Urbanization is a by-product of economic development. The urban population is rising faster than overall population growth even in those Asian countries with abundant land resources (Human Development Report 1999: p. 231-234).
As countries develop, urban areas account for an increasing share of the gross national product (GNP). The growth sectors of an economy, particularly manufacturing (including food processing) and services, are generally located in cities where they benefit from agglomeration economies, ample markets for inputs and outputs and readily available labour. These urban agglomerations are also areas where ideas and knowledge are rapidly diffused. According to Shukla (1996) productivity rises with city size, e.g. a typical firm will see its productivity climb between 5 and 10 percent if city size and scale of local industry double.
Although the definition differs, most countries call settlements between 2,500 and 25,000 people urban areas. Regardless of the criteria used, the number of people living in large cities is on the rise (World Development Report 1999/2000). Very often, the majority of the urban population lives in the capital city, e.g. Bangkok, Manila, Jakarta.
The share of agriculture in GDP is declining as a result of higher overall growth rates in the manufacturing and service sectors. The income elasticity of demand, as a measure of responsiveness of consumers to changes in their income, is higher for non-agricultural products. It is generally lower and decreasing for food products. Hence a dollar invested in industrial development is expected to yield higher returns than one invested in agriculture. For economic reasons industrialization takes place in urban areas where the agglomeration of production factors such as labour and infrastructure as well as the output of markets generate economies of scale. The accumulation of a growing share of the population in urban agglomerations has generated a political economy where the agricultural sector became taxed by the rest of the economy (Krueger et al. 1992). Overvalued exchange rates and government administered food prices were set below world market levels (Schiff and Valdés 1995) and have generated disincentives for farmers to produce more, to innovate, to adopt new technology and to invest. The pressure that the urban population can put on governments effectively has resulted in a cheap food policy that invariably has brought about a conflict of interest between urban and rural (Lipton 1977). While the importance for coordinated complimentary investments across sectors as a substitute for inefficient subsidies has been addressed in the context of the so-called big push strategies (Murphy et al. 1989), the specific role of agriculture was not mentioned.
By and large, development policy suffers from an urban bias that has an empirical as well as a theoretical base. Empirically, food production has outpaced population growth resulting in declining food prices. On the theoretical side, the root for a bias toward agriculture is the Arthur Lewis Dual Economy Theory (Lewis 1958). His model is based on the assumption that the major role of the agricultural sector in a developing economy is to supply surplus labour to a growing industrial sector. This theory relies on the perception that agriculture is characterized by inefficiency and low labour productivity. Investing in agriculture was regarded as investment in poverty. Modernizing agriculture on the other hand was assumed to require industrialization first. Only through the process of industrialization would the traditional equity-based wage of a feudalistic agricultural society (Schäfer 1983) be replaced by an economic price for labour, i.e. one based on supply and demand in the labour market of the industrial sector (Ranis and Fei 1961). While the positive contribution of agriculture in early phases of economic development was recognised, low prices for agricultural products relative to industrial products were believed to be a necessary pre-condition for rapid industrialization (Schäfer 1983). Food prices in countries where incomes are low are wage goods, i.e. as people spend a large share of their income for food, the price of food determines their true earnings.
While the early industrialization strategy worked well in some countries it failed in others. To date, there are large differences among Asian countries. The share of agriculture in GDP has declined in all countries taken into consideration in this paper (World Bank 2000). However this decline does not correspond very clearly with the overall socio-economic well being of a country as expressed by the Human Development Index (UNDP 1999).
It is now clear that the dual economy model is too unspecific for designing policy recommendations (Bhadra and Brandao 1993). A policy of protecting a growing manufacturing and service industry on the one hand (infant industry-argument) while taxing agriculture on the other was not always effective in reaching the dual purpose of raising consumer income and enhancing agricultural productivity. Government programmes to compensate farmers for low output prices through input subsidies for seeds, fertilizer and pesticides in several instances (e.g. the Philippines, Indonesia) have failed to achieve food security and have resulted in significant negative externalities. For example, Rola and Pingali (1993) established that farmers in Philippine rice production experience health costs at a ratio of 1:1 to their expenditures on insecticides.
The reliance on external chemical inputs and the promotion of monoculture has not only led to natural resource degradation and environmental damage but has also contributed to a negative image of the farming community. Farmers are often blamed for pollution of water bodies, erosion and forest encroachment. They are sometimes misused as an easy scapegoat for governmental policy failures. For example, farmers in the mountain areas of Northern Vietnam grow upland rice for household food security under a swidden agriculture system (Pemsl 2000). These swidden agriculturalists are not only blamed for deforestation but are also being accused as the cause for low rice prices during a period of growing national rice production (Pandey 2000).
Another lesson learned from the now outdated dual economy paradigm is that food prices are an insufficient indicator of food security. Although food prices are low to date, having decreased by 50 percent in real terms between 1960 and 1990 (McCalla 1998), there is no decline in absolute poverty measured in income terms. To date, an estimated 1.2 billion people live on less than one dollar per day and almost three billion have less than two dollars a day (World Bank 1999). Many of these people are unable to benefit from lower food prices and the increase in agricultural production. Sen (1981) showed that famines happen despite high aggregate food supply. Apparently the market is not able to solve this problem. Hunger in a broader definition, i.e. when including all kinds of social and biological disadvantages associated with inadequate food intake (Drèze and Sen 1989) requires public action that goes beyond food production. The lessons learned from misguided development interventions during the past provide some hint about how the process of urban-rural relations in the context of food supply and sustainable development can be efficiently steered.