having a large middle class is also critical for fostering good governance. Middle-class citizens want the stability and predictability that come from a political system that promotes fair competition, in which the very rich cannot rely on insider privileges to accumulate unearned wealth[6]. Middle-class people are less vulnerable than the poor to pressure to pay into patronage networks and are more likely to support governments that protect private property and encourage private investment. When the middle class reaches a certain size—perhaps 30 percent of the population is enough—its members can start to identify with one another and to use their collective power to demand that the state spend their taxes to finance public services, security, and other critical public goods[7]. Finally, members of a prospering middle class are unlikely to be drawn into the kinds of ethnic and religious rivalries that spur political instability.

Of course, having a large middle class is no guarantee that a country will enjoy political stability and democratic (or even accountable) government. By the early 1980s, Venezuela’s middle class had grown to include around half of the country’s population, thanks mostly to the strength of the state-controlled oil sector. But unlike revenue from tax-paying middle classes, easy oil income tends to enrich governments without forcing them to become more accountable, and that is precisely what happened in Venezuela. In recent decades, poor governance has contributed to economic decline, and by 2006 (the most recent year for which data are available), the middle class had shrunk to 40 percent of the population. In the past ten years, it has almost surely shrunk even further. In 2012, more than 50 percent of Thailand’s population belonged to the $10-per-day middle class. But the following year, the country boiled over into political chaos that ended in a military coup. Meanwhile, under President Vladimir Putin, oil-rich Russia has developed a big middle class and a stable government, but Putin’s regime has successfully resisted pressure to become accountable. (It’s also worth noting that a large middle class is not a prerequisite for stability. Rwanda, where less than ten percent of the population belongs to the $10-per-day middle class, has enjoyed a stable government and rapid, widely shared growth for more than two decades under President Paul Kagame.)

The point is that when it comes to the middle class, size matters, but it is not everything. For example, if a middle class grows large but then feels threatened during a major economic downturn, its members may succumb to demagogic and populist appeals—from the right or the left. In Argentina, a decade of inflation and a debt crisis in 2001–2 paved the way for the revival of Peronist populism, which shaped the policies of Néstor Kirchner, who served as president from 2003 to 2007, and of his wife, Cristina, who succeeded her husband and served until last year, when she was defeated in a bid for a third term. This dynamic is hardly exclusive to the developing world: a 2015 Pew Research Center study concluded that the size of the U.S. middle class and its share of the country’s income and wealth are shrinking[8], which might partly explain the appeal of “outsider,” non-establishment candidates in this year’s presidential race. And in Europe, the fear of slow growth and worries about a “new machine age,” in which automation and robots will eliminate jobs now held by well-educated members of the middle class, help account for the growing influence of anti-immigrant right-wing political parties.

Put simply, to constitute a politically positive force, a middle class must be not only large relative to a country’s other classes but also prospering and feeling confident[9]. That is not surprising: behavioral studies show that for most people, losing ground is more troubling than never gaining it, a tendency known as “loss aversion.” Widespread fears of looming losses undermine the sense of security and the expectations of a better future that characterize the middle class.

A Middle-Class World

Twenty-five years ago, hardly any developing countries had large, growing middle classes. Most people in the world still lived in places where the distribution of income could be characterized (with only slight exaggeration) as bimodal: a small elite lived in comfort, while the vast majority of people were poor. There were exceptions, including Singapore, South Korea, and a number of Latin American countries in which industrialization had begun before World War II. By 1990, South Korea had experienced 30 years of extraordinary growth. As a result, more than 60 percent of its population earned an annual household income of $10,000 or more in today’s U.S. dollars. South Korea had, in effect, already become a middle-class society; at the same moment, it was also completing a transition to democracy after decades of military rule.

But across most of the developing world, the $10-per-day middle class was still tiny. In China, India, and sub-Saharan Africa, it represented less than two percent of the population—and in Africa, that number was probably made up mostly of civil servants and the employees of international organizations and Western aid groups. Most people in Asia and Africa were still either terribly poor or just getting by.

Then, in the early 1990s, growth took off across the developing world and accelerated further during the first decade of this century, as low interest rates and a commodities boom benefited many low- and middle-income countries. Between 1990 and 2015, around one billion people escaped poverty, including about 650 million in China and India. During the same period, some 900 million people entered the $10-per-day middle class[10].

The most extraordinary middle-class growth has come in urban areas of China. In 1990, the $10-a-day middle class comprised an estimated 0.3 percent of China’s urban population—about one million people. By 2010, it had grown to 35 percent of China’s now much larger urban population—about 220 million people. By 2015, the figure had reached an estimated 340 million[11]. China may not be taking the road that brought South Korea to democracy in the 1970s and 1980s. But even the Chinese government has had to become far more responsive to an economically independent middle class that is unhappy about problems such as air pollution and corruption.

Brazil is another place where the impact of a growing middle class has been undeniable. In the first decade of this century, low interest rates and iron ore exports to China boosted Brazilian growth and domestic investment, including in job-intensive sectors such as construction. Partly as a result, Brazil’s $10-per-day middle class more than doubled between 1990 and 2015, from 20 percent of the population to almost 50 percent, and began flexing its political muscles. This newly empowered middle class has lent implicit but important support to the recent indictment of Brazilian officials accused of corruption in the Petrobras scandal, and its members will likely balk at any policies that might resurrect the destructive inflation that held the country back in the 1990s.

In the last two decades, meanwhile, Chile, Iran, and Malaysia have watched their middle classes grow to encompass almost 60 percent of their populations. And Mexico (close to 40 percent) and Peru (50 percent) have also witnessed major gains. This change bodes well for economic and political stability in all five of those countries—even in Iran, where President Hassan Rouhani’s modest but promising opening to the West has stemmed in part from his need to win the political support of a bigger, better-educated middle class.

Middle classes have grown in poorer countries as well, although they started from much lower bases and have reached much smaller sizes. Middle classes still comprise less than ten percent of the populations of many countries in South Asia and sub-Saharan Africa; the same is true in rural China. Even with healthy growth, the middle classes will be unlikely to reach 30 percent of the populations in those places during the next 20 years[12]. Among the non-oil Arab countries, only Morocco and Tunisia have sizable middle classes; in Egypt, by far the largest Arab country, just six percent of the population lives above the $10-per-day threshold.

In a few developing countries, such as India, Kenya, Nigeria, and Tanzania, middle classes have appeared but have not grown nearly large enough to effect significant political change. In those places, creating a virtuous cycle of middle-class growth and accountable governance remains a long-term development challenge. So although it makes sense to cheer the existence of modern shopping malls serving new middle classes in Lagos and Bangalore, it does not make sense to assume that every country with a lot of new malls is on a steady, predictable road to good governance and liberal democracy.

First the West, Now the Rest?

The last 25 years have been an exceptional period for the developing world and might eventually prove comparable to the surge in economic growth in the West that began with British industrialization in the nineteenth century and eventually allowed liberal democracy to spread throughout Europe and North America. During the twentieth century, the West left the rest of the world behind. The ratio of the median household income in the rich countries of the West to that in the rest of the world grew from about five to one in 1900 to around 20 to 1 in 2000[13]. The West experienced the twentieth century as one long virtuous cycle—interrupted by war and depression, of course—in which economic growth nurtured middle classes that in turn fought and paid for the state-led foundations of continuing growth: the rule of law, institutions that created the environment for entrepreneurship and innovation, and well-regulated markets.

The past few decades might prove to be an early chapter in a similar story for the developing world. For one thing, the globalization of markets may be speeding up the process. Globalization has favored the middle class by creating economies that richly reward educated workers, making it easier to obtain mortgage loans and other forms of credit, and generating manufacturing and retail supply chains that offer plenty of good jobs for skilled people. Meanwhile, advances in communications technology—the Internet, mobile devices, and social media—have empowered middle classes around the world to organize and advocate corporate and government accountability.

Optimists see those changes as major factors driving current political trends in some countries. In Turkey, factions within the urban middle class have resisted the creeping authoritarianism of President Recep Tayyip Erdogan. In Argentina, a large and relatively resilient middle class contributed to the recent defeat of former President Cristina Fernández de Kirchner’s handpicked successor, rejecting the costs of continued economic populism. And perhaps it’s no accident that Tunisia—where about 30 percent of the population belongs to the middle class (a very large proportion compared with most of the Arab world)—is the only country to have emerged from the Arab revolts of 2010–11 with something resembling democratic rule.

Soft in the Middle

The trouble is that the ongoing conversion of economic gains into political progress requires continued growth, and the global slowdown now threatens that process. Middle classes in Brazil, urban China, and Turkey are big but still new; the endurance of the political and social benefits they have provided depends on their institutionalization over the long term and the adoption of customs and rules that take a long time to harden into habits. A prolonged downturn in growth will complicate things in those countries—far more so than in the United States and Western Europe, where middle classes are suffering but the institutions built around them are well established and relatively strong.

In most emerging markets, high growth during the last decade depended on commodity exports and low interest rates. High profits and easy credit created retail and public-service jobs for graduates of secondary schools but did not necessarily raise productivity in manufacturing and large-scale agriculture. The economist Dani Rodrik worries about what he calls “premature deindustrialization” in the developing world, given that manufacturing—the setting for the productivity increases and struggles between labor and capital that helped produce democratic politics in the West—has already peaked at 15 percent of employment in Brazil and India, far lower than the 30 percent level found in South Korea in the late 1980s[14]. The fear is that the new middle classes will be hit hard if it turns out that global growth was built too much on easy credit and commodity booms and too little on the productivity gains that raise incomes and living standards for everyone.

If the middle class and those struggling to join it see their incomes stagnate or fall, they are less likely to support the economic and regulatory policies that over time increase the size of the overall economic pie. Instead, they are likely to embrace short-term, populist measures they believe will help them retain their gains and meet their raised expectations. In short, slow growth (or, worse, an economic collapse) could erode middle-class support for good governance, a broad social contract, and the economic reforms that sustain the opportunities on which the middle class depends.

Brazil might prove vulnerable to that dynamic. When the economy was growing rapidly and steadily, the middle class supported President Luiz Inácio Lula da Silva’s impressive program of cash transfers to the very poor. Although most middle-class families responded to weak public schooling by sending their children to private schools, they did not resist educational reforms to improve public schooling. In leaner times, however, a beleaguered middle class might be less tolerant of programs that benefit the poor and the working class and might politically ally itself with the rich instead. A similar shift could occur in every country with a large but relatively new middle class.

It takes several decades to develop and solidify the responsive state institutions that the middle class wants and on which it relies. And even then, a large middle class does not guarantee that democratic institutions can survive hard times. Germany in the 1920s and 1930s provides the quintessential cautionary tale. Devastated by the country’s defeat in World War I and suffering from runaway inflation, the German economy tanked, robbing the middle class of the sense of security that had bound it to the common good and opening the door to dangerous populism, Nazi demagoguery, and, finally, autocracy and the genocidal scapegoating of the Jews.

Protecting Fragile Gains

In a hyperconnected global economy, lower growth in China, Japan, and Europe and economic fragility in Brazil and other big emerging markets spell trouble everywhere. To avoid the worst outcomes, countries with emerging middle classes cannot take shortcuts. That means eschewing irresponsible fiscal policies and other missteps that could generate inflation and hurt everyone. Developing countries should also consider reforming their health-care, pension, and unemployment programs, which underpin citizens’ confidence in a secure future—not only within growing middle classes but also among people who have escaped absolute poverty and now aspire to middle-class security and status. Above all, middle classes in developing countries would benefit from reforms to educational systems and increased investment in infrastructure. Good schools and roads offer high returns to everyone, but they especially encourage private investment and bring the productivity gains on which middle classes build and prosper[15].

High-income countries can also play a role in building middle-class societies in the developing world. Development aid is not enough, however, as it ultimately has a minimal effect in recipient countries. Wealthy countries should instead focus on removing the obstacles they have created to healthy, productivity-driven growth—by cracking down on tax avoidance and evasion on the part of major multinationals, which reduce tax revenue in developing countries[16]; by fixing privacy laws that have made it too easy to hide stolen assets abroad; by enforcing anti-bribery rules[17]; by ending protectionist policies in agriculture and textiles[18]; and by improving the management of their immigration systems[19].

The rich world can also lead by coordinating responses to collective problems that no single country has an incentive to address on its own. The most immediate danger comes from the risk of another financial panic that might spread globally. Such turmoil would do tremendous damage to incipient middle classes and to the millions of workers on the verge of moving from low-productivity, informal jobs into steady and reliable positions. Even more troubling for the long run is climate change, which threatens economic development everywhere and will surely go unsolved without leadership and financing from wealthy countries.

Rich individuals and corporations can also do their part by continuing to create new opportunities around the world, especially through investment in new technologies. Brazil’s huge middle class, concentrated in the country’s south, is in part the product of public and corporate research and investment that dramatically increased the region’s yields of soy, apples, and other crops. Mobile technology is helping create middle-class opportunities in poor countries. And the philanthropist Bill Gates’ recent launch of a $2 billion initiative to research and develop clean energy will indirectly create new green industries and jobs for middle-class workers everywhere.

None of those steps, of course, will completely offset the ill effects that long-term stagnation might have on the world’s burgeoning middle classes. Only strong growth can do that. But doing nothing at all would risk allowing the world’s new middle classes—one of the most hopeful developments of the past 30 years—to turn into a source of division and instability.

[1] I am enormously grateful to Anna Diofasi for her assistance in accessing and developing the data invoked throughout this essay.

[2] Lopez-Calva, L. F. and Ortiz-Juarez, E. (2011). “A Vulnerability Approach to the Definition of the Middle Class”. World Bank Policy Research Working Paper.

[3] Birdsall, N. (2012). “A Note on the Middle Class in Latin America”. CGD Working Paper. /publication/note-middle-class-latin-america-working-paper-303.

[4] Estimates about the population share of the middle class throughout this papers are based on income or consumption survey data as available via the World Bank’s PovcalNet database. All those with an income/consumption level of above $10 a day (using 2011 PPPs) are categorized as ‘middle class’. All estimates cited are approximate; the relevant survey data come from different years and no adjustment was made in comparing countries with consumption-based vs. income-based survey estimates.    

[5] All median income and consumption figures cited in this paper are based on data from PovcalNet. See also the global median database published by Diofasi and Birdsall (2016).

[6] Loayza, N., Rigolini, J. and G. Llorente. (2012) provide suggestive evidence in “Do Middle Classes Bring Institutional Reforms?” IZA Discussion Paper Series.

[7] See: Sokoloff, K. L. and S. L. Engerman (2000). “Institutions, Factor Endowments, and Paths of Development in the New World.” Journal of Economic Perspectives, 14(3):217-232; and Birdsall, N. (2010). The (Indispensable) Middle Class in Developing Countries; or, the Rich and the Rest, Not the Poor and the Rest. CGD Working Paper. /publication/indispensable-middle-class-developing-countries-or-rich-and-rest-not-poor-and-rest.

[8] Pew Research Center. (2015). “The American Middle Class Is Losing Ground”.

[9] See also: Friedman, B. M. (2006). The Moral Consequences of Economic Growth. New York: Vintage Books.

[10] Author’s estimates based on data from PovcalNet. See footnote 4 for a more detailed description.

[11] Author’s estimates based on data from PovcalNet.

[12] See: Birdsall, N, Lustig, N. and C. J. Meyer (2013). “The Strugglers: The New Poor in Latin America?” CGD Working Paper. /publication/strugglers-new-poor-latin-america-working-paper-337.

[13] Historic median data based on approximation from GDP per capita data in the Maddison Project database:

[14] Rodrik, D. (2015). “Premature Deindustrialization”. NBER Working Paper.

[15] Investment in public infrastructure has probably been less significant in developing countries than in most developed ones. However, returns, given lower capital base, are relatively higher in developing countries as also noted by Berg et al. (2015) "Some Misconceptions About Public Investment Efficiency and Growth," IMF Working Paper. Moreover, the IMF (2015) sees large potential to improve efficiency: "Making Public Investment More Efficient",

[16] See on these and other non-aid policies the papers cited as background to the Commitment to Development Index at On tax issues, see also Forstater and Ramachandran (2015): /blog/taxing-multinationals-there-pot-gold-finance-development.

[17] See, for example, Moran, T. (2008). “Combatting Corrupt Payments in Foreign Investments Concessions: Closing the Loopholes, Extending the Tools”. /sites/default/files/15197_file_CombatingCorruption.pdf

[18] Elliott (2015) outlines specific recommendations for the US in “Taking the Lead on Trade and Development”. /publication/ft/taking-lead-trade-and-development.

[19]See: Pritchett, L. (2006). Let their people come: Breaking the Gridlock on International Labor Mobility. Washington, DC: Center for Global Development. /publication/9781933286105-let-their-people-come-breaking-gridlock-global-labor-mobility.

Rights & Permissions

You may use and disseminate CGD’s publications under these conditions.