Tuesday 29 January 2013

A growing evidence base: yet more inconvenient truths

Last week in the latest of the Sokwanele “land debate” contributions, Dale Doré used his slot to critique our work in Masvingo. Since the publication of the book, Zimbabwe's Land Reform: Myths and Realities, exactly two years ago we have had plenty of reviews, and a number of critiques. Most common is the refrain, that Masvingo is different to other areas (of course it is: see the blog on Masvingo exceptionalism). Others have focused on the credentials and backgrounds of the research team, while others have questioned our sampling and methodology. Still others have called us names familiar to the discourse from the liberation struggle (sell-outs, collaborators, sympathisers, liberals, apologists and so on). Others have been plain bonkers or simply abusive, and I won't share these, in case there is a family readership of this blog.

All this shows the heated nature of the debate, and frustrations felt. Doré's piece focuses on methodology, while offering no new data to counter our arguments. He questions our approach to the study of complexity in particular which aimed at discovering emergent patterns from diverse data, arguing instead for a model-driven reductionism. In this regard he has problems with our chapters on labour and markets, suggesting that they are neither novel nor revealing. Well, others disagree, and so do I. This data offers, I would argue, fundamentally new insights into labour regimes and market processes, which have not been discussed before, and certainly both chapters analyse the processes and outcomes in great detail. The frustrations Doré feels may be due to disciplinary preferences (he’s an economist), but exploring patterns and processes on the ground in great detail, I believe has important merits, and reductionist approaches may do violence to the complexity observed.

Also, as part of his methodological assault, he disputes our use of baselines against which change is measured. But if you read the book you can see we were careful on this – using data on nearby communal areas, the past work of Bill Kinsey and colleagues on old resettlements, and the limited available data on the production and economics of commercial farms. And in relation to the baseline costs on investments, I am afraid he missed the detail in the footnotes which contains all the assumptions: the analysis cannot thus so easily be dismissed as 'sheer nonsense' Doré goes on to accuse us of simply creating 'straw men' myths to ease the flow of our narrative. This is an argument I have heard before. Surely, people have argued, no-one ever believed these myths! Well, just take a look at any media commentary, donor document and many academic pieces and you will see these myths (and many more) are alive and well. A particularly pure form appeared in the press recently penned by UZ Professor Tony Hawkins if you need convincing further. Later, in the piece Doré also accuses us of lack of triangulation, an approach to probing the robustness of data. Triangulation may be of methods (and we used every method, qualitative and quantitative, we found appropriate) or of cases (and again the site comparisons, within and between clusters, was central in the book), although we do admit that we did find it difficult to gain perspectives from former farm owners and workers, despite many attempts. Finally, Doré accuses us of making 'egregious' 'false claims' about the process of land reform. Again, I beg to differ. Our book offered the stories of what happened on 16 farms – all were different (as is clear from studies from elsewhere). The simplistic picture Doré paints, backed up not by empirical information but by broad proclamations, is not enough to understand the diversity of settings, processes and outcomes of Zimbabwe’s land reform.

Two years on (and why did it take this long for this review to emerge?), we actually have many more cases to compare with, improving possibilities of triangulation. In several talks last week in Harare I presented the following map, showing all the studies I know about which have looked at what has happened in the new resettlements since 2000. These include our Masvingo studies (green), the African Institute of Agrarian Studies district level research (purple), the Ruzivo Trust studies (now a book, yellow), the Livelihoods after Land Reform small grant studies (light blue) and a growing number of PhD studies (pink), some which were reported on in the Journal of Peasant Studies special issue. It is an impressive array, with pretty good geographical coverage, although clearly still some gaps. This is definitely an incomplete picture, so please let me know if you are doing something that is not captured here, as it is an important base for comparative analysis and reflection, both on commonalities and differences.


While there are important variations across sites, there is an emerging, common story that Doré and others still find difficult to accept. These are indeed inconvenient truths. The accumulating and converging evidence points to the following:
  • A1 farms are doing relatively well (although could do better), with a solid 'middle farmer' group within them who are reinvesting profits from agriculture in their farms. By contrast, A2 farms have struggled, although things have improved since the end of hyperinflation and in the multicurrency environment since 2009. They have been greatly assisted by contract farming arrangements that have provided much needed capital and inputs.
  • Private and community investment in the resettlement areas is significant, especially in the A1 sites. But more needs to be done, with clear needs for public investment in infrastructure.
  • Capture of farms by high level, politically-connected elites has taken place, and this varies between different parts of the country, especially in relation to proximity to Harare. However even in these areas, the dominant story remains small and medium scale A1 and A2 farmers. A1 farmers, particularly on land that was invaded and occupied, are largely from nearby communal areas and small towns, while A2 farmers are predominantly former or serving civil servants, teachers and business people, with urban connections.
  • The potential for production across the resettlements is far from being realised due to inefficiencies in input markets, a lack of credit and rural finance and the high costs of transition in infrastructure, and up and downstream industries. However, production has not collapsed, and is booming in some commodities and areas. Markets may be informal, but they generate employment and spin-off benefits from economic linkages in an area.
There are nuances and variations – yes complexity – but the picture is increasingly clear, as are the policy challenges. The now infamous five myths we set out to examine in Masvingo are rejected countrywide, although with important qualifications – as indeed we offered in the 288 pages of small type in our book for Masvingo.

In stark contrast to the Dale Doré diatribe on the Sokwanele site, at the SAPES Trust Policy Dialogue I spoke at last week, I was pleasantly surprised by the tone of the discussions. A sense of pragmatism and realism prevailed (mostly). The room was packed, with over 100 people attending from all sides of the debate – the CFU was represented in force, including the current President, as was the MDC, with the Director of Policy and Research, Charles Mangongera, offering the discussant's comments. And representatives from the Ministry of Lands were there too, including the Minister, Herbert Murerwa. Mandi Rukuni chaired the debate superbly, and it was clear that there was more agreement than many would expect.

As Zimbabwe moves into a new phase, and a new election settlement some time next year, the more consensus building and solid debate around facts and evidence that occurs the better. Ibbo Mandaza's SAPES Dialogues are good examples of such fora. Sadly unfounded accusations and gratuitous swipes, as in some of Doré's piece, are not.

This post was written by Ian Scoones and originally appeared on Zimbabweland