Previous PageTable Of ContentsNext Page


4.5.2 Yield gaps

Despite the increases in land under cultivation in the land-abundant countries, much of agricultural production growth has been based on the growth of yields, and will increasingly need to do so. What is the potential for a continuation of yield growth? In countries and localities where the potential of existing technology is being exploited fully, subject to the agro-ecological constraints specific to each locality, further growth, or even maintenance, of current yield levels will depend crucially on further progress in agricultural research. In places where yields are already near the ceilings obtained on research stations, the scope for raising yields is widely believed to be much more limited than in the past (see, for example, Sinclair, 1998). However, this has been true for some time now, but average yields have continued to increase, albeit at a decelerating rate. For example, wheat yields in South Asia, which accounts for about a third of the developing countries' area under wheat, increased by 45 kg p.a. in the 1960s, 35 kg in the 1970s, 55 kg in the 1980s and 45 kg in 1990-99. Yields are projected to grow by 41 kg per year over 1997/99 to 2030.

Intercountry differences in yields remain very wide, however. This can be illustrated for wheat and rice in the developing countries. Current yields in the 10 percent of countries with the lowest yields (excluding countries with less than 50000 ha under the crop), is less than one-fifth of the yields of the best performers comprising the top decile (Table 4.13). If subnational data were available, a similar pattern would probably be seen for intranational differences as well. For wheat this gap between worst and best performers is projected to persist until 2030, while for rice the gap between the top and bottom deciles may be somewhat narrowed by 2030, with yields in the bottom decile reaching 27 percent of yields in the top decile. This may reflect the fact that the scope for raising yields of top rice performers is more limited than in the past. However, countries included in the bottom and top deciles account for only a minor share of the total production of wheat and rice. Therefore it is more important to examine what will happen to the yield levels obtained by the countries which account for the bulk of wheat and rice production. Current unweighted average yields of the largest producers,10 are about half the yields achieved by the top performers (Table 4.13). In spite of continuing yield growth in these largest producing countries, this situation will remain essentially unchanged by 2030 for wheat, with rice yields reaching about 60 percent of the top performers' yields.

Table 4.13: Average wheat and rice yields for selected country groups

 

1961/63

1997/99

2030

tonnes/ ha

as % of top decile

tonnes/ ha

as %of top decile

tonnes/ ha

as % of top decile

Wheat

No. of developing countries included

32

 

32

 

33

 

Top decile

2.15

100

5.31

100

7.44

100

Bottom decile

0.40

18

0.80

15

1.25

17

Decile of largest producers (by area)

0.87

40

2.60

49

3.89

52

All countries included

0.97

45

2.15

41

3.11

42

Major developed country exporters

1.59

 

3.19

 

4.13

 

World

1.23

 

2.55

 

3.47

 

Rice (paddy)

No. of developing countries included

44

 

52

 

55

 

Top decile

4.51

100

6.57

100

7.93

100

Bottom decile

0.72

16

1.14

17

2.12

27

Decile of largest producers (by area)

1.82

40

3.51

53

4.84

61

All countries included

1.88

42

3.17

48

4.30

54

World

2.07

 

3.43

 

4.52

 

Notes: Only countries with over 50 000 harvested ha are included. Countries included in the deciles are not necessarily the same for all years.

Average yields are simple averages, not weighted by area.

Based on this analysis, a prima facie case could be made that there has been and still is, considerable slack in the agricultural sectors of the different countries. This slack could be exploited if economic incentives so dictated. However, the fact that yield differences among the major cereal producing countries are very wide does not necessarily imply that the lagging countries have scope for yield increases equal to intercountry yield gaps. Part of these differences may simply reflect differing agro-ecological conditions. For example, the low average yields in Mexico of its basic food crop, maize (currently 2.4 tonnes/ha), are largely attributable to agro-ecological constraints that render it unsuited for widespread use of the major yield-increasing technology, hybrid seeds, a technology which underlies the average 8.3 tonnes/ha of the United States. Hybrids are at present used in Mexico on about 1.2 million ha, out of a total harvested area under maize of 7 million ha, while the area suitable for hybrid seed use is estimated to be about 3 million ha (see Commission for Environmental Cooperation, 1999, p.137-138).

However, not all, or perhaps not even the major part, of yield differences can be ascribed to such conditions. Wide yield differences are present even among countries with fairly similar agro-ecological environments. In such cases, differences in the socio-economic and policy environments probably play a major role. The literature on yield gaps (see, for example, Duwayri, Tran and Nguyen, 1999) distinguishes two components of yield gaps: one due to agro-environmental and other non-transferable factors (these gaps cannot be narrowed); and another component due to differences in crop management practices such as suboptimal use of inputs and other cultural practices. This second component can be narrowed provided that it is economic to do so and therefore is termed the«exploitable yield gap». Duwayri, Tran and Nguyen (1999) state that the theoretical maximum yields for both wheat and rice are probably in the order of 20 tonnes/ha. On experimental stations, yields of 17 tonnes/ha have been reached in subtropical climates and of 10 tonnes/ha in the tropics. FAO (1999c) reports that concerted efforts in Australia to reduce the exploitable yield gap increased rice yields from 6.8 tonnes/ha in 1985/89 to 8.4 tonnes/ha in 1995/99, with many individual farmers obtaining 10 to 12 tonnes/ha.

In order to draw conclusions on the scope for narrowing the yield gap, one needs to separate its«non-transferable» part from the«exploitable» part. One way to do so is to compare yields obtained from the same crop varieties grown on different locations of land that are fairly homogeneous with respect to their physical characteristics (climate, soil and terrain), which would eliminate the«non-transferable» part in the comparison. One can go some way in that direction by examining the data on the suitability of land in the different countries for producing any given crop under specified technology packages. The required data comes from the GAEZ analysis discussed in Section 4.3.1. These data make it possible to derive a«national maximum obtainable yield» by weighting the yield obtainable in each of the suitability classes with the estimated land area in each suitability class. The derived national obtainable yield can then be compared with data on the actual national average yields. This comparison is somewhat distorted since the GAEZ analysis deals only with rainfed agriculture, while the national statistics include irrigated agriculture as well. However, the findings seem to confirm the hypothesis that a good part of the yield gap is of the second, exploitable type. For a further discussion on this topic, see Section 11.1 in Chapter 11.

4.6 Input use

4.6.1 Fertilizer consumption

As discussed in Section 4.2, the bulk of the projected increases in crop production will have to come from higher yields, with the remaining part coming from an expansion in harvested area. Both higher yields, which normally demand higher fertilizer application rates, and land expansion will lead to an increase in fertilizer use. Increases in biomass require additional uptake of nutrients which may come from both organic and mineral sources. Unfortunately, for most crops there are not enough data to estimate the relation between mineral fertilizer consumption and biomass increases. The historical relationship between cereal production and mineral fertilizer consumption is better known. One-third of the increase in cereal production worldwide and half of the increase in India's grain production during the 1970s and 1980s have been attributed to increased fertilizer consumption. The application of mineral fertilizers needed to obtain higher yields should complement nutrients available from other sources and match the needs of individual crop varieties.

Increased use of fertilizer is becoming even more crucial in view of other factors, such as the impact on soil fertility of more intensive cultivation practices and the shortening of fallow periods. There is empirical evidence that nutrient budgets11 change over time and that higher yields can be achieved through reduction of nutrient losses within cropping systems. That is, increases in food production can be obtained with a less than proportional increase in fertilizer nutrient use. Frink, Waggoner and Ausubel (1998) showed this situation for maize in North America. Farmers achieve such increased nutrient use efficiency by adopting improved and more precise management practices. Socolow (1998) suggests that management techniques such as precision agriculture offer abundant opportunities to substitute information for fertilizer. It is expected that this trend of increasing efficiency of nutrient use through better nutrient management, by improving the efficiency of nutrient balances and the timing and placement of fertilizers, will continue and accelerate in the future.

Projections for fertilizer consumption have been derived on the basis of the relationship between yields and fertilizer application rates that existed during 1995/97. Data on fertilizer use by crop and fertilizer application rates (kg of fertilizer per ha) are available for all major countries and crops, accounting for 97 percent of global fertilizer use in 1995/97 (FAO/IFA/IFDC, 1999 and Harris, 1997). This relationship is estimated on a cross-section basis for the crops for which data are available and is assumed to hold also over time as yields increase (see Daberkov et al., 1999). It provides a basis for estimating future fertilizer application rates required to obtain the projected increase in yields for most of the crops covered in this study. It implicitly assumes that improvements in nutrient use efficiency will continue to occur as embodied in the relationship between yields and fertilizer application rates (fertilizer response coefficients) estimated for 1995/97. For some crop categories such as citrus, vegetables, fruit and«other cereals», fertilizer consumption growth is assumed to be equal to the growth in crop production: i.e. for these crops, the base year input-output relationship between fertilizer use and crop production is assumed to remain constant over the projection period. To account fully for all fertilizer consumption, including its use for crops not covered in this study, fertilizer applications on fodder crops were assumed to grow at the same rate as projected growth for livestock (meat and milk) production, and fertilizer applications on«other crops» is at the average rate for all crops covered in the study.

The overall result, aggregated over all crops, is that fertilizer consumption will increase by 1.0 percent p.a., rising from 138 million tonnes in 1997/99 to 188 million tonnes in 2030 (Table 4.14). This is much slower than in the past for the reasons explained below. Wheat, rice and maize, which together at present account for over half of global fertilizer use, will continue to do so, at least until 2030. By 2015 maize will rival wheat as the top fertilizer user because of the projected increase in maize demand for feeding purposes in developing countries (see Chapter 3). Fertilizer applications to oilseeds (soybeans and rapeseed) are expected to grow fastest.

Table 4.14: Fertilizer consumption by major crops

 

1997/99

1997/99

2015

2030

1997/99-2030

Share (%) in total

Nutrients, million tonnes

% p.a.

Wheat

18.4

25.3

30.4

34.9

1.0

Rice

17.3

23.8

26.5

28.1

0.5

Maize

16.3

22.5

29.0

34.5

1.3

Fodder

6.2

8.5

9.3

10.0

0.5

Seed cotton

3.5

4.9

6.2

7.1

1.2

Soybeans

3.4

4.6

7.6

11.5

2.9

Vegetables

3.3

4.6

5.3

6.1

0.9

Sugar cane

3.2

4.4

5.5

6.6

1.3

Fruit

2.9

4.1

4.3

7.5

1.9

Barley

2.9

4.0

4.4

4.8

0.6

Other cereals

2.9

3.9

9.2

8.3

2.3

Potato

2.0

2.7

3.3

3.8

1.1

Rapeseed

1.5

2.1

3.5

5.1

2.8

Sweet potato

1.3

1.8

2.0

2.1

0.5

Sugar beet

1.0

1.4

1.6

1.7

0.6

All cereals

 

79.5

99.5

110.6

1.0

% of total

57.7

57.7

64.8

58.8

 

All crops above

 

118.5

148.2

172.1

1.2

% of total

86.0

86.0

89.8

91.5

 

World total

 

137.7

165.1

188.0

1.0

Notes: Crops with a 1997/99 share of at least 1 percent, ordered according to their 1997/99 share in fertilizer use.

North America, western Europe, East and South Asia accounted for over 80 percent of all fertilizer use in 1997/99. Growth in fertilizer use in the industrial countries, especially in western Europe, is expected to lag significantly behind growth in other regions of the world (Table 4.15). The maturing of fertilizer markets during the 1980s in North America and western Europe, two of the major fertilizer consuming regions of the world, account for much of the projected slowdown in fertilizer consumption growth. In the more recent past, changes in agricultural policies, in particular reductions in support measures, contributed to a slowdown or even decline in fertilizer use in this group of countries. Increasing awareness of and concern about the environmental impacts of fertilizer use are also likely to hold back future growth in fertilizer use (see Chapter 12).

Over the past few decades, the use of mineral fertilizers has been growing rapidly in developing countries starting, of course, from a low base (Table 4.15). This has been particularly so in East and South Asia following the introduction of high-yielding varieties. East Asia (mainly China) is likely to continue to dwarf the fertilizer consumption of the other developing regions. For sub-Saharan Africa, above average growth rates are foreseen, starting from a very low base, but fertilizer consumption per hectare is expected to remain at a relatively low level. The latter probably reflects large areas with no fertilizer use at all, combined with small areas of commercial farming with high levels of fertilizer use, and could be seen as a sign of nutrient mining (see also Henao and Baanante, 1999).

Table 4.15: Fertilizer consumption: past and projected

 

1961
/63

1979
/81

1997
/99

2015

2030

1961
-1999

1989
-1999

1997/99
-2030

Total

Nutrients, million tonnes

% p.a.

Sub-Saharan Africa

0.2

0.9

1.1

1.8

2.6

5.3

-1.8

2.7

Latin America and the Caribbean

1.1

6.8

11.3

13.1

16.3

6.1

4.4

1.2

Near East/ North Africa

0.5

3.5

6.1

7.5

9.1

7.3

0.8

1.3

South Asia

0.6

7.3

21.3

24.1

28.9

9.6

4.5

1.0

excl. India

0.2

1.6

4.2

5.4

6.9

9.2

4.6

1.5

East Asia

1.7

18.2

45.0

56.9

63.0

9.3

3.8

1.1

excl. China

0.9

4.1

9.4

13.8

10.3

7.0

3.2

0.3

 

All above

4.1

36.7

84.8

103.5

119.9

8.5

3.7

1.1

excl. China

3.3

22.6

49.2

60.4

67.3

7.6

3.5

1.0

excl. China and India

2.9

16.9

32.1

41.6

45.3

6.9

3.1

1.1

 

Industrial countries

24.3

49.1

45.2

52.3

58.0

1.4

0.1

0.8

Transition countries

5.6

28.4

7.6

9.3

10.1

0.7

-14.9

0.9

 

World

34.1

114.2

137.7

165.1

188.0

3.6

0.2

1.0

Per hectare

kg/ha (arable land)

% p.a.

Sub-Saharan Africa

1

7

5

7

9

4.5

-2.4

1.9

Latin America and the Caribbean

11

50

56

59

67

6.0

0.0

0.6

Near East/North Africa

6

38

71

84

99

5.7

3.9

1.0

South Asia

6

36

103

115

134

9.5

4.5

0.8

excl. India

6

48

113

142

178

8.8

4.3

1.4

East Asia

10

100

194

244

266

8.3

3.6

1.0

excl. China

12

50

96

131

92

6.1

3.3

-0.1

 

All above

6

49

89

102

111

7.7

3.3

0.7

excl. China

6

35

60

68

71

6.9

3.2

0.5

excl. China and India

7

35

49

58

58

6.0

2.6

0.5

 

Industrial countries

64

124

117

   

1.3

0.3

 

Transition countries

19

101

29

   

0.9

-14.4

 
 

World

25

80

92

   

3.3

0.1

 

Note: Kg/ha for 1997/99 are for developing countries calculated on the basis of “adjusted” arable land data. For industrial and transition countries no projections of arable land were made.

Average fertilizer productivity, as measured by kg of product obtained per kg of nutrient, shows considerable variation across countries. This reflects a host of factors such as differences in agro-ecological resources (soil, terrain and climate), in management practices and skills and in economic incentives. Fertilizer productivity is also strongly related to soil moisture availability. For example, irrigated wheat production in Zimbabwe and Saudi Arabia shows a ratio of 40 kg wheat per kg fertilizer nutrient at yield levels of 4.5 tonnes/ha. Similar yields in Norway and the Czech Republic require twice as large fertilizer application rates, reflecting a considerably different agro-ecological resource base. Furthermore, a high yield/fertilizer ratio may also indicate that fertilizer use is not widespread among farmers (e.g. wheat in Russia, Ethiopia and Algeria), or that high yields are obtained with nutrients other than mineral fertilizer (e.g. manure is estimated to provide almost half of all external nutrient inputs in the EU). Notwithstanding this variability, in many cases the scope for raising fertilizer productivity is substantial. The degree to which such productivity gains will be pursued depends to a great extent on economic incentives.

The projected slowdown in the growth of fertilizer consumption is due mainly to the expected slowdown in crop production growth (Table 4.1). The reasons for this have been explained in Chapter 3. Again, this is not a sudden change but a gradual process already under way for some time, as illustrated by the annual growth rates for the last ten years (1989-99) shown in Table 4.15. In some cases it would even represent a«recovery» as compared with recent developments. As mentioned, fertilizer is most productive in the absence of moisture constraints, i.e. when applied to irrigated crops. For this reason, the expected slowdown in irrigation expansion (Section 4.4.1) will also slow the growth of fertilizer consumption. The continuing trend to increase fertilizer use efficiency, partly driven by new techniques such as biotechnology and precision agriculture, will also reduce mineral fertilizer needs per unit of crop output. There is an increasing concern about the negative environmental impact of high rates of mineral fertilizer use. Finally there is the spread of organic agriculture, and the increasing availability of non-mineral nutrient sources such as manure; recycled human, industrial and agricultural waste; and crop by-products. All these factors will tend to reduce growth in fertilizer consumption.

4.6.2 Farm power

Human labour, draught animals and engine-driven machinery are an integral part of the agricultural production process. They provide the motive power for land clearance and preparation, for planting, fertilizing, weeding and irrigation, and for harvesting, transport and processing. This section focuses on the use of power for primary tillage. Land preparation represents one of the most significant uses of power. Since land preparation is power intensive (as opposed to control intensive), it is usually one of the first operations to benefit from mechanization (Rijk, 1989). Hence any change in the use of different power sources for land cultivation may act as an indicator for similar changes in other parts of the production process.

Regional estimates of the relative contributions of different power sources to land cultivation have been developed from estimates initially generated at the country level. On the basis of existing data and expert opinion, individual countries were classified into one of six farm power categories according to the proportion of area cultivated by different power sources, at present and projected to 2030. The categories range from those where hand power predominates, through those where draught animals are the main source of power, to those where most land is cultivated by tractors. The figures were subsequently aggregated to estimate the harvested area cultivated by different power sources for each region (see Box 4.5 for details of the methodology).

Box 4.5 Methodology to estimate farm power category

Individual countries were classified by expert opinion into one of six farm power categories according to the proportion of area cultivated by different power sources. The six categories identified are given in the table below. The percentages of the area cultivated by different power sources are indicative only and refer to harvested land, which represents the actual area cultivated in any year, taking into account multiple cropping and short-term fallow. Upper and lower limits were set for the area cultivated by each power source (bottom row in the table).

Farm power category at country level

Percentage of area cultivated by each power source

Hand

Draught animals

Tractors

A

humans are the predominant source of power, with modest contributions from draught animals and tractors

>80

< 20

<5

B

significant use is made of draught animals, although humans are still the most important power source

45-80

20-40

<20

C

draught animals are the principal power source

15-45

>40

<20

D

significant use is made of motorized power,
including both two-wheel and
four-wheel tractors

20-50

15-30

20-50

E

tractors are the dominant power source

<5

<25

50-80

F

fully motorized

<10

<10

>80

Minimum and maximum percentage

5-90

5-70

2-90

Where possible, country classifications were verified against existing data. Sources included a number of country farm power assessment studies commissioned by the Agricultural Engineering Branch of FAO and published reports. The classifications from various sources proved to be fairly consistent. The categories were converted into the physical area cultivated by each power source by multiplying the estimated percentage figure for each power source with the data for harvested area for the base year 1997/99 and the projected area for 2030. The country figures were subsequently aggregated to estimate the harvested area cultivated by different power sources for subregions and regions.

This approach has several advantages. First, it highlights the role of humans as a source of power in those parts of the world where they are responsible for much of the land preparation. This is essential for understanding concerns arising from future projections about the size and composition of the agricultural labour force, particularly in countries where a sizeable share of the population is expected to be affected by HIV/AIDS within the next 20 years. The significance of humans as a power source can be easily overlooked if their contribution is expressed solely as a percentage of the total power input, rather than the area they cultivate. Second, when projecting future combinations of farm power inputs at the country level, account can be taken, albeit by way of expert judgement, of the developments in the overall economy and in the agricultural sector, competing claims on resource use, and opportunities for substitution between power sources. Third, this approach is independent of estimates of inventories of draught animals and tractors, data for which are often unreliable or not readily available. The number of tractors and draught animals working in agriculture may vary considerably from published data. These numbers depend on several unknown variables, such as the working life, the proportion working in agriculture as opposed to off-farm activities, and the proportion in an operational state. Finally, the process of converting different power sources into a common power equivalent is a process fraught with difficulties and inaccuracies.

Nevertheless, there are also several limitations associated with the methodology. First, only one farm power category is selected to represent the power use for land cultivation within an entire country. This overlooks the diversity that exists inside many countries, particularly when the use of a specific power source is highly influenced by soil and terrain constraints, by cropping patterns, or is highly differentiated between commercial/ estate and smallholder sectors. A second limitation is the use of a single average percentage figure for each power source to convert categories to harvested area, rather than actual percentages.

Overall results. It was estimated that in 1997/99, in developing countries as a whole, the proportion of land cultivated by each of the three power sources was broadly similar. Of the total harvested area in developing countries (excluding China), 35 percent was prepared by hand, 30 percent by draught animals and 35 percent by tractors (Table 4.16). By 2030, 55 percent of the harvested area is expected to be tilled by tractors. Hand power will account for approximately 25 percent of the harvested area and draught animal power (DAP) for approximately 20 percent.

Table 4.16: Proportion of area cultivated by different power sources, 1997/99 and 2030

Region

Percentage of area cultivated by different power sources

Hand

Draught animal

Tractor

All developing countries

1997/99

35

30

35

2030

25

20

55

Sub-Saharan Africa

1997/99

65

25

10

2030

45

30

25

Near East/North Africa

1997/99

20

20

60

2030

10

15

75

Latin America and the Caribbean

1997/99

25

25

50

2030

15

15

70

South Asia

1997/99

30

35

35

2030

15

15

70

East Asia

1997/99

40

40

20

2030

25

25

50

Notes: Figures have been rounded to the nearest 5 percent. China has been excluded from the analysis because its size and diversity made it impossible to estimate a single farm power category for the country.

There are marked regional differences in the relative contributions of the power sources, both at present and in the future. Tractors are already a significant source of power in the Near East/North Africa region and in Latin America and the Caribbean: approximately half of the harvested area is currently prepared by tractor in these regions. This is expected to rise to at least 70 percent of harvested area by 2030. Draught animals are at present relatively important sources of power in the rice and mixed farming systems of South and East Asia, accounting for over one-third of harvested area. However, the shift to motorized power by 2030 will be substantial. The area cultivated by tractors will rise in South Asia from 35 percent of harvested area to 70 percent, and in East Asia (excluding China) from 20 percent to over 50 percent. This increase in area cultivated by tractor arises from two factors: an increase in total harvested area at the country level, combined with a reduction in the area cultivated by humans and draught animals as a result of substitution between power sources.

In contrast, humans are and will continue to be the main power source in sub-Saharan Africa. Almost two-thirds of the harvested area is prepared by hand at present and although this will fall to 50 percent by 2030, the physical area involved will remain broadly constant. The area cultivated by draught animals and tractors is expected to increase (both in physical area and proportional terms) but they will not offset the dominance of hand power.

When countries are classified by farm power category, some common characteristics can be observed:

Forces changing the composition of farm power inputs. The stimulus to change the composition of farm power inputs will come from either changes in the demand for farm power or from supply-side changes, or both. Any increase in total agricultural output (be it from area expansion, an increase in cropping intensity or an increase in yield) requires additional power, if not for technology application then for handling and processing increased volumes. Similarly, land improvements (such as terracing, drainage or irrigation structures), soil conservation and water harvesting techniques frequently place additional demands on the power resource.

In response, farmers can either increase their inputs of farm power or increase the productivity of existing inputs through the use of improved tools and equipment. Alternatively, adopting different practices or changing cropping patterns may reduce power requirements. For example, the use of no-till and direct seeding practices eliminates the need for conventional land preparation and tillage; broadcasting rice overcomes the labour-intensive activity of transplanting seedlings, and the use of draught animal power, benevolent herbicides or no-tillage with continuous soil cover (see Chapter 11) can overcome labour bottlenecks associated with weeding.

Motivations to mechanize may also arise from supply-side changes in the availability and productivity of farm power inputs, as well as a wish to reduce the drudgery of farm work. The health, nutritional status and age of the workforce affect the productivity of labour. The availability of household members for farm work is influenced by other claims on their time, such as household tasks, schooling and opportunities for off-farm work. The household composition also changes through rural-urban migration or the death of key household members. The productivity of draught animals is affected by their health and nutrition, the training of animals, operator skills and availability of appropriate implements. Productive and sustainable use of motorized inputs is dependent on operator skills, appropriate equipment and an infrastructure capable of providing timely and cost-effective access to repair and maintenance services.

The changing composition of farm power inputs will also have an impact on the division of agricultural tasks among household members. The range of tasks performed by different members of the household varies according to sex, age, culture, ethnicity and religion (see Box 4.6). It also varies according to the specific crop or livestock, sources of power input and equipment used.

Box 4.6 Gender roles and the feminization of agriculture

Women are key players in both cash and subsistence agriculture. Their daily workload is characterized by long hours, typically 12- to 14-hour days, with little seasonal variation. The use of their time tends to be fragmented, mixing farm work with household duties and other off-farm activities. The range of agricultural tasks performed by different household members varies according to sex, age, culture, ethnicity and religion. It also varies according to the specific crop or livestock activities, sources of power input and equipment used. Hence gender roles differ markedly, not only between regions and countries, but also within countries between neighbouring communities.

In sub-Saharan Africa, women contribute between 60 and 80 percent of the labour in food production. While there are significant variations in gender roles, women overall play a major role in planting, weeding, application of fertilizers and pesticides, harvesting, threshing, food processing, transporting and marketing. Men are largely responsible for clearing and preparing the land, and ploughing. They also participate alongside women in many of the other activities. In many countries men are responsible for large livestock and women for smaller animals, such as poultry, sheep and goats. Women are usually most active in collecting natural products, such as wild foods, fodder and fuelwood. Men are usually associated with the use of draught animals or tractors. However, with appropriate training and implements, women also prove to be very effective operators of mechanized inputs.

In Asia, women account for 35 to 60 percent of the agricultural labour force. Women and men often play complementary roles with a division of labour similar to that found in sub-Saharan Africa. In many Asian countries women are also very active with livestock, collecting fodder, preparing buffaloes for ploughing, feeding and cleaning other cattle, and milking. In Southeast Asia they play a major role in rice production, particularly in sowing, transplanting, harvesting and processing. Women also supply a significant amount of labour to tea, rubber and fruit plantations.

In the Near East, women contribute up to 50 percent of the agricultural workforce. They are mainly responsible for the more time-consuming and labour-intensive tasks that are carried out manually or with the use of simple tools. In Latin America and the Caribbean, the rural population has been decreasing in recent decades. Women are mainly engaged in subsistence farming, particularly horticulture, poultry and raising small livestock for home consumption.

Gender roles are dynamic, responding to changing economic, social and cultural forces. The rural exodus in search of income-earning opportunities outside agriculture, usually dominated by men, has resulted in increasing numbers of female-headed households and the«feminization» of agriculture. Similar patterns arise from the death of male heads of household. Women are being left to carry out agricultural work on their own, changing the traditional pattern of farming and the division of tasks among household members. For example, women in female-headed households without recourse to adult male labour may clear and prepare land, including ploughing with oxen (tasks which traditionally would have been performed by men).

The feminization of agriculture, most pronounced in sub-Saharan Africa but also a growing phenomenon in other parts of the world, has significant implications for the development of agriculture. The needs and priorities of both rural women and men must be taken into account in any initiative to support and strengthen the sector.

Source: Based on FAO (1998a).

Patterns of mechanization up to 2030. Most of the changes in farm power categories during the next 30 years are expected to occur in countries that already make significant use of tractors. By 2030, tractors will be the dominant source of power for land preparation in southern Africa, North Africa/Near East, South Asia, East Asia and Latin America and the Caribbean. Southeast Asia is also expected to shift from draught animals to making greater use of tractors. The reasons underlying these shifts are explained below.

In a few countries, it is expected that the present composition of farm power inputs is not sustainable. In eastern Africa, for example, the number of draught animals has been decimated in some areas through livestock disease and cattle rustling, thereby removing a principal power source from certain farming systems. The sustainability of tractor-based systems is highly dependent on the profitability of agriculture and an infrastructure capable of providing timely access to fuel and inputs for repairs and maintenance. The failure of government-based initiatives often results from introducing a level of mechanization that is inappropriate for the state of economic development and political stability. As a result, in the absence of further government interventions, it is expected that some countries will revert to increasing the use of hand or draught animal power during the next 30 years.

Persistence of hand and animal power in sub-Saharan Africa. Human labour is the most significant power source throughout sub-Saharan Africa. The human contribution is most pronounced in Central and western Africa where it accounts for 85 and 70 percent of harvested area, respectively. These areas include the forest-based farming systems of Central Africa, characterized by shifting cultivation and the gathering of forest products, the root crop system stretching across West Africa, Central Africa and parts of East Africa, and the cash tree crop system in West Africa (FAO, 2001d). A relatively high proportion of rainfed land is under cultivation (45 percent of the potential area) but it is not used intensively as reflected in a relatively low cropping intensity of 60 percent. The presence of trees, stumps and shrubs makes it difficult to use ploughs without considerable investment of time and effort in land clearance (Boserup, 1965; Pingali, Bigot and Binswanger, 1987). Moreover, the incidence of tsetse fly (which breeds in tropical forests and forest margins) makes the area unsuitable for many types of draught animals. There is very little irrigated land.

Draught animals (predominantly work oxen) are concentrated on rainfed land in the cereal-based farming systems in the northern parts of West Africa, throughout the maize mixed systems of eastern Africa and the highland mixed systems of Ethiopia. Countries making significant use of tractors are scattered throughout the region.

Two-thirds of the countries in sub-Saharan Africa are not expected to change their power category by 2030. Although there will be some movement in the relative contributions of hand, draught animal and tractor power to land preparation, much of the region will continue to be cultivated using hand and animal power. All countries that are expected to change either from hand power to draught animal power, or from DAP to tractors, will experience lower population growth rates, higher incomes per head and higher income growth rates than those countries with the same farm power category at present that are not expected to change.

The process of urbanization in this region will provide some stimulus to switch power sources, as it not only draws labour away from the agricultural sector but also has implications for wage levels and the composition of the remaining labour force. Typically the young, able-bodied, educated and skilled migrate. The shift to urbanization is most pronounced in countries switching from DAP to tractors or already using tractors as the dominant power source. In these countries, a growth of almost 30 percent in the proportion of the population living in urban areas is expected by 2030, twice the rate of urbanization expected in countries not switching power source or shifting to DAP. Countries that will continue to use draught animals as a significant source of power will remain predominantly rural.

Another factor driving the process of change in eastern and southern Africa will be the impact of HIV/AIDS on the workforce (Box 4.7). Those countries that are expected to switch from hand power to DAP are projected to lose almost 20 percent of their agricultural labour to AIDS by 2020, that is, more than twice as much as those countries continuing to use hand power. Similarly, those shifting from DAP to tractors are expected to experience higher losses in their labour force (12 percent by 2020) than countries continuing with DAP. Some of the highest losses (16 percent by 2020) are projected for countries already making significant use of tractors. Thus the impact of HIV/AIDS will make it vital for many countries to change their source of farm power in order to overcome serious labour shortages at critical times of the farming year.

Box 4.7 Household vulnerability to the loss of human and draught animal power

Households reliant on human power, and draught animals to a lesser extent, are extremely vulnerable to the loss of their principal power source. More than 15 countries in sub-Saharan Africa are projected to lose at least 5 percent of their workforce to HIV/AIDS by 2020. The pandemic will impact heavily in the agricultural sector where losses will typically account for at least 10 percent of the workforce and, in at least five countries, more than 20 percent. In a region where people are a significant, and often the dominant, source of power for both household and farm activities, this loss of labour will have a dramatic impact on rural livelihoods.

HIV/AIDS usually strikes at the heart of the household, killing women and men in their economic prime. Not only do households lose key family members but they also lose time spent by other household members caring for the sick. The situation is exacerbated by urban dwellers returning to their villages to be cared for when they become ill, thereby placing further strain on rural households. In addition to the immediate emotional, physical and financial stresses, the remaining family members have to take on the long-term care of orphan children. In some cultures, widows also have to cope with the threat of property-grabbing by the relatives of the deceased.

In parts of eastern and southern Africa, the vulnerability of rural livelihoods has been worsened by the decimation of the DAP base caused by the switch from hardy local breeds to cross-breeds, coupled with the failure to carry out regular healthcare practices and increased livestock susceptibility to disease (such as East Coast fever). Cattle rustling is also a threat, particularly in areas close to international borders. In the absence of alternative power sources, such as tractor hire, households have reverted to hand power. Areas under cultivation have fallen significantly and households that once were food self-sufficient and producers of surplus for sale, now regularly experience food shortages. Household transport has become more problematic and the opportunity to earn additional income from hiring out draught animals has also disappeared.

Food insecurity, arising from the inability to produce or purchase sufficient food for the household throughout the year, is a persistent characteristic of subsistence agriculture. Short-term coping strategies include reducing the number of meals eaten per day, with very poor households spending up to two days between meals, or by switching to less nutritious foods. The poor may gather and sell natural products (such as wild fruits, mushrooms, tubers, firewood and grass thatch) or beg for food. Households may also engage in off-farm activities, trading and making handicrafts, or rely on remittances from family members living elsewhere. Some survival strategies, such as the sale of assets to buy food, taking out loans to purchase inputs, or hiring out family labour to work on other farms, invariably place the household at greater risk in subsequent seasons. Longer-term adaptive strategies to overcome labour shortages include reallocating tasks between household members, using labour-saving technologies and switching to less labour-intensive cropping patterns and practices.

Increasing use of tractors in the Near East/North Africa and Asia. The development of regional markets and strong links with Europe are expected to be important engines of growth for North African countries. Oil wealth will continue to underpin development in the Near East. Economic development will be coupled with continued growth in non-agricultural employment and the migration of people from the land to urban areas. By 2030, over 75 percent of the population in the Near East/ North Africa region will be living in urban areas. The option of using tractors becomes more viable with increasing costs of labour and increasing shortage of land for fodder production for draught animals.

Prospects for mechanization in Asia are based on projections of buoyant economies and high rates of growth in income per capita, but the process of urbanization would not appear to be so significant. More than half of the population in South Asia will continue to be based in rural areas by 2030. Single-axle tractors will be an increasingly important form of farm power in irrigated farming systems, which are suited to their use. The process of mechanization will be facilitated in this region by proximity to sources of manufacture, namely India (the world's largest manufacturer of tractors) and China (a source of low-cost power tillers).

Stable use of tractors in Latin America and the Caribbean. Projections for economic growth in Latin America and the Caribbean are on a par with other regions and per capita incomes are among the highest in the developing world. However, almost half of the countries in the region are not expected to change farm power categories during the next 30 years. Several countries are at the limits of technical change in terms of farm power. Much of their agricultural sector is already fully mechanized and any further expansion in the use of tractors will be largely constrained by topographical features, notably the Amazon basin and the mountainous regions of the Andes. In other countries (such as El Salvador, Guatemala, Mexico and Paraguay) the shift towards no-till farming and conservation agriculture may reduce or eliminate the need for the increased use of tractors. For a few countries, economic conditions are determining factors and stagnant incomes in the smallholder sector inhibit any increase in the use of tractors.


10Top 10 percent of countries ranked according to area allocated to the crop examined: China, India and Turkey for wheat; and India, China, Indonesia, Bangladesh and Thailand for rice.
11A nutrient budget is defined as the balance of nutrient inputs such as mineral fertilizers, manure, deposition, biological nitrogen fixation and sedimentation, and nutrient outputs (crops harvested, crop residues, leaching, gaseous losses and erosion).


Previous PageTop Of PageNext Page