The shifting ideological winds of foreign aid donors have driven their policy towards governments in poor countries. Donors supported state-led development policies in poor countries from the 1940s to the 1970s; market and private-sector driven reforms during the 1980s and 1990s; and returned their attention to the state with an emphasis on governance and government social spending thereafter. Poor countries — sometimes called “low-income economies” or “least-developed countries” — have over the decades been a proxy battleground for the Western left and the right, with heated debates about the merits of infant industry protection, privatization of public utilities, government-provided health care, and the role of government more generally.

Both liberals and conservatives in the West have more in common than they realize, however. Their policy preferences have evolved in the richest countries in the world, with well-financed governments that carry out an ever-growing number of functions under the rule of law. Much of their sense of what is minimally acceptable in government is a relatively recent product of wealth. Yet they do not hesitate to apply their home-grown standards to governments in poor countries.

While it may seem obvious that the governments of poor countries are themselves poor, donors have largely failed to appreciate the enormity of the gap between the revenues of poor country governments and of their own governments. They have taken for granted the government-provided prerequisites that make their favorite policies work at home. When their policy prescriptions fail in poor countries, donors blame the failure on corruption and weak political will for implementation. But even with clean government and the best will, many of their policy prescriptions are unlikely to succeed. The poorest countries do not have the resources to build weaker versions of Western governments. Their attempt to do so spreads government too thin, making it impossible for the government to deliver on its commitments and making both law and government policy declarations aspirational. Instead, poor countries will have to build sustainable governments commensurate with their resources, which will necessarily be governments of more limited function. Donors shrink before the hard choices, and their reluctance to acknowledge them undermines the quality of government in poor countries and deepens poor country aid dependence.

Development and government

Books and articles that trace the intellectual history of development tell how development policy has been shaped by ideas of the role of government in richer countries. The story they tell divides development thinking into three eras: an initial era of state-led development; the “Washington Consensus” era, which focused on government retrenchment and market solutions; and the current era, which has been characterized as post-consensus or no consensus, but in which the pendulum swung back towards a more important role for government.

The idea that government had an important role to play in managing the economy, educating the people, providing social safety nets and reducing inequality was well accepted in the United States and other industrialized countries when most poor countries gained independence. The Great Depression and World War II led the United States to adopt Keynesian economic policies, intervene in the economy, and begin to create social safety nets first for returning veterans and then for others. Alternative models available to leaders of newly independent countries were even more statist — European and Latin American socialism, or Soviet or Chinese communism.

Initially, poor governments attempted to fill an expansive role in leading development. They regulated international trade, currency exchange, and domestic markets; established state-owned enterprises; sought to protect and nourish infant industries; and sought to raise revenue for infrastructure construction and the provision of social services, such as education and the delivery of clean water. But by the 1980s, many began to have problems even paying the salaries of their civil servants and extending law and order to their territories.

As bankrupt poor governments turned to the World Bank and the International Monetary Fund for loans, conservatives came to power in the United States and Britain. The international financial institutions began pressing for loan recipients to reduce the reach of the state and to adopt reforms, including privatization of state-owned enterprises, liberalization of trade and markets, civil service wage bill freezes, and fiscal discipline or austerity. Market ideologues claimed that those policies would spur growth and lower corruption, but the results were mixed at best in poor countries. The era of the Washington Consensus devolved into finger pointing, with the left accusing the World Bank and the imf of foisting unsound policies on poor countries in crisis, and the right accusing the governments of poor countries of having failed to implement the reforms properly.

The current era is sometimes called one of post-consensus, or no consensus, but the pendulum has arguably swung left again. The debt relief initiatives that sought to reduce the debt stock of bankrupt poor countries were conditioned on the implementation of poverty reduction strategies, which tended to focus heavily on government provision of social services. In the United Nations Millennium Declaration governments pledged by 2015 to halve poverty levels, create universal primary education, reduce maternal mortality, reverse major diseases (including hiv/aids and malaria), provide assistance to orphans, improve the lives of slum dwellers, and promote gender equality, among other goals. Implicitly, this assumes a role for government in delivering or accomplishing these objectives. To better enable poor governments to make effective use of aid and to carry out such objectives, donors since the 1990s have focused on improving the quality of poor country governance through technical assistance and donor projects.

This historical narrative of left, right, left is oft-told and familiar, and it is not wrong. However, it misses the biggest and most influential change in western ideas about government in the 20th century: the increased expectations of government function, quality of service, and the expansion of human rights norms concomitant with the tremendous growth in the size and functions of government.

Expanding ideas of government function

While liberals and conservatives have different preferences regarding the size and role of government, they are like two people sitting in a boat arguing about which way to row as the boat is carried rapidly downstream. In fact, the 20th century saw an explosive growth in government spending, size, and functions throughout the industrialized world.

A fundamental obstacle to development is the inability of poor-country governments to implement the policies that donors push.

In 1902 in the United States, for example, total government spending was about 0.8 percent of the gross domestic product, or about $36.7 billion in 2011 dollars. By 2010, total government spending had risen to 35 percent of gdp, or about $5 trillion. This expansion of spending was accompanied by an expansion in the number and the extent of government functions. A few of the new U.S. government activities of the 20th century include the creation of a social safety net through the Social Security Act of 1965 (including the creation of Medicare and Medicaid) and unemployment compensation, federal aid to public education and federal financial aid for students, and a host of new regulatory activities and the bureaucracies to carry them out — e.g., protections against discrimination on the basis of race or gender, labor laws such as those providing for workplace safety and restricting child labor, environmental laws and regulations including the Clean Air, Water Quality, and Clean Water Restoration Acts and the Endangered Species Act, and market-supporting laws and regulations such as antitrust, securities, and consumer protection laws. In addition to new functions, older functions of government grew in extent, complexity, and quality. Staff-patient ratios dropped in hospitals and student-teacher ratios dropped in schools where computers are increasingly seen as necessary teaching technology.

Nor was this expansion of government peculiar to the United States. Increased public sector spending prompted economist Adolph Wagner to propose Wagner’s Law in 1893. Wagner’s Law states that there is a long-run tendency for the public sector to grow in relationship to the size of the economy in industrialized countries.

Advocates have pressed to expand notions of human rights, with concomitant obligations for governments. The Universal Declaration of Human Rights, drafted shortly after World War II with strong participation from socialist countries, specifies in Article 25 a person’s right to “food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.” Similarly, Article 26 provides that everyone has the right to free and compulsory elementary education and to technical and professional education. While the drafting committee was deliberately ambiguous about the corresponding duty of states to supply these goods, these articles have been used to support an argument of government duty where such a duty has not simply been assumed.

The very wealth of rich countries is used as an argument for the emergence of new rights. In the United States, Jesse Jackson Jr. argued in a 2006 op-ed that government-provided health care “of the highest quality” is a human right on the grounds that “we can afford health care for all our citizens.” The Poverty and Race Research Action Council has argued for a “right to housing” because “America has the resources to guarantee” it.1 

People in rich liberal democracies demand more, receive more, and expect more from government than they ever have before. These expectations have arisen in parallel with increased wealth and increased government revenue that allow governments to deliver new and better services. However, both aid donors and recipients err when they transpose this understanding of government functions derived from rich countries uncritically to poor governments.

The revenue gap

The donor community has come to see the inability of poor governments to implement the policies that donors consider to be essential as a fundamental obstacle to development. Donors are focused intently on strengthening and reforming poor governments, by financing projects and providing technical assistance. They condition aid on the creation of new institutions, civil service reform, strengthening of budget and public expenditure systems, training of political parties and election monitoring, promotion of decentralization, and improving public service delivery. As a consequence, the “good governance” agenda for developing countries grew in importance. In 2010, World Bank lending for the Public Administration, Law, and Justice sector comprised 18 percent of the bank’s loans. In 2009 in Niger, for example, the United Nations, the World Bank, France, Germany, Japan, Canada, Italy, the United States, Spain, Sweden, Luxembourg, and Norway were funding more than 100 projects aimed at reforming or strengthening governance and civil society. Included were projects focused on promoting children’s rights, promoting women’s rights, reforming the juvenile justice system, anticorruption, conflict prevention, promoting the free flow of information, decentralization, and government capacity building.

Most donor attention has focused on what is termed the “lack of capacity” of recipient governments, meaning the lack of trained personnel, equipment, and infrastructure. Donors have attempted to build recipient capacity through training programs, “equipment drops,” the construction of infrastructure such as schools, hospitals, government buildings, and roads, incentive programs, and direct budget support.

Less attention has been given to the political constraints on the achievement of the good governance agenda because they are politically inconvenient. Many poor governments depend critically on the political support of persons who expect private benefits from government, such as government jobs, contracts, or opportunities for income generation through sale of influence or access to government property. While donors presume that the purpose of government is the delivery of public goods and services, this purpose is a distant second priority for those governments whose philosophy is, in the words of one development professional speaking privately, “Capture the state, bleed it dry, help your friends, and screw your enemies.” This is certainly an important constraint, because it means that there is fundamental disagreement between donors and many aid recipients about the purpose of government.

Donors may overlook fiscal constraints because there are no commonly used metrics for comparing government wealth.

Donors sometimes blame poor country governments for their “lack of political will” to implement agreed reforms and projects. But there is an additional and equally important constraint that prevents poor governments from implementing the donors’ policy agenda, including the governance agenda — a constraint that is simultaneously very obvious and somehow routinely overlooked. This is the fiscal constraint. The governments of poor countries are themselves very poor — too poor to afford the type of government institutions and services upon which donors often insist.

One reason why donors may be insufficiently focused on the fiscal constraint is that there are no commonly used metrics for cross-country comparisons of government wealth. Government budgets are too large to grasp intuitively. They are in different currencies. And as the countries themselves are not easily comparable, there is no reason to imagine that their governments should have the same budgets.

A metric that could give an appreciation for available government resources for the provision of public goods and services might be revenue per capita. It is an admittedly crude measure because the size of the population is not the only factor that could influence the cost of providing government services, which is also influenced by factors such as quality of governance, area, population density, geographic remoteness, or the differing mix of public and private goods offered by different governments. But by looking at best estimates of revenue per capita, one can start to get an appreciation for the enormity of the gap between the resources available to donor and recipient governments. By way of comparison, in 1902 in the United States total government spending at all levels was about $438 per capita in 2011 dollars. In 2010, it was $16,417 per capita. Our data on government revenue is sketchy for poor countries, but drawing on data from the Central Intelligence Agency revenue per capita was about $77 in Niger in 2011, $64 in the Democratic Republic of Congo, $52 in Afghanistan, and $44 in Burundi.

It is easy to see how important aid is to recipient governments, increasing their resource envelope by a hundred percent or more. Using 2010 data on aid from the OECD, Niger had $77 per person per year to spend in revenue, but if aid is included (although in practice, aid is often not controlled by the recipient government) this total is $122. Uganda had about $72 per person; with aid, it rises to $124. Afghanistan, with $52 per person, has $263 per person with aid, reflecting large aid flows to support the government as the war continues. However, even with these increases, the gap between the resources available to poor governments and those available to rich governments is monumental. The U.S. government still has 132 times the amount to spend per person that Uganda has, even counting the aid that Uganda receives.

The unfunded mandate

Both poor governments and donors have expanded the mandate of government far beyond what can be financed with poor country domestic government revenues. The unfunded mandate undermines the quality of governance.

Poor governments are adopting policies and institutions that they cannot afford to implement or sustain, often with donor encouragement, if not long-term commitments of support. The process began at independence, when former colonies adopted constitutions and formal government institutions modeled on those of their former colonial powers, and it continues today. In some cases, it is motivated by poor governments’ desire to provide their citizens with the very best — which is taken by definition to be whatever is provided in rich countries — without an understanding of or willingness to accept the constraint of poverty.

Democratic pressures also play a role. For example, Uganda adopted universal primary education in 1997 as the fulfillment of a popular campaign promise. Primary education was extended more broadly at the expense of quality of education, leading one early observer to comment that he could not reject the null hypothesis that, taking the deterioration of quality into account, the cost of education had not changed at all. While donors provided some support for universal primary education, lack of resources meant that while children would not be turned away for lack of school fees, neither were they guaranteed a teacher, a classroom or a book. The worst of the overcrowding was eased by donor-financed classroom construction, but teacher morale has plummeted and teacher absenteeism rates are among the highest in the world. (As the prescient British parliamentary undersecretary of state for the colonies observed in 1965 regarding African independence, “The moment you get fully representative government — one man, one vote — any political party, any government, is going to be under terrific pressure to spread education universally, but very, very thin. And nothing eats money more rapidly than universal free primary education. And yet, you may say, it’s the inalienable right of every child to have it.”2 ) Uganda went on to adopt universal secondary education in 2007.

Donors also contribute to the problem by failing to consider the cost to the recipient government of implementing donor supported and demanded policies, reforms, projects, and programs. And even where donors consider the initial costs to government of implementation, they often fail to consider recurring costs. For more than 30 years aid practitioners and scholars have called attention to the “recurrent cost problem” in aid, although donor awareness has not often translated into donor action. Donor funds pay for the construction of roads or schools but the future operation, maintenance, and repair costs fall on the aid recipient government. Too often the recipient government is either unwilling or unable to maintain the donor “gift,” and both donors and recipient governments ignore the liabilities that they are creating for future governments. As recently as 2005, donors issued the Paris Declaration in which they committed to provide support for recurrent costs, but there are few incentives for follow-through. Donors’ own budget cycles make it difficult for them to make long-term financial commitments, and supporting initial construction results in more visible and tangible outcomes that can be used to demonstrate donor effectiveness to their own domestic constituents. Efforts to convince donors to finance poor government budgets directly have been unsuccessful.  

While the recurrent cost problem is most obvious in infrastructure, where roads crumble and must be rebuilt after every rainy season, nearly every government policy and reform has implementation and recurrent costs. For example, donors support and demand legal changes, but rarely carry out an analysis of the cost of implementation of the policies for which the laws are vehicles. Antony Allott, a law professor at the University of London’s School of Oriental and African Studies, described in 1968 the failure to consider recurrent costs of legislative reform in the African context:

Whatever else African countries may lack in the way of modern armies, a literate population, an adequate infrastructure, large capital resources, and an experienced cadre of leaders at all levels, there is one thing they all have, which can be as rapidly and cheaply manufactured as paper money and which has the same tempting property of seeming to be available to solve all problems: that is legislative power.3

Allott described “phantom legislation”: “A law is enacted, but the regional facilities for its application are as weak as are the administrative cadres generally, and nothing of significance happens.”

Even democracy is expensive, as Ottaway and Chung pointed out more than a decade ago.4 They urged donors to consider the sustainability and recurrent costs of their democratization work, and in particular elections, political party financing, and civil society support. Elections are costly and the cost goes up dramatically in the presence of conflict. The 1994 elections in Mozambique cost $65.5 million and the 1996 elections in Nicaragua cost $45 million; the bulk of these costs were donor-financed. Cambodia spent $24 million on its national elections in 1998, $15 million in 2002 on commune elections, and another $11 million in 2003 on National Assembly elections for a total of $50 million on one election cycle, or in total more than 10 percent of the government’s budget in 2003.5 More than half of that cost was paid for by donors.

The mismatch between poor governments' resources and their mandates undermines the quality of governance.

The enormous mismatch between poor governments’ resources and their mandates undermines the quality of governance in a number of important ways. The adoption of laws and policies that cannot be implemented means that rule of law and equality before the law become impossible. The unfunded mandate increases government discretion as the government necessarily picks and chooses which of its policies or laws to implement, at what time, and for whose benefit. This discretion in turn allows government officials to sell government implementation (or non-implementation) of laws and policies to the highest bidder, fueling corruption.

It also makes government accountability impossible. When people are tasked with responsibilities that they could not possibly perform, how can they be blamed when they fail? When a recipient government fails to provide adequate education to those in its public education system, adequate health care to those in the public hospitals, or adequate care for those in prisons and it could not conceivably do so with the resources at hand, who can blame it? How can citizens judge the performance of government as a supplier of public goods and services? Instead, citizens are obliged to judge government by other criteria, such as its ability to deliver private goods and benefits that they can more easily monitor.

Lastly, the adoption of policies that cannot be implemented and institutions that cannot be sustained without external funding deepens and continues poor country aid dependence. Even the very functioning of government depends on donors, who are not able to commit to long-term reliable streams of funding. The dependence of poor country governments on donor funds gives donors strong leverage, which they in turn often use to press for expanded government obligations in a damaging downward spiral.

Delicate sensibilities

The premise of development is that poor countries’ economies will grow, and with economic growth comes increased revenue and a wider range of government policy choices. Poverty is not destiny, and there are a number of countries that, since the 1950s, have succeeded both in terms of economic growth and improved governance and graduated from the group of the poorest countries. But this still does not mean that Western governments are the ideal models for poor country governments, like hand-me-down pants into which one day every country will grow. There is no reason to imagine that every country in the world will one day be as rich as the richest countries, and even if so, it will not happen soon. If Burundi enjoyed a very respectable 5 percent growth in gross national income per capita year on year, it could aspire to be Mauritania in about 30 years and Botswana in 70. In the interim, rich countries’ economies will have been growing too, and doubtless their ideas of rights, entitlements, and government will have expanded further, forever moving the goalposts.

While paying them substantial lip service, donors have largely failed to consider the implications for government function of the limited and unreliable revenue streams available to poor country governments. To be sure, there have been calls for greater donor harmonization, so that the ever growing number of aid donors does not collectively overwhelm understaffed poor governments with its projects and demands. And there have also been calls for more thoughtful sequencing and prioritization of the seemingly all-encompassing donor governance agenda. But the problem is not just one of harmonization, or of sequencing or prioritization, which suggest that donors should put more thought into the question of how one builds a modern Western government. It is impossible for poor country governments to have modern Western governments with budgets of under $100 per person per year. Instead, sustainable and effective government in poor countries is likely to be very different in kind, with different and less expensive institutions and much more limited functions. Donors and aid recipients need to think about what an effective $100 per person government looks like, and what functions it will serve.

The fiscal constraint is not the only constraint that prevents effective government in poor countries, but it is so important that even if there were no other constraints it would be enough to prevent poor governments from effectively carrying out their current mandates. Why then haven’t donors and recipients grappled with the question of what kind of government poor countries can really afford?

It is impossible for poor-country governments to have modern Western governments with budgets under $100 per person.

One reason is that the idea that poor countries cannot afford the same kind and quality of government as rich countries is unpopular with both donors and recipients. Advocates framed education, health care, a pension, representation by a free attorney, a hearing by a judge, a minimal standard of living and care in prison, or government information on demand as human rights in order to raise them in priority over other demands on government resources and capacity. In doing so, they deliberately delegitimized the kind of cost-benefit analysis and balancing that can be used to make decisions about less important entitlements. This approach works as long as the government has enough money to provide these rights-based entitlements but leaves no guidance about how to make hard choices when the government does not have enough money to provide universal education, health, pensions or free legal representation. Donors cannot bring themselves to advise poor governments to adopt policies that would infringe on the entitlements they have come to see as human rights or to recommend policies that they themselves would not be willing to accept or that their constituents would find unconscionable. And recipients bristle at any implication that their lives or liberties might be worth less than those of people in rich countries, sometimes insisting on adopting the “best” policies from the rich countries because they know that they are not less deserving. Neither donor nor recipient governments are able to face the terrible choices that poverty requires.

The ideal of egalitarianism can be seen to require that every citizen benefit equally from government services. Article 21 of the Universal Declaration of Human Rights states, “Everyone has the right of equal access to public service in his country,” and this theme is repeated throughout. But egalitarianism that results in the overextension of government can mean that no citizens get adequate services if public services have “threshold effects” whereby a minimum must be delivered for a recipient to get a benefit. Three days of schooling, ten feet of road, or a dose of antibiotics short of curative do no benefit and the latter is positively harmful. And where a public service is a service in name only, extension should not be the first priority. For example, donors have been pressing for equal citizen access to the formal justice system in poor countries. But in many poor countries, the formal legal system does not serve anyone well, is feared for its corruption, arbitrariness, and use of torture, and barely functions in the capital. Extending coverage more broadly before quality issues are addressed is a dubious benefit and could even have the effect of lowering the quality of justice further as thin resources are spread thinner. But the alternative is to accept that some will lack access to the system, a difficult proposition.

Ottaway and Chung, after pointing out the unsustainable cost of some of the democratic institutions donors have worked to build or support in developing countries, cautioned:

Our argument is not that [less-developed countries] do not need or cannot afford democracy, any more than our comments about [the sustainability of infrastructure] development projects should be interpreted as a claim that these countries did not need schools or roads . . . Poor countries need democracy, but the democratic institutions and processes they can afford are limited, different from those in the established democracies, and probably less than ideal.6

While they were quick to disavow any claim that some poor countries cannot afford democracy, democratic rights, like any other rights, require money for meaningful implementation, and this means that it is indeed possible to be too poor to be a domestically sustainable democracy. If donors are to be effective they must be able to consider this and similarly awful possibilities in all seriousness, no matter how precious the value and no matter how uncomfortable the consideration. In fact, it seems likely that some of the poorest governments have insufficient resources to mount an effective government of even the most limited scope and functions. In that case, we must ask whether they are self-governing in any meaningful way, whether they can be in the foreseeable future, and what expectations donors and their own citizens should properly have of them.

Government “lite”

By pressing for broad unfunded mandates for poor governments, donors have helped set those governments up for failure and helped make rule of law and public accountability impossibilities, even as they claimed to be sponsoring interventions to strengthen rule of law and public accountability. Donors need to take poor government fiscal constraints seriously, and this will involve changing how they do business in a number of ways.

Donors should ensure that all donor-supported or donor-demanded reforms or proposed interventions be accompanied by publicly available fiscal analyses that spell out the costs to the government (and implicitly, the public), including long-term recurrent costs. Such analyses are a standard part of the legislative and rulemaking process in the United States, but donors’ fiscal analyses tend not to look beyond the period of anticipated donor involvement. Poor governments lack the capacity to carry out such analyses themselves.

Current thinking in development is that donors should not run parallel programs and agencies that compete with the government for scarce technical staff and risk marginalizing government in the eyes of its own citizens. Instead, poor country governments should run the programs. But unless such programs are accompanied by adequate long-term donor finance, making the government publicly responsible for delivering services and programs that it has no money or capacity to deliver leaves the government to be scapegoated when such programs end or fail to deliver; and at the same time, it fails to create real accountability, as the government had no capacity to deliver such programs in the first place. Rather than assume that all service delivery should be done by the government, donors and recipient governments should consider a clear public demarcation of donor and government responsibilities that ensures that governments are not publicly committed beyond their own ability to deliver. This in turn would allow governments to be held accountable for what they can and do deliver.

Recipients and donors together will need to reinvent government for poor countries, with the full understanding that such government will not be very similar to government in rich countries. Government that is fiscally sustainable with domestic resources will be much more limited in function, scope, quality, and the number of government-enforced rights. This is not a question of ideology; it is a question of pragmatism.

The recognition that poor countries cannot afford a wide range of universal public services and well implemented rights should not mean a surrender of aspirations or a renunciation of the belief in basic human rights and dignities. Government law and policy can be guided by those aspirations and beliefs even if governments are never in a position to give them full voice. It may not be possible to give quality free primary education to every child, for example, but it may be possible to give it to one or more children in every family or every village. (Uganda’s original target for its Universal Primary Education program was four children per family.) It may not be possible to assure that every defendant has a right to counsel, but it may be possible to assure that those facing the death penalty or sentences of life imprisonment do. It may be that the formal court system cannot adjudicate every type of dispute, but it may seek to assure a minimum quality of justice for the disputes that it can adjudicate. Only by bringing government functions and obligations back in line with government resources and capacities can poor countries reap the benefits of accountable and effective government.

1 Chester Hartman, “The Case for a Right to Housing,” Housing Policy Debate 9:2 (1998).

2 Eirene White, “Last Steps Towards Independence: The Three African Protectorates,” African Affairs 64:257 (1965).

3 Antony Allott, “The unification of laws in Africa,” American Journal of Comparative Law 16:51 (1968).

4 Marina Ottaway and Theresa Chung, “Toward a new paradigm,” Journal of Democracy 10:4 (1999).

5 See (accessed September 5, 2012).

6 Ottaway and Chung, “Toward a new paradigm.”

overlay image