Keywords

Performance indicators, inspired by benchmarking techniques invented in the private sector, are increasingly used in public management, according to the notion that in order to survive in a competitive environment, countries must continually improve their organization by importing “best practices” that prevail among their “competitors” (OECD & World Bank, 2008). Alain Supiot calls this the “dogmatics of universal competition” (Supiot, 2009). These indicators are the cornerstone of results-based management (Managing for Development Results, MfDR) and have become the key criteria for measuring the effectiveness of public policies, on the basis of quantitative targets. Governments are anxious to gain points in international rankings by any means, focusing especially on the Human Sustainable Development Index compiled by the United Nations Development Programme (UNDP). In this index, countries are ranked according to a series of indicators that implicitly set a norm, suggesting that the “lowest performing” countries need to better their scores. In this framework performance indicators have become a formidable tool in the hands of a few international experts, who insidiously impose far-reaching public policy choices on states, in the absence of any real debate among citizens.

In Third World countries these indicators are in effect used by international funding bodies to orient development aid in the context of a change of strategy, shifting from “project aid” to budget support, in keeping with the Paris Declaration of 2 March 2005.1 There is now an international consensus that management of project aid, largely controlled by funding bodies, undermines the responsibility of recipient countries, hampers mobilization of resources and raises transaction costs. In response to this criticism, the ministers in charge of development from developed and developing countries, and officials of bilateral and multilateral funding organizations convened in Paris on 2 March 2005. At this meeting they resolved to reform delivery and management of aid as stated in the Paris Declaration on Aid Effectiveness. To make aid more effective the Paris Declaration underscored the following five principles:

  • Ownership: Recipient countries set their own public policy and strategy in the framework of Poverty Reduction Strategies (PRS), which are the reference documents of public policy;

  • Alignment: Donor countries align their support with the recipient countries’ national priorities, systems and procedures of public financial management, instead of each following its own system;

  • Harmonization: Donor countries coordinate their choice of sectors to support (education, public health, etc.), control and reporting procedures to be put into place, and share information to avoid overlapping action;

  • Results: Recipient countries work with partners to define results expected from funded programmes and on their evaluation;

  • Mutual accountability: Donors and recipient countries are both accountable for results achieved in development action.

The general budget support recommended by the Paris Declaration is intended to improve mobilization of resources and foster greater mutual responsibility of donors and beneficiaries of aid.2

In the course of the 1990s, some African countries, among them Tanzania, Uganda, Mozambique, were able to shift from project aid to budget support. In West Africa, Mali was a pioneer of budget support aid, under a Framework Arrangement for general budget support agreed in March 2006 by the government of Mali and funding bodies. Agreements for sectoral budget support in the areas of education, public health and decentralization and institutional development were concluded respectively on 18 July 2006, 19 July 2006 and 11 September 2009. Under these agreements, the disbursement of aid in these sectors is contingent upon attaining quantitative figures for a number of indicators.3 Advocates of this approach see in it a way to bring the authorities in recipient countries to report to funding institutions on the results of development programmes, in exchange for greater (although “under surveillance”) responsibility in the use of the resources attributed. To link action undertaken to results obtained and more systematically monitor performance, the funding bodies outline results chains in loan agreements. These results chains form part of the logical framework that connects the objectives, the activities pursued to attain them, the resources deployed and the results obtained in relation to those expected, and the impact achieved.

This chapter illustrates how these performance indicators have contributed to undermine public policies in Africa. We look at several examples drawn from experience in Mali, particularly in education and public health. This research highlights the harmful effects of performance indicators (Sect. “The Harmful Effects of Performance Indicators”), queries the meaning of management by indicators (Sect. “Querying Performance-based Management in Third World Countries”) and takes up the issue of the autonomy of aid recipient countries in management of their public policy (Sect. “Giving Aid Recipient Countries Greater Autonomy to Conduct Their Public Policy”).

The Harmful Effects of Performance Indicators

Upon close examination it becomes clear that the use of performance indicators in managing public affairs often leads to outcomes that are contrary to the stated objectives (Desrosières, 2014). In the education sector in Mali, while indicators pertaining to the gross enrolment rate in the primary education fundamental cycle (71.0% in 2013) and to the completion rate for the fundamental cycle (51.0% in 2013) (Ministry of the Economy & Finances, 2014) attest to progress in universal schooling, they have nonetheless had undesirable effects on the education system.

It should be remembered that the widespread introduction of performance indicators for education in Africa was based on thinking that lent credence to the idea that education in Africa suffered more from poor governance than from inadequate funding (Boone, 1996; Mingat & Tan, 1998). This neoliberal theory, very much in vogue at the World Bank at the time, was supported by international comparisons of the internal efficiency of education systems, measuring the number of years in school against the share of spending for education in GDP (see Mingat & Suchaut, 2000; Olivier & Orivel, 1999).

International rankings do indeed reveal the low efficiency of African education systems in the 1990s, compared to Asian and Latin American countries where pupils stayed in school twice as long, on average, for the same proportion of GDP spent on education. This theory, positing that school performance is not fundamentally related to the volume of financial resources allocated to education, had earlier been advanced in other contexts in Brazil and in the United States, most notably in the Coleman Report in 1966 (Hanushek, 1997; Harbison & Hanushek, 1992). Subsequently, this theory served as the “scientific” basis for a number of decisions that can be seen as factors in the present crisis of education systems in Africa.

Under agreements with international aid donors, the national education ministers of African countries were pushed to apply brutal reforms with the sole aim of improving the internal yield of the systems, without regard for the quality of teaching. In this context various highly debatable measures were taken in order to maintain the largest body of teachers possible with the lowest possible payroll costs. Continuing to follow international comparisons, we recall that in the 1980s the Bretton Woods institutions worked hard to demonstrate that African teachers were overpaid, and that this hampered universal schooling. They strongly recommended that these countries recruit more teachers, for the same aggregate payroll expenditure. On the strength of this advice salaries were sometimes drastically cut, as in Cameroon, resulting in a demotivation of the teaching body, with foreseeable consequences for the quality of teaching. This theory was also the basis for the creation of different categories of teaching personnel in all African countries (non-civil-servant status, contract workers, volunteers, etc.). The first non-civil-servant teachers were hired in Mali in 1991, in general, technical and vocational secondary education. This practice was extended to primary education in 1992, and at an ever greater pace thereafter. In 2002, 95% of new teachers hired were contract workers.

The only justification for the introduction of this new type of staff was the desire to create new categories of teachers, without civil service status and trade union representation, and poorly paid, in order to attain the targets and improve the internal efficiency of the system as measured by indicators. The aim was to formally achieve universal school attendance, to improve ranking under the Human Sustainable Development Index (UNDP) as well as according to the Millennium Development Goals standard (MDG).

Today the undesirable effects of this policy cannot be denied. Impartial observers recognize that non-civil-servant teachers lack both the training and professional ethics required for this function. Corruption in schools is largely due to the status of this group of teachers. Poorly paid at the outset, poorly trained, receiving little consideration and without any real perspective for professional advancement, these teachers had few reasons to observe ethical standards, in which they had never been inculcated in any event. Having never felt included in this profession, they brought the educational system to its knees.4 In hoping to economize on payroll costs, the country had disinvested in its essential resource, teachers.

Another unexpected effect of the massive introduction of non-professionals into the teaching body, in order to boost performance indicators, was to discourage civil-servant teachers. Many ultimately gave up in the face of the unorthodox practices of their young colleagues recruited from any number of different trades (accountants, plumbers, electricians, even individuals without degree or diploma) whose only motivation was to avoid unemployment. The most surprising aspect of this is that the Bretton Woods institutions, with all the brainpower within their ranks, did not foresee that once non-professional teachers had attained a level of critical mass within the system they too would seek to have the same advantages accorded to their civil-servant colleagues, which would wipe out the payroll savings purportedly achieved by hiring non-professionals. For quite a few years in Mali now, the issues pursued by teachers’ unions have focused mainly on acquiring the same remuneration for non-civil servants and for civil servants.

Another paradox of this system is that the conditions for the application of structural adjustment programmes led countries like Mali to close most teacher training schools in the 1980s, and to implement voluntary early retirement schemes for civil servants in order to reduce overall remuneration costs, thus creating a shortage of qualified personnel in the education sector.5 When in the 1990s social services were granted priority for hiring civil servants, there were no more qualified teachers in the job market. This led to a massive influx of insufficiently trained people in the categories of teachers created outside of the civil service, provoking serious disturbances in the education system (strikes, successive “blank” academic years, etc.). On top of this, in their desire to mechanically raise the schooling rate, education authorities were less rigorous in their oversight of ethnic-community-based schools and medersas, for instance, schooling dispensed in Arabic. In a country where French is still the official language this could only aggravate youth unemployment (Ministry of the Economy & Finances, 2014).

It was also with encouragement from the World Bank that African governments fostered the development of community schools, financed by village communities. These schools were housed in straw huts, with pupils seated on grass mats, without textbooks or school supplies. The teachers sometimes had not even completed primary school, and with salaries three to five times lower than remuneration in public schools, the quality of teaching was severely affected. Other highly debatable teaching reforms were undertaken in the pursuit of good indicator scores. In Mali, for example, one teacher might be assigned to teach two grade levels at the same time in the same room, the pupils doubling up on the benches (double load), or to teach two groups of pupils of the same grade level alternating morning and afternoon (double time).

The internal efficiency of a teaching cycle is also measured by the capacity to retain pupils from the beginning to the end of the cycle. To obtain a good score the system is likely to artificially limit the number of pupils repeating a year, even if the pupils have failed to improve, for the simple reason that repeating a year is considered to be a waste of public money (Mingat & Suchaut, 2000). Brutally applied, this way of thinking has without any doubt led to bargain-basement teaching in countries like Mali. To boost the completion rate for the fundamental cycle of primary education the National Education Ministry in Mali instituted in the 1990s a policy designed to pass as many pupils as possible to the next higher grade level, even with very poor results (as low as 2/10). This manifestly eroded the overall achievement level. Testing to measure basic skills (reading, writing, arithmetic) has shown a notable drop in achievement among pupils in Mali, and even a critical failure in subjects such as Mathematics and French (CONFEMEN, 2004). What is worse, and again with the notion of boosting indicator scores related to internal system efficiency, many African countries formed class groups of 100 or 200 pupils, assigned to harassed teachers, with disastrous consequences for the quality of teaching.

The consequences of these policies on teaching are such that today we question them, in pursuit of a more coherent system. Education systems are too complex to apply an analysis based on quantitative factors, as for factory production. If efficiency is defined as the ratio between a given result (duration of schooling) and financial resources consumed to attain it, this indicator is not in itself significant. We must look farther and measure the impact of the system as a whole. More broadly, one of the effects of mechanical application of performance indicators has been the creation of a two-tier education system in all African countries; public schools are open to all, with the prime mission of raising the schooling rate, while schools in the private sector are reserved for a rich minority of the population who seek a certain level of quality, and at the same time lessen pressure on the public purse.

The pernicious effects of performance-based management can be seen through indicators that measure budgetary expenditure in sectors deemed to be high priority. While indicators that measure the share of the national budget devoted to so-called high priority sectors (education, public health, agriculture) provide an incentive to concentrate resources in these sectors, they also have the drawback of pushing the government administrations in question to spend public money on measures that do not always improve the provision of services to the public.

In addition to questions of the quality of public expenditure and the capacity of these sectors to absorb this money, the use of these indicators drives ministries to analyse their budgets not in terms of needs, but in terms of percentage of the national budget, leading to sterile quarrels and competition throughout the budget process. In this respect it can be observed that priorities have shifted to infrastructure investment in the most recent generation of Poverty Reduction Strategies. As grain prices have been rising on world markets, agriculture, which was until recently neglected, is moving up in the list of priorities. While one can only express satisfaction with this particular evolution, as a general rule we observe regular changes in priority goals, an illustration of the fragility of objectives when they are not popularly supported.

Successive reforms, sometimes abandoned before completion, often for reasons related to the international agenda, without consultation of the people, is a source of real instability in public policy in developing countries. In this respect Africa appears to be a laboratory for all sorts of experiments.

In the area of justice, indicators pertaining to the number of judges per capita and to construction and renovation of judicial infrastructure (courthouses, prisons, housing for magistrates, etc.) give an idea of public budget support for this sector. These indicators do not, however, in themselves give information on the quality of justice, or on the independence of the judiciary, and even less on the level of satisfaction of those who have recourse to the courts. Quite the contrary, application of these two indicators combined pushes governments to concentrate their efforts on hiring judges and building courthouses, to the detriment of other crucial measures to institute fair remuneration of judicial personnel and to strengthen the ethics of the judicial apparatus as a whole.

One of the great weaknesses of “trigger indicators”, i.e. those that determine the amounts of aid to be paid annually under funding agreements, lies precisely in the fact that recipient countries have incentives to focus on attaining the objectives that have been given, to the detriment of other aspects of public policy. The quest to reach the target figure set for the indicator becomes something of an obsession, detached from the policy it is meant to support. The finance ministry, anxious to cash in on the money promised by donors, puts pressure on other government administrations to attain the key indicator levels, and there is a great temptation to doctor the figures when objectives are not reached.

But a national indicator, by nature, does not reflect the conditions and the results of a public policy in distinct geographic areas. There are indeed strong disparities between regions. Looking at education in Mali, while the national rate of completion of the first cycle of schooling was 51.0% in 2013, this average encompassed significant differences across regions. The Bamako district registered the highest completion rate, with 77.9%, while in the districts of Kayes, Koulikoro, SĂ©gou and Sikasso this rate stood at 44.4, 64.9, 49.5 and 53.5% respectively (Ministry of the Economy & Finances, 2014). The same disparities are found for public health indicators and access to drinking water. Globally speaking, the Bamako district is situated well above the national average, while interior regions, in particular in the north of the country, have lower scores. This underscores the dilemma of the northern part of the country, where repeated and ongoing rebellions are in part linked to differences in development between regions. These differences are not reflected in the national indicators published in international rankings.

Along the same lines, certain indicators must be carefully examined, and cross-analysed in conjunction with other data. While the pupil/teacher ratio for fundamental primary education (first cycle) is a relevant indicator, in that it describes the level of adult interaction with pupils, it must nonetheless be counterbalanced with the fact that in the schools with the most teachers (in cities) many of them are deployed in administrative positions, rather than in classrooms, or are simply extra staff on hand.

The public health indicator stating the percentage of the population that lives within five kilometres of a functional healthcare centre (56% in 2013) does not inform us as to the real accessibility of healthcare (Ministry of the Economy & Finances, 2014). We know that due to the paucity of medicines available in healthcare centres, and frequently to understaffing (as centres are located farther from large cities), the local population has little reason to go to these centres. People, especially the poorest, are likely to have recourse to traditional medicine, or worse to obtain remedies from “itinerant pharmacists” who are quacks and deliver counterfeit drugs. According to the 2013 report on implementation of PRS credits, barely one in three people in Mali visited a healthcare centre in the course of the year (Ministry of the Economy & Finances, 2014). It is rare that someone living in Mali would not have a health problem at least once a year, especially as malaria is endemic. The fact is that the economic accessibility of healthcare is the real challenge, in all African countries.

Furthermore, some indicators are difficult to quantify, contrary to what some may think. Two examples in the area of national health insurance (“sécurité sociale”) coverage are the percentage of the population covered by secondary insurance (“mutuelles”) and the percentage of the poor population that has medical coverage. This is a real preoccupation, as traditional social safety nets are becoming weaker, and are less and less able to cushion the social problems of the poor. However, it is difficult to compile reliable information on secondary health insurers, their subscribers and the services offered, as this is an informal sector that has sprung up in a legal and institutional framework that is inadequate.

It can also be noted that indicators are sometimes far from comparable between countries, contrary to what is intended. The 2013 report on implementation of PRS credits gives an unemployment rate of 7.9%, which is quite surprising when compared with that of developed countries (Ministry of the Economy & Finances, 2014). Knowing that this figure refers to persons of working age who meet both of the following conditions—without economic activity during the period, or without formal employment and availability to take a job–it is easy to understand that the indicator masks massive under-employment, to the order of 80% in most African countries. The vast majority of people considered to be unaffected by unemployment in fact work in subsistence agriculture or in the informal sector, and work only a few months in the year. The expansion of the informal economy is one of the key mutations of the labour market in Africa over the last two decades. The structure of employment has been considerably transformed by a shift from jobs in the formal sector to work in the informal sector, which has become a sponge absorbing the urban workforce. The informal sector adapts quickly to the real circumstances of the labour market, and the ease of taking and leaving work makes it attractive, in particular for vulnerable groups such as women. Research has shown that even within a single country, comparison of unemployment figures over the long term is problematic, because of the transformation of the phenomenon itself (on this issue see Salais, 1986; Thélot, 1985).

Querying Performance-Based Management in Third World Countries

In technocratic circles in the Third World, performance indicators are considered to be the miracle cure that will ensure proper use of public money. The most sceptical of technocrats see these tools as a practice imposed by international donors to force the leaders of recipient countries to finally be accountable for their management. But few people seek to understand what is really at stake. Indeed, these new tools of international cooperation based on quantitative objectives may appear comfortable for managers. A time frame and evaluation grid that are known in advance are components that can be reassuring for policy chiefs, who are looking for immediate results.

It is nonetheless true that a mechanical application of these tools leaves no room for in-depth thinking to accompany complex processes that involve fundamental changes, sometimes even a cultural revolution, setting off profound movements in society. One of the problems of performance-based management is to define relevant and legitimate targets. Reaching a consensus on targets is truly a challenge, as it is difficult to set priorities that are acceptable to a broad majority. While it may be easy to agree on objectives in a corporation (lowering costs, raising productivity and profits, etc.), finding a consensus is much more arduous in the public sector. The aims of public policy are multiple, and sometimes contradictory, so reaching consensus can be extremely laborious. It is true that it is the duty of those who govern to manage public resources properly and report on their use to citizens, and this makes the principle of Managing for Development Results (MfDR) legitimate. But first there must be agreement on the meaning and substance of this concept, for the results culture is not the same everywhere on the globe.

Let there be no mistake, the choice of indicators is not politically neutral. Since the beginning of the 1990s, in the wake of the fall of the Berlin Wall and with the growing move to democracy in Africa, “governance” has become the central theme of international aid and cooperation, the lens through which the relations between international donors and aid recipients are seen (James, 1998; Olivier de Sardan, 2007). A technocratic conception, inspired by the aims of reducing deficits and restricting the role of state government, now governs the relations between aid donors and recipients under cooperation agreements, with considerable room for sanctions against countries that deviate from prescribed practice in terms of good governance (on this issue see Meisel & Ould, 2008). In reality in Western countries, fervent advocates of MfDR (Canada, the United Kingdom, the United States, etc.), this reform inspired by neoliberal economic thought is intimately tied to systematic budget cuts as part of a move to clean up public finances, and these cuts become an end in and of themselves. Without a doubt the neoliberal free market movement of the 1980s that propounded the notion of “less government” and relegated social and cultural issues to the back burner had major repercussions on public policy in countries that receive aid.

What should we make of the fact that a country like Mali, heir to an ancient culture, home to world heritage sites in the old cities of Djenné, Tombouctou and Gao, guardian of thousands of ancient manuscripts, has no public policy indicator pertaining to the safeguarding of these treasures?

Likewise, social protection is analysed within the confines of the official outward-looking system, that posits the modern system as its model and refers only to the tools used in official regimes, which are meant to be extended more or less as is, to the rest of the population without taking into account peoples’ needs, or existing mechanisms. It has to be recognized, however, that the state of healthcare coverage in Africa cannot be remedied using only existing tools associated with formal insurance regimens that cover between 5 and 10% of the population. There are other mechanisms of solidarity and redistribution, however. Research has described the faltering of traditional solidarity systems, under the effect of increasing focus on the individual, a process in African societies driven by monetization, urbanization and the consequences of neoliberal economics. This has led some authors, such as Axelle Kabou, to question the capacity of African society to stand up in the face of these social changes (Kabou, 1991). Others, including Vimard and Locoh, are willing to bet on the adaptive capacity of African solidarity mechanisms (Locoh, 1988, 1993a, b, 1995; Vimard & N’cho, 1997;). A number of solutions can be proposed as ways to integrate traditional solidarity into the global system of social protection. It is clear that indicators pertaining only to these covered by modern social security systems cannot be expected to reflect the reality of social protection in Africa.

In this context, how can a new social contract be devised, setting reciprocal obligations for families, communities, local authorities, civil society and the national government, in order to ensure that the poorest members of society have effective rights to social protection? What forms of consultation and cooperation between all the players are needed to renew thinking and means of action to address the real needs in the sector? These are the fundamental questions that must be answered to set up a complete and coherent social protection system.

In truth, the notions of ownership and responsibility touted by the Paris Declaration have not yielded “true” appropriation, but have resulted in national domestic policy piloted by donor countries. The Paris Declaration does not afford truly autonomous action, without which there can be no responsibility on the part of recipient countries. With a knife to their throats, the recipient states must sometimes accept to see their public policies guided by the key indicators that are imposed, with the hope that they may regain some degree of control later. Furthermore, while donor countries have made significant progress in coordinating their action, much remains to be done by the recipient countries to improve their internal coordination (among different government ministries) and set their own priorities for discussion with donors. This is due to weak national leadership, and to the weak institutional capacity of national administrations that do not yet possess the means to implement results-based management.

There is abundant literature that shows how institutional variables affect the performance of public policy (Burki & Perry, 1998; Fukuyama, 2004; Persson & Tabellini, 1997). The quality or weakness of institutions are factors that explain time and time again the success or failure of many aspects of reform in states that receive development aid (Burnside & Dollar, 1997; Dollar & Easterly, 1999; Sindzingre, 1998). On this point, it can only be regretted that under structural adjustment programmes in the 1980s these countries were led to drastically “shrink” their rosters of civil servants (which were in fact quite meagre already), and to create numerous programme management units. This was not conducive to a stable institutional environment. In Mali, for instance, all administrative reform measures in this period were undertaken in the name of “management of structures and staffing”, a culture that continues to mark government administration in the country.

Likewise, and even if some donor countries have aligned practices with those of recipient countries, different types of aid may coexist, sometimes within the same programme, and with the same donor. This coexistence may impose considerable burdens on the national government in terms of record-keeping and reporting, taking time away from action to provide services to the people, which after all is the purpose of the aid. States deemed to be fragile, with little human resources, are forcibly given sophisticated tools that they are not well equipped to implement. In the same way, programme managers in aid recipient countries spend a good deal of their time collecting data on progress to attain “target values” for indicators, on reporting tasks and on discussion of these issues with representatives of donor countries. Clearly, this is time not devoted to work managing the programmes themselves, undermining the objectives of results-based management. In fact, performance-based management, invented in the private sector, is not readily transposed to the public sector, where it is less effective by reason of the very nature of public service. It would be pointless to try to measure with strict accuracy the cost of public action, precisely because these services are provided for all citizens.

In the Sahel countries, for example, government administration does not adequately cover the vast desert areas, and consequently the poor level of service to citizens (healthcare, education, safety and security of people and property, etc.) are factors of instability. In Mali, the sparse population in the north means low resources for local authorities as well as particular constraints in terms of economic, social and territorial organization (highly dispersed population, little infrastructure, open spaces, nomadic peoples). Low population density goes hand in hand with underprivileged areas that are poorly served in terms of basic amenities (water, schools, healthcare) and are remote and isolated. With little capacity to harness financial resources and hence low investment, low population and remote location, these territories are subject to additional costs that should not be borne by the local population alone. As equal access to public services is an imperative principle of republican government, it is not reasonable to apply across the board the same performance indicators–and most notably the ratio of public expenditure per capita–for public service throughout the country.

Referring to other contexts, Daniel Kaufman has given a very good description of the limits of management by indicators and this modelling of public action (Kaufman & Kraay, 2007). All this is to say that the real challenge facing developing countries, particularly in Africa, is to rebuild the capacities undermined by decades of structural adjustment and to regain a measure of negotiating power in their relations with international donors.

Giving Aid Recipient Countries Greater Autonomy to Conduct Their Public Policy

The dependence of developing countries with respect to donors is a notion that encompasses several different aspects (Naudet, 1999). In the most common sense, this dependence refers to the situation in which outside aid continues to weigh heavily in national budgets, while levels of development and internal financial capacity stagnate (see Azam et al., 1999). Dependence can also be apprehended through indicators, such as the share of aid in GDP, or in the conditions of aid allocation. Above all, dependence is characterized by little room to negotiate with donors. In this light the faltering capacity of recipient countries, the “fragile states”, takes on new meaning, particularly in Africa. Chronic political instability, military coups and armed rebellion are among the ills that plague African states, described as oligarchic, neo-patrimonial, fragile, failed, among the long string of adjectives that flourish in the “post-colonial” approaches that predominate in political science literature since the 1980s (on this issue see Haut Conseil de la Coopération International, 2005; World Bank, 2006; but also Badie, 1992; Devarajan et al., 1999).

With very few exceptions (Cabo Verde, Botswana, South Africa, Rwanda), African countries struggle to establish state structures that are able to conduct strong public policy on their own, and to meet the needs of citizens, while resisting the demands of international donors (on this issue see Joseph, 1999; Mhonimpa, 1994; Prest et al., 2005). On this continent more than elsewhere, the relations between donors and recipient countries are skewed, and even coercive, with stringent conditions and ever more exacting timetables for implementation. Indeed, while in the spirit of the Paris Declaration indicators are to be chosen from the grids in PRS documents or from the logical frameworks of sectoral programmes, in truth it is the weight of the funding institutions that is the determining factor, in particular for setting the target values to be attained. Furthermore, this is the justification for the fact that negotiations on the budgetary arrangements that frame budgetary support last for months and absorb all the energy of programme managers. These arrangements set, among others, the “trigger indicators” that authorize disbursement of variable tranches of aid. The choice of “target values”, i.e. the quantitative objectives to be reached each year as policy is implemented, are the focus of especially tough discussions. Be as it may, in the end it is always the donors who have their way, recalling the famous saying of the Malian philosopher Amadou Hampaté Bâ, who noted that “the hand that gives is always above the hand that receives”. This situation is illustrated by the words of a representative of donor countries:

The Malian government pilots the plane; we make sure there is a flight plan and that the altitude is properly measured; if the needle of the altimeter is not good, we will find better instruments for the pilot. (Bergamashi, 2009, p. 27)

Coordination between donors and the International Monetary Fund (IMF) imposes cross-linked conditions, increasing the pressure on the recipient countries. This type of condition–measures creating the obligation to link disbursement to progress markers set in advance under funding agreements–is in fact embedded in the mandate of international financial institutions, and is considered to be a sign of their credibility (Kanbur et al., 1999). It is hard to comply with these conditions, which explains the ambivalent attitude of recipient countries towards the reforms imposed by donors. Their formal acceptance often gives way to resistance in different forms when it comes to implementation (see Azam et al., 1999; Collier, 1997; Kahler, 1992). Faced with the difficulty of meeting some of the required conditions, recipient governments may turn to various subterfuges to postpone, circumvent or vitiate these rules. In some cases, conditions are suspended, knowing that from the outset recipient governments evaluate the costs of not complying with the conditions (see Bird, 1998; Bayart et al., 2002).

Paradoxically, the states with the least institutional capacity to manage the weighty apparatus of management by indicators are those on whom the donors impose the most stringent terms of conditionality, because they lack the capacity to negotiate. By setting ambitious goals the donors perhaps thought they would lift the recipients out of the danger zone; in fact, they have pushed them in deeper. In this respect, the annual joint review of aid by both donor and recipient countries are truly examinations, intended to assess the level attained for the indicators retained under the funding agreements. They are above all a humiliating experience for the recipient governments, when ordinary civil servants from aid agencies conduct themselves as judges, pronouncing sentences without appeal, setting the level of aid that will be disbursed, according to their appraisal of the results obtained.

It must also be said, however, that conditionality is sometimes exploited by the recipient governments themselves, for internal political purposes; donors can be held up as scapegoats, to justify unpopular (even if often essential) government decisions in the eyes of national public opinion (Collier, 1998; Rodrik, 1996). In reality, behind the pretext of granting greater autonomy to Third World countries to manage the aid they receive, the new approach based on budgetary support is a significant step towards a more political concept of aid, as part of a process to transform the political and social landscape of recipient countries, following pathways already defined by the donor countries. The donors’ injunctions are based on a transfer of models, replacing the sociological foundation of states, as if an idealized and disembodied institutional reorganization could be decreed, instead of being worked out by interaction between public authorities and civil society. This forcible transfer of model aggravates the deficit of legitimacy of the state already found in African countries, and further undermines civic and citizen responsibility, by discrediting government action and widening the gap between citizens and the state.

Each donor country has its own world vision, its own framework of reference, that is vaunted in the country’s relations with aid recipient countries. The desired transformations are imposed from the outside. This process can even go as far as direct supervision of entire sections of government, as in Sierra Leone, Iraq and Afghanistan. All this is to say that recipient countries should be able to construct their own public policies, in the context of their specific circumstances, without being forced to follow an outside model. Countries may be inspired by successful practices elsewhere, but freely and voluntarily, making the necessary adjustments. It is a matter of respecting the dignity of states, especially the most fragile among them. This movement must be advocated by democratic powers with real popular support and legitimacy from a fully involved civil society.

The fundamental issue is that donor countries must turn away from what RĂ©gis Debray, referring to Europe, has called a culture imbued with its own formulas when in dialogue with other countries (Debray, 2007). This is an enormous challenge, because it will oblige donors to change their intellectual processes and accept cultural diversity, something for which Western elites are not necessarily prepared. Experience teaches us that good governance is not achieved by decree. It is constructed by society in each country, in its specific circumstances. Political dialogue between donor and recipient countries naturally has a legitimate place, but the terms of this dialogue must be balanced. Each country should be allowed to pursue its own trajectory, according to its circumstances. The notions of the common good, the general interest and even of solidarity are not given the same meaning everywhere. Levy and Fukuyama have clearly shown how approaches diverge, depending on the situations found in each country (Levy & Fukuyama, 2008).

The concept of the effectiveness of public action is framed differently in the United States, in Europe and in Africa. Whereas Americans appraise the effectiveness of public policy through the lens of Schumpeterian “creative destruction” that enables a system to remain dynamic by eliminating its least effective agents, Europeans give preference to public funding based on a social contract. To promote economic and social development some Asian countries have emphasized the reinforcement of the state apparatus, while others, notably in Latin America, have preferred to bolster democracy in the political system.

Conclusion

The growing interest in using indicators to allocate development aid is closely connected to the needs of world capitalism. In this light, indicators are needed by investors, including international aid agencies that act as lenders, as a way to gather information on the quality of governance, and “country risk”. The underlying idea for investors is to be assured of “good governance” and of a functioning market economy that will guarantee returns on investment. It is therefore hardly surprising to note that indicators are biased, even if the bias is hidden. This makes use of indicators problematic. In the absence of a universal common reference framework, countries do not necessarily share the same vision of development.

Despite these imperfections, the concept of performance indicators cannot be totally rejected. The problems arise when they are mechanically applied to rank countries, and even to sanction countries depending on their rank. This practice is less and less acceptable, even in schools for evaluating pupils. An indicator should be just that, literally furnishing an indication of the quality of governance, but not in itself a full appraisal of quality. From this point of view, indicators can send a warning on the state of public policy, and foster discussion, without issuing a judgement with no appeal, and are much less a guilty sentence.

Management by performance must in no case be allowed to be a new ideology, or a tool in the hands of a few technocrats who decide what is good for entire nations and impose profound changes on society. This approach can be legitimate only as an inclusive process intended to support discussion and joint efforts to seek the best use of public resources, according to criteria of intelligence, balance and fairness. Elected officials and citizens, through their organizations and associations, must take their full place in this process. To allow people to appropriate this discussion they must be properly informed as to the stakes and outcomes, with access to forums where views can be freely expressed, and the full range of opinion heard, in particular those voiced by the weakest groups in society.

Notes

  1. 1.

    Project aid, whether managed by project unit or through institutional support more or less directly administered by funding bodies, is criticized for low disbursement rates, high overhead costs and weak appropriation by recipient countries. For these reasons a number of donors, including the European Union, are turning more and more often to budget support, i.e. aid paid directly into national budgets to fund all activities in the budget without specification (general budget support) or to support certain targeted sectors (sectoral budget support), most often public health and education (Feyzioglu et al., 1996).

  2. 2.

    On issues related to budget support see the study carried out by researchers at the University of Birmingham for the OECD Development Assistance Committee (DAC) and other donors: IDD and Associates, Evaluation of General Budget Support: Synthesis Report. A joint Evaluation of General Budget Support 1994–2004, May 2006 (IDD & Associates, 2006).

  3. 3.

    On the choice of indicators for budget support, see: Introduction d’indicateurs de Résultat en Matière d’Appui aux Programmes d’Ajustement Structurel dans les Pays ACP, Etude réalisée à la demande de la Commission Européenne, Rapport de Synthèse, Vol.1, CERDI, June 2002.

  4. 4.

    For a discussion of teachers’ capacities in Mali see: La formation des enseignants dans la francophonie: diversités, défis et stratégies d’action (2007). Montreal: AUF. Profil de l’enseignant de qualité au Mali. Oxfam, IE, OPT, SNEC, MEALN, undated. Principes directeurs pour un profil de compétences national des enseignants du primaire. Internationale de l’Education, Oxfam Novib, 2012.

  5. 5.

    In principle, teachers and public health personnel were not eligible for early retirement, so as to preserve these social services, but in fact they were the most severely affected, because workers in these sectors suffered from more difficult conditions than civil servants in other sectors.