My argument is that while enormous progress has been made during the post–World War II “development era” during which there was substantial multilateral and bilateral development cooperation, or “foreign aid,” that era is coming to a close. The main driving forces of the end of the relevance of development cooperation to the overall development endeavor is that the politics of the rich Organisation for Economic Co-operation and Development countries has shifted such that the broad development agenda of interest to the leaders (and citizens) of the “recipient” countries can no longer muster support. Instead, the mainstream, legacy development organizations supported by the Organisation for Economic Co-operation and Development (e.g., the World Bank) are caught in a cycle of increasing irrelevance and decline, and I argue that both the external political and the internal organizational dynamics of development organizations are past the point of no return.

The period from 1950 (or so) to 2020 (or so), which I will call the “development decades,” brought more progress in the human material condition for those living in the “developing” countries than all of previous human history combined (Pritchett 2022). This is true measured as overall economic condition, by material deprivation measures like extreme poverty, or by physical measures like health (child mortality) or schooling or access to basic infrastructure. For instance, the average number of years of schooling of an adult in the developing world in 1950 was 1.6; sixty years later, in 2010, the average was 7.4, an increase of 5.9 years for the average adult (Pritchett 2013). Since the stock of adult schooling in 1950 was the cumulative result of all previous human endeavor, this means the progress in just the sixty years from 1950 to 2010, 5.9 years, was 3.7 times larger (=5.9/1.6) than in all previous human history.

While it will almost certainly be impossible to parse out with any certainty how much the broad notion of “development cooperation” or “development assistance” or the field of development (in the broad, perhaps Bourdieuan, sense of “field”) contributed to that enormous progress in material well-being, both did happen. There was a set of organizations and agencies, both multilateral and national (bilateral), that had as an explicit goal the promotion of development. And development—by any measure—happened; not everywhere and not on everything, but in general, progress in the human material condition in the development decades has been phenomenal.

However, I suspect the combination of historical circumstances that made “development cooperation” a truly global and multilateral effort is nearing an end (Pritchett 2015a). The field of development is headed to a sharp bifurcation. The gap between what the countries in “the North” or “the West” (or the “old” Organisation for Economic Co-operation and Development, the twenty-four countries that joined the OECD before 1973—historically, “donor” countries) wish to support and what countries in “the South” (historically, “recipient” countries) want has grown inexorably wider, and the gap is now too wide to sustain a truly global cooperation in supporting broad-based development efforts.

My view is that existing multilateral and bilateral organizations for development that are primarily led and/or dominated by the (old) OECD countries are in a long, slow, inexorable decline. The “beneficiary” countries have largely lost interest in mainstream “development cooperation” as the PA/VA ratio for the “beneficiary” countries of working with these organizations has exceeded the tolerable threshold in all but the most desperate of circumstances. The PA/VA ratio is defined by how much “value added” there is for the “recipient” country versus how much a “pain in the ass” it is for the recipient country in actually getting that value added from the “donor.” A large part of the reduction in the value added of the legacy development cooperation agencies is that, in spite of occasional rhetorical protestations to the contrary, such as the Paris Declaration in 2005, they have lost interest in advancing the priorities of the recipient/beneficiary/partner countries and instead insist on a very narrow set of priorities of their own.

First, in 2019 the government of Ghana released a document outlining a “national agenda”—meant to go beyond a “government” agenda—titled Ghana Beyond Aid Charter and Strategy Document. This made it an explicit agenda of the Ghanaian government to end their receipt of aid. According to the OECD Development Assistance Committee (DAC), over the previous five years Ghana had received $6.3 billion in net official development assistance (ODA). While the document claims it is not “anti-aid,” a major (and popular) platform of the current president (who was reelected in 2020) has the explicit notion of getting “beyond aid.” The president’s national strategy could have ignored aid altogether or hailed donors as partners in the national strategy but rather chose to highlight “beyond aid” as a rallying cry. And it is not as if Ghana is a rich country—in GNI per capita (at official exchange rates) it is 48th poorest of 120 aid recipients (on a par with Côte d’Ivoire or Honduras) and still at only a fifth the level of income of Argentina or Malaysia. How much of a collective pain in the ass do you have to be such that an elected poor country government builds into its national strategy a goal to stop getting a billion dollars a year from you? Second, in June 2021 I was the external peer reviewer of a World Bank strategy document that meant to help an (unnamed) country with an effective strategy for growth. An explicit objective of helping the current government to accelerate growth was to legitimate the democratic politics of the country as there was a growing fear that the inability of “democracy”—that had arrived through popular protest a decade ago—to deliver on growth and jobs and rising incomes was threatening a reversal in democracy in the very near term (weeks or months not years). Yet as I listened in on the organization-wide review meeting, I think that something like 40 percent of all comments were concerns that there was not enough attention in the growth strategy to climate change. Whatever your views about the consequences of climate change, one might have hoped that people in a development organization could understand that in giving advice to a government whose realistic horizon to deliver on concrete benefits was at most a year, whether or not their actions were “green growth” or just any color growth had to be a secondary issue. But they couldn’t ignore their overarching climate agenda even for a minute. The government that the report was being prepared to help was in fact removed, irregularly, two weeks after the meeting—before, I suppose, the myriad of “more climate change” comments could be adequately incorporated into the document. Third, the shift in mainstream development organizations from a “national development” approach to a “charity work” approach that merely seeks to “cost-effectively” mitigate the consequences of the lack of national development is illustrated again and again with the sheer audacious madness of what the randomista movement chooses to evaluate (Pritchett 2020). A recent paper published in Nature (Bossuroy et al. 2022)—with no fewer than ten coauthors—used an RCT to evaluate the impact of different “treatment” arms in a “cash plus program” cash transfer programs in Niger. One of the elements of one of the treatment arms was “psycho-social interventions.” The abstract says: “The results highlight the value of addressing psychosocial constraints as well as capital constraints in government implemented poverty reduction programs.” I am worried we development types are all so inured to a programmatic, targeted approach to an arbitrary low-bar definition of poverty that this example doesn’t shock you. In Maddison-style very long-run historical purchasing power parity (PPP)-adjusted GDP per capita, the estimate for Niger’s 2018 GDP per capita is M$965. GDP per capita in France in 1 CE during the Roman Empire was M$956, and in the United Kingdom at 1000 CE (around the invasion of William the Conqueror) it was higher, at M$1,151. According to the World Bank 2014 consumption data, with a global moderate poverty rate of 2011 PPP of P$5.5/day (around P$2,000/year), it is estimated that 93.6 percent of the population of Niger is poor. People in Niger are poor because they live in an extremely low-productivity economy. People are not poor in Niger because of the choices they make, they are poor because of the choices they have.

Governments cannot “transfer” their citizens out of poverty if the underlying economy is unproductive. The Penn World Tables 10.0 estimate of national accounts government spending (the G in C+I+G+X-M) is P$149 per person per year. This is what the government of Niger deploys to address all of the needs and challenges of the people of Niger: security, law and order, education, health, infrastructure (of all types), regulation, administration, tax collection, etc. The only way Niger can reduce poverty (at any poverty line) to low levels is by massive increases in the productivity of their economy. And the “policy advice” of ten global economists is that there is a high benefit/cost ratio to giving citizens of Niger “psycho-social interventions” along with cash transfers. This is not development advice, it is “coping with a lack of development with a tiny amount of foreign funds” advice or “how to be more effective at charity work” advice. You can see where the government of Niger (or any person dedicated to the development of Niger) would be less than thrilled that this tiny tidbit about coping tactics is what is on offer as “development” from the World Bank. Fourth, when China and other developing countries created new development organizations, there was widespread enthusiasm, and a large number of countries joined in these new organizations. When the Asian Infrastructure Investment Bank was formed, even the United Kingdom joined up, even though it was widely (and correctly, in my view) seen as an alternative source of funding created in part due to borrowing countries’ unhappiness with the restrictions on infrastructure lending from the World Bank—restrictions created and promoted by shareholder countries like the United Kingdom. Let me lay out my argument, very briefly and schematically, with an alert to the reader that I have laid out these arguments more completely and more technically elsewhere and the references to those previous works account for what would otherwise be an absurd degree of self-citation. One, the original commitment to development was to “national development,” which was about a fourfold transformation at the level of country/society/economy toward (i) a more productive economy, (ii) a more capable state/administration, (iii) a more representative polity, and (iv) a society with a greater shared identity and more equal treatment of citizens. The early postcolonial leaders of the newly decolonized countries saw development as the natural counterpart of formal, legal decolonization, which was the transformation of their countries into fully “developed” countries that would play an equal role on the global stage with the then “developed” countries. The post–World War II/independence leaders of India (Nehru, Gandhi, Ambedkar), Indonesia (Soekarno), Egypt (Nasser), Vietnam (Ho Chi Minh), Ghana (Nkrumah), and Tanzania (Nyerere) did not see development as a minor improvement in specific indicators but rather a national and then a global transformation. While of course the ideological specifics of the path to national development and of the new global order were hotly contested, the leaders of the West saw that national development was the agenda, if not altruistically, then at least since national development was seen as a necessary element of the increasingly high-stakes Cold War. Two, national development was seen as instrumental to achieving the normatively important goals of higher human well-being or “human development” and solving the important and pressing problems of material deprivation (e.g., malnutrition, lack of clean water and sanitation, safe cooking fuels, adequate housing, healthy lives and health care, education, etc.). It was believed that higher levels of output per person, more capable states, more responsive governments, and a more cohesive society would be powerful mechanisms for nominating and solving the challenges confronting the developing world. On this point, the view that national development delivers on human well-being has been, and remains, strongly true as national development is an empirically necessary and sufficient condition for achieving high levels of human well-being, however measured (Pritchett 2022, 2022). Three, in the post–Cold War world, after the dissolution of the Soviet empire (and of the Soviet Union itself), there was a threefold dynamic. The first stage of the dynamic was that with the end of the Cold War, the political coalition of political Left and Right shifted such that there was much less bipartisan support for development assistance. There was a legitimate worry that part of the “peace dividend” of the end of the Cold War would be large cuts in development assistance. This led the advocates for continued development assistance to develop new arguments and appeals. The second stage of the dynamic was a new set of development goals and appeals for development assistance that ultimately “defined development down.” Rather than being about national development, development assistance was seen as increasingly about promoting a quite narrow agenda of specific goals drawn with very “low-bar” goals. Development was not seen as a shared donor/recipient agenda to promote national development that would lead to countries with greater and greater ability to improve the well-being of their citizens—and to ultimately be equal participants and partners on the global stage. Nor was it a movement to promote human development, defined as broad-based gains in human material conditions and improved well-being. Rather, the twin rise of narrow, low-bar development objectives framed around “extreme poverty” and the Millennium Development Goals (MDGs) attempted to create a vision of development that could be sold to voters and politicians in the donor countries. But this penurious vision did not at all reflect the broad range of needs/wants/wishes/aspirations of those living in developing countries. Let me offer the very briefest of explications of the long-run negative consequences for the field of development of the advent of “extreme poverty” and of the MDGs. The “dollar a day” standard as a global definition of poverty or “extreme poverty” was launched with the World Bank’s World Development Report 1990 (World Bank 1990). This measure acquired more and more attention over time, and the elimination of extreme poverty gained traction as if not the, then at least a central organizing objective of development efforts. The “dollar a day” standard was chosen literally because it was the lowest a poverty line could be and be defensible. The rationale was that national poverty lines tended to increase as countries’ GDPPC increased above a threshold level but that there was a rough floor to poverty lines, and no matter how low GDPPC was, countries tended to not adopt poverty lines below that threshold, and the “dollar a day” standard was roughly the poverty line of the ten lowest-income countries with poverty lines (Ravallion, Datt, and Walle 1991). The obvious converse of this logic is that the World Bank was advocating a poverty line that was lower than the actual poverty line of all but ten countries. I am not arguing against the idea of poverty or the idea of the elimination of poverty as a major organizing goal of development efforts. I am arguing against the combination of poverty as an objective and the adoption of a completely arbitrary, low-bar, penurious standard of “a dollar a day” (which is now, with inflation,$1.9/day) as the more or less sole definition of poverty and/or the adoption of the elimination of “extreme poverty” as the organizing goal of development and development assistance. The obvious alternative to a goal of poverty reduction with emphasis on a single low-bar poverty definition was a development strategy putting emphasis on reducing (i) “extreme poverty” using the penurious poverty line, (ii) “national poverty” at the poverty lines countries draw for themselves, and (iii) “global poverty” using the aspirational poverty lines of creating prosperity and a global middle class by adopting the poverty lines that the donor countries use for their own citizens as a global standard (Pritchett 2006).

There are a variety of arguments against a penurious poverty line, but perhaps the most important is that it is impossible to be simultaneously committed to three things: (i) the primary or overarching goal of development should be to eliminate extreme poverty, (ii) recipient governments should be the main drivers of the development agenda that donors support, and (iii) democracy in developing countries. When I was working in India in the early 2000s, my friend Montek Ahluwalia, then the deputy chair of the Planning Commission and a major force in India’s successful economic reforms of the 1990s, said to me (and though this is presented in quotes as they are roughly his words, I am paraphrasing from memory): “All you people at the World Bank want to talk about is poverty, poverty, poverty. But on your current definition only about a quarter of Indians are ‘poor.’ I realize that as international bureaucrats you don’t have to pay much attention to politics, but you must realize that to get elected in a democracy a government needs 50 percent of the votes. We cannot get 50 percent of the votes with an agenda for only a quarter of the population. Moreover, once a government is elected it is the government of all Indians. It cannot say to 75 percent of the population that their concerns are not legitimate and pressing concerns of the government. The more you insist on your narrow vision of poverty the less help you are to a democratically elected government of India” (see Pritchett 2013).

In Pritchett (2014) I show that more than 60 percent of the population is not in “dollar a day” poverty in seventeen of the top twenty largest-population developing countries. Of the top twenty largest-population countries, in only Nigeria, Democratic Republic of Congo, and Bangladesh (at the time) was the median voter also “dollar a day” poor. The “dollar a day” standard was so penurious it would prevent nearly all developing-country governments from adopting “eliminating poverty” as a political platform in the direct interest of the median voter. In contrast, at a poverty line even half that of the US poverty line, in only three countries was 60 percent of the population not poor (Mexico, Russia, Turkey).

The adoption and continued use of the “dollar a day” poverty standard was a political act of the rich countries that was believed to be advantageous in sustaining a political coalition in those countries for supporting development, but at the cost of defining a poverty line so penurious that it undercut what most developing-country governments (and certainly those in a competitive political environment) could adopt as the organizing goal or strategy for their own national development.

Similarly, the MDGs were an attempt to sustain support for development assistance in the traditional donor countries in a post–Cold War world by signaling global political support for a set of goals and then creating the idea that development assistance could be effective and “accountable” for progress on those goals. However, again the MDGs were both too narrow (both across and within sectors) and too low to truly be a developing-country government’s agenda. Let me briefly illustrate what I mean by (i) too narrow across sectors, (ii) too narrow within sectors, and (iii) too low.

Too narrow across sectors. The MDGs consisted of only eight goals. This meant that all kinds of issues of importance to people in developing countries were not on the agenda. As part of the deliberations in the run-up to the new Sustainable Development Goals (SDGs), there was a survey in which people could name what was in their top six of sixteen possible domains. The obvious areas of health and education had goals, targets, and indicators (though even these were narrow—see below), but many areas that were of concern to more than a third of the survey participants, such as “reliable energy at home” and “better roads and transport,” had no MDG goal, no MDG target, no MDG indicator. Even concerns that were in the top six for more than half, such as “better job opportunities” and “honest and responsive government,” were not MDG goals and had no specific indicator. Note that the two environmental issues in the sixteen priority areas were the least likely to be a priority for participants from the poorer,low Human Development Index (HDI) countries, but these were part of a goal and had indicators in the MDGs.

Table 1.
The MDGs were narrow compared to the expressed priorities of developing countries in the UN “My World” survey
 Sixteen possible priority areas in My World survey Percent naming as one of six top priorities Included in MDGs? Low HDI countries Medium HDI countries Good education 71% 58.3% Goal, target, and indicator, but only primary school and nothing about “good” education (no quality target or indicator) Better health care 59% 51.1% Goals (4, 5, and 6), targets, and indicators, but only on specific diseases/conditions or ages (under five) and nothing about “health care” or “better” care Better job opportunities 53% 50.3% No goal, a target (1.B), but no numerical indicators An honest and responsive government 50% 41.8% No goal, a subclause of target 8A, no indicator Affordable and nutritious food 42% 38.6% Hunger goal, target, indicator on malnutrition and intake Reliable energy at home 39% 25.3% No goal, no target, no indicator Better transport and roads 36% 33.4% No goal, no target, no indicator Protection against crime and violence 35% 41.0% No goal, no target, no indicator Political freedoms 33% 25.1% No goal, no target, no indicator Access to clean water and sanitation 33% 43.2% No goal, target (7.C) under “Environmental sustainability” goal, indicator Support for people who cannot work 32% 37.8% No goal, no target, no indicator Phone and internet access 29% 24.5% No goal, nonnumeric target, indicators Equality between men and women 29% 35.0% Goal, target, indicator Freedom from discrimination and persecutions 23% 31.0% No goal, no target, no indicator Protecting forests, rivers, and oceans 15% 33.0% Goal, target (7.A), indicators for forest and water Action taken on climate change 11% 33.4% No goal, part of target (7.A), indicator
 Sixteen possible priority areas in My World survey Percent naming as one of six top priorities Included in MDGs? Low HDI countries Medium HDI countries Good education 71% 58.3% Goal, target, and indicator, but only primary school and nothing about “good” education (no quality target or indicator) Better health care 59% 51.1% Goals (4, 5, and 6), targets, and indicators, but only on specific diseases/conditions or ages (under five) and nothing about “health care” or “better” care Better job opportunities 53% 50.3% No goal, a target (1.B), but no numerical indicators An honest and responsive government 50% 41.8% No goal, a subclause of target 8A, no indicator Affordable and nutritious food 42% 38.6% Hunger goal, target, indicator on malnutrition and intake Reliable energy at home 39% 25.3% No goal, no target, no indicator Better transport and roads 36% 33.4% No goal, no target, no indicator Protection against crime and violence 35% 41.0% No goal, no target, no indicator Political freedoms 33% 25.1% No goal, no target, no indicator Access to clean water and sanitation 33% 43.2% No goal, target (7.C) under “Environmental sustainability” goal, indicator Support for people who cannot work 32% 37.8% No goal, no target, no indicator Phone and internet access 29% 24.5% No goal, nonnumeric target, indicators Equality between men and women 29% 35.0% Goal, target, indicator Freedom from discrimination and persecutions 23% 31.0% No goal, no target, no indicator Protecting forests, rivers, and oceans 15% 33.0% Goal, target (7.A), indicators for forest and water Action taken on climate change 11% 33.4% No goal, part of target (7.A), indicator

Source: Pritchett (2015a), using data from http://data.myworld2015.org/

Too narrow and too “low-bar” within sectors. Education had a goal, which was Achieve universal primary education, and a target: Ensure that, by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary schooling. This is obviously only one part of what the education agenda of any developing country has to be, in two important senses. One, while there is debate about the relative priority and budget allocations, governments have to provide secondary schooling, they have to be engaged in higher education, they have to worry about vocational training. Two, early global fora on education, such as the Jomtien Declaration on Education for All, spoke eloquently of the need to both expand enrollment and grade attainment but also of the need to improve quality of teaching and learning in schools. This was not mentioned at all in a supposed “education” goal and target (which were really just “schooling” goals and targets).

Donor support organized around the MDGs could not be effective support to governments in improving their education systems to meet their development agenda and needs—which was also true of the health goals and the gender goals (Pritchett 2015a).

The third stage of the post–Cold War development cooperation dynamic, and the situation we are now in, is that many developing countries—particularly the two largest, China and India—having experienced rapid and sustained economic growth since the 1980s or 1990s, decided that if what “development” meant was a crimped and penurious vision of programmatic charity work aimed at appealing to a narrow partisan group in the West, then they had no stake or interest in it.

The Sustainable Development Goals of 2015 were the end of the pretense there was a global consensus about the “kinky” development of low-bar goals (Pritchett 2015b). Collectively, the developing countries decided they would rather have a set of development goals that truly reflected their broad and high aspirations than have a document that might be effective in raising a few more billions for development organizations to then bring to them with their increasingly high PA/VA ratios. Donors and philanthropists complained that the SDGs were too many, too broad, too aspirational to prioritize and focalize development actions—but they apparently did not realize that was a feature of the SDGs for many countries in the world, not a bug.

In the donor countries, the tactic of moving to a narrow development agenda to create a firmer political base created such a narrow base that the freestanding development agency no longer had sufficient political power to remain autonomous and has been merged into (more bluntly, “swallowed up by”) larger organizations. The Department for International Development (DfID) in the United Kingdom was recently merged to create the Foreign, Commonwealth, and Development Office (FCDO); in Australia, AusAid was merged into the Department of Foreign Affairs and Trade (DFAT); in Canada, the Canadian International Development Agency (CIDA) was merged into Global Affairs Canada. In none of those instances (or others) was there any indication these moves had “prodevelopment” intent or effect.

In the United States, the advocacy group the Modernizing Foreign Assistance Network had some hope that in an Obama administration, the head of USAID would be raised to cabinet status. However, while the administration made some early efforts at a “whole of government” approach to foreign assistance, these efforts largely petered out.

The World Bank remains primer et pares among the multilateral development agencies but seems on a path of decline in at least relative power (if not absolute size) and influence and relevance to the agenda of the large developing countries. Attempts to constitute a new agenda for the organization around “global public goods” seem doomed to fail, and even if the bank succeeded, seem likely to make it an even more minor tool serving the agendas of what the rich nations deem “public goods”—rather that its original mission as the International Bank for Reconstruction and Development. A colleague of mine who has been a prominent government official in a “developing” country recently said: “Every time I attend a World Bank meeting my blood boils—the sense of complacency, the ugly bureaucratic jargon, the obsessive political correctness, etc. Royal pain in the ….” But it is hard to see the World Bank turning itself around to either reduce the PA or increase the VA—hence it will likely stay on its current path of dignified irrelevance.

The author has no competing interests to disclose.

Lant Pritchett is a development economist from Idaho. He graduated from BYU in 1983 and received his PhD in economics from MIT in 1988. He worked for the World Bank from 1988 to 2007, living in Indonesia 1998–2000 and India 2004-2007. He taught at the Harvard Kennedy School from 2000 to 2019 and was the faculty chair of the MPA/ID degree program. Having twice retired, he is currently affiliated with Oxford’s Blavatnik School of Government as the research director of the RISE Programme, is the research director at LaMP (Labor Mobility Partnerships), and is a fellow at the London School of Economics School of Public Policy. He has published over a hundred works with over fifty coauthors, and his work spans a wide range of development topics including economic growth, state capability, education, labor mobility, development assistance, and more.

Bossuroy, Thomas, Markus Goldstein, Bassirou Karimou, Dean Karlan, Harounan Kazianga, William Parienté, Patrick Premand, et al. 2022. “Tackling Psychosocial and Capital Constraints to Alleviate Poverty.” Nature 605 (7909): 291–97. https://doi.org/10.1038/s41586-022-04647-8.
Pritchett, L. 2006. “Who Is Not Poor? Dreaming of a World Truly Free of Poverty.” The World Bank Research Observer 21 (1): 1–23. https://doi.org/10.1093/wbro/lkj002.
———. 2013. “The Dangerous Seduction of Kinks in Development.” Lecture given at the Center for International Development, Harvard Kennedy School, as Part of the Global Empowerment Meeting. 2013. https://www.youtube.com/watch?v=nUeikupfGds.
———. 2014. “A Development Agenda without Developing Countries? The Politics of Penurious Poverty Lines (Part I).” Center for Global Development. 2014. https://www.cgdev.org/blog/development-agenda-without-developing-countries-politics-penurious-poverty-lines-part-i.
———. 2015a. “Can Rich Countries Be Reliable Partners for National Development?” Horizons 2 (Winter).
———. 2015b. “The New Global Goals Spell the End of Kinky Development.” Center for Global Development. 2015. https://www.cgdev.org/blog/new-global-goals-spell-end-kinky-development.
———. 2020. “Randomizing Development: Method or Madness?” In Randomized Control Trials in the Field of Development: A Critical Perspective, edited by Florent Bédécarrats, Isabelle Guérin, and François Roubaud. Oxford University Press. https://doi.org/10.1093/oso/9780198865360.003.0004.
———. 2022. “Development Happened: Did Aid Help.” In Elgar Handbook of Development and Aid, edited by Raj Desai, Shanta Devarajan, and Jennifer Tobin.
Ravallion, Martin, Gaurav Datt, and Dominique Walle. 1991. “Quantifying Absolute Poverty in the Developing World.” Review of Income and Wealth 37 (4): 345–61. https://doi.org/10.1111/j.1475-4991.1991.tb00378.x.
World Bank. 1990. “World Development Report 1990: Poverty.” New York: Oxford University Press.