3. REVIEW RESULTS OVERALL

Status of the reports

CCA reports

Of all the 50 countries with a CCA report, three countries had performed a second country assessment. The Development Assistant Framework of the UN (UNDAF) was completed before 1 September 2002 in two-thirds of the countries (33).

PRSPs

Of the 25 PRSPs, only 11 had been completed. The remaining 14 were interim PRSPs. All PRSPs from Latin America and the Caribbean and half of the PRSPs from Africa had been completed, while this was true for only one report in West Asia and North Africa, and one in Asia and the Pacific. In the Europe and CIS region, all PRSPs were interim papers. In seven, predominantly European, countries, the UNDAF had not been completed by 1 September 2002.

The status of interim PRSPs differs from country to country, depending on the purpose. Some are draft documents, while others resemble project descriptions that need to be carried out to develop a PRSP.

Report preparation process

CCA reports

UN agencies have been involved in the execution of all CCA reports (see Table 4), with 36 reports specifically mentioning which organizations of the UN family have been involved, ranging from 4 to 17 UN institutions. Government involvement is mentioned in 39 country reports. In at least 16 of these reports, the government has been actively involved in the entire process of report preparation, while 11 of them do not mention in what way government has been involved. The involvement of government institutions is specified in 14 reports, with the number of institutions ranging from 2 to 17. Two-thirds of the reports refer to civil society involvement. The extent of civil society involvement is barely reported, but, in seven cases, NGOs were mentioned as having participated actively in the entire process. Over one-third of the cases report some involvement of (bilateral) aid agencies and 14 cases mention the involvement of the private sector. The involvement of research institutes and universities is mentioned in 6 reports and 8 PRSPs. Some reports state the involvement of other organizations and individuals, such as consultants and some specific national institutions, like the supreme council, transitional organizations and churches.

The involvement of FAO is mentioned in 68 percent of all reports. In most of the remaining reports the involvement of FAO is unclear, as the cooperating UN agencies are not specified. When a report indicated that FAO was actively involved, its degree of involvement in the whole process was usually high. FAO was actively involved in 19 out of 34 CCA reports.

PRSPs

UN organizations were involved in almost half the PRSPs (see Table 4). In a quarter of the papers, the UN participated actively in the entire process, and in the other cases it participated either in the problem or data analysis, or by providing comments. Approximately half the reports mention the involvement of bilateral aid agencies and the private sector. University and research institute involvement is low. Civil society participated in the preparation of three-quarters of the reports, to varying degrees. In 16 percent of the cases, the involvement of other institutions is mentioned, mainly other types of donor. The involvement of FAO was mentioned in only one case (Nicaragua), where comments had been provided.

Table 4. Stated involvement of institutions in preparation of country reports (percentage)

Despite 23 of the countries included in this review having PRSPs that had been prepared after or in the same year as CCA reports, only two PRSPs refer to the respective CCA report. The PRSP of the Democratic Republic of the Congo presents some data on poverty indicators drawn from the CCA published the previous year. The report from Niger states in its introduction that the preparation of its Poverty Reduction Strategy coincided with the Common Country Assessment, which made it possible to have a fruitful exchange of views with the United Nations Development Programme (UNDP). According to the PRSP, UNDP also provided the thematic groups with very useful documentation.

Food security in the country reports

Most CCAs and completed PRSPs deal with food security in one way or another, whereas only half of the interim PRSPs do so (see Table 5). The remaining reports ignore food security. Moreover, Table 5 shows that most CCA reports assess food security in a separate section. In one CCA report, food security was not discussed in a separate section, but the issue was found scattered in the report. The PRSPs show a greater tendency to discuss food security, although in a fragmented way.

Table 5. Assessment of food security in CCA reports and PRSPs

Table 6 shows the (sometimes shortened) titles of the sections in which food security is discussed. The amount of text that is dedicated to food security varies from a small paragraph to a complete chapter. While 16 CCA reports include a specific section on food security, other CCA reports have food security as an element of nutrition/health, agriculture or another heading.

Table 6. Titles of sections in which food security is discussed

Notes:

¹ The right to health and survival; the right to adequate food; world food summit; food security indicators; the state of human development; food supply; maternal and child health.

² Expanding social protection programmes.

³ Sustainable and equitable economic growth; vulnerability; the agriculture and rural development sector.

Definitions used for food security and poverty

Definitions used for food security

According to the FAO definition, food security exists when all people, at all times, have physical and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life (FAO, 1996).

All country reports were analysed with regard to the explicit definition of food security. Table 7 shows that 28 percent of the CCA reports and only 8 percent of the PRSPs provide a definition of food security, with 10 percent of CCA reports using the complete FAO definition of food security, while the definitions used in the other reports are less comprehensive. If a definition is given, most CCA reports use the concepts of ‘all people’ and ‘all times’, and to a lesser extent ‘access (both physical and economic)’ and ‘sufficient food’. Concepts like ‘nutritious food’, ‘meeting dietary needs’ and ‘meeting food preferences’ are virtually absent.

Table 7. Food Security definitions and related terms used in country reports (percentage)

The reports were also screened for the use of the following dimensions in describing food security: availability, access, utilization, stability and sustainability. Availability in our context refers to the physical presence of food for consumption. Access refers to the financial ability to acquire food. Utilization refers to the adequate dietary composition of the available and acquired food, to the proper use of this food and to the effective digestion and absorption of what is consumed. Stability refers to the year-round ability of households to meet their dietary needs, even in situations of stress like seasonal food shortages. Sustainability refers to the capacity of households to ensure the long-term stability of the household food supply.

In the reports, the dimensions “availability” and “access” are mentioned in a few cases, while other dimensions are missing. Hence, food security is mainly seen as a problem of food access and food availability. The concepts of utilization, stability and sustainability are hardly addressed.

Definitions used for poverty

Poverty is defined in a quarter of the CCA reports (24 percent) and in less than half of the PRSPs (44 percent) (see Table 8).

Table 8. Defining poverty in the country reports (percentage)

This apparent lack of a definition of poverty is compensated by an almost universal use of one or more poverty lines, income or consumption levels that demarcate a certain degree of poverty. In most of the reports, the poverty lines were defined in national currency to suit the country-specific situation. The use of poverty lines in terms of US$ 1 or 2 per person per day is meant for broad international comparison and has less domestic relevance. Whenever described, poverty lines are phrased in terms related to food consumption, such as calorie intake, food, the breadbasket or the proportion of income spent on food.

Combined use of food security and poverty definitions

A small minority of the CCA and PRSPs provide definitions for both food security and poverty. The majority, however, do not give definitions for either term (see Table 9).

Table 9. Defining food security and poverty in country reports (percentage)

Development priorities

Development priorities are stated in all reports, but generally not in a separate section where the full set of priorities is discussed. Food insecurity and vulnerability is stated as a priority in 62 percent of the CCA reports and 16 percent in the PRSPs (see Table 10). In more than half of these CCA reports and PRSPs, food insecurity and vulnerability is mentioned in the context of a broader development priority such as poverty reduction, equity or human development.

Table 10. Development priorities in country reports (percentage)

Note:

¹ Most frequently stated priorities only.

A comparison of food insecurity and vulnerability with other development priorities shows that food insecurity and vulnerability ranks among the most frequently mentioned priorities in CCA reports. In PRSPs, the focus is generally on priorities phrased in economic terms, with poverty reduction as the most widely stated development priority.

The data: collection and presentation

Collection of data on food insecurity

Primary versus secondary data on food insecurity

The review reveals that data are predominantly secondary data, i.e. collected and recorded for purposes other than the CCA reports or PRSPs themselves. Only the CCA reports of the Gambia, Liberia and Namibia use specially collected primary data to supplement secondary data in the CCA reports.

Specification of sources of data on food insecurity

Table 11 shows that report authors are often negligent in specifying the sources of their data on food insecurity. This is especially true of PRSPs. A reason for limited data specifications could be the use of data that has not been published and therefore cannot be cited.

Table 11. Specification of data sources in country reports (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Sources of food insecurity data

The government is the most important source of data in the country papers, followed by UN organizations. FAO is mentioned explicitly in only a few documents. The actual proportions in Table 12 might be higher as data sources are often not specified in the country papers. In 38 percent of the CCA reports and 20 percent of the PRSPs, data are used that have been collected as a result of collaboration between at least two organizations. These efforts mostly concern secondary data.

Table 12. Organizations stated as providers of data on food insecurity (percentage)¹

Note:

¹ 14 percent of the CCA reports and 56 percent of the PRSPs do not either contain or specify any data on food insecurity.

Data collection methods

Information describing the data collection methods used is not widely stated in the country reports. In part of the reports, surveys are mentioned. Table 13 indicates the percentage of reports in which a given survey is stated as the source for at least one indicator.

Table 13. Surveys stated in country reports for obtaining data on food insecurity (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

The country reports have also been screened for the use of rapid, qualitative techniques to gather information. This screening refers to primary data as well as to secondary data in the report, and not specifically to data on food insecurity (see Table 14).

Table 14. Participatory and rapid assessment methods stated (percentage)

Note:

¹ Among which: Rapid Assessments of street children; of economic and social conditions; of agriculture; of drug use.

The figures presented in Table 14 might be an underestimation because the use of data derived from these methods could only be detected if the methods were explicitly mentioned, or if the reference list of the country papers indicated reports with titles that alluded to participatory or rapid assessment methods.

To assess the representativeness of national data used in the country studies, sampling strategies were considered (see Table 15). Approximately half of the country reports state the sampling strategy used, which refers to part of the data only.

Table 15. Sampling strategies stated in country reports for obtaining data on food insecurity (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Summarizing the information stated on data collection methods it can be concluded that the country studies do not provide comprehensive insight into the data that form the basis for the analyses.

Indicators used for food insecurity and poverty

Improving food insecurity and vulnerability is not feasible without exact definitions and reliable numbers. While the presence of definitions and collection of data was looked at in the previous sections, this section pays attention to whether the country reports clearly make an effort to estimate the extent of food insecurity and vulnerability, and of poverty.

Food insecurity and vulnerability

Both CCA reports and PRSPs were screened on the use of five types of indicators of food insecurity, all measuring different but complementary aspects of the food insecurity and vulnerability problem (for a detailed description of the indicators, see Appendix VII). These indicators estimate either dietary intake, food production, qualitative measurements of food security, nutritional status or mortality. All these are related to different components of the conceptual framework for understanding possible causes of low food consumption and poor nutritional status (FAO, 1998) (see Appendix V).

Most of the CCA reports and PRSPs use one or more of these five indicators to describe the food insecurity and vulnerability problem in their country. Only one CCA report and three interim PRSPs ignore the indicators mentioned above (see Table 16).

The indicators most frequently used to describe food insecurity and vulnerability in both CCA reports and PRSPs are mortality rates (infant mortality rate (IMR), under-five mortality rate (U5MR) and maternal mortality rate (MMR)) and indicators of child malnutrition. Less than half of the country reports use dietary intake indicators like per capita dietary energy supply (DES), or food production indicators (mainly CCA reports). Qualitative measures of food insecurity are hardly used in any of the country reports.

There is no absolute measure or golden standard, but all the indicators mentioned assess different aspects of food insecurity and vulnerability. For example, as shown in the conceptual framework developed by FAO (1998), mortality rates and the percentage of malnourished children not only reflect the food insecurity problem, but are also very often related to health and health care. Therefore, to estimate the extent of the food insecurity and vulnerability problem, the use of a combination of the above indicators is advisable in order to capture the different aspects of food insecurity and vulnerability.

Indeed, 68 percent of the PRSPs and 92 percent of the CCA reports employ a combination of two or more types of indicators, but none of the PRSPs and only two CCA reports use all five types. For CCA reports, about one-third use four of the five indicators (except the qualitative indicators), and another 24 percent combine the use of mortality rates, malnourished children and dietary intake indicators. For PRSPs, 24 percent have a combination of mortality rates and child malnutrition indicators and 16 percent use mortality rates, child malnutrition and indicators of dietary intake.

In the country reports, an analysis was also made of the use of indicators that were relevant to the Millennium Development Goals (MDGs) and the CCA indicator framework (CIF) (see Table 16). The MDGs do not explicitly deal with food insecurity and (nutritional) vulnerability as such. However, the objective of MDG 1 is to eradicate extreme poverty and hunger. Three indicators are mentioned in respect of hunger: (i) prevalence of underweight children, (ii) proportion of household income spent on food for the poorest quintile of the population, and (iii) proportion of the population who consume less than the minimum level of daily dietary energy. Mortality rates are used as indicators for goals that explicitly aim at reducing mortality. The goal to reduce child mortality (MDG 4), for example, uses both infant mortality rates and under-five mortality rates. The goal to improve maternal health (MDG 5) suggests using maternal mortality rates as an indicator. The CIF does have “food security and nutrition” under a specific heading, and uses the same three indicators as identified for the MDGs. In the CIF, mortality indicators are clustered under “Health and Mortality” for child mortality, and “Reproductive Health” for maternal mortality.

Table 16. Indicators used to describe food insecurity and vulnerability (percentage)¹

Notes:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

² This column indicates whether the indicators used are part of the official set of indicators that monitor the Millennium Development Goals (MDG) and the CCA indicator framework (CIF).

³ Low weight for age.

⁴ Low weight for height, reflecting acute malnutrition.

⁵ Low height for age, reflecting chronic malnutrition.

With respect to mortality rates, the CCA reports, and to a lesser extent also the PRSPs, show consistency with the indicators to monitor the MDGs and CIF. The percentage of underweight children (MDG indicator) is only addressed in 50 percent of the CCA reports and 40 percent of the PRSPs, while most CCA reports (88 percent) and almost two-thirds of the PRSPs (64 percent) describe the prevalence of undernutrition (related to CIF). Most of the country reports do not show coherence with the indicator ‘Proportion of population below minimum level of dietary energy consumption’, as only 24 percent of the CCA reports and 8 percent of the PRSPs use this indicator.

Poverty indicators

The country reports use a range of poverty indicators. Table 17 gives those most frequently used.

Table 17. Indicators used to describe poverty in country reports (percentage)

Notes:

¹ This column indicates whether the indicators used are part of the official set of indicators that monitor the Millennium Development Goals (MDG) and the CCA indicator framework (CIF).

² Incidence x depth of poverty.

³ Poverty disaggregated by region, age, urban/rural, or socio-economic group.

Both CCA reports and PRSPs make use of the poverty indicators recommended in CIF and MDG, with the poverty headcount ratio related to the national poverty line being the one most frequently used. Other indicators, however, receive similar attention. The CCA reports and the PRSPs use approximately three and four poverty indicators respectively per study, with more or less comparable frequencies. The exception is the indicator referring to income spent on food by the poorest quintile, which is used in all PRSPs and in a few CCA reports. This is despite this indicator being mentioned under the Food Security and Nutrition indicators in the CIF. Disaggregation of poverty - important for the formulation of effective interventions - receives attention in most of the CCA reports and PRSPs.

Table 18. Presentation forms for data on food insecurity in country reports (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Presentation of data on food insecurity

Presentation formats for food insecurity data

Most information on food insecurity and vulnerability is provided as text. Three-quarters of the CCA reports, and little more than half the PRSPs provide data on food insecurity and vulnerability in tables. Graphs, charts and maps are used much less often.

Table 19 shows the proportional use of national and subnational data in text and tables. It can be concluded that all forms of presentation are mainly used to present national-level data.

Table 19. Presentation format used for national and subnational data on food insecurity in country reports (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Disaggregation of food insecurity data

The general picture that emerges is that most data on food insecurity and vulnerability are not disaggregated. When disaggregated, the data mostly address gender; rural vs urban population; or administrative subdivisions (see Table 20). Furthermore, disaggregation is applied only to some indicators. Most country reports do not explicitly describe the administrative structure of the country. If disaggregation occurs, it usually concerns regions, districts, provinces, etc. Disaggregation by cities or households occurs in three CCA reports. One PRSP addresses cities. The country reports do not describe methodologies related to disaggregation.

Table 20. Disaggregation of data on food insecurity in country reports (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Temporal dimension of food insecurity data

All the reports were published in the period 1997-2002. To assess the timeframe of the data, the reports were screened for the most recent and the oldest data on food insecurity. The implication of this screening method is that it does not supply us with information on data in between these limits. However, most data stem from the late 1990s.

Table 21 shows that 70 percent of the CCA reports and 64 percent of the PRSPs contain at least some data from 1998 onwards. In interpreting these data, the year of publication of the reports should be taken into consideration. Due to the more recent release of PRSPs, data for 2001 can be found. Only one PRSP had most-recent data older than 6 years: in this case it was 13 years old.

Table 21. Temporal dimension of data on food insecurity used in country reports (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Data from prior to 1990 appeared in 52 percent of the CCA reports and 20 percent of the PRSPs. Some of this older data could date back to the 1940s, although in such cases it was usually part of a time series.

Predictions on food insecurity and vulnerability are made in 14 percent of the CCA reports and in 28 percent of the PRSPs. The majority of the cases concern predictions for 1-5 years ahead, and some for 5-10 or even more than 10 years ahead.

Time series of food insecurity data

Most CCA reports and PRSPs provide at least one time series (see Table 22).

Table 22. Time series used in country reports for presenting data on food insecurity (percentage)¹

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Table 23 shows that the typical time step in which data on food insecurity and vulnerability are presented is an annual or a 2-5 year step. Also a “more than 5 year” time step is frequently used in CCA reports. The presentation of data for periods of under a year is uncommon. A monthly presentation of data is found only in the CCA report for Kosovo. A seasonal presentation of data occurs in only two of the 50 CCA reports (Burundi and Eritrea) and in one PRSP (the Gambia).

Table 23. Specified time steps in country reports for presenting data on food insecurity (percentage)1

Note:

¹ 2 percent of the CCA reports and 12 percent of the PRSPs contain no data on food insecurity.

Causes and explanations

About half of the CCA reports (56 percent) and 20 percent of the PRSPs present a causal model for food insecurity and vulnerability. In almost all cases, the causal model is not explained. Only one CCA report and two PRSPs explicitly describe the model.

Nevertheless, 80 percent of the CCA reports and 40 percent of the PRSPs explain the causes of food insecurity and vulnerability. Not one single report addresses all the components of the conceptual framework, as developed by FIVIMS (FAO, 1998), but all components do appear if the entire set of reports are taken together. In general, CCA reports provide a wider variety of explanations of food insecurity and vulnerability. Further, as might be expected, the completed PRSPs discuss more explanations of food insecurity and vulnerability than the interim PRSP reports.

Most explanations contained in the CCA reports (68 percent) and PRSPs (32 percent) were related to the socio-economic and political environment (see Table 24). Of these explanations, natural disasters and the agricultural sector were seen as a main cause of food insecurity and vulnerability (see Appendix VIII). Significantly, 26 percent of the CCA reports implicate the policy environment as a possible cause, although this aspect is never mentioned in the PRSPs. Furthermore, a number of CCA reports indicate that causes of food insecurity and vulnerability are related to the macro-economy and the national natural resource endowment. The PRSPs hardly ever mention these factors.

Table 24. Explanations regarding the causes of food insecurity and vulnerability in country reports (percentage)

The problem of availability of food is identified in 56 percent of the CCA reports, particularly in relation to food production. These aspects are hardly mentioned in the PRSPs. Stability of food supplies and access, mainly related to income fluctuations, are discussed in 52 percent and 24 percent of the CCA reports and PRSPs, respectively.

Explanations related to access to food and to food consumption are indicated in CCA reports, but hardly ever in PRSPs. In some reports, food insecurity and vulnerability are explained in terms of care practices, health and sanitation, and food utilization.

Food insecurity and vulnerability do not exist in isolation from other problems, so reports were analysed for linkages between food insecurity and other problems, including poverty, human rights, gender, HIV/Aids, women education, good governance and environment. Most of the CCA reports (68 percent) and PRSPs (64 percent) relate food insecurity to poverty. As far as the linkages between food security and other factors are concerned, CCA reports mention more linkages than the PRSPs, and 33 percent of CCA reports relate food insecurity to gender issues and 25 percent point to the relationship between environmental problems and food insecurity. Links between food security and HIV/Aids, good governance, and female education are only weakly suggested and there is virtually no link made with the issue of human rights.

Most of the explanations offered and links made are not supported by an analysis of past trends or projected data. Moreover, the explanations given are mostly supported by qualitative descriptions and not quantitative data.

Policies and strategies

In most of the country reports, the reduction or eradication of food insecurity and vulnerability, and poverty are stated as development priorities. The determination of priorities, the formulation of policies and strategies, and the development and implementation of interventions, can be seen as three steps in the policy formulation cycle. In this section, the elaboration of these priorities into policies and strategies, as reflected in the reports, is analysed. The interventions will be reviewed subsequently.

Table 25. Linkage between food insecurity and vulnerability and other problems (percentage)

Policies form an elaboration of priorities set, and most of the country reports with stated priorities regarding food insecurity and vulnerability, and poverty reduction state more or less elaborated policies in these two fields (see Section “From priority to intervention”, page 30, for an analysis of the policy formulation cycle in country reports). Food insecurity and vulnerability policies are stated in about half of the country reports, and most of these reports also contain poverty eradication policies (see Table 26).

Table 26. Policy formulation regarding food insecurity and vulnerability, and poverty eradication (percentage)

Note:

¹ Two PRSPs (out of 25) were without general development policies, but were interim reports.

Poverty eradication policies are stated in 73 percent of the country reports, sometimes alongside food insecurity and vulnerability policies. Thus almost half of the country reports contain policies for both food insecurity and vulnerability, and poverty eradication, although both policies focus on largely the same segment of the population.

The common ground shared by food insecurity and vulnerability policies and poverty eradication policies becomes clearer in the review of the strategies formulated. The strategies stated in the reports that are relevant in the context of policies aimed at reducing, or even eradicating, food insecurity are grouped into ten categories under three headings. Categories and headings relate to the linkages between household food security, nutrition and poverty, as stated in the conceptual framework for understanding possible causes of low food consumption and poor nutritional status (FAO, 1998). The first heading, socio-economic and political environment, focuses on the national and subnational levels that underlie the complex of reasons that lead to (low) food consumption. The second heading deals with the components of food consumption. The third heading focuses on health and care as this affects food utilization by the human body. Table 27 gives the strategies in the CCA reports and PRSPs.

Table 27. Stated strategies related to food insecurity and poverty eradication (percentage)

In 86 percent of the CCA reports and in 96 percent of the PRSPs, one or more of these strategies are proposed. The country reports generally state a combination of strategies, and in 45 percent of the country reports a combination of strategies was stated, covering the three broad fields mentioned in the table: socio-economic and political environment, food consumption, and health and care. The CCA reports and the PRSPs show a remarkable concurrence in strategies, especially when one considers the unavoidably arbitrary element in the classification of strategies applied.

Interventions

The interventions stated in the reports are grouped into the same categories as the strategies in the previous section (see Table 28).

Table 28. Interventions related to food insecurity and poverty eradication in country reports (percentage)

Interventions related to food insecurity or poverty eradication are described in 39 percent of all country reports, but the PRSPs state interventions more often than CCA reports. Most of the country studies that mention interventions state a combination of interventions related to food insecurity and poverty eradication. Interventions related to food consumption are stated in 24 percent of the country reports, and generally cover several interventions that fall into four categories. Interventions related to the socio-economic and political environment are more frequently stated (33 percent), whereas interventions regarding health and care are stated less often (23 percent) than interventions related to food consumption.