and processing exportable crops. A second was the transportation workers’ union, whose members transported the crops from the agricultural regions to the port cities of Dar es Salaam, Tanga, and Mtwara. The third was the dockworkers’ union, whose members then loaded the country’s agricultural exports onto ships. Taken together, the membership and influence of these unions far outweighed those of workers in the industrial sector. Policies that diminished the well being of export-oriented farmers would inevitably have a similar effect on these workers as well. The farmer-worker coalition, though unusual in that it jointed together employer and employee, constituted a powerful opposition to policies that imposed costs on the agricultural sector.
Bates’s urban bias approach reversed cause and effect. As a matter of historical sequence, urban industrial workers and managers did not become a well-organized political force until state-led industrialization was well underway and until some of the larger publicly owned industries, such as wearing apparel, footwear, brewing, and cigarettes, had grown to the point where they employed a large workforce. In other words, the industrial unions that became a powerful force for the policy bias against agriculture only emerged after the industrialization policy had begun and after the factories that employed large numbers of industrial workers had completed their hiring and begun operations. For Tanzania and a host of other African countries, then, the presence of powerful industrial unions does not explain why the policy of taxing agriculture to fund industries first began. However, it does help explain why this policy remained in place so long after its harmful effects had become visible.
The key question, then, suggests itself: why did Africa’s leaders adopt a set of policies so harmful to the development of their societies? The most persuasive answer is the influence of prevailing economic ideas. During the roughly two-decade period from the mid-1950s through the mid-1970s, innumerable African governments derived their policy preferences from a subfield of economics known as development economics.40 Tanzania was among these, and its policy choices throughout these two decades reflected the profound influence of this field. The development economists were intellectually joined by their mutual concern with the question of how primarily agricultural societies might best achieve sustained economic growth. They were pessimistic about the growth benefits of free markets as the best economic model for developing countries; they were pessimistic about the possibility that the agricultural sector might provide the basis for broad-gauged growth. In addition, they had doubts about whether free trade would provide sustainable development for countries dependent upon the export of primary commodities. Their economic strategies derived from these views. They believed industry, not agriculture, should have the highest priority; they believed protectionism, not free trade, would help industry to develop; and they believed the best use of the agricultural sector was as a resource base to provide the input needed to create industries.
The most widely discussed basis for the development economists’ agricultural pessimism was the notion of falling terms of trade for agricultural products. Many of these economists believed and tried to show that the prices developing countries received for their agricultural exports would tend to fall relative to the prices of the industrial products and consumer goods they needed to import. To the extent that this was true, rising levels of agricultural exports could only sustain a constant level of imports, and this would cause each producing country to seek to increase its export levels.41 Agricultural exporters would find themselves in a repetitive cycle of diminished well-being because the downward pressure on agricultural prices would cause them to fall relative to the costs of industrial imports. The difficulty in increasing agricultural prices at a pace commensurate with price increases for industrial goods had to do with the ease with which second- and third-party producers could enter global agricultural markets: it was far easier to capitalize and operate a new coffee plantation than a new automobile industry. The validity of this idea continues to be an unresolved topic of discussion among economists.42 But so long as it held intellectual sway, it discouraged any tendency to view agriculture as a possible source of robust economic growth.
The development economists were convinced that the best use of the agricultural sector was as a source of resources, such as capital and labor that could be invested in industry. To reinforce this viewpoint, they assembled an entire repertoire of concepts that cast doubt on the wisdom of investing in the agricultural sector. Among other ideas, for example, the development economists stressed the low marginal productivity of labor in agriculture; the idea that labor productivity in agriculture would be less, at the margin, than labor productivity in new industries. The ideas of W. Arthur Lewis were especially influential, and his most famous research sought to show that agriculture could provide a labor supply for industry without adverse effects on agricultural production.43 Benno Ndulu, today governor of the Central Bank of Tanzania, has stated Lewis’s idea succinctly: “Lewis’ seminal paper on the dual economy provided the rationale for perfectly elastic supply of labor released from agriculture where its marginal product was zero or released at a constant, institutionally set wage below the industrial wage.”44 Those responsible for framing economic policy in Tanzania and elsewhere interpreted Lewis’s research to mean that they could siphon labor out of agriculture without adverse effects on the agricultural sector and deploy that labor to industrial production, where its contribution to economic growth would be much greater.
One of the most influential of the development economists’ ideas about agriculture was the so-called backward bending supply curve of agricultural production. According to this notion, smallholder farmers in developing countries did not behave as income maximizers in the traditional economic sense, that is, by increasing their marketed production in response to favorable price incentives and then decreasing their production when prices fell. Instead, the development economists portrayed smallholder farmers as target workers whose participation in the marketplace was motivated by the need to have a certain level of cash income for specific purposes, such as to pay school fees for children, to purchase a bicycle or concrete flooring, or to pay local taxes or medical fees. Once farmers acquired that amount of cash, they would stop producing for the market and withdraw into a local economy of subsistence and barter. If this theory was correct, it meant that farmers benefiting from a price increase might well produce less of the good rather than more since their cash needs would be satisfied with a lower marketed volume. Similarly, if the farmers’ price for a good were to decline, they would have an incentive to produce more of the good, rather than less, in order to attain the required level of cash income.
This concept provided the theoretical foundation for policies that increased taxes on farmers. What the development economists appeared to be saying was that governments could adopt tax measures that lowered the producers’ net return on an agricultural good without risking a falloff in production for the marketplace. This idea became a fundamental conviction among many of the economists and policy advisors involved in the development process and set the stage for the post-independence era of escalating taxes on agricultural producers. Convinced such taxes might actually yield higher levels of production, many African countries began to impose a whole set of additional taxes on agricultural products. In Tanzania, these additional taxes began with agricultural exports but quickly embraced food staples as well.
During the 1950s and 1960s, the development economists’ prescriptions exercised a powerful influence over the policy choices of a host of developing countries, not only throughout sub-Saharan Africa but also in Latin America and South and Southeast Asia. In conception, the ISI strategy of development was simple: it reinvents the old idea of the infant industries approach that was popular in the early industrial history of the United States. The infant industries would receive an incubator of protection until they had matured into adolescent or even full-grown industries, able to stand on their own in a competitive world. Both the ISI strategy and the infant industries approach emphasize the benefits of substituting domestic industrial production for imports, and both emphasize the need to provide the new industries with trade protection during their formative phase to insulate them from international competition by more mature firms. To political leaders throughout the developing world, both in sub-Saharan Africa and in other developing regions, the key challenge for policy-makers was to find ways to transfer productive resources from agriculture to industry.
