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

4. Baseline Projections of Production and Use

Changes in the volume, rate of growth, form, and location of R&T production and use have continued to evolve in a highly heterogeneous fashion over the last two decades. Two aspects of this evolution have been particularly noteworthy: the versatile uses of R&T and the adaptability of these crops to the emerging needs of local food systems in developing countries. These aspects of R&T will be even more important in the face of future increases in population, urbanization, and persistent poverty in the midst of rising incomes. Assessing the future role of R&T in the global food system requires careful consideration of the relative capacity of these different food commodities to respond to the challenges ahead. For this assessment to be transparent, key assumptions about the most likely track for the global economy and population growth, as well as the most likely responses of particular food commodities to these trends, need to be made explicit. The rationale behind these assumptions also need to be delineated.

In this and the following chapter two alternative scenarios for the role of R&T in the food systems of developing countries up to 2020 will be analyzed. The baseline projections are presented in detail in this chapter, and the alternative high demand and production growth projections are presented in Chapter 5. These global projections of food supply and demand are based on IFPRI's International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT), which is continually refined and updated and covers 37 countries or country groups and 18 commodities, including all cereals, soybeans, the major R&T, meats, and milk (see Box 3). The assumptions behind the estimates are based on assessments of the future outlook for production and use of R&T; the prospects for expansion in area planted and increases in productivity; and the implications of these growth patterns for future net trade, international prices, and value of production of these commodities.

Baseline Scenario to 2020

Under the baseline scenario, projections for R&T are driven by conservative estimates of the effects of income growth on the demand for these commodities (see Appendix, Table 26) and of the effects of technological change and other parameters on increases in production and yield. As a corollary, the rate of growth of output is modest in relation to recent historical trends.

Total use of R&T in developing countries is projected to increase by 232 million mt to 635 million mt, or by 58 percent, between 1993 (1992-94 average) and 2020 (Table 8). The largest increase in terms of volume is projected for cassava: 103 million mt, or 44 percent of the total increase in R&T use over the period. Potato ranks second, with 68 million mt, or 29 percent of the increase in R&T use. Sweetpotato and yam will account for an additional 62 million mt, about 27 percent.10

10 IMPACT implicitly assumes that the relative importance of the volume of production (fresh weight) of sweetpotato versus yam in Sub-Saharan Africa will remain roughly constant during 1993-2020. The results presented here are based on the June 1998 IMPACT baseline.

Total demand for cassava is expected to increase at 1.9 percent per year during 1993-2020 in developing countries; potato at 2.0 percent per year; and sweetpotato and yam combined at 1.3 percent per year. These rates of growth compare well with projected increases in demand of major cereals during this period: demand for wheat is projected to grow at 1.8 percent annually; maize at 2.2 percent annually; and rice at 1.2 percent annually (Tables 9 and 10).

Box 3: The IMPACT Model

IFPRI's IMPACT model is specified as a set of country-level demand and supply equations linked to the rest of the world through trade. Food demand is a function of commodity prices, per capita income, and population growth. Feed demand is a function of livestock production, feed prices, and feeding efficiency. Total demand equals the sum of food, feed, and other demand. Crop production is determined by the area and yield response functions; area is projected as a function of crop prices, investment in irrigation, and estimated rates of loss of land to urbanization and land degradation. Crop yield is a function of crop price, input price, investment in irrigation, and yield growth due to technological change. Growth in productivity due to technological change is, in turn, estimated by its component sources, including advances in management research and, in the case of food crops, plant breeding research. Other sources of growth considered in the model include private-sector investments in agricultural research and development, agricultural extension and education, markets, infrastructure, and irrigation (see Rosegrant, Agcaoili-Sombilla, and Perez [1995] for details on the methodology).

The projections presented in this paper go beyond past estimates of future R&T supply and demand in a number of important respects. Previous attempts typically focus on a single root and tuber crop, for example, potato (see FAO 1995b; Henry and Gottret 1996) or aggregate R&T into one commodity. These approaches do not allow for the estimation of the possible linkages among R&T and between R&T and other food commodities.11 Previous projections have relied heavily on past commodity trends and are seldom explicit about key parameters, such as income elasticities of demand. Given concerns about the accuracy of time-series data on production and use of R&T (Alexandratos 1995, 100), IMPACT integrates an analysis of past trends and projections with a synthesis of surveys and case studies of these commodities. Previous attempts at multicommodity projections were often carried out without the full collaboration of R&T specialists. Given the relative shortage of published information on projections of supply and demand for R&T, consultation with specialists, as is the case here, represents an important aspect of any modeling exercise for these commodities.

11 Although sweetpotato and yam are combined in this analysis some results are disaggregated outside IMPACT when discussing the findings and associated contributing factors at the regional and subregional level. Cassava refers to cassava and other roots and tubers, including aroids such as taro; however, cassava alone accounts for more than 97 percent of the cassava total.

Demand for potato in developing countries is expected to increase by 2.3 percent annually for food and 0.4 percent annually for feed during 1993-2020 (Table 9). These projected food and feed growth rates are well below the annual rates achieved during 1983-96 of 4.1 percent and 4.6 percent, respectively. The combined annual growth in use of sweetpotato and yam is projected at 0.4 percent for food and 1.8 percent for feed over the next two decades compared with annual growth in human consumption of -1.8 percent and 8.6 percent per year, and annual growth in feed demand of 3.4 percent and 2.7 percent during 1983-96. According to IMPACT, cassava demand will grow at 2.0 percent annually for food and 1.6 percent per year for feed in developing countries (Table 9). These are virtually the same rates of growth achieved during 1983-96.

Per capita consumption of cereals as food is expected to decline slightly in both developed and developing countries, from 144 kg in 1993 to 140 kg in 2020, and from 172 kg to 170 kg, respectively (Table 11). But per capita consumption of R&T as food is projected to increase, albeit marginally, in both developed and developing countries, from 77 kg to 78 kg and from 56 kg to 58 kg, respectively. Declines in per capita R&T demand in China, Southeast Asia, Latin America, and Sub-Saharan Africa will be more than offset by increases in India and other South Asian countries, and in East Asian countries other than China.

Sub-Saharan Africa

Sub-Saharan Africa is expected to experience the most rapid growth in food demand in all R&T categories with the total R&T growth rate averaging 2.6 percent per year through 2020. Growth in total use (food, feed, and other uses) in Sub-Saharan Africa will account for nearly 122 million mt or 53 percent of the increase in demand for all R&T crops in developing countries during 1993-2020. The increase in use will come largely from cassava, 80 million mt (66 percent of the total), and yam roughly 33 million mt (31 percent) (Table 8) and will be overwhelmingly for food (Table 12).

Table 8 - Total use of roots and tubers in 1993, and projected to 2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T

Country/region

1993

2020

1993

2020

1993

2020

1993

2020


(million metric tons)

China

5.1

6.4

42.7

63.3

108.0

126.9

155.9

196.7

Other East Asia

1.8

1.9

2.6

3.7

0.9

1.2

5.4

6.7

India

5.7

7.3

16.3

37.1

1.2

1.2

23.2

45.6

Other South Asia

0.9

1.4

3.5

7.6

0.5

0.7

4.9

9.7

Southeast Asia

18.9

24.4

1.4

2.7

5.3

7.7

25.6

34.8

Latin America

30.3

42.9

13.0

20.4

2.5

3.6

45.8

67.0

WANA

0.9

1.0

12.8

22.0

0.1

0.2

13.8

23.2

Sub-Saharan Africa

87.7

168.1

2.8

6.3

36.0

74.5

126.4

248.9

Developing

152.0

254.6

95.2

163.2

155.5

217.3

402.7

635.1

Developed

20.7

20.5

190.1

206.2

2.5

2.7

213.3

229.4

World

172.7

275.1

285.3

369.4

158.0

220.0

616.0

864.5

Source:

IMPACT Simulations, June 1998.

Notes:

Total use includes food, feed, and other uses. WANA is West Asia and North Africa. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

Table 9 - Projected annual growth rates for food, feed, and total use of roots and tubers, 1993-2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T

Country/region

Food

Feed

Total

Food

Feed

Total

Food

Feed

Total

Food

Feed

Total


(percent per year)

China

0.17

1.61

0.84

2.20

0.27

1.47

-1.02

1.81

0.60

0.00

1.55

0.86

Other East Asia

0.83

0.21

0.05

1.31

1.20

1.29

0.84

1.47

0.86

1.22

1.20

0.83

India

0.93

na

0.93

3.09

na

3.09

0.14

na

0.14

2.42

na

2.54

Other South Asia

2.03

na

1.62

2.97

na

2.95

1.31

na

1.18

2.63

na

2.58

Southeast Asia

0.97

0.89

0.96

2.31

2.58

2.30

1.31

2.41

1.39

1.13

1.45

1.14

Latin America

0.70

1.75

1.30

1.69

1.62

1.69

1.09

2.01

1.32

1.18

1.75

1.42

WANA

1.34

0.43

0.68

2.02

1.59

2.02

1.52

na

1.51

2.00

0.60

1.95

Sub-Saharan Africa

2.49

1.53

2.44

3.10

1.81

3.10

2.74

1.89

2.73

2.55

1.56

2.54

Developing

1.99

1.62

1.93

2.33

0.37

2.02

0.44

1.81

1.25

1.62

1.57

1.70

Developed

-0.50

0.01

-0.04

0.37

0.22

0.30

0.28

0.61

0.33

0.36

0.15

0.27

World

1.98

0.95

1.74

1.20

0.26

0.96

0.43

1.80

1.23

1.30

1.07

1.26

Source:

IMPACT Simulations, June 1998.

Notes:

na signifies no recorded use. Total use includes food, feed, and other uses. WANA is West Asia and North Africa. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

Table 10 - Projected annual growth rates for food, feed, and total use of wheat, maize, rice, and all cereals, 1993-2020, baseline scenario


Wheat

Maize

Rice

All Cereals

Country/region

Food

Feed

Total

Food

Feed

Total

Total

Food

Feed

Total


(percent per year)

China

0.90

3.46

1.11

-0.50

3.46

2.53

0.58

0.58

3.37

1.37

Other East Asia

1.20

2.55

1.64

0.29

2.15

1.86

0.44

0.64

2.22

1.44

India

1.90

3.65

1.98

0.81

7.39

2.44

1.56

1.56

4.96

1.69

Other South Asia

2.77

2.83

2.77

2.21

2.88

2.34

1.84

2.30

2.85

2.32

Southeast Asia

2.28

2.46

2.29

0.84

2.81

2.27

1.23

1.31

2.66

1.53

Latin America

1.36

2.19

1.44

1.22

2.00

1.74

1.66

1.36

2.04

1.70

WANA

2.02

2,45

2.05

1.22

2.40

2.01

2.19

1.98

2.47

2.12

Sub-Saharan Africa

3.30

3.65

3.30

2.58

3.43

2.64

3.20

2.91

3.45

2.92

Developing

1.67

2.92

1.77

1.07

2.92

2.24

1.23

1.43

2.81

1.75

Developed

0.31

0.42

0.35

0.00

0.70

0.66

0.29

0.21

0.59

0.49

World

1,29

1.08

1.22

0.94

1.72

1.49

1.19

1.21

1.41

1.27

Source:

IMPACT Simulations, June 1998.

Note:

Total use includes food, feed, and other uses. Rice demand for animal feed is negligible. All cereals includes wheat, maize, rice, and other coarse grains. See Table 2 footnote for regional breakdown.

Table 11 - Per capita use of roots and tubers and cereals as food in 1993, and projected to 2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T

Cereals

Country/region

1993

2020

1993

2020

1993

2020

1993

2020

1993

2020


(kilograms per year)

China

2

2

14

20

45

28

61

50

214

206

Other East Asia

1

1

18

21

6

6

24

27

157

149

India

6

5

13

21

1

1

20

27

163

175

Other South Asia

3

3

9

11

2

1

13

15

159

170

Southeast Asia

32

30

3

3

10

10

45

43

169

169

Latin America

25

21

22

24

3

3

50

48

128

129

WANA

1

1

28

28

...

...

29

29

214

210

Sub-Saharan Africa

131

124

3

3

36

36

169

164

112

119

Developing

24

28

13

16

19

14

56

58

172

170

Developed

...

...

75

77

1

1

77

78

144

140

World

19

23

27

27

15

12

61

62

165

165

Source:

IMPACT Simulations, June 1998.

Notes:

Ellipses (...) signify very small values. WANA is West Asia and North Africa. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

Continued strong growth in food demand for cassava (2.5 percent per year) reflects the important role that cassava plays in many African diets and the relatively high rates of population growth projected for the region. The growth rate in food demand also stipulates that cassava will maintain its importance in regional diets as Sub-Saharan Africa continues to urbanize and increase its share of processed food products for consumers in the countryside and the cities. In major cassava-producing countries such as Nigeria, the bulk of annual production is already processed into food products (Adeniji et al. 1997). Finally, despite what some might consider relatively high growth rates, annual average per capita consumption of cassava as food in Sub-Saharan Africa is actually projected to decline slightly, from 131 kg per year in the base period to 124 kg per year by 2020 (Table 11).

Table 12 - Food and feed use of roots and tubers in 1993, and projected to 2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T

Country/region

1993

2020

1993

2020

1993

2020

1993

2020


(million metric tons)

Food


China

2.7

2.8

15.9

28.5

53.2

40.3

71.7

71.6

Other East Asia

0.1

0.1

1.8

2.5

0.5

0.7

2.4

3.2

India

5.4

6.9

11.8

26.9

1.1

1.2

18.4

35.0

Other South Asia

0.7

1.3

2.5

5.6

0.4

0.6

3.7

7.4

Southeast Asia

15.0

19.5

1.2

2.2

4.6

6.6

20.8

28.2

Latin America

11.5

13.9

9.9

15.6

1.6

2.1

23.0

31.6

WANA

0.2

0.3

10.5

18.0

0.1

0.2

10.8

18.5

Sub-Saharan Africa

67.0

130.2

1.4

3.1

18.2

37.9

86.6

171.2

Developing

103.3

175.9

55.0

102.5

80.5

90.5

238.8

368.9

Developed

0.4

0.4

96.2

106.2

1.7

1.8

98.3

108.4

World

103.7

176.3

151.2

208.7

82.2

92.3

337.1

477.3

Feed


China

1.9

3.0

12.3

13.3

49.4

80.2

63.7

96.4

Other East Asia

...

...

0.4

0.6

0.1

0.1

0.5

0.7

India

na

na

na

na

na

na

na

na

Other South Asia

0.1

...

na

na

...

...

0.1

...

Southeast Asia

0.7

0.9

...

...

0.3

0.6

1.0

1.5

Latin America

13.7

21.9

0.4

0.7

0.4

0.6

14.5

23.2

WANA

0.6

0.7

0.1

0.1

na

na

0.7

0.8

Sub-Saharan Africa

5.0

7.5

...

...

0.4

0.7

5.4

8.2

Developing

22.0

33.9

13.3

14.7

50.7

82.3

86.0

130.9

Developed

19.4

19.4

37.0

39.3

0.4

0.5

56.8

59.2

World

41.4

53.4

50.3

53.9

51.1

82.8

142.8

190.1

Source:

IMPACT Simulations, June 1998.

Notes:

Ellipses (...) signify very small values; na signifies no recorded use. WANA is West Asia and North Africa. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

The projected rate of growth for sweetpotato and yam as food is 2.7 percent per year during 1993-2020 (Table 9). This increase is driven largely by the projected high population growth rate and modest per capita income growth in West Africa, where production of these crops, particularly yam, will remain concentrated. Additional factors that will contribute to the growth rate include increased purchases of fresh sweetpotato by low-income, urban consumers as a cheap, starchy staple (see, for example, Hall, Bockett, and Nahdy 1998); the recurrent use of sweetpotato as a crop for food security and disaster relief; and, the moderate expansion of the use of processed food commodities made from sweetpotato as market niches emerge with population growth and urbanization (Gatumbi and Hagenimana 1998). The projected annual growth rate of potato as food (3.1 percent) reflects increasing urbanization and changes in tastes in the region and the relatively low level of per capita demand for potato as food (2.7 kg per year) in the base period (Tables 9 and 11).

Asia

Potato for food and sweetpotato for feed - the latter almost exclusively in China - dominate projected use patterns for R&T in Asia. Of the projected total increase in demand of 79 million mt of R&T in Asia by 2020, some 48 million mt (61 percent) will be contributed by potato, another 22 million mt (28 percent) by sweetpotato, and the remaining 9 million mt (11 percent) by cassava. Annual growth rates in R&T food demand will be driven largely by increased consumption of potatoes (Table 9). Growing urbanization, rising incomes, and a desire by consumers to increasingly diversify diets will help spur continued growth in demand for processed potato products (Pacific-Vision 1995a, 1995b; Scott 1994a; VIPDT 1999; Ye and Rozelle 1993; Zhang et al. 1999). Estimated growth rates of roughly 3.1 percent per year for India and 2.2 percent for China conform with the potato's status as one of the most preferred of the complementary vegetables in Asia and, in some areas in South Asia, as a seasonal staple. Continued increases in potato demand are also consistent with the effects on consumption of past increases in income and estimated income elasticities of demand (see, for example, Bouis and Scott 1996; Goletti 1993). Results from IMPACT suggest that per capita food demand for potato in developing Asia will increase from 11 kg in 1993 to 17 kg in 2020, on average, and growth in food demand for potato is projected at higher levels than growth for the major cereals (Tables 9 and 10). However, even in those countries with projected high growth rates in potato demand, estimated per capita consumption levels in 2020 will still be a third or less of current consumption levels in developed countries (Table 11).

Growth in sweetpotato use will be concentrated in China (which produces very little yam) and, to a lesser extent, in Southeast Asia. Projected growth in feed demand for sweetpotato in China (1.8 percent per year) continues a trend already well documented in field studies, which describe a growing tendency of farmers and small-scale village enterprises to use both roots and vines as animal feed, particularly in Sichuan province (see, for example, Jiang, Rozelle, and Huang 1996; Peters 1997). Despite overall negative growth in total demand for sweetpotato as food (-1.0 percent per year), some local processing of sweetpotato into food will likely continue (Fuglie et al. 1999; Jiang, Rozelle, and Huang 1996; Zhang 1999), given that these activities in many instances complement sweetpotato processing for feed at the small-entrepreneur level (Peters 1997). As for cassava, feed use and processed products will increasingly replace direct consumption as food. This pattern has already become evident in Vietnam (Goletti and Wheatley 1999; Howeler 1996), Thailand (Titapiwatanakun 1998), and Indonesia (Wheatley and Scott 1994).

Latin America

In Latin America, the increase in total R&T demand will be dominated by cassava (12.6 million mt or about 60 percent of the total increase in R&T demand) and potato (7.4 million mt or 35 percent of the total increase). Demand growth for cassava as feed (1.8 percent per year) is projected to be stronger than growth for food (0.7 percent per year) (Table 9). This slow growth in food demand will follow recent trends: per capita food demand declined from 29 kg in 1983 to 25 kg in 1993 and is projected to decline further to 21 kg by 2020 (Table 11). Lynam (1989a, 1989b) and Ostertag and Herrera (1992) point to the increasing availability and use of cassava substitutes such as wheat flour, especially in urban areas, a trend facilitated in recent years by trade liberalization. Although new or improved forms of processed cassava for human consumption have been developed, their entry into the market has been sufficient only to slow the decline in per capita food demand for cassava. Feed use has been and will continue to be more dynamic, because cereal-feed-deficit countries (such as Colombia; see, for example, Balcazar 1997) or regions within countries (like northeast Brazil; see, for example, Ospina and Wheatley 1992) exploit cassava as a local substitute for maize.

Per capita food demand for potato stood at 22 kg in 1993 and is projected to rise, albeit modestly, to 24 kg in 2020 (Table 11). Population and income growth combined with high levels of urbanization, will spur a greater intake of processed potatoes in countries like Colombia (Rodriguez and Rodriguez 1992), following a trend already manifest in some developed countries.12 Potato will not be used much for animal feed in the region: a mere 0.6 million mt out of a total use of 15.6 million mt in 2020; nor is this projected to change in the future (Table 12). Per capita consumption levels for sweetpotato and yam as food (roughly 3 kg in both 1993 and 2020) are low when compared to cassava and potato (Table 11). The growth rate in food demand for potato is highest of all R&T and surpasses those for the cereals (Tables 9 and 10).

12 Lucier et al. (1991) report that over 52 percent of the U.S. potato crop is used for processed food products. Hesen (1991) describes similar trends for Western Europe.

Baseline Projections for Production, Area, and Yield

According to the baseline scenario, production of cassava in developing countries will grow at an annual average rate of 1.7 percent during 1993-2020 (Table 13) compared to the annual growth rate of 1.8 percent achieved during 1983-96 (Table 5). Production growth for sweetpotato and yam is projected at 1.3 percent per year during 1993-2020, and for potato at 2.0 percent. Thus, growth in potato production is expected to slow down considerably from the rate of 4.1 percent per year during 1983-96.

Total R&T production in developing countries will increase by some 230 million mt by 2020 according to the baseline scenario (Table 13). More than half of that increase (122 million mt) is projected to occur in Sub-Saharan Africa and will consist largely of cassava (81 million mt) and yam (roughly 30 million mt).13 Over the next two decades, the source of output growth will shift away from expansion in area planted toward increases in yields (66 percent of total output growth) for total R&T (Tables 14 and 15). The projected growth rate for cassava production in Sub-Saharan Africa of 2.5 percent per year actually constitutes a 33 percent decline compared to the rate of increase during 1983-96 (Table 5). The growth rate for sweetpotato and yam production is also projected to decline between 1993 and 2020 to 2.7 percent annually (Table 13). Discrepancies between production and use data for yam in Nigeria (Bricas and Attaie 1998), agroclimatic constraints on further expansion in area planted for yam (M. Bokanga, personal communication, September 1998), and slower development of high-yielding varieties (Spencer and Badiane 1995) all support the more conservative projected growth rate for sweetpotato and yam.

13 The figure of 30 million mt is for yam alone and assumes that the division of total sweetpotato and yam production conforms to the proportion in 1993.

Projected increases for R&T production in Asia will be led by potato. In fact, about 48 million mt out of the 68 million mt increase in potato production in developing countries will come from developing Asia. The bulk of this production increase will be driven by improvements in yields (Tables 14 and 15). With access to irrigation, chemical inputs, and relatively abundant labor, prospects for yield improvements are good. In Bangladesh, India, and Pakistan, the introduction of improved varieties and the expansion of cold storage facilities have provided additional incentives to growers (FAO 1995b). The liberalization of the internal market for cold storage has brought an expansion and modernization of India's storage industry that appears capable of keeping up with farm production of potatoes (Fuglie et al. 1997). Some farmers in Bangladesh and India, in fact, currently report yields double their respective national averages of 11 mt/ha and 16 mt/ha (see, for example, Dahiya et al. 1997; Khatana et al. 1997). Given that anywhere from 65 to 90 percent of the potato harvest in South Asia is sold for cash (Bottema et al. 1989; Dahiya and Sharma 1994; Fuglie et al. 1997; Kokab and Smith 1989), growers have a strong commercial incentive to improve productivity.

Sweetpotato production in China is projected to grow at 0.6 percent per year, due entirely to increases in yields because area planted is projected to contract. As irrigation and improved road networks penetrate into more isolated areas, growers will switch from sweetpotato to production of higher value-added crops unless new technology (for example improved varieties for use as feed), institutional development (such as improvements in small agro-enterprise management and marketing), and appropriate policies in support of these initiatives are implemented. Increases in production of cassava will also largely come from increases in yield.

Table 13 - Production levels and annual growth rates of production for roots and tubers, 1993-2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T


Production

Growth rate

Production

Growth rate

Production

Growth rate

Production

Growth rate

Country/region

1993

2020


1993

2020


1993

2020


1993

2020



(million mt)

(percent/yr)

(million mt)

(percent/yr)

(million mt)

(percent/yr)

(million mt)

(percent/yr)

China

4.8

6.5

1.18

42.5

63.4

1.49

108.5

128.3

0.62

155.8

198.3

0.90

Other East Asia

na

na

na

2.4

3.2

1.13

0.8

1.1

1.18

3.2

4.3

1.14

India

5.8

7.0

0.71

16.3

37.3

3.10

1.2

1.2

0.12

23.3

45.5

2.51

Other South Asia

0.8

1.3

1.56

3.5

7.5

2.90

0.5

0.7

1.00

4.8

9.4

2.53

Southeast Asia

42.0

48.2

0.51

1.3

2.3

2.06

5.3

7.5

1.29

48.6

58.0

0.65

Latin America

30.3

41.7

1.19

12.6

20.0

1.72

2.6

3.5

1.19

45.5

65.3

1.35

WANA

0.1

0.2

1.60

13.0

22.4

2.02

0.1

0.2

1.35

13.3

22.8

2.01

Sub-Saharan Africa

87.8

168.6

2.45

2.6

5.9

3.01

36.0

74.2

2.71

126.4

248.7

2.54

Developing

172.4

274.7

1.74

94.3

162.0

2.02

155.9

217.8

1.25

422.6

654.5

1.63

Developed

0.4

0.4

0.68

191.0

207.4

0.31

2.1

2.1

0.12

193.4

210.0

0.31

World

172.7

275.1

1.74

285.3

369.4

0.96

158.0

220.0

1.23

616.0

864.5

1.26

Source:

IMPACT Simulations, June 1998.

Notes:

na signifies no recorded production. WANA is West Asia and North Africa. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

Table 14 - Area planted and annual growth rates in area planted for roots and tubers, 1993-2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T


Area

Growth rate

Area

Growth rate

Area

Growth rate

Area

Growth rate

Country/region

1993

2020


1993

2020


1993

2020


1993

2020



(million ha)

(percent/yr)

(million ha)

(percent/yr)

(million ha)

(percent/yr)

(million ha)

(percent/yr)

China

0.3

0.3

0.08

3.1

3.2

0.17

6.2

5.6

-0.39

9.6

9.1

-0.19

Other East Asia

na

na

na

0.2

0.2

-0.41


0.1

0.30

0.2

0.2

-0.25

India

0.2

0.2

0.02

1.0

1.4

1.19

0.1

0.1

-0.16

1.4

1.8

0.90

Other South Asia

0.1

0.1

0.21

0.3

0.4

0.89

0.1

0.1

0.10

0.5

0.6

0.67

Southeast Asia

3.5

3.5

0.04

0.1

0.1

0.59

0.8

0.8

0.04

4.4

4.4

0.06

Latin America

2.7

2.7

-0.01

1.0

1.1

0.41

0.3

0.3

-0.23

4.0

4.1

0.07

WANA

...

...

0.12

0.7

0.8

0.55

...

...

0.13

0.7

0.8

0.54

Sub-Saharan Africa

11.9

15.9

1.09

0.4

0.6

1.25

4.2

5.7

1.16

16.5

22.2

1.11

Developing

18.8

22.9

0.73

6.8

7.8

0.51

11.9

12.8

0.27

37.5

43.5

0.55

Developed

...

...

-0.04

11.6

11.0

-0.19

0.1

0.1

-0.09

11.7

11.1

-0.19

World

18.8

22.9

0.73

18.4

18.8

0.09

12.0

12.9

0.27

49.2

54.6

0.39

Source:

IMPACT Simulations, June 1998.

Notes:

Ellipses (...) signify very small values; na signifies no recorded production. WANA is West Asia and North Africa, ha stands for hectares. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

Table 15 - Yield and annual growth rates in yields for roots and tubers, 1993-2020, baseline scenario


Cassavaa

Potato

Sweetpotato and yamb

All R&T


Yield

Growth rate

Yield

Growth rate

Yield

Growth rate

Growth rate

Country/region

1993

2020


1993

2020


1993

2020




(mt/ha)

(percent/yr)

(mt/ha)

(percent/yr)

(mt/ha)

(percent/yr)

(percent/yr)

China

15.1

20.2

1.10

13.7

19.6

1.32

17.5

23.1

1.02

1.09

Other East Asia

na

na

na

13.2

20.0

1.55

15.9

20.2

0.89

1.40

India

23.6

28.4

0.69

15.6

25.9

1.89

8.4

9.1

0.29

1.60

Other South Asia

9.4

13.5

1.35

10.9

18.5

1.99

9.1

11.6

0.90

1.84

Southeast Asia

12.1

13.7

0.46

12.5

18.6

1.46

6.8

9.5

1.25

0.60

Latin America

11.3

15.6

1.21

12.9

18.3

1.30

7.6

11.2

1.43

1.27

WANA

32.5c

48.3c

1.48

18.6c

27.6c

1.46

17.8

24.7

1.23

1.46

Sub-Saharan Africa

7.4

10.6

1.34

6.5

10.4

1.74

8.6

12.9

1.54

1.41

Developing

9.2

12.0

1.00

13.8

20.7

1.50

13.1

17,0

0.97

1.08

Developed

12.1

14.7

0.72

16.5

18.9

0.50

17.9

18.9

0.22

0.49

World

9.2

12.0

1.00

15.5

19.6

0.87

13.2

17.0

0.96

0.87

Source:

IMPACT Simulations, June 1998.

Notes:

na signifies no recorded production. WANA is West Asia and North Africa; ha is hectares. See Table 2 footnote for regional breakdown.

a These figures are for cassava and other roots and tubers such as taro. For developing countries, cassava alone accounts for over 97 percent of the total.

b Estimates for Sub-Saharan Africa are largely for yam, given the roughly 80/20 distribution in favor of yam production in the region, according to FAO 1999a. Estimates for Asia and WANA are for sweetpotato only, and in Latin America estimates are 68/32 for sweetpotato versus yam.

c FAO indicates very high yields in Egypt in small areas.

In Latin America, cassava production will increase at a moderate rate (1.2 percent per year) during 1993-2020. This increase will be driven by growth in yields, even as market demand for traditional cassava production remains weak, compared to potential growth opportunities in alternative markets. A similar scenario plays out in the case of sweetpotato and yam, but from a much lower level of output in the base period. For potato, annual growth in yields and area planted, 1.3 and 0.4 percent, respectively, combine to form the annual production growth rate of 1.7 percent. Improvements in the management of pests and diseases - late blight, for example - will facilitate productivity increases. Demand for both fresh and processed potato products will induce farmers to adopt yield-increasing technologies - a tendency supported by falling tariffs (Scott, Basay, and Maldonado 1997).

Baseline Projections for World Prices and International Trade

IMPACT projections indicate that global production of R&T will grow fast enough for real world prices of these commodities to fall between 1993 and 2020. According to the baseline scenario, average root and tuber prices are projected to decline by 19 percent: potato prices by 14 percent; sweet-potato and yam by 23 percent; and cassava and other roots and tubers by 15 percent (Table 16).

This slow decline in prices will not be accompanied by a significant increase in aggregate world trade of major R&T according to the baseline scenario. In fact, as a share of developing country R&T production, trade in R&T is projected to decline over time.

Developing countries as a group will remain net R&T exporters and developed countries net importers. Total net exports are projected to decrease slightly from 19.9 million mt in 1993 to 19.4 million mt in 2020 under the baseline scenario. The slight decrease in net cassava exports largely reflects the current declining trend in cassava exports from Southeast Asia to the European Union and the almost offsetting exports of cassava products to other regions (Goletti and Wheatley 1999; Titapiwatanakun 1998). Potato imports by the developing world will show the biggest absolute increase in R&T trade, rising by nearly a third, from 0.9 million mt in 1993 to 1.2 million mt in 2020. The biggest increases in potato imports are projected for Southeast Asia and Sub-Saharan Africa (0.3 million mt each), the former region will import mostly processed food products and the latter a mix of food products and seed. China and India are expected to become net exporters on the order of 100,000 mt per year, but in absolute percentage terms this will represent less than one half of one percent of their projected domestic production in 2020.

Table 16 - Estimated world prices for roots and tubers and selected other foods, late 1980s, 1993, and projected to 2020, baseline and HDP scenarios


Price (US$ per metric ton)

Price changes (percent)

Commodity

1987/89a

1989/91b

1993c

2020Ad

2020Be

1993/2020Af

1993/2020Bg

Potato

180

110

160

137

145

-14

-9

Sweetpotato and yam

-

-

91h

70h

82h

-23

-10

Sweetpotatoi

82

76

80

56

69

-30

-14

Yami

105

137

135

105

115

-22

-15

Cassava and other R&T

-

-

54j

46j

48j

-15

-11

Cassava

66

68

-

-

-

-

-

All roots and tubers

-

-

113

91

99

-19

-12

Wheat

144

144

148

133

133

-10

-10

Maize

104

124

126

123

123

-2

-2

Other grains

-

-

122h

105h

106h

-14

-13

Barley

128

114

-

-

-

-

-

Sorghum

93

124

-

-

-

-

-

Millet

132

158

-

-

-

-

-

Milled rice

284

292

286

265

266

-7

-7

Soybean

265

234

263

234

235

-11

-11

Beef

-

-

2,023

1,768

1,771

-13

-12

Beef and buffalo meat

1,458h

2,226h

-

-

-

-

-

Pigmeat

-

-

1,366

1,209

1,212

-11

-11

Sheep and goat meat

1,6521h

2,099h

2,032h

1,842h

1,845h

-9

-9

Poultry

-

-

1,300

1,157

1,159

-11

-11

Sources:

1987/89: TAC (1996, Annex II); 1989/91: Rao (1993); 1993, 2020A, 2020B: IMPACT Simulations, June 1998.

Note:

- signifies not applicable/available.

a Prices used in 1992 analysis (TAC 1996, Annex II, Table 9).

b Prices used by Rao (1993).

c Base period prices used in present study.

d Baseline scenario.

e High demand and production growth scenario.

f 1993/baseline scenario percent change in price, rounded to nearest percent.

g 1993/high demand and production growth scenario percent change in price, rounded to nearest percent.

h Composite price.

i Disaggregation of sweetpotato and yam outside of IMPACT, but based on historical trends and IMPACT simulations.

j Composite price: cassava and other roots and tubers such as taro.

Net exports of sweetpotato and yam by developing countries are projected to increase by 40 percent or 162,000 mt, a tiny fraction of total production in these countries. It should be noted that trade among developing countries causes their aggregate net exports of sweetpotato and yam (and other commodities) to be lower than the exports of some individual developing countries. China, for example, is expected to export 1.4 million mt of sweetpotato by 2020.

Net exports of cassava and other roots and tubers by developing countries are projected to decrease slightly, by 290,000 mt or 1.4 percent, from 1993 levels. This decrease will come about largely because India will switch from being a small net exporter to a small net importer. The opposite trend is projected for China. Latin America will see imports of cassava and other roots and tubers rise to 1.2 million mt by 2020. The major beneficiaries of a more diversified world market for cassava and other roots and tubers will be Southeast Asia and Sub-Saharan Africa.

Table 17 - Total value of selected IMPACT commodities for developing countries in 1993 and projected to 2020, baseline and HDP scenarios


1993

2020A

2020B


Valuea

Percent of total

Percent of subtotal

Valuea

Percent of total

Percent of subtotal

Valuea

Percent of total

Percent of subtotal

Commodity

(US$ million)



(US$ million)



(US$ million)



Potato

15,094

4.1

6.6

22,193

3.9

7.0

28,131

4.9

8.6

Sweetpotato and yam

14,185

3.9

6.2

15,248

2.7

4.8

18,879

3.3

5.8

Yam aloneb

4,209

1.1

1.8

6,642

1.2

2.1

7,693

1.4

2.4

Cassava and other R&T

9,307

2.5

4.0

12,636

2.2

4.0

13,937

2.4

4.2

All R&T

38,586

10.5

16.7

50,076

8.8

15.8

60,946

10.5

18.6

All cereals

176,622

48.0

76.7

241,253

42.6

76.3

242,195

41.9

73.8

Soybeans

15,176

4.1

6.6

24,839

4.4

7.9

24,958

4.3

7.6


Subtotal

230,384



316,168



328,099



All meat

137,752

37.4


249,862

44.1


250,467

43.3



Total

368,136

100.0

100.0

566,030

100.0

100.0

578,567

100.0

100.0

Source:

IMPACT Simulations, June 1998.

Note:

2020A: baseline scenario; 2020B: high demand and production growth scenario (HDP).

a Value is calculated using production data (1993: IMPACT base year values; 2020A: IMPACT baseline scenario; 2020B: IMPACT HDP scenario) multiplied by price (see Table 16).

b Disaggregation for yam outside of IMPACT, but based on historical trends and IMPACT simulations.

An important development for the international R&T market is the projected increase in total use of R&T in some developing country regions at rates higher than the projected production growth rate in these regions. This trend will benefit traditional net potato exporters such as Canada, the United States, and the Netherlands. Eastern Europe will become a net exporter of potato at 411,000 mt, while Japan will increase its imports of all roots and tubers by 23 percent, from 1.5 million mt in 1993 to 1.9 million mt by 2020, under the baseline scenario.

The Value of R&T Crop Production in the Baseline Scenario

Policymakers and researchers are also interested in the projected value of root and tuber crops in the decades ahead. The projections of the rate of growth of supply, demand, and world market prices for R&T permit the estimation of the future value of root and tuber crops alone and in relation with the estimated values for other major food commodities produced in developing countries.

The IMPACT baseline scenario shows that the share of R&T in the total value of the major food commodities will decrease from 10.5 percent to 8.8 percent between 1993 and 2020, and R&T's share in the major food and feed crops will shrink from 16.7 to 15.8 percent (Table 17). The bulk of the change in the projected future values of R&T is due to the decline in the estimated world prices for cassava and sweetpotato and the relatively strong projected prices for maize and rice in 2020.

In summary, the projected baseline growth rates for supply and demand of R&T vary according to crop, region, and use. Furthermore, in several instances the projected growth rates in supply are rather conservative when compared with recent historical trends. This is most notably the case for potato. Nevertheless, even the baseline estimates indicate that the role of R&T in the food systems of developing countries will not deteriorate significantly over the next two decades.

The next chapter presents the outcome of an alternative scenario, based on more optimistic growth prospects for R&T.