FARMERS' AND PUBLIC RESPONSES TO THE 1994-95 DROUGHT IN BANGLADESH: A CASE STUDY
Return to Hazards Center Home Page
Return to Quick Response Paper Index
This material is based upon work supported by the National Science Foundation under Grant No. CMS-9632458. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
The continued drought in the northwestern districts of Bangladesh
led to a shortfall of rice production of 3.5 million tons (Rahman
and Biswas 1995, 7).(1) These districts are considered
granary of Bangladesh and produce surplus rice - the main staple of the
country. However, by early 1995, the government food stock fell the
lowest level in the last five years. The government had agreed to
import 0.2 million tons of rice to offset the shortage in
government stock and meet the country's requirement on an emergency
basis (Rahman and Biswas 1995, 9). A significant quantity of food
grain has already reached the country.
Despite the recurrent and devastating nature of drought in
Bangladesh, it has received much less attention from researchers
than floods and cyclones (Brammer 1987, 21; also see Alexander
1995, 2). In a recent annotated bibliography of social science
literature on natural disaster in Bangladesh, Alam (1995, 6) listed
only 11 titles on drought as against 156 for floods and 54 for
cyclones. The primary objective of this study was to explore and
analyze the ways and means by which residents of drought affected
areas adjust to drought conditions. The extent of damaged caused by
the 1994-95 drought is also examined with the help of data
collected from a sample survey conducted in the drought-affected
northern districts of Bangladesh.
The adjustment strategies adopted by the people in rural Bangladesh will be studied using a structuralist political-economy approach (Emel and Peet 1989, 50; Zaman 1989, 198). This approach claims that people affected by environmental hazards respond in different ways, depending on their economic position, and social and political linkages. Variables important in this context are occupational characteristics, landholding size, tenancy status, and years of schooling. Some of these variables make some people more vulnerable to drought than others (see Liverman 1990, 50).(2) The household responses to the 1994-95 drought will be examined in relation to the above variables.
An additional variable is also considered in order to see whether or not institutional membership status influences adjustment strategies of the affected people. In the context of flood adjustments, it has been found that the households that were members of institutionalized groups demonstrated a better performance compared to their counterpart nonmembers (Haque 1993, 384). The institutionalized groups are defined as the registered target groups of government and nongovernment organizations (NGOs) such as agricultural cooperatives, credit unions, and women's groups.
Before dealing with the research design and findings of the
sample survey, definitions of drought and its effects are
conceptualized in the next section. This is followed by a section
which provides a brief accounts of drought occurrences in
Bangladesh. These two sections are very relevant to understand
drought as an extreme event to which Bangladesh is prone. The next
section deals with the survey design of the study and the profile
of the sample households. The main section reports the results of
the sample survey. The concluding remarks are presented in the
Drought definitions abound and are used to meet specific goals
such as agricultural development planning or water resource
management (Giambelluca et al. 1988, 406; Jallow 1995, 24). In the
context of Bangladesh, Brammer (1987, 21) defines drought as a
period when soil moisture supply is less than what is required for
satisfactory crop growth during a season when crops normally are
grown. This definition resembles one provided by Heathcote (1974).
He defines drought as a "shortage of water harmful to man's
agricultural activities. It occurs as an interaction between
agricultural activity (i.e., demand) and natural events (i.e.,
supply), which results in a water volume or quality inadequate for
plant and/or animal needs." Thus, a drought is not absolute in the
sense of there being a total lack of rainfall during a normal crop
The impacts of drought are diverse and often ripple through the economy. Impacts can be classified as economic, environmental, and social. They are often referred to as direct or indirect, or they are assigned an order of propagation (i.e., first-, or second- order) (see Kates et al. 1985). In a society where agriculture is the main economic activity, a direct or first-order impact of drought is observed in the form of decrease in food production via decrease in area and yield. The second-order impact is decreased employment and income. The delay in sowing and transplanting crops reduces agricultural employment. Employment opportunities are further reduced due to diminished need for weeding and harvesting.
Because of reduced food production, prices of foodgrains usually rise rapidly following a drought (Ghose 1982, 389). Decreased food production, abnormal increases in foodgrain prices, and non-availability of jobs reduce the food entitlement of rural people, especially the small farmers and landless laborers. At this stage, drought victims often are compelled to buy food by selling their lands, household goods, and livestock at distressed prices (Reardon et al. 1988, 1065).(3) People start to consume wild plants, tubers, and leaves not normally eaten (Jallow 1995, 35). This provides an 'early warning' of famine. In this stage government and NGOs need to mobilize additional food from different sources and distribute it free of cost or at subsidized price to the affected people and provide additional employment opportunities or financial aids to the drought victims. Failing these responses famine becomes unavoidable.
People adapt various strategies to cope with the effects of drought. At the household level, people intend to reduce the effects of drought hazard by using two types of drought-mitigating techniques. These techniques are referred to as agricultural and non-agricultural adjustments. People usually practice agricultural adjustments to compensate for crop loss. Without such adjustments, people will get lower than expected food production, which can threaten their food security. People practice some agricultural adjustments, such as resowing of crops, in order to compensate for the reduction in the crop area, and others, such as application of irrigation water, to increase crop yield (Brammer 1987, 24-25). Both adjustments are practiced for the same purpose, i.e., to attain food production at or near the level of normal year.
Because of many fold rise in the prices of foodgrains during the drought period, people need additional cash to buy food crops for consumption. For this reason, they generally practice non- agricultural adjustments. The need for cash is further aggravated due to remarkable decrease in demand for agricultural wage laborer. As a result, people either sell and/or mortgage their land and livestock, and sell their belongings to earn additional cash. The community in which the drought victims live also helps in coping with the negative impact of the hazard. All members of the affected community are not equally vulnerable to drought. At the community level, friends, neighbors, relatives, and affluent members of the samaj may help the drought victims by providing cash, loans, food, and clothes.(4) In the same way, the local government and various NGOs can also help to avert the impact of the drought.
Beyond the community level, the national government as well as friends and relatives of the drought victims who live outside the victims' community can play key roles by providing financial and other support to overcome the hardships of the drought victims as well as to halt the occurrence of famine. Distribution of free food, clothes, medicine, and other relief goods is the appropriate public response to drought hazard. The government can also minimize hardships by creating employment opportunities for the drought victims and providing financial assistance to them.
The non-local NGOs may also extend their support to the
drought victims to cope with drought losses. Indeed, the impact
of the drought can be reduced significantly if all parties
respond to the hazard adequately in appropriate time.
Otherwise drought victims will face hardship in coping with the
hazard. The government's interventions are particularly needed
to avert famine and minimize the hardship of the drought
DROUGHTS IN BANGLADESH
Since independence, Bangladesh has experienced droughts of
major magnitude in 1973, 1978, 1979, 1981, 1982, 1989, 1992, 1994,
and 1995 (Adnan 1993, 1; Hossain 1990, 33). Although droughts are
not always continuous in any area, they do occur sometimes in the
low rainfall zones of the country. As listed above, Bangladesh
experienced consecutive droughts in 1978 and 1979, 1981 and 1982,
and 1994 and 1995. The 1973 drought was labelled 'the worst in
recent history,' 1979 drought was dubbed 'the worst in living
memory,' (see Murshid 1987, 35) and 1994-95 drought 'the worst in
this century' (Rahman 1985, 8).
Drought severely affects crop output in Bangladesh. Because of nonavailability of relevant data, the figures on the annual drought-related loss of crop production cannot be presented except for the 1982 drought. The total loss of rice production due to drought in 1982 was 52,896 metric tons (BBS 1986, 287-90). This accounted for about 41% of the total damage caused by all types of environmental hazards (cyclones, hailstorms, heavy rains, floods, and drought) that occurred in that year. The 1982 flood damaged about 36,000 metric tons of rice, much lower than the damage done by drought. Brammer (1987, 21) claimed that the 1978-79 drought reduced rice production by an estimated two million tons. It directly affected about 42% of the cultivated land and 44% of the population (Ericksen et al. 1993, 5). Ahmed and Bernard (1989, 40) and Hossain (1990, 37) contend that during the 1973-87 period, crop losses to drought were almost as severe as the losses attributed to floods. About 2.18 million tons of rice were damaged due to drought in the above period. The corresponding flood loss was 2.38 million tons.
Drought adversely affects all three rice varieties (aman, aus, and boro) grown in three different cropping seasons in Bangladesh.(5) It also causes damage to jute, the country's main cash crop, and other crops such as pulses, potatoes, oilseeds, minor grains, winter vegetables, and sugarcane. Rice alone accounts for more than 80% of the total cultivated land of the country. Droughts in March-April prevent land preparation and plowing activities from being conducted on time. As a result, broadcast aman, aus, and jute cannot be sown on schedule. Droughts in May and June destroy broadcast aman, aus, and jute plants. Inadequate rains in August delay transplantation of aman in high land areas, while droughts in September and October reduce yield of both broadcast and transplanted aman and delay the sowing of pulses and potatoes. Boro, wheat, and other crops grown in the dry season are also periodically affected by drought. Fruit trees, such as jackfruit, litchi, and banana, often die during drought. But the loss of rice production is the most costly damage incurred by droughts in Bangladesh.
The impact of drought spreads disproportionately amongst
regions of Bangladesh. There is a popular impression in Bangladesh
that the northwestern districts of Rajshahi, Dinajpur, Rangpur,
Bogra, and Pabna are particularly drought-prone (Murshid 1987,
38).(6) The northwestern districts are relatively dry,
receiving only 50 inches of rainfall annually. The eastern districts, in
contrast, receive more than 80 inches of rainfall. But drought can
hit both drought-prone and nondrought-prone areas (see Murshid
1987, 38; Paul 1995).
A sample size of 320 households was covered and the heads of the sample households were interviewed with the help of a structured questionnaire (see Appendix). The interview was supplemented by informal post-interview discussion. The questionnaire comprises two parts. The first part contains questions on the extent of crop and other damages caused by the 1994-95 drought and on coping strategies adopted by the drought victims to mitigate the effects of the hazard. The second part focuses on sociodemographic background of the respondent family.
A complete list of all households of the selected villages was compiled and the appropriate number of samples was then chosen from each selected village using a random procedure. Ten trained field investigators conducted the field survey during the month of May in 1995. The PI was in Bangladesh at the time of field survey and participated in the field work. Most field investigators have already earned Master's degree either from the University of Dhaka or Jahangirnagar University. Others are students of one of the above two universities.
Based on land ownership, the latest agricultural census of Bangladesh classified farmers into three categories: small farmers (up to 2.4 acres), medium farmers (2.5-7.4 acres) and large farmers (7.5 acres or more). They account for 62%, 32%, and 7% of the total farmers of the former districts of Rangpur and Dinajpur (BBS 1994, 158). The first category also includes landless households. The proportion of respondents interviewed under three landholding categories differs from the above percentages (see Table 1) because the study area has relatively larger landholding size compared to the other parts of Rangpur and Dinajpur districts. In fact, the average landholding size of both Dinajpur (3.3 acres) and Rangpur (2.3 acres) districts is higher than the national average (2.2 acres) (BBS 1994, 158).
Based on their tenancy status, the respondent households are also classified into two groups: owner farmers and tenant farmers. Irrespective of their landholding size and occupational category, the respondent households are considered owner farmers if they themselves cultivate their farm lands with the aid of family labor and/or hired labor. Tenant farmers are those who, along with their own land, also cultivate lands of others as share croppers, or rent out some of their own lands to others. Table 1 shows that 58% respondent households are classified as owner farmers while the remaining 42% as tenant farmers.
Data on educational attainment indicate that 68% of the total
heads of the sample households were literate. Given the country's
overall literacy rate of 32.4 in 1991 (BBS 1994, 263), the sample
represents a higher literacy rate. It is noteworthy that the male
literacy rate (38.9) in Bangladesh is much higher than the female
literacy rate (25.5%) (BBS 1994, 263). With one exception, all
heads of the households in the study villages were male. Nearly 21%
of the respondents had one to five years of schooling while 47.2%
had more than five years of schooling. Only 9% respondent
households were members of the various registered organizations.
Similar to the rest of Bangladesh, agriculture is the
principal economic activity of the vast majority of the people in
the study area. The main crops of the study area are all three
varieties of rice, jute, wheat, pulses, and potatoes. Boro
rice and other minor crops of the dry season are grown with
the aid of irrigation. Lack of moisture often causes damages to the
other two rice varieties, particularly if no rain occurs during the
growth period of rice plants or at the flowering stage. As a
result, crop damages constitute the major damage caused by drought
Among all the respondent households, ten had no farm land.
Only one of the ten landless households was employed in farming as
pure tenant farmer.(10) Of the 292 respondent
households who either
owned land and/or associated with farming as a share cropper, as
many as 290 experienced crop damage due to the drought of 1994-95.
The damage was so widespread because the 1994-95 drought expanded
over all three cropping seasons and, thus, all three rice varieties
were affected by it. No loss of human life was caused by the
The survey data shows that as many as 15 different crops wereaffected by the drought of 1994-95 (Table 2 ). A large number of crops were damaged because the drought period corresponds with 1995 sowing period of aus, aman, and boro rice, jute, and (summer) chillies; 1994 sowing period of winter crops (e.g., vegetables, potatoes, pulses, til, kaon, onion/garlic, chillies, and wheat); 1994 harvesting period of aman rice; and 1995 harvesting period of boro rice, and winter crops including wheat. The loss in crops was attributed to both decrease in acreage and yield of crops.
Aman was the most affected crop by the 1994-95 drought. Of the 290 respondent households, 281 (97%) reported loss of aman rice (Table 2 ). Like most other parts of Bangladesh aman is the main rice crop of the study area. It accounts for nearly 56% of the total rice acreage of Bangladesh (see BBS 1994, 180). Another reason for widespread damage of aman rice was that both sowing and harvesting periods of the crop correspond with the drought period. The percentages of the total aman acreage damaged ranged from 45% to 100%, with average damage in the vicinity of 75%.
In terms of extend of crop damage, aus rice ranks second followed by jute and boro rice (Table 2 ). Two hundred thirty-two of the 290 respondent households reported loss of aus rice acreage due to 1994-95 drought. Based on the responses of the sample households, nearly 65% and 55% of the aus and jute acreages were respectively damaged by the drought. Other crops damaged by the 1994-95 drought were: wheat, potatoes, kaon, sugarcane, pulses, til, vegetables, onion/garlic, chillies, and groundnut (see Table 2 ). Kaon is a drought resistent minor grain crop while til is one type of oilseed. Sugarcane is an annual crop, sown in October and November and harvested between the following December and March. Two respondent households left their entire farm land fallow during the 1995 sowing period of aus, aman rice, and jute.
The percentage of crop acreage damaged by the 1994-95 drought
as reported by the respondents seems consistent with the reports
published in the national dailies. They claim that the crop yield
in the drought affected areas of the greater Rangpur and Dinajpur
districts was reduced by 60-70% (see Rahman 1995, 8). The decrease
in production was more than the decrease in yields because of
reduction in cropped area. A substantial amount of arable land
remained unsown in the study area due to lack of water.
In addition to crop loss, 178 (59%) respondent households
experienced other types of damages due to the 1994-95 drought. The
extreme heat and lack of moisture caused loss of trees of as many
as 163 respondent households. The respondents lost different types
of trees such banana, mango, bamboo, coconut, bettlenut, and
jackfruit trees. Twenty-two respondents reported that they lost one
or more livestock. Another 18 experienced loss of poultry. Dried up
pond beds caused loss of fish of 9 respondents.
Drought Adjustment Mechanisms
In order to cope with the adverse effects of the 1994-95
drought, the affected households practiced various adjustments at
the household level. Beyond the household levels, they also
received support from both formal and informal sources.
A number of agricultural adjustments practiced during the drought period in other parts of Bangladesh were not reported by the respondent households (see Brammer 1987). One traditional adjustment farmers usually practice if drought occurs during early kharif (March-April) is the conservation of the soil moisture provided by occasional showers. After each shower, farmers quickly plough or handweed their fields. These operations reduce moisture losses by evaporation and evapotranspiration, and prepare the soil to absorb the next shower quickly and deeply. The study area did not experience any rainfall and probably this is why this adjustment technique was not employed.
Resowing of crops was not also reported by the respondent households. This is an adjustment usually practiced if drought occur in April after aus, aman, and jute have been sown. The young plants may die due to lack of moisture. In such a situation, farmers often resow the crops in May and June. As the 1994-95 drought lasted since October of 1994, respondents had limited opportunity to resow crops. Agricultural adjustments to drought are not confined only during the drought period. To compensate loss of crop production, farmers devote more land to crops in the post-drought period.
Since only a few households practiced agricultural adjustments to drought, no attempt is made to explore the relationship between the adjustment and the selected characteristics of the respondent households. The analysis of reasons for not practicing agricultural adjustments, however, suggests that the respondents of the middle and large landholding categories were in a better position to practice agricultural adjustment compared to their counterparts, the respondents of the small landholding class. One hundred twenty- four respondent households reported that they could not practice agricultural adjustment because of financial reasons. Of them 71 (61%) fall in the small landholding size category.
Resowing or irrigating crops requires additional money, which many respondent households, particularly the poor ones, could ill afford during the disaster period. There are two sources of irrigation for crops in the study area. Application of water in the crop field by fetching water from nearby sources, such as wells, tanks, or hand pump tubewells. An alternative is to install a deep or shallow tubewell in the crop field. The former requires additional labor while the latter demands large capital investment. Even if some villagers are financially able to invest capital in a well or could get institutional finance to sink a well, still there is no certainty that they will get water in the well due to decrease of the water table. In fact 124 respondent households mentioned lack of water as the principal reason for nonadoption of agricultural adjustment. This resulted in the decrease of gross irrigated areas. Another 33 respondents did not practice adjustments because they suffered from indecision. A considerable number of respondents gave more than one reasons.
One important point emerged from the extent and type of nonagricultural adjustments made by the respondent households. The 1994-95 drought was a severe one since 72% of the respondents had to sell and/or mortgage out their lands to cope with the hazard. As mentioned above, villagers usually sell and mortgage out land only in extreme circumstances. It is worthwhile to mention that all respondents who had practiced agricultural adjustments also practiced non-agricultural adjustments. This suggests that they were the group most affected by the 1994-95 drought and they were compelled to adopt both types of individual level adjustments in order to ensure their survival against the devastating effects of the 1994-95 drought.
Table 4 presents individual level adjustments by selected socio-economic and demographic characteristics of the head of the respondent households. The X2 statistic in the table shows that the practice of both agricultural and non-agricultural adjustments differs significantly with respect to occupational characteristics of the head of the sample households. Farmers were more likely to adopt adjustment at the individual level compared to businessmen and service holders. Because of their educational attainment and access to government and other sources, businessmen and service holders are in a better position compared to the farmers to receive support from various sources. For this reason, they seem less willing to make individual level adjustments to cope with drought hazard.(13)
Adoption of individual level adjustments to drought also significantly differs according to landholding size and tenancy status of the sample households. As expected, the small and middle farmers adopted adjustments relatively in greater proportion compared to the large farmers. Ninety-six percent of all tenant farmers adopted individual level adjustments. The corresponding percentage is 82 for owner farmers (Table 4 ). More than two-thirds of the tenant farmers are also small farmers, which might explain why they practiced individual level adjustments in larger numbers compared to their counterparts.
Data presented in Table 4 further suggests that illiterate respondents practiced individual level drought adjustments in larger proportion than the literate respondents. This is consistent with the earlier findings reported on occupational categories and adoption of adjustments. Illiterate respondents have the least access to various sources involved in supporting the drought victims and therefore they are compelled to make individual level adjustments to mitigate the effects of the hazard. No statistically significant difference is observed with respect to drought adjustments between member respondents and non-member respondents of institutional organizations.
The items of assistance received from the above six sources were cash loans, food, seeds, and fertilizer. One hundred seven (36%) respondent households received cash loans from government banks, relatives, friends, and NGOs (Table 5 ). Similar to the national government, NGOs' role was limited to provide cash loans to the victims. The respondent households who were successful in obtaining cash loans from formal sources, on an average, received Tk. 9,140.00 (US $228.00). The amount of loans ranged from only Tk. 200.00 (US $5.00) to Tk. 1,00,000 (US $2,500.00). The median per capita household loan received from government sources was Tk. 5,000.00 (US $125.00). The amount of loans provided by friends, relatives, and other villages was much lower compared to the corresponding amount provided by the formal sources. The average amount of money borrowed from the informal sources was only Tk. 2,500 (US $62.50) per borrower household. Local government, friends, relatives, and other villagers were the sources for other items offered to the drought victims. Only 10 respondents received other items of supports.
Table 6 suggests that the support receipt from both governmental and nongovernmental sources differ significantly according to the occupation, land ownership, tenancy, and years of schooling of the respondents. Contrary to the expectation, no difference exists in terms of receiving support from different sources between farmers, and businessmen and service holders. The statistically significant X2 value is found because of the laborer group. Only 11% of all laborers received support from government and non-government sources (Table 6).
But when only governmental sources are considered, the businessman and service holders were the overrepresented groups. As indicated earlier, members of these two groups are more educated and also own relatively more land compared to their counterpart groups. Additionally, they are acquainted with local and thana levels government officers and have regular contact and/or personal relationship with the officers, bank managers, other key officials involved in providing support to mitigate hazard loss. Because of their connections and influence, they not only were overrepresented in receiving government support, but received larger amounts of support compared to their counterparts. The average cash loan received from the government banks was about Tk. 20,000 (US $500.00) for respondent households engaged in business or employed in the service sector.
Since the respondents owning moderate and large landholdings are more influential than their counterparts, the former two groups were better represented in receiving supports beyond the household level (Table 6). The survey data indicates that a considerable number of middle and large landowners rented out some of their lands to tenant farmers. Additionally, some educated respondents who are employed in nonagricultural sectors also rented out lands to tenant farmers. As noted earlier, both rented out and rented in tenants are considered tenant farmers. For this reason, a strong positive relationship is also evident between receiving support and the type of tenancy.
In terms of ability to secure support from various
sources, the respondent households that were members of
institutionalized groups did better compared to their
counterpart nonmembers (see Table 6).
Eighty-two per cent of members of institutional organizations
received support from various government and nongovernment
sources. Only 18% of all nonmember respondents received such
support. The above finding corroborates the contention that the
development of social institutional networks can effectively
lessen hazard efforts. A similar observation was also made by
Haque (1993, 384) in the context of the 1988 catastrophic flood
Droughts are not only climatic, meteorological, and hydraulic,
but also socio-political phenomena. The government has a
responsibility to minimize hardships of the hazard-affected people
by organizing relief work, providing loans, and generating
employment schemes for drought victims. Evidence presented in this
paper suggests that the national government was involved in
mitigating the effect of drought in northwestern Bangladesh only by
providing financial assistant to 21% of the respondent households.
The government did not supply emergency relief goods to the drought
victims and no measure was undertaken to create employment
opportunities for the affected people. Although the daily news
papers published from Dhaka and affected districts contained
reports of drought occurrence, the government did not pay any
attention to these reports. In fact, at the initial stage of the
drought, the government denied the occurrence of such a hazard in
the country. Thus, the government's response to the drought was
late and inadequate. For this reason, 88% of the respondents
practiced household level adjustments and were compelled to draw
more on their household resources to cope with the drought.
Local government and NGOs also played limited roles in
drought-mitigating efforts. Victims also received some support from
friends and relatives living either in their own community or
outside of it. However, the support they received from various
sources not only helped them to survive through the disaster
period, but also assisted them in either not selling or only
selling some of their belongings at nominal prices. In fact, the
support of both institutional and non-institutional sources helped
halt the process of marginization of many victims. But the
experience of the victims should not be catalogued as a complete
While both poor and rich households were affected by the 1994- 95 drought in the study area, the analysis of adjustment strategies adopted to cope with drought situation suggests that the households that belonged to the former socio-economic group suffered the most. Among all households they received the least support from the government and were hurt most from sharp increase of foodgrain prices. This finding contrasts with the studies dealing with the flood adjustment strategies of the farmers in Bangladesh (see Paul 1995) but supports the contention of researchers who use social- historical and political-economic perspectives in studying environmental hazards.
One surprising finding is that irrespective of socio-economic conditions of the affected households, drought victims were able to maintain their consumption pattern of a normal year. This occurred in spite of marked increase in rice prices and lack of widespread availability of consumption credit from the government. During the drought period, people usually consume less amounts of rice and depend largely on famine foods such as coarse wheat bread, vegetables, and wild leaves. Drought victims in the study area did not consume famine foods because the affected area produces surplus rice. Many households consumed food from their previous year's stock which was stored for selling in the market. There was an acute shortage of water for irrigation. But the study area did not suffer seriously for shortages of drinking water. Probably because of low disruption to the consumption pattern and drinking water, the incidence of diseases, such as diarrhoea, dysentery, other intestinal diseases, scabies, skin diseases, and diseases related to nutritional deficiencies was not widespread in the study area. In general, the 1994-95 drought did not cause a severe worsening of physical health status of the people of the affected area.
Had the 1994-95 drought struck other parts of Bangladesh, its
effects would have been more severe. The characteristics of the
people and the affected area helped in dissipating the devastating
effects of the drought. The people of the drought affected area are
wealthier than the people of the rest of Bangladesh. Moreover, the
study area belongs to the region which produces surplus rice.
Although the study area experienced severe shortage of irrigation
water, its drinking water sources were less affected by the
drought. This is because the 1994-95 drought occurred after the
installation of hand pump tubewell, which is the main source of
drinking water in the study area.
Results of a sample survey conducted in Bangladesh to examine
the extent of damage caused by the 1994-95 drought and the
adjustment mechanisms adopted by rural households to mitigate its
effect have been presented in this paper. While drought victims
practiced an array of adjustments to cope with the drought, the
public responses were delayed and inadequate to provide employment
for the affected population and to compensate for the eroded
income. This contrasted sharply with the overwhelming attention and
enthusiasm devoted to controlling floods.
In spite of all the adjustment mechanisms used by the drought victims, their sufferings were substantial, particularly for the socio-economic groups with little or no land, assets, and education. They have few resources with which to buffer themselves against adversity. In order to alleviate worse effects of drought, the government should be prepared for the hazard before it occurs. Projects to be implemented during drought periods should be developed in advance of drought. It is an important lesson gained from this study which can be useful for hazard management programs in other countries, including the United States.
In the virtual absence of empirical research on drought in
Bangladesh, the findings of the present study may provide useful
information on the survival strategies used to combat drought at
the individual and community levels. This information is crucial
for planners, administrators, extension officials, and NGOs to
improve responses to future drought occurrences and thus help to
minimize resulting hardships. The conceptual framework developed to
study drought mitigating techniques adopted by drought victims and
the support they received from various levels can provide important
insights into how humans subject to different levels of
vulnerability respond to an extreme natural event, like a drought.
1. A district is the second largest administrative unit in
2. Natural hazard "vulnerability" is broadly defined as the characteristics of places or people that are likely to be harmed by extreme natural events (Liverman 1990, 50).
3. Droughts also affect livestock by reducing the availability of fodder and grazing lands.
4. The samaj is an informal, predominantly social grouping based on kinship, social, and religious interests of its members.
5. Three rice varieties (aman, aus, and boro) are grown in three different cropping seasons in Bangladesh. Aus and jute are the crops of kharif season (late March to early September), while aman is grown in haimantic season (August to early December) and boro in rabi or dry season (late November to early April). These seasons partly overlap. It is important to note that there are two aman varieties: broadcast aman is sown in April to May and harvested in December, while transplanted aman is transplanted in June to July and harvested in December.
6. At present, the country is divided into 64 administrative districts which have been created from the former nineteen district. The latter districts are also referred to as greater districts.
7. The 1995-94 drought affected 11% of the total area and population of the country.
8. Thanas are the third largest administrative unit in Bangladesh.
9. On behalf of the Principal Investigator (PI), Dr. Maudud Elahi, Professor of Geography, Jahangirnagar University, Savar, Dhaka, also personally visited some of these thanas in order to select sample villages. Dr. Elahi has vast experience in conducting field work in rural Bangladesh.
10. A pure tenant farmer does not own any farm land but cultivates lands of others as share cropper.
11. The cultivation of rice usually requires more water than other crops such as kaon, jute, and wheat.
12. For some, dwindling fodder availability was also a reason for selling livestock.
13. For this reason, they are grouped into one category in order to calculate the X2 value.
Adnan, S. 1993. Living without Floods: Lessons from the
Drought of 1992. Dhaka: Research and Advisory Services.
Ahmed, R. and Bernard, A. 1989. Rice Price Fluctuation and an Approach to Price Stabilization in Bangladesh. Washington, D.C.: International Food Policy Research Institute.
Alam, S.M.N. 1995. Social Science Literature on Natural Disaster in Bangladesh: A Source Book. Dhaka: PACT Bangladesh/PRIP.
Alexander, D. 1995. Changing Perspectives on Natural Hazards in Bangladesh. Natural Hazards Observer 10(1):1-2.
BBS (Bangladesh Bureau of Statistics). 1994. 1993 Statistical Yearbook of Bangladesh. Dhaka.
BBS. 1986. The Bangladesh Census of Agriculture and Livestock: 1983-84: Structure of Agricultural Holdings and Livestock Population. Dhaka.
Brammer, H. 1987. Drought in Bangladesh: Lessons for Planners and Administrators. Disasters 11(1):21-29.
Emel, J. and Peet, R. 1989. Resource Management and Natural Hazards. In New Models in Geography, edited by Peet, R. and Thrift, N., 49-76. London: Unwin Hyman.
Erickson, N.J., Ahmad, Q.K. and Chowdhury, A.R. 1993. Socio- Economic Implications of Climate Change for Bangladesh. Dhaka: Bangladesh Unnayan Parishad.
Ghose, A.K. 1982. Food Supply and Starvation: A Study of Famines with Reference to the Indian Sub-Continent. Oxford Economic Papers 34(2):389-398.
Giambeluca, T.W., Nullet, D. and Nullet, A. 1988. Agricultural Drought to South-Central Pacific Islands. The Professional Geographer 40(4):404-415.
Haque, C.E. 1993. Flood Prevention and Mitigation Actions in Bangladesh: The "Sustainable Floodplain Development" Approach. Impact Assessment. 11(4):367-390.
Heathcote, R.L. 1974. Drought in South Australia. In Natural Hazards: Local, National, Global, edited by White, G.F. New York: Oxford University Press.
Hossain, M. 1990. Natural Calamities, Instability in Production and Food Policy in Bangladesh. The Bangladesh Development Studies. 18(1):33-54.
Jallow, S.S. 1995. Identification of and Response to Drought by Local Communities in Fulladu West District, the Gambia. Singapore Journal of Tropical Geography 16(1):22-41.
Kates, R.W., Ausubel, J.H. and Berberian, M. 1985. Climate Impact Assessment. New York: John Wiley & Sons, Inc.
Liverman, D.M. 1990. Drought Impact in Mexico: Climate, Agriculture, Technology, and Land Tenure in Sonora and Puebla. Annals of the Association of American Geographers 80(1):49-72.
Murshid, K.A.S. 1987. Weather, New Technology and Instability in Foodgrain Production in Bangladesh. The Bangladesh Development Studies 15(1):31-56.
Paul, B.K. 1995. "Flood Research in Bangladesh: Major Findings and Future Research Direction." A paper presented at the 1995 Annual Meeting of the Association of American Geographers held in Chicago, IL, March 14-18, 1995.
Rahman, A. and Biswas, P.R. 1995. Devours Resources. Dhaka
Reardon, T., Matlon, P.J. and Delgado, C. 1988. Coping with Household-Level Food Insecurity in Drought Affected Areas of Burkina Faso. World Development 16(9):1065-1074.
Zaman, M.Q. 1989. The Social and Political Context of
Adjustment to Riverbank Erosion: Hazard and Population
Resettlement in Bangladesh. Human Organization
Table 1. Some Selected Characteristics of the
Heads of the Sample Households
Characteristic Number Percentage
Farming 217 72.1
Service 48 15.9
Business 17 5.7
Laborer 19 6.3
Total 301 100.9
Small 126 41.9
Medium 137 45.5
Large 38 12.6
Total 301 100.0
Owner 175 58.1
Tenant 126 41.9
Total 301 100.0
Illiterate 97 32.2
1-5 yrs. of schooling 62 20.6
Above 5 yrs. of schooling 142 47.2
Total 301 100.0
Yes 28 9.3
No 273 90.7
Total 301 100.0
Table 2. Crop Damage*
Crop Number Percentage**
Aman 281 97
Aus 232 80
Jute 199 69
Boro 147 51
Wheat 39 13
Potatoes 36 12
Kaon 30 10
Sugarcane 24 8
Pulses 17 6
Til 17 6
Vegetables 9 3
Onion/Garlic 8 3
Chillies 7 2
Groundnut 6 2
*N=290. Multiple responses are possible.
**Rounded to the nearest whole number.
Table 3. Distribution of Non-agricultural Adjustments
Adopted by the Respondent Households*
Type Number Percentage**
Sold Livestock 166 55
Sold Land 112 37
Mortgaged land 106 35
Mortgaged livestock 2 1
Sold Belongings 26 9
Moved family members to other area 1 -
*N=265. Multiple responses are possible.
**Rounded to the nearest whole number.
Table 4. Distribution of Respondent Households Who Practiced
Individual Level Adjustments by Selected Characteristics
Yes No Total
Farming 206 (95.0) 11 (5.0) 217 (100.0)
Business* 43 (89.6) 5 (10.4) 48 (100.0)
Service* 9 (52.9) 8 (47.1) 17 (100.0)
Laborer 7 (36.8) 12 (63.2) 19 (100.0)
Total 265 (88.04) 36 (12.0) 301 (100.0)
X2=61.074 (p=0.001) d.f.=2
Small 108 (85.7) 18 (14.3) 126 (100.0)
Medium 127 (92.7) 10 (7.3) 137 (100.0)
Large 30 (79.0) 8 (21.1) 38 (100.0)
Total 265 (88.04) 36 (12.0) 301 (100.0)
X2=6.457 (p=0.040) d.f.=2
Owner 143 (81.7) 32 (18.3) 175 (100.0)
Tenant 122 (96.28) 4 (3.2) 126 (100.0)
Total 265 (88.04) 36 (12.0) 301 (100.0)
X2-15.886 (p=0.001) d.f.=1
Years of Schooling
Illiterate 90 (92.8) 7 (7.2) 97 (100.0)
1-5 49 (79.0) 13 (21.0) 62 (100.0)
>5 126 (88.7) 16 (11.3) 142 (100.0)
Total 265 (88.04) 36 (12.0) 301 (100.0)
X2=6.995 (p=0.072) d.f.=3
Yes 25 (89.3) 3 (10.7) 28 (100.0)
No 240 (87.9) 33 (12.09) 273 (100.0)
Total 265 (100.0) 36 (12.0) 301 (100.0)
X2=0.046 (p=0.831) d.f.=1
*Business and service are aggregated to calculate X2 value.
Table 5. Sources of Support by Items of Support
Source Cash Loan Food Fertilizer Seed Total
Government 63 - - - 63
government - 3 1 - 4
NGO 9 - - - 9
Friend 14 1 - 1 16
Relative 16 2 - - 18
Villagers 5 - - 2 7
Total 107 6 1 3 117
Table 6. Distribution of Respondent Households
Who Received Assistance by Selected Characteristics
Yes No Total
Farming* 86 (39.6) 131 (60.4) 217 (100.0)
Business 6 (35.3) 11 (64.7) 17 (100.0)
Service 19 (39.6) 29 (60.4) 48 (100.0)
Laborer* 2 (10.5) 17 (89.5) 19 (100.0)
Total 113 (37.5) 188 (62.5) 301 (100.0)
X2=9.297 (p=.010) d.f.=2
Small** 31 (24.6) 95 (75.4) 126 (100.0)
Medium 67 (49.9) 70 (51.1) 137 (100.0)
Large 15 (39.5) 23 (60.5) 38 (100.0)
Total 113 (37.5) 188 (62.5) 301 (100.0)
X2=16.601 (p=.001) d.f.=2
Owner 54 (30.8) 121 (69.2) 175 (100.0)
Tenant 59 (46.8) 67 (63.2) 126 (100.0)
Total 113 (37.5) 188 (62.5) 301 (100.0)
X2=7.966 (p=.005) d.f.=1
Years of Schooling
Illiterate 20 (20.6) 77 (79.4) 97 (100.0)
1-5 19 (30.7) 43 (69.3) 62 (100.0)
>5 74 (52.1) 68 (47.9) 142 (100.0)
Total 113 (37.5) 188 (62.5) 301 (100.0)
X2=25.963 (p=.001) d.f.=2
Yes 23 (82.1) 5 (17.9) 28 (100.0)
No 90 (33.0) 183 (67.0) 273 (100.0)
Total 113 (37.5) 188 (62.5) 301 (100.0)
X2=26.191 (p=.001) d.f.=1
*Farming and Laborer are aggregated to calculate X2 value.
**Including landless households.
_________________________________________________________________ District Name: _______________ Thana Name: _________________ Village Name: ________________ Respondent's Name: __________ Sample No.: __________________ _________________________________________________________________ Part l: 1. Do you consider drought to be a serious environmental hazard? Yes: ______ No: _____ 2. Why do you think so? Specify reasons: __________________________ __________________________ __________________________ 3. How often does drought occur in your locality? Once in every ____ years 4. When did the last drought occur in your locality? Specify the year: _________ 5. Were your crops damaged due to the 1994-95 drought? Yes: ______ No: ______ (go to question 8) If answer is yes: 6. What percentages of the total acreage were damaged due to the 1994-95 drought? Aus: ______________ Aman: _____________ Boro: _____________ Jute: _____________ Other Crops (specify): ____________ 7. What adjustments did you practice to reduce crop loss? (specify by crop varieties) _____________________________ _____________________________ _____________________________ Did nothing (specify reasons): ________________ 8. Were you experienced other damages? Yes: ______ No: _______ (go to question 10) 9. List items damaged due to the 1994-95 drough: _______________________________ _______________________________ _______________________________ 10. Did you receive any financial and other forms of support from the government and other sources? Yes: _________ No: ___________ (go to question 12) If answer is yes: 11. What type of support did you receive? (specify by sources): Sources Types of Support (e.g., food, clothes, cash, seeds, housing materials, loans, free labor) Relatives _________________________ Friends _________________________ Other Villagers _________________________ Local Government _________________________ Government _________________________ NGOs _________________________ Others (specify) _________________________ 12. What are the other adjustments (e.g., sale of land, livestock, and belongings, mortgaged land, dismantled housing structure, borrowed money, spent previous savings, family members migrated to other areas) did you adopt to cope with the 1994 drought? ___________________________________ ___________________________________ ___________________________________ Nothing: _______________ Part ll: 13. What is the current family size of your household? ____________ 14. For each member of your household, provide the information (including the yourself): Name Age Sex Yr. of Schooling Occupation __________________ ___ ___ ________________ __________ __________________ ___ ___ ________________ __________ __________________ ___ ___ ________________ __________ 15. What is the landholding size of your family? Total Farm Land Owned: _______ Bigha Farm Land Rented Out: _______ Bigha Farm Land Rented In: _______ Bigha Total Non-Farm Land Owned: _______ Bigha 16. If possible can you tell us approximate monthly income of your family? _________ Takas. 17. Are you a member of institutional group? Yes: _______ No: __________ If answer is yes: 18. What is the name of the group? _______________________________ Thank You!
Return to Hazards Center Home Page