Thursday, September 9, 2010

New World Bank Study on Large-Scale "Land Grabs"

More mistaken reports of Chinese investment. photo: farmlandgrab.org

Two days ago, on September 7, 2010, the World Bank released its long-awaited study of large scale land investments. The report overall seems quite balanced and contains the expected mix of concern and pragmatism. In some countries with ample land and low population densities, commercial investment might provide benefits, if done with concern for mitigating social and environmental impact, and within the rule of law. The report emphasizes that far too often, this is not being done, and the potential for harm is immense.

Although I think it will be a helpful contribution, overall I was a bit disappointed in the study, for several reasons.

First, they actually made use of the media reports collected by GRAIN at farmlandgrab.org, put all of them into a database, and performed econometric analysis on them, without checking the veracity of these reports. Yet what does this really tell us about the realities of large-scale land investments when so many of the media reports are wrong?

With regard to Pakistan, for example, World Bank researchers did fieldwork to cross-check the media reports collected by GRAIN: "In none of these cases could any evidence of investment be found" (p. 40). Yet, apparently, even these spurious reports still went into the databases and number crunching.

This seems to me a bit like doing econometric analysis of media reports of all the suspected locations of Saddam Hussein's weapons of mass destruction in Iraq circa 2002 and then saying something useful about the threat these posed to US national security.

Second, the report is very bland and addresses the issue at such a high level of aggregation that for anyone interested in learning specifics about the reality of Chinese activity in agricultural investment abroad, there was nothing whatsoever to learn. This is very different from the report produced by the researchers at FAO, IFAD, and IIEE in 2009 which admittedly had as one purpose looking into the veracity of these reports. Indeed, the researchers at Grain.org criticized the report for the very same reason, saying:
Had the Bank really wanted to shed light on this new investment trend it would have at least peeled back the curtain on the investors. Who are they? What are they after?
But perhaps other readers will have different opinions? Click here for an ongoing discussion on this topic.  

Keywords: China, Africa, World Bank, land grab

7 comments:

Sigrid Ekman said...

Wow...econometrics has hit a new low.

Anonymous said...

I disagree completely about including the GRAIN reports. The Bank faced a real risk with this report of coming across as too sympathetic to international investors and too dismissive of the (real and imagined) victims of land grabbing. I think it was wise of them to include all these reports to show that they take the problem of land grabbing seriously. Being excessively credulous about the concerns of small, powerless farmers in developing countries is not a complaint the Bank gets very often, and one I'm sure it welcomes.

Carlos Oya said...

I agree. You're spot on. It's classic WB empirical approach, basically the popular model "garbage in = garbage out". They do it too often. This time, despite the importance of the topic and the dubious quality of underlying data, they didn't even bother to make an effort. Independent researchers will have to do that.

Deborah Brautigam said...

Thanks for the comments. I think there is a difference between taking the concerns of small farmers and landowners seriously -- an excellent and worthy goal that I think I address in my book (see chapters 7 & 8 in particular on China's agricultural engagement in Africa) -- and using poor "data", apparently without any effort to "clean" it by throwing out spurious cases, in order to perform econometric analysis intended to inform those concerned about this issue.

Matthew Robertson said...

The point about the inaccuracy of media reports makes sense, but the alternative is not clear. How should people who want to get a grip on what is going on here treat the statistics culled from those reports?

Unfortunately both the donorplatform links are dead (webcitation.org is a brilliant solution to this problem) Is anyone doing more serious research into the question of Chinese land acquisitions abroad?

I participated in a teleconference discussion today with Lester Brown and brought up this question. He said that they had purchased land as big as Belgium sometimes (unclear whether this meant in aggregate or in one gulp)--but added that the projects often faced enormous infrastructural hurdles, like fuel storage facilities, places to repair the tractors, and the whole gamut of equipment associated with modern agricultural production. He did not get into the details of where these places were, however, and I couldn't ask a follow-up question.

Deborah Brautigam said...

Statistics culled from media reports can only give us statistics on media reports, not on the underlying phenomenon. I'd say use the media reports as a starting point and have in country teams check them out. It would take some travel but I can imagine dozens of student interns would jump at the opportunity to do this kind of summer field research for expenses.

On Lester Brown, is he in a position to have good statistics on Chinese land purchases? I'm not confident he is ... and the description of a land deal "the size of Belgium" is often associated with the infamous Daewoo (Korean) land deal in Madagascar ... but hey, Chinese, Koreans ... they all look alike.

Deborah Brautigam said...

Matthew thanks for pointing out the non-working links. We've fixed them and they should be working now.