Chair:

Simon Maxwell, ODI

Speakers:

Michael Lipton, PRUS, University of Sussex.

Jonathan Kydd, Imperial College.

Peter Gibbon, Centre for Development Research, Copenhagen.

Simon Maxwell introduced the discussion. The previous week's meeting had debated the role of agriculture in rural development, and had concluded that the direct and indirect poverty linkages were strong - provided that the economic health of the sector could be preserved. This did not mean, however, that a small-farm development option was either feasible or desirable. The case needed to be made.

Michael Lipton set out the issues. His background notes pointed to the historical evidence in support of the thesis that small farm development was poverty-reducing, and to the contemporary importance of small farm agriculture to rural livelihoods in developing countries. In labour-surplus countries, the evidence suggested that an agricultural focus was both efficient and equitable. However, agricultural growth had slowed, so had employment intensity, and, perhaps as a result, so had the rate of poverty reduction.

Globalisation and technical change had both been implicated in the reduced contribution of small farm development to poverty reduction. On its own, in theory, neither should have had negative effects. Globalisation, after all, should favour labour-intensive production in labour-rich countries; and technical change should respond to local factor availabilities. In practice, however, the interlock of technical change and market liberalisation could be prejudicial to small farms: research, for example, was increasingly in the private domain, and targeted to those with ability to pay. Unless protected by transport costs and remoteness, few unskilled farmers would survive and compete. Policy, however, could make a difference.

Jonathan Kydd (who also offered some background notes) framed the debate as being between those who thought that small farm development represented a win-win option for both efficiency and equity, and those who thought that equity considerations might require a subsidy to small farms. He then introduced a framework for thinking through the issues, using perspectives from new institutional economics (concerned with transformation and transactions costs), neoclassical economics (concerned with factor scarcities and prices), and 'dynamic' issues, including trade liberalisation, technical change, changes in supply chains etc . . . He noted especially that the rapid rate of change in the world meant that past conclusions about small farm development might not necessarily hold in the future.

A more detailed analysis of transformation and transactions costs suggested that small farms could be expected to have some advantages - for example, lower costs of hiring and supervising labour - but also faced substantial disadvantages - for example in credit markets. The main, somewhat speculative, conclusion, was that the areas of the globe where support for small farmers is 'win-win' is diminishing quite fast. The 'win-win frontier' is contracting. Again, public intervention can counter-act some negative aspects, for example interventions in technology, a greater focus on the transactions costs of small farmers, and some rethinking of the liberalisation agenda.

Peter Gibbon based his remarks on a background note about agro-commodity chains, particularly those for traditional export crops (e.g. coffee, cotton). These chains had evolved in three important ways since about 1980, in each case with potentially problematic implications for small farms. First, there had been a shift from horizontal and public coordination (for example through marketing boards) to chains that were market-based and vertically-organised (often by transnational actors). Second, there had been a rapid differentiation of quality conventions, with more and more complex standards being introduced for both product (e.g. quality) and process (e.g. pesticide residue). Thirdly, the sector was marked by rapid technical change in production processes, which tended to favour the large-scale (for example, genetic engineering of coffee bushes which ensured that all berries ripened simultaneously, allowing a single strip harvest, rather than repeat selection of ripe berries).

Gibbon argued that classical agro-commodity chains offered smallholders broadly-based upgrading possibilities. New chains offered potentially deeper rewards, but more difficult to access and for much smaller numbers. The policy challenge was to devise new forms of regulation that could make new opportunities more socially inclusive and more predictable.

Simon Maxwell opened the discussion by reminding participants that many development specialists, not least in the NGO community, regarded the small farm development model as sacrosanct. The sober presentations given by the speakers should be regarded as deeply worrying by those who held to this view. Was there now a deep divide between optimists and pessimists? The discussion largely addressed this question:

• One or two interventions pointed out that investing in small-farm agriculture could still be highly 'profitable' - economically and socially. NGO projects in this sector, presumably within Kydd's win-win frontier, often showed high returns.
  

• It was also true that small farmers were not, as Michael Lipton reminded the audience, 'being pushed against the wall'. Numbers were not declining. On the other hand, as the discussion the previous week had shown, the rural poor were diversifying out of agriculture almost everywhere, and many small farmers should now be considered as part-time.
  

• In terms of the future, it was important to develop further the agenda of positive interventions that public agencies and others could make: apart from reversing the decline in support to agriculture, rich countries could help by supporting fair trade and environmental branding (though these were not unproblematic, and were likely to remain small in scale).
  

• These more optimistic conclusions, notwithstanding, the future of small-scale farming remained far from certain. It should not be forgotten, for example, that the findings of commodity chain analysis applied as much to grains as they did to coffee or cotton - increasingly so as the developing world became more urban, and as food systems became more industrialised.