Which method, proxy means test or community targeting,
performed best at identifying free health care card beneficiaries?
For fighting against poverty governments and organizationsi2
work together and they conduct poverty alleviation programs. However,
government and organization’s resources aren’t unlimited and it requires that
the poor individuals are identified correctly and that their needs are well
understood. Thus, targeted social safety net programs have become an
increasingly common tool to address poverty (Coady, Grosh, and Hoddinott 2004).
In developed countries, income is observable so mean testing is enough to
determine eligibility thresholds. However, most of developing countries, income
may not directly observable because of lack of records and unstable income.
Most of people work in agricultural or farming sector so, they are open natural
shock. Their income fluctuates year by year even month, so it is hard to
determine certain threshold leveli3 .
For these reasons, identifying beneficiaries is a big
challenge. My research area will be Burkina Faso which is one of the poorest
countries in the world with just under half its population living under the
poverty line. In this research, I will study two methods that are commonly used
to target poor households in this context: PMT and CBT for determining
According to 2015 WHO data of Burkina Faso only 5% of
GDP allocate for health expenditures. Also, life expectancy is around 60 years
for both sexes, under 5-year mortality rate around 32i4 %.
All these numbers are evidence of weak access to proper health care. This research
aims to provide a new set of poverty criteria based on community perception and
Proxy mean method and understand which method is more successful for
determining real beneficiaries in case of giving free health care card.
Cash transfers have been increasingly adopted by low-
and middle-income countries as central elements of their poverty reduction and
social protection strategies (Barrientos, 2013; DFID, 2011; Hanlon et al.,
2010; Honorati et al., 2015; ILO, 2014). However, there is other programsi5
which have other objectives than transferring money to poorest.
These programs provide in-kind benefits. Unlike cash
transfer programs, these benefits are not paid directly in cash. Instead of
this, poor individual receives good or services free or reduced rate. For
Medicare which subsidized health care for elderly people and disabled. Other
example is food stamps. It is federal aid program, designed to temporarily help
people paid for groceriesi6 . Although,
cash transfers provide freedom of choice to people, in-kind programs better
help to reach program’s aim. Because, in-kind benefits must be used for just
targeted uses. In my research, there will be in-kind benefits for poor
households. Poor people with poor health condition will get health card which
ensures free health services. It is important to specify objective of program
and its implementation before choosing targeting method.
In the growing literature, most of study1
shows that CBT has several advantages; involving community may lead to monitoring
and accountability. “Local
community agents often have better information on household characteristics,
needs and recent events upon which to condition beneficiary eligibility than do
outside welfare agents who must often rely on crude and outdated proxy
indicators” (Cremer, Estache and Seabright 1996). However,
CBT is not always and everywhere the optimal policy. Community decision can be
less transparent and manipulable when there is inequality and stratification in
the society. In a study of community heterogeneity and inequality in rural
villages in Tanzania, La Ferrara (1999), found evidence higher inequality cause
less democratic decision-making processi7 .
PMT is another method which helps to measure people
welfare. Chile is the first country to use this approach in Ficha CAS program
in 1980. Then it spreads other countries; Colombia (subsidized health insurance,
conditional cash transfer programs), Mexico (conditional cash transfer programs
(PROGRESA)). In PMT, variables are chosen which are corelated well with
poverty. And these variables use for well-being measure. It is less open to
manipulation however, it can’t measure recent shocks that household has
experienced. One example is father of home is get sick and he isn’t able to
work for 1 month. In one month, family still have good assets, but their income
dramatically decreases that’s why after couple months if this condition
continues, they will be in poverty. PMT can’t capture these peoplei8 .
we see, that each method has advantages and disadvantages. According to
community and program design we should find which method is best for picking
In this research has descriptive and exploratory study i11 using qualitative methodsi12 . This research data is based on
household surveys which conducted in dry seasons of Burkina Faso. Participants answered questionnaire which
includes household composition,
assets, and another dimension. For measuring which method is capturing poor
better we should check exclusive (rich but receive health card) and inclusive
error (poor but not received health card) rates. To identify the determinants
of exclusion and inclusion errors, I will use probit model. Probit model gives
result of probabilities of non-poor households being wrongly included and poor
household being wrongly exluded.
Nacional de Solidaridad in Mexico (PRONASOL), Uzbekistan(1994), Albania(Ndihme
Ekonomika), Armenia and China(Five Gurantee)
motivations is quite good.
You could add a section after the literature review
which is “contribution” (i.e. contribution to the literature). There, you can
explain what is different / original in your research.
maybe 32 per 1000, and still high
the US? Elsewhere?
US examples can be interesting, but in general
developing country examples are more interesting.
but you could have used more studies from the literatüre. You actually know
have another paragraph about the existing studies comparing them (like yours)
is too short. This is the most important section.