Cover Image
fechar este livroRoots and Tubers for the 21st Century - Trends, Projections, and Policy Options. 2020 Vision for Food, Agriculture, and the Environment. Discussion Paper 31 (IFPRI, 2000, 72 p.)
Ver o documento(introduction...)
Ver o documentoForeword
Ver o documentoAcknowledgments
Ver o documento1. Introduction
Ver o documento2. Trends in the Use of Roots and Tubers
Ver o documento3. Trends in the Supply of Roots and Tubers
Ver o documento4. Baseline Projections of Production and Use
Ver o documento5. High Demand and Production Growth Scenario
Ver o documento6. Roots, Tubers, and the Environment
Ver o documento7. Conclusions and Recommendations
Ver o documentoAppendix: Supplementary Tables
Ver o documentoReferences

3. Trends in the Supply of Roots and Tubers

Growth in Production of R&T

Rates of growth for area, yield, and production of major R&T have varied substantially by crop and country over the past two decades. Annual production growth for R&T during 1983-96 averaged a modest 2.1 percent in developing countries. This overall rate of growth masks significant differences across R&T and regions ranging from a rapid 8.3 percent annual rate of growth for yam in Sub-Saharan Africa, to a -6.8 percent per year decline for sweetpotato in East Asian countries other than China (Table 5). Thus, the analysis of the dynamics of R&T production requires a variety of vantage points from which to draw a comprehensive picture of the evolution of these crops.

In the last two decades, yam and potato achieved the highest annual growth rates in production among R&T in developing countries: 8.0 percent and 4.1 percent, respectively (Table 5). Yam production grew from a small base and increased largely in one region (West Africa). Cassava production grew at a more modest pace, 1.8 percent annually. Growth in sweetpotato production was flat over the period, with an initial decline in production followed by a recovery to earlier levels.

In developing countries, total production of R&T crops increased by 30 percent, from 344 million mt in 1983 to 449 million mt in 1996 (Table 6). Production increases varied substantially by commodity. The production increase was largest for yam in percentage terms. Output reached 32 million mt in 1996, a 170 percent increase over the 13-year period, albeit from low production levels. The crop contributed 19 percent to the total increase in R&T output. Sweetpotato production, on the other hand, barely increased during the same period, rising by 1.8 percent to 134 million mt in 1996. Cassava and potato contributed 33 percent and 42 percent, respectively, to growth in R&T output. Production of cassava grew by 27 percent between 1983 and 1996 to 164 million mt, and production of potato grew by 68 percent to 108 million mt.

On a per capita basis, production of R&T in developing countries increased from 99 kg in 1983 to 101 kg in 1996. Developing-country production of cassava remained virtually constant at 37 kg per capita, supported mainly by the per capita production growth in Sub-Saharan Africa. Per capita production of potato increased by 6 kg to reach 24 kg in developing countries in 1996, and yam production per capita grew by 4 kg to reach 7 kg. Yam production is significant only in Sub-Saharan Africa where it reached 56 kg per capita in 1996, up from 28 kg per capita in 1983. Between 1983 and 1996 sweetpotato production declined from 38 kg to 30 kg per capita.

Production of individual R&T tends to be highly skewed toward particular countries and regions. Figure 5 shows the locations of R&T production in 1996. More than 60 percent of global potato production was harvested in developed countries, followed by China with a 17 percent share and India with 7 percent. Potato production has been shifting back toward developing countries; they have increased their share of global output from 11 percent in 1961-63 to 37 percent in 1995-97 (Scott and Maldonado 1998). Slightly more than half the global production of cassava takes place in Sub-Saharan Africa, followed by Southeast Asia with 23 percent and Latin America with 20 percent. Sweet-potato production is concentrated in China, which has an 88 percent share of global production. Ninety-six percent of the world's yam is produced in Sub-Saharan Africa (mostly West Africa).

The importance of particular crops in specific regions and subregions can also be seen from the share of each crop in total R&T production (Figure 6). In WANA, virtually all R&T production consists of potato. Potato is also of relatively high importance in India and other South Asian countries and in East Asian countries other than China. Production of cassava plays a major role in Southeast Asia (86 percent of R&T production), Latin America (65 percent), and Sub-Saharan Africa (68 percent). Yam production accounts only for 24 percent of R&T output in Sub-Saharan Africa, but it has become increasingly concentrated in this region. China continues to dominate sweetpotato production, which accounts for almost 70 percent of the country's R&T output. Sweetpotato also plays a major role in R&T production in the rest of East Asia.

Table 5 - Annual growth rates in area planted with and production of roots and tubers by commodity and region, 1983-96


Cassava

Potato

Sweetpotato

Yam

All R&Ta

Country/region

Production

Area

Production

Area

Production

Area

Production

Area

Production

Area


(percent per year)

China

-0.53

-0.47

4.60

3.03

0.21

-0.84

na

na

1.20

0.30

Other East Asia

na

na

-1.00

-2.09

-6.77

-5.69

na

na

-3.24

-3.26

India

0.30

-1.94

5.12

3.77

-2.50

-3.33

na

na

3.35

1.73

Other South Asia

-6.64

-4.69

3.73

2.69

-4.25

-3.16

na

na

1.50

1.13

Southeast Asia

0.17

0.18

5.42

2.53

-0.73

-1.31

4.28

-0.55

0.22

-0.02

Latin America

0.80

0.02

1.95

0.32

-1.18

-1.60

2.97

1.22

1.09

0.05

WANA

na

na

4,75

2.59

3.67

3.46

na

na

4.71

2.59

Sub-Saharan Africa

3.56

2.38

0.62

0.70

1.75

2.64

8.32

4.70

4.32

2.75

Developing

1.83

1.37

4.08

2.42

0.14

-0.50

7.96

4.49

2.06

1.31

Developed

na

na

-0.83

-1.40

-1.00

-2.00

2.73

1.61

-0.84

-1.40

World

1.83

1.37

0.60

-0.07

0.12

-0.52

7.91

4.48

1.07

0.61

Source:

FAO 1999a.

Note:

na signifies no recorded production. WANA is West Asia and North Africa. Data for 1983 are averages for 1982-84 and data for 1996 are averages for 1995-97. See Table 2 footnote for regional breakdown.

a All R&T includes cassava, potato, sweetpotato, yam, and other R&T such as taro.

Table 6 - Production of roots and tubers by commodity and region, 1983 and 1996


Cassava

Potato

Sweetpotato

Yam

All

R&Ta

Country/region

1983

1996

1983

1996

1983

1996

1983

1996

1983

1996


(million metric tons)

China

3.8

3.6

27.3

48.9

114.6

117.8

na

na

147.1

171.7

Other East Asia

na

na

1.2

1.1

1.1

0.5

na

na

2.4

1.5

India

5.5

5.7

10.7

20.4

1.6

1.2

na

na

17.8

27.3

Other South Asia

0.7

0.3

2.4

3.8

0.9

0.5

na

na

4.2

5.1

Southeast Asia

36.9

37.7

0.8

1.6

5.1

4.6

...

...

43.4

44.6

Latin America

28.9

32.1

11.4

14.7

2.2

1.9

0.7

1.0

43.9

50.5

WANA

na

na

8.3

15.1

0.1

0.2

na

na

8.5

15.4

Sub-Saharan Africa

53.8

84.7

2.3

2.5

5.4

6.8

10.7

30.3

75.6

131.0

Developing

129.8

164.3

64,3

108.1

131.6

133.9

11.7

31.6

344.4

448.8

Developed

na

na

209.2

187.6

2.2

1.9

0.1

0.2

211.9

190.0

World

129.8

164.3

273.5

295.6

133.7

135.8

11.8

31.8

556.4

638.8

Source:

FAO 1999a.

Note:

Ellipses (...) signify very small values; na signifies no recorded production. WANA is West Asia and North Africa. 1983 is average for 1982-84 and 1996 is average for 1995-97. See Table 2 footnote for regional breakdown.

a All R&T includes cassava, potato, sweetpotato, yam, and other R&T such as taro.


Figure 5 - Location of root and tuber production, 1996 - Potato


Figure 5 - Location of root and tuber production, 1996 - Cassava


Figure 5 - Location of root and tuber production, 1996 - Sweetpotato


Figure 5 - Location of root and tuber production, 1996 - Yam


Figure 5 - Location of root and tuber production, 1996 - All roots and tubers

Source:

FAO 1999a.

Note:

1996 is average of 1995-97. See Table 2 footnote for regional breakdown.


Figure 6 - Relative importance of major roots and tubers in countries and regions, 1996, based on production volumes

Source:

FAO 1999a.

Note:

1996 is average of 1995-97. See Table 2 footnote for regional breakdown.

Sources of Growth in Output

Expansion in area and higher yields have contributed almost equally to output growth for R&T during 1983-96. Increase in area accounted for 57 percent of total output growth and yield improvements accounted for the remaining 43 percent. The role of area expansion as a source of output growth is significantly larger for R&T than for other major food crops. This is due, in part, to the location of a significant share of the harvested area for R&T in Sub-Saharan Africa. Twenty-six and 34 percent of total R&T area could be found in this region in 1983 and 1996, respectively. Other factors contributing to this evolution include the relatively low investments in agricultural R&D for these crops compared to investments in wheat, maize, and rice, for example. The greater adaptability of some R&T to marginal areas, combined with their flexible growth cycle, also facilitated area expansion in some countries and regions.

Globally, the area harvested in R&T increased during 1983-96 from 45.8 million hectares to 49.5 million hectares. The largest expansion of R&T area occurred in Sub-Saharan Africa, from 11.9 million hectares in 1983 to 16.9 million hectares in 1996. The distribution of area planted in R&T across regions largely coincides with that of production and use levels because only a small proportion of R&T output is traded internationally.

Sub-Saharan Africa

In Sub-Saharan Africa, the increase in cassava output has been driven largely by growth in area planted. A synthesis of the results from the comprehensive COSCA study found that the most important reasons for farmers to increase cassava production are famine, hunger, and drought (Spencer and Associates 1997). Cassava's low input requirements, a trait that fits well with the region's resource endowments (relatively abundant land, relatively scarce labor), make it suitable for the difficulties African farmers face. The shortage of chemical inputs and organic matter and the limited irrigation facilities in the region also make cassava a crop of choice for African farmers. Moreover, as farm size shrinks under population pressure, food requirements per hectare of land cultivated rise, increasing the prospects that farmers will shift to crops with higher output of energy per hectare as one strategy for overcoming hunger.7 Food shortages precipitated by a combination of political and civil unrest, wars, economic stagnation, erratic rainfall patterns, and rapid population growth have had a much greater influence on R&T production in this region than anywhere else.

7 Ruthenberg (1980, 361) predicted a shift in cropping patterns to more energy productive crops as farm size in developing countries declined and population growth pushed up food (that is energy production per hectare) requirements per given land area.

Another important reason for cassava's burgeoning presence in Sub-Saharan Africa is the crop's resistance to pests and diseases (Spencer and Associates 1997). Higher prices, increased market access for farmers, and higher yields have also played a role in cassava's emergence as a cash crop in much of the region (Nweke 1992). This commercialization of the crop is particularly significant, given that the share of the urban population is expected to increase from 30 to 50 percent by 2020 (FAO 1998b).

Table 7 - Yields and annual growth rates in yield for roots and tubers, 1983-96


Cassava

Potato

Sweetpotato

Yam

All R&Ta


Yield

Growth

Yield

Growth

Yield

Growth

Yield

Growth

Yield

Growth

Country/region

1983

1996

rate

1983

1996

rate

1983

1996

rate

1983

1996

rate

1983

1996

rate


(mt/ hectare)

(percent per year)

(mt/hectare)

(percent per year)

(mt/hectare)

(percent per year)

(mt/ hectare)

(percent per year)

(mt/ hectare)

(percent per year)

China

15.6

15.5

-0.06

11.3

13.8

1.52

16,7

19.1

1.06

na

na

na

15.3

17.1

0.89

Other East Asia

na

na

na

13.3

15.4

1.12

20.0

17.2

-1.15

na

na

na

15.8

15.9

0.02

India

17.5

23.5

2.28

14.0

16.5

1.30

7.4

8.3

0.85

na

na

na

13.7

16.9

1.60

Other South Asia

11.5

8.8

-2.05

9.7

11.1

1.01

10.4

9.0

-1.12

na

na

na

10.1

10.6

0.36

Southeast Asia

12.1

12.1

-0.01

9.1

13.1

2.82

6.2

6.7

0.59

2.6

4.7

4.85

10.7

11.0

0.24

Latin America

10.8

12.0

0.78

11.1

13.7

1.63

7.2

7.6

0.43

6.7

8.4

1.73

10.3

11.8

1.04

WANA

na

na

na

14.8b

19.4b

2.10

21.8b

22.4b

0.21

na

na

na

15.0

19.5

2.06

Sub-Saharan Africa

7.1

8.2

1.15

6.1

6.0

-0.08

5.0

4.5

-0.86

6.4

9.9

3.46

6.4

7.8

1.53

Developing

9.3

9.9

0.46

11.6

14.2

1.62

13.8

15.0

0.64

6.5

9.9

3.32

10.6

11.7

0.74

Developed

na

na

na

16.0

17.2

0.57

16.3

18.6

1.02

18.1

20.9

1.10

16.0

17.2

0.57

World

9.3

9.9

0.46

14.6

16.0

0.67

13.8

15.0

0.64

6.5

9.9

3.29

12.2

12.9

0.46

Source:

FAO 1999a.

Note:

na signifies no recorded production, mt is metric ton. WANA is West Asia and North Africa. 1983 is average for 1982-84 and 1996 is average for 1995-97. See Table 2 footnote for regional breakdown.

a All R&T includes cassava, potato, sweetpotato, yam, and other R&T such as taro.
b FAO indicates very high yields in Egypt on small areas.

Evidence from the COSCA study further indicates that the increased area planted is in many instances replacing fallow land. New cassava production is also crowding out other crops, especially yam in the humid zone and maize in the nonhumid zone (Spencer and Associates 1997).

On the post-production side, a key food security role played by cassava in Sub-Saharan Africa is its ability to be stored in the ground for 36 months or more after the formation of the edible roots is complete. Hence, cassava cultivation serves as something of a household food bank that can be drawn upon when adverse agroclimatic conditions or civil unrest limit the availability of and access to other foods. The wide variety of food products that are made from the roots and the highly nutritious leaves widely consumed as a regular part of the diet, particularly in Central, Southern, and West Africa, are added reasons why cassava cultivation is expanding (NRI 1992).

About 95 percent of the world's yam output is produced in Sub-Saharan Africa. Nigeria, Ghana, and Cd’Ivoire account for more than 80 percent of the worldwide yam harvest (Lev and Shriver 1998). The growth in area planted to yam accelerated in recent years and accounted for 56 percent of total output growth during 1983-96.8 In West Africa, particularly in Nigeria, the increase in area planted represents an expansion in yam cultivation from its traditional growing area in the humid forests to the moist savannahs (Manyong et al. 1996). Higher solar radiation, less pressure from pests and diseases, and lower costs of production due to less labor-intensive cultivation practices appear to have induced this shift.

8 Global and regional statistics for yam are highly influenced by data for Nigeria that show serious inconsistencies between production and consumption estimates (see Bricas and Attaie 1998 and Dorosh 1988).

Increases in yam area and production have also been driven by strong demand for the tubers in fresh form as food and a growing interest in their use as raw material for processed food products (Attaie, Zakhia, and Bricas 1998).

Sweetpotato production in Sub-Saharan Africa during 1983-96 has been entirely driven by growth in area planted; average yields actually contracted in this period (Tables 5 and 7). Most of the growth in production occurred in Eastern, Central, and Southern Africa in response to steadily increasing pressure on local food systems due to population growth, civil war, and economic hardship (see, for example, Bashaasha et al. 1995; Tardif-Douglin 1991). In the Kivu region, Democratic Republic of the Congo, for example, sweetpotato has been used as a staple food for disaster relief (Tanganik et al. 1999). Declines or stagnation in output of other staples have also contributed to the interest by farmers and consumers in sweetpotato in some countries, for example Malawi (Phiri 1998). Cash sales of the roots and a nascent processing sector have added to the momentum in production in Uganda (Scott et al. 1999) and Kenya (Gatumbi and Hagenimana 1998).

For potato in Sub-Saharan Africa, pressure on land to produce more food and the absence of government intervention in output markets for table potatoes (Rasolo et al. 1987; Scott 1988b, 1994b) have led to an increase in area planted of 0.7 percent per year during 1983-96. However, average yields remain too low to foster greater market participation by small-scale farmers in East and Southern Africa, mainly due to unfavorable growing conditions and lack of access to improved seed and chemical fertilizers.

Asia

In 1996, 29 percent of the global area harvested for potato was located in developing Asia, up from 19 percent in 1983. Most of the rest, 59 percent, was harvested in the industrialized countries. Following the break-up of the Former Soviet Union, China became the world's largest potato producer. In 1997, India ranked third after the Russian Federation. In China, India, and Asia in general, expansion in area planted has been driven in large part by strong off-farm demand. On the supply side, the potato's highly flexible vegetative cycle, which allows it to fit into a wide variety of cropping systems, has been another factor influencing area expansion. In the Indian Indo-Gangetic plain, where area expansion has been particularly rapid, potato can be harvested between rice and other crop harvests (Bardhan Roy et al. 1999). The spread of potato cultivation in the region has also been facilitated by ample availability of irrigation, chemical fertilizers, and cold storage facilities, and improvements in cultivation techniques and road and rail transport. And with relatively high yields in 100-120 days, potato represents an extremely lucrative crop for even the small farmers who dominate production in many Asian countries (Bardhan Roy et al. 1999; Bottema et al. 1989; Dahiya and Sharma 1994; Scott 1983b, 1988a).

Seventy-eight percent of global area planted to sweetpotato is located in Asia, 68 percent alone in China, down from 83 percent in 1983. Some of the same factors that contributed to expanded potato production have led to a decline in area planted in sweetpotato. With the spread of irrigation, farmers in some parts of China (Stone 1984; Ye and Rozelle 1993; Zhang 1999, 46) and Korea (Chin 1989) switched to crops with higher returns per hectare. Moreover, with increasing economic growth and rapid urbanization in many parts of Asia, consumers decreased their demand for traditional starchy staples, such as fresh sweetpotato, in favor of meat, bread, potato, and other preferred foods. Hybrid maize or imported feed rations displaced sweetpotato as a feed source as countries such as Korea (Chin 1989) and Taiwan (Chiang 1992) became more integrated into the global economy. An exception is sweetpotato production in Sichuan province, China (see Box 2). Bulkiness, perishability, and erratic year-to-year, season-to-season movements in supply and prices made it difficult to establish local sweetpotato-based agroindustries in the Philippines (Cabanilla 1996), Indonesia (Setyono, Damardjati, and Malian 1992), and elsewhere in Asia (Woolfe 1992).

Growth in area planted to cassava in Asia has been negative primarily due to the contraction in demand for cassava chips and pellets in the European Union. Stagnant demand for meat in industrialized countries, in particular pork, has also contributed to this trend (Delgado et al. 1999). Use of cassava as feed in developed countries in 1996 was less than half the 1983 level of 19 million mt (Table 3). In India, cassava has come under increasing pressure from competing raw materials in the markets for processed products (Balagopalan, Padmaja, and Kurup 1992) and has been hurt by adverse policies, including subsidies favoring substitute crops such as Hevea brasiliensis (natural rubber) (Best 1996).

Since the early 1990s, however, the cassava industry in South and particularly Southeast Asia has aggressively pursued alternative market outlets (Dang, Le, and Henry 1996; FAO 1994, 1995a; Titapiwatanakun 1996). The growth in the use of cassava for starch and as raw material for livestock production has raised expectations that the sector's decline may have bottomed out, and that cassava production may rebound, capitalizing on the same favorable set of factors that led to spectacular growth in the 1970s and 1980s (Konjing 1989). These factors include climate conditions that facilitate low-cost solar drying, ample transportation infrastructure, technology transfer through joint ventures, well-organized commodity associations, and attractive returns to production (Titapiwatanakun 1998).

Latin America

In Latin America, area planted to R&T has been either flat or negative. Production of cassava and sweetpotato stagnated or contracted due to urbanization and its associated shifts in eating habits. Moreover, imports of wheat flour for food, and maize or concentrates for feed, provided stifling competition for R&T in countries such as Peru (Blondet and Espinola 1998; Meerdink 1995). In other countries, such as Argentina, sweetpotato use was confined to niches for processed products in the domestic market or to exports of fresh roots. Weak demand, high relative prices, and more attractive returns to other crops have dampened potential production increases (Brescia and Parellada 1994; Maggi 1990). Where cassava production did increase - on the north coast of Colombia and in northeast Brazil - it followed the pattern first established in Thailand. Markets for cassava as animal feed and links between small farmers and these alternative commercial outlets provided the incentives for growers to expand output, often through yield-increasing technologies. This pattern offers promise to cassava-producing countries where the absence of emerging markets has led to a decline in cassava production (Henry 1992).

Potato production in Latin America increased by more than 3 million mt and yam by 0.3 million mt during 1983-96, with growth in area accounting for 16 percent and 41 percent of this increase, respectively. The continued popularity of yam as a food item and source of cash income was largely confined to Jamaica and Haiti. The emergence of a fast-food sector and processing industry in the region spurred domestic output of potato (Scott, Basay, and Maldonado 1997). However, the elimination of credit schemes (as part of a broader effort to reduce public spending and government participation in the marketplace) raised the costs of production per hectare, driving many smaller potato producers out of the sector (Rodriguez 1996). The more efficient farms were able to respond to growing market opportunities by expanding area planted.

Sources of Productivity Growth

Although growth in area planted has contributed more to increases in R&T production in developing countries than improvements in yields, noteworthy increases in productivity have taken place in some countries and regions (Table 7).

Sub-Saharan Africa

Yield growth rates in Sub-Saharan Africa have been disappointing except in the case of yam. Increases in yield are often difficult to achieve in the region because of nutrient-poor soils, lack of irrigation, and weak infrastructure (Spencer and Badiane 1995). In addition, R&T have suffered from the tendency of governments, with a few noteworthy exceptions, to focus their policies and resources on cash crops for export, or, in parts of East and Southern Africa, on cereals. One consequence of this relative neglect is that national research programs for R&T are often poorly funded and understaffed.

But as population growth and urbanization have continued apace, many governments and researchers are reappraising the potential of R&T to help meet food, feed, and income requirements in the decades ahead (Adeniji et al. 1997; Bashaasha and Mwanga 1992). Research on R&T has focused more on efforts to control pests and diseases through a combination of better biological control, improved cultural practices, and the introduction of disease-resistant varieties. Several of these production interventions have been successful (see, for example, Rueda et al. 1996). But with the noteworthy exception of integrated pest management for cassava mealybug (Norgaard 1988), the area cultivated using these new technologies has been limited to date.

Asia

Yields have increased more rapidly in Asia. In the case of potato, yield increases have been catalyzed in part by the introduction of high-yielding varieties, which made the crop more profitable for farmers (see, for example, Bofu et al. 1996; Scott 1988a). Introduction of improved seed multiplication techniques meant that farmers could achieve higher yields by having seed available at optimum planting time. In the Indo-Gangetic plain, these techniques were complemented by the expansion of cold storage facilities for seed and table potatoes (FAO 1995b). Potato yields also benefited from the earlier introduction of improved rice and wheat varieties. Successful adoption of the cereals prompted increases in the supply of chemical fertilizers, irrigation, and rural infrastructure, with subsequent spillover benefits to potato, which also is an input-intensive crop. After taking hold in prime locations, potato production expanded to somewhat less favorable soils, which led to a slowdown in yield improvement.

Production growth for sweetpotato is now positive in China, although it had contracted. The rebound is largely due to the explosive demand for meat and animal feed in the feed-deficit, inland, sweetpotato production centers. Growth in demand both at home and abroad for processed food products made from sweetpotato has also contributed to the upsurge in sweetpotato output (Fuglie et al. 1999; Zhang 1999). Improved, small-scale processing of sweetpotato roots has also boosted production by making household or village-level processing less onerous and more profitable (Wheatley, Liping, and Bofu 1997). In addition, new varieties have been adopted more widely, in part because of the rebound in off-farm demand. But China's earlier isolation from Western science and sweetpotato's much lower priority than cereals or industrial crops such as cotton have handicapped more rapid improvements in productivity.

Cassava productivity has followed trends similar to sweetpotato productivity. Cassava and, even more so, sweetpotato have been neglected in hard-pressed national agricultural research programs. Policies that have favored cheap imports or domestic products that could serve as substitutes have also hurt the potential for improving cassava productivity.

Latin America

For the region as a whole, cassava and sweetpotato productivity has been affected by weak demand. Most farmers have had little incentive to use yield-increasing technologies because potential commercial opportunities have yet to be exploited. Existing market outlets are limited and relatively thin, with the exception of cassava for feed or processed food in Colombia and northeast Brazil. Remaining growers are generally resource-poor farmers who choose to cultivate these commodities in part to avoid the financial risks associated with more input- and cash-intensive crops. As area planted with cassava and sweetpotato has declined or at best stagnated, cultivation often has been pushed onto or confined to more marginal soils. Productivity increases therefore have become more difficult to achieve. With a few notable exceptions, for example cassava in Colombia, weak national research programs have further handicapped productivity improvements for these crops. R&T are also often overlooked in policy deliberations regarding exchange rates and trade and tariff agreements. Because the farmers who make up these commodity subsectors typically have smallholdings and are poorly organized, they lack the political clout of more formidable farmers' groups, such as the national rice-growers' associations, to press for more public and private sector investments.

For potato and yam, improvements in yields have been larger. Average potato yields in Mexico are now 20 mt per hectare, after increasing at more than 4.1 percent per year during the last decade. Strong off-farm demand, fueled by contracts with agroindustry, have helped catalyze this upward trend. Potato productivity in Mexico has also benefited from the introduction of improved varieties; the spread of potato cultivation to lowland, irrigated farming areas; the high returns per hectare that have attracted large, highly technical commercial growers to the sector; and the spillover benefits - fertilizers, pesticides, and local infrastructure - associated with the spread of improved cereal varieties (Biarn Colin, and Santiago Cruz 1995). Colombia has witnessed similar developments, though production remains concentrated in highland and rainfed areas (Rodriguez 1996). In contrast, yield growth for potato in Bolivia and Ecuador, for example, has been negative or stagnant during the last decade, with average productivity in Ecuador still nearly 30 percent below levels reached in the early 1980s. Complex agroecologies; a heterogeneous set of target farmers in terms of size, education, and market orientation; slower introduction of and demand for new products; and trade and exchange rate policies that have favored cereal crops have all played a role in this negative trend (Zevallos 1997).

In summary, trends in production, area planted, and yield for R&T have been highly variable by crop and by region. Increases in potato and yam production have been most impressive in Asia and Sub-Saharan Africa, respectively. Cassava output growth has been strongest in Sub-Saharan Africa.9

9 In WANA, potato production nearly doubled during 1983-96, from 8.3 million mt to 15.1 million mt (Table 6). Area planted expanded at an annual rate of 2.6 percent in response to both strong domestic demand (see, for example, Fuglie 1994) and shifts in government policy in countries like Egypt that served to spur exports (Pautsch and Abdelrahman 1998). Yields increased at a rate of 2.1 percent per year. With access to irrigation facilities, chemical fertilizers, and improved seed (from foreign and local sources) farmers in the region had both the incentive (strong off-farm demand) and technical capability to raise productivity.

For cassava and sweetpotato in Asia and Latin America, trends have been more mixed. Issues that need to be explored include the potential for increased productivity, the reduction in per unit costs to make these crops economically attractive sources of raw material, and the increased ability of existing public and private organizations to respond to the demand for new uses. The following chapter will address these issues and assess future prospects for R&T in a framework that encompasses all the major food commodities.