- Email: [email protected]
Nutritional functional classification study of Panama
Nutritional functional classification study of Panama
David L. Franklin, Marielouise W. Harrell and Cutberto Parillon
Despite significant progress, widespread malnutrition persists in Panama. This article describes a study which found that over half the malnutrition in the country is clustered in functional groups economically connected to the agricultural sector and which indicates that food availability is a serious problem for 25% of the population. Food intake deficits were equally prevalent among the urban and rural populations. Dietary composition was not seen as the principal problem; rather malnutrition problems are linked to low wages and poor access to public services. A nutrition policy for Panama must be centred first on employment and income generation, and then on extending and maintaining public health and nutrition services, particularly for the extremely poor in the central provinces. Food policy per se would not seem a priority area for public intervention. Keywords: policy
Panama; Nutrition; Public
David L. Franklin and Marielouise W. Harrell are with the Sigma One Corporation PO Box 12425. Raleiah. NC. USA. Cutberto Parillon works in tie’Directorate of Nutrition, Ministry of Health, PO Box 2048, Zone 1, Panama, Republic of Panama. The authors would like to acknowledge the collaboration and comments of Victor ValVerde of INCAP in Guatemala, Elena Brineman and Ben Severn of the Agency for International Development (AID), and the financial support of the Office of Nutrition of AID and of the government of Panama.
0306-9192/85/010063-11$3.00
0
This article presents the results of a functional classification study Panama which has been undertaken with the following objectives: 0 0
of
to provide an analytical and quantitative basis for nutrition planning; to provide information which might be useful in improving existing nutrition programmes and projects, and for incorporating nutritional activities into other developmental initiatives.
The study is based on data developed by the 1980 National Survey as part of the ‘Multi-sectoral analysis of the food and nutrition situation in Panama’, and on available secondary data from Panama. The approach of the study is to apply the concepts of the ‘functional classification approach as developed by Joy and Payne,‘, JoY,~ and Payne,3 and as implemented by Valverde et al.” The specific steps in the functional classification approach are: Define distinct categories of population groups in poverty conditions according to their connection to the economic and productive processes of the society in which they operate. These groups are called functional groups because an important characteristic of their definition is the occupation of the head of household. Locate these groups geographically and by administrative/political subdivisions within a country, or regions in that country. Enumerate or otherwise estimate the size of each group in terms of number of persons affected of each group in each geographic or political/administrative subdivision. Describe the social and economic characteristics that may be causally related to specific nutritional problems for each functional group.
The nutrition
problem
in Panama
More than 25% of the households in Panama have important nutritional problems in that at least 30% of the country’s preschool-aged children are affected by the growth retardation effects of malnutrition. Among
1985 Butterworth
& Co (Publishers)
Ltd
63
Nutritional
functional
classification study of Panama
‘J.L. Joy, ‘Economic aspects of food and nutrition planning’, First Asian Conference on Nutrition, 1971; J.L. Joy and P.R. Payne, ‘La nutrition y la planificacion del desarrollo national’, Alimenfacion y M&icion (FAO), Vol 1, No 4, 1975. “J.L. Joy, ‘Food and nutrition planning’, Journal of Agricultural Economics, No 24, 1973. 3P.R. Payne, ‘Nutrition planning and food policy’, Food Policy, Vol 1, No 2, 1976. W. Valverde eta/, ‘Classification funcional de poblaciones desnutridas en la republica de El Salvador’, Archives Latin0 Americanos de Nutrition, 1978. 50.1.C. Bermudez, ‘Estado nutritional de la poblacion adulta en la republica de Panama’, INCAP, Guatemala, MS Thesis, Universidad San Carlos, 1980. ‘C. Parillon, M. Harrell, B. Frazao, D. Franklin, Vial de Valdes, ‘Alimentacion y nutrition en Panama: la situation actual’, Ministerio de Salud, Republica de Panama, 1982. 7D.L. Franklin and M. Harrell, ‘Nutrition evaluation methods: a household production approach’, Proceedings of the Latin American Econometric Society, Santiago, Chile, 1983.
the adult population, 22.8% of males and 24% of females are classified as malnourished.” These nutritional problems are, in part, associated with levels of food consumption which are below internationally recommended levels of calorie intake. A large number of females in Panama use the bulk of their total real resources to obtain diets which are mainly below international standards of adequacy for food energy. These apparent deficits in food energy are associated with other indicators of poverty which, jointly with poor diets, have been shown to be important determinants of the prevalence and persistence of malnutrition in Panama.” Among the indicators of poverty that have been shown to be statistically and causally associated with households with malnutrition in Panama are: social and cultural characteristics of the household; variables such as low calorie intake levels, mothers’ opportunity cost of time, low income, and occupational characteristics of the working age adults in the household; health and environmental variables such as poor housing quality, lack of access to potable water, poor sanitary conditions, high levels of childhood morbidity from measles and diarrhoea, and poor access to the health services.’
Food consumption
aspects of the nutritional problem
Data from the 1980 National Nutrition Survey indicate that for more than half the households in the sample, calorie intakes adjusted for age and sex composition are below the FAO reference value of 2700 calories per calorie equivalent person. As many as 25% of the households have daily calorie intakes one-third lower than the reference value. There do not appear to be appreciable differences between the rural and urban distributions of nutrients among households. Diet composition is not a principal problem. Diets below the calorie reference level are more prevalent than those below the protein recommendations, and protein consumption does not appear to be a major problem for most households. The composition of urban and rural diets is presented in Tables 1 and 2. Rice is the important staple in the Panamanian diet; on a national level it accounts for one-third of calories, one-quarter of protein, and one-sixth of food expenditures. The importance of rice is greater in the diets of rural than of urban households. The diets of the rural population are composed of rice, oils Table 1. Calorie, protein and expenditure households (n=418).
64
shares for principal
food commodities
- rural
Average % of daily calories
Average % of daily protein
Average % of daily food expenditures
Corn Cereals Breads Beans Rice Fish Milk Poultry/eggs Beef Pork Oils Vegetables/fruits Sugars Others
3.6 03 5.6 6.1 36.0 2.1 3.0 3.2 48 1.o 11.3 14.5 5.3 3.2
2.7 0.4 5.9 13.2 26 9 10.5 5.3 8.0 12.4 1.8 0.0 8.0 0.0 4.9
2.8 0.4 4.3 5.7 20.0 6.2 5.5 9.0 10.6 2.2 6.7 14.1 2.5 10.0
Totals
100.0
100.0
100.0
FOOD POLICY February
1985
Nutritional Table 2. Calorie, protein and expenditure households (n=192).
functional
classification
study
of Panama
shares for principal food commodities
- urban
Average % of daily calories
Average % of daily protein
Average % of daily food expenditures
Corn Cereals Breads Beans Rice Fish Milk Poultry/eggs Beef Pork Oils Vegetablesifrults Sugars Others
1.8 0.7 11.4 3.8 28.3 1.0 6.3 5.2 9.6 1 .o 13.0 7.3 6.2 4.4
1.3 0.7 11.9 7.7 17.9 4.7 10.1 12.2 22.4 1.6 0.0 3.6 0.0 5.9
1.o 0.5 6.9 3.4 9.6 3.2 9.4 11.7 21.0 1.8 5.8 10.1 2.0 13.6
Totals
100.0
100.0
100.0
fats, fruits and vegetables, including roots, tubers, plantains and bananas. The urban diets are relatively more intense in beef and poultry as complements to the rice-based diet. The Directorate for Nutrition of the Ministry of Health in Panama publishes periodically an estimate of the per person food costs necessary to obtain a nutritionally adequate diet from the typical food-use pattern in Panama. This cost is known as the cost of the ‘basic food basket’ and was estimated as US$1.06 per adult per day at the time of the 1980 survey. This value was used to stratify the data from the survey households according to income and a criterion suggested by the Economic Commission for Latin America (ECLA). This criterion establishes three income cut-off points for defining levels of poverty; incomes below the cost of a basic food basket are said to reflect extreme poverty; incomes greater than the cost of one basic basket but lower than twice the monetary value of two basic baskets are said to reflect poverty; incomes at levels between twice and three times the cost of the basic basket are considered low but adequate, and income levels about three times the cost of the basic basket are considered as not poor.’ Table 3 presents the stratification of household incomes according to this criterion and the associated average values for food costs, calorie and protein intakes, and number of calorie equivalent persons per household. Table 3 suggests that low incomes are associated with low levels of food expenditure and food consumption. It also highlights the great disparity between urban and rural incomes - 23% of the urban sample households are judged to be in extreme poverty, whereas 54% of the rural households are judged to be in extreme poverty according to the ECLA criterion. and
Functional classification approach
8Sergio Molina, ‘La pobreza: description y analisis de politicas para superarla’, Revkfade La CEPAL, Santiago, Chile, December 1982.
FOOD POLICY
February
1985
The purpose of a nutritional functional classification study is to provide diagnostic information to assist planners and service providers in locating and identifying population groups with important nutritional problems. The basic approach consists of identifying the geographic and politico-administrative location of population groups whose sociodemographic characteristics are causally or statistically related to the prevalence of important nutritional problems. The methods are a planning rather than an analytical tool. These methods permit an
Nutritional
functional
classification
study of‘ Panama Table 3. Averages of consumption variables and other selected household variables for urban and rural households in the food consumption subsample of the National Nutrition Survey.”
Averages for urban households (n= 166)
Notes: ‘The food consumption subsample consisted of a 10% sample in every segment in the sample; as such, it is representative at the district, province and national levels. bCEP = Calorie equivalent person; weights each household member by the ratio of the age-sex calorie recommendation relative to the male adult recommendation. CMarket surplus = ratio of value of agricultural sales to value of agricultural production for each household. dMeasures of poverty: extreme poverty = gross income < 1 basic basket: poverty = 1 basic basket < gross income < 2 basic baskets; adequate = 2 basic baskets < gross income i 3 basic baskets; non-poor = gross income L 3 basic baskets. $1.06 = daily cost of basic food basket from ‘Determination y cost0 de la canasta basica de alimentos para Panama’, Ministerio de Salud, Panama.
Poverty level&
n
%
Money value of daily food consumption (household)
Extreme poverty Poverty Adequate Non-poor
38 32 23 73
23 19 14 44
5.28 7.06 6.64 5.76
166
100
Totals
CEPb per household
Daily calories per CEP
Daily protein per CEP (grams)
5.3 5.2 4.4 3.8
2126 2671 2586 2490
66.2 60.9 83.6 83.4
Averages for rural households (n=38 1)
Poverty levelsd
n
Extreme poverty Poverty Adequate Non-poor
204 81 34 62
54 21 9 16
Totals
361
100
%
Money value of daily food consumption (household) 3.96 4.29 5.27 4.54
CEPb HH
Daily calories per CEP
Protein per CEP (grams)
MarketedC surplus
% off-farm income
5.4 4.7 3.9 3.5
2230 2403 2667 2696
61.8 69.4 88.5 89.1
0.26 0.27 0.19 0.31
36 66 72 75
per
identification of the characteristics of population groups which allows the targetting of specific remedial activities, and provides estimates of the number of persons in each group and setting to facilitate the estimation of costs and level of effort required to deliver services or other remedial actions. The technique differs significantly from conventional nutrition planning approaches” in that these latter have been based principally on sociodemographic and physiological or health characteristics of persons at nutritional risk. By contrast, the functional classification technique attempts to relate the prevalence of malnutrition in each identified functional group by characteristics of that functional group that are associated with their connection to the productive process in the economy. In this manner the interventions or remedial actions that can be instituted include those which can be promoted through policy mechanisms. Therefore, a functional classification study identifies, quantifies and localizes groups of families according to their life pattern, social, economic and cultural problems, and the level of resources that are available to them. They are clustered according to these characteristics under the assumption that they will respond in a similar manner to specific policies and programmes. The design of such interventions should, of course, be the result of rigorous economic and social cost-benefit analysis.
Definition of functional groups
‘J.M. Pines, ‘National nutrition planning lessons of experience’, Food PO/icy, Vol7, No 4, 1982; FAO, Food and Nutrition Planning, Nutrition Consultants Report Series (Rome), No 35.
66
The 1980 National Nutrition Survey was designed with the same sampling frame as the World Fertility Survey for Panama; as such it is representative of the population in each administrative district as a self-weighting sample. Occupational questions in that survey were used to identifv each head of household with the occupation codes used by the International Labour Office. The groups thus-formed were further aggregated into larger groups with similar socioeconomic characteristics.
FOOD POLICY February 1985
Nutritionalfunctional classificationstudy of Panama
For the agricultural sector, the data from the 1980 survey and analyses of variance techniques were used to form seven distinct functional groups from those households engaged in agricultural production activities. The analyses of variance procedure was used to determine the extent to which the type and number of crops grown by the household served as an indicator of the prevalence of malnutrition. The seven functional groups for agricultural households were defined as follows: Small
horticultural producers. This group consists of households which farm more than half a hectare of land and produce fruits and vegetables but do not grow any rice or maize. They are generally located in the very remote areas or close to urban centres. Those located in remote areas consume almost all their own produce; those close to urban areas sell fruits and some vegetables for urban consumption. Their median land area sown to crops is 2.5 hectares. Small cassava producers. This group of farmers specializes in the production of cassava and their farms have a median crop area of 1 hectare. Small diversified producers. This group of small farmers produces at least three crops, of which two are maize and rice. They generally produce for subsistence consumption and their median crop area is 5 hectares. Small maize producers. These producers are specialized in maize production on their own farm. Some farms may be as large as 20 hectares in size, although the median farm size is 5 hectares. This group was isolated as a distinct group primarily to assess the impact of the maize price support policy which the country has pursued since 1974. Small rice producers. This group specializes in rice production and includes farms of up to 20 hectares, although the median farm size is 2.5 hectares. This group was formed to assess the impact of the rice support policy. Small rice and maize producers. These producers are similar in all respects to the previous two, except that the farm produces both maize and rice on farms whose median size is 5 hectares. Farmers with employees. This group consists of all other households that are engaged in agricultural production in farms 20 hectares or larger. This group has also identified itself as using hired labour on a permanent basis.
The above groups exhaust the households that live on farms; in general they represent small landholders with the following distribution of farm sizes: 25% farm less than 0.5 hectares; 50% farm less than 2.0 hectares; 75% farm less than 10.0 hectares; 90% farm less than 30 hectares; and only the top 5% farm 50 hectares or more. Even the so-called ‘farmers with employees’ represent relatively small landholdings, with median size of 32.5 hectares. It is important to note that this sample represents households whose residence is on or near the land which they farm. Households dwelling on large farms would typically be the households of hired caretakers. Those households not classified by the agricultural production activities were classified by occupation of the head of household:
l
Rural workers.
principally
FOOD POLICY
February
1985
This group is composed of households which reside in rural areas and whose income is derived principally
67
Nutritional functional
classification
study of Panama
Table 4. Definition of malnutrition indicator based on weight-for-height and height-forage. Weightfor-height
Low Normal
High
Height-for-age Low
Normal
High
Acute Chronic Chronic
Acute Normal Normal
Acute Normal Normal
from wage work. This group includes a few dwellers in peripheral urban areas whose economic activity is agricultural wage work. This group probably includes a large number of households whose head migrates seasonally or even continuously seeking employment on large farms, banana or sugar estates or in the urban areas. Salaried urban workers. This group consists of households whose head is permanently employed in an urban job. It includes persons engaged in construction, services, and industry. 50% of this group is concentrated in commercial or industrial occupations. Self-employed. This group of households represents those who identified themselves as self-employed and who do not engage in agricultural production activities. The majority reside in rural areas and tend to be engaged in commercial activities such as marketing of agricultural products. Commercial and transportation employees. This group includes permanently employed clerks and drivers, as opposed to ‘occasional’ workers in this sector. Professionals, office and financial employees. Households in this group include all professionals (ie college-trained persons who did not identify themselves in any other group), and ‘white collar’ office and financial employees. Skilled workers. This group includes all skilled occupations including barbers, craftsmen, carpenters etc. Unskilled workers. This group represents households whose heads are unskilled construction workers, loaders working in the transport sector, and other ‘occasionally employed’ workers. Government employees. Households in this group have heads who identify themselves as working for the government.
Distribution of nutritional status across functional group
“The heiaht-far-aae measurement is derived by ;ompariig the child’s present height to a standard reference and then classifying it as low (below two standard deviations (SD) of the standard reference), normal (between -2SD below the reference standard and +2SD above the reference standard), and high (above +2SD above the standard reference). In like fashion, the weight-for-height indicator is classified as low, normal or high based on comparing the child’s given height. These standards are stricter than conventional criteria such as the Gomez standard.
For the purpose of this analysis, nutritional status is established at the household level by determining the nutritional status of the children aged nine years or younger in each household. If any child in the household is ‘malnourished’, as measured by anthropometric indicators, the household is classified as ‘malnourished’. By focusing on this age range rather than on older children and adults, it is possible to concentrate the analysis on programmes and policies of the past decade. Households without young children may also have nutritional problems, but, according to analyses of variance, they exhibit higher incomes, smaller families and lesser problems with health and access to the health system than the households with children. On other socioeconomic variables, they tend to be closer to the households with acutely malnourished children than to the households with normal or chronically malnourished children. To the extent that they belong to functional groups with malnourished children, they would, of course, also benefit from any intervention. The malnutrition indicator for each child is a function of both the child’s height-for-age measurement and weight-for-height measurement. The malnutrition indicator (Table 4) combines both the weightfor-height indicator (low, normal, high) and the height-for-age indicator (low, normal, high).“’ Children with a low value of weight-for-height, which measures present malnutrition, are ‘acutely malnourished’; children with a normal or high value of weight-for-height and a low
FOOD POLICY
February
1985
Nutritional functional classification study of Panama of height-for-age are ‘chronically malnourished’; and children with normal or high values for both weight-for-height and height-for-age are classified as ‘normal’. For the sake of this analysis, high values on both measures are classified as normal since the household malnutrition indicator was used to identify households likely to be at risk from substandard intakes and the wasting effects of poor health. The health problems associated with obesity are not considered in this study. Table 5 presents the functional groups ordered in descending order of the combined prevalence of acute and chronic malnutrition among children of nought to nine years old. Clearly, agricultural and rural occupations define the functional groups with the highest prevalence of malnutrition. In addition, the analyses showed that unskilled workers are also concentrated in the rural area. This strong rural-urban dichotomy on the prevalence of malnutrition among children also reveals itself in the data from the 1980 survey which was used to classify the adult population. The rural provinces have higher prevalence of adult malnutrition (Veraguas, in particular) and the predominantly urban provinces of Panama and Colon have the lowest adult malnutrition. l1 All the agricultural or rural functional groups rank above all the urban functional groups in terms of malnutrition. Almost three-quarters of all the malnourished children in Panama live in households whose head is dependent on the agricultural sector for his or her income and employment. The two groups having the most malnutrition (diversified farmers and horticulturalists) are those whose production is highly diversified. These two groups account for 34% of rural households as well as 35% of the malnutrition in the whole country. Diversified production can usually be attributed to portfolio behaviour in the presence of risk, or associated with variability in soil quality, differences in timing of cropping tasks and with market imperfections such as very poor market access in product and factor markets. l2 Both acute and chronic prevalence rates are higher among the rural functional groups, but the difference in the prevalence of chronic rates is more marked between the rural and urban groups. This would suggest that urban children that become malnourished are able to recover more quickly and remain well nourished
value
“Bermudez, op tit, Ref 5. “James A. Roumasset, ‘Unimportance of risk for technology design and agricultural development policy’, in A. Valdes, G.M. Scobie and J.L. Dillon, eds, Economics and the Design of Small-Farmer Technology, Iowa State University Press, 1979.
Table 5. Functional groups ordered by prevalence of acute or chronic malnutrition in children aged g-9 years. Houses
Functional group
aAcute malnutrition is defined as below two standard deviations below the WHO reference value for weight-for-height, and chronic malnutrition is defined as below two standard deviations below the WHO reference value for height-forage. Vfow percentages
FOOD POLICY
February
1985
Acute or chronic= Acute n O/O~ n %
without Chronic n %
Normal n %
children n %
Total households
218 234 124 57 58 305 212 43 46 283 108 101 42 197 221
286 29.9 236 37.6 128 38.1 62 39.5 72 42.4 259 36.1 141 31.7 63 47.7 22 26.2 152 29.5 121 446 125 47.9 26 33.3 163 40.2 135 33.7
717 627 336 157 170 718 445 132 84 516 271 261 78 406 401
Diversified small farms 213 157 Small horhculturalists Maize and rice producers 84 38 Solely rice producers Farmers with employees 40 154 Agricultural workers Unskilled workers 92 26 Maize alone 18 Cassava producers 81 Skilled workers 42 Salaried urban 35 Self-employed 10 Government 46 Commercial, transport 45 Professional, office
29.7 25.1 25.0 24.2 23.5 21.4 20.7 19.7 19.1 15.7 15.5 13.4 12.8 11.3 11.2
30 18 13 10 8 21 21 4 3 17 7 3 1 13 12
4.2 2.9 3.9 6.4 4.7 2.9 4.7 3.0 3.6 3.3 2.6 1.1 1.3 3.2 3.0
183 139 71 28 32 133 71 22 13 64 35 32 9 33 33
25.5 22.2 21.1 17.8 18.8 18.5 16.0 16.7 15.5 12.4 12.9 12.3 11.5 8.1 8.2
Total
20.3
181 3.4
898
16.9 2249
1079
30.4 37.3 36.9 36.3 34.1 42.5 47.6 32.6 54.7 54.8 39.9 38.7 53.9 48.5 55.1
42.3 1991
37.4 5319
69
Nutritional
functional
classification
study
of Panamu
more so than rural children. Rural children malnutrition persists, or at least recurs.
live in conditions
in which
Socioeconomic characteristics of the functional groups
‘3Parillon
et al, op tit,
Ref 6.
Low incomes, large families, health costs, unemployment, poor sanitation, lack of potable water, and illiteracy have been found to be the social, economic and health factors that are statistically related to malnutrition in Panama;‘” these are presented in Tables 6, 7 and 8. Table 6 presents selected variables by the nutritional status indicator. Table 7 presents the functional groups ranked by the prevalence of malnutrition, and associates each group with its average per cupita income, average family size, literacy of the head of household (an indicator of whether the household has an employed adult). and, for the agricultural households, the percentage of income and how much of that production is marketed by the household. Table 8 presents variables associated with the health system. Results from Table 7 indicate that lower per cupitu incomes are concentrated in the rural and agricultural functional groups, but within these there does not seem to be an obvious correlation between income Table 6. Socioeconomic
variations
Household variable
Nutritional Acute
Average per capita household income ($/month) Average family size Average mother’s education (years) Average father’s education (years) % with father employed % with health access problems % with recent morbidity % with access to potable water
Table 7. Means of income
Functional group Diversified small farms Horticulturalists Maize and rice Solely rice Farmers with employees Agricultural workers Unskilled workers Solely maize producers Cassava producers Skilled workers Salaried urban workers Self-employed Government Commercial, transport Professionals, office workers National
average
by nutritional
status
indicator
for Panama in 1980. Households without
status of children Chronic
55 5.9
68 6.7
85 5.8
193 3.2
6.4
5.1
6.6
7.3
5.7 87 2.1 46 70
4.7 85 3.8 45 56
6.2 85 2.4 43 80
5.9 76 13 34 79
and selected
socioeconomic
indicators
by functional
group.
Monthly per capita income (B/ma)
Family size (no persons)
Literacy of head of household (% households)
% households have adults employed
% income from agricultural production
% marketed surplus
127.00 87.00 56.47 66 10
5.4 5.2 5.0 4.9
80 80 77 75
73 69 77 74
37 50 54 55
34 59 53 53
11284
5.0
79
77
60
58
114.45 95 75
5.1 5.1
83 86
70 77
13 6
49 52
78.00 60 00 149 65
4.9 4.9 5.4
82 82 94
66 63 82
41 28 _
63 35 _
150.48 80.06 122.00
4.8 4.7 5.4
95 81 96
61 71 83
288.00
4.8
93
77
2
48
198.00
4.7
97
87
122.19
4.9
87
68
YB = iUS$.
70
FOOD POLICY
February
1985
Nutritional functional classification study of Panama Table 8. Distribution of health system indicators by functional group.
Diversified small farms Small horticulturalists Maize and rice producers Solely rice producers Farmers with employees Agricultural workers Unskilled workers Maize alone Cassava oroducers Skilled workers Salaried urban Self-employed Government Commercial, transport Professional, office National average
14Franklin and Harrell,
FOOD POLICY
op tit, Ref 7.
February
1985
(%)
Health money problems W)
Sick persons (“/)
52.02
5.29
11.64
45.50
31.8
55.73
3.79
7.25
40.69
46.0 42.8
48.23 56.34
3.44 4.03
9.38 6.71
41.56 40.94
42.2 25.6 11.8 31.5 19.2 5.7 0.8 31.5 2.7
49.69 68.89 83.78 63.78 62.34 94.25 99.22 65.73 97.22
4.24 2.54 1.35 2.38 2.53 1.55 0.74 3.13 0.00
4.85 7.06 4.72 6.35 2.53 2.52 3.33 3.91 1.28
46.06 41.24 41.35 40.48 40.51 43.02 39.63 34.38 38.46
94.88
96.69
0.74 0.24
1.49 1.46
37.13 35.77
75.83
2.20
5.50
39.90
High risk sanitation (W
Good
33.9
5.8 4.7 na
Health access problems
and the prevalence of nutritional problems. For the country as a whole, income is significantly associated with food consumption levels and with nutritional status. l4 Hence, this result suggests that for some functional groups the central cause of malnutrition is now low income per se. For example, the average per capita incomes of diversified small farmers are the highest for all agricultural functional groups, and this group has the highest prevalence of malnutrition when chronic and acute cases are combined. The causes of the nutritional problems for this group are undoubtedly related to factors other than incomes and food consumption levels. Health access problems, inadequate water and recent morbidity are high for this group. Apparently, health problems are such that they lead these households to indicate that lack of money causes problems with their access to health services. On the other hand, in the absence of other information it would appear that the nutritional problems of rice and maize producers are associated with comparatively low levels of income. In a similar fashion, the self-employed in the urban area report low incomes, yet exhibit low rates of malnutrition. The rural functional groups show a slight tendency for larger families than the urban functional groups, and there is also a slight correlation between larger families and malnutrition. Although this may reflect a causal relationship between close child spacing, high fertility and malnutrition, it is unlikely that a major effort in birth control services is warranted per se. Similarly, the association between literacy and malnutrition may reflect historical conditions rather than current conditions of access to schooling. In Panama a very large percentage of the school-aged population and of the 15-2.5 years age group is currently enrolled in some form of schooling. The more important information is revealed by the employment and income source data. The number of households that report an adult as ‘formally’ employed within the past week is low, ranging from 61% for households headed by urban, unskilled workers to 87% for white collar workers’ households. The source of income information reveals that more than half of rural incomes are derived from sources other than agricultural production on their own farms.
Nutritional fundonal
classification siudy of Panama
Table 8 presents the health experience and access characteristics for the functional groups. As can be seen, the diversified small farmers have higher rates of access problems to the health system, have a higher than average rate of high-risk sanitation facilities, and have relatively low coverage of adequate water supplies. Given their higher incidence of sick persons in the household, it appears then that malnutrition problems of this functional group are associated more with poor health than with low incomes, although it would appear that the problems with water, sanitation and morbidity cannot be overcome by their relatively higher incomes. It is precisely in this sense that the functional classification approach provides a diagnostic and targetting tool for nutrition planning. The diversified small farmers are able to obtain adequate though low incomes through off-farm employment, and through this income they are generally able to obtain adequate diets since their per cupifu income would classify them as not poor by the ECLA criterion. Nevertheless, they have the highest rates of malnutrition because of poor health and inadequate access to the public health system resources. A food- or income-oriented intervention would do little to ameliorate their what would be required would be nutritional problems; rather, improved access to the health system, including improved coverage of functioning of the water and sanitation systems.
Distribution of malnutrition
and functional groups
Analyses of the distribution of functional groups by district designated 20 districts as requiring priority attention from the public sector. The criteria for selection included, but were not limited to. malnutrition prevalence rates. For example, San Miguelito and La Chorrera have relatively low prevalences of malnutrition, but being populous have a high absolute number of households with malnourished children. They also have high concentrations of the functional groups with nutritional problems. These districts and their functional group composition are presented in Table 9. Among the districts requiring priority attention are ten from the central provinces, particularly from Veraguas Province. These districts have high concentrations of the highly diversified subsistence maize and rice producers. For example, the two districts requiring most attention, Canazas in Veraguas and La Pintada in Cocle, are inhabited predominantly by the functional group of highly diversified maize and rice producers that has the highest prevalence of malnutrition. In Canazas. per capitu incomes are among the lowest in the country, literacy is low. coverage of health system is low, and the households depend on off-farm work for over two-thirds of their income. In La Pintada, incomes and other indicators are higher, with a higher proportion (half) of income being derived from agricultural sources. In La Pintada, marketed surplus is also higher. These results suggest that an agricultural production/marketing type intervention would have a higher probability of impact in La Pintada than in Canazas. In Canazas the principal means for increasing incomes and improving nutrition would be through improved employment opportunities. In both districts access to public health services is low. The results illustrated by these two districts highlight the usefulness of the functional classification approach. Both districts are inhabited by a
FOOD POLICY
February
1985
Nutritional Table 9. Socioeconomic
District Canazas La Pintada Chagres Santa Fe La Mesa San Bias Las Palmas Renacimiento Boqueron Tole Ola Montijo Anton Penonome Prnogana Donoso Sona Santiago San Miguelito La Chorrera
characteristics
N 51 67 71 30 42 96 38 53 33 57 29 63 103 117 96 65 94 189 369 209
functional
classification
study of Panama
in 20 districts with high prevalence of malnutrition.
Average family size
% literacy of head of household
% head of household employed
Average per capita monthly income
5.53 5.43 5.62 5.40 4.79 5.76 5.05 5.96 5.58 5.07 4.97 5.00 5.72 5.17 4.74 4.52 5.57 5.46 6.04 5.11
61 79 90 90 95 80 80 77 71 50 85 87 83 92 74 82 86 92 96 90
89 87 91 96 97 92 91 83 72 77 83 83 73 80 85 89 90 87 84 78
27.17 52.48 64.99 49.61 37.89 64.38 14.34 52.87 63.77 59.32 88.77 63.53 44.86 92.71 33.27 134.60 56.07 97.93 111.39 230.99
% %
%
from agriculture
marketed surplus
adequate
32 53 62 87 26 31 84 34 28 12 99 30 16 43 89 67 18 19
29 51 73 52 21 56 30 37 47 44 76 37 28 53 50 74 40 42 16 49
25 35 56 20 56 59 26 31 52 72 43 89 82 61 3 56 73 77 100 96
4
functional group (diversified small farmers) whose national average per capita income is not low; yet in these districts incomes are very low, a quarter of the national average for Canazas and less than half the national average for La Pintada. Although both districts have health a near-term intervention would focus on income access problems, generation and would require substantially different forms in each district. A food-for-work programme during periods of seasonal unemployment might be effective in Canazas but would probably compete with other income-generating activities in La Pintada. Some districts are characterized by their remoteness and the difficulty in delivering public services to them. In each of these cases the specific interventions may differ, but much emphasis will need to be given towards promoting self-sufficiency and self-reliance through a community self-help approach. The role for direct public programme and policy intervention is limited by the difficulty of access. The districts of Santiago, La Chorrera and San Miguelito contain urban and peri-urban unskilled workers whose nutritional health can best be enhanced by ensuring stable urban employment through specific skill training and employment promotion policies.
Food policy issues The food policy issues are complex. At present there exists a policy of supporting maize and rice prices above import parity which results in higher incomes to maize and rice producers and may result in higher wages for rural labourers, but which results in higher food costs for market-dependent households. Economic efficiency costs and nutritional costs to non-producing households are an argument for removing the ‘support’ component of the maize and rice price policies. Removal of the price supports would result in lower food prices and farm incomes. The incidence of these lower incomes cannot be ascertained with the analyses undertaken thus far. Small rice and maize producers tend to be more dependent on their production for their income and they tend to sell a greater portion of their production. Thus, they are likely to be benefitting from the higher product prices; removal of the support
FOOD POLICY
February
1985
73
Nutritional functional classification study of Panama
element would tend to make them even poorer and perhaps exacerbate their nutritional problems. On the other hand, the impact of the higher prices is greater for the larger farmer because of higher yields and greater areas under production. The bulk of domestic rice production comes from large, mechanized farms in the valleys of the central region, whereas the bulk of the rural population lives on the hillsides and mountains of the central region. Larger farmers are also the employers of rural labourers. Removal of price supports could also affect the incomes of rural wage workers to the extent that it puts downward pressure on wages and the demand for labour. While it is reasonable to believe that the country-wide economic and nutritional benefits of removing the high support price levels for maize and rice will, in the long run. outweigh nutritional and income losses to low-income rural dwellers, provision must be made for protecting the incomes of the rural poor.
Towards a nutrition policy for Panama Panama has not had a nutrition policy; it has applied agricultural policy instruments to boost the incomes of maize and rice producers through price supports and institutional services in the agricultural sector,and through policies to put downward pressure on wage goods through retail price controls, particularly on beef. It is not clear that the farm income policies have been effective in raising the incomes of poor farmers; in fact, it is possible that the price support policies may have induced resource misallocations within the agricultural sector. To wit, high rice and maize prices may have forced small landholders to produce these grains, whereas in the absence of the support prices these grains might have been available at lower prices as consumer goods for farmers. and the farmers could have used resources currently allocated to maize and rice production to grow other more economically appropriate crops. Such a policy reversal could have also released labour for sale in rural labour markets, and, in the absence of employment sources, could have aggravated rural incomes. A nutrition policy for Panama must be based on three clear facts that arise from this study. First, malnutrition is overwhelmingly concentrated in functional groups whose principal economic activities are linked to the agricultural sector. Second, although for some groups and some districts the prevalence of malnutrition is principally linked to low incomes, these are being determined in large part by rural labour markets rather than by agricultural product markets. Third, for an important number of rural dwellers nutritional problems are still associated with inadequate provision or access to health services, potable water and sanitation. This exists despite significant long-term public investments in improving access to these systems. A nutrition policy in Panama would focus primarily on two aspects income generation through off-farm employment and health service provision. Only secondarily would a nutrition policy focus on increased food output. The focus on agricultural policy would be to remove the distortions impeding optimal resource allocation in the rural sector. An important caveat is that optimal resource allocation may imply accelerated rural-to-urban migration.
74
FOOD POLICY
February
1985
David L. Franklin, Marielouise W. Harrell and Cutberto Parillon
Despite significant progress, widespread malnutrition persists in Panama. This article describes a study which found that over half the malnutrition in the country is clustered in functional groups economically connected to the agricultural sector and which indicates that food availability is a serious problem for 25% of the population. Food intake deficits were equally prevalent among the urban and rural populations. Dietary composition was not seen as the principal problem; rather malnutrition problems are linked to low wages and poor access to public services. A nutrition policy for Panama must be centred first on employment and income generation, and then on extending and maintaining public health and nutrition services, particularly for the extremely poor in the central provinces. Food policy per se would not seem a priority area for public intervention. Keywords: policy
Panama; Nutrition; Public
David L. Franklin and Marielouise W. Harrell are with the Sigma One Corporation PO Box 12425. Raleiah. NC. USA. Cutberto Parillon works in tie’Directorate of Nutrition, Ministry of Health, PO Box 2048, Zone 1, Panama, Republic of Panama. The authors would like to acknowledge the collaboration and comments of Victor ValVerde of INCAP in Guatemala, Elena Brineman and Ben Severn of the Agency for International Development (AID), and the financial support of the Office of Nutrition of AID and of the government of Panama.
0306-9192/85/010063-11$3.00
0
This article presents the results of a functional classification study Panama which has been undertaken with the following objectives: 0 0
of
to provide an analytical and quantitative basis for nutrition planning; to provide information which might be useful in improving existing nutrition programmes and projects, and for incorporating nutritional activities into other developmental initiatives.
The study is based on data developed by the 1980 National Survey as part of the ‘Multi-sectoral analysis of the food and nutrition situation in Panama’, and on available secondary data from Panama. The approach of the study is to apply the concepts of the ‘functional classification approach as developed by Joy and Payne,‘, JoY,~ and Payne,3 and as implemented by Valverde et al.” The specific steps in the functional classification approach are: Define distinct categories of population groups in poverty conditions according to their connection to the economic and productive processes of the society in which they operate. These groups are called functional groups because an important characteristic of their definition is the occupation of the head of household. Locate these groups geographically and by administrative/political subdivisions within a country, or regions in that country. Enumerate or otherwise estimate the size of each group in terms of number of persons affected of each group in each geographic or political/administrative subdivision. Describe the social and economic characteristics that may be causally related to specific nutritional problems for each functional group.
The nutrition
problem
in Panama
More than 25% of the households in Panama have important nutritional problems in that at least 30% of the country’s preschool-aged children are affected by the growth retardation effects of malnutrition. Among
1985 Butterworth
& Co (Publishers)
Ltd
63
Nutritional
functional
classification study of Panama
‘J.L. Joy, ‘Economic aspects of food and nutrition planning’, First Asian Conference on Nutrition, 1971; J.L. Joy and P.R. Payne, ‘La nutrition y la planificacion del desarrollo national’, Alimenfacion y M&icion (FAO), Vol 1, No 4, 1975. “J.L. Joy, ‘Food and nutrition planning’, Journal of Agricultural Economics, No 24, 1973. 3P.R. Payne, ‘Nutrition planning and food policy’, Food Policy, Vol 1, No 2, 1976. W. Valverde eta/, ‘Classification funcional de poblaciones desnutridas en la republica de El Salvador’, Archives Latin0 Americanos de Nutrition, 1978. 50.1.C. Bermudez, ‘Estado nutritional de la poblacion adulta en la republica de Panama’, INCAP, Guatemala, MS Thesis, Universidad San Carlos, 1980. ‘C. Parillon, M. Harrell, B. Frazao, D. Franklin, Vial de Valdes, ‘Alimentacion y nutrition en Panama: la situation actual’, Ministerio de Salud, Republica de Panama, 1982. 7D.L. Franklin and M. Harrell, ‘Nutrition evaluation methods: a household production approach’, Proceedings of the Latin American Econometric Society, Santiago, Chile, 1983.
the adult population, 22.8% of males and 24% of females are classified as malnourished.” These nutritional problems are, in part, associated with levels of food consumption which are below internationally recommended levels of calorie intake. A large number of females in Panama use the bulk of their total real resources to obtain diets which are mainly below international standards of adequacy for food energy. These apparent deficits in food energy are associated with other indicators of poverty which, jointly with poor diets, have been shown to be important determinants of the prevalence and persistence of malnutrition in Panama.” Among the indicators of poverty that have been shown to be statistically and causally associated with households with malnutrition in Panama are: social and cultural characteristics of the household; variables such as low calorie intake levels, mothers’ opportunity cost of time, low income, and occupational characteristics of the working age adults in the household; health and environmental variables such as poor housing quality, lack of access to potable water, poor sanitary conditions, high levels of childhood morbidity from measles and diarrhoea, and poor access to the health services.’
Food consumption
aspects of the nutritional problem
Data from the 1980 National Nutrition Survey indicate that for more than half the households in the sample, calorie intakes adjusted for age and sex composition are below the FAO reference value of 2700 calories per calorie equivalent person. As many as 25% of the households have daily calorie intakes one-third lower than the reference value. There do not appear to be appreciable differences between the rural and urban distributions of nutrients among households. Diet composition is not a principal problem. Diets below the calorie reference level are more prevalent than those below the protein recommendations, and protein consumption does not appear to be a major problem for most households. The composition of urban and rural diets is presented in Tables 1 and 2. Rice is the important staple in the Panamanian diet; on a national level it accounts for one-third of calories, one-quarter of protein, and one-sixth of food expenditures. The importance of rice is greater in the diets of rural than of urban households. The diets of the rural population are composed of rice, oils Table 1. Calorie, protein and expenditure households (n=418).
64
shares for principal
food commodities
- rural
Average % of daily calories
Average % of daily protein
Average % of daily food expenditures
Corn Cereals Breads Beans Rice Fish Milk Poultry/eggs Beef Pork Oils Vegetables/fruits Sugars Others
3.6 03 5.6 6.1 36.0 2.1 3.0 3.2 48 1.o 11.3 14.5 5.3 3.2
2.7 0.4 5.9 13.2 26 9 10.5 5.3 8.0 12.4 1.8 0.0 8.0 0.0 4.9
2.8 0.4 4.3 5.7 20.0 6.2 5.5 9.0 10.6 2.2 6.7 14.1 2.5 10.0
Totals
100.0
100.0
100.0
FOOD POLICY February
1985
Nutritional Table 2. Calorie, protein and expenditure households (n=192).
functional
classification
study
of Panama
shares for principal food commodities
- urban
Average % of daily calories
Average % of daily protein
Average % of daily food expenditures
Corn Cereals Breads Beans Rice Fish Milk Poultry/eggs Beef Pork Oils Vegetablesifrults Sugars Others
1.8 0.7 11.4 3.8 28.3 1.0 6.3 5.2 9.6 1 .o 13.0 7.3 6.2 4.4
1.3 0.7 11.9 7.7 17.9 4.7 10.1 12.2 22.4 1.6 0.0 3.6 0.0 5.9
1.o 0.5 6.9 3.4 9.6 3.2 9.4 11.7 21.0 1.8 5.8 10.1 2.0 13.6
Totals
100.0
100.0
100.0
fats, fruits and vegetables, including roots, tubers, plantains and bananas. The urban diets are relatively more intense in beef and poultry as complements to the rice-based diet. The Directorate for Nutrition of the Ministry of Health in Panama publishes periodically an estimate of the per person food costs necessary to obtain a nutritionally adequate diet from the typical food-use pattern in Panama. This cost is known as the cost of the ‘basic food basket’ and was estimated as US$1.06 per adult per day at the time of the 1980 survey. This value was used to stratify the data from the survey households according to income and a criterion suggested by the Economic Commission for Latin America (ECLA). This criterion establishes three income cut-off points for defining levels of poverty; incomes below the cost of a basic food basket are said to reflect extreme poverty; incomes greater than the cost of one basic basket but lower than twice the monetary value of two basic baskets are said to reflect poverty; incomes at levels between twice and three times the cost of the basic basket are considered low but adequate, and income levels about three times the cost of the basic basket are considered as not poor.’ Table 3 presents the stratification of household incomes according to this criterion and the associated average values for food costs, calorie and protein intakes, and number of calorie equivalent persons per household. Table 3 suggests that low incomes are associated with low levels of food expenditure and food consumption. It also highlights the great disparity between urban and rural incomes - 23% of the urban sample households are judged to be in extreme poverty, whereas 54% of the rural households are judged to be in extreme poverty according to the ECLA criterion. and
Functional classification approach
8Sergio Molina, ‘La pobreza: description y analisis de politicas para superarla’, Revkfade La CEPAL, Santiago, Chile, December 1982.
FOOD POLICY
February
1985
The purpose of a nutritional functional classification study is to provide diagnostic information to assist planners and service providers in locating and identifying population groups with important nutritional problems. The basic approach consists of identifying the geographic and politico-administrative location of population groups whose sociodemographic characteristics are causally or statistically related to the prevalence of important nutritional problems. The methods are a planning rather than an analytical tool. These methods permit an
Nutritional
functional
classification
study of‘ Panama Table 3. Averages of consumption variables and other selected household variables for urban and rural households in the food consumption subsample of the National Nutrition Survey.”
Averages for urban households (n= 166)
Notes: ‘The food consumption subsample consisted of a 10% sample in every segment in the sample; as such, it is representative at the district, province and national levels. bCEP = Calorie equivalent person; weights each household member by the ratio of the age-sex calorie recommendation relative to the male adult recommendation. CMarket surplus = ratio of value of agricultural sales to value of agricultural production for each household. dMeasures of poverty: extreme poverty = gross income < 1 basic basket: poverty = 1 basic basket < gross income < 2 basic baskets; adequate = 2 basic baskets < gross income i 3 basic baskets; non-poor = gross income L 3 basic baskets. $1.06 = daily cost of basic food basket from ‘Determination y cost0 de la canasta basica de alimentos para Panama’, Ministerio de Salud, Panama.
Poverty level&
n
%
Money value of daily food consumption (household)
Extreme poverty Poverty Adequate Non-poor
38 32 23 73
23 19 14 44
5.28 7.06 6.64 5.76
166
100
Totals
CEPb per household
Daily calories per CEP
Daily protein per CEP (grams)
5.3 5.2 4.4 3.8
2126 2671 2586 2490
66.2 60.9 83.6 83.4
Averages for rural households (n=38 1)
Poverty levelsd
n
Extreme poverty Poverty Adequate Non-poor
204 81 34 62
54 21 9 16
Totals
361
100
%
Money value of daily food consumption (household) 3.96 4.29 5.27 4.54
CEPb HH
Daily calories per CEP
Protein per CEP (grams)
MarketedC surplus
% off-farm income
5.4 4.7 3.9 3.5
2230 2403 2667 2696
61.8 69.4 88.5 89.1
0.26 0.27 0.19 0.31
36 66 72 75
per
identification of the characteristics of population groups which allows the targetting of specific remedial activities, and provides estimates of the number of persons in each group and setting to facilitate the estimation of costs and level of effort required to deliver services or other remedial actions. The technique differs significantly from conventional nutrition planning approaches” in that these latter have been based principally on sociodemographic and physiological or health characteristics of persons at nutritional risk. By contrast, the functional classification technique attempts to relate the prevalence of malnutrition in each identified functional group by characteristics of that functional group that are associated with their connection to the productive process in the economy. In this manner the interventions or remedial actions that can be instituted include those which can be promoted through policy mechanisms. Therefore, a functional classification study identifies, quantifies and localizes groups of families according to their life pattern, social, economic and cultural problems, and the level of resources that are available to them. They are clustered according to these characteristics under the assumption that they will respond in a similar manner to specific policies and programmes. The design of such interventions should, of course, be the result of rigorous economic and social cost-benefit analysis.
Definition of functional groups
‘J.M. Pines, ‘National nutrition planning lessons of experience’, Food PO/icy, Vol7, No 4, 1982; FAO, Food and Nutrition Planning, Nutrition Consultants Report Series (Rome), No 35.
66
The 1980 National Nutrition Survey was designed with the same sampling frame as the World Fertility Survey for Panama; as such it is representative of the population in each administrative district as a self-weighting sample. Occupational questions in that survey were used to identifv each head of household with the occupation codes used by the International Labour Office. The groups thus-formed were further aggregated into larger groups with similar socioeconomic characteristics.
FOOD POLICY February 1985
Nutritionalfunctional classificationstudy of Panama
For the agricultural sector, the data from the 1980 survey and analyses of variance techniques were used to form seven distinct functional groups from those households engaged in agricultural production activities. The analyses of variance procedure was used to determine the extent to which the type and number of crops grown by the household served as an indicator of the prevalence of malnutrition. The seven functional groups for agricultural households were defined as follows: Small
horticultural producers. This group consists of households which farm more than half a hectare of land and produce fruits and vegetables but do not grow any rice or maize. They are generally located in the very remote areas or close to urban centres. Those located in remote areas consume almost all their own produce; those close to urban areas sell fruits and some vegetables for urban consumption. Their median land area sown to crops is 2.5 hectares. Small cassava producers. This group of farmers specializes in the production of cassava and their farms have a median crop area of 1 hectare. Small diversified producers. This group of small farmers produces at least three crops, of which two are maize and rice. They generally produce for subsistence consumption and their median crop area is 5 hectares. Small maize producers. These producers are specialized in maize production on their own farm. Some farms may be as large as 20 hectares in size, although the median farm size is 5 hectares. This group was isolated as a distinct group primarily to assess the impact of the maize price support policy which the country has pursued since 1974. Small rice producers. This group specializes in rice production and includes farms of up to 20 hectares, although the median farm size is 2.5 hectares. This group was formed to assess the impact of the rice support policy. Small rice and maize producers. These producers are similar in all respects to the previous two, except that the farm produces both maize and rice on farms whose median size is 5 hectares. Farmers with employees. This group consists of all other households that are engaged in agricultural production in farms 20 hectares or larger. This group has also identified itself as using hired labour on a permanent basis.
The above groups exhaust the households that live on farms; in general they represent small landholders with the following distribution of farm sizes: 25% farm less than 0.5 hectares; 50% farm less than 2.0 hectares; 75% farm less than 10.0 hectares; 90% farm less than 30 hectares; and only the top 5% farm 50 hectares or more. Even the so-called ‘farmers with employees’ represent relatively small landholdings, with median size of 32.5 hectares. It is important to note that this sample represents households whose residence is on or near the land which they farm. Households dwelling on large farms would typically be the households of hired caretakers. Those households not classified by the agricultural production activities were classified by occupation of the head of household:
l
Rural workers.
principally
FOOD POLICY
February
1985
This group is composed of households which reside in rural areas and whose income is derived principally
67
Nutritional functional
classification
study of Panama
Table 4. Definition of malnutrition indicator based on weight-for-height and height-forage. Weightfor-height
Low Normal
High
Height-for-age Low
Normal
High
Acute Chronic Chronic
Acute Normal Normal
Acute Normal Normal
from wage work. This group includes a few dwellers in peripheral urban areas whose economic activity is agricultural wage work. This group probably includes a large number of households whose head migrates seasonally or even continuously seeking employment on large farms, banana or sugar estates or in the urban areas. Salaried urban workers. This group consists of households whose head is permanently employed in an urban job. It includes persons engaged in construction, services, and industry. 50% of this group is concentrated in commercial or industrial occupations. Self-employed. This group of households represents those who identified themselves as self-employed and who do not engage in agricultural production activities. The majority reside in rural areas and tend to be engaged in commercial activities such as marketing of agricultural products. Commercial and transportation employees. This group includes permanently employed clerks and drivers, as opposed to ‘occasional’ workers in this sector. Professionals, office and financial employees. Households in this group include all professionals (ie college-trained persons who did not identify themselves in any other group), and ‘white collar’ office and financial employees. Skilled workers. This group includes all skilled occupations including barbers, craftsmen, carpenters etc. Unskilled workers. This group represents households whose heads are unskilled construction workers, loaders working in the transport sector, and other ‘occasionally employed’ workers. Government employees. Households in this group have heads who identify themselves as working for the government.
Distribution of nutritional status across functional group
“The heiaht-far-aae measurement is derived by ;ompariig the child’s present height to a standard reference and then classifying it as low (below two standard deviations (SD) of the standard reference), normal (between -2SD below the reference standard and +2SD above the reference standard), and high (above +2SD above the standard reference). In like fashion, the weight-for-height indicator is classified as low, normal or high based on comparing the child’s given height. These standards are stricter than conventional criteria such as the Gomez standard.
For the purpose of this analysis, nutritional status is established at the household level by determining the nutritional status of the children aged nine years or younger in each household. If any child in the household is ‘malnourished’, as measured by anthropometric indicators, the household is classified as ‘malnourished’. By focusing on this age range rather than on older children and adults, it is possible to concentrate the analysis on programmes and policies of the past decade. Households without young children may also have nutritional problems, but, according to analyses of variance, they exhibit higher incomes, smaller families and lesser problems with health and access to the health system than the households with children. On other socioeconomic variables, they tend to be closer to the households with acutely malnourished children than to the households with normal or chronically malnourished children. To the extent that they belong to functional groups with malnourished children, they would, of course, also benefit from any intervention. The malnutrition indicator for each child is a function of both the child’s height-for-age measurement and weight-for-height measurement. The malnutrition indicator (Table 4) combines both the weightfor-height indicator (low, normal, high) and the height-for-age indicator (low, normal, high).“’ Children with a low value of weight-for-height, which measures present malnutrition, are ‘acutely malnourished’; children with a normal or high value of weight-for-height and a low
FOOD POLICY
February
1985
Nutritional functional classification study of Panama of height-for-age are ‘chronically malnourished’; and children with normal or high values for both weight-for-height and height-for-age are classified as ‘normal’. For the sake of this analysis, high values on both measures are classified as normal since the household malnutrition indicator was used to identify households likely to be at risk from substandard intakes and the wasting effects of poor health. The health problems associated with obesity are not considered in this study. Table 5 presents the functional groups ordered in descending order of the combined prevalence of acute and chronic malnutrition among children of nought to nine years old. Clearly, agricultural and rural occupations define the functional groups with the highest prevalence of malnutrition. In addition, the analyses showed that unskilled workers are also concentrated in the rural area. This strong rural-urban dichotomy on the prevalence of malnutrition among children also reveals itself in the data from the 1980 survey which was used to classify the adult population. The rural provinces have higher prevalence of adult malnutrition (Veraguas, in particular) and the predominantly urban provinces of Panama and Colon have the lowest adult malnutrition. l1 All the agricultural or rural functional groups rank above all the urban functional groups in terms of malnutrition. Almost three-quarters of all the malnourished children in Panama live in households whose head is dependent on the agricultural sector for his or her income and employment. The two groups having the most malnutrition (diversified farmers and horticulturalists) are those whose production is highly diversified. These two groups account for 34% of rural households as well as 35% of the malnutrition in the whole country. Diversified production can usually be attributed to portfolio behaviour in the presence of risk, or associated with variability in soil quality, differences in timing of cropping tasks and with market imperfections such as very poor market access in product and factor markets. l2 Both acute and chronic prevalence rates are higher among the rural functional groups, but the difference in the prevalence of chronic rates is more marked between the rural and urban groups. This would suggest that urban children that become malnourished are able to recover more quickly and remain well nourished
value
“Bermudez, op tit, Ref 5. “James A. Roumasset, ‘Unimportance of risk for technology design and agricultural development policy’, in A. Valdes, G.M. Scobie and J.L. Dillon, eds, Economics and the Design of Small-Farmer Technology, Iowa State University Press, 1979.
Table 5. Functional groups ordered by prevalence of acute or chronic malnutrition in children aged g-9 years. Houses
Functional group
aAcute malnutrition is defined as below two standard deviations below the WHO reference value for weight-for-height, and chronic malnutrition is defined as below two standard deviations below the WHO reference value for height-forage. Vfow percentages
FOOD POLICY
February
1985
Acute or chronic= Acute n O/O~ n %
without Chronic n %
Normal n %
children n %
Total households
218 234 124 57 58 305 212 43 46 283 108 101 42 197 221
286 29.9 236 37.6 128 38.1 62 39.5 72 42.4 259 36.1 141 31.7 63 47.7 22 26.2 152 29.5 121 446 125 47.9 26 33.3 163 40.2 135 33.7
717 627 336 157 170 718 445 132 84 516 271 261 78 406 401
Diversified small farms 213 157 Small horhculturalists Maize and rice producers 84 38 Solely rice producers Farmers with employees 40 154 Agricultural workers Unskilled workers 92 26 Maize alone 18 Cassava producers 81 Skilled workers 42 Salaried urban 35 Self-employed 10 Government 46 Commercial, transport 45 Professional, office
29.7 25.1 25.0 24.2 23.5 21.4 20.7 19.7 19.1 15.7 15.5 13.4 12.8 11.3 11.2
30 18 13 10 8 21 21 4 3 17 7 3 1 13 12
4.2 2.9 3.9 6.4 4.7 2.9 4.7 3.0 3.6 3.3 2.6 1.1 1.3 3.2 3.0
183 139 71 28 32 133 71 22 13 64 35 32 9 33 33
25.5 22.2 21.1 17.8 18.8 18.5 16.0 16.7 15.5 12.4 12.9 12.3 11.5 8.1 8.2
Total
20.3
181 3.4
898
16.9 2249
1079
30.4 37.3 36.9 36.3 34.1 42.5 47.6 32.6 54.7 54.8 39.9 38.7 53.9 48.5 55.1
42.3 1991
37.4 5319
69
Nutritional
functional
classification
study
of Panamu
more so than rural children. Rural children malnutrition persists, or at least recurs.
live in conditions
in which
Socioeconomic characteristics of the functional groups
‘3Parillon
et al, op tit,
Ref 6.
Low incomes, large families, health costs, unemployment, poor sanitation, lack of potable water, and illiteracy have been found to be the social, economic and health factors that are statistically related to malnutrition in Panama;‘” these are presented in Tables 6, 7 and 8. Table 6 presents selected variables by the nutritional status indicator. Table 7 presents the functional groups ranked by the prevalence of malnutrition, and associates each group with its average per cupita income, average family size, literacy of the head of household (an indicator of whether the household has an employed adult). and, for the agricultural households, the percentage of income and how much of that production is marketed by the household. Table 8 presents variables associated with the health system. Results from Table 7 indicate that lower per cupitu incomes are concentrated in the rural and agricultural functional groups, but within these there does not seem to be an obvious correlation between income Table 6. Socioeconomic
variations
Household variable
Nutritional Acute
Average per capita household income ($/month) Average family size Average mother’s education (years) Average father’s education (years) % with father employed % with health access problems % with recent morbidity % with access to potable water
Table 7. Means of income
Functional group Diversified small farms Horticulturalists Maize and rice Solely rice Farmers with employees Agricultural workers Unskilled workers Solely maize producers Cassava producers Skilled workers Salaried urban workers Self-employed Government Commercial, transport Professionals, office workers National
average
by nutritional
status
indicator
for Panama in 1980. Households without
status of children Chronic
55 5.9
68 6.7
85 5.8
193 3.2
6.4
5.1
6.6
7.3
5.7 87 2.1 46 70
4.7 85 3.8 45 56
6.2 85 2.4 43 80
5.9 76 13 34 79
and selected
socioeconomic
indicators
by functional
group.
Monthly per capita income (B/ma)
Family size (no persons)
Literacy of head of household (% households)
% households have adults employed
% income from agricultural production
% marketed surplus
127.00 87.00 56.47 66 10
5.4 5.2 5.0 4.9
80 80 77 75
73 69 77 74
37 50 54 55
34 59 53 53
11284
5.0
79
77
60
58
114.45 95 75
5.1 5.1
83 86
70 77
13 6
49 52
78.00 60 00 149 65
4.9 4.9 5.4
82 82 94
66 63 82
41 28 _
63 35 _
150.48 80.06 122.00
4.8 4.7 5.4
95 81 96
61 71 83
288.00
4.8
93
77
2
48
198.00
4.7
97
87
122.19
4.9
87
68
YB = iUS$.
70
FOOD POLICY
February
1985
Nutritional functional classification study of Panama Table 8. Distribution of health system indicators by functional group.
Diversified small farms Small horticulturalists Maize and rice producers Solely rice producers Farmers with employees Agricultural workers Unskilled workers Maize alone Cassava oroducers Skilled workers Salaried urban Self-employed Government Commercial, transport Professional, office National average
14Franklin and Harrell,
FOOD POLICY
op tit, Ref 7.
February
1985
(%)
Health money problems W)
Sick persons (“/)
52.02
5.29
11.64
45.50
31.8
55.73
3.79
7.25
40.69
46.0 42.8
48.23 56.34
3.44 4.03
9.38 6.71
41.56 40.94
42.2 25.6 11.8 31.5 19.2 5.7 0.8 31.5 2.7
49.69 68.89 83.78 63.78 62.34 94.25 99.22 65.73 97.22
4.24 2.54 1.35 2.38 2.53 1.55 0.74 3.13 0.00
4.85 7.06 4.72 6.35 2.53 2.52 3.33 3.91 1.28
46.06 41.24 41.35 40.48 40.51 43.02 39.63 34.38 38.46
94.88
96.69
0.74 0.24
1.49 1.46
37.13 35.77
75.83
2.20
5.50
39.90
High risk sanitation (W
Good
33.9
5.8 4.7 na
Health access problems
and the prevalence of nutritional problems. For the country as a whole, income is significantly associated with food consumption levels and with nutritional status. l4 Hence, this result suggests that for some functional groups the central cause of malnutrition is now low income per se. For example, the average per capita incomes of diversified small farmers are the highest for all agricultural functional groups, and this group has the highest prevalence of malnutrition when chronic and acute cases are combined. The causes of the nutritional problems for this group are undoubtedly related to factors other than incomes and food consumption levels. Health access problems, inadequate water and recent morbidity are high for this group. Apparently, health problems are such that they lead these households to indicate that lack of money causes problems with their access to health services. On the other hand, in the absence of other information it would appear that the nutritional problems of rice and maize producers are associated with comparatively low levels of income. In a similar fashion, the self-employed in the urban area report low incomes, yet exhibit low rates of malnutrition. The rural functional groups show a slight tendency for larger families than the urban functional groups, and there is also a slight correlation between larger families and malnutrition. Although this may reflect a causal relationship between close child spacing, high fertility and malnutrition, it is unlikely that a major effort in birth control services is warranted per se. Similarly, the association between literacy and malnutrition may reflect historical conditions rather than current conditions of access to schooling. In Panama a very large percentage of the school-aged population and of the 15-2.5 years age group is currently enrolled in some form of schooling. The more important information is revealed by the employment and income source data. The number of households that report an adult as ‘formally’ employed within the past week is low, ranging from 61% for households headed by urban, unskilled workers to 87% for white collar workers’ households. The source of income information reveals that more than half of rural incomes are derived from sources other than agricultural production on their own farms.
Nutritional fundonal
classification siudy of Panama
Table 8 presents the health experience and access characteristics for the functional groups. As can be seen, the diversified small farmers have higher rates of access problems to the health system, have a higher than average rate of high-risk sanitation facilities, and have relatively low coverage of adequate water supplies. Given their higher incidence of sick persons in the household, it appears then that malnutrition problems of this functional group are associated more with poor health than with low incomes, although it would appear that the problems with water, sanitation and morbidity cannot be overcome by their relatively higher incomes. It is precisely in this sense that the functional classification approach provides a diagnostic and targetting tool for nutrition planning. The diversified small farmers are able to obtain adequate though low incomes through off-farm employment, and through this income they are generally able to obtain adequate diets since their per cupifu income would classify them as not poor by the ECLA criterion. Nevertheless, they have the highest rates of malnutrition because of poor health and inadequate access to the public health system resources. A food- or income-oriented intervention would do little to ameliorate their what would be required would be nutritional problems; rather, improved access to the health system, including improved coverage of functioning of the water and sanitation systems.
Distribution of malnutrition
and functional groups
Analyses of the distribution of functional groups by district designated 20 districts as requiring priority attention from the public sector. The criteria for selection included, but were not limited to. malnutrition prevalence rates. For example, San Miguelito and La Chorrera have relatively low prevalences of malnutrition, but being populous have a high absolute number of households with malnourished children. They also have high concentrations of the functional groups with nutritional problems. These districts and their functional group composition are presented in Table 9. Among the districts requiring priority attention are ten from the central provinces, particularly from Veraguas Province. These districts have high concentrations of the highly diversified subsistence maize and rice producers. For example, the two districts requiring most attention, Canazas in Veraguas and La Pintada in Cocle, are inhabited predominantly by the functional group of highly diversified maize and rice producers that has the highest prevalence of malnutrition. In Canazas. per capitu incomes are among the lowest in the country, literacy is low. coverage of health system is low, and the households depend on off-farm work for over two-thirds of their income. In La Pintada, incomes and other indicators are higher, with a higher proportion (half) of income being derived from agricultural sources. In La Pintada, marketed surplus is also higher. These results suggest that an agricultural production/marketing type intervention would have a higher probability of impact in La Pintada than in Canazas. In Canazas the principal means for increasing incomes and improving nutrition would be through improved employment opportunities. In both districts access to public health services is low. The results illustrated by these two districts highlight the usefulness of the functional classification approach. Both districts are inhabited by a
FOOD POLICY
February
1985
Nutritional Table 9. Socioeconomic
District Canazas La Pintada Chagres Santa Fe La Mesa San Bias Las Palmas Renacimiento Boqueron Tole Ola Montijo Anton Penonome Prnogana Donoso Sona Santiago San Miguelito La Chorrera
characteristics
N 51 67 71 30 42 96 38 53 33 57 29 63 103 117 96 65 94 189 369 209
functional
classification
study of Panama
in 20 districts with high prevalence of malnutrition.
Average family size
% literacy of head of household
% head of household employed
Average per capita monthly income
5.53 5.43 5.62 5.40 4.79 5.76 5.05 5.96 5.58 5.07 4.97 5.00 5.72 5.17 4.74 4.52 5.57 5.46 6.04 5.11
61 79 90 90 95 80 80 77 71 50 85 87 83 92 74 82 86 92 96 90
89 87 91 96 97 92 91 83 72 77 83 83 73 80 85 89 90 87 84 78
27.17 52.48 64.99 49.61 37.89 64.38 14.34 52.87 63.77 59.32 88.77 63.53 44.86 92.71 33.27 134.60 56.07 97.93 111.39 230.99
% %
%
from agriculture
marketed surplus
adequate
32 53 62 87 26 31 84 34 28 12 99 30 16 43 89 67 18 19
29 51 73 52 21 56 30 37 47 44 76 37 28 53 50 74 40 42 16 49
25 35 56 20 56 59 26 31 52 72 43 89 82 61 3 56 73 77 100 96
4
functional group (diversified small farmers) whose national average per capita income is not low; yet in these districts incomes are very low, a quarter of the national average for Canazas and less than half the national average for La Pintada. Although both districts have health a near-term intervention would focus on income access problems, generation and would require substantially different forms in each district. A food-for-work programme during periods of seasonal unemployment might be effective in Canazas but would probably compete with other income-generating activities in La Pintada. Some districts are characterized by their remoteness and the difficulty in delivering public services to them. In each of these cases the specific interventions may differ, but much emphasis will need to be given towards promoting self-sufficiency and self-reliance through a community self-help approach. The role for direct public programme and policy intervention is limited by the difficulty of access. The districts of Santiago, La Chorrera and San Miguelito contain urban and peri-urban unskilled workers whose nutritional health can best be enhanced by ensuring stable urban employment through specific skill training and employment promotion policies.
Food policy issues The food policy issues are complex. At present there exists a policy of supporting maize and rice prices above import parity which results in higher incomes to maize and rice producers and may result in higher wages for rural labourers, but which results in higher food costs for market-dependent households. Economic efficiency costs and nutritional costs to non-producing households are an argument for removing the ‘support’ component of the maize and rice price policies. Removal of the price supports would result in lower food prices and farm incomes. The incidence of these lower incomes cannot be ascertained with the analyses undertaken thus far. Small rice and maize producers tend to be more dependent on their production for their income and they tend to sell a greater portion of their production. Thus, they are likely to be benefitting from the higher product prices; removal of the support
FOOD POLICY
February
1985
73
Nutritional functional classification study of Panama
element would tend to make them even poorer and perhaps exacerbate their nutritional problems. On the other hand, the impact of the higher prices is greater for the larger farmer because of higher yields and greater areas under production. The bulk of domestic rice production comes from large, mechanized farms in the valleys of the central region, whereas the bulk of the rural population lives on the hillsides and mountains of the central region. Larger farmers are also the employers of rural labourers. Removal of price supports could also affect the incomes of rural wage workers to the extent that it puts downward pressure on wages and the demand for labour. While it is reasonable to believe that the country-wide economic and nutritional benefits of removing the high support price levels for maize and rice will, in the long run. outweigh nutritional and income losses to low-income rural dwellers, provision must be made for protecting the incomes of the rural poor.
Towards a nutrition policy for Panama Panama has not had a nutrition policy; it has applied agricultural policy instruments to boost the incomes of maize and rice producers through price supports and institutional services in the agricultural sector,and through policies to put downward pressure on wage goods through retail price controls, particularly on beef. It is not clear that the farm income policies have been effective in raising the incomes of poor farmers; in fact, it is possible that the price support policies may have induced resource misallocations within the agricultural sector. To wit, high rice and maize prices may have forced small landholders to produce these grains, whereas in the absence of the support prices these grains might have been available at lower prices as consumer goods for farmers. and the farmers could have used resources currently allocated to maize and rice production to grow other more economically appropriate crops. Such a policy reversal could have also released labour for sale in rural labour markets, and, in the absence of employment sources, could have aggravated rural incomes. A nutrition policy for Panama must be based on three clear facts that arise from this study. First, malnutrition is overwhelmingly concentrated in functional groups whose principal economic activities are linked to the agricultural sector. Second, although for some groups and some districts the prevalence of malnutrition is principally linked to low incomes, these are being determined in large part by rural labour markets rather than by agricultural product markets. Third, for an important number of rural dwellers nutritional problems are still associated with inadequate provision or access to health services, potable water and sanitation. This exists despite significant long-term public investments in improving access to these systems. A nutrition policy in Panama would focus primarily on two aspects income generation through off-farm employment and health service provision. Only secondarily would a nutrition policy focus on increased food output. The focus on agricultural policy would be to remove the distortions impeding optimal resource allocation in the rural sector. An important caveat is that optimal resource allocation may imply accelerated rural-to-urban migration.
74
FOOD POLICY
February
1985