We Promised Development Would Follow Growth. The Money Stayed at the Top.

There is a hotel in Dhaka where the breakfast buffet costs more than a garment worker earns in a week. I know this because I have stayed there, and because I have met the workers who sew the clothes sold in the countries where people like me live. The hotel and the garment factories exist in the same economy. We were told this arrangement would eventually benefit everyone.

The promise was simple: prioritize economic growth, attract foreign investment, liberalize markets, and prosperity would trickle down to the poorest. For half a century, this has been the dominant theory of development economics. It shaped World Bank lending, IMF structural adjustment programs, and the development strategies of nations from Indonesia to Ghana to Peru. It was presented not as ideology but as science.

The problem is that it did not work. The wealth did not trickle down. And the people who designed these policies had access to the data that showed it would not.

The Promise

The theory emerged in the 1970s and 1980s, codified in what economist John Williamson termed the "Washington Consensus" in 1989. Ten policy prescriptions: fiscal discipline, tax reform, trade liberalization, privatization, deregulation. The assumption was that economic growth and poverty reduction were essentially the same project. If GDP grew by six percent annually, the benefits would cascade through the economy. A rising tide lifts all boats.

Between 1990 and 2019, the global economy grew by 147 percent in real terms. During the same period, according to the World Bank's own data, the number of people living in extreme poverty fell from 1.9 billion to 648 million—a genuine achievement. But the distribution of that growth tells a different story. The richest one percent captured 27 percent of total income growth between 1980 and 2016, according to the World Inequality Database. The bottom 50 percent captured 12 percent.

In individual countries, the pattern is starker. India's economy grew ninefold between 1991 and 2020 after liberalization. The number of billionaires rose from zero to 102. Real agricultural wages—the income of the poorest 40 percent of workers—grew by 2.1 percent annually, far below the overall growth rate of 6.5 percent. The boats rose. Some rose faster than others. Many stayed exactly where they were.

What the Architects Knew

The failure was not an accident of implementation. It was baked into the design. In 1974, economist Hollis Chenery, then chief economist at the World Bank, published "Redistribution with Growth," arguing that deliberate redistributive policies were necessary alongside growth. The report was shelved. By the 1980s, under pressure from the Reagan administration and Thatcher's Britain, the Bank had moved in the opposite direction.

Joseph Stiglitz, who served as the World Bank's chief economist from 1997 to 2000, later wrote that the institution knew structural adjustment programs were failing the poor but continued to impose them because they served the interests of creditors and multinational corporations. In a 2002 book, he described how the IMF's policies in East Asia during the 1997 financial crisis prioritized debt repayment over social spending, deepening inequality even as economies recovered.

Internal World Bank evaluations tell the same story. A 2006 review of structural adjustment lending in 36 countries found that poverty reduction was "not systematically monitored" and that distributional impacts were "rarely assessed." The evaluators recommended that future programs explicitly target inequality. The recommendation was noted. The programs continued largely unchanged.

The Alternative That Worked

The cruelest irony is that we already knew what worked. Between 1960 and 1990, South Korea, Taiwan, and Japan achieved both rapid growth and dramatic reductions in inequality—not by following the Washington Consensus, which did not yet exist, but by violating nearly every principle it would later enshrine.

They imposed capital controls. They protected infant industries. They used industrial policy to direct investment toward sectors that employed large numbers of workers. Most importantly, they pursued land reform and invested heavily in education and healthcare before liberalizing trade. When South Korea's economy grew by an average of 8.6 percent annually from 1960 to 1990, its Gini coefficient fell from 0.36 to 0.31. Income inequality declined as the economy grew.

More recently, Ethiopia's Productive Safety Net Programme, which provides cash transfers and public works employment to eight million people, lifted 1.4 million out of poverty between 2005 and 2015, according to independent evaluations by the International Food Policy Research Institute. The program cost $500 million annually—less than the country spent servicing external debt.

The difference is clear: programs that directly target poverty reduction work. Programs that assume growth will automatically reduce poverty do not. And yet the latter continue to dominate development lending.

Who Benefits

The persistence of trickle-down development makes sense once you ask who profits from it. Trade liberalization in the 1990s opened markets in the Global South to multinational corporations, which increased their share of developing-country retail sales from 23 percent in 1990 to 67 percent in 2015, according to UNCTAD data. Privatization of state enterprises transferred assets worth an estimated $3.3 trillion to private investors between 1988 and 2008, often at below-market valuations.

Financial liberalization allowed foreign banks into developing economies, where they now control 40 percent of banking assets in Latin America and 52 percent in sub-Saharan Africa. The profits flow outward: in 2019, multinational corporations repatriated $1.3 trillion from developing countries in dividends and interest payments, more than double the $570 billion those countries received in foreign direct investment and development aid combined.

Tax policy tells the same story. Between 1990 and 2018, the average corporate tax rate in developing countries fell from 38 percent to 23 percent, according to the IMF, often as a condition of World Bank loans or in competition to attract investment. The Tax Justice Network estimates that developing countries lose $483 billion annually to corporate tax avoidance and evasion—more than they receive in development assistance.

The arrangement is self-reinforcing. Countries compete to offer lower taxes and fewer regulations. Corporations relocate to wherever the offer is most favorable. The promised jobs arrive, but the tax revenue to fund schools and hospitals does not. The growth happens. The development does not.

The Human Cost

I return to the hotel in Dhaka, because it is a useful image. The workers who make the clothes are employed. Bangladesh's garment industry employs four million people, mostly women, and accounts for 84 percent of the country's export earnings. By the metrics of trickle-down development, this is success. GDP growth averaged 6.2 percent annually from 2010 to 2019.

The workers earn an average of $95 per month, according to the Clean Clothes Campaign. This is above the $1.90-per-day extreme poverty line, which is why they do not appear in the poverty statistics. It is not enough to afford adequate housing, healthcare, or education for their children. When the Rana Plaza factory collapsed in 2013, killing 1,134 workers, the compensation fund took three years to raise $30 million—less than the annual profit of a single mid-sized Western retailer that sourced from the factories inside.

This is what trickle-down development produces: employment without dignity, growth without equity, integration into the global economy on terms set entirely by those who already have capital. The workers are not unemployed. They are simply poor in a way that is statistically convenient.

What We Owe

The defense offered by the architects of trickle-down development is that the alternatives were worse. Central planning failed. State-led industrialization produced inefficiency and corruption. Market-based growth, however unequal, at least delivered growth. This is true as far as it goes.

But the choice was never binary. The East Asian development model demonstrated that state capacity and market mechanisms could coexist. The Nordic countries showed that high taxation and economic dynamism were compatible. More recently, conditional cash transfer programs in Brazil, Mexico, and elsewhere have proven that direct anti-poverty measures can operate alongside market economies without undermining growth.

The money exists. The question is whether we choose to redistribute it. For fifty years, we have chosen not to, while telling ourselves—and the people whose labor enriches us—that redistribution was unnecessary because the market would eventually do the work for us.

It has not. It will not. The trickling has stopped, if it ever began. What flows upward, toward capital and away from labor, is not a bug in the system but its central feature. We built this. We have maintained it. And the evidence that it does not work—that it has never worked—has been available the entire time.

I am not sure what I expected when I started paying attention to how development economics actually functioned, but it was not this: the cold realization that the failure was deliberate, that the architects knew, and that the suffering of billions of people was considered an acceptable cost of doing business. The hotel in Dhaka is still there. So are the factories. So is the theory that one day, if we wait long enough, the wealth will trickle down.