Previous Page Table of Contents Next Page


Chapter 3
Social, economic and policy environment


Improving the livelihoods of the people in the Limpopo River Basin in a sustainable manner is one of the objectives of the integrated approach to catchment basin management. Understanding the human dimension is critical to designing projects and programmes that will secure livelihoods based on stable, productive and profitable use of natural resources, particularly land use. This chapter describes the social, economic, policy and institutional dimensions that affect the capacity of the people living in the Limpopo River Basin to live with recurrent drought and climate variability.

Social and economic characteristics

Population characteristics

Chapter 2 described the biophysical diversity and challenges facing integrated management in the Limpopo River Basin. These challenges are reflected in the diversity of rural versus urban populations of the basin. In Botswana and South Africa, the capital cities are situated in, and some of the largest urban populations reside within, the basin boundaries, such as Gaborone, Francistown, Pretoria, Polokwane, Thohoyandu and Witbank. Not reflected in the basin population figures (Table 15) are other large urban centres (Johannesburg, Maputo and Bulawayo) that lie on the fringes of the basin and influence, or are influenced by, events and activities within the basin.

Aside from these urban centres, which for the most part are located at the headwaters (or near the mouth in the case of Maputo), the Limpopo River Basin is predominantly rural. On average, at the national level, the population in the Limpopo River Basin countries is just more than 50 percent rural - ranging from 31 percent in Botswana to 66 percent in Zimbabwe. However, at the subnational and district level within the basin, the population is predominantly rural - more than 60 percent. About 8 million people live in rural areas within the Limpopo River Basin (Table 16).

Botswana

According to the 1991 census, nearly 46 percent of Botswana’s population was urban-based, an increase of 18 percent over the 1981 estimates (GOB, 1992b). While much of this growth is a consequence of the continued expansion of the larger urban areas such as Gaborone and Francistown, most of the increase is attributed to a reclassification of some larger villages from rural to urban.

Eighty-three percent of Botswana’s people live in the eastern hardveldt region with its characteristics of better surface water and groundwater availability, good communications, proximity to South Africa, access to markets, and better-quality soils compared with the rest of the country. The Limpopo River Basin covers most of the hardveldt and falls within the rural administrative districts of North East, Central, Kgatleng and Kweneng. This catchment area accounts for 59 percent of the country’s population and 28 percent of its area. The average national population density is only 2.3 persons/km2, although the highest population density is in the districts of Kweneng (21 persons/km2) and Kgatleng (15 persons/km2) because of the location effects of greater Gaborone (GOB, 1992b).

TABLE 15
Selected statistics of the four basin countries

Country

Total area

Area of country within basin

As % of total basin area

As % of total country area

Population of country in 1998

Population in basin

As % of country population

Population density in basin

(km2)

(km2)

(%)

(%)

(million)

(million)

(%)

(persons/km2)

Botswana

581 730

80 118

19

14

1.6

1.0

59

12.5

Mozambique

801 590

84 981

21

11

16.5

1.3

7

15.3

South Africa

1 221 040

185 298

45

15

42.1

10.7

24

57.7

Zimbabwe

390 760

62 541

15

16

11.4

1.0

9

16.0

Total


412 938



71.6

14.0


33.9

TABLE 16
Estimates of rural population in the Limpopo River Basin

Country/Province

2001 rural population estimates for basin provinces

Botswana


835 000


Central

540 000



Kwaneng

190 000



Kgatleng

60 000



North East

45 000


Mozambique


1 045 000


Gaza

1 000 000



Inhambane

45 000


South Africa


5 400 000


Gauteng

100 000



Limpopo

4 300 000



Mpumalanga

500 000



North West

500 000


Zimbabwe


900 000


Matabeleland South

550 000



Masvingo

250 000



Midlands

100 000


Total


8 180 000

Note: Estimates based on provincial and not basin-specific figures.

Although the population of Botswana is small, it has had one of the highest growth rates in Africa (3.5 percent). This high growth rate has created an age structure with a very high ratio of children (unproductive) to adults (productive). Nearly 43 percent of the country’s population was below the age of 15 years in 1991. The young age structure of the population will probably persist for some time owing to high fertility levels and increasing infant and child survival rates. In the period 1992-2001, the growth rate was 2.67 percent (GOB, 2002).

Migration has played an important role in the dynamics of the country’s population through either permanent urban migration or through seasonally changing settlement patterns. However, the effect of the latter is less than it was during the 1970s and 1980s as a result of more settlements being provided with piped water and other social services. In 1981, some 50 percent of the population were enumerated within 200 km of Gaborone. By 1991, 50 percent of the country’s population resided within 100 km of the capital.

Mozambique

The Limpopo River Basin in Mozambique falls almost entirely within Gaza Province. It also covers portions of three districts in Inhambane Province. The catchment area of the Limpopo River Basin represents 11 percent of the total national land area and includes just more than 7 percent of the total population. The density of the population within the Limpopo River Basin ranges from 1 person/km2 in Chigubo District to more than 100 persons/km2 in rural Xai-Xai District. The basinwide average in Mozambique of 15 persons/km2 compares with a national population density of 21 persons/km2. The population density in Gaza Province follows very closely the major agro-ecological zones and biophysical characteristics (Chapter 2). Market accessibility and infrastructure have also had a strong influence on where settlements have developed.

The percentage of the population living in urban and rural areas in Gaza Province depends in part on the definition of an urban area. According to GOM-INE (1999b), nearly 93 percent of Gaza Province was considered rural, with the city of Xai-Xai on the coast the only urban centre. The 1997 census (GOM-INE, 1999a) included more of the major towns in the definition of an urban area (e.g. Chibuto, Chokwé and Macia), giving a rural percentage of 75 percent. Either way, the area is predominantly rural and the population density decreases significantly with distance from the coast and the Chokwé area.

South Africa

Nearly 45 percent of the total population of South Africa lives in the four northern provinces (Limpopo, North West, Mpumalanga and Gauteng). Limpopo Province is almost completely contained within the Limpopo River Basin, as are parts of North West, Mpumalanga and Gauteng Provinces. Limpopo Province is by far the most dominant administrative zone in the basin - representing more than half of the rural population - for a total of more than 4 million people.

The average population density of South Africa is about 33 persons/km2 (GOSA-StatsSA, 2001b). The distribution is very skewed, with the majority of the population living in the coastal provinces and around Johannesburg and Pretoria. The population density ranges from less than 2 persons/km2 in Northern Cape, to 40 persons/km2 in Limpopo Province and more than 430 persons/km2 in Gauteng.

Population density is particularly high in the former homelands. This is largely attributed to past migration policies, which prevented the African population from migrating out of the homeland areas. Migration has also played a significant role in redistributing South Africa’s population, especially since the formation of the new republic in 1994. Interprovincial migration data for 1992-96 indicate that Limpopo Province and Gauteng Province lost a large number of people through outmigration. However, Gauteng Province was also the most popular migration destination in the country (GOSA-NPU, 2000).

Zimbabwe

According to the population census of 1992, about 8 percent of Zimbabwe’s population resided in the Limpopo River Basin, then estimated at 850 000 people, and currently estimated at just more than 1 million. The Limpopo catchment area in Zimbabwe falls predominantly in Matabeleland South Province, as well as portions of two districts in Masvingo Province (Mwenezi and Chiredzi) and one district in Midlands Province (Mberengwa).

The average population density in the semi-arid regions of the Limpopo River Basin is generally low, ranging from 6 persons/km2 in Beitbridge District to 23 persons/km2 in Umzingwane, adjacent to Bulawayo, compared with 30 persons/km2 at the national level. However, the primarily rural district of Mberengwa has a population density of 37 persons/km2, comparable with the more densely populated rural areas around Harare.

HIV/AIDS

Sub-Saharan Africa (SSA) is home to about 70 percent of the 36 million people currently living with HIV/AIDS worldwide. In this region, an estimated 3.8 million adults and children became infected with HIV in 2000, bringing the total number of people living there with HIV/AIDS to 25.3 million. In the same period, millions of Africans infected in earlier years began experiencing ill health, and 2.4 million people at a more advanced stage of infection died of HIV-related illness.

Although SSA heads the list as the region with the largest annual number of new infections, there may be a new trend on the horizon: regional HIV incidence appears to be stabilizing. Because the long-standing African epidemics have already reached large numbers of people whose behaviour exposes them to HIV, and because effective prevention measures in some countries have enabled people to reduce their risk of exposure, the annual number of new infections has stabilized or even fallen in many countries. These decreases have now begun to balance out rising infection rates in other parts of Africa, particularly the southern part of the continent.

Among the countries of the world hardest hit by HIV/AIDS are South Africa and Botswana (Table 17). In Botswana, almost 36 percent of all adults are infected with HIV/AIDS. Life expectancy has dropped from 61 years seven years ago to 39 years today, and the figure is projected to fall below 29 years by 2010. Without HIV/AIDS, it would have been more than 66 years. The epidemic began in South Africa much later than it did in other countries in Africa. By the mid-1990s, infection rates among pregnant woman were increasing rapidly. South Africa is now facing one of the most serious epidemics in the world (U.S. Census Bureau, 2001).

The provinces that are within or largely within the basin may not always be representative of the basin country at large. In Botswana, the highly populated areas of the basin districts (parts or all of the Kweneng, Southern, South-East, Kgatleng and Central Districts) are relatively highly affected. In Mozambique, Gaza Province (covering the basin area) appears not to be among the most highly affected. In South Africa, Limpopo Province is less affected than the provinces of Mpumalanga, Gauteng and North West (parts of which are within the basin). Urban and mining centres (with the exception of the Western Cape) appear to be most affected. In Zimbabwe, Matabeleland South is slightly less affected than Manicaland and Masvingo Provinces in the east (U.S. Census Bureau, 2001).

TABLE 17
HIV prevalence in countries in the Limpopo River Basin


Botswana

Mozambique1

South Africa

Zimbabwe

Population

1 597 000

19 105 000

39 900 000

11 343 000

Population aged 15-49 years

786 000

No data

20 982 000

No data

Percent of total population with HIV

18


11


HIV infected people

290 000


4 200 000


Adult prevalence rate of HIV (%)

36

13

20

25.12

Adult female HIV population: % of total

51

No data

55

No data

Pregnant women HIV + prevalence (%)

43


19


AIDS orphans (living)

55 000


338 000


1 2001 figures (American Friends Service Committee, 2003).

2 CIA World Factbook gives a figure of 33.7 percent (CIA, 2003); 24 percent quoted by SADC-FANR (2003)

Main source: U.S. Census Bureau (2001)

The effects of HIV/AIDS are far reaching. It has a severe effect on affected households, children and their schooling, the level of services rendered, and on business and economic growth. HIV/AIDS creates new pockets of poverty where parents and breadwinners die and children leave school earlier in order to support the remaining children. Affected households bear the brunt of the misery caused by the epidemic. Rising sickness and death often take place against a background of deteriorating public services, poor employment prospects and endemic poverty that are not directly related to the HIV epidemic, but that may be exacerbated by it (UNAIDS-WHO, 2000).

HIV is reducing the numbers of children in school. HIV-positive women have fewer babies, in part because they may die before the end of their childbearing years, and up to one-third of their children are themselves infected and may not survive to school age. Many children who have lost their parents to AIDS, or are living in households which have taken in AIDS orphans, may be forced to drop out of school in order to start earning money, or simply because school fees have become unaffordable. Teacher shortages may be looming (UNAIDS-WHO, 2000).

Some recent survey results show just how great the future impact of HIV is likely to be on business. A 1999 study among miners in southern Africa found that more than one-third of employees in their late 20s and 30s were infected with HIV, as were one-quarter of young and older employees. Rates among workers in other sectors are similarly high, at least in South Africa. For example, in a sugar mill, 26 percent of all workers were living with HIV. There, as in the mining industry, HIV rates were higher among unskilled workers than among managerial-level workers (UNAIDS-WHO, 2000).

It remains exceptionally difficult to gauge the macroeconomic impact of the epidemic. Despite incomplete data, there is growing evidence that as HIV prevalence rates rise, both total and growth in national income (GDP) fall significantly. In South Africa, where per capita income is six times the average for SSA and the national economy accounts for 40 percent of the total economic output of the region, the overall economic growth rate in the next decade is likely to be 0.3 to 0.4 percentage points lower every year than it would have been without AIDS. Cumulating the slower economic growth over time, by 2010, real GDP may well be 17 percent lower than it would have been in the absence of AIDS. In terms of current value, that would wipe US$22 000 million off South Africa’s economy - more than twice the entire national production of any other country in the region except Nigeria (UNAIDS-WHO, 2000).

AIDS is likely to cause skills shortages in most sectors of the economy, creating major bottlenecks in business and production. This will be exacerbated by its undermining effect on education, and on the potential to expand skills as quickly as they are needed. HIV infection rates are highest among individuals in the workforce without special skills; so are the unemployment rates (about 30 percent in South Africa). Thus, in theory, individuals who are not currently employed may replace unskilled workers dying of AIDS. The skills shortage can be expected to be even more acute in neighbouring

Botswana, which is already importing white-collar workers (UNAIDS-WHO, 2000). There are indications that AIDS is starting to have negative affects on resource-poor agriculture. For example, with respect to Zimbabwe, the United Nations (UN) Relief and Recovery Unit noted “productivity has been severely affected in the agriculture sector as a direct result of the HIV/AIDS pandemic in the country” (UN-IRIN, 2003).

Human development and poverty

The rural population of about 8 million people living in the Limpopo River Basin face special challenges to make a living. The biophysical description in Chapter 2 showed that this area is predominantly semi-arid, with little arable land and very low potential for agriculture. Two related concepts that are useful for measuring “how people are doing” are human development and poverty. Human development can be thought of as the process of enlarging people’s choices so that they can live longer and healthier lives (SADC-UNDP, 1998). This is usually measured in terms of educational and health opportunities, as well as some measure of wealth or standard of living (Box 14). Poverty is usually defined as living below a certain income or income-poverty threshold, but it has many dimensions (IFAD, 2001).

BOX 14

Human development and poverty indices

The Human Development Index (HDI) was introduced in 1990 as an attempt to measure and rank countries according to progress in human development beyond a simple gross national product figure. Instead of using only a financial measure, the HDI is a composite of three basic components of human development:

longevity - as measured by life expectancy;

knowledge - as measured by a combination of adult literacy (two-thirds weight) and mean years of schooling (one-third weight);

standard of living - as measured by purchasing power, based on real GDP per capita adjusted for the local cost of living (purchasing power parity).

The index ranges from 0 to 1. Although useful at the global scale, indices at the national level can conceal much that is happening within the country. The best solution would be to create separate HDIs for the most significant groups, e.g. by gender or by income group. Separate HDIs would reveal a more detailed profile of human deprivation in each country and disaggregated HDIs have been conducted in some SADC countries.

Similarly, human poverty is usually defined as living below a certain income level or income poverty line. From a human development perspective, poverty means the denial of choices and opportunities most basic to human development, including deprivation in health and survival, lack of knowledge, denial of opportunities for a creative and productive life, social exclusion, lack of freedom, as well as deprivation in income. Therefore, the Human Poverty Index (HPI) was proposed in 1997 as a new way of measuring poverty in developing countries. The index measures the proportion of the population affected by three key deprivations affecting their lives:

deprivation in survival - measured by the percentage of people expected to die before age 40;

deprivation in knowledge - measured by the percentage of illiterate adults;

deprivation in “economic provisioning” - measured by the percentage of people without access to health services and safe water, as well as the percentage of underweight children under five.

The index ranges from 0 (low) to 100 (high). No class limits are given for low, moderate and high.

The HDI and HPI can be useful alternatives to gross national product for measuring the relative socioeconomic progress of nations. They enable people and their governments to evaluate progress over time - and to determine priorities for policy interventions. They also enable instructive comparisons of the experiences in different countries.

Source: SADC-UNDP (1998); UNDP (2003).

Table 18 gives selected national-level HDI values (see Box 14) for the countries in the Limpopo River Basin. At the national level, as of 1998, three of the four countries were in the medium (values between 0.500 and 0.790) human development range: South Africa, Botswana, and Zimbabwe (SADC-UNDP, 1998). Mozambique has the lowest HDI value in the region (0.374), although in percentage terms, the situation has been improving steadily (GOM-UNDP, 1999). The impact of the HIV/AIDS pandemic on life expectancy is an important factor in slowing the growth, or reducing the HDI, especially for Botswana, South Africa and Zimbabwe.

As income is an important component of the HDI, an unequal distribution of income may skew the results, especially at the national level. Income inequality can be estimated by the Gini coefficient, which is defined as the maximum vertical deviation between the perfect diagonal and the Lorenz curve, which is a graphical representation of the proportionality of a distribution. The higher the Gini coefficient, the greater the inequality. Southern Africa has some of the highest Gini coefficients in the world and, therefore, it is appropriate to apply an adjustment for income inequality to the HDI values. The national-level HDI values for Botswana, South Africa and Zimbabwe drop dramatically, by more than 20 percent, when adjusted for income inequality. Thus, Botswana and Zimbabwe drop to an HDI of about 0.50 and South Africa moves to 0.60, which is perhaps more representative of the majority of the people living in the rural areas of these countries.

Although it is difficult to make comparisons across the four countries at the subnational level, there is also great disparity in HDI values between geographical regions and urban and rural areas.

TABLE 18
Selected human development and poverty indicators for the basin countries

Country

Global HDI
(1998)

Global HPI-1
(1998)

People not
expected to live to
age 40 (1998)
(%)

Adult illiteracy
rate
(%)

Population without
access to safe
water (1998)
(%)

Underweight
children under
age five (1998)
(%)

Botswana

0.593

28.3

37.1

24.4

10.0

17.0

Mozambique

0.341

50.7

41.9

57.4

54.0

26.0

South Africa

0.697

20.2

25.9

15.4

13.0

9.0

Zimbabwe

0.555

30.0

41.0

12.8

21.0

15.0

However, the HDI is generally lower (higher HPI) in more remote rural areas, where education, health and employment opportunities are more limited.

Botswana

Botswana has completed two national-level human development reports in recent years. A 1997 report recommended nine issues to be considered for future human development studies, the first being conducted in 2000 on the theme Towards an AIDS free generation (GOB-UNDP, 2000). This report provides a detailed assessment of the impact of the HIV/AIDS epidemic on Botswana’s society. With the highest reported HIV prevalence rate in the world, Botswana has recognized the importance of integrating a HIV/AIDS strategy with national development and poverty reduction.

The Botswana HPI was calculated using the percentage of children that die before the age of 5 years as a measure of a long and healthy life, not the percentage of people who will not survive 40 years of age. The HPI reveals that more than 25 percent of the population live in human poverty. The HPI values for rural areas are nearly double those in the urban areas - 39.0 compared with 16.8. These disparities are closely linked to available services such as schooling, water supply and health care (GOB-UNDP, 2000).

Mozambique

Mozambique has also produced several national-level human development reports in recent years, although the values are at the provincial level (GOM-UNDP, 1998 and 1999). Severe problems of poverty still exist in Mozambique, affecting nearly 70 percent of the population, or 10.9 million people, according to the latest poverty assessment (GOM-UNDP, 1998). The incidence is higher in rural than in urban areas, with rural headcount reaching 71.2 percent compared with 62.0 percent in urban areas. The incidence of poverty is highest in the central region, whereas the north and south are nearly equal. However, if Maputo city - which has low rates relative to the rest of the country - is excluded from the southern region, the remainder of the southern region (including Gaza Province) has poverty rates higher than the northern region, and not significantly different from the central region.

Nationally, the average household size is 4.8 persons, but among the poor the average household size is 5.6 persons compared with 3.6 persons for the non-poor. The difference is more pronounced in the rural areas (5.5 persons for the poor, 3.3 persons for the non-poor) than in urban areas (6.0 persons for the poor, 4.7 persons for non-poor). Only 32 percent of the adult rural population and 71 percent of the adult urban population are literate. The differences are greater between regions and sexes than between levels of poverty; adult poor = 54 percent literate, adult non-poor = 63 percent, whereas, males = 59 percent and females = 24 percent (change of 36 percent) and urban rural change is 39 percentage points. Given the dependence of the population on agricultural production, and the important role played by women in agricultural activity, the extremely low literacy rate of rural women has serious implications for agricultural productivity in the country (GOM-UNDP, 1998).

South Africa

The national-level HDI varies across geographical regions in South Africa. Gauteng Province has the highest HDI in South Africa (0.717) while Limpopo Province has the lowest at 0.531 (GOSA-StatsSA, 2001a). There is also a close relationship between HDI values, rural areas, and former homelands. For example, Limpopo Province has the highest percentage of rural population (89 percent) and the highest percentage of the population living in former homeland areas and the lowest HDI value at the national level.

According to a recent poverty study (Whiteford and Van Seventer, 1999), 45 percent of South Africans are poor. The figure is even higher in mainly rural areas, and Limpopo Province has the highest poverty rate in the country - nearly 80 percent - compared with 45 percent nationwide, and 32 percent in Gauteng (GOSA-StatsSA, 2001a). Another national report (GOSA-NPU, 2000) further emphasizes the relative poverty and lack of human development in Limpopo Province. Some characteristics of Limpopo Province taken from this report include:

Zimbabwe

Zimbabwe has also produced two national-level human development reports, one focusing on poverty (GOZ-UNDP, 1998) and one focusing on globalization (GOZ-UNDP, 1999). The 1998 report discusses the relationship between poverty and health issues and economic development. The report states that although some progress has been made, the 1990s witnessed decreased income levels, a contraction of social expenditure, and low levels of economic growth. HIV/AIDS is also taking its toll and reducing life expectancy.

In terms of HDI and HPI characteristics, Zimbabwe also exhibits a disparity between geographical regions and between urban and rural communities. At the national level, Matabeleland South Province ranks highest (best) in terms of HDI or HPI. Four of the six districts in this province have HDI values greater than 0.60, which is near the national average of 0.62 (Table 19). Beitbridge District is the lowest in the Province, comparable with Mberengwa District (Midlands Province). The two districts in Masvingo Province, Mwenezi and Chiredzi, are ranked 74 and 65, respectively, out of the 77 districts and urban centres listed in the 1999 report.

Gwanda (urban) ranked first out of all 77 districts and urban centres in terms of lowest (best) HPI, higher than Harare or Bulawayo urban centres. This indicates good access to infrastructure (markets, schools, health clinics, water, and electricity) as compared with the more remote districts, which tend to be the poorest and least developed.

Livelihoods and food security

Understanding how rural populations live and maintain their livelihoods is crucial to understanding food security. As with many sectors presented in this situation analysis, there has been no systematic analysis of livelihoods and food-insecure populations conducted across the four countries of the Limpopo River Basin. However, using these concepts in their broadest sense (Box 15), the information that was obtained is presented to highlight the general types of livelihood systems in the Limpopo River Basin. This information is also useful for determining which populations are likely to be the most chronically food-insecure as well as at risk of drought and other climate-induced events.

TABLE 19
Selected poverty comparisons for Zimbabwe districts in the Limpopo River Basin

District

Non-survival to 40 years of age %

Illiteracy %

Underweight children %

Non-access to clean water %

No access to health care %

Living standard deprivation %

HPI

HDI

Mwenezi

22.0

39.1

30.3

31.7

0.7

20.9

29.8

0.44

Chiredzi

22.0

38.8

20.0

9.7

2.3

10.7

28.6

0.52

Beitbridge

16.9

37.8

10.0

14.2

3.4

9.3

27.1

0.55

Mberengwa

15.6

24.6

12.8

41.0

8.6

20.8

21.0

0.55

Bulilimamangwe

9.7

24.7

6.4

37.5

11.7

18.5

19.5

0.59

Matobo

7.7

18.5

9.0

34.1

9.3

17.7

15.9

0.60

Insiza

9.7

19.1

5.2

34.1

6.9

15.4

15.7

0.60

Gwanda rural

10.8

17.9

2.8

30.7

5.4

13.0

14.5

0.60

Umzingwane

9.7

14.7

9.3

18.7

1.0

9.6

11.9

0.62

Gwanda urban

12.0

6.6

0.0

0.8

0.0

0.3

8.7

0.67

Zimbabwe urban

16.9

19.6

10.0

1.0

8.8

3.7

16.0

0.62

Zimbabwe rural

16.9

19.6

14.7

36.5

8.8

17.1

17.9

0.62

Source: GOZ-UNDP (1999).

BOX 15

Livelihoods and vulnerability assessments

Research in recent decades has led to a wealth of methods and approaches for analysing and monitoring livelihoods and food security. The concept of using some form of livelihood system (LHS) is becoming common as the basis for development planning, understanding food security, as well as responding to various types of emergencies for many development and response agencies and organizations. Many agencies have incorporated LHS concepts into their programming cycles and development programmes (e.g. FAO and UNDP).

In general, the LHS approach is a systematic and structured way to understand how people make their living so that development and emergency response interventions can be matched more appropriately with their real needs. The basic objective is to analyse in a holistic way the various components (physical, natural, social, economic and human) that make up the livelihood structure (see DFID, 1998). These components, and the linkages and interactions between them, can than be studied to determine which factors are the most important, and which ones are least stable. In this way, more appropriate interventions can be developed. Participatory methods are encouraged so that the perspectives of the people are captured in the process.

Vulnerability assessment methods and techniques have been developed to help identify and understand food-insecure populations, generally using LHS concepts as the basis. Most methods respond to the basic definition of food security as given by the World Food Summit in 1996: “when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food which meets their dietary needs and food preferences for an active and healthy life.” Most use some combination of the generally accepted “pillars” of food security (food availability, access and utilization) in order to identify who does not meet the criterion.

As with livelihood analysis, there is no standardized approach to vulnerability assessments, although there is general agreement on basic concepts and components to consider. The outcomes may vary in terms of scale and level of detail, but the results of a vulnerability assessment should identify: (i) who and where are the most food insecure or vulnerable to becoming food insecure (to help prioritize/target populations for development/emergency interventions); and (ii) an understanding of why they are vulnerable (to help prioritize the appropriate type of intervention to meet their needs).

Sources and further information: DFID (http://www.dfid.gov.uk/); ODI (http://www.odi.org.uk/publications/susliv.html); FIVIMS (http://www.fivims.org/); SCF-UK (http://www.dinf.ne.jp/doc/japanese/twg/eng/contact/scf.html); USAID-EWS NET (http://www.fews.net/); UNDP (http://www.undp.org/); WFP (http://www.wfp.org/); IFPRI (http://www.ifpri.org); FAO (http://www.fao.org/); USAID/OFDA (http://www.usaid.gov/ofda/).

The analysis of livelihoods and vulnerability are also linked closely to some of the previous sections that discussed various biophysical and socio-economic characteristics, such as the risk of a drought occurring, the type of land use and farming systems, and poverty characteristics. More detailed surveys are needed to better understand the dynamics of livelihood systems and the relationship to food security, for example, between the various sources of own production and income, especially in these drought-prone, marginal production areas.

Various organizations listed in the sources under Box 15, as well as many others at national and local levels, are involved in livelihoods and vulnerability analysis. Efforts are underway to harmonize approaches and build capacity within the SADC region in order to build baseline information and expertise in this area (SADC, 2000).

Botswana

Although the agriculture sector of Botswana contributes only 4 percent to the national GDP and formal sector employment, 65 percent of the population within the Limpopo catchment area live on agricultural holdings and derive their livelihood mainly from agricultural activities (Table 20). More than 70 percent of these agricultural holders can be found in Central District and in Kweneng District.

TABLE 20
Profile of the traditional agriculture sector in Botswana

Characteristic

Administrative districts

Limpopo River Basin

Botswana total

North East

Central

Kgatleng

Kweneng

Population on agricultural holdings

79 200

161 170

28 150

116 200

384 720

584 280

Households with cattle (%)

39.9

60.9

60.1

46.2

52.8

53.1

Households with land (%)

94.2

63.8

79.3

72.7

73.5

69.6

Full-time farmers (%)

83.9

77.2

73.9

82.3

79.5

80.4

Source: GOB-FAO (1995).

Increasing demographic pressures are causing farm holdings to decrease in size, resulting in reduced options for grazing livestock, and consequently small ruminants are replacing cattle. The trends have created peri-urban production systems that include larger-scale commercial and intensive enterprises characterized by specialization and intensive market orientation, e.g. poultry, pigs, milk and beef feedlot systems, generally located near centres of consumption.

There is very little formal sector employment in Botswana rural areas. One report (GOB-CSO, 1996) indicates that for all rural areas in 1994, business profits accounted for only 15 percent of average household income, compared with 43 percent from cash earnings, 19 percent from the value of own produce consumed, and 14 percent from remittances from family members in urban formal employment. Income and expenditure data contained within the report show that the average rural household in Botswana was 51 percent self-sufficient in 1993/94 in terms of food requirement by value, compared with 68 percent in 1985/86.

The interplay between agricultural and nonagricultural activities has long existed in rural areas. The World Bank (1990) and others have suggested that the core strategy to alleviate poverty must be to create employment and increase rural incomes.

The 1993/94 Household Income and Expenditure Survey (HIES) (GOB-CSO, undated) indicated:

Mozambique

Mozambique has been involved in livelihood and food security analysis for many years. In the early 1990s, Medicins Sans Frontiere (MSF) developed a system to classify each district in Mozambique according to several vulnerability indicators. These structural vulnerability assessments are used to determine those areas or populations that are faced with chronic food insecurity. Table 21 describes structural vulnerability for Gaza Province. Only four variables are listed here to illustrate the methodology. Severe risk of drought includes those districts with less than 600 mm/year of rainfall, where extended dry periods are typical. At moderate risk of drought are those districts that normally receive more than 600 mm/year of rainfall but can be adversely affected by regular climatic variations. At risk of flood are those areas that are low-lying and experience periodic flooding that can be detrimental to crop production. Self-sufficiency is a measure of the total food produced (cereals, tubers, etc.) from own production from the two agricultural seasons, in terms of months of consumption. Finally, structural vulnerability is determined by combining all of the different data layers listed above, i.e. climate, agricultural production, livestock, sources of income, access to markets, and coping strategies. The result is a relative scale ranging from none to food crisis.

TABLE 21
Structural vulnerability in Gaza Province

District

Severe risk of drought

Moderate risk of drought

Risk of flood

Self-sufficiency (months)

Structural vulnerability

Xai-Xai


X

X

10-12

None

Bilene Macia


X

X

10-12

None

Manjacaze


X

X

10-12

None

Chokwé

X


X

10-12

None

Chibuto

X


X

10-12

None

Guija

X



7-9

Slight

Massingir

X



7-9

Slight

Mabalane

X



5-6

Economic

Massangena

X



5-6

Economic

Chicualacuala

X



5-6

Food insecure

Chigubo

X



3-4

Food insecure

Source: MSF-CIS (1998).

Gaza Province has the most districts of all provinces in Mozambique that suffer from structural vulnerability. This is especially the case in the four northernmost districts of the province - Massangena, Chicualacuala, Mabalane and Chigubo. These areas are particularly arid, with an increased risk of drought, very poor soils and low agricultural potential. The people living there typically produce only half of their annual food consumption needs (less than 6 months). With poor access to markets and limited alternative sources of income the people in these areas are chronically food insecure (Box 16).

For Gaza Province, changing eating habits (reduced meals) and intensifying the search for part-time work to generate income were identified as principal coping strategies. Other strategies employed (depending on the district) include: intensifying fishing and hunting; sale of charcoal and fuelwood; sale of livestock; and in the worst case, moving the family (MSF-CIS, 1998). Other livelihood surveys conducted in Gaza Province highlight the importance of remittances from South Africa to maintain household food security (Diriba, Getachew and Cooke, 1995; FEWS NET, 2001). These studies also discuss the importance of diversified farming systems, i.e. planting various types of crops in at least two fields (one in the more productive, but flood-prone lowlands, as well as one on higher ground) to spread the risk from drought or floods and enhance household income.

Another system that was developed was the Food Security and Nutrition Survey (FSNS) to monitor, collect, analyse and interpret information about the food security and nutritional status at subdistrict level (FAO, 1997). The FSNS characterizes the basic livelihood economies and the factors that influence production, sales, consumption, food needs, and health. In recognition of the variability within a district, the data collected for each district are further subdivided into three wealth classes: poor, medium, and rich. The FSNS subdivides Gaza Province into two zones: a productive coastal zone, and an arid zone.

In the productive coastal zone (Bilene-Macia, Xai-Xai, Manjacaze, Chokwé, Guijá and Chibuto), conditions are favourable for agricultural production with “relatively” fertile soils and climate. The principal staple crops are maize, manioc and rice, and cash crops include fruits, such as mafurra (castor beans), mango, oranges, cashew, tobacco, cotton and sugar cane. The local markets in these areas are fairly well developed and accessible. Poor families in this area manage to produce 50-60 percent of their basic needs, and the rest comes from the sale of cash crops, animals, local beer, working as a labourer in nearby fields, donations, and remittances from family members in South Africa.

The arid zone (Massangena, Chicualacuala, Chigubo, Mabalane and Massingir) has poor soils for agriculture and the rainfall is low and irregular. Most of the agriculture is along the rivers, and the main crops are maize, sorghum, millet and cassava. Livestock was an important activity in the area but suffered heavily during the war. The roads are not so developed and they are in poor condition, which makes transport and marketing difficult. Poor households manage to acquire food through purchases, remittances and donations. Other principal sources of household income are the sale of traditional beverage, charcoal, fuelwood and animals.

BOX 16

Livelihood study in Massangena District, Gaza Province, Mozambique

FAO conducted livelihood studies in several provinces in Mozambique in 1998. The objective was to obtain comprehensive information about traditional farming practices in areas subject to tsetse and trypanosomiasis. Farm and non-farm activities were analysed to identify the linkages that could affect (directly or indirectly) the impact of the disease. An additional objective was to examine the scope for poverty alleviation through livestock and crop development strategies.

The survey was conducted in August and September 1998 in three districts in south-central Mozambique. The total number of households surveyed was 2 231, with 548 in Massangena District, the northernmost district in Gaza Province. This district is mostly in the Limpopo River Basin and partly in the Save River Basin, adjacent to Gonarezhou National Park in Zimbabwe. There were 2 694 households within the district at the time of the survey. At 6.6 persons per household, this converts to 17 780 people.

Some salient points derived from this survey are:

The district is very remote and transport conditions are generally poor.

Average annual rainfall is very low (400-600 mm) and soils have very low agricultural potential.

Main crop is finger millet - grown by 93 percent of households - with an average yield about 400-500 kg/ha.

Maize, although technically unsuited, is grown by 80 percent of households, with crop failure common.

The average household grows 7 crops - more than 50 percent of households grow 7 or more crops, a risk reduction strategy to cope with erratic rainfall patterns.

Crops ranked in terms of area planted were: finger millet, cowpeas, grain sorghum, maize, beans, groundnuts, cassava, pumpkin, and sweet potatoes.

Only 10 percent of households sell millet regularly, and in small amounts to local farmers.

Overall, less than 5 percent of farmers use any type of fertilizer; cattle owners (8 percent), non-cattle owners (1.7 percent).

Overall, nearly 70 percent of households meet their household food needs regularly; only 8 percent were regularly dependent on food aid.

Overall, 35 percent of adults generated income from non-agricultural sources such as trading, off-farm employment, small business activities, handicrafts and brewing.

In Massangena District, the earnings from these sources in 1997 were about US$250 000.

39 percent earned income from handicrafts; 37 percent from brewing; 47 percent from remittances - although in small amounts and no more than twice a year.

Regarding livestock ownership:

Cattle are owned by only 9 percent of households.

32 percent claimed they had previously owned cattle but lost them during the war.

For those that own or hold cattle, the average herd size is 11 head.

69 percent did not own goats, and 10 percent of households owned about one-third of all goats.

Chickens were owned by 75 percent of households, and ownership was again skewed, with 10 percent of the population owning 59 percent of the chickens.

66 percent of households owned only chickens and no other forms of livestock.

Less than 1 percent owned pigs.

More than 90 percent of the cattle sold were sold to obtain cash for a specific purpose, e.g. to pay for clothes, school fees, purchase food, buy more cattle, purchase farm assets (ploughs), or pay medical expenses.

Cattle-owning households grow significantly more crops than do non-cattle owning households.

Source: FAO (1999b).

South Africa

One study of the livelihood conditions in North West Province examined the population structure according to four categories: rural dwellers, rural producers, self-sufficient “subsistence” farmers, and farmers (Data Research Africa, 1995). Similar studies have not been conducted for the entire basin area, but the categories and general relationships probably apply throughout. These subcategories are described below (with the relative percentage in each category in the North West Province study area in parenthesis).

GOSA-StatsSA also conducted a rural survey in 1997. The aim of the survey was to better understand the economies of the rural population to determine, especially: their reliance on subsistence agriculture, the impacts of high unemployment, low-income levels, and, poor infrastructure and service provision. Five of the study areas in the rural survey were in North West and Limpopo Provinces. Results from this study revealed:


Previous Page Top of Page Next Page