A METHODOLOGY FOR RANKING COUNTRIES ACCORDING TO RELATIVE FOOD

A METHODOLOGY FOR RANKING

COUNTRIES ACCORDING TO RELATIVE FOOD INSECURITY

by

Jerre A. ManarollaFVA/PPl'1

Presented at ~he A.I.D.Economists Conference

November 12-17, 1989

TABLE OF CONTENTS

FOOD SECURITY: b DEVELOPMENT PARADIGM FOR THEu.s. FOOD AID PROGRAM 1

A. METHODLOGICAL APPROACH1. Purpose and Objectives of the Analysis 42. Measuring Hunger 43. The Concept and Dimeilsions of Food Security 54. Food Security and Food Aid Allocations

a. Ranking countries according to food security 6b. Allocating food aid Oil the ba~is of food security 6

B. DETAILED DESCRIPTION OF THE INDICES1. National Food Self Reliance Index (NRI)2. Household Food Access Index (HAl)3. Performance Index (PI)4. Food Security Index (FSI)

689

10

C. FOOD SECURITY ~OUNTRY RANKINGSOverview of Statistical Ranking Techniques Used 11

2. The Sum-of-individual-ranks Technique 123. The Normalized Value Technique 124. Calculating a Composite Index Using Both Techniq~vs 135. Results of the Statistical Rankings 13

D. ALLOCATING the P.L. 480 BUDGET to FOOD INSECURE COUNTRIES1. The P.L. 480 Budget: Actual and Hypothetical

Allocations 142. Proposed Allocation of a Food Security Fund 15

~1 Billion Food Security FundSelf. Reliance IndexSelf Reliance Index -- Adjusted for Debt

Figure 2:Table I:Table IA:

Table II:Table III:Table IV:

ANNEX OF FIGURES AND TABLESFigure 1: 75 Developing Countries Grouped According to Relative

Food SecurityAllocation ofNational FoodNational FoodServiceHousehold Food Access IndexPerformance IndexFood Security Index

A METHODOLOGY for RANKINGCOUNTHI~S ACCORDING to RELATIVE FOOD INSECURITY

FOOD SECURITY: A DEVELOPMENT PARADIGM FOR THE U.S. FOOD AID PROGRAM

Although the primary rationale for a U.S. Government food aidprogram is the developing country hunger problem, the current P.L.480 program doe~ not address that problem with sufficientdirectness. In fact, much of our food aid does not go to hungrypeople. Rather, it is provided to achieve macroeconomicstabilization, economic policy reform, market development, U.S.foreign policy objectives, and to satisfy the particular interestsof commodity producers, value added food processors and shippers.

Estimates of the world's hungry range from 500 million on up. Theworld hungeI problem is no longer viewed as primarily a supplyproblem, as it was in the 1950s, 1960s and ever the 1970s. For manyin developing countries, the problem is lack of ~conomic access tofood for many low income households in develcping countries aroundthe world. This fundamental reorientation has been captured underthe rubric of "food security", a term which marries food supplyconcerns with economic growth, access and safety net concerns.

Developing country food security is understood to depend on theamount of food available in-country and economic access to theavailable food by the people living in that country. More peoplehave ~dequate access to food if incomes are high, if incomeincreases arJ broadly shared, and if food prices are low andstable. More food will be available in-country if domestic foodproduction is large and t:.le country's ;.;apaci ty to import food(foreign exchange earnings) is strong.

The food security paradigm not only provides a deeper understandingof hunger in the developing world, but it may also reflect a morerealistic view of the current constituency for food aid within theUnited States. Agricultural commodity groups have increasingly lostinterest in food aid, seemingly preferring explicit, short-termexport promotion programs like the export enhancement (EEP) and GSMcredity guarantee programs. To the extent that commodity groupshave lost interest in food aid, o~e of the main sources of politicalsupport for food aid will have disappear. That leaves the "hungerlobby". Support for food aid from that quarter could bestrengthened if food aid were more effectively focussed on povertyand hunger and it it fostered participation by a broader range ofU.S. private voluntary organizations (PVOs). Food aid will find iteasier to survive politically with a renewed focus on a compellingissue that sparks the imagination of the American people, yet doesnot threaten traditional P.L. 480 supporters.

The food security paradigm appeals to diverse groups interested in,or dependant on, food aid. It melds five major concerns of food aidconstituents.

- First, a concern for sustainable agriculture (the role of theagricultural technologist and ·the environmentalist);

Second, a concern for employment and incomes -- recognizingthat access is as important as supply (grist for the economist);

- Third, a concern for the safety net, for those left out of thegrowth process -- but a sustainable safety net that carefullyconsiders who should be included, at what costs and with whatkinds of financing (enter the humanitarian and congressionalbudget-watchers);

Fourth, a concern for expanding world trade in agriculturalproducts (welcome U.S. exporters); and

- Fifth, a concern for pva disenchangment with traditional relieffeeding in favor of more developmental efforts.

Melding those five elements -- a concern for sustainable production,for employment and incomes, for the safety net, for trade expansion,and for pva vitality -- could pull together a sustainable food aidcoalition as well as provide the maklllgs of a new developmentstrategy rooted in our ethical and humanitarian concern for theworld's hungry, in sound economic approaches to the resolution ofhunger, and in the self interest of U.S. agriculture.

What might that new development strategy look like if the food aidprogram were rooted in food security? Let's start with the countryallocations. A first step would be to rank developing countriesaccording to relative food insecurity, where food security dependson the developing country's ability to produce or import food andeconomic access to the available food on the part of the population.

Ideally, estimates of food need for each developing country shouldbe based on the numbers of undernourished people in the country andby how much they are under nourished. But this data is simply notavailable for enough countries. .In many cases, the survey data thatis available is 10-20 years old.

As a second-best, one can use indirect measures of the severity of anation's hunger or malnutrition problem. As conceived in thisstudy, food security embraces two separate components, one at thecountry level and the other at the individual or household level.At the country level, food self reliance represents a country'scapacity to provide enough food for its citizens. At the householdlevel, the critical issue is economic access to food supplies

- J -

·<.ivai l'.3ble in-count,ry. The operat.ional question. then becomes how tomea Sll re t.hese t,,10 componen ts . Two rE;ad i 1y ava i 1. able indicators werellsed to measure a country's ability to provide food to its citizens,that is. food self reliance: per capita domestic food production andper capitel foreign exchange earnings. Three readily availableindicato~s were used to measure economic access to food: percapitaGNP, average daily per capita counsumption of calories and themortlity rate of infants and children under five years. Why werethese indicators chosen?

Food self reliance is conceived as a measure of country levelcapacity to provide food rather than as the actual amount of foodavailable in-country. Whereas actual in-country food supplies wouldconsist of the sum of domestic f00d production, commercial importsof food and food aid, food self reliance uses domestic foodproduction and foreign exchange earnings to measure a country'scapacity to provide food to its citizens. Again, data are notavailable for all countries: FAD data on cereal production areavailable through 1986 for all countries except Cape Verde whileWorld Bank/IMF data on total foreign exchange earnings areunavailable for Afghanistan, Angola, Bhutan, Guinea and Lebanon.

Economic access to food is conceived as a measure of income leveland dis~ribution and food price level and stability. Although across-country comparision of food prices is not possible because noappropriate data series exist for a large enough cross-section ofdeveloping countries, three readily available indicators were chosento represent average food consumption and income levels and howequitably econo~ic access to food is distributed within thepopulation. Av~rage per capita calorie consumption is conceptuallya good measure of the generalized level of malnourishment and datais available for all the 75 countries included in this analysis.However, average per capita calorie consumption does not sayanything about the distribution of food consumption; that is, whatportion of the population consumes significantly less than theaverage. By adding per capita GNP, a measure of the averagepurchasing power of citizens in each country can be introduced. Butper capita GNP data are not available for Afghanistan, Lebanon andAngola because of the difficulty of making such estimates underconditions of civil war. Like average calorie consumption, however,per capita GNP says nothing about how income is distributed; thatis, what portion of the population lives on significantly lessincome than this average level. The under five mortality rate wasintroduced because it was felt that higher mortality rates indicatea more unequal distribution of living standards, be they measured interms of food consumption, income (GNP) or some other variable.Like average calo~ie consumption, under five mortality is availablefor all 75 co~ntries.

- 3 -

f\. METHODOLOGICAL APPHOACH

'fhe purpose of the analysis is two-fold: to develop criteria forallocatlng food aid resources by country and to outline adevelopment strategy for using U.S. food aid resources once incountry. both based on the principles of food security.

The food aid allocation objectives are threefold: (i) identify thosedeveloping countries that are most food insecure and most dependanton concessional food imports; (ii) identify those developingcountries in transition to trading partner status (as commercialpurchasers of agricultural commodity imports); ~nd (iii) indicatehow A.I.D.'s food aid budget might be allocated to these countries,especially the food insecure countries.

As for how to use food aid resources, the objectives are alslthree-fold: (i) categorize developing countries according to themain causes of food insecurity (whether a problem of not enoughfood, poor access to food, economic stagnation, etc); (ii) provide acoherent focus for the use of food aid resources, centered onreducing world hunger; and (iii) outline a strategy for adaptingspecific uses of food aid resources to the different developmentsituations faced by individual countries.

2. Measuring Hunger

Objectively, a person is hungry when he does not consume enough foodto maintain health and a normal level of activity. At the countrylevel, the degree of hunger conceptually depends on the number ofpeople that consume less than the minimum standard and how far belowthe standard actual food consumption of that hungry group is.However, no single measure of country-level hunger exists for enoughcountries to be useful in this analysis. A 1988 study conducted forthe FVA Bureau provides direct malnutrition data for only 54countries, and the dates of the survey data cited range from themid-1960s to the present.

The most widely used measure of food consumption is daily calorieintake where a minimum standard daily calorie requirement is assumednecessary in order for individuals to remain healthy and able towork. The U.N.'s Food and Agriculture Organization defines this percapita standard at 2300 calories per day, while others argue that itshould be lower. Average country-level malnutrition can beestimated by comparing this minimum standard to estimated actualconsumption levels -- the larger the gap the hungrier people in thatcountry are assumed to be. However, this estimate of averagemalnutrition at the country level does not take into account greater

- 4-

(,)' lesser c1iffel'encc5 in aCCE:~·;S to fooel by the poorest, sectors ofsociety, nor does it place enough direct emphasis on the country'sproductive capasjty.

In the absence of an ideal measure of malnourishment with broadenough country covprage, this analysis must adopt more indirectmethods. The food security index was developed to indicate th~

combined irr.pact of country-level capacity to make enough foodavailable on the domestic food market and household access to theavailable food,

3. GQncept and Dimensions of Food Security

The paradigm of food security provides the conceptual framework forthis analysis of the macroeconomic causes of hunger andmalnutrition. The analysis looks at food security from the nationaland household levels, similar to the way Brown Unversity's 1989World Hunger Report conceives of world hunger, save th~t Brown alsoidentifies a third dimension of hunger at the intra-household level.

This paper argues that income, employment, food prices, foodproduction and food imports are the main determinants of hunger indeveloping countries. A useful way to define food security at thenational level is to think in terms of food self relidnce -- can thecountry make enough food available to its citizens. The concept offood self reliance has become a tenet of A.I.D.'s agriculturaldevelopment strategy, embracing as it does a country's ability toproduce food domestically and its capacity to import commerciallythe food it cannot or does not want to produce. Food self relianceis juxtaposed against the more simplistic and shortsighted conceptof food self sufficiency. Many development economists now arguethat a focus on greater food self reliance should be at the heart ofcountry development strategies to attain food security, that is, toeliminate hunger. An equally useful way to look at household levelfood security is in terms of economic access to enough food throughthe market. The concept of economic access to food, defined interms of incomes, employment and food prices, is at the heart ofA.I.D.'s increased emphasis on growth through revitalized privatesector institutions.

In summary, countries that are more self reliant in their ability toprovide domestic populations with adequate supplies of food andwhere most citizens have adequate economic access ·to available foodsupplies are defined as relatively food secure. Countries withsevere access problems and that lack the capacity to produce orimport adequate food supplies are considered to be the most foodinsecure. But the methodological question remains: how to developindicators of national food self reliance and household food accessfor each country of interest to A.I.D.

a) ranking countries according to food security

Countries are ranked according to a statistical index of foodsecurity. Operationally, the food security index is conceived as acomposite index of two elemental indices, one representingcountry-level capacity to provide enough food supplies (nationalfood self reliance index) and the other, household economic accessto available food suppli~s (household food access index). Fivesocia-economic indicators cLrrently make up the food security index,two representing food self reliance and three representing questionsof access.

b) allocating food aid on the basis of food security

Since the 75 countries included in the analysis range in size fromIndia, with a population wall over 700 million, to Cape Verde, witha population of around 300 thousand, food security is expressed inper capita terms. In general, it is assumed that the most foodinsecure countries need more economic assistance per capita, andthat food aid can be especially useful in helping these countriesbecome more food secure. From a food ai0 programmatic view, implythat the most food insecure countries need more food aid per capitathan the others, but that the larger countries may still need largervolumes of food aid. This preliminary conclusion that the more foodinsecure countries need more food aid, aud that the U.S. Governmentshould provide it, must eventually be checked against a country'seconomic policy stance and the appropriateness of food aid becauseof high transport costs or other inefficiencies. A country's totalfood aid needs would have to take into account the country's degreeof food insecurity and its population.

B. DETAILED DESCRIPTION OF THE INDICES

1. National Food Self Reliance Index (NRI)

National food self reliance is defined as a country's capacity toproduce food plus its capacity to import food it does not or cannotproduce. The NRI is one of three elemental indices that con~titute

the bas~s of this analysis. In line Hith this genera! definition,the analysis uses per capita food production and gross foreignexchange earnings data to compose the NRI.

Given the debt problems experienced by so many developing co~ntries

during the 1980s, some economists argue that gross foreign exchangeearnings should be adjusted for debt service. Therefore, twoversions of the NRI were calculated, one with gross foreign exchangeearnings and another with debt service obligations subtracted from

the gross figul"e. Ta ble 1 presents the count,ry "ran!{ings for the NRIusing gross foreign exchange earnings while Table IA ranks countriesusing gross earnings adjusted for debt service obligations. As itturns out, adjusting for debt service does not make much differenceto the rank ordering of countries in the NRI; only one countrychanges its rank by more than 10 places.

Other critics might argue that the definition of national food selfreliance should be extended to include variability of domestic foodproduction and foreign exchange earnings as well as average levelsof these indicators. Future evolution of the analysis will move inthat direction.

Specifically, two socio-economic indicators are used to calcula~e

the food access index:

i) FOODPROD, defined as the average annual per capita foodprciuction (measured in terms of cereal equivalent) for theperiod 1984-86 (from the World Bank, based on FAO country-leveldata aud the Bank's population estimates); and

ii) FOREX, defined as the average annual per capita grossforeign exchange earnings for the period 1984-86 (using WorldBank balance of payments and population data) or

ii') AFOREX, defined a5 the averate annual per capit grossforeign exchange earnings for the period 1984-86 adjusted forcebt service (incorporating World Bank estimates of debt serviceobligations)

Gross foreign exchange earnings includes the earnings from exportsof merchandise and services (non-factor and factor services) plusthe inflow of private unrequited transfers from abroad. Grossfigures are available for merchandise exports and non-factor andother services, but the only data available in the World Tables forprivate unrequited transfers is either partial (worker remittances)or net of transfers to abroad (net current transfers). Data used inthe analysis to represent transfers is a mixture of net currenttransfers and worker remittances. Net current transfers is usedunless the figure for worker remi·ttances is larger, in which case,the figure for worker remittances is used. Although this techniquestill underestimates gross private unrequited transfers, itconstitutes the best data available.

Five countries lack adequate data on gross foreign exchange earningsfor the period 1984-86: Afghanis·tan, Angola, Bhutan, Guinea andLebanon. Although Burma lacks data on non-factor and otherservices, the numbers are assumed to be small in any case.Therefore, Burma was included in the analysis using the three-yearaverage of merchandise exports and net current transfers. One

- 7 -

~dditional country, Cape Verde. Jacks ~AO data on domestic foodproduction. Thus, the food self reliance index can only becalculated for sixty-nine of the 75 countries.

2. liQ).J.13.ci'»l.d F0 0 d A~~.5.....l.ru:i.ex-U1Ail

Economic access to food (food access) is conceptually defined interms of incomes. employment and food prices. The HAl is one ofthree elemental indices that constitute the basis of this analysis.Because appropriate and comparable data for food prices andemployment are not available, the HAl is operationally defined interms of average income and food consumption levels and the rate ofchild mortality, where the latter is a proxy for the distribution ofincome and food consumption. Future evolution of this analysisshould attempt to incorporate food prices.

i) GNP, defined as per capita GNP (from the World Bank's WorldDevelopment Report);

ii) CALORIE, defined as the average daily calorie consumptionper capita (also from the World Development Report); and

iii) INFANTS, defined as the mortality rate per thousand forchildren under the age of five years (from UNICEF's State of theWorld's Children).

Although the average daily per capita consumption of food indicateswhich countries have the most severe hunger problem, the average percapita figure can mask differences among countries due to a more orless unequal distribution of food consumption across populationgroups. Per capita GNP is a reasonably good measure of averageaccess to food on the demand side, but again, does not provide anyindication of the distribution of income. The inability to includea direct measure of the distribution of income or food access is adefinite shortcoming because a more unequal distribution means thata larger segment of the population will be hungry for a givenaverage level of i~come or food cousumption. The under fivemortality rate attempts to get at the distribution question.Although mortality rates may be more sensitive to broadly definedhealth factors than to strictly nutritional causes, they may be abetter measure of access to food, income and health services by thepoorest groups than any other indicator with data avaliable for alarge cross-section of countries.

Data for per capita calorie intake and child mortality rates areavailable for all 75 countries included in the analysis, around theyear 1985. However, per capita GNP data are not available for threeimportant, or potentially important, A.I.D. countries: Afghanistan,

- rg -

Angola and L~banon. ~or the other 72 countries, per capita GNP isavailable through 1987. Thus, seventy-two of the 75 countriesincluded in the analysis have 1985 data for each of the threevariables that make up the food access index.

A country can improve its food security status by improving thatstatus over time, by "performing" well with regard to the indicatorsused to calculate food security status. The PI is one of threeelemental indices that constitute the basis of this analysis.Ideally, the Performance Index should measure growth or improvementin ~ of the five variables comprising the NRI and the HAT.

So far, the analysis does not include time series data for averagedaily calorie intake and child mortality. The PI explicityrepresents growth in only three of the five variables, growth of percapita domestic food production, foreign exchange earnings and GDP.GDP is used in the growth context, whereas GNP was used in theHousehold Food Access Index. Future evolution of the analysis willattempt to broaden the scope of the Performance Index. Growth datafor average calorie intake and child mortality, or some otherappropriate access variables, would allow the analysis todistinguish between performance in making more food avaialble andperformcnce in improving economic access to food.

Specifically, the three variables that comprise the PI include the:

i) FOODINDX, defined as the index of per capita domestic foodproduction, based on average annual food production (in metrictons of cereal equivalent) over the three-year period 1984-86compared to average production for the three-year period1979-81, taking account of 1980 and 1985 population estimates;

ii) FOREXIND, defined as the index of per capita foreignexchange earnings, based on average annual earnings over thethree-year period 1984-86 compared to average earnings for thethree-year period 1979-81; and

iii) GDPGRWTH, defined as the average annual growth rate of percapita GDP over the period 1980-85 (from the World DevelopmentReport) .

GDP growth rate data are available for the same 72 countries as isper capita GNP. Data on growth in domestic food production (theindex of domestic food production) are available for 74 countries,all but Cape Verde. However, data for the index of average percapita foreign exchange earnings, an indicator of the growth inforeign exchange earnings, are unavailable for a total of six

- 9 -

countries. Five countries (Afghanistan, Angola, Bhutan. Guinea andLebanon) lacked data on foreign exchange earnings for the recentperiod 1984-86. Those five countries also lack data for the earlierperiod 1979-81, as does Guinea-Bissau, bringing the total to sixcontries lacking data on growth in foreign exchange earnings. Thus,only 68 countries are included in the Performance Index.

4. Food Security Index (ESI)

A country is more food insecure to the extent that the country'seconomic system is unable to provide enough food to feed its people(the country is not food self relic.nt) and citizens of that countryhave inadequate economic access to the available food supplies. TheFSI is a composite.indeY obtained from the combined effect of twoelemental indice~ representing each of these concepts: the NationalFood Self Reliance Index (NRI) and the Household Food Access Index(HAl).

The food security index can be calculated from the five variablesthat make up these two indices in a number of ways; two areconsidered. The most conceptually neutral method is to weight eachcomponent index equally, where the ESI value for each country iscalculated as the sum of that country's rank or normalized "z value"for each component index. Alternatively, one could weight eachvariable equally, where ~he FSI value for each country is calculatedas the sum of that country's rank or normalized "z value" for eachcomponent index. The implication of the latter approach would be toassert that household economic access to food is 50% more importantthan national food self reliance when adding up a country's overallfood security.

Specifically, one can:

i) Compose a composite FSI by combining the two-variable NRl~lith the three-variable HAl; that is, weighting each componentindex equally. The equation for this option would be:

FSl = HAl + NRI.

ii) Compose a five-variable FSI by treating each of the fivefood access and food self reliance variables separately;weighting each variable equally. The equation for this optionwould be:

FSI = GNP + CALORIES + INFANTS + FOODPROD + EOREX

Sixty-nine of the 75 countries have data available for all fivevariables used to calculate the ESI. Eight food aid recipients(Gaza, West Bank, Kampuchea, Laos, St. Kitts, St. Nevis, andSeychelles) are excluded from the analysis altogether.

- /0-

C. FOOl) ~.\t:ClJh' i"y COlJNTRY HANKI NGS

Multivdriate statistical ranking techniques. required when using twoor more variables as ranking criteria, are employed to calculateeach of the four indices used in this analysis. Five different percapita variables representing status in 1985 (along with growth inthree of those status variables) are used to estimate relative foodinsecurity among countries. Such multivariate techniques cannot usenominal values of these variables in the direct calculation of anindex used to rank countries. These nominal values must first bestatistically manipulated before being used. Why is this so?

If nominal values of the variables are used to calculate the index,variables with larger absolute values and with wider ranges betweenlargest and smallest values will have more influence on the finalcountry ranking than variables with smaller absolute values andnarrower ranges. For example, if the Household Food Access Indexwere calculated by adding the nominal values of the variables foreach country, per capita GNP (ranging from 160 to 2500) would have astronger influence on the final ranking than would the under fivemortality rate (with a range from 40 to 330). It may well be thatone variable should be weighted more heavily than another. However,the analyst should be required to explicitly specify the relativeweighting, rather than allowing the way a variable is measuredimplicitly determine, from "behind the scenes," how each variablewill be weighted.

After all the data is collected, those countries that lack data forone or more of the five variables used to calculate the foodsecurity inde~ are excluded from the analysis. Six countries areeycluded from the food security ranking for lack of data, leaving 69countries included in each of the three related indices: HAl, NRIand FSI. One additional country is excluded from the performanceindex (leaving 68) because foreign exchange earnings data were notavailable back far enough to calculate change in foreign exchangeearnings since 1980. The included countries are then rankedaccording to several different combinations of variables, eachrepresenting a particular aspect of food security, or combination ofaspects.

Two sim?le multivariate statistical ranking techniques are availablethat avoid the problems caused by using nominal values of thevariables: the normalized value and the surn-of-in2ividual-rankstechniques. Each technique has advantages and disadvantagescompared to the other, but each is a statistically valid techniquefor calculating indices incorporat_ng two or more individualvariables.

-II -

Three elemental indices are calculated. based on valve~ of theindividual variables, using both of the above statisticaltechniques. The National Food Self Reliance Index is based on percapita measures of domestic food production and gross foreignexchange earnings and the Household Food Access Index is based onper capita GNP, per capita daily calorie intake and child mortalityrates while the Performance Index is based on growth in per capitaGDP, domestic food production and gross foreign exchange earnings.

2. The Sum-of-indiyidual-ranks Techn~

The first step in the sum-of-individual-ranks technique is to rankthe countries according to the nominal value of each individualvariable that comprise the index. In the case of the Food AccessIndex, for example, all countries are ranked according to each ofthree variables: per capita GNP, per capita daily calorie intake andmortality rate per 1000 infants and children under five years ofage. This generates three separate rank orderings of the includedcountries. The next step is to sum the ranks, by country, for eachof the three variables that comprise the index. After the ranks onindividual variables have been summed for each country, thecountries are reranked according to this sum of individual ranks.

The sum-of-individual-ranks technique eliminates the unduly largeimpact that extreme values have on the eventual ranking and has theadded advantage that it is easy to understand. A disadvantage isthat the relative difference between the nominal valuescorresponding to countries ranked above or below a given country,for each of the individual variables, is lost. Instead, therelative difference between the nominal values for two countries,for a given variable, is translated into an integer r0presenting thenumber of countries with intermediate nominal values for thatvariable. For example,

3. The Normalized Value Technique

The normalized value technique assumes that each of the variablescomprising the index follows a "normal" distribution. To the extentthis assumption is a reasonable approximation of fact, it justifiesthe calculation of a new set of. '.'normalized" values that replaceeach of the nominal values, using the following formula:

Xni = (Xi - mean of X)/standard deviation of X, whereXi = the nominal value of variable X for country i andXni = the normalized value of variable X for country i.

The normalized values corresponding to a given country are thensummed for each variable comprising the index and the countries areranked according to +~e sum of the normallzed values.

- 1d- -

Tb~ advantage of using normalized values (variable Xn) instead ofnominal values (variable X) is that the mean value Gf Xn is zero andthe standard deviation is 1. Thus. the absolute values and therange of values are approximately the same for each normalizedvariable. At the same time, normalized variables maintain the samerelative difference between values for any two countries as existswith the nominal values of ~he basic data.

4. Cal~ng a CQIDPQsite Index Using BQL~ Technigues

The Food Security Index can be calculated ~~ing either statisticaltechnique applied to the five variables separately: per capita GNP,domestic food production, gross foreign exchange earnings, dailycalorie intake and child mortality rates. Each individual variableis weighted equally in this calculation. Alternatively, the FoodSecurity Index can be calculated as a composite index using eitherstatistical technique applied to the two elemental indices thatcomprise it: the Naticnal Food Self Reliance Index and the HouseholdFood Access Index. This latter approach is conceptually closer tothe original definition of food security, as a combinaticn ofnational-level food self reliance and household-level access ·tofood; it takes both statistical ranking techniques one step further.

A composite index uses other indices to create a new index, ratherthan the individual varlables that make up both elemental indices.III order to calculate the so called double sum-of-individual-ranks(or "double soir") rank, the NRl and HAl ranks initially obtained byapplying the sum-of-individual-ranks method are, in turn, sumed.The new sum-of-individual-sums variable is then used to rank thecountries, under the name "double soir" rank. In order to calculatethe 50 called double normalized (or "double norm") rank, the twosums of individual normalized variables used as the basis forcalculating each of the elemental indices are themselvesnormalized. The values of these normalized sums, corresponding toeach country, are then added.

In the first case, the sum of two index ranks is used to calculatethe "double soir" rank. In the second case, the sum of twonormalized sums of normalized variahles is used to calculate the"double norm" rank. Each of these "double" ranks weights theelemental indices equally (rather than weighting the individualvariables equally) in calculating the composite index.

5. Results of the Statist~cal Ranking~

Country rankings were calculated for each of the th: 'ee elementalindices: National Food Self Reliance, Household Fooil Access andPerformance. Tables I (and IA), II and III lj.st the countries withadequate data (69 countries in the case of the NRI and HAl and 68countries in the case of the PI), in rank order according to thenormalized statistical rank technique. The sum-of-individual-ranksranking is indicated in the adjacent eolumn.

- /3 -,

'i'clb.le IV presents the results of t.he Food Security Index. which wascalculated four different methods. Two methods treat the nationalfood self reliance and household food access indices separately: thedouble normal ized sums (the "double norm") ranldng technique and thedouble sum-of-individual-rankc; (the "double soir") rankingtechnlque. The other two methods treat the five variablesseparately (rather than the indices), based on either the sum ofnormalized values or the sum of individual ranks. Table 4 indicatesthe country rank ordering according to each of th~=e fourtechniques, under the heading Composite Food Security Index. TableIV lists the 69 countries with adequate data in rank order accordingto a weighted sum of the four ranks; the "double norm" method isgiven double weight while the other three methods are given singleweight.

T~is rank ordering was used to identify a core group of the mostfood insecure countries and another core group of relatively foodsecure countries. The core groups were selected from the set of 69countries. The core group of the 32 most food insecure countries issupplemented with four of the six excluded countries that, byguesstimate, seem to belong with the core group of food insecurecountries. Similarly, the core group of relatively food securecountries consists of the 22 countries ranked highest on the foodsecurity index. An intermediate group of 15 borderline countriesare also identified, to which are added the two remainig excludedcountries. Some of these borderline countries may eventually beincluded among the more food insecure countries when a finaldecision on food aid eligibility is made. Figure 1 groups all 75countries according to relative food insecurity. The 69 countriesstatistically ranked are unmarked. The six countries lacking dataon one or more variable are placed in one of the three groups by"guesstimate" and are denoted with an asterisk (*).

D. ALLOCATING the P.L. 480 BUDGET to FOOD INSECURE COUNTRIES

1. The P.L. 480 Budget: Actual anQHypothetical Allocations

Actual FY 1988 P.L. 480 program budget allocations are compared witha hypothetical allocation scheme that directs more food aidresources to the most food insecure countries. The actual P.L. 480budget was defined to include all Title I and Title II regularcommodity and transport costs; but it excludes Title II emergencyand all Section 416 program costs. Title I and Title II regularcommodity plus transport costs totaled approximately $1.26 billionin FY 1988. By including commodity and transport costs of Title IIemergency programs, the FY 1988 total would have reached $1.4billion.

- /y -

The hypothetical allocation scheme excludes the relatively foodsecure country grouping (22 out of 75 countries), allocating theentire $1.26 billion on a per capita basis to the relatively foodinsecure and the borderline country groupings (a total of 53countries). The hypothetical allocation scheme allocates twice asmuch, on a per capita basis, to the 36 most food insecure countriesas it does to the 17 borderline countries.

2. Proposed Allocation of the Food Security Fund

For illustrative purposes, assume that the U.S. food aid program hasbeen divided into two funding "pools", a food security fund for foodinsecure countries and a market development fund for countries intransition from a dependent status as aid recipients to coequaltrade partner. A hypothetical $1 billion Food Security Fund,providing mainly grant food aid, would reserve $100 million forworldwide emergencies. Thus, the non-emergency commodity plustransport value of the Food Security Fund budget would beapproximately $900 million. Although the hypothetical Food SecurityFund would have significanatly less food aid resources at itsdisposal than does the present P.L 480 program ($1.1 billion incommodity costs plus another $300 million for transport costs in FY1988), it would be focused on fewer countries (the 53 relativelyfood insecure and borderline countries). The hypothetical andactual food aid allocation numbers are contained in Figure 2.

The same hypothetical allocation criterion is used to allocate FoodSecurity Fund resources and the resulting allocation is compared tothe actual FY 1988 allocation. A fourth column of adjustedhypothetical allocations represents expert judgements by FVA Bureaumanagement that take into account Regional Bureau decisions toexclude food aid from certain countries (e.g. Papua New Guinea,Nepal, Rwanda, Burundi, Central African Republic, Zimbabwe andCameroon) as well as political realities in Central and SouthAmerica that require larger food aid allocations than the foodsecurity rankings would indicate. As a result of these assumptions,in the aggregate, the relatively food insecure countries wouldreceive slightly more food aid under the hypothetical and adjustedallocations of the Food Security Fund than they actually received inFY 1988.

- 15-

It 1 ~3 T lil Vl~

11/8/e9li'igure 1

75 DEVELOPING COUNTRIESGROUPED ACCORDING TO RELATIVE FOOD SECURITY

MOST FOOD INSECURE COUNTRIES

MozambiqueEthiopiaChadSierra LeoneSomaliaMaliBangladeshSudanBurkina FasoHaitiMauritaniaGuinea Bissau

TanzaniaNigeriaSenegalPeruCameroonPapua New Guinea

Central African RepublicYemen Arab RepublicNepalRwandaGhanaZairePakistanBoliviaAfghanistan*Angola*Bhutan*Guinea*

BORDERLINE COUNTRIES

Sri Lanl{aEI SalvadorSwazilandCape Verde*Lebanon*

BurundiKenyaIndiaMalawiBeninGambiaLiberiaLesothoNigerZambiaUgandaTogo

HondurasMadagascarBotswanaEcuadorNicaraguaDominican Republic

RELATIVELY FOOD SECURE COUNTRIES

BurmaGuatemalaIndonesiaPhilippinesZimbabweIvory CoastCongo

ColombiaMauritiusMoroccoEgyptJamaicaAlgeriaTunisiaGuyana

JordanBrazilThailandCosta RicaParaguayPanamaMexico

NOTE: * Six countries lack data for one or more of the fivevariables used to rank countries according to food security.Therefore, the six countries marked with an (*) could not beincluded in the statistical ranking.

SOURCE: U.S. Agency for International Development, Bureau for Foodfor Peace and Private Voluntary Assistance.

FOODFUND10/21/89

Countries

Figure 2ALLOCATION OF $1 BILLION FOOD FUND

AND$450 MILLION EXPORT MARKET FUND

Excludes $100 Million Emergency Reserve$ millions

Real/Hypothetical Food Aid AllocationsPopu- Actual Food ** Expert

lation FY 1988 Security Opinion

FOOD FUND COUNTRIES (MOST FOOD INSECURE)

MozambiqueEthiopiaChadSierra LeoneSomaliaMaliBurkina FasoBangladeshMauritaniaHaitiSudanGuinea BissauCentral African RepublicNepalRwandaYemen Arab RepublicGhanaZaireBoliviaPakistanBurundiIndiaGambiaKenyaMalawiBeninNigerLiberiaZambiaLesothoUgandaTogoAfghanistan*Angola*Bhutan*Guinea*

Total Most Food Insecure

13.842.3

53.75.47.57.9

100.61.75.9

21.90.92.6

16.568

12.730.66.4

96.24.7

765.10.7

20.474

6.42.26.71.5

14.73

16.58.81.26.2

1264.1

9816

1814

776

61040

1131

10182347

1151

1235

11121

1010

754oo1

13

601

928

32455

6614

1412

11458

204

633

501o

13534141

102

11614

828

1515

56

151010

1106

1530

2

41018253585

2155

415

3

810

511

33010

113

686

FOODFUND10/21/89

Countries

B'igure 2ALLOCATION OF $1 BILLION FOOD FUND





BORDERLINE COUNTRIES (MODERATELY FOOD INSECURE)

Senegal 6.6 10 2Nigeria 99.7 0 33Tanzania 22.2 8 7Peru 18.6 40 6Cameroon 10.2 0 3Papua New Guinea 3.5 0 1Sri Lanka 15.8 28 5El Salvador 4.8 56 2Swaziland 0.8 1 0Honduras 4.4 22 1Madagascar 10.2 6 3Botswana 1.1 4 0Ecuador 9.4 2 3Dominican Republic 6.4 22 2Nicaragua 3.3 0 1Cape Verde* 0.3 3 0Lebanon* 2.7 1 1

Total Moderately Food Insecure 220.0 205 72

Total Food Fund Countries 1484.7 813 900

8401035

2245

120

63

20

31

214

900

FOODFUND10/21/89

Countries

Figure 2ALLOCATION OF $1 BILLION FOOD FUND





MARKET DEVELOPMENT COUNTRIES (RELATIVELY FOOD SECURE)

Indonesia 162.2 25 0Philippines 54.7 67 0Burma 36.9 0 0Guatemala 8 24 0Zimbabwe 8.4 0 0Ivory Coast 10.1 0 0Congo 1.9 0 0Morocco 21.9 55 0Colombia 28.4 0 0Egypt 48.5 181 0Mauritius 1 1 0Jamaica 2.2 37 0Tunisia 7.1 32 0Guyana 0.8 7 0Algeria 21.9 0 0Jordan 3.5 0 0Brazil 135.6 0 0Thailand 51. 7 0 0Costa Rica 2.6 11 0Paraguay 3.7 0 0Mexico 78.8 4 0Panama 2.2 0 0

Total Relatively Food Secure 692.1 444 0

Grand Total 2176.8 1257 900------------------

3070

25

160

1155

2403510

1

15

5

450

1350

NOTE: * Six countries lack data for one or more of the fivevariables used to rank countries according to food security.Therefore, the six countries market with an (*) could not beincluded in the statistical ranking. They are placed by"guesstimate" in the first and second country groupings.

** The column titled Food Security allocates twice the per capitadollar value of food aid from the $1 billion Food Fund to the mostfood insecure countries as it does to the moderately food insecure.Relatively food secure countries receive no Food Fund allocations.

SOURCE: Bureau for Food for Peace and Private Voluntary Assistance,Office of Program, Policy and Management, October 1989.

Table I.NATIONAL FOOD SELF RELIANCE

Ranking of Selected Developing Countriesfrom least to most self reliance

RELYlr1/7/89

food productionfore x earnings

COUNTRIES

PERCAPITAFOODPROD

kg/cap1984-86

9/

PERCAPITAFOREX

EARNINGS$/cap

1984-864/

NATIONALFOOD SELF

RELIANCE INDEXnorma- sum oflized indivvalue ranlts

Yemen Arab RepublicMauritaniaSomaliaMozambiqueEthiopiaLesothoSierra LeoneHaitiMauritiusPeruChadKenyaSudanSri LankaDominican RepublicBangl~desh

El SalvadorPakistanZambiaColombiaPapua New GuineaBoliviaGhanaJamaicaMaliLiberiaCentral African Rep.NicaraguaIndiaGuatemalaZaireGuinea BissauRwandaEcuadorBotswanaBurkina FasoHondurasNepalCameroonSwazilandEgyptNigeriaTogoAlgeriaPhilippinesBenin

6752

119125149101145141

15135179169177160128204151195178161140188204

54213170209200224191221232229160

26232190240181140183223224122229262

144232

341418

2214571

659202

278856

125252

17234

68137212308121

60660

37210

66112

21157

662033

316858

43217

21260432264108128572153

63

1

3-156789

10111213141516171819202122232425262728293031323334353637383940414243444546

7.514.5

213

16.54.51.5

29.516.54.5

10.59

12.525

637

20.523

33.5312319

40.512.533.5

2326.510.540.526.514.520.548.5

3128521853455437

48.542.555.5

46

Table INATIONAL FOOD SELF RELIANCE

Ranking of Selected Developing Countriesfrom least to most self reliance

RIU,Y 111/7/89

food productionforex earnings

COUNTRIES

PERCAPITAFOODPROD

kg/cap1984-86

9/

PERCAPITAFOREX

EARNINGS$/cap

1984-864/

NATIONALFOOD SELF

RELIANCE INDEXnorma- sum oflized indivvalue ranks

CongoTanzaniaMadagascarSenegalIndonesiaBurundiJordanTunisiaNigerMalawiUgandaMoroccoIvory CoastGambiaCosta RicaBurmaZimbabweBrazilGuyanaMexicoParaguayThailandPanama

131277281255263292

22194296303318277257307229380344334335340528525222

5902840

145120

281112

435484332

198328128531

11167209322367190212

2881

4748495051525354555657585960616263646566676869

48.529.542.5

5155.5

31375851

48.544606559

62.53261

62.56869656765

NOTES:4/ World Bank, World Tables, years 1987 & 1988-89.9/ IFPRI, FAD data from data tape, 1989.

SOURCE: Agency for International Development, Bureau for Food forPeace and Private Voluntary Assistance.

.RELY1A11/7/89

Table IANA1IONAL FOOD SELF RELIANCE

Ranking of Selected Developing Countriesfrom least to most. food self reliant

food product.ionadjusted ForEx

COUNTRIES

PERCAPITAFOODPROD

kg/cap1984-86

9/

ADJUSTEDPERCAPITA

FOREXEARNINGS

$/cap1984-864/ ,10/

ADJUSTEDNATIONALFOOD SELF

RELIANCE INDEXnorma- sum oflized indivvalue ranks

Yemen Arab RepublicMauritaniaMozambiqueSomaliaEthiopiaSierra LeoneLesothoHaitiMauritiusPeruKenyaChadDominican RepublicJamaicaSri Lank~.

SudanPapua New GuineaColombiaEl SalvadorBangladeshBoliviaPakistanZambiaGhanaAlgeriaMaliEcuadorGuatemalaLiberiaCentral African Rep.NicaraguaIndiaGuinea BissauZaireHondurasRwandaCameroonBurkina l!asoCongoEgyptNigeriaNepalTogoSwazilandBotswana

61.451.6

124.9119.3148.8145.1100.5141.314.6

134.9168.5119.3128.454.0

160.1171.1140.5161.2151.1203.1188.2195.2118.4204.3121.1212.6159.9191.4110.2'209.4199.7224·.3232.4220.8189.7229.3180.5231.6130.9182.6223.5239.8224.2139.926.5

134.0193.5

9.832.015.641.2

209.661.9

585.1155.3

61.725.5

214.8481.0109.6

49.6199.6153.5190.514.613.456.9

119.850.7

361.132.4

229.0122.4199.1

59.791.917.613.656.2

172.831.1

208.439.9

405.2219.9

78.120.094.2

403.3821.3

123456789

101112131415161718192021222324252627282930313233343536313839404142434445

913.5

1324

237

3218.5

105

3238.513.5

828.528.534.5

618.518.5

2118.538.5

1551.5

3648.524.5

321112

24.551.5

2253.528.518 5

5641164650

38.5

rmLY1A11/7/89

food productionadjusted ForEx

COUNTRIES

PhilippinesBeninTanzaniaTunisiaIndonesiaMadagascarSenegalBurundiJordanNigerMalawiIvory CoastMoroccoUgandaCosta RicaGambiaBrazilZimbabweBurmaMexicoGuyanaParaguayThailandPanama

Table IANATIONAL FOOD SELF RELIANCE

Ranking of Selected Developing Countriesfrom least to most food self reliant

ADJUSTED ADJUSTEDPERCAPITA PERCAPITA NATIONAL

FOOD FOREX FOOD SELFPROD EARNINGS RELIANCE INDEX

kg/cap $/cap norma- sum of1984-86 1984-86 lized indiv

9/ 4/,10/ value ranks------- ------- ----- -----

228.6 121.1 46 53.5261.5 51.4 47 47277.0 25.0 48 ~~

L.

193.8 326.7 49 58263.2 90.4 50 55280.7 31.6 51 43255.1 125.2 52 57292.4 23.0 53 28.5

21.7 1021.1 54 38.5296.3 30.7 55 43302.6 30.3 56 43256.5 202.0 57 62276.6 147.4 58 6031'7.6 27.9 59 45228.5 366.3 60 64306.8 120.3 61 59334.5 129.0 62 61344.0 130.4 63 63379.7 5.6 64 34.5339.7 186.4 65 67335.1 283.1 66 69528.1 142.7 67 66525.0 161. 2 68 68221.6 2666.4 69 65

NOTES:4/ World Bank, World Tables, years 1987 & 1988-89.9/ IFPRI, FAD data from data tape, 1989.10/ World Bank, World Debt Tables, 1989.

SOURCE: Agency for International Developmen~, Bureau for Food forPeace and Private Voluntary Assistance .

. .I]

Table II

ACCESS HOUSEHOLD FOOD ACCESS INDEX11/7/89 Ranking of Selected Developing Countries

from least to most access

HOUSEHOLDGNP CALORIES UNDER FOOD ACCESSPER PER FIVE INDEX

CAPITA CAPITA MORTAL norma- sum of$ per day per 1000 lized indiv

COUNTRIES 2/ 3/ 5/ value ranks--------- ----- ------- ------ ----- -----

Mali 150 1,788 302 1 2Chad 130 1,504 232 2 3Ethiopia 110 1,681 257 3 1Mozambique 160 1,6'78 252 4 4

Sierra Leone 350 1,817 302 5 7Burkina Faso 1;)0 1,924 245 6 5Gambia 230 2,~52 292 7 15Somalia 280 2,072 257 8 10Sudan 300 1,737 187 9 13Bangladesh 150 1,899 196 10 6Guinea Bissau 180 2,073 232 11 9Rwanda 280 1,919 214 12 12Central African Republic 260 2,050 232 13 11Nepal 160 2,034 206 14 8Haiti 310 1,855 180 15 18Malawi 170 2,448 275 16 17Ghana 380 1,747 153 17 23Niger 250 2,250 237 18 19Mauritania 420 2,078 223 19 20Burundi 230 2,116 200 20 14Uganda 230 2,083 178 21 16Benin 260 2,173 193 22 22Zaire 110 2,154 170 23 21Senegal 370 2,342 231 24 26Pakistan 380 2,159 174 25 28Bolivia 470 2,146 184 26 25Liberia 470 2,311 215 27 30India 270 2,189 158 28 27Yemen Arab Republic 550 ·2,250 210 29 31Togo 230 2,236 160 30 24Tanzania 290 2,335 183 31 29Zambia 390 2,137 135 32 33Nigeria 800 2,085 182 33 34Kenya 290 2,151 121 34 32Cameroon 810 2,089 162 35 35Zimbabwe 680 2,089 121 36 36Lesotho 470 2,358 144 37 37Swaziland 670 2,556 182 38 42Honduras 720 2,211 116 39 41Madagascar 240 2,469 97 40 38Papua New Guinea 680 2,181 94 41 44Ivory Coast 660 2,505 157 42 45Peru 1,010 2,171 133 43 40Burma 190 2,547 91 44 39El Salvador 820 2,148 91 45 46Indonesia , 530 2,533 126 46 47

ACCESS11/7/89

Table I IHOUSEHOLD FOOD ACCESS INDEX

Ranking of Selerted Developing Countriesfrom lea~t to most access

COUNTRIES

BotswanaPhilippinesEcuadorSri LankaMoroccoNicaraguaDominican RepublicGuyanaGuatemalaCongoThailandEgyptJamaicaTunisiaColombiaParaguayMauritiusBrazilCosta RicaPanamaJordanAlgeriaMexico

GNPPER

CAPITA$2/

840580

1,160380560110190500

1,2501,100

800610940

1,1901,320

8601,0901,6401,3002,1001,5602,5502,080

CALORIESPER

CAPITAper day

3/

2,2192,3412,0542,3852,6182,4252,4612,4512,?942,5492,4623,2632,5852,8362,5142,8132,1402,6572,8032,4192,9472,1993,126

UNDERFIVE

MORTALper 1000

5/

99789248

130104

8841

109122

55136

25110

12643291233565

11773

HOUSEHOLDFOOD ACCESS

INDEXnorma- sum oflized indivvalue ranks

47 4848 5149 4350 4951 5052 5253 5654 5555 5456 5157 5858 5359 6460 5961 6362 6563 6664 6265 6966 6167 6768 GO69 68

NOTES:Data for 1985. Pulled together by Resources for the Future in

"Malnutrition: Opportunities and Challengse for A.I.D.," 1988.2/ World Bank, World Development Report, years 1981 & 1988.3/ World Bank, Social Indicators of Development, 19875/ UNICEF, The State of the World's Children, 1987


GROWTH11/7/89

Table IIIPERFORMANCE INDEX

Ranking of 68 Developing Countriesfrom slowest to fastest growth

GDP rowth ratefood prod indexforex index

COUNTRIES

ANNUALPERCAPITA

GDPGROWTH

rate1980-852/ & 4/

PERCAPITA PERCAPITAFOOD FOREXPROD REVENUE

INDEX INDEX1984-86 1984-88

1979-81=1004/ 4/

PERFORMANCETNWU~

:>It,t}tlt;!- !=H·1fft (=:Ift,ieal iwjiV

rank ranks

MozambiqueNigerNicaraguaGuyanaNigeriaBoliviaTogoZambiaLiberiaGuatemalaMadagascarTanzaniaKenyaHondurasEl SalvadorRwandaSomaliaPeruSierra LeoneMalawiCentral African Rep.EthiopiaPhilippinesGhanaZimbabweIvory CoastCosta RicaLesothoZaireSudanPanamaEcuadorHaitiPapua New GuineaMaliSwazilandDominican RepublicJamaicaSenegalColombiaYemen Arab RepublicMauritaniaBurkina FasoBurundiBenin

-13.6-·6.7-3.1-7.3-7.3-7.0-5.6-4.1-6.4-4.3-6.1-3.1-1. 7-2.6-3.1-1.50.6

-4.2-0.2-0.6-1. 5-2.0-3.4-3.90.0

-5.2-2.73.4

-3.8-4.2-0.2-2.4-2.5-1. 6-3.0

2.4-0.8-3.10.0

-0.50.9

-0.7-1. 3-0.8

0.1

85.385.076.181.0

103.293.190.995.998.697.098.191.887.385.789.787.398.599.997.190.193.687.494.3

108.792.4

104.891.882.499.696.297.999.596.398.9

100.9103.7100.4103.0102.095.8

109.487.7

111.797.5

113.5

38.244.348.966.541.968.366.855.270.265.674.271.874.387.485.587.152.980.162.584.683.6

101.394.965.379.683.099.283.691.7

102.576.990.298.389.994.262.490.098.686.7

104.867.0

126.976.8

108.670.8

1234()

6789

101112131415161718192021222324252627282930313233343536373839404142434445

1

4::I

10657

128

139

111514162419172018262122232725323028293331343541393743384436404245

J..). ii',

,/ .lJ/

Table I I IGROWTH PERFORMANCE INDEX11/7/89 nanking of 68 Developing Count-ries

from ~lowest to fastest growth

GDP rowth rate ANNUAL PERCAPITA PERCAPITAfoodprod index PERCAPITA FOOD FOREXforex index GDP PROD REVENUE PERFORMANCE

GROHTH INDEX INDEX INDEXrate 1984-86 1984-86 statis- sum of

1980-85 1979-81=100 tical indivCOUNTRIEf. 2/ & 4/ 4/ 4/ rank ranks--------- ------- ------- ------- -----Mexico 0.8 96.6 107.6 46 46Paraguay 1.4 106.1 84.6 47 52Tunisia 1.4 107.6 83.2 48 49Algeria 4.9 100 ..6 78.9 49 47Botswana 7.4 76.2 120.3 50 48Sri Lanka 3.2 85.3 127.6 51 50Nepal 0.8 102.0 104.7 52 53Congo 4.9 92.9 102.3 53 51Morocco 0.1 108.6 99.1 54 54Mauritius 2.3 99.7 110.8 55 59Jordan 1.5 109.1 95.7 56 60Cameroon 4.5 94.4 113.2 57 57Egypt 1.3 104.8 108.9 58 61Bangladesh 0.9 97.7 129.1 59 55Brazil 1.3 106.4 108.9 60 62Indonesia 2.3 117.2 84.4 61 58Pakistan 2.8 103.6 117.8 62 64Burma 3.3 124,0 70.7 63 56India 3.1 111.7 101.1 64 65Chad 1.8 99.8 143.1 65 63Uganda 2.2 110.5 120.0 67 66Thailand 2.6 109.0 121.5 66 67Gambia 4.9 120.2 109.0 68 68--------------------

NOTES:Columns A, Band C provide data for 1985.2/ World Bank, World Development Report, 1982, 1986, 1987, 1988.4/ World Bank, World Tables, 1988-89.


SKCUHE~ Table IVLl /1 /89 rOO 0 S KCUB I TY ! M 0 KX'

Ranking of 69 Oeveloping Countriesfrom least to most food secure

GHP CALOBIKS OHDKR BOOSKBOLD PKBCAPITA NATIONAL COHPOSITK FOOD SKCOBITYPKR PKR FUK FOOD ACCKSS FOOD FORKX FOOD SKLF IHDK!

CAPITA CAPITA MORTAL IKDK! PROD KARHIHGS RKLIAKCK IHDKI lIeight lleight$ per day per 1000 nona- SUI of kglcap S/cap nona- SUI of indices equal variables equal

1985 1985 1985 lized indiv 1984-86 1984-86 lized indiv double SUI of norla- SUI ofCOUHTRIKS 21 31 51 value ranks 91 4/ value ranks non SUIS lized indiv---------Hozalbique 160 1678 252 4 4 125 14 4 1 1 1 1 1Ktbiopia 110 1661 257 3 1 149 18 5 3 2 2 2 2Chad 130 1504 232 2 3 179 27 11 5 4 3 3 3Sierra Leone 350 1811 302 5 7 145 45 7 4 3 5 4 6Soaalia 280 2072 257 8 10 119 34 3 2 5 4 6 4Kali 150 1188 302 1 2 213 37 25 12 6 6 5 5Bangladesh 150 1899 196 10 6 204 17 16 6 11 7 11 1Sudan 300 1131 181 9 13 111 56 13 9 10 9 9 11Burkina Vaso 150 1924 245 6 5 232 43 36 28 12 10 8 8Baiti 310 1855 180 15 16 141 71 8 6 8 12 10 12Mauritania 420 2016 223 19 20 52 232 2 15 1 15 1 15Guinea Bissau 160 20n 232 11 9 232 20 32 14 14 8 13 9Central African Rep. 260 2050 232 13 11 209 66 27 22 13 13 12 13Yelen Arab Republic 550 2250 219 29 31 61 144 1 1 9 11 15 20Nepal 160 2034 206 14 6 240 21 38 18 11 11 16 10Rllanda 280 1919 214 12 12 229 33 33 20 15 14 14 14Ghana 380 1141 153 11 23 204 60 23 19 16 18 11 11laire 110 2154 110 23 21 221 66 31 26 22 20 19 18Pakistan 380 2159 114 25 28 195 68 18 21 19 25 20 21Bolivia 410 2146 164 26 25 168 121 22 24 21 23 21 25Burundi 230 2116 200 20 14 292 28 52 31 29 16 23 16Kenya 290 2151 121 34 32 169 68 12 10 20 21 26 24India 210 2169 156 26 27 224 21 29 11 24 19 21 21Kalalli 110 2448 2'(5 16 17 303 43 56 49 30 24 22 22Benin 260 2113 193 22 2? 262 63 46 46 26 21 26 26Gaabia 230 2252 292 1 15 307 126 60 59 27 32 18 28Liberia 410 2311 215 21 30 170 210 26 33 23 30 25 32Lesotho 410 2356 144 37 3'{ 101 221 6 16 18 33 32 34Niger 250 2250 231 16 ·19 296 48 55 51 31 26 24 23Zaabia 390 2131 135 32 33 118 137 19 23 25 28 29 29Uganda 230 2083 118 21 16 318 32 51 44 31 22 33 19Togo ~30 2236 160 30 24 224 126 43 41 28 31 31 31Tanzania 290 2335 163 31 29 217 28 48 30 36 29 35 30Nigeria 800 2085 182 33 3. 223 106 42 36 33 34 34 33Senegal 370 2342 231 24 26 255 145 50 51 34 36 30 35Peru 1010 2171 133 43 40 135 202 10 17 32 35 31 36Calleroon 610 2069 162 35 35 181 260 39 53 35 39 36 37Papua Hew Guinea 660 2161 94 41 44 140 306 21 31 38 41 36 41Sri Lanka 360 2365 46 50 49 160 125 14 13 40 37 42 40HI Salvador 820 2148 91 45 46 151 234 11 39 39 43 40 44SlIalliland 670 2556 182 38 42 140 432 40 45 42 42 41 42Honduras 720 2211 116 39 41 190 217 31 52 41 47 39 45Madagascar 240 2469 97 40 38 281 40 49 42 47 40 44 39Botsllana 640 2219 99 41 48 26 856 35 35 44 45 43 46Kcuador 1160 2054 92 49 43 160 316 34 50 46 45 45 43Nicaragua 710 2425 104 52 52 200 112 28 21 45 44 41 41

fl

SECURE5 Table IV11/7/89 F0 0 D SEC URI TY I HDKX

Ranking of 69 Developing Countriesfrol least to most food secure

GNP CALORIKS UNDKR HOUSHHOLD PHRCAPITA NATIONAL COMPOSITE FOOD SECURITY~HR PHR FIVE FOOD ACCESS FOOD FOREI FOOD SKLF INDKI

CAPITA CAPITA MORTAL INDEI PROD KARNINGS RELIANCK IHDU lIeight lIelght$ per day per 1000 nona- SUI of kg/cap S/cap noru- SUI of indices equal variables equal

1985 1985 1985 lized Indlv 1984-86 1984-86 liaed Indiv double SUI of norla- SUI ofCOUHTRIKS 2/ 3/ 5/ value ranks 9/ 4/ value ranks nOli SUI!S lhed Indlv---------DOllnican Republic 790 2461 88 53 56 128 2~2 15 25 43 48 46 50Buna 190 2547 91 44 39 380 11 62 32 59 38 54 36Guatelala 1250 2294 109 55 54 191 157 30 41 48 49 52 51Indonesia 530 2533 126 46 41 263 120 51 56 50 50 49 49Philippines 580 2341 78 48 51 229 153 45 55 49 51 48 52Zilbablle 680 2089 121 35 36 344 167 63 61 56 52 50 48Ivory Coast 660 2505 157 42 45 257 328 59 64 52 60 51 56Congo 1100 ~549 122 56 57 131 590 47 48 55 53 55 54Cololbia 1320 2574 12 61 63 161 212 20 34 53 56 57 58Mauritius 1090 2740 32 63 66 15 659 9 29 51 57 59 59Morocco 560 2678 130 51 50 277 198 58 60 51 58 53 55Egypt 610 3263 136 58 53 183 264 41 54 58 54 58 53Jalaica 940 2585 25 59 64 54 660 24 40 54 59 56 60Algeria 2550 2799 117 68 60 122 572 H 43 63 55 66 57Tunisia 1190 2836 110 60 59 1S4 435 54 58 60 62 61 62Uuyana 500 2451 41 5: 55 335 322 65 68 61 63 60 63Jordan 1560 2941 65 67 67 22 1112 53 38 62 61 63 61Brazil 1640 2657 91 64 62 334 209 64 62 64 65 62 65Thailand 800 2462 55 51 58 525 212 68 67 66 64 64 64Costa Rica 1300 2803 23 65 69 229 531 61 63 65 68 65 68Paraguay 860 2873 64 62 65 528 190 67 65 68 67 67 67Panala 2100 2419 35 66 61 222 2881 69 66 69 66 69 66Mexico 2080 3126 73 69 68 340 361 66 69 67 69 68 69--------------------

HOrKS:Data for 1985. Pulled together in Resources for the Future in "Malnutrition: Opportunities and Challenges for A.I.D.,· 1988.2/ World Bank, World Developaent Report, years 1987 &1988.3/ World Bank, Social Indicators of Developlent, 19874/ World Bank, World Tables, 1988-89.5/ UNICEF, The State of the World's Children, 19819/ IFPRI, FAO data frol data tape, 1989.

SOURCH: Agency for International Development, Bureau for Food for Peace and Private Voluntary Assistance.

I

Documents

A METHODOLOGY FOR RANKING COUNTRIES ACCORDING TO RELATIVE FOOD