Microeconomic Aspects of Development
UNIT 1 - NOTES
Debraj Ray
2.2.1 - Measurement Issues
Low per capita incomes are an important feature of economic underdevelopment - distribution of
income is extraordinarily skewed across countries. Comparison between countries - each
country's income converted to a common currency and divided by population to get a measure
of per capita income. Such GNP per capita estimates serve as a yardstick for comparison
between countries and their incomes. Though disparities are enormous, a better understanding
of the degree of international variation through a reliable analysis that recognizes the biases of
these estimates:
(1) Underreporting of income - Tax collection systems are not equally efficient across
countries (more efficient in industrialized market economies and incentive to
under-report for tax purposes), national income accounts may not be comprehensive,
self consumption - relatively high in developing countries, so subsistence output is not
adequately reported. Though an educated guess about relative underestimation can be
made - very little can be done to correct problem
(2) Prices don’t reflect exchange rates - Exchange rates reflect prices of only tradable goods
and not non-tradable goods (which are typically cheaper in poor countries), so people in
poor countries have a higher purchasing power (as non-tradable goods are cheaper)
which is not reflected in the exchange rates, so it underestimates the real income and
real living standards of people in poor countries. PPP adjusts for differences in price
levels across countries and gives a more accurate measure of real income and living
standards
(3) Market prices reflect people’s preferences as well as relative scarcities, however not all
markets are perfectly competitive, and neither are prices fully flexible. Prevalent prices
do not capture the true marginal social value or cost of a good or a service. (Positive and
negative externalities of goods and services like government spending and pollution may
not reflect the true value of goods and services)
2.1
Economic development is the primary objective of the majority of countries. To raise the
well-being and economic capabilities of people is the most crucial social task for everyone.
Intuitively people have different notions of development, but most would agree that “a minimal
requirement for a “developed” nation is that the physical quality of life be high, and be so
uniformly, rather than being restricted to an incongruously affluent minority.” Development
further encompasses political rights and freedoms (relate it to what Sen talks about), intellectual
and cultural development, stability, low crime rates, etc. GNP per capita, is just a unilateral
measure of development that barely touches one node. “Along with higher income, one needs
removal of under nutrition, an increase in life expectancy, access to sanitation, health services,
reduction in infant mortality, increased access to quality education,” a step towards
understanding the multidimensionality of development. People believe that, “the universal
features of economic development—health, life expectancy, literacy, and so on—follow in some
natural way from the growth of per capita GNP.” However, correlation between GNP per capita
and other indicators is not as forward as presented and in some cases may be non-existent.
GNP per capita is not an adequate overall measure and must be supplemented with other
measures. The book focuses on how the average levels and distribution of economic attainment
influences development.
2.4
Limitations of GNP per capita being the sole measure of development - income distribution and
social factors like health, education, and equality (gender), also are vital indicators of well being.
Example: Guatemala and SriLanka, even with similar per capita income levels, had vastly
different outcomes in HDI indicators like infant mortality, life expectancy, and literacy rates, due
to the difference in income distribution. HDI, a composite measure of life expectancy (health),
education (2/3rd weight to adult literacy and 1/3rd weight to enrollment), and adjusted
(diminishing marginal utility of income) income per capita. Per capita income has a strong
correlation with other indicators of development (like life expectancy, literacy, and so on).
Generally countries with higher incomes tend to have better HDI scores (outcomes of the
indicators). However this correlation is not perfect, as there are examples to the contrary like Sri
Lanks and Guatemala or Sri Lanka and Pakistan. Income therefore does not capture all aspects
of development, nevertheless is a useful measure and proxy for development.
8.1 and 8.2
Most visible characteristic of economic underdevelopment - Poverty (overgrown layers of
inequality). “Poverty and its correlates - illiteracy, undernutrition, ill health, bleak future.” Poverty
has both intrinsic and functional significance in how it takes away opportunities and freedoms
from people. Economic development - fundamental goal removal of poverty.
“Poverty line: a critical threshold of income, consumption, or, more generally, access to goods
and services below which individuals are declared to be poor.” Poverty line represents a minimal
level of ‘acceptable’ economic participation. Various estimates of poverty line - Income, nutrition
based, and minimum wage. Nutrition-based poverty lines like minimum calorie consumption are
commonplace enough. Fundamental problems across measurement of poverty:
(a) Who is considered poor? When their actual observed consumption basket falls below the
threshold or when the expenditure of the person falls below the minimum standard. How
is the consumption standard defined? Income elasticity of nutrition may not be high.
(b) There is an absolute sense of poverty - adequate levels of food, clothing, shelter.
However, what is considered ‘adequate’ is relative to interpretation based on the various
prevailing socio-economic standards. Though poverty lines should incorporate relative
notions of what is necessary and what constitutes basic needs, they need to fulfill the
basic necessities in some absolute capacity. (absolute vs relative notion of poverty)
(c) Some people who live close to a state of poverty are subject to fluctuations in income
(economic exogenous shocks push them below the line), like weather dependent small
agri farmers. So notions of chronic and structural poverty must include the vulnerability
to poverty of people. This is because they need different policies to combat temporary vs
chronic poverty.
(d) Though we use household level expenditure data for understanding poverty, the
allocation of expenditure within households is often skewed (gender inequality,
discrimination increases with destitution). Large household also have higher no. of
children, this skews consumption as children consume less than adults (soln adult
equivalence scales)
“It is important to realize that poverty lines are always approximations to a threshold that is truly
fuzzy, more because the effects of sustained deprivation are often felt at a later point in time.”
Headcount ratio: Head count (given by the number of individuals i income of i < p) / n
(population)
Poverty gap ratio (measure of resources required to eradicate poverty): m = income
(denominator is mean income, numerator is extra consumption needed to get all poor ppl to p)
Income gap ratio (captures acuteness of poverty more directly as it measures poverty relative to
total income needed to make poverty go away): HC = head count of poor;
Sachin and Nagendra
Expert group - Chairman Suresh D Tendulkar - Planning Commission, 2005 - Mandated to do:
(a) Issues related to comparability around consumer expenditure data and suggest methods
methods to derive comparability with past and future survey data
(b) Review alternatives conceptualizations of poverty and associated technical procedures
of measurement and empirical estimation
(c) Recommend any changes to existing procedures of official estimates of poverty
In the report that the Tendulkar committee produced 3 deficiencies in the Lakdawala poverty
lines were noted:
(a) Poverty line baskets remained tied to consumption patterns observed in 1973-74
however the baskets shifted even for the poor. CPI of agriculture workers understated
true price increase (Estimated poverty ratios understated rural poverty). The assumption
that govt provides health and education is not true as pvt exp had risen considerably
even for the poor (not reflected in the Lakdawala PL).
(b) Converted URP consumption based PL to MRP-consumption based PL
Martinez
Governments, in general, have an incentive to overestimate their GDP for various socio-political reasons.
Democracies have a solid system of checks and balances in the form of opposition parties, the media, the
judiciary and the public that constrains their ability to manipulate data. Autocratic regimes on the other
hand restrict civil liberties that curbs their ability. The main research question being examined in this
paper is to check if there is a systematic overstatement of GDP in autocratic regimes compared to
democratic regimes. This is done by comparing the data about Night Time Lights (NTL) recorded by
satellites and self reported GDP figures. The underlying hypothesis is to see if the elasticity of GDP with
respect to NTL differs between autocracies and democracies.
Reported GDP growth rate (g hat) is dependent not only on true gdp gr (yit) but also on who is reporting it
(the autocracy variable)
(Regression Framework)
The average NTL elasticity of GDP is 0.3. Conditional to NTL growth when FiW index is added as an
additional regressor, GDP growth is found to be smaller in countries with more authoritarian regimes. This
does not take into account proportional overstatement of GDP growth rate in autocracies. To address this
limitation, an interaction term between ln NTL and FiW is added. This causes the point estimate of FiW to
be insignificant. The coefficient of interaction term on the other hand is positive and significant. A unit
increase in FiW is associated with an increase of 0.02 units in the NTL elasticity of GDP. Autocracies
overstate yearly GDP by 35% (i.e., a GDP of 1% is reported as 1.35%).
(Robustness checks)
“Overstatement of GDP is the underlying factor of the autocracy gradient in the NTL elasticity of GDP”.
Three possible mechanisms that the author discusses in support of this statement are as follows:
(a) Sub components of GDP - GDP is decomposed in terms of expenditure into sub components.
These sub components consist of investment, government spending, private consumption,
exports and imports. The author conducts an analysis of the sub components to identify which of
the sub components are drivers of the autocracy gradient in the NTL elasticity of GDP. The
results of the analysis show that the overstatement of GDP in autocratic governments is
concentrated in the sub components of investment and public spending. This intuitively makes
sense, as governments are the primary source of information for public spending and
investments, so it is easier to manipulate this data. Comparatively, private consumption can be
verified through retail sales data and surveys. Exports also have little room for inflation as it is
hard to align balance sheets with trade partners.
(b) Relative Economic Underperformance - The author examines whether the autocracy gradient in
NTL elasticity of GDP is larger when an economy is underperforming compared to the rest of the
world. In the analysis, the dependent variable in all equations is ln(GDP). It includes a dummy for
low growth, ln(NTL) and FiW index (Freedom in World - range from democracy to autocracy), and
a triple interaction between these other variables. Even when NTL elasticity of GDP is allowed to
vary with years of low growth the baseline results remain the same. That is, they suggest that the
incentive to overstate GDP is higher when the economy is relatively underperforming.
(c) Foreign Aid Eligibility - The author also examines the impact of changes in concessions and loans
offered by IDA (International Development Association) on the autocracy gradient in the NTL
elasticity of GDP. The results show that the autocracy gradient in the NTL elasticity of GDP is
observed only for the subset of developing countries that cross the threshold and become
ineligible for grants and loans from IDA. This implies that countries that are close to the
ineligibility mark have a weaker incentive to overstate GDP as doing so will cause a large
reduction in funds in the form of foreign aid.
Though Night Time Light (NTL) as a variable is not as noisy (not endogenous) as GDP and a good proxy
for economic development, there are some potential issues that arise due to the use of night lights. The
elasticity of population density and NTL is not captured (areas with high population might have fewer
lights and vice versa). The landmass and population distribution across countries also creates confusion
in understanding the reality of development especially in distinguishing different levels. That is, it is hard
to gauge relative development and depth of underdevelopment. It is also easy to confuse urbanization
and development, as even if there is correlation it is not always exclusive (that is urbanization does not
imply development). Infrastructural factors of how the government chooses to spend and cultural factors
are also not highlighted in this proxy.
ILO
How one may construct summary measures of the extent of poverty?
(1) Head count ratio - measures proportion of population that is poor. Though it is easy to
understand and measure, it does not indicate the depth of poverty, i.e., how poor the
poor are. If yi < z then I is 1. And summation of all Is gives us Np - total no. of poor.
(2) Poverty gap index measures the extent to which individuals fall below the poverty line as
a proportion of the poverty line. The sum of poverty gaps gives the minimum cost of
eliminating poverty (This interpretation only has meaning if transfers are perfectly
efficient. This does not measure the inequality among the poor. Gi = Poverty gap. The
poverty gap measure has the virtue that it does not imply that there is a discontinuity
(“jump”) at the poverty line.
(3) The squared poverty gap index averages the squares of poverty gaps relative to the
poverty line. By squaring the poverty gap index, the measure implicitly puts more weight
on observations that fall well below the poverty line. α is a measure of the sensitivity of
the index to poverty. When parameter α=0, P0 is simply the head-count index. This is a
poverty severity index.
HCR - violates transfer principle (transfer from richer to a poorer person improves welfare), but if
a poor person gives to a poorer person HCR does not change even though overall poverty
lessened. HCR does not indicate depth of poverty. Poverty estimates should be individual
based not household based (as ratio may be higher in the former’s case).
UNIT 2
Banerjee and Duflo (Chp - 1)
“Poor countries are poor because they are hot, infertile, malaria infested, often landlocked; this
makes it hard for them to be productive without an initial large investment to help them deal with
these endemic problems. But they cannot pay for the investments precisely because they are
poor—they are in what economists call a poverty trap.” Foreign aid - kickstart a virtuous cycle -
making investment in critical areas.
When markets are free and incentives are right people can find ways to solve their problems.
But in this sense aid pessimists are quite optimistic about how the word works (ironic?)
Aid beneficial / not necessary - two sides of the debate (on one side countries that received aid
did not grow faster than countries without but this could also mean aid helped them avoid a
major disaster - speculation on large scale)
It matters where the aid money goes - food, health? Poverty - not having the capability to realize
one’s potential
Trapped in Poverty - They are poor because they are poor (need to be dislodged from where
they are stuck - Sachs emphasis on a big push). “ There will be a poverty trap whenever the
scope for growing income or wealth at a very fast rate is limited for those who have too little to
invest, but expands dramatically for those who can invest a bit more.”
S Curve - S shape source of poverty trap (45 degree line: income today = income tomorrow),
the very poor in the poverty trap zone income in the future is lesser than income today (curve
below diagonal line)
Balboni et al.
Two views for why people stay poor:
Difference in fundamentals (lack of motivation, ability & talent)
Poverty trap (difference in opportunities due to lack of access to wealth)
Data supports poverty trap view - they identify a threshold level of assets above which
households accumulate assets, take better occupations and grow out of poverty. For those
below the threshold the reverse happens.
Data collected to evaluate BRAC’s Targeting the Ultra-poor (TUP) Program in Bangladesh
In the context of asset transfers, such as giving cows to poor households, the success of
beneficiaries in escaping poverty often depends on their pre-existing assets. For example, if
reaching the market requires a cart, only those who can afford a cart may succeed in running a
livestock business, illustrating the threshold required to escape poverty.
This leads to a pattern consistent with poverty traps, where those below a certain asset
threshold lose wealth over time, while those above accumulate more. Over four years,
households that start with very low assets and receive transfers often fall back into poverty,
while those who surpass the threshold continue to accumulate assets, improve their livelihoods,
and increase consumption. Over an 11-year period, the two groups diverge even more: the
wealthier group moves into more productive occupations and escapes poverty, while the poorer
group remains stuck.
The study accounts for variations in human capital (e.g., earning potential and saving rates) and
finds that individuals with higher earning potential have lower asset thresholds to escape
poverty. However, without surpassing their personal thresholds, even asset transfers may not lift
the poorest out of poverty, as they lack productive assets like livestock. Their labor is often
wasted in low-wage, unreliable jobs, preventing them from saving and investing in better
opportunities.
The S-shaped curve of asset accumulation shows two stable states, where households with
assets below a critical threshold (Kc) will lose wealth and stay poor, while those with assets
above it can escape poverty. This suggests that the poor remain trapped not because of their
abilities but because of a lack of opportunities.
Large, one-time asset transfers can have lasting impacts, enabling people to change
occupations and escape poverty permanently, unlike smaller transfers that only increase
short-term consumption. These "big push" policies can create long-term benefits by addressing
labor misallocation, as many poor individuals remain stuck in low-productivity wage labor,
despite having the potential to thrive in more productive activities like livestock rearing.
Reetika Khera
Disconnect between theoretical and empirical work on in-kind transfers.
Methodology - The paper is based on a household survey conducted with more than 1200 rural
households who currently have access to subsidized food transfers from PDS. The PDS has
been criticized as an inefficient and expensive mechanism of enhancing food security. Poor
targeting, corruption (or ‘diversion’), and high operational costs are among reasons used to
justify switching from universal PDS to Targeted PDS.
Investigators in the survey sought to see whether respondents prefer existing in-kind transfers
of subsidized food or a cash equivalent.
Concerns related to PDS:
Corruption, Irregularity, Unpredictability, Poor quality of grain, hassles in getting grain
Advantages of PDS:
Food security, protection from misuse of money, familiarity, convenience, feeding guests,
lower traction costs
Concerns related to cash:
Cash will be dissipated, inflation, local hoarding and shortage, corruption, distance to
markets and banks, hassle of collecting money, unfamiliarity of cash system, cash
maybe stolen, transactional costs
Advantages of cash:
Fungibility, diversified diets, better quality of food, no fraud, convenience, allows to save
Food over cash existing theoretical arguments - Paternalism (“intimately related to the idea of
merit goods and merit wants and as a means of steering ‘people’s choices in the directions that
will improve their lives”); Food security (source of security, an assurance they wouldn’t sleep
hungry), Protection from exploitation - (cash swindles out of people more easily than food),
Self-control (cash spent unnecessarily on alcohol or wasteful expenses). Interdependent
preferences that give rise to consumption externality also justifies in-kind transfers (cash for
themselves but food for others as it is more secure). Pecuniary benefits - in-kind transfers
protect from inflationary pressures (inflation leads to erosion of purchasing power), inflow of
cash through cash transfers can itself contribute to inflation (if supply is inelastic). Poverty and
food security, poorly developed rural markets (greater uncertainty while converting cash
transfers into food grains due to lack of market, possibility of hoarding and artificial storages
when markets are poorly developed, localized inflation due to hoarding), access to banks
(greater flexibility with rations, familiarity and convenience), intra-household inequality and
gender (lack of physical mobility for women), transactional costs and transitional costs
(accountability of the government);
Cash over food: Tamil Nadu has a universal PDS (therefore contrasting interests of poor and
non poor). Fungibility of cash, quality of grain, regularity, corruption (as they receive money
directly).
Jayachandran and Pande (Why are Indian children so short?)
Height deficit for India compared to Africa is more pronounced in girls than boys
General son preference
● Sibling Rivalry (Birth order - Older brother competing for resources - son
preference)
● Fertility stopping behavior (desire for son implies family size may increase
beyond the ‘desired’ size, fewer early life resources, subsequent order of birth for
girls = fewer resources)
UNIT 3
Dreze and Deaton
In low income countries, there is a very clear correlation between incidence of poverty and
incidence of under-nutrition.
However, the relation between increase in income and increase in nutrition may or may not be
as strong.
Increase in incomes may raise nutritional levels through consumption of more quantity of food.
Preferences also come into play and the kind of foods consumed may depend on social factors,
availability, visibility etc.
Best measure - income elasticity of calorie/nutrition intake
Even though per capita consumption has been steadily rising since the 1980s, calorie intake is
falling. Attributed to changing consumption patterns - Steady decline in calorie & protein
consumption (rural very close to urban).
-> Richer households consumer more calories (higher income higher consumption)
-> Overtime consumption has declined for each level of expenditure, especially in rural areas
Standard measures of nutrition
(a) Height for age - stunting
(b) Weight for height - wasting
(c) Weight for age - underweight
Since 1998-99, stunting has gone down but wasting increased
Why care about nutrition? - Intrinsic value of being nourished and healthy and instrumental
value of capacity to work
Calorie intake declined not because its expensive or impoverishment, but due to change in
consumption patterns and declining need for calories
Calorie intake - poor indicator of nutrition
Jayachandran and Pande
Eldest son pref- patrilocal and patrilineal Hindu kinship system. States like Kerala, birth order
gradient shallower. Muslim children- shallower birth order gradient.
son-preference mechanism predicts that the drop-off in height across successive children will
depend on their older siblings’ gender composition. a son born at birth order 2 is taller in India
than Africa if and only if his older sibling is a girl: i.e., he is the family’s eldest son.
Younger daughters- late birth order and sibling rivalry and fertility-stopping mechanism- has to
compete with older siblings ( who may be brothers) for family resources. When a daughter is
born into a family with only girls, her parents are likely to continue having children in their quest
for a son, exceeding their originally desired family size.
Daughters with no elder brother- lack of sibling rivalry will increase both prenatal and postnatal
investments but fertility-stopping effect reduces the postnatal investment they receive - parents
try for a boy.
Prenatal behavior includes the number of prenatal care visits, whether the pregnant woman
received tetanus shots and iron supplementation, and delivery at a facility; India typically
outperforms Africa on these measures
In figure 2, An Indian deficit emerges at birth order 2 and widens for birth order 3 and higher.
HFA imc = α1 I c + α2 I c × 2ndChild imc + α3 I c × 3rd+Child imc + β1 2ndChild imc + β2
3rd+Child imc + γ X imc + ϵim where I c is an indicator for Indian children; α1 is the India gap for
first-born children (omitted birth order category), and α2 and α3 capture how the gap differs for
second-born children and third-and-higher birth order children.
Endogenity concerns- householdsmust have completed fertility. Higher birth order children are
more likely to come from larger families, and family size could be correlated with child height;
family size could affect child height via its effect on the available resources per child, plus larger
families tend to be poorer. So fam size held fixed when making sibling comparisons.