WPS5238


Policy Research Working Paper                       5238




        Assessing Ex Ante the Poverty
and Distributional Impact of the Global Crisis
          in a Developing Country
                 A Micro-simulation Approach
                 with Application to Bangladesh

                                 Bilal Habib
                              Ambar Narayan
                               Sergio Olivieri
                          Carolina Sanchez-Paramo




The World Bank
Poverty Reduction and Economic Management Network
Poverty Reduction and Equity Group
March 2010
Policy Research Working Paper 5238


  Abstract
  Measuring the poverty and distributional impact of the                             with pre-crisis household data. The approach is then
  global crisis for developing countries is not easy, given the                      applied to Bangladesh to assess the potential impact of
  multiple channels of impact and the limited availability                           the slowdown on poverty and income distribution across
  of real-time data. Commonly-used approaches are of                                 different groups and regions. A validation exercise using
  limited use in addressing questions like who are being                             past data from Bangladesh finds that the model generates
  affected by the crisis and by how much, and who are                                projections that compare well with actual estimates from
  vulnerable to falling into poverty if the crisis deepens?                          household data. The results can inform the design of
  This paper develops a simple micro-simulation method,                              crisis monitoring tools and policies in Bangladesh, and
  modifying models from existing economic literature,                                also illustrate the kind of analysis that is possible in other
  to measure the poverty and distributional impact of                                developing countries with similar data availability.
  macroeconomic shocks by linking macro projections




  This paper--a product of the Poverty Reduction and Equity Group, Poverty Reduction and Economic Management
  Network--is part of a larger effort in the department to analyze the poverty and distributional impact of macroeconomic
  shocks.. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be
  contacted at csanchezparamo@worldbank.org and anarayan@worldbank.org.




          The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
          issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
          names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
          of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
          its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.


                                                        Produced by the Research Support Team
                                                                           

    Assessing Ex Ante the Poverty and Distributional Impact of the Global 
                        Crisis in a Developing Country 
                       A Microsimulation Approach with Application to Bangladesh 
                                                                           
                      Bilal Habib, Ambar Narayan, Sergio Olivieri and Carolina SanchezParamo* 
        

        

        

        

        

        

        

        

        

        

        

        

        

        

               Poverty Reduction and Equity Group, Poverty Reduction and Economic Management 
                                                               World Bank, Washington DC 
                                                            
*
   This  paper  is  the  first  in  a  series  of  country  notes  to  be  prepared  under  the  work  program  on  Analyzing  the 
Distributional Impact of the Financial Crisis, in the Poverty Reduction Group of the PREM Network. We gratefully 
acknowledge  the  data,  inputs  and  comments  received  from  the  Bangladesh  PREM  Team  of  the World  Bank  ­  in 
particular  Zahid  Hussain,  Sanjay  Kathuria,  Lalita  Moorty,  Nobuo  Yoshida  and  Sanjana  Zaman.  We  also  thank  the 
participants  at  a  seminar  organized  in  South  Asia  Region  of  the  World  Bank,  Washington  DC  for  their  insightful 
comments on the interim results of this work.  

                                                                                                                                   
 
       1. Introduction 

What began as a financial crisis in a few industrialized countries is quickly turning into a job crisis, with 
contractions  in  growth  taking  their  toll  on  developed  and  developing  countries  alike.  The  projected 
economic  slowdown  across  the  world,  recent  World  Bank  global  estimates  for  poverty  suggest,  could 
add an extra 50 million to the number of people living below $1.25 a day and 64 million to the number 
below $2 a day, compared to a scenario of uninterrupted growth in the absence of the crisis. By 2010, 
the crisis is projected to add an estimated 89 million to the number of people living below $1.25 a day 
and  120  million  to  the  number  below  $2  a  day  (Chen  and  Ravallion,  2009).  Along  with  the  impact  on 
poverty numbers, the crisis is also likely to have significant impacts on the distribution of  income and 
consumption among the poor and nonpoor, both within and between countries.  

Evidence  from  previous  crises  suggests  that  the  output  elasticity  of  wages  tends  to  be  larger  during 
downturns  than  during  booms,  and  that  relative  inequality  falls  about  as  often  as  it  rises  during 
aggregate  contractions  (Paci  et  al,  2008).  Further,  the  crisis  is  rapidly  shifting  across  countries  ­  via 
trade, financing, and remittances ­ as well as within countries ­ via adjustments in domestic credit and 
labor  markets  and  fiscal  policies.  As  a  result,  it  is  difficult  to  predict  the  distributional  impacts  of  the 
crisis with a high degree of confidence. 

There  are  some  hypotheses  about  how  the  impact  of  the  crisis  is  likely  to  evolve  between  different 
groups  in  developing  countries.  The  emerging  consensus  is  that  the  initial  impact  is  likely  to  be  seen 
mainly  on  the  emerging  middleclass,  since  they  are  more  likely  to  be  employed  in  exportoriented 
industries  and  salaried  jobs  in  the  services  sector,  which  appear  to  have  suffered  the  largest  labor 
market  shocks.1  Approximately  90%  of  households  entering  the  middleclass  during  19902005  joined 
the  lower  tier  ($29/day)  (Ravallion,  2009),  and  thus  risk  falling  back  into  poverty  if  they  face  large 
employment or earnings shocks. The initial impact on the poor may be limited due to the very isolation 
from  the  global  markets  (and  the  formal  sector  that  gains  the  most  from  such  linkages)  in  many 
countries that have prevented them from exiting poverty in the past.  

As the crisis unravels, however, the poor in developing countries are likely to be increasingly impacted. 
With  labor  markets  in  the  formal  sector  being  affected,  job  opportunities  and  wages  are  likely  to  fall, 
pushing more people into the informal sector, which could depress earnings in the informal sector. This 
can be accompanied by reverse migration from urban to rural areas, increasing the burden on poor rural 
households and drying up remittances from workers previously employed in the formal sector, leading 
to higher and deeper poverty.  

The  distributional  impact  of  the  crisis  is  thus  likely  to  be  complex  and  dynamic.  Some  of  the  key 
questions an analysis of distributional impacts would need to address are: how are the impacts going to 

                                                            
1
  There is little consensus in the development literature about who the "middle class" are. A number of different 
definitions have been used to study the evolution of the middleclass, such us Blackburn and Bloom (1987), Beach 
(1989),  Levy  and  Murnane  (1992),  Jenkins  (1995)  and  Burkhauser  et.  al  (1999).  For  our  purpose,  we  define  the 
middleclass  broadly,  as  those  above  the  poverty  line  but  below  the  richest  decile  (10%)  of  the  population.  In 
Bangladesh, this is the group between the 40th and 90th percentiles of per capita expenditure or income.  

                                                                                                                            2 
 
be shared across the distribution of income or consumption, which sectors, areas and regions are likely 
to  be  impacted,  and  what  are  the  characteristics  of  those  who  would  become  poor  as  a  result  of  the 
crisis? In order to be useful to policymakers in countries, these questions would have to be analyzed ex 
ante, with available data that in most cases predate the crisis, rather than be delayed till the time post 
(or during) crisis data become available. Even in cases where some realtime data is available from crisis
affected  sectors  or  countries,  an  ex  ante  approach  can  be  useful  to  simulate  future  impacts  for 
hypothetical  scenarios  that  are  not  available  from  realtime  data,  especially  in  countries  where  the 
situation on the ground is changing rapidly. 

Current  approaches  to  analyze  ex  ante  the  impact  of  a  macroeconomic  shock,  with  the  limited  data 
available in most countries, are somewhat inadequate in addressing the kind of questions posed above. 
To  improve  upon  existing  approaches  given  the  typical  data  constraints  seen  in  most  developing 
countries,  we  develop  in  this  paper  a  microsimulation  model  to  evaluate  exante  the  distributional 
impacts of the crisis. Section 2 outlines a rationale for building such a model, including the value added 
from  this  approach  compared  to  existing  methods.    Section  3  presents  the  model,  starting  with  an 
overview  of  the  approach  and  subsequently  outlining  each  step  in  detail,  and  ending  with  the  key 
limitations and caveats to the model. In Section 4, we discuss a country case study with an application of 
our model (the case of Bangladesh). Section 5 concludes with implications of the results for Bangladesh 
and possible areas for extension in the course of future work. 

    2. Rationale for our approach 

Currently,  there  are  two  approaches  commonly  used  within  the  World  Bank  to  assess  the  welfare 
(primarily  consumption  or  income  poverty)  impacts  of  the  crisis:  the  output  elasticity  of  poverty 
method, and PovStat (World Bank PovertyNet). The elasticity method involves using historical trends of 
output  and  poverty  to  determine  the  responsiveness  of  poverty  rates  to  growth  in  output  (and 
consumption),  which  is  then  combined  with  macroeconomic  projections  to  estimate  the  impacts  of 
future  reduced  growth  on  poverty.  Although  this  method  is  easy  to  implement  and  serves  as  a 
convenient  benchmark,  it  is  limited  in  its  predictive  capability  since  it  yields  only  aggregate  poverty 
impacts, with no account of the broader distributional effects. It may also prove deficient in predicting 
poverty  impacts  during  a  crisis  that  affects  output  growth  in  a  way  not  entirely  consistent  with  the 
recent  growth  experience  in  a  country.  This  crisis,  for  example,  is  likely  to  impact  some  sectors  more 
than others and have a disproportionate impact on inflows like external remittances that directly impact 
household income.   

PovStat  is  an  EXCELbased  World  Bank  simulation  package,  which  uses  household  survey  data  and 
macroeconomic  projections  as  inputs  and  estimates  changes  in  poverty  and  inequality  indicators. 
Although it allows for the impacts to occur through multiple channels, it offers no easy way to account 
for changes in nonlabor income (such as remittances), which has important implications in the context 
of  this  crisis  in  some  countries.  By  focusing  exclusively  on  household  heads  (and  ignoring  the 
employment status of other household members), PovStat also does not allow for a full accounting of 
labor market impacts. Perhaps the most important limitation of PovStat is that it generates estimates for 
poverty and inequality (aggregate or disaggregated by regions/groups) but not the kind of distributional 

                                                                                                                    3 
 
results that require individual or household level projections. For example, an important distributional 
question  like  how  the  impact  of  the  crisis  is  likely  to  be  distributed  across  different  groups 
(differentiated  by  income  status,  sector,  region  or  any  other  relevant  attribute)  cannot  be  answered 
with PovStat. 

More  sophisticated  simulation  approaches  than  Povstat  have  been  used  in  some  cases  by  the  World 
Bank (see Bourguignon et al, 2008).2 All of them are based on a Computable General Equilibrium (CGE) 
or General Equilibrium macroeconomic model that demand a lot of information (for constructing Social 
Accounting  Matrices  or  timeseries  of  macroeconomic  data)  in  order  to  create  the  "linkage  aggregate 
variables" (LAVS) that are fed into the microsimulation model. At the same time, these models do not 
allow for changes in some key features of the population, such as gender or age composition, and the 
economy (Ferreira et al, 2008). The main advantages of these models are related to improved accuracy 
of  the  counterfactual  and  consistency  of  the  analysis.  Notably,  the  information  demands  of  these 
models  make  them  hard  to  apply  in  most  developing  countries,  and  calls  for  an  approach  that  is 
workable with available data and macroeconomic projections.  

Because the nature of the crisis is difficult to pin down, any valid impact assessment needs to account 
for  multiple  transmission  mechanisms  and  capture  impacts  at  the  micro  level  over  the  entire  income 
distribution.  In  the  microsimulation  model  presented  here,  we  do  this  by  focusing  on  labor  market 
adjustments  in  employment  and  earnings,  nonlabor  income,  and  price  changes  (with  a  view  to  the 
variation in food/nonfood prices). Note that for the purposes of this exercise, we use the terms "labor 
income" and "earnings" interchangeably.3  

Our  proposed  methodology  can  be  seen  as  a  compromise  between  the  "simple"  methodologies  like 
PovStat and the "complex" ones using general equilibrium models ­ with the attendant advantages and 
disadvantages. Compared against the simpler approaches, the value added by our approach is twofold. 
First, the model is able to generate estimates for individuals and households across the distribution with 
and  without  the  crisis,  which  can  be  used  for  detailed  poverty  and  distributional  analyses  of  impact. 
Second, the model uses income rather than consumption; this allows us to model labor and nonlabor 
incomes  separately,  which  is  particularly  important  for  countries  in  which  remittances  and  public 
transfers form a large proportion of incomes. 

In  comparing  our  approach  with  PovStat,  the  pros  and  cons  of  either  method  should  be  kept  in 
perspective.  Our  model  in  its  most  basic  form  has  data  requirements  similar  to  that  for  PovStat,  and 
adds value along the two important dimensions stated above. At the same time, it requires considerably 
more computational resources and time than PovStat, which can be run as a simple Excelbased model 
once the household and macroeconomic data are ready to be fed into the model. Compared with the 
complex  approaches,  the  primary  advantage  of  our  approach  is  the  lower  demand  for  information, 

                                                            
2
    These  include  the  micro  accounting  approach  (Computable  General  EquilibriumRepresentative  Household 
Groups), topdown micro simulations models (CGEMicro or Macro models) and Feedback loops from bottomtop 
(Bourguignon et. al 2008). 
3
   In fact, "labor income" is defined here as the total income earned from any sort of labor activity by all members 
of a household, including wages and profits.  

                                                                                                                   4 
 
which would make it practical to apply in many cases where the more complex approaches would have 
to be ruled out. 

    3. Proposed approach 

We propose to use a microsimulation model that combines macroeconomic projections with precrisis 
micro  data  from  household  and/or  labor  force  surveys  to  predict  income  and  consumption  at  the 
individual  and  household  level  under  different  scenarios,  which  can  then  be  compared  to  measure 
poverty  and  distributional  impacts.  Comparisons  will  be  made  between  different  projected  scenarios 
(most  commonly  a  crisis  or  low  scenario  and  a  benchmark  or  base  scenario)  rather  than  a  time 
comparison  (i.e.  a  comparison  between  2005  and  2009  or  2010  in  the  case  of  Bangladesh).  Figure  1 
presents a stylized representation of the methodology. 

The  model  focuses  on  labor  markets  and               Figure 1: Microsimulation methodology 
migration  as  transmission  mechanisms  and 
allows  for  two  types  of  shocks:  shocks  to  labor 
income,  modeled  as  employment  shocks, 
earnings  shocks  or  a  combination  of  both;  and 
shocks to nonlabor income, modeled as a shock 
to  remittances.  Shocks  can  be  positive  or 
negative  depending  on  the  trends  outlined  by 
the  macroeconomic  projections.  In  most  cases 
labor  income  and  remittances  account  for  at 
least  7580%  of  household  income.  Minimum 
assumptions  are  made  about  other  sources, 
such  as  capital  and  financial  income  or  public 
transfers, as discussed below. 

The  data  requirements  can  be  summarized  as  follows.  At  the  macro  level,  information  is  needed  on 
projected  (i)  output,  employment,  remittances  and  (ideally)  labor  earnings  growth;  (ii)  population 
growth and (iii) predicted price changes. At the micro level, information is needed on (i) labor and non
labor  income  and  consumption,  and  (ii)  labor  force  status  and  basic  job  characteristics,  including 
earnings. Needless to say, the reliability and accuracy of the simulation results is a direct function of the 
quality and level of detail of the information available at the macro and micro levels. 

Finally,  the  income  and  consumption  projections  from  the  model  can  be  used  to  produce  a  variety  of 
outputs,  including  aggregate  poverty  and  inequality  comparisons  across  scenarios,  poverty  and 
vulnerability profiling of specific groups and/or areas, and various summary measures of distributional 
impacts,  such  as  growth  incidence  curves  and  state  transition  matrices.  We  comment  on  these 
extensively for the case of Bangladesh below. 




                                                                                                                 5 
 
3.1           Overview of simulation exercise 

In this Section we provide a brief overview of the mechanics underpinning the simulation exercise. The 
exercise  can  be  broken  down  into  three  distinct  steps:  calibration,  simulation  and  assessment  of 
impacts.  A  description  of  each  step  follows  and  a  schematic  of  the  complete  model  is  presented  in 
Figure 2. 

              Figure 2: Schematic of the microsimulation model




Calibration.  Calibration  is  the  process  by  which  household  and  individuallevel  information  is  used  to 
model labor market behavior and outcomes and to predict the likelihood of receiving remittances.4 This 
is  done  in  three  steps.  First,  we  model  labor  force  status  for  all  working  age  individuals  (1564)  as  a 
function  of  household  and  individual  characteristics,  where  labor  force  status  can  be  nonemployed,5 
employed  in  agriculture,  industry  or  services.  Although  ideally  we  would  like  to  work  with  a  more 
detailed menu of options (e.g. "employed in tradeables" and "employed in nontradeables" instead of 
"employed in industry"), the number of labor force states that can be considered is dictated by the level 
of disaggregation available for the macro projections. We then use a multinomial logit to estimate the 
parameters  of  the  model,  as  well  as  the  individuallevel  probability  of  remaining  in  a  particular  state 
and/or changing to a different one, as given by (1). The approach is similar to that used in Ferreira et al 
(2009).  We  estimate  the  model  separately  for  high  and  lowskill  individuals  to  allow  for  structural 
differences in the labor market behavior of the two groups.6 


                                                            
4
    We  estimate  a  reduced  form  of  the  household  incomegeneration  model  which  is  based  on  Bourguignon  and 
Ferreira (2005) and Alatas and Bourguignon (2005)  
5
   This includes "out of the labor force" and "unemployed". The decision to pool both states into a single category is 
motivated by the fact that the unemployment rate is extremely low in Bangladesh, even during crisis times. 
6
   For Bangladesh, low and highskilled refer to individuals with 0 ­ 9 and >10 years of education, respectively. 

                                                                                                                      6 
 
   ,                
I iGj  s  Ind a s  zi b s  uis  a j  zi b j  uij | j  s                                                                     (1) 

where s = Labor force status; G = labor skill level (high/low); z = gender, age, education, region, remittances, and 
land ownership. 

Second, we model labor earnings for all employed individuals ages 15 to 64 as a function of individual 
and job characteristics and use a standard Mincerian OLS regression to estimate the parameters of the 
model, as given by (2) (similar to Ferreira et al. 2008). We use a fairly broad definition of labor earnings 
for the purpose of the exercise that includes wages and salaries, but also income from selfemployment. 
This  is  particularly  important  in  the  case  of agriculture  and  for  economies  with  large  informal  sectors, 
such as Bangladesh, since wage and salaried workers constitute a limited fraction of those employed in 
these  sectors.  It  may  lead  however  to  a  loss  in  precision  and/or  predictive  power  given  that  the 
structural  relationship  between  individual  and  job  characteristics  and  earnings  could  be  different  for 
salaried and nonsalaried workers. To allow for maximum flexibility and (indirectly) account for some of 
these  differences  we  estimate  the  model  separately  for  agriculture,  industry  and  services  and  for  low 
and highskill workers.7 
log wiG   sG  xi  sG  vsG,i                                                                                                 (2) 

where  x  =gender,  age,  education,  region,  land  ownership,  and  indicators  for  export  industry,  salaried  and  public 
employment. 

The results of the estimation of equations (1) and (2) and a full description of all variables can be found 
in Annex 1. 

Finally, we model nonlabor income with a focus on international remittances and make some minimal 
assumptions  about  other  sources  of  nonlabor  income.  Ideally,  we  would  estimate  a  probit  model  to 
estimate the probability of how likely a household would be to receive international remittances, given 
its  characteristics.  However,  if  the  migrationrelated  information  in  the  survey  is  poor  or  insufficient 
and/or the predictive power of probability model is low (as is the case for Bangladesh), we are betteroff 
relying  on  a  simple  nonparametric  assignment  rule  that  is  consistent  with  the  existing  evidence  (the 
specific rule used for Bangladesh is discussed in more detail when we describe the simulation process).  

Simulation. Simulation is the process by which information on aggregate projected changes in output, 
employment and remittances is used to generate changes in labor and nonlabor income at the micro 
level using the structural models developed as part of the calibration.8 This is done in four steps.9 First, 

                                                            
7
    Notice  that,  although  we  could  estimate  separate  models  for  salaried  and  nonsalaried  workers  based  on  the 
information from the household survey, we would not be able to use these models for the purpose of predicting 
future  earnings  since  we  do  not  have  earnings  and  employment  information  disaggregated  by  salaried/non
salaried workers from macro data.  
8
   We do not assure consistency (i.e that absolute aggregate magnitudes are equal) between the data set used at 
the two modeling stages (see Bourguignon et. al 2008). Additionally, we assume equal changes at macro and micro 
levels. We cannot run a test if macro changes are similar or not to micro changes because of lack of a panel data at 
micro level (see Deaton 2001 and Bourguignon et al 2008). 
9
   This sequence for introducing changes in the model is based on Vos et al (2002) 

                                                                                                                             7 
 
we use population growth projections to adjust for demographic changes between 2005 (base year) and 
20092010. This adjustment is particularly important in the case of Bangladesh because fertility rates are 
still high, which implies that the number of labor market entrants rises faster than overall population, 
and  the  baseline  household  survey  is  relatively  old.  In  practical  terms,  taking  into  account  population 
growth allows us to explicitly take into changes in the size of the working age population, and hence to 
distinguish between employment growth driven (or rather absorbed) by demographic trends and net (or 
additional) employment growth. 

Secondly,  we  use  the  projections  from  the  labor  force  status  and  labor  earnings  models  to  replicate 
projected changes in aggregate total and sectoral output and employment. We start with employment 
and calculate the total number of individuals that need to be reassigned between employment and non
employment and across employment sectors in order to match projected aggregate changes in total and 
sectoral  employment.  We  then  use  the  estimated  probabilities  from  the  multinomial  model  to  select 
candidates  for  reassignment.10  The  direction  and  magnitude  of  flows  between  employment  and  non
employment  and  across  sectors  of  employment  is  given  by  changes  in  the  relative  shares  of  different 
status with respect to  the reference  population.  For instance, whether individuals must be reassigned 
from  nonemployment  to  employment  or  viceversa  depends  on  whether  the  employment  rate  of 
individuals ages 15 to 64 is increasing or decreasing. Similarly, workers are expelled from sectors whose 
relative share in total employment is declining and absorbed into sectors whose relative share in total 
employment is increasing will absorb workers (note that this is independent of whether employment in 
a sector is growing or contracting in absolute terms). 

The  sequence  in  which  individuals  are  reassigned  across  states  and  sectors  matters  for  the  final 
simulation results so we briefly describe here the procedure we follow: 

              Step  1    Flows  between  employment  and  nonemployment:  If  the  employment  rate  is 
              increasing,  nonemployed  individuals  with  the  lowest  predicted  probability  of  being  non
              employed will be reassigned. If the employment rate is declining, employed individuals with the 
              highest probability of being nonemployed will be reassigned. Reassignments will continue up to 
              the point where the change in the employment rate at the micro level matches the change at 
              the macro level. 
              Step  2    Flows  out  of  contracting  sectors:  For  sectors  whose  share  of  total  employment  is 
              declining,  those  individuals  with  the  lowest  predicted  probability  of  being  employed  in  the 
              sector  will  be  selected  out  and  added  to  the  pool  of  "eligible"  workers  to  be  employed  in 
              growing  sectors  (notice  this  pool  also  contains  those  who  has  been  reassigned  from  non
              employment  if  the  total  employment  rate  is  growing).  Reassignments  out  of  each  sector  will 
              continue up to the point where the change in the sector employment share at the micro level 
              matches the change at the macro level. 


                                                            
10
   We add error terms which represent the unobserved heterogeneity of agents' labor supply behavior. These lead 
to  some  disparateness  in  responses  to  a  change  in  the  labor  demand,  capturing  the  fact  that  in  the  real  world 
individuals who are identical in observables might still respond differently the same change in labor demand. 

                                                                                                                              8 
 
              Step 3 ­ Flows into growing sectors: Individuals in the pool of "eligible" workers will be assigned 
              to growing sectors on the basis of their predicted probability of being employed in each sector. 
              Assignments are made sequentially with the sector whose employment share is growing fastest 
              absorbing  workers  first  and  the  sector  whose  employment  share  is  growing  the  slowest 
              absorbing  workers  last.  Reassignments  to  each  sector  will  continue  up  to  the  point  where  the 
              change  in  the  sector  employment  share  at  the  micro  level  matches  the  change  at  the  macro 
              level. 

There  are  a  few  important  features  of  this  process  that  are  worth  mentioning.  The  reassignments 
described in steps 1 to 3 are calibrated so as to replicate net aggregate flows between employment and 
nonemployment  and  across  sectors.  In  reality,  movements  across  these  different  states  are  quite 
significant so that gross flows are usually larger than net flows. The order of proposed steps is such that 
it allows for nonemployed individuals to become employed and employed individuals to become non
employed, but also for individuals to change sectors. In doing this we try to capture the fact that highly 
"employable" individuals are more likely to remain employed in one sector or another, at times at the 
expense  of  less  "employable"  workers  (i.e.  highly  "employable"  workers  will  crowd  others  out  when 
employment opportunities become relatively more scarce). 

We next use the earnings model estimated as part of the calibration to predict earnings for two groups 
of workers: those with no previous earning history (i.e. nonemployed in 2005) and those who change 
sector of employment. Because earnings are a function of both observable and unobservable individual 
and job characteristics, we add a random element to the predicted earnings produced by the model to 
account  for  unobserved  heterogeneity.11  For  all  other  individuals,  we  use  the  earnings  information 
available in the household survey. 

Once all workers have been assigned positive labor earnings, total earnings in a sector are adjusted to 
match aggregate projected changes in output. This step relies on the fact that that projected changes in 
sectoral output can be explained by projected changes in sectoral employment and projected changes in 
sectoral earnings and profits, and assumes that earnings and profits grow at the same rate. 

The treatment of public sector workers and those with more than one job differs slightly from what we 
just  described.  Total  public  sector  employment  is  assumed  to  remain  constant  (i.e.  no  individuals  are 
assigned to or out of the public sector) and labor earnings of public sector workers are adjusted in line 
with their sectoral mapping (agriculture, industry or services). Similarly for those holding more than one 
job, we assume the sector of employment of their secondary activity remains constant while earnings 
are adjusted in line with sectoral changes. 

The third step in the simulation process pertains to changes in international remittances. As mentioned 
above, in the case of Bangladesh we simulate these changes following a very simple allocation rule. We 
calculate the total change in international remittances between 2005 and 20092010,12 using actual and 

                                                            
11
   Specifically, we draw an individual error from the error distribution generated during the estimation. 
12
   The nominal (in dollar terms) growth in external remittances is adjusted using the taka/$ exchange rates of the 
relevant years, to arrive at the value of change in remittances in constant 2005 takas.  

                                                                                                                       9 
 
projected data (this change is positive under both the crisis and the benchmark scenarios) and allocate 
the dividend as follows: (i) across divisions, remittances are allocated proportionally to the 2005 across
division  distribution;  (ii)  among  households  within  divisions:  recipient  households  are  selected  at 
random and given a remittance transfer equivalent in real terms to the median remittance transfer in 
that division in 2005, with the number of total transfers to be made within each division being equal to 
the total amount of additional remittances to be distributed divided by the 2005 median value.13 As a 
result of this process the overall distribution of remittances across divisions remains unchanged; while 
there is an increase in number of recipient households (both types of households, those who did or did 
not receive remittances in 2005, can receive new remittances in 200910). 

Finally we simulate changes in other sources of nonlabor income. For this we assume that capital and 
financial income grow at the same rate as real GDP, public transfers (primarily Social Security transfers) 
remain constant in real terms at 2005 levels, and domestic remittances change at the same rate as labor 
income.  These  assumptions  appear  to  be  reasonable  for  Bangladesh,  and  can  be  modified  for  other 
countries depending on country circumstances. 

Assessment of Impacts. Impact assessment is the process by which we use the information on individual 
employment  status  and  labor  income,  as  well  as  on  nonlabor  income  at  the  household  level,  to 
generate  income  and  consumption  distributions  and  construct  various  poverty  and  distributional 
measures  that  can  then  be  used  to  compare  the  crisis  and  the  benchmark  scenarios.  This  is  done  in 
three steps. 

First, we account for the fact that between 2005 and 20092010, food prices have increased at a higher 
rate than other prices. To ensure that the same food basket is affordable at the new prices, we adjust 
the  2005  poverty  line  to  reflect  the  increase  in  food  prices  relative  to  that  of  other  prices  (for  more 
details, see Section 4.4 below) 14.  

Second,  we  calculate  total  household  income  by  aggregating  labor  income  across  all  employed 
individuals  and  adding  nonlabor  income,  and  then  use  information  on  household  size  to  construct  a 
measure of percapita household income, as in (3).  
        km * *    NL 
         wi I i  ym * 
                                                                                                                                 (3) 
PCI m                 
    *   i 1

            km

Third, because poverty in Bangladesh is measured on the basis of percapita consumption, we need to 
map income to consumption. We do this by assuming that the householdlevel consumptiontoincome 

                                                            
13
     To  illustrate,  23%  of  additional  remittances  are  allocated  to  rural  Dhaka,  equal  to  its  share  of  remittances  in 
2005.  Every  randomly  selected  recipient  household  then  receives  5,417  taka/month,  equivalent  to  the  median 
international  remittance  transfer  in  the  division  in  2005  (<U$  90  dollars  per  month  with  an  exchange  rate  of 
tk.61/U$S) 
14
    An alternative, more sophisticated way of dealing with this problem would be to construct householdspecific 
consumption deflators that take into account differences in consumption patterns across the income/consumption 
distribution.  

                                                                                                                                  10 
 
ratio remains unchanged between 2005 and 2010. This is a strong assumption, but the best we can do 
given the available information. We may try to refine this step in future iterations of the exercise. 

                      PCExpm 2005                                                                                (4) 
    PCExpm  PCI m *
         *      *
                           2005
                       PCI m
 

Finally, we use information on household and individual income and consumption levels to evaluate the 
poverty and distributional impacts of the crisis by comparing poverty and other outcomes between the 
benchmark (without crisis) and "with crisis" scenarios. 

3.2       Limitations and caveats of the simulation exercise 

The  proposed  approach  has  some  appealing  features,  the  primary  one  being  its  capacity  to  generate 
income and consumption counterfactuals at the individual and household levels that can then be used 
to  assess  impacts  across  the  entire  distribution.  However,  it  also  has  some  important  limitations  that 
must be taken into account when interpreting the results presented below. We discuss these below. 

Firstly, the quality and accuracy of the simulation output is a function of the nature and quality of data 
underpinning the exercise. More specifically, the results would depend  not only on the micromodels, 
but also on the macro projections of the crisis and the benchmark or nocrisis scenarios. In a typical ex
ante assessment of this type, building the counterfactual to evaluate impact is especially tricky because 
the  comparison  between  the  situations  "with"  and  "without"  the  "treatment"  (the  macro  shock)  is 
purely  virtual  or  notional.  This  is  particularly  important  with  regard  to  the  output  and  employment 
projections since they are key drivers of the results in the absence of a CGE or similar model. In addition: 

          The ability to account for heterogeneity across sectors, groups, and others depends on the level 
          of disaggregation of the available macro projection. For instance, the behavior of the tradeable 
          and  nontradeable  sectors  within  industry  can  only  be  modeled  separately  if  output  growth 
          projections are available for each sector. 
          The  ability  to  accurately  predict  employment  and  earning  changes  depends  on  the  available 
          information and on the assumptions needed to correct for information gaps. For instance, in the 
          absence of projections on total and sectoral earnings growth, we need to assume that earnings 
          and profits grow at same rate within a sector. How realistic this assumption is would depend on 
          the country and sectoral context. 
          The ability to model remittances depends on the quantity and quality of the available data on 
          migrants  and  remittances,  particularly  for  countries  with  rapid  and/or  volatile  growth  of 
          remittances. 

Secondly,  the  simulations  implicitly  assume  that  the  structural  relationships  estimated  as  part  of  the 
calibration  process  on  the  baseline  data  continue  to  be  valid  in  the  future  years  for  which  the 
projections  are  made.  The  more  distant  in  the  past  the  baseline  year  is,  the  more  questionable  this 
assumption is likely to be. In the case of Bangladesh, for example, the baseline year is 2005, which is a 
full four to five years away from the prediction years (2009 and 2010). This particular caveat, however, 


                                                                                                                 11 
 
links directly back to the constraints imposed by availability of data. In most countries, household survey 
data that is available for analysis and processed to the extent necessary for the analysis is likely to pre
date the prediction years (usually 2010) by at least 3 to 4 years.  

The  third  caveat  relates  to  our  decision  to  work  with  income,  rather  than  consumption  data.  The 
advantage  of  using  income  is  that  it  allows  us  to  link  welfare  impact  on  households  directly  with 
potential  channels  of  impact,  which  are  employment,  labor  earnings  and  remittances.  There  are  two 
primary caveats to working with income data: (i) income data often tends to  be of lower quality  than 
consumption  data,  which  introduces  an  element  of  noise  into  the  analysis  due  to  the  unobserved 
presence  of  measurement  error;  (ii)  certain  assumptions,  which  can  be  challenged  on  the  grounds  of 
realism,  are  needed  to  convert  predicted  income  levels  into  consumption  and  consumptionbased 
welfare measures. It is important to note, however, that (ii) would not be necessary for countries that 
use an incomebased measure of poverty, which is the case in most Latin American countries. 

The  approach  we  have  adopted  to  convert  income  into  consumption  assumes  that  the  ratio  of 
consumption to income is unchanged for every household between the baseline and prediction years. 
The  constant  savings  rate  that  this  assumption  implies  is  probably  more  realistic  for  poor  households 
than for betteroff households. This also implies that our approach may overestimate the consumption 
impact of the crisis on betteroff households, since such households may compensate for an income loss 
by reducing savings (or dissaving), resulting in a smaller impact on consumption.  

Fourthly,  our  model  does  not  explicitly  account  for  labor  demand  at  the  sectoral  level  and  instead 
assumes  that  the  labor  market  conditions  mirror  (or  are  proxied  by)  the  macroeconomic  projections. 
The  simple  approach  we  adopt  implicitly  assumes  stable  relationships  between  output,  demand  for 
labor  and  labor  earnings,  which  may  not  hold  due  to  the  distortions  (such  as  segmentation  and 
downward stickiness of nominal wages) that typically exist in the labor market and are likely to affect 
adjustments during a crisis. 

Related to the above point, the simplifying assumptions for the labor market also do not account for the 
possibility  of  structural  shifts  in  labor  demand  due  to  the  crisis.  Sectoral  movements  of  labor  are 
modeled  as  depending  only  on  individual/household  characteristics  (through  the  multinomial  logit 
model)  and  population  growth.  This  cannot  take  into  account  the  kind  of  structural  shifts  that  have 
apparently  been  observed  in  some  countries,  such  as  a  reduction  in  the  relative  demand  for  skilled 
labor. That said, structural changes (for example, in relative demand for skills) can be incorporated into 
our model if there exists prior analytical work that provides parametric estimates of how these changes 
may have affected labor earnings. 

Fifthly, our model is limited in its ability to account for shifts in relative prices between different sectors 
of the economy as a result of the shock. The model does take into account the impact of shifts in the 
price of food relative to general inflation on poverty measures, by using the simple device of adjusting 
poverty lines that are anchored to a fixed food basket. It can be argued that changes in the relative price 
of  food  represent  the  most  significant  source  of  price  impact  on  poor  or  nearpoor  households  in  a 
country like Bangladesh, as the recent food crisis has shown. At the same time, there are other potential 


                                                                                                                12 
 
sources of price impacts ­ for example, the effect of a change in the terms of trade between agriculture 
and other sectors on real household incomes in all sectors ­ our approach does not take into account. 
Unfortunately, in the absence of a CGE model to link up to, it is nearly impossible to explicitly model for 
changes in terms of trade between sectors. 

The final limitation, related to validation of hypothesis, applies to all exante approaches including ours. 
The  only  validation  or  test  for  our  simulation  model  is  to  combine  exante  and  expost  analysis  (see 
Bourguignon  and  Ferreira,  2003).  Since  ex  post  data  will  not  be  available  for  some  time,  some 
uncertainty about the validity of the simulations generated by this exante method is bound to remain. 

       4. The Bangladesh case study 

Bangladesh is a good case study for our model for a number of reasons. Most importantly, it is a low
income, highgrowth country with an export sector that is significant but not predominant. As a result, 
its  exposure  to  the  direct  impact  of  the  crisis  is  on  the  moderate  to  low  side,  which  is  typical  of  a 
majority  of  developing  countries.  The  relatively  low  exposure  also  makes  it  possible  to  see  how  the 
model predicts the impact of relatively small shocks to the economy and how sensitive the results are to 
changes  in  macroeconomic  projections  and  assumptions.  Bangladesh  also  has  a  reasonably  balanced 
economy,  with  all  three  principal  sectors  contributing  positively  to  GDP  growth.  The  importance  of 
agriculture  has  been  declining  gradually  over  the  last  15  or  so  years  (Paci  and  Sasin  2008),  similar  to 
what is seen for most developing countries. In sum, Bangladesh appears to have many of the features 
that  are  typical  to  a  developing  country.  It  is  also  an  important  country  from  a  global  poverty 
perspective,  given  its  relatively  high  poverty  rate  and  large  population,  which  translate  to  a  large 
number of poor people.  

In terms of data availability, Bangladesh measures up quite well to most lowincome countries. A fairly 
recent household survey is available (Household Income and Expenditure Survey/HIES of 2005), which is 
the  latest  in  a  long  series of  surveys  that  have  been  used  to  track  poverty  and  related  estimates.  The 
availability of a recent Poverty Assessment report, based primarily on HIES 2000 and 2005, is an added 
plus since it provides information on poverty and labor market  that can inform the specifications of our 
empirical  models.  The  available  macroeconomic  information  is  however  limited,  with  disaggregated 
output  growth  projections  available  at  the  sectoral  level  but  not  for  key  subsectors.  Moreover,  while 
the household survey is detailed enough to apply our model, the base year (2005) is far enough from the 
target years for projections (2009 and 2010) that we cannot rule out structural changes not captured by 
our model. 

4.1           Background: poverty reduction in Bangladesh15 

Growth  and  Poverty  in  Bangladesh.  Bangladesh  has  enjoyed  a  fairly  stable  macroeconomic 
environment  with  high  growth  and  significant  poverty  reduction  over  the  past  15  years.  Poverty 
headcount  rate  declined  from  57%  in  the  early  1990s  to  49%  in  2000  and  40%  in  2005.  The  rate  of 

                                                            
15
  The discussion in this subsection is based on findings from World Bank (2007, 2008) and Paci and Sasin (2009). 
All figures cited are from World Bank (2008). 

                                                                                                                      13 
 
reduction during 20002005 was one of the highest in South Asia region for comparable periods. Poverty 
reduction was driven by robust GDP growth, which averaged above 5% annually during 200005, along 
with  stable  relative  inequality  (the  Gini  of  per  capita  consumption  remained  unchanged  at  0.31  since 
mid1990s).  While  all  sectors  ­  agriculture,  industry  and  services  ­  are  expanding,  agricultural  growth 
has been lagging behind that of other sectors, with the result that the contribution of agriculture to GDP 
fell from around 30% in the beginning of 1990s to just above 20% in 2005. 

Poverty  reduction  and  the  labor  market.  Poverty  reduction  between  2000  and  2005  was  linked  to 
significant  transformations  in  the  labor  market  and  rapid  increases  in  remittances.  Economic  growth, 
driven mainly by factor accumulation in the private sector and rapid growth of the export sector, led to 
wage growth and enhanced labor productivity. Between 2000 and 2005, the share of agriculture in total 
employment  declined  ­  agricultural  employment  grew  at  0.7  percent  per  year  compared  to  5  and  2.8 
percent for services and manufacturing respectively. There was a movement away from low productivity 
jobs  in  agriculture  to  more  productive  jobs  in  the  nonfarm  private  sector,  particularly  in  urban  areas. 
These  trends  are  consistent  with  the  shrinking  share  of  agriculture  in  GDP  while  the  share  of  services 
and industry was increasing.  

Increasing  flow  of  remittances  (20%  growth  annually  during  20002005)  was  another  key  factor 
contributing  to  poverty  reduction.16  Government  statistics  on  documented  migration  estimate  that 
more than 3.7 million Bangladeshi have emigrated during the past 30 years and about 3 million ­ or 6% 
of the incountry economically active population ­ were estimated to be living abroad in 2005. 

Notwithstanding  the  progress,  however,  Bangladesh  had  an  estimated  56  million  people  in  poverty  in 
2005  and  wide  disparities  across  income  and  occupational  groups,  gender,  and  regions.    Poverty  was 
most  prevalent  among  daily  agricultural  wage  workers  and  subsistence  farmers  while  the  betteroff 
tended to be engaged in salaried employment or nonfarm selfemployment.  The likelihood of poverty 
was also higher when a household had a larger number of dependents and low levels of education, and 
did  not  receive  remittances.  The  poverty  rate  in  2005  among  households  receiving  remittances  from 
abroad was 17% compared to 42% among the rest.    

Regional differences in poverty reduction. Underlying the national poverty story were vast differences 
between regions. Dhaka, Chittagong, and Sylhet divisions in the eastern part of the country experienced 
rapid poverty reduction between 2000 and 2005. Dhaka and Chittagong divisions, with just over half the 
country's  population  in  2000,  contributed  nearly  80%  of  the  reduction  in  national  poverty  during  this 
period.  In the West, meanwhile, gains were much smaller for Rajshahi and nonexistent for Barisal and 
Khulna. As a result, in 2005, the poverty rate was between 30% and 35% in Dhaka, Chittagong and Sylhet 
divisions, and more than 45% in Khulna, Barisal and Rajshahi (more than 50% in Barisal and Rajshahi). 



                                                            
16
   Equally important were some of the forces that have emerged from social transformations occurring over time.  
A  fall  in  the  number  of  dependents  in  a  household,  linked  to  past  reductions  in  fertility,  has  been an  important 
contributor  in  raising  per  capita  incomes,  as  has  been  increases  in  labor  force  participation  and  educational 
attainment, particularly among women. 

                                                                                                                              14 
 
Differences in wage growth, jobcreation patterns and how remittances are distributed help explain the 
widening  EastWest  gap  in  poverty.    Wages  grew  robustly  in  the  eastern  part  of  the  country  but 
stagnated in the West.  While both East and West created employment to match the rise in working age 
population,  the  East  created  many  more  jobs  that  are  better  paid  and  in  a  robustly  growing  non
agricultural sector.  In contrast, a large proportion of jobs created in the West consisted of daily wage 
work  or  agricultural  selfemployment.  The  eastern  region  has  increasingly  benefited  from  integration 
with growth poles, namely Dhaka and Chittagong urban areas, in contrast to the more isolated West and 
Southwest.   These two  cities have emerged as the  main centers of economic activity of  the country ­ 
Dhaka  alone  accounts  for  80%  of  the  country's  Ready  Made  Garments  (RMG)  output  and  half  of 
manufacturing sector employment. There is wide EastWest disparity in the distribution of remittances 
as  well:  in  2005,  24%  of  households  in  Chittagong  division  and  16%  in  Sylhet  received  remittances, 
compared to less than 5% in Rajshahi, Khulna, and Barisal. 

The recent shocks. The rapid pace of poverty reduction is likely to have slowed during 200708, due to 
the cumulative impact of natural disasters in 2007 (a severe flood and a cyclone) and more significantly 
for  the  population  as  a  whole,  the  foodprice  shock  during  200708.  The  food  (mainly  rice  for 
Bangladesh) price shock is estimated to have increased the poverty headcount rate in 2008 by around 3 
percentage points compared to what it would have been in the absence of the shock (World Bank 2008). 
While the effect of the rice price shock is likely to be temporary, with prices reportedly coming back to 
their  trend  levels  in  2009,  the  financial  crisis  is  expected  to  have  some  adverse  impact  on  poverty 
reduction in 2009 and 2010. The cumulative impact of successive shocks on the poor is also a cause for 
concern.  The  poor  may  have  exhausted  their  options  in  terms  of  increasing  their  labor  supply  to  deal 
with rising food prices, which could worsen the impacts of the current crisis.17 

The  story  of  poverty  reduction  outlined  above  also  suggests  the  likely  channels  through  which  the 
financial  crisis  can  slow  the  pace  of  poverty  reduction.  These  are  likely  to  be:  (i)  labor  market 
adjustments due to the impact of the crisis in the manufacturing sector, which includes RMG exports; 
and  (ii)  adverse  impact  on  remittances.  Some  impact  can  be  expected  in  the  service  sector  as  well, 
primarily due to the forward linkages between this sector and manufacturing.  

Given  that  the  poor  were  less  likely  to  be  employed  in  the  nonfarm  sector  and  receiving  external 
remittances in the first place, we would also expect that the impact of the crisis will be skewed towards 
those  who  are  relatively  betteroff.  Moreover,  given  that  the  industry  sector  and  remittances  are 
concentrated  more  in  the  eastern  part  of  the  country  than  the  west,  the  impacts  are  also  likely  to  be 
higher  in  the  east.  Our  simulation  results  that  follow  confirm,  to  a  large  extent,  these  intuitions  and 
provide  an  outlook  of  how  the  impacts  are  likely  to  be  distributed,  across  different  income  groups, 
regions and urban/rural areas.  



                                                            
17
   A World Bank Rapid Survey of Food Price Impact found that 95% of households in the sample were adversely 
affected.  Most  had  to  adopt some  coping  strategy,  such as  reducing  their  food  intake  (76%),  switching  to  lower 
quality food (88%), cutting education expenses (39%), taking out loans (55%) and drawing down savings (45%). A 
third of all households attempted to cope with higher prices by increasing their work effort (World Bank 2009). 

                                                                                                                          15 
 
4.2           Macroeconomic projections of crisis impact           

Projections of aggregate and sectorspecific growth rates for 2009 and 2010, obtained from the World 
Bank Bangladesh country office, are used as inputs into our microsimulation model (see Annex 2, Table 
A.2 for more detailed data). In conducting simulations, two scenarios are considered for each year: (i) 
the  scenario  with  crisis,  and  (ii)  the  benchmark  or  "nocrisis"  scenario  (namely  projections  that  were 
made before the crisis hit).  

Aggregate GDP growth is expected to  be 0.8 and 1.5 percentage points lower with crisis  compared to 
the  benchmark  or  nocrisis 
                                         Table 1: Macroeconomic projections 
scenario,  in  2009  and  2010 
                                                                                                         Impact of Crisis
respectively  (Table  1).  Growth  of                        Baseline           With Crisis          (Crisis ­ Benchmark) 
output  in  industry  and  services  is                         2005         2009        2010           2009        2010
projected to slow down by 1.1 and        Output growth (%)
1.2  percentage  points  respectively  Total GDP                 5.9          5.8         5.6            0.8        1.5 
in  2009  and  by  3.3  and  1               Agriculture         2.2          4.6         3.6            0.6        0.3 
percentage  points  respectively  in 
                                             Industry            8.3          5.9         6.0            1.1        3.3 
2010.  The  impact  is  most 
significant  in  the  industry  sector       Services            6.4          6.3         6.1            1.2        1.0 
primarily  due  to  a  fall  in  demand  Remittances
for manufacturing exports and the  (US$ millions)18 3,848                   9,689       10,872          311        1,228 
spillover  effect  this  may  have  on  Source: World Bank Bangladesh country office (as of September 2009) 
ancillary  industries.  The  impact  is  Note: 1) Benchmark projections are the projections made for the relevant years 
much smaller for the service sector  at a time before the financial crisis occurred. 
                                         2) Crisis projections take into account the impact of the crisis. 
since  services  in  Bangladesh  are 
generally not traded and therefore less susceptible to global trends.  

Importantly, remittances are expected to be lower by about 10% due to the crisis in 2010, compared to 
the  nocrisis  scenario.  In  2009  however,  the  available  figures  suggest  that  the  actual  growth  in 
remittances  in  2009  was  about  the  same  as  it  was  predicted  before  the  crisis.  Even  with  the  crisis, 
remittances  are  expected  to  grow  by  12%  between  2009  and  2010  in  dollar  terms,  which  is  however 
quite a bit lower than the 21% growth that was expected had there been no crisis.  

In sum, the macroeconomic impacts of the crisis, according to the latest projections, are estimated to be 
relatively  small  in  Bangladesh  and  occur  mainly  in  the  industry  sector  and  remittances.  This  suggests 
that the impact on labor earnings of household is likely to be small in both 2009 and 2010 and skewed 
towards  households  earning  income  from  industry  sector.  The  adverse  impact  on  remittance  growth 
would lead to some losses in household nonlabor income for year 2010, but not for year 2009.  


                                                            
18
   For  the  purposes  of  the  simulation,  the  remittance  projections  are  converted  into  Takas  using  the  taka:US$ 
exchange rate for each year, and then adjusted to 2005 takas using each year's CPI (adjusted to base year 2005). In 
other words, the following formula is applied to each year's remittance projection: RR= RN. /CPI, where RR is the 
amount in real takas (at 2005 prices), RN is the amount in nominal US dollars, and  is that year's exchange rate. 

                                                                                                                         16 
 
4.3           Labor market projections and impact of crisis  

Labor market status in 2005. In 2005 (according to HIES  Figure 3:  Population by age cohorts (2005)
data),  about  48%  of  the  workingage  (1564  years  of         95-99
                                                                  90-94                     2005
age)  population  was  employed.19  Among  those                  85-89
                                                                  80-84
employed,  44%  was  employed  in  agriculture,  24%  in          75-79
                                                                  70-74
industry,  and  32%  in  services.  55%  of  those  employed      65-69
                                                                  60-64
                                                                  55-59
were  salaried  workers,  and  75%  of  the  labor  force had     50-54                           Women
                                                                  45-49
less  than  10  years  of  education  (defined  as  "low          40-44                           Men
                                                                  35-39
skilled").  The  employed  population  includes  those            30-34
                                                                  25-29
working  for  wages/salaries,  the  selfemployed  and             20-24
                                                                  15-19
                                                                  10-14
those  contributing  labor  to  householdbased                      5-9
                                                                    0-4
enterprises  or  activities  that  produce  income  for  the 
                                                                  milion-10 -8 -6 -4 -2 0 2 4 6 8 10
household.  On  average,  68%  of  household  income 
comes  from  labor  earnings  and  9%  comes  from 
                                                               Source: HIES 2005
remittances (from abroad and domestic), which in turn 
constitutes about 39% of nonlabor income (see Annex 2, Table B for detailed figures). 

                                                                     Projections  for  2009  and  2010.  The 
    Table 2: Employment projections (% of people age 1564 yrs)
                                                                     population  growth  rate  between  2005 
                                                   Impact of Crisis  and 2010 is projected to be 7.4%, with a 
                  Baseline      With Crisis  (Crisis ­ Benchmark)
                                                                     higher  growth rate  (11.9%) projected  for 
                    2005       2009  2010          2009        2010
                                                                     the  working  age  population.  This  is  a 
Employed (%)        48.4       49.4     51.0        0.3        0.6 
                                                                     result  of  the  age  distribution  of  the 
Sectoral shares (% of employed in each sector)                       population  in  2005.  The  "bulge"  among 
  Agriculture       44.3       40.4     39.5        0.3         0.5  the  514  years  age  group  in  the  2005 
  Industry          23.8       23.8     22.5         0          1.4 
                                                                     population  (Figure  3)  indicates  that  the 
  Services          32.0       35.8     38.0        0.3         0.8 
                                                                     working  age  population  will  expand 
Source: Own calculations using GDP data and HIES (2000, 2005) 
                                                                     rapidly  over  the  subsequent  510  years, 
imposing a burden on the labor market to absorb the new entrants (see Annex 2, Table A.1). 

Employment  growth  projections  for  2009  and  2010  are  computed  using  past  (20002005)  trends  in 
output and employment.20 Given the relatively small impact of the crisis on projected GDP growth rates 
(particularly  for  2009),  the  impact  on  overall  employment  is  expected  to  be  very  small  ­  employment 
rate  would  be  lower  by  just  0.3  and  0.6  percentage  points  compared  to  the  benchmark  (nocrisis) 
scenario in 2009 and 2010 respectively (Table 2).  



                                                            
19
    The remaining 52% of the working age population comprises of those who are "inactive" or unemployed. We 
consider an individual to be inactive if s/he has not worked for income (wages, or contribution to family enterprise 
that produced income) during the last year and is not looking for work. Out of the working age population, 2% are 
unemployed and 50% are inactive. 
20
    See Annex 2, Tables A.2 and A.3 for detailed projections. 

                                                                                                                 17 
 
The  impact  on  sectoral  shares  in  employment  is  more  significant,  since  the  crisis  does  not  affect  all 
sectors equally. In 2010, the share of industry in total employment is projected to be lower in the crisis 
scenario by 1.4 percentage points, while those of services and agriculture are expected to be 0.8 and 0.5 
percentage points higher, respectively (Table 2). 

4.4      Impact on household income, poverty and distribution: results from microsimulation 

It  is  important  to  note  that  the  results  of  the  simulation  exercise  are  entirely  dependent  on  the 
macroeconomic  projections  (in  Section  4.2  above),  which  can  change  as  more  information  becomes 
available. If the actual impact of the crisis on remittance growth in 2010, for example, were to be less 
than the projections shown in Table 1, the impact on poverty and household incomes will also be lower 
than what is estimated here. The primary purpose of this exercise should therefore be seen not as an 
attempt to definitively predict future poverty in the highly uncertain environment of a global economic 
crisis, but rather as illustrating the kinds of results and insights this simulation model can yield, given the 
most recently available projections of macroeconomic impact of the crisis. 

 Average household income is expected to be 0.7% lower in 2009 and 2.4% lower in 2010 as a result of 
the crisis (compared to the benchmark case), applying the microsimulation method described in Section 
3.  The  drop  in  2009  income  is  attributed  to  a  0.5%  drop  in  labor  income  and  1.5%  drop  in  nonlabor 
income, while in 2010 labor and nonlabor income is projected to drop by 1.5% and 5% respectively. The 
significantly higher loss in nonlabor income in 2010 is due to a 9% loss in remittances in 2010 compared 
to 3% in 2009 (see Annex 2, Table B for the detailed estimates on household income). 

 Impact on poverty and inequality. As expected, and consistent with the impacts on household income, 
the impact of the crisis on poverty measures is low in 2009 and more significant in 2010. Table 3 shows 
the  projections  for  poverty  and  inequality  measures  for  2009  and  2010  with  crisis  and  the  projected 
impact of the crisis for each year  Table 3: Impact of financial crisis on poverty and inequality indices
(i.e. the difference between crisis                                                              Impact of Crisis
and benchmark scenarios).                                       Baseline    With Crisis       (Crisis ­ Benchmark)
                                                                   2005     2009      2010        2009       2010
The  poverty  headcount  rate  Moderate poverty 
(based on the upper poverty line)    Headcount rate (%)          40.0     28.6       25.8         0.4          1.2 
and  extreme  poverty  rate  (based    Poverty gap                9.0      6.0        5.3         0.1          0.3 
on  the  lower  poverty  line)  are     Squared poverty gap 2.9            1.9        1.6         0.1          0.1 
expected  to  be  0.4  and  0.2  Extreme poverty 
percentage points higher in 2009,    Headcount rate (%)          25.1     16.7       14.8         0.2          0.9 
respectively,  as  a  result  of  the    Poverty gap              4.7      3.0        2.5         0.1          0.1 
crisis.  In  2010,  the  poverty     Squared poverty gap 1.3               0.8        0.7           0          0.1 
headcount rate is expected to be  Inequality percap. Exp
1.2 percentage points higher and    Gini                         0.31     0.34       0.32         0.02          0 
the  extreme  poverty  rate  0.9    Theil                        0.19     0.22       0.19         0.02        0.01 
percentage  points  higher  due  to  Source: Microsimulations using macro projections (see Table 1) and HIES (2005) 
the crisis. The impacts on poverty gap (depth of poverty) and squared poverty gap (severity) are small 


                                                                                                                  18 
 
for both years, but slightly higher for 2010 than 2009. The impact of the crisis on aggregate measures of 
inequality is negligible (see Annex 2, Table D.1 for detailed results). 

The poverty  rate in  the absence of the crisis would have declined by 15.4 percentage  points between 
2005 and 2010. With the crisis affecting GDP and remittance growth in 2009 and 2010, the poverty rate 
is  projected  to  decline  by  14.2  percentage  points  between  2005  and  2010.21  The  difference  of  1.2 
percentage points between the two scenarios translate to around 2 million additional poor individuals in 
2010 due to the impact of the crisis. 

Comparison  with  other  simulation  methods.  Our  results  on  the  poverty  impact  of  the  crisis  are 
comparable  to  results  obtained  using  other  commonly  used  methods,  for  the  same  macroeconomic 
projections  (Table  4).  Using  Povstat  and  the  aggregate  elasticity  of  poverty  to  growth,  the  poverty 
impacts are 0.2 and 0.6 percentage points for 2009, respectively, compared to a 0.4 point impact using 
our micromacro simulation method. For 2010, both the alternate approaches yield a poverty impact of 
1.2 percentage points, identical to the impact estimated by our approach. 

For  every  year  and  scenario,  the  poverty  rates  projected  by  the  three  approaches  are  within  2 
percentage  points  of  each  other  (for  Table 4: Comparing poverty headcount rate (%) using different 
more  detailed  results,  see  Annex  2,  projection methods
Table  D2).  Our  method  consistently                                                              Impact of Crisis
                                                                  Baseline      With Crisis     (Crisis ­ Benchmark)
yields  poverty  rates  that  are 
                                                                    2005     2009      2010        2009       2010 
somewhere          inbetween         those 
                                            Povstat                 40.0      28.9     26.3         0.2         1.2 
generated  by  Povstat  and  the 
                                            Elasticity of 
elasticity  method  (Table  4).  The  poverty to growth  40.0                 28.0     25.6         0.6         1.2 
reasons why these small differences in 
                                            Our simulation          40.0      28.6     25.8         0.4         1.2 
predictions occur are hard to pinpoint 
                                            Source:  Simulations/computations using macro projections and HIES (2005)
since  the  other  two  approaches  are 
substantively different from ours, as well as from each other. A quick comparison between Povstat and 
our  approach,  however,  is  useful  since  both  approaches  involve  simulations  linking  micro  data  with 
macro changes. One key difference between Povstat and our approach is the treatment of remittances. 
Povstat  implicitly  assumes  that  remittances  grow  at  the  same  rate  as  other  incomes,  whereas  (actual 
and  projected)  growth  in  remittances  has  been  much  higher  than  that  of  GDP  after  2005.  Our  model 
explicitly  takes  into  account  actual  and  projected  growth  in  remittances  since  2005  and  as  a 
consequence, generates larger improvements in household income and lower poverty for all scenarios 
compared to Povstat.    

Price  adjustments  underlying  the  poverty  projections.  All  projected  expenditures  are  at  2005  prices, 
adjusted for spatial differences.22 This would normally imply that no adjustments are necessary for the 

                                                            
21
     The  decline  in  poverty  rate  for  both  benchmark  and  crisis  scenarios  would  be  an  improvement  upon  past 
performance ­ poverty fell by 9 percentage points between 2000 and 2005, a period when annual average growth 
in GDP and remittances was lower than the 20052010 annual average growth for both scenarios.   
22
     All  expenditures  and  the  poverty  line  for  the  baseline  year  (2005)  are  adjusted  to  2005  rural  Dhaka  division 
prices, which is one of the strata in the sample of HIES 2005. 

                                                                                                                               19 
 
poverty line (which is defined at 2005 prices) as well, had it not been for the fact that food prices have 
risen at a faster rate than the general  consumer price index (CPI) since 2005  (by 52% as compared  to 
45%). Given that the poverty line is anchored to a fixed basket of food items necessary to meet the basic 
minimum  calorie  needs  of  an  individual  in  2005,  a  rise  in  the  relative  price  of  food  would  necessarily 
imply  that  the  same  poverty  line  will  no  longer  be  sufficient  to  purchase  the  basic  food  basket.23 
Therefore,  the  poverty  line  for  2009  and  2010  must  be  adjusted  upwards  to  reflect  the  same  level  of 
welfare as in 2005.  

We adjust the 2005 poverty line upward using the food and nonfood CPI for 2009 (or 2010), weighted 
by the food and nonfood shares in the poverty line. This yields a 2009 (or 2010) poverty line at current 
prices, which is then deflated to 2005 prices using the appropriate general CPIs (see Annex 2, Table C). 
The poverty line that results ­ effectively the poverty line for 2009 (or 2010) at 2005 prices ­ is slightly 
higher than the original 2005 poverty line since the weight of food in the poverty line is higher than that 
in  the  CPI.24  The  adjusted  poverty  line  for  each  year  reflects  the  number  of  takas  required  by  an 
individual to be able to afford the same food basket despite the disproportionate rise in food prices. 

Food prices ­ most significantly rice prices ­ in Bangladesh rose sharply during 2008 in response to the 
worldwide food price shock, which likely led to a temporary spike in poverty rate (as mentioned earlier).  
Since rice prices have moderated since late 2008, the food CPI projections for 2009 and 2010 are likely 
to  reflect  the  expected  trend  of  prices  rather  than  the  temporary  spike  in  2008.  One  question  that 
arises,  however,  is  how  the  poverty  projections  will  change  if  the  financial  crisis  were  to  have  a 
significant dampening effect on domestic food prices. No alternate projections for food and general CPI, 
which take into account the dampening effect of the financial crisis on domestic prices in Bangladesh, is 
available at this stage. Our only recourse appears to be the world food price projections by FAO, which 
show  a  sharp  downward  movement  of  rice  prices  in  2010  (see  Box  1  for  a  description  of  alternate 
poverty projections based on FAO world price projections). 

Box 1: Alternate poverty projections for 2010 using FAO data.  
We  compute  the  alternate  poverty  estimates  for  2010  under  the  assumption  that  rice  prices  in  Bangladesh  will 
follow the trend in world prices projected by FAO.  Under this assumption, the food price inflation during 2005
2010 will be 5% lower than general inflation, which would require a small downward adjustment of the poverty 
line  to  ensure  that  it  represents  the  same  purchasing  power  with  regard  to  the  food  basket.25  With  this 
adjustment, the poverty rate projected for 2010 under the crisis scenario is slightly lower ­ 25.4%, compared to 
the  earlier  projection  of  25.8%  (see  Annex  2,  Table  D.1,  last  column).  This  translates  to  a  0.8  percentage  point 
increase in poverty rate as a result of the financial crisis, instead of the 1.2 percentage point impact seen earlier.  
                                                            
23
    The implicit assumption here is that all nominal incomes (that translate to expenditures) would rise over time at 
the rate of inflation given by the general CPI. The higher rate of inflation in food prices, however, would mean that 
the nominal income of an individual who was consuming at the poverty line level in 2005 would lag behind the rate 
at which the food component of the poverty line will inflate. Thus the poverty line would have to be shifted up to 
reflect the fact that such an individual will no longer be able to afford the basic minimum food needs.   
24
    Note that this adjustment would not be necessary if the share/weight of food in the poverty line were identical 
to that in the CPI. 
25
    Note that a decline in food prices will also imply a decline in general CPI, albeit at a lower rate. Our adjustment 
to the poverty line follows a method identical to the one described earlier. 

                                                                                                                               20 
 
These  projections  are  just  for  illustrative  purpose  and  should  not  be  considered  as  robust  for  two  key  reasons. 
Firstly,  there  is  little  information  available  on  the  underlying  assumptions  behind  the  FAO  projections,  for  us  to 
ascertain how robust these projections are. Secondly, the assumption that domestic rice prices in Bangladesh will 
move in sync with world prices is not realistic for a number of reasons. Rice is a thinly traded commodity in the 
world market, with the result that there is often a high "wedge" between domestic and world prices. In the case of 
Bangladesh,  domestic  demand  and  production  trends  ­  as  well  as  distortions  created  by  export  restrictions  in 
countries that Bangladesh typically import from ­ are likely to play a key role in shortterm price movements. In 
sum,  the  assumption of a steep decline in domestic rice prices in 2010 based on FAO projections appears to be 
highly speculative. 
In spite of these caveats, our alternate projections under a scenario of rice price decline serve a useful purpose ­ of 
illustrating the impact that rice prices would have on poverty in Bangladesh, at least in the shortterm. The lesson 
to take away from the alternate poverty projections is that the ultimate impact of the financial crisis on poverty 
will depend to some extent on how food prices move in the near future.     

4.5                   The impact of the crisis across regions and urban/rural areas 

The EastWest gap in poverty reduction between 2000 and 2005, as described in Section 4.1 earlier, is a 
result of significant differences in how industrialization and rapid growth in remittances have evolved in 
different parts of the country. Given this background, it is important to look at how the impact of the 
crisis is likely to be distributed between the two regions. We would expect the impact to be higher in the 
East  than  the  West  because  of  two  primary  reasons.  First,  the  East  has  a  greater  concentration  of 
industry, which is the worstaffected sector; and second, the eastern region ­ and Sylhet and Chittagong 
divisions in particular ­ has much higher incidence of remittances. 

    Figure 4: Impact of crisis on income and poverty ­ by region (2010)
            (a) % change in average household income              (b) % change in poverty rate (upper PL) between 
                     between benchmark and crisis                              benchmark and crisis 
                0.0                                                                                            15


                1.0
                                                                                                               10
     % change




                                                                                                    % change




                2.0

                                                                                                                5
                3.0


                4.0                                                                                             0
                                                                                                                                                         Sylhet
                                                                                                                                    Chittagong




                                                                                                                                                                            Khulna
                                                                                                                                                 Dhaka




                                                                                                                                                                                     Rajshahi
                                                                                                                            Urban
                                                             Sylhet




                                                                                Khulna
                                        Chittagong

                                                     Dhaka
                                Urban




                                                                                         Rajshahi




                                                                                                                    Rural




                                                                                                                                                                  Barisal
                        Rural




                                                                      Barisal




    Source: Own simulations based on HIES (2005)
    Note: Vertical axis measures the difference between crisis and benchmark poverty rates (as % of benchmark rate)
EastWest differences in income losses and poverty impact. Our microsimulation results confirm these 
expectations for 2010 (see Annex 2, Table G for all results). Sylhet, Chittagong and Dhaka are the worst
affected  divisions  in  terms  of  loss  of  average  income.  The  average  household  income  in  Sylhet, 
Chittagong and Dhaka are projected to be lower by 4%, 3% and 2.5% respectively in the crisis scenario 

                                                                                                                                                                                            21 
 
compared to the benchmark in 2010, while the losses are less than 2% for Khulna, Rajshahi and Barisal 
(Figure 4a). This is because the East is projected to suffer larger losses in remittances and labor incomes 
compared to the West (see Annex 2, Table G). The fact that remittances are more common in the East 
also implies that loss in remittances affects a larger share of the population than in the West. 

Consistent  with  the  projected  income  losses,  the  impact  on  poverty  is  higher  in  the  East  as  well 
compared  to  the  West.  The  percentage  increase  in  poverty  rate  due  to  the  crisis  (compared  to  the 
benchmark) is projected to be the highest in Chittagong followed by Sylhet and Dhaka, and smaller for 
the other divisions (Figure 4b). The percentage increase in extreme poverty rate (not shown here) is high 
for Chittagong and Sylhet divisions, and much lower for the other divisions including Dhaka. Chittagong 
and  Sylhet  are  the  highest  recipients  of  international  remittances  by  a  large  margin.26  As  a  result  the 
projected income loss in 2009 and 2010 due to the crisis (relative to the benchmark) is the highest in 
Chittagong and Sylhet, which in turn leads to high poverty impact in these divisions. 

Even after taking into account the significantly higher impact of the crisis in the East, the EastWest gap 
in poverty and income levels would continue to remain large in 2010. This is because the gaps in 2005 ­ 
the baseline year for all the simulations ­ were large to start with (Figure 5). Khulna, Rajshahi and Barisal 
are projected to remain as the poorer divisions of the country due to a slow rate of poverty reduction 
during 20052010. The crisis is however expected to moderate the rapid poverty reduction that would 
have  otherwise  occurred  in  Sylhet,  Chittagong  and  Dhaka  due  to  rapid  industrial  growth  and  rise  in 
remittances. 

    Figure 5: Poverty headcount rates for divisions (%)                                      Figure 6: Poverty headcount rates for urban/rural
    2005 estimates & 2010 projections                                                        2005 estimates & 2010 projections 
                         60                                                                                     50 
                         50                                                                                     40 
       Poverty HCR (%)




                                                                                              Poverty HCR (%)




                         40 
                                                                                                                30 
                         30 
                                                                                                                20 
                         20 
                                                                                                                10 
                         10 
                                                                                                                 
                                                                                                                            Rural                    Urban
                                                               Barisal
                                                    Sylhet
                                            Dhaka
                               Chittagong




                                                                                  Rajshahi
                                                                         Khulna




                                                                                                                 2005   2010 benchmark (no crisis)    2010crisis

    Source: Estimates and simulations based on HIES (2005)

Ruralurban  differences  in  impact.    Our  model  does  not  project  stark  differences  in  the  income  and 
poverty impact of the crisis between rural and urban areas. The loss in average household income due 
to  the  crisis,  as  a  percentage  of  benchmark  income,  is  slightly  higher  in  rural  areas  (2.5%)  than  urban 
areas  (2.3%).  The  main  source  of  income  loss  is  the  drop  in  remittances,  which  is  slightly  higher  (as  a 
percentage  of  benchmark  remittances)  in  rural  areas  than  in  urban  areas.  The  projected  loss  in  labor 

                                                            
26
   For  example  in  2005,  24%  and  16%  of  households  in  Chittagong  received  remittances,  compared  to  9%  of 
households nationally (World Bank 2008). 

                                                                                                                                                                   22 
 
income, which accounts for a smaller share of income losses than remittances, is around 1.9% in urban 
areas and 1.4% in rural areas (Annex 2, Table E). 

The poverty headcount rate is expected to increase by 1.3 and 1.1 percentage points in rural and urban 
areas, respectively, due to the crisis (Figure 6; full results in Annex 2, Table E). The extreme poverty rate 
will  be  raised  by  1.0  and  0.7  percentage  points,  respectively.  However,  as  a  percentage  of  the 
benchmark poverty rates, the rate of increase in poverty and extreme poverty is higher in urban areas 
than rural areas (Figure 4b). This happens because even though the amounts of change are similar for 
urban and rural areas, the benchmark poverty rates are significantly lower in urban areas. 

4.6           Distributional impact of the crisis ­ going beyond poverty and inequality indices 

By  generating  predicted  levels  of  income  and  consumption  for  all  households  in  benchmark  and  crisis 
scenarios,  our  simulation  model  allows  us  to  examine  the  type  of  households  that  are  likely  to  be 
affected by the crisis, the primary channels of impact and their relative importance, and the distribution 
of  the  impact  across  different  income/consumption  groups.  Here  we  present  the  results  from  three 
types of analysis that have been selected primarily for illustrative purposes, taking into account some of 
the  key  issues  relevant  to  Bangladesh.  While  the  same  analysis  can  be  done  for  both  years,  for 
illustrative purposes we choose to present the results for 2010 only, when the relatively larger impact of 
the crisis is expected.   

First,  we  examine  the  characteristics  of  the  group  we  will  call  "crisisvulnerable",  which  refers  to 
households that would not have been poor in 2010 had there been no crisis. Second, we use the well
known analytical device of growth incidence curves to see how change in consumption, as a result of the 
crisis,  is  distributed  across  the  distribution  and  between  urban/rural  areas  and  regions.  Third,  we 
construct  a  few  transition  matrices  to  look  at  upward  and  downward  movement  of  households  as  a 
result of the crisis, compared to the benchmark. These are but examples of what is possible in the way 
of distributional analysis with the results of our model; the choice of what type of analytics needs to be 
done for a certain country would depend on the specific country context and policy concerns. 

A profile of the "crisisvulnerable". Households that are expected to be in poverty in the crisis scenario 
but  not  in  the  benchmark  scenario  in  2010  constitute  around  1.2%  of  the  population  (1.8  million 
individuals).27  These households, who are the "crisisvulnerable", are projected to suffer large income 
losses  due  to  the  crisis,  relative  to  the  benchmark  scenario  ­  with  a  23%  drop  in  average  household 
income that translates to a 39% fall in average per capita consumption. The income loss is mainly due to 
the loss of external remittances. While labor income loss amounts to around 2% of benchmark income, 
the  loss  in  nonlabor  income  is  around  67%,  which  is  in  turn  due  to  a  combination  of  declines  in  the 
share  of  households  receiving  remittances  and  the  average  amount  of  remittances  conditional  on 


                                                            
27
    Our  simulation  method,  because  of  the  way  remittances  are  assigned  with  and  without  crisis,  ensures  that 
almost  every  household  who  is  poor  in  the  benchmark  scenario  will  be  poor  in  the  crisis  scenario  as  well,  and 
conversely,  almost  every  household  that  is  nonpoor  in  the  crisis  scenario  will  be  nonpoor  in  the  benchmark 
scenario as well. For the sake of simplicity and realism alike, this appears to be a reasonable result. 

                                                                                                                             23 
 
receiving. The key characteristics of crisisvulnerable households, relative to the rest of the population, 
are as below (see Annex 2, Table F for all results).  

Firstly,  the  crisisvulnerable  households  have  much  higher  incidence  of  external  remittances  than  the 
general  population­  46%  of  the  crisisvulnerable  households  receive  remittances  from  abroad  in  the 
benchmark  scenario  compared  to  10%  of  the  general  population.  Secondly,  the  crisisvulnerable  have 
larger household size and dependency than the rest of the population. Thirdly, they have lower skills on 
the average ­ 89% of household heads have education of 09 years compared to the national average of 
79%. Fourthly, a much  higher share of adults in  crisisvulnerable households are employed in  industry 
and  services  (73%  of  total  adults  employed)  compared  to  the  general  population  (61%  of  total  adults 
employed).  Finally,  they  look  fairly  similar  to  the  general  population  in  terms  of  their  ruralurban 
composition, with 80% from rural areas compared to 75% for the general population. 

How do the crisisvulnerable households compare with the permanently or structurally poor (defined as 
the  households  who  are  poor  in  both  benchmark  and  crisis  scenarios  in  2010)?  Crisisvulnerable 
households  have,  on  the  average,  lower  dependency  and  higher  skill  level  than  the  structurally  poor. 
Their  employment  rate  in  industry  and  services  is  also  much  higher  ­  73%  of  all  employed  adults  in 
crisisvulnerable households are employed in these two sectors compared with 54% in the case of the 
structurally  poor.  The  crisisvulnerable  households  are  also significantly  less  rural  than  the  structurally 
poor households (see Annex 2, Table F).  

In sum, the loss of (or reduction in) remittances from abroad is the most significant channel of impact 
for households who are predicted to become poor as a direct result of the crisis. In fact, nearly 60% of 
the  aggregate  poverty  impact  of  the  crisis  in  2010  is  due  to  the  loss  in  remittances.  It  is  important  to 
note  that  "loss  of  remittances"  actually  refers  to  the  reduction  in  potential  remittances  in  2010  ­ 
remittances, while still growing even during the crisis, will do so at a lower rate than what was expected 
without the crisis (resulting in 10% lower remittances). 

The  profile  of  crisisvulnerable  households  is  consistent  with  our  a  priori  intuition.  Compared  to  the 
general  population,  households  vulnerable  to  the  crisis  are  disproportionately  employed  in  the  non
agricultural sectors because output and income losses are expected to occur in these sectors. They also 
have  "worse"  attributes  (larger  households,  higher  dependency  and  lower  skills)  compared  to  the 
general population ­ attributes that would make them vulnerable to being poor as a result of the crisis. 
The  crisisvulnerable  or  the  "newly  poor"  households  as  a  result  of  the  crisis  are  also  different  in  a 
number of key aspects from the more permanently poor. Compared to the structurally poor, they have 
"better" characteristics (lower dependency, higher skills, receiving remittances, living in urban areas and 
employed outside agriculture) ­ characteristics that our simulations indicate would have taken them out 
of poverty had there been no crisis. 

Growth incidence between benchmark and crisis scenarios. We use growth incidence curves (GICs) to 
examine  how  the  reductions  in  per  capita  consumption  and  income  between  benchmark  and  crisis 
scenarios  in  2010  are  distributed.  GICs  are  commonly  used  to  look  at  the  distribution  of  growth  over 
time; however the same concept is useful to see changes between two "states" of the world as well, in 


                                                                                                                        24 
 
this case the benchmark (nocrisis) and crisis states for the same point in time. A GIC in this case will plot 
the  percentage  change  in  per  capita  consumption  between  2010  benchmark  and  crisis  scenarios  for 
every centile in the benchmark distribution of per capita consumption/income.28 

Taking  the  country  as  a  whole,  the  reduction  in  per  capita  expenditures  due  to  the  crisis  (as  a 
percentage  of  benchmark  expenditures)  rises  as  one  moves  up  the  distribution,  with  the  largest 
difference (a drop of 45% in per capita consumption) occurring for households above the 70th percentile 
of  the  benchmark  distribution  (Figure  7a).  The  drop  in  per  capita  consumption  is  less  than  1%  for 
households in the bottom 10% of the distribution, but 34% for households in the 50th to 70th percentile. 

There  are  interesting  differences  between  how  the  impact  of  the  crisis  would  be  distributed  within 
urban and rural areas (Figure 7b). Three key differences stand out: (i) in rural areas, the largest impact is 
seen among the richest (above the 80th percentile), while in urban areas those between the 50th and 80th 
percentiles suffer the largest loss; (ii) the rural rich suffer a much steeper drop in consumption than the 
urban rich; (iii) the moderately poor and nearpoor in urban areas (10th to 40th percentiles) are affected 
more than the corresponding group in the rural areas. These differences are mainly explained by high 
dependence  of  the  rural  rich  on  external  remittances,  and  the  fact  that  the  moderately  poor  and  the 
middleclass  in  urban  areas  depend  more  on  income  from  the  industry  sector  than  do  their  rural 
counterparts.29 

                                                 Figure 7: Growth incidence curves (crisis vs. benchmark, 2010) 
                 (a)             Per capita expenditure: Bangladesh                                  (b) Per capita expenditure: urban and rural
                0.00




                                                                                                    0.00
                -1.00




                                                                                                    -1.00
                                                                                                    -2.00
                -2.00




                                                                                         % Change
     % Change




                                                                                                    -3.00
                -3.00




                                                                                                    -4.00
                -4.00




                                                                                                    -5.00
                -5.00




                                                                                                    -6.00
                -6.00




                        0   10    20   30   40        50       60   70   80   90   100                      0   10   20   30   40           50     60      70   80   90   100

                                                  Percentile
                                                                                                                                      Percentile

                                                                                                                                    Urban               Rural




 Source: Own simulations based on HIES (2005)
Note: 1) GICs have been smoothed to eliminate sharp kinks in the original curves 
2) Percentiles on the Xaxis are applicable to the relevant group (national, urban, rural, etc.)   
Thus  the  impact  of  the  crisis  is  higher  for  the  betteroff  and  particularly  for  the  top  30%  of  the 
distribution. The loss in welfare among the richest 10% of the rural population is larger than that for the 
                                                            
28
   A "centile" refers to 1% of the population, when the households are ordered by per capita consumption/income. 
29
   It is important to note that the percentiles of expenditures are defined with reference to each area (urban or 
rural). Since urban areas are better off on the average than rural areas, the kth percentile of urban households is 
better off than (and thus not strictly comparable with) the same percentile of rural households. 

                                                                                                                                                                           25 
 
equivalent group in urban areas, while the losses among the moderate and near poor in urban areas are 
higher  than  that  among  the  equivalent  group  in  rural  areas.  The  impact  is  lowest  among  the  extreme 
poor  in  both  urban  and  rural  areas,  just  because  the  extreme  poor  are  the  least  likely  to  receive 
remittances or work in industry even in the absence of the crisis. 

GICs  are  also  used  to  examine  the  incidence  of  change  in  per  capita  labor  income  in  each  sector  (see 
Annex  2,  Figure  A1).  As  expected,  there  is  no  change  in  labor  income  for  agricultural  workers.  Labor 
income  in  industry  falls  across  the  board,  which  suggests  that  losses  are  distributed  at  a  similar  rate 
(relative to their income in the nocrisis scenario) among all those who are employed in this sector. The 
loss of income in the service sector is much lower and significant for only the bottom 10% of those who 
earn any income from this sector. 

    Figure 8: Growth incidence curves (crisis vs. benchmark, 2010) by regions (East and West) 
                           (a) Per capita expenditure by region                                                                            (b)Per capita labor income by region
               0.00




                                                                                                                               0.00
               -1.00




                                                                                                                               -1.00
               -2.00




                                                                                                                               -2.00
                                                                                                                    % Change
    % Change




                                                                                                                               -3.00
               -3.00




                                                                                                                               -4.00
               -4.00




                                                                                                                               -5.00
               -5.00




                                                                                                                               -6.00
               -6.00




                       0   10        20       30       40         50      60       70       80        90      100                      0   10        20       30        40         50     60       70       80       90       100

                                                             Percentile                                                                                                      Percentile
                                West: Barisal,Khulna & Rajshahi                East: Chittagong, Dhaka & Sylhet
                                                                                                                                                West: Barisal, Khulna & Rajshahi               East: Chittagong, Dhaka & Sylhet

                                                                                                                                                                                                                                     
    Source: Own simulations based on HIES (2005) 
    Note: 1) GICs have been smoothed to eliminate sharp kinks in the original curves
    2) The percentiles of consumption/income on the horizontal axis are for each region (East/West) 
Finally, Figure 8(a) shows how the change in consumption due to the crisis is distributed within the two 
regions (East and West). The results in Section 4.5 have shown that the adverse impact on incomes and 
poverty is projected to be larger in the East than the West, with the most significant impacts expected in 
Sylhet and Chittagong due to their high dependence on remittances. Figure 8(a) shows that the negative 
impact of the crisis on consumption is larger in the East all along the distribution and worsens sharply as 
one moves up the distribution, while the impact in the West is smaller and more evenly distributed. This 
implies that the EastWest gap with regard to impact of the crisis widens as consumption levels increase, 
and is particularly significant for the 60th  95th percentiles of the distributions.  

Figure  8(b)  shows  that  the  EastWest  gap  is  not  primarily  due  to  differences  in  the  household  labor 
income between the two regions, which are quite small. That said, richer individuals in the East (those in 
the  60th90th  percentile)  suffer  larger  labor  income  losses  than  their  counterparts  in  the  West,  most 
likely due to the concentration of industry in the East. The small gaps in labor income also suggest that 


                                                                                                                                                                                                                                    26 
 
the  EastWest  differences  in  Figure  8(a)  are  in  a  large  part  due  to  the  impact  of  remittance  decline, 
which affects the East much more than the West. 

Movements  of  households  up  or  down  the  distribution  due  to  the  crisis.  The  GICs  described  above 
provide a clear picture of the incidence of the losses in consumption due to the crisis across the entire 
distribution.  They  also  lead  to  some  important  followup  questions  from  a  distributional  perspective: 
where would the households suffering these losses end up relative to other households as a result of the 
crisis,  and  what  is  the  extent  of  "churning"  that  occurs  in  the  distribution  as  a  result  of  the  crisis? 
Transition  matrices  are  useful  to  address  such  questions,  by  indicating  movements  up  or  down  the 
income/consumption  distribution  between  the  crisis  and  benchmark  scenarios.  These  matrices  are 
constructed  for  deciles  of  per  capita  income  and  labor  income,  keeping  the  upper  and  lower  limits  of 
each  decile  fixed  at  the  benchmark  income  levels.  This  implies  that  movements  up  or  down  by 
households  in  these  matrices  reflect  the  shifts  that  occur  as  a  result  of  the  crisis,  relative  to  the 
benchmark (nocrisis) income distribution (see Annex 2, Tables G.1 and G.2).  

Figure 9 shows the share of households in each decile that (i) remain in the same decile, (ii) move up to 
a higher decile or (iii) move down to a lower decile as a result of the crisis. Summing across all deciles, 
93% of households are found to remain in the same decile by per capita income, while 96% do so for 
deciles  by  per  capita  labor  income.  The  deciles  are  derived  from  the  benchmark  distribution  of 
consumption or income, and kept unchanged for the crisis distribution. This implies that since the crisis 
yields an overall loss of income, movements down are much more common than movements up. 

The  most  significant  downward  movements  occur  for  households  in  the  upper  middle  range  of  the 
distribution,  namely  in  the  6th9th  deciles  by  per  capita  total  income,  with  the  largest  movement 
occurring  for  those  in  the  7th  decile.  The  movements  between  deciles  of  per  capita  labor  income  are 
smaller,  with  the  largest  movements  occurring  for  the  7th  and  8th  deciles.  Upward  movements  are 
extremely rare. 

Figure 9: Movements up and down the distribution  due to crisis compared to the benchmark in 2010 
                     By deciles of per capita income                                 By deciles of per capita labor income

    100%                                                               100%

    80%                                                                80%

    60%                                                                60%

    40%                                                                40%

    20%                                                                20%

     0%                                                                 0%
              1     2      3     4      5     6    7     8    9   10             1     2      3     4      5     6    7     8    9   10
                                       Deciles                                                            Deciles 
           Stay in same decile       Movement up       Movement down          Stay in same decile       Movement up       Movement down

Source: Own simulations based on HIES (2005)
Thus the negative impact of the crisis is high for the middle and upper middle class relative to that on 
the  poor  and  the  richest.  Between  812%  of  households  in  the  5th  to  9th  deciles  of  per  capita  income 
move down to a lower decile as a result of the crisis. Labor income losses (that are significant enough to 


                                                                                                                                          27 
 
lead  to  downward  movements)  are  more  narrowly  concentrated  among  the  upper  middle  class,  likely 
because  the  losses  occur  in  the  sectors  (primarily  industry)  that  produce  relatively  high  incomes.  The 
fact  that  movements  are  less  frequent  when  households  are  grouped  by  per  capita  labor  income 
suggests that a significant part of the movement in terms of total income is due to loss of remittances. 
While there are some movements, more than 90% of all households remain in the same decile because 
the  aggregate  income  shock  due  to  the  crisis  is  quite  low,  due  to  the  limited  macroeconomic  impact 
projected for Bangladesh. 

4.7      A limited validation of the model using precrisis data for Bangladesh 

How well does our model perform in estimating the impact of macroeconomic changes on poverty and 
distribution? Addressing this question to our satisfaction will require a validation exercise that compares 
our projections with actual post (or during) crisis estimates from a household survey, which must await 
the availability of a survey conducted during or after the crisis. That said, a limited validation exercise 
can be undertaken with available precrisis data.  

In case of Bangladesh, such a validation involves using household data from an earlier year (HIES round 
of  2000)  to  "project"  poverty  and  other  distributional  estimates  for  2005,  and  comparing  these 
projections to the actual estimates from HIES 2005. In practical terms this implies replicating the exact 
same  simulation  method  described  in  this  paper  on  HIES  2000  data,  using  actual  macroeconomic 
statistics (growth rates of sectoral and aggregate GDP and remittance, and general and food CPI figures) 
for the period 20002005, which are analogous to the macro projections used for the forwardlooking 
projections described above. 

 Table  5  shows  the  results  of  this  simulation  and 
                                                             Table 5: Poverty and inequality estimates for 2005 ­ 
compares  them  with  the  actual  estimates  for            simulated and actual
2005. The simulated poverty headcount rates are                                             2000           2005
within 1 percentage point of the actual estimates                                                   Simulated Actual
for that year, with the difference between actual            Moderate poverty 
                                                               Headcount rate               48.9       39.3          40.0 
and simulated rates being 0.7 and 0.3 percentage 
                                                               Poverty gap                  12.8        9.2          9.0 
points  for  moderate  and  extreme  poverty  rates, 
                                                               Squared poverty gap           4.6        3.1          2.9 
respectively.  The  simulated  results  are  also  close 
                                                             Extreme poverty 
to  the  actual  indices  for  depth  (poverty  gap)  and 
                                                               Headcount rate               34.3       25.4          25.1 
severity  (squared  gap)  of  moderate  and  extreme 
                                                               Poverty gap/depth             7.5        5.1          4.7 
poverty, as well as for the Gini index of inequality 
                                                               Squared poverty gap           2.4        1.5          1.3 
in per capita expenditure.  
                                                             Inequality percap. Exp 
The results suggest that the model does quite well    Gini                              0.31       0.33        0.31 
in simulating the distribution of expenditures and    Theil                             0.18       0.22        0.19 
poverty  and  inequality  indices.  Interestingly,  all   Source: HIES 2005; microsimulations using macro statistics 
                                                          and HIES 2000
the  simulated  indices,  with  the  sole  exception  of 
the (moderate) poverty headcount rate, are slightly above the actual values. This seems to indicate that 
the  simulation  model  underestimates  the  consumption  growth  for  the  poorest  groups  between  2000 


                                                                                                                             28 
 
and  2005.  This  is  understandable,  given  that  the  poorest  groups  experienced  unusually  high 
consumption  growth  during  this  period  ­  per  capita  real  consumption  grew  annually  at  2.9%  for  the 
bottom decile compared to 2.4% for the population.  

Thus  the  validation  based  on  historical  data  provides  some  confidence  in  the  ability  of  the  simulation 
model to project future outcomes, linking macro statistics with econometric models built on household 
data. However, important caveats apply to these results. Firstly, the validation has the benefit of using 
actual  macro  data,  whereas  the  20092010  projections  rely  on  macro  projections.  The  ability  of  the 
model  to  predict  the  future  would  then  critically  depend  on  the  accuracy  of  the  macro  projections. 
Secondly,  the  outputemployment  elasticities  used  for  the  validation  are  derived  from  the  actual 
employment data from HIES 2000 and 2005 and macro output statistics. The same elasticities are used 
for our simulations for future years, but whether these will continue to be valid ­ particularly in a crisis 
scenario that may lead to structural changes in the economy ­ is an important question that will matter 
for  the  accuracy  of  our  projections.  Taken  together,  these  caveats  basically  imply  that  with  reliable 
macroeconomic inputs and elasticities our model does well with household data for Bangladesh. While 
this  does  not  necessarily  validate  our  future  projections  of  poverty  and  distributional  impact,  a  full 
validation will only become possible when household data from 2009 or beyond becomes available. 

    5. Conclusion: implications of our results and extensions 

Even with the caveats and limitations discussed in Section 3, our model appears to yield reasonable and 
intuitively appealing results, based on a transparent and flexible approach using macro projections and 
household  survey  data  that  are  typically  available  in  a  developing  country.  The  limited  validation 
exercise  we  conduct  on  precrisis  data  appears  to  suggest  that  the  simulation  results  on  poverty  and 
inequality from the model, based on realworld macroeconomic data for relevant years, are close to the 
actual results.  The impacts of the crisis on aggregate poverty and inequality estimates projected by the 
model also appear to be consistent with those obtained using alternate, simpler approaches commonly 
used by the Bank. 

The value that our model adds to those simpler models is the ability to analyze potential distributional 
impacts in detail, linking these changes to the channels through which the impact of the crisis is likely to 
flow. The application of our model to Bangladesh also yields some useful insights relevant to the country 
specifically, in two main areas discussed below: monitoring of impacts of the crisis and possible policies 
to address these impacts. 

Implications for Bangladesh in the context of the crisis. Firstly, our results suggest that in Bangladesh, a 
select few indicators are prime candidates to monitor as realtime or "earlywarning" indicators for the 
poverty  and  distributional  impact  of  the  crisis.  To  be  useful  for  rapid  monitoring  of  poverty  and 
distributional  impacts,  the  indicators  must  be  easily  obtained  and  measurable,  sensitive  to  changes  in 
economic conditions, and correlated with changes in poverty and distribution. By these criteria, changes 
in  remittance  flows  from  abroad,  the  movement  of  prices  (including  food  prices)  and  movement  of 
wages  by  sector  can  serve  as  useful  indicators  for  rapid  monitoring.  The  extent  of  losses  in  foreign 
remittances  and  labor  income,  and  the  price  of  food  relative  to  other  prices  emerge  from  our 


                                                                                                                  29 
 
simulations  as  key  determinants  of  the  impact  of  the  crisis  on  poverty.30  Loss  of  labor  income  in  the 
industry sector emerges as an important determinant of the impact as well. These indicators also have 
the  advantage  of  being  relatively  easy  to  monitor  ­  with  administrative  or  financial  data  (for 
remittances), and quick market or household surveys (prices and wages).  

The argument for tracking wages as a part of rapid monitoring, rather than other types of incomes or 
employment,  is  twofold.  First,  growth  slowdown  in  a  country  like  Bangladesh  is  more  likely  to  be 
manifested in lower earnings rather than open unemployment ­ a hypothesis supported by our results 
as well. Second, while wages represent only a part of the labor earnings from a sector, wages are much 
easier to track (e.g. through quarterly  labor force or even quick  enterprise surveys) than income from 
household  enterprises  and  selfemployment.  Wages  are  also  likely  to  be  reasonably  correlated  with 
broader measures of labor income, particularly in the industry sector where most of the labor market 
impact is expected.  

The  same  set  of  indicators  is  likely  to  be  useful  for  tracking  shortterm  poverty  trends  in  Bangladesh 
even  if  there  were  no  crisis.  Monitoring  shortterm  changes  in  poverty  and  distribution,  however, 
becomes especially important during an economic shock or a crisis when rapid changes in welfare are 
possible in a short timeframe that one would not expect in a "normal" year. The indicators suggested 
here can be useful in predicting such changes when a crisis hits, as well as the size of these impacts and 
how they are likely to be distributed in the population ­ given that realtime data on household income 
or consumption to measure these impacts directly are usually unavailable.       

Secondly, our results indicate that just focusing on poverty numbers would provide a partial view of the 
distributional  impact  of  the  crisis  in  Bangladesh.  The  consumption/income  impact  is  particularly 
significant  for  the  top  half  of  the  distribution  and  becomes  progressively  higher  further  up  the 
distribution.  While public  programs may not have much role to  play to cushion the losses suffered by 
the upper half of the distribution, these losses may have important political economy implications. The 
prediction that the maximum impact in urban areas is likely to be on middle class households (50th to 
80th percentiles of per capita consumption) may be particularly significant, since this group can have a 
strong influence on public perceptions about economic hardship. 

Thirdly,  our  results  provide  useful  insights  on  the  types  of  households  who  are  likely  to  be  poor  as  a 
direct  result  of  the  crisis.  Such  households  are  found  to  have  characteristics  somewhat  different  from 
those  of  the  structurally  poor,  including  a  higher  share  of  urban  and  nonagricultural  sector  workers. 
This  would  imply  that  any  intervention  to  mitigate  the  impact  of  the  crisis  on  vulnerable  households 
would need to be modified (or designed differently) from that of the traditional safety net programs.  

A discussion of what kinds of public policies and programs are best suited to mitigate the impact of the 
crisis in Bangladesh is beyond the scope of this paper. A few broad statements can however be made 
based  on  our  results  and  what  is  known  about  existing  safety  net  programs.  Given  that  safety  net 

                                                            
30
    Around  70%  of  the  aggregate  poverty  impact  is  driven  by  slowdown  in  growth  of  external  remittances;  the 
impact  of  the  crisis  in  2010  would  be  lower  (by  almost  a  third  of  the  projected  impact)  if  food  prices  were  to 
decline relative to other prices (Box 1). 

                                                                                                                                30 
 
coverage is quite limited relative to the size of the poor population,31 there is little justification for using 
the  traditional  programs  to  specifically  target  crisisaffected  households,  who  appear  to  be  better
placed  than  the  structurally  poor  households  in  terms  of  their  endowments  and  access  to  economic 
opportunities.  As  the  impacts  of  the  crisis  take  hold,  there  may  be  a  case  though  for  extending  or 
deepening certain types of existing programs (e.g. public works programs) in specific areas, regions or 
sectors. Extending coverage among the urban poor, for instance, may address an important need, given 
that the urban poor (10th to 40th percentiles) is likely to be one of the key affected groups and coverage 
of existing safety net programs is extremely low in urban areas.  

Implications for simulations in other countries. Clearly, our model is applicable to a number of country 
contexts and for addressing a range of questions related to distributional impacts of an economic shock. 
In  terms  of  where  this  approach  can  be  applied,  the  minimum  data  requirements  include  uptodate 
macroeconomic projections by sector, with and without crisis, and a fairly recent household survey with 
income data that can be classified by sector. In addition to Bangladesh, this approach is currently being 
attempted to assess crisis impacts in Philippines, Mexico and Egypt. While these countries share some of 
the  conditions  with  Bangladesh  in  terms  of  the  potential  channels  of  impact,  the  specific  country  and 
data circumstances will also require the approach to be tailored somewhat to fit the relevant context. 

Given the choice of models available to simulate the distributional impact of the crisis, it is also useful to 
revisit  the  comparison  between  existing  approaches  and  the  model  we  have  proposed  here.  The 
comparison between our model and PovStat is the most relevant, given that both models have similar 
data  requirements.  In  our  view,  the  choice  of  which  model  to  use  should  be  guided  by  the  kind  of 
projections one needs to generate. In a situation where the impact on poverty and inequality measures 
at  the  national  and  regional  levels  are  all  that  is  required  and/or  resources  in  terms  of  personnel  and 
time is limited, PovStat is likely to be the better and more practical choice. On the other hand, if more 
detailed distributional impacts ­ including estimates on how the impact will be distributed across groups 
or types of households and the sources of these impacts ­ are a concern, our model provides a feasible 
approach. Even in that case, the robustness of the results can be tested by comparing the poverty and 
inequality estimates generated from our model with those using PovStat, as it was done for Bangladesh. 

Finally, decisions to extend or nuance the type of modeling approach used here would depend on the 
availability of data and the specific country context. For example, the availability of more disaggregated 
macroeconomic  projections  would  allow  for  a  more  nuanced  distributional  analysis  that  is  able  to 
capture movements within sectors and/or types of employment. The availability of better migration and 
remittance data, particularly in the household survey, could allow for explicit modeling of remittances 
and  migration  rather  than  the  somewhat  arbitrary  assignment  rule  used  here.  Availability  of  surveys 
conducted in 2009 (e.g. any labor force or rapid monitoring survey) that hint at some of the impacts of 
the ongoing crisis would be useful in finetuning some of our analysis, particularly to take into account 
any sudden price movements or structural changes in the labor market. 


                                                            
31
   Around  13% of  households,  including  around 5%  of  urban  households,  benefited  from  at  least  one  safety  net 
program in 2005; 8% of all households benefited from targeted programs (Chapter 6, World Bank 2008).  

                                                                                                                       31 
 
                                                    References 

Alatas,  V.,  and  F.  Bourguignon  (2005).  "The  evolution  of  Income  distribution  during  Indonesia's  fast 
    growth,  198096"  in  The  microeconomics  of  income  distribution  dynamics  in  East  Asia  and  Latin 
    America, eds. Bourguignon, F., Ferreira, F., and Lustig, N., Washington DC: Oxford University Press 
    and The World Bank. 
Beach,  Ch.  (1989).  "Dollars  and  Dreams:  A  reduced  middle  class?  Alternative  Explanations",  Journal  of 
   Human Resources, 24, 162193 
Blackburn, M. and D. Bloom (1987). "Earnings and income inequality in the United States", Population 
    Development Review, pp. 575609 
Bourguignon,  F.  and  F.  Ferreira  (2005).  "Decomposing  Changes  in  the  Distribution  of  Household 
   Incomes:  Methodological  Aspects"  in  The  microeconomics  of  income  distribution  dynamics  in  East 
   Asia  and  Latin  America,  eds.  Bourguignon,  F.,  Ferreira,  F.,  and  Lustig,  N.,  Washington  DC:  Oxford 
   University Press and The World Bank. 
Bourguignon, F. and F. Ferreira (2003). "Exante evaluation of policy reforms using behavioral models" in 
   The Impact of Economic Policies on Poverty and Income Distribution, Bourguignon, F. and Pereira da 
   Silva, L. (eds), World BankOxford University Press. 
Bourguignon,  F.,  M.  Bussolo  and  L.  Pereira  da  Silva  (2008).  "Introduction:  Evaluating  the  impact  of 
   Macroeconomic  Policies  on  Poverty  and  Income  Distribution",  in  The  Impact  of  Macroeconomic 
   Policies on poverty and Income Distribution, eds. Bourguinon, F., Bussolo, M. and Pereira da Silva, L. 
   Washington DC, The World Bank. 
Burkhauser, R., A. Crews, A. Daly and S. Jenkins (1999). "Testing the significance of income distribution 
   changes  over  the  1980s  business  cycle:  a  crossnational  comparison",  Journal  of  Applied 
   Econometrics, 14, pp 253272 
Chen,  S.  and  M.  Ravallion  (2009).  "The  Impact  of  the  Global  Financial  Crisis  on  the  World's  Poorest".  
   World Bank Development Research Group. 
Deaton, A. (2001). Intervention in a panel discussion on the International Monetary Fund ­ sponsored 
   conference  on  "Macroeconomic  Policies  and  Poverty  Reduction"  April,  13,  2001,  Washington  D.C; 
   transcript at http://www.imf.org/external/np/tr/2001/tr0104143.htm 
Ferreira,  F.,  P.  Leite,  L.  Pereira  da  Silva  and  P.  Picchetti  (2008).  "Can  the  Distributional  Impacts  of 
    Macroeconomic  Shocks  Be  Predicted?  A  Comparison  on  TopDown  MacroMicro  Models  with 
    Historical  Data  for  Brazil"  in  The  Impact  of  Macroeconomic  Policies  on  poverty  and  Income 
    Distribution,  eds.  Bourguinon,  F.,  Bussolo,  M.  and  Pereira  da  Silva,  L.  Washington  DC,  The  World 
    Bank. 
Jenkins,  S.  (1995).  "Did  the  middle  class  shrink  during  the  1980s?  UK  evidence  from  kernel  density 
    estimates", Economics letters, 49, 407413 
Levy, F. and R. Murnane (1992). "U.S. Earnings levels and earnings inequality: A review of recent trends 
    and proposed explanations", Journal of Economic Literature, Vol XXX 

                                                                                                                    32 
 
Paci,  P.  and  M.  Sasin  (2009).  "Making  Work  Pay  in  Bangladesh:  Employment,  Growth,  and  Poverty 
    Reduction". World Bank Directions in Development: Poverty. 
Paci, P., B. Rijkers, and A. Seinaert (2008). "Crunch Time in the Developing World Too? Policy options for 
    dealing  with  the  potential  poverty  and  distributional  impact  of  the  financial  crisis".  World  Bank 
    Poverty Reduction Group, PRMVP. 
Ravallion,  M.  (2009).  "The  Developing  World's  Bulging  (But  Vulnerable)  Middle  Class".  Policy  Research 
   Working Paper 4816. World Bank. Washington DC 
Vos, R., L.  Taylor, L and  R. Paes de Barro eds.  (2002)  Economic liberalization,  distribution and poverty. 
    Latin America in the 1990's. UNDP 
World Bank (2007). Bangladesh: Strategy for Sustained Growth. Bangladesh Development Series. Paper 
   No. 18. Dhaka. 
World Bank (2008). Poverty Assessment for Bangladesh: Creating Opportunities and Bridging the  East
   West Divide. Report No. 44321BD. Washington DC. 
World  Bank  (2009).  Welfare  Impact  of  Rising  Food  Prices  in  South  Asian  Countries.  Policy  Note,  South 
   Asia Economic Policy and Poverty Unit. Washington DC 
World Bank PovertyNet website. "PovStat" (Accessed August 12, 2009).  
http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPOVERTY/EXTPSIA/0,,contentMDK:20483029
    ~isCURL:Y~menuPK:1108016~pagePK:148956~piPK:216618~theSitePK:490130,00.html 

 

 

                                                          




                                                                                                                33 
 
                                                                     Annex 1 

OCCUPATIONAL DECISION MODEL FOR BANGLADESH 
Individuals between 15 and 64 years old 
                                        Low skills (*)                                      High skills (**)
                             Agric           Industry       Services              Agric        Industry         Services

    male                   5.609***         4.557***       5.192***            5.991***       4.570***          3.414***
                             (0.142)          (0.132)        (0.146)            (0.477)        (0.229)           (0.162)
    yredu                   0.0864*         0.177***         0.0645              0.385         0.0728           0.854***
                            (0.0498)         (0.0492)       (0.0505)            (0.413)        (0.375)           (0.318)
    yredu2                 0.00906         0.0296***      0.0182***             0.0183        0.00203           0.0255*
                           (0.00693)        (0.00683)      (0.00699)           (0.0172)       (0.0155)          (0.0131)
    age                    0.249***         0.265***       0.291***            0.237***       0.324***          0.351***
                            (0.0183)         (0.0190)       (0.0195)           (0.0409)       (0.0400)          (0.0349)
    age2                  0.00340***       0.00399***     0.00430***          0.00307***     0.00459***        0.00469***
                          (0.000245)       (0.000260)     (0.000263)          (0.000549)     (0.000550)        (0.000489)
    hhead                  7.017***         5.722***       6.407***            9.948***       6.191***          6.331***
                             (0.225)          (0.242)        (0.240)            (0.959)        (1.073)           (0.828)
    sylh                   0.685***         0.605***        0.413**              0.232          0.232             0.117
                             (0.164)          (0.166)        (0.167)            (0.317)        (0.296)           (0.246)
    bari                     0.150          0.639***          0.126              0.319          0.392            0.0346
                             (0.155)          (0.164)        (0.159)            (0.256)        (0.243)           (0.206)
    chit                   0.504***         0.760***       0.804***            0.472**         0.0690            0.0293
                             (0.111)          (0.113)        (0.116)            (0.196)        (0.177)           (0.153)
    khul                    0.0316          0.341***         0.0130              0.194          0.214             0.225
                             (0.121)          (0.123)        (0.123)            (0.215)        (0.212)           (0.179)
    rajs                     0.189*           0.0419          0.152            0.639***         0.207            0.0682
                             (0.102)          (0.103)        (0.107)            (0.187)        (0.182)           (0.164)
    remitt                 0.563***           0.0503          0.211             0.0616          0.273            0.371*
                             (0.213)          (0.155)        (0.211)            (0.849)        (0.372)           (0.206)
    depen                  2.163***         2.067***       2.263***            1.274***       0.993***          1.255***
                             (0.202)          (0.208)        (0.212)            (0.386)        (0.375)           (0.329)
    perce                  9.412***         10.06***       9.349***            10.14***       10.84***          10.10***
                             (0.236)          (0.237)        (0.241)            (0.462)        (0.438)           (0.390)
    gen_hhd                2.462***         1.183***       1.733***            5.343***        1.967*           2.221***
                             (0.284)          (0.299)        (0.298)            (0.970)        (1.086)           (0.843)
    gen_rem                0.675***           0.0229          0.369              0.182          0.104             0.509*
                             (0.250)          (0.209)        (0.253)            (0.874)        (0.422)           (0.269)
    hhd_rem                1.853***            0.482          0.582              1.073          19.46             1.557
                             (0.333)          (0.372)        (0.400)            (1.274)          (0)             (0.977)
    gen_hhd_rem            2.087***            0.266          0.373              1.734        20.38***           2.276**
                             (0.455)          (0.491)        (0.510)            (1.394)        (0.571)           (1.121)
    low_l                  0.320***         0.872***       0.836***
                            (0.0840)         (0.0864)       (0.0877)
    hig_l                  0.665***         1.809***       1.879***
                             (0.147)          (0.196)        (0.195)
    oth_pub                1.760***         1.908***          0.444            1.375***          0.215         0.601***
                             (0.460)          (0.454)        (0.368)            (0.414)         (0.254)         (0.189)
    urban                  0.787***         0.426***       0.383***            1.502***         0.0404           0.108
                            (0.0877)         (0.0836)       (0.0855)            (0.142)         (0.133)         (0.116)
    enrolled                                                                   4.425***        4.598***        3.519***
                                                                                (0.447)         (0.442)         (0.234)
    Constant               10.72***         10.05***       10.94***            13.68***        12.19***        16.75***
                            (0.369)          (0.367)        (0.384)             (2.443)         (2.216)         (1.894)

    Observations             20331            20331          20331                7568           7568            7568
    Pseudo R2                0.515            0.515          0.515                0.517          0.517           0.517
                                                                                                                             
Notes: Dhaka is the base region; NonEmployed is the base category; Nonland is the base category. 
(*) Low skills includes those individuals with 08 years of education; (**) High skills includes those individuals with >9 years of education 
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 
Source: Own estimations based on HIES 2005 


                                                                                                                                                i 
 
 
Variable definitions for Occupational Decision Model 

     Variable                        Definition 

     male                            = 1 if gender is male

     yredu                           Maximum years of education

     yredu2                          Square of maximum years of education

     age                             Age of the individual

     age2                            Age squared 

     enrolled                        = 1 if she is currently attending school

     hhead                           = 1 if she is household head

     remit                           = 1 if the household receive any kind of remittances (domestic or abroad)

     perce                           =  ratio  between  total  number  of  income  preceptors  in  the  household  1  and 
                                     the total number of potential income preceptors within the household 

     depen                           =  dependency  ratio  between  total  number  of  members  <15  or  >64 years  old 
                                     and total number of members of the household 

     hig_l (*)                       = 1 if the household operates high areas of land

     low_l (*)                       = 1 if the household operates low areas of land

     oth_pub                         = 1 if there is another member of the household with a public job 

     urban                           = 1 if living in urban area

     sylh, bari, chit, khul, rajs    Regional dichotomic variables

     gen_hhd;       gen_rem; 
                              Interaction terms 
     hhd_rem; gen_hhd_rem 

    Note:  
    (*) Land: 
      1 Per  adult  operating  land:  we  divide  the  total  amount  of  operating  land  by  the  total  number  of 
           members above 15 years old; 
      2 Categories: we generate quintiles of per adult operating land and separate the distribution into two 
           categories:  low  operating  areas  (low_l)  includes  those  household  between  the  first  and  forth 
           quintile and high operating areas (hig_l) concentrates the fifth quintile; 
      3 The base category is households with no operating land 
 
 

                                                                                                                             ii 
 
LOG EARNING EQUATIONS (Individuals between 15 and 64 years old) 
                                                 Low skills (*)                                  High skills (**)
                                   Agric              Industry          Services      Agric         Industry         Services

    male                         1.271***            0.915***    0.785***            1.072***      0.751***           0.245***
                                 (0.0405)            (0.0407)     (0.0481)            (0.214)       (0.128)            (0.0517)
    yredu                       0.0228***           0.0396***   0.0357***            0.128***      0.119***          0.0934***
                                (0.00481)           (0.00474)    (0.00464)           (0.0256)      (0.0148)           (0.00737)
    age                         0.0549***           0.0555***   0.0730***           0.0678***     0.0996***          0.0625***
                                (0.00559)           (0.00702)   (0.00739)            (0.0221)      (0.0187)            (0.0103)
    age2                       0.000666***         0.000683*** 0.000867***         0.000831***    0.00114***        0.000582***
                                (7.12e05)           (9.20e05)   (9.47e05)           (0.000273)    (0.000240)         (0.000130)
    sylh                          0.0877              0.0466       0.0690              0.205       0.414***           0.208***
                                 (0.0608)            (0.0617)     (0.0569)            (0.229)       (0.146)            (0.0705)
    bari                          0.0335            0.222***      0.0960*              0.211        0.0728              0.0678
                                 (0.0528)            (0.0638)     (0.0522)            (0.177)       (0.115)            (0.0603)
    chit                         0.175***            0.176***    0.298***             0.291**      0.261***           0.267***
                                 (0.0391)            (0.0410)     (0.0418)            (0.143)      (0.0837)            (0.0472)
    khul                         0.119***           0.332***      0.102**              0.264*      0.269***           0.174***
                                 (0.0425)            (0.0458)     (0.0411)            (0.138)       (0.103)            (0.0519)
    rajs                         0.127***           0.375***    0.239***               0.103       0.254***           0.191***
                                 (0.0345)            (0.0356)     (0.0376)            (0.114)      (0.0801)            (0.0493)
    sala                         0.228***           0.158***    0.190***             0.656***      0.179***           0.222***
                                 (0.0303)            (0.0314)     (0.0287)            (0.138)      (0.0613)            (0.0401)
    low_l                        0.200***                                            0.729***
                                 (0.0310)                                             (0.121)
    hig_l                        0.750***                                            1.506***
                                 (0.0454)                                             (0.135)
    urban                        0.00610              0.0702**          0.0653**     0.378***       0.133**          0.163***
                                 (0.0364)             (0.0278)          (0.0272)      (0.104)       (0.0628)         (0.0352)
    hhead                                             0.180***           0.0656                      0.0806           0.0452
                                                      (0.0406)          (0.0418)                    (0.0812)         (0.0468)
    ind_exp                                            0.0627*                                       0.0698
                                                      (0.0331)                                      (0.0696)
    public                                                              0.404***                                     0.254***
                                                                        (0.0724)                                     (0.0426)
    Constant                     4.594***             5.847***          5.593***    2.889***        4.284***         5.597***
                                  (0.115)              (0.132)           (0.138)     (0.495)         (0.353)          (0.195)

    Observations                   4734                 2384              2700         802             759             1938
    Rsquared                       0.272                0.376             0.262       0.247           0.319            0.281
    Adj Rsquared                   0.270                0.372             0.259       0.235           0.307            0.277
                                                                                                                                    
Notes: Dhaka is the base region;  
Nonland is the base category 
(*) Low skills includes those individuals with 08 years of education  
(**) High skills includes those individuals with more than 9 years of education 
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 
Source: Own estimations based on HIES 2005 




                                                                                                                                iii 
 
Variable definitions for Earning Equation 

     Variable                        Definition 

     male                            = 1 if gender is male

     yredu                           Maximum years of education

     age                             Age of the individual

     age2                            Age squared 

     hhead                           = 1 if she is household head

     hig_l (*)                       = 1 if the household operates high areas of land

     low_l (*)                       = 1 if the household operates low areas of land

     ind_exp                         =  1  if  she  works  in  an  industry  which  belongs  to  the  export  sector  (clothes 
                                     production; garments; skins and chemicals) 

     public                          = 1 if she works in the public sector (government organization, public factory 
                                     or local government) 

     urban                           = 1 if living in urban area

     sylh, bari, chit, khul, rajs    Regional dichotomic variables

    Note:  
    (*) Land: 
      1 Per  adult  operating  land:  we  divide  the  total  amount  of  operating  land  by  the  total  number  of 
           members above 15 years old; 
      2 Categories: we generate quintiles of per adult operating land and separate the distribution into two 
           categories:  low  operating  areas  (low_l)  includes  those  household  between  the  first  and  forth 
           quintile and high operating areas (hig_l) concentrates the fifth quintile; 
      3 The base category is households with no operating land 
 
 
 
 




                                                                                                                                  iv 
 
                                                                                                      Annex 2 

    Table A.1: POPULATION GROWTH CHANGE 
    (in millions  medium variant)                                                                                
                                                     014                     1564                      +65                    Total

                       Total                                51.81
                                                                                    95.76                       5.55               153.12
                                                                                                                                    
            2005       Male                                 26.36
                                                                                    48.36                       2.76                 77.48
                                                                                                                                      
                       Female                               25.45
                                                                                    47.40                       2.79                 75.64
                                                                                                                                      
                       Total                                50.76
                                                                                 107.15
                                                                                                                6.51               164.42
                                                                                                                                    
            2010       Male                                 25.83
                                                                                    54.18                       3.13                 83.13
                                                                                                                                      
                       Female                               24.93
                                                                                    52.98                       3.39                 81.29
                                                                                                                                      
                       Total                                 (2.03)             11.89
                                                                                                            17.39                       7.38
             %         Male                                  (2.02)             12.02
                                                                                                            13.46                       7.29
                       Female                                (2.04)             11.76
                                                                                                            21.37                       7.47
    Source:World Population Prospects  The 2008 revision  UN Population Division                                                                  

    Table A.2: SECTORAL OUTPUT GROWTH CHANGE 
    (Base: 199596 = 100; Taka in Millions)                                                                                                     
                                                                        Benchmark                                           Crisis                           % Benchmark                                 % Crisis
                                               2005
                                                                     2009       2010                             2009                   2010              09 vs 05 10 vs 05                       09 vs 05 10 vs 05
    TOTAL                                     2,560,897
                                                                    3,311,735             3,546,604              3,286,805               
                                                                                                                                        3,469,363                   29.3                 38.5               28.3                35.5
    GDP in           Agriculture          570,367                       
                                                                       674,336               700,754                678,417                 702,840                 18.2                 22.9               18.9                23.2
    constant         Industry                     724,890              985,917
                                                                                          1,077,113                 976,016             1,034,577
                                                                                                                                                                    36.0                 48.6               34.6                42.7
    prices           Services                 1,265,640
                                                                    1,651,482             1,768,737              1,632,371              1,731,946
                                                                                                                                                                    30.5                 39.8               29.0                36.8


                     Total                               5.93                 6.62
                                                                                                    7.09                   5.82                    5.55              (0.8)               (1.5)
                     Agriculture                 2.21                         4.00
                                                                                                    3.92                   4.63                    3.60               0.6                 (0.3)
    Annual 
                     Industry                            8.28                 7.00
                                                                                                    9.25                   5.93                    6.00              (1.1)               (3.2)
    Growth rate
                     Services                            6.36                 7.50
                                                                                                    7.10                   6.25                    6.10              (1.2)               (1.0)


                                            3,848
    Remittances (US$ millions)                                              9,689              10,872
                                                                                                                      10,000
                                                                                                                                              12,100              
                                                                                                                                                                 151.8                182.5
                                                                                                                                                                                                         159.9
                                                                                                                                                                                                                             214.4
                                                                                                                                                                                                                              
    Source: BBS  GDP of Blangadesh 2008 and Projections                                                                                                                                                                                 

 
Table A.3: EMPLOYMENT SECTORAL GROWTH CHANGE 
    HIES 2005
    Individuals 1564 years old
                                        2005                                                      Benchmark                                                                               Crisis
                                                                                2009                                         2010                                       2009                                   2010
                      Millions                   %                Millions                   %                Millions                     %              Millions                 %              Millions                %
    Total                       79.7                                          89.6
                                                                                                                           89.6                                     89.6                                    89.6


    Nonemployed            41.1                          51.6                 45.1
                                                                                                    50.3                   43.3                    48.4             45.3                 50.6               43.9                49.0
    Employed                    38.5                     48.4                 44.5
                                                                                                    49.7                   46.2                    51.6             44.3                 49.4               45.7                51.0


    Agriculture                 17.1                     44.3                 17.9
                                                                                                    40.1                   18.0                    39.0             17.9                 40.4               18.0                39.5
    Industry                      9.2                    23.8                 10.6
                                                                                                    23.8                   11.0                    23.9             10.5                 23.8               10.3                22.5
    Services                    12.3                     32.0                 16.1
                                                                                                    36.1                   17.2                    37.2             15.9                 35.8               17.4                38.0
    Source: BBS  GDP of Blangadesh 2008 and Projections                                                                                                                                                                                 

 


                                                                                                                                                                                                                                  v 
 
Table B DEMOGRAPHIC, HOUSEHOLD INCOME & LABOR MARKET OVERVIEW 
                                                                                                                           Benchmark                                                                     Crisis
                                                                     2005
                                                                                                             2009                                 2010                              2009                                 2010
                                                           Quantity               %                Quantity              %              Quantity              %           Quantity             %                  Quantity          %
    Population (million)
        Total                                                      138.8                                   149.1                                149.1                            149.1
                                                                                                                                                                                                                      149.1
                                                                                                                                                                                                                       
        Urban                                                        34.3           24.7                     36.9          24.8
                                                                                                                                                  36.9          24.8                36.9         24.8                    36.9         24.8
                              (1)
       Workingage                                                    81.9           59.0                     92.0          61.7
                                                                                                                                                  92.0          61.7                92.0         61.7                    92.0         61.7

    Household Income (Tk/month)
       Total                                                   7,229.5
                                                                                  100.0                8,290.6
                                                                                                                         100.0               
                                                                                                                                            8,751.2           100.0           8,229.0          100.0               8,539.1          100.0
                               (2)
       Labor income                                            4,933.2
                                                                                    68.2               5,522.8
                                                                                                                            
                                                                                                                           66.6             5,792.8
                                                                                                                                                                66.2          5,493.9            66.8              5,701.7            66.8
                                       (3)
       Nonlabor income                                         1,757.5
                                                                                    24.3               2,226.2
                                                                                                                            
                                                                                                                           26.9             2,416.8
                                                                                                                                                                27.6          2,193.5            26.7              2,295.9            26.9
          Remittances                                              676.7              9.4
                                                                                                       1,102.5
                                                                                                                            
                                                                                                                           13.3             1,262.8
                                                                                                                                                                14.4          1,073.0            13.0              1,151.8            13.5
       Implicit rent                                               538.9              7.5
                                                                                                           541.5              6.5               541.5              
                                                                                                                                                                  6.2            541.5
                                                                                                                                                                                                   6.6
                                                                                                                                                                                                                      541.5
                                                                                                                                                                                                                                        6.3
                                                                                                                                                                                                                                         
     
                                         (1) 
    Workingage (million) 
        Labor Force                                                  79.7         100.0                      89.6        100.0                    89.6        100.0                 89.6       100.0                     89.6       100.0
            Nonemployed                                              41.1           51.6                     45.1          50.3
                                                                                                                                                  43.3          48.4                45.3         50.6                    43.9         49.0
            Employed                                                 38.5           48.4                     44.5          49.7
                                                                                                                                                  46.2          51.6                44.3         49.4                    45.7         51.0
        Mean earnings per worker 
                       (2)
    (Tk/month)                                                 3,523.4
                                                                                                        
                                                                                                       3,979.7                              4,174.9
                                                                                                                                                                              3,958.2                              4,108.0

        Education status                                             79.6         100.0                      89.5        100.0                    89.5        100.0                 89.5       100.0                     89.5       100.0
                              (4)
            Low skilled                                              59.5           74.8                     67.4          75.3
                                                                                                                                                  67.4          75.3                67.4         75.3                    67.4         75.3
                              (5)
            High skilled                                             20.1           25.2                     22.1          24.7
                                                                                                                                                  22.1          24.7                22.1         24.7                    22.1         24.7

        Employment                                                   38.5         100.0                      44.5        100.0                    46.2        100.0                 44.3       100.0                     45.7       100.0
                        (6)
            Salaried                                                 21.1           54.8                     24.2          54.5
                                                                                                                                                  25.4          54.9                24.1         54.4                    25.0         54.7
            Selfemployment                                           17.4           45.2                     20.3          45.5
                                                                                                                                                  20.8          45.1                20.2         45.6                    20.7         45.3

        Economic sectors                                             38.5         100.0                      44.5        100.0                    46.2        100.0                 44.3       100.0                     45.7       100.0
            Agriculture                                              17.1           44.3                     17.9          40.1
                                                                                                                                                  18.0          39.0                17.9         40.4                    18.0         39.5
            Industry                                                   9.2
                                                                                    23.8                     10.6           
                                                                                                                           23.8                   11.0          23.9                10.5         23.8                    10.3         22.5
            Services                                                 12.3           32.0                     16.1          36.1
                                                                                                                                                  17.2          37.2                15.9         35.8                    17.4         38.0
    Notes: (1) Individuals in 1564 age category                                               (2) Labor earnings from all activities
                     (3) Includes capital(rent land, property, profits, etc), remittances, social(insurances, charity, etc) and other nonlabor income                                    (4) Lowskilled = 09 years of education
                     (5) Highskilled = +10 years of education                                 (6) Include daily wage and salaried workers
    Source: Own estimations based on HIES 2005 and projections                                                                                                                                                                                 

 
Table C FOOD PRICES COMPARED TO OVERALL CPI 
 
                                                      CPI  Rural
                                    General            Food           NonFood      Food vs Gral
                                      (1)               (2)              (3)                                                    Weight of food in CPI and Poverty Line (PL) in rural areas
                   2005                 100.0                   100.0                       100.0                1.81           CPI                      Food                        62.96
                                                                                                                                                                                      
    Projection                                                                                                                                           Nonfood                     37.04
                                                                                                                                                                                      
                   2009                 135.7                   140.7                       126.7                5.58           PL Rural Dhaka Food                                  67.10
                                                                                                                                                                                      
                   2010                 144.8                   152.0                       131.9                6.86                                    Nonfood                     32.90
                                                                                                                                                                                      
              2010 (*)                  134.3                   132.5                       137.4                0.49                % variation in PL '09                         0.3093
    Variation                                                                                                                        % variation in PL '10                         0.4150
             2009/05            35.7                               40.7                    41.7
                                                                                                                                     % variation in PL '10 *                      (0.1086)
             2010/05            44.8                               52.0                    31.9
                                                                                                                                (*) Considers the international rice price forecast of FAO for 2010
         2010/05(*)            34.3                                32.5                    37.4
                                                                                                                               Source: BBS  Consumer Price Index (CPI) and Inflation Rate several reports                                      
 




                                                                                                                                                                                                                                        vi 
 
Table D.1: MAIN DISTRIBUTIVE RESULTS OF MACROMICRO SIMULATIONS 

                                                                     Benchmark                                         Crisis
                                                 2005
                                                                     2009          2010                2009           2010             2010*
    Moderate poverty
      Headcount rate                                 40.0                28.2        24.6                   28.6         25.8              25.4
      Poverty gap                                         9.0               5.9        5.0                   6.0          5.3                5.2
      Severity of poverty                                 2.9               1.8        1.5                   1.9          1.6                1.6
    Extreme poverty
      Headcount rate                                 25.1                16.5        13.9                   16.7         14.8              14.5
      Poverty gap                                         4.7               2.9        2.4                   3.0          2.5                2.5
      Severity of poverty                                 1.3               0.8        0.6                   0.8          0.7                0.7
    Inequality percap. Exp
      Gini                                         0.310               0.321        0.324                  0.341       0.320             0.320
      Theil                                        0.186               0.197        0.199                  0.222       0.194             0.194
    Notes: Benchmark = economy growth on trend
                     Crisis = financial crisis scenario
                   (*) Crisis with FAO rice prices forecast
     (1) Per capita Expenditure  
    (2) Total labor income 
    (3) Main labor income 
    Source: Own estimations                                                                                                                          

Table D.2: MAIN POVERTY & INEQUALITY RESULTS  DIFFERENT METHODOLGIES 
                                 Benchmark              Crisis
                        2005
                                                                             2009              2010                    2009             2010

    Headcount Ratio
    Povstat                                                  40.0                    28.7
                                                                                                   
                                                                                                  25.1                        28.9
                                                                                                                                           26.3
                                                                                                                                            
                     (1)
    Elasticity                                               40.0                    27.4
                                                                                                   
                                                                                                  24.4                        28.0
                                                                                                                                           25.6
                                                                                                                                            
    Macromicro simulation                                    40.0                    28.2
                                                                                                   
                                                                                                  24.6                        28.6
                                                                                                                                           25.8
                                                                                                                                            


    Poverty Gap
    Povstat                                                    9.0                      5.8          4.8                         5.8          5.1
    Macromicro simulation                                      9.0                      5.9          5.0                         6.0          5.3


    Severity of poverty
    Povstat                                                    2.9                      1.7          1.4                         1.7          1.5
    Macromicro simulation                                      2.9                      1.8          1.5                         1.9          1.6

                                                   (2)
    Inequality percap. Exp 
    Gini
    Povstat                                                     0.01                 0.31         0.31                        0.31         0.31
    Macromicro simulation                                       0.33                 0.32         0.32                        0.34         0.32


    Theil
    Povstat                                                     0.19                 0.19         0.19                        0.18         0.18
    Macromicro simulation                                       0.21                 0.20         0.20                        0.22         0.19
    Notes: (1) No food prices adjustment
                    (2)  Noncomparable because Povstat calculates over household head distribution
    Source: Own estimations & World Bank (2009)                                                                                                      
 

                                                                                                                                                        vii 
 
TABLE E POVERTY IMPACT BY AREA & REGION 
 
Table E.1: Household income (mean Tk/month at constant 2005 prices) 
                                               AREA                                                                                  REGION
                                TOTAL
                                                                  Rural         Urban           Barisal       Chittagong         Dhaka         Khulna        Rajshahi         Sylhet
    Benchmark
                           Total income                 8,751         7,506     12,421                 7,602         11,114       9,720       6,906         6,439     11,600
             2010




                                 Labor income           5,793         4,764       8,826                5,188           6,958       6,316       4,695         4,852       6,450
                                 Remittances            1,263         1,326       1,078                   881           2,514       1,318          563            296       3,461
    Crisis
                           Total income                 8,539         7,318     12,137                 7,456         10,748       9,480       6,793         6,353     11,159
             2010




                                 Labor income           5,702         4,699       8,657                5,115           6,836       6,208       4,633         4,788       6,347
                                 Remittances            1,152         1,209           984                 819           2,280       1,200          520            280       3,135
    Loss of income with crisis (% of benchmark)          2.4           2.5             2.3             1.9              3.3            2.5           1.6             1.3            3.8
    Loss of  labor income  (% of benchmark)              1.6           1.4             1.9             1.4              1.8            1.7           1.3             1.3            1.6
    Loss of  remittance   (% of benchmark)               8.8           8.8             8.7             7.1              9.3            8.9           7.6             5.5            9.4  
 
 
Table E.2: Poverty rate (% of population under Upper PL) 
                                                                          AREA                                                       REGION
                                                   TOTAL
                                                                  Rural         Urban          Barisal        Chittagong         Dhaka         Khulna        Rajshahi         Sylhet
    HEADCOUNT
                                     2005               40.0          43.8            28.4            52.0             34.0           32.0          45.7            51.2           33.8
    Benchmark [B]                    2010               24.6          26.7            18.1            37.8             12.4           19.2          31.7            36.5           16.3
    Crisis [C]                       2010               25.8          28.0            19.2            39.5             13.7           20.1          32.8            38.1           17.4
  [C]  [B] as % of [B] 2010            5.1        4.9    5.9                                            4.6            10.2             4.8           3.7             4.5             6.5  
 
 
Table E.3: Extreme poverty rate (% of population under Lower PL) 
                                                                          AREA                                                       REGION
                                                   TOTAL
                                                                  Rural         Urban          Barisal        Chittagong         Dhaka         Khulna        Rajshahi         Sylhet
    HEADCOUNT
                                     2005                 25.1          28.6          14.6               35.6             16.1         19.9         31.6           34.5          20.8
    Benchmark [B]                    2010                 13.9          15.9            7.7              24.7                3.4         11.2         19.5           21.5            8.3
    Crisis [C]                       2010                 14.8          16.9            8.4              25.8                4.3         11.9         20.7           22.6            9.3
    [C]  [B] as % of [B]             2010                 6.6           6.4            8.4              4.7            26.5             5.6           5.9             5.1          13.0  
 




                                                                                                                                                                              viii 
 
TABLE F PEOPLE WHO ARE POOR "DUE" TO CRISIS  2010 

                                                                     Crisis                General               Structurally 
    Household characteristics                                      Vulnerable             population                 poor
       Rural (%)                                             79.5
                                                                                                         74.7                  82.1
       No. of members                                        4.90
                                                                                                         4.86                  5.19
       Dependency ratio                                      0.42
                                                                                                         0.39                  0.48
       Employed (individuals)*                               54.8
                                                                                                         50.3                  53.3
       Sectoral share of employment (% of total employed)
            Agriculture                                      26.7
                                                                                                         39.0                  45.6
            Industry                                         27.1
                                                                                                         23.9                  19.4
            Services                                         46.2
                                                                                                         37.2                  35.0
    Household head
       Age (mean)                                          40.98                                      45.57                
                                                                                                                            43.47
       Male (%)                                              95.0
                                                                                                         89.7                  91.3
       Education level 09 years                              87.9
                                                                                                         79.1                  94.3

    Household income (CrisisVulnerable)                            Benchmark                   Crisis
       Income (mean)
         Total Income (tk. 2005 prices)                                      6,741.2          5,165.2
                                                                                                                                 23.4
                                                                                                                                  
         Labor income                                                        4,113.8          4,046.0
                                                                                                                                    1.6
         Nonlabor income                                                     2,268.0              
                                                                                                 759.9                           66.5
                                                                                                                                  
             Remittances from abroad
              % of household  receiving                                            45.7
                                                                                                           9.9                   78.5
                                                                                                                                  
              Mean conditional on receiving                                  1,681.1
                                                                                                      178.9
                                                                                                                                 89.4
                                                                                                                                  
              Implicit rent                                                      359.4                359.4
                                                                                                                                   
                                                                                                                                    
       Percapita Expenditure (mean)                                          1,278.9
                                                                                                      785.9
                                                                                                                                 38.5  
                                                                                                                                  
 
 
TABLE G INCOME STATUS "CHANGES" 
    Table G.1: Transition matrix: Benchmark  Crisis
    Percapita income
                                                                           Crisis
                            1        2          3          4           5            6          7            8           9           10
                   1     100.0
                   2       2.79
                                    97.21
                   3       0.68
                                      4.21
                                               95.11
                   4       0.34
                                      0.46
                                                 6.18
                                                          93.03
      Benchmark




                   5       0.69
                                      0.68
                                                  
                                                 1.33       5.79
                                                                      91.50
                   6       0.60
                                      0.35
                                                  
                                                 0.39       0.58
                                                                        6.89   91.18
                                                                         
                   7       0.33
                                      0.14
                                                  
                                                 0.71       0.38
                                                                        2.62      
                                                                                 7.26   88.55
                   8       0.25
                                      0.49
                                                 0.27
                                                             
                                                            0.92        0.93      
                                                                                          7.34   88.58      
                                                                                 1.10                       0.12
                   9       0.11
                                                 0.23
                                                             
                                                            0.32        0.49      
                                                                                          1.32      
                                                                                 0.32              6.30   90.91
                  10       0.11
                                                                                  
                                                                                 0.04      
                                                                                          0.17              4.72
                                                                                                   0.61                             94.35

    Table G.2: Transition matrix: Benchmark  Crisis
    Percapita Labor income 
                                                                Crisis
                          1        2        3        4        5          6        7        8        9       10
                   1   100.0
                         1.59      
                   2              98.4
                   3              2.78      
                                           97.2
                   4                       5.03   94.85      
                                                             0.12
      Benchmark




                   5                                5.30      
                                                             94.7
                   6                                         6.87      
                                                                        93.1
                   7                                                    6.95      
                                                                                 92.9              0.12
                                                                                                    
                   8                                                             7.72   92.28
                                                                                  
                         0.07
                   9                                                                      4.11      
                                                                                                   95.8
                  10                                                                               2.82      
                                                                                                            97.2                             

                                                                                                                                                ix 
 
 
 
 
    Figure A1: GICs (crisis vs. benchmark, 2010) 
    Per capita labor income by sector




                0.00
                -1.00
                -2.00
     % Change
                -3.00
                -4.00
                -5.00
                -6.00




                        0   10   20       30        40       50        60    70   80         90   100

                                                         Percentile

                                      Agriculture                 Industry        Services

                                                                                                         
    Source: Own simulations based on HIES (2005)
    Note: 1) GICs have been smoothed 2) Percentiles of income (x
    axis) refer to the relevant sector
 




                                                                                                            x