Chronic Vulnerability to
Food Insecurity in Kenya--2001
A WFP Pilot Study for Improving Vulnerability Analysis
Nicholas Haan, consultant
Graham Farmer, consultant
Robin Wheeler, Regional VAM Officer
Glossary
Page
Executive Summary4
Introduction4
There is a Food Security Problem in Kenya6
National Food Balance6
Nutritional Effects5
Future Trends8
Recent Food Aid Intervention9
How Do We Understand the Problem?10
Conceptual Framework10
Examining the Problem at the National Level12
Goals, Data, Methods12
Secondary Data Analysis Results14
Interpretation and Discussion15
HIV/AIDS Analysis with District Data18
Examining the Problem at the Community and
Livelihood Zone Level19
Goals, Methods, and Sampling19
Overview of PVP Results23
HIV/AIDS Analysis at Community Level26
Problem Analysis and Interventions28
Implications for WFP30
Conclusions and Further Recommendations31
Need to Link Community Food Insecurity
To Specific Contextual Issues31
Need for Shift to Problem Solving Approach32
Possibility for More Formal Collaboration33
Appendix AParticipants and Institutions Involved
Appendix BLiterature Review
Appendix CData Used for Secondary Data Analysis
Appendix DSDA Data Sources and Technical Notes
Appendix EPVP Methodology Overview
Appendix FPVP Research Instruments
Appendix GLivelihood Zone Summary Results from PVP
Appendix HPVP Methodology Feedback and Review
Appendix IBibliography
List of Illustrations
Table 1National Food Balance for Key Food Security Crops
Table 2Percentage of Under 5 Children who are Undernourished
Map1Wasting: % below 2 SD
Map 2Life Expectancy
Table 3Population vs. Selected Crop Production for 1989 – 2000
Graph 1Trends in Population and National Food Production
Map 3Recent EMOP Distributions
Diagram 1Conceptual Framework
Table 4Variables Used for Secondary Data Analysis
Map 4SDA results from z-score/compositing
Map 5SDA results from Principal Components Analysis
Map 6Results from 1998 FEWS Study
Map 7Comparison of PCA and FEWS results
Table 5Districts in Highest Vulnerability Category
Map 8Clusters of Similar Districts
Diagram 2Cluster Interpretation Matrix
Map 9Incidence of HIV/AIDS
Map 11Intersection of High HIV/AIDS and Chronic Food Insecurity
Map 12PVP Sample Districts and Villages
Map 13Livelihood Zones Created for Study Area
Table 6Livelihood Zone Descriptions
Map 14Relative Drought Risk by Livelihood Zone
Table 7Intervention Opportunities and Selected Structural Problems
Executive Summary
Currently in Kenya over three million people are receiving food aid in twenty-five districts covering over three-quarters of the land area. Food aid recipients are over 10% of the national population, and in the nine worst affected districts over 75% of the people are receiving nearly 100% rations. While the need for this massive intervention is a function of the drought these past three years, the situation is part of a long-term pattern of periodic--indeed increasing--reliance on food aid for the majority of people in Kenya’s more food insecure districts. In addition to the lost lives and nutritional-cum-developmental stunting during such periods of food stress, cultural and environmental foundations are shocked, undermining future food security. Emergency food aid, while critical for immediate survival, must be in conjunction with interventions that address the causes of chronic humanitarian crises, less it become part of the problem itself.
This report analyzes chronic food insecurity, a “bigger picture” companion to analyses of the emergency situation. Based on a literature review, secondary data (satellite imagery, socio-economic indicators, and others), and primary data (collected in 79 villages, 12 districts); the aim is to understand characteristics and causes of chronic vulnerability to food insecurity for a cross section of Kenya’s population, as well as identify intervention opportunities. The analysis makes initial links between community-level food insecurity and underlying causes and contextual issues that, if left in place, will continue to undermine both indigenous livelihoods and coping capacities, along with even the most well designed community development projects.
The report is divided into a core document and a number of appendices with more details concerning results, technical notes, and information about the study. Following specific results for targeting and strategizing interventions, the final recommendations include thought provoking ideas regarding WFP’s strategic role in situations of chronic food insecurity.
Given the history, scope, and complexity of the food security situation in Kenya, it is hoped that this report compliments numerous academic, government, UN, and NGO efforts as we collectively increase our depth of understanding and, more importantly, our resolve to act.
I.Introduction
Background
Organizations concerned with food security of chronically hungry peoples (i.e. repeatedly over a number of years) are interested in knowing where, when, and who they are as well as the immediate and underlying causes of this unnecessary phenomenon. With a mandate not just to feed the hungry poor, but also to do something about it, the United Nations World Food Programme (WFP), through its Vulnerability Analysis and Mapping Unit (VAM), is initiating an effort to improve the quality of vulnerability analysis and strengthen links with intervention design.
This report on a Chronic Vulnerability to Food Insecurity Analysis (CVA) for Kenya is part of the overall VAM quality improvement initiative called the Standard Analytical Framework (SAF). As a first round, the SAF has been concurrently piloted in Mali, Bolivia, and Kenya (see Appendix A for SAF overview). Because chronic food insecurity is a cross-sectoral problem, the results from the CVAs should be of interest not only for WFP (i.e. direct input into the Country Strategy Outline, Country Planning, and Activity Planning), but to many NGOs, government agencies, and international organizations as well.
As a pilot effort, this report on the Kenya CVA should be read with an eye towards: (1) the findings on food security for Kenya, (2) the tools and methodologies employed, and (3) the institutional implications within WFP and inter-agency.
For WFP, the goals of the SAF are both institutional and operational, as follows:
WFP Institutional Goals
- To improve links between VAM analysis of vulnerability to food insecurity and WFP programme development (e.g. CSO, CP, and Activity Planning)
- To bring together "best practices" of vulnerability analysis methodologies that combine use of secondary and primary data
- To improve and somewhat standardize vulnerability analysis methodologies in all WFP countries
WFP Operational Goals
- To characterize the "where, who, how, and when" of chronic vulnerability to food insecurity
- To understand causes of food insecurity from both community and national perspectives
- To provide recommendations for long- and short-term programme design (e.g. CSO, CP, and Activity Planning)
While this study is part of an overall effort to improve vulnerability analysis within WFP, the complexity of food security analysis mandates integration of a variety of techniques and methodologies as well as close collaboration with various stakeholders. As such, this study benefits from the ideas and involvement of many UN, government, NGO, and international organizations (see Appendix A for List of Participants and Organizations).
Stages of Analysis in Kenya CVA
The Kenya CVA began in mid-February 2001 and the final presentation in country will be June 21st, 2001. Following the same general stages as the other pilot countries, the Kenya CVA included: (1) a literature review, (2) secondary data analysis, (3) community-based analysis, and (4) final analysis and write-up. A SAF-wide comparison and synthesis with the other pilot countries occurred in July 2001 in Italy.
The literature review summarizes available published and gray literature concerning food security in Kenya. The main goals were to understand contextual issues at the national level, enable the study to build from previous research/assessment efforts, and direct identification of relevant indicators and data needs (see Appendix B for Literature Summary and Appendix I for the Bibliography).
The secondary data analysis (SDA) utilized district, in the absence of sufficient division level, spatial/statistical data (18 selected variables) to: (1) identify relative differences in vulnerability to food insecurity between districts, (2) characterize contributing factors to vulnerability at the district level, and (3) prioritize/stratify districts for subsequent community-based analysis. A variety of data sets and techniques were employed, allowing for verification of results and a mixture of interpretations.
The community-based analysis, called Participatory Vulnerability Profiles (PVP), involved participatory research in 79 villages stratified by livelihood zones in 12 districts selected based on the SDA results and subsequent key informant discussions. The goals of the PVP were to: (1) develop relatively homogenous livelihood zones in the study districts, (2) verify and further disaggregate results of the SDA, (3) characterize community vulnerabilities to food insecurity based on livelihood zones, (4) characterize and identify proportions of more vulnerable populations, (5) identify both community-level and macro, or structural causes of food insecurity, and (6) identify intervention opportunities with and without the use of food aid.
Organization of the Report
Due to the volume of details of both results and methodological techniques, the bulk of the information is organized in the appendices. This main report provides an overview the methods, results, and intervention implications. A reader more interested in knowing specifics of an analytical technique or results from a particular livelihood zone, for example, will find these in their appropriate appendices.
Reviewing the highlights, this main report tells a story along the lines of: There is a food security problem in Kenya…How do we understand the problem and its causes?…Examining the problem at national/district scales…Examining the problem at livelihood zone/community scales…Implications for WFP…Further Recommendations.
II.There is a Food Security Problem in Kenya
National Food Balance
Kenya is a net importer of food and is likely to remain so for some time (see Table 1). The biggest deficit in absolute terms has been in coarse grains, which are the leading staple in the country. Imports as a proportion of total production, however, has varied for the three types of grains. It has been 170% for wheat, 444% for rice and only 28% for coarse grains in the past five years. For the years 2000/2001 imports as a proportion of domestic production will be 333% for wheat, 606% for rice and 98.4% for coarse grains. The increased import requirements are no doubt the effect of the drought the country has been experiencing since 1998.
Table 1: National Food Balance for Key Food Security Crops (in ‘000 Tonnes)
Wheat / Rice / Coarse grainsa / TotalPrevious five years average production / 267 / 50 / 2,588 / 2,905
Previous five years average imports / 456 / 222 / 745 / 1,423
2000/01 Domestic Availability / 180 / 33 / 1,986 / 2,199
2000/01 Utilization / 780 / 233 / 3,350 / 4,363
2000/01 Import Requirement / 600 / 200 / 1,364 / 2,164
Estimated per cap. consumption (Kg/year) / 25 / 7 / 98 / 130
Source: FAO/ GIEWS – December 2000
a includes maize and other edible grains
This lack of self-sufficiency in national food production equates to a serious indicator of food insecurity given the importance of subsistence agriculture and pastoralism as a proportion of rural household livelihoods and the minimal purchasing power of such households. The differential affects of this shortfall on places and social groups are a subject of this study.
Nutritional Effects
A recent joint UNICEF/GOK Multiple Indicator Cluster Survey (December 2000) confirms the nutritional affects related to the shortfall in food production. “Almost 23 percent of children under the age of five in Kenya are underweight and 6.5% are classified a severely underweight. Thirty-sevenpercent of children are stunted or too short for their age and 6% are wasted or too thin for their height[1]” (CBS 2000). The distribution by province of child malnutrition is an indicator of such distributions for the population as a whole, as displayed in Table 2.
Table 2: Percentage of Under 5 Children who are Undernourished
Province / Weight forAge: -2SD / Height for
Age: -2SD / Weight for
Height: -2SD
Nairobi / 13.1 / 29.3 / 3.0
Central / 18.8 / 30.3 / 5.6
Coast / 25.6 / 38.5 / 7.8
Eastern / 29.3 / 44.3 / 7.9
North Eastern (Urban) / 13.0 / 34.4 / 4.8
Nyanza / 21.4 / 37.9 / 5.8
Rift Valley / 25.5 / 37.9 / 7.4
Western / 22.1 / 39.2 / 5.5
Source: UNICEF/GOK 2000
While recent district level nutritional data proved difficult to attain in Kenya, the following maps indicate the district distributions of the nutritional situation in Kenya. Note that nutritional values fluctuate constantly, but this data indicates both the degree of severity and the rough spatial distribution of the problem.
Future Trends
Food insecurity in Kenya is both an acute and chronic problem. Of the numerous structural issues that lead to food insecurity in Kenya, population growth is one that is certain to put increasing pressure on food availability. Table 3 indicates the trend in relationships between population growth and key food crop production levels from 1989 to 2000. While one can analyze the specific trends for particular crops, the main point is well illustrated in Graph 1—since 1989 (at least) there has been an overall neutral or downward trend in production of key crops (e.g. maize and beans) while at the same time population has steadily increased. Thus not only is the current food balance in Kenya of serious concern, but the trend has been long term and is set to worsen in the future. While HIV/AIDS is also set to take its toll on population growth rates, this epidemic primarily affects the productive age sets of the population and will tend to only exacerbate the food security problem (see sections IV and V for further discussion on HIV/AIDS).
Table 3: Population vs. Selected Crop Production for 1989 – 2000
Year / Population (Million) / Maize (T) / Wheat (T) / Beans (T) / Rice (MT) / Sorghum (T) / Irish Pot (MT) / Swt Pot (MT) / Cassava (MT)1989 / 21.40 / 2610000 / 34722 / 422235 / 56680 / 138036 / 815920 / 262280 / 796770
1990 / 23.90 / 2250000 / 249411 / 420034 / 41088 / 99952 / 779190 / 241884 / 683000
1991 / 24.20 / 2340000 / 264457 / 400107 / 50500 / 100982 / 987828 / 405444 / 984168
1992 / 24.70 / 2430000 / 297000 / 449828 / 44203 / 128950 / 632572 / 249863 / 565295
1993 / 25.70 / 1755000 / 212776 / 408681 / 47000 / 89578 / 1033514 / 315140 / 631300
1994 / 26.10 / 3060000 / 297000 / 279000 / 64480 / 108000 / 806000 / 285000 / 780000
1995 / 26.60 / 2698863 / 312644 / 398361 / 44000 / 79495 / 928744 / 315461 / 435244
1996 / 27.00 / 2160000 / 315000 / 207000 / 50620 / 36180 / 744000 / 200400 / 330000
1997 / 27.60 / 2214000 / 252000 / 398201 / 48620 / 35167 / 376612 / 299520 / 219648
1998 / 28.10 / 2461598 / 270810 / 269764 / 41829 / 90306 / 679738 / 397469 / 650500
1999* / 28.70 / 2250000 / 334808 / 360000 / 46787 / 81000 / 625018 / 322632 / 338520
2000* / 29.60 / 2333892 / 350122 / 287713 / 46402 / 50686 / 597128 / 327248 / 289597
* Projections
Source: CBS; Economic Survey; Oduor-Otieno and Okwiri, 2001
Recent Food Aid Intervention
While this study focuses on chronic, not current, food insecurity, further evidence of the problem in Kenya is the response that the current situation has warranted by the Government of Kenya and the international community. The tonnage and distribution of the recent and ongoing WFP-led emergency operation (EMOP) can be seen in Map 3 (this data is for August 2000, and has changed slightly since then). This massive response is a critical short-term intervention into what would otherwise be a catastrophic humanitarian crisis, but it is not the answer. If anything, the size of the EMOP is in direct proportion to the size and complexity of the food security problem, and is an indicator of the need to better understand the nature of the problem.
III.How Do We Understand the Problem?
Conceptual Framework
There are a number of different approaches to understanding vulnerability to food insecurity, and this is not the forum for a complete review and critique. At the core of these approaches is an attempt to understand peoples access to food (including social entitlements, physical access, etc.), the availability of food (supply versus demand), and people’s utilization of food (preparation, health, etc.). What variables are examined, what is measured, and how to do that is a source of healthy debate amongst practitioners and academics alike.
Many organizations (e.g. FIVIMS, the Food Economy Group, Save the Children, FEWSNET, CARE, TANGO, WFP, and others) and academics (e.g. Amartra Sen, Michael Watts, Pierce Blakie, Patrick Webb, Robert Chambers, Max Dilley, Thomas Downing, and others) have contributed to our overall understanding of food security analysis.
One dimension of food security analysis that seems under emphasized, however, is the fact that household food security is a function of inter-related processes that happen at different geographic scales: international, national, regional/district, community, and intra-household. Each of these scales is important for understanding food insecurity, and each of these scales lends themselves to different types of analysis and methodologies. A challenge for a study such as this, which is national in scope but ultimately interested in household food security, is to have a conceptual framework to guide the analysis that is comprehensive and with direct links to the methodologies employed.
The conceptual framework developed for this study[2] is a composit of a variety of perspectives with a twist here and there, and is illustrated in Diagram 1. This framework incorporates both national and community/livelihood zone scales of analysis. At the national scale, in order to characterize vulnerability to food insecurity one must understand the political, economic, social, and environmental contexts—these can or cannot provide an enabling environment for communities to realize their own food security. These are the structural issues referred to throughout this text that can enable or undermine community food security.
At the community and livelihood zone scale (see below for discussion on livelihood zones), the conceptual framework identifies five factors that directly affect access, availability, and utilization of food for households. These are inter-related and include: resource endowment, livelihood strategies, hazards, coping strategies, and social dynamics (including gender, community cooperation, culture, religion, etc.). The core story-line requiring explanation for a characterization of vulnerability roughly follows:
…a community of people live in a place with a certain resource endowment; they have developed locally adaptive livelihood strategies for that place; occasionally they experience hazards that pose shocks to their livelihood system; in the face of such shocks they have developed coping strategies; and the efficacy of interactions amongst community members is affected by a suite of social dynamics; furthermore, these community scale dynamics are embedded in contextual dynamics occurring at a broader scale…
For each of these factors there are a set of key questions that need to be answered so as to characterize vulnerability to food insecurity (see Appendix E). These key questions are the critical link between the conceptual framework and the field instruments.