-
2004 Floods in Bangladesh
Damage and Needs Assessment and Proposed Recovery Program
PART 11 - SECTOR ANNEXES
A Joint Report by The Asian Development Bank
and The World Bank
To the Government of Bangladesh January 5,2005
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Administrator31628 V. 2
-
CURRENCY EQUIVALENTS Exchange rate effective September
26,2004
Currency Unit = Taka (Tk) 1 Taka = U S Dollar 0.0 16
1 U S Dollar = Taka 59
Acronyms and Abbreviations
ADB ADP AusAID BADC BMDA BMD BRDB BWDB CI DER DflD DPHE
DWASA ECLAC EMOP FA0 FAP FBCCI FFW FFWC FSS FY GDP GOB IDA E O
IMF I-PRSP LCG LGED
MDG MFA MFDM NBR N C B NGO NHA OCHA PEDP 11 PFDS
Asian Development Bank Annual Development Plan Australian Agency
for International Development Bangladesh Agriculture Development
Corporation Barind Multipurpose Development Authority Bangladesh
Meteorological Department Bangladesh Rural Development Board
Bangladesh Water Development Board Cormgated Iron Disaster
Emergency Response UK Department for International Development
Department o f Public Health Engineering, Ministry o f Local
Govemment, Rural Development and Cooperatives Dhaka Water and
Sewerage Authority UN Economic Commission for Latin America and the
Caribbean Emergency Operation Food and Agriculture Organization o f
the United Nations Flood Action Plan Federation o f Bangladesh
Chamber o f Commerce and Industry Food for Work Flood Forecasting
and Waming Center Flood Forecasting System Fiscal Year Gross
Domestic Product Govemment o f Bangladesh International Development
Association International Labor Organization International Monetary
Fund Interim Poverty Reduction Strategy Paper Local Consultative
Group Local Govemment Engineering Department, Ministry o f Local
Govemment, Rural Development and Cooperatives Millennium
Development Goals Multi-Fiber Arrangement Ministry o f Food and
Disaster Management National Board o f Revenue Nationalized
Commercial Bank Non-governmental Organization National Housing
Authority UN Office for Coordination o f Humanitarian Affairs
Primary Education Development Project I1 Public Food Distribution
System
-
Acronyms and Abbreviations (Cont’d)
PKSF PO PRGF PRSP PWD RHD SME SPARRSO SWAP TA UN UNDP UNESCO
UNICEF USAID UZR UR VGF WB WFP
Palli Karma-Sahayak Foundation Partner Organizations Poverty
Reduction and Growth Facility Poverty Reduction Strategy Paper
Public Works Department, Ministry o f Housing and Public Works
Roads and Highways Department, Ministry o f Communications Small
and Medium Enterprises Bangladesh Space Research and Remote Sensing
Organization Sector Wide Approach Technical Assistance United
Nations United Nations Development Programme United Nations
Educational and Scientific Organization United Nations Children’s
Fund United States Agency for International Development Upazila
Roads Union Roads Vulnerable Group Feeding World Bank World Food
Program
.. 11
-
2004 FLOODS IN BANGLADESH DAMAGE AND NEEDS ASSESSMENT AND
PROPOSED RECOVERY PROGRAM
PART - SECTOR ANNEXES
Preface
This report presents the preliminary assessment o f damage and
needs resulting from the floods that affected Bangladesh from July
through September 2004, and the proposed recovery and long- term
mitigation program. Part I constitutes the main report, while this
volume includes annexes with detailed assessments o f selected
sectors.
The report was prepared by a joint team from the Asian
Development Bank and the World Bank, in collaboration with the
Govemment o f Bangladesh. Team members from the ADB include: Hans
Carlsson (Joint Mission Team Leader), Kamal Ahmad, Faruque Ahmed,
Shamsuddin Ahmed, Stefan Ekelund, Ki Hee Eye, Zahid Hossain, Nao
Ikemoto, Eafiqul Islam, Eezaul Kar im Khan, Jamal Mahmood, S.
Motin, Arun Saha, Ferdousi Sultana, and Tomoo Ueeda. Team members
from the World Bank include: Mohinder S. Mudahar and Enrique
Pantoja (Joint Mission Team Leaders), Nilufar Ahmad, G. M. Khurshid
Alam, Mahmudul Alam, Mi tch Asada, Rafael Cortez, Subrata Dhar,
Paul Dorosh, Eaihan Elahi, Nuran Ercan, Fabio Galli, Chandra
Godavitame, Zahid Hussain, Zahed H. Khan, Kar in Kemper, Qaiser
Khan, Khwaja Minnatullah, Jelena Pantelic, Christoph Pusch, Moh i U
z Zaman Quazi, S.A.M. Eafiquzzaman, and Zoe Elena Trohanis.
To complete the assessment, the joint team worked closely with
GOB and consulted widely with non-governmental organizations, other
bilateral and intemational agencies, members o f academia, and the
private sector. The team also visited several flood affected
districts, including Bogra, Comilla, Gaibandha, Manikganj ,
Maulvibazar, Munshiganj , Mymensingh, Narayanganj , Sirajganj,
Sunamganj, and Sylhet to meet with local authorities and
communities and collect damage information.
The assessment team i s grateful to GOB for i t s considerable
work to produce the preliminary damage assessments for al l o f the
affected sectors, and i ts valuable comments on the draft Report.
The team gratefully acknowledges the comprehensive reports prepared
by the Disaster Emergency Response (DER) o f the Local Consultative
Group (LCG) and other organizations that provided valuable
insights, information and data on the impacts o f the floods.
The interim conclusions o f the assessment were discussed with
GOB at the end o f September, and the draft report shared in mid
November 2004. It should be kept in mind, however, that this i s a
preliminary assessment based on the information available at the
moment o f completion o f the report. Understanding o f the effects
o f the floods can be expected to evolve as more information
becomes available.
iii
-
TABLE OF CONTENTS
PART 11 . SECTOR -XES
ANNEX 1 . MACROECONOMIC IMPACTS
......................................................................................
1 Short-term Macroeconomic Impact
....................................................................................
1 fiscal implications
...............................................................................................................
4 Economic Growth ........ :
......................................................................................................
7 Developing an Appropriate Policy Response
...................................................................
10
A. B . C . D.
ANNEX 2 . SOCIAL IMPACTS
.......................................................................................................
11 A . B . C . D .
Food Security: Availability and Prices o f Food
................................................................ 11
Impact o f the 2004 floods on Assets. Employment. and Incomes
.................................... 13 Government Response: Food
Distribution Programs
....................................................... 14 Special
Needs o f Women and Children
............................................................................
15
ANNEX 3 . RAPID ENVIRONMENTAL IMPACT ASSESSMENT
..................................................... 18 A . B
.
Environmental Impacts
.....................................................................................................
18 Needs and Actions
............................................................................................................
19
ANNEX 4 . HOUSING
....................................................................................................................
20 A . Estimated Housing Damage
.............................................................................................
20 B . Estimated ReconstructionlRecovery Needs..
....................................................................
22 C . Response to Housing Needs
.............................................................................................
22 D . Recommendations
.............................................................................................................
23
ANNEX 5 . TRANSPORT INFRASTRUCTURE
................................................................................
25 Damage Assessment
.........................................................................................................
25 Reconstruction and Recovery Needs
................................................................................
29 Social and Environmental Aspects
...................................................................................
30
A . B . C .
ANNEX 6 . WATER SUPPLY AND SANITATION
............................................................................
31 Damage Assessment
.........................................................................................................
31
Implementation Arrangements and Coordination
............................................................. 33
Environmental and Social Aspects
...................................................................................
34
A . B . C . D .
Reconstruction And recovery Needs
................................................................................
33
ANNEX 7 . PRIMARY AND POST-PRIMARY EDUCATION
............................................................ 35
Damage Assessment
.........................................................................................................
35 Reconstruction and Recovery needs
.................................................................................
36 Preparing for better flood r i sk management
.....................................................................
36
A . B. C .
ANNEX 8 . HEALTH AND NUTRITION
..........................................................................................
37 Response to Health Risks due to the Floods
.....................................................................
37 Damage Assessment
.........................................................................................................
37 Preparing for Better Flood Risk Management
..................................................................
38
A . B . C .
iv
-
ANNEX 9 . WATER RESOURCES MANAGEMENT
........................................................................
39 Reconstruction And Recovery Needs
...............................................................................
40
A . B .
Damage Assessment
.........................................................................................................
39
ANNEX 10 . URBAN AND MUNICIPAL INFRASTRUCTURE
.......................................................... 42
Damage Assessment
.........................................................................................................
42 Reconstruction and Recovery Needs
................................................................................
42
A . B .
ANNEX 11 . TRADE AND INDUSTRY
............................................................................................
44 Damage and Output loss Assessment
...............................................................................
44 Recovery Strategy
............................................................................................................
46
A . B .
ANNEX 12 . AGRICULTURE (CROPS. LMSTOCK. AND FISHER~~ES)
......................................... 48 A . B . C .
Damage and Loss Assessment
..........................................................................................
48 Flood-Related Govemment Programs
..............................................................................
49 Recovery Needs and Strategy
...........................................................................................
50
ANNEX 13 . MICROCREDIT
..........................................................................................................
52 Possible Impact ofthe Floods
...........................................................................................
52 Funding Needs
..................................................................................................................
52 Policy Implications
...........................................................................................................
52
A . B . C .
V
-
ANNEX 1 - MACROECONOMIC IMPACTS
1.1 In addition to causing invaluable losses o f human lives and
physical damage, the 2004 floods might impact efforts to meet some
o f the Millennium Development Goals (MDGs). More specifically,
total assets and output losses due to the floods i s estimated at
around Tk 134 bi l l ion (US$2.3 billion), constituting 4.0 percent
of GDP (see Table 1.1). While flooding significantly affected
assets and outputs, the extent o f the former i s much larger than
that o f the latter (damage to assets was almost 1.5 times as large
as output losses), The private sector suffered almost al l o f the
output losses, with agriculture accounting for nearly half o f the
total losses, and 57 percent o f total asset losses. Housing,
transport infrastructure, and agriculture suffered the most
damage.
Table 1.1: Preliminary Estimates of Asset and Output Loss due to
the 2004 Floods (Tk in million)
Housing
Transport Infrastructure
BWDB’s FCDf i Schemes
Water Supply and Sanitation
Dhaka W A S A
Urban Municipalities
Primary Schools
Post-Primary (Sec. schools and colleges)
Health
Agriculture, Livestock & Fisheries
Industry
Wholesale & Retail Trade
Power
Others (small losses from several agencies)
Total in Tk Million
Total in US$ Million
Percent of (FY04) GDP
Asset Loss 1 Output Loss I
27,500 i i
20,000 1 11,500 1
4,000 i 900 . I 1,400
3,100
2,400
1,800
400
1 1,800
4,000 ’
1,600
900
79,800
26,600
5,600
1 1,000
I 54,700 1,350 1 930 2.4 I I 1.6
i
Total 27,500
3 1,500
4,000
900
1,400
3,100
2,400
1,800
400
38,400
9,600
1 1,000
1,600
900
134,500
2,280
4.0
YO of Total 20.4
23.4
3.0
0.7
1 .o 2.3
1.8
1.3
0.3
28.6
7.1
8.2
1.2
0.7
100
Sources: Based on GOB data & Asian Development Bank I World
Bank Staff estimates. Notes: Table includes public and private
sector losses; the nominal exchange rate i s Tk 59 per U S
dollar.
A. SHORT-TERM MACROECONOMIC IMPACT
1.2 T h i s section focuses on the short-term (2004105)
macroeconomic impact o f the floods on Bangladesh’s economy,
particularly on the balance o f payments, the central government
budget, economic growth, inflation, and food security. Table 1.2
summarizes the impact on key economic indicators.
1
-
Table 1.2: Impact of Floods on Key Economic Indicators
GDP Growth ( percent) Agriculture ( percent) Manufacturing (
percent) CPI Inflation Balance of Payments Exports
FY02 1 FY03 I FY04 I FYO5 4.4 i 5.3 i 5.5 / 5.5 / 5.2 .01 1 3.1
2.7 I 3.3 1 3.0
' 1 I I Pre-Flood , PostFlood
5.5 1 6.8 I 7.4 1 6.2 1 5.3 2.8 , 4.4 I 5.8 1 5.5 I 7.5 $ in
billions 5.9 1 6.5 1 7.6 I 7.9 1 8.4
(Annual percent change)
(Annual percent change) Gross official reserves Reserves in
months o f imports
Imports ! 12.0 I 11.3 I 15.0 1 3.2 / 2.9
-7.6 1 9.5 1 16.1 I 5.1 7.7 i 9.7 10.9 I 11.1 -8.7 1 13.0 12.5 I
10.0 1.6 . 1 2.5 1 2.7 1 3.0 2.1 , 2.9 I 2.7 j 3.0
~
Central government operations Total Revenue Total
Expenditure
Domestic financing Overall budget deficit
Money and credit Broad money Private sector credit
1.3 The balance of payments position may worsen somewhat on
account of the floods. Export growth reached 16.1 percent in FY04,
which was stronger than expected as compared to 9.5 percent in
FY03. Growth was driven by the ready-made garment (RMG) and frozen
food industries. Import growth was also buoyant, mainly reflecting
increases in inputs required by the garment sector. Combined with
continued strong remittances, the extemal current account balance
had a surplus equivalent to 0.1 percent o f GDP in FY04, compared
to 0.6 percent surplus in FY03. FY05 started with improved current,
account surplus amounting to US$200 mil l ion in July 2004,
compared to US$158 mil l ion in July 2003. This reflected 28
percent growth in exports and nearly 11 percent growth in
remittances, which swamped 19 percent growth in imports. However,
for the year as a whole before the flood, the current account was
projected to have a deficit equivalent to 0.9 percent o f GDP, as a
slowdown in export growth was expected with the W A phase-out. The
flood i s l ikely to exacerbate this slowdown.
Percent of GDP
14.9 i 13.7 ' 13.3 15.5 ! 15.4 2.6 I 1.3 , 2.1 , 1.9 , 2.4 End
of year, percent Change 13.1 15.6 1 13.4 j 12.6 1 14.0 13.9 12.6 1
12.0 * 13.3 1 13.8
10.2 , 10.3 1 10.1 1 11.1 1 10.7 4.7 3.4 1 3.2 1 4.3 i 4.7
1.4 A reduced export growth rate and additional flood-induced
imports could result in a widening o f the current account deficit
to 1.1 percent o f GDP in FY05, even after taking into account
increases in remittances' and current official grants in response
to the UN Flash Appeal. Assuming an unchanged pre-flood foreign
exchange reserve target o f US$3 bi l l ion by end-June FY05, the
extemal financing gap could be in the order o f US$SOO mill ion. If
not met by additional assistance, this would create pressure on
reserves. Fortunately, unlike in 1998, this year Bangladesh has in
place a floating exchange rate regime which will absorb the
pressure on reserves. At the end o f June in 1998, Bangladesh's
gross reserves were US$1.8 bil l ion (equivalent to 2.8 months o f
imports), whereas now gross reserves are US$3.1 bi l l ion
(equivalent to 3.7 months o f imports). The Government i s
committed to this regime and should continue to confine i t s
interventions to countering disorderly conditions and building
reserves to desired levels.
Remittances during July-August 2004 had 15.7 percent growth
relative to the corresponding period ofthe previous year.
2
-
1.5 I m p o r t levels are projected to increase, although the
impact of the floods on import demand cannot be easily estimated.
On the one hand, post-flood relief and rehabilitation i s l ikely
to require the import o f selected products, particularly
construction materials (CI sheets, cement, and steel) for making
repairs to road links, rehabilitating damaged embankments and
repairing schools, and health and sanitation facilities. Aided food
imports are also l ikely to be higher than they otherwise would
have been in the absence o f floods. On the other hand, a fall in
real income and import-intensive exports has a depressing effect on
import demand. Historically, however, import growth in dollar terms
has tended to increase in the aftermath o f floods, both in 1987188
and 1998. Import growth in FY05 was projected to decline to 10.0
percent, compared to the 13.6 percent achieved in FY04, because o f
a decline in export growth. Given current trends, this figure i s
likely to be higher at about 15.0 percent.
Box 1: Impact of the Flood on External Sector Indicators Floods
temporarily reduce export growth, increase imports, remittances and
foreign assistance.
Chart 1.1: Export growth (%) I 1-Annual QfOwth -$year moving
average g m ~ ( h
Chart 1.3: RMG Export growth FY90-04
I
40
20
0
-7a -Annual gmWm -&year moving average growth
Chart 1.5: Remittance growth (%)
Chart 1.2: RMG Export growth FY81-89
200 100 h d 0
Chart 1.4: Import growth
50 .
-Annual growth -5-year moving average growth
Chart 1.6: Foreign Assistance growth (%)
3
-
1.6 Export performance i s likely to be affected, but not as
much as after the 1998 floods. Export growth in July 2004 relative
to July 2003 was 28 percent, driven by 27.4 percent growth in
export volume (Box 1). Knitwear exports +creased by nearly 48
percent and woven garment exports by almost 18 percent. Frozen food
exports declined by 0.6 percent. Exports to United States markets
increased by 27 percent and to the European Union (EU) by 28
percent. The floods clearly did not affect July exports. However,
production in many o f the RMG units in Dhaka and Narayanganj were
disrupted by the floods for 3 to 4 weeks. The fisheries and shrimp
sectors also suffered. A number o f export-oriented units suffered
indirectly because their workers were unable to come to the
factories andor because supply o f raw materials was disrupted.
1.7 While export growth has tended to decline in the flood year,
it has historically bounced back the year after. Export growth in
FY99 declined to 3.5 percent, compared to 17.7 percent in FY98
since the 1998 floods made both the production and shipment o f
goods extremely difficult in FY99. A large number o f garment and
knitwear factories were under water for over 8 weeks on average,
Most road and rai l links were cut-off, including the critical
Dhaka-Chittagong route, making the shipment o f goods difficult and
expensive, Some export orders were affected due to buyers’ loss o f
confidence in the ability o f the exporters to meet delivery
schedules.
1.8 The above adversities seemed to have been o f a smaller
scale during the 2004 floods. Because o f the shorter duration and
smaller extent o f flooding, the export-oriented manufacturing
sector was not affected as much in 2004 as it was in 1998. The
garment sector i s more resilient now and can be expected to make
up for the production time lost during July 2004, as it has done in
previous years when the sector was adversely affected by hartah
(strikes) and political agitation. Export growth was projected to
decline from the 16.1 percent achieved in FY04 to 5.1 percent in
FY05 because o f the MFA phase out. Recent trends have been more
positive than expected, however, and overall export growth could
reach at least 12.0 percent this fiscal year.
B. FISCAL IMPLICATIONS
The Budget
1.9 The relief effort, expansion in food-assisted safety nets,
repairs to public property and the impact o f the floods on
economic growth are l ikely to put pressure on both the revenue and
expenditure sides o f the budget. The overall budget deficit can be
expected to rise from 3.2 percent o f GDP in FY04 to 4.7 percent in
FY05, overshooting the original FY05 target o f 4.3 percent o f
GDP. As a result, a financing gap i s l ikely to emerge, given the
ambitious external financing target in the FY05 budget. A redeeming
feature o f the fiscal situation i s that the FY05 budget already
has an ostensible pro-agriculture, pro-poor stance. Increased
budgetary allocations towards supporting agriculture and rural
development and expanding the targeted poverty alleviation programs
are already in place. Full and effective utilization o f these
funds can go a long way towards the rehabilitation o f the rural
economy and restoring the livelihoods o f the flood-affected
poor.
1.10 The experience o f both the 1988 and 1998 floods show that
current expenditures tend to rise significantly in the flood year,
but then growth declines in the succeeding year (Box 2). There i s
some probability that the revenue ,expenditures would overshoot the
overall target despite the allocations that exist in FY05 Revenue
Budget and the FY05 Food Account. These include the Tk 6 bil l ion
provided for farm subsidies, Tk 12 bi l l ion for Unforeseen
Expenditure Management, Tk 4.6 bi l l ion for “Others” Block
Allocations, Tk 2 bi l l ion for the Micro Entrepreneur Development
Fund, and an additional Tk 4 bi l l ion allocated for a food-based
safety net program in the food budget.
4
-
Box 2: Impact of the Flood on Fiscal Indicators Tax revenues are
adversely affected by floods and both recurrent and ADP
expenditures get a boost.
Chart 2.1: Tax revenue growth
--
-Annual growth -&year moving average
Chart 2.3: ADP growth (%)
I 30
20
10
0
-10
-20
Chart 2.2: Revenue expenditure growth
-Annual growth -5-yr moving average Ir.
Chart 2.4 Growth in credit to Central Govt
loo I
-Annual growth -5-yr moving eveage I I -Annual gmwth -5-y moving
average
1.1 1 The Government, however, i s committed to meet
pre-existing, competing demands, including subsidies to electricity
consumption in irrigation and agriculture-related activities,
increased cash incentives (from 25 to 30 percent) for export o f
agricultural products (especially fruits and vegetables), expansion
in programs for the socially disadvantaged, expanded microcredit
programs through Government departments and the Palli Karma-Sahayak
Foundation (PKSF), Pay Commission awards, and the retraining and
creation o f employment opportunities for voluntarily-retired state
owned enterprise (SOE) employees and unemployed garment
workers.
1.12 The floods created additional, urgent demands. The
Government will have to help the flood-affected farmers commence
production. The most immediate need was to re-plant rice for the
winter crops under the Transplanted aman rice cultivation program.
The farmers needed seeds, fertilizer, farm machines and fuel.
Farmers may also use STWIDTW-based irrigation wate? for the winter
crops. In addition to providing 5 kg seed o f aman rice and 25 kg
fertilizers each to 1.9 mil l ion marginal farmers, the Government
needs to ensure adequate supplies o f seeds o f maize, pulses,
wheat, vegetables and bovo so that these are available to the
farmers in time.
1.13 Another area where the farmers need assistance i s with the
health o f their cattle. Thousands o f cattle have already died,
which might create a draught power crisis in the coming season, The
surviving ones in the affected areas are l ikely to suffer from
malnutrition and disease.
STW shallow tubewell; DTW deep tubewell.
5
-
The Govemment needed to provide essential veterinary medicines
and vaccines to the seriously affected areas. The FY05 target for
food distribution through the Public Food Distribution System
(PFDS) i s 1.3 mill ion mt. This will be revised upwards to take
into account the additional food distributed as flood relief. Other
costs l ikely to increase current expenditures include the need to
provide assistance to the manufacturing sector for rehabilitation;
grants to dislocated people; and other increased operations and
maintenance (O&M) expenditures to repair infrastructure
damage.
1.14 The Government would redirect expenditures from the Annual
Development Program (ADP) towards the repair of damaged
infrastructure. Actual ADP implementation i s usually wel l below
the budgeted ADP. Floods tend to reduce the ADP shortfall by
augmenting demand for fast disbursing expenditures. Considering t h
i s and assuming that the Govemment would be able to achieve
significant reallocation o f ADP funds to finance non-food imports
and pay for repairing damages to public assets, it seems reasonable
to assume that the actual ADP expenditures can st i l l be
contained within the FY05 budget target Tk 220 billion.
Implementation shortfall o f the ADP i s normally about 10-15
percent relative to the original budget target, but this tends to
improve in flood years. ADP utilization was almost 100 percent o f
the original budget target in both FY89 and FY90. In FY99, ADP
implementation shortfall dropped to 8 percent relative to the
original budget target, compared with 14 percent the previous year.
Th is year the funds that can possibly be reallocated are
potentially sizeable. The Govemment has indicated willingness to
reallocate 10 percent o f ADP for post-flood rehabilitation and
reconstruction activities. As a result, i t i s therefore
reasonable to assume that the ADP implementation shortfall could
decline to less than 10 percent.
1.15 The flood i s also likely to have an adverse impact on
revenue collections. The FY05 budget revenue targets are ambitious.
The revenue-GDP ratio i s projected to increase from 10.5 percent
in FY04 to 11 percent in FY05, implying a revenue growth o f about
20 percent relative to the FY04 revenue outturn. The largest
increase i s assumed to be from taxes, which are projected to grow
by nearly 23 percent relative to the actual FY04 tax collection
estimate. While this i s not unprecedented, achieving it would
require diligent implementation o f the strategy adopted for the
expansion o f tax base, simplification and rationalization o f the
tax system, increased transparency and efficiency in revenue
administration, creation o f a Central Intelligence Cell in the
National Board o f Revenue (NBR), decentralization o f revenue
administration, reduction o f the discretionary powers o f tax
officials, increase in manpower o f the revenue collection
authorities, and efficient application o f information technology
for strengthening monitoring activities and preventing tax evasion.
Historically, in both 1988 and 1998, growth in revenue collections
declined in both the flood year and the subsequent one. Thus the
challenge o f achieving the FY05 budget revenue target has become
al l the more formidable with the shrinking o f the tax base
resulting from lower than expected growth because o f the floods.
In the absence o f new measures to broaden the tax base, strengthen
the incentive for compliance and plug leakage, i t will be
difficult for the Government to even come close to achieving the
revenue targets.
Money and Credit
1.16 Monetary accommodation may need to continue. Monetary
policy stance started to become somewhat accommodative toward erid
June 2004. However, for the year as a whole, monetary growth
declined from 15.6 percent in FY03 to 13.4 percent in FY04. Private
sector credit grew by 13.3 percent in FY04, driven mostly by
lending o f private banks.3 Monetary growth was targeted to decline
further to 12.6 percent in FY05 under the monetary program
Lending from nationalized commercial banks (NCBs) was restrained
according to the memoranda o f understanding (MOU) with the
individual banks.
6
-
agreed as part o f the Fund’s Poverty Reduction and Growth
Facility (PRGF). Th is figure required adjustment to accommodate
the rise in private sector credit needs. The Government has already
taken steps for increased credit disbursement towards agricultural
rehabilitation. The Finance Ministry has issued guidelines for
Bangladesh Bank, the three NCBs, and two specialized banks to
disburse new credit and suspend realization o f farm loans in the
flood-affected areas for a year, including rescheduling all
outstanding loans and relaxing rules for down-payment by the
defaulters. The NCBs, specialized banks and state-run non-bank
financial institutions increased their agricultural loan
disbursement target for the fiscal year by 25 percent.
1.17 A moderate expansionary stance in monetary management may
thus be desirable for the sake o f economic recovery if the
increased disbursements were to meet the genuine short-run working
capital needs and if the credit was paid back in time. Given the
need to ensure adequate credit for agriculture in the aftermath o f
the flood, a modest increase in the FY05 targets for private sector
credit from 13.4 percent to 15.0 percent, and broad money growth
from 12.6 percent to 14.0 percent should be considered. At the
same
Impact of Flood on Private Sector Credit Growth
70 60 I
I -Annual growth -5-yr moving average fime, the authorities need
to maintain caution to make sure that the suspension o f loan
recoveries i s strictly contained. Given that, a modest upward
revision o f the net lending limit by each N C B from the current 5
percent to 8 percent, in line with the upward revision in private
sector credit growth target, could be considered.
C. ECONOMIC GROWTH
1.18 Although the floods caused enormous suffering,
impoverishment, and extensive damage to real assets in the public
and private sectors, i t s impact on economic growth i s not l
ikely to be dramatically large, provided that the economic recovery
effort i s wel l managed and additional shocks do not occur in the
next few months. Assumptions on the likely growth in the 15
national accounts sectors are based on past flood experience and
recent growth trends (Box 1.3). The overall GDP growth for FY05 i s
projected at 5.2 percent, lowered than the pre-flood projection o f
5.5 percent and the Government target o f 6 percent. The overall
growth impact does not reflect the proportional impact o f the
floods on the poor. While farmers can replant crops and businesses
can recoup losses by working extra shifts, those earning their
livelihoods on a daily basis, such as day laborers, are the worst
affected by the disruption in economic activities.
1.19 The immediate, most visible impact of the floods i s the
direct and indirect effects on the economy’s output. Output losses
in agriculture and industry and disruption in the provision o f
services during the floods directly affect growth. The indirect
effects arise from the damage caused to the stock o f capital,
infrastructure assets, raw and intermediate materials and erosion
and sedimentation o f land. In general, the magnitude o f losses
depends on the extent o f flooding, i t s duration, and water
levels during the floods. The impact o f these losses on growth
depends on the timing o f the floods, the recession o f water, and
the pace o f reconstruction and rehabilitation o f infrastructure,
as well as overall economic management, In comparison to the flood
o f 1998 and 1987188, the floods in 2004 have been less severe in
terms of the extent o f area inundated, duration o f flood, and
persons affected, but more severe in terms o f damage caused to
areas adjacent to major rivers. Preliminary estimates indicate that
direct output losses due to floods in 2004 may have been about 1.5
percent o f GDP, almost entirely in the private sector, Losses
to
7
-
assets may have been in the order o f 2.3 percent o f GDP, o f
which 43 percent i s in the public sector on account o f damage to
infrastructure.
Box 1.3: Impact of the Flood on Overall and Sectoral Growth
Rates
]-A& growth -$yea mwim averagegrowth
Chart 3.1: GDP growth
I
+0' +B" 60' ++ +@ .@ #at -Annual gmMh -5-9 moving average
growth (%)
I
Chart 3.5: Transport & communication I
I ' 0 1 I
O l ....................... 1 $4 +"" *.81 e"" *493 e"" e@
+d"
-Annual growth -.%year moving average gnrnnh
manufacturing growth (%)
cc"' *& *
-
and 30. If the availability and distribution o f seeds,
fertilizer and credit i s managed satisfactorily, there i s a
possibility of partly (if not completely) recovering the damage to
crop sector growth caused by the floods. After the 1998 floods,
crop growth in FY99 was higher than in FY98. Damage to poultry and
livestock were about the same this year as in FY98. Growth in this
sector in FY99 was about the same as in FY98. Pond fisheries were
severely affected, causing losses at the individual level, but not
necessarily to the economy. The fish have been released to open
water where they will breed and grow. Due to the vast flooded area
available as habitat, the actual fish catch may increase after the
recession o f flood water.
1.2 1 According to the Board o f Investment (BOI), damage to
industry this year has not been as severe as in 1998. Knit
industries in Narayanganj and Narsingdi and small industries
(handlooms) in Sirajgong have been affected. Small scale
manufacturing growth i s likely to be hit harder than the large
scale manufacturing sector, as took place both during the 1998 and
1987188 floods. In 1998, over 5,000 small, medium and large
industries were inundated by 4 to 6 feet o f water for about 8-10
weeks. Rapid submersion o f smaller scale units located in rural
and semi-urban areas resulted in closures, damage to machinery, and
production losses.
1.22 The road network was badly affected in 2004, but unlike
during the 1998 floods, neither the Dhaka-Chittagong road link nor
the rai l link was disrupted. Thus, the overall impact on the
transportation sector should not be too negative, The
reconstruction effort could provide a boon to construction, but all
o f i t may not necessarily take place this fiscal year. Floods
reportedly slowed activities in commercial banks, but this will
change when rehabilitation activities start in the rural areas.
Wholesale and retail trade may have been disrupted, but not for a
long period. Loss in growth on account o f disruption during the
floods should at least be partially recovered due to increased
reconstruction and rehabilitation-related trading activities.
Inflation
1.23 Floods have tended to be inflationary in the past, and one
similar jump in inflation was expected in 2004, particularly in
food prices? Although July inflation was not affected by the
floods, by October inflation had reached a six- year high o f 7.9
percent, reflecting an increase in food inflation o f 10.5 ~ e r c
e n t . ~ This i s a pattern similar to the aftermath o f the 1998
floods. Specifically, the year-on-year inflation in June 1998 was
6.8 percent. Inflation decreased slightly to 6.7 percent in July,
jumping to 7.8 percent in August and peaking at 8.4 percent in
Impact of Flood on Inflation
A I
El I -Annual inflation rate -5-yr moving average September. As i
s the case in 2004, these increases were driven mostly by food
price inflation, which was already on the high side before the
onset o f the floods. Containing further food price increases
should be one o f the immediate policy goals.
1.24 Before the floods, the inflation rate, which has been on
the rise during last three years, appeared to be stabilizing. When
the floods started, the year-on-year inflation rate (July 2004 over
July 2003) was 5.6 percent, same as in M a y and June 2004,
although higher than the 5.1 percent
Annual inflation was about 3 percentage point higher than its
long run trend rate (5-year moving average) in both FY88 and FY89.
Comparatively, inflation was 6.2 percent - and food inflation 7.1
percent - in October 2003.
9
-
year-on-year inflation rate in July 2003. Food inflation rate in
July 2004 was 6.5 percent, compared with 6.6 percent in May-June
2004 and 5.9 percent in July 2003. Nonfood inflation was stable: i
t was 4.4 percent in July 2004, compared with 4.4 and 4.3 percent
respectively in May and June 2004 and 4.1 percent in July 2003
D. DEVELOPING AN APPROPRIATE POLICY RTiSPONSE
1.25 The near-term economic prospects o f Bangladesh have
suffered a setback due to the floods. Growth prospects have been
shadowed by the impact o f the devastating monsoon floods. In
addition, the multi fiber arrangement (MFA) phase out at the end o
f 2004 and higher o i l prices represent significant risks for
growth and inflation. Business confidence i s also weakened by
renewed security concerns and increased political confrontations
between the major political parties. Given these adversities, the
Govemment’s immediate policy response should focus on mitigating
the sufferings o f the flood-affected poor and helping them recoup
losses incurred. It should also try to enable the garment industry
to better prepare for the post-MFA era by removing policy
distortions, such as restricting the import o f yam through the
Benapole border.
1.26 The Government would need to revise the medium-term
macroeconomic framework to incorporate the impact o f floods and
the persistence o f higher o i l prices, Fortunately, the impact o
f these shocks are most l ikely to be temporary, implying that the
medium and long-term strategy for poverty reducing growth does not
need to change from what has been envisaged in the structural and
policy reform program agreed under the Fund’s PRGF and the Bank’s
Development Support Credits. Thus, the medium-term fiscal strategy
should remain centered on stronger revenue mobilization and
reorientation o f expenditures to better support growth and the
goals o f the interim Poverty Reduction Strategy Paper (I-PRSP),
while maintaining fiscal sustainability. A cautious monetary policy
i s necessary to balance support for coping with the floods while
containing inflation. Interventions in the foreign exchange markets
need to remain confined to countering disorderly conditions and
building reserves to a more comfortable level, Further
liberalization o f the foreign trade and investment regime remain a
high priority. Implementation o f structural reforms to improve the
performance o f SOEs, NCBs, tax administration and to address
infrastructure bottlenecks to growth and poverty reduction must be
more vigorously pursued. Finally, concerted actions to develop an
anti-corruption strategy and improve the law and order situation
remain top priorities.
10
-
ANNEX 2 - SOCIAL IMPACTS
2.1 Though floods may affect both poor and non-poor households,
poor households are generally least able to withstand the negative
impacts o f floods on prices o f food and other essentials,
household income, individual health and nutritional outcomes. T h i
s Annex covers the impacts o f the 2004 floods on several key
aspects o f social and individual welfare, including food prices
and food security, household income and wealth losses, household
coping strategies, and special concerns related to women and
children.
A. FOOD SECURITY: AVAILABILITY AND PRICES OF FOOD
2.2 The recent floods in Bangladesh have caused less damage to
rice production than the 1998 floods, private sector imports are
helping to stabilize prices, and government stocks are sufficient
for planned distribution.6 Availability o f food (either in terms o
f total availability or government stocks) i s not currently a
constraint on food security. Instead, access to food for poor
households, particularly severely flood-affected households, may be
constrained due to lost production opportunities, loss of assets,
and lack o f employment, which limits their purchasing power.
2.3 In spite o f severe damage to selected areas, the total rice
crop production loss associated with 2004 floods appears to be
relatively small - on the order o f perhaps 1 mil l ion tons: equal
to about 4 percent o f total production in 2003104 o f 26.4 mil l
ion tons. By comparison, the 1998 floods resulted in a 2.2 mil l
ion ton reduction (relative to original targets) to the aus and
aman crops.’ (See Annex Table 2.1.)
2.4 The aus rice crop, typically planted in MarchfApril and
harvested in JulylAugust9 suffered the greatest damage - 400
thousand tons (a 20-25 percent production loss compared to 2003104
production o f 1.832 mil l ion tons), Damage to the monsoon season
aman rice crop i s estimated at about 400 thousand tons o f
broadcast (generally deepwater) aman, with perhaps a 100 thousand
ton loss in aman production due to reduced yields in some areas.
The boro rice crop (planted in DecemberlJanuary and harvested in
May1June) i s l ikely to be somewhat greater than the 2003104
harvest, however, due to , increased soil fertility resulting from
sediment deposited from the 2004 floods.” These rice production
losses to the aman and aus rice crops were particularly severe in
the northeast o f Bangladesh, and along major rivers.
2.5 Government stocks are sufficient for planned distribution.
Due mainly to large-scale procurement o f the boro rice crop in M a
y and June o f 2004 (490 thousand tons in these months),
This analysis i s based on data available in mid-September 2004.
Loss estimates o f the Ministry o f Agriculture as o f September
14,2004 were: aus 394 thousand tons, broadcast aman 335 thousand
tons and transplanted aman 94 thousand tons, totaling 823 thousand
tons. Th is larger 1 million ton estimate allows for additional
damage caused by heavy rains during the week o f September 12, 2004
and regional flooding in the Jessore area in mid-September, It
should be noted that the Ministry o f Agriculture estimates are
subject to reyision. The flood damage to the aus and aman crop was
equal in 1998199 was 11 percent o f the original 1998199 target
rice production o f 19.2 mill ion tons Crop planting and harvesting
times vary somewhat across Bangladesh. The months indicated in this
paragraph are indicative o f the major months o f planting and
harvest.
'%ate, however, that though there was a substantial jump in boro
production following the 1988 and 1998 floods, a part o f this
effect may have been due to increased market prices in these years
and extensive losses of the aman crop. Given that market prices are
not substantially increased.and at flood damage was more restricted
in 2004 than in 1988 and 1998, the increase in boro production may
not be as large as in these latter years.
11
-
gross stocks at the end of the 2003-04 fiscal year (end-June
2004) were 839 thousand tons (net stocks were about 100 thousand
tons less). Given the additional 213 thousand tons o f rice
procured from the boro and aus crops in July and August 2004,
stocks at the end o f August 2004 were even higher: 942 thousand
tons (about 842 thousand tons net). These stocks are more than
sufficient to cover planned distribution from September through
March (less than 700 thousand tons), even without food aid inflows
expected at about 448 thousand tons over this period, Taking into
account the timing o f expected food aid inflows, other
procurement, and distribution, the end-month stocks are expected to
vary between about 580 thousand and 1.1 mil l ion tons between
September 2004 and June 2005.
2.6 Private Imports. Following the 1998 floods, more than 2 mil
l ion tons o f private imports o f rice (almost exclusively from
India) supplied domestic markets and stabilized market prices.
Changes in Bangladesh wheat production, and prices o f grain in
India and in the broader world market have led to a shift in market
incentives, but private sector imports, now mainly o f wheat, are
again a major source o f domestic supply and price stability.
2.7 In 2003104, the private sector imported 797 thousand tons o
f rice and 1.684 mil l ion tons o f wheat, mainly from India. Most
o f this rice trade occurred from July 2003 through March 2004;
from April through August 2004, only about 82 thousand tons o f
rice have been imported (an average o f 16 thousand tons per month
compared with 84 thousand tons per month). Given the bumper boro
crop and relatively small production losses associated with the
flood, domestic prices o f rice remain relatively l ow (14.1 Twkg
national average wholesale, DG Food, 19 September 2004), and there
i s little incentive for larger scale private sector imports as o f
mid- September. A seasonal price rise i s to be expected, however,
and prices could increase to import parity levels, providing
incentives for larger scale private imports of rice."
2.8 Private sector wheat imports are currently several times
larger than are rice imports, due to lower domestic production,
increased domestic food and feed demand, and relative price
changes, Wheat production in Bangladesh has declined steadily in
recent years, from 1.67 mil l ion tons in 2000101 to only 1.25 mil
l ion tons in 2003104 as maize production (mainly destined for the
growing poultry feed indus!xy) has increased. Meanwhile, wheat
demand, both for food and for animal feed (again, mainly poultry),
has increased. India's pricing policies have played a major role,
as well, as export subsidies to private traders exporting Food
Corporation of India excess stocks have encouraged exports. Private
sector wheat imports have averaged 142 thousand tons per month from
April through August 2004, (80 percent o f which came across land
borders with India), These wheat imports can be expected at
approximately these levels barring trade restrictions or shifts in
intemational prices. Government policy has implicit ly promoted
this trade through only minimal tari f fs on rice and wheat
imports. Maintaining these incentives for private sector imports i
s crucial to ensuring adequate availability and keeping market
prices stable.
2.9 Rice and wheat prices in markets across Bangladesh have
generally been stable since the floods began - consistent with the
relatively small size o f the flood-related estimated aus and aman
rice production shortfall and with continuing large-scale private
sector grain imports. Rice prices in some markets rose slightly at
the height o f the floods due to transport and other disruptions,
but returned to approximate pre-flood levels thereafter. Overall,
the wholesale national average price o f coarse rice rose only
marginally (3.6 percent) from 12.62 Tk/kg in June (at the time o f
the boro harvest) to 13.07 Tk/kg. Note that in real terms (adjusted
for inflation
" In 1998-99 private sector imports o f r ice from India
averaged over 200 thousand tons o f rice, and stabilized domestic
rice prices at about 14.5 M g (wholesale, Dhaka).
12
-
using the non-food consumer price index) i s about 30 percent
lower than the September 1998 real price o f coarse rice in 1998-99
(14.2 Tkkg in 1998 prices; 18.1 Tklkg in 2004 prices).
3. IMPACT OF THE 2004 FLOODS ON ASSETS, EMPLOYMENT, AND
INCOMES
2.10 Even with adequate availability o f rice and wheat in
markets at approximately the same prices as before the floods,
household food security and overall welfare i s threatened by
losses o f assets, employment and other incomes. These problems can
be especially acute for poor households, particularly those living
in char areas.
2.1 1 Poverty in Bangladesh i s concentrated in rural areas:
according to the Bangladesh Bureau o f Statistics, 53 percent o f
rural households are poor (compared to 37 percent o f urban
households) and 85 percent o f the country’s poor l ive in rural
areas. The sample size and sampling frame do not permit
statistically valid information at the District level, so it i s
not possible to assess average incomes o f households in
flood-affected districts.
2.12 In rural areas, households living in the Sylhet basin howar
areas o f northeast Bangladesh and near major rivers suffered from
the longest, and generally the most severe floods. Within these
areas, landless laborers and small farmers who lost crops were l
ikely the most affected. Although average agricultural crop and
livestock losses for the year are l ikely to have been small,
individual losses were much larger. Data from a sample survey
following the 1998 flood indicated production losses (including all
crops) for households that cultivated crops averaged 7,500
Tkhousehold, about 48 percent o f expected production. Moreover,
total asset loss among the 55 percent o f rural households in
flood-affected areas in 1998 that lost assets averaged 6,900
Tkhousehold, 16 percent of average household assets. Given that the
onset o f floods in 2004 was reportedly more rapid than in 1998,
loss o f assets per flood-affected household may have been greater,
though the number o f flood-affected households may be less.
2,13 Coping strategies to rescue assets. People living in the
flood-prone areas, especially the chars, have a coping strategy for
saving their assets and themselves during floods. Over the years,
they have learned how to identify precursors to floods - dark
clouds in the north signaling impending heavy monsoon rains and
increasing currents in the rivers. Within the household, men,
women, boys and girls each have responsibilities for saving assets.
Landless, poor families usually do not have many assets and build
their houses with flexible materials (bamboo, grass, straw, mud),
so that when they see water i s rising they are able to dismantle
their houses, take their children and livestock, and move upland.
Men have the responsibility o f dismantling and moving the house
materials, while adolescent boys and girls rescue livestock and
poultry. Women typically keep a packet ready in the machang (a
bamboo platform in the room for storage) with a potable cooking
stove, matches, clothes, and food, and usually leave higher grounds
with the packet, food, and children. Some households move to
embankments and highways, and construct makeshift homes with their
own materials and, plastic sheets. Women will stay with the
children, while men go to nearby towns to look for jobs and relief.
If the flood water i s very high, public buildings, such as schools
and clinics, will be opened as shelters, where people will take
refuge.
2.14 Households with more substantial assets have a different
coping strategy because they have less flexibility than the
landless people. Their houses are built with stronger materials and
they have some furniture, so they usually remain at their house in
order to protect their assets. Women, children and elderly to
relatives and young men may stay either on the machang or on the
roof. Cattle will be taken to higher ground. If fields are flooded
for a long time (2-3 months in 1998), farmers will sow aman paddy
seeds in makeshift seedbeds made out o f water hyacinth,
13
-
jute sticks and mud. Women will sow seeds o f vegetables,
especially creepers in earthen pots and tie them on trees.
Seedlings are planted in the fields when flood waters recede.
2.15 Employment. Loss o f employment i s l ikely to be far less
than it was in 1998, when average monthly wage earnings during
July-October 1998 were 46 percent below their average monthly
earnings one year earlier. Given that ,the duration o f the 2004
floods was approximately one-half that o f 1998, and that the
damage to the aman rice crop and overall economy was also about
one-half the size, labor earnings losses may also have been
proportionately less severe than in 2004.
2.16 Urban households. In urban areas, slum dwellers typically
live in poorly drained areas that suffered from long periods o f
standing water that led to increased prevalence o f diarrheal
disease with particularly adverse effects on children. Flood waters
also reduced employment opportunities and incomes o f these
households. Moreover, though the value o f their housing i s not
large (according to NGO estimates, the cost o f rebuilding a simple
house made o f corrugated iron and bamboo panel walls using new
materials i s only about Tk 11,800 (US$200), loss o f housing
assets bring about severe hardships for poor households.
C. GOVERNMENT RESPONSE: FOOD DISTRIBUTION PROGRAMS
2.17 Government relief programs to provide for the immediate
needs o f flood-affected households consisted mainly o f food
transfers (mainly rice) through the Gratuitous Relief (GR) program.
Survey data from 1998 indicate that GR distribution was
well-targeted to flood-affect, though not necessarily poor,
households in 1998, and most reports suggest that these programs
were in general well-targeted in 2004, as well. These relief
programs involved only a small amount o f resources transferred per
participant, however.
2.18 Two food-assisted programs, Vulnerable Group Feeding and
Food For Work, are major components o f medium-term efforts to
assist the poor who suffered from the floods. In order to increase
access to food by poor rural households, the government’s
Vulnerable Group Feeding program has been expanded by 81 thousand
tons from an initial budget allocation o f about 90 thousand tons
to 170 thousand tons. This increase is, in principle, supported by
food aid resources under the Emergency Operation (EMOP) o f the
World Food Program.” Under this program, a general one-time food
ration o f 30 kgs o f grainI3 i s to be provided to one mil l ion
card-holding households (reaching approximately 5 mil l ion
people). Timing o f this ration i s to be geographically based,
with six districts (Sirajganj, Netrokana, Sunamganj, Habibganj,
Kishoreganj, and Brahmanbaria, al l in the flood-affected northeast
region of Bangladesh) receiving their ration in September, and
districts in various other parts o f Bangladesh receiving their
rations in subsequent months from October through January 2005.
Cards will be allocated according to the same criteria used in the
regular VGF program.
2.19 Because the number o f households that meet the criteria
for VGF exceeds the planned program size, there will be
considerable discretion amongst union parishad authorities
regarding who receives these cards. Nonetheless, data from the 1998
flood suggests that the VGF criteria
”The EMOP also contains resources for expanded Food for Work
from October 2004 to May 2005 through the Local Govemment
Engineering Department (LGED). The Bangladesh Water Development
Board and the Department of Forestry and Environment may also be
involved in special interventions, as well.
I3Given a shortage o f Govemment wheat stocks, but ample rice
stocks, VGF distribution i s likely to be entirely in the form of
rice, with WFP food aid wheat swapped with the GOB on a 1: 1 basis
that reflects the approximate current costs of the two commodities.
In previous years, such swaps typically involved a higher ration of
wheat to rice, e.g. 1.25: 1 that reflected lower costs of wheat
than of rice.
14
-
effectively target the poor, though not necessarily the most
flood-affected among flood-exposed households, 35 percent o f
households in the bottom quintile o f expenditureslcapita received
grain transfers compared to 18 percent and 8 percent in the top two
quintiles. Thus, the 1998 experience suggests that if similar
mechanisms are employed, VGF may be relatively well-targeted to
poor households, though not necessarily to the most flood-affected,
Expanded Food for Work programs will begin mostly in January 2005
after water logged soils have partially drained and heavy earthwork
involved in building roads i s feasible.
2.20 The cost o f the proposed expansion in VGF and FFW will be
met by food aid resources under the Emergency Operation (EMOP) o f
the World Food Program. As o f mid-September, however, only about
US$24 mil l ion o f the US$74 mil l ion in resources has been
committed by donors. Nonetheless, the Government currently has
sufficient stocks for this program. Food aid in kind i s not
necessary for the expansion in program, nor i s there a market
shortfall o f grain. Market prices have remained relatively stable
since the floods (the wholesale national average price o f coarse
rice rose only 3.6 percent between June and August). Given ample
availability o f foodgrain, stable prices and the prospect for
continued private sector imports, cash transfers or cash for work
public employment schemes are a promising alternative to food
transfers in-kind in the coming months. The implication i s that
the major contribution o f EMOP food aid i s as a fiscal support to
targeted transfer programs.I4
2.21 Even with the benefits o f Government and private
transfers, existing microcredit programs, and stable food prices,
many households that suffered from lost labor earnings, illness,
crop losses, and damage to productive assets and housing, resorted
to informal sector loans as a major thrust o f their coping
strategies. Though these loans can be instrumental in preventing
serious reductions in food consumption and health status, the
burden o f these debts i s l ikely to remain long after the visible
effects o f the flood are gone. Following the 1998 floods,
borrowing was also a major coping strategy, with households
borrowing the equivalent o f about 1 month o f consumption. Th is
debt remained essentially intact one year later, however,
suggesting the need for additional mechanisms to provide credit or
retire loans o f poor flood-affected households.
2.22 The Need for Continued Monitoring. To date, the policies o
f the Government o f Bangladesh, and in particular the Ministry o f
Food and Disaster Management, have been highly effective in helping
to ensure sufficient availability o f food. Stocks o f grain are
sufficient for planned distribution, and a policy o f promoting
private sector international trade has helped stabilize market
supplies and prices. Food markets are very fluid, however,
necessitating continued monitoring and analysis to adjust policies
and distribution to avoid further losses in expected production or
disruptions in markets. In addition, because food insecurity
resulting from lack o f purchasing power i s less obvious than
price movements, indicators o f worsening deprivation, such as
increased incidence o f migration, asset sales, and child
malnutrition, should be carefully monitored.
D. SPECIAL NEEDS OF WOMEN AND CHILDREN
2.23 Relief agencies (Government, NGOs, and donors) do not
always adequately address the special needs o f women and children,
who tend to suffer adverse consequences both in the short
l4 In principle, in the absence o f food aid, the Ministry o f
Food and Disaster Management can float tenders for imports o f
wheat. The most recent tender for wheat was largely unsuccessful,
however, (resulting in contracts for only about 10,000 tons out of
50,000 tons in the tender), because o f a lack o f bidders. Since
India has been subsidizing exports and their export prices are
lower than intemational prices, international trading companies may
have been reluctant to submit bids. In this situation, local
tenders or a change o f tender specifications may be alternative
means of government procurement.
15
-
and long term. Relief food (usually rice, wheat flour, pulses)
does not provide essential nutrition for infants and children.
Lactating women may not have milk due to flood related stress and
lack o f nutritious food - so infants and children under two years
o f age become malnourished quickly. Government and NGO nutrition
and feeding programs for pregnant and lactating mothers and young
children oRen stop during floods, with adverse nutritional
consequences. WFP estimates that nearly one mil l ion children
suffered from malnutrition during the 2004 floods because o f
inadequate intake o f appropriate food.
2.24 Lack o f sanitation facilities also causes severe problems
to women at the time o f floods. Reports indicate that majority o f
shelters do not have latrine andlor the latrines were submerged
under water. Elsewhere, even where latrines were available, the
supply was inadequate: one latrine for over 500 families,
Furthermore, during the floods, essential health services
discontinued and women were unable to get their regular
contraceptives supply, potentially resulting in thousands o f
unwanted pregnancies, botched abortions, increased morbidity for
women, and reduced health outcomes.
2.25 Security o f Women, Adolescent Girls and Children. Social
cohesion i s st i l l effective in rural areas o f Bangladesh and,
in general, provided security to women, adolescent girls and
children during the floods. Security in the rural shelters i s
provided by the local leaders (e.g. Union Parishad chairlmember,
school teachers, and religious leaders) and there were no reports o
f robberies or harassment at rural shelters during the 2004 floods.
People living on roads and embanhents, however, were much more
vulnerable. Social capital i s lacking in urban areas, as people
from different districts l ive in slums and usually do not
socialize with each other. As a result, inadequate security o f
women, adolescent girls and children was reported as a problem in
urban shelters, particularly harassment o f women and girls.
2.26 programs for women, children and other vulnerable groups,
including:
Several measures could be undertaken to improve post-flood
relief and recovery
(a) Increase security for women, girls and children in flood
shelters, especially in urban areas. The local govemment
representatives, school teachers, religious leaders and NGOs could
be mobilized and form flood mitigation committees to manage local
flood shelters with the help o f civi l administration and to
provide security with the assistance o f police, village defense
party (VDP), community police, and guards (chowkidars).
(b) Assist people with disabilities. People with disabilities
(PWD) require special attention for their rescue, as well as in
shelters. Rescue workers and shelter volunteers need to be trained
to assist with PMb. Within shelters, mothers and wives o f PWD are
also vulnerable, as they are unable to search for relief or work.
The locations o f shelters should be well publicized so that family
and friends can take people with disabilities.
(c) Special attention to other vulnerable groups, Stigmatized
groups, such as commercial sex workers and transgender people, are
usually driven from shelters andlor harassed if their identity i s
known. Shelter management committees should provide security to
these (and other) vulnerable groups.
16
-
Table 1: Production Losses due to Floods, 1998 and 2004
1998-99 Pre-flood estimateharget Estimated loss Revised
Production Estimate Losseshitial Production Est.
2004-05 Pre-flood estimateitarget Estimated loss Revised
Production Estimate Lossesflnitial Production Est.
Aus, Total Aus Atnan Atnan B O r O Total Rice Wheat
Foodgrain
1.9 9.5 11.4 7.8 19.2 1.8 21 -0.3 -1.76 -2.06 2.25 0.19 0.11 0.3
1.6 7.74 9.34 10.05 19.39 1.91 21.3 -16% -19% -18% 29% 1% 6% 1%
1.80 12.00 13.8 13.20 27.00 1.20 28.20 -0.40 -0.60 -1.00 0 -1.00
0.00 -1.00 1.40 11.40 , 12.80 13.20 26.00 1.20 27.20 -22 Yo -5 % -7
% 0 % -4 Yo 0 % -4 %
Note: 1998-99 audaman loss as share of initial production
estimate: -1 1 percent. Source: del Ninno, Dorosh, Smith and Roy
(2001); Food Planning and Monitoring Unit.
17
-
ANNEX 3 - RAPID ENVIRONMENTAL IMTACT ASSESSMENT
3.1 The field assessment was conducted during 18-21 September
2004 to assess the scope o f sectoral flood damage. Preliminary
environmental impacts o f flood were assessed based on the
observation o f the damaged sites visited during the field
assessment period, discussion with World Bank and ADB team members,
and evaluation o f available documents and other information. The
following i s the summary o f the preliminary environmental impacts
and recommended actions.
A. ENVIF~ONMENTAL IMPACTS
3.2 Riverbank Erosion and Siltation. Most o f the rivers are
morphologically very active. All major rivers such as Meghna,
Brahmaputra, and Ganges, carry huge sediment loads from the large
catchments that l i e in neighboring countries. Shifting o f river
course i s a common problem and contributes to the riverbank
erosion and siltation. The recent flood accelerated riverbank
erosion and siltation in a number o f locations, which was observed
during the field visits. This erosion caused substantial irrigation
loss and displacement o f large number o f people due to loss o f
land, crops, and housing. In addition, siltation caused closing o f
the offtakes o f the distributory channels.
3.3 Soil Erosion. Overall, soil erosion i s apparent along the
river embankments. Soil erosion along the road and railway
embankments was most severe in the areas situated close to major
rivers, where road sections had been completely washed away by the
erosion. The overflow o f the river water, together with strong
currents, contributed to the physical damage o f the embankments
and sedimentation in adjoining rice paddy fields. The continuous
use o f damaged roads, encroachment, debris, uprooted trees on the
roadside, and sandy soil base in the embankment, have al l
exacerbated the soil erosion o f the damaged areas and are posing
safety r isks for communities.
3.4 Loss o f Flood Flows due to Encroachment. Sedimentation and
encroachment that has occurred over the years has contributed to
the aggravation o f flood damage. Lakes, wetlands, and natural
drainage have been encroached upon by earth filling, conversion to
agricultural land, depositing o f city garbage and construction o f
buildings and roads, As a result, natural water retaining ponds
have disappeared and many o f the drainage channels lost their
original conveyance capacity, which has caused stagnation o f flood
flows. The flood and resulting inundation o f the land caused
socioeconomic impacts such as loss o f crops and agricultural land,
livestock, housing, and the interruption o f industrial production
activities.
3.5 Water Contamination and Health Impacts. The impact o f the
flood on water quality has posed major health concerns. Due to
mismanagement o f collection and disposal o f solid waste in urban
areas, especially in Dhaka, considerable amounts o f waste were
disposed in the storm drainage system, which caused blockage o f
storm drains and overflow o f the storm water. Urban environmental
facilities including production wells, tube wells, pit latrines,
and ground level water reservoirs in the flooded areas were
submerged. The drinking water in the flooded areas i s then
contaminated by flood water and industrial storm run-off. Due to
lack o f adequate water and wastewater treatment facilities, the
contamination o f surface and ground water, particularly in the
urban areas, spreads diseases and has become a major public health
concem, particularly for children.
18
-
3. NEEDS AND ACTIONS
3.6 Environmental Assessment Framework. The needed immediate
response to flood damage i s to restore andlor improve the damaged
facilities with flood protection measures. While rehabilitation and
protection work will result in major positive environmental
impacts, some negative impacts may occur due to the site selection
and during the construction. To ensure that environmental
consideration i s taken into account the rehabilitation and
protection work, an appropriate environmental assessment framework
should be developed in accordance with the environmental assessment
requirements o f the Government. The EA framework should include
the following environmental assessment requirements: the
development o f environmental site selection criteria; preparation
o f environmental assessment reports; review and clearance;
institutional responsibilities; estimated budgets for environmental
mitigation measures and monitoring; and staffing and training
needs. I
3.7 Environmental Monitoring. There are a several environmental
monitoring needs which are required to: (i) test the drinking water
quality; (ii) monitor the efficacy o f the erosion protection
measures for embankments; and (iii) monitor river hydrology and
morphology, While it should be inspected urgently, well water used
for drinking purposes should undergo ongoing water quality tests to
check the health impacts and reduce health risks. Soil erosion
monitoring should be conducted during the rehabilitation work and
also as part o f ongoing monitoring. As a long-term monitoring
initiative, online monitoring o f river erosion and platform
changes, as well as morphological changes, should be developed and
institutionalized among relevant agencies in order to utilize the
flood information database in their design and installation
structure o f roads, railways and bridges, and for planning flood
preparedness activities.
3.8 Community Disaster Preparedness and Management. Due to poor
socioeconomic conditions, the poor mostly l ive in the
disaster-prone areas and are more vulnerable to the effects o f
disasters. Since the poor depend on the natural resources, they
cannot recover from the damage o f natural disasters only on relief
resources. Awareness building o f the flood related environmental r
isks i s important to understand the linkage between environmental
degradation and natural disaster. Enhancement o f people’s capacity
to protect and manage natural resources and ecosystem i s essential
to cope with natural disasters and sustain their livelihoods.
Community-based awareness and capacity building activities should
include not only emergency response activities, but also proactive
community-based environmental management approaches.
19
-
ANNEX 4 - HOUSING
4.1 This housing damage assessment i s based on damage reports
provided by the Ministry o f Food and Disaster Management (MFDM);
the. most recent figures were given on September 21, 2004. Further
information was available through other sources and reports,
including the Disaster & Emergency Response (DER)
Sub-group.
4.2 In addition, the joint A D B M mission visited the
Narayanganj and Manikganj Districts between September 18 and 19,
2004 to review the damage and discuss with local officials the need
for reconstruction and recovery.
A. ESTIMATED HOUSING DAMAGE
4.3 Reported Damage. The 2004 flood severely impacted the
housing stock and left hundreds of thousands of homes destroyed or
heavily damaged. According to MFDM, 39 districts were affected with
total o f 896,000 un i ts fully damaged and 3.39 mil l ion uni ts
partially damaged,
4.4 The flood washed away homes, particularly at the riverbanks.
Due to the severe erosion, not only the home disappeared but also
the land was washed away. Many o f the affected people took shelter
in their relatives’ homes or in schools, religious institutions, or
other community buildings. A considerable portion o f flood victims
are st i l l occupying the temporary shelters on the flood
protection embankments. Lack o f access to safe drinking water and
lack o f sanitary infrastructure within flood shelters contributed
to the critical condition. In rural areas, villagers who did not
leave their homes built bamboo platforms at a higher level or
raised the height o f their beds by tightening them with bamboo and
wooden poles.
4.5 Housing types. Three different type o f housing can be found
in the flood affected areas:
(a) Kutcha houses are mainly constructed o f corrugated iron
sheets or bamboo. Structural support material used for this type i
s bamboo or wood. Kutcha houses have no foundation and some o f
them are constructed on an elevated mud base or bamboo columns to
offer a minimum protection against raising water. Due to the
structural weakness, this type o f housing does not provide a high
degree o f protection against floods.
(b) Semi-pucca houses use a combination o f materials such as
bricks, iron sheets and concrete. I t can be observed throughout
the region that the use o f low quality construction material, and
the lack. o f structural binding material and proper bracing,
increases this housing type’s vulnerability to floods.
(c) Pucca houses are in general masonry structures, with mainly
reinforced concrete foundations. These houses have either load
bearing brick walls with bracings or reinforced frames. The
construction quality seems to be weak. However, compared with the
previous described housing types, Pucca houses offer a higher
degree of flood protection. Flood damage was reported mainly with
regard to the Kutcha and Semi-pucca housing.
20
-
Pre-flood Asset Value
Type of Damage Unit Number Average Value Total (Thousands) Taka
Million Taka
4.6 Heavily Damaged Units. The average construction cost o f new
homes varies widely throughout the region. According to the
information provided by local officials during the site visits and
the latest damage reports prepared by the Disaster Emergency
Response (DER) sub- group, the average replacement cost o f a
minimum standard house i s about Tk 16,500 +/- 7,000. In addition,
a depreciation factor o f 20 percent was introduced to determine
the pre-flood value. Thus, the average pre-flood asset value i s
estimated at 13,200 TWunit. Th is amount was confirmed with local
NGOs, which determined the pre-flood asset value to be in the range
o f Tk12,OOO to 15,000.
Total Million US$
4.7 Partially Damaged Units. The flood affected the stability o
f several houses by washing out or decreasing the bearing capacity
o f the soil beneath the foundation, causing the homes to either
fall down or tilt sideways. Partially damaged houses maintained
their structural integrity, but the non-structural elements were
affected. Thus, damage i s specific to the individual unit. In
order to calculate a damage estimate, a ratio o f 25 percent o f
the average replacement cost has been applied in this assessment
for partially damaged housing units.
Heavily Damaged Units Partially Damaged Units Loss o f
House-hold Goods
Loss of Goods 4.8 The houses visited during site visits revealed
that the majority o f the damaged homes met minimal human needs and
did not have in-house water and sanitation facilities, Especially
in rural areas, homes did not have washing machines, TVs,
refiigerators, etc. Although the houses contained extremely basic
necessities (spread beds, cloths, few kitchen goods), these goods
must be replaced. The value o f these goods may be estimated at Tk
1,500 for a family living in a heavily damaged house.
896.0 13,200 11,827 200.5
3,391 .O 4,125 13,988 237.1
896.0 1,500 1,344 22.8
Total Estimated Damage 4.9 Based on the calculation assumption
above, the total damage i s estimated as follows:
Sum
Table 2: Housing Damage Estimate
27,159 460.4
4.10 Given the high uncertainty in the number o f damaged units,
a damage ratio has been introduced. Experience f iom other
assessments carried out by the team has shown that the damage
reports prepared without a detailed validation process result in a
20 and 30 percent overestimation, in some cases even 50 percent. It
was thus decided to present the init ial damage estimate in a range
between -20 percent and +5 percent. The high case o f 5 percent has
been introduced to reflect the additional damage caused by the
September flood. Based on these assumptions, the total damage i s
estimated between US$370 mil l ion and US$480 million.
21
-
Table 3: Range of Estimated Damage
Heavily Damaged Units Partially Damaged Units
" ...........
1 High Case +5 percent I Base 1 LOW Case -20 percent
Unit Number Average Value Total Total (Thousands) Tk Million Tk
Million US$
896 19,800 11,827 301
3,391 4,125 13,988 237 "" ......... " ........ I" ....... "" "
.... ....... l..".l l.l.l ..... "" ........ " .....I..."....I.."
.... ",..I ".."....."......".I...I
US$370 mill ion 1 US$460 million US$480 mill ion I
Sum
B. ESTIMATED RECONSTRUCTION/RECOVERY NEEDS
25,815 538
4.11 Reconstruction Costs. The condition o f the housing
buildings i s poor. During the reconstruction scheme, at least to
some extent, safety measures should be considered in the designs
and the construction methods to reduce the vulnerability o f the
houses to future floods and other hazards.
Low Case -20 percent
US$430 Mil l ion
4.12 The additional cost o f these measures depends on the level
o f improvement to be introduced. At a minimum, the buildings
should be reconstructed on a relatively elevated foundation,
resistant to certain level o f scouring affect o f flood. The cost
o f such improvement might be anticipated to increase the
reconstruction cost o f Tk 16,500 by 20 percent. Th is results in
an average reconstruction cost o f Tk19,800. However, the repair
cost would remain the same.
Base High Case +5 percent
US$S38 Mil l ion US$S60 Mil l ion
Table 4: Estimated Reconstruction Cost (including improved
design standards)
4.13 Following the rationale presented in section B, a range
between -20 percent and +5 percent should be considered to reflect
uncertainty, particularly with regard to the reported damaged
units. Thus, the total reconstruction cost i s estimated between
US$430 mil l ion and US$560 million.
C. RESPONSE TO HOUSING NEEDS
4.14 Repair and reconstruction o f damaged homes has been
supported by the Government's recovery program. As part o f the
flood rehabilitation program, the Ministry o f Food and Disaster
Management distributed 10,800 bundles o f corrugated iron (CI)
sheets and Tk 540,000 (US$91,525) in the affected districts as
house building grants. Additional distribution o f C I sheets
22
-
i s foreseen in the amount of Tk 200 mill ion (US$3.4 million).
UNDP i s providing an additional 100,000 bundles o f CI.
4.15 In addition, a Disaster Risk Mitigation fund was
established to support victims who have lost their income
generating assets due to the flood disaster. An amount o f Tk 750
mil l ion (US$12.7 million) has been allocated to this fund under
the current budget. In the United Nations Bangladesh Floods Flash
Appeal 2004, family shelter and non-food item needs total US$34
million.
4.16 International NGOs, such as CARE, CONCERN, and Oxfam, have
also provided emergency shelter. CARE-Bangladesh has provided
5,OOOMT o f plastic sheeting, Oxfam has provided
reconstructiodrehabilitation support to 16,000 flood affected
households in Sunamganj and Gaibandha Districts (as o f the end o f
July) and plans to extend i ts support for the
reconstructionhehabilitation o f another 7,500 homes in Faridpured
and Rajbari.
D. RECOMMENDATIONS
Short-term 4.17 Although the exact number o f affected
households cannot be verified, the demand for housing repair and
reconstruction i s large. Aside from supplying building materials,
the Government should consider offering assistance targeted to the
poorest flood victims. A program could be created following the
criteria o f the VGF program, providing cash grants to those who
lost their homes and other assets. The cash grants would allow
beneficiaries to decide directly on how best to restore their
livelihood and homes. However, technical guidance i s also
recommended.
Mid-to long-term 4.18 The incorporation o f multihazard risk
mitigation measures in the design o f infrastructure systems,
housing, health, and education facilities i s essential to reducing
damage to the built environment.
4.19 A mechanism for housing reconstruction should be considered
to allow affected households to receive financial assistance soon
after a disaster. The funding scheme could combine micro-credit
with a grant element to make the financing affordable. The maturity
o f the loan as well the collateral issue would need to be
addressed. Furthermore, a financial r i s k mitigation mechanism
should be incorporated to strengthen the sustainability o f such a
program.
4.20 Further support should be given to NGO and local
initiatives that have introduced innovative design approaches for
minimum standard housing. The design increases the structural
integrity and allows easy dismantling o f the house in case o f a
flood, thus reducing future vulnerability.
23
-
3 5 Jaipurhat 36 Barisal 9 37 Noakhali - 38 Kushtia 39 Laxmipur
7
TOTAL 896
24
4 4 103 112 17 17 4 4 19 26 3,391 4,287
-
ANNEX 5 - TRANSPORT INFRASTRUCTURE
5.1 The preliminary assessment o f the transport sector covers
road, bridge, ra i l and inland water transport infrastructure. The
joint team reviewed field data and preliminary assessment reports
prepared by the Roads and Highway Department (RHD), Local
Government Engineering Department (LGED), Bangladesh Railway (BR),
Dhaka City Corporation (DCC) and Ministry o f Shipping (MOS). The
team visited the Districts o f Bogra, Brahmanbaria, Habiganj,
Jamalpur, Mymensingh, Natore, Sirajganj, Sunamganj, and Sylhet to
assess flood-related damage.
A. DAMAGE ASSESSMENT
5.2 The transport sector, specifically the road sector, was the
most seriously affected o f the infrastructure sectors. The 2004
floods caused extensive damage to the primary road network, managed
by the RHD, and to the secondary road network, managed by the LGED.
About 2,970 km, or 14 percent, o f the primary road network was
affected by the floods. Damage was particularly severe on some road
sections because incidences o f flash flooding led to wave action
and erosion o f many embankments. About 25,000 km, or 25 percent, o
f the secondary road network comprising Upazila Roads (UZRs) and
Union Roads (URs) , were affected by the floods. Many UZRs and U R
s were found to have failed sections and structures that were
washed away. This was partly due to l ow design standards. Dhaka
division suffered the most damage in the road sector, including
urban roads. Railways suffered damage on several network sections.
Inland waterways and airports were marginally affected by the
floods, according to information received from the Government. The
team noted that insufficient maintenance adversely impacted and
increased the damage on road and railway infrastructure.
5.3 The preliminary cost o f asset damage caused by the 2004
floods to the transport infrastructure sector has been assessed at
about U S 2 6 0 mil l ion (Tk 15 billion). The asset replacement
cost i s estimated to be U S 3 4 0 mil l ion (Tk 20 billion) to
US$380 mil l ion (Tk 22 billion), depending on the engineering
standards adopted during the reconstruction phase.
Road Sector Infrastructure
5.4 Road transport infrastructure suffered extensive localized
damage in al l o f the 39 flood- affected districts. During the
field visits, the team observed: (a) damage to'road pavements; (b)
damage to bridges and culverts; (c) damage to embankments; (d)
total failure o f embankments including pavements; and (e) total
failure o f bridges and culverts. The damage was caused by a
combination o f factors, including heavy persistent rainfall
combined with flooding, and in some cases, flash flooding.
5.5 The institutions (RHD, LGED and DCC) responsible for most o
f the country's road infrastructure provided the mission with their
preliminary assessments. Total damage was estimated at about US$236
mil l ion (Tk 14 billion).l5 The breakdown o f the estimated damage
by type o f road asset i s as follows: (a) National Highways (NHs),
Regional Roads (RHs), District Roads (DRs) and related structures -
about US$111 mi l l ion (Tk 6.5 billion); (b) WZRs, U R s and
related bridgeslculverts - about US109 mil l ion (Tk. 6.4 billion);
and (c) Dhaka City urban roads - about US$16 mil l ion (Tk 944
million).
"This amount does not include the damage caused by the flooding
to Village Roads A and B, which are the responsibility o f Local
Government Institutions (LGIs).
25
-
Primary Road Network
zone
5.6 RHD i s responsible for the management o f the primary road
network, which i s broken down into 3,473 km o f NHs, 4,254 km o f
RHs, and 13,724 km of DRs, totaling about 21,45 1 km of roads, The
preliminary flood damage assessment for the RHD road network i s
summarized as follows:
Affected road Affected structures Total damage Assessed
Replacement length bridgeslculverts (US$ million) Cost
Affected Road Length: 2,970 km Affected Number o f Bridges:
171
= Affected Number o f Culverts: 135 Affected Number o f River
Jetties: 26
Dhaka Comilla Rangpur
Barisal Total
Rajshahi
5.7 The R H s and DRs sections o f the primary road network
suffered more extensive and structural damage than the N H s . This
i s due to their much