Zimbabwe’s resettlement experience from 1980 to 2000

My recent blog on ‘phases of land reform’ focused on the post-2000 land reform period, but of course there was resettlement before 2000 as part of the post-Independence land reform efforts from 1980 to 2000. This too had phases, again affected by the wider political economy, especially via the legal framework for land appropriation and redistribution by the state.

Bill Kinsey who, together with colleagues, examined post-Independence resettlement in great detail through an important longitudinal study, commented on the earlier blog, asking what about the post-1980 resettlement experience? A good point. So, over this last week I have spent some (limited) time going back over the literature on this period – including important work from Bill and team (reflections back and links to the post-2000 land reform are here and here), and of course Sam Moyo (see also here, here, here and here, amongst many others), as well as the vital paper by Robin Palmer mentioned last week. Apologies for the length of this blog, but I hope Bill and others will use the comments to correct me and add to the information!

Phases of land reform post 1980

A basic story emerges, again in phases set in this case by changing legislative provisions over land. After Independence, the new government was committed to resettlement but was constrained by the Lancaster House Agreement that stipulated only ‘willing seller-willing buyer’ arrangements with large-scale commercial farmers. This meant that only certain land came onto the market, often not good quality and far from services and infrastructure making resettlement challenging. For example, by 1983 only 4% of land allocated for resettlement was in Region II (a high potential area) and none in Region I, with most medium-low potential areas in Regions III and IV. Following an initial phase of enthusiasm for resettlement this waned as the state focused more on investing in the communal areas as part of its wider political pacts with the people.

With the end of the strictures imposed by Lancaster House Agreement in 1990, it was possible for the state to acquire land and the Land Acquisition Act was passed in 1992. However, this was the period of the economic structural adjustment programme (the dreaded ESAP, which started in 1991), and in order to comply with the conditions set by the International Finance Institutions the government was reluctant to rock the boat, and anyway it didn’t have any money as the state contracted. This was the period when private transactions in land increased, but not for resettlement, as it was black business and political elites who acquired farms through the market.

By the end of the 1990s another period emerged, marked by a famous dispute with the new UK Prime Minister Tony Blair and his development secretary Clare Short. In 1997 they made it quite clear that the British were not going to pay for land for resettlement and that the UK claimed it did not have an obligation towards its former economy. This of course infuriated President Mugabe and a more radical stance on land ensued. This was particularly prompted by the rise of the war veterans as a political force once more, and their demands on the state for compensation for their war efforts. A huge payout was agreed also in 1997, which created turmoil in the economy. Around this time, a new opposition party, the Movement for Democratic Change, became a significant force especially in urban areas, signalling discontent with the ruling party. In the period that followed various land invasions occurred that were the prequel to the major invasions that happened after 2000.   

Welfare outcomes: it takes time!

What were the outcomes of land reform over these periods? The research tells an interesting story, with many parallels now becoming clear with the post-2000 period. A key point made by Bill Kinsey and others, based on wider experience of land reform, is that it takes time for the results of redistribution to show.  In a 2003 piece, Kinsey comments:

“Experience shows that welfare levels are almost universally lower following resettlement than before. The period following resettlement is one of stress and adjustment from which most—but not all—households will recover. There is then an upturn as farmers complete the post-relocation adjustment process and begin to reap benefits from their enhanced resource base. As experience accumulates and collaborative efforts begin, benefits continue to grow— often quite rapidly. At some point, as the potential of the resource base and the technologies employed are more fully realised, the rate of growth of benefits slows”.

He offered a simple graph to illustrate the point (explained in terms of ‘stages’ in this later paper). In other words, it takes a decade to find out if resettlement has had a positive impact, and longer-term impacts will largely depend on the wider economy, and the way technologies, markets and so on become available, as with standard patterns of agriculture-led growth (see also other chapters in the book, and the earlier LTC offering from 1994).

Let’s now look at the different ‘phases’, starting with the post-Independence period.

Post-Independence resettlement: big ambitions but limited implementation

The newly independent government had big plans for resettlement, and with large areas of land abandoned due to the war and later due to drought in 1982-84 there was land to acquire at reasonable prices. Up from an initial target of only 18,000 for the intensive programme, a larger target of 162,000 households to be settled was set by the accelerated programme. In practice over the following decade only 70,000 households were settled as the programme ran out of steam, but at the beginning there was great enthusiasm.

Initially the targets for resettlement were the poor, landless, refugees and those displaced from the war. After 1984, ‘Master Farmers’ were included, intended to offer a demonstration of ‘good farming’ to the settlers, and later the programme shifted much more towards ‘qualified’ farmers with the aim of boosting productivity. The overall aim was to provide adequate levels of agricultural income on ‘viable’ holdings, specified in dollar and hectare terms in the initial programme documents.

At the outset there were four different ‘models’ – the Model A schemes, which were villagized settlement of smallholders, comparable to A1 schemes; Model B schemes, which were collective, cooperative schemes where a socialist form of production was to be carried out, involving particularly demobbed war veterans; Model C schemes, which were ‘commercial farm settlement schemes’ linked to core ARDA estates and Model D schemes, designed for extending grazing areas for livestock, particularly in Matabeleland. These were technical schemes under direct jurisdiction of the Ministry of Lands and Agriculture, problematically putting resettlement areas outside the new administrative structures of independent Zimbabwe.

In the end, it was Model A smallholder schemes that became the dominant approach (93% of 71,000 settled by 1999), with Model B and C schemes being quickly abandoned and converted into more individualised, village-based arrangements, while Model D schemes did occur but had mixed success. Bill Kinsey and team studied three Model A schemes across three agroecological zones (II-IV), collecting data from 400 resettlement households from 1983 and 100 comparator communal area households since 1997. It is an impressive data set, building on a strong tradition of rural household studies in Zimbabwe in the 1980s and 90s, and we used the questionnaire as a template for our studies in the 2000s (although a much shorter version!).

In the early years, the studies highlighted the struggles of establishment. Unlike the ‘fast-track’ schemes of the post-2000s these were supposed to be planned schemes with services supplied. As Kinsey noted in 1982: “Settlement schemes are currently being designed and implemented under great time pressure and with a unidimensional emphasis on short-term technical aspects rather than longer-social and economic issues”. Settlers complained about lack of safe drinking water (and so poor health), poor transport networks and food shortages. In a short paper from 1984, Kinsey explains the enthusiasm of the new settlers, noting that despite being in the midst of a major drought, the word ‘happy’ mostly described settlers’ outlook, with most saying that “they have come to feel at home on their schemes despite early difficulties”. Nevertheless, in this early establishment phase, “settlers faced a host of problems, including clearing of land, moving onto their land late and having to build houses–in the rains–while trying to get a crop planted, building new schools, wild animals’ eating crops, lack of infrastructure and finally drought–the one theme that dominates any discussion with settlers. They had come from many different parts of the country and had to form communities and collaborate.

At this stage (the mid-1980s), there was other spontaneous resettlement (aka ‘squatting’) going on around the country, most notably to the frontier lands of Gokwe and Hurungwe. As a result of state-supported efforts to reduce tsetse fly infestation and so trypanosomiasis, huge areas became available for farming using draft power as cattle did not die. This resulted in the cotton boom in these areas and a massive expansion of agricultural activity and increasing wealth for the new settlers. In contrast to the resettlement areas, which were few and far between, these were patterns that I was more familiar with whilst doing my PhD research in Zvishavane district.  

In both cases, but through different routes, there was a growth in income for settlers, set in a context of agricultural ‘success’ in the smallholder sector more broadly. For the formal resettlement areas, this was well documented by Kinsey and team in a series of influential papers. From a rather pessimistic tone struck in his papers of 1982 and 1983 where questions were raised about government capacity for implementation and the overall economic returns of resettlement, a much more positive perspective (with qualifications) emerged over time. This was reinforced by an evaluation carried out for the British aid agency, published in 1988. This showed returns of 20%, prompting The Economist to comment that this must be “one of the most successful aid schemes in Africa”.

In an important paper published in 1999, Kinsey showed that household production and income was superior to communal area comparators, with asset accumulation of cattle being notable. However, nutritional indicators were much more similar suggesting that the gains of land reform were not translating into individual nutritional wellbeing. As a follow up paper surmises this was in particular because of larger household sizes and lower incomes from remittances, consumption levels of resettlement households were found not to be higher than communal areas. This 2001 paper corrected for the agroecological bias in the sample and found similar if less dramatic results. The paper nevertheless shows that “Land reform beneficiaries cultivate nearly 50 percent more land than non-beneficiaries, obtain four times as much in crop revenues, own substantially more livestock, and have expenditures that are higher by 50 percent”. A paper from 2004 looked at returns to land and a more detailed comparison of land reform beneficiaries and communal area comparators (using propensity score matching). This again showed positive returns to land reform ranging from 5% to 15% per annum, but these estimates are dependant on assumptions about the opportunity cost of land and the economic role of the additional household members attracted to land reform areas.

This work challenged the assumption that resettlement was a waste of time, more a political sop to liberation war rhetoric than anything that could achieve ‘growth with equity’ (as the slogan for the first national plan proclaimed). Instead, resettlement was driving growth and increases in welfare, something that the team working on land in the World Bank (notably Klaus Deininger and Hans Binswanger) was interested in, especially in the context of post-1994 South Africa. This growth of production, assets and income on the resettlements became a magnet for others, and many household sizes grew as relatives and others moved to the resettlement areas. Household sizes by the 1990s were significantly higher in resettlement areas compared to communal areas, a big change from the early 1980s. Patterns of productivity however were highly uneven both within and across schemes, reflecting processes of social differentiation as well as different production systems (with some irrigated schemes being the star performers).

Structural adjustment: redistribution takes a back seat

Enthusiasm for the resettlement programme had tailed off before the initiation of ESAP in 1991, but the lack of government funds and a lack of enthusiasm from international donors for resettlement meant that the programme effectively ground to a halt, despite the fact that from 1992 the Land Acquisition Act meant that land could be acquired by the state.

Even though the ambitious targets for expanding the programme were not reached there were still around half a million people living on the resettlements. With the decline in government interest, the planning requirements of the early phases were relaxed. Resettlement officers no longer insisted on every regulation, including requirements for cropping patterns, carrying capacities for livestock and so on. The restriction on taking external employment was also released (or ignored) after 1992, as this was really limited opportunities. The idea that farming in Zimbabwe had to be ‘full-time’ and could not migrate was absurd – not even the large-scale commercial farmers were restricted to their farms as the only source of income. However, as resettlement households diversified there were tensions between ‘survival’ and ‘growth’.

By 1997, the government had acquired 3,498,444 ha and resettled about 71,000 families. In this phase, the farmers were able to invest and expand. The graph of increasing welfare was being followed. The studies showed how both agricultural production and income as well as assets increased, notably cattle. While droughts (notably in 1982-84 and 1991-94) affected resettlement households, but as they increased livestock assets they were better able to cope. Generational transfers were a challenge as there were restrictions on subdivision, but a more flexible approach was adopted and the next generation was often accommodated.

The politics of land and resettlement post 1997

The period after 1997 was highly volatile and land became a big issue. The Clare Short letter, the war veterans pay-out and the sporadic land invasions – most famously in Svosve but in 15 other sites too – became a sign that the failure to address resettlement effectively over the previous 17 years was a mistake. With a further change in land acquisition legislation and the increasingly vociferous demands to address the land question, the donors and the government, together with the commercial farmers, came together to discuss in 1998. The land conference was an attempt to map out an orderly way forward – with two models proposed (A1 and A2, small-scale and medium-scale) and a plan to transfer 1471 farms that were deemed ‘underutilised’. However the Inception Phase of the Land Reform and Resettlement Programme Phase II (LRRP II) that followed only managed to allocate a further 168,264 hectares to 4697 families, way below the target of 1 million ha and 118,000 families settled envisaged at the land conference.

In this period – and indeed for much of the 1990s – the ‘old resettlements’ were barely part of the debate. The work of Bill Kinsey and colleagues helped make the case that land reform could have welfare benefits and that it was not such a disaster as many feared (far from it), but the key question became how to gain access to sufficient land through compulsory acquisition with compensation so that a significant land reform could be realised. The late 1990s plan never came to be as the donors, the government and the large-scale farmers failed to come up with a realistic deal. The rest, as they say, is history.

This blog was written by Ian Scoones and first appeared on Zimbabweland