Property Rights and Labour Relations: Explaining the Relative Success of Native Purchase Area Farmers in Southern Rhodesia, 1930–1965

Abstract

In the 1930s the colonial authorities in Zimbabwe set aside geographical areas where Africans were allowed to purchase land. Despite having private property rights to land, a rare occurrence among Africans in colonial times, the performance of this group of farmers has rarely been investigated. In this article, we show that the average group of ‘native purchase’ farming households performed far better than the average African farmer in the native reserves. We do more, by offering one of the first explanations behind the ‘success’ of this group of farmers. We argue that the explanation for this is not that private property rights were more secure than other forms of land rights as argued in mainstream economics. The farmers who owned land performed better than those who did not because private property rights changed social relations in a wider sense of the term. Private property rights enabled the emergence of various forms of non-family labour relations including sharecropping and wage labour that the landowner could exploit to increase production.

Keywords:

• capital
• land
• labour
• native purchase farmers
• productivity
• property rights
• agricultural investment
• Southern Rhodesia
• Zimbabwe

Introduction

As part of a larger package of further reinforcing racial segregation the authorities in colonial Zimbabwe (henceforth Southern Rhodesia) implemented the Land Apportionment Act of 1931. The prime aim of the act was to legalise the already existing division of the country’s land and water resources between black and white people. Yet it also allowed a small group of wealthier Africans to buy land, in so-called Native Purchase Areas (NPAs). In total, 7.4 million acres were set aside as NPAs in the period 1931 to 1962. The adoption of such provisions was no less than an admission by the colonial government that the rise of an African rural bourgeois class was inevitable.¹ This arrangement was virtually unique to colonial Africa. In most places, Africans were formally not allowed to own landed property. Despite its distinctiveness, the performance of this group of farmers has rarely been systematically investigated.² This study is an attempt to fill that gap. More specifically, we use previous research, colonial records and agricultural surveys to analyse both the ways in which native purchase (NP) farming households organised production from 1930 to 1965, and the performance of production in these households. We offer one of the first analyses of the factors that allowed the NP farming households to perform better on average than the farmers in communal areas that were known as ‘native reserves’ during the colonial period. We end in the mid 1960s because this period marked a shift in policies related to the NPA farmers. The unilateral declaration of independence (UDI) by the Ian Smith government in 1965 was followed by a more hostile and less supportive approach towards the NP farming households.³ The year 1965, therefore, marked the advent of a new agricultural policy under the UDI or Rhodesian Front (RF) government.

Notably, a well-defined and secure property rights system (read: private property rights) is, in the ‘new’ institutional economic history and in mainstream economics, often identified as a key institution in enhancing long-term agricultural development. The reason for this is simple – farming households that hold secure rights over land are more likely to invest more in the land, which will lead to increased productivity.⁴ Does this mean that rural households in NPAs fared better compared to those households that accessed land in the reserves? Ultimately, a detailed comparison between NP farming households and farmers in the reserves would be the best way to go about exploring this question, but limited data prevent a systematic comparison. Having said that, the survey and scattered aggregated data all point in the same direction: yields were higher on farms in the NPAs compared to those in the reserves. One may conclude that this provides support for the claim that private property rights are superior to any other form of landed property rights.

Does this mean that proponents of private property rights are correct? It is not obvious that this would be the case. A number of scholars have criticised the idea that ‘traditional’ land rights – today known as customary rights – impeded agricultural development in Africa. They based this conclusion on the many failed land-titling programmes in postcolonial Africa, where ‘formalisation’ and individualisation of land rights did not boost agricultural growth.⁵ A possible reason for this is provided by economic historians who argue that secure land rights would only matter in cases where land is scarce. If land is abundant and free for all to use, there would be no need to implement secure property rights. In most times and places in Africa land has been available in abundance and the NPAs are in this regard no exception. Population densities varied between the different NPAs, but in 1936 the average population density was two people per square kilometre. In 1958 the average population density was just 6.6 people per square kilometre.⁶ Under such conditions, the need for the privatisation of landed property rights is limited and would have no major impact on agricultural output.⁷

Given labour scarcity we would expect farming households in the NPAs to employ a labour-saving and more capital-intensive mode of production than the average farms in the reserves. This was in part the case. Still, the major difference was not the share of capital in the production process, but the hours worked. Farming households in NPAs spent significantly more working hours on farming than the average farm in the reserves. Thus, rather than just shifting the production function upwards through technological changes and improvements, they intensified land use by adding more labour at diminishing rates of return.

To understand why the farming households in the NPAs on average performed better than those in the reserves we need to understand the role of landed property rights from a wider perspective. Anthropologists have rightly pointed out that landed property rights cannot be analysed in isolation from other social and political relations.⁸ Changes in the governance of one factor of production will eventually affect social relations in a broader sense of the term. We argue that changes in landed property rights led to shifts in labour relations. Private property rights enabled the NP farmers to outperform the average farmers in the native reserves, not by substituting labour for capital but because the tenure regime allowed them to access additional labour needed to pursue land-intensive growth. In technical terms this can be described by using the concept of interlinked (or interlocking) factor markets. It is a concept that describes the simultaneous fixing of transactions in more than one market.⁹ It is often used to explain tenancy and sharecropping contracts, where for example transactions in land are connected to transactions in crops, labour services and/or credit. We argue that being able to buy relatively large tracts of land enabled the farming households in the NPAs to access a larger labour pool by attracting migrants – both relatives and non-relatives – to settle on the NP farms. In exchange for access to and use of land, the migrants provided labour services and/or crops to the farming households in the NPAs. These families became sharecroppers or labour tenants on the land of the NP farmers. In short, private property rights facilitated the rise of new labour relations that could be exploited by the NP farmers to their own benefit.

This article is structured as follows: we begin with a brief background of the rise of NPAs in Southern Rhodesia. The second section focuses on the performance of NP farmers. In the third section we then compare the production systems of an average NP farm with an average communal farm with a focus on capital and labour inputs. In the fourth section, we analyse the role of landed property rights for the success of NP farmers. We argue that these enabled the NP farmers to increase the total hours worked on the farm by attracting additional labourers. This allowed for increased output through a shift to a more labour-intensive mode of production. In the final section, we conclude that the success of the NP farmers stemmed from their ability to combine a shift upwards in the production function through technological change and improvements, with land intensification by adding more labour at diminishing rates of return.

The Rise of NPAs in Southern Rhodesia

The creation of the NPAs was a political move, which explicitly differentiated freehold among Africans from the same pattern of land tenure among white settlers. The original concept of the NPAs was due partly to political expediency and partly to the genuine belief by the government that selected African farmers could successfully be introduced to a system of freehold land tenure. African farming played an important role in the early colonial Southern Rhodesian economy. Africans were the main supplier of food to the growing mining and urban centres. By 1904, Africans produced more than 90 per cent of the country’s marketed output.¹⁰ During the First World War Southern Rhodesia became more isolated and the domestic market contracted. Despite these developments, sections of the African population continued to do relatively well. The Chief Native Commissioner (CNC) reported in 1919 that ‘although no figures are available showing the exact amount contributed by the native in indirect taxation through custom dues, the indications are that the importance of the native as a consumer of imported goods is increasing, and that there is a growing native demand for better class articles’.¹¹ In the CNC annual report of 1920, it was stated that in the more prosperous districts the enormous prices do not appear to have prevented the ‘natives’ from purchasing what they required. The Superintendent of Natives in Bulawayo wrote: ‘in the same district [Bubi] ploughs are reported to be almost universal; in Matobo there is now a plough for every twelve souls of the population; in Bulalima [Bulilima-Mangwe] fully 1,000 ploughs were purchased [representing] fully 100 per cent’.¹²

As the European farming sector grew in the inter-war period, concerns were raised among the white settlers that vibrant African commercial farming was threatening their further development.¹³ Unlike in most other parts of colonial Africa, Africans were not legally prevented from purchasing land. The Southern Rhodesia Order in Council (1898) gave Africans the right to own and purchase land anywhere in the colony. The majority of Africans who had been able to purchase land were of South African origin.¹⁴ As a matter of fact, the British South Africa Company early on tried to attract numbers of Mfengu (Fingo) from South Africa to settle in Southern Rhodesia by giving them the right to ten acres of freehold land in exchange for three months’ labour. The colonial authorities were less enthusiastic about granting native Africans in Southern Rhodesia the same opportunity to buy land – more than once Africans’ applications to buy crown land were turned down. However, the authorities could not prevent them from buying land owned by white settlers. In the early years of the colony, when settler farmers were struggling to establish profitable businesses and much of the land was controlled by absentee landlords, there were opportunities for a small group of relatively wealthy Africans to buy white-owned land. However, with the rapid growth of settler farming, a larger number of settlers became increasingly critical of and hostile to the freedom of Africans to buy land wherever they wanted. Up until this point, the number of Africans who had been able to purchase land was very small, but sections of the settler community as well as some colonial officials feared that that was about to change.¹⁵

Settler farmers began to pressurise the authorities to further tighten Africans’ access and right to landed property in the inter-war period. The colonial authorities, however, remained reluctant to consider the idea of complete institutional segregation of landholding rights between white settlers and Africans simply because the settlers depended on the labour provided by the Africans residing there as labour tenants.¹⁶ This is partly why the Land Commission of 1925 recommended a compromise, namely that specific portions of land, so-called Native Purchase Areas, should be set aside specifically for Africans who wished – and had the means – to buy agricultural land under freehold title.¹⁷ After debating the issue for nearly five years, the NPAs were formally recognised by the authorities under the Land Apportionment Act of 1930/31, which worked in conjunction with the Native Land Board.¹⁸ The Act, which legalised the division of the country’s land and water resources between black and white communities, made provisions among other things for wealthier Africans to buy land in the NPA.¹⁹ In total, 7.4 million acres were set aside as NPAs in the period 1931 to 1962 (see Figure 1) and by the end of the 1960s about two per cent of the African population were part of NP farming households.²⁰

Figure 1. Map showing the NPAs, Reserves and Crown Land in Southern Rhodesia, 1931–1962. (Source: Adapted from Native Department, reproduced by H.A. Cartography, April 2020.)

Figure 2. Number of Africans in NPAs in Southern Rhodesia, 1936–1960. (Source: PRO DO 64/25-88, Chief Native Commissioner Reports, 1936–1960.)

Purchasing land required access to capital. Unsurprisingly the group of African farmers buying land in the NPAs, therefore, tended to be over-represented by relatively wealthy Africans, such as chiefs’ families and successful businessmen.²¹ While settlers who bought land only required a deposit of five per cent and were allowed 19 years to repay, the Africans had to pay a deposit of ten per cent and pay the balance within ten to 15 years. According to Robin Palmer, this implied that the prospective farming household was required to have accumulated between £10 and £100 in cash, cattle or small stock, which at the time was substantial wealth. The applicant should also own farm implements, such as a cultivator and a harrow, not as a legal requirement but as an implied need.²² In Marirangwe, for example, the average cost of farms was level with that of white settler farms at 12 shillings per acre.²³ It was initially believed by the authorities that only a very small number of Africans of this middle-class category with abundant land resources relative to labour would have the resources or incentives to move to the NPAs.²⁴ The authorities seem to have been correct in their initial predictions, as the number of applicants remained low in the 1930s. According to Palmer and Mark Nyandoro, however, the lack of interest among Africans in buying land was not that they lacked the means.²⁵ Instead, the chief reason was that the allocated land was on average inferior, lacked reliable water sources and was located far from major markets.²⁶

By the end of the 1930s things were about to change. The number of NP farmers increased significantly and they bought up larger tracts of land. In 1940 there were 2,022 NP farms. By 1955 the number had increased to 9,244 while there were more than 5,000 Africans on the waiting list to become NPA farmers.²⁷ The calculated average NPA farm size was between 200 and 250 acres depending on the area in which it was located, but sometimes a farming household in the NPAs could own up to 400 acres.²⁸ By contrast, the average farm sizes in the reserves were approximately between three and 10 acres of land.²⁹

The Performance of NP Farmers

After the establishment of the Purchase Areas in 1930, 11 years passed before any official mention was made of agricultural practices and production in these areas.³⁰ In 1942 the Native Land Board noted in its annual report that only a third of the NP farming households had adopted improved agricultural techniques such as applying manure, compost or following a crop rotation plan. Erosion and overstocking were causing concern to some officials, but the tone of this report does not convey any serious anxiety.³¹

Subsequent reports voiced reservations about production techniques and output, but it was not until 1948 that the Native Land Board became more selective in its allocation of farmers, ‘discouraging applications by Natives with no agricultural background’.³² In 1953, recognised agricultural training became a prerequisite for all applicants for farms.³³ Such training itself was, however, hardly tailored to the requirements of managing a farm of 200 acres or more, since it was designed for the rural households in the reserves working up to 10 acres.³⁴ In addition, to become an NP farmer one now had not only to have enough financial means to buy land but also be in possession of £300 above the purchase price of the farm and a Master Farmer certificate.³⁵

While it may have been true that most of the selected NP farmers in the early years mainly used land as security rather than as a source to accumulate wealth, this had all changed by the 1950s. Table 1 shows the estimates of average maize yields in NPAs and native reserves in the 1950s. The estimates reveal that the farming households in the NPAs performed significantly better than the farmers in the reserves.

Table 1. Estimated maize yields per acre (203-lb bags) in native reserves and Native Purchase Areas (NPAs), 1951–1958

Year	Total production: native reserves (203-lb bags)	Average yields (bags/acre)	Total production: NPAs (203-lb bags)	Average yields (bags/acre)
1953	5,734,900	2.3	260,400	3.3
1954	5,798,225	2.2	294,213	4.1
1955	5,861,550	2.1	320,000	3.5
1956	5,924,875	2.0	474,254	4.7
1957	5,988,200	1.9	n/a	4.1
1958	6,051,527	1.8	430,278	3.5

Source: Andersson and Green, ‘Development Under the Surface’, p. 140.

We need to treat these estimates with a great deal of caution as they are averages. The performances varied significantly between the reserves and the NPAs and within both, although the limited number of estimates prevents us from studying these variations in any great detail. Estimates made by the colonial authorities in the Budjga NPA, for example, reveal high levels of income stratification in the 1950s among NPA farming households, with a range of between R$316 and R$2,750 (Rhodesian dollars) annually. Further, some farming households in the reserves clearly outperformed the average NP farming households in yields per unit land, just as the performance between NPA farming households varied.³⁶

In general, the quality of the soils and access to water resources was better in NPAs than in the reserves, but inferior to the regions where most white commercial farmers operated.³⁷ Not only variation in the environment mattered, but also access to markets. A recent study by Tawanda Chingozha and Dieter von Fintel found that farming households in the NPAs situated closer to main markets performed better, measured as more land under cultivation.³⁸

The director of Native Agriculture expressed enthusiasm about the performance of NP farming households in the 1950s in his confidential correspondence.³⁹ However, he also raised concerns that the productivity levels of the NP farms were below their potential and that the NP farming households had not fully embraced the extension education, which some African farmers resisted as a colonial prescription.⁴⁰ Other observers provide a more optimistic account of the farming methods applied by the farming households in the NPAs. Ian Scoones, for example, claims that at least some NP farms, especially in Mshagashe in Masvingo province operated as profitable commercial enterprises.⁴¹ There are also several reports that speak of a marked improvement in productivity in areas where there was ‘a swing towards cash crops and improved livestock management’ and in areas ‘where effective extension effort [was] applied’ compared to the reserves.⁴² In 1969 the colonial director of Native Affairs, R.M. Davies, was impressed by the performance of the farming households in the NPAs as they increasingly invested in more productive capacity. He observed that:

The gap between the high standards applying mainly to the European farming sector and the farming activities in the Purchase Area [was] being significantly narrowed … factors such as the lack of access to credit and the shortage of supervision and guidance which, in the past, seriously retarded the development of these areas [were] no longer obstacles to real progress.⁴³

To get a more in-depth understanding and picture of the differences between native purchase farms and the farms in the native reserves we can make use of Benton Massell and R.W.M. Johnson’s detailed comparison between the performances of the Mount Darwin NPA with Chiweshe Native Reserve in Mashonaland Central province in Southern Rhodesia.⁴⁴ One hundred and eighteen farms from five different villages in Chiweshe and 25 farms in Darwin were surveyed. During crop year 1960 to 1961 these farms were visited once per week and the surveyors collected vital farm-level statistics on output and input.⁴⁵ Comparing yields between only two districts must be read with a great deal of caution. It is by no means a representative sample. Darwin was, together with, for example, Makoni, Msengezi and Marirangwe, one of the more successful NPAs.⁴⁶ Still, given the limited data, the Massell and Johnson survey provides detailed information with regard to yield differences that point in the same direction as the more aggregate data one can find in the archival sources.

In 1960/1961 farming households in NPA Darwin produced nine times as much per farm, and sales were 150 times as high as those of the combined value of the production of maize, groundnuts and millet from the Chiweshe Native Reserve. Part of the gap in output is explained by differences in farm sizes, these being significantly larger in the NPAs. However, the yield gap remains significant after controlling for land size. It is estimated that NPA farmers were, on average, almost four times as productive (measured in the value of yields per acre) as master farmers in the Chiweshe Reserve.⁴⁷ Having said that, the yield gap between the two districts could as well be an outcome of different agro-ecological conditions. We have already noted that the quality of the soils in NPA Darwin was comparatively high. Unlike most Chiweshe Reserve was also in a fertile and favourable agro-climatic region where farmers planted their crops on sand and red loam soils, with higher yields recorded on red loam than on sand soils. Darwin NPA possessed four distinct soil types, namely brown sand, black cotton, clay and red loams. The average yield on each type varied but was not significantly higher than in Chiweshe. The difference in yields cannot be in the intrinsic fertility of soils, but in that the Chiweshe Reserve has been cultivated by a large population for a greater number of years, with inadequate soil conservation. By contrast, the farming households in NPA Darwin were settled more recently on ‘virgin soil’ that was more fertile than in Chiweshe, hence their greater productive capacity.⁴⁸ Population pressure and levels of soil exhaustion rather than the quality of the soils can explain part of the yield gap between the two districts.

All in all, the scattered sources point in the same direction. On average, the farming households in the NPAs produced substantially higher yields than those in the communal farms in the reserves. As shall be shown below, the NP farmers produced higher yields as they used both capital and labour more intensively. While the former is in line with our expectations the latter is somewhat surprising and requires further investigation, which we will return to in the final section.

Labour and Capital on NP Farms

An average and typical farming household in the NPAs benefited from land abundance and the relative abundance of capital, but faced, as their chief challenge, labour shortages. Population densities were significantly lower in the NPAs than in the reserves, and the very nature of how land was allocated in the NPAs meant that farming households were not only far apart but also that few NPAs lived side by side with close relatives or friends.⁴⁹ Sometimes there was competition for seasonal labour between the NPAs and white settler farmers, mainly during peak periods, which further reinforced the scarcity of labour.⁵⁰ Given the relative abundance of capital compared to farming households in the reserves the most obvious strategy for the NPA farming households would be to invest in labour-saving technologies to improve their land and increase output.

Generally, data on agricultural inputs among African farming households in Southern Rhodesia are scarce, to say the least. The farming households in the NPAs are no exception. Inputs on NPA farms were not systematically recorded in the statistics and we consequently have to rely on Massell and Johnson’s cross-sectional comparison between Chiweshe Native Reserve and Darwin Native Purchase Area. Table 2 shows the prevalence of the most common farm implements in the two areas in the mid 1960s.

Table 2. Farm implements in Darwin and Chiweshe

Implement	Average number per farm		Percentage of farms with at least one
	Chiweshe	Darwin	Chiweshe	Darwin
Single furrow plow	1.04	1.7	100	100
Disc plow	0	0.05	0	5
Ridge plow	0	0.05	0	5
Double furrow plow	0	0.05	0	5
Planter	0.02	0.7	2	70
Cultivator	0.77	1.6	83	100
Simple harrow	0.29	0.9	27	85
Disc harrow	0	0.05	0	5
Sheller	0.02	0.25	2	25
Scotch cart	0.21	0.85	25	85
Water cart	0	0.15	0	15
Mower	0	0.05	0	5
Tractor	0	0.1	0	5

Source: Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’, p. 18.

As Table 2 shows, farming households in the NPA Darwin used more sophisticated and more varied farming equipment compared to the surveyed farms in Chiweshe Reserve, which is in line with expectations. As discussed above, the NPA farming household belonged to a group of relatively wealthy Africans. Given this, the interesting question is not whether the farmers invested more capital in their farms, but whether they invested in order to save labour, the presumably scarce factor of production.

One way of looking into this is to compare working hours. Table 3 shows the total hours worked in the Darwin NPA and the reserve in Chiweshe. It reveals that the farming households in Darwin spent significantly more working hours compared to the farming households in Chiweshe. On average the farming households in Darwin spent five times as much time on farming activities. The significant difference in hours worked was, in part, an outcome of variations of landholding sizes. The NPA farms were much larger than the farms in the reserves. To control for that we compared hours worked per acre. The estimates are presented in Table 4.

Table 3. Total hours worked per crop in Chiweshe and Darwin

Crop	Chiweshe	Darwin
Maize	610	2,666
Groundnuts	310	1,833
Millet	187	587
Total	1,107	5,096

Source: Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’, p. 22.

Table 4. Hours worked per acre, crop and operation in Chiweshe and Darwin

Operation	Corn		Peanuts		Millet		Total
	Chiweshe	Darwin	Chiweshe	Darwin	Chiweshe	Darwin	Chiweshe	Darwin
Manure application	6.73	18.55		0	0	0	6.73	18.55
Plowing and planting	15.79	20.35	37.89	42.12	25.47	14.24	20.01	24.74
Weeding, transplanting and cultivating	34.49	66.25	64.79	75.66	79.37	77.14	43.62	70.07
Harvesting	19.87	68.17	98.22	215.76	61.06	109.37	36.45	107.81
Total	76.88	173.32	200.90	333.54	165.90	200.75	106.81	221.17

Source: Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’, p. 23.

Looking at hours worked per acre shows that farming households in Darwin still worked twice as many hours compared to the farming households in Chiweshe. The peak period for labour was between November to December when land was ploughed and sowed or planted and weeded, and again from April to May when crops were harvested. The gender division of labour was similar among NPA farming households to that found in the reserves. Men and boys were mostly responsible for ploughing. The men led the plough, while the boys guided the oxen. All members of the household were involved in planting and sowing, while women and girls were mainly in charge of weeding.⁵¹

Massell and Johnson’s survey further reveals that farmers in Chiweshe relied solely on family labour (see Table 5).⁵² Family labour, including child labour, constituted the main share of labour inputs among the farmers in Darwin NPA as well. The importance of child labour among NPA farming households is further supported in a survey conducted by the Ministry of Agriculture in the mid 1960s. In NPAs such as Mshagashe, Zowa and Chitomborwizi (Mashonaland West) and Chesa (Mount Darwin, Mashonaland Central) children were recorded as working more hours in tending crops and cattle than in any other labour category, their contribution accounting for 24 per cent and 35 per cent of the hours of work respectively.⁵³ However, all members of the family contributed in terms of labour, including the children; women in general spent more hours in the fields than men.⁵⁴

Table 5. Hours worked per farm by labour group and crop

Crop	Men		Women		Children		Hired labourers		Social
	Chiweshe	Darwin	Chiweshe	Darwin	Chiweshe	Darwin	Chiweshe	Darwin	Chiweshe	Darwin
Maize	226	1,103	287	1,252	191	265	0	156	0	44
Groundnuts	93	748	177	945	79	207	0	30	0	10
Millet	61	231	99	306	54	45	0	12	0	85
Total	380	2,082	563	2,503	324	517	0	198	0	85

Source: Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’, p. 22.

While family members constituted the main source of labour for the farming households in Darwin NPA, these farmers, in contrast to those in the Chiweshe Reserve, also employed hired labour and ‘social labour’ as Table 5 reveals. Massell and Johnson define the latter as all forms of additional labour accessed via social networks and not markets that is, reciprocal labour arrangements, help from friends and relatives. So-called work parties, where a farming household invited nearby farmers to a party in exchange for the labour service, were commonly applied during the peak seasons. Usually, 15 to 20 people would show up at a work party.⁵⁵

Hired and social labour, however, only accounted for about 3.8 per cent of the total hours worked, according to the survey. There are, however, reasons to believe that the survey underestimates the reliance on non-family labour among the NP farming rural households. Looking at the reported total hours worked by men and assuming that a farm in Darwin was managed by a nuclear family leads to unrealistic estimates. Dividing the total hours by 52 weeks shows that on average men worked on the farm for more than 40 hours per week. For women, the figure is even higher, that is, on average nearly 50 hours per week throughout the year. Given the seasonal fluctuations in labour demand and hours needed on domestic work and other agricultural-related activities like transport and marketing, we believe that Massell and Johnson overestimate the hours that the husband and wife spent on the farm.

Part of the gap can be explained by the prevalence of polygyny. Polygyny was more common among farming households in the NPAs than those in the reserves. Some men in the NPAs such as Nyahunda in Bikita (Masvingo province) were polygamous to allow for a larger pool of family labour, which would be able to do more work than a smaller family unit.⁵⁶ For example, 37 per cent of all farm owners in Guruuswa NPA were polygamists while in Mutadza NPA the share of polygamists was 43 per cent. Most polygamists had two or three wives, but there are cases where an NP farm owner had married more than five wives.⁵⁷ Polygamy can in part explain the number of hours of family labour spent on farming, but it is not the whole story. The majority of the NPA farming households were not polygamists and none of the surveyed farms in Chiweshe and Darwin was reported to consist of households of one husband and several wives. The recorded hours worked by men and women include labour by people outside the nuclear family constellation; it is likely that the NPA farming households were less inclined to report these workers because the colonial officials held that this group of non-NP farmers had no right to live in the NPAs.

That employment of non-family labour was common among NP farmers is confirmed by other sources. In the mid 1970s, Cheater estimated that 85 per cent of all farming households in Msengezi NPA employed non-family labour.⁵⁸ A majority were employed on short-term or casual contracts. Only 4.3 per cent of the farmers hired wage labourers on a permanent basis. Weeding and harvesting (which took only a few weeks) were the two most crucial tasks when additional labour was needed. For both tasks speed was more important than accuracy. Under these circumstances, casual labour was a more suitable form of labour than permanent wage labour, especially when the remuneration was directly related to the completion of a specific task, a form of labour known as piecework.⁵⁹

The picture that emerges is one of different labour relations among farming households in NPAs compared to farming households in the reserves. NPA farmers were to a larger degree dependent on non-family labour. Where did these labourers come from? On paper, NPA areas were only meant for Africans who had the means to buy land. In practice, the establishment of NP farms in the NPAs was followed by an increased inflow of non-NP households into these areas. We argue that having access to larger tracts of land enabled the farming households in the NPA to attract migrants, including relatives and non-relatives who would provide their labour service in exchange for access to land. Having access to this additional source of labour enabled the farming households in the NPA to increase yields by applying more labour to the land.

Private Property Rights, Interlinked Markets and the Success of NP Farmers

The establishment of NPAs created tension between those Africans who were able to purchase land and those Africans who were either evicted from newly established NPAs or lived in reserves near to an NPA. Scholars like Norma Kriger, Gerald Mazarire, Terence Ranger and Allison Shutt provide examples of the tense relationship between the Africans who were able to purchase land in the NPAs and the Africans in the reserves. In some cases, Africans were evicted from land that would become an NPA and forced to move into the reserves. In other cases, Africans lived in overpopulated reserves so their chances of accessing land in an NPA area were limited. Whether this was true or not, there was a widespread perception among Africans in the reserves that the farming households in the NPAs were siding with the colonial authorities. Occasionally, there were clashes between evicted Africans and the newly arrived NPA farmers.⁶⁰ At the same time, as will be shown below, the establishment of NPAs attracted an inflow of Africans from the reserves who became tied to the NPAs farming households as farm labourers and labour tenants.

When establishing the NPAs, the colonial authorities envisioned that the farmers who bought land would develop into a new class of ‘yeoman’ farmers referring to a group of independent small-scale farmers in 17th-century England and the 18th-century United States. These farmers practised land-intensive farming, through a combination of capital investments and use of nuclear family labour.⁶¹ It was envisaged that the farming households in the NPA areas would develop into ‘single family farms’ that combined rotation agriculture with pastoralism.⁶² This vision proved difficult to realise because the combined activities required either labour-saving technologies or more labour or both. In the early years (that is, the 1930s) many farming households in the NPAs faced serious economic problems. Instalment payments were delayed and many of the farmers had to keep their urban jobs in order to survive. The colonial authorities believed that the ‘failure’ of the farming households in the NPAs was because the selected NP farmers were not investing enough in farming activities. They criticised them for combining urban employment with farming instead of concentrating solely on farming.⁶³ On similar lines, historians have generally downplayed the economic performance of the farming households in the NPAs and argued that they bought land for political and social rather than economic reasons.⁶⁴ In part, this is to confuse cause and effect.

There are indicators that the challenges the farming households in the NPAs faced in the initial years were rooted in real economic problems. By the mid 1930s the Native Land Board began to realise that the call for intensive farming based on nuclear family labour was not realistic due to labour shortages. The latter implied that the selected NP farmers, instead of following the stipulated approach of using intensive crop rotation farming methods, continued with extensive methods cultivating only parcels of the most fertile land, taking advantage of differences in soil fertility by planting in patches.⁶⁵ Things were, however, about to change with the inflow of migrants into the NPAs, which allowed the farming households there to increase yields by utilising the labour services of migrants. Figure 2 shows the estimated population growth of the NPAs provided by Martin Andersson and Erik Green.⁶⁶ The total population was significantly larger than the number of selected purchase farmers. In 1957 it was estimated that 187,505 people were living in the NPAs of Southern Rhodesia. Of this number, approximately 160,000 were living without legal title. The increase in annual population growth in the 1940s cannot be explained by a natural growth in population. Instead, the figures reveal an inflow of people from other areas. More importantly, the figures indicate that the growth of the population in the NPA areas increased significantly in the 1940s, just as the farming households in the NPAs were recorded by the colonial authorities to be profitable.

From the mid 1940s, the farming households in the NPAs mainly tapped into a new social network for their labour needs in addition to relying on extended kin or men and women living nearby.⁶⁷ Since the mid 1940s their network extended to people who migrated from the southern parts of the country to settle and work for landholders, particularly those willing to offer land for work:⁶⁸ any labour that was appropriated by NP farming households from their dependants and kin, it would seem, had access to land as a form of remuneration. In freehold areas, it was quite common that only those who had close kinship links to landowners obtained land to work for themselves in return for labour.⁶⁹

Based on in-depth interviews with farming households in Marirangwe NPA, Shutt argues that immigrants to the NPA areas comprised two groups: renters and ‘squatters’.⁷⁰ Both groups came to the area with the long-term intention of buying or leasing land from the NPA farming households. The ‘squatters’ initially found a piece of land where they planted vegetables to sell in town that they combined with working on the NP farms. The money saved was later used to buy or rent land from the NPA farmers. The renters were in practice tenants who provided labour services to the farming households in the NPAs in exchange for access to land. In some cases, the contract was transformed as the renter/tenant managed to buy the land after a couple of years.⁷¹ Mazarire concludes that in a typical NPA ‘portions of surveyed and demarcated holdings existed side by side with pockets of clusters of local people living variously around these plots as “squatters”’.⁷²

The colonial authorities continuously raised concerns about what they regarded as illegal ‘squatting’ in the NPAs. The Native Land Board, after an inspection tour of the Mshagashe NPA in 1936, concluded that development in the area looked promising, but overstocking and ‘squatting’ was a potential threat to the success of NP farming.⁷³ In the 1967 annual report the Rural Land Board concluded that: ‘this squatter problem [in the Native Purchase Areas] [was] more dangerous than low productivity – the net result [was] illegal sub-division and reduction of the economic potential of the farm’.⁷⁴ The irony is that the constant inflow of people into the NPAs enabled the farming households to increase output by employing more people to till the land.

The picture that emerges is one of different labour relations among farming households in NPAs compared to farming households in the reserves. NPA farmers, to a larger degree, depended on non-family labour. Having access to this additional source of labour enabled the farming households in the NPA to increase yields by applying more labour to the land. This therefore specifically reveals and confirms the labour dynamics within the NPAs, linked to the space afforded for incorporating certain kinds of ‘migrants’. The private property rights that the farming households in the NPAs had claimed helped them to attract much-needed labour, by enabling the farmers to give portions of their allocated land to Africans who moved into the NPA areas in exchange for their labour; hence the success of these farmers. Theoretically, the private property rights enabled the rural farming households to exploit the opportunities of interlinked factor markets. The landed private property rights shaped socio-economic relations in a wider sense. It created an incentive for the emergence of various forms of non-family labour relations in the NPAs including sharecropping and wage labour. It was the effect that the landed property rights had on labour relations that can explain the relative success of the farming households in the NPAs and not – as mainstream economics assumes – that private property rights are more secure than alternative forms of property rights.

Conclusion

On the whole, this article poses questions about the role of property rights in agricultural development, using an exceptional case of land tenure reforms in rural Africa, namely the African NPAs in colonial Zimbabwe. The NPAs, first, differed from other titling programmes by giving the farmers complete freehold rights over land in locations outside customary jurisdiction. Second, to become a NP farmer one had to meet certain liquidity targets and have the means to buy the land. Finally, the NPA farmers did perform significantly better than the farmers working on customary land in the so-called native reserves.

At first glance, this seems to lend support to the view within mainstream economics that private property rights were a superior form of tenure rights as they enhanced security and incentivised increased capital investments in land among the property holders. However, analysing investment patterns among NP farming households reveals a different pattern. The chief difference between the NPA farming households and the farming households in the native reserves was not capital investment, but the amount of labour that was invested in the production process. The NPA farming households invested considerably more working time in farming. This could not have been done without access to non-family labour. Labour was scarcer in the NPAs than in the reserves, so it was necessary for the NP farming households to access non-family labour. It is from this perspective one has to understand the role of private property rights and markets in the NPAs. To explain this, one has to take account of the fact that changes in landed property rights altered social relations in the wider sense of the term. The private property rights enabled the emergence of various forms of non-family labour relations – including sharecropping, labour tenancy and wage labour – that the landowner could exploit to increase production.

Having complete ownership over relatively large tracts of land enabled the NP farming households to attract migrants to settle on their land under various forms of tenancy contracts. Thus, rather than just shifting the production function upwards through technological change/improvements, they combined this with land intensification by adding more labour at diminishing rates of return. Landed private property rights in that regard changed the economic and social landscape in the NPA in a more fundamental way. It led to a stricter division between those who controlled the means of production and those who only controlled their own labour. The NP farming households could exploit these divisions to their own benefit, which thus explains why NP farming households produced higher yields than the farmers in the reserves.

Acknowledgements

This article was presented at the 6th annual meeting of the African Economic History Network Workshop at the University of Sussex, Brighton, England, 21–22 October 2016 and at the European Social Science Conference, Belfast, 4–7 April 2018. We are grateful for comments we received from the audiences. The article has also benefited from feedback from Gareth Austin and Mandivamba Rukuni. It has been published in the African Economic History Working Paper Series, no. 62/2021.

Additional information

Notes on contributors

Erik Green

Erik Green Associate Professor/Senior Lecturer, Department of Economic History, Lund University, PO Box 7083, 220 07 Lund, Sweden; Fellow at LEAP, Department of Economics, Stellenbosch University, Private Bag X1, Matieland, 7602, South Africa. Email: [email protected] https://orcid.org/0000-0002-7805-4696

Mark Nyandoro

Mark Nyandoro Professor of Economic History/Senior Lecturer, Department of History, Heritage and Knowledge Systems, University of Zimbabwe, PO Box MP 167, Mount Pleasant, Harare, Zimbabwe; Extraordinary Professor Research, School of Social Sciences, Faculty of Humanities, North-West University, Vaal Triangle Campus, Building 11B, Office 029, Vanderbijlpark, South Africa. Email: [email protected] https://orcid.org/0000-0002-1716-0102

Notes

1 M. Nyandoro, ‘Land and Agrarian Policy in Colonial Zimbabwe: Reordering of African Society and Development in Sanyati, 1950–1966’, Historia, 64, 1 (2019), pp. 111–39.

2 Three notable exceptions are B.F. Massell and R.W.M. Johnson, ‘Economics of Smallholder Farming in Rhodesia: A Cross-Section Analysis of Two Areas’, supplement to volume VIII, Food Research Institute Studies in Agricultural Economics, Trade and Development (California, Stanford University, 1968); A.K.H. Weinrich, African Farmers in Rhodesia – Old and New Peasant Communities in Karangaland (Oxford, Oxford University Press, 1975) and N. Kriger Zimbabwe’s Guerrilla War: Peasant Voices (Cambridge, Cambridge University Press, 1992).

3 M. Nyandoro, ‘Development and Differentiation: The Case of TILCOR/ARDA Irrigation Activities in Sanyati (Zimbabwe), 1939 to 2000’ (PhD thesis, University of Pretoria, 2007); M. Nyandoro, ‘Development and Differentiation in the Post-Independence Era: Continuity or Change in ARDA-Sanyati Irrigation in Zimbabwe (1980–1990)’, African Historical Review, 41, 1 (July 2009), pp. 51–89; M. Nyandoro, ‘Emerging Smallholder Cotton Irrigation Agriculture and Tensions with Estate Labour Requirements in Sanyati, Zimbabwe, 1967–1990’, Journal of Southern African Studies, 48, 3 (2022), pp. 453–72.

4 See D.C. North and R.P. Thomas, The Rise of the Western World: A New Economic History (Cambridge, Cambridge University Press, 1973); D.C. North, Institutions, Institutional Change, and Economic Performance (Cambridge, Cambridge University Press, 1990); G. Feder and D. Feeny, ‘Land Tenure and Property Rights: Theory and Implications for Development Policy’, World Bank Economic Review, 5, 1 (1991), pp. 135–53; H. de Soto, The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else (London, Black Swan, 2000); D. Acemoglu, D.H. Autor and D. Lyle, ‘Women, War, and Wages: The Effect of Female Labor Supply on the Wage Structure at Mid-Century’, Journal of Political Economy, 112, 3 (2004), pp. 497–551. On key institutions enhancing property rights and agricultural intensification, see also R. Bubb, ‘The Evolution of Property Rights: State Law or Informal Norms?’ The Journal of Law and Economics, 56, 3 (August 2013), pp. 555–94; M. Goldstein and C. Udry, ‘The Profits of Power: Land Rights and Agricultural Investment in Ghana’, Journal of Political Economy, 116, 6 (February 2008), pp. 981–1022.

5 For example, D.A. Atwood, ‘Land Registration in Africa: The Impact on Agricultural Production’, World Development, 18, 5 (1990), pp. 659–71; J. Ensminger, ‘Changing Property Rights: Reconciling Formal and Informal Rights to Land in Africa’, in J.N. Drobak and J.V.C. Nye (eds), The Frontiers of the New Institutional Economics (San Diego, Academic Press, 1997), pp. 165–96; J.P. Platteau, ‘The Gradual Erosion of the Social Security Function of Customary Land Tenure Arrangements in Lineage-Based Societies’, United Nations University/World Institute for Development Economics Research (Insurance Against Poverty Project), Discussion Paper no. 26 (2002); P.E. Peters, ‘Conflicts over Land and Threats to Customary Tenure in Africa’, African Affairs, 112, 449 (2013), pp. 543–62.

6 B.N. Floyd, ‘Land Apportionment in Southern Rhodesia’, Geographical Review, 52, 4 (1962), pp. 566–82; W. Roder, ‘The Division of Land Resources in Southern Rhodesia’, Annals of the Association of American Geographers, 54, 1 (1964), pp. 41–52.

7 G. Austin, ‘Resources, Techniques, and Strategies South of the Sahara: Revising the Factor Endowments Perspective on African Economic Development, 1500–2000’, The Economic History Review, 61, 3 (2008), pp. 587–624. James Fenske, ‘Land Abundance and Economic Institutions: Egba Land and Slavery, 1830–1914’, The Economic History Review, 65, 2 (2012), pp. 527–55.

8 E. Green, ‘State-led Agricultural Intensification and Labour Relations: The Case of Lilongwe Land Development Program in Malawi, 1968–1981’, International Review of Social History, 55, 3 (2010), pp. 413–46; P. Woodhouse, ‘African Enclosures: A Default Mode of Development’, World Development, 31, 10 (2003), pp. 1705–20; A.O. Chimhowu and P. Woodhouse, ‘Customary vs Private Property Rights? Dynamics and Trajectories of Vernacular Land Markets in Sub-Saharan Africa’, Journal of Agrarian Change, 6, 3 (2006), pp. 346–71; P. Peters, ‘Inequality and Social Conflict Over Land in Africa’, Journal of Agrarian Change, 4, 3 (2004), pp. 269–314; J.P. Colin and P. Woodhouse, ‘Interpreting Land Markets in Africa’, Africa, 80, 1 (2010), pp. 1–13; E. Green and M. Norberg, ‘Traditional Landholding Certificates in Zambia: Preventing or Reinforcing Commodification and Inequality’, Journal of Southern African Studies, 44, 4 (2018), pp. 613–28.

9 F. Ellis, ‘Household Strategies and Rural Livelihood Diversification’, Journal of Development Studies, 35, 1 (1998), pp. 1–38; see also G. Feder and R. Noronha, ‘Land Rights Systems and Agricultural Development in Sub-Saharan Africa’, World Bank Research Observer, 2, 2 (1987), pp. 143–69.

10 G. Arrighi, ‘The Political Economy of Rhodesia’, New Left Review, 1, 39 (1966), pp. 35–65; I.R. Phimister, An Economic and Social History of Zimbabwe, 1890–1948: Capital Accumulation and Class Struggle (London, Longman, 1988).

11 Cited in M. Andersson and E. Green, ‘Development Under the Surface: The Rise of Indigenous Agriculture as an Unintended Consequence of Settler Institutions in Southern Rhodesia, 1896–1962’, Journal of International Development, 28, 1 (2016), pp. 135, 127–46.

12 Cited in Andersson and Green, ‘Development Under the Surface’, p. 135.

13 Ibid. See also M. Nyandoro, ‘Development and Differentiation’.

14 J. Mujere, ‘Land, Gender and Inheritance Disputes Among the Basotho in the Dewure Purchase Areas, Colonial Zimbabwe’, South African Historical Journal, 66, 4 (2014), pp. 699–716.

15 J. Mujere and A. Mseba, ‘The Politics of African Freehold Land Ownership in Early Colonial Zimbabwe’, African Economic History, 41, 1 (2019), pp. 32–53.

16 E. Green, ‘Development of Settler Agriculture in British Africa Revisited: Estimating the Role of Tenant Labour in Southern Rhodesia, c.1900–1960’, African Economic History Working Paper Series, 29, 2016, pp. 1–32; Nyandoro, ‘Land and Agrarian Policy in Colonial Zimbabwe’; Nyandoro, ‘Emerging Smallholder Cotton Irrigation Agriculture’.

17 H.V. Moyana, The Political Economy of Land in Zimbabwe (Gweru, Mambo Press, 1984); B. Floyd, ‘Land Apportionment in Southern Rhodesia’, in R. Prothero (ed.), People and Land in Africa South of the Sahara (Oxford, Oxford University Press, 1972); L.H. Gann, ‘The Southern Rhodesian Land Apportionment Act, 1930: An Essay in Trusteeship’, The National Archives of Rhodesia and Nyasaland, Occasional Paper, 1 (1963); M.L. Rifkind, ‘Land Apportionment in Perspective’, Rhodesian History, 3 (1972), pp. 33–62; M. Nyandoro, ‘Zimbabwe’s Land Struggles and Land Rights in Historical Perspective – The Case of Gowe-Sanyati Irrigation (1950–2000)’, Historia, 57, 2 (2012), pp. 298–349.

18 The Native Land Board was an Advisory Board of the Native Affairs Department. It recommended land prices and instituted the necessary land surveys. The Board was empowered to prescribe conditions of tenure in the Native Area, processed individual applications for land and also placed African farmers on holdings.

19 A.K. Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia, c.1931–1952’, The International Journal of African Historical Studies, 30, 3 (1997), pp. 555–81; Nyandoro, ‘Land and Agrarian Policy in Colonial Zimbabwe’.

20 Kriger, Guerrilla War.

21 O.B. Pollak, ‘Black Farmers and White Politics in Rhodesia’, African Affairs, 74, 296 (1975), p. 265.

22 R.H. Palmer, Land and Racial Domination in Rhodesia (London, Heinemann, 1977), pp. 213–14.

23 Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’.

24 E. Punt, ‘The Development of African Agriculture in Southern Rhodesia with Particular Reference to the Interwar Years’ (MA thesis, University of Natal, 1979); Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’; Nyandoro, ‘Development and Differentiation’.

25 Palmer, Land and Racial Domination in Rhodesia; Nyandoro, ‘Development and Differentiation’.

26 See also T. Chingozha and D. von Fintel, ‘The Complementarity Between Property Rights and Market Access for Crop Cultivation in Southern Rhodesia: Evidence from Historical Satellite Data’, Economic History of Developing Regions, 34, 2 (2019), pp. 132–55.

27 National Archives of Zimbabwe (Records Centre) hereafter NAZ (RC), Purchase Areas GEN., Volume 4, APL. 69–SEPT.70, Box 135183, Location 4.11.7R, File BJA – Budjga, Volume 1, August 1964 – October 1970, ‘Native Purchase Area Category: Lease in terms of Section 39 of the Rural Land Act, 1963: Crown Land Holding Budjga 87 Mtoko District: 151.9626 acres’, for Director of Lands (Office of the Director of Lands) to Mr Amos (RC No. X 22076 Mtoko), Ref. No. Budjga 87, 7 March 1966.

28 A.P. Cheater, Idioms of Accumulation: Rural Development and Class Formation among Freeholders in Modern Zimbabwe (Gweru, Mambo Press, 1984), p. 39; R.W.M. Johnson, African Agricultural Development in Southern Rhodesia, 1945–60 (California, Stanford University Food Research Institute, 1964), p. 216. NAZ (RC), Purchase Areas GEN., Volume 4, APL. 69–SEPT. 70, Box 135183, Location 4.11.7R, File BJA, Budjga, Volume 1, August 1964–October 1970, ‘Native Purchase Area Category: Lease in terms of Section 39 of the Rural Land Act, 1963: Crown Land Holding Budjga 87 Mtoko District: 151.9626 acres’, for Director of Lands (Office of the Director of Lands) to Mr Amos (RC No. X 22076 Mtoko), Ref. No. Budjga 87, 7 March 1966.

29 V.E.M. Machingaidze, ‘Agrarian Change from Above: The Southern Rhodesia Native Land Husbandry Act and African Response’, International Journal of African Historical Studies, 24, 3 (1991), pp. 557–88; Nyandoro, ‘Development and Differentiation’.

30 Cheater, Idioms of Accumulation, p. 7.

31 Ibid.

32 Ibid., pp. 7–8.

33 Ibid., p. 8.

34 Cheater, Idioms of Accumulation; M. Rukuni, ‘The Development of Zimbabwe’s Agriculture: 1890–1990’, Working Paper AEE 7/90, Department of Agricultural Economics and Extension, Harare, Faculty of Agriculture, University of Zimbabwe, 1990.

35 W.R. Duggan, ‘The Native Land Husbandry Act of 1951 and the Rural African Middle Class of Southern Rhodesia’, African Affairs, 79, 315 (1980), pp. 227–39. For approval of NPA land allocation see, for example, NAZ (RC), Ministry of Internal Affairs, Box 158077, Location C19.2.10R, File: LAN 8, 1961–1964, ‘Holdings Applications and Approvals’, G.A. Barlow (NC Gatooma) to the PNC Mashonaland West, 29 June 1959; NAZ (RC), Ministry of Internal Affairs, Box 158088, Location C19.6.8R, File: PTR 2 October 1951 – October 1973, ‘Appendix “A” Crop Production: Gokwe TTL – Copper Queen and Chenjiri Purchase Areas, 1973’; and Nyandoro, ‘Development and Differentiation’.

36 Kriger, Zimbabwe’s Guerrilla War; A.K.H. Weinrich, African Farmers.

37 Nyandoro, ‘Development and Differentiation’.

38 Chingozha and von Fintel, ‘The Complementarity Between Property Rights and Market Access’.

39 NAZ, S160 DMN 2/9/51 – Confidential Correspondence, ‘Ministerial Economic Co-ordinating Committee’, Director of Native Agriculture to The Assistant Secretary, Native Economic Development, 30 March 1953, pp. 1–2.

40 Ibid.; H. Dunlop, ‘Efficiency Criteria in Primary Marketing: An Analysis of African Marketing Policies in Rhodesia’, Rhodesian Journal of Economics, 4, 3 (1970), pp. 10–20; A.F. Hunt, ‘Native Purchase Area Farms: An Economic Appraisal’ (Salisbury, Department of Native Economics and Markets, 1960); R.W.M. Johnson, African Agricultural Development in Southern Rhodesia, 1945–60 (California, Stanford University Food Research Institute, 1964); D.G. Matondo, An Analysis of Some African Purchase Area Farms: Physical and Financial Records, 1964–65 and 1966–67 (Salisbury, Economics and Markets Branch, Ministry of Agriculture, 1970); M.G. Paraiwa, An Analysis of Some African Purchase Area Farms’ Physical and Financial Records, 1964–1965 and 1965–1966 (Salisbury, Ministry of Agriculture, Economics and Markets Branch, 1970); M.G. Paraiwa, A Two-Year Comparison of Agricultural Production in African Purchase Areas: 1968/69 and 1969/70 (Salisbury, Rhodesia, Ministry of Agriculture, 1972); Nyandoro, ‘Development and Differentiation’.

41 I. Scoones, B. Mavedzenge and F. Murimbarimba, ‘Medium-Scale Commercial Farms in Africa: The Experience of the “Native Purchase Areas” in Zimbabwe’, Africa, 88, 3 (2018), pp. 597–619.

42 NAZ (RC), Purchase Areas ICG5, volume 3, no. 2, ‘Report of the Rural Land Board (RLB) Committee on Productivity Problems in the Native Purchase Areas’, Box 129755, Location 28.2.8R, 1969–1971, p. 1.

43 R.M. Davies, ‘Farming Progress in the African Purchase Areas’, NADA, X, 1 (1969), pp. 21–2.

44 Mount Darwin district contains two NPAs – Chesa and Karuyana – but here we simply refer to both the NPAs as ‘Darwin’.

45 Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’.

46 See T.O. Ranger, Peasant Consciousness and Guerrilla War in Zimbabwe: A Comparative Study (London, James Currey, 1985); Cheater, Idioms of Accumulation; Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’; A.K. Shutt, ‘We are the Best Poor Farmers: Purchase Area Farmers and Economic Differentiation in Southern Rhodesia, 1925–1980’ (PhD thesis, University of California, 1995).

47 Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’.

48 Ibid.

49 Kriger, Zimbabwe’s Guerrilla War; Weinrich, African Farmers.

50 NAZ, S160, DB 104/1, LDO’s Report for the Month of October 1950, Bikita District, 1950.

51 Weinrich, African Farmers.

52 Massell and Johnson, ‘Economics of Smallholder Farming in Rhodesia’.

53 NAZ, RG-3/AGR-5, no. 20453, An Analysis of Some African Purchase Area Farms’ Physical and Financial Records, 1964/65 and 1965/66, Economics and Markets Branch, Ministry of Agriculture, August 1970, p. 5. Besides children, NP labour categories mainly included the farmer/manager, wife/wives, adult males, adult females, regular labour and others.

54 Interview with Mandivamba Rukuni, director, Barefoot Education for Africa Trust (BEAT), Marlborough, Harare, 17 September 2015; telephone interview with Mandivamba Rukuni, director, Barefoot Education for Africa Trust (BEAT), Harare, 12 April 2020; interview with Caiphas T. Nziramasanga, curriculum and Social Studies education specialist, University of Zimbabwe, Teaching and Learning Centre, Mt Pleasant, Harare, 16 January 2017. All interviews were conducted by Mark Nyandoro.

55 NAZ, RG-3/AGR-5, no. 20453, An Analysis of Some African Purchase Area Farms’ Physical and Financial Records, 1964/65 and 1965/66.

56 Interview with Rukuni, 17 September 2015 and interview with Rukuni, 12 April 2020.

57 Weinrich, African Farmers.

58 Cheater, Idioms of Accumulation.

59 Ibid.

60 Kriger, Zimbabwe’s Guerrilla War; G.C. Mazarire, ‘A Social and Political History of Chishanga: South-Central Zimbabwe, c. 1750–2000’ (PhD thesis, University of Zimbabwe, 2009); Ranger, Peasant Consciousness; Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’.

61 R.C. Allen, Enclosure and the Yeoman (Oxford, Clarendon Press, 1992); R.C. Allen, ‘Economic Structure and Agricultural Productivity in Europe, 1300–1800’, European Review of Economic History, 4, 1 (2000), pp. 1–25.

62 A.K. Shutt, ‘Pioneer Farmers and Family Dynasties in Marirangwe Purchase Area, Colonial Zimbabwe, 1931–1947’, African Studies Review, 43, 3 (2000) pp. 59–80.

63 Ibid., p. 64.

64 See, for example, Arrighi, ‘The Political Economy of Rhodesia’; K. Good, ‘The Direction of Agricultural Development in Zambia, Zimbabwe and Malawi’, in Z.A. Konczacki, J.L. Parpart and T.M. Shaw (eds), Studies in the Economic History of Southern Africa. Volume 1: The Front Line States (London, Frank Cass & Co, 1990), pp. 127–58; Palmer, Land and Racial Domination in Rhodesia; Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’; Shutt, ‘Pioneer Farmers and Family Dynasties in Marirangwe Purchase Area’; M.O. West, The Rise of an African Middle Class: Colonial Zimbabwe 1898–1965 (Bloomington, Indiana University Press, 2002).

65 Shutt, ‘Pioneer Farmers and Family Dynasties in Marirangwe Purchase Area’, pp. 68–9; A.K. Shutt, ‘Squatters, Land Sales and Intensification in Marirangwe Purchase Area, Colonial Zimbabwe, 1931–65’, Journal of African History, 43, 3 (2002), pp. 473–98; Scoones et al., ‘Medium-Scale Commercial Farms in Africa’, p. 602.

66 Andersson and Green, ‘Development Under the Surface’.

67 Shutt, ‘Squatters, Land Sales and Intensification in Marirangwe Purchase Area’.

68 Ibid.

69 Cheater, Idioms of Accumulation.

70 Shutt, ‘Squatters, Land Sales and Intensification in Marirangwe Purchase Area’. The epithet ‘squatter’ was initially applied, by Government officials, to all residents on NP farms who were not members of the landholders’ immediate families or known dependents. Over the years, this usage was extended to all adults to whom the landholder granted rights of cultivation on his farm, including his sons, whether married or not. However, NP freeholders resented or deplored the fact that their children and dependents were called ‘squatters’ who had to be contracted on the farm as paid labourers without which they would have to face immediate eviction; see A.P. Cheater, ‘Formal and Informal Rights to Land in Zimbabwe’s Black Freehold Areas: A Case-Study from Msengezi’, Africa: Journal of the International African Institute, 52, 3 (1982), pp. 79, 77–91; see also Nyandoro, ‘Zimbabwe’s Land Struggles and Land Rights in Historical Perspective’, p. 318.

71 Shutt, ‘We are the Best Poor Farmers’; Shutt, ‘Purchase Area Farmers and the Middle Class of Southern Rhodesia’; Scoones et al., ‘Medium-Scale Commercial Farms in Africa’.

72 Mazarire, ‘A Social and Political History of Chishanga’, p. 212.

73 W.B. Mazobere, ‘The Native Purchase Areas: A Success in the Development of African Farming? A Case Study of Mshagashe Purchase Area’ (BA Honours dissertation, University of Zimbabwe, 1985), p. 30.

74 Annual Report quoted in Mazobere, ‘The Native Purchase Areas’, p. 20.