Redistributive Land Reform: No April Rose. The Poverty of Berry and Cline and GKI on the Inverse Relationship

Abstract

At the theoretical heart of the Griffin, Khan and Ickowitz (GKI) case for redistributive land reform (‘a many‐splendoured thing’) lies the highly influential study by Albert Berry and William Cline, Agrarian Structure and Productivity in Developing Countries, published for the ILO in 1979. That study is regarded by many as the definitive work on the inverse relationship between farm size and land productivity. This paper subjects Berry and Cline, and by extension GKI, to critical scrutiny with respect to their policy implications, theoretical framework and empirical evidence. It also provides an alternative class‐theoretic approach to understanding the inverse relationship which undermines the use of the latter as the central rationale for redistributive land reform. If the approach of Berry and Cline can be shown to be theoretically, methodologically and empirically flawed, then perforce the argument and policy recommendations of GKI, who replicate that approach, can be shown to be fundamentally defective.

Leaving the Household Out of Family Labour? The Implications for the Size–Efficiency Debate

“…For example, Griffin et al (2004: 371) note the U-shaped findings in the study of centre-west Brazil mentioned by Dyer (2004), but they also report on a study of north-east Brazil that found a clear inverse relationship after controlling for soil conditions. Using dummies for the type of production undertaken (to control for land quality differences), Deininger et al (2003Deininger et al ( : 1395 find a large and statistically important negative relationship between farm profits per acre and size in Nicaragua.…”

“…However, Dyer (2004) has made a damaging methodological critique of these two studies. He argues that a crucial distinction must be made between aggregated levels of analysis (for countries or regions) and disaggregated studies.…”

“…Dyer shows that the problem of over-aggregation is present and furthermore, that the most recent data quoted show a 'breakdown of the inverse-relationship', which appears to be due to the introduction of new technology (2004: 65). Dyer (2004) then presents a range of newer studies that he argues show more methodological soundness by, inter alia, disaggregating data to show land quality and water access differences. A case study of districts in Pakistan shows that some districts exhibited a positive relationship between output per acre and size (2004: 65), a study in centre-west Brazil shows a U-shaped relationship (2004: 64) and newer studies of India show a weakened inverse relationship (2004: 68).…”

New investment, old challenges. Land deals and the water constraint in African agriculture

“…One the one hand, advocates of small-scale production units argue that small farms are more labourintensive and hence more 'efficient', producing greater yield per unit area as well as greater employment (Lipton 1977, Griffin et al 2002. On the other hand, critics argue higher land productivity on smaller farms may be illusory, reflecting denser population in more fertile areas (Dyer 2004), or that it arises not from greater efficiency, but from greater selfexploitation, of family labour (Byres 2004). This latter point is illustrated by a World Bank study comparing production costs in Brazil, Thailand and four African countries, which observed:…”

Access to Land, Rural Development and Public Action: The When and the How

“…Therefore, disentangling the possible causes of variation, we find that the impact of land distribution from large/rich to small/poor landowners has no predefined effect on growth per se, but depends on the relative importance of other economic factors present in each specific location. In fact, we can find new empirical evidence on the size-productivity relationship in developing countries showing all the different relationships that one can imagine (see, for example, Dyer, 2004). The conclusion is straightforward: the need to distribute land when ownership is skewed cannot be universally justified from a microeconomic analysis perspective.…”