Reports from Union of African Population Studies / L'Union pour l'Etude de la Population Africaine Num. 31, 1998

Union for African Population Studies, Rapport de Synthese / Summary Report, Numéro/Number 31, 1998

POPULATION GROWTH, LAND USE AND FOOD SELF-SUFFICIENCY IN KENYA: A COMPARATIVE ANALYSIS OF SMALL AND MEDIUM-LARGE LAND HOLDINGS IN KISII AND NYAMIRA DISTRICTS

Mary Omosa

Research Fellow, Institute for Development Studies, University of Naïrobi / KENYA

Code Number: uaps98031

TABLE OF CONTENTS

List of Tables
List of Figures
Summary
1 - Introduction
2 - Theoretical framework

3 - Findings

4 - Conclusions and policy implications
Bibliography

List of Tables

Table 1: cropping patterns
Table 2: Maize output

List of Figures

Appendix 1: acreage under food crops relative to cash crops
Appendix 2: number of bags of maize harvested in the 1993 season
Appendix 3: number of months that the 1993 maize harvest lasted

Summary

Literature has conceptualized the relationship between population growth and food self-sufficiency at two different levels. One school of thought assumes that once population growth outstrips food production, an agricultural crisis will ensue. On the other hand, it is anticipated that population growth is likely to stimulate agricultural production and in particular, food production. This study hypothesized that population growth was a major driving force behind land use, and this process in turn influenced food self-sufficiency at the household level. Study findings however show that this relationship was mediated by several factors. Nevertheless, such patterns do impact on food output and, subsequently, the ability to be food self-sufficient. Households with a high dependency ratio tended to devote less land to food production. In cases where land resources were limited, cash crop farming competed with food crop production but, off-farm incomes enhanced food production. And, whenever land resources were plenty, acreage under cultivation was no longer a function of population parameters. This study, therefore, recommends a greater synchronization between agricultural policies and rural development. It is assumed that where physical infrastructure is in place, agricultural production and fertility behaviour will respond accordingly, both simultaneously and in isolation.

1 - INTRODUCTION

Food self-sufficiency can be achieved through increased production, the offering of better prices, availability of agricultural inputs and diversified research and extension services. However, the food situation in high density rural areas does indicate the need to look beyond increased production. It is therefore being suggested by this study that we look into the possibility of supplementing the above efforts by emphasizing enhanced production alongside planned infrastructure. This additional aspect has not featured prominently, as it should in official documents, as a solution to the food problem which culminated already in food shortage in Kenya in 1981 and 1984 (World Bank, 1986; ROK, Sessional Paper n° 4, 1981; ROK, Sessional Paper n° 2, 1994) and the current acute food shortage in many parts of the country. In this connection, the government's suggestion that the sub-division of land should be discouraged inspite of the growing population on land can no longer be taken for granted (ROK, Sessional Paper n° 1, 1986). Indeed, if the agricultural sector is to continue to lead the country toward sustainable development by providing among other things, enough food for an estimated population of 35 million by the year 2000, then household fertility behaviour should as a matter of urgency correspond to available resources. The first essential step should be the ability to balance food supply and demand, hence, the focus on population growth, land use and food-sufficiency in Kenya.

The basic issue to be addressed in this study is the effect of population growth on rural development, the latter refers to those activities and processes of productive change which lead to increased incomes and improved social wlefare. The key result of any rural development endeavour is a higher standard of living for the majority of the population (Johnston and clark, 1982; Chambers, 1983, Kiros, 1985; Alila, 1989). This development is measured in terms of the extent to which the social and economic needs of the population are met (Bryant and White, 1982; Eicher and Baker, 1982; Omosa, 1985). In the rural areas, there is an overall major concern with meeting the basic needs of its dwellers. One of the primary considerations in this regard is people's ability to maintain a balance between the supply and demand for food (Mujihahuzi, 1985; Barker, 1984; Lappe et al., 1980; Timmer et al., 1983). The levels and structure of rural population are therefore of major importance for development efforts aimed at meeting food requirements and other basic needs within the rural sector.

The present study focuses on the supply and demand for food in Kisii and Nyamira districts, as a basis for examining the nature of the relationship between population growth, land use and agricultural development. To assess the effect of population growth on rural development, we examine the current demographic profile, evaluate on-going land use practices as well as the stability of on-farm food supplies.

2 - THEORETICAL FRAMEWORK

2.1 - Population growth and food shortages

Although the world food problem has always existed, its magnitude has changed because the number of people in the world is increasing rapidly and this is occurring at a time when idle suitable land for cultivation is rapidly diminishing (Brown 1963). This makes heavy dependence on agricultural incomes a problem, especially if the rate of population growth continues to remain high (Okoth-Ogendo & Oucho 1993). This situation appears grim all the more because the land's capacity to produce is ebbing away while the majority of the population in Africa continue to depend on land for their livelihood (FAO 1985; World Bank 1986; MacNamara 1990). The subdivision of farms means that many of them are now too small to provide the minimum subsistence (Grigg 1980). Although Kenya seems to have done well in agriculture, the country's production has fallen well behind the rapidly increasing population. It is even predicted that over time, 60 per cent of Kenya's population will be too young to contribute to the national economy (FAO 1985).

Census statistics indicate that Kenya's population growth rate rose from 2.5 per cent in 1948 to 3.8 per cent in 1979. And, although recent studies have shown a decline in fertility for the first time in Kenya's demographic history, from a level of 7.9 children per woman in the late 1970s to 6.7 in the late 1980s (ROK KNCPD & IRD 1989), total fertility increased from 6.7 children per woman in 1948 to 7.9 by 1979. Indeed, in spite of observable population control policies and projected decreases in the growth rate, Kenya's current population, estimated at 25 million (a 42 per cent increase above the 1980 level) still remains high by world standards. This is accentuated by the fact that Kenya's population is still concentrated in the younger age brackets and will therefore continue to grow for many years to come. It is projected that between 1990 and 2020, Kenya's population will increase by 33 million, more than twice the number of people that were living in the country in 1980. In a country where less than 10 per cent of the arable land is high potential, population growth is bound to influence land use patterns. A basic premise of this study is that the effect of population growth will be most glaringly manifested in the agricultural sector.

The basic issue to be addressed therefore, is the effect of population growth on household level food self-sufficiency. One of the primary considerations is people's ability to maintain a balance between the supply and demand for food (Mujihahuzi, 1985; Barker, 1984; Lappe et al, 1980; Timmer et al, 1983). The size and the structure of rural population is of major importance in meeting food requirements and other basic needs within the rural sector. Whereas better health care and education have been accompanied by falling population growth, in Asia and Latin America, the reverse has been true in Africa (MacNamara, 1990).

2.2 - Land carrying capacity and farming practices

Improvements in the quality of life in many developing nations is viewed as a race between agricultural production and population growth (Spore, 1994). Modest gains in agricultural production are lost to increased population and any significant improvement is constrained by over-use. In the agricultural areas, population growth is a challenge to land tenure and land use practices. Pressure on land leads to fragmentation and more intensive use. This intensive use of land requires new techniques some of which are not always accessible to the small scale farmer. In Kenya, rapid population growth has resulted in the scarcity of the most natural resource, soil (and water) on which agriculture depends (Meyers 1982).

Land use as we see it today is therefore in many ways a combination of both its natural genesis and the human influences which have been brought to bear on it in the past and of those which are still active in the present. This refers to any kind of permanent or cyclic human intervention. Land use then becomes the application of human controls, in a relatively systematic manner, to the key elements within any ecosystem, in order to derive benefit from it (Lundgren 1975).

2.2.1 - Destruction through population growth

According to Malthus, over-population occurs when growth in numbers causes output per head to fall to the subsistence level until rising mortality causes this growth to cease. Responding to William Godwin's enquiry concerning Political Justice, where he argued that human misery, and in this case lack of sufficient agricultural production were due to the institutional imperfections of society, Malthus believed that no change in institutions could improve man's lot, because whichever way that society was organized, there would always be a tendency for population to outrun the growth in the means of subsistence, and:

Writers of the optimum theory have therefore argued that over-population will occur whenever population exceeds the optimum (Grigg 1980). The optimum theory assumes that while population changes, other factors that influence productivity are held constant thus, the amount of land does not change, capital available remains constant, technology does not advance, the working population remains a constant proportion of the total population, the distribution of income within the population remains unchanged, and for analytical purposes, output can be regarded as one product (Grigg 1980).

But, we know that population pressure is not the sole or indeed the most important cause of rural misery. Poor agricultural output is also the result of limited resources, following the concentration of land (and other natural assets), in the hands of a few. In Lesotho where soil and pasture productivity continued to fall due to intensified land use, the carrying capacity of the country's natural pastures reduced by half as early as the 1930s (UNEP 1982). Following apartheid policies, people were crowded in a small country and forced to settle in steep, easily erodible, inhospitable mountains. Agricultural fallow periods were eliminated, mono-culture cultivation increased. Similarly, in the West African Sahel where overgrazing is cited as a major cause of rangelands degradation (UNEP 1982), this can however be linked to the introduction of the monetary economy. In order to increase cash income and meet food needs, farmers were forced to expand production to marginal lands.

Furthermore, Malthus neglects the fact that as population grows, technology can change and maintain output per head. The optimum theory is therefore generally static because there are several other possibilities, in the face of population growth. Indeed, Africa is not the most populous region in the world, but it certainly has the largest number of poor people. Some countries with a much higher density have remained food secure while Africa has not. In fact, although a chinese farmer has on average half as much land as an Indian farmer, India has 200 million malnourished people (Bachuman & Paulino 1979).

2.2.2 - A prerequisite for development

A major proposition in recent theories on population growth is that population pressure is necessary for development to take place (Boserup 1965; Samir Amin, 1972). A high population growth rate increases resources by way of innovations and therefore promotes development. Samir Amin has argued that land carrying capacity is not an issue in Africa because there is still plenty of land. In fact, the current land pressure in Africa is attributed to unequal distribution (Alila et al., 1985). The United Nations (1973) threshold hypothesis further suggests that socio-economic development is likely to have an effect on fertility and not vice versa. It has also been argued that with relevant technology, coupled with training, credit, access to the necessary inputs, appropriate pricing policies and assured access to markets, there are a wide variety of farming systems by which one hectare of land or less could provide a family with sufficient food and also supply a cash income for purchasing the necessities of life (Smith 1976).

As such, the imbalance between food supply and demand in Kenya can be attributed to the tenure system. Bondestand and Berston (1981) have actually argued that pressure on land has little to do with population growth and more to do with the inequitable distribution of this resource. For example, data show that between 1965 and 1985, food production in Sub-Saharan Africa grew by 54 per cent but per capita food production declined by 12 per cent, a stagnation that has been attributed to technical and political reasons (WRI, 1988-89).

Boserup has argued, contrary to Malthus, that population growth is crucial in effecting economic growth, rather than vice versa. She therefore views population growth as a necessary step towards innovation and increased economic growth. In the agricultural sector, this innovation could be witnessed in the (dynamic) land use patterns.

2.2.3 - Impact on farming practices

In many farming societies, the amount of land available to the farmer is of great importance, particularly where the main factors of production are land and labour. While a number of processes determine the size of farm holding, family size is perhaps the single most important. Where population is increasing with no commensurate increase in the cultivated area, farms may get smaller. This is especially the case where partible inheritance is practised as opposed to primogeniture (whereby the eldest son inherits a whole piece of land without subdivision). In case of the latter, rapid population growth is not necessarily accompanied by the subdivision of farms.

Grigg (1980) has even argued that sub-division of farms, largely a culturally derived practice, does not necessarily lead to a fall in output. Productivity per hectare is higher on small than on large farms, as farmers are compelled to work their land more intensively or switch to high value crops to raise their incomes. Therefore, population density is not the greatest enemy of agricultural productivity. Indeed, agriculturally successful countries like Japan and Taiwan have twice as many agricultural workers per acre of land than the Philippines and India. Yet, the value of production per acre in Japan is seven times that of the Philippines and ten times that of India (Lappe & Collins 1977). But, it is also true that subdivision of land could reduce farm size to a level where it is not possible to provide an adequate subsistence.

When small farms are continuously subdivided, they eventually get too small to provide livelihood, and so a quasi-landless population grows, or where primogeniture is practised, the younger sons swell the population of the landless (Grigg 1980). But, this does not mean that we have run out of resources. Africa still has considerable agricultural potential. It is estimated that about 840 million hectares i.e. 30 per cent of Africa's land area is potentially cultivable but only less than one third of this has actually been cultivated (FAO 1989).

2.2.4 - Population growth and land use

The effects of population growth are certainly most evident in land use. Increasing population leads to the clearing of woodlands and waste land for extra arable land. Population growth may also affect the type of crops grown. In pre-industrial Europe, farmers grew more cereals and small vegetables. But, with the rise in population, there was a tendency to grow hitherto inferior crops such as potatoes and maize, until then grown only as livestock feed. This was because these crops had the advantage of higher yields (Grigg 1980). It is however also true that farmers are influenced by market demand. A study on the sweet potato potential in Kenya found that the root crop was neglected in spite of its being highly ecologically adaptive (Omosa 1994).

In discussing the interdependence between land use and technology, Boserup has attributed technological change to population dynamics. Pre-industrial peasant societies increased output through expansion of cultivated area, increasing the frequency of cropping initially through reduced fallow periods and later on, through multiple cropping, a shift to higher yielding crops, increasing labour inputs, technological advancement, division of labour and regional specialization, domestic industry and seasonal migration. This means that population is a challenge to growth but, one that is likely to lead to innovation and higher technological advancement (and hopefully, better standards of living). Boserup has however stated that relative to forest fallow, output per man-hour is likely to decline until a point is reached where higher output per man-hour could only be obtained by the introduction of animal-drawn ploughs or introduction of other techniques. Under such conditions, the introduction of the plough can be regarded as a means to prevent a fall in output per man-hour rather than as a means to raise it (Boserup 1965).

Intensification of agriculture then becomes a gradual change towards patterns of land use which make it possible to crop a given area of land more frequently than before, in order that productivity can be enhanced. She however also admits that where cultivation under long fallow has developed into the intensive type, it may be impossible to relieve population pressure by a change to short fallow. In such communities, a change to annual cropping or multi-cropping imposes itself (Boserup 1965).

But, in western Europe, change to annual cropping is described as a result of the discovery of the possibility of cropping the land without fallow by the use of crop rotations with fodder plants, of which some were leguminous. Rapid growth in population is then considered to be the result rather than the cause of this change. Other evidence however suggests that some of the intensive practices introduced in the fields during the agricultural revolution in Europe had been used in gardens, and were extended to fields, only after the rapid increase in numbers and the need for more output. This suggests that the transition in Europe from short fallow to annual cropping was not the result of contemporary inventions. Instead, it is the spread of various methods of intensive cultivation, most of which:

Nevertheless, Boserup ignored the fact that adoption of agricultural tools depends on the given context, much more than the system of land use in place and even then, some technical changes can only materialise if the community chooses to adopt them. In addition, these changes only come into effect if they are accompanied by the economic means to acquire them. She has however rightly emphasised that achieving the "threshold" in terms of population density does not necessarily transforms into a more efficient land use and food self-sufficiency. Instead, just at this critical point, some communities have lost their ability to produce and reproduce, a process that has turned one-time food self-sufficient households into food aid dependants.

2.3 - Hypotheses

Literature has brought out two main assumptions regarding direction and outcome, in the relationship between population growth and land use. One school of thought (Malthusian) argues that population growth will translate into an agricultural crisis, evidenced by land fragmentation, environmental deterioration, famine and general poverty. However, an alternative school of thought (Boserupian) suggests that population increase will stimulate agricultural growth through the intensification of agriculture, leading to improvements in food availability and general economic development.

This study argues that the outcome of the relationship between population growth and food production is not linear. It is dependent on the social structure, mainly the policies governing the allocation and distribution of resources and the level of technological applications. As such, the relationship between population growth and food production is conceptualized as one that is mediated by social, political and economic processes.

The study hypothesizes that:

1 - Rural land use practices vary with population dynamics;

2 - Rural household food production is a function of household composition;

3 - Food self-sufficiency is dependent on household fertility behaviour;

4 - Population growth has an effect on rural agricultural development;

5 - Natural resource management is dependent on population growth.

2.4 - Methodology

2.4.1 - The study area

This study was carried out in two high potential agricultural districts with contrasting land sizes. While Nyansiongo location in Borabu division of Nyamira District was relatively less densely populated, the opposite was true of Nyakoe location in Marani division of Kisii District. The two locations therefore gave us the desired strata with regard to land holdings.

Both Kisii and Nyamira districts are mainly inhabited by the Abagusii, a bantu-speaking people who occupy the fertile South Western slopes of the Mau Escarpment, overlooking Lake Victoria to the West. The area's vegetation is woody and bushed grassland with scattered or grouped trees. Most of the vegetation has however been replaced by crops and exotic trees (District Dev. Plan 1994-98).

The region lies on a highland equatorial climate. It receives rain almost throughout the year with two major rainy seasons. The high and reliable rainfall supports crops such as tea, coffee, pyrethrum, maize, beans, finger millet, potatoes and dairy farming. The most important economic resource and therefore the basis of economic development is the high potential agricultural land, comprising of good soils and supported by abundant rainfall, geographically well distributed throughout the year.

2.4.2 - Data collection

The household was the sampling unit. Proportionate random sampling was used to select a total of 118 respondents using the most recent land registration list. The selection was based on land holdings which were stratified into large-medium (Nyansiongo in Nyamira district) and small (Nyakoe in Kisii district) scale holdings. Attempts were made to reconcile this with on-ground observations so as to capture those who may have undergone sub-division in the recent past. This sampling procedure was adopted because land was a central variable. A questionnaire, made up of both closed and open-ended questions was administrated to the 118 randomly selected households. Questions covered background information, agricultural practices and family planning history.

Focus group discussions (FGD) were also conducted, mainly to explain, reinforce and/or enrich survey results. Two focus group discussions were conducted in each of the sample strata, i.e large-medium and small scale farm holdings. A total of 32 households were randomly picked and categorized into male only and female only to form two groups of eight persons per study stratum. Thereafter, each of the groups was met separately and was engaged in a free discussion centering on population growth, land use and food sufficiency. However, in order to answer the research questions, probing questions were used. These questions were drawn from the key survey themes, namely; farming practices, population dynamics, food-self-sufficiency, fertlity behaviour and the effect of population growth on agricultural development in the study areas.

A documentary search on population dynamics, land use and production was equally carried out. In addition, district level records on agricultural output were consulted.

For data analysis, averages, range, percentage and correlation coefficients have been extensively used for variables such as household size, occupation, sex ratio, land holding, agricultural output, food consumption, etc...

2.5 - Population status

In 1969, the larger Kisii district had a total fertility rate (TFR) of 7.3 children per woman, far exceeding the average national rate of 6.6 children per woman. The 1979 census revealed an even higher level of 8.7 births per woman. Maximum fertility is observed among females between 25 and 29 years with the cohorts aged 15-34 years accounting for 73.7 per cent of total fertility. Conclusively, the larger Kisii district is the most populous of the regions that make up Nyanza province. With a total land area of only 2,196 sq. kms, the area is the most populated after Mombasa and Nairobi, and it ranks top as the most densely populated rural area in Kenya (RoK Census 1989).

Topographical maps reveal an unevan population distribution due to the undulating nature of the landscape. Settlement is confined to the slopes of the plateau and river valley, while rocky mountain tops and extensive swampy areas are avoided (District Dev. Plan 1994-98).

Both Kisii and Nyamira districts report the lowest level of mortality in Nyamira Province. According to the 1979 census, the area had a child mortality rate of 101 deaths per 1.000 births, during the first two years of life. Further estimates show a general life expectancy of 54 years and a crude death rate of about 15 deaths per 1.000 of the total population.

From the survey data the following results were obtained. The household size for Nyansiongo ranged from 6 to 10 and that of Nyakoe from 4 to 10, giving an average household size of 7. The correlation coefficient between household size and the number of children aged fifteen years and below was found to be 0.7184 for Nyakoe and 0.1943 for Nyansiongo. Most rural residents are mainly farmers by occupation. A few have other sources of income.

Educational attainment was generally low. The type of housing suggests that the majority of people in the Kisii sample have limited economic means. About 57.6 per cent lived in iron roofed mud houses, 20.3 per cent in grass thatched houses and 22.1 per cent live in semi-permanent and permanent dwellings. In Nyamira, 63 per cent lived in permanent or semi-permanent dwellings and only 5 per cent have grass-thatched homes.

3 - Findings

This study set out to investigate the linkage between population growth, land use and food self-sufficiency. The rest of what follows is a discussion on the relationship between population growth, land holdings, cropping practices and food output. A comparison is made between small-scale and medium-large scale holdings in Kisii and Nyamira districts, respectively. The overall observation is that the relationship between population growth and food self-sufficiency is mediated by several factors, among which resource endowments.

3.1 - Land holdings

In agriculture based economies, land is a source of production and reproduction, and for this, its ownership is economically, socially, politically, and emotionally driven.

In general, land holdings ranged from a low of a quarter of an acre to over 50. But, some households in the medium-large scale area had as much as 200 acres of land. In addition, households augmented their holdings by hiring in land. This was more common among the small holders (59 per cent) as compared to households with medium-large scale holdings (16 per cent).

Hiring land is a recent phenomena. Among small holders, most of the households (31 per cent) hired land for the first time in 1993. Although the first person did so in 1970, this practice seemed to have become widespread only after the mid 1980s. Eighty three per cent of the households explained that they found it necessary to hire additional land because they required more food for their families. Only 17 per cent stated that this was prompted by the need to realize a surplus for the market. In the medium-large scale area, only two people had hired in land.

3.2 - Cropping patterns and practices

Most of the agricultural land was under cropped. In small holder areas, an average of 1.7 acres of land had been put under food crops and the majority (36 per cent) of the households had two acres of land under food crops. In large-medium scale farms, an average of 10 acres were under food crop production. One person did not grow any food crops in the 1993 period, while about 17 per cent of the people had put two acres of their land under food crops, and this rose to a maximum of 80 acres (see Figure 1). In both regions, most people had two to four acres of land under food crops and only households with medium-large scale holdings had more than five acres of land under food crops. Even for them, this was limited to ten acres (Table 1).

Cash crops were grown by about 70 per cent of the small scale holders. They had up to 3 acres of their land under cash crops, although this fell to a low of one tenth of an acre. On average, in the large-medium scale area, cash crops occupied a much smaller proportion of the land (~4 acres). In fact, 14 per cent of the holdings had no cash crops in spite of their relatively large land sizes.

Table 1: cropping patterns

Limitations in land use were even more obvious when we looked at land under pasture. For 36 per cent of the small scale holdings, no land was set aside for pasture or left fallow in the 1993 season. The rest had minimal amounts, ranging from 0.1 to 2 acres. Contrary to this, an average of 12.6 acres of land were under pasture or fallow in the large-medium scale holdings during the same season. Similarly, only less than 40 per cent of the small scale farms had some of their land under bush. Even then, for many of them, this was less than an acre. In the large-medium scale farms, a relatively less densely populated area, 85 per cent of the holdings had some land under bush. This ranged from 0.3 to 8 acres.

3.3 - Food self-sufficiency

3.3.1 Inputs

Both improved seeds and the application of fertilisers were widely used. Over 86 and 94 per cent of the households in the small scale farms used improved seeds and fertilizers respectively. In large-medium scale holdings, virtually all 98 per cent of the households used improved seeds and fertilizers for their food crops. However, labour input was low as most households experienced shortages (70 and 66 per cent in the small and medium-large holdings, respectively). Labour constraints were higher in the small scale farms because they relied totally on rudimentary tools. All households used the hoe (jembe) for ploughing. However, in the medium-large scale farms, some households (14 per cent) used oxen-ploughs or tractors.

3.3.2 - Household food output

Food security levels depend on availability of resources necessary to obtain the required amounts of food. Among the farming communities, it is expected that they grow their own food, although it is also hoped that they will concentrate on areas where they feel they enjoy a comparative and competitive advantage. For small holders, this advantage points to on-farm production and self-sufficiency.

Table 2: Maize output

Maize, the staple food crop was cultivated by all the households. In 1993, the small scale farmers realized an average harvest of 8 bags but the majority harvested only 4 bags. Only 20 per cent of them realized a harvest of 10-40 bags of maize. In large-medium scale holdings, an average of 84 bags of maize were harvested and only 16 per cent of the households had below 10 bags. Further, over 50 per cent of the households harvested more than 20 bags of maize and the highest 20 per cent got over 100 bags (Table 2; see also Figure 2).

3.3.3 - Sufficiency of Food Output

The 1993 food harvest lasted for an average of 8.7 months for small holder producers. Out of 59 respondents, only seven (12 per cent) had a harvest that lasted them for a full year. Eighteen households (31 per cent) exhausted their domestic supply by the first six months after harvest. In the large-medium scale holdings, their harvest lasted an average of 10.5 months and for the majority of such households (68 per cent), their harvest lasted an entire 12 months. Only less than 8 per cent of the households exhausted their maize within the first six months of harvest (see Figure 3).

Unlike in the large-medium scale holdings where association was insignificant, in the small scale holdings, the number of months that their harvest lasted depended on household size (.+7273). This suggests that when food output is limited, self-sufficiency is dependent on consumption, unlike in resource endowed areas where supply was not easily exhausted and therefore unlikely to be dependent on the number of regular consumers.

3.3.4 - Food Shortages

Most of the small holder producers (76 per cent) indicated that there were months in a year during which they were unable to have the required number of meals. Most (30 per cent) people cited the month of May as the most critical month of the year. May is the weeding/pre-harvest season when many granaries are already empty. As would be expected, large-medium scale farmers did not face food shortages. Although they all indicated that they ate only two meals a day, none of them cited any month in a year when they were unable to get the required number of meals or when consumption was poor.

Food shortages among small scale farmers were mainly attributed to sales (29 per cent) and land shortage (28 per cent). Only 17 per cent of the households attributed this to climatic factors that result in poor harvest. The rest felt that food shortages were as a result of limited capital, poor farm management and/or crop husbandry. In spite of their ability to explain food shortages, 55 per cent of the small holders had not taken any step to remedy the situation. Thirty one per cent indicated that they had increased inputs. The rest had re-allocated cropping by increasing acreage under maize (food crops), they had hired more land or they instituted better crop husbandry through frequent weeding and proper timing.

Only 20 per cent of the large-medium scale holders attributed food shortages to land scarcity. The rest explained food shortages in terms of climatic limitations, laziness, sales, lack of capital, inadequate labour and poor farm management. Unlike small holders, only 15 per cent of the large-medium scale farmers that anticipated food shortages had not made any effort to boast their food production. The rest of them had re-allocated cropping, hired more land, increased use of inputs, and undertaken proper crop husbandry.

More than 63 per cent of the small holder farmers depended on the market for additional food (maize). In 1993, an average of 3.5 bags of maize were purchased. Considering that households indicated an average consumption of five bags of maize, it can be construed that over 50 per cent of their supply had to be met outside domestic production. But, not one household in the large-medium scale holdings depended on the market for their food supply.

4 - Conclusions and policy implications

This study hypothesized that population growth was a major driving force behind land use, and this process in turn influenced food self-sufficiency at the household level. Study findings show that this relationship is nevertheless mediated by several factors, including the social structure of a given population, the specific land use patterns and the technological applications in place. These relationships were however not linear hence, intensified cultivation preceded population growth and vice versa, just like lack of self-sufficiency in food production occurred even in the absence of population growth.

Households with a high dependency ratio tended to devote less land to food production. Farmers in large-medium scale areas had five times more land under food crop production compared to small scale holders. However, in both small and large-medium scale holdings, settlement patterns seemed to be culturally determined. In both places, a similar size of land (~ 2 acres) had been set aside for the home compound even when total land sizes were themselves dissimilar.

In cases where land resources were limited, cash crop farming competed with food crop production but, off-farm incomes enhanced food production. However, whenever land resources were plenty, acreage under cultivation was not influenced by population size. Households that were too large but nevertheless able to hire additional land did not experience a high production shortfall. Indeed, only less than 10 per cent of the medium-large scale holders had leased out land in 1993 compared to 29 per cent among small scale farmers.

Whereas variation in household size does suggest that people in high density regions were already taking some action, it is still potentially worrying that those currently enjoying abundance are not "reading the signs". In the near future however, large-medium scale holdings are likely to continue enjoying relatively good harvests because a high dependency ratio after fifteen years of age is less likely. A sizeable number of their children were in some training/employment compared to small holders, among whom this proportion was highly negligible. Thus, even if land sub-division were to continue in the large-medium scale holdings, off-farm incomes might come to positively bear on production. This is unlikely to happen among small holders because a great majority of them were dependent only on tiny land parcels.

This study recommends a greater synchronization between agricultural policies and rural development. This entails provision of basic physical infrastructure upon which farm households can necessarily base their commitment to agriculture. In our understanding, once the physical infrastructure is in place, agricultural production and fertility behaviour will respond accordingly, both simultaneously and in isolation. Indeed, demographic considerations alone cannot explain the poor distribution of food resources. To believe that any arbitrary stabilisation of world population or even its reduction could directly solve the problem of hunger would nonetheless be an illusion: without young people's work, without the contribution of scientific research, without solidarity between people's and generations, agricultural and nutritional resources would probably become less and less reliable and the poorest categories would remain below the poverty line and excluded from economic circulation (Pope John Paul ll, address to the World Food Summit, Rome November 1996).

BIBLIOGRAPHY

• Adlakha, A.(1992) Population Trends, Kenya. Centre for International Research. U.S. Bureau of the Census. Washington D.C.
• Alila, P.O. (1985) Rural Landlessness Situation in Kenya, Report for FAO Expert consultation on Landlessness; Rome: FAO.
• Alila, P.O. (1989) "Changing Rural Development Perceptions and Social Science Research Contribution", in A.C.Akello-Ogutu (ed), Kenya's Rural Research Priorities, Occasional paper No. 53, Nairobi:IDS.
• Amin, S. (1972) Under-Populated Africa. Montreal, McGill University Centre for Developing Area Studies.
• Bachuman, K.L.; and Paulino, L.A., (1979) Rapid Food Production Growth in Selected Development Countries: A comparative Analysis of Underlying Trends, 1961-67. Research Report no 11, IFRI: Washington, D.C.
• Barker, J [ed] (1984) The Politics of Agriculture in Tropical Africa. Beverly Hills: SAGE Publications.
• Bongaarts J., Odile F., Lestheghae R.(1984) The Proximate Determinants of Fertility in Africa South of the Sahara. In Journal of Population and Development, vol. 10 No.3.
• Bonderstand, L.R. & Berston, S. (1981) Poverty and Population Control. Academic Press, New York.
• Boserup, E. (1965) The Conditions of Agricultural Growth. The Economics of Agrarian Change under Population Pressure. London: George Allen & Unwin Limited.
• Boserup, E. (1981) Population and Technology. Oxford: Basil Blackwell Publishers.
• Bryant C. and White L.G. (1982) Managing Development in the Third World. Boulder, Colorado: Westview Press.
• Brown, L.R. (1963) "Man, Land and Food. Looking Ahead at World Food Needs". Foreign Agricultural Economic Report n° .11.
• Caldwell, J.C. (1976) "Towards a Restatement of Demographic Transition Theory". In Population and Development Journal, vol. 2, n° s. 3-4.
• Chambers, R. (1983) Rural Development: Putting the Last First. Longman. New York:
• Collier, P. and Lal, D. (1980) Poverty and Growth in Kenya. WorldBank Staff Working Paper no. 389. Washington, D.C.: The World Bank.
• Davis, A.K. and Blake, J. (1956) Social "Structure: An Analytical Framework", in Journal of Economic Development and Cultural Change. Vol. 4, pp. 211-235.
• Eicher, C.K. and Baker D.C. [eds] (1982) "Research on Agricultural Development in Sub-Saharan Africa: A Critical survey". Michigan State University, International Development Paper 1.
• FAO (1985) 20 Years of Co-operation. 40th Anniversary of FAO. Rome.
• FAO (1989) Food and the Environment. World Food Day.Rome:FAO.
• FAO (1993) The Conservation and rehabilitation of African Lands. An International Scheme. Rome:FAO.
• Grigg, D. (1980) Population Growth and Agrarian Change. A Historical Perspective. Cambridge: Cambridge University Press.
• Haugerud, A. (1981) "Development and Household Economy in two eco-zones of Embu district". Working paper n° . 382. Nairobi: Institute for Development Studies, University of Nairobi.
• Hyden, G. Kates, R.W. & Turner ll, B (1993) "Beyond Intensification" in Turner ll, B.L.; Hyden, G. & Kates, R. Population Growth and Agricultural Change in Africa. University Press of Florida, Miami. pp. 401-439
• Johnston V.F and Clark W.C. (1982) Redesigning Rural Development: A Strategic Perspective. Baltimore: The John Hopkins University Press.
• Kenya, Republic of (1966), 1966-70 Development Plan, Nairobi: Government Printer.
• Kenya, Republic of, (1981), National Food Policy, Sessional Paper N° .4. Nairobi: Government Printer.
• Kenya, Republic of, (1984), 1984-88 Kisii District Development Plan. Nairobi: Government Printer.
• Kenya, Republic of, (1986), Kisii District Socio-cultural Profile. Ministry of Planning and National Development and Institute of African Studies, University of Nairobi. Nairobi: Lengo Press.
• Kenya, Republic of, (1989a), 1989-93 Development Plan. Nairobi: Government Printer.
• Kenya, Republic of, (1989b), Demographic and Health Survey(KDHS) 1989. Kenya National Council for Population and Institute for Development. Nairobi: Government Printer.
• Kenya, Republic of, (1994a), 1994-98 Development Plan. Nairobi: Government Printer.
• Kenya, Republic of, (1994b) 1994-98 Kisii District Development Plan. Nairobi: Government Printer.
• Kenya, Republic of, (1994c) 1994-98 Nyamira District Development Plan. Nairobi: Government Printer.
• Kenya, Republic of, (1994d), Kenya Population Census 1989. Nairobi: Government Printer.
• Kiros, F.G [ed] (1985) Challenging Rural Poverty: Experiences in Institution Building and Population Participation for Rural Development in Eastern Africa. Trenton and N.J: Africa World Press.
• Kocher, J.E. (1974) "A Description of the Study of Rural Development and Demographic Change in Four Densely Settled Areas of Moshi and Lusotho Districts". Working Paper number 2. Demographic Unit, Bureau of resource Assessment and Land Use Planning, University of Dar es Salaam.
• Lanting, H. (1977) Land Utilization Types in Kisii District. A Preliminary Study. preliminary Report no. 15. Training Project in Pedology, Kisii. Wageningen: Agricultural University.
• Lappe, F.M., Collins J. and Hinley, D. (1980) Aid as Obstacle. San Francisco: Institute for Food and Development Policy.
• Lundgren, B.[ed] (1975) Land Use in Kenya and Tanzania. The Physical Background and Present Situation and an Analysis of the Needs for Its Rational Planning. Royal College of Forestry. Stockholm: International Rural Development Division.
• Malthus, T.R. (1872) An Essay on the Principle of Population. Reprints of Economic Classics. New York: Augustus M. kelley Publishers. Seventh Edition Reprints 1971.
• MacNamara, R.S. (1990) Africa's Development Crisis: Agricultural Stagnation, population Explosion, and Environmental Degradation. Address to the Africa Leadership Forum, Ota Nigeria. June 21.
• Meyers, R.L. (1982) A Sociological Approach to Farming Systems in Kenya. Cornell International Agriculture Mimeograph 94. Bulletin no. 124. Ithaca: Cornell university.
• Mujihahuzi, M.R. (1985) "The Impact of Population Growth on Food Production in Tanzania: Problems and Prospect", in F. Kiros (ed.) op. cit.
• Okoth-Ogendo, H.W.O. & Oucho, J.O. (1993) "Population Growth and Agricultural Change in Kisii District, Kenya: A Sustained Symbiosis?" in Turner ll, B.L.; Hyden, G. & Kates, R. Population Growth and Agricultural Change in Africa. University Press of Florida, Miami. pp. 187-205
• Ominde, S.H. (1975) The Population of Kenya, Tanzania and Uganda. Heinnmann, East Africa.
• Omosa, M. (1992) "The Food Situation in Africa: A Historical Analysis of the Food Situation in Kenya". in Resources Review Journal vol.4 no.2, Washington, D.C.
• Omosa, M. (1994) Current and Potential demand for Fresh and Processed Sweet potato Products in Nairobi and Kisumu, kenya. A Research Report submitted to the International Potato Centre (CIP), Nairobi.
• Omosa, M. (1996) "Rural Household Food Security. A focus on Small Scale Farmers in Kisii District, Western Kenya". Institute for Development Studies WP NO. 510, Nairobi.
• Schultz, T.W. (1964) Transforming Traditional Agriculture. Chicago: The University of Chicago Press.
• Smith, T.L. (1976) The Race between Population and Food Supply in Latin America. University of New Mexico Press, Albuquerque.
• Spore, (1994) Sustaining Soil Productivity in intensive African Agriculture. Bi-Monthly bulletin of the Technical Centre for Agricultural and Rural Cooperation, CTA no. 49, February.
• Steel, R.W. (1965) "Population Increase and Food Production in Tropical Africa" in Africa Affairs, pp. 55-68.
• Third World Development (1984) "Industrial Development and Agriculture", in Third World Development Journal. Grosvenor Press.
• Timmer, C.P., Falcon, W.P. and Pearson, S.R. (1983) Food Policy Analysis, The John Hopkins University Press. Baltimore.
• UNEP, (1982) Desertification Control. Nairobi United Nations (1973), Determinants and Consequences of Population Trends. Vol. 1. New York.
• World Bank (1986), Population Growth and Policies in Sub-Saharan Africa. Washington D.C.
• World Resources Institute-WRI, (1988) World Resources 1988-89. International Institute for Environment and Development and UNEP. Basic Books. New York.
• Wortman, S. and Cummings, R.W. (1978) To Feed This World: The Challenge and the Strategy. The John Hopkins University Press. Baltimore.
  
  
  Copyright 1998 - Union for African Population Studies.