Select Committee on International Development Minutes of Evidence

Supplementary memorandum submitted by Professor Paul Mosley, University of Sheffield


  Overseas aid has had a bad press, and many blame it for the frequent failures of development policy in poorer countries. A large literature over the last twenty years has found it difficult to demonstrate the positive impacts of aid statistically, and casual empiricism does little better—witness the lack of correlation between aid flows into Africa and economic development in that continent during the years of economic reform in the 1980s and 1990s.

  More recently the World Bank—in its book Assessing Aid (1998) has claimed that overseas aid is effective in the sense of increasing the growth rates of recipient countries if and only if economic policies are "good". Economic policies, on this approach, are conceived as a blend of measures to ensure budgetary and macro-economic stability, and measures to keep the economy open. The prescription arising from this approach, which has become a conventional wisdom, is to confine aid to countries which practice "good policy" in the sense above defined. This prescription can easily be combined with poverty focus by limiting aid to poor countries with good policies as advocated for example by the World Bank in their recent (2000/01) World Development Report.

  Unfortunately the entire approach of the World Bank appears to rest on a statistical misinterpretation. For as demonstrated by the Danish economists Henrik Hansen and Finn Tarp (Journal of Development Economics, September 2001), aid flows now appear to be a significant influence on growth across the average of all developing countries, with or without "good policies" in the Bank's sense. The Bank's error appears to have arisen from failing to acknowledge that, within any individual recipient country, the effectiveness of aid tails off as the aid flow is increased. Once this factor is included in the analysis, the influence of "good policies" disappears. The message, so far as the growth objective is concerned, would appear to be that aid is now able in a majority of developing countries to improve performance by working on long-period factors—investment, infrastructure and the skill base—and that in the short term there is relatively little which policy can do to make or mar the effectiveness of aid. Given the still defective quality of policy and institutions in a number of aid-recipient countries, this is encouraging news for donors.

  Of course, donors are additionally concerned to reduce poverty by means of aid. They can do this, of course, simply through its impact on growth earlier mentioned, since in fast-growing recipient countries such as China, poverty reduces growth automatically so long as that growth provides employment and livelihoods to low-income people. Additional leverage on poverty can be, and is being, exerted by donors through a new style of conditionality in which ultimatum is avoided, and the pattern of public expenditure is used as the focus of policy dialogue. Two illustrations of this process, as mentioned in my verbal evidence to the Committee, are Uganda and Ethiopia. In both countries in the early to middle 1990s:

    —  advice was given by donors to liberalise policy and at the same time to move public expenditure in a more poverty-focussed direction, favouring primary health and education, agricultural research and extension, and rural water and sanitation;

    —  the response by the recipients was not immediate and in some ways was perverse, with both countries persisting with fixed exchange-rate policies and export taxation well into the 1990s;

    —  nonetheless, dialogue, trust and an aid flow were sustained;

    —  and eventually, the trust of the donors was rewarded, with the needed changes in expenditure policy eventually materialising and poverty falling—dramatically so in Uganda, from 50 per cent to 32 per cent in the decade to 2000.

  The message coming out of all this is that aid donors do have influence, but that it is not influence of a conventional kind. They have contributed to growth in developing countries by building up their stocks of physical and human capital, and they have helped to reduce poverty both by this route and through policy dialogue aimed at a pro-poor pattern of public expenditures. But the influence of more conventional policy dialogue, aimed purely at macro-stability and economic liberalisation, would appear to have been much less significant.

Professor Paul Mosley

University of Sheffield

May 2002

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 24 July 2002