Guest post by Wytske Chamberlain and Wegayehu Fitawek of Land Matrix Africa, hosted by the University of Pretoria 

Remember the global riots over food set off by sharp spikes in commodity prices in 2008? The biofuel hype as THE solution to dirty oil? And the financial crisis that drove investors to look for alternative assets to invest their dollars, euros and pounds? As these developments came together in a perfect storm and a wave of “land grabbing” ensued, it came as no surprise that the African continent got thrown into the ring. With its seemingly cheap land (and labour), much of which was not being used, and need for investments, foreign exchange, and jobs, these land deals were a win-win situation, right?

Ten years down the line, as the recently published book The Transnational Land Rush in Africa – A Decade After the Spike reveals, vulnerable people were very much the losers. Yet, while the outcomes from individual projects are starting to crystallise, many studies that look into how much land is affected give vastly diverging numbers.

In Ethiopia, for example, three different sources reporting on land deals in the country cite three completely different total sizes for these acquisitions: 1.2 million hectares (ha), just over 600,000 ha, and 3.6 million ha, respectively.

These disparities cause confusion on the scale of land acquisition and raise an important question: “What is a land deal?” Classifying exactly what constitutes a land deal is essential if we are to be able to not only quantify the extent of these types of investments, but also, crucially, their consequences.

This may sound straightforward, but is actually a complex issue, on which the Land Matrix, an independent global land monitoring initiative established in 2009 to address the lack of robust data on large-scale land acquisitions (LSLAs), can shed some light.

Take a look at this common scenario, for instance. The transaction process begins simply enough, with an investor signing a Memorandum of Understanding with a host government. The investor then finds a suitable piece of land with the approval of the traditional authorities, who need to consult with their communities (although, in reality, community consultation is woefully inadequate). The host government may also request an environmental impact assessment for the selected area.

During each of these steps in the process, negotiations frequently alter the agreed size of the acquisition. The final contract typically ends up much smaller than the initial intention. As if this was not complicated enough, the government may add a clause in the agreement stipulating increases or decreases in land size depending on the performance of the investor.

Secondly, reported numbers may include forestry or mining concessions, which give investors the right to manage resources on the land, but do not confer any property rights. These concessions generally cover massive areas, and therefore including them substantially increases the reported size of the land rush. Similarly, areas for conservation, tourism, industrial activities and the like are sometimes also included.

A third major consideration when calculating LSLAs is whether to include domestic investors. This is critical, given that local elites and diaspora investors are known for controlling large areas in their home countries, and especially since their activities tend to be even less transparent than those of international investors. Including domestic investors also greatly increases the resources needed to capture these LSLAs, and therefore many studies choose not to do so.

Lastly, there is the question of what the minimum size of a deal should be in order to be included (ranging from as little as 20 ha to as much as 200 ha, or even larger), as well as from which year to begin tracking these transactions (often set at the start of the millennium).

Many other challenges around reporting exacerbate the problem. Contracts are considered highly confidential and land registration systems are not publicly accessible, if contracts are registered at all. Indeed, in cases where oral agreements are entered into between investors and traditional authorities, no written records exist of any kind.

Then there are the difficulties in continuing to monitor what happens after a deal has been concluded. An investor abandoning a project certainly doesn’t receive as much attention as the announcement of the signing. And what happens with all the land under contract that the investor is unable to develop? These factors all bring further nuances to how the numbers can be interpreted, and even manipulated, as “evidence” to promote a particular message. Defendants of LSLAs, for instance, will often exclude many elements to argue that the land rush is not such a big deal. Opponents, on the other hand, include more aspects, potentially inflating the size to strengthen their argument against the land rush.

Taking into account all these elements, although we know that the demand for land and natural resources has significantly accelerated in the last decade, it remains difficult to gauge the exact size of the land rush.  Even so, the Land Matrix allows users to explore a wealth of individual deals and investors through its platform. Deals can be filtered by a wide range of variables, such as location, negotiation and implementation status, and intended use.

Users can also delve deeper into specific deals to reveal even more detailed data to put together a comprehensive picture of the situation on the ground. For example, size details, contract particulars and information relating to local communities, including compensation, consultation, and displacement are all captured.

In this way, the platform allows for adaptation of the data, depending on the objectives and needs of the user. While it may not be perfect, it certainly provides a much more balanced and informed depiction of the data, and makes it possible to identify, and thus compare, apples with apples, so to speak.

So, when next you read how many hectares have been “grabbed”, “invested in”, or “developed”, be critical of the data – after all, to quote Paolo Magrassi, “If somebody tortures the data enough, it will confess anything”. Ultimately, in the words of Chip and Dan Heath, “Data are just summaries of thousands of stories – tell a few of those stories to help make the data meaningful.”