Human Capital or Human Development? Search for a Knowledge Paradigm for Education and Development
Santosh Mehrotra

This paper compares human capital theory with the capability approach and lays out the problems with the theory. As a knowledge paradigm for education and development, it finds the theory wanting. However, it has remained the foundation for sectoral work in education and health by international financial institutions. The paper spells out the problems, historically, with World Bank lending in the education sector, some of which follow from human capital theory, while others follow from a broader neoliberal agenda. It concludes by delineating the foundational elements of an alternative knowledge paradigm for 'education for all', based on the capability approach and its extension.

This paper is about knowledge paradigms and development, but it is also about institutions which espouse certain kinds of knowledge about world development, and especially education – and how those paradigms affect the lives of people in the hands of powerful institutions. One should not underestimate the power and influence of knowledge paradigms – or as John Maynard Keynes (1947) said, “The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood… Practical men, who believe themselves to be exempt from any intellectual influences, are usually slaves of some defunct economist.”¹

Section I of the paper compares human capital (HC) theory and the notion of human development (HD) underlying it, which is the human capability approach. Section II lays out the problems with HC theory. Section III addresses some of the education sector policies that have derived, at least partly, from HC theory, and where they are not those that are part of a larger neoliberal policy framework. The final Section IV spells out what may be foundational elements of a new knowledge paradigm at least for ‘education for all’ (EFA), if not for social development. It confines itself to issues of education, although the implications can be easily generalised to the provision of other basic services.

Theodore Schultz’s (1963) The Economic Value of Education – with its exposition of the concept of human capital and programme of evaluating its returns – focused the attention of economists on human capital. The notion acquired greater salience in the late 1980s to early 1990s of the 20th century and there was a shift in contemporary economic analysis, from seeing capital accumulation in largely physical terms to viewing it as a process in which the productive quality of human beings is integrally involved. The new growth theories emphasised that through education, learning, and skill formation, people can become much more productive, and this contributes significantly to the process of economic growth [Barro and Sala-I-Martin 1995; Barro 1996]. Studies of economic growth re-examined the experiences of Japan and newly industrialising countries of east Asia (and Europe and north America) and emphasised the role of ‘human capital’ to a greater extent than had been the case earlier [Amsden 1989, Wade 1990, World Bank 1993]. This new emphasis on human capital was reinforced by education sector analytical work of some international financial institutions (IFIs) – these were believers in human capital theory [World Bank 1995]

At the beginning of the 1990s there was another major event in development theory. The work of Amartya Sen and others on human capability [Sen 1985, 2000], over two decades, resulted in the gradual emergence of a human development paradigm – partly manifested in the Human Development Reports (of UNDP), and the much used composite index of ranking countries, the human development index (HDI). In fact, ‘human development’ became an overly popular term after 1990 – popular even with IFIs, who reorganised departments and called some of them human development networks. However, the consensus within which the term ‘human development’ was used, was neoliberal, and Washington-based. For the adherents of the consensus, the term human development was used, but the theory underlying it was human capital theory, not the capability approach, which gave birth to the term. The critique of economic growth as a measure of development success predates the emergence of the HD paradigm by at least three decades; the IFIs did not subscribe to that critique. Yet the term human development has been appropriated by all and sundry.

HD is a much richer notion than human capital. The human capability approach has intrinsic value for the well-being of people; an indirect role in influencing social change; and an indirect role in influencing economic output [Sen 2000]. Besides, the notion of human capital can barely be a subset of the notion of HD, since HC theory sits firmly within neoclassical economics – which HD certainly does not.²  However, human development as a paradigm offers more than HC theory; it is only within the former, buttressed by the capability approach that it is possible to integrate economic and social policy – which should be the ultimate goal of good governments. We return to this issue later.

Not only does human capital come out looking like a narrower notion than human development, there are several problems with HC theory on its own terms, and with the IFIs’ approach to education, The latter is heavily dependent for its justification upon the application of human capital theory. These problems are summarised here. First, HC theory has been used in education to generate rates of return and cost-benefit analyses, which have been subjected to severe criticism by many [Lauglo 1996; Bennell 1996; Samoff 1996]. Cost-benefit analysis in education grew out of the World Bank’s (WB) devotion to human capital theory. It is interesting that while cost-benefit analysis as applied to education has been popularised by the World Bank since the early 1980s (thanks primarily to the work of Psacharapolous), in the early 1960s (when WB lending to education began) it was frowned upon by its economics department: “…Its (CBA) operational value is very low because it fails to inform us about the rate of return on future investments; in the absence of marginal ROR calculations we still do now know how much to invest in one direction before the ROR on investments at the next lower place in the scale” [cited in Jones 1992].

Second, HC theory treats, schooling as a ‘black box’, a technical relationship between inputs and outputs. In this respect, it is similar to the neoclassical theory of production with its idea of fixed technical relationships between inputs and outputs, as expressed in the form of a production function. This completely ignores the fact that the whole structure of education – from pre-school, elementary and secondary school, vocational and then schools of higher education – is a system; it is a system of power and ideology, which is itself deeply rooted in society’s cultural norms [Fine et al 2001]. Anyone who doubts this should only look at the deep segmentation of say, on the one hand, the Indian school system, with its missionary-run, English-medium private schools producing ‘brown sahibs’ who effectively run the country, and on the other, the majority of rural government schools with one or two teachers teaching five grades in the vernacular language or as many as 150 children, perhaps a third of them from the scheduled castes or tribes. Or contrast that to another system: elementary schools in Thailand, where a more egalitarian system, largely lacking a language barrier, ensures schooling for all, and where the private elementary school is resorted to only by children unable to survive the government school system. There could not be a better example of schooling being a system of power and ideology than that of Francophone Africa. For over three decades after independence, French has remained the language of instruction in the earliest grades of primary school (and for higher grades) even in the remotest parts of rural Sahel – despite the overwhelming evidence from pedagogical theory that children learn best in the earliest years in their mother tongue. The alien school system was one factor underlying French west Africa’s very low primary enrolment rates, which remain till date the lowest for any region of the world.

Third, not only is production of schooling treated like a technical relationship to produce human capital, the use of human capital as an input into production is also unsatisfactory in HC theory. In the theory, human capital is said to affect economic outcomes, in particular wage outcomes. However, other factors affect wage outcomes too: gender, race, capital intensity, degree of monopoly, all of which mutually condition one another. But HC theory assumes that the labour market is a perfectly competitive one with almost no interaction between these other labour market variables. Thus, the impact of human capital on wages and growth assumes that the economy is at full employment and perfectly competitive. Otherwise, estimates of wage outcomes will be conflating the effects of human capital with those of excess capacity and price distortions.

Despite these and other criticisms, the World Bank’s economists do not see HC theory as controversial [Jones 1997]. The WB’s own rationale for education lending continues to be based, for over three decades, on HC theory. Even as a considerable shift has occurred at least among those who believed in the Washington consensus to what is now termed a post-Washington consensus, HC theory remains entrenched. Stiglitz’s (1998) critique of the Washington consensus did not amount to a rejection of human capital theory. The justification for a focus on education and the rationale for government financing (if not provision) remains based, in Stiglitz’s recent writings, on the concept of human capital and its relationship with growth.

At best, the IFIs’ use of human development as an objective provided the rationale for one of the prongs of the pro-poor growth strategy in the World Development Report (1990 and 2000) on poverty, and the consequential increase in population, health, nutrition, and education lending. However for multilateral financial institutions HD remained a term to be used, while development thinking within these institutions remained within the confines of human capital theory.³  Nevertheless, given this paradigmatic melange, one can well question whether the IFIs can be brokers for what is and is not knowledge for development.

Turning to education, the questions that have dominated the 1990s – on account of the brokerage of IFIs – usually tended to be the wrong ones (even though they were about relevant themes), or theoretical underpinnings and resulting policy implications have sent mixed signals to governments. Even neoliberals recognise the existence of market failure, and state involvement in education provision and financing has been seen as a second best solution. The degree to which the state should subsidise education has, nevertheless, been subject to contending opinions. This resulted in conflicting messages throughout the 1980s from IFIs about the state and education. On the one hand, education was a priority for development. On the other, structural adjustment loans required cuts in government spending and greater market reliance, including privatisation. It is not that these loans required education and health expenditure reductions, but these sectors suffered a cut regardless.⁴  Thus in India, in seven out of eight states examined, education expenditures showed a decline in the 1990s (after the onset of structural adjustment in 1991) which was statistically significant compared to the preceding 15-year period [Mehrotra et al, 2005].⁵

Another example of mixed signals emerging from IFIs in respect of education (and also the health sector) was in respect of user fees. Along with concern about public sector spending in the 1980s, WB economists developed an economic model illustrating that, where there is excess demand for education, charging user fees at all levels of education would be advantageous from both an efficiency and equity perspective [Thobani 1983, Tan et al 1984; World Bank 1986; Jimenez 1986, 1987; Psacharapolous 1986]. In fact, World Bank (1986) strongly advocated user charges in Malawi at the secondary level, arguing that with user charges education can expand with no loss of equity. Remarkably, in 1994, when Malawi eliminated fees and uniforms, enrolments increased dramatically throughout the school system [Unicef 1998].

The World Bank’s policy note (1986) argued strongly that governments ‘do not tap the willingness of households to contribute resources directly to education’. This policy was most convincing (and actually persuasive in Sub-Saharan Africa through the instrument of IMF stabilisation and WB adjustment loans) for higher education. However, after sharp criticism of the consequences for elementary schooling, by the early to mid-1990s, the WB distanced itself from user charges at the primary level [Tilak 1997], but not before doing some damage.

Similarly, in order to ensure better learning outcomes, the WB’s knowledge dictated a predominant role for textbooks in the policy package from the mid-1970s on. Not that the emphasis on textbooks per se was wrong, but the emphasis was misplaced relative to the neglect of the role of the teacher. For much of the 1960s – when WB lending to education began – to the early 1970s, lending to education was a brick-and-mortar affair, with most lending going to building school infrastructure. In fact, most of the staff in the WB’s education department were architects [Jones 1992]. Bank loans mainly provide the foreign exchange cost of projects, and hence were limited in their applications early on, as foreign exchange costs of primary education projects are normally low. Once construction activities in borrowers declined, and WB developed a greater interest in schoolwide quality, it was a challenge to find new areas within education to loan monies to. One aspect of pedagogy – textbooks – helped serve the need for new, tangible areas of lending from the mid-1970s. Such bankable avenues of loaning money and the WB’s emphasis on them can mask dangers to borrowers who may mistakenly look to the WB for a comprehensive and balanced view of educational development. In fact, one result was that in many countries the issue of teacher remuneration, development and incentives was ignored – with serious adverse consequences for school effectiveness. Teacher salaries had been systematically declining in many Sub-Saharan and Latin American countries through the 1980s and the much of the 1990s, adversely affecting teacher motivation.⁶  In the late 1990s, this imbalance in emphasis in WB sectoral analysis was corrected but too late in the day.

On the role of the private sector in education, the research question regarding the relative advantages and disadvantages of private versus public schooling was often framed in terms of: do children in private schools have a current learning or future wage advantage than those in public schools? However, the overwhelming historical evidence is that neither in industrialised countries nor in contemporary high-achieving developing ones did the private sector play a predominant role in universalising schooling. In fact World Bank (1986) had argued strongly for encouraging the expansion of non-government schools, suggesting, “in (government) schools resources are not being used as efficiently as they might be, (a) problem reinforced by the lack of competition between schools” [cited in Jones 1992]. The WB’s Education Sector Strategy 1999 noted that one of the drivers of change in the education sector was that public/private roles were changing, the state was no longer the only provider, and that increasingly there were in education a spectrum of providers – private for-profit, private not-for-profit, and community-level organisations.

Since 2001, however, there is a new emphasis on supporting private sector growth in education. Two major documents emerged that will drive this agenda. First the Private Sector Development Strategy – Directions for the World Bank Group [World Bank 2001] and Investing in Private Education: IFC’s Strategic Directions [IFC 2001]. The WB strategy proposed the following measures to extend the reach of markets: (a) conducting systematic surveys of the investment climates for the private sector, including comparisons of countries in respect of the climate; (b) continued direct support to firms, especially small and medium enterprises. The areas of focus for extending the reach of markets were infrastructure and social sectors (health, education, water and sanitation); output-based aid was identified as a cross-cutting objective. For the IFC, five sectors of emphasis were identified: domestic financial institutions, infrastructure, information technology and communications, small and medium enterprises, and the social sectors (health and education). This emphasis on the social sector as a ‘frontier’ area was adopted despite the admission: “there are as yet no evaluation results from Bank projects supporting private participation in the social sectors” [World Bank 2001: 44].

For the IFC, in education the higher investment probability areas are: tertiary education, technical and vocational training, technology-based education and distance education; and student financing through loan schemes to middle and lower-income families. The IFC strategy notes that early childhood schooling and primary and secondary education projects usually call for only smaller scale investments, which limits IFC investment until it develops better mechanisms to deal with small projects. While IFC investments are likely to be concentrated in middle income countries (as they have been so far in Latin America), it is likely that the World Bank group as a whole, and IDA in particular, will work in a coordinated manner to encourage the private sector at all levels of education.

The World Bank’s investment climate surveys (of which 20 are planned per year, starting 2003 over the next five years) are to be integrated into products such as the Country Economic Memorandum and to inform the development of Country Assistance Strategies (CAS). CAS and the Poverty Reduction Strategy Papers (PRSPs) will be the main vehicles by which investment climate assessments will be integrated with lending projects and technical assistance operations, including in health and education. The implementation of the PSD strategy will be coordinated between IBRD, IDA and IFC. Further, some donors indicated that such surveys will be a ‘useful basis’ for their own programme decisions and they would participate in a joint mechanism to develop and finance them. Such coordinated action may well influence the course of decision-making in the future in both low and middle income countries.

Output based aid, another new area of emphasis, is offered as “a strategy for improving the delivery of services that depend at least in part on public financing”. In contrast to traditional approaches in channelling support to inputs consumed by public sector providers, output-based aid intends to delegate service delivery to third parties under contracts that tie payment to the results or outputs actually delivered. It may be unobjectionable that output based approaches intend to improve efficiency and accountability for service provision by shifting more responsibilities to the service provider. However, the point is that “what output-based approaches offer beyond other efforts that focus on results, is the use of contractual discipline and the scope for competition for service provision” [World Bank 2001: 59]. In fact the approach is intended to shift service delivery to the private sector, while shifting performance risk to it as well.

As noted earlier, the contemporary promotion of the private sector in education ignores the historical experience of Europe and north America. In the 19th century, in the now industrialised countries, nation states that were leaders in mass schooling – Prussia and northern US – the state played a pre-eminent role from the beginning. In fact, where the state did not play a pre-eminent role – in the southern cotton slave belt in US, in France until 1870,⁷  and the UK till later – mass schooling lagged behind the other major states.

There were a number of measures that were common to all states as the 19th century progressed. First, the state created a public system of schooling that was financed from taxes. The rising tide of public investment is evident for all.⁸ Second, gradually the state worked towards the elimination of school fees. Third, the state made elementary education compulsory. Finally, varying degrees of centralisation emerged, with bureaucracies to run the public school system, including the training of teachers and creation of an education department in government [West 1965; Stephens 1998; Green 1990; Maddison, 1987]. Countries which developed state funded systems of public education early (e g Prussia, northern US, Switzerland, Holland) achieved higher levels of enrolment and literacy. Countries (e g, England and Wales) with no public system until quite late fared worse in terms of those indicators [Lindert 2000].

Similarly, we have argued elsewhere [Mehrotra and Jolly 1997; Mehrotra 1998] that in contemporary high-achieving developing countries the private sector played a minor role in universalising elementary schooling. Thus in some east Asian countries there is strong evidence that while all primary school-age children were in public schools, most secondary and tertiary level students went to private schools. In Korea 99.5 per cent of primary school children in 1965 were enrolled in publicly funded schools, while only 55 per cent of lower secondary and 42 per cent of higher secondary students were in public schools; the share was even lower (27 per cent) for universities [Mason et al 1980].

In fact, in India the private sector plays not a complementary role, but a substitutive role – with serious adverse equity and efficiency effects; such questions are never raised in the sectoral analytical work by IFIs. An analysis of the National Sample Survey of India (1995-96) data on enrolment by type of school management – government, government-aided private, and private unaided – shows that private aided schools’ share in enrolment tends to rise with the level of education from primary to upper primary to secondary. As a result of political lobbying, private school managers/teachers convince governments to provide subsidies to these schools, without any performance guarantees, with two results. One is that contrary to the principle that a fiscally squeezed state should be targeting its subsidies to the poor, the state actually stops cost recovery from a section of the population which is able to pay, and subsidises them. This happens because it is mainly the non-poor who are able to make their way past the elementary school barrier (8th grade), and the private government-aided schools predominate at the secondary level. A second result is that, after conversion to being aided schools, teachers in private aided schools begin to be paid higher government school salaries; a significant proportion of government expenditure at the secondary level is devoted to this kind of subsidy of the non-poor, which leaves fewer public resources for government elementary schools.⁹ Contrary to this historical evidence, World Bank education loans have required the promotion of the privatisation of education.

Given these many concerns, can, or rather, should IFIs be brokers for what is and is not knowledge for education and development? There may possibly be practical reasons for accepting IFIs as such brokers. The IFIs are big lenders, the biggest of any provider of external finance for education; they have practical experience spanning all continents. But the problem with IFIs’ knowledge is that it is usually ahistorical. The IFIs have shown a singular incapacity to undertake research of a historical nature – social or economic history – which might enable policy lessons to be drawn for contemporary development problems.¹⁰ Yet, the most powerful arguments come from historical analyses, not econometric analyses of cross-country data for one point of time.

We know that education for all cannot be achieved unless the level, efficiency and equity of government education spending improves. However, rarely is the case for such spending made by examining the historical pattern of spending in the now industrialised countries in the 19th century

An examination of that spending pattern would reveal what governments considered important, and with what outcomes. Table 1 presents the share of government spending circa 1900 in the now industrialised countries. By 1900, most of North American and European countries, as well as Australia and New Zealand had significantly expanded elementary enrolment: per 100 children of age 5-14 year, in Canada primary and secondary enrolment was 90; in New Zealand, 77; in US, 88; in Australia, 87 in elementary and secondary; in Germany, 76; and in UK, 72. However, despite the high levels of enrolment every country was still allocating 90 per cent of education spending to elementary and secondary education.
 
 

Table 1: Now-industrialised Countries Share of Public Education Expenditures by Level (Circa 1900)
Country	Elementary	elementary and Secondary
Australia	-	95.9
Belgium	94.2	96.7
Canada	-	94.2
France	72.5	90.3
Germany	77.8	88.7
Japan	85.2	88.7
Netherlands	77.1	92.5
New Zealand	-	95.3
Norway	83.7	90.7
UK	80.4	-
US	-	93.6
Source: Lindert 2004, Volume 2, Appendix C.

 

In fact, even more striking is the fact that in industrialised countries, over three-fourth of public education spending was going to the elementary level, which is higher than India’s allocation to the same (grades 1-8) over the entire 50 years since independence, and in most developing countries spending for primary education (usually grades 1-5).¹¹

All this raises serious doubts about whose knowledge will count when it comes to the sector-wide approach, the new form of programme aid. Yet, the heavily indebted poor countries (HIPCs) – mostly the least developed ones – and other low-income states (eligible for the IMF’s Poverty Reduction Growth Facility, or the erstwhile Extended Structural Adjustment Facility) have to undergo the PRSP process. And the content of macroeconomic policy in the PRSPs as well as IMF lending has not changed much from over the 1990s. If anything, the Independent Evaluation Office (IEO) of the IMF has said even in 2004 that little had changed from the past. Given that the Bretton Woods Institutions (BWI) strategy has failed in terms of its own criteria of promoting growth and thus reducing poverty (except in China and India), it might be worth taking seriously the IEO’s suggestion to expand the space for local decisions in macroeconomic policy-making. First, it says the IMF should let, “authorities have the initial responsibility for proposing a reform programme, which should be the starting point for negotiations”. Second, the core elements of the programme should be “subject first to a domestic policy debate within the member country’s own policy-making institutions”. And third, Article IV surveillance reports should “actively seek to present alternative policy options and to analyse the trade-off between them so as to encourage an open debate on alternative policy options” [IMF 2004].

However, until any of this happens, it is clear that for most countries who come to the IMF for funds, it is a non-question as to whose knowledge counts. IFIs are the largest providers of funds for education. Bilaterals do not have sector expertise to the same extent, and hence must rely on the IFIs. However, all the discussion of ‘ownership’ notwithstanding, these international agencies have limited local knowledge, and are pumping significant funds into education, often creating imbalances in the policy package, as we saw earlier.

To return this discussion to knowledge paradigms, and their importance for development, the capability approach holds out much greater promise than human capital theory, both from a normative and a policy perspective. First, in HC theory the building of human capital is only a means to an end – the expansion of production and of economic growth. But does it tell us why increased production is good thing, why it should be sought after? As Sen says, if economic growth expands the, “human freedom to live the kind of lives that people have reason to value, then the role of economic growth in expanding these opportunities has to be integrated into that more foundational understanding of the process of development as the expansion of human capability to lead more worthwhile and more free lives” [Sen 2000].

Second, beyond this abstract construction, there is a more practical, policy concern that makes the capability approach a more appropriate paradigm for development. In most governments there is a hierarchy of policy – with macroeconomic policy being determined first and social policy in the role of follower. This hierarchy of policy is reflected in the hierarchy of government service, where a career in the ministry of finance is typically preferred to one in a social affairs/education/health ministry. Atkinson (1999) notes that this lead/follower model applies not just to national governments but also to international agencies. “It makes no sense,” he remarks, “for macroeconomic stabilisation policies to be determined by Bretton Woods institutions, and then for the social consequences to be addressed by other agencies”. This has been partly recognised by entry of IFIs into social fields, but that has its own problems – some of which we identified earlier.

The separation of the ‘economic’ from the ‘social’ discourse is similarly inherent in the Washington consensus (notwithstanding the Comprehensive Development Framework). The emphasis of the consensus, maintained in the post-Washington consensus, on increasing income as a means of reducing poverty is partly inherent in the definition of inadequate income as a strong predisposing condition for an impoverished life. This approach has shortcomings: it is inefficient in achieving macroeconomic objectives, as it underplays human welfare outcomes, often unintended, or inadequately anticipated, on account of initial conceptual errors in the model used for prediction of outcomes or on account of imperfect information during formulation. In such circumstances, social funds and education and health ministries are left to take care of the consequences of macroeconomic policy mistakes – essentially, to pick up the pieces. The separation of economic and social discourse in governments (or in international agencies) has another consequence: where economic and social policies are in conflict, they can only be resolved by assessing their impact on human welfare, not and on economic growth rates. The setting of macroeconomic goals as the dominant ones does not allow such a direct assessment. The recognition of the goal of expanding human capabilities, on the other hand, allows this separation of economic and social goals to disappear, as human capabilities are intrinsically important to the well-being of people, but they also encourage social change to occur (e g, female education will reduce gender inequality, fertility rates, and child mortality rates, improve the quality of public debates), quite apart from increasing human productivity in an economic sense.

What is remarkable is that the disjunction between the economic and social was not characteristic of policy-making in industrialised countries in the heydays of the welfare state. In most industrialised countries, the post-1945 expansion of the welfare state contributed to the macroeconomic objective of avoiding the mass unemployment of the 1930s. The disjunction is a more recent phenomenon, characteristic of the revival of neoliberal orthodoxy in both north America as well as Europe.¹²

Assuming that the human development paradigm has many advantages over the alternative, the question is: where do we go from here? The question may well be raised whether the capability approach offers an operational framework beyond its conceptual advantages. The most important question in this context is: does the capability approach offer insights for ensuring that all children out of school access school, complete at least an elementary education, and reach minimum levels of learning? The short answer is yes, but it is incomplete.

The capability approach is essentially an evaluative one, and thus can be and has been used for normative purposes.¹³  However, an essential element in making the approach operational, in the sense of being helpful to the policy-maker, is to define the conditions that would lead to the realisation of simple functioning (being able to read and write, being adequately nourished and free of avoidable disease) and complex ones (like being able to take part in the life of the community and having self-respect) – which is the stuff of human capabilities.

In our elaboration on the capability approach we have argued [Mehrotra and Delamonica forthcoming) for three extensions in order to make it operational. First, Sen’s distinction between simple and complex functioning is too watertight; in real life, there is mutual interdependence between them that Sen does not recognise. Second, Sen’s formulation of the capability approach focuses exclusively on the individual, ignoring the collective, i e, the voice of an organised community of people. Third, Sen’s articulation of democracy as a desirable condition (made famous in the contrast with China, where the absence of democracy made famines possible, unlike in India) for enhancing human capabilities is mistakenly conceived of only at the national level, when what matters most for genuine participation is local participation, only realised through deep democratic decentralisation.¹⁴  To bring all three points together, we suggest that the complex functioning of participation the approach postulates needs to be contextualised at the level of the community – collective voice and collective action – to have operational use. In fact, the efficiency of service delivery by state functionaries would be greatly enhanced if collective voice could be enabled, and accountability to the community also improved. Unless the complex functioning of participation is perceived in this way, none of the simple functioning (e g, the ability to read and write) are likely to be realised, even in democratic states.

It is not democracy at the national level – with a once in five-year vote – that is necessary to enable simple functioning to be realised; what is needed is deep democratic decentralisation. Public actions relating to valued, but elementary, functioning may be difficult to achieve without the more complex functioning of participation in the life of the community. The capability approach is focused on the individual’s capabilities and functioning. However, poor individuals are powerless to realise that functioning, even if a distant central (or state government in a federal country) government was willing and able to finance/provide services (through the government functionaries) – which are their basis. Participation by the poor has to be at the collective level. Without collective action, the voices of the poor can rarely be heard; without the state ‘enabling’ collective action, which emerges as a counterweight to intermediaries, the delivery of services, and hence the functioning cannot be realised. This requires democratic decentralisation. Without voice, the supply of services will remain of poor quality and thus ineffective.

History is in favour of such local democracy and such decentralisation. Thus Lindert (2000) argues that before 1880 there was so little social spending of any kind (in industrialised countries, i e, tax-based public spending on health care, low-income housing subsidies, pensions, welfare and unemployment compensation) mainly because political voice was so restricted. Voting rights were restricted by law, limiting the franchise to those who owned some land, earned some minimum income, or paid some minimum value of direct taxes. As voting rights became less restricted, a shift occurred towards progressive taxation, enabling social spending to grow. Education spending by the state was the main form that social spending took prior to 1880 in the now industrialised countries. However, there was a clear difference between countries. Where democracy and decentralisation was more advanced, public action by the state in the sphere of elementary education was most advanced.

In fact, the interests of those with voice explains why Prussia and laissez-faire North America pioneered public schools financed from taxes and why Britain fell behind (despite the latter being regarded as the workshop of the world). Lindert (2000) suggests two reasons, which derive from the political economy of these regions. First, there was a systematic influence of the spread of voting rights upon primary school enrolments (as on social transfer spending as a share of GDP). Countries where a majority of adults (North America, Australia, France) voted had many more children in school than either non-democracies or countries where only a wealthy minority could vote (e g, Britain, southern USA).

A second cause was decentralisation. Prussia/Germany and Northern America left the decision of how much tax to pay for schools up to the localities. Even though the German national government was undemocratic, it allowed local governments to raise taxes and locally elected and appointed officials ran the schools. This was unlike Britain, France, and other industrialising countries, where local governments had much less power and where education correspondingly lagged behind.

Decision-making informed by local knowledge, collective voice and collective action – in sync with central government action – can trigger the transformation necessary to achieve education for all. The trick is to combine the language of the Millennium Development Goals (or the EFA targets) – perfectly laudable goals provided they come with additional external resources – with local knowledge and community action to pressure government functionaries to deliver basic services effectively. Examples are growing of successful local contributions to the policy process (from India and elsewhere).¹⁵ The history of industrialised countries is in favour of such democracy and decentralisation. Since the poor cannot ‘exit’ in the Hirschman sense, ‘voice’ alone works – but it must be a collective voice; hierarchical control through the government bureaucracy will not suffice nor will the stick of social conditionalities from the BWIs, combined with the carrot of external assistance. There lies the possibility that human development, conceived as both means and end, and underlying capability theory, is established as the knowledge paradigm for development.

To sum up, it has been argued here that for education for all to be realised, the knowledge paradigm needs to be human development and capability theory, not human capital theory. Many examples were noted why imported wisdom needed to be treated with caution – partly because of the kinship of some of this wisdom to HC theory, and partly because much of it suffers from multiple biases: ahistoricism, empiricism and a bias in favour of market-driven solutions, if not biases driven entirely by the need for a bank to keep lending.

There are at least four foundational elements of a knowledge paradigm for EFA: local knowledge to inform policy decisions specific to the context; historical knowledge; institutions enabling deep democratic decentralisation, and institutions that enable the articulation of collective voice and collective action.

Address for correspondence:
santosh.mehrotra@undp.org

[Thanks are due to Chris Colclough and Enrique Delamonica for comments on the paper. Views expressed here are those of the author, and should not be associated with that of any organisation.]

 1 Keynes continues: “Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back” [Keynes 1947, Ch 24].
 2 The widespread acceptance of HC theory is on account of the increasing conservatism of the economics profession and uncritical use of the notion of HC by other social scientists. See Fine et al (2001) for this view. Non-economists have been struggling to be taken seriously by economists for long, and have accepted the term human capital, just as they approve of the use of the term ‘social capital’ by economists – regardless of the weaknesses of the latter notion and the theoretical underpinnings of the former.
 3 It is possible that within the World Bank the multidimensionality of the concept of HD was the basis of its attractiveness, compared to the unidimensionality of the concept of income-poverty.
 4 The evidence is ample in WB’s own publications. See, for instance, Jayarajah et al (1996).
 5 Ravi Srivastava argues in the chapter analysing public spending in eight states – Andhra Pradesh, Assam, Bihar, Madhya Pradesh, Rajasthan, Uttar Pradesh, West Bengal, and the relative high-achiever, Tamil Nadu – that except in Rajasthan, the growth rates of public spending on elementary education declined after the reforms began in 1991.
 6 See Mehrotra and Buckland (2001) for an analysis.
 7 The defeat of France at the hands of Prussia in 1870 led to a sudden stirring of efforts by the French state in favour of education, and a sharp increase in public education expenditure [Prost 1981]
 8 In 1860 Germany (then Prussia) was spending 1 per cent of NNP, a share that nearly doubled by the end of the century. The share of French government expenditure allocated to education and science increased sharply between 1872 and 1880, and then nearly doubled between 1880 and 1890. In the UK, the laggard, it increased from 0.9 per cent in 1880 to 1.3 per cent in 1890 of GNP; prior to 1880 the state had mainly subsidised private schools with grants, but after that the state school system expanded rapidly [data drawn from Fishlow 1966; Sanderson 1983; West 1994].
 9 For a further analysis, see Mehrotra et al, 2005. For similar conclusions, see Kingdon (1994) and Tilak (2000).
10 An exception is the World Bank History project; however, the project’s output is concentrated on the WB’s own history.
11 In India, the share of education spending allocated to higher education over the first 40 years since planned development began (1951-90) was much higher than in industrialised countries circa 1900, and even in the last 10 years has continued to be higher than that in industrialised countries around 1900. For a further analysis, see Mehrotra and Delomonica (forthcoming).
12 The emergence of this orthodoxy was coterminus with the electoral victories of Ronald Reagan, Margaret Thatcher and Helmut Kohl at the beginning of the 1980s.
13 If specific functionings (or their elements and concomitant indicators) are defined, then it is possible to measure the distribution of that functioning (or elements) in the population. For instance, Brandolini and D’Alessio (1998) use such components like education and skills (with their typical indicator years of education, level of education reached), health and access to health care (indicator: contacts with doctors and nurses), and so on. These are used in the Swedish Level of Living Surveys.
14 For an extended discussion of the successes of deep democratic decentralisation, see Tendler (1997) and Mehrotra (2002).
15 See Mehrotra (2002) for an analysis of examples of the effectiveness of such deep democratic decentralisation.

Amsden, A (1989): Asia’s Next Giant: South Korea and Late Industrialisation, Oxford University Press, New York.
Atkinson, A B (1999): ‘Macroeconomics and the Social Dimension’, Experts Discuss Some Critical Social Development Issues, ESA/DSPD/bp2, Division for Social Policy and Development, Department of Economic and Social Affairs, United Nations, New York, May.
Barro, R J and Sala-i-Martin, X (1995): Economic Growth, McGraw-Hill, New York.
Barro, R J (1996): Getting It Right: Markets and Choices in a Free Society, MIT Press, Cambridge.
Bennell, P (1996): ‘Using and Abusing Rates of Return: A Critique of the World Bank’s 1995 Education Sector Review’, International Journal of Educational Development, Vol 16, No 3, pp 235-48.
Brandolini, A and d’Alessio, G (1998): ‘Measuring Well-Being in the Functioning Space’, mimeo, Bank of Italy.
Fine, B, Lapavitsas, C, and Pincus, J (eds) (2001): Development Policy in the Twenty-first Century. Beyond the Post-Washington Consensus, Routledge, London.
Fishlon, A (1996): ‘Inequality, Poverty and Growth: Where Do We Stand?’ in Bruno, M and Pleskovic, B (eds), Annual World Bank Conference on Development Economics, Washington, DC.
Green, A (1990): Education and State Formation: The Rise of Education Systems in England, France and in the USA, St Martin’s Press, New York.
IMF (2004): Evaluation of PRSPs and PRGFs, IMF, Independent Evaluation Office, Washington DC.
IFC (2001): Investing in Private Education: IFC’s Strategic Directions, International Finance Corporation, Washington DC.
Jayarajah, C, W Branson, and B Sen (1996): ‘Social Dimensions of Adjustment: World Bank Experience, 1980-1993’, World Bank Operations Evaluations Study, World Bank, Washington DC.
Jimenez, E (1986): ‘The Public Subsidisation of Education and Health in Developing Countries: A Review of Equity and Efficiency’, World Bank Research Observer, Vol 1, No 1, pp 110-129.
– (1987): Pricing Policy in the Social Sectors: Cost Recovery for Education and Health in Developing Countries,
Jones, P (1992): World Bank Financing of Education: Lending, Learning and Development, Routledge, London.
– (1997): ‘World Bank (1995) Policies and Strategies for Education: A World Bank Review’ in Comparative Education, Vol 33, No 1, 1997, pp 117-129.
Keynes, J M (1947): The General Theory of Employment, Interest and Money, Prometheus Books, UK.
Kingdon, G G (1994): An Economic Evaluation of School Management Types in Urban India – A Case Study of Uttar Pradesh, DPhil Thesis, University of Oxford, Oxford.
Lauglo, J (1996): ‘Banking on Education and the Uses of Research: A Critique of World Bank Priorities and Strategies for Education’, International Journal of Educational Development, 16, pp 221-233.
Lindert, P H (2000): ‘What Drives Social Spending, 1980-2020’, mimeo, University of California, Davis.
– (2004): Growing Public: Social Spending and Economic
Growth since the Eighteenth Century, Cambridge University Press.
Maddison, A (1987): The Cambridge Economic History of Europe, Cambridge University Press.
Mason, E S, Kim, Maha Je, Perkins, Dwight H, Kim, K S, and Cole, David C (1980): The Economic and Social Modernisation of the Republic of Korea, Harvard University Press, Cambridge, Mass.
Mehrotra, S (1998): ‘Education for All: Policy Lessons from High Achieving Countries’, International Review of Education, Vol 44, 5/6, pp 461-484.
– (2002): ‘Basic Services for All? Ensuring Accountability through Deep Democratic Decentralisation’, http://hdr.undp.org/docs/publications/background_papers/2002/Mehrotra_2002.pdf.
Mehrotra, S and Buckland, P (2001): ‘Managing Teacher Costs for Access and Quality’, Economic and Political Weekly, December 4.
Mehrotra S, and Jolly, R (1997): Development with A Human Face: Experiences in Social Achievement and Economic Growth, Clarendon Press, Oxford.
Mehrotra, S, P R Panchamukhi, Ranjana Srivastava and Ravi Srivastava (2005): Financing Elementary Education in India: Uncaging the “Tiger” Economy, Oxford University Press, New Delhi.
Mehrotra, S, and Delamonica, E (forthcoming): Public Spending for the Poor: Getting the Fundamentals Right on Social and Macroeconomic Policy, Oxford University Press.
Prost, A (1981): Histoire Generale de l’Enseignement en France, 1800-1967, Armand Colin, Paris.
Psacharapolous, G, Tan, J P and Jimenez, E (1986): Financing Education in Developing Countries, World Bank, Washington DC.
Samoff, J (1996): ‘Which Priorities and Strategies for Education?’ International Journal of Educational Development, 16/3, pp 249-71.
Sanderson, M (1983): Education, Economic Change and Society in England, 1780-1870, Macmillian, London.
Schullz, Theodore (1963):The Economic Value of Education, Columbia University Press, New York.
Sen, A (1985): Commodities and Capabilities, Oxford University Press.
– (2000): Development as Freedom, Oxford University Press.
Stephens, W B (1998): Education in Britain, 1750-1914, St Martin’s Press, New York.
Stiglitz, J (1998): ‘More Instruments and Broader Goals: Moving Toward the Post-Washington Consensus’, Annual Lecture 2, World Institute for Development Economics Research (WIDER), Helsinki, Finland.
Tan, J P et al (1984): ‘User Charges for Education: The Ability and Willingness to Pay in Malawi’, World Bank Staff Working Paper, No 661, World Bank, Washington DC.
Tendler, J (1997): Governance in the Tropics, Princeton University Press.
Tilak, J B G (1997): ‘Lessons from Cost Recovery in Education’ in Colclough (ed), Marketising Education and Health in Developing Countries: Miracle or Mirage?, Clarendon Press, Oxford.
– (2000):‘ Financing of Elementary Education in India, Year 2000 Assessment: EFA’, Government of India and National Institute of Educational Planning and Administration, New Delhi.
Thobani, M (1983): ‘Charging User Fees for Social Services: The Case of Education in Malawi’, World Bank Staff Working Paper, No 572, World Bank, Washington DC.
UNICEF (1998): Malawi: A Success Story in Education, New York.
Wade, R (1990): Governing the Market: Economic Theory and the Role of the Government in East Asian Industrialisation, Princeton University Press, Princeton.
West, E G (1965): Education and the State, Institute of Economic Affairs, London.
– (1994): Education and the State, Liberty Fund, Indianapolis.
World Bank (1986): Financing Education in Developing Countries: An Exploration of Policy Options, World Bank, Washington DC.
– (1993): The East Asian Miracle, World Bank, Washington DC.
– (1995): Priorities and Strategies for Education, World Bank, Washington, DC.
– (1999): Knowledge for Development, World Development Report, World Bank, Washington DC.
– (2001): Private Sector Development Strategy – Directions for the World Bank Group, Washington DC.

© Copyright 2001 The Economic and Political Weekly. All rights reserved.