Public Subsidy / Private Accumulation: The Political Economy of Singapore’s Public Housing, by Chua Beng Huat. 2024.

For a certain kind of housing policy nerd, Singapore’s model of public housing is the stuff of legend. 76% of Singaporeans live in public housing units, commonly called HDBs after the government’s Housing and Development Board. Unlike public housing in America, which sadly evokes cockroaches and black mold, HDB units are generally modern, clean, and safe. The homeownership rate is over 90%. Homelessness is rare.

Since their institution in the 1960s, the HDBs have been the gold standard for how to house a rapidly growing urban population in a developing country. (This blog’s first post explored public housing as a development policy.) Now even rich Westerners, beset by housing crises, peer enviously on.

From afar, Singapore is often flattened into a Utopia, some kind of Platonic technocracy where the grime of politics has been abolished by fiat. Indeed, it’s tempting to view Singapore’s housing policy as a purely technical exercise, reducible to a set of savings schemes and planning policies that can be turned like dials to maximize social efficiency.

Public Subsidy / Private Accumulation, by Chua Beng Huat, a professor emeritus of sociology at the National University of Singapore, complicates this sterile picture. It’s a concise, lucid account of the politics of HDBs, and a natural entry-point to learn about Singapore.

Subscribe now

Building a Nation

When Lee Kuan Yew’s People’s Action Party (PAP) came to power in 1959, most Singaporeans lacked decent housing. A third of the population lived in crowded shophouse apartments in the city center, and perhaps another third lived in kampongs (Malay for “villages”)—the kind of informal settlements that crop up on the margins of cities everywhere in the developing world. Pigs and chickens roamed around amongst the human residents; there was no sewage or electricity; hygiene was poor; fire safety was worse. A 1961 fire in the kampong of Bukit Ho Swee destroyed 2,800 homes, and left around 16,000 people homeless.

The Housing Development Board had been created in 1960, but the disaster of Bukit Ho Swee only underscored the political urgency of addressing housing. Lee and Finance Minister Goh Keng Swee—both schooled in the British Fabian tradition—were likely influenced by Britain’s postwar expansion of social housing under Labour. Governments that failed to deliver on housing would be punished; even victorious Churchill had been kicked to the curb in a remarkable act of democratic unsentimentality.

Moreover, for young Singapore—ethnically fractious, (post-1965) newly independent, still rather poor—providing housing was far more important than just winning elections. In his memoirs, Lee gives a famous account of the HDB as central to nation-building:

… [I] was determined that our householders should become home owners, otherwise we would not have political stability. My other important motive was to give all parents whose sons would have to do national service a stake in the Singapore their sons had to defend. If the soldier’s family did not own their homes, he would soon conclude he would be fighting to protect the properties of the wealthy. I believe this sense of ownership was vital for our new society which had no deep roots in a common historical experience.

These lofty goals were constrained by the fiscal realities of a developing country. The 1959 Housing and Development Board Act stipulated that public housing had to reclaim a “reasonable return to capital invested”, meaning that it could not be sustained on public subsidy. Nominally a democratic socialist, Lee was still opposed to a strong social safety net on principle: Singapore would have no minimum wage and minimal unemployment benefits.

The HDB system needed to square the circle: addressing a real social need without breaking the bank. Two policy instruments were crucial.

The first was a scheme to make units affordable on the demand side. To keep costs down, the first HDB units (called “emergency flats”) were spartan one-bedroom flats with shared toilets and kitchens. The next tranche were more spacious two-bedrooms, which were tied to 99-year leases which simulated homeownership. Demand was tepid to start; few could afford the up-front payments. But, after 1968, when citizens were allowed to direct their contributions to the Central Provident Fund (CPF)—the compulsory social security scheme—towards an HDB purchase, demand for units shot up, from 3,000 in 1967 to 70,000 in 1968.

The other key policy ingredient was on the supply side—namely, the 1966 Land Acquisition Act, which gave the government sweeping powers to acquire any land it needed for “national development”. Critically, compensation was tied to the prices at the date of the statute, meaning that as the economy developed the state could buy land at below-market rates. The government’s principles were clear:

No landowner should benefit from development which has taken place at public expense… [and] the price paid on acquisition for land should not be higher than what the land would have been worth had the Government not carried out development generally in the area. (p. 47)

Henry George could not have put it better.

The victory of Singapore’s HDB was near-total. By the mid-1970s, “a quarter of a million people… living in degenerated city slums and another one-third of a million in squatter areas” had been rehoused (p. 55). By the mid-1980s, 85-90% of people living in HDBs were homeowners.

Victory was not costless. Chua (who grew up in the burned-down Bukit Ho Swee) is clearly nostalgic for the lost kampong communities. An older, messier Singapore, with its own illegible village traditions, has been replaced by the constructions of a ubiquitous, all-seeing state. Indeed, the HDB is the site of social engineering that would make most libertarians retch. Blocks and neighborhoods follow racial quotas based on national demographics (75% Chinese, 15% Malay, 8% Indian), to break up ethnic enclaves and promote the government’s goal of racial harmony. Eligibility for a unit means conforming to the government’s social preferences, like for married two-parent, heterosexual households.

Moreover, as Singapore got richer, the original intent of the HDB became diluted. In Chua’s account, the Original Sin was committed in 1971, when HDB owners were first allowed to resell their homes on an open market (after a required one-year timeout before they could apply for another unit). Before, resale units had to be sold at their original price, net of depreciation; now, the market was allowed to set prices. After 1979, the one-year timeout was abolished, and households began to profit from rising housing prices by quickly turning over resale units. Gradually, what began as an “emergency” good, supported by the state’s ability to seize land for the public good, morphed into what Chua calls an “asset-based welfare product”.

This shift can be seen in the changing rhetoric at the highest levels of power. Compare Lee’s description of housing as a cornerstone of nation-building, to the 2011 words of his son, Prime Minister Lee Hsien Loong:

The house is much more than a secure roof over their heads… [it] is also a major way to level up the less successful and to give them a valuable asset and a retirement nest egg. (p. 57)

Subscribe now

A Polity Without Politics?

The coalition of homeowners proved to be electoral dynamite for Lee’s People Action Party. 1963’s Operation Cold Store (see my previous post) broke the back of the opposition through force; but the PAP’s promise of fast economic growth, clean governance, and public housing was also genuinely popular. Over sixty years, the PAP has won every election—including thirteen years (1968-81) of parliaments without a single opposition member. Because of this continuous success, it’s easy to misread Singapore as a polity without politics, and to see the PAP as gliding through elections as placidly as a duck over water.

But there’s some furious paddling going on beneath the surface.

Chua’s careful survey of Singaporean electoral history reveals that housing is a recurrent cause of political discontent. In 1981, Worker Party candidate J. B. Jeyaretnam won a chance by-election in Anson Constituency to become Singapore’s first opposition MP in sixteen years. This minor setback for the PAP can be traced directly back to housing: over 700 families in Anson had been cleared from their homes to make way for a new port facility, and the government had refused to prioritize residents for resettlement.1

In 1984, after the opposition Singapore Democratic Party also won a seat in Potong Pasir, the Minister of National Development threatened that any housing estate in a constituency that voted against the PAP would be last in line to be served by the HDB. This punitive policy stood for almost thirty years, excluding opposition constituencies from the HDB’s annual billion-dollar upgrading program. The threat was clear—in the words of Prime Minister Goh Chok Tong in 1996, “your estate through your own choice will be left behind. They’ll become slums.”

The 2011 general election—the greatest scare for the PAP in forty years—is perhaps the starkest case of housing intruding on national politics. Because Singapore had weathered the global recession relatively well, foreign capital and high-skilled workers had flowed in, driving up housing prices. The opposition called for reforms, like pricing new units based on costs, to make housing more affordable. The government’s response was tone-deaf—the Minister of National Development (who oversees the HDB) argued that such reforms would be an “illegal raid” on the national reserves, since it would lower the rents the government could charge private developers on public land. The PAP lost seven seats, their worst electoral performance since independence.

To their credit, the PAP was quick to respond to voters’ rebuke. A new Minister of National Development, Khaw Boon Wan, introduced a set of “market cooling” measures: he untethered the price of new HDB units from the market price, increased annual housing production to 25,000 units, and symbolically restricted expatriates’ ability to speculate on public housing units. These measures were broadly successful at lowering prices, and in the 2015 elections the PAP won a resounding victory, with its vote share rising from 60 to 70 percent.

Subscribe now

Global Singapore

This brings us to the present day, where Singapore’s peculiar housing system has become a model for the world. This is true in two ways.

The first is that the Singapore housing model has become a widespread object of envy. It would be malpractice to ignore how well Singapore has managed rising housing costs compared to other rich countries. As of 2024, the average home costs 3.8 times the median income, compared to 4.8 times in the United States, 9.7 times in Australia, and 16.7(!) times in Hong Kong (Bird 2025, p. 238). Housing ownership is broad-based: 25% of Singapore’s housing wealth is owned by the bottom 50% of asset owners (Bird 2025, p. 252). This performance has won admirers across the West, particularly in the Yes In My Backyard (YIMBY) movement. Another eager pupil is China, which has built whole model cities with Singaporean advice.

But Singapore is instructive to the world in another sense. There is trouble in YIMBYLand: since the Covid pandemic, the price of resale HDBs has risen over 50%. The fact that price increases are less bad than in peer countries is cold comfort to voters; Singapore is not exempt from the generational and class conflict over housing affordability that is playing itself out across the developed world. The old, who depend on housing as a retirement asset, want prices to go up. The young, who want to move out and start families, want prices to go down. Both groups cannot be satisfied at once.

The government recognizes the problem, but its muddled response reflects its conflicting allegiances. One scheme is to give cash grants to subsidize first-time homebuyers, which in mainstream economics is a little like pouring gasoline onto the fire. A better approach is to boost supply by building more HDBs—but this requires patience. Singapore builds quickly (the HDB added 73,000 units from 2021-4), but not fast enough to keep up with the demands of its citizens.

Ultimately, Chua argues, the PAP’s loyalties lie with the HDB homeowners who form the foundation of their voting base, rather than with younger, new entrants. It simply cannot allow housing prices to fall significantly. Having tied people’s welfare to the performance of the market, the PAP is now beholden to the political consequences.

Allow me to close with a digression. Despite the strangeness of Singapore’s institutional structure, the political economy resembles nothing other than the United States, where, in the absence of a strong welfare estate, housing remains the most important retirement asset, and homeowners a politically powerful group. One person who has notably grasped this logic is Donald Trump, who said in February:

Every time you make it more and more and more affordable for somebody to buy a house cheaply, you’re actually hurting the value of those houses. And I don’t want to do anything that’s going to hurt the value of people that own a house, who for the first time in their lives are walking around the streets of whatever city they’re in, very proud that their house is worth $500,000, $600,000, $700,000.

The conversation around housing feels to me eerily reminiscent of that of global trade around 2015. On one side, there is the policy consensus: every supply-and-demand graph, every academic journal, every white-paper, every respectable economist you’ve ever heard of. On the other, you have a certain Manhattan property developer with a preternatural sense for American grievance. That he has identified homeowners as a potential constituency should give us YIMBYs some pause. Even in seemingly quiescent Singapore, the politics of housing is fraught enough to make the mighty PAP quake in its boots.

How Africa Works: Success and Failure on the World’s Last Developmental Frontier, by Joe Studwell. Atlantic Monthly Press. 2026.

When Joe Studwell’s How Asia Works came out in 2013, it was a book deeply out of consensus. In an age of randomized control trials and micro-interventions, it resurrected macro policies—land redistribution, industrial policy—that had virtually disappeared from mainstream development economics. Moreover, it returned East Asia, the only developing region in the world to successfully make the postwar climb out of poverty, back to the center of debate.

Thirteen years on, Studwell probably deserves some kind of triumphal march. Industrial policy is back in a big way. Through How Asia Works’s influence on Noah Smith and a host of bloggers, a generation of young tech-adjacent males were primed to rant about semiconductor subsidies at parties.

I am no exception. Reading How Asia Works was a formative intellectual experience for me—a jolt out of the mathematical slumber of PhD coursework. I have a complex relationship with the text (more on this in a moment), but I still recommend it effusively to anyone who wants to learn about East Asia.

Now, thirteen years later, Joe Studwell is back. How Africa Works aims to do for Africa what he achieved for Asia—becoming the natural first stop for readers who want to learn about the economics of the continent.

Subscribe now

A Dismal Inheritance

The first part of How Africa Works addresses the perennial question: why is Africa poor?

Historically, low population density, induced by pests like the tse-tse fly, discouraged the formation of large urban centers. The slave trade—first Arab, then Western—further depopulated the continent, breaking down social bonds. When Europeans arrived in force in the 19th century, they did colonialism on the cheap, with few policemen and even fewer schools. (I wrote in depth about this a few months ago.) Unlike (say) the Japanese in Korea or Taiwan, the colonial state rarely penetrated much farther than the capital or key ports, leaving governance in the vast hinterland to invented or upjumped chiefs.

Decolonization left a dismal inheritance. In spite of superficial similarities in GDP with East or South Asia, Africa had far more problems on its plate. Levels of education were far too low to sustain an effective civil service, let alone communities of engineers or innovators. Incoherent states encased in inappropriate borders meant Africa’s founding fathers had to stitch nations together from unrelated ethnic groups.

Studwell’s diagnosis of Africa’s problems is steadfastly conventional, leaning heavily on the academic consensus established by Jeffrey Herbst, Robert Bates, Nicolas van de Walle, Leonard Wantchekon, among others. This is no dig; Studwell is an elegant synthesizer. I have some quibbles around the margins—the underrating of precolonial Africa reflects some lingering Western state-centric bias—but as a diagnosis for Africa’s poverty this is a far richer, textured, and more accurate account than the memelike “extractive institutions”.

Subscribe now

Four Success Stories

Having set the scene, Studwell turns to four successful case studies: Botswana, Mauritius, Ethiopia, and Rwanda. This itself is a refreshing approach to economic analysis of Africa, which so often wants to dwell on failure. Unlike Taiwan or South Korea, none of these countries is an unqualified developmental miracle, but their relative success provides clues to how an African economic transformation might take place.

Botswana

Botswana is Studwell’s poster child for a successful democratic developmental coalition. (For this reason, it featured heavily in Acemoglu and Robinson’s Why Nations Fail as an example of “inclusive institutions”.)

Under the sound leadership of Seretse Khama, local chiefs were carefully co-opted at independence and the Botswana Democratic Party built up into a genuine national force. Khama also created a capable civil service, initially staffed by remaining Europeans, but gradually Africanized with sterling Batswana talent. This meant that when diamonds were discovered just around independence, the windfall was carefully managed, avoiding the worst effects of Dutch Disease. These mining revenues helped raise Botswana to upper middle-income status, making it the fourth-richest country in continental Africa.

Botswana’s chief failing, in Studwell’s view, was adhering too much to responsible policy orthodoxy—i.e., not enough industrial policy. There was no vision for large-scale industrialization, no coherent plan to create large numbers of factory jobs. Moreover, the political dominance of large cattle owners (Botswana was a society of pastoralists rather than farmers) meant that redistribution was never in the cards. The result is a relatively rich society, but one that is highly unequal.

Mauritius

Mauritius, which is often not thought of as an African country, is perhaps the most unusual choice. An uninhabited island before Dutch colonization in the 17th century, its ethnic makeup of Indians and Creoles resembles the Caribbean more than continental Africa. Moreover, Mauritius became independent in 1968 at an income level that most contemporary Africans would envy (see chart above).

Nonetheless, Mauritius’s developmental record is impressive. Originally a sugar colony, a tax on sugar receipts was used to funnel landowners’ capital from agriculture to manufacturing. In the subsequent manufacturing drive, powered by the exports of apparel and textiles, GDP rose 6% a year. With egalitarian, broad-based growth, poverty was virtually eradicated.

However, Mauritius was unable to make the leap from garments to higher-value manufacturing, and the sector’s share of GDP has since halved from over 20 percent to just 11 percent by 2020. Alongside Seychelles, it is one of only two African countries ranked “very high” on the UN’s Human Development Index.

Rwanda

Ethiopia and Rwanda, as recent developmental darlings and conscious emulators of the East Asian example, are perhaps the least surprising inclusions in Studwell’s list.

Under President Paul Kagame, Rwanda has explicitly modeled itself after Singapore (including Lee Kuan Yew’s authoritarian tendencies). At first blush, this struck me as absurd: Singapore is an island state on the crossroads of the world’s richest sea lanes; Rwanda is a landlocked country in poor central Africa.

Studwell’s account convinced me there is an economic logic to this strategy. The high cost of road transport means that importing goods into Central Africa is prohibitively expensive. Rwanda does not necessarily need to compete with the world; by delivering on infrastructure projects and maintaining rare political stability, it can attract investment as a kind of entrepot to Africa’s Great Lakes. Under this formula (with perhaps some slight fudging of the numbers), Rwanda has maintained impressive 7% growth for the past decade.

The big question surrounding Rwanda is if the growth coalition can hold together. Nowhere else in Africa is the tragic legacy of ethnic division more apparent; the present Kagame regime took power by overthrowing the perpetrators of the infamous 1994 genocide. Rwanda’s military involvement in the Eastern Congo, which represents both a source of raw materials and a lucrative market of 30 million, adds a further dark cast to its developmental success.

Ethiopia

It is Ethiopia that comes the closest to achieving all parts of Studwell’s formula. As a country of 135 million people, it has the scale to set a major example to the world and to take a serious bite out of Africa’s poverty all on its own.

Meles Zenawi, prime minister from 1995 to 2012, was an avid student of East Asia. (His thesis outline is available online; for any economist with a wavering faith in the power of ideas, read the bibliography.) Under his leadership, the Ethiopian state invested heavily in agricultural extension and irrigation, improving the yields of smallholder farmers. It began (with Chinese support) building industrial parks to support an export manufacturing base. Most ambitious of all, it began work on the Grand Ethiopian Renaissance Dam, one of the largest hydropower projects in the world, to find a permanent solution to Ethiopia’s energy woes.

No student of How Asia Works could have done better. Had How Africa Works been published before November 2020, it’s easy to see how a celebration of Ethiopia might have occupied most of the book. But the outbreak of civil war derailed Ethiopia’s progress, demonstrating the continuing risk of ethnic conflict to the prospects of economic growth.

Subscribe now

Mashamba Na Viwanda

Unlike East Asia, Africa has no unqualified economic miracles to point to. The result is a book that is more diffuse in its rhetorical impact than How Asia Works, but also one that is perhaps more realistic about the constraints. Some of the swaggering confidence that marked the Asian Triple Growth Formula is gone.

Nonetheless, Studwell insists that universal prescriptions still exist:

… despite the radically different context, I have found that the policies that were most effective in East Asia in producing economic transformation are the same ones that have worked in the handful of cases of early success in Africa. In this respect, there is no African exceptionalism.

As a recap, these policies were smallholder agriculture with state support, industrial policy to support export-led manufacturing, and tight government control of finance to support all these aims. To these three, Studwell adds the extra ingredient of a “developmental political coalition”—taken largely for granted in the relatively homogenous, authoritarian states of East Asia, but far from table stakes in ethnically fractious, democratic Africa.

I’m no expert on any of the four countries Studwell discussed. But let me comment on two of Studwell’s key pillars from an economic lens: agriculture and manufacturing.

Agriculture

Like in How Asia Works, Studwell advocates for smallholder farming in Africa, citing the familiar evidence that small farms grow more crops per acre than big ones. (In jargon, this is the “inverse farm size - yield relationship”.) In theory, then, redistributing land from big landowners to smallholders should improve aggregate productivity.

Smallholder farming may be desirable for political, distributional, or social reasons. In most societies, owning your own plot of land naturally has enormous psychological value. In Kenya, for instance, having a rural shamba is a source of social status and, in urban downturns, acts as a form of social insurance. Regimes that ignore this basic fact invite unrest: the anti-communist regimes of East Asia likely had to do some form of land redistribution or risk being thrown into the sea.

But on the narrow point of efficiency I am more agnostic. I mentioned earlier my complex relationship with How Asia Works; my academic work finds that the major land distributions in Taiwan and Mainland China had smaller yield effects than previously thought. Having pondered this question for years, it now seems to me simplistic to expect there to be a universal Platonic relationship where the smaller the farm, the higher the yields. Far more likely that this relationship depends on the crop, the soil, and the available infrastructure. Wheat yields in Europe, for instance, seem to be the highest on large farms, while rice yields in Asia can grow on tiny plots with the near-endless application of labor.

But on Studwell’s broader theme, of a renewed developmental focus on agriculture, I am in complete agreement. African smallholders, ignored by their states and deprived of support, are struggling. According to the best available data, stretching from 2008 to 2019, both smallholder yields and total factor productivity have been declining by around 3 to 4% a year.1 It’s difficult to envision lifting 460 million Africans out of extreme poverty without improving the meager returns from their primary occupation.

Manufacturing

The other noteworthy component of the Studwellian recipe is a heavy emphasis on growing manufacturing, fostered by state industrial policy.

What’s so special about manufacturing? Studwell leans heavily on an influential 2013 paper by Dani Rodrik, who argues that manufacturing possesses the unique property of “unconditional convergence”. Unlike other sectors, manufacturing in developing countries appears to catch up quickly to the global frontier of productivity. Intuitively, because most manufactured goods are tradable, manufacturing firms are more exposed to the pressure of international competition, forcing them to innovate; moreover, manufacturing processes (compared to, say, crop growing practices) are readily transferrable across borders.

I was long a True Believer in this thesis, but have recently had my faith shaken. New empirical work, forthcoming in the American Economic Review: Insights, suggests that unconditional convergence in manufacturing may partly have been an illusion of the data. (In that paper, somewhat cheekily, it turns out that agriculture and services display convergence, but manufacturing does not.)

Of course, no one’s worldview is really determined by a paper based on a few cross-country regressions. (Even one by Dani Rodrik.) What convinces most is how central manufacturing was to the East Asian miracle, still the only region of the world to ride the escalator up from poverty to riches.

This relates to a deeper problem with the prospects for African industrialization: namely, that industrialization never happens in a vacuum. A successful domestic manufacturing base is a product not only of your own industrial policy, but global market conditions and the strength of your competitors.

One obvious risk is automation, which threatens the manufacturing sector’s absorption of labor, and may help keep Chinese factories globally competitive despite rising wages. Studwell quickly brushes off these concerns (“[the] labour cost in a country like Madagascar is US$65 a month… the cost of an advanced industrial robot in the apparel sector is over US$100,000”). In my view they deserve deeper inspection.

Moreover, even if Studwell’s right, Africa has strong competitors in the race to claim China’s manufacturing share: South and Southeast Asia, with their large urban populations and increasingly capable states. Studwell notes optimistically that Africa has finally caught up to the educational attainment of East Asia in 1960; he fails to note that South and Southeast Asia have long exceeded that level.

Subscribe now

A Bias For Hope

Longtime readers of Studwell’s writing—from 2003’s The China Dream to 2013’s How Asia Works to the present volume—will know that he has a strong contrarian streak. The China Dream was notably downbeat about China’s development prospects just as the largest export boom in history was getting started (p. xii: “the economic foundations of contemporary China have been laid on sand and [are] constructed from the kind of hubris that drove the Soviet Union in the 1950s”). How Asia Works was stridently dirigiste, right at the high-water mark of the Neoliberal Age.

By contrast, Studwell sounds unusually optimistic about Africa, where, post-aid cuts, the pendulum of international opinion has swung decisively towards gloom. State-led improvements in health and (to a lesser extent) education, supported in part by international aid, have eroded some of Africa’s historical disadvantages. Most of all, the demographic boon of the world’s youngest population will give growth efforts a brief but powerful tailwind.

As an analysis of what makes countries grow, the Studwellian formula is of course incomplete—but, with 54 countries and 1.6 billion people, how could it not be? What makes Studwell nonetheless compelling to read is his steadfast belief that poverty is a product of policy decisions. Analytically, this is of course not quite right: as the first part discussed, strong historical and geographic factors condition what’s possible. But, for a practitioner, such belief—what Albert Hirschman once called a bias for hope—is surely a necessary condition for action.

Development is ultimately an act of imagination, of envisioning what’s not yet there. Sound policy requires that these visions be supported by durable political coalitions and exist within states’ capabilities. (A latent motif of the book is eager states overreaching with megaprojects, in a vain attempt to leapfrog their peers.) But even the mixed success of import substitution industrialization or the follies of incomplete irrigation megaworks seem preferable to the status quo of seeking rents while sitting on one’s hands.

On one final point I am in wholehearted agreement.

At various points Studwell discusses “demonstration effects”: the positive influence one country can have on peer states. Demonstration effects are essentially impossible to falsify in the modern language of econometrics, but are unmistakeable in the real world. (If you disagree, take a look again at Meles Zenawi’s library.) The world really only has two industrial clusters: one began in 18th century Britain, and grew to encompass most of Europe and its colonial offshoots; the other started in Meiji Japan and spread throughout East Asia. In both cases, culturally similar neighbors saw what was possible and copied the recipe.

In one sense, this is a note for pessimism. If history is any guide, the great global factory complex will first stretch down from China through to mainland Southeast Asia and Indonesia, and westward through Bangladesh and India, long before it ever reaches Africa. But in another, it sounds a note of hope. If even one African country manages to sustain the kind of broad-based growth that Studwell describes, it could do for its neighbors what Meiji Japan accomplished for the rest of Asia. It may only take one resounding success to shatter the illusion—fed by sixty years of disappointment, egged on by lingering prejudice—that Africans are incapable of achieving economic prosperity.

A story getting covered in The New York Times is usually a lagging indicator: a good way to learn that something horrible has happened, long after you could conceivably profit from the news.

So it was an ill portent when, earlier this month, The Times finally wrote about what’s been going wrong with the job market for economists:

… Universities and nonprofits have scaled back hiring amid declining state budgets and federal funding cuts. At the same time, the Trump administration has laid off government economists and frozen hiring for new ones.

Tech companies also have grown stingier, and their need for high-level economists — once seemingly insatiable — has waned. Other firms have slowed hiring in response to the economic uncertainty introduced by President Trump’s tariffs and the possibility that artificial intelligence will replace their workers, even if those workers have a doctoral degree.

On a normal day, this blog is about the poorest people in the world. The median economist made $115,440 in 2024. Until recently, this was a profession with an employment rate of 100%. Most people might ask: why should we care?

Of course, I’m selfishly interested—it’s a bit harrowing when your chosen profession is nuked from orbit. (With the gutting of USAID, development economics has been particularly hard-hit.)

But there are more dispassionate reasons to care. Among academics, Econs are unique for the power we wield. We control the financial weather; we recommend bombing campaigns to presidents. The market for Econs (with all its institutional strangeness) is more than just a potential curiosity for anthropologists; it puts its unmistakeable stamp on the systems of knowledge and expertise that, without exaggeration, have governed the postwar global order.

This post explores this looming crisis, in two parts. Is this down labor market really the Twilight of the Econs? And what does this mean for Economics as a science, and how Economics acts upon the world?

Subscribe now

What’s Going On With The Econ Job Market?

Even before the cuts to the National Science Foundation and the National Institutes of Health, the economics job market was cooling.

The number of openings each year from Job Openings for Economists (JOE), the American Economic Association’s centralized job board, have fallen precipitously over the past two years:

In the halcyon days of 2015-19, openings on the economics job market hovered at around 1900 per year. In 2020, Covid was a major shock, but the market bounced back quickly in 2021 and 2022. Since then, though, the market has clearly been in a funk. 2023, my job market year, saw a sudden dip in postings. 2024 was even worse, with openings falling 16% lower than the 2015-19 average.

At the time, the sudden fall in 2023 seemed mysterious—it was an otherwise healthy year for the broader labor market. In hindsight, it seems like the 2021-22 recovery masked some underlying weakness. The 2020 job market had 500 fewer openings than the 2014-19 average; 2021 and 2022 together produced only around 100 more jobs than the 2014-19 average. In other words, the recovery never made up for the pandemic; by this crude logic, around 400 economist jobs were “destroyed”.

Economists like to think of themselves as exceptional (more on this in a minute), but this slowdown is part of a general decline in academic hiring. Disciplines like history and anthropology have been suffering for some time, with the production of PhDs outpacing the number of tenure-track jobs. Life in STEM is not much better; by one calculation, only 13% of PhD graduates will land a tenure-track position in the United States. The drying-up of tenure lines has led to the “adjunctification” of the academy, with more university courses taught by lower-paid, contingent workers.

I still haven’t seen a convincing empirical breakdown of how much over-supply versus under-demand have driven the breakdown of the academic market. But surely an important overarching factor is the demographic decline in the number of college-age adults, a trend that will only accelerate over the next decade:

And of course, all of this decline occurred before the litany of disasters that have recently hit the Econ job market. In May, Jerome Powell announced that the Federal Reserve—perhaps the largest employer of economists in America—would cut its workforce by 10%. The federal government has frozen hiring, as has the World Bank. Hit by the dual threat of fines and looming cuts to federal funding, Harvard, MIT, the University of Washington, Notre Dame, Northwestern University, among others, have announced hiring freezes and budget cuts.1

The 2025 market has not yet started in earnest, but the early indications are bad. Job postings are currently below those at the same point in 2023 and 2024:

A potential 20% drop in the job market is a serious historical shock, and a moment to pause and take stock of what is at threat. What does it mean to be an economist in 2025?

Subscribe now

What Does It Mean To Be an Economist?

The best answer to that question remains Marion Fourcade’s 2010 classic Economists and Societies, a comparative sociology of the economics profession in America, Britain, and France.

Fourcade’s central point is that, far from being a universal science, the profession and practice of economics reflect the societies in which economists live. This is news perhaps only to economists. But “mere” description can be quite illuminating, because it reminds us that things do not have to be this way.

In Britain, midcentury economics was the province of gentleman scholars, who often lacked formal credentials but influenced policy through intra-elite personal networks. (Keynes and Kaldor are exemplars.) In France, postwar economics developed from the tradition of state engineers, growing out of a cadre of public managers and technicians trained by the administrative state. It was in America that the Economics PhD first emerged as a professional credential—a union card if you wanted to be a serious policymaker.

How did this come about? Fourcade describes how, in the early 20th century, demand for technical expertise and the forces of academic hierarchy interacted to produce a discipline with an unusually tight grip on the bounds of orthodoxy:

In the opening decades of the twentieth century, public officials in the American administrative institutions at the local, state, and federal levels created demand for unpoliticized, technical expertise mostly drawn from the academic professions… In the absence of an elite of public technocrats… they explicitly relied on academic economists to carry out technical tasks… This created a strong institutional basis for an economics profession that is profoundly rooted in the academic world, and in the imperatives of empirical relevance and scientific quantification. A small elite of professors within top universities exerts efficient control over the rest of the field and defines the boundaries of what constitutes acceptable economic expertise. Commanding widespread respect (both nationally and internationally) from the lower strata of the field, it also holds institutionalized access to prestigious appointments in government and international organizations (pg. 8).

Let me offer a couple of observations, particularly for non-economists, to underline the strength of the academic economic hierarchy—and its connection to the levers of power.

According to a 2022 study in Nature, 6% of academic economists place at a higher-ranked school than their PhD, third-last among all fields after Classics & Religious Studies, and less than half the average for social sciences:

Concentration only gets higher at the top of the profession. By my count, only one out of the last 25 presidents of the American Economic Association was from a program outside the top ten—Tom Sargent, at NYU. By comparison, the American Political Science Association had twelve.2

Why is prestige so concentrated in a few top departments? Surely a large part of this is that economists at top departments are smart and work frankly insane hours. But surely another part is institutional. Promotion and tenure are largely determined by publishing articles in prestigious journals, but editorial boards are dominated by professors at top schools. In the extreme, two of the Econ’s Top Five are “house” journals, Harvard’s Quarterly Journal of Economics (QJE) and Chicago’s Journal of Political Economy (JPE). Bethmann et al (2023) find that QJE articles by authors with ties to Harvard and MIT are cited systematically less than those by authors from other top 10 schools. If citations are a crude proxy for article quality, this suggest some measure of home bias—as if the Yankees got to change out of their uniforms mid-game and call balls and strikes.

(Interestingly, JPE articles by authors affiliated with Chicago are cited systematically more.)

This kind of minutiae matters because, as Fourcade describes, academic hierarchy is tightly intertwined with the apparatus of American government. The most powerful instance of this is, of course, the Federal Reserve, which has largely been staffed and led by PhD economists from top programs.3 With its near-instantaneous influence on global financial conditions, there is no remote equivalent for political scientists or sociologists. (It’s worth meditating on how strange a sociology equivalent in government would seem—which, on further thought, makes the Fed seem even more remarkable.)

Then, on top of the Fed, there is the glittering march of star advisors to the White House—the Galbraiths and Rostows, the Furmans and Summerses. The joke was that every time a Democrat won the Electoral College, Harvard’s Littauer Center would have to start looking for guest lecturers; for a time, with Marty Feldstein and Greg Mankiw, this was even true for Republicans as well. At the risk of being charged with first-degree name-dropping, my own experience in academia reflects this extreme concentration. As a Harvard and Berkeley grad, the last four chief economists of the International Monetary Fund were all at one time my professors. (Incidentally, Marion Fourcade is married to Pierre-Olivier Gourinchas, the current Chief Economist.)

That institutional power and academic prestige should be so intertwined deserves some pause.

Subscribe now

The Electric Kool-Aid Market Test

But the technocratic-academic nexus (say that five times fast) is not the only thing that makes American economics so unique, or so hegemonic. Its second characteristic might be called the Market Test. Fourcade again:

The combination of the definition of the economist by a technical, measurable form of competence, of a certain consumer orientation within academic institutions, and of institutionalized competition among professions has produced a situation where economic knowledge has been more “market-oriented”… than elsewhere…. The inscription of American economics in the market system has served as a basis for a gradual expansion of the profession’s jurisdictional claims, through the commercialization of economic ideas and tools (pg. 9).

Put crudely, the one thing economists could always lord over other academic disciplines was their employability. During a golden age that spanned somewhere from prehistory to around 2019, economists were in hot demand from consultancies, investment banks, and tech companies. This raised wages (economics professors make more than other academics), and soaked in talent which otherwise might have gone to other disciplines. It speaks to the level of security (and entitlement) of that era that the running advice to prospective PhDs was that if you flamed out of the academic job market, you could at least settle for a job at Facebook.

With the fall in job postings, we Econs are sounding a lot less smug.

Between threats to Fed independence, cuts to academia, pullbacks in industry, the institutional features that Fourcade argues enabled the ascendance of American economics—the technocratic deference to economists in government and the demand for economic tools in the private sector—are all currently at risk, in a way that they never were in the 20th century.

An additional complication, which raises the stakes, is that the American system of economics has conquered the world. Fourcade’s book came out in 2009. Since then, much of the distinctiveness of the national economicses in her account has been flattened out. The Paris School of Economics has emerged as one of the world’s leading departments, regularly placing students in top American jobs. Oxford and Cambridge’s training, after some resistance, has converged with the American model. (LSE was an early adopter.) The European academic job market and the American job market are now joined at the hip; applying for jobs in both is trivial.

In short, as usual, America won.

But this convergence now exposes the profession to common shocks. My sense is that supply of US-style, non-US trained economics PhDs has recently increased. (Unfortunately I wasn’t able to wrangle the European data to confirm this; if you can get this to work please reach out.) If those positions were created with the expectation that American demand would soak up supply, the market has a rude awakening on its hands. The looming crisis of the Econs is not American, but global.

Subscribe now

What Is To Be Done?

In any crisis, there’s a temptation to blame impending downfall on one’s sins. Perhaps economics got too woke and invited right-wing backlash. Or perhaps it got lost in a fog of frictionless theorizing, and was caught flatfooted by the surprises of the Great Recession, the China shock, the Trump election, and the Trump election.

Close readers of this blog will note that I’m fairly ambivalent, even critical, about the overall impact of my field. But none of these theories quite ring true to me.

Economics may have given birth to neoliberalism, but—if Fourcade’s account is right—its ascendance was also the product of social forces far larger than any academic discipline. The expanding 20th century American state needed a source of technocratic expertise in economic governance, and academic economists happened to elbow out the other contenders—the lawyers, the businesspeople, the political scientists, the sociologists—to supply it. Put simply, nothing says that the economic policy has to be made by professors. The mid-21st century may chart a different course.

What about the more urgent question I started with. Do I think the PhD job market will bounce back?

Prognosticating too eagerly is a good way to land yourself a place in the Irving Fisher Hall of Forever Being Remembered For Having Said One Stupid Thing.4 A 16% fall in jobs, while devastating, is not yet apocalyptic. (By comparison, historian job ads have fallen closer to 50% since their 2008 peak.) But for things to get better requires a causal mechanism. Reinstating science and academic funding would require either Republicans to reverse their stance on the value of higher education, or for Democrats to win back the Senate. I don’t have a great sense of if either will happen.5

In this case, prediction may be less important than preparation. Placement chairs need to own up to the harshness of the labor market, and urge job market candidates to start prepping non-academic options. (Better yet, admissions chairs should consider paring back cohort sizes.) Candidates who would like a proper job after graduating should be networking, hard. And candidates resolutely committed to academia should steel themselves for long hibernations as post docs, to wait out the coming storm.

On second thought, I will venture one dark prediction, for at least the near future.

We’re going to see a lot more Substacks.

PS: If you’re on the market this year, and would like to chat, please reach out. You can DM me on Twitter at @oliverwkim.

If you liked this article, you might also enjoy:

• A historical deep-dive into the role of economists—and statistical malpractice!—in prolonging the Vietnam War

• The early life of Albert Hirschman—resistance hero and development economist—and his remarkable Strategy of Economic Development

• “You Can Even Kill Them”: a UN development advisor’s role in shaping Singapore’s development—and its authoritarian turn

During my PhD, a recurrent complaint I had when studying Macroeconomics was that it really should be called Rich World Macro. Independent central banks, solidly anchored inflation expectations, and the steady rule of law are institutions particular to a small club of advanced economies. A more realistic and globally representative macroeconomics would engage with the institutional messiness of developing contexts—most of all, the political-economic pressures that most central banks face.

Somewhere, as I griped, a monkey’s paw curled.

As we react to Monday night’s news, I thought a quick post on what we know about central banks in developing countries would be a useful intervention.

Subscribe now

The March of Independent Central Banking

Davide Romelli at Trinity College Dublin has made a career out of measuring central bank independence around the world. His global dataset on de jure Central Bank Independence will, I’m sure, soon be much-downloaded.

Romelli (2024) documents that, since the 1990s, the world has seen a general trend towards central bank legal independence:

The peak of these reforms was in 1998, when the European Central Bank took over monetary policy in the Euro Area. But this reform process was not restricted to poor countries. In the age of the Washington Consensus, developing countries also increasingly adopted reforms towards central bank independence, captured here by Romelli’s Central Bank Independence index:

Surely no small part of central bank reform has been, to use Lant Pritchett’s phrase, isomorphic mimicry—the de jure replication of the form of rich-world institutions, if not quite their function. With the United States trending in the opposite direction, I’m doubtful that this wave of reform will be fully reversed, but it seems like a fair bet that would-be autocrats around the world will look at Trump’s growing influence over the Fed and start getting ideas.

Subscribe now

You Don't Know What You've Got Till It's Gone

But, beyond mimicry, another reason central bank independence has become so widespread is that it actually seems to, well, work.

Athanasopoulos, Masciandaro, and Romelli have a well-timed April 2025 working paper showing that central bank independence has a long-run negative effect on inflation, averaging more than 10 percentage points over 10 years:

Similarly, Athanasopoulos, Fraccaroli, Kern, and Romelli have a working paper showing that higher restrictions on a central bank’s ability to lend to the government (i.e., greater independence) means higher debt-to-GDP and lower interest rates in developing countries:

The authors' intuition is that central bank independence appears to help poorer countries sustain higher debt loads without increased borrowing costs—which would make sense if one’s constrained oneself against the temptation just using the central bank to finance the debt.

The results of these working papers are far from final, so they’re worth taking with a grain of salt. (Arguing for a true, exogenous “natural experiment” in a time series macro setting is always a tricky business.) My gut sense is that an effect size of 10 percentage points of inflation seems quite high.

It’s also important to note that, up to this point, we’ve considered only de jure reforms. Romelli’s Central Bank Independence Index measures legal characteristics like who gets to appoint the central bank board, the level of required disclosure, etc. These do not capture all the other kinds of malfeasance and informal shoulder-leaning that can transpire outside the letter of the law.

A good example of that, in a working paper that will surely be getting a lot of press, is Drechsel (2024). Using a remarkable dataset compiled from presidential diaries, Drechsel analyzes the effect of political pressure on Fed behavior—in particular, Nixon’s 1971 pressuring of Fed Chair Arthur Burns to ease policy to aid his 1972 reelection. According to Drechsel’s data, Nixon met Burns 34 times(!) in the latter half of 1971. (By comparison, Bill Clinton met with Fed officials a total of 6 times during his presidency.)

Burns appears to have complied—in the Romer and Romer (2004) dataset of monetary policy shocks, the Fed Funds Rate was hit by a negative 150 basis point shock. In a straightforward event study, Drechsel finds a significant ~2 percentage point increase in the price level several years after the shock:

Embedding this event study into a model estimated on the broader sample of Fed-Presidential interactions, Drechsel finds that “exerting political pressure 50% as much as Nixon did, over a period of six months, permanently increases the U.S. price level by more than 8% after several years”.

These graphs remind us that the effects of interfering with central banks, like all monetary policy, can appear with long and variable lags. But if the empirics of central bank independence are to be believed, expect higher inflation and higher borrowing costs on the horizon.

Subscribe now

Silver Linings?

That’s all rather gloomy. Do I have a silver lining to offer?

In his survey of central banking reforms, Romelli (2022) offers one. Central bank reforms, he argues, are an endogenous development after periods of high inflation—an institutional solution to a perennial economic problem.

If we go further down the road of lessened Fed independence, the good news is that, if developing countries are any guide, that these institutions can re-emerge. The bad news is that it may take long bouts of economic pain to remind ourselves why we had them in the first place.

It’s now become an economic truism that China underconsumes: that Chinese households consume far less than those from countries with similar incomes.

Here’s the problem, illustrated in this New York Fed graph of middle-income countries’ consumption versus their GDP per capita:

If we trace the red Chinese triangle up to the line of best fit which prevails for other countries, China’s consumption is too low by around half.

Underconsumption is far from an econ nerd curio. It’s the stuff of geoeconomic—and, dare I say it—world-historical importance. By underconsuming and oversaving, the theory goes, China can pump more of its GDP into investment, building the factories and laying down the infrastructure that reinforce its dreaded production “overcapacity”. And since the total amount of savings and investment in the world have to equal, in this viewpoint Chinese underconsumption has a bizarro reflection in American overconsumption, whose symptoms are chronic trade deficits and the decline of manufacturing.

Addressing foreign imbalances has recently, if you’ve noticed, become a major theme of American economic policy. But it’s worth dwelling a bit on what underconsumption means from the Chinese side. What goods are Chinese people not buying enough of?

As it turns out, by following this trail and employing a bit of economic detective work, we can shed new light on the underconsumption thesis.

Subscribe now

Turning Up the Volume

Last month, the China Underconsumption debate was jolted back to life by a report in the newsletter Pekingnology, by Yu Fei and Guo Kai of the China Finance 40 Forum:

Pekingnology

China's Consumption Is Not Nearly as Low as It Appears: CF40 Policy Brief

There is a broad international consensus among economists, policy analysts, and multilateral institutions that China’s level of household consumption is significantly lower than what would be expected for a country of its economic size and development…

Read more

10 months ago · 35 likes · 5 comments · Zichen Wang

The heart of CF40’s analysis is this table:

The table is dense, but its message is simple. If we sidestep all the technicalities of cross-country price conversions (what is a purchasing power parity dollar, anyway?) and simply compare the raw volumes of goods that households use, China looks like a fairly consumption-rich society.

In dollar terms, Chinese people consume half as much ($4,936) as Mexicans ($9,688). But Chinese households eat more protein, have more mobile phones (280 per 1,000 vs. 218 per 1,000 in Mexico), and almost as many refrigerators (35.4 per 1,000 vs. 37.8). There are some categories where Mexico is ahead, like items of clothing apparel and laptops, but across the board this does not look like a comparison where one country is half as rich as the other.

The conclusions the authors lay out are stark:

Overall, we believe that China's per capita consumption level should actually be higher than that of Mexico (approximately $10,000), rather than just 50% as indicated by the per capita consumption expenditure. It should also reach at least 40%-50% of the levels of major developed countries such as Japan, Germany, and France, rather than the 20%-30% shown by the expenditure figures. If measured in terms of physical consumption volume, China's consumption level might be about double the current level indicated by expenditure data.

These are massive claims. Here’s a quick bit of napkin math using the Penn World Tables: 39% consumption share * $20 trillion GDP = $7.8 trillion in 2017 PPP dollars. If China’s consumption is underestimated by half, we’re undercounting global GDP by around 6%. It’s as if, when tallying up the global economy, we forgot to count both Germany (GDP: $4.3 trillion) and the United Kingdom ($3.1 trillion), with room to spare for Peru ($400 billion).

Subscribe now

What’s Your Engel?

The Pekingnology newsletter sparked an idea. Can we look more deeply at spending behavior on a key good to see if it sheds light on overall Chinese consumption?

A handy trick in the development economist’s toolkit is the Engel Curve: the observation that, as households and countries get richer, they spend a smaller share of their consumption on food.1 This is one of those rare economic regularities that crops up again and again in the data:

Nakamura, Steinsson, and Liu (2016) show how we can deploy Engel curves to sniff out anomalies in Chinese price data.2 (Skip to the *** below if you don’t want the technical details.) Below are Chinese Engel curves in 1995, 1998, and 2000, plotting the share of household consumption spent on food against real total expenditures:

Notice how the lines shift down and leftward from 1995 to 2000? A lower Engel curve means that households spend a smaller share on food for a given amount of total real expenditures (which are deflated using official Chinese inflation data).

Why did the Engel curves shift at such a fast pace? It doesn’t seem likely that the consumption habits of Chinese households should vary so much from year to year. The parsimonious explanation, Nakamura, Steinsson, and Liu argue, is that the official Chinese price data is biased. In this case, if the inflation used to deflate total expenditures is systematically overestimated, then real expenditures are underestimated—which would explain why the Engel curve has shifted down!

(It’s worth noting that biased inflation is not a problem specific to China; this method was invented to detect upward biases in US CPI.)

By observing how Chinese Engel curves shift up and down, Nakamura, Steinsson, and Liu can back out their own measure of inflation, completely independent of the Chinese official statistics. They find inflation that is significantly more volatile than the official metrics:3

These inflation series can then be used to deflate nominal expenditures to get real consumption.

***Wonky bit ends here***

Nakamura, Steinsson, and Liu’s headline finding is that Chinese consumption growth was likely significantly understated in the 1990s, and overstated in the 2000s:

Official consumption growth was 163% between 1995 and 2011. Tallying up all the fluctuations, Nakamura, Steinsson and Liu’s series implies that actual consumption growth was 210%, or 47 percentage points higher.4

Some caveats are in order. Note that an Engel curve is not based on any theory, but an empirical regularity that we’ve reified, solidified, congealed into a stylized fact. Perhaps it is true that Engel curves move around; those shifts are indistinguishable in our data from price changes. The paper’s data also spans only 1995-2011. Perhaps the original price level in 1995 was overestimated. Most of all, there’s been another decade of Chinese growth since—it’s possible that recent inflation has been underestimated, cancelling out these gains.5

On the other hand, if the present results hold, then we’re halfway to explaining China’s consumption gap with Mexico.

Subscribe now

The View From Outer Space

Notice that up to this point we’ve been solely focused on consumption. But underconsumption is a relative concept; what matters is consumption relative to other countries at similar incomes.

Estimating the true level of Chinese GDP is unfortunately not just a well-established problem; it is a cottage industry keeping an army of intelligence analysts gainfully employed. Even official Chinese sources sometimes have trouble keeping things straight. For a time, provincial GDPs added up to a number larger than national GDP, and former Premier Li Keqiang is said to have preferred an index of railway cargo, electricity consumption, and bank loans to the official GDP numbers.

Perhaps the most interesting recent academic attempt is Martinez (2022), which uses nighttime satellite imagery to measure national GDP. I have some statistical qualms with economists’ growing infatuation with nightlights as a data source; that’s a subject for another post. But it’s generally true that, as places grow richer, putting up streetlights and neon signs, they appear brighter at night:

Martinez’s discovery was to realize that autocracies and democracies behave differently when viewed from outer space. For each percentage increase in GDP, nighttime lights in autocracies grow less than those in democracies. If we believe that nightlights are an impartial read of economic activity on the ground, while GDP statistics are fashioned from the crooked timber of humanity, this suggests, not implausibly, that autocratic regimes are more likely to be fudging their GDP numbers.

Moreover, of the components of GDP, investment and government spending appear to be systematically overestimated in autocracies, while consumption is not (Table 2). This is doubly helpful. First, investment and government spending are plausibly easier variables for governments to manipulate than household spending, strengthening the evidence in favor of autocratic meddling. And second, if GDP is overestimated versus nightlights, while consumption is not, this may help explain China’s purported underconsumption.

After statistically adjusting for autocracies’ bias, Martinez finds that China’s GDP growth from 1992 to 2013 was likely overstated by 60%:

I have some skepticism about the scale of this figure—60% is a huge revision. But if we take this adjustment seriously, combining it with the hypothesized 47% increase in consumption from Nakamura, Steinsson, and Liu (2016), we can return to the chart we began with, putting China in a corrected place:

Perhaps China is not underconsuming at all. Perhaps it is in fact overconsuming, relative to an income level that is much lower than is officially reported.

None of this is meant to be definitive. What I mean to highlight, through all this statistical speculation, is the sensitivity of our presumptions to alternative looks at the data; even the grandest theories are built atop prosaic questions of measurement. There’s a lot more to mine in the theoretical implications of under-consumption. (If there’s reader demand, I can pick this up in a Part II.) But I’ll leave it here for now.

That Chinese underconsumption is possibly the product of undercounting consumption and overestimating GDP strikes me as the perfect sort of compromise—the kind which leaves everyone a little unhappy.

The “Global Value Chain” is now an exhausted metaphor, but it’s worth reminding ourselves now and then what it means in human terms.

At the bottom are the raw materials producers, among them the child slaves in the Congo digging out cobalt with their bare hands. Somewhere in the middle are the gray towns like Luoyang and Dongguan where ores are melted down and workers put their byproducts into little cases to be shipped abroad. Then at the top are the people like you and me, clacking away at the machines with cobalt-cathodes humming inside, who are so inconceivably productive we can spend most our days Just Reaching Out and Circling Back and Following Up Again.

In our world, we know of no escape from poverty other than climbing the value chain. And perhaps the boldest attempt in recent memory is Indonesia’s gamble on nickel.

Subscribe now

If I Had a Nickel Every Time Said Industrial Policy

Nickel is a key component in the lithium-ion batteries used by electric vehicles (EVs). With global EV sales surging amid the global green transition—17 million were sold in 2024, a 25% increase over 2023—nickel has become a critical link in the global value chain.

Indonesia mined around 1.8 million tons of nickel in 2023, more than any other country, and nearly half of global production. But for decades, it has been stuck at the bottom of the value chain. Raw Indonesian ores (saprolite and limonite) were sent abroad, largely to China, where they were smelted, processed, refined, and manufactured into higher-value goods like batteries and stainless steel, fetching higher-value prices.

The chart below, from Wijaya and Jones (2024)’s excellent analysis of the geopolitics of Indonesia’s nickel industry, shows just how many stages follow mining nickel from the earth:

For many developing countries, extracting rents from raw minerals is the full extent of their policy ambition. But, over the past few decades, Indonesian policymakers have tried to leverage their commanding market position to begin a slow, tentative ascent up the chain.

Starting in 2009, foreign-owned mining firms were required to divest up to 51% of their ownership to domestic firms. In 2014, a ban on the export of nickel ore went into effect, excepting some low concentrations. And in 2020, under President Jokowi, the export of all unprocessed nickel ore was banned outright. At the same time, Indonesia has doggedly courted foreign capital, attracting around $30 billion in investment by 2023 to help build nickel processing factories.

Central Sulawesi—along with Halmahera Island, home of the majority of Indonesia’s nickel deposits—has received around $7.5 billion in foreign direct investment, the lion’s share of which is from Chinese firms. Likely the largest investment is the 10,000-acre Morowali Industrial Park in central Sulawesi ( red), which processes the ores right by where it is mined:

The massive growth of the Morowali complex can be easily seen from space, as shown below in night-time satellites between 2014 (left) and 2023 (right):

Indonesia’s Gambit

Thus far, Indonesia’s gambit has paid off handsomely. Processed nickel have grown to be almost 10% of exports as of 2023, and Indonesia has become the largest exporter of stainless steel in the world.

In 2019, the European Union raised a complaint under the General Agreement on Tariffs and Trade (GATT), arguing that Indonesia’s export restrictions were were inconsistent. Indonesia appealed in 2022; the dispute has been stuck in limbo due to vacancies on the World Trade Organization’s Appellate Body.

Yet Indonesian leadership remains proudly defiant. Here’s Muhammad Lutfi, Minister of Trade under Jokowi, at Davos 2022. It’s worth listening in full as it lays bare the power dynamics of global trade:

Journalist: Mr. Lutfi, you wanted to triple your economy by 2030. That is a lot of consumption and a huge amount of demand. Is the world geared to support your growth ambitions?

Mr. Lutfi: Well, I don't know about the world, but I want to continue what the minister from Peru says. If we just rely on selling a basic commodity, the country cannot sustain [itself], I guarantee you that. But I want to tell you a story. I don't want to name the commodity, but we stopped the export of one particular commodity. [Ed.: this is, of course, nickel.] We got about $1.1 billion in 2019 selling that commodity. We stopped it and forced people to make value additions to that. We created stainless steel. Suddenly, Indonesia became the number two stainless steel producer in the world.

Do you know how much the multiplication is? 10.86 billion dollars that very year after. Double that in 2021 to be $21 billion. This year it will be 10% of Indonesian exports coming from stainless steel. The recipe is value added, but like my friend says, when a developing country creates something, the developed countries are not happy. So right now, I'm having dispute settlements with the developed countries almost like every other day, but for Indonesia we have no other choice but to fight it.

[The exchange continues with some pointed questions from the Davos interview about resource nationalism.]

Pause for a moment and consider the position of asking someone from a country with less than $5,000 in GDP per capita, where around half the population lives on less than $3.65 PPP a day, if the world is ready for its “growth ambitions”. Should poor countries wait to ask for permission to start growing?

But Western skepticism of Indonesia’s nickel push is less rare than you might think. You can detect a similar tone from this major 2023 photo feature by the New York Times, which examined the dark side of the nickel boom on Sulawesi:

[Mr. Jamal] and his family complain about the dust pouring off piles of waste, the belching smokestacks, and trucks rumbling past at all hours bearing fresh ore. On the worst days, residents don masks and struggle to breathe. People go to clinics with lung problems.

“What can we do?” Mr. Jamal said. “The air is not good, but we have better living standards.”

Here is the crux of the deal that Indonesian officials have cut with deep-pocketed Chinese companies now dominating the nickel industry: pollution and social strife in exchange for upward mobility.

What Mr. Jamal and his community have had to endure is appalling. But I have to admit, somewhat sheepishly, that my first thought was how the Times’s account smacked of typical privileged Western handwringing about industrialization, when the alternative is brutal poverty. The transition to factory work is never a cakewalk—but here is a case of a poor country trying to wield the tools of industrial policy, however imperfectly, to climb from the lowest place in the value chain. If it succeeds, there is perhaps no better blueprint for other resource exporters looking to do the same. Who are we to dismiss it?

Subscribe now

Where’s the Beef [Rendang]?

But my second reaction was to suspend judgment. Is it actually true that nickel mining has relieved poverty?

At the national level, Indonesia has taken significant economic strides, averaging around 5% growth in real GDP and around 3% in per capita terms in recent years. By basically every international standard, poverty has fallen:

But what has it meant for Sulawesi? The chart below shows poverty rates across regencies in Indonesia, with Morowali Regency—the home of the Morowali Industrial Park—highlighted in red:1

Morowali’s poverty rate has seen a modest decline since 2016, from around 15% to 12%, but this puts it squarely in the middle of the pack of other Sulawesi districts—and not far from the decline in the national average, which fell from 10.9% to 8.6% over the same period. When set against the threefold increase in Indonesia’s nickel export value, it’s tiny.

This is not a new finding: a 2024 report by the Lowy Institute also found limited poverty reduction in the Konawe district, home of a major nickel smelter, despite significant local economic growth. But the enormous growth of the industrial parks—Morowali Industrial Park alone employs 38,000 workers, 6,000 of them Chinese—have hardly moved the needle on local poverty.

Nickel has likely been a boon to the broader Indonesian economy, and perhaps areas outside Sulawesi; its precise effects are hard to identify. But it has almost surely benefitted Indonesia’s business elites. Members of President Jokowi’s inner circle have developed mining and processing interests, and over half of Indonesia’s top 50 billionaires are now involved in nickel (Wijaya and Jones 2025, pg. 11). Local elites, too, have gotten in on the action, taking bribes in exchange for licenses and security. Meanwhile, nickel workers—both Chinese and Indonesian—work in dangerous conditions, with allegations of frequent accidents, withheld pay, and suppression of unions.

Critically, domestic inequality also intersects with geopolitics. Despite Indonesian attempts to hedge between the West and China, Chinese firms dominate the space. China accounts for around 90% of Indonesia’s nickel exports. Mandala (2024) estimates that around 75% of direct profits accrue to foreign—likely Chinese—shareholders, with the rest flowing to Indonesian elites. By binding together Chinese capital with Indonesian politics, the nickel industry is becoming a gravitational force in the tricky nexus of Chinese-Indonesian relations.

Kicking Away the Ladder?

With its bet on nickel, Indonesia is now the world’s industrial policy darling—it already has at least one international emulator in Zimbabwe, which banned lithium exports in 2022. After its success inching downstream, President Prabowo Subianto is now leading a new push to build a domestic electric vehicle industry. For a country that currently does little more than raw materials assembling and parts assembly, jumping to manufacturing is (needless to say) a bold leap.

I don’t want to pre-judge. But, peering past the hype around Morowali reminds us that we should be careful before assuming that industrial policy must be pro-poor. Poverty reduction is never a purely technocratic exercise, and questions about industrial policy can never be entirely divorced from those of political economy—including those of global geopolitics. Development economists do themselves an analytical disservice when they fail to ask: who gains from the present state of affairs?

It’s one thing if residents of Sulawesi like Mr. Jamal and his family have had to risk exposure to toxic pollutants in a successful escape from poverty. It’s another thing entirely if those bearing most of the burden see few of the benefits.

I started this Substack exactly a year ago, on February 19.

For a blog called “Global Developments”, I largely managed to avoid writing anything about current events. Partly this was a conscious choice, to offer a differentiated product—essays, not newsflashes. Mostly it was my inability to write fast enough.

But a nice side-effect is that almost every post over the last year was evergreen. This is a look back on the last year of Global Developments; for those of you who are new, it might also be a gentle introduction to the blog, to see what you’ve missed. Annual rankings are boring, but I thought it would be nice to seat the blog in a chaise longue, pull out my psychiatrist’s legal pad, and see if I can discern a pattern to its obsessions.

Several themes emerge:

Subscribe now

The Limits of Economic Measurement

Anyone who’s worked seriously with economic data knows what a cockroach-infested sausage factory they come from. Amateurs gobble down the outputs cheerfully; professionals check the innards with their forks.

This was clearly my primary obsession over the year, sparked by reading Morten Jerven’s Poor Numbers. My meta-concern is that most of the work academic economists do—all the attention given to careful identification and theoretical explication—is downstream of basic questions of measurement. This doesn’t mean we should descend into total nihilism about the value of quantification, but careful attention to the limits of our knowledge can prevent us from being fooled by bad statistics.

• “How Much Should We Trust Developing Country GDP?” was my most-read piece of the last year. Credit mostly goes to Morten Jerven, who wrote Poor Numbers; I just plucked out the best bits from that book. Once you learn that the national accounts in Zambia were compiled by one (1) person in 2010, it’s hard to look at a GDP statistic in the same way again.

• “No, South Korea Was Not Poorer Than Kenya in 1960.” The Korea-Africa economic comparison is a personal obsession of mine, having grown up on my Mom’s stories of her childhood in Zaire. This little bit of economic history tries to pick the grounds of this comparison apart; Kenya and Korea’s similar GDPs in 1960 conceal deep historical differences.

• “Beyond GDP: One Table of Neoclassical Economics Solves the Europe vs. America Debate, For All Time, Forever and Ever, Amen”. As an earlier post, this didn’t get as many eyeballs, but I think it’s a nice subversive take from within the framework of squarely neoclassical welfare economics on measuring what actually matters. I hope it gets more airplay when Europeans and Americans resume their seasonal Twitter argument over whose society is more dysfunctional.

I also had drafts about P.C. Mahalanobis, statistics under Chinese socialism, and measuring GDP from satellite nighttime lights that I shelved; you may see these over the coming year.

Subscribe now

The role of foreign advisors in development

Policy advice has typically been the main way that economists interact with developing countries, but on closer examination the historical record of foreign advisors at promoting development has been spotty.

This year, I wrote about three Western development experts: Albert Winsemius, Edward Mitchell, and Albert Hirschman. General themes are that advice is never apolitical, and that a close attention to local context is necessary for success. Expect to see more in the coming year on this theme, including (potentially) on W. Arthur Lewis in Ghana and Wolf Ladejinsky in Japan.

• “‘You Can Even Kill Them’: Albert Winsemius and the Rise of Singapore.” Lee Kuan Yew’s Singapore is now a touchstone for would-be autocrats touting economic growth. This post traces the influence of the Dutch economist Albert Winsemius on Singapore’s economics—and, at one critical moment, its authoritarian political turn.

• “Quiet Americans: Research Malpractice and the Vietnam War.” This was my favorite piece of the year. Careless application of bad econometrics—springing from a young, ambitious economist’s ignorance of context—may have led to a prolonging of the Vietnam War.

• “The Real Development was the Friends We Made Along the Way.” Albert Hirschman’s emphasis on close observation, his self-subversive distaste for sweeping theorizing, and his lifelong commitment to the possibilities of reform, make him the patron saint of this blog. This post chronicles his remarkable early life as an anti-fascist résistant, and his classic Strategy of Economic Development.

Subscribe now

The possibility of radical reform

The tension between What Is and What Can Be makes development intellectually fascinating. Academics naturally gravitate towards What Is, since empirical research is inherently backwards-looking. But development, when it is successful, is inherently a process of radical change. Part of the job of economists, then, is to enlarge the space of possibilities and imagine outcomes beyond what is immediately feasible, so that when the envelope expands, places are ready to seize the opportunity.

• “Leasing Like A State, or: Public Housing is Development Policy.” My first piece, arguing that there is a potential connection between the manufacturing success of Hong Kong and Singapore and their ambitious public housing programs. African cities are also falling far short of their population’s growing housing needs—could public spending on housing be a way of improving economic competitiveness?

• “The Beginning of the End of Africa’s Last Colonial Currency.” Since independence, the CFA Franc has bound the currencies of 14 West and Central African countries to France’s. This piece examines the potential macroeconomic and developmental impacts of this strange institution, in the broader context of the history of colonizer and colonized.

• “What is it like to work in an Ethiopian factory?” This was one of my favorite pieces, but the most difficult to write. Ethiopia’s ambitious push to industrialize has stalled, with factories failing to absorb labor—in part, perhaps, because working conditions and wages are so poor. This piece documents some of these workers’ experiences, and entertains the possibility that improving conditions for workers may help the cause of Ethiopian manufacturing growth.

Subscribe now

A Call for Feedback

The main goal for Global Developments is to write the kind of blog that I’d like to read; I’m grateful that so many others have joined along for the ride. I’d like to take this opportunity to request feedback. I’m perpetually surprised by what posts are popular, and which are not. If you have opinions—what you dislike about the blog, pieces you’ve enjoyed, what you would like to see more of—please let me know in the comments section below. (Yes, I will try to write more, and faster.)

Fawning praise will also be accepted—building a Substack readership from scratch can be tough work. Early on, I was reminded of an old joke about the great director Orson Welles. Looking at an empty theater, the audience diminished by a snowstorm, Welles is said to have asked: “Why are there so many of me, and so few of you?”

Now there are a lot more of you. Thanks to Tyler Cowen, Noah Smith, Brad DeLong, Alice Evans, Adam Tooze, Ken Opalo, Davis Kedrosky, Lauren Gilbert, Angus Bylsma, Deena Mousa, Yaw Asamoah, among others, for spreading the word. And thanks to you, reader, for being here.

Most of my readers are likely deeply concerned about the wholesale destruction of USAID. They don’t need to be convinced further with statistics—that 19 million lives have been saved by PEPFAR, that 434 of 634 vital soup kitchens in Khartoum will have to shut down, that $500 million of food aid is stuck rotting in warehouses rather than going to the hungry.

Just to lay all my cards on the table, I’m gutted.

Ken Opalo, Lauren Gilbert, Kelsey Piper, among others, have already done an excellent job analyzing the developmental and humanitarian consequences of this rollback. Pushback has begun, as have attempts to mitigate the damage.

Of course, there’s almost no way to fill a $72 billion hole, and with the full force of the Trump-Musk movement concentrated on USAID, it’s clear that the political terrain around aid has been fundamentally altered. Moreover, the same impulses that are driving the defunding of USAID will soon gear up in other rich countries.

As the initial shock has worn off, and as the longer-term implications have become clear, in this short post I thought it would be worth taking a step back to think about US AID in broader historical perspective.

Weeks Where Decades Happen

Let me do my best Adam Tooze impression and say that the last few weeks represent an epochal break with basically the entirety of post-World War II American foreign policy.

Looking at the official spending data from ForeignAssistance.gov, the striking fact is the remarkable stability of American aid spending in real terms since 1945:

Yes, there have been fluctuations, but look at the scale: for the better part of a century, aid obligations have hovered around a $40-80 billion band. Remarkably, the year the US spent the most in dollar terms on aid was… 1948.

Even as GDP has grown tenfold, US aid budgets have remained basically flat in real terms—the result (one might guess) of a delicate balancing act between bipartisan consensus among elites and a wariness of attracting too much negative attention. The longevity of that consensus is itself remarkable: in 79 years of data, Republican presidents spent an average of $34 billion a year on aid, while Democrats spent $40 billion. For all the stereotypes of bleeding-heart liberalism, the largest modern ramp-up of foreign aid occurred under a Republican, George W. Bush, who launched the President’s Emergency Plan for AIDS Relief (PEPFAR).1

In a few short weeks, that political equilibrium has now been disrupted, I fear permanently.

Subscribe now

What Aid Can Accomplish

Measuring the empirical effect of aid on development is a real hornet’s nest, which I can’t do justice to in this post. But it’s worth reminding ourselves what, in the best-case scenarios, aid can accomplish.

A few weeks ago, I wrote about the enormous differences in starting conditions between newly independent East Asia and Africa, that sent them on vastly different developmental trajectories.

All that still holds true, but I could have just as easily written about the historical gap in their incoming aid flows. Amid all the debate about industrial policy and Confucian culture, the sheer quantity of American aid to Korea and Taiwan often gets lost in the discussion about what set East Asia apart.

Well into the late 1970s, South Korea was getting as much American aid as all of Sub-Saharan Africa combined:

Just as Taiwan was into the early 1960s:

Am I claiming that American aid was the main factor behind the East Asian Miracle? No. (Before you start typing in the comments, read last month’s piece.)

Much of this was military aid, and pretty useless for economic development. If the volume of American aid were the sole determinant of growth, then the developmental champion of the 20th century would have been a country that no longer exists: South Vietnam, which absorbed half ($20 billion) of all American aid spending in 1973.2

But take it from an expert in the niche subfield of 1950s Taiwanese development: American aid helped a lot, too. In both countries, food and military aid supported countries which might not otherwise have survived. In Taiwan, it was American funding of the Joint Committee on Rural Reconstruction that enabled the famous 1950s land reform, and paid for an army of agricultural extension workers to canvas the country spreading best practices. (It was the latter, more than the former, which likely kickstarted Taiwan’s agrarian development.) In South Korea during the late 1960s, USAID gave technical assistance to the infant export industry, and USAID procurement contracts helped provide a crucial demand stimulus for Korean firms. (Keep an eye out for my ongoing research with Nathan Lane, Philipp Barteska, and Seung Joo Lee on this topic.)

All this to say: capable states with ambitious developmental plans, when hit with a bazooka of money, can seize the opportunity to transform themselves.

Subscribe now

A World Without Aid?

Of course, states capable of emulating Korea and Taiwan have turned out to be pretty rare. As scholars like Nicolas van de Walle have pointed out, aid can also foster an unhealthy dependency—in his coinage, many African states live in a state of “permanent crisis”, where aid is large enough to reshape (and warp) the political economy around it, but not large enough to be fundamentally transform countries’ economies.

American aid obligations in Sub-Saharan Africa, for instance, have never exceeded $15 per person:

Of course, not all that money was spent effectively. But a solid chunk—roughly $5 of those $15—went to PEPFAR, one of the most effective public health programs around. And even if you poured the money into the highest-return, most cost-effective programs, $15 is still not a whole lot—less than 1% of Kenyan GDP per capita, or likely less than 3% in Malawi.3 In the early 70s, South Korea got more like $180 in 2023 dollars, or closer to 10% of GDP per capita.

The potential end of USAID is an unfolding public health and humanitarian disaster, but it does not permanently doom the task of growth—in large part because these aid flows were far too small to begin with. For people like me working on global development in the West, the bitter pill to swallow is that aid is not development. NGO leaders and aid workers and development economists were never the protagonists; the key players are still there, in their home countries, scrimping and saving and hustling, and the task ahead for them remains largely unchanged.

A world without aid is closer than you think—because we’ve been living so close to it this whole time.

Note (Feb 12, 2025): an earlier version of the first graph understated post-2000 aid. Thanks to Lauren Gilbert for pointing this out.

Before I get to the mechanics of exchange rates and trade—somewhat in the news these days—let me start off with some fortune cookie koans about the nature of Truth:

• Lawyers: what’s provable is True.

• Artists: what’s authentic is True.

• Politicians: what’s popular is True.

• Practitioners: what works is True.

• Academics: what’s correct is True.

And so on.

By the last of these, I mean that there’s a significant difference between knowledge that can be exhaustively proven to an academic’s satisfaction and what most people are willing to act on in the world.

And there’s probably no topic with greater distance between what academics Know and what the world believes Works than the effects of exchange rate movements on the economy.

Folk Tales of Competitive Devaluation

The practitioner’s view of exchange rates goes something like this.

Let’s imagine two fictional countries: call them Poland and Japan. Imagine the Polish złoty falls in value relative to the Yen. The prices of Polish pierogi and kielbasa are in złoty, so from the perspective of a Japanese consumer, they can buy more kielbasa for the same amount of yen. Poland’s exports to Japan increase. On the other hand, Japan’s exports are listed in Yen; from the perspective of a Polish consumer, umeboshi and sake have become more expensive because of the fall of the złoty. Poland’s imports from Japan decrease.

The folk view of exchange rates concludes that devaluing your exchange rate is good for exports, but bad for imports. Choosing which exchange rate to pursue is thus a matter of political priorities. Historically, developing-country elites from 1950s Taiwan to present-day Senegal have favored overvalued exchange rates, because they like cheaper foreign goods; more recently, fast growers like post-1950s Taiwan and China have favored undervalued exchange rates to promote their exports.

In case this all seems a bit abstract, here’s noted practitioner Donald Trump, laying out the same logic:

So we have a big currency problem because the depth of the currency now in terms of strong dollar/weak yen, weak yuan, is massive… That’s a tremendous burden on our companies that try and sell tractors and other things to other places outside of this country. It’s a tremendous burden . . . I think you’re going to see some very bad things happen in a little while. I’ve been talking to manufacturers, they say we cannot get, nobody wants to buy our product because it’s too expensive.

There’s a whole menagerie of different names for this phenomenon, from “competitive devaluation” to “expenditure switching”. In its crudest form, it’s often reduced to the caricature that a weaker currency boosts GDP by boosting net exports. (This is incorrect as a matter of accounting.) In development, perhaps the most sophisticated perspective comes from Rodrik (2008), who argues that devaluing one’s real exchange rate promotes the relative growth of the tradable sector—which usually means manufacturing.1

But all these views are predicated on the same story, that changing the exchange rate shifts the relative price seen by buyers and sellers, which in turn affects quantities of actual goods in the real economy.

Only problem is—is this actually true?

Subscribe now

Exports, Invoices, and Loyalty

One immediate wrinkle is that firms, being firms, can smell a buck to be made. If the złoty falls—in effect, setting an international discount on Polish goods—they may raise their złoty prices a tad, offsetting a bit of the depreciation but allowing them to increase their profits. So some of the exchange rate depreciation is eaten by firms; the “passthrough” to prices observed by importers may be incomplete.

But there’s a deeper issue. With its matrices and subindices and trilemmas, international economics is so complex that one can stumble around a boneheadedly obvious point for years before colliding with it like a mountainside.

To wit: how are prices actually listed?

The chart below, from the IMF’s Boz et al (2022), compiles data on the currencies in which global exports are invoiced:

Around 40% of all global trade is invoiced in US dollars, even though only 10% of global trade is destined for the United States. A similarly large share of world trade (around 46%) is invoiced in the Euro, but this is much more in line with the Euro Area’s share of world trade (37%). Over 80% of global goods and services are thus listed in terms of just these two currencies—and, in particular, 30% of global trade is invoiced in US dollars, even when America is not necessarily involved in the transaction.

This has immediate implications for my little folk tale of Poland and Japan. A hidden assumption behind that story was “producer currency pricing”—that Poland’s exports are listed in złoty while Japan’s are listed in Yen. One could have just as easily imagined the reverse “local currency pricing”, where the importing country’s currency is used—kielbasa priced in Yen, sake priced in złoty. But reality has other ideas. We live instead in a world of dominant currency pricing, where the US dollar reigns.

We can see the consequences first in prices. Here’s a table from Gopinath and Itskhoki (2021), measuring the responses of import prices to exchange rate movements:

Columns 1 and 4 show the conventional view, regressing import prices on just the exchange rate between the two trading countries. Passthrough is not 100%, but it’s pretty high; a 10% decrease in exchange rates is associated with a 7.5% increase in prices. High passthrough, in turn, implies a world operating close to producer currency pricing.

But the moment you control for the movements of the US dollar, as in columns 2 and 5, passthrough disappears: a 10% depreciation in the bilateral exchange rate is associated only with a 1.6% increase in prices. And what’s really striking is what’s going on with the US dollar exchange rate in the third row: a 10% depreciation in the US dollar is associated with a 7.8% increases in prices—even if the US is not involved in the transaction! If not quite a smoking gun, it’s a strong hint that the world we live in is ruled by a dominant currency paradigm.

So much for prices. What does this actually mean for trade volumes—tangible goods and services? Gopinath and Itskhoki (2021) run the same regression as before, but putting trade volumes on the left hand side:

Column 1 here is the conventional wisdom about exchange rates: when a currency depreciates by 10%, trade volumes increase by around 1.2%—not a whole lot, but still a significant positive relationship. But if we control for the US dollar exchange rate, as in column 2, the coefficient shrinks tremendously. Now when the exchange rate depreciates by 10%, trade volumes increase by only 0.3%!

The folk wisdom thus appears to be quite mistaken: exchange rates matter a lot less for trade movements—and, by implication, the real economy—than one might think.

It Was the Practitioners All Along

Careful readers will note that I used non-causal words like “associated” to describe the tables above. Applied microeconomists, who are used to cleaner experimental settings, are usually aghast when they discover the Wuhan wet market that is international macro. The honest truth is that very little in international economics is “exogenous”, since nothing moves on its own. Shifts to global trade flows—including, say, by a 25% tariff on Canadian goods—can be the causes of exchange rate movements, just as often as they are their effects.

Perhaps the best attempt to isolate the effects of exchange rates is a recent working paper by my teachers, Emi Nakamura and Jón Steinsson, with their coauthor Masao Fukui. They observe that countries that peg their currencies to the US dollar are subjected to exchange rate shocks that often have little to do with their local conditions. (As recent experience has shown, the United States doesn’t give much of a damn about what happens to the rest of the world when it sets its policies.) So the differential effect between countries who float and countries who peg—whose currencies are dragged around by the vagaries of the US dollar—gives you a kind of natural experiment for an exchange rate movement.

Fukui, Nakamura, and Steinsson then trace the response of several key macro variables to a 1% depreciation of the US dollar. First, the nominal and real exchange rates:

After a 1% depreciation of the US dollar, the nominal and real exchange rates of pegs immediately depreciate relative to the floats (though slightly less than one-to-one, since very few currency pegs are strict). Then, over the following years, the exchange rate slowly starts to appreciate again. So this experiment works— pegged countries have their exchange rates yanked around by the dollar.

Next, macro variables:

The bottom row shows that, just as with the previous table, the response of exports and imports to a depreciation is pretty muted. Exports hardly move at all; imports in the pegs increase by about .3 percentage points of GDP after 3 years. Net exports (exports minus imports) thus fall ever so slightly after a depreciation, by about .3 percentage points of GDP. So far, so good: in a cleaner quasi-experimental setting, we’ve confirmed the Gopinath and Itskhoki result that exchange rates hardly seem to move trade volumes around at all, consistent with dominant currency pricing.

But how can we explain that graph in the top-left: why does investment— typically thought of as a domestic variable—go up because of a currency depreciation?

And it’s not just investment—GDP and consumption also increase:

In response to a 1% dollar depreciation, GDP in peg countries rises by around 0.4% after 5 years—quite a large estimated effect for a currency that can move by 10% in the space of a year. In the paper, they find that this is largely driven by growth in the non-tradables services sector, not manufacturing like the Rodrik (2008) view might predict.

We have a strange pattern of results where domestic macro variables appear to respond more to exchange rates than trade itself.

The authors’ answer to the puzzle lies in a market we’ve been studiously ignoring: capital. Applying their methodology to nominal interest rates and inflation, they find (somewhat noisy) evidence that interest rates appear to increase in response to a depreciation:

For currency traders holding assets in a pegged currency, this is a great deal. After the initial shock fall of the exchange rate, the currency slowly and steadily appreciates. Factoring in the (noisy) increase in nominal interest rates, this means an increase in the relative return of assets in the pegged currency. In finance jargon, this violates uncovered interest parity. In plain English, traders smell money.

And this brings us full circle.

Currency depreciations, Fukui, Nakamura, and Steinsson argue, appear to juice the economy by raising the return on assets in the depreciating currency. This brings in the “hot money” from days of yore: global capital flows in and borrowing increases, boosting consumption, investment, and service sector growth. Returning to our fortune cookie wisdom at the top, it may perversely be the Practitioners’ belief that Depreciations Work that actually powers the boom.

What Works Is True turns out to be correct. Just not in the way that anyone intended.

Please consider subscribing to Global Developments to receive more essays on global economics, development, and poverty. Some recent posts:

• In 2010, Zambia’s national accounts were assembled by one person. How Much Should We Trust Developing Country GDP?

• No, South Korea Was Not Poorer than Kenya in 1960: a quick tour of historical legacies, developmental states, and family history.

• The early life of Albert Hirschman—resistance hero and development economist—and his remarkable Strategy of Economic Development

Sixty years ago, East Asia and Africa were both desperately poor. Then East Asia experienced a miracle and Africa didn’t. Why?

The space is now open for the author to offer their favored explanation, which (logically) occurred in the past sixty years.1

Here’s a recent example:

The Taiwanese miracle ranks among the most dramatic examples of modern economic development. In 1950, Taiwan was a poor, primarily agrarian economy, with a real GDP per capita lower than Ghana or Afghanistan. Today, after 50 years of over 5% growth in GDP per capita, it is a manufacturing powerhouse with living standards comparable to France or Germany.

The author, of course, was me, in a coauthored paper with Jen Kuan Wang on Taiwan’s 1950s land reform. Mea culpa! We’ve since revised it. But far more important people than me have tried this sleight of hand. Here’s a 2021 quote from Uhuru Kenyatta, when he was President of Kenya:

If the Asian Tigers were at the same level of development as Kenya in 1963 and Kenya actually lent South Korea some money to implement its development plan in 1963, the question is what happened? Why were we unable to develop as fast as they did?

If Kenyan presidents are getting in on the action, it’s time for some good old-fashioned economic history myth-busting. Was South Korea really poorer than Kenya at independence?

It’s a simple question, but doggedly following the thread leads us quickly down the rabbit-hole of economic history—and, for me, some personal history as well.

A Brief Education

Let me start off with a confession: I lied. It is literally true in the data that Kenya had a higher GDP per capita than South Korea in 1960. Here’s real GDP per capita from the Penn World Tables:

While heeding all the warnings from a few weeks back about not taking any of these GDP estimates too seriously, it’s clear that up until around 1970, South Korean and Kenyan income levels were at least in the same ballpark.

But there are enormous historical differences hidden behind these GDP figures.

Consider education. At independence, South Korea and Taiwan had far higher rates of primary education than most of Sub-Saharan Africa. We can see this in a plot of countries’ 1960 GDPs against their primary school completion rates (with the word of caution that the data is likely not very good):

South Korea and Taiwan (red) had unusually high levels of primary education for their relatively low levels of GDP—over 30% of people completed primary school, at annual incomes of just $1224 for Korea and $2553 for Taiwan. By contrast, most African states (blue) had lower educational endowments than their incomes might imply, falling below the line of best fit. In the Barro and Lee data, only one country, Senegal, had a primary school completion rate over 20%. Kenya was at just under 9%—less than a third of Korea or Taiwan.

Secondary and tertiary (university) schooling were also relatively common in Korea and Taiwan, and rare in Sub-Saharan Africa:

Why does education matter for development? It almost feels ridiculous to type out that sentence—education is a good in itself—but it’s worth a standing-on-one-foot recap of the economics.

The first, obvious answer is that education creates better workers. At a minimum, knowing how to read and write and add and subtract makes you more productive. During the second half of the twentieth century, East Asia’s combination of a relatively productive workforce with low levels of wages turned out to be economic dynamite. As the West grew richer and shifted into services, a channel opened up for cheaper foreign manufacturers—and capital chose to go where labor was cheap, pliant, and productive.2 As the East Asian economies rose, they needed educated engineers, inventors, entrepreneurs, etc. to do the engineering, inventing, entrepreneuring, and et cetering—which the educational system duly supplied.

By contrast, in Africa, skilled labor was often in short supply. In 1964, only 53% of Kenyan administrative, executive, and managerial positions were held by Africans. Even in 1963-66, only 38 out of 98 graduates of University College, Nairobi were African (the vast majority were Kenyan Asians). As a settler-colony, Kenya actually had a relatively high level of colonial investment; in the most infamous case, at independence the Belgians left only 30 people in the Democratic Republic of the Congo with two years of schooling at the university level.

One Level Deeper: State Capacity

The second, more subtle channel for education’s importance is its relationship with state capacity. Here, cause and effect are also intermingled: you ideally want well-trained people to manage your government, but high rates of education are as much the products of effective states as they are its inputs.

Catch-up growth in the 20th century was almost entirely state-led. The South Korean and Taiwanese states intervened extensively in markets, promoting agricultural extension, fostering new industries, and sending bureaucrats abroad to ply their country’s wares. When they pursued these policies, they could draw from a deep well of former bureaucrats trained by the Japanese colonial state. According to Atul Kohli, “there were 40,000 Koreans qualified as government officials on the eve of the second world war”.3 Koreans were barred from the higher levels of government—in the Japanese empire, as elsewhere, colonialism was not intended to benefit the colonized—but at independence there was a relatively large stock of administrators with at least some government experience.

The Japanese empire reached far deeper into its colonies than any European empire ever did in Sub-Saharan Africa. In 1941, there were more than 60,000 police in Korea, or 2 for every 1000 people; at the end of World War II, there were roughly 5,000 police in Kenya, or 1 for every 1000. As a settler colony, Kenya was again relatively well-policed; in the extreme, British Nyasaland (now Malawi) had just 500 police in 1945 for an estimated population of 2 million—a ratio of 0.2 per 1000.

It’s important to remember that colonial police were, first and foremost, tools of coercion and control. In Korea, they committed hideous abuses of power, including sexually enslaving comfort women and trying to stamp out the Korean language and culture. Nor did the relative scarcity of the state preempt colonial cruelty in Africa, from the Belgian atrocities in the Congo to the British concentration camps in Kenya. Nothing says that police states should be developmental—but in the history of statehood, gendarmes have usually come before teachers and doctors. If the police are the knife’s edge of the state, the blade did not penetrate deeply. The European imperialists in Africa tried to do “administration on the cheap”—and the newly independent African states inherited the result.4

To be clear, I am not saying that well-educated bureaucrats were the only things separating effective Asian states from ineffective African ones. (I’ll go into some of these deeper differences in a moment.) Nor is it to say that newly independent Africa did not have first-rate administrators—for instance, under the skillful leadership of Charles Karanja, Kenya’s Tea Development Authority made the country the second-largest tea exporter in the world.

But these were the exceptions to the rule. The fact remains that without a stock of capable administrators, industrial policy and export targeting and infrastructure-building—or even Adam Smith’s minimum of “peace, easy taxes, and a tolerable administration of justice”—are a nonstarter. And regardless of your theory of development, 30 university graduates is not the correct number.

It’s Deep Roots All The Way Down

Let me go just one level deeper. Surely some substantial chunk of state capacity had to do with pre-colonial state capacities. Japan’s colonial project in Korea lasted around forty years; British rule in Kenya lasted for just sixty. In each case, the colonial regime had to build atop of—and confront—indigenous political institutions that had existed for far longer. (This explains why the Japanese needed so many police.)

In East Asia, colonization confronted some of the oldest states in the world. China of course was the forerunner, but by AD 400-800 Korea and Japan had formed cohesive states, with clear territorial boundaries and sophisticated administrative capacities. To give one striking data point: in the 16th century, East Asian armies were fielding thousands of well-drilled musketeers, decades before the practice caught on in Europe.5

Precisely how much the ancient continuity of East Asian states matters for development is still a little murky, and the delineation with colonial legacies is still an area of active debate—but the balance of evidence suggests that the legacies of these pre-colonial states stretched through the colonial period, extending even into the post-independence era. For example, places in Korea with more Joseon-era civil service exam passers saw higher levels of education during the colonial era, and areas in Vietnam governed by the Confucian Dai Viet state have higher modern household consumption.

By contrast, before colonization, in large parts of Africa, people lived outside of state-level units, as illustrated by this map by Jack Paine, Xiaoyan Qiu, and Joan Ricart-Huguet (2024):

The relative absence of pre-colonial states meant that the newly independent African states were largely European inventions; modern Kenya, for instance, inherits the boundaries of British East Africa, which was created by pulling together the Kikuyu, Luo, Luhya, Kalenjin, and Kamba ethnic groups. Paine, Qiu, and Ricart-Huguet have shown that colonial boundaries followed the frontiers of pre-colonial states more than once thought—but they could still stir conflict by lumping together disparate ethnic groups, and encasing them in newly invented units. Moreover, colonial policies often worked to exacerbate and accentuate these ethnic distinctions, making ethnic tensions not a bug, but a feature. Africa’s long history of civil wars and ethnic conflict are their tragic legacy.

I’ve just written several hundred words on how central states are to development, so allow me to briefly subvert my main point. It’s important to not succumb entirely to the Eurocentric trap that Africa before colonization was a static, stagnant monolith because of its lack of states. African states formed where it was rational to do so, in places like the Ethiopian highlands where population density was high and the disease burden was relatively low.6 Otherwise, decentralized gerontocratic or democratic systems of governance often suited local conditions better. Moreover, the best available guesswork by Broadberry and Gardner estimates incomes in parts of Africa were likely above subsistence levels at the beginning of large-scale colonization, suggesting that some societies may have already departed from a Malthusian economic regime.

Nonetheless, to the extent that states were the vehicles of rapid development in the 20th century, the newly independent African states were dealt a bad hand. While Korean nation-builders could rely on hundreds of years of shared statehood, Kenya’s founding fathers had to build atop states with far less internal coherency—while starting out with a shallower pool of educated personnel.

This is not to say that post-1960 policies didn’t matter for the economic divergence of East Asia from Africa, and everything was pre-ordained. Nor is this an argument that education and state capacity are all that matters; I’ve ignored gender, religion, culture, the international environment, and probably a half-dozen major research literatures who are all furiously typing away in the comments. But even through this relatively narrow lens, it’s hopefully self-evident that there’s a fundamental shallowness to pulling a GDP estimate, and assuming that it tells you all there is to know about a country’s development potential.

A Personal Coda

Having fallen into the East Asia - Africa Trap before, I can tell you precisely why so many people make this mistake. Simply put, it is rhetorically useful to emphasize that countries can alter their economic destinies. The growth South Korea saw over the past hundred years is almost inconceivable—and if Korea could transform itself, why not Kenya? Emphasizing agency and the possibility of radical transformation is at least a refreshing antidote to the fracasomania that can often set in in development circles.

But the temptation to juxtapose Korea and Africa also has, for me, a personal dimension.

I’m Korean-American-Canadian, by way of Hong Kong and Singapore, but my Mom grew up partly in Zaire, now the Democratic Republic of the Congo. Her father, my grandfather, was a career diplomat for the South Korean government. Much of the Sixties and Seventies was spent shuttling back and forth between Seoul and Paris, until at last the family landed in Kinshasa, where my grandfather was appointed South Korea’s Ambassador to Zaire, then ruled by the murderous Mobutu Sese Seko.

By the children’s accounts it was a happy time: unsupervised, free. They went to a French lycée in Kinshasa, surrounded by the remaining colonialists, largely unaware of what was going on outside of their diplomatic bubble. (A rare sign of trouble was when a child snatched my mother’s necklace in a crowded marketplace.) I would happily wager that they were the only Koreans in town—save, of course, for the North Koreans, who would occasionally drive by to shake their fists at my uncle playing Cowboys and Indians in the yard.

At the time, neither Korea was represented in the United Nations. Both North and South were engaged in a mini Great Game for recognition, sending diplomats like my grandfather canvassing across Africa. In cables from the archives, my grandfather can be spotted in Chad in 1974, delivering a donation of 10,000 vaccines in response to a cholera outbreak. A week later, he can be seen fleeing Benin after a pro-Kim Il Sung editorial appeared in the Daho Express.

I have to admit that the image of besuited Korean diplomats chasing each other across Africa is now faintly funny. What could South Korea (1980 GDP: $5097 in 2017 USD) and Zaire (1980 GDP: $2219) possibly have to talk about? But members of my grandparents’ generation had seen their country come within a hair’s breadth of destruction; diplomacy was serious business.

My grandfather, Kee Heum Shin, was born in 1928 in what is now North Korea. He must have been forced to learn Japanese in school, though I never heard him speak it. At independence, his family joined the exodus of Christians and former landowners fleeing south. Because he was lucky enough to know some English, he served out the war far to the south, as a translator at an American air force base. In the crushing poverty after the war, he served as the main support in a family of three boys and three girls. He studied hard, and went on to a promising career in Korea’s diplomatic service—to postings in France, Morocco, Zaire, Nigeria, and perhaps others I have not heard of.

To be frank, diplomacy was probably never his strong suit. As a father and a husband, tact was certainly not a quality he was known for. When he was Ambassador to Nigeria in the early 80s, he fell afoul of Chun Doo-hwan’s all-powerful KCIA, which led to the family’s emigration to the United States. That was alright, because money was his real passion. In America, he found a second wind as a liquor store owner, plowing the earnings obsessively into the stock market. Even as his health failed, and his interest in all else faded, he would get up early every morning to watch the little Tesla and Apple sparklines tick up.

Later in life, he confessed that he should have always been an economist. (Our sole economic debate occurred when I confessed, to his great disappointment, that I was a Keynesian.) A large part of this, I’m sure, was his love of money talking. But there was something else.

As diplomats, my grandparents spent most of their adult lives abroad. The family they built would spread across three continents, where if everyone was together the dinner-table talk would mingle into a creole of Korean, French, and English. (It was only a few years ago when I realized that dejeuner was not a Korean word.) The Korea they mostly knew had been bombed to ruins. One in ten people were dead. Despite all the data I showed, for the people who lived it there were few inklings of the remarkable transformation that was to come. They were frankly eager to get out.

As a child at my grandparents’ house in Washington DC, the nightly ritual was to put in a videotape we would rent from the Korean grocer. The soap operas were always a week or two behind what was being shown in Korea, but my grandparents would follow the plot-lines religiously, as closely as they watched KBS news immediately after (when, bored, I would trundle off to bed). Each year, the quality of the productions seemed to improve: the budgets got larger, the sets more ornate. Even the actors and actresses seemed to get prettier, their jawlines shrinking, their eyes widening inconceivably.

I didn’t realize it then, but some of my happiest childhood memories were sitting with my grandparents, as they watched the country they once knew transform beyond all recognition.

My grandfather passed away on Sunday, at the age of 96.

Why did the Industrial Revolution happen? Why, for the past two centuries, have hundreds of millions of people moved off the farm and into factories?

Stop and think about it for a second. Most people, including myself, will likely pull from some scattered half-remembered facts about the Industrial Revolution from high school: something about enclosures, Eli Whitney, the seed drill. Coursework in undergraduate economics is usually no help—the Solow growth model of Econ 101 only has one sector.

Here’s my quick primer on the textbook neoclassical theory of industrialization: how it can be used to interpret the last 200 years of global economic growth, and—by applying it to the recent rise of China—explore important ways that that theory may be wrong.

You’ll know just enough theory to impress people at parties, for the first time in history someone asks this question. (Thanks will be forwarded on to Professor Andrés Rodríguez-Clare and his Econ 270C module.)

A Quick Primer on Theories of Industrialization

Let’s call “industrialization” growth in the share of the labor force in manufacturing. Neoclassical theory typically thinks of industrialization as the product of two different forces.1

The first is different rates of productivity growth between sectors. This story is probably intuitive for most people: new inventions—the spinning jenny, the steam engine, the iPod Nano—raised productivity in manufacturing more than in agriculture and pulled people out of farms into the factories.

The second, more subtle force is based on the observation that as people get richer, they spend a smaller share of their income on food. This is a phenomenon known as Engel’s Law:2

Engel’s Law (no, not that Engels) implies that any form of productivity growth—even agricultural—can produce industrialization, so long as it raises overall incomes. People can only eat so much; when they get richer, they start demanding stuff other than food, like better housing and clothes. The relative price of these non-agricultural products rises, pushing labor off the farm.

Frankly, I’ve always found the Engel’s Law story somewhat puzzling, as it implies industrialization can happen without any of the scientific and technological inventions typically associated with the Industrial Revolution. But if you stop and think about it, the differential-productivity-growth story is also strange. What pulls people from farms to factories are the higher wages induced by higher manufacturing productivity. But we also have a story of how the Agricultural Revolution preceded the Industrial Revolution: if (say) productivity in agriculture rose faster than in manufacturing, relative agricultural wages should rise and pull people back onto the farm.3

How can we reconcile all these facts?

Subscribe now

Push or Pull?

A solid framework to think through these issues is “Labor Push versus Labor Pull” by Alvarez-Cuadrado and Poschke (2010), who develop the simplest possible model—a closed economy, where labor can move freely between sectors—and take it to data for a set of countries from the 19th century to today.

For non-economists, a useful rule of thumb is that prices and productivity move in opposite directions: all else equal, if we get much more productive at squeezing out lemonade, lemonade prices go down.

Applying this logic to the relative price of manufactures to agriculture (Pm / Pa) helps us learn something about the productivity changes churning underneath:

• If the relative price of manufactures to agriculture is falling, manufacturing productivity must be rising faster than agricultural productivity, and “labor pull” is the main story.

• If the relative price of manufactures to agriculture rises, the story is a bit more ambiguous.

  • If the manufactures are growing only weakly more expensive than that of agriculture, then we can’t conclude anything. Because of Engel’s law, productivity growth will weakly push up demand for manufacturing, and the relative price of manufactures, regardless of which sector is growing.

  • If the price of manufactures is rising much more than that of agriculture, then productivity growth in agriculture is clearly outpacing productivity growth in manufacturing. “Labor push” has to be the main story.

If it helps, you can think of “labor push” very loosely as pushing up the supply of labor, since as incomes grow there’s less need for everyone to be farming;4 and you can similarly think of “labor pull” as pushing up labor demand, since the development of new technologies in manufacturing pulls people to the factories.

Alvarez-Cuadrado and Poschke take this to the data. They compile price and employment data for manufacturing and agriculture for 12 now-industrialized countries (most of Western Europe plus Canada, the USA, Japan, and South Korea), then examine how the relative prices of manufactures to agriculture evolved throughout the past 200 years:

The relative price of manufactures to agriculture (Pm / Pa) fell for most countries from 1800 until about 1920. Then, after 1920, the price of manufactures started to rise relative to agricultural products.

So, we have two different answers for two different time periods. “Pull” factors dominated throughout the Long Nineteenth Century: manufacturing productivity must have been rising relative to agricultural productivity. Then, after World War II, “push” factors dominated, with general productivity growth causing a shift of labor out of agriculture.

Subscribe now

Some Further Wrinkles

It’s worth pausing here and thinking about the ways this simple model might be wrong. One is the assumption that workers can move freely between sectors, and are paid the marginal product of their labor—a famous man with an enormous beard wrote a long book about this in 1870, and people are still arguing over its implications.

Another is that prices are assumed to be determined entirely within the closed system of the domestic economy. Given the minor phenomenon known as the global trading system, this is unlikely to be true. Kiminori Matsuyama has an elegant 1992 paper showing that, in autarky, agricultural productivity growth has a positive effect on growth—but when the economy is open to trade, agricultural growth has a negative effect. The intuition is that a small developing country that’s open to trade will be forced by the global market into its comparative advantage—which, given its lack of development, will usually be farming.5

(As a quick aside, this model gives a theoretical channel for developmentist industrial policy to work. And it falls out of a standard neoclassical model with limited modification!)

But a third complication, which I’ll focus on in this short piece, is the idea of push and pull as distinct, separable forces.

To make this a bit more concrete, let’s focus on the biggest Industrial Revolution in recent memory: the rise of China. (The macroeconomist Basil Halperin has done several great Twitter threads digging into the Chinese miracle.) A very live question is the extent to which China’s rise was driven by agriculture or manufacturing. This has critical implications for other developing countries—if state resources are limited, which should policymakers prioritize?

This debate stems from an iconoclastic 2003 paper by Alwyn Young. Young showed that if you account for the enormous quantities of labor that were moving from farms to factories, Chinese non-agricultural productivity growth must have actually been fairly pedestrian, around 1.4% per year:

Think back to the push-pull framework. If all these people were leaving the farms but non-agricultural productivity growth was low, Young’s result implies that the engine of China’s rapid growth must have been in agriculture, not manufacturing.

Not so fast, say Loren Brandt and Xiaodong Zhu in 2010. If you write down a more sophisticated version of the push-pull model, and take the observed rates of productivity growth in each sector as given, then non-agricultural productivity growth must have been the key driver. Without it, China couldn’t have sustained such a massive movement of labor into manufacturing without crashing into severely diminishing returns; if those workers weren’t productive, industrialization wouldn’t have been profitable.

What’s going on? We’ve ping-ponged back and forth between several reasonable-sounding theories. The sleight of hand—one that we’ve implicitly accepted from the start in the “push-pull” framework—is that Brandt and Zhu assume that we can take each sector’s productivity as given. In a more realistic model, the rate of technology growth in a sector is surely shaped by the number of people working in it. If nobody ever left China’s collective farms and joined the cities, then surely the rate of technological growth would drop—there would be fewer scientists, inventors, engineers, entrepreneurs to generate TFP growth in the first place. Incorporating this TFP response would likely shift the burden of explanation back towards whatever was driving people from farming—decollectivization, procurement price reforms, etc.

We’ve come full circle. Separating out Push and Pull was a helpful starting analytical device—but a more accurate model of growth would endogenize the rates of technological progress. Technological development does not fall like Manna from Heaven, but is the organic result of people in each sectors experimenting and inventing; Push and Pull are clasped together, intertwined, inseparable.

Where does this leave us? An elevated sense of confusion: somehow, with more sophisticated modeling, and a head-on confrontation with two centuries of empirical data, we know less than when we started.

If you like this feeling, consider applying to grad school.

“If I took you and blindfolded you and took you to LaGuardia Airport in New York, you must think I must be in some Third World country.”

— Joe Biden, 2014.

“Underdeveloped”, “developing”, “emerging markets”—it’s remarkable how much of the economic history of the last century can be seen just in the words we use to describe global poverty.

Here’s the frequency of these three words for “poor country” in print books, from Google n-grams:

“Underdeveloped country” was the most common term in the era of early decolonization, up till the 1960s. The more positive “developing country” took over in the optimistic 1960s, Kennedy’s Decade of Development. It remained dominant for forty years, even as the notion that African and Latin American countries were progressing economically became harder and harder to justify. (East Asia, of course, was the bright spot that truly deserved the label.)

This changed in the fat years of the 1990s. As the Chinese Miracle gathered steam, fueling a global commodity markets boom, “emerging market” took off: now there was money to be made. In the mid 2000s, “emerging markets” even threatened to overtake “developing countries”, until the disappointment of the financial crisis. China grew so successfully it graduated into its own category; elsewhere, promised markets largely failed to emerge.

What of the poor countries as a whole? What do we call this bloc?

The dominant term of art for most of the 20th century was the “Third World”—a 1952 coinage of the French demographer Alfred Sauvy, in conscious reference to Abbe Sieyés’s revolutionary pamphlet “What is the Third Estate?”.

Far from an anti-colonialist, Sauvy conceived of the “Third World” in opposition to American notions of “underdevelopment”, hoping that old colonial powers like France might make the “vital investments” that would shepherd their development.1 Though Sauvy’s original intentions were forgotten, the “Third World” caught on as a term, peaking somewhere in the mid 1980s—the height of the confrontation between the capitalist First and communist Second Worlds, but also (perhaps more significantly) the age of Live Aid and “Do They Know It’s Christmas?”. As the levels on the graph show, more than any other term here, Third World has a claim to being a household word.

But, much like being “homeless”, the stigma of poverty soon transformed a neutral term into a pejorative. “Third World” is now what Americans say when the airport bathroom leaks, or when the 08:15 at Penn Station is late, or when it’s people in Kansas and not Kinshasa who are bleeding out while waiting to see a doctor.

If the Google n-grams data (which ends in 2022) is any guide, Global South is quickly becoming the new polite term of art. Here’s Joe Biden stumbling over this semantic shift in a 2023 speech in Hanoi:

THE PRESIDENT: We talked about what we talked about at the conference overall. We talked about stability. We talked about making sure that the Third World — the — excuse me — “Third World” — the — the — the Southern Hemisphere had access to change, it had access —

Global South, by the way, is a 1969 invention of Carl Oglesby, American academic and one-time president of the Students for a Democratic Society. It fits neatly into the dependency theory of Prebisch and Singer and notions of solidarity between the poor countries of the world. At root, it reflects the common left-academic hope that terminology can reshape thought, that the right word can shock us back into collective action.

The jury is still out on “Global South”—as Joe Biden just demonstrated, there’s some general confusion about if Southernness is about economics or geography. But it is certainly true that terminology can create its own reality—at least when it is backed by the force of states.

Over the last decade, until at least 2018, the use of “BRICS” has risen steadily alongside “Global South”. The acronym “BRIC” (Brazil, Russia, India, and China) was originally coined by Goldman Sachs economist Jim O’Neill for a November 2001 research report identifying promising growing economies. Brazil and Russia’s foreign ministers picked up on the term, and the BRIC countries started to meet on the margins of the UN General Assembly. The first formal BRIC meeting was held in Russia in 2009; and in 2010, the “S” for South Africa was tacked on after a formal invitation by China, presumably to give the body some African representation.

BRICS, of course, is almost entirely a diplomatic invention: very little unites these countries, other than the fact that China would like to hold them close. But China’s inclusion of South Africa is telling. Even as China’s GDP has steadily climbed up the global rankings, Xi Jinping has been careful to position China in solidarity with the poor countries: in his words, “China has been and will always be a developing country (发展中国家, fazhanzhong guojia)”. This embrace of the “developing” label stands in a stark contrast with the rhetoric of American politicians, for whom “developing countries” serve mostly as useful metaphors for shoddy infrastructure like LaGuardia—or, in the extreme, “shithole” places to openly detest.

The LaGuardia-Africa parallel is particularly apt. We live in a world of appalling inequality, where the 20 million people served by LaGuardia collectively earn more than the 1.2 billion people who live in all of Sub-Saharan Africa:2

I’m deeply sympathetic to the view that words should emphasize the dignity and humanity of the people they are describing. Better still, we should use their words for themselves—in this sense, BRICS (for all its faults) may be the most vital and apt term here.

But the brutal truth is that words reflect power much more than they can reshape it. The inequalities in the map above are far beyond the power of any terminological reshuffling; eventually the stigma of poverty always catches up. Without transformative economic growth in Africa, South Asia, Latin America, and elsewhere, “Global South” will likely assume its place alongside “developing” and “Third World”—as an easy put-down from the mouths of the rich.

Poor Numbers: How We Are Misled by African Development Statistics and What to Do About It, by Morten Jerven. 2013.

I think most users of GDP presume that, when they download an Excel spreadsheet from a slick-looking official website like the World Bank or the International Monetary Fund, what they get is something FDA-approved for human consumption—that, somewhere down the line, maybe in some other room down the hall, the real experts have done the shoe-leather work to make sure that the numbers are accurate and reliable.

Morten Jerven’s 2013 book, Poor Numbers, is about why that isn’t true.

Subscribe now

The Late 1980s Tanzanian Growth Catastrophe

Perhaps the best illustration comes from Tanzania, where the socialist leader Julius Nyerere famously pursued a policy of ujamaa—“brotherhood” in Swahili—in which the newly independent state assumed control of large portions of the economy.

Initially, Nyerere’s ambitious project of socialist nation-building captured the global imagination. In some respects, it was a remarkable success—to this day, Tanzanians have a stronger sense of shared national identity than their neighbors. But it also had a dark side: a process of forced villagization, where Tanzania’s scattered rural population was coercively—sometimes violently—resettled into villages planned by the state.

By the 1980s, it was clear that ujamaa had fallen short in terms of delivering development. Under pressure from the IMF and the World Bank, Tanzania began to implement neoliberal reforms, loosening the state’s control of the economy.

According to the Penn World Tables, one of academic economists’ favorite datasets, between 1985 and 1986, GDP per capita fell from $1,717 to $1,130 in 2017 international dollars—a 33% collapse, larger than what the United States experienced from 1929-33 during the Great Depression:

The only problem? It likely didn’t happen.

The Late 1980s Tanzanian Growth Catastrophe is largely a statistical artifact—a product of the simple truth that GDP measures what the state can see.

As ujamaa collapsed, the Tanzanian state atrophied, and economic activity retreated into the informal sector, with farmers switching from export to subsistence crops. But the GDP statistics largely missed this shift. In 1997, the GDP methodology was finally revised, using 1992 prices as a base. With this revision, the Bureau of Statistics admitted that GDP had previously been underestimated by anywhere between 30 and 200(!) percent.1

This left the question of how to reconcile the new numbers with the old ones. For somewhat murky reasons, the Penn World Tables chose to stitch the two series together in such a way to create the Late 1980s Tanzanian Growth Collapse. Not only that, but to get the reconciliation to work, the Penn data also shows a miraculous recovery over the late 1980s, with per capita growth rates averaging over 5% a year from 1995 to 1999. (This recovery has erroneously been used to tout the success of Tanzania’s structural adjustment program.)

But no other estimate of Tanzanian GDP shows such dramatic volatility:

In the current Penn World Tables, Tanzania’s one-year GDP collapse is a large outlier—a three-standard deviation event. But it is by no means alone. In 2013, Nigeria revised its GDP estimates and, overnight, Nigerian GDP jumped by 89%, leapfrogging South Africa to become Africa’s largest economy. In 2010, a similar revision by Ghana caused its GDP to increase by 60%.2

In a sense, drastic revisions like the Late 1980s Tanzanian Growth Collapse or the Great 2010 Ghanaian Rebasing Jump are straightforward to deal with. If you know about them, you can drop them from your data. The scary thing is that decisions like these likely undergird most GDP numbers from poor countries. Your real concern should be the dogs that don’t bark—it should be the data fictions that lurk in plain sight.

Subscribe now

“What Happens If I Disappear?”

Through a series of country case studies, Poor Numbers documents how African economic statistics are actually produced. Jerven’s examples, though not comprehensive, are collectively shocking. Hollowed out by years of state neglect, African statistical agencies are now often unable to conduct basic survey and sampling work. Jerven writes:

In 2010, I returned to Zambia and found that the national accounts now were prepared by one man alone… Until very recently he had had one colleague, but that man was removed from the National Accounts Division to work on the 2010 population census. To make matters worse, lack of personnel in the section for industrial statistics and public finances meant that the only statistician left in the National Accounts Division was responsible for these data as well. (pg. x)

Without the staff to collect and analyze survey data, statistical agencies are usually forced to improvise, guessing the size of the economy from population figures, which are themselves extrapolated from censuses that are decades-old.3

To name one prominent example, Nigeria (thought to be the world’s sixth-largest country) has not conducted a national census since 2006. With almost twenty years since the last data collection, the population figure of 230,842,743 that you can currently pull from Wikipedia is a quixotic monument to false precision.

Indeed, it would not be a great exaggeration to say that we have never known how large Nigeria is. The first census after independence, in 1962, was marred by regional infighting, because of its role in determining federal spending and assembly seats. The initial results, which reduced the population share of the politically dominant north, were rejected by the Northern People’s Congress. A redo in 1963 was also rejected as fraudulent—this time largely by southern politicians. The 1973 census was largely rejected as illegitimate; the 1980s and 1990s saw no censuses under military dictatorship; and the 2001 census produced an implausibly low number.4 Even in the relatively successful 2006 attempt, census enumerators were attacked and killed, and the results were still disputed.

Many African states are failing at the basic task of knowing how many people live in their borders—let alone accurately measuring their economic activity. The vast, unobserved informal sector (which includes subsistence farming, and something like 60% of working people) is usually estimated just as a direct function of population.5 Lacking direct harvest yields, estimates of agricultural output are often produced using FAO models based on planting-season rainfall data.6 Even the minimal task of measuring the goods traveling across borders—in theory, the easiest thing for a sovereign state to accomplish—is occasionally beyond the reach of statistical agencies. Until 2008, landlocked Uganda only collected trade data on goods that eventually passed through the Kenyan port of Mombasa, ignoring the four other countries on its borders.7

In the absence of good underlying data, the prevailing approach for GDP in developing Africa can be summarized as:

Income estimates… derived by multiplying up per capita averages of doubtful accuracy by population estimates equally subject to error. (p. 39)

This is not the fault of the statisticians, who are doing the best they can with shoestring budgets. It is because of their work that we have estimates at all for what are some of the poorest countries in the world—and the most vulnerable. Earlier this year, Oxfam estimated that 84 out of Zambia’s 114 districts were hit by drought, leaving around a million farming households at risk of hunger. We are left with imprecise statements ("around a million”) and poor targeting of those in need, precisely because of our statistical ignorance.

I’m haunted by the words of the lone Zambian statistician, sitting in his empty office, who asks Jerven plaintively: “What happens if I disappear?”

Subscribe now

A Marginal Proposal

Poor Numbers came out in 2013, attracting a wave of scholarly and policy attention (including by Bill Gates). Once you’ve heard its arguments, it’s virtually impossible to look at a GDP statistic the same way again.

But what actual progress has been made in the statistical capacity of nations?

Seemingly, not much. In late 2014, perhaps in response to Jerven’s book, the World Bank relaunched its website for its Statistical Capacity Indicator—a metric on a 0-100 scale which scores countries based on the strength of their “Methodology”, “Source Data”, and “Periodicity and Timeliness”. But even by this clunky internal metric, progress has been glacially slow: in 2004, the average score for African countries was 58.2; in 2019, it was just 61.4.

Moreover, over six years of an economics PhD, I have never heard of any economist using this statistic. Poor Numbers is well-cited and well-read (at least by Africa specialists), but its lessons about the fundamental unreliability of statistics have largely not been absorbed in how we actually do economics.8 A separate data series with a clunky, hard-to-interpret scale is simply not going to cut it. The problem, I think, is that we economists need a tool that is internal to our statistical and quantitative frameworks.

And we actually have one: margins of error.

Strangely, economists pay a great deal of attention to statistical error when it comes to causal inference or survey data they collect themselves—but very little when it comes to national statistics. The presumption is the one with which I began this piece—and I shared, until fairly recently—that somewhere down the line, surely somebody has to have checked the work on something as important as GDP.

Here’s a simple bounding exercise. Remember Nigeria’s 89% GDP revision? If we assumed that Nigeria’s GDP estimate was exactly correct every other year since 1960, the implied margin of error of GDP would be around 20%.9 This aligns roughly with the pioneering work of the statistician Derek Blades, who estimated in 1980 that the errors in aggregate GDP for Botswana, Kenya, Malawi, Nigeria, and Tanzania ranged from 20-35%, and the errors in GDP growth rates were at least 3%.10 (The correlation of year-to-year GDP estimates helps bring down the margin of error when computing growth rates.)

I am writing this in November 2024, not long after the bulk of the news-reading population of the United States just got a crash course in statistical literacy. (There will be just enough time in four years for everyone to forget again.) But hopefully we’ve learned enough to realize that, with a 3% margin of error, empirical arguments about policies that might raise GDP growth by around 0.5% are—like Kamala Harris’s 2-point polling lead in Pennsylvania—largely academic. We can recommend good policies on the merits, but their growth effects are essentially undetectable with our crude instruments.

Problems of statistical volatility of macro data extend, by the way, to the recent past of developed countries. A famous paper on the US by Christina Romer shows that much of the perceived postwar fall in unemployment volatility may simply be a statistical artifact—applying the old techniques to new data yielded far more volatile series. Developing countries and the developed countries of the past exist on a continuum of statistical precision.

The far-and-away first-best option, of course, would be for African countries to sustainably develop their indigenous statistical capacities. (Never forget that large-scale sample surveying was an invention of the Global South!) But a second-best approach would be for researchers and policymakers to begin developing better estimates of the statistical error of GDP. An excellent habit would be to mimic political polling, and to start printing margins of error alongside GDP growth estimates. Seeing that Nigeria posted GDP growth of 2.9% in 2023 feels substantially different from seeing that Nigeria posted GDP growth of 2.9 ± 5%. The former is a reified certainty; the latter, a statistical estimate.

The world is fiendishly complicated: fuzzy, gooey, hairy, spiky, oozing with contradiction. The job of statistical measurement is to squish all this unimaginable complexity down to dimensions that human beings can manage. This exercise is never value-free, nor one immune to error. I haven’t even gotten into all the thorny ideological complications of GDP; proposing margins of error is a purely internal statistical critique—a modest first step. Our goal should be a social scientific approach that doesn’t ignore this messiness, but is honest about how the data is made—one with a little less certainty and a little more epistemic humility—one that, when confronted with a graph of a particularly dubious GDP estimate, always remembers to draw out the missing confidence intervals with the mind’s eye.

Apologies for the long delay between posts. I was, erm, finishing my PhD.

I’ve since started as a Research Fellow at Open Philanthropy, helping to do research on causes in global health and development. We’re launching a new program on economic growth in low and middle income countries, and hiring a program officer to manage this grant portfolio.

You can apply here. The deadline is 11:59pm Pacific Time, December 1.

Short of waking up one morning from unsettling dreams to find yourself transformed into Brunei, there is only one way we know of to rapidly lift a poor country out of poverty: manufacturing growth.

Ethiopia’s government has staked its national development strategy on this idea—most of all in the textile industry, where it has built vast industrial parks, subsidized electricity, and offered generous tax incentives to attract foreign investors. It is the highest-profile attempt by a country in Sub-Saharan Africa to adopt the East Asian model of export-led manufacturing growth. If Ethiopia succeeds, no doubt others will follow.

Ethiopia’s original goal was to create 350,000 jobs by 2022; so far, it has managed around 150,000. Political events have disrupted these grand designs. First came the outbreak of civil war in late 2020. Then, because of wartime human rights violations, the United States terminated Ethopia’s beneficiary status under AGOA, a law that gives developing African countries tariff exemptions. But even before its recent political turmoil, Ethiopia’s experiment was not yielding the expected results: headline GDP growth was good, but was largely driven by other sectors. Like cold fusion or the end of my PhD, the promised manufacturing renaissance always seemed just a few years away. Why?

The answer, I think, has been partly obscured by how we sometimes talk about development, including at the start of this post: as something done by states, the product of Five Year Plans and Export Promotion Targets. It’s sometimes easy to forget that the real protagonists of development are not governments or their leaders or the bureaucrats who enact their policies, but the people toiling in fields and factories.

I have never worked in a factory. Nor have I even been to Ethiopia—a research trip I planned in 2021 was cancelled because of the war. But I have long been fascinated by the growing body of research that documents the perspectives of Ethiopian manufacturing workers. If we want to understand what is going wrong, we need to listen to their voices.

The story of Ethiopian manufacturing—its rise, its faltering, and its potential for renewal—is an example, I believe, of where a little more empathy can lead to better economics.

Subscribe now

The View from the Factory Floor

Some of the first signs of trouble could be seen in a 2010-13 field experiment by the economists Chris Blattman and Stefan Dercon. Around a thousand applicants to five industrial firms were randomly sorted into three groups: a third were offered a factory job, a third were offered a $300 grant to start a business and five days of entrepreneurship training, and a third were sorted into a control group. Blattman and Dercon then followed up a year later to see how each group was doing economically.

Most economists (including me) would have predicted that those who got factory jobs would have done best. In standard neoclassical theories of industrialization, workers are pulled into factories through the draw of higher wages—since the modern, industrial sector is typically more productive, factory wages should be higher than in agriculture.

But that’s not what Blattman and Dercon found.

Their survey data showed that factory jobs paid less, not more, than other jobs—wages were 25% less than informal pay. In their randomized experiment, offering people factory jobs had no significant effect on incomes by any measure. By contrast, the entrepreneurship grant increased earnings by one-third. (Sadly, in Ethiopia, that is just US$1 a week.)

Even more strikingly, few of the factory jobs stuck. A third of those offered a factory job quit within the first month; another 77% left within the first year. Beyond low pay, workers frequently mentioned health problems: after a year, “repetitive stress, kidney, and respiratory issues” were common. With each additional month of factory work, workers were 1.1 percentage points more likely to report a disability. (A silver lining is that these negative health effects did not seem to persist after five years.)

Blattman and Dercon’s paper is not alone in its findings. Last year, the sociologists Reimer Gronemeyer and Michaela Fink compiled an invaluable report based on interviews with garment workers, their families, and their managers, largely from the state-of-the-art Bole Lemi and Hawassa Industrial Parks—the gleaming vanguard of Ethiopia’s industrialization push. Built in 2014, the Bole Lemi park is on the outskirts of Ethiopia’s capital Addis Ababa—not far from the new Chinese-built airport terminal—and has the capacity for 35,000 workers. The Hawassa park, built in 2016 around 270 kilometers south of Addis, is even larger, with capacity for 60,000.

The crucial data in the report comes from a survey of 456 workers by Tesfaye Semela, Setisemhal Getachew and Daniel Semela, scholars at Hawassa University.

Average monthly base pay was 1449 birr, or just $25 USD. Pay rose a bit with bonuses: around 80% of workers reported receiving attendance bonuses, which averaged $5, and a third received productivity bonuses, with a mean of $6. But, even accounting for these, most salaries were eaten up by the relatively high cost of living close to the factories: respondents spent an average of $8 birr on rent, $15.50 birr on food, and $5 birr on transportation.1

A female textile worker says:

“How could a person live in Addis Ababa with a salary of 1500 birr? [$26 USD.] How could a person paying 1000 Birr for the housing rent survive with 500 Birr being left from the housing rent? What kind of life is it? If somebody eats breakfast in the morning, he/she may not eat lunch. What is being paid does not match our effort.” — p. 49

On the factory floor, workers find the environment stifling and oppressive. Workers complain that the hours are long—eight hours a day (eleven or twelve if one includes the long commutes to the parks), six days a week. When orders come due, there is enormous pressure to work overtime (p. 48). Over half of workers said they suffered from kidney pain (p. 92), possibly developed because of the unnatural sitting positions, or perhaps because of the tightly controlled toilet breaks. Workers also complain of a lack of understanding around holidays and sick leave—particularly for women (p. 50)—leading to chronic absenteeism.

There is an unmistakable gender dimension to all this: the vast majority of workers are single young women. Historically, Alice Evans reminds us, factory work has been a great engine of female empowerment: around 74% of respondents said that “financial independence” was their initial motivation to get a factory job. But, far from becoming breadwinners, 67% reported receiving financial support from their families. Workers also report cases of robbery and sexual assault on the way home, traveling after dark after long hours spent at work (p. 169).

It’s not hard to see why so many want to leave. 62% rejected the statement that “I feel that my life has improved since I worked in the factory”.

One former factory worker says:

“Since I was jobless, like everyone else, I was thinking of changing my life. Even if there are people who helped themselves in that situation, the salary was not equivalent to the time and workforce spent there… My expectation and what I found was different. There was labor exploitation, high workload, there was no freedom, everything was tough. — p. 2

Historically, factory labor has always been grueling—but this does not sound like a system that is successfully absorbing labor. Over 60% disagreed or strongly disagreed with the statement “I plan to stay with this company for a long time”.

Why are wages so low? Why, at a minimum, are working conditions so bad?

For one thing, workers lack a voice. Workers who organize or complain or ask too many questions are usually fired (p. 144). Formal unions appear to be largely disconnected from the workforce, with labor organizations like the Confederation of Ethiopian Trade Unions (CETU) often acting as a late-coming mediator after industrial disputes have begun. (There are, however, documented instances of wildcat strikes outside this organized structure).

A man from Tafo, a slum on the edge of Addis Ababa, whose wife works in one of the factories, sums up the dire situation well:

“The girls are suffering a lot. Yet, the factories do not pay enough. Moreover, the factories do not respect the rights of the workers. At any time and moment, they order them to do that and this as they wish. The workers do not have the right to say no and reject the orders being given by the bosses. If the workers reject any order, they immediately fire them. If you look at the job properly, it has features of slavery.” — p.123

Subscribe now

Turning to Exit

With poor wages and worse conditions, and without the voice to improve their lot, many workers have turned to exit as their only option. The rapid turnover that Blattman and Dercon observed ten years ago has become a serious systemic problem.

A member of the Bole Lemi’s Investors Association says:

“We are not able to fix the problems of absenteeism and attrition… It is the most serious issue in every industrial park. [Ed: Emphasis mine.] We tried to evaluate it and are currently working on it, but we have yet to find an answer for it.” — p.133-134

They are not being hyperbolic. One study of the Hawassa Industrial Park found that 10% of workers quit on their first day, and 25% in their first month. Similarly, another study by NYU Stern researchers found that the attrition rate is 5-10% a month, or 60-120% per year (p. 129). At the root of the problem are low wages: 70% of workers reported “inadequate pay” as a major reason for labor turnover at their company (p. 103).

But this points to a central tension in Ethiopia’s strategy—the situation could reasonably be described as a pernicious low-wage, low-productivity equilibrium. From the workers’ point of view, low wages make an already intolerable work environment unacceptable. The resulting attrition erodes efficiency—every new worker who has to be recruited, trained, brought up to speed is a cost to the firm. But with chronically low productivity, Ethiopia only has one carrot with which to draw international investors: low labor costs.

And yet—in the marketplace of global capitalism, what choice does Ethiopia have?

A good case in point is PVH Corp., the world’s second-largest apparel firm. In 2016, in a major coup, Ethiopia successfully lured PVH, which owns Calvin Klein and Tommy Hilfiger, to Hawassa. A World Bank report cited low wages as a major factor:

The cost of labor was an important consideration in PVH’s decision-making process. “In the fabric side of the business, apart from the raw material, the highest cost is energy. In the garment side of the business, apart from the cost of raw material, the highest cost is labor” (Ashurst, personal communication). PVH compared the cost of labor in Ethiopia with other key countries, including Bangladesh, Cambodia, India, Kenya, and Uganda. Ethiopia offered significant labor cost advantages.

What this highlights, rather coldly, is that Ethiopia is not making these decisions in isolation. It is competing in a global arena against other developing countries, most of which would also happily welcome foreign investment to help their economies grow. Given the right package of subsidies and tax incentives, many of these manufacturers surely would be happy to up sticks and migrate to Bangladesh or Vietnam. (PVH left in November 2021, just two weeks after Ethiopia lost its AGOA privileges.)

Comparing Ethiopia and its peer countries, a World Bank report by Caria (2019) finds that sales per worker in Ethiopian firms are lower than their peers in Kenya, Bangladesh, India, and Vietnam—even after controlling for differences in firm size and category and capital stock:

These figures deserve a big orange biohazard sticker: the sample is small; Ethiopia and Vietnam lack data on land and capital rents so these have to be filled in using a model; and all this sales data doesn’t account for taxes. But they suggest that Ethiopian firms are less productive in a way that cannot be accounted for by just capital or the sectors they work in.

The Ethiopian state has staked much of its precious capital on the bet that manufacturing growth is its Way Out. It has set aside land and borrowed money to lay out modern industrial parks. It has plied foreign firms with tax incentives, attracting major international brands. It has underwritten their electricity bills, to avoid Nigeria’s chronic blackouts or Kenya’s exorbitant power costs. But, in a sense, what all these subsidies have done is reveal a contradiction at the heart of the enterprise. Every factor of production—land, capital, energy—has been carefully supported and accounted for, save one: labor.

Without workers moving into the factories, a manufacturing takeoff with enough thrust to meaningfully uplift large numbers of the poor is a long way away. In 2019, Myanmar had around 1.2 million garment factory workers; Bangladesh, Pakistan, and Vietnam each had around 5 million. For all of the Ethiopian state’s efforts, in a country of over 120 million people, 150,000 workers amounts to close to a rounding error.

Subscribe now

Cutting the Knot

One reasonable, though tragic, response to these figures would be for the state to abandon the manufacturing drive altogether. If a barrage of electricity subsidies and land grants and tax incentives and tariff exemptions still have not worked, perhaps this points to deeper economic problems. (One of these may be Africa’s relatively high prices.) Perhaps the Ethiopian state’s resources would be better spent developing agriculture instead—when one day, growing farm productivity may free workers from the land, and push them towards the factories.

A better solution may be to cut straight through the Gordian knot, and raise wages—perhaps, as Gronemeyer and Fink suggest, through a state-set minimum wage. In the short run, higher wages could be partly offset by the savings from less churn. In the medium run, there may be latent productivity gains from retaining workers for longer. Caria observes that, in the World Bank’s Enterprise Survey, Ethiopian firms have far lower management scores than their peers in Kenya, India, and Vietnam. Most of their shortfall comes from their poor human resources (HR) practices. “Ethiopian firms perform poorly,” he writes, “in terms of their strategies to identify and retain high performers within the organisation, and to deal with poor performers.” If workers are leaving after a few weeks or months, they never have a chance to develop their skills—and managers do not have the time to identify the best talent. Higher retention could lead to greater skill accumulation and a more efficient allocation of workers to tasks.

As I said at the outset, I am far from an expert on Ethiopia. I’m not sure if the correct prescription is a minimum wage, some form of wage subsidies, or greater experimentation on the part of individual firms; I hope this piece spurs others to look more deeply into this question. (If I’ve said anything completely moronic, please let me know in the comments.)

For all its faults, it’s still hard not to be struck by the ambition behind Ethiopia’s manufacturing push. The goal is nothing less than the transformation of the economic destiny of one of the poorest countries in the world; to abandon it outright would be a travesty.

But, in pursuit of that national dream, we ignore the voices on the ground at our own peril. In the past, the engine of industrialization has always been the movement of people from farms to factories. What pulls them there is the hope for a better life, or at least for better lives for their children.

Without an answer to those hopes, the transformation cannot even get started.

If you enjoyed this post, please consider subscribing to Global Developments to receive more regular essays on global development and poverty. Some recent posts:

• “You Can Even Kill Them”: a UN development advisor’s role in shaping Singapore’s development—and its authoritarian turn

• The early life of Albert Hirschman—resistance hero and development economist—and his remarkable Strategy of Economic Development

• A historical deep-dive into the role of economists—and statistical malpractice!—in prolonging the Vietnam War

• Analysis of the developmental effects of the CFA franc—“Africa’s last colonial currency”

Whether China will democratize is one of the great unanswerable questions of our time. I won’t venture a guess as to if or when this will happen: Western analysis of Chinese politics, including mine, is often hopelessly clouded by wishful thinking or its dark mirror, knee-jerk pessimism. A fool and his China prediction soon are parted.

But a productive, and perhaps more analytic, way forward is to backwards induct: to reason about what conditions would likely exist in the world where Chinese democracy was already a reality.

The first question to ask if the transition will be violent. Political transitions from autocracy to democracy are often bloody—and a democratization won through civil conflict, in a country as large as China, would spill a horrific amount of blood. Moreover, the evidence suggests that nonviolent transitions are more likely to be successful.

Let’s thus restrict ourselves to the set of worlds where the democratic transition is largely nonviolent. We could set as a baseline the democratic transition in Russia, which saw two coup attempts that killed a total of several hundred people, but no violence on the scale of the 1917 Revolution. (Russia’s transition ultimately failed, but happier outcomes were at least conceivable.)

Almost immediately, this tells us a major feature of the democratic China: the Chinese Communist Party (CCP) or its successor party will likely persist as a major political force. The CCP’s hundred million members will not simply evaporate. The Party’s structure is so inextricably linked to the state that even if the bureaucracy is liberalized, there will be large constituencies—former cadres, subsidy-receiving farmers, state-owned enterprise employees, and most of all, the People’s Liberation Army, which serves the Party—who will continue to feel a strong loyalty to the CCP. They will not give up without a fight. If we want our imagined Chinese democratic transition to be peaceful, all these groups will have to be accommodated.

Most of all, for the new democratic regime to be durable, the leaders of the new CCP will also have to believe that they can win elections, or they will be tempted to simply overthrow the government by force—as the hardline Communists tried in Russia, twice. Ironically, the ideal model may be Taiwan’s Kuomintang, the CCP’s old arch-rival. Feeling the pressure from benshengren Taiwanese, the KMT voluntarily acceded to free elections in the late 1980s—and proceeded to win under Lee Teng-Hui. More than thirty years after transition, it remains one of Taiwan’s two major political parties, and for all its flaws tries to win people’s votes through the democratic process.

This points to a rather tragic unexpected consequence. Just as the KMT’s electoral survival makes impossible a full reckoning with its crimes committed under martial law, the persistence of the CCP in political life will likely complicate the process of justice for its victims, among them the Uyghurs, the Tibetans, the Hong Kongers, Falun Gong, the Tiananmen protesters. An honest reckoning may not be possible during their lifetimes.

The other critical feature for the health and legitimacy of democracy will be economic growth. If the initial record of the democratic era is instability, stagnation, and inflation, it is easy to see how nostalgia for the Good Old Days of the high-growth CCP era will be revived. The new democratic regime will likely have to deliver development and stability. A hundred years on, Mainland China’s most significant experiment with representative democracy, its 1912 National Assembly election, is still remembered (somewhat unfairly) as a failure on this count.

This points to a tension with the nationalist turn in American economic policy. While China moves towards democracy—a process which Acemoglu et al. tell us, on average, involves a 10% fall in GDP in the years before the transition—there will be an enormous temptation in the US to exploit China’s turmoil for the benefit of American industry. On the Chinese side, the pressure to generate growth will be great, and anti-competitive practices from foreign powers will always be an easy political target. In other words, democracy or not, zero-sum conflicts between great powers will remain zero-sum—the United States will likely face a choice between its economic interests and its stated values.

And this is where I find the backwards induction exercise prove so helpful: exorcizing wishful thinking. When American observers say they hope that China will democratize, this often hides further, unstated hopes: that the CCP will be punished in some way, and that China will oppose the West less. Perhaps in the long, long run, that is possible.1 But backwards induction suggests that in the possible worlds where the Chinese democratization is peaceful, Chinese policy will not become automatically friendly to the United States, and the CCP will not simply go away. What is good for the Chinese people may not coincide with what is good for America’s national interest.2

Is a world with Chinese democracy still worthwhile? I think so. But if we hope for Chinese democracy, we should do so on the grounds that the billion-plus Chinese citizens deserve to have a voice—not because of what it will yield for the West.

Why is African manufacturing failing to take off?

Manufacturing in Sub-Saharan Africa is stuck at less than 10% of GDP, compared to 24% in East Asia and even 17% in Latin America and the Caribbean. Any discussion of a China-style takeoff in the region is a nonstarter without understanding what is holding back manufacturing.

A plausible answer to this question comes from this 2013 paper by Alan Gelb, Christian Meyer, and Vijaya Ramachandran at the Center for Global Development. They argue that though African countries are quite poor, African factory workers’ wages are relatively high. More specifically, industrial labor costs in Africa are high compared to non-African countries at similar levels of GDP. In this sense, paradoxically, “poor does not mean cheap”.

Okay, then, that raises the next question. Why are industrial labor costs relatively high in Africa?

Gelb, Meyer, and Ramachandran argue that surely it has something to do with prices. If African prices are high, relatively high wages might make sense, to cover living costs.1

Shifting the puzzle from high wages to high prices seems to kick the can down the road one more time: but here at least I think there’s some intellectual progress to be made with a little application of old-school trade theory. This is my attempt at a short and unabashedly econ-wonky post, in the proud tradition of vintage 2010s-era Krugman / DeLong / Smith. More literary programming will return in the following weeks.

Poor Does Not Mean Cheap

Here’s a depiction of the problem of “high African prices” in graph form.

Gelb, Meyer, and Ramachandran plot the classic Balassa-Samuelson relationship: real GDP on the x-axis, price level on the y-axis, with each country in the Penn World Tables a single point. This should tell us if prices are relatively high in Africa for a given level of GDP:

Lo and behold—sub-Saharan African countries, that cluster of red dots, all largely lie above the red line of best fit. So for a given level of GDP, African countries have higher prices than a linear model would predict for the rest of the world.

Another way to visualize this is by plotting the residuals, the vertical distance between each country’s price level and their price level predicted by the Balassa-Samuelson line of best fit:

There’s a lot of red at the top of that graph: African countries really seem to have prices that are too high compared to the linear fit.

We’ve replaced one puzzle with another, and another again. What was once a puzzle about Africa’s low manufacturing shares and then a puzzle about manufacturing wages is now really one about African prices being too high relative to the Balassa-Samuelson relationship. The great Pseudoerasmus opines that this seems like a neglected puzzle to solve for growth and development.

But is there really a puzzle at all?

A Quick Balassa-Samuelson Recap

Let’s rewind a bit here and think through some of the fundamentals. (Feel free to skip to *** if you don’t like math.)

The Balassa-Samuelson relationship falls out of a simple two-country, two-sector neoclassical model.

Let’s say there are two countries, A and B. Each economy has two sectors: tradable and non-tradable. For simplicity, let’s say labor is the only factor of production. For even further simplicity, let’s set the marginal productivity of labor in the non-tradable sector in countries A and B both equal to 1,

where the subscript NT means “non-tradable” and A and B are the countries.

Since we’re in Neoclassical Land, workers are paid their marginal product of labor times the price in the tradable sector:

and in the nontradable sector too (which we can simplify since MPL_NT=1):

Thus, wages equal the price of the non-tradable good. Workers are also free to move between the tradable and non-tradable sectors, which equalizes the wage between T and NT, but not between countries. And the price of the tradable, being tradable, is equalized across countries. Putting this all together and solving for the ratio of non-tradable prices between A and B gives

Let’s assume, without loss of generality, that country A is richer and more productive than B, i.e. MPL_{T,A} > MPL_{T,B}. Then the price of non-tradables in country A is going to be higher than in B, and thus the price level in country A is also going to be higher than in B.2

*** This tiniest of models gives us the Balassa-Samuelson prediction shown in that scatter plot: richer, more productive countries should have higher prices. Poorer countries should have lower prices. The line of best fit should slope upwards.

Intuitively, what’s going on is that higher productivity in rich countries increases wages for both the tradable and non-tradable sectors, making everything more expensive. When factory wages go up because of technological progress, this also pulls up the wages of hairdressers and barbers, raising prices for everyone.

Now we should ask—where’s the model going wrong?

One Man’s Points-Lying-Above-A-Line is Another Man’s Nonlinearity

Hassan (2016), which I think is a severely underrated paper, starts out with the acute observation that maybe the price level-GDP relationship isn’t a linear relationship at all. Here’s his version of the scatter plot using a 2011 cross-section of countries:

If you scroll back up to the original Gelb et al chart and squint, you can make out the same banana shape. One man’s points-lying-above-the-line is another man’s nonlinearity.

Hassan lays out a simple model to explain this U shape, which I think captures a key feature of the world missing in the simple Balassa-Samuelson model.

When thinking about the differences between rich countries, middle income countries, and the poorest countries, dividing the economy into two sectors—Tradable and Non-Tradable—like the Balassa-Samuelson model isn’t the most obvious way to slice things. Instead, in development, we typically think of three big sectors: agriculture, manufacturing, and services. The story of development is largely the story of structural transformation between these three sectors: the poorest countries farm; middle-income countries like China and Mexico manufacture goods; and most of the rich West does services.

How do these three sectors map onto the Tradable vs. Non-Tradable distinction? Manufacturing is obviously tradable. Services are obviously non-tradable. But agriculture is a mixed bag. Hassan chooses to model it as non-tradable: in practice, most of the poorest countries survive on subsistence agriculture, and hardly trade their agricultural produce at all. And for the richest countries, agriculture ends up being a relatively minor component of both output and the price basket anyway.

In other words, contrary to the baseline Balassa-Samuelson setup, developing economies go from non-tradables to tradables and then back to non-tradable services again, once they’re really rich.

These assumptions, and a little algebra, yield the augmented Balassa-Samuelson equation:

which, when fed data on labor shares l, expenditure share γ, and productivity A, voila, predicts the observed nonlinear relationship between price and GDP:3

Intuitively, poorer countries have more of their labor forces and consumption baskets tied up in non-tradable subsistence agriculture, so when agricultural productivity grows, it lowers the price level—the exact reverse of the conventional Balassa-Samuelson relationship. But at some point, enough people have moved into manufacturing that agricultural productivity growth becomes less relevant, and the traditional upward-sloping Balassa-Samuelson relationship reasserts itself. Hence: a U-shape between incomes and prices.

The Upshot for African Prices

We’ve taken a long detour from the original goal, understanding if African industrial labor costs are relatively high. Let’s bring it back.

In their CGDev paper, Gelb et al. argue that perhaps there are Africa-specific explanations for why African countries appear to have relatively high prices—these might include natural resource exports or aid. What Hassan (2016)’s model shows is that it’s possible to explain why African prices are high relative to GDP, without needing to add an epicycle specifically for Africa. (Gelb et al. also ultimately land at a similar place, positing that it likely has to do with low-productivity but high-employment agriculture.)

The clincher for me is actually less Hassan’s model, which shows one simple mechanism that is consistent with the data (albeit one more realistic than baseline Balassa-Samuelson). It’s what happens when Hassan plots prices against incomes from 1950 to 2011 as a panel, and not just in a single cross-section of time:

There’s that familiar U-shaped pattern again.

Perhaps African prices aren’t so high after all. What may seem at first a problem unique to Africa may simply be a reflection of a common experience of all developing countries as they undergo structural change.

Postscript (12/8/2024)

There was some Twitter discussion after the original post, asking: if all the points on the leftward points on the “U” (i.e., the poorest countries) are African, is the puzzle preserved?

Without good long-run panel price data, this will remain something of an open question. But since the World Bank released its 2021 International Comparison Program data I decided to take it for a spin. Below are my replications of the Hassan regression using the 2017 and 2021 vintages of the data:

The left two columns are with the linear Balassa-Samuelson, and the right two are with the quadratic term added. The key takeaway is that the Africa indicator disappears with the addition of the square term.

It’s not conclusive, but it suggests there’s juice in the nonlinear Balassa-Samuelson view.

There are two things that everybody knows about Singapore. One is that it is fabulously rich. The other is that it is a one party-state which subscribes to a meticulous brand of authoritarianism—something to do with chewing gum and the death penalty.

Then, I suppose, there is a third idea, dimly perceived—that the prosperity and authoritarianism are, in some murky but indisputable way, causally linked. It is this idea that makes Singapore globally important, far beyond even its sovereign wealth or GDP.

To be clear, Singapore has elections that are widely regarded as free and transparent, but they have been won entirely by the People’s Action Party over the past sixty years. Opposition parties exist, but tight restrictions on free speech, a close nexus between elites and the state, and electoral rules that favor incumbents mean that they never seriously challenge for power. In a few weeks, Singapore will see just its fourth Prime Minister since independence—and only its second from outside the Lee family.

How did Singapore get this way?

This is a story about a small incident, in a small country, that happened many years ago—but one that illuminates a great deal about development, authoritarianism, and the surprising role of Western advice in shaping both.

Subscribe now

Enter Winsemius

The founding father of modern Singapore was Lee Kuan Yew, Prime Minister from 1959 to 1990. Because of Singapore’s success, Lee has become a model for would-be enlightened despots from Kigali to Moscow—a figure of admiration for those who like their development with an extra dose of the cane.1 Lee’s singleminded focus on growth, his intolerance of dissent, and even his stranger proclivities (like his love of trees, or more darkly, his fondness for pseudoscientific race science) have left an indelible mark on Singapore’s design.

But, next to Lee and his lieutenants, one foreigner has an important claim to the coauthorship of modern Singapore: the Dutch economist Albert Winsemius, who was chief economic advisor for two decades of miraculous growth from 1961 to 1983.

In 1958, Britain granted Singapore internal self-government, and in 1959 Lee Kuan Yew’s People’s Action Party won a strong majority in the Legislative Assembly elections. Prime Minister Lee asked the United Nations to send an advisory mission to help Singapore industrialize. 

Winsemius answered the call.

The Winsemius Report

Before coming to Singapore, Albert Winsemius had overseen Marshall Plan spending in the Netherlands, helping his home country rebuild its manufacturing base after the Second World War. He had also advised the governments of Jamaica, Greece, and Portugal, giving him some experience with less developed countries.

In 1960 and then again in 1961, he visited Singapore with a research team to make his appraisal. The product of Winsemius’s two missions was a 1963 UN report, “A Proposed Industrialization Programme for the State of Singapore”.

The central problem was that Singapore needed to create jobs, fast. Winsemius guessed that the unemployment rate was somewhere around 30%, with informal service work like food hawking concealing high underemployment. Rapid population growth, boosted by huge inflows of migration, was only exacerbating the shortage of jobs.

And, in Winsemius’s eyes, the labor problem was not just economic. A strong trade union movement had taken root in Singapore’s working class population, particularly among its Chinese majority. Led by the charismatic Lim Chin Siong, the trade unions were at the vanguard of anti-colonial agitation. Recognizing their potency, Lee Kuan Yew had recruited Lim and the trade unionists into his People’s Action Party (PAP), folding their working-class support into a coalition that won the 1959 election in a landslide.

But because of the popularity of their strong anti-colonial stance, and perhaps because of their ethnic ties to Mao’s China, the labor unions were suspected to have communist links. The early 1960s saw substantial labor unrest; 1961 alone saw 153 strikes. To Winsemius, organized labor represented an existential threat to Singapore’s development project. He recounted in 1982:

It was bewildering. There were strikes about nothing. There were communist-inspired riots almost every day and everywhere. In the beginning, one has to very careful about passing any judgement—one does not know the country, one does not know the people, one does not know the men and women who are trying to steer this rudderless ship. But after a couple of months the pessimism within our commission reached appalling heights. We saw how a country can be demolished by unreal antitheses. The general opinion was: Singapore is going down the drain, it is a poor little market in a dark corner of Asia.

The Winsemius plan sought to strike at the heart of Singapore’s labor problems. Its centerpiece was a “crash programme” to encourage the growth of low-value added manufacturing, like making shirts and underwear, that could absorb large amounts of labor. This, at least, would take some pressure off of unemployment. In the medium run, Winsemius identified key sectors, like chemicals and electrical equipment, in which Singapore might have a comparative advantage. He also developed plans to encourage inflows of foreign capital, like the creation of the Economic Development Board (EDB), which eventually became an important vehicle for promoting investment in Singapore.

The report was sound economics. Winsemius correctly identified Singapore’s strengths—its large stocks of cheap but relatively skilled labor, its strategic position relative to larger export markets—and developed a complementary strategy of attracting foreign capital. And over the next few decades, following the rough contours of this plan, Singapore would grow at breakneck speeds, graduating from low value-added textiles to higher-value added industries, vindicating Winsemius’s basic analysis.

The Winsemius report marked the beginning of a fruitful twenty-year collaboration—a rare happy marriage between foreign advisor and government, in one of the great success stories of rapid economic development.

Operation Cold Store

But there was another crucial component to Winsemius’s advice, that went unmentioned in his official report to the UN.

In his memoirs, Lee Kuan Yew recounts:

I remember his first report to me in 1961 when he laid two pre­conditions for Singapore’s success: first, to eliminate the communists who made any economic progress impossible; second, not to remove the statue of Stamford Raffles. To tell me in 1961, when the communist united front was at the height of its power and pulverizing the PAP government day after day, that I should eliminate the communists left me speechless as I laughed at the absurdity of his simple solution.2

It's a remarkable admission. Here was a United Nations advisor, telling an ostensibly democratic government to “eliminate” its main political rivals.

In a 1982 interview, which you can listen to on Singapore’s National Archive website, Winsemius elaborated on what he meant by “eliminate”:

… as an economist, I’m not interested in what you do with them. You can throw them in jail, you can throw them out of the country, you can even kill them. As an economist it does not interest me, but I have to tell you, if you don’t eliminate them in government, in unions, in the streets, forget about economic development.

Of course, Lee Kuan Yew’s People Action Party had risen to power in coalition with the trade unions. But it had always been an uneasy alliance, and in July 1961, thirteen left-wing PAP members were expelled after abstaining or voting against a confidence motion in the government. They went on to form their own party, the Barisan Socialis (Socialist Front, in Malay), bringing a hefty chunk of the PAP organization with them. The PAP was left with the thinnest of majorities—26 seats in the 51-seat Legislative Assembly.

True to Lee’s word, the Barisan Socialis were indeed battering the PAP in the legislature, constituting a genuine threat to their hold on power. The common view—held by Lee himself—was that if an election were held before 1964, Barisan Socialis would win.

Compounding Lee’s problems were frustrating negotiations with the British and the Malayans for a potential merger. Lee’s steadfast view was that Singapore would not be viable as an independent city-state; by attaching itself to Malaya, it could ensure its security and guarantee an internal market for its exports. (This diagnosis of course proved incorrect; Singapore was expelled from Malaysia in 1965, and as of 2024 has a GDP per capita that is over seven times higher. But we are getting ahead of ourselves.)

The Malayans were anxious about adding this fractious, strike-prone city to their federation, which would also tilt the ethnic balance away from Malays to Chinese.3 As a precondition to go ahead with the merger, the Tunku of Malaya demanded that Lee deal with Barisan Socialis. The British, the Singaporeans, and the Malays worked out a plan—according to British documents, as early as April 1962, arrest orders for key opposition leaders had been drawn up by Singapore’s Internal Security Council (Jones 2000, p. 91). What they needed was a pretext.

On December 8th, 1962, left-wing, pro-republican forces rebelled against the British in Brunei. Barisan Socialis issued a statement of solidarity with the anti-colonial forces. Lim Chin Siong, it was reported, had also recently met with the leader of the Brunei revolt. It was all that the Internal Security Council needed. “Developments in Brunei had made it necessary to initiate action against the communists,” wrote Lee Kuan Yew in his memoirs, “and the statement by the Barisan supporting the revolt had provided the opportunity.”

The wheels of Singapore’s security apparatus were set into motion. On the morning of February 2nd, 1963, starting at around 3 AM, the police began rounding up suspected communist sympathizers. All told, over a hundred political leaders and trade unionists were arrested and detained without trial. 24 of them were members of Barisan Socialis—including the party’s secretary-general, Lim Chin Siong. Lim would remain in prison for six years, without ever facing trial.

Operation Cold Store, as it became known, broke the back of Barisan Socialis. The People’s Action Party went on to win 47% of the vote in the 1963 general election in September. With most of its leaders in prison, Barisan Socialis still came in second with 33% of the vote.

It remains the best electoral performance by an opposition party in Singapore since independence.

If you seek his monument…

It’s hard not to see Cold Store as a foundational event in Singapore’s constrained political life, where critics of the government are routinely sued into oblivion for ordinary acts of political speech, and the gnat-like opposition can never seriously challenge the PAP’s monopoly on power. Even the study of Cold Store itself remains politically sensitive topic—in 2018, the historian P. J. Thum was subjected to six hours of critical questioning in Parliament for his research.

The official purpose of Cold Store was to “safeguard against any attempt by the Communists to mount violence or disorder in the closing stages of the establishment of the Federation of Malaysia”. Sixty years on, the Singaporean government has not released any evidence demonstrating Barisan Socialis’s communist links, citing the need to maintain the confidentiality of its sources (most of whom are now dead). With no evidence forthcoming, it is impossible for any independent researcher to verify claims of a communist plot. For their part, contemporary British sources concluded that evidence that Barisan Socialis leaders Lim Chin Siong and Fong Swee Suan were communists was “very stale”, noting further “that there has been no recent proof of Communist activity or allegiance on their part” (Jones 2000, p. 92).

Now, was Winsemius’s advice really decisive? Probably not. With Barisan Socialis a political thorn in the side of the PAP, and the Malayans demanding their elimination as a precursor for merger, Lee didn’t need more reasons to get rid of them. But it is certainly remarkable that the first mention of the idea to eliminate Singapore’s political opposition, both in Lee’s recollection and Winsemius’s, came from this foreign economic advisor—a representative of the United Nations. If Winsemius deserves credit for envisioning Singapore’s economic miracle, he also shares responsibility for conceptualizing its authoritarian political turn.

Perhaps, without the crushing of it militant labor movement, Singapore’s headline growth numbers posted through the 1960s and 70s would have been less impressive. In the 1960s, when Fairchild Semiconductor—a precursor to Intel—was deciding where to place its second Asian factory (after Hong Kong), it chose Singapore for its low labor costs.4 Wages in Singapore were 11 American cents an hour, compared to 25 cents in Hong Kong, 19 cents in Taiwan, and 15 cents in Malaysia. A Fairchild semiconductor employee noted approvingly that the country had “pretty much outlawed” unions.

But it’s also possible to imagine a different path for Singapore, one perhaps with a little less headline growth, but one with a little more attention to those at the bottom of the income distribution, and—most of all—one where a thriving opposition can keep the government accountable. Instead, what Singapore got is an authoritarian party-state that, while responsive to the people’s needs, is never at serious risk of losing power—an example not lost on Deng Xiaoping when he visited in 1978.

Sixty years on, Singapore, next to China, has become one of the world’s bywords for authoritarian development. That it ever had a vibrant political opposition now seems almost unthinkable. That Western hands, too, had some authorship in the disappearance of this opposition, has been lost in time. Like learning that the Sahara was once an ocean, it simply doesn’t register beyond a remote intellectual level. The slate has been wiped clean.

Thank you for reading. Please consider subscribing to Global Developments to receive more regular essays on global economics, development, and poverty. Some recent posts:

• A historical deep-dive into the role of economists—and statistical malpractice!—in prolonging the Vietnam War

• Analysis of the developmental effects of the CFA franc—“Africa’s last colonial currency”

• The early life of Albert Hirschman—resistance hero and development economist—and his remarkable Strategy of Economic Development

This is a new blog about development economics, and there’s surely no better way to start than by writing about Albert Hirschman—one of the pioneers of the field, and quite possibly the most interesting person to ever become an economist. Originally I set out to write a review of Albert Hirschman’s 1958 Strategy of Economic Development, but the tale grew in the telling—into a biography of Hirschman’s fascinating early life, then a broader history of development economics, and at last into a reflection on the current state of the field.

Hirschman was a master of those little ideas that, once you’ve heard them, are impossible to shake loose from your head. Because of his unconventional life story (more on that in a moment), he was an outsider for much of his academic career. Unusual for his era, he did extensive fieldwork, which helped him to develop a wry literary style that emphasized in-person observation and an allergy to sweeping mathematical models that smoothed away the messiness of human behavior. Because of these proclivities, Hirschman and his ideas were basically exiled from the mainstream of economics in the 1960s and 1970s, swept away by the rising tide of mathematization.

And precisely because of these tendencies, I think his ideas are ripe for a comeback.

The Strategy of Economic Development is a classic of offbeat Hirschmanian insight. His main idea is simple: that the “ability to make development decisions”—not capital, or labor, or any other material factor—is the limited resource constraining economic development.

If that made you scratch your head, you’re not alone. For someone conditioned to think in Lagrangians, pointing to the lack of “decision-making” as the main obstacle to economic growth feels fuzzy around the edges, a little like declaring that the real development was the friends we made along the way.

But I’ll admit that at the end of the Strategy, I was won over. This is one of those rare economics books that reconfigured my view of the world, made me reconsider views that I once considered silly, and adopt much of its language in my own thinking–so much so that I’m trying to evangelize it to you right now. How did Hirschman do it?

Hirschman’s Odyssey

It is unfashionable to talk about economists’ life stories—and truthfully, you don’t usually get more out of the latest instrumental variables procedure by knowing about the author’s childhood. But Albert Hirschman lived a life inseparable from his ideals, and to omit it would deprive you of a fascinating story.

Hirschman was born in 1915 to a well-to-do Jewish family in Berlin. He was still a teenager when he joined the Social Democratic Party, the main political opposition to the rising Nazi Party. After the Reichstag fire, with Hitler threatening to take dictatorial control, Hirschman’s cell of young activists took to the streets, passing flyers and urging people to resist. But resistance soon collapsed. With several of his friends arrested, a seventeen-year-old Hirschman was forced to flee to Paris.

In Paris, Hirschman studied at HEC, one of France’s top business schools, but found the courses dull. When a fellowship at the London School of Economics opened up, he jumped at the opportunity. As the home of Keynes, England was the shining center of the economics universe; but Hirschman instead gravitated more towards the lectures of the free-marketeers Friedrich Hayek and Lionel Robbins—an early sign of his trademark anti-ideological streak.

In 1936, the Spanish Civil War broke out. Almost without hesitation, Hirschman went to Spain to fight for the POUM, a Marxist group on the Republican side. Hirschman rarely spoke about his wartime experiences, even to his family, but biographer Jeremy Adelman places him at the brutal battle of Monte Pelado, a costly victory for the poorly trained Republican volunteers. He left after three months, just as another famous POUM volunteer, George Orwell, was arriving in Catalonia.

Hirschman next went to Trieste, Italy to join his sister Ursula (a noted intellectual in her own right) and his brother-in-law Eugenio Corloni. With Corloni, he became involved in the anti-Mussolini underground, carrying secret documents between France and Italy in a false-bottomed suitcase. Somehow he also found the time to finish a thesis on international economics and earn his doctorate.

Resurgent antisemitism in Italy sent him back to France. With war against Germany looming, Hirschman joined the French Army, and was trained up just in time for France to surrender. His Jewish background and history of anti-fascism made him a natural target for the Nazis, so he abandoned his uniform, took a false identity, and fled to Marseilles, which had become a haven for other refugees trying to get out of Europe.

There he met Varian Fry, an American journalist who was organizing an underground movement helping Jews and anti-Nazi refugees to escape France. Hirschman’s quick thinking, many languages, and easy charm made him an invaluable asset. Recruited by Fry, he soon became indispensable to the movement. Hirschman’s portfolio from his Marseilles days reads like something out of Casablanca: he forged passports, smuggled funds through the French criminal underground, and scoped out routes for the refugees through the Pyrenees. Only when the Vichy police began asking questions about Hirschman himself did he decide to make his discreet exit. He hiked his way over the mountains to Spain, made his way to Portugal, and from Lisbon took a ship to the United States, where a research fellowship at Berkeley was awaiting his arrival.

Over the past five years, Hirschman had fought fascism in four European countries, earning two graduate degrees along the way. Over five months in Marseilles alone, he had helped thousands of mostly Jewish refugees–including Hannah Arendt, Marc Chagall, and Marcel Duchamp–escape to safety.

Albert Hirschman was twenty-five years old.

Perhaps inevitably, Berkeley was something of a comedown from Europe. The Berkeley economics department was still frankly a backwater—the Keynesian revolution had not yet fully reached California. But Hirschman made two important acquaintances: a young scholar called Alexander Gerschenkron, also fleeing Nazi persecution, also destined to become an eminent economist; and, in the Berkeley International House cafeteria, Sarah Chapiro, a Jewish emigre from France studying literature and philosophy.

Albert and Sarah married in 1941. When it was America’s turn to go to war, Hirschman enlisted (his third army in five years), and served as a translator during the Allied invasion of Italy. Bored and underused throughout the campaign, his most notable experience was at the end of the war, when he served as the interpreter for the German general Anton Dostler during the very first Allied war crimes trial.

After returning to America, Hirschman had trouble finding work—unbeknownst to him, he was under suspicion with the FBI for his wartime leftism—but he got a hand from his old colleague Gerschenkron, who gave him a job working on the Marshall Plan at the Federal Reserve Board. Hirschman spent six productive years at the Fed, where he helped to lay the groundwork for postwar European integration. But the Hirschmans came to feel suffocated by the Washington suburbs, which were coming under the grip of McCarthy’s Red Scare. In 1951, desperate to get out, Hirschman made the surprise decision to take a job with the World Bank in Bogotá, Colombia. It would prove to be a central junction in his career.

The Big Push

Here begins Albert Hirschman’s connection to development economics. For four years Hirschman worked as an economic advisor to the Colombian government, first as a representative of the World Bank, then as a sort of freelance consultant.

By all accounts, Hirschman’s years in Bogotá were happy ones—he liked Colombia, as did Sarah and his two young girls. Despite the tragic violence of the ongoing La Violencia, Colombia was experiencing rapid economic growth, and Hirschman got the chance to work on practical problems, like advising low-cost urban housing projects and helping the Colombian government write up loan proposals for the World Bank.

But professionally, Hirschman found himself adrift. He frequently clashed with his superiors at the Bank, who condescended to their Colombian counterparts and fixated on grand economic plans based on silly formulas. Moreover, while he had aspirations of becoming a public intellectual, he was cut off from the mainstream of American economics. So when a call came from Yale offering a visiting appointment, Hirschman seized the opportunity.

His timing couldn’t have been better. Development economics was just beginning to emerge as an independent field amidst the collapse of the European colonial empires, but there was a serious shortage of economists in American schools who had actually spent time in the field. Hirschman fit the bill.

Coming into academia as an outsider, he did not like what he saw. Development economics was dominated at the time by big, sweeping theories of growth, with one-size-fits-all prescriptions.

Perhaps the most influential these was the theory of the Big Push, which Paul Rosenstein-Rodan first laid out in 1943. Think of a poor, developing country. Put down a single shoe factory in the fields, and it will likely fail due to a lack of demand, since most people will be too poor to afford the factory’s shoes. But build a cluster of factories, and you’ve given them a fighting chance—with enough factories, enough workers being paid industrial wages, and enough of them buying each other’s stuff, the market might be large enough to support itself.

The Big Push is a demand-side case of “balanced growth” theory—the idea that different parts of the economy should remain in sync as they develop. Balanced growth theories can also take supply-side forms, like stressing that manufacturing should not race ahead of agriculture, or that proper roads and infrastructure should be laid down before industrialization takes place. But, whether they’re on the demand- or supply-side, balanced growth theories largely call for the same remedy: vast programs of coordinated investment, to try and jump-start the “modern sectors” all at once.

There’s a certain theoretical elegance to the Big Push/balanced growth view of the world. It’s a close cousin of the classic Keynesian story of depressions, where if everyone tries to save at once, all we end up doing is depriving each other of income (what modern economists call a coordination failure). And, of course, there’s an undeniable political appeal. For America and its allied development agencies, looking for intellectual justifications to shore up their Cold War allies, the Big Push was music to their ears.

Subscribe now

Slaying the Dragon

Hirschman, however, was not so easily impressed. Having seen firsthand the difficulties of getting things done on the ground—and the arrogance of foreign planners—he was skeptical of the grand promises made by the balanced growth theorists. During his fellowship at Yale, he worked furiously at a rebuttal–a project that eventually became his 1958 classic, The Strategy of Economic Development. In it, Hirschman set out to slay the dragon of the Big Push, and replace it with a theory more grounded in reality.

Hirschman’s main critique of balanced growth stems from a simple observation. “If a country were ready to apply the doctrine of balanced growth,” he writes, “then it would not be underdeveloped in the first place.” The capabilities needed to successfully kickstart the “modern sector” in one go–the systems of organization, the scientific and technical knowledge–are precisely the ones that need to be nurtured by the development process. In other words, balanced growth theory provides the correct but unhelpful diagnosis that a country’s failure to modernize stems from its lack of modernity.

Rosenstein-Rodan’s Big Push tries to solve the problem by starting the modern sector all at once, grafting it like foreign tissue onto the traditional agrarian economy. But even if the graft somehow sticks, Hirschman observes that the results can be quite unpleasant. A common outcome is a dualist economy, which you’ll still often see today around the world—a gleaming modern sector of skyscrapers and shopping malls, next to a traditional agrarian sector that remains desperately poor. Further examples are the persistent gulf in Latin America between indígenas and mestizos, and the foreign-owned plantations that are islands of “modernity” surrounded by seas of poverty.

Crucially, in the Big Push mindset of precise coordination, failure can always be chalked up to the shortage of some crucial factor (usually imported from abroad). Perhaps there was not enough cutting-edge machinery to make the power plant viable; or perhaps there was enough machinery, but more foreign experts were needed to get it running. There are so many lurking pitfalls that one can easily fall into a kind of decision paralysis.

Albert Hirschman’s main goal in the Strategy is to sidestep all of that. Here, at last, we can return to his strange central argument–that the scarce resource constraining development is not natural resources or capital or schooling or technology or any other tangible factor, but the ability to make development decisions.

Now, hopefully, this makes a little bit more sense. Rather than endlessly debate which prerequisites for growth a poor country is missing, Hirschman shifts the focus to figuring out how to make the best use of the resources it already has. But, without the guidance of a grand plan, how should policy makers decide how to allocate their attention?

The solution Hirschman proposes is unbalanced growth. Instead of trying to solve all problems all at once, policy-makers should push forward in a limited number of sectors, and use the reactions and disequilibria created by those interventions to inform their next move.

Take the example of an industrial district. With limited resources, a policymaker may have to choose between building the actual factories, or laying down the highways and power-plants (what he calls Social Overhead Capital) to supply it. Based on his theory, Hirschman makes the provocative case that infrastructure should sometimes follow, not lead, private investment:

… if we endow an underdeveloped country with a first-class highway network, with extensive hydroelectric and perhaps irrigation facilities, can we be certain that industrial and agricultural activity will expand in the wake of these improvements? Would it not be less risky and more economical first to make sure of such activity… and let the ensuing pressures determine the appropriate outlays for SOC [Social Overhead Capital] and its location?

In this example, once the industrial zone is built, it’s easy to imagine frequent blackouts and bottlenecks as trucks clog up the unimproved road. A Big Push theorist might see the disequilibria created by these initial ventures (what Hirschman calls “pressure points”) as evidence that their plan has failed. But Hirschman encourages us to flip the script. Rather than see bottlenecks and shortages as embarrassing signs of failure, policy makers can use these pressure points to identify opportunities where an additional investment might go the farthest.

With the road congested, it’s much easier to see it needs to be improved–and far better to upgrade a well-used road than build a highway to nowhere. The policy-maker, once inundated with choices, now has a clear guide on how to act. Even better, pressure points may organically summon other, non-state actors to try and address the now-glaring shortfall. Perhaps an independent power producer will spy a market opportunity, or a private railway may decide to step in and connect the factory to their network.

Hirschman formalizes this insight with his famous notion of backward and forward linkages. Backward linkages are the demand created by a new industry for its inputs, like steel for an auto plant or milk for a cheesemaker. Forward linkages are the reverse—the knock-on effects of a new industry’s outputs on the firms it supplies. Backward linkages, Hirschman goes on to explain, are better at spurring growth than forward ones. Rather than plopping down a steel factory somewhere, with no customers assured, it is far easier to build the auto plant first, sourcing the car parts from other countries as needed, then gradually entice local producers to enter the market. Instead of a Big Push across all industries at once, Hirschman calls for the Targeted Strike–choose the sectors with the most potential to create demand for other inputs, and support those.

Ahead of his time for economics, Hirschman also argues that the “nonmarket” responses induced by a policy change may be just as important as market ones. If, say, the factories in the industrial zone face a shortage of trained workers, the locals may clamor for new schools. Or if the trucks congest their neighborhood roads, they may pressure their local officials to improve the highway. Politics cannot be separated from economics when thinking about developmental choices.

Hirschman’s theory of unbalanced growth is rooted in empiricism, allowing policy makers to test and gauge the reactions of the specific context rather than applying some universal formula. It recognizes that development is naturally a chaotic, messy process, much closer to a “chain of disequilibria” than the result of a master plan. To paraphrase another important development thinker, Hirschman’s unbalanced growth is the modest call to cross the river by feeling the stones, one intervention at a time.

I could go on about the Strategy. Most modern economics papers, despite their ever-increasing length, contain only one or two ideas—the rest is just exhaustively dotting i’s and crossing t’s. Hirschman builds the Strategy around three or four big, field-changing ideas, then proceeds to drop a minor gem every couple of pages. I haven’t even mentioned his sly anthropological observations on nepotistic factory managers (“the last-ditch stand for [the] right to a quiet, incompetent existence”), or his canny prediction that developing countries should promote their exports in order to ease their balance-of-payments problems. There is a wealth of insight in The Strategy of Economic Development—one of those rare books that makes it impossible to see the world in the same way after reading it.

So why, then, has it been out of print for decades?

The Wilderness

Hirschman is still read today in political science and sociology, but hardly ever in economics, and even less in his home discipline, development. Backward and forward linkages have been incorporated into the conceptual language of economics, but Hirschman’s original, broader usage is long forgotten. Probably the only time his name actually comes up is the Herfindahl-Hirschman index used to study industrial concentration–but often we just shorten it to the Herfindahl index. (For the record, Hirschman came up with it first.)

In his classic 1995 essay, “the Fall and Rise of Development Economics”, Paul Krugman carved out the epitaph for Hirschman’s intellectual legacy—what he called “the High Development Theory”, of which the Strategy represented the high-water mark. The coup de grace had been delivered long ago, by economics’s mathematical turn:

High development theorists were having a hard time expressing their ideas in the kind of tightly specified models that were increasingly becoming the unique language of discourse of economic analysis. They were faced with the choice of either adopting that increasingly dominant intellectual style, or finding themselves pushed into the intellectual periphery. They didn’t make the transition, and as a result high development theory was largely purged from economics, even development economics…

Along with some others, notably Myrdal, Hirschman didn’t wait for intellectual exile: he proudly gathered up his followers and led them into the wilderness himself. Unfortunately, they perished there.

For his part, Hirschman knew the direction economics was heading. Hirschman’s biographer Jeremy Adelman notes that he was anxious to show his mathematical chops—the Strategy is scattered with stray equations as if to remind you of its seriousness. (Graduate students everywhere can surely relate.) But the math in the Strategy is ultimately ornamental; nothing is conveyed that isn’t already in the words. Hirschman eventually gave up on formal economic modeling, a path Krugman believes led to academic obscurity.

The bulk of Krugman’s essay is concerned with showing that this exile was unnecessary—many of the ideas of High Development could have easily been translated to the language of mathematics. To show just how easy it is, in a few pages Krugman works out Murphy, Shleifer and Vishny’s 1989 model of industrialization, which formalizes Rosenstein-Rodan’s 1943 insight about the Big Push. The algebra is straightforward; the model elegantly simple–nothing that the perfectly capable theoreticians of the 1950s couldn’t have done, had they known a few small tricks. (A minor inconvenience for Krugman’s account is that Rosenstein-Rodan’s “Big Push” was precisely the theory Hirschman’s Strategy was trying to debunk. But no need to stand in the way of a good story.)

Reading Krugman carefully, one can sense the intellectual tension: he sympathizes with Hirschman and admires his insight, but his loyalties ultimately lie with the advances in mathematical theory that seemed so triumphant in the 1990s. As a result, he concludes on a strangely ambivalent note:

The truth is, I fear, that there’s not much that can be done about the kind of apparent intellectual waste that took place during the fall and rise of development economics. A temporary evolution of ignorance may be the price of progress, an inevitable part of what happens when we try to make sense of the world’s complexity.

But what Krugman sees as a sad inevitability of the march of progress could just as easily be recast as a disastrous loss of knowledge. Nothing prevents economists from reading old books!

Within a few years of Krugman’s article, it was clear that development economics was stuck in a rut. Read William Easterly’s 2001 The Elusive Quest for Growth, a good summation of development thinking at the time, and the prevailing sense one gets is of frustration. According to Easterly, nothing seemed to work—not aid, not debt relief, and certainly not any of Jeffrey Sachs’s ideas. The highly-touted growth models of the 1990s proved to be not much help. Murphy, Shleifer, and Vishny 1989 is still well-cited, but has not become one of the standard models economists use to think about development. It is still respectfully included in many graduate syllabi, but more as a kind of curio, an evolutionary dead-end. More to the point, it’s doubtful that the paper’s much-vaunted algebra swayed any politician who wasn’t already convinced by the rhetoric of the Big Push.

In hindsight the unlearning of Hirschman’s ideas seem more like victims of economists’ rigid insistence that evidence take the form of mathematical models than necessary casualties of science. Indeed, post-2008 Krugman has made a second career publicly decrying the Dark Age of Macroeconomics—the deliberate forgetting of the hard-won Keynesian ideas of the Great Depression. It would be interesting to hear what he would make of this essay today.

The Friends We Made Along The Way

Far too many Wisdom of the Ancients pieces—those op-eds that claim Adam Smith had it all right, if only you hadn’t dozed through The Wealth of Nations; or Marx saw it all coming, it’s just all in the endnotes to Capital Volume III–end up wringing their hands in despair, bemoaning what their chosen prophet would think if they saw the state of economics today.

But not so with Albert Hirschman.

Since Krugman wrote his 1995 essay, development economics has undergone another intellectual upheaval. Highly formalized growth models are still around, but the center of gravity has shifted decisively towards randomized control trials (RCTs) and micro-level evidence—a more Hirschmanian approach if there ever was one. The 2019 Nobel Prize to Abhijit Banerjee, Esther Duflo, and Michael Kremer was, in a sense, just a belated recognition of this de facto intellectual triumph.

The methodological switch to RCTs has not been without controversy; I’ll leave that whole debate aside for another time. But the randomista revolution remains underrated in one crucial respect. Running RCTs has forced a whole generation of economists to leave their desks, go to the countries they study, and talk to the people who live there. In Chris Blattman’s words, we’re all Hirschman now—tramping through fields, collecting our own data, trying our best to not let theory obscure the evidence of our eyes. It’s not hard to see Hirschman’s ghost wandering through the Busia offices of Innovations for Poverty Action, nodding his head in wry approval.

Not to say that we’re perfect; nor even to say that we should seek approval from this defunct economist. For his part, I suspect that Hirschman would like development to use more qualitative evidence, and speak more to other social sciences, like sociology and anthropology. He would also probably wish that more economists finally learned to write better.

On our side, we should not underrate what we’ve learned since Hirschman’s time. For instance, while Hirschman takes the novel step of integrating politics into his analysis of pressure points, he tends to view it in rather antiseptic terms, treating it as a force that responds neatly to demands and pressures from the population. Writing in 1958, his view of development does not yet anticipate the likes of Mobutu Sese Seko or Ferdinand Marcos–or the continued US support that made their misrule possible. (Hirschman later returned to partly correct this oversight in his classic Exit, Voice, and Loyalty.)

On the other hand, failure to predict the future is forgivable—after all, we’re economists here. The more severe omission of the Strategy comes from Hirschman’s failure to look backwards. Hirschman was by temperament an optimist, more interested in finding solutions for poverty than wondering why people were poor in the first place. But the developing world of the 1950s was far from a blank slate. History is almost entirely absent from the Strategy. There are snippets, here and there, that suggest a broader awareness of the scars wrought by colonialism—like when Hirschman laments the division between indígenas and mestizos in Peru—but he never really advances the view that the lack of decision-making capacity may stem from these deep historical causes. It’s a shame.

Still, writing as a young economist, the continued relevance of the Strategy to development seems clear. He encourages us to see the inevitable pitfalls and stumbles of the growth process not as disappointments, but as opportunities, and gives us a conceptual language to identify them. For randomistas-in-training, steeped in the world of deworming and bed-nets and pre-analysis plans, Hirschman also reminds us that we need to step back from individual interventions more often and think more about development strategies–not just how different projects can complement each other, but also how each project might organically summon market and non-market forces to help growth along.

The Credibility Revolution has yielded, perhaps for the first time, robust evidence for individual program effects. The time is ripe, not to copy Hirschman’s ideas wholesale, but to borrow his clear-eyed approach and think carefully about how projects can be brought together, pressure point by pressure point, into programs for sustained development.

Subscribe now

This post originally appeared on my academic website in 2021. If you enjoyed this post, I highly recommend Jeremy Adelman’s biography of Hirschman, Worldly Philosopher, which was my main source. Here’s Adelman talking to Cardiff Garcia on one of the FT’s podcasts.

Also, please consider subscribing to Global Developments to receive more weekly essays on global economics, development, and poverty. Some recent posts:

• A historical deep-dive into the role of economists—and statistical malpractice!—in prolonging the Vietnam War

• Analysis of the developmental effects of the CFA franc—“Africa’s last colonial currency”

• Lessons in public housing policy from Hong Kong, Singapore, and Addis Abbaba for the developing world

After a nerve-wracking delay to the election, Senegal at last appears to have a new president-elect: Bassirou Diomaye Faye, the 44-year old opposition candidate.

Faye’s victory is welcome news for Senegalese democracy, which under the incumbent Macky Sall appeared at serious risk of backsliding. (We may want to postpone celebrations, though, until Faye is safely seated.) Faye is an intriguing candidate. Now the youngest elected president in Africa, he has vowed to govern with “humility and transparency”—starting with renegotiations of Senegal’s new foreign oil and gas contracts.

But perhaps the most intriguing part of Faye’s program is his popular call to reform the CFA franc.

The CFA franc zone is likely the most important developing country institution you’ve never heard of. Its fourteen countries, all in sub-Saharan Africa, peg their currencies to France’s—once the French franc, now the Euro—making it the second-largest currency union in the world. (More precisely, it is two currency unions—eight countries in West Africa, six in Central Africa—but their exchange rate is the same.) Over 160 million people live in countries that use the CFA franc, with a total GDP of about $283 billion. All but two countries, Guinea-Bissau and Equatorial Guinea, are former French colonies.

That both Faye’s platform and his opponent’s included reform of the CFA highlights its growing unpopularity—and the growing potential for meaningful change, which may even extend to dissolution. This raises the question: what have the implications of the CFA franc been for development?

Fixed or Floating?

The heart of the CFA franc is the fixed exchange rate: France guarantees that CFA francs can be freely exchanged at the set rate of 655.957 francs to the Euro. Pegged currencies often have short life expectancies, but the CFA franc has proved unusually stable, thanks to the backing of the French treasury. Outside of a single devaluation—in 1994, by 50%—both the West African and Central African CFA francs have tracked the French franc and the Euro almost one-to-one:

While other stipulations of the CFA zone have attracted controversy—most notably, the requirement until recently that the West and Central African central banks deposit their foreign reserves in the French Treasury—the fixed exchange rate is by far the most macroeconomically consequential part of the CFA regime.

In mainstream economic theory, fixing your exchange rate can have two major benefits over floating:

• Fixed exchange rates promote growth in an optimal currency area. Trade and capital flow more easily between countries when you don’t have to worry about fluctuating exchange rates, so fixed exchange rates are good for growth.

• Fixed exchange rates constrain the central bank, preventing inflation. Fixing your currency to a foreign one ties the hands of the central bank, which has to keep local monetary policy in step or lose the currency peg. (In the case of CFA Franc, individual countries no longer even have central banks.) Reckless, inflationary money-printing becomes impossible.

But each of these advantages also has a dark mirror:

• Fixed exchange rates promote growth… under certain conditions. If economic conditions between countries are not similar, there may be no positive growth effect. The exchange rate may even be fixed at an inappropriate level that is detrimental for growth.

• Fixed exchange rates tie the central bank’s hands… preventing responses to local shocks. Tying one’s hands to prevent reckless behavior is good, until you need those hands in an emergency. Central banks who peg their exchange rates to foreign currencies cannot use monetary policy to respond to local shocks. Growth may suffer.

Which of these conditions are likely to hold? To understand the full effects of the CFA franc, we can tackle each of them in turn.

Subscribe now

Suboptimal Currency Areas

Economists since the Nobel Laureate Robert Mundell have thought long and hard about what constitutes an “optimum currency area”—the conditions for a shared currency like the CFA franc to make economic sense.

In a seminal article from 1961, Mundell laid out four criteria for a successful currency union: common business cycles, labor mobility, capital mobility, and a fiscal transfer mechanism. The intuition is straightforward: if the exchange rate is bound across multiple countries, economic conditions need to be relatively similar for it to be appropriate for everyone. If not, then another policy instrument like fiscal transfers needs to compensate, or people need the freedom to adjust on their own, either by moving their bodies or their capital.

It is self-evident that the first condition, relatively similar economic cycles, does not hold. GDP per capita in the CFA zone is less than $1,500 per year. The CFA countries, particularly in the Sahel, are among the poorest in the world, with subsistence farming and herding livestock among the primary occupations. The price of their currency is set by a country on another continent, where incomes are 30 to 40 times higher. Mundell’s conditions were discussed endlessly during the 2010 Euro crisis, when the question was if Greece met the conditions for an optimal currency area with Germany. Whatever the arguments for union were then, they must surely be less tenable when considering the gulf between Germany and Burkina Faso.

The three remaining elements are labor mobility, free capital flows, and a system of fiscal transfers to buffer against shocks. While there is free movement within much of West Africa, migration from Africa to Europe remains a deeply contentious political issue, and there is no prospect of that changing in the near future. Nor do the meager transfers of Agence Française de Development count as anything close to a fiscal union. The CFA zone thus meets just one of Mundell’s conditions, capital mobility.

The proof is in the pudding. French colonies have fallen far behind their former British counterparts in real GDP per capita (below). Over fifty years of integration, the CFA’s trade with France has actually collapsed, from over 25% in the mid-1990s to just 5% today. More than that, CFA countries hardly even trade with each other—intra-regional trade is only 9% of their total trade. By any measure, if the goal was growth through economic integration, the CFA zone is a dismal failure.

Subscribe now

Free Hands Can Be Useful in a Fire

Far more often than trade, the most common argument you’ll hear in favor of the CFA is that it has helped countries put a lid on inflation. By tying the hands of their central banks and preventing reckless money-printing, the argument goes, the CFA countries have avoided a common scourge of developing economies.

And indeed, on this metric, CFA countries look pretty good. No CFA country has ever suffered a hyperinflation. Compared to their peers, CFA countries have also weathered the recent global inflationary shock relatively well:

But keeping inflation low cannot be the only goal of a central bank. Inflation must be traded off with growth. If it wanted, the Federal Reserve could raise interest rates to 15% and get inflation close to 0%. The problem is that doing so would take down the US economy.

Also, avoiding unnecessary recessions is one thing in the United States, with a GDP per capita north of $80,000. It is another thing entirely in countries with GDP per capita less than $1,500, who have struggled for decades to initiate the kind of meaningful economic transformation that could end extreme poverty.

How can we think more systematically through the growth-inflation tradeoff? In a 1991 paper for the World Bank, Shantayanan Devarajan and Dani Rodrik examine the tradeoff between monetary stability and lack of policy autonomy within the CFA franc system.

Rodrik and Devarajan tackle the question from an interesting angle: take the inflation and GDP numbers of the CFA countries as given, assume that their decision-making is optimal, and work backwards to figure out what policymakers’ preferences must be. For these poor, mostly rural countries, a tradeoff of 10 percentage points of higher inflation for 1 percentage point in higher real GDP growth seems reasonable. (Remember that going from 2% GDP growth to 3% means incomes double 12 years sooner.)

But, under standard assumptions about the sensitivity of output to the real exchange rate and terms of trade, CFA franc countries seemed to be trading 1 percentage point of GDP growth for just 1 percentage point of inflation reduction.1 In the most egregious case, Gabon was losing 1 percentage point of GDP growth to shave just half a percentage point off inflation. 

Either leaders of CFA countries have deeply weird preferences about inflation and growth, or the overarching policy regime is not rational. Devarajan and Rodrik conclude:

… our highly stylized calculations suggest that fixed exchange rates have been, on the whole, a bad bargain for the CFA member countries. For most of the CFA members, the inflation benefits do not appear to have been large enough to offset the costs on the output side. Under “reasonable” output-inflation tradeoffs, these countries would have been better off having the flexibility to adjust to external shocks.

In other words, the CFA franc is keeping a lid on inflation, but strangling growth in the process. And the problems go just beyond just the management of aggregate demand. As Dani Rodrik has argued, strategic devaluation of the real exchange rate was a key ingredient in the export booms that lifted Taiwan, Korea, and China out of poverty. Under the CFA franc, this kind of policy is a total nonstarter—one cannot complain about the incoherence of African industrial policy and the weakness of manufacturing growth without acknowledging that CFA countries are missing one of the key policy instruments.

Devarajan and Rodrik’s article came out in 1991. How have the CFA economies fared since then? Here’s that graph of the GDP divergence between former British and French colonies again:

I want to be careful about putting too much analytical weight on a single time series plot. But is it a coincidence that the GDP of former French colonies slid steadily downwards throughout the late 1980s, as the French Franc appreciated post-Plaza Accords—and that that decline halted around 1994, the year of the CFA Franc’s sole devaluation?

Nothing I’ve laid out here is a slam-dunk case; the developmental effects of this crucial monetary institution surely deserve a closer statistical examination. However, everything I’ve described is a fairly mainstream analysis of fixed exchange rates. In my view, the burden of proof is on defenders of the zone to show why the conventional rules don’t apply.

(Teaser: I’ve been working on an empirical paper with my coauthors Joel Ferguson and Abdoulaye Cissé that we hope to share in the coming months.)

Subscribe now

Africa’s Last Colonial Currency?

But to conduct a purely economic analysis of the CFA is also to miss the point. The problem of the CFA is, at its heart, political.

One central notion is sovereignty: a nation’s ability to control its own currency. Europeans clearly take that right very seriously. It took national referenda for countries to join the European Union, which carried an obligation to join the Eurozone and abandon their local currencies. Needless to say, these decisions remain deeply contentious. But the citizens of the 14 CFA Franc countries were never even given that choice.

A further ugly complication is that almost all the CFA countries have surrendered their sovereignty to their former colonizer.2 The map above shows the obvious continuity between France’s African empire and the modern West African and Central African CFA zones. Even the name CFA—which you may notice I haven’t yet defined—originally stood for Colonies françaises d’Afrique. After 1945, the acronym was left unchanged: CFA became the Communauté financière d’Afrique for 8 West African countries and Coopération financière en Afrique centrale for 6 Central African ones.

On the French side, the political economy for maintaining the CFA zone is fairly obvious. At independence, a 1960 article in the French newspaper La Croix was disarmingly frank about the currency’s economic benefits to the old colonial power:

It has allowed France to obtain supplies of various raw materials (lead, zinc, manganese, nickel, wood, phosphates, oilseeds...) without the need to disburse any foreign currency... It is estimated that the livelihood of 500,000 French people from the metropole depends on the economy of the franc zone.

(quoted in Pigeaud and Sylla, p. 32)

In the present day, I doubt that most of the French electorate could explain what the CFA Franc is, or articulate how it serves their material interests. But the fixed exchange rate certainly suits French special interests—most notably Orano, which supplies 10% of France’s civilian uranium demand with its mines in Niger—who can do business in Africa without a pesky currency conversion. The CFA also acts as one of the central pillars of Françafrique, France’s African sphere of influence, binding its former colonies diplomatically and militarily to Paris. The political scientist Ken Opalo has joked that, without nukes and its African sphere, France would just be Italy. In a nation that harbors Gaullist and Macronian notions of grandeur, this simply cannot stand.

On the African side, historically, the political elites of Françafrique enjoyed a close relationship with the metropole. The fixed exchange rate, generally overvalued, suited elites because it made imported goods cheap and eased the transfer of assets abroad. When there was popular discontent, and this arrangement was threatened, France was not afraid to enforce it. When Guinea’s Sékou Touré sought to leave the CFA franc upon independence, the French government launched Operation Persil, a covert operation to introduce counterfeit banknotes to destabilize the new Guinean currency. When landlocked Mali left the CFA zone in 1962, it found itself economically isolated from its neighbors, who raised commercial barriers. It returned to the CFA franc in 1984.

And these are the situations that are relatively transparent to prove.

The Road Ahead

But the geopolitical logic that held Françafrique together is starting to disintegrate. In June 2019, the Economic Community of West African States (ECOWAS) announced plans for a new common currency, the eco, which would exist independently of France and would potentially include non-Francophone countries, like Nigeria and Ghana. In December of that same year, Emmanuel Macron announced reforms to the West African CFA franc, relaxing the requirement that they deposit their reserves in the French treasury and renaming the West African CFA franc… to the eco.

Now, with a wave of military coups across West and Central Africa, ECOWAS is itself starting to fray. Last month, in February 2024, Burkina Faso, Mali, and Niger announced their withdrawal, in response to ECOWAS’s condemnation of their recent coups. The three countries are reportedly considering a new currency union of their own, separate from the CFA. And, on Sunday, we had the electoral victory of a candidate who has promised to pull Senegal out of the CFA franc. The political situation is incredibly fluid.

What remains true, however, is that even in countries that decide to leave the zone, France retains an enormous degree of influence—de jure, through its seats on the board of both CFA central banks and its hold on their reserves in Paris; and de facto, through its still-significant links to governing elites. If countries like Senegal choose to leave, it can make the transition process more difficult than it needs to be. It shouldn’t. One of the main defenses of the CFA seems to be its obscurity: French voters need to wake up, and demand an international policy more in line with France’s stated values.

The political change roiling West Africa also represents an opportunity for the United States. America should decouple its Sahel policy from France’s, as Georgetown’s Ken Opalo has argued, and adopt a stance more in line with what the people living there have been demanding, putting further pressure on Paris. In short, the West should pay attention to what African experts—from economists like Ndongo Samba Sylla to activists like Kémi Séba—have been saying about the CFA franc all along. (A good place to start reading, if you’re interested, is Sylla’s Africa’s Last Colonial Currency, coauthored with Fanny Pigeaud.)

It’s worth noting that leaving the CFA zone will likely be no growth panacea. As noted above, with the limited market integration of West African economies, even the proposed eco may not constitute an optimal currency area. Countries must also plan careful exits to avoid the risk of a spiral into inflation—perhaps an intermediate arrangement, like a shared peg to a more relevant basket of countries, may be in order.

But countries in the CFA zone deserve to make the choice of those arrangements themselves, free from foreign coercion. If the CFA zone continues, and if the population projections are correct, by 2100 France, a country of 74 million, will control the currency of a region of 800 million—a region which, over a century, it conquered, brutalized, and exploited, largely without repercussions.

Whether this is an economically efficient arrangement at least remains something of an open question. But whether it is a just one seems to me a closed book.

It always shocks me to remember how recent democracy in South Korea is. I was born in 1994. The events of the coup in 12.12: The Day—tanks in Seoul’s streets, gunfights between opposing military factions—occurred just 15 years before I was born, in 1979. 15 years ago from today is… 2009.

The rather drily titled 12.12: The Day—the highest-grossing Korean movie of 2023—depicts how Chun Doo-hwan, South Korea’s military dictator from 1980-7, seized power in a midnight coup, not long after the assassination of the previous dictator Park Chung-hee.1 Chun is possibly the most reviled figure in modern South Korean history, in no small part for ordering the massacre of hundreds of pro-democracy protestors in Gwangju in 1980. It speaks to the trauma of Chun’s dictatorship that 12.12: The Day has sold around 12 million tickets—a quarter of the country’s population.

Unfortunately, as a movie, 12.12: The Day is pretty bang average. Like Ridley Scott’s recent Napoleon, it falls squarely in the genre of the Serious Historical Drama, with title cards announcing the arrival of Important Historical Figures and the passage of Key Historical Events. There are good military guys (the fictionalized Lee Tae-shin) and bad military guys (Chun Doo-hwan) and a chorus of feckless senior bureaucrats gumming up the works.2 At the successful conclusion of the coup, the camera lingers on Chun triumphantly pissing at a urinal, emitting a laugh that Dr. Evil might have thought a bit much. The movie has basically no interest in ordinary civilians, or for that matter women: the only female speaking role beyond a few lines is Lee Tae-shin’s doting wife, who shows up at work to dutifully deliver her husband a scarf, I suppose to remind us that he is, indeed, a Good Guy.

But this is a blog about development, not the movies. Let me slip off my film critic beret, douse my cigarette, and put on the thick-rimmed glasses of political-economic analysis. There were two points relevant to development that I thought 12.12 captured particularly well. (Spoilers follow.)

Subscribe now

State Survivorship and Development

The first major theme I got from the movie is the ever-present threat of North Korea, which I think is essential to understanding South Korea’s developmental experience.

A major plot point in the film is if airborne divisions guarding the border with North Korea will be pulled away to join the ongoing coup in Seoul. For the generals on both sides, this is like crossing the Rubicon. Diverting troops from the frontline risked not just escalating the internal conflict, but inviting a North Korean invasion—and, possibly, the extinction of the South Korean state. Invasion was a credible threat. North and South Korea were far closer to military parity in 1980 than they are today; even with the breakneck growth under Park Chung-hee, South Korean GDP per capita only surpassed North Korea’s in 1974.

This points to a central theme of modern East Asian history: that bordering a stronger neighbor threatening to swallow it whole disciplines both political and economic policy. In a classic, Charles Tilly, states-make-war and war-makes-the-state sense, the South Korean state could not afford a Mobutu Sese Seko or a prolonged internal conflict, because otherwise it might cease to exist. Politically, the generals in 12.12 know this and bend over backwards to avoid a full-on shooting war. Economically, it’s also no coincidence that Park Chung-hee launched his famous Heavy and Chemical Industry drive, which essentially built the South Korean steel industry from scratch, just after the United States threatened to withdraw entirely from Asia. (You need a steel industry to make artillery shells and tanks.)

Taiwan, of course, faced a similar peril with Mainland China, which forced the Kuomintang to finally get serious about development. The exception that proves the rule is South Vietnam, the quintessential example of an American Third World client state—which collapsed because it was unable to develop a large enough power base willing to defend it. (My post last week discussed the failure of South Vietnamese land reform.) It’s a bit reductive given how few data points we have, but I’ll say it anyway—there’s a sense in which the developmental success of East Asia reflects a giant experiment in state survivorship, in which South Korea and Taiwan are the winners, and South Vietnam is the plane that didn’t make it back:

The Mechanics of a Coup

But back to the film. (Like Godwin’s Law, all social science discussions seem to lead eventually to the survivorship-bias plane.)

The second major theme, which I think the movie portrayed skillfully and absorbingly, is the actual mechanics of engineering a coup, particularly in a political context with weak institutions. I’m not enough of an expert on the precise play-by-play of the night of 12/12 to comment on its exact accuracy; but the dynamics it portrays are commonplace.

One critical aspect is the importance of informal networks when formal institutions are weak. Chun is the ringleader of Hanahoe, a secret society of officers largely from the 1955 class of the Korean Military Academy. From Gaddafi’s Free Officers Movement in Libya to the Regime of the Colonels in Greece, cabals of disgruntled mid-level officers are a perennial threat to weak regimes.

At one point, with the coup hanging in the balance, Hanahoe officers desperately work the phones, calling every contact they can to convince them to join them. This intervention proves crucial. Subordinates mutiny against their superiors or simply refuse to fight; pro-government forces are literally stopped in their tracks. These informal connections ultimately overwhelm the formal military chain of command.

Another aspect common to many developing countries is the tenuous intersection between civilian and military authority.

As the plot is being hatched, the senior generals insist that they first obtain legal authorization, either from the President or the Minister of Defense—even under Korea’s authoritarian Yushin Constitution, there were at least gestures towards civilian legal authority. Chun needs the generals’ support, and spends much of the coup’s early stages supplicating to President Choi in his office, trying to get him to sign the orders to arrest the Army Chief of Staff, his major obstacle to power. The President refuses. (This refusal is historically accurate.)

The film captures nicely the evolving logic of the plotters. At first, drawn in by promises of promotions and material advancement, they want only a limited sort of mutiny, deposing just the Army Chief of Staff with the formal protection of the law. The crucial turning point is when loyalist officers declare Chun and his allies outlaws, and orders their arrest. The senior generals on Chun’s side, who were skittish about a full-on coup, realize that they have crossed a point of no return.

This helped answer for me why so many military coups go through the hassle of seeking legal authority, when they are the ones holding the guns. Civilian institutions are weak, but not so weak that they can completely be ignored by any individual. When the plot is in its fragile infancy, the generals want the protection of the law, unsure of if they will succeed. (An unspoken assumption, I think, is that the Americans would have something to say about a right-wing coup without at least a fig leaf of legal legitimacy.) Once they are outlaws, the only viable option left is to go all the way, and depose the government.

At the end of the movie, with the loyalist forces neutralized, Chun returns to the President’s office, surrounded by his clique of officers. But by this point, the President’s signature is just a rubber stamp, authorizing events that have already occurred. Chun’s era of military rule has begun.