Agriculture growth over the next decade will have to depend on increasing yields per acre rather than by bringing more land under cultivation. This means that the growth of input productivity (seeds, water, fertilizers) would have to accelerate. Yet the evidence shows that the growth of factor productivity in agriculture has been declining. It is crucial for Pakistan's future to reverse this trend and place agriculture on the path of sustainable growth. In the pursuit of this objective, agriculture research and extension could play a key role. In this article we will analyze the imperatives, and identify both the issues, as well as the institutional framework for more effective agriculture research.

It is well known that high yielding varieties of seeds introduced at a point in time, gradually lose their potency through reuse, changing micro structure of soils, and changing ecology of micro organisms in the top soil. Therefore, breeding of more vigorous seed varieties adapted to local environmental conditions and their diffusion amongst farmers is required through an effective research and extension program.

At the moment, there is no organized seed industry in Pakistan to meet the needs of farmers for the supply of vigorous varieties of seeds for even the major crops. The existing institutional framework for agriculture research suffers from a proliferation of research institutes, which are inadequately funded, often lack professional expertise, proper equipment and the research environment necessary to produce significant results. Finally, there is considerable overlapping of research responsibilities across institutes. Consequently, research has by and large failed to produce operationally usable results much less increase input efficiency. Seed varieties research where it has produced new seeds has involved a large time lag in getting research to the farmers. In wheat, for example, the average age of seed varieties is 11 years compared to 7 years for all developing countries. It has been shown that there was a sharp decline in growth of total factor productivity in Pakistan after 1975. According to a study by Rosegrant and Evenson Pakistan's lower factor productivity growth compared to India can be attributed to the poorer level of research and extension in Pakistan compared to India.

A new dimension to the imperative of improving research capability in the crop sector is indicated by the possibility of declining yields per acre related with global warming. According to Qureshi, Ata and Iglesias, given the sensitivity of wheat seed to temperature increase, even a 2-degree centigrade increase in average summer temperatures could mean an absolute yield decline of between 10 to 16 percent during the 21st century. With a 2.8 percent population growth, even a decline of 5 percent in yield per acre associated with global warming, could mean serious food deficits for Pakistan. It is, therefore, necessary to develop heat resistant varieties of food grains.

The current ineffectiveness of agriculture research and poor diffusion amongst farmers is a cause for concern. This is particularly so in a situation where future agriculture growth and labour absorption will have to depend more on input efficiency than on enlargement of irrigated acreage and input intensification which were the major sources of agriculture growth in the past.

The following issues may be identified for research and subsequent diffusion:

i) Development of (a) new heat resistant varieties of wheat, (b) short duration and late sowing varieties of wheat to enable multiple cropping, (c) heat resistant, pest resistant, short duration and high yielding varieties of cotton, (d) development of short duration and high yielding varieties of pulses and oilseeds to fit into cropping pattern of late Rabi/late Kharif crops.

ii) Rehabilitation of agriculture in saline and water logged soils.

iii) Improvement of barani agriculture by research into designs of low-cost rainwater reservoirs and chalk dams for water-run off, for irrigation and soil conservation.

iv) Efficient harvest and post harvest handling technology for different crops.

v) Development of hybrid seeds of Maize, Sorghum, Sunflower and other crops.

On the basis of my wide ranging consultations with farmers in the Punjab and Frontier provinces as well as officials concerned with agriculture research during the last two years, it can be proposed that for more effective research and its diffusion amongst farmers it may be necessary to take the following initiatives:

i) Restructure the existing institutional framework for seed development in particular and agriculture research in general to (a) bring greater professionalism, (b) improve monitoring and evaluation of research work.

ii) Focus research on improving input productivity at the region specific level. In this context the outcomes of controlled experiments at research institutes need to be tailored to the specific needs and conditions of farms in the region.

iii) Maintain regular interaction between research workers and farmers in the field by establishing an institutional linkage between research and village level institutions for diffusion of such research. Village level community organizations of farmers for participatory development have demonstrated the ability for a fruitful dialogue with research and extension institutions in the government and non-governmental sector.

Such interactions amongst farmers and agricultural researchers could redirect research to the specific needs and conditions of farmers and also enable a more rapid adoption of improved seeds and agricultural practices. Support organizations such as the PRSP in the Punjab (during August to October 1998), and AKRSP in the northern areas (during the early 1990s) have demonstrated an ability for facilitating the adoption of new knowledge and agricultural practices, particularly in cases where such adoption requires training and credit to farmers at the village level.