8—
The Economic Development of the Homelands

South Africa's modern economic history began with the mineral discoveries of the mid-nineteenth century. The exploitation of diamonds, gold, and other minerals offered substantial employment outside agriculture to whites and blacks for the first time. New transport networks linked port cities to growing inland centers, and blacks were drawn to the towns to work in mining, commerce, and industry, and on white-owned farms. However, this new economic growth produced a pattern of infrastructural development, population settlement, and location of employment that for the most part bypassed the African reserves. The economies of the reserves became dependent on the white-dominated core economy and increasingly lagged behind the rest of South Africa. Predominantly pastoral and agricultural, the reserves offered only a drastically restricted range of opportunities for economic security and achievement for their peoples. In explicitly dedicating its policies to the development of these backward areas, the Republic is undertaking an immensely complex task of economic reconstruction as well as trying to alter an historic pattern sustained by a variety of powerful interests.

Agriculture

The forces that have shaped the growth of agriculture in the Republic over the last century have had little positive effect upon the traditional subsistence economy of the homelands. As South African society became more urban and industrial, the demand for food and raw materials grew. Tariffs protected the home market and export subsidies opened the way for larger overseas sales of maize, sugar, wine, wool, and dairy goods. Sources of credit were provided by the Land and Agricultural Bank of South Africa, created in 1912, and by the Cooperatives Act of 1922. The extension of railways and roads brought isolated farmsteads into closer contact with city markets, sources of supplies, and technical and price information. As a result the productivity and profitability of agriculture in the white areas was increased

and farming became a commercial undertaking.^[1] In the African reserves, however, these changes had little impact because blacks were barred from participating in most of the government's programs and because the systems of land and labor use in the homeland agriculture were geared, not to money gain, but to the subsistence production of food — to be supplemented by wage income when necessary — and the attainment of a measure of security in the form of a family plot and a few head of livestock grazing on common pasture.

As in other parts of the economy, Africans responded to the constrained opportunities that were open and contributed to South Africa's agricultural growth as laborers, tenants, foremen, and managers, and, before being barred in 1913 from land ownership and working "on halves," Africans were successful sharecroppers and owners.^[2] Today about 30 percent of the African population lives and works on white farms, and this exposure, added to the background of decades of participation in a modernizing South African agriculture, makes the stagnation — actually the retrogression — of the homeland's primary sector all the more striking. A number of farms in white areas are occupied and run by blacks for white absentee owners. The backwardness of homeland agriculture must therefore be explained by the unique structure of that sector, not by an inherent unresponsiveness or disinterest on the part of Africans.

A major reason for the agricultural underdevelopment of the homelands is the long-standing neglect of the needs of black farmers by the Union and Republican governments, a neglect that contrasts sharply with the zealous attention devoted to white South African farmers. After 1910 there were enough problems arising from the changing structure of white agriculture, and in the related rapid exodus of poor whites to the cities, to occupy the Union government and turn its attention away from the politically impotent reserves. Yet, many of these stresses were felt as keenly, or more keenly, in the reserves. Growing populations and redundant farm families, overgrazing and improper burning of pasturage, recurrent droughts, raging epidemics of rinderpest and other diseases, wide swings in prices, and the vicissitudes of war and depression afflicted South African farmers of all colors. Agricultural education and extension services were slow in developing, at first even in white areas, and poor cropping practices abused the soil and its cover. Erosion became a national affliction. The stabilization of tenure in 1913 and the measures taken over the next several decades to ensure a supply of cheap black labor to white farms, in combination with measures to provide better

[1] For a description of agricultural development in South Africa after 1866, see Francis Wilson, "Farming, 1866–1966" in Monica Wilson and Leonard Thompson (eds.), Oxford History of South Africa (Oxford, 1971), II, 104–171.

[2] Colin Bundy, "The Emergence and Decline of a South African Peasantry," African Affairs, LXXI (1972) 269–288.

technologies, prices, and open marketing channels, pushed white agriculture down an uneven road to progress. Farmers in the reserves were left to their own devices to try to live as best they could with a set of agrarian conditions inappropriate to growth. Ever larger numbers of men and their livestock had to be absorbed on steadily shrinking and depleting allotments of farmland and pasturage.

By 1929 the Union government had become alarmed enough about the deteriorating agricultural conditions in the African reserves to establish a Native Agricultural and Lands Branch within the Department of Native Affairs. Having an extremely limited budget and an equally narrow range of responsibilities, this branch focused its activities on soil conservation and the stabilization of grazing. After the South African Native Trust was formed in 1936, most of the agricultural branch's attention was directed to the newly acquired white farmlands, with the hope that these tracts could be preserved until resources for development became available.^[3] Some very limited extension work was essayed, including herd culling, stock dipping, and inoculation. Quality breeding stock was provided. There were scattered demonstration projects and sales of subsidized implements. After the interruption of World War II, conservation efforts were renewed, but progress throughout the 1950s and 1960s was slow since the program was confined largely to trying to increase the proportion of land brought under strict land-use management. Ideally at least, this "physical planning" was intended to limit herd sizes, set up a rotational grazing plan, and allocate house and garden plots to the residents of rural locations.

The state of agriculture and animal husbandry in the reserves was not affected to any great degree by these programs, patently insufficient as they were in relation to the magnitude of the task. They were also unimaginative and unattuned to African economic and social conditions. In the managed or planned areas, people had to move to new housing, change the sites of their fields, and practice unfamiliar rotational grazing on fenced pasturage. There was no particular gain to an individual from doing any of these things—indeed, there were private gains in violating the rules—and the program was never adequately explained to the majority of Africans. Resistance to relocation sometimes became violent, but most often took the form of open gates, cut fences, and disregard for herd size guidelines. Even if the program had been enthusiastically adopted, it would have done nothing to raise farm or pasturage productivity since no complementary extension

[3] Summary of Report, 74–75. Since 1948 there has been an extensive renaming of institutions and legislation, principally by the substitution of "Bantu" for "Native." Thus the South African Native Trust is now the South African Bantu Trust, and the Native Agricultural and Lands Branch of the Department of Native Affairs is now the Agricultural and Development Branch of the Department of Bantu Administration and Development.

advice, credit, new technologies, or other measures were forthcoming. The Tomlinson Commission reported:

[It] is clear that the existing land settlement policy in the betterment areas of the Reserves brings no alleviation of the agricultural poverty of the Bantu, and that it remains necessary for the family head and, possibly, for other members of the family as well, to sell their labour outside the Reserves in order to provide for the needs of the family.^[4]

Output and Yields

By everyone's admission homeland crop yields and stock turnover rates remain low in the mid-1970s and renewed efforts to grow more food, develop commercial crops, and increase livestock productivity have had only spotty success. The homeland governments, the Bantu Investment Corporation, and the Agricultural and Development Branch of the Department of Bantu Administration and Development have not yet overcome the inertia of the agricultural sector.

Maize accounts for about 70 percent of the output of food grains, by weight, in Bophuthatswana, and about 80 percent in KwaZulu. In the former homeland, the remainder of production is divided between sorghum and dry or irrigated wheat. In the latter, sorghum is almost exclusively the alternative food grain. There are negligible amounts of cowpeas, beans, peanuts, potatoes, and millet grown.^[5] Table 8.1 shows the trends of maize output, area, and average yields for Bophuthatswana, KwaZulu, and all the homelands. the quality of the data is poor, but differences among the homelands and over time are probably in the right directions even if the precise numerical values are subject to a margin of error.^[6] It should be recognized, too, that averages are highly misleading. Often the best cultivators, and particularly those with access to water and better land, do much better

[4] Ibid., 77.

[5] These proportions are calculated from data available in Republic of South Africa, Department of Bantu Administration and Development, Annual Report, 1970, 32, 99, and Republic of South Africa, Bophuthatswana, Department of Agriculture, Annual Report, 1972, 15–22. It is unlikely that there has been much change.

[6] Agricultural statistics for herds and crops are assembled from the estimates of regional field personnel. Although some do this chore conscientiously, others are more casual, and all are busy with other duties. In the data there are unwarranted fluctuations in crop output or herd sizes from year to year, and some five-figure numbers do not change at all. Private studies have found considerable errors in official local crop, herd, and human population statistics, but they are too few to permit conjectures about general biases. The shortcomings of African herd and crop statistics are well known to the government, and rapid improvement in their quality is universally desired so that they can be used for planning and evaluation. Small-area sampling could be used to provide additional information and to correct for systematic biases in the aggregate data.

than average. Also, larger scale operations organized on the European model, and possibly managed by a government-paid white overseer, have high yields. Dry-cropped maize and sorghum are very susceptible to variations in rainfall. This factor further complicates the ascertainment of trends in yields.

The data suggest that from 1968 through 1972 Bophuthatswana devoted an increasing area to maize, eventually almost tripling cropping. Output rose by over five times, so that average yields almost doubled. There was a sharp fall off in plantings in 1973, apparently due to inadequate rainfall. In 1974 the area sown rebounded to 60 percent of the 1972 peak, but output was higher and yields were substantially above previous levels. A speculative explanation is that the larger government and private farms and the better small holder areas were drawn back into production and received enough fertilizer and other inputs to raise yields. More marginal areas had not recovered from the previous drought and were not replanted. Thus these gains were highly localized.

Only about 6 to 7 percent of the land area of Bophuthatswana is arable, but in KwaZulu a much larger share of the land can be used for perennial farming. In recent years, KwaZulu's cultivators have planted in maize an area four to ten times the size of the similarly planted area in Bophuthatswana. Most fields are quite small, some only garden plots, others amounting to a few acres. But the aggregate output of these household plots and small peasant holdings is a major determinant of nutritional and economic welfare in the homeland. Weather is more predictable than in Bophuthatswana, and the area planted in maize has fluctuated between 232,000 and 301,000 hectares. Output more than doubled between 1968 and 1972 and yields rose steadily, until the decline of 1973. Some notion of the benefits of irrigation can be obtained from the high yields obtained on watered lands, although it is also obvious how small a fraction (about 0.5 percent) of the arable area is presently irrigated. On white farms, depending on the area, maize yields are 1,600 to 2,300 kilograms per hectare, six to seven times average homeland yields.

The longer time series available for all the homelands shows some tendency for yields to rise, especially after 1965. The stabilizing and productivity-enhancing effects of irrigation are again readily apparent. Bophuthatswana's yields are higher than those of KwaZulu's, which are only about average for all homelands. This is not a sign of agricultural strength, but rather of the fact that intensive farming is practiced in those few areas where conditions are most favorable.

There has been considerable speculation about agricultural trends in the homelands. The Tomlinson Commission estimated that yields averaged about 264 kilograms per hectare in fiscal 1949/50, this amount being exceeded by

any margin only in 1971 and 1972 among the years covered by table 8.1.^[7] Allowing for regional differences and the effects of the weather, it appears safe to conclude that yields fell unevenly until the mid-1960s.

Since that time there may have been some improvement. This improvement has come about, however, under favorable conditions such as are found on stabilized Bantu Trust farms, in irrigated garden schemes, and where a few African farmers have had close ties to homeland agricultural extension officers. In Bophuthatswana, for instance, one-third of the crop comes from the Ditsobotla district, south of Mafeking. In those areas yields are three to five times those of other districts.^[8] These higher yields are encouraging because they indicate what is feasible given a solid resource base, modern inputs, and extension services. But since they also imply that most women who have household gardens, and many other small farmers, are probably not participating in the gains, they also measure the failure to universalize agricultural reforms. As is generally the case in underdeveloped areas, if all regions and farmers could attain the standards set by their most productive neighbors, the problem of realizing agricultural growth would be solved.

There are projects underway in both homelands to broaden the agricultural base by developing commercial crops. Following a recommendation of the Tomlinson Commission, fibre crops have been introduced. There are three small sisal projects in Bophuthatswana, covering about 1,200 hectares; the yield in 1972 was 212 tons of fibre, up from 130 tons in 1970. A canal irrigation project at Taung covers almost 1,600 hectares, and there are other irrigated areas in the Kuruman, Thaba 'Nchu, and Lehurutse districts.^[9] Vegetables, lucerne, and root crops are produced for use and sale. Phorium tenax (flax) is grown in KwaZulu on 3,551 hectares. Coconuts, cashew nuts, coffee, and tea are in experimental stages. Irrigated areas are used for gardens of specialty crops. There are also plans to utilize small reservoirs as fish farms in KwaZulu; there is an established fish farm at the Klipvoor Dam near Mabopane and GaRankuwa in Bophuthatswana.

Since its agricultural division was formed in 1973, the Bantu Investment Corporation has begun to assume responsibility for large-scale projects, such as plantations, irrigation works, and trust farms, that require capital investment and management. Most of the new commercial crop projects have been transferred to the corporation from the agriculture departments of Bophuthatswana and KwaZulu. Also, the corporation has assumed control of the Taung

[7] Summary of the Report , 84–85; also see J. A. Lombard and P. J. van der Merwe, "Central Problems of the Economic Development of the Bantu Homelands," Finance and Trade Review , X (1972), 37; Muriel Horrell, The African Homelands of South Africa (Johannesburg, 1973), 86.

[8] Republic of South Africa, Bophuthatswana, Department of Agriculture, Annual Report, 1972 , 16.

[9] Ibid. , tables on irrigated land production.

irrigation project, where a feed lot will be established to use fodder production for cattle and to fatten cattle sold through a stock sale scheme. A similarly integrated plan is being established for KwaZulu, where the agricultural division of the corporation has taken over the long-established project now embracing the Tugela and Ferry estates, where cotton, citrus, vegetables, and fodder crops are grown experimentally. The plan is to establish market outlets in homeland townships for the produce of these farm plots. In KwaZulu, where there is considerable potential for more irrigation, there is hope that the Pongola River dam may eventually provide irrigation water.

The Bantu Investment Corporation wants to support relatively large-scale projects with outside management and adequate amounts of capital. If these projects become self-sustaining, they will be handed over, presumably on favorable terms, to Africans. In some cases, the larger units may be broken up into smaller parcels held by private owners who compose a cooperative. In others, control may pass to single owners or possibly to corporate ownership. No commitment has been made to any particular schedule or to any method of transfer. It is uncertain whether direct control of large farms and plantation-style units by white managers using hired African labor will permit the rapid transfer of skills and managerial ability.

The only cash crop that is currently important in either homeland is sugar cane in KwaZulu, although some cotton is planted in both. The Natal coast is a prime area for sugar cane, and parts of KwaZulu, the only remaining lands suitable for sugar cane expansion, could become very productive. High world prices have spurred an interest in expanding sugar cane acreages and improving yields. A quota system is in effect in the Republic and KwaZulu has been granted an allotment. This is divided and allocated to farmers by the KwaZulu government on the advice of white extension agents. Although allotments usually comprise only a few hectares, the returns to sugar cane farmers are lucrative and assured. The crop takes over a year to mature, however, and, in contrast to subsistence farming (or working in industry), a threshold amount of capital is needed to plant the crop, hire labor, buy fertilizer and other inputs, and support the farmer and his family while the crop matures. Understandably, all of KwaZulu's quota has not always been utilized, but the South African Sugar Association has helped initiate a credit and training program to assist small producers, and the homeland agricultural department is attempting to devote more effort to cane production. If large enough acreages can be granted to Zulu farmers, and if they are supplied with credit and extension guidance, they can earn individual incomes that are attractive compared to those earned in factory work.

In the six-year period, from fiscal 1968/69 to fiscal 1973/74, about 3,300 Zulu farmers supplied sugar cane to mills. Output has varied from about 275,000 to 420,000 tons, with no clear trend. The recent higher price for

sugar has, however, almost doubled the earnings of Zulu farmers from $1.5 million in fiscal 1968/69 to $2.8 million in fiscal 1973/74.^[10] Sugar cane output amounts to about 10 percent of the total value of KwaZulu's agricultural output, but is roughly half of industrial output (more than R5 million). It appears that relatively small expenditures on sugar cane can have a profound effect on KwaZulu's total domestic product.

Animal products and livestock sales are a major component of agricultural production. It is well known that cattle play an important role in African society, serving as a measure of wealth and status and as a means of extending marriage and kinship alliances. Less clear, however, are the effects of these traditional factors on present-day practices. It is arguable that veterinary, extension, and marketing shortcomings, contrasted with the persistence of tradition, are at least as important in explaining the underutilization of African herds. There is some evidence that initial expenditures on improving livestock quality and on bettering market facilities are yielding substantial gains. The statistics are subject to error, but the cattle population of the homelands probably reached a plateau several decades ago and has since fluctuated with weather and grazing conditions. Between 1968 and 1973 the number of cattle in Bophuthatswana was in the range of 345 thousand to 384 thousand; in KwaZulu, there were 1.2 to 1.3 million.^[11] Much of Bophuthatswana is suited for at least light grazing, and cattle numbers there are in line with an estimated carrying capacity of about 400,000 head, allowing for other stock; KwaZulu, however, has double the number of cattle it should have. Again, averages may be misleading, since there are probably parts of both homelands that are more severely overstocked than others, and there may be tracts, even in KwaZulu, where levels are acceptable or even below capacity.^[12]

Any increase in herd sizes would obviously be undesirable, but two reasonable aims of a stock development program are to improve animal quality and to increase the proportion of the herd marketed annually. Livestock experts believe that gradual modernization of homeland animal husbandry is taking place. Knowledge of the purposes of stock dipping and preventive inoculation is spreading, and less pressure is needed to secure compliance. They say that herd sizes are better controlled, that the value of quality breeding stock is appreciated, and that animal size and quality have improved. The African herdsman does take pride in good stock, and the few breeding bulls that are sold each year draw premium prices.

[10] BENBO, KwaZulu, Economic Revue, 1975 (Pretoria, 1976), 40.

[11] Ibid. , 38; BENBO, Bophuthatswana, Economic Revue, 1975 (Pretoria, 1976), 34.

[12] Bantu Investment Corporation, Homelands: the Role of the Corporations in the Republic of South Africa (Pretoria, 1975), 2nd ed., 66.

In the last few years these older extension programs have been supplemented by better marketing and finishing facilities. The homeland government agriculture departments and the Bantu Investment Corporation have established new stock marketing projects in both homelands, at Taung in Bophuthatswana and at several sites in KwaZulu. There is an open competitive auction, sellers are offered at least a minimum price, and substandard stock are fattened before final sale. The initial response, boosted by rising world prices for beef in 1973 and 1974, was very positive and strongly suggests that fair prices, good marketing and finishing facilities, and continued attention to improved quality may encourage a more commercial approach to livestock management. The data show a very strong positive supply response to higher prices, belying the often heard maxim that Africans will not treat their cattle as economic goods.^[13] Increased turnover rates—selling animals as they mature, and not carrying them to old age and death—could add millions of rand of almost costless income to the homelands. But much must still be done to improve the handling of cattle and to upgrade marketing before the potential economic value locked up in the herds of the homelands can be released.

Another aspect of livestock management where improvement is needed is dairying. The history of dairying in the homelands has been disappointing, as early efforts to form cooperatives languished in the 1950s.^[14] The Bantu Investment Corporation estimates that 107 million litres of milk are required each year by the towns of the homelands, but only 613,000 litres—0.6 percent of the needs—are currently being produced in organized schemes. In 1974 a dairy unit of 320 cows was started in Bophuthatswana to service part of the needs of GaRankuwa; similar projects are planned for KwaZulu.^[15] The corporation appears ready to make a commitment to large commercial dairies, which have good marketing potential and good backward linkages to suppliers of fodder crops.

There are over 800,000 sheep and goats in each homeland, and large numbers of pigs, horses, mules, donkeys, and chickens. Little is known of the economic worth of these animals, their importance in the African diet, and their role in the ecological system. Wool and skins are sold, and some sheep and goats are marketed or slaughtered privately.

The agricultural development of the homelands depends upon improving the output of food crops, commercial crops, and animal products. An overview of growth patterns in these areas from 1965 to 1973 is presented in

[13] Ibid. , 75. Data available on increased sales since 1973 and 1974 from KwaZulu Department of Agriculture, Pietermaritzburg, discussions with officials in homeland governments, and the Bantu Investment Corporation.

[14] H. Hamburger, "Animal Production in Developing Communities," Proceedings of the South African Society for Animal Production (1968), 29–40.

[15] Bantu Investment Corporation, Homelands , 2nd edition, 73.

table 8.2. Once again the estimates must be regarded as tentative approximations, but they are a considerable improvement over the fragmentary statistics previously available. Although in the previous year the gain was only 27 percent, the value of agricultural output in Bophuthatswana had risen by 96.5 percent in 1973 over the 1965/67 base. In KwaZulu, the gain over the period was 54 percent, but again, through 1972, growth amounted only to 32 percent. The 1972 figures imply low compound growth rates and it would be necessary to correct these by an appropriate price index before reaching any conclusion about volume movements. The two key components in both homelands are animal products and food grains. Growth in food grains in Bophuthatswana has been negligible, and only the higher prices of 1973 and possibly increased sales due to poor rainfall and grazing buoyed the value of animal production above the flat trend of earlier years. The results in KwaZulu are similar for livestock, but there does appear to have been modest growth in the value of food grain production. The utilization of forests for firewood and timber has declined.

The most obvious gains have come in the most recent few years and in specialized areas. The sisal projects, and stimulus to fodder and horticultural products provided at Taung, have elevated the value of these relatively minor items in Bophuthatswana. KwaZulu has benefited from higher prices for sugar, the main component of commercial crops, and there has evidently been some progress in horticulture.

Overall, there has not been any sort of decisive improvement in food grain farming or in livestock management in the two homelands. Most gains in these two critical areas have occurred unevenly and have been concentrated in few areas and—probably—among relatively few persons. There are signs of retrogression, too, where herds are too large for grazing lands to sustain, where fencing and conservation works are not maintained, and where more and more people try to eke out a subsistence existence in overcrowded rural locations. Success and failure often lie side by side. Tractors, trucks, and modern equipment are found in many rural areas. Several Tswana ranchers have built up herds of registered stock. In Natal, there is an emerging group of sugar cane growers making an acceptable living out of farming.

Where resources and personnel have been committed, as in sisal, phorium tenax, vegetable gardening, and stock sales, there have been some measurable gains—gains large enough to hold out some hope that with more resources and more personnel even more growth may be realized. The nagging question is whether the management of these pilot projects can be transferred to local people. Expenditures on agricultural development have been small, and the fact that measurable gains have resulted at all is noteworthy. It would therefore be surprising if cost-effectiveness studies did not show large returns to more expenditures. On balance, however, there simply has not been enough forward movement to make a real difference: cumulative growth has not been

initiated in homeland agriculture. And if such growth depends on major inputs from outside, what will be the reaction of the white farming community to special prices, marketing arrangements, and subsidies? And how will the homeland urban markets be shared, and what of Soweto and the other African townships of South Africa?

Causes of Agricultural Underdevelopment

The agricultural problems of the homelands are not unique, although they do differ in some ways from those of other underdeveloped areas. Nor, if properly diagnosed and dealt with by applying a coherent set of policies, are they insoluble. The causes of low productivity and stagnation in homeland agriculture are obvious and numerous–indeed, the only real difficulty is to determine which should be given the greatest weight. They may be grouped under three headings: 1) inadequate modern inputs and poor infrastructure; 2) obstructive land tenure and labor practices; and 3) shortcomings in extension and policy.

The shortage of water in Bophuthatswana and the lack of control over water in both homelands keeps total production of food, other crops, and livestock down and reduces the productivity of land and labor. The western areas of Bophuthatswana average ten to fifteen inches of rain a year, and the Transvaal blocks receive only twenty to twenty-five inches. But nowhere in this homeland is rainfall reliable. Only 6 to 7 percent of the area is arable, and only about half of this is cultivated. Outside the eastern Molopo and Ditsobotla areas, and Thaba 'Nchu, there is little agricultural potential except in ranching. Small irrigation projects, such as one at Taung, may make possible a limited amount of intensive gardening or small-scale truck farming. But there is little ground water potential and few sites for riverine canals. In some areas it will not be easy even to keep people and livestock supplied all year with potable water, Since Bophuthatswana must compete with the cities, mines, and industries of the Pretoria-Johannesburg region for water, and with white farmers, it is improbable that its agricultural needs will be given high priority. The proportion of the work force that can be retained in herding and light farming is likely to be very small.

In KwaZulu the agricultural potential is great, but there may be a tendency to overestimate what can be done by way of providing incomes for numbers of workers that can compare with those offered in industry, government, or education. Rainfall averages thirty to fifty inches almost everywhere and the growing season is long. The northern part, between Swaziland and the coast, is not useful except for the grazing of livestock; most of the inland regions can sustain herding and intermittent cultivation, merging into mixed farming regions near the coast and on the interior plains. A major problem is the rolling, hilly terrain, which heavy rains scour when protective vegetation

has been destroyed. Reclamation and runoff control can add substantially to the cost of sustaining stable farm operations. Apart from herding and farming, there are the essential water and land resources, and climatic conditions conducive to fruit and plantation crops. KwaZulu has the potential for the expansion of timber production. There is no environmental reason for KwaZulu's agriculture to remain as unproductive as it is, but considerable investment, adaptive ingenuity to bring in new crops and technologies, and good private and public resource management will be needed before agriculture can advance.

Information about the use of modern inputs is difficult to obtain. Some chemical fertilizer is applied in both homelands, mostly in the Ditsobotla maize and Taung irrigated areas in Bophuthatswana and in the sugar cane areas of KwaZulu. Some high yielding seeds are used, and crops and animals are protected from diseases with modern chemical products. Plowing is still more likely to be done with oxen than with owned or rented tractors. Some dairy and other cooperatives have been formed in both homelands, but few have survived. In 1973 provision was made for the creation of a Bophuthatswana Agricultural Board, with the power to establish district advisory committees. The homeland minister of agriculture may act on the recommendation of the board to extend credit and other forms of assistance to cooperatives. Bophuthatswana's fiscal 1975/76 budget set aside R369,000 for loans to agricultural cooperatives, up from R100,000 of the previous year. If the Bantu Investment Corporation also moves ahead with plans to provide certain kinds of credit, then at least a few farmers and stockmen may be able to obtain the financial assistance they need to survive and expand. On the whole, however, shortages of credit, modern inputs, and controllable water supplies remain major, immediate causes of low yields and poor farmer responsiveness.

At the core of the process of agricultural production is the organization of labor and land. In the homelands both of these major inputs are subject to social and political forces that make it impossible for them to move flexibly into (or out of) agriculture. Rural and urban land use is not determined by economic advantage, but by factors intrinsic to South African society and the homelands' place in that society. Similarly, labor use is affected by the absentee labor system and by the limitations on labor mobility imposed by influx control. Neither of the two main factors of production is allocated by its returns in the market system. In these circumstances, and given the dearth of modern inputs and market outlets, agriculture cannot be anything but a subsubsistence, noncommercial activity. There is neither incentive for the average person on the land to do things any differently, nor, if he wanted to, is there any way in which he could bring together in profitable production the land, labor, capital, and supplementary inputs needed. It is commonly

recognized that Africans on the land are in some sense noneconomic, but this behavior is often wrongly attributed to motivational, psychological, or cultural factors beyond the immediate control of the government and its extension agents. That this diagnosis is outdated and wrong is gradually being understood, but neither the real reasons for the persistence of subsubsistence practices nor the means of breaking them down are widely apprehended.

The use of labor in agriculture is subject to several constraints. In traditional times there existed a division of work within the household between husbands and wives, boys and girls. Each family unit was to a large extent self-reliant, and its members were responsible for and skilled in a number of tasks. It is difficult to gauge to what degree the customary division of labor endures in Tswana or Zulu society and to what extent, if any, its lingering inhibitions affect attitudes to agricultural work. As the reports of Schapera and others make clear, modern influences, such as the plow and Christianity, and economic necessity, have altered attitudes toward the work roles of men and women.^[16]

There is a good deal of variation by area in the degree to which customs have broken down, but such older social restrictions are now much less important than economic considerations. For most men the returns to full-time labor in agriculture are not commensurate with earnings outside. The labor pattern looks much the same as before, since women often handle agricultural tasks, but the root cause is not custom but economic calculation acting under political constraint. As long as influx laws keep women from moving to their husbands' place of work, much of the minor agriculture will remain in their hands. Homeland agriculture thus has a reciprocal or symbiotic relationship with the absentee labor system, economically and socially, the one being the obverse of the other. Families straddle both economic sectors, trying to build a partial subsistence base and security in house, land, and livestock on one side and to realize adequate cash income on the other. This is individually rational, but amounts to collective suicide insofar as developing a modern agricultural sector is concerned.

The absentee labor system and restrictions on the movement of workers and their families act in effect as a rural "depressor," overloading grazing and farm lands in the homelands with redundant labor that, given the shortage of other resources to work with, is casual and unproductive. A change in the system of labor would draw the excess population off the land, and, with new land tenure laws, make it easier and cheaper for government action to raise incomes for full-time farmers to parity with earnings from absentee work. Doing the reverse, further restraining the flow of African

[16] Isaac Schapera, The Tswana (London, 1971), 27.

workers to the cities as population expands without creating enough non-agricultural jobs in the homelands, will force more women and men into subsubsistence agriculture and make raising average productivity impossibly expensive. Plots will become smaller — perhaps shrink to only a few hectares — and gains could come only with the addition of enough inputs — water, fertilizer, and credit — and sufficient training to create a Japanese-style agriculture in a part of the world where such intensive methods are alien.

Many experts believe that another fundamental barrier to the development of homeland agriculture is the system of land tenure. The present system is widely and erroneously regarded as a carefully preserved relic of earlier times. As with the division of labor, the modern pattern of allocating rights in land bears but a superficial resemblance to and incorporates vestiges of traditional practice. For the most part land is not allocated through the market but by chiefs and headmen working with the approval of the white authorities. White officials created the present tenure system, oversee the distribution of lands in many localities through the "physical planning" or "betterment" program, and have the power to force black individuals and groups to abandon homes and lands and move elsewhere.

To simplify, some lands are owned both by Africans with private property rights and by collectivities like churches or missions. Most land in Bophuthatswana and KwaZulu is predominately held under what is usually known as traditional communal tenure, the most important feature of which is the prerogative of the chief to give rights to land which supposedly include a homesite, adequate fields for crops, and grazing rights in the common pasturage. Yet a continuation of the traditional ideal has become impossible with the growth of population, and fields are now mostly too small to support a family. Grazing rights are insufficient for everyone's stock. The main feature of the existing tenure system is its enormous elasticity in absorbing population, although unlike an Asian intensive agriculture based on small holdings, it does so only with diminishing returns and ecological damage. The system is made flexible by its very inchoateness: white commissioners and their deputies, and chiefs and headmen acting in their designated capacities as land donors, can accept, or at least tolerate, squatters, migrants from other parts of the homeland, or alien blacks drifting in from other homelands, white farms, or foreign countries. The rigidity of the Republic's influx controls, "black spot" removals, and homeland consolidation plans is matched and made workable by the fluidity of homeland tenure rights.

Although it is often said that the chief has the power to evict persons, most people in fact acquire hereditary rights of usufruct. Under the homeland concept, every African in South Africa is obligated to identify himself with a homeland constituency and vote as if he were from its area. As a citizen-"resident" it would appear that no person could be denied access by a chief

or by any white authority to a modicum of living space in his official homeland area, regardless of what a chief says. This is just one of the many unresolved contradictions and anomalies involved in the system of homeland tenure rights. The government and local chiefs, who often collect "rents," also condone squatting — such as exists in the sprawling collection of shanty towns north of Pretoria — although it has no legal basis.

Nominally, chiefs still control land distribution, but the titular chief of all Africans is the president of the Republic, who is invested with discretionary powers to alter African land occupation and use. Under the Republic's law a black person or persons can be administratively evicted and told to live elsewhere — this is the basis for "black spot" removals and resettlements. This power will also be used to consolidate the homelands without African approval.

For a time many tribes were indifferent to or resisted the planning of their areas, but acquiescence is not required and local wishes can be overriden. Control over land-use patterns — where people live, how much farm land they have — ultimately rests with executive authorities who can act in unconstrained ways. No traditional or modern rights of blacks on the land are strong enough to provide secure immunity from the executive powers held by whites.

Because land rights are granted by legal and political authority, there is no land market. Africans are not free to buy and sell land in white areas or in homelands, except where there are rare privately held farms and in certain townships where housing plots are available. Otherwise access to land is obtained through inheritance rights or by obtaining additional grants from tribal authorities. Agricultural extension agents think that this tenure system, which attempts to provide each rural family with a full or partial subsistence base, does not permit efficient farmers to add to their holdings, and in no way discourages casual, ineffective land use. The traditional grant of access to unlimited grazing likewise impedes attempts to improve herding methods. The lack of private ownership makes it impossible for farmers to put up land as security for loans. Although loans on crops alone could be feasible in some cases, such as on sugar cane, where output and prices can be predicted, credit-granting institutions, such as cooperative or agricultural banks, will otherwise have difficulty making loans unsecured by mortgages on land.

A central problem in homeland agriculture is to break down the barriers that now exist to the economic allocation of land and labor in production. Land is distributed on the basis of social and political principles. Labor is drawn off into the commercial economy, leaving agriculture as a predominately subsubsistence occupation for dependents and those temporarily withdrawn from the absentee labor system. There is no capital or credit market. There is simply no way for the enterprising farmer to acquire the

land and credit he needs, in union with his own and any hired labor, to make a commercial success of growing crops or raising livestock. Furthermore, as matters now stand, he will find it difficult to secure advice, modern inputs, technology, and access to product markets. The market system is not a panacea, but the creation of at least limited markets for land and capital is essential if farming in the homelands is to become economically oriented.

Suggestions for tenure reform have been numerous. The Tomlinson Commission called for the survey and allocation or sale of subsistence-sized plots to absorb the maximum amount of African labor in agriculture and for the creation of a land market. The creation of at least a limited land market is favored today by most extension workers. KwaZulu has appointed a commission to study land tenure. Bophuthatswana is interested in a similar review of land policy. It would be possible to put a ceiling on land holdings and to limit tenancy so that the stratification of rural society into a small class of large holders and a mass of unpropertied laborers would be avoided. Experiments with cooperative or corporate farming might not be out of the question.

Resistance to change in African tenure comes from the very highest levels of the South African government. The reasons usually given are the need to preserve the culture of Africans and a desire to maintain the positions of chiefs and induna. The first objection is transparently false, but the second demands some attention. In earlier days, the power of the local ruler allowed him to allocate rights to land, but the establishment of homeland governments has effectively modernized the administrative and political position of the chiefs by specifying their functions and by awarding many of them ex officio seats in the assemblies. It may no longer be necessary to impede agricultural progress to secure the chiefs' political position. In any event, compensation in the form of cash, bonds, or lifetime annuities could assure the chiefs' compliance. Tenure reforms would make it easier for homeland Africans to acquire township housing and business sites, subject to zoning laws or other types of regulation, and could accelerate urban as well as rural development.

Some observers fear that the commercialization of land would lead to radical changes in African life and create new burdens on the homeland and Republican governments. Presently, many of an African migrant worker's needs are met by his right to a homestead in the homelands. He would have a house, his family would raise at least a portion of its food needs, and he would have a place to go when sick or old. In theory he belongs somewhere, and kin and family give him comfort and support. If land rights can be disposed of, however, the Republican and homeland governments must cope with a mounting need for expensive social programs — social security, unemployment benefits, health care, housing, and poverty relief. Access to land

in a designated homeland is one of the few unambiguous rights Africans enjoy, and unless job benefits and government social services can replace the security conferred by that right, a change in the tenure system might prove socially harmful.

Three agencies are involved in setting and implementing agricultural policy in the homelands: homeland government departments, the advisory services branch of the Department of Bantu Administration and Development, and the Bantu Investment Corporation. Because of the division of control over agricultural policy, it is not possible to describe and evaluate a single strategy of development. Homeland government departments continue to concern themselves with thinly spread extension work aimed at improving crops, gardens, and herds and with the erosion and grazing controls implicit in physical planning. The Department of Bantu Administration and Development advocates traditional policies including betterment. The Bantu Investment Corporation favors large-scale projects.

The extension work of homeland governments has not been very effective, despite the expenditure of much energy. Even the showpiece sisal and phorium tenax projects in Bophuthatswana and KwaZulu are for the most part unprofitable. In the irrigated areas, such as at Taung in Bophuthatswana or along the Tugela River in KwaZulu, allotments are tiny and often neglected despite the efforts of extension workers. African extension workers in rural districts find it difficult to convince people to adopt new seeds or techniques. The extension staffs of both homelands work in an atmosphere of defeat and frustration countered by individual concern and hard work. There is not enough money to do those things that promise gains, and barriers such as the lack of credit, the rigidity of the land system, and the absence of transport and marketing outlets, make real progress difficult. Large commercial projects mean nothing to Africans who cannot command the capital, land, and labor needed to establish similar enterprises. Nor can irrigated plots, no matter how fertile and profitable, signify progress in parts of the homelands where water is scarce.

Extension staff is spread unbelievably thin. Lipton reported that "while 90,000 rich, educated white farmers have 3,000 extension officers (plus enormous injections of easy credit, marketing facilities, and guaranteed prices) 600,000 black farmers have less than 1,000 extension officers: and these hopelessly overstretched men (and their small budgets) have been concentrated on the irrigation schemes." She adds, "the black peasant had not yet been given a chance."^[17] Some of the senior white officers are well informed, but as Lipton also points out, others are seriously out of touch with advances in agriculture in Africa, Asia, and the rest of the world during the

[17] Merle Lipton, "The South African Census and the Bantustan Policy,," The World Today , XXXVIII (1972), 266.

past few decades. One of the great advantages of détente between the Republic and the rest of Africa would be openings created for the staffs of the homeland departments—and of the development corporations, too, for that matter—to travel and visit countries where similar developmental problems are being attacked and sometimes solved.

The belief that Africans in agriculture will not react to high prices or to the opportunity to make profits is inaccurate, as the recent response to better stock prices shows and as isolated individuals of surpassing enterprise have proven. Unfortunately, belief in the unresponsiveness of the black farmers appears to be confirmed by the poor results of current extension policies and by the general sluggishness of homeland agriculture. Yet these are not the result of African conservatism, but rather a positive response to the income rewards of absentee labor, the inability to obtain and hold enough land, and the absence of controlled water supplies, capital, and markets. Until stable, certain economic opportunities are provided, Africans will not risk committing themselves to new practices.

One brighter spot is the output of African extension officers from the agricultural colleges at Taung, for Bophuthatswana, and Cwaka, for KwaZulu. The graduates of these institutions are well trained (under strong paternalistic white supervision) and have positions virtually guaranteed for them in the homeland extension services.^[18] The training programs appear well enough designed, although they may lack enough attention to the human and cultural sides of extension work. The quality of incoming student material has not always been high, and some young men doubtless regard the work as unappealing. Agreeable conditions in the colleges, larger stipends, and higher pay after graduation can ensure that these important institutions do not operate under capacity.

For over forty years resources have been poured into betterment schemes in order to stabilize and develop agriculture in the homelands. But the program has had no visible positive impact on production in the homelands. Moreover, because it has been widely circumvented, it is doubtful that it has really done much to prevent erosion or control the size of herds. It has had the dubious negative virtue of poisoning relations between agricultural field workers and local farm communities. The percentages of the homelands planned, and the miles of fence and grass strips annually laid down, have stood for many years as the leading benchmarks of progress in homeland agriculture.

[18] H. W. L. Lilley reports that Cwaka graduates scored better on a test of farming knowledge than did older African extension officers in Natal and the Ciskei and white farm overseers; they were equalled only by a group of whites holding agricultural diplomas. "B.E.O. Knowledge and Understanding of Agriculture and Extension: A Comparative Study," unpub. paper (January, 1972), mimeo.

Physical planning involves the reorganization of existing land-use patterns into clearly demarcated residential, cropping, and fenced grazing areas.^[19] There are high social costs to rural Africans of these physical rearrangements, but they are not considered in relation to the conservation benefits perceived by the planners.^[20]

Lilley provides an account of the reaction of local Zulu to the planning and resettlement of a location near Port Shepstone. After little consultation it was proclaimed a betterment area in 1954 and, by the middle of 1960s, was divided into the prescribed three parts. One clan, the Ngeleka, was moved away from the homesites to which they had unusually strong emotional claims. People did not comprehend what was being done to them and felt cramped in the residential sites. Stories circulated of people who had lost land through the resettlement scheme. A small tax was imposed at about the same time for unrelated reasons, but was felt to be part of the program. Everything came by fiat from the government and was part of the same incomprehensible scheme.^[21]

At the end of 1974 Bophuthatswana was 57 percent and KwaZulu 49.1 percent planned.^[22] Yet there often is a considerable lag between the planning on paper of an area and its actual resettlement. More tribal authorities in Bophuthatswana are now said to desire planning than the Department of Bantu Administration and Development can oblige, while the Zulu are regarded as uncooperative. By the department's own standards, the progress of physical planning has been too slow, its explanation being the resistance by some peoples to the policy and a shortage of available funds. The subsistence plots sought by the Tomlinson Commission have been compromised by population pressure, and families have received only one-half, one-third, or even one-quarter of a recommended unit.

It is difficult to see why the betterment policy has been pursued for so long. An explanation may lie in the commitment made to the program long ago

[19] Since physical planning is a spatial reorganization of houses and fields it is best understood in terms of maps. See J. S. Murray, "The Development of Agriculture in the Bantu Homelands," Tegnikon (March, 1967), 30–36; P. Smit, "Physical Planning in the Bantu Homelands of South Africa," ibid. , 5–17. Both Smit and Murray have been heavily involved in the direction of physical planning.

[20] The attitude of the Tomlinson Commission was that "All of these stabilization measures can be undertaken without the approval of the inhabitants of the Reserves in question, since the necessary powers will have been obtained by legislation. . . . Planning for stabilization and the carrying out of such plans must, wherever possible, be undertaken with the consent and cooperation of the reserves concerned; but where protection of the soil demands it, stabilization must be carried out even without such consent." Summary of the Report , 118.

[21] H. W. L. Lilley, "Characteristics and Motivational Orientation of the Amazimakwe Land-Occupiers of Location 4B," master's thesis (University of Pretoria, 1967).

[22] Survey of Race Relations, 1975 , 142.

by senior personnel who continue to head the agricultural branch of the department and who have passed the policy on to the relevant homeland departments. As the Tomlinson Commission emphasized, physical planning affords no solution to the problem of raising African farm productivity. It stated bluntly that the mere rearrangement of the spatial structure of an area had not up to that time and could not in the future make any lasting differences.^[23]

The reasons are obvious. Physical planning entails a once-and-for-all change in production conditions; under the best circumstances, it would allow for slightly higher crop and herd productivity. But there is no fundamental change in the organization of the patterns of use of land and labor, and if much greater quantities of modern inputs together with credit and extension advice are not simultaneously added, sustained growth will not be initiated.

Agriculture must be the key sector in any attempt to develop the homelands as hospitable living areas and to create growing regions increasingly self-sufficient in food, raw materials, and jobs. Of all the homelands, Bophuthatswana probably has the lowest potential. KwaZulu is at the other end of the continuum. Its agricultural sector can probably support a significant number of full-time farmers and herdsmen as well as provide some cash income for dependents of absentee workers and part-time agriculturists. It would be wrong, however, to talk in static terms about the carrying capacity of the homelands and to try to estimate how many farm families earning subsistence incomes could be packed into them. The homeland economies are going to become urbanized at a very rapid pace and a larger share of their work force is going to move into industry and service jobs, regardless of whether they are located in white areas, in border industrial areas, or inside the homelands. Incomes in farming will have to rise in step with those in other sectors; even dependents and part-time workers will withdraw from agriculture as wages rise in the nonagricultural occupations. With development, a smaller share of area national income will come from homeland agriculture. The Tomlinson vision of a static, subsistence agricultural sector was utopian even in its time.

The bias in the Bantu Investment Corporation today is in the direction of larger farm units; this is probably also the dominant sentiment of the homeland governments' field staffs. It is very likely that homeland agriculture will come increasingly to resemble white agriculture, because it is the model known to white and black farmers and extension advisers. Black farms, because of population pressures, inflexible land tenures, and capital shortages, will probably remain smaller and will yield less output per unit of labor

[23] Summary of the Report , 77.

input than white farms, but productivity per unit of land will gradually rise. In the next decade and a half, if the flow of inputs is allowed to grow, there should be rather large absolute increases in output, mostly because levels are currently so abysmally low. In the near term, an effort could be made to reach part-time farmers, many of whom are women or older men. This step would not generate the highly visible employment and income that the Republic wants, but the homeland governments could scarcely do anything that would do more to improve the quality of diet and the output of food.

Returns to investment in homeland agriculture, either in larger scale units or in the household sector, will be great if projects are carefully planned, in the technical sense, and if they are suited to local conditions and provide real incentives for black enterprise. There is no better way to meet those conditions than to rely as heavily as possible on the insights and opinions of the black farmers, extension staff, and technical experts, as they are trained, and on political leaders, all of whom will have to make the programs work.

Industry and Commerce

It has long been recognized that the development of the homelands cannot depend exclusively upon the agricultural sector. Even the Tomlinson Commission's optimistically low homeland population projections exceeded by a wide margin its estimates of the employment absorptive capacity in agriculture. Further, in order to halt and if possible reverse the exodus of labor to white areas, the commission acknowledged that a large number of job opportunities would have to be created in mining, commerce, and industry in the homelands. But Verwoerd, then minister of native affairs, rejected the report's suggestion that white capital be allowed to invest in the homelands. Instead he preferred to see the creation of industrial centers on the homeland borders. This decision blocked industry from moving into the African areas for fifteen years.

South Africa's industries are heavily concentrated in four regions: Pretoria-Witwatersrand-Vereeniging, the southwestern Cape, Durban-Pinetown, and Port Elizabeth-Uitenhage. Covering only 4 percent of the land area, these regions contain 75 percent of all manufacturing enterprises and generate 80 percent of total industrial production. Half the industrial output of the country arises in the first region alone.

In mid-1960 the Permanent Committee for the Location of Industry was formed to coordinate the government's actions, but its weak inducements could not appreciably affect private business location decisions. The Physical Planning and Utilisation of Resources Act of 1967 gave the minister of planning substantial power to regulate the location and expansion of industries. Private industrialists were initially unsettled by the uncertainties arising

from the government's new controls over their investment decisions, but the Report of the Interdepartmental Committee on the Decentralisation of Industries (the Riekert Committee) and the accompanying white paper, both appearing in 1971, made clearer the standards and procedures that would be applied. Concessions were spelled out in detail and a number of border industry locations were affirmed or designated.

In 1969 the government decided to permit white investment in the homelands on an agency basis. Additional incentives were given to attract investors, but all contracts provided for the eventual handing over to Africans of the ownership and management of the enterprises. The Industrial Development Corporation had earlier been given primary responsibility for creating the infrastructure, providing housing, and making loans in the border industrial areas. The Bantu Investment Corporation similarly took on these tasks in the homelands and was also to build and lease factory buildings and factory flats. Various Republican and homeland departments would assist the corporations in providing water, power, roads, rail connections, and other services.

In 1975 the government set a target of 6.4 percent annual growth of gross domestic product, a rate that could be attained only by moving African workers into more highly skilled jobs. The continuity of the government's economic thinking was evident: it sought rapid economic growth, sufficient job creation for Africans to reduce black unemployment, and even greater emphasis on establishing poles of development away from present centers and for the most part outside but close to the homelands.

The rationale for the decentralization program is complex, although the main intention remains clearly to stem the movement of black workers to the older white centers. Military safety, pollution, rail and traffic congestion, and water shortages are sometimes advanced as justifications, and cannot be completely dismissed, but the policy of decentralization is primarily a social policy that has economic effects. The restrictions and concessions of the program create incentives designed to induce labor-intensive industry to move away from its existing focal points toward the hinterlands.

Its significance to white society explains why the industrial relocation program remains under the tight control of the Republic's government and why the homeland governments have only the smallest say in their own industrial development. The internal industrial bases and the associated border industries of Bophuthatswana and KwaZulu will thus be planned in accord with the spatial and ethnic group employment designs of the Republic rather than coordinated with their own internal agricultural, mineral, demographic, and infrastructural resources.

Over the past fifteen years approximately fifty small towns and cities near the African homelands have been designated as border industrial areas.

There is no single reason why certain centers have been chosen and others neglected. In the early 1960s border areas were chosen that were merely extensions of the major economic regions. Industrialists received the benefit of the already existing positive external benefits—transport, skilled labor, water, power, housing—of the older urban complexes and, in addition, benefited from the concessions allowed by the government. Both Tswana and Zulu areas were affected by this mild relocation, since Rosslyn, on the northern fringes of Pretoria, and Hammarsdale, between Durban and Pietermaritzburg, and Pietermaritzburg itself, were so designated. Rosslyn and Hammarsdale have impressive industrial parks housing a number of major national and international firms. Rosslyn is a center for automobile assembly, and may have provided 10,000 jobs for Africans, many of whom commute from the GaRankuwa or Mabopane townships. By the late 1960s these areas had filled up and emphasis had shifted to more remote centers: Brits, Rustenburg, Zeerust, and Mafeking in the Tswana region; Empangeni-Richards Bay, Ladysmith-Colenso, and Newcastle near the Zulu territories. A 1975 estimate by the minister of planning and statistics claimed 18,058 jobs had been created at Rosslyn, Hammarsdale, and Pietermaritzburg. Over the last eleven years, he said, 68,685 people had been given employment in border industry areas.^[24]

With others, Lombard has criticized the results of the first decade of decentralization. He found no evidence of substantial industrial relocation. The first border industry sites were chosen to minimize the short-run direct economic costs to the government of the program. The Tswana and Zulu areas happened to be most affected, but regions with equally underemployed labor in the Transkei and northern Transvaal were ignored. Many non-Tswana and Tswana were encouraged to move as close as possible to Pretoria. Many Zulu were drawn toward the Durban-Pietermaritzburg area. These population movements did not represent departures from the previous situation. The close proximity of the border industrial areas and their satellite African townships to major towns meant that commuters still spent most of their incomes in white shopping areas. More remote industrial development, and the provision of shopping facilities in the African townships, would have reduced the immediate income leakages back to the white areas and stimulated homeland development and employment.^[25] The most

[24] Survey of Race Relations, 1975, 180.

[25] J. A. Lombard, "Background to Planning the Development of Bantu Homelands in South Africa," unpub. paper, 1972. Elsewhere, Lombard has said: "I have already stated that this whole matter has a political motivation, that the statecraft should take cognizance of the economically oriented basis of the origin of this problem, and that it will have to condition the economic system of South Africa with its profit motive towards the automatic achievement of its objectives. . . . Insofar as the strategy must rest on industrial decentralization it is becoming more and moreclear to me personally that concentration on growth points in border areas alone will not have the desired effect politically. The process will have to penetrate to the heartlands of the homelands, to places such as Butterworth and Umtata in the Transkei, Sithebe, and perhaps other points in . . . Zululand as well, . . . ." "Political and Administrative Principles of Homeland Development," Bantu, XVIII (Feb., 1971), 28.

mobile and responsive firms have already moved to the border areas. It is thus becoming more and more difficult to induce white capitalists to move farther and farther away from the historically dominant economic regions.

The concessions to industrialists for moving to border growth areas were numerous and have been further liberalized from time to time. Loans are available at low interest rates (2 to 7 percent) for land, factory buildings, and equipment. Housing for whites is subsidized with loans offered at 2.5 percent below the market mortgage rate. Part of African (or coloured) wages and capital investment can be set against income taxes. There is a cash grant to cover the actual costs of moving from the older regions. Rebates on rail and harbor charges may be allowed. Concessions of up to 10 percent are given on bids for purchases from the government.

It is difficult to estimate the cost of the decentralization scheme or to calculate its effects on black employment and the distribution of industrial activity. Costs that appear high to some are reasonable to others; gains that seem large to proponents of the program are dismissed as inadequate by critics. The various concessions are hard to monetize and overhead costs of power, water, and transport in a region are not easily allocated to the border industrial sites. Some investment would have taken place in these cities anyway, even without incentives, and some black employment would have been generated. It is therefore almost impossible to say with precision what the net costs and net employment changes have been.

According to the Decentralisation Board, a total of at least R637 million was invested in the decentralized areas by the public and private sectors between 1960 and the end of 1973. The Industrial Development Corporation had committed R159 million and corporation-aided industries had invested R349 million. In addition, R500 thousand had been paid for moving industrialists from the older regions, tax and interest concessions totalled R37 million, and rail and harbor rebates amounted to R6 million. Other funds had been spent on townships, water, and power supplies. About 92,000 African jobs had been generated, an average of about 7,000 per year.^[26]

Through March 1974 the Industrial Development Corporation had invested R33.3 million in industrial building, loans and share capital, and white housing in Brits, Rosslyn, Rustenburg, and other border areas near

[26] Survey of Race Relations, 1974, 274–275. Spending on interior growth points is included. Also see ibid., 1975, 180–181.

Bophuthatswana. Private industrialists attracted to these zones added over R76 million of their own capital, R43 million of which went into Rosslyn. A total of 12,208 jobs for Africans were created; there were 1,771 new white jobs. These were presumably directly created work opportunities and did not include secondary multiplier effects. Over 10,000 of the African jobs were in metalworking or machinery and equipment. Approximately 1,000 jobs were added in each of five other classifications: food, beverages, and tobacco; textiles and apparel; wood and wood products (furniture); chemicals, petroleum, coal, rubber, and plastic products; nonmetallic mineral products. In border areas near KwaZulu—Elangeni, Ladysmith, Newcastle, Pietermaritzburg, Richards Bay, and others—the Industrial Development Corporation spent R70.1 million: R28 million on industrial buildings, R33 million on loans and share capital, and R9 million on white housing. Private capital inflows amounted to R163 million and 19,000 African and 2,000 white jobs were created.^[27]

There has been sterile debate in South African political and academic circles about the number of jobs decentralization has managed to create and their total, average, and marginal costs. Generally, all have agreed that the additional employment has not been sufficient to absorb the annual increments of the total or homeland African labor force, leaving a fraction to spill over into the historic focal points of the absentee labor system. The incremental cost has been higher than that of merely adding additional workers in established areas. The head of the Bantu Investment Corporation has estimated that 20,000 jobs need to be created in the border areas (or in the homelands) to meet the goals of employment relocation.^[28] Bell says that of the 87,500 jobs created between 1960 and 1970 in the border areas, 64,000 would have spontaneously arisen. Therefore, only 2,200 jobs per year can be attributed to the government's efforts. Obviously the cost of creating one of these incremental jobs was far higher than an estimate of the average cost per job using the whole total as the divisor.^[29]

One estimate projected an annual increase in the homeland labor force of 66,000 males and 39,000 females in the period from 1974 to 1976; during the same period in the white areas, 60,000 males and 38,000 females were estimated to be added to the labor force yearly.^[30] KwaZulu is responsible for the largest additions to the labor force of any homeland. Another source

[27] BENBO, Bophuthatswana, Economic Revue, 1975, 40; idem, KwaZulu, Economic Revue, 1975, 46.

[28] Survey of Race Relations, 1974, 275.

[29] Trevor Bell, "Bantustan Economic Development," Third World, II (June, 1973), 30. See also Bell, Industrial Decentralisation in South Africa (Cape Town, 1973).

[30] University of Pretoria, Department of Economics, Focus on Key Economic Issues, 9, The Homelands (1974), figure 7.

placed the present annual African male labor force increment at 120,000, of which 60,000 will be in the homelands.^[31] Sadie predicted that the total African labor force would expand by 159,200 annually in the 1970s.^[32]

BENBO calculates that from 1973 to 1976 the total Tswana labor force will have risen by 72,200: 44,400 in the common area, 27,300 in Bophuthatswana, and 3,000 in other homelands. For the Zulu from 1974 to 1976, 164,000 additional workers would have been seeking employment: 70,100 in the common area, 91,200 in KwaZulu, and 3,300 in other homelands. According to BENBO's planners, all of these figures will be 30 to 40 percent higher in the next three-year period—from 1977 to 1980.^[33] Clearly the addition of new labor from the homeland populations puts enormous pressures on capital formation and leaves little leeway for the luxury of high-cost ideological preference for shifting employment from one region to another. If a recent R8,500 estimated average cost per job in the border industries has some validity, the size of the capital formation requirements becomes obvious.^[34]

In 1970, Riekert said that adding 44,000 jobs per year in the border areas and the homelands would absorb all new homeland workers; 36,000 jobs would keep the number working in white areas constant (since some would naturally withdraw from the migrant labor force, leaving room for some new workers); and 21,000 jobs would succed in stabilizing the proportion of the African work force in white areas. Allowing for jobs arising in agriculture, services, and other industry, the figures would be reduced to 21,000, 15,000, and 5,000 new industrial employment opportunities.^[35] Riekert said that the lowest goal had already been attained, although he did not adduce any evidence in support of his assumption that 16,000 jobs were being created annually in other sectors. These numbers are now too small, but since Riekert was a key figure at the time he spoke, they do indicate that government thinking in the early 1970s was already far removed from any attempt to do more than stabilize or moderately reduce the proportion of African

[31] Bantu Investment Corporation, Homelands (Pretoria, 1973), 1st ed., 125.

[32] J. L. Sadie, in an address to the National Labour Conference, Cape Town (April, 1971), quoted in Gavin Maasdorp, "Targets of Development in Relation to Population Trends and Needs," unpub. paper (1972), 6.

[33] BENBO, Bophuthatswana, Economic Revue, 1975, 20; idem, KwaZulu Economic Revue, 1975, 22. The statistical bases are not consistent in these reports. The Bophuthatswana figures are given as 1970 to 1973, 1973 to 1976, and 1976 to 1980; the KwaZulu figures are given as 1971 to 1973, 1974 to 1976 and 1977 to 1980. We assume that inclusive figures, as given for KwaZulu, are correct for Bophuthatswana as well.

[34] Survey of Race Relations, 1975, 180.

[35] P. J. Riekert, "The Economy of the Republic (with Special Reference to Homeland and Border Industrial Development and the Economics of Southern Africa)," Institute of Race Relations (1970), 14, mimeo.

workers who worked in established centers. The number of workers in core areas would at most be held constant, or limited to population increase, while the number and perhaps the proportion of African workers in the Republic would actually rise because of the border industries.

What has emerged from this debate is the conclusion that the border industries program is at best stemming only a fraction of the outflow of absentee labor from the homelands each year. Even if each of the 6,000 to 8,000 jobs arising annually in the border areas generates 2 or 3 more jobs in agriculture, services, or other industries—a very generous assumption—only one-third of the increase in the homeland male labor force can be absorbed. (The Industrial Development Corporation puts the multiplier at only 1.5 additional jobs.)^[36] No women are included in the base amount, nor is any provision made for Africans who are now moving out of white agricultural areas, for reductions in unemployment and underemployment, or for the repatriation of urban black workers to the homelands. The numbers are large enough, however, to have an impact on homeland living and working patterns, shifting them somewhat toward commuter communities linked to small- or medium-sized white cities. These satellite townships will be additions to the present distributions of workers and living areas, fed essentially by new population growth, and not offering alternatives to migrancy, employment within the homelands, or permanent work in the core areas.

Developmentally, the border industry scheme provides dubious blessings for the homelands. It does not generate new jobs so much as move employment closer to the homelands. Relocation makes it less necessary than before for African workers to migrate to distant white centers for jobs. Families will rent or own homes in townships built by the homeland governments and white governmental agencies and commute on a daily or perhaps weekly basis to their places of work. Because no one has undertaken a systematic study of its impact on the movements, living conditions, and working habits of black labor, the extent to which this has already happened is unknown. The homelands acquire no property interest in the border enterprises, nor will they have any powers to tax them. Moreover, the new firms are not necessarily linked to the mineral, agricultural, or secondary industrial bases of the homeland economies. The homelands cannot plan or control their economic impacts on their own internal economies. On the contrary, the homelands must constantly face the threat that their own best and most skilled workers, administrators, and technicians will be enticed away by offers of higher earnings in the border areas.

Given the openness of the homeland economies and their lack of policy instruments such as tariffs or currency control, their economic positions are

[36] Survey of Race Relations, 1975, 180.

highly vulnerable. Workers will spend large fractions of their incomes in the white towns. Funds that the homelands devote to education, urban amenities, and the social welfare of workers will result in direct or indirect benefits for white firms that are under no obligation to pay taxes in return. Unlike the internal growth points, the border industries are merely devices for extracting the cheapest possible labor from African territories with the minimum of inconvenience for the white firms and the white public sector. Self-interest is likely to lead the homeland governments to try to internalize the economic activities now being brought to peripheral growth points, to acquire control over their planning and operation, and vigorously to represent the economic interests of their work forces.

The internal development of industry in Bophuthatswana and KwaZulu has been sluggish, even in contrast to poorer underdeveloped nations. Given the size of their populations and their per capita incomes, the homelands could be expected to have a number of commercial and industrial centers: certainly a large city or two, some rural market towns, and many shops, mills, and backyard workshops distributed throughout small settlements and along country crossroads. Several million persons, each of whom receives an average of over $100 per year in cash, are theoretically capable of supporting a sizeable commercial class and at least a few medium-sized industrial firms. Tea shops, restaurants, laundries, clothing shops, hardware stores, petrol stations, metal workshops, and automobile and farm machinery repair facilities ought to provide visible evidence of economic activity. It is therefore surprising to drive through the homelands without finding many of these signs of commerce or industry. Even the larger settlements lack the commonplace business establishments so necessary to satisfy the routine needs of their residents. Isolated general dealerships, selling gasoline, soft drinks, tinned goods, staples and condiments, household hardware and utensils, and ready-made clothes are often the only manifestation of business activity. From their small spaces they fill many needs, but they are poor substitutes for a series of specialized shops. Often expensive, they provide convenience when travel is costly and tiresome. Each is a small-volume monopoly working on the margin between the dealer's wholesale cost of goods and the price of a roundtrip bus ticket to the nearest market town.

It is not hard to explain why there has been so little business development in the homelands. Black South Africans lack the access to capital, credit, technology, and markets that enables the white, and even the politically impotent Asian, community to engage in commerce. There has not been any easy manner in which black South Africans could acquire the managerial, accounting, engineering, and marketing skills needed to prosper in business.

Many white businessmen made their fortunes from the exploitation of

mineral rights, land speculation, or sharp business trading. Others made money out of inventions, patents, and foreign collaborations. Most had the support of the government, which provided an infrastructure, erected tariff barriers, and sponsored a legal framework conducive to business enterprise. Africans have never been able to deal in land, mineral rights, capital, and acquire rights to technology on an equal basis with whites. They have been denied acccess to the clubs, bars, and hotels where the social life of business is arranged. They could not easily marry wealth, urge their children to save, or pass on their prosperity to their progeny. Trading rights—the rights to do business—have been zealously rationed by white authorities outside the homelands and by chiefs inside. A wealthy professional man, such as a physician, is hard pressed to find a way of conserving or increasing his wealth. The various laws that prevent African land ownership in the cities, and the tenure system of the reserves, mean that Africans cannot readily become housing contractors or real estate agents. Nor can they build a block of flats or shops for rental income. Those redoubtable and notorious entrepreneurs of the townships, the shebeen women, are harassed by police. In the open, they would constitute a successful business class, and it would be surprising if a few did not move their capital into hotels, restaurants, or other service-oriented businesses.

Now that the Republic's developmental agencies have turned their attentions to the homelands, they have realized that, if the homelands are to become more self-reliant, agricultural development must be increasingly supplemented by progress in other spheres. The agencies have thus embarked upon an aggressive program to implant commerce and industry in the townships and rural areas. As with agriculture, however, the making of policy and its administration is all too frequently based on the premise that Africans are noneconomic and nonentrepreneurial.

The Bantu Investment Corporation was established in 1959 as the vehicle for encouraging homeland commercial and industrial development. Given the extremely deficient industrial position of the Tswana and Zulu homelands, the recent history of their business development is virtually identical with the activities of the corporation. When it began operations the corporation had a staff of six whites and one black. Its share capital totaled R1 million. In the first nine months of its existence it considered 246 applications for loans from Africans, conducted 36 complete credit investigations, and made 5 loan commitments totaling less than R40,000. By March 1973, fourteen annual reports later, share capital stood at R40,830,000. Through March 1974, 1,577 loans worth R11 million had been granted, but four-fifths were for commercial undertakings, and only two dozen were for industries. The enlargement of its staff to 1,021 whites and 6,804 Africans and the

mounting complexity of its internal organization in the last seventeen years are in accord with the forty-fold growth in its share capital. The corporation has evolved rapidly into a many-branched industrial empire. It reaches into every modern economic activity in the homelands: factories, shops, transportation, farming, construction, housing, banking, insurance, and training. No black has, however, served in an executive or technical advisory capacity.

Under its charter, the Bantu Investment Corporation can stimulate the industrial and commercial growth of the homelands in several ways. It can assist African businessmen by providing them with advice, training, and initial or additional capital. It can purchase existing white or Asian enterprises, such as general dealerships, and transfer them to Africans. The corporation can purchase or construct businesses and operate them for profit (or loss) until such time as it believes an African or Africans can assume ownership and management responsibility. The corporation has the responsibility for encouraging and assisting white businesses to enter the homelands, on an agency basis.

The corporation stands ready to advise any African businessmen seeking management, marketing, or other assistance. It builds and leases shopping centers, garages, factories, and office blocks. Several kinds of training are provided: on-the-job, or "in-the-shop" training in building trades and subcontracting, and training in business skills. The corporation operates fifty-six savings banks. It has lent over R1 million to more affluent Africans for private building and home construction in homeland townships. The new agricultural investments branch finances larger scale projects, including irrigation systems, and engages in the provision of agricultural credit. The corporation channeled R66 million into the homelands during fiscal 1975/76 (over 50 percent more than in fiscal 1974/75) but about R26 million of this amount came from private investment. Eleven bus companies have been started or expanded, and to some of these black directors were appointed for the first time. It is further intended that individual firms can invest directly in the homelands after negotiating directly with the homeland government in question: the Bantu Investment Corporation and the minister of Bantu administration and development would have to be notified, however, and such firms would not qualify for incentives or guarantees.^[37]

Under its program to finance African businesses the Bantu Investment Corporation has assisted a disproportionate number of Tswana when compared to the number of Zulu. Through the end of the fiscal 1973/74 accounting period, 423 loans, totaling R3.7 million, had gone to Bophuthatswana, and 417 loans, totaling R3.1 million, to KwaZulu. Since there are roughly

[37] Data on the Bantu Investment Corporation are taken from its annual reports and the annual Survey of Race Relations summaries of those reports. For most recent data, see Survey of Race Relations, 1975, 145–149.

two and one-half times as many Zulu as Tswana in the homelands, Bophuthatswana has been a more active field than KwaZulu for the corporation. In fact, Tswana received 27 percent of the loans made and these amounted to 32 percent of the total value of all loans (through 1974). The Zulu obtained 26 percent of the loans extended, and 28 percent of the total loan value.

In all of the homelands, the Bantu Investment Corporation had financed 1,300 concerns through the end of the 1974 fiscal year. Of these 1,112 (86 percent) were "commercial," including general dealers (759), cafés (117), butcher shops (66), and liquor outlets (71). "Service industries" numbered 164 (13 percent), the most common types being bus services (47) and garages (19). The 24 (2 percent) manufacturing enterprises started included 5 tailors, 5 brick manufacturers, 4 furniture makers, 3 cabinet makers, and 4 dry cleaners. The mixture of these businesses shows that the corporation has concentrated above all on trying to put general dealerships and other types of family stores into African hands. Yet such single proprietorships are obviously going to generate little employment in relation to the amount of money devoted to the shop, its inventory, and the training and supervision of its operation. Until recently it was corporation policy not to permit African individuals to own and operate more than one business. There remains a "fear of scale" built into the thinking of the corporation that appears to restrain it from encouraging Africans to expand and pyramid businesses. It is difficult to tell whether the corporation and the government do not want to foster large-scale black businesses or whether they actually confuse the attributes of small shopkeeping with those of a large-scale business enterprise.

By the middle of 1974 the corporation owned over 100 large commercial enterprises that it had begun or taken over. These were to be held in trust until such time as an African owner or group of operators was ready to assume control. In general, these enterprises are larger and more complex than the average business or shop. They include, for example, breweries, beer gardens, bus services, arts and crafts centers, furniture factories, garages and petrol stations, holiday resorts, wholesale dealers, savings banks, and movie theaters. There is no explicit timetable for turning over these large concerns, many of which are profitable, to Africans. One major enterprise, the Bophuthatswana Transport Company, is apparently nearly autonomous. It has 221 buses and carries over 1.6 million passengers per month.

At the end of fiscal 1973/74, 552 Africans and 26 whites were working under corporation auspices; the three largest areas of employment were manufacture of food (203), retail and allied services (155), and refreshment and accommodation (100). In KwaZulu the scale of operations was almost exactly double with 1075 Africans and 53 whites in Bantu Investment Corporation enterprises: the three largest areas of employment were building construction (322), wholesale trade (238) and food, beverages, and tobacco

(188).^[38] The range of industrial activities is far larger in KwaZulu; it is not clear whether or not this difference is due to variations in policy between the regional offices of the corporation.

The most ambitious undertakings of the corporation are its homeland growth points, which are small industrial parks located inside the homelands. Considerable amounts of money are invested by the corporation in readying the sites of these parks. All transport, water, and power facilities are available and factory sheds may be built or leased, White industrialists are encouraged to locate in these growth points by a number of inducements slightly more attractive than those for the border industries, but they must ultimately turn over the ownership of their operations to Africans. In the growth points African workers are generally freed from the job discrimination restrictions that operate in the white economy. They may theoretically advance to highly skilled jobs and be trained as apprentices in crafts normally closed to them. There are two types of labor cost advantages for the employer. He first obtains immediate access at the lowest market wages to a reservoir of unskilled labor. He also may use trained Africans in positions normally occupied by coloureds, Asians, or whites, and should enjoy substantial wage savings. Firms that require large amounts of labor may thus be drawn to the homeland growth points. The average ratio of white to black workers is said to be 1:30 in the homeland growth points as opposed to 1:3 in the core area.

Three of the six homeland growth points being actively developed by the corporation are in Bophuthatswana and KwaZulu. One of these, Montshiwa, just outside Mafeking, is not yet important, but the other two have become showplaces. Babelegi is about thirty-five miles north of Pretoria and lies adjacent to major rail and road links. An African township, Temba, has been built nearby and is the site of several other corporation projects, including a cane furniture factory, a modern new shopping center, and the corporation's business training school. The development of Babelegi has proceeded rapidly since 1970, when the first ten enterprises were initiated. By March 1974 there were fifty-six agency enterprises in Babelegi, employing 5,559 Africans and 186 whites. About R10.2 million had been spent on infrastructure, buildings, and white housing by the corporation. About one-half the employment had been provided by textile, wearing apparel, and leather firms—traditionally employers of abundant, cheap, mostly female labor. Over 1,000 persons worked in metal fabrication and related firms, making such things as aluminum window frames. All sites are now allocated.^[39]

[38] BENBO, Bophuthatswana, Economic Revue, 1975, 40; idem, KwaZulu Economic Revue, 1975, 48.

[39] Idem, Bophuthatswana, Economic Revue, 1975, 38.

In Isithebe, located about sixty-five miles north of Durban on the Durban-Richards Bay rail line, complete infrastructural facilities are available at forty-nine sites. The nearness of the port of Richards Bay is considered an advantage for export industries. In fiscal 1973/74 sixteen businesses employing 983 black and 42 white workers were located there. At the end of March 1975, employment of 1,558 Africans was reported.^[40]

The interior growth point program as it is structured does seem capable of attracting a number of small- to medium-sized firms, such as clothing makers, wig makers, and metalworking and assembly shops. These centers and their associated residential enclaves are an important addition to the homeland economies, but it is unlikely that they will absorb more than a fraction of their burgeoning labor forces. Compared to the border industries, however, the growth points have substantially more linkage effects to the local economies. They provide a potential for recouping public expenditures on infrastructure, housing, health, and education if and when tax reform occurs. Ultimately the growth points may contribute to African business ownership.

It is in many ways premature to judge the efficacy of the corporation's various programs. It has grown so quickly and assumed so many responsibilities that it would be surprising had there been no setbacks. The Tswana are reputed to be dissatisfied with the corporation because of its extremely close supervision of client enterprises. It is said that one or two general dealers may have had their assets disposed of in an arbitrary fashion in order to meet loan obligations. There is the larger problem of the degree to which the homeland governments should be involved in the corporation's loan and investment decisions and in the supervision of its wider activities. Both Mangope and Buthelezi have argued vigorously for independent investment corporations, and in 1976 they had their requests honored. Both would prefer to have themselves, cabinet ministers, and some homeland officials on the boards of their investment banks, but the Bantu Investment Corporation has acceded only in allowing the homelands to name one-half of the board of directors. Political appointments will be discouraged. Yet, strong links still need be built between the public allocative process and the homeland corporations so that greater coordination becomes possible and homeland politics can influence economic decisions.

The homelands are not yet permitted departments of industry and commerce, although Mangope has made a strong plea for one. Control over industrial policy still rests with white authorities, in large measure with the corporation. Except for the control that they exercise over the granting of

[40] Idem, KwaZulu, Economic Revue, 1975, 46.

"trading rights," which enable Africans to start shops or businesses, the governments of Bophuthatswana and KwaZulu have little influence over any decisions of the corporation.

In 1974 Vorster said that homeland governments would gradually acquire more say over industry and held out the possibility that at least a few white concerns could operate on an ownership, or partial ownership, basis rather than under the agency system. Tomlinson, speaking publicly on homeland policy for the first time in twenty years, lamented that it had taken that much time before the government acceded to a major recommendation of his commission.^[41]

As the Republic's primary agency for restraining the outflow of workers from the homelands, the central priority of the corporation is supposed to be employment generation rather than multifaceted development. The activities of the organization have in practice become extremely diffuse; it has been compelled to assume a host of banking, investment, and training responsibilities. In so doing, the corporation has drifted closer toward serving the balanced needs of the homelands than its creators intended. The great tragedy is that time and resources are being devoted to programs to force black development into approved channels rather than allocated in response to community wishes and needs and to build on existing strengths. Hart, who has made one of the few studies of black entrepreneurship in South Africa, found that spontaneous initiative was being quashed in the urban townships while at the same time the corporation and its sister organization, the Xhosa Development Corporation, were trying with mixed success to infuse commercial spirit into single-proprietor dealerships in rural areas.^[42]

The firms operating on an agency basis are not always committing substantial capital and new equipment. The corporation owns many of the factory sheds, no one owns the land, and equipment is being depreciated physically and financially. Whatever remains at the termination of the agency will not easily provide the basis for viable businesses. The timing of withdrawal is vague and the terms of transfer are unspecific. On its own account, the corporation is expanding rapidly and shedding enterprises only gradually. The funding of the border industries has been ten times as great as that of the growth points and the business response much stronger. Without genuine black participation at all levels, and the tying of the growth points to the infrastructure, towns, human resources, minerals, agricultural production, and markets of the homelands, they may not readily serve as springboards for more development.

[41] Survey of Race Relations, 1974, 215.

[42] Gillian Hart, Some Socio-economic Aspects of African Entrepreneurship (With Special Reference to the Transkei) (Grahamstown, 1972).

Mining

Another potential avenue for development in Bophuthatswana and KwaZulu is mining. Bophuthatswana is known to have reserves of platinum, nickel, vanadium, asbestos, iron, limestone, diamonds, chrome, granite, calcite, manganese, and fluorspar. KwaZulu may have important coal, titanium, and gold resources, but presently only dolerite, sand, kaolin, stone, and coal are being extracted.

The Bantu Mining Corporation was instituted in 1969 to guide greater white and black exploitation of homeland mineral resources. Many of the homelands had not been systematically surveyed for mineral wealth, and mineral rights were held by individual blacks, tribal communities, the Bantu Trust, and sometimes by the state. The Bantu Mining Corporation was to oversee existing operations and act as agent for individual Africans and communities in their negotiations with white companies. Only South African companies were to be permitted into the homelands. In addition, the homeland governments were not empowered to negotiate on behalf of themselves or their citizens. The Bantu Mining Corporation could undertake explorations on its own account and could establish its own enterprises or permit white firms to operate on an agency basis. It was also to encourage black ownership and to train Africans to assume control of at least some mineral firms.

The corporation has acted in its brief history principally to give white mining interests access to homeland mineral deposits. As with industry, mineral development is a vital national interest of the Republic, and the homeland governments have no say regarding mineral leasing or the terms of exploitation. In the underdeveloped world, mining is at best a classic enclave industry, employing local workers at minimal wages held down by the threat of imported or recruited migrant labor. Unless special efforts are taken by an independent government, raw minerals or semiprocessed ores are often shipped away to processing centers outside the country. There are few internal forward linkages into processing and fabrication. The homeland governments are not, however, in a position to regulate the operations of mines, nor can they easily try to develop processing and fabricating industries that would be based on ore production.

In mid-1974 there were sixty-three mines operating in the homelands; of these twenty-five were in Bophuthatswana and nine were in KwaZulu. Approximately 78,000 African workers were employed, earning an aggregate payroll of R31.8 million. Of these 61,000 or 78 percent were working in Bophuthatswana, where they earned R27.0 million.

Through 1974 the Bantu Mining Corporation had granted prospecting and mining rights to 145 applicants. It appears that these numbers do not

include firms that had been mining actively prior to the formation of the corporation in 1969. In 1973 only four Africans had prospecting rights, and one African was mining, although what he was mining has not been stated.^[43] The corporation is supposed to encourage African mining, but it is apparent from statements in its annual reports that the corporation envisages Africans as small-scale miners using labor-intensive methods on residual holdings that whites are not interested in developing. The assumption is that Africans are satisfied with lower levels of income or profits than whites. Yet the corporation explores on its own account, and, if it chooses, could presumably develop a discovery in a relatively short time for an African takeover.

There are eleven times as many workers in Bophuthatswana's mines as in Babelegi; there are eighty-seven times as many mine workers as persons engaged in commercial farming on Bantu Investment Corporation projects. KwaZulu's mineral resources employed only 307 workers, who earned R128,000 in fiscal 1973/74.^[44] The most sought after mineral in Bophuthatswana is platinum, the extraction of which engages 58,349 African miners or 95.6 percent of the mining labor force. Very few Tswana work in the three platinum mines in the homeland, and it is said that this low rate of participation is due to "cultural reasons." The real reason may be that wages in relation to working conditions are superior for most Tswana in the Pretoria-Brits-Rustenburg area. Production of platinum has risen sharply since 1972 to meet the needs of American automobile producers, who use the metal in emission control devices and who have signed contracts worth several hundred million rand with the mine operators. There are very poor linkage effects from these and other mines to the Tswana economy.

At present the Bophuthatswana government receives only indirect and marginal benefits from these ventures, which remain under the control of the Bantu Mining Corporation, an agency of the Republic, and the licensee firms. The terms of the contracts, insofar as they are known, are not favorable to the homeland when contrasted with agreements reached normally between foreign investors and independent nations. The royalties paid are small and go mostly to the Fokeng, and to the Bantu Trust. There are limited employment and income multiplier effects, because few Tswana are employed and the workers do not spend the bulk of their funds in Tswana shops or stores. The power of the white mining unions is such that they have been able to enforce job reservation in the homeland mines despite the government's general policy that discrimination will not be allowed in the homelands.

[43] Bantu Investment Corporation, Homelands, 2nd ed., 85.

[44] Ibid., 85–87.

Developmental Prospects

The development of the homelands by the Republic, acting primarily through its development and mining corporations, must overcome the cumulative disabilities of the homeland economies. Deficient infrastructures, inadequate human resources, ecological collapse, and the competitive attractions of the core area and of the border industrial sites make it exceedingly difficult to create a vigorous commercializing farm sector and profitable industries in the homelands. Further, many programs—indeed the general policy framework itself—are dictated from Pretoria and frequently are based on erroneous suppositions about African responses. Black leaders, civil servants, teachers, extension workers, businessmen, and farmers are imperfectly involved in decision making.

Without case studies it is hard to be certain when projects are yielding spontaneous and sustained results, and when they are merely being kept going by support from one of the developmental agencies. In agriculture, sugar cane growing in KwaZulu and cattle sales in both homelands appear to provide positive signs. Yet irrigation projects at Taung and Tugela have not so far been structured so that land, water, credit, and other inputs can be organized by individuals to yield secure returns in accessible markets. More importantly, widespread gains in garden and grain cropping would seem to be beyond the scope of present extension and support service levels. Too much effort is being devoted to physical planning, which is of transitory usefulness without simultaneous improvements in the extension, technological, and infrastructural spheres.

In industry, the growth points have succeeded in bringing some labor-intensive industries into the homelands, but otherwise industry is very much underdeveloped. Commercial and autonomous craft and service undertakings handled independently by blacks remain rare, and thus gains in industrial employment have limited effects on the local economy. Small shops are important, but a means has not yet been found to put large amounts of capital into African hands. There are few financial institutions to mobilize savings and take advantage of new investment opportunities, although the urbanization of the homelands would appear to offer considerable potential for blacks in construction, housing, and urban services.

In mining, the position of the homelands is peripheral and unsatisfactory. Unless there are substantial institutional changes, Bophuthatswana can only look on as its depletable resources are exhausted to enrich others.

An independent Bophuthatswana could use the extractive sector as the basis for further development, as Zambia, the oil countries of the Middle East, and—not to be forgotten—South Africa itself have done. There is a substantial conflict of interest between Bophuthatswana and the Bantu

Mining Corporation acting as the agent of the Republic and white firms. What is in practice elsewhere essentially an adversary relationship should not be left to a single agency.

The elaborate organizational system created by the Republic over the last decade or so has taken white South Africans, with their own economic interests and developmental models and their ingrained view of African culture and African personal potentials, very deep into the last enclosures of African society and economy. These thrusts in essence complete, although in a different form, a process begun centuries ago. Whether by acting in this manner whites will at long last be able to assist the Africans in the homelands more than harm them, and whether they can successfully dismantle their now burgeoning bureaucracy and disentangle themselves with grace, while ceding their powers to Africans with the necessary alacrity, are prime questions to ask in evaluating the homeland developmental policy in the years ahead. Success would mean, at most, that these small backward economic regions become able to provide more employment, higher incomes, and better living conditions in a manner to be determined by their elected political leaders. There can be no serious hope that an augmented homeland development program will effect a major restructuring of existing South African economic and employment patterns; therefore, success within the context of this policy depends upon a recognition that the economic needs of African residents in urban South Africa will have to be met there and not in the homelands.