We Promised Growth Would Lift All Boats. The Water Never Rose.

There is a photograph I keep returning to, taken in Lagos in 1987. It shows a construction site for what would become one of West Africa's first shopping malls. In the foreground, a woman sells oranges from a wooden crate. Behind her, a billboard promises 'Modern Prosperity for a Modern Nigeria.' The mall opened eighteen months later. The woman, according to a photographer's note I found years afterward, was still selling oranges from the same spot in 2004. The crate had been replaced, but the economics had not.

I know what I am talking about here. I have spent two decades covering what we politely call 'emerging markets'—a phrase that contains its own assumption about where these places are headed and why. The assumption is this: economic growth, properly structured and adequately incentivised, will reduce poverty through a process so natural it requires no deliberate intervention. Build the infrastructure. Attract the capital. Let the market do its work. The benefits, we were told, would trickle down.

They did not.

The Promise

The theory had a certain elegant simplicity. It was articulated most clearly by Simon Kuznets in 1955, refined by the World Bank in the 1980s, and embedded in structural adjustment programs across three continents. Growth would initially increase inequality—this was acknowledged, even expected—but as economies matured, as capital accumulated and wages rose, the curve would bend back. The rich would get richer first, yes, but eventually the poor would catch up. The Kuznets curve promised an inevitable convergence.

What made the theory so appealing was not just its logic but its convenience. It required nothing of the wealthy except patience. It required nothing of governments except restraint. It transformed inequality from a moral problem into a transitional phase. All you had to do was wait.

We waited. The curve did not bend.

What Actually Happened

By the early 2000s, the evidence was becoming difficult to ignore. Countries that had followed the prescription—liberalised trade, privatised state assets, removed capital controls—had grown. But the growth had accumulated in predictable places. In Brazil, GDP per capita doubled between 1995 and 2019. The income of the top 10 percent grew by 84 percent. The income of the bottom 50 percent grew by 23 percent. In India, between 1991 and 2021, the number of billionaires increased from two to 140. The proportion of the population living on less than $2.15 a day fell, but remained above 12 percent—a number that in absolute terms meant more human beings in extreme poverty than lived in the entire United Kingdom.

The pattern was global. It was not a failure of implementation. It was the system working as designed.

Milanović has spent thirty years studying global inequality. What he found, and what the World Inequality Database has now documented across 175 countries, is that growth does not automatically reduce inequality. It can—under specific conditions, with specific policies—but left to its own devices, capital concentrates. It flows to where returns are highest, which is almost never where need is greatest. The structures we built to encourage growth—low corporate taxes, flexible labour markets, privatised services—were also structures that prevented redistribution.

The Arrangement

There is a reason the theory persisted long after the evidence turned against it. It served too many interests. For international financial institutions, it justified loans conditioned on policies that opened markets to foreign capital. For multinational corporations, it legitimised labour practices that would have been unacceptable in their home countries. For political elites in developing nations, it provided cover for arrangements that enriched a narrow class while delivering just enough poverty reduction—through targeted programs, conditional cash transfers, microfinance schemes—to claim progress.

I am not suggesting conspiracy. I am suggesting incentive. When the World Bank measures success by GDP growth rather than income distribution, when credit rating agencies reward fiscal austerity rather than social spending, when trade agreements protect intellectual property more rigorously than labour rights, you do not need conspiracy. You need only to follow the logic of the system to its conclusion.

The woman selling oranges in Lagos was not waiting for trickle-down. She was living inside the mechanism that prevented it.

What the Randomistas Found

By 2010, a new generation of development economists had arrived at a different approach. Esther Duflo, Abhijit Banerjee, Michael Kremer—they did not start with theory. They started with experiments. Randomised controlled trials in villages, schools, health clinics. The questions were smaller: Does deworming medication improve school attendance? Do bed nets reduce malaria? Do conditional cash transfers change behaviour?

The answers were precise. Often they were yes. Sometimes they were no. What they were not was automatic. Poverty, it turned out, was not solved by waiting for growth. It was solved by interventions—specific, tested, funded. It required deliberate action, not theoretical patience.

The 2019 Nobel Prize in Economics went to Duflo, Banerjee, and Kremer for this work. The citation praised their 'experimental approach to alleviating global poverty.' What it did not say, but what was implicit in every study they published, was that the previous approach—the one that had dominated development policy for half a century—had failed.

Why It Mattered Who Believed It

I keep thinking about the timing. Trickle-down development theory achieved its greatest influence in the 1980s and 1990s—precisely the decades when inequality within countries began to accelerate. The Gini coefficient, which measures income inequality, rose in 73 percent of countries between 1988 and 2008. It was not a coincidence. The policies designed to promote growth—deregulation, privatisation, tax cuts for capital—were also policies that weakened the mechanisms through which growth might have been shared.

In the Philippines, GDP growth averaged 5.1 percent annually between 2010 and 2019. The number of billionaires increased from seven to nineteen. The share of national income held by the bottom 50 percent fell from 17.4 percent to 14.2 percent. In Indonesia, growth averaged 5.4 percent over the same period. The top 1 percent captured 26 percent of all income. The bottom 50 percent captured 13 percent.

These are not aberrations. They are the pattern. And the pattern was predicted—not by critics on the margins, but by economists working within the institutions that set development policy. A 2014 International Monetary Fund study found that income inequality drags down growth and shortens economic expansions. A 2018 World Bank report concluded that 'growth is necessary but not sufficient' for poverty reduction. By then, seventy years had passed since the modern development project began. Billions of dollars had been lent, spent, lost. We knew the theory was wrong. We kept using it anyway.

What It Means to a Nurse in Dhaka

In 2019, I interviewed a woman named Nasrin Akhter at a public hospital in Dhaka. She had been a nurse for twenty-two years. Her salary, in nominal terms, had increased by 340 percent since she started in 1997. Her rent, over the same period, had increased by 890 percent. The cost of rice had increased by 420 percent. She worked sixty-hour weeks. She sent her daughter to a school she could not afford because the public schools, she said, 'no longer function.'

Bangladesh, over those same twenty-two years, had become the development community's favourite success story. GDP growth averaged 6.2 percent annually. Poverty rates fell from 48.9 percent to 20.5 percent. Textile exports made the country the world's second-largest garment producer. The World Bank called it 'an economic miracle.' Nasrin Akhter, who treated the workers who sewed the garments, who lived in the city that hosted the miracle, did not feel miraculous. She felt, she said, 'like I am running faster just to stay still.'

That is what trickle-down economics produces. Not stagnation—growth. Not collapse—rising GDP. What it produces is a prosperity that belongs to someone else.

The Reckoning

I am not sure what I expected when I started reporting on development twenty years ago, but it was not this. I expected complexity, yes. I expected failure in implementation, corruption, bad governance, bad luck. What I did not expect was a system that worked exactly as designed and still produced outcomes its architects claimed to oppose. Trickle-down development was not undermined by kleptocrats or sabotaged by inefficiency. It failed because it was based on a theory that prioritised growth over distribution, efficiency over equity, and market logic over human need.

The evidence-based development movement that emerged in the 2000s offered something different. Not grand theories about how economies evolve, but specific findings about what actually reduces poverty. Not patience, but intervention. Not waiting for markets to self-correct, but acting as though poverty is a design problem that requires deliberate solutions.

The shift has been slow. The World Bank still measures success primarily by GDP growth. Credit rating agencies still penalise governments that increase social spending. Trade agreements still prioritise investor rights over labour protections. But the intellectual foundation has cracked. We can no longer pretend we do not know. The data is public. The studies are published. The woman selling oranges in Lagos is still there, and we know why.

Seventy years ago, we made a choice. We chose to believe that inequality was a transitional problem, that growth would eventually deliver justice, that the market needed only time and freedom to lift all boats. The water never rose. It pooled at the top. And every year we waited was a year someone lived in poverty that policy could have addressed, that resources could have relieved, that deliberate action could have prevented.

We tell ourselves stories in order to live. Trickle-down was a story we told to avoid a harder truth: that reducing poverty requires taking from those who have enough and giving to those who do not. It requires redistribution, regulation, taxation, enforcement. It requires treating inequality not as a stage of development but as a policy failure. The question now is not whether the theory works. We know it does not. The question is whether we are willing to act as if we know.

The photograph from Lagos is dated 1987. The woman is nameless. She exists in my files as evidence of something we already knew and chose not to see. I keep returning to it because it reminds me what was always required and what we refused to do. We could have built a different system. We built this one instead. That was not inevitable. It was a choice. And every day we keep choosing it is another day the water does not rise.