CHAPTER 1 - A CHANGING WORLD

1.1 A Dynamic Rice Economy
1.2 Evolving Research Issues
1.3 Growing Concern for the Environment
1.4 A Changing Research Climate

1.1 A Dynamic Rice Economy

Rice is the most important food crop in developing countries. It is the major staple for 2.7 billion people in Asia, providing between 35 and 60 percent of the calories they consume. In sub-Saharan Africa and Latin America, rice provides approximately 8 percent of the food energy, and is a relatively new staple in the diet. Demand for rice in these regions is increasing rapidly. In West Africa, per capita consumption has doubled over the past two decades, and now accounts for almost 20 percent of calorie intake. In Latin America it increased by about 15 percent. Rice is a relatively minor crop in developed countries, which together account for less than 5 percent of global production.

Rice production is concentrated in Asia which accounts for 93 percent of output and 90 percent of rice area cultivated in developing countries (Table 1.1). China alone accounts for 23 percent of the rice area and 38 percent of rice output. Latin America and the Caribbean (LAC) account for 3.9 percent of rice production, sub-Saharan Africa (SSA) for 1.6 percent and West Asia-North Africa (WANA) for 1.1 percent. The world's biggest producers are China, India and Indonesia. Total production of sub-Saharan Africa approximately equals that of the Republic of Korea and is only half of that of Vietnam.

Rice is grown in four major production environments or "ecosystems" which are broadly defined on the basis of water regime; irrigated (IRR), rainfed lowlands (RFL), uplands (UPL), and deepwater/tidal (DWT). Irrigated areas account for 71 percent of rice output, the rainfed lowlands for 19 percent, uplands for 7 percent and deepwater/tidal areas for 4 percent (Table 1.2). Table 1.2 presents an overview of area, yield and output of rice by ecosystem. A breakdown of this information by region is provided in Appendix Table A.1.

Table 1.1: Area, Yield and Output of Rice by Region, 1986-88

Region		Area	Share of Area Total	Yield	Output	Share of Output Total
		m.ha	%	t/ha	m.t.	%
Asia		128.5	90.3	3.37	433.2	93.4
	of which China	52.7	22.9	5.35	174.7	37.7
SSA		5.1	3.6	1.49	7.6	1.6
	of which West Africa	3.1	2.1	1.43	4.5	1.0
LAC		7.9	5.5	2.32	18.4	3.9
WANA		0.8	0.6	5.72	4.8	1.1
	Developing Countries	142.3	100.0	3.26	464.0	100.0
	Developed Countries	1.7	-	5.42	9.5	-
World Total		144.0	-	3.29	473.5	-

Source: IRRI, Rice Facts, November 1990

Table 1.2: Area, Yield and Output of Rice by Ecosystem: Developing Countries 1986-88

Ecosystem	Area	Share of Area Total	Yield	Output	Share of Output Total
	m.ha	%	t/ha	m.t	%
IRR	67.5	47.4	4.88	329.7	71.1
RFL	39.4	27.6	2.20	86.5	18.6
UPL	23.6	16.6	1.28	30.2	6.5
DWT	11.9	8.4	1.49	17.7	3.8
Total	142.3	100.0	3.26	464.1	100.0

Source: CIAT (1992) on the basis of IRRI statistics

Note totals in this and subsequent tables may not add exactly due to rounding.

Average yields vary widely among regions, countries and ecosystems. In 1986-88, they ranged from 5.72 t/ha in WANA to 1.5 t/ha in sub-Saharan Africa, and from an average of 4.88 t/ha in irrigated systems to 1.3 t/ha in upland areas. Developing countries with the highest average yields at that time were Korea, DPR (7.1 t/ha), Republic of Korea (6.4 t/ha), Egypt (5.7 t/ha) and China (5.3 t/ha).

Over the past few decades, the nature of growth in rice production has changed significantly. Since the 1960s, the major source of output growth has shifted from expansion in crop area to increase in yield (David, 1991). Overall, adoption of modem rice varieties, higher application of fertilizers, and expansion of irrigated area contributed about equally to the increases in productivity in South and Southeast Asia. Annual growth rates of rice production, area and yield by region are summarized in Table 1.3.

Table 1.3: Annual Growth Rates of Rice Production, Area and Yield: 1950-1989 Percent Per Annum

Region	1950-65			1965-80			1980-89
	P1	A1	Y1	P	A	Y	P	A	Y
Asia	2.2	1.1	1.1	2.9	0.9	2.0	2.7	0.2	2.5
		(50.0)2	(50.0)		(31.0)	(69.0)		(7.0)	(95.0)
Sub-Saharan Africa	2.6	0.9	1.7	3.2	2.8	0.4	3.0	1.3	1.7
		(35.0)	(65.0)		(88.0)	(12.0)		(43.0)	(57.0)
Latin America	5.6	5.3	0.3	3.7	2.5	0.9	2.4	-0.4	2.8
		(95.0)	(5.0)		(74.0)	(26.0)		(-17.0)	(117.0)
World	2.5	1.2	1.3	2.9	1.0	1.9	2.5	0.2	2.3
		(48.0)	(52.0)		(38.0)	(62.0)		(8.0)	(92.0)

¹ P = Production; A = Area; Y = Yield
² Figures in parenthesis are percent of growth in rice production due to changes in area and yield

Source: David (1991) on the basis of FAO data

During the last decade, the rate of growth of yield increments has slowed down considerably in the favourable areas of Asia. Overall, there has been a slow down in production from 2.8 percent (1974-82) to 2.0 percent (1982-1989), mainly due to the decrease in yield growth from 2.6 percent p.a. to 1.8 percent during these two periods (David and Rosegrant, 1991). In all major producing areas of Asia, with the exception of India, the rate of growth in yields has fallen in the last five years. As indicated in the Fourth External Review report of IRRI (1992), irrigated area yields are not only constrained by a technological ceiling but that ceiling itself may be slowly coming down. A further factor contributing to the slowdown in yields has been the reduced investment in irrigation (Rosegrant and Pingali, 1991).

While rice consumption is increasing steadily, the rate of growth of demand for rice is declining in Asia. This is due both to decreases in growth rate of population, and declining per capita consumption of rice. Sheer population growth accounted for 80 percent of increases in total consumption, but in the three largest rice consuming countries (India, China and Indonesia) rice consumption continued to grow by 0.8 to 1.2 percent per year in the 1980s. In West Africa, overall consumption of rice grew by 5.9 percent p.a. between 1970 and 1989 and by 2.7 percent on a per capita basis (WARDA, 1992). Imports of rice now amount to approximately 1 million tons of paddy annually. In Latin America per capita consumption increased by 50 percent between 1950 and 1990 (Sanint, 1992).

Between 1986-88 and 2030, production in developing countries will need to more than double from 464 to 1,000 million tons to meet expected demand. This projected increase in demand is estimated more precisely in Chapter 3.

Most of the rice output is consumed within the country or region where it is produced. Only about 4 percent of world production is traded internationally.

However, differential growth rates in rice production and demand across regions have resulted in changing patterns in the regional distribution of world trade. The share of Asia in rice imports declined from 60 percent in 1970 to 22 percent by 1987. In contrast, that region now accounts for two thirds of exports. Thailand is the leading exporter, accounting for 40 percent of world trade. Among exporters, this country is the lowest cost producer of rice (Yap, 1991). In 1987, sub-Saharan Africa accounted for 23 percent of all imports, the Middle East for 19 percent and Latin America for 10 percent (David, 1991).

Given the small volume of world trade relative to production, world market rice prices show strong fluctuations. Over the past two decades, real price of rice has continued a downward trend, partly because of increasing self-sufficiency ratios of major traditional rice importers (Siamwalla and Haykins, 1983). It is a remarkable feature of the Asian rice economy that growth in rice production has been sustained despite these declining prices. Domestic prices of rice which are usually well above world prices have also declined in several Asian countries. While this has benefited consumers in terms of lower prices, it may also imply reduced incentives to producers, particularly those for whom new technology is not available.

If the declining level of real rice prices continues in the future, the sustainability of the past rapid productivity growth may be threatened.

Overall, government policies have not been favourable to rice production. David and Rosegrant (1991) conclude that domestic rice production in Southern and Southeast Asia, except in Japan, South Korea and Indonesia has been generally lower than it would have been without policy distortions. In sub-Saharan Africa, particularly in West Africa, imports were encouraged by governments through low import tariffs and low farmgate prices for domestic production. In effect, domestic producers were taxed in order to subsidize foreign producers.

The ability of the rice production sector in part to meet growing demand through increased production and trade will depend on a climate of economic liberalization. Reduced government interventions will typically stimulate exports, encourage technological change and allow for increasing investments in the irrigation sector.

1.2 Evolving Research Issues

The recent history of rice research in developing countries logically begins with the establishment of IRRI in the early 1960s, against a background of imminent food crises in Asia. The initial research objective was very specific: to increase crop yields in the high-potential, irrigated and favourable lowland areas. This was achieved by introducing the short, erect, fertilizer responsive high-yielding plant type of the japonica rice into the tropical indicas. A rapid, dramatic breakthrough was achieved by the mid-1960s with the release of the first of the modem plant type varieties, IR8. The non-photosensitive nature of IR8 made it widely adaptable across the favourable environments and it became the basis of the green revolution.

Following the release of IR8 with its high yield ceiling, attention shifted to utilizing more fully that potential by improving resistance to biological and physical stresses, together with enhancing grain quality. Noteworthy landmarks have been the release of IR36, an extremely popular variety in many areas, with high multiple disease and pest resistance, and more recently IR64, already the most widely grown rice in the tropics. A further important feature of many of the new varieties, has been earlier maturity without loss in yield potential. Growth durations as short as 100-110 days have provided increased opportunities for growing additional crops of rice, or other crops, during the year. By the 1980s, primarily because of these new varieties combined with the management practices developed for them, many Asian countries approached or became self-sufficient in rice. Even more remarkable has been the emergence of countries such as Vietnam as rice exporters.

Latin America and the Caribbean has experienced a similar green revolution in rice, starting with the introduction of IR8 in the late 1960s. Irrigated yields doubled between 1966 and 1980, and LAC production between 1966 and 1990. In contrast, there has been limited spread of modem rice varieties in sub-Saharan Africa. Reasons for this include the low proportion of irrigated rice; the area-specific pest and disease problems such as diopsis fly, rice yellow mottle virus, and unique races of blast; the complex nature of the farming systems in which rice is grown in rainfed environments; the poor development of the agricultural infrastructure, and policies that favour rice imports.

IRRI has systematically collected and characterized rice germplasm, assembling 80,000 accessions from 113 countries by 1992. It has also facilitated the worldwide evaluation and dissemination of genetic material through international networks, initially the International Rice Testing Programme (IRTP), latterly the International Network for Genetic Evaluation of Rice (INGER). By 1992, these networks had led to the release of 225 new varieties and breeding lines (Evenson and Gollin, 1991). The proportion of the total rice-growing area under modem varieties had exceeded 50 percent by the early 1980s (Dalrymple, 1986). By 1987, some proportions in South and Southeast Asia were India 69 percent, Indonesia 75 percent, and the Philippines 85 percent (David and Rosegrant, 1991).

Following this impact in the more favourable environments, research in recent years has given more explicit attention to poorer environments. The need for this has been argued primarily on equity grounds. Although there have been benefits from green revolution technologies for poor growers in the more favourable environments (Hazell and Ramasamay, 1991) there has been little benefit in the very poor environments such as the uplands (David and Otsuka, 1990). There are also growing concerns about the degradation of poor environments. Further factors are the declining investment in irrigation, the evidence of degradation in existing irrigated areas, and the evidence from very intensive rice producing areas that yields may be already approaching a plateau, or even decreasing in some instances, resulting in declining factor productivity (Rosegrant and Pingali, 1992).

Increasing recognition of the complexity of farming systems in poorer environments has strengthened the need for research on rice improvement to be set within the context of improvement of the farming system as a whole. Recent research agendas have therefore been widened to include greater attention to crop and resource management, to socioeconomic factors, including the role of women, and to the need to integrate rice with other components of the farming system.

1.3 Growing Concern for the Environment

Providing food for the world's population in 30 years time will place enormous demands on the environment. IRRI estimates (1989) that the global supply of rice will need to grow at a rate of 1.7 percent per annum, or by a massive 65 percent by the year 2020, in order to feed the increased population. The projections used in this study (Chapter 3) suggest the need to double output by 2030.

Particularly in Asia where more than 90 percent of the world's rice is produced and consumed, increases of the magnitude required will have to be achieved largely by increased production per hectare, i.e. by intensification of cropping and/or increased yield. Furthermore, such increases must be achieved without irreversible depletion or damage to the environment. The CGIAR has called for much greater attention to activities in the area of conservation and management of natural resources including germplasm conservation (biodiversity) (TAC/CGIAR, 1992).

While there is some potential for increased areas of land to be used for rice growing in other regions of the world, intensification of cropping is already taking place and will undoubtedly continue in many countries of those regions. Such intensification in the absence of sustainable technology could have serious environmental consequences. This is a particularly serious consideration in the uplands where soils are often subject to erosion and to severe nutrient depletion leading to irreversible degradation.

In irrigated and rainfed lowland areas, environmental concerns relate more to the use of high rates of fertilizers and widespread use of pesticides, with the accompanying threat to human health because of accumulation of residues in soil and water. In some regions additional ponding of water in itself increases human health risks due to diseases such as malaria and schistosomiasis.

The Panel notes that attention is being given to quantifying the contribution of rice to global warming through increased methane emission and to examining the effects of increases in ultraviolet radiation on rice yields.

The key issue is that short-term production gains should not be derived at the expense of long-term productivity of the resource base. Research in the future will need to have a clear focus on resource management and be guided by a strong farming systems perspective.

1.4 A Changing Research Climate

The global rice research system encompasses the activities of the national agricultural research systems including universities, the CGIAR, bilateral and international agencies, the private sector and advanced research institutes. The overall pattern of global investment in public agricultural research has changed dramatically over the past two decades (Pardey, Roseboom and Anderson, 1991). Global agricultural research capacity has grown substantially; most notably, the share of developing countries has increased significantly. However, investment in agricultural research in developing countries appears to be slowing recently, particularly in sub-Saharan Africa and Latin America.

Pardey, Roseboom and Anderson (1991) estimated output intensity ratios, which express agricultural research expenditures as a percentage of the value of agricultural output, for each of the regions for the 1982-85 period. Asian countries typically have values of below 0.5 percent. In sub-Saharan Africa the average was 0.5 percent, with the SADCC countries and the former French colonies in West and Central Africa well above this figure. In Latin America, average spending amounted to 0.8 percent. In contrast, in developed countries these intensity ratios are usually well above 2 percent. This illustrates the chronic under investment in agricultural research in developing countries. Moreover, in most developing countries, a declining trend in public expenditures for agricultural research has been observed during the last decade (Roe and Pardey, 1991).

The bulk of rice research in developing countries is undertaken by the national research systems, which typically focus on adaptive and applied aspects of research. However, several of the major rice dependent countries already have strong, independent rice research programmes. Countries like China, India and Brazil have considerable scientific strengths.

Among the bilateral agencies, the major actor in rice research has been France, which has a major involvement particularly in sub-Saharan Africa, but also in Latin America and Southern Asia. In addition, France has close linkages with the CGIAR centres and has staff posted at IRRI and WARDA. The French rice research efforts are focused principally on the upland, highland and temperate (Mediterranean climate) areas. In total, their current international rice research programmes involve 45 scientific staff (de Lattre, personal communication, 1992 - French Ministries of Cooperation and of Research, 1986). This is a significant element of the total research effort on rice. While international agencies like FAO and the World Bank do not have rice research programmes per se, they provide financial and technical assistance at the development project level. Both also have an important function in policy analysis while FAO is the major source of data relating to rice production, consumption and trade.

Basic rice research is conducted primarily at advanced research institutes primarily in developed countries. For example, many of the biotechnology tools and basic research on physiology and ecology are being developed at universities in USA, Japan, France, Italy, Germany, The Netherlands and Belgium. The Rockefeller Foundation has made substantial contributions through the sponsoring of its rice biotechnology network which has led to substantial strengthening of national capacities in this field.

Prior to World War II, rice research was primarily concentrated in Japan, South Korea and Taiwan where population pressure was severe and almost all of the rice area irrigated, from early this century. As rice self-sufficiency became an objective in many newly de colonized countries, efforts to raise productivity increased and investment in rice research accelerated during the late 1960s and 1970s (David and Evenson, 1992). Around that time, the CGIAR was established and initially invested about 22 percent of its resources in rice research. Since 1976, the share of CGIAR core research operating expenditures for rice research has remained constant at just above 17 percent (Gryseels and Anderson, 1991). Technological breakthroughs in international rice research induced greater investments in national agricultural research by demonstrating that rice research can be an important source of growth (Pray, 1991).

Table 1.4: Distribution of Agricultural GDP Population, Poor, Rice Production, and CGIAR Expenditures Among Developing Countries: By Region

Region	Agricultural GDP	Share of Population	Distribution of Poor	Share of Rice Production	Total CGIAR Funding
	(%)	(%)	(%)	(%)	(%)
Asia (excl. China)	26	40	53	56	29
China	41	29	19	38	0
SSA	8	12	16	2	43
LAC	15	11	6	4	15
WANA	9	8	5	1	13

Source: Gryseels and Anderson (1991); TAC/CGIAR (1992); IRRI (1991).

Table 1.4 presents data on the distribution of agricultural GDP, number of poor, rice production and CGIAR core expenditures by region and in China. The data suggest that Asia and China are dominant in terms of agricultural GDP, share of population, number of poor and share of rice produced. The data suggest that the CGIAR could well be under investing in Asia, particularly if the needs of China are taken into account. TAC has already recommended a moderate increase in the priority the CGIAR currently allocates to Asia and a corresponding reduction in the priority of sub-Saharan Africa (TAC/CGIAR, 1992).

The CGIAR currently allocates just over US$39 million to rice research annually. This represents approximately 19 percent of the System's core resources allocated to commodity research during 1991 (TAC/CGIAR, 1992). The share of the CGIAR in total resources allocated to rice research in the global rice research system for developing countries can be estimated at approximately 6 percent. This was calculated as follows. Pardey and Roseboom (1991) have estimated that total expenditures of national research systems on crops research amount to US$2,480 million a year in 1980 purchase parity prices (PPP). Assuming that these funds were allocated in proportion to the share of the value of production, rice would have received approximately 23.6 percent of total expenditures (TAC/CGIAR, 1992). The allocation of national research systems would then amount to US$585 million a year in 1980 PPP. According to the Agreement of Cooperation between WARDA and the CIRAD, INRA and ORSTOM Group signed on 26 June 1992, the latter three organizations in 1990 devoted 45 man-years to rice research in some fifteen countries in Europe, Africa, Asia and Latin America. This would represent an investment of approximately US$16 million annually. This estimate does not include contributions by other organizations in France or of sub-contracts by the Ministry of Cooperation. Assuming that advanced research institutes and other agencies assign a total of approximately US$10 million to rice research, then total public sector expenditures would amount to US$585 million (NARS) plus US$16 million (France) plus US$10 million (others) plus US$39 million (CGIAR), or a total of about US$650 million annually. The share of the CGIAR in total public sector expenditure for rice research in developing countries would then amount to approximately 6 percent. This estimate does not include contributions to rice research by the private sector.

This contribution of US$39 million in 1991 by the CGIAR only relates to direct core expenditures for commodity-related research by IRRI, WARDA, CIAT and IITA. It does not include expenditures by IFPRI on policy research, by IBPGR for genetic conservation of rice plant genetic resources, by IIMI for rice-related irrigation research or by ISNAR for strengthening of capacity for rice research in national systems. The current organization of rice research in the CGIAR, particularly with respect to the commodity improvement aspects, is summarized in Appendix D. The allocation of CGIAR core expenditures for rice by region and by ecosystem is examined in detail in Chapter 3.

The growth of research expenditures appears to have been declining both at the levels of the national and international systems. In recent years, the availability of CGIAR funds has been constant, declining in nominal and real terms since 1991. With both national and international funding for research coming under increasing pressure, the need is even greater to ensure that those funds are efficiently allocated.