Abstract
After three generations of changing paradigms, development economists know more today about the deep determinants of economic development than at any time before. This knowledge has helped to devise policies and programmes that contributed to the promotion of economic and human development in many parts of the world. But in almost all of these cases, internationally assisted development policies complemented already existing, home-grown development processes. In no case have development economists been able to devise programmes or policies that succeeded in igniting previously inexistent processes of sustained, broad-based growth. Of course, such economic miracles did happen in the last decades, most notably in East and South Asia, and have helped there to significantly lower the proportion of people that live in extreme poverty. But, again, these sustained growth accelerations were mostly driven by indigenous factors and unorthodox development strategies that dissented from the prevailing paradigms in the development literature.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Evidence for this statement is implicitly provided by a number of influential studies on the effectiveness of international aid (see, e.g. World Bank 1998; Burnside and Dollar 2000). By concluding that aid does work, but only in countries that have good economic policies and institutions, these studies actually admit that development assistance cannot ignite growth-promoting policies, but can only support existing policy reform processes that are driven by strong government ownership on the part of the recipient country. This is a notable implication of the studies’ conclusions. For their detailed discussion see Sect. 9.2.1.
- 2.
Accepting this challenge requires realising that igniting economic growth in low-income countries and sustaining existing growth dynamics in more advanced developing countries are two different enterprises which require differentiated policy agendas. Moreover, it has to be noted that igniting growth, as understood here, is a long-term process that involves the formation of an institutional environment that provides the preconditions and incentives for economic activity to grow on a broad base beyond subsistence levels.
- 3.
In the last decade, many African low-income countries south of the Sahara have experienced hitherto unknown rates of economic growth. But in very few of them has this growth actually been of a sustained nature and resulted from economic activity in diverse fields of the economy and with broad participation of the poor. Instead, according to the World Bank’s African Development Indicators (World Bank 2008), growth was highest in oil-exporting low-income countries like Equatorial Guinea (30.8% average GDP growth per year in the period 1996–2005), Chad (9%), Angola (8.5%), Sudan (6.3%), Nigeria (4.3%) and the Democratic Republic of Congo (3.4%). This implies that in many of these countries, the challenge of igniting broad-based, sustained growth represents a challenge in spite of their positive aggregate growth rates.
- 4.
The notion of high quality growth has entered the rhetoric at international development institutions in the last decade, meaning mostly economic growth that brings lasting employment gains and poverty reduction, provides greater equality of income through broader access, especially for women, to markets, education and finance and protects the environment. It has frequently been invoked as an important objective of development policy, with very little mention of how it is to be achieved, though. For further reference on the quality dimensions of growth, see Barro (2002) and World Bank (2000); and for a summary of the discussion on pro-poor growth, see Ravallion (2004b), and Lopez (2005).
- 5.
The economists whose contributions to development economics earned them a Noble Prize are Jan Tinbergen, Noble Laureate in Economics in 1969; Simon Kuznets in 1971; Gunnar Myrdal in 1974; Arthur Lewis in 1979; James Mirrlees in 1996; Amartya Sen in 1998 and Joseph Stiglitz in 2001.
- 6.
Noble-Prize-winning research in economics that has fruitfully been applied to the study of problems of developing countries include Friedrich von Hayek’s analysis of the interdependence of economic, social and institutional phenomena (1973 Noble Prize in Economics); Herbert Simon’s research into the decision-making process within economic organisations (1978); George Stigler’s studies of industrial structures, functioning of markets and causes and effects of public regulation (1982); James M. Buchanan’s development of the contractual and constitutional bases for the theory of economic and political decision-making (1986); Robert Solow’s contributions to the theory of economic growth (1987); Ronald Coase’s clarification of the significance of transaction costs and property rights for the institutional structure of an economy (1991); Gary Becker’s extension of the domain of microeconomic analysis to a wide range of human behaviour and interaction, including non-market behaviour (1992) and Douglass North’s renewal of research in economic history to explain economic and institutional change (1993).
- 7.
For instance, the function of legal and juridical institutions to ensure compliance in contractual relations is often thought to correspond exclusively with the institutional form of a state-backed system of formal laws and public courts.
- 8.
See Sect. 9.2.2 for a detailed discussion of the methodological problems with this approach and their implications for development policy.
- 9.
Determining the goodness of fit of a policy of institutional design with local circumstances requires understanding the dynamic interplay between the economic system and the social, political and cultural spheres of life. It is in the latter where people’s mental models, traditions and customs are shaped. These informal institutions ultimately decide about the compatibility, and hence effectiveness, of an imported institutional design or policy reform. See Sect. 13.2.1 for a detailed discussion of this argument.
- 10.
For a collection of country studies that discuss such unconventional but successful development strategies, see Rodrik (2003).
- 11.
It is relevant here to recall from Sect. 6.2.1 that a similar view had already been adopted by Albert O. Hirschman in the 1950s. He was interested in uncovering the “hidden rationality” of seemingly odd, irrational or reprehensible social behaviour that determines development outcomes in low-income countries, and aspired to exploit these countries’ indigenous resources and structures for the design of context-specific development policies. Hirschman thus pioneered an approach the time of which had yet to come. At his time, however, development economics evolved into the opposite direction and became, with the second generation’s neoclassical turn, increasingly preoccupied with the obvious, i.e., with the visible and measurable characteristics of developing countries.
- 12.
Regression analysis is applied to estimate economic relations since the 1930s. Early examples include the study of demand for agricultural products by Schultz (1938) or the pioneering monograph on statistical testing of business cycle theories by Tinbergen (1939). Albeit the scepticism especially voiced by Keynes (1939) in the Keynes–Tinbergen debate, the use of econometric methods in empirical development research has risen continuously, partly because of the increasing formalisation of theoretical research in development economics, and in part also due to improvements in the quantity and quality of statistical data from developing countries. Today, cross-country regressions are the method of choice for almost all of the empirical literature on institutions and growth (see Sect. 8.3).
- 13.
Studies that use cross-country econometric analysis to test the empirical relationship between growth and trade openness include Dollar and Kraay (2004);Lopez (2005) and Yanikkaya (2003). For a discussion of the empirical literature on growth and institutions, see Sect. 8.3. Dollar and Kraay (2002), Cord et al. (2004) and Kraay (2006) use cross-country growth regressions to assess the empirical relationship between growth and poverty reduction.
- 14.
Rodrik (2005) discusses the implications of regressing economic growth on policy variables. He notes that in these cases, problems of reverse causality (or endogeneity) are bound to happen as the policy variable is an integral part of the null hypothesis that is being tested and thus cannot be treated as if it were exogenous or random. He concludes that in such a setting, any interpretations of the results from cross-country growth regressions must lead policymakers astray.
- 15.
Many further methodological problems challenge the robustness and reliability of the study’s findings. For instance, the results become untenable after adding four more years to the sample, or after the use of more plausible measures of aid and policies (see Easterly et al. 2003; The Economist 2007c). Similar points of critique were raised in an evaluation of the World Bank’s in-house research commissioned by the bank to a group of top academic economists chaired by Angus Deaton of Princeton University (Banerjee et al. 2006). Further critical reviews of the Burnside/Dollar and Collier/Dollar aid allocation models can be found in Beynon (1999, 2003).
- 16.
The Monterrey Consensus was the outcome of the 2002 United Nations International Conference on Financing for Development, in which over 50 Heads of State and 200 ministers pledged, among others, to increase their official development assistance to developing countries. The Consensus codified officially the call for effectiveness of foreign aid. The Millennium Challenge Corporation, announced by George W. Bush in 2002, was set up in 2004 as a new mechanism for granting up to 5 billion US dollars per year in additional US bilateral aid to well-governed developing countries. Eligible countries are selected on the basis of a number of independent policy indicators.
- 17.
For further critical reviews of the empirical growth literature, see Temple (1999), Glaeser et al. (2004) and Easterly (2005), as well as Sect. 8.3. Technical limitations of using cross-country regressions in explaining economic growth are discussed in Rodríguez (2007).
- 18.
See Sect. 6.2 for a detailed discussion of their contributions to development economics.
- 19.
See Hirschman (1994) and Sect. 6.2.1 (especially footnote 13) above.
- 20.
Clear cause-and-effect relations are also an indispensable precondition for the planning, implementation and evaluation of development assistance measures. For this purpose, development institutions often apply the so-called Logical Framework approach, a project management tool that involves identifying strategic elements (inputs, outputs, outcomes and impact) and their causal relations, together with performance indicators, as well as risks and assumptions that may influence project results. Without clear guidance as to the direction of causality between relevant objectives and the intervention strategy, as well as the causal chain of individual factors and conditions that need to be addressed, development projects cannot be planned and implemented in a meaningful way.
- 21.
This possibility had first been considered by Ragnar Nurkse (see Sect. 6.1.3).
- 22.
Similar examples can be found in almost all areas of development policy. For instance, the theoretical finding that high growth correlates positively with a high exposure to international trade does convey little policy-relevant information per se. What policymakers are actually interested to know is if they can expect an economy to grow if they abolish import restrictions or if they may instead be better advised to strengthen the competitiveness of its domestic industry, for instance, through investments in the education system that increases labour productivity and in turn enables domestic firms to increase production and exports. Another example is the theoretical finding that good institutions matter for growth. In order to translate this conclusion into effective development policies, additional knowledge is needed, for instance, on how high quality institutions evolve in a specific country context or on how to determine appropriate institutional forms and governance structures that a policy should promote to increase their long-term effectiveness and legitimacy. Finally, the empirical evidence on the link between economic growth and poverty reduction provides relatively little operational guidance as to how effective growth strategies can be designed that create the means and incentives for the poor to contribute to, and benefit from, economic growth.
- 23.
Premises represent general laws about the expected behaviour of individuals or groups (e.g. the postulates of rationality or the efficiency rule), as well as assumptions about the conditions under which they interact (e.g. the available production technologies or the constraints that economic actors face).
- 24.
Such self-deprecatory attitudes have already been noted by Hirschman (1961), and are more recently also contended by African leaders themselves. For instance, the Asantehene Otumfuo Osei Tutu II, traditional king of the Ashanti in Ghana, urged Africans in a speech at the Third German-African Forum on November 3, 2007 to realise the importance of their tradition in addressing their situation, and continued that “[w]e have allowed ourselves to be persuaded that there is nothing good in our tradition. As a consequence, we have almost jettisoned our tradition and culture, thinking that everything African is bad and everything foreign is good.” [quoted from Kumi (2007)]. A similar attitude can also be observed with regard to the research orientation of economists from developing countries. Instead of taking advantage of their familiarity with the reality of their home countries, they often turn towards the pursuit of economics as a science of universal laws in order to meet the high standards of formal rigour and innovative use of econometric methods that is still demanded by most international top journals. On the dualist structures that are evolving in many African societies as a consequence of the juxtaposition of traditional and modern influences, see Sect. 13.2.1.
- 25.
For a recent deliberation on the role of explorative data analysis in the process of theory formation in development economics, see Mookherjee (2005).
- 26.
See also Sect. 13.1.4.
- 27.
- 28.
Inductive research is actually also the favoured method of scientific inquiry by many authors of genuine African culture. Tévoédjrè (1978), for instance, calls for a rethinking of the economy along inductive lines.
- 29.
The growing body of the aid effectiveness literature ranges from historical assessments of the effectiveness of development assistance (e.g. Tarp 2000; Easterly 2001; Goldin et al. 2002) over econometric analyses of the effects of aid on growth and poverty reduction (e.g. World Bank 1998; Burnside and Dollar 2000; Collier and Dollar 2004) to in-house evaluations by development institutions of the effectiveness of their own projects and programmes (e.g. the Annual Reviews of Development Effectiveness by the World Bank’s Independent Evaluation Group or the Development Effectiveness Reports by the UNDP Evaluation Office).
- 30.
In his study, Peirce formulates a large number of hypotheses about the causes of bureaucratic failure which he illustrates by means of 11 case studies that deal with various instances of bureaucratic failures at the level of the US federal government. In spite of his focus on the United States, several aspects of his work, as well as the conceptual framework of the study, are useful and applicable also to the analysis of the institutional and behavioural determinants of bureaucratic failure in the foreign aid system.
- 31.
According to the 2007 OECD Development Cooperation Report (OECD DAC 2008), the proportion of financial aid from Western donors recorded as untied increased from 42.5% in 2002 to 53.0% in 2006 (the remaining share accounts for tied aid or aid those status is not reported). Despite the increasing untying of aid, the report also notes that the latest figures suggest that a large proportion of contracts financed from untied aid are still going to suppliers in donor countries.
- 32.
The OECD estimates that tying aid to specific commodities and services, or to procurement in a specific country or region, increases development project costs by as much as 20–30% (Jepma 1991).
- 33.
In order to mitigate theses sources of bureaucratic failure, aid agencies increasingly aim at obtaining feedback from independent evaluations of their projects and programmes. These evaluations often take the form of programme assessments including organisational reviews and process monitoring, which cannot estimate the magnitude of effects of an intervention with clear causation. Causal analyses that help to understand the relative role of alternative interventions require assessing the counterfactual (i.e., the situation of a target group without the intervention). This is inherently difficult and adequate approaches are just about to be developed and tested. For a discussion of some of these new techniques see Sect. 10.4.
- 34.
See Sect. 10.1.
- 35.
Uphoff and Combs (2001) provide two interesting case studies which demonstrate how “paradigm blockages” can obstruct the view on efficient but by the standard of the prevailing paradigm seemingly impossible solutions. In one case, they report of unrecognised opportunities for alternative growing techniques of rice in Madagascar which depart from international best practice but produce even higher crop yields in the Madagascan context. In another case they discuss how childhood rickets going unseen as a disease in Bangladesh because conventional wisdom stipulates that rickets does not occur in the tropics.
- 36.
The slow adaptability of the World Bank and other aid agencies to new development approaches has already more than two decades ago been noted by Ascher (1983) who found that employees at the Bank are much more guided in their decision-making by their ideologies and professional role models than by the political agreements that set out the institution’s goals and strategies.
- 37.
See also Kremer (2002) who extends the focus on the role of institutions and incentives on development outcomes to the design of development assistance. Easterly (2001) makes the argument that foreign aid doesn’t work because it doesn’t take into account the fact that people respond to incentives. For further elaborations on his arguments, see Sect. 10.2.
- 38.
The following examples are drawn in part from Ostrom et al. (2002).
- 39.
The “Samaritan’s Dilemma” has earlier been analysed by Buchanan (1975). For an application of the dilemma in a study on the effectiveness of development assistance, see Pedersen (2001).
- 40.
A further problem associated with the Samaritan’s Dilemma and the ability of recipients to reallocate resources to other uses (i.e., the problem of the fungibility of aid) is that recipients are more likely to become dependent on aid.
- 41.
For a more detailed discussion of the problem of corruption in both donor and recipient countries, see Shleifer and Vishny (1993, 1998), Hancock (1998), Johnson et al. (1998), Klitgaard (1988) and Wade (1982).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Altmann, M.P. (2011). Conclusions from the Past and the Agenda for a New Generation of Development Economics. In: Contextual Development Economics. The European Heritage in Economics and the Social Sciences, vol 8. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-7231-6_9
Download citation
DOI: https://doi.org/10.1007/978-1-4419-7231-6_9
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-7230-9
Online ISBN: 978-1-4419-7231-6
eBook Packages: Business and EconomicsEconomics and Finance (R0)