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

Special Publication 38

Impacts

What would become Hurricane Georges was born off the coast of Africa in mid-September 1998 and followed a typical path across the Atlantic and then through the Caribbean (see Figure 1). It first hit the eastern Caribbean islands of St. Kitts, Nevis, Antigua, and Barbuda on September 20 and 21. According to the final USAID/ OFDA (U.S. Agency for International Development, Office of U.S. Foreign Disaster Assistance) Fact Sheet #9 on Hurricane Georges, the storm caused five deaths on St. Kitts and Nevis and three deaths and two serious injuries on Antigua and Barbuda. The most affected of the early islands was St. Kitts, where total dollar damage was estimated at $445 million, with major infrastructure losses, half the sugar crop gone, 25% of the housing stock destroyed, and 60% of the housing stock missing roofs.

Hurricane Georges then struck Puerto Rico and, on September 22, the island of Hispañola, which is divided roughly in half into the Dominican Republic in the east and Haiti in the west. Georges then made the usual turn north, raked Cuba, and crossed the Florida Keys, ultimately striking the U.S. mainland near Pascagoula, Mississippi.

On the island of Hispañola, Haiti suffered 147 killed, 40 missing, and 34 serious injuries, with 4,500 homeless (combining reports from OFDA and the International Federation of Red Cross and Red Crescent Societies). The focus of this chapter, however, is the impacts on the Dominican Republic.

Sometimes not fully appreciated in the U.S., one of the problems for Caribbean and Central American nations is that they are often physically smaller than the hurricanes that hit them. From a satellite perspective, these "city-states" can virtually disappear under a hurricane for a day or more. (See Photos 1 and 2 - two September 22 satellite photos, six hours apart, of Hurricane Georges. Without the islands drawn in, neither Puerto Rico nor the Dominican Republic would be visible in the first; Haiti "disappears" in the second.)

As a result of this size mismatch, disaster response in the Caribbean and Central America is often slow, and damage assessments can be delayed 36 hours or more. For example, the OFDA Fact Sheet #1 of September 24 was actually quite brief regarding both the Dominican Republic and Haiti:

Dominican Republic: President Leonel Fernandez reports that 70 individuals have died as a direct result of Hurricane Georges and another 100,000 people are homeless. A full assessment has not yet taken place to ascertain the extent of damage and immediate needs in the Dominican Republic; however preliminary reports indicate that heavy wind damage and flooding have occurred in Santo Domingo. The airport is not fully functional, electricity is not available, and telephone services are limited. A curfew is being enforced by soldiers in Santo Domingo, following reports of looting and street violence.

Haiti: A damage and needs assessment also has not yet taken place in Haiti. The northern coastal areas that extend from Cap-Haitien to Gonaives are most damaged-mainly as a result of flooding. Some damage to roofs has been reported. The Haitian Civil Protection Directorate reports 27 deaths, 29 serious injuries, and 9 missing persons.

It should be noted, however, that U.S. Chargé d'Affaires Linda Watt in the Dominican Republic anticipated the developing damage pattern and declared a disaster on September 23 (as did U.S. Chargé d'Affaires Margaret Jones in Haiti). Such a declaration is formally required for the United States government to begin responding to an overseas disaster.

Indeed, OFDA Fact Sheet #3, on September 26, reflected a worsening picture. It reported that the Dominican Department of Defense Damage and Needs Assessment Com-mission was listing 201 killed, 551 injured, and 90 missing. OFDA also said that "the final death toll is almost certain to exceed 500." The casualty figures remained essentially unchanged until Fact Sheet #6 on September 30, when OFDA began to note, and report on, data problems:

Reports of hurricane-related damages continue to vary. Figures from the State Secretary's Public Health and Welfare Office indicate 208 deaths and 134,836 displaced persons . . . whereas the Pan American Health Organization (PAHO) reports 865,510 displaced and 400,000 homeless.

The final OFDA Fact Sheet for Georges (#9) repeated that the casualty figures continued to "vary." It then added, however, that "the American Red Cross reports 2,000 missing persons and 300 deaths," but the detail was even more interesting:

The death toll is almost certain to rise [to or past the previously mentioned 500] because many unregistered migrant workers [a code term for Haitians] are missing after being swept away by flood waters and mudslides. The areas hardest hit by the hurricane include La Romana, San Juan de la Maguana, and San Cristobal. While the greatest wind damage to housing is in the coastal areas of Santo Domingo and to the east, most flood damage to homes is in the south and west. Critical shortages of food, water, and shelter have resulted in poor urban neighborhoods and rural areas throughout the country. . . . Infrastructure damage includes schools, hospitals, health clinics, homes, the main airport, and municipal water systems. In addition, approximately 100% of the road network and 60% of the bridges are damaged. . . . Total damage to the power system is $1.2 billion. . . . The Dominican Secretary of Agriculture conservatively estimates $260 million worth of damage or 90% destruction to the agricultural sector.

As one might suspect from the above reports, the figures concerning losses from Hurricane Georges in the Dominican Republic, especially the human losses, became a political football. The Fernandez government (from the PLD party, Partido de la Liberación Dominicana) consistently reported relatively low human loss figures (the 200-plus killed), while the congress, which was controlled by the opposition PRD party (Partido de la Revolución Dominicana), publicly discussed more than a thousand killed. It is, of course, not surprising that the incumbent administration and the political opposition would disagree over casualty estimates. In this case, the opposition wanted a high death count to make the government look inept or uncaring, while the Fernandez administration was trying to "construct" a lesser disaster.

The leading nongovernmental organizations (NGOs) were rather in the middle regarding the casualty figure (500 or so killed). On the street, however, death toll talk reached 2,000. Interestingly, this pattern is very reminiscent of the 1985 Mexico City earthquakes, in the aftermath of which, for political reasons, the PRI government held the human loss figure consistently below 10,000 while playing up the dollar damage figures.

The bottom line is that no one will ever know how many people died in the Dominican Republic as a result of Hurricane Georges. Setting aside the casualty figures, however, it is interesting to take a broader look at the storm's impacts on the country. In a late November 1998 internal report, a donor agency summarized them very succinctly:

Hurricane Georges passed over the Dominican Republic on September 22, 1998, as a category III hurricane with winds reaching 130 miles per hour. The eye of the hurricane entered the southeast portion of the country in the morning, traversed the country at approximately 6 mph on a northwest path and arrived at the border with Haiti in the even ing, downgraded to category I. The destructive winds destroyed housing, agricultural and industrial infrastructure, uprooted trees and destroyed crops, mainly in the eastern part of the country. The heavy rain was centered in the south/southwest areas of the country and led to floods and rivers overflowing with water and mud, which destroyed bridges, homes, and household and farming equipment; damaged roads, schools, health clinics and water supply systems; and washed away crops, including sugar, bananas, yucca, coffee and vegetables. Because of its diameter, the hurricane affected at least 70% of the country, equivalent to 34,000 square kilometers. . . . The death toll now stands at 235.

While prior to the hurricane the Dominican Republic had been enjoying a high rate of GDP growth, agriculture in the country sustained severe immediate losses from Hurricane Georges. The Economist Intelligence Unit [EIU] Country Report for the Dominican Republic (Second Quarter, 1999, p. 15) estimated total disaster losses at $2 billion and noted:

The agricultural sector was clearly hard hit. Whereas crop production had experienced year on year growth of 5.4% in the first nine months of 1998, production declined by 1.4% over the year as a whole. The hurricane forced the government to resort to massive imports of basic foodstuffs such as rice, in order to prevent major price increases. Nonetheless, the effects of supply shortages were still visible in some domestic foodstuffs, such as plantains. The livestock sector, likewise, finished the year with unimpressive growth of 1.5% after a buoyant performance in the first nine months. Chicken production was severely affected by the hurricane, while pork production was hit by an outbreak of swine fever.

Another post-impact problem was inflation:

Before Hurricane Georges at the end of September, inflation had reached its lowest level for some years. Accumulated inflation in the first nine months of the year was just 2.25%, with year-on-year inflation in September of 2.91%. Most categories of goods in the basic consumption basket experienced even lower price pressures, with prices of clothing and foodstuffs rising by an accumulated 1.26% and 1.64% respectively in the January-September period. An average increase of 3.34% in the price of housing, however, was the main factor dragging the price index upwards. . . . Inflation rose significantly in the wake of the hurricane, however, with monthly rates of 2.2% in October and 1.8% in November bringing the year-on-year rate in the latter month to 6.6%. [EIU (Dominican Republic), First Quarter, 1999, pp. 18-19]

Disasters are never lose-lose situations; in fact, some individuals and economic sectors benefit tremendously. Typical was a post-Hurricane Georges construction boom in the Dominican Republic:

Although growth in the construction sector in the first nine months of 1998 was buoyant at 7.5%, it was below the 19% experienced in the same period of 1997. The slowdown was to be short-lived, however. In the final quarter of 1998, construction output grew by 49.2% year on year, on the back of the massive reconstruction effort in the wake of Hurricane Georges. [EIU (Dominican Republic ), Second Quarter, 1999, p. 16]

Overall, however, Hurricane Georges affected the nation's balance of payments in a very direct and negative way (although it was far from being the sole culprit in the balance problem):

The Dominican Republic's current-account deficit widened sharply in 1998, from a revised total of $163m (1.1% of GDP) in 1997 to $387m (2.5% of GDP). Most of the deterioration was accounted for by the widening trade deficit, which rose from $2bn to $2.6bn in 1998 as imports continued to grow rapidly and domestic exports contracted sharply. The poor outturn for the non-free-zone trade balance can be attributed largely to the impact of Hurricane Georges: import growth rose from 13.5% in the first nine months of the year to 16.8% for the year as a whole, while the decline in domestic export earnings went from -7.7% to -12.7%. [EIU (Dominican Republic), Second Quarter, 1999, p. 17].

Importantly, however, the same EIU report (p. 3) expected the Dominican Republic-"buoyed by post-hurricane reconstruction"-to maintain consistent positive GDP growth rates. The EIU forecast the economy to show 6.4% growth in 1999 before falling to a still quite respectable 5.3% in 2000.

Economic impacts aside, a recurrent problem is the widely varying political contexts within which disasters occur. The 1998 political situation in the Dominican Republic was tense and conflictive, and that context continued right into 1999. Despite the disaster, substantive collaboration between the major political parties was elusive at best. As the EIU noted:

The PRD-dominated Senate withheld ratification of emergency loans worth a total of $215m from the World Bank and the Inter-American Development Bank, destined for post-hurricane repairs and balance-of-payments support. Although the upper house softened its position under severe pressure from public opinion, and approved a number of laws governing the economy, by mid-March the crucial loan legislation had still not been debated. [EIU (Dominican Republic), Second Quarter, 1999, pp. 10-11].

Politics in the Dominican Republic is far from simple, and understanding the political terrain helps to further contextualize the problems with the government of the Dominican Republic's institutional response to the disaster.

To start, democracy is not fundamentally rooted in the Dominican Republic, and, as noted above, the Fernandez government faced a vocal opposition majority in the congress. The EIU described the post-Georges situation this way:

The coming year will be an important one for Dominican democracy, which has been in a state of halting transition for the past 35 years. Tension between the ruling Partido de la Liberación Dominicana (PLD) and the majority party in Congress, the Partido Revolucionario Dominicano (PRD), has heightened considerably in recent months. . . . But the persistence of such disputes-stemming from the struggle for political power rather than from any ideological differences-flies in the face of the country's many pressing needs, both in terms of reconstruction in the wake of September's Hurricane Georges and in terms of economic and political modernisation. The population seems increasingly disillusioned with the country's political leaders. [EIU (Dominican Republic), First Quarter, 1999, p. 7].

Because public opinion and media coverage go hand in hand, even in a fledgling democracy, below we take a look at how the Dominican newspapers covered Hurricane Georges.

Domestic Media Treatment

Covering a 13-week period (September 22-December 22, 1998) and 10 major Dominican newspapers that chronicled Hurricane Georges in the Dominican Republic, Table 4 reports the total number of Hurricane Georges stories published per week.

Table 4

Hurricane Georges Coverage:
Ten Dominican Newspapers

Beginning with the pre-impact week and following the evolution of the post-impact efforts, a rather astonishing 5,497 stories were published detailing Georges's track toward and then its impact on the Dominican Republic. Figure 2 graphs the same coverage data. Interestingly, it is an almost perfect hyperbolic curve from high/intense attention to virtual issue disappearance.

More specifically, during the first four weeks of the disaster, 4,567 stories were published, in sharp contrast to 898 published in the subsequent nine weeks. Week four to week five appears to be a break point, Georges-related stories dropping more than half, from 633 to 302.

To the extent that media play an important role in the creation of the collective memory of a disaster, the break point is interesting. Unlike media treatment of the 1997-1998 El Niño (more properly, ENSO [El Niño-Southern Oscillation]) in the Andean countries, coverage that went up and down with specific impacts and lasted months, the attention span of domestic media to the quick-onset disaster of Hurricane Georges was really only a few weeks, a month at most. This relatively brief but intense span of attention perhaps explains why disasters become so political so quickly: the window for both official and competing explanations of what happened and why is not open very long.

To better understand the role of the media, we also examined the way in which international disaster assistance was reported and specifically how much attention was given to various donors. Tables 5 and 6 are "story counts" of how assistance by source (donor) was reported. Table 5 contains the story counts for assistance identified by country of origin. The most noted donor was the United States, which was featured in 116 stories, followed by Spain in 47 stories, and France in 44. Among the 116 stories that mentioned the United States, interestingly, OFDA was specifically mentioned in 12 stories, USAID was noted in 29, and the Peace Corps was mentioned in 25. The remainder simply cited the United States or the Clinton administration as the donor.

Table 5

Hurricane Georges:
Dominican Media Recognition of Assistance, by Donor Nation

Table 6 contains the story counts for assistance from nongovernmental organizations (NGOs), international intergovernmental organizations (IGOs), and multinational corporations (MNCs). Most often mentioned were Dominicans in New York (in 28 stories), followed by Dominicans in Miami (in 10 stories). Interestingly, the Inter-American Development Bank (IADB) was also mentioned in 10 stories.

Table 6

Hurricane Georges:
Dominican Media Recognition of Assistance,
by NGOs, IGOs, and MNCs

Finally, we also assessed, if only generally, how the domestic Dominican media attempted to apportion blame for the disaster. To appreciate those findings, however, one must first analyze the government's disaster response, which left a great, great deal to be desired.

Responding to Georges: A Tragedy of Errors

Keeping in mind the context outlined above, let us return to our focal questions: How well did the government of the Dominican Republic respond to the threat and then the actuality of Hurricane Georges? What was learned and/or changed as a result of the experience? How well does the disaster agency marginalization or "sidelining" proposition hold in this case?

To begin, we have, in typically understated bureaucratic language, a November 1998 international donor agency report. It notes that the government of the Dominican Republic "acknowledged . . . serious shortcomings in its emergency management" and would be seeking external assistance for a reconfiguration. Given what happened, this should not come as a surprise. Hurricane Georges exposed major intraorganizational and interorganizational shortcomings.

The line between denial/wishful think-ing on one hand and poor judgment on the other is often blurry in a disaster situation, and the government's problems with the threat of Hurricane Georges were no exception. The first set of problems was "intra-scientific." Several interviews and a review of Dominican newspapers in the week prior to impact reveal major differences between the assessments and forecasts of the U.S. National Hurricane Center (NHC) in Miami and the government of the Dominican Republic's National Meteorological Office (Oficina Nacional de Meterologia, ONM). The same review also shows major inconsistencies in the way that the ONM was assessing the threat posed by Hurricane Georges.

While as a standard practice the NHC in Miami shows what they consider the most likely future track for a particular hurricane, they are careful to include a "fan" of possibilities that widens with time (up to 72 hours). They also emphasize that it is dangerous to focus attention on the storm's eye, often reiterating that more important is the totality of the storm (wind, storm surge, rain, flooding). Indeed, as would be the case with Mitch in Nicaragua, a hurricane can kill thousands without the eye even touching national territory.

At any rate, the initial problem was that the Dominican media were reporting, as late as September 20, contradictory statements from the ONM, sometimes in the same article ("prepare" versus "it will just be some rain and thunder"). The media were also picking up contradictions between the ONM ("Georges will probably turn north and won't affect the country") and the NHC in Miami (which was repeating that the Dominican Republic was well within the 72 hour fan of track possibilities). The Weather Channel in the U.S. was taking the same position as the NHC.

To make a long story short, the ONM maintained its optimistic forecast until the trees literally began going down on the eastern tip of the island. Interestingly, it was revealed in a September 22 Dominican newspaper that the ONM was completely without an operating Doppler radar. (It had one, a gift from Germany, but it had been shut down for three years for lack of both technical personnel and an adequate maintenance budget.) That is, the ONM was relying on the same images and data as the NHC and the Weather Channel but was coming to different conclusions. It turned out that the chief of the ONM, Felix Abel Abreu, was rejecting forecasts from his own people that agreed with the NHC. He was quoted in defense as saying that he "didn't want to alarm the people" (Abreu was later dismissed from his position).

Unfortunately, this desire to avoid a possible "cry wolf" problem also directly affected, and reinforced a bias within, the Dominican government's Civil Defense (CD) system. That story is also very interesting and exemplifies both interorganizational and intraorganizational problems in responding to Georges.

Although the NHC was issuing strike probabilities for the Dominican Republic for days prior to impact and at least some people in the ONM agreed that the threat from Hurricane Georges was real, Dominican Civil Defense was slow to heed the warnings. Indeed, two days before Georges struck, the Dominican media reported that Civil Defense was saying that the Dominican Republic was in no danger. That the newly appointed head of Civil Defense, Elpidio Báez (a former communications professor and journalist) had no formal training in meteorology or disaster management only contributed to the problem. (It was reported that Báez himself was surprised when he was designated as the head of Civil Defense.)

Until very near impact, Báez publicly insisted that Georges would turn northward, as the majority of hurricanes had done in the past-despite information made available by the NHC, CNN, and the Internet. Even as the hurricane was making landfall, Civil Defense was still claiming that that they could not be sure of the hurricane's trajectory.

Accused of hiding the true gravity of the situation, Báez did a poor job of defending himself and demonstrated how little he knew of disaster management, meteorology, and hurricane strike probabilities. He was quoted in every major newspaper as saying, "El Meteorólogo dio su opinión, opinó y teorizó, yo no me meto en eso, yo me metí en mi prevención" ("The Meteorological Service gave their opinions and theorized; I didn't involve myself in that; I immersed myself in prevention").

Apparently Báez did not understand the difference between opinion and scientifically determined probabilities, as well as the importance of advance warning. Fearful of "panic," Civil Defense issued no official warning until the storm was literally on top of the country.

Civil Defense also seemed to confuse the eye of the storm with the potential impact area. Baez himself stated on television that "we have to wait and see what path it takes" even as the outer bands of Georges were beginning to impact the country. "They were looking at the eye's wobble and not the entire storm," one NHC official in Miami said, adding that he had told officials in the Dominican Republic, "Forget the eye. A storm the size of Texas is headed your way." As the EIU summed it up later:

The government has been heavily criticised for its handling of the hurricane, which hit the island on September 22^nd, killing at least 280 people and leaving a further 300,000 homeless. The authorities contradicted the forecasts of the National Hurricane Center in Miami, which warned of a direct hit on the island and failed to inform the poor of arrangements for shelter. [EIU (Dominican Republic), Fourth Quarter, 1998, p. 29]

Especially troubling was the Civil Defense delay in divulging information about the location of evacuation sites. The list of approved evacuation sites was kept confidential until, in some areas, less than 24 hours before impact. Apparently, the government was fearful that "professional victims" (refugiados profesionales) would take advantage of the shelters and set up permanent occupancy (this concern had some basis in fact, given the response to Hurricane David in 1979). In order not to give away the location of the shelters until it was absolutely necessary, the shelters were also not stocked adequately with food, water, and medical supplies ("people would follow the trucks and find the shelters"). When Hurricane Georges eventually struck the island, people became desperate when they could not locate the shelters or found them ill-equipped at best.

In the end, the Dominican Republic was ill prepared to deal with the disaster not only because vital information was not disseminated soon enough to the population, but also because Dominican Civil Defense had never been allocated sufficient funding. In fact, at one point and under media and political attack, Báez stated that the organization had only its "hands" to work with, lacking even the most basic resources to deal with a major hurricane.

To be fair, in early September 1998, soon after being appointed, Báez had proposed the creation of a more professional and better funded civil defense structure, including a Comisión de Prevención y Mitigación de Desastres (Disaster Prevention and Mitigation Commission) and a much better equipped emergency operations center. Nothing had been done, of course; it was a proposal by an official very far down in the political hierarchy.

With Civil Defense failing and being faulted everywhere, responsibility quickly gravi-tated to the Armed Forces, which set up its own operations center in the capital to deal with Georges and became the de facto leader of the governmental response.

In conclusion, it is safe to say that Dominican Civil Defense was underfunded, understaffed, and organizationally, administratively, and politically weak prior to Hurricane Georges-and thereby unable to effectively coordinate response to the hurricane. Civil Defense and its chief, Elpidio Báez, bore the brunt of criticism, especially from the media and the opposition, and became the scapegoat for the poor disaster response. Indeed, for several days after impact, Báez was reported to be staying in a series of hotels because he was afraid to go to his house. (He was also later dismissed from his position.)

Interestingly, President Fernandez did his best to distance himself from Civil Defense, instead focusing his attention on international aid and reconstruction issues. While not within the scope of this study, Fernandez presents an interesting contrast to other, more "activist" presidents of disaster-stricken countries (Fujimori in Peru would be a near polar opposite given his handling of the 1997-1998 El Niño).

Blame Themes

It is clear that opportunities for political or social credit in the Georges disaster were nonexistent in the Dominican Republic, which is hardly surprising because nearly everything went wrong. Post-disaster blame, however, is always abundant, and reviewing the major Dominican newspapers, we isolated a set of five themes that appeared repeatedly in the three months following Georges's impact.

The first blame theme revolved around the forecasting problems noted above. The local media were scathing in their attacks on the national weather service, the Oficina Nacional de Meterología, accusing them of ignoring the National Hurricane Center in Miami as well as the Weather Channel "track forecasts" that showed Georges bearing down on the country. (As suggested above, this criticism was partly because of serious disagreements within the ONM.) According to some press accounts, ONM "incompetence" cost lives.

The second blame theme encompassed the warning and evacuation problems. The media devoted considerable attention to the failure of the government, especially Civil Defense, to issue timely warnings. The problem of unclear or inconsistent mandates and responsibilities surfaced, and the heads of both the ONM and Civil Defense were faced with explaining why it took so long for an official warning to be issued and why so little information was disseminated regarding the location of evacuation sites.

The third blame theme focused on dams and flooding, especially in the high casualty areas around San Juan de la Maguana and Mesopotamia, which are below a dam. Dam officials had to explain why there was so much death and destruction, even after they supposedly undertook normal precautionary measures. Stories in the press attacked dam officials and the government for negligence, incompetence, and even manslaughter.

Fourth, the seemingly inevitable theme of corruption surfaced. Numerous stories appeared questioning the apparent discrepancy between what was arriving from international sources and what was being distributed and where. However, it should be noted that the international donor community was not being questioned or blamed. Rather, the questions concerned what happened to assistance once it arrived in-country and was in Dominican hands.

Finally, the media also questioned the actual response capability of the government, appropriate roles of the president and the military, and the general lack of "leadership" (liderazgo) in dealing with Georges. Indeed, the best way to close the chapter on Hurricane Georges and the Dominican Republic is with a summary of a September 26, 1998, article from the country's leading daily newspaper, the Listin Diario. The piece captured a number of the problems. Entitled (our translation) "Poor Government, a Hurricane's Best Ally," the article opened with a blast at both government and individual (but unnamed) authorities:

The lack of institutionalization in the Dominican state and the irresponsibility of some of its officials were the best allies of Hurricane "Georges," which has caused at least 125 deaths although hundreds of people are still missing.

The article continued by detailing the ONM's ignoring of the strike probabilities and the head-in-the-sand attitude of Civil Defense. It then turned to the issue of the government's keeping the list of evacuation centers "confidencial." The article also excoriated the military, noting that "evacuation orders for the most at risk zones coincided with the military going into their barracks [acuartelamiento]" and taking their trucks with them. Before closing with a general attack on "the absolute lack of official information" about the storm and its impacts, the article even managed to work in the problem of officials giving inadequate warning to San Juan de la Maguana. In the end, no part of the government escaped this article, but one could hardly call the story unfair.

Back to Table of Contents
Forward to Part III - Hurricane Mitch and Honduras
Return to the Hazards Center Home Page