Natural Hazards Research and Applications Information Center Institute
of Behavioral Science University of Colorado

Special Publication 36

The structure of this study is as follows: After framing the background and then dealing with the lessons learned/not learned from the 1982-1983 ENSO, we focus on what happened in the three Andean countries between roughly June 1997 and the end of August 1998-and why. The final section discusses policy and program implications, and a post-script brings the analysis to December 1999. Two major findings, however, should be noted at the out-set; multiple field visits to the three countries between 1997 and 1999 revealed an interesting pair of commonalties:

1. It took international news media reporting to stimulate/convince the governments of Ecuador, Peru, and Bolivia to attend to the implications of the developing 1997-1998 ENSO. That is, it took "bounce back" foreign coverage to put the ENSO threat on the domestic policy agendas in the three countries and to induce the governments to pay more attention to their own scientific communities.

2. As 1997-1998 ENSO impacts deepened in all three countries, each government set up new response organizations and structures-in some cases several in sequence-that sidelined and effectively demoralized the organization (in all three cases, the national civil defense office) previously and supposedly charged with disaster response. Several reasons were adduced for the marginalization of civil defense in the three countries, both for and off the record, and they are discussed in greater depth below. Nonetheless, despite differences in detail, the overall governmental-institutional response pattern in the three countries was consistent, clear, and striking.

It will hardly come as a surprise, but the various governmental responses to the 1997-1998 ENSO were influenced and complicated by a host of socioeconomic and, especially, political factors. In two cases, the 1997-1998 ENSO became part of electoral campaigns, one official (Ecuador), the other unofficial (Peru). In the third case (Bolivia), ENSO impacts and responses were influenced by a transition of administrations and a certain amount of political maneuvering common to coalition governments where ministries are divided among various political parties and factions. Therefore, two assumptions underlying this analysis should be made explicit at the outset:

1. Institutional readiness to deal with disaster on the part of any government is not primarily a technical or administrative issue. Rather, it is a political and policy decision be-cause it authoritatively allocates scarce resources (personnel, budget, access, public profile) among competing agencies, priorities, and values.

2. Disasters themselves are innately political events because they place enormous demand and decision-making stresses on governments that rather suddenly find themselves in situations of fluctuating resources. Some resources are lost in disaster events, but others are freed up or derive from external donors, which generates both cooperation and conflict a) within host governments, b) between host and donor governments, and c) among host governments, non-governmental organizations (NGOs), international organizations, and the private sector. The types and levels of cooperation and conflict are deter-mined by disaster needs and the socioeconomic and political context of the country at the time of impact.

The weather phenomenon known as "El Niño" (ENSO) is certainly not new (Spanish explorers reported the occurrence in 1525-1526), but after wreaking its destruction relatively quietly for hundreds of years, it has become suddenly famous in the last few years. Indeed, during the latter months of 1997 and especially in the first quarter of 1998, it was difficult to find a U.S. news network that did not have an ENSO report or even a feature article five to six times a week. "El Niño" became a household word in much of the Western Hemisphere. Of course, the populations of nations (especially Ecuador, Peru, and Bolivia) that have historically borne the more noticeable brunt of significant ENSOs had been very familiar with the event for decades-but not explicitly as a problem for public policy.

Deceptively mild sounding, an ENSO comprises the appearance of unusually warm sea surface water in the central and eastern Pacific Ocean near the equator. ENSOs can range from very weak (1-2 degrees centigrade above normal) to very strong (7 or more degrees centigrade above normal). An ENSO occurs every half decade or so, but most are weak to moderate (sea surface water 2-3 degrees centigrade above normal) and have only local effects. A major ENSO, however, changes weather patterns globally and creates "anomalies" (storms, floods, droughts) not only in the Americas (including the Andean countries and Brazil, Paraguay, and Argentina) but also, through atmospheric "teleconnections," in Africa, Asia, and even Australia (see Glantz, 1996, for a full treatment).

The only certain silver lining to an ENSO is that it dampens the Atlantic hurricane season, but ENSO impacts in some countries also appear to affect world commodity markets and therefore benefit specific economic sectors in other countries. Finally, some agricultural sectors (e.g., prawn production in Ecuador) benefit from ENSO-induced weather changes, as does the construction industry in affected countries, especially during reconstruction. Nonetheless, all other ENSO effects are in-variably negative and qualify major ENSOs as "disasters." For our purposes, conceptualizing ENSO events as disasters allows them to be analyzed in a particularly useful way.

Extending Fritz's classic (1961) definition, Kreps (1989, p. 219) has defined disasters as non-routine events in which societies or their larger subsystems (e.g., regions, communities) are socially disrupted and physically harmed. The key defining characteristics of such events are 1) length of forewarning, 2) magnitude of impact, 3) scope of impact, and 4) duration of impact. [original emphasis]

As noted above, until media attention to their truly global impacts really took hold in 1997-1998, ENSOs were relatively unknown except by 1) atmospheric and related field scientists, and 2) the populations of the most dramatically and immediately affected nations, especially the Andean countries. The rest of the world remained largely ignorant of the nature and global reach of ENSOs. Indeed, in the only book-length and "user-friendly" treatment of the phenomenon, Glantz (1996, p. 2) points out quite emphatically that the underlying purpose and theme for his book is that, given that ENSOs are globally important, "we, as members of different societies, need to have more than passing, intermittent interest in it, limited for the most part to when it occurs every few years or so."

Taking the Kreps definition of disaster and combining it with what we know about ENSOs in general, we can state with confidence that 1) although appearing periodically and with relative predictability, ENSOs are "nonroutine" because they vary tremendously across time, and 2) historically, very strong ENSOs have been enormously destructive. For analyzing ENSOs, however, the more interesting aspects of the Kreps definition of disaster lie in the four, more specific, "defining characteristics" of disaster: length of forewarning, magnitude of impact, scope of impact, and duration of impact.

Various monitoring technologies now provide literally months (often up to six months) of notice that an ENSO is building. In that sense, ENSOs are slow-onset disasters, and we generally know the regions of greatest impact. The problem is that scientists necessarily work at a level of abstraction, but people in potentially affected areas have more concrete concerns, as Betsill, Glantz, and Crandall (1997, p. 7) point out:

While scientists have been most concerned with determining the probability of an ENSO event, people living in the affected area are far more interested in knowing that such an event has actually begun. It is this in-formation, not necessarily a statistically valid forecast based on probabilities, that will en-courage them to prepare for the likely consequences.

A general forewarning is the easy part because, as noted above, ENSOs actually occur on average every four to five years with a range of between two and ten years. Major events occur on average every five to seven years. For practical and policymaking purposes and recalling the Kreps definition, therefore, it is ENSO variance along the dimensions of magnitude, scope, and duration that is important.

A major (strong to very strong) ENSO event produces severe weather anomalies whose costs are ecological (often affecting entire species), human, and economic. Countries dependent on agriculture and/or fisheries (prime examples are Ecuador, Peru, and Bolivia) are especially vulnerable to ENSOs, as Golnaraghi and Kaul (1995, p. 16) note in their treatment of the phenomenon:

Societies whose wealth is derived primarily from agricultural and hydrological resources are particularly vulnerable to dramatic inter-annual climate disruptions.

The significant problem, of course, is that population and economic growth in many countries has placed increasing numbers of people and investment in vulnerable areas, essentially in harm's way, so the magnitudes of ENSO impacts are increasing. While national economies at the aggregate level are in some ways now better able to absorb ENSO losses (due to higher levels of development compared to 10-15 years ago), it is a fact that economic development has been highly uneven, and traditionally disadvantaged parts of the population are, in fact, worse off. Indeed, it is now almost a cliché to say that "no one eats the GDP [Gross Domestic Product], just their share," and the problem is exacerbated by the fact that most nation states, especially in the Third World, have drastically reduced social spending in the last decade. Whatever safety nets that may have existed are now pretty much gone.

Therefore, because 1) economic development has been highly skewed and favorable to only certain parts of the population, 2) state capabilities have been reduced, 3) population and urbanization have increased, and 4) hazard mitigation has not been high on the public policy agendas of most countries, ENSO events of the same size are potentially more destructive-in absolute terms-than they were a century ago.

An ENSO's scope of impact must be understood in two senses-geographic and sectoral. Geographically, the impact area for a significant ENSO event is huge. While most attention is focused on the impact areas of the eastern Pacific (especially the west coast of South America and, increasingly, the United States), the atmospheric teleconnections of a strong to very strong ENSO extend its impacts globally. Again, the problem is level of abstraction. The scientific community tends to focus on regional impacts, but it is the specific local effects that can be so devastating-and local effects are much less predictable from one ENSO to another. For example, a valley unscathed in one ENSO might experience severe flooding in the next, a decade later.

Sectorally, ENSOs can have devastating impacts on agriculture, fisheries, transportation, finance, housing, and a host of other areas. Indeed, in Andean countries experiencing a major ENSO event, it is difficult to find an economic sector that is not affected to one degree or another.

In sum, it is possible to argue that ENSOs are so complex that they really should not be defined as "events." It is probably more correct to say that major ENSOs are temporary climatic conditions that create a multiplicity of local effects that extend, however, in some cases over entire nations. Moreover, because of their varied and multiple effects, ENSOs reveal and deepen weaknesses in economies, societies, and governments. With governmental structures especially, and at the risk of anthropomorphism, ENSOs seem almost to home in on the interstices between national, provincial, and local governments (exposing and exacerbating vertical coordination problems) and between ministries and offices at the national level (ex-posing and exacerbating horizontal coordination problems).

ENSOs build, peak, and decline over many months, and they bridge calendar years, so their duration of impact is often discussed in two-year cycles, especially when they are strong to very strong (e.g., 1972-1973, 1982-1983, 1997-1998). The problem with identifying them in that fashion, however, is a tendency to underestimate the fact that ENSO impacts can be very long lasting. Infrastructure damage in Peru, for example, may actually run from one strong ENSO event through several smaller ones. The same could be said of ENSO-induced drought in the high valleys of Bolivia, where it may take several years for agriculture to recover.

Second, because of the wide scope of ENSO impacts, especially sectoral impacts, mitigating an ENSO requires an inclusive and highly coordinated approach involving multiple government agencies, the private sector, and national and international intergovernmental and nongovernmental organizations (NGO coordination is an entirely separate issue; see McEntire [1998] on Peru). That is, managing ENSOs requires the same approach (ideally) as managing development in general. While economic growth/prosperity may ameliorate ENSO impacts, or at least facilitate recovery at the aggregate level, only a highly coordinated international, governmental, and civil society approach will provide some protection for the poorest and most vulnerable of a nation's population.

Third, and most importantly, because of their recurrent, if somewhat unpredictable, nature and the increasing length of forewarning provided by new technologies, ENSOs are prime candidates for systematic mitigation. After all, compared to major earthquakes, which often have recurrence intervals measured in hundreds of years, major ENSOs recur every decade or so. While that still exceeds the life expectancy of most governments, a typical adult might expect to see three or four major ENSOs during his/ her lifetime and will probably experience at least some ENSO impacts. Therefore it bears repeating that, at least for the Andean countries, time is either "ENSO" or "inter-ENSO."

Institutional development and capability to deal with disasters in the three Andean countries also showed no serious learning from the 1982-1983 ENSO. While civil defense organizations and systems were strengthened a tiny bit, they were given no mitigation mandates. That is, they were starved for resources and remained low-profile response systems with virtually zero preevent coordinating or policy influencing capabilities.

The greatest learning from the 1982-1983 ENSO occurred in the scientific communities, but more in Peru than in either Ecuador or Bolivia. Referring to Peru, Golnaraghi and Kaul (1995, p. 41) noted:

The estimated loss of $2 billion in damages to agriculture, industry, transportation, and other economic sectors during the 1982-1983 event led the government to set up a network incorporating scientific expertise into the policy process.

Field observations during the 1997-1998 ENSO, however, revealed a somewhat different picture, where the lack of permanent institutional channels (and scientific disagreements in early 1997) retarded the flow of in-formation to key ministries and the presidency. One critic noted that "they [the government of Peru] only listened to part of what we were saying," which may explain one of the most striking 1997-1998 ENSO problems in Peru: a governmental misreading of the appropriate prior experience.

President Alberto Fujimori and his government were conspicuously active in mid- to late 1997, engaging in a variety of physical mitigation activities (e.g., clearing channels, repairing dikes). Unfortunately, they were assuming that the developing 1997-1998 ENSO would be a repeat of 1982-1983, whereas in fact it turned out to be most similar to the 1925-1926 event. The result was that much of the mitigation activity was misplaced. Not misplaced, however, were the political motivations, as The Economist (May 9, 1998, p. 38) observed:

The true artist of El Niño has been Peru's President Alberto Fujimori. His government spent $300m in advance (not all in the right places, but at least the ones that looked right at the time); and El Niño struck as he rushed about frenetically taking personal charge of relief efforts, even rescue attempts. Too frenetically, say some critics, who claim presidential efforts are muddling those of people on the spot. Maybe, maybe not; but his poll ratings, 30% in mid-1997, now stand at 45%.

The larger lesson, as noted previously, is that while ENSOs are regional to global phenomena, it is their specific local effects that are so problematic, and these local effects 1) vary tremendously from ENSO to ENSO, 2) are difficult to predict except in the very short-term (hours to a day or two), and 3) constitute the actual disasters/catastrophes associated with ENSO events. Therefore, ENSO mitigation in the Andean countries must be long-term, national in scope, and historically informed by the scientific community. To put it bluntly, countries that regularly experience significant ENSO impacts should always be systematically readying themselves (mitigating and preparing) for the next event.

Given an admitted inability to calculate losses to the environment, the CEPAL team simply noted that were such losses to be included, "the figures would go higher."

Aggregate economic losses are an in-complete (and rather cold) way to capture disaster impacts, a point not lost on the CEPAL team. The CEPAL report noted that seven million people-60% of the population of Ecuador-were affected in one way or another by the 1997-1998 ENSO but that the primary impacts were in the coastal and southern provinces of the country. While the numbers of killed (286), injured (162), and missing (36) were not large, the major problem was the number of families permanently or temporarily homeless due to riverine and coastal flooding. Combining the two categories, CEPAL reported more than 18,000 families (nearly 90,000 people) homeless at one time or another during the 1997-1998 ENSO. Of that number, more than 6,000 families (29,000 people) lost their homes completely.

As is common in disasters, the poor, especially owners of small farms and day laborers in the southern and coastal rural areas, suffered the greatest losses. The CEPAL team noted that many males from these groups were migrating to the cities, leaving female heads of household alone in disaster zones, in economically precarious and unsanitary conditions. The CEPAL report offered the following general assessment:

El Niño has generated a migratory wave of vast consequences. Thousands of families have been displaced by the destruction of homes, loss of crops, loss of work, or to seek protection in shelters. In Guayaquil alone, 18 square kilometers . . . have been occupied [by homeless from the affected region] . . . These are families who lost or abandoned in the majority of cases very humble dwellings and now look for something as humble-or even poorer. [CEPAL, 1998, p. 21]

Field observations corroborated this point, as does an assessment by the special "Intelligence Unit" of The Economist, which releases quarterly country reports published in country series format as The Economist Intelligence Unit [EIU] Reports. The second quarter 1998 Ecuador re-port tied together the country's chronic unemployment problems and the more specific impacts of the 1997-1998 ENSO:

Unemployment and underemployment are likely to have increased during the remainder of 1997 and in early 1998. Since the end of 1997 El Niño accelerated emigration from the countryside to cities has aggravated unemployment and underemployment. [EIU (Ecuador), 1998, Second Quarter, p. 17]

In terms of infrastructure, the CEPAL report identified the greatest losses in water systems, sewage treatment, communications, and transport, especially roads. The education infrastructure also suffered damage, as some schools were lost to flooding and many others had been used as emergency shelters for dozens to hundreds of families.

Until the delivery of the CEPAL study, the government of Ecuador loss estimates were generally lower, but in June 1998, the vice president of Ecuador reported figures in a meeting with the World Bank in Washington that closely approximated those of the CEPAL team. Total losses were estimated at US$2.5 billion. By categories, "Roads and Bridges" constituted US$1 billion of the damage. "Housing" losses were estimated at US$23 million. For "Schools" the estimate was US$15 million. The most impacted sector, "Agriculture," accounted for US$1.5 billion.

In this latter sector, heavy rains hit hard the provinces of El Oro, Guayas, and Los Rios, where agricultural production is concentrated. The rains destroyed banana, coffee, cocoa, and rice crops. As early as the fourth quarter 1997 Ecuador report, the EIU expected (p. 8) "GDP growth in the second half of 1997 to be well below that of the first half and growth of 3% is forecast for this year as a whole." Moreover (p. 7), "Higher prices of imports in domestic currency will feed inflation pressure already created by fiscal imbalance and scarcity of basic food as a result of El Niño." In its second quarter 1998 report, the EIU estimated:

The government's target of 30% inflation for 1998 is likely to be missed by a wide margin. Accumulated inflation between January and April 1998 was 15.6% and year-on-year inflation was 33.4% in April. Shortages caused by El Niño were a major cause of inflation, pushing up food prices, which have a 32.1% weight in the consumer basket. Food prices increased by 15% during January-April, bringing year-on-year inflation for that category in April to 43.5%. [EIU (Ecuador), 1998, Second Quarter, p. 16]

Interestingly, in late 1997 in its fourth quarter report, the EIU was guardedly optimistic about one of Ecuador's principal agricultural exports:

Ecuador's banana production has so far been unaffected by the changes associated with El Niño. Local experts believe that banana production will remain unaffected for the duration of the climatic changes if the excess rainfall continues to be concentrated in the northern coastal region, away from the main banana-producing areas such as El Oro. As a safeguard measure formulated following the 1982/83 El Niño producers have built and maintained drainage channels and containing walls to prevent the flooding of plantations. [EIU (Ecuador), 1997, Fourth Quarter, p. 18]

As noted above, however, luck did not hold and the bananero mitigation works proved inadequate when the rains moved south. The EIU reported export revenues falling by 5% in the first quarter of 1998, compared to the same period a year earlier. Heavy rains and humidity also caused the resurgence of the Sigatoka Negra fungus that, together with weather damage, destroyed an estimated 6 to 12% of the banana crop (EIU [Ecuador], 1998, Second Quarter, p. 20).

Peru has long been adversely affected by ENSOs, and the 1982-1983 ENSO contracted Peru's GDP by over 10% (the estimated losses were US$2 billion caused by heavy rains on the northern coast and drought in the southern highlands). Despite the significance of past ENSO impacts, the government of Peru did not request a CEPAL, or for that matter any other external, evaluation of 1997-1998 ENSO damage. Indeed, the entire topic became quite sensitive. On January 9, 1998, the government's Presidencia del Consejo de Ministros (PCM, Office of the President of the Council of Ministers-i.e., the prime minister) removed control of disaster information from the Instituto Nacional de Defensa Civil (INDECI, the National Civil Defense Institute). Then, rather than continuing to provide damage statistics, the PCM began reporting government response activities-not the same thing at all. Over time this led to some conflict between the national government and various NGOs as to the severity of ENSO impacts.

Similar to Ecuador, Peru's primary problems from the 1997-1998 ENSO were coastal storms and riverine flooding, especially in the areas around the northern cities of Piura and Tumbes and the mid-south city of Ica. A preliminary (June 1998) government damage census by the Instituto Nacional de Estadística e Información (INEI, National Institute of Statistics and Information) reported 529,000 people "affected" by the 1997-1998 ENSO (396,000 urban, 133,000 rural). The INEI report also listed nearly 31,000 homes destroyed or otherwise uninhabitable and another 32,000 partially destroyed. INEI's figures for infrastructure and economic losses were hotly contested by several government ministries, but damage to roads and bridges was at least the US$120 million announced for reconstruction. Fisheries were showing at least a 20% drop in production from the previous year, but it should be noted that fisheries and fishmeal only account for 1% of Peru's GDP.

Agriculture, however, accounts for approximately 12% of Peru's GDP, and that sector began experiencing ENSO effects as early as August 1997. While the minister of agriculture maintained as late as June 1998 that agricultural growth would hit 5% for the year, most private-sector groups estimated only a 2% gain. Perishable products were a particular concern. Various private-sector groups cited a 33% drop in fruit and a 9% drop in vegetables reaching market in the first four months of 1998, compared to the same period of 1997. For the first half of 1998, the EIU reported that agricultural production had shrunk by almost 4% year-on-year and that huge cultivation areas had been destroyed:

Some 17,000 ha [hectares] of crops, representing 1.4% of the total national area under cultivation, have been lost because of rains and flooding from El Niño, according to the Congressional Agricultural Commission. The areas most affected by El Niño are Tumbes in the north and Ica in the south. The commission president, Carlos Blanco, has explained that in the 1997 planting season, an additional 25,000 ha were planted nation-wide, which should help to compensate for crop losses. [EIU (Peru), 1998, Second Quarter, p. 23]

In early March 1998, President Fujimori even called on the Peruvian navy to transport fruits and vegetables from northern Peru to the area around Lima. Heavy rains and landslides had made overland transportation impossible or exorbitantly costly. The EIU stated that "the products were distributed to the wholesale market in a bid to prevent a rapid surge in food prices which would have fuelled consumer price inflation" (EIU [Peru], 1998, Second Quarter, p. 23).

NGOs in Peru consistently reported higher loss figures than the government for the 1997-1998 ENSO (and had noted the effects much earlier as well). The Centre for the Study and Prevention of Disaster (PREDES, an NGO with considerable field experience and contacts) estimated ENSO losses as 374 people killed, 412 injured, 35,669 homes destroyed, 74,000 homes damaged, and a total affected population of nearly 600,000. CARITAS (the Catholic relief agency) reported 40,549 homes destroyed and another 36,699 seriously damaged.

PREDES estimated the total cost of ENSO damage at US$1.8 billion-considerably higher than the government of Peru's reconstruction estimate of US$620 million. The discrepancy is largely due to the government's almost exclusive focus on infrastructure projects, versus the various NGOs' concern that small farmers, artisans, and the poor (which the NGOs argued were disproportionately hard hit by the 1997-1998 ENSO), were ignored in the accounting and left out of the reconstruction financing plans.

Summarizing prospects for 1998 in its second quarter 1998 report (p. 8), the EIU could find only spotty positive aspects:

Economic prospects for 1998 indicate a mediocre year by recent historical standards. El Niño and the Asian crisis are likely to depress growth to around 4% this year and raise inflation to just under 10% by the close of the year. There are signs that El Niño is starting to subside but its overall impact on the economy is still difficult to predict. The government has estimated the total cost of the damages caused by El Niño at more than $600m, and the destruction of the infra-structure will certainly stifle production in agriculture. However, in terms of income flows other sectors, notably construction, will benefit as the repair efforts continue.

Because it is landlocked and very poor (the lowest per capita income in South America), Bolivia experiences ENSOs somewhat differently than do Ecuador and Peru. While Bolivia also incurred some unexpected rains and flooding, the country's major problem was drought in the Andean high valleys, where the reduced precipitation primarily affected the economically weakest part of the indigenous population, the poorest of the poor of a poor country, as well as one of their staple crops, potatoes.

As in Ecuador and Peru, the government of Bolivia was hesitant to officially recognize the full extent of the disaster. Indeed, it was USAID-Bolivia and its mission field investigations that by February 1998 identified drought and falling agricultural production (and possible famine) in the high valleys as the real problem posed by the 1997-1998 ENSO. Until then, public, national government, and inter-national attention was focused on rain in the lowlands and some local flooding, which was in fact not unusual.

Drought and drought effects on agriculture are not as photogenic as violent storms, mud-slides (huaicos), and floods. Hunger is invisible until it reaches the famine stage, and by then it is often too late for most victims. As a result, ENSO effects in Bolivia were not as obvious as in either Ecuador or, especially, Peru. In addition, the government of Bolivia downplayed the impact of the 1997-1998 ENSO on food production, saying as late as March 1998 that crop losses were only the equivalent of US$131 million and production was off no more than 40% for potatoes, corn, barley, quinoa, and wheat. In contrast, USAID-Bolivia estimated crop losses "in the US$200 million range" and production down astoundingly "over 60%." To be fair, several Bolivian officials understood that ENSO effects were more severe than they would admit publicly, but they held to their lower estimates in an attempt to, as one official explained, "limit price speculation in food-stuffs." The problem was, of course, that holding to the underestimation also undercut the sense of urgency and retarded response planning.

In agriculturally based economies such as Bolivia's (16% of the GDP is agriculture), adverse effects on food production and distribution tend to quickly drive up inflation. In its fourth quarter 1997 Bolivia report (pp. 16-17), the EIU credited the government of Bolivia with keeping the lid on inflation but also noted that ENSO effects were catching the poor in a particular, very cruel economic vise:

Prices have been held down by the authorities' tight monetary policy and the firm exchange rate. There may also have been some downward pressure on meat prices as farmers slaughtered stock in anticipation of difficulties brought on by the El Niño climatic phenomenon, which will reach its peak at the end of the year. With agricultural production expected to be affected by the unusual weather patterns, there may be a temporary increase in consumer prices of as much as 3 percentage points in the first half of 1998. A study by the Unidad de Análisis de la Política Social (UDAPSO, the Social Policy Analysis Unit) concludes that the increase in the prices of staple foods in the first half of 1998 as a result of El Niño will have a regressive effect on income distribution, as the cost of living of the poor will rise more sharply than that of consumers with more diversified consumption.

The situation deteriorated as Bolivia moved into 1998 and ENSO effects became more pronounced-or at least more visible. In its first quarter report for 1998 (p. 16), the EIU summarized ENSO's estimated impacts on legal (i.e., excluding coca) agriculture:

The El Niño phenomenon has been blamed for severe drought in western parts of the country and serious flood damage in the eastern lowlands during the second half of last year. A report on Oruro province released in January found severe drought in six eastern provinces of the department. By exhausting supplies of both drinking water for livestock and irrigation water, the drought in Oruro is estimated to have damaged 20% of the agricultural crops planted and 30% of the livestock herd. Meanwhile the Cámara Agropecuaria de Oriente (CAO, Eastern Agricultural Chamber) estimated that El Niño had caused at least $17m worth of losses to the maize crop in the eastern lowland area around Santa Cruz. Flooding has made 60,000 ha [hectares] uncultivable, representing over one-half of the 110,000 ha originally planned for cultivation. As a result, output for the season is not expected to exceed 200,000 tonnes, less than half of domestic demand. Soya, rice, sugar and cotton crops have also been badly affected, according to the CAO. Torrential rains and flooding be-tween October and December 1997-attributed to El Niño-have also caused serious damage to the fruit-growing sector in Tarija.

Economic losses of such magnitude quickly become social catastrophes in Bolivia, and field investigations confirmed that the vast majority of the victims were the poorest of the poor in the highlands, whose standard of living was abysmally low before the onset of the 1997-1998 ENSO. By early 1998, NGO workers were reporting families dividing and increased migration from the western highlands 1) to the cities of Cochabamba, Santa Cruz, and La Paz, and 2) more troubling, to the coca-growing region of Chapare (this migration was con-firmed by the EIU in its second quarter 1998 report for Bolivia).

Drawing from the local press and government estimates for its second quarter 1998 report, the EIU offered the following analysis:

The government estimated in mid-April that over 40% of the population of some provinces will require emergency assistance to help cope with El Niño-related losses. The government has not specified the nature of the losses, which appear to cover a wide range-and degree-of problems including illness, damage to dwellings and crop destruction. Attending to those affected will be a priority in the coming months in order to prevent a wider social dislocation as a result of El Niño. The government has noted a sharp drop in school attendance in these areas as families cope with illness and clearing up the material damage of recent severe weather. According to government estimates, the worst-affected areas were Oruro, Chuquisaca and Potosí.

Explaining the need for donor coordination and a food security program in the high valleys, a May 1998 internal USAID report captured the overall situation both clearly and succinctly:

The peasant farmers in the highlands and high valleys, the most impoverished people in South America, have become poorer be-cause of the drought. They are consuming whatever food surpluses they have available from the previous year's production and are selling part of their livestock, usually their only capital asset. Migration to Santa Cruz, other urban centers, the Chapare, and neighboring countries is on the rise, reaching 50 percent of persons, both men and women, between 15 and 30 years old.

Bolivia makes it even clearer than do Ecuador or Peru that the fundamental problem with ENSOs is not solely their aggregate economic impact. As national economies develop, they are better able to absorb losses and make compensatory adjustments. Aggregate ENSO losses in the 1997-1998 event in the three Andean countries were probably not as devastating as those in 1982-1983, when the economies were smaller and much less resilient. The problem with future ENSOs will increasingly revolve around how the periodic losses (and gains) are distributed among social groups/classes. Essentially a moral issue, this point is hauntingly familiar. One only has to return to the work of Cuny (1983) on disasters and (uneven) development and, more recently, Albala-Bertrand (1993) on disasters and differential social group "entitlements" to the basic necessities of life to see the problem.

Back to Contents
Forward to Part II
Return to the Hazards Center Home Page