Tim Miller > Curriculum Vitae > Malnutrition and mortality among Bolivian children > Chapter Two

Bolivia is one of the Andean nations of South America. (See map in Figure 2.1). Many of its current boundaries have been shaped by military losses. It has been landlocked since the War of the Pacific when in 1880 it lost its coast to Chile. The Bolivian Military still maintains a naval force on Lake Titicaca -- the largest lake in South America. In the North, its military was defeated in a rebellion of rubber tappers (mostly Brazilian migrants supported covertly by the Brazilian government) resulting in annexation of the Acre region to Brazil in 1903. In the southeast, Bolivia fought and lost a disastrous war with Paraguay over the Chaco region. The war ended in 1937 with the loss of the entire region to Paraguay.

The defining feature of Bolivian geography is altitude. The country can be divided into three large regions: the Altiplano, the highland valleys, and the tropical lowlands. The Altiplano is an elevated plateau averaging between 3,500 and 4,200 meters in elevation framed by the eastern and western mountain ranges of the Andes. Subsistence farming on the Altiplano consists of crops consumed on the farm and livestock raised for off-farm sales. The main crops grown are potatoes, quinua (a cereal grain), broad bean, and barley.

Livestock consists of llamas, sheep, cattle, pigs, and chickens. In addition to agriculture, there are extensive mining operations in the eastern ranges of the Andes. Mining of Bolivia's extensive silver deposits dates back to the Spanish conquest. Bolivia is rich in a variety of other metals, chiefly tin, and various rare metals such as bismuth, zinc, and antimony.

Farther to the east are the highland valleys. The valleys are mainly temperate climates in which the main crops are corn, potatoes, and other vegetables. There are several large corporate farms involved in dairy and poultry production.

There are two important humid tropical valleys in Bolivia: the Chapare region in Cochabamba and the Yungas region in La Paz. The major crop in both regions is coca. Coffee and various tropical fruits such as bananas, yuccas, and citrus fruits are also grown.

Still farther east are the tropical lowlands. Agriculture there is a mixture of slash-and-burn technology for small farmers and mechanized production on large commercial farms. The main crops are soybeans, maize, cotton, sugar cane, and wheat. Various tropical fruits are also grown such as citrus, mango, banana, and pineapple. In the northern Beni, cattle production is the main agricultural activity.

Throughout its history the Bolivian economy has been characterized by its agricultural and mining activities. The focus of its monetized economy has been on the export of raw materials, usually dominated by a single commodity. This resulted in a small elite export sector with minimal domestic linkages and strong international dependency for both sales and capital. From the time of the Spanish conquest until the beginning of this century, silver had been the main export.

Tin, a by-product of the mining of silver in Bolivia, began to take on an increasingly important role in export. By the 1930's tin accounted for 50% of export. Bolivian economic development focused almost exclusively on the mining of tin and its export. The collapse of world tin prices contributed to a major crisis of the Bolivian economy in the early 1980s.

In the 1980s, demand for cocaine in developed countries stimulated production of the coca crop in Bolivia. For the first time in Bolivian history, agricultural products dominated mining as the chief export. Growing coca is a legal enterprise in Bolivia. Coca is a traditional staple of Bolivian food culture. Chewing coca leaves relieves hunger and staves off fatigue; in addition, it is used in a variety of herbal treatments. Coca has much the same effects, usage, and acceptance as caffeine (coffee) in American culture. Cocaine and coca paste are illegal and are a major source of power and foreign revenue in Bolivia. It is difficult to estimate how much money this export represents, for obvious reasons. One estimate placed this in the range of $2.5 billion in 1986 -- making it almost as large as the entire formal sector of the Bolivian economy.

The dependency of the Bolivian economy on external economies is also seen in its dependency on foreign capital. In the 1970s Bolivia -- like other developing nations -- received large foreign loans for development projects. By 1986 Bolivian public debt stood at nearly $3 billion dollars -- owed mainly to private U.S. banks. A large drop off in foreign loans in the 1980s produced various supply shocks in the Bolivian economy which led to the worst economic crisis in Bolivian history.

There are two important events in the recent economic history of Bolivia. The most profound change in the Bolivian economy took place in the 1952 Revolution. The 1951 elections had been won by Víctor Paz Estenssoro and the Nationalist Revolutionary Movement (MNR). The military moved to annul the elections and assume control of the government. The MNR initiated an armed conflict to gain control of the government. The armed rebellion of miners and farmers which eventually led in 1952 to the installation of the Paz Estenssoro government was a violent turning point in Bolivian history, which has been compared to "The Great Fear" peasant movement of the French Revolution. The peasants seized the land of the hacienda system, resulting in a de facto land redistribution. Meanwhile, in the urban sector, the government moved to nationalize the mines. This initiated a period of heavy government involvement in the economy which was brought to an end in the early 1980s.

The crisis of the 1980s began with the decline in foreign lending which led to falls in imports causing various supply shocks in the Bolivian economy. The world price of tin began to drop, leading to a precipitous fall in export earnings. The government made up for the shortfall by printing money -- touching off a hyperinflation. During the first 9 months of 1985, inflation in Bolivia was running at an annualized rate of 28,000%. The Bolivian peso virtually ceased to function as a means of exchange.

Paz Estenssoro re-entered the political life of Bolivia in August 1985, elected President in the first peaceful exchange of power between opposite political parties in Bolivian history. The government adopted an austerity plan consisting of a freeze of public sector wages, a huge increase in public sector prices, and a free-floating exchange rate. Paz Estenssoro, who had presided over the nationalization of the Bolivian economy following the 1952 Revolution, now set the economy on a new course. An IMF Stabilization Plan was adopted which called for significantly reducing the role of government in the national economy by privatization of state holdings -- mainly through sale to foreign multinationals.

The military has played a strong role in Bolivian society. There have been 189 coups in the 168 years since Bolivian independence. The various military juntas have ranged the political spectrum from reactionary to radical. The chief attribute of the government as a social institution has been its instability.

The defining moment of recent Bolivian history was the 1952 Revolution. Economically, the revolution resulted in the state's seizure of the mines and a huge land redistribution. Politically, it resulted in the enfranchisement of the peasants -- when the literacy voting requirement was eliminated. This event marked the beginnings of self-government and the entrance of most Bolivians into the political and economic life of their country. At the local level, worker organizations such as the unions in the mines and the sindicatos in the rural areas began to play an important role in Bolivian life.

The expansion of the educational system in Bolivia is arguably the chief lasting legacy of the Revolution. Today it is estimated that 94% of the population aged 5 to 14 in urban areas and 55% in rural areas are enrolled in school. While the quantitative change is evident, unfortunately we have no way of assessing the quality of these schools. In many rural areas, the quality may be assumed to be quite low -- often monolingual Quechua or Aymara students have their first contact with the Spanish language in schools. Thus, the first years are spent struggling to learn Spanish.

Bolivia is a rarity among the nations of South America in that the majority of the population is indigenous. It has been estimated that at contact between the 15th century worlds of America and Europe there was a population of approximately 50 million in the South American continent. Owing mainly to disease epidemics of European origin, most indigenous populations were decimated. There remain three concentrations of indigenous peoples in the Americas, centered near the locations of the three great civilizations of the Americas: the Aztec in Mexico, the Mayan in Central America, and the Incan in the Andean nations. Current estimates of the indigenous population of Latin America indicate a population recovery during the last 500 years to the initial level of 50 million people.

There are important problems in defining a post-contact indigenous population. The processes of intermarriage and socialization blur distinctions among peoples. As noted by CELADE researchers in their study of indigenous peoples of Latin America: "The main problem is the proper definition of what is considered an indigenous population, both as a concept as well as an operational indicator for identifying them in censuses and surveys." The principal definition chosen in their study and in this dissertation is the language spoken. The indigenous population is defined as comprising anyone who speaks an indigenous language (i.e., monolingual and bilingual speakers).

Figure 2.2 presents an age pyramid for Quechua speakers based upon the 1988 National Population and Housing Survey (ENPV). When this population is divided into monolingual and bilingual groups (Figures 2.3 and 2.4), several interesting patterns emerge. In Figure 2.3 one first notes the large number of monolingual speakers aged 5-9 compared to those aged 10-14. This difference reflects the effect of schooling as monolingual children become exposed to Spanish in schools. Second, there are many more female than male monolingual speakers. This asymmetry reflects differences in education as boys are more likely than girls to attend schools. Third, there is a strong age pattern -- with increasing numbers of monolingual speakers in the older age ranges. This reflects the expansion of the educational system over time.

According to the results of the 1976 census and 1988 population survey, a phenomenal growth occurred among monolingual Spanish speakers who recorded an increase of 921,000 to their numbers -- growing at an annual rate of 4.3%. But, the number of monolingual Quechua speakers declined by 237,000 and monolingual Aymara speakers declined by 160,000. The number of bilingual Quechua grew by 413,000 -- for a net increase in Quechua speakers of 176,000 and an implied annual growth rate of 1.0%. The number of bilingual Aymara rose by 296,000 -- for a net increase of 137,000 people and an implied annual growth rate of 1.1%.

These differences in growth are probably not due to differences in mortality and fertility between these language groups. Rather, they are reflections of a process of socialization -- closely associated with the expansion of schools in Bolivia. This has caused a shift from monolingual indigenous speakers to bilingual speakers as children from monolingual indigenous homes become bilingual at school. If this trend continues, then for the first time in Bolivian history more Bolivians will be speaking Spanish than an indigenous tongue.

According to the 1988 Population and Housing Survey, there are approximately 6.7 million Bolivians. The population has doubled in the 38 years since the census of 1950 which recorded a population of 2.7 million. The population is currently growing at a rate of nearly 2.4% implying another doubling in size in the next 30 years. Accompanying this rapid growth is rapid urbanization. For the first time in Bolivia's history, more Bolivians live in cities (51%) than in rural areas. This urbanization has been concentrated in six urban centers, in which a third of the entire Bolivian population lives.

Bolivia is at the initial stages of a demographic transition characterized by a reduction in fertility and mortality. Fertility is falling much more slowly than mortality resulting the rapid population growth noted above. The TFR currently stands at 5.8 births per woman, having declined from 6.5 in 1976 and 6.8 in 1950. Mortality rates have been falling as indicated by the rise in the expectation of life at birth from 40 years in 1950 to 56 years in 1988.

The DHS Program provides the largest global database of demographic and health information in the world. Over 50 national surveys have been conducted. Sponsored by AID as a follow-up to the World Fertility Survey, this program is aimed at providing information about population and health based on interviews with nationally-representative samples of women aged 15-49. Comparability among results is assured through administration of a set of standardized surveys in all countries.

In Bolivia, the DHS was administered by the Instituto Nacional de Estadística, the chief statistical agency in the Bolivian government. The preliminary interview sites were chosen based upon the National Population and Housing Survey which was carried out a year earlier. Interviews with a total of 8,538 women between the ages of 15 and 49 were located in an initial household survey of these sites. Of these, 7,923 women consented and completed the full interviews. The two main reasons for non-response were absence of the woman from the household on the survey date (304 women) and refusal to be interviewed (107 women).

Those women who had given birth (5,542) provided complete birth histories on a total of 22,338 births. Questions about prenatal care and delivery assistance were asked of those who gave birth in the last five years. A total of 3,681 mothers gave information on 5,804 births. These records are used as the basis for the analysis of neonatal and postneonatal mortality presented in Chapter Five. Children aged 3 to 36 months at the time of the survey were measured (height and weight). There were 3,085 children in this age range. Of these, 403 were not measured and 70 had measurements judged to be grossly inaccurate. This yields a sample of 2,612 children which forms the basis for the analysis of child stunting (low height-for-age) presented in Chapter Four.

There are four major advantages to large-scale surveys such as the DHS: standardization of approach, cost-effectiveness, large sample size combined with wide coverage of the population, and statistical reliability (Assogba, Campbell, and Hill; 1991). These surveys have permitted cross-national comparisons of fertility and mortality. These are necessary for program assessment by international aid agencies which must decide where to invest their efforts. They are also important in revealing worldwide patterns in mortality and fertility.

There are four major limitations of these DHS Surveys. First, the standardization of surveys means that country-specific factors are ignored. Many of the factors which make Bolivia unique -- information on ethnicity, altitude, or indigenous herbal medicine -- are not measured or, at best, poorly measured in the Bolivian DHS.

Second, the use of retrospective data introduces several problems. One problem is the mixing of retrospective data on fertility and health outcomes with current period data on household characteristics. Most analyses assume, with little empirical support, that household characteristics have been unchanging over time. Another set of problems with retrospective problems surrounds recall error, such as forgetting events and misstating the time of events.

The third limitation concerns the use of cross-sectional data. It is difficult to study a longitudinal problem using cross-sectional data. Demographers generally use a synthetic cohort assumption to make inferences about longitudinal data based upon cross-sectional data (e.g. period life tables). In studying health outcomes such as mortality or malnutrition which are the results of a series of events, there are serious selection bias problems which are likely to invalidate the synthetic cohort assumption.

Fourth and most importantly, there are major problems concerning the validity of large scale surveys. Surveys can be judged according to two criteria: reliability and validity (Scrimshaw, 1991). Reliability refers to how consistently an instrument measures something. A highly reliable instrument will consistently yield the same answer. The DHS survey has high reliability as evidenced by its large sample size, random sampling, and standardized questionnaires. Validity refers to the accuracy of an instrument. A highly accurate instrument will measure what you claim to be measuring. Measures in the DHS survey may have very low validity. The greatest danger of the DHS surveys is that they may be reliably giving us the wrong answers.

Based on anthropological accounts, one would assume that traditional healing rather than modern medicine is the primary response of most Bolivians to child sickness. However, one would not reach this conclusion based upon the Bolivian DHS. First, there are not many questions about traditional treatments. For example, there are no questions about the use of mates (herbal teas). Second, the questions are often biased. For example, mothers were asked: "Did you bring this child to a hospital, a health center, a clinic, or some other place so that they could treat him when he had this cough?" If she answered yes, she was then asked where the child was brought. Among the possible responses was the choice: "traditional healer." By focusing the question on places (like clinics and hospitals), it is difficult to elicit a response of "traditional healer" -- who is a person and not a place. Given this type of question, it is not surprising to find that despite Bolivia's strong indigenous medical tradition, less than 2% of respondents indicated "traditional healer."

In the three analytical chapters that follow, the DHS surveys will be analyzed in order to understand malnutrition and mortality among Bolivian children in the late 1980s and changes in their mortality risks before this period. In doing so, we must bear in mind the limitations imposed on the analysis as outline above.