Lecture 01
Economics of Poverty, Inequality
and Gender
Chapter name: Measuring
Welfare
While money can’t buy
happiness, it certainly lets
you choose your
own form of misery.
- Groucho Marx
Background:
At the foundation of most measures of poverty and
inequality is a concept of individual welfare.
In economics, “welfare” (or “well-being,” which is
used interchangeably here) is generally equated with
“utility”—a subjective assessment of all the things a
person cares about.
Economists have often tried to infer what those things
are from behavior, and one aspect of behavior in
particular: what people choose to buy and sell in
markets.
A broader concept of welfare has also been sought,
allowing for external evaluations of a person’s welfare
that may or may not accord with their utility, defined
as whatever people maximize.
Differences over how one thinks about welfare can
matter greatly to the descriptive and normative
claims made about poverty.
Concepts of Welfare
Approaches to welfare measurement differ in terms of the
importance attached to the individual’s own judgments about
his or her well-being.
They also differ in terms of the factors they try to include
within a single measure.
It is very widely agreed that individual welfare depends in
part on household command over commodities, but it is also
widely agreed that it depends on other things as well.
The debates are mainly about what other factors are relevant
and how they should be weighted.
Welfarism
The standard economic approach to monitoring social progress
overall and assessing policies aims to rely solely on the
individual welfare levels in the relevant population.
Social states are judged by (and only by) individual welfare
levels. (This is sometimes called “individualistic.”) This
approach has its roots in classical utilitarianism.
One specific definition of welfarism says that we should strive
to base welfare comparisons and public policy decisions solely
on the individual utilities, defined as what people maximize in
their own choices.
It is clearly a big step to equate the personal objective that
guides one’s choices— the utility function that represents the
set of indifference curves.
Welfarism
An important message of this version of welfarism is that, in assessing individual
well-being, one should avoid making judgments that are inconsistent with the
preferences that guide people’s own choices.
Each person is presumed to be a rational actor maximizing his or her utility.
This approach will include all those commodities that people chose to consume in
assessing their welfare.
To say that this only includes market commodities is an unjustified specialization.
As long as markets exist and are deemed to be competitive, prevailing prices can
be used for aggregating the commodities consumed and in deflating for differences
in the cost-of-living to derive the welfare metric.
However, this is a partial welfare metric to the extent that people also care about
non-market goods.
Utility based approach
The utility-based approach draws on a model of rational
consumer choice. The essence of the approach is the idea of a
utility function.
This serves two distinct roles in utility-based welfarism:
First: It is a convenient representation of the
consumer’s preferences over her affordable
consumption bundles.
The consumer is presumed to be able to order those bundles
from the best to the worst and pick the best among the
feasible options.
In this first role, the utility function is nothing more than an
analytically convenient way of representing the consumer’s
preferences.
Consumer Choice
To keep things simple, suppose that there are two
goods:
Food and Clothing
They are consumed in the amounts QF and QC with
prices PF and PC (“F” for food and “C” for clothing).
Total spending on the two goods is denoted Y, and
this is held constant in this thought experiment.
The affordable sets of bundles (QF,QC) are those for
which:
PFQF + PCQC ≤ Y.
Consumer choice
The consumer is able to rank the affordable combinations,
(QF,QC), satisfying this equation.
Assuming that more of good is better (often called the
“non-satiation” assumption).
We only need to consider the budget allocations that
exactly absorb the available budget (since if one is at an
interior point where PFQF + PCQC < Y, then it is possible
to afford more of one good with no less of the other).
The consumer is assumed to be rational, meaning that she
picks her preferred bundle.
And the economy is competitive, in that the consumer
cannot alter the prices faced.
Consumer choice
The slope of the indifference curve is called the
marginal rate of substitution (MRS), which is
defined as the increment to consumption of one good
that is needed to compensate for one less unit of
another good, while keeping utility constant.
The quantities of food and clothing that maximize
utility are denoted Q∗F and Q∗C.
These will clearly depend on the prices, PF and PC,
and of course income, Y.
Utility maximization
Critics of Consumer choice
Critics of this model have pointed to situations where
personal choices do not appear to be rational.
There is a risk here that what is seen to be “irrational” may
well reflect an overly narrow view of what people care
about.
For example, if we ignore the fact that people (including
poor people) have concerns about their relative position in
society, as well as their absolute level of living, we may
misunderstand their behavior, such as when they spend
their scarce resources on celebrations.
To give another example, people may derive utility today
from knowing that they will be less poor in the future, and
this may influence inter-temporal decision-making.
Utility based approach
Second: The utility function is assumed to provide
sufficient information for assessing whether a person
is better off over time, or after a policy change, or in
determining one person’s welfare relative to another.
This latter role has proved to be contentious.
One critique questions whether personal preferences
should be given this status in assessing welfare.
Some observers have questioned whether the choices are
morally sound.
For example, it is sometimes argued that the decision to
buy some luxury good is not ethically defensible when
people (including children) are dying from poverty-related
causes in the world.
Utility based approach
Another objection is that “utility” is not something we
observe.
That is true, but we can still use the idea to motivate
welfare comparisons in more familiar monetary terms—in
the “income space.”
However, we must then be confident that our measure
based on the observables is calibrated to be consistent with
our concept of utility.
When it is a monetary measure and it is consistent with
utility it is called a money-metric of utility (or
sometimes equivalent income).
This can be readily defined in theory, by finding the income
equivalent to utility at fixed reference prices and personal
characteristics.
Utility based approach
Arguably the bigger problem is heterogeneity across people in
their welfare-relevant non-market characteristics.
People differ in the utility they can be expected to derive from a
given consumption bundle.
Some people have characteristics—being elderly, or disabled, or
living in a cold climate, or living in a place where public services
are poor—whereby they need more of certain market goods to
attain the same level of utility.
We may well be able to find a utility function consistent with that
behavior, but it will not be unique; there can be many such
functions, varying with personal characteristics.
So the idea that one can infer utilities of heterogeneous
individuals from looking solely at their demand and supply
behavior is easily ruled out.
The Challenge of Inferring Utility from
Behavior in Markets
A fundamental premise of economics is that the observed
commodity demands, and labor supplies of households are
utility maximizing.
The basic model assumes a utility function that depends on the
quantities consumed of all goods and services and the leisure time
left after working.
Recall that this function represents consumer preferences in that
the utility function ranks commodity bundles identically with how
the consumer ranks them.
This utility function is then maximized subject to the budget
constraint, which says that the total expenditure on commodities
plus the imputed value of leisure time (time taken for leisure
times the market wage rate) cannot exceed “full income” given by
the value of the time endowment plus all other income (including
profits from own enterprises).
The Challenge of Inferring Utility from
Behavior in Markets
If demands and supplies are consistent with this model,
then we can in general solve backwards from the observed
demands and supplies of a given individual to recover a utility
function that is consistent with the choices made.
When we come to compare people in different households,
with different sizes and demographic compositions, and
differences in other characteristics (such as health and
disability), we must allow the possibility that these differences
matter to both the observed demands and to the level of utility
attained given those demands.
However, we cannot expect that the ways those differences
influence utility at given demands (and supplies) will be properly
reflected in the observed demand and supply behavior.
The Challenge of Inferring Utility from
Behavior in Markets
In general there will be multitude utility functions
(reflecting the heterogeneous characteristics) that can
support the observed behavior as an optimum.
Thus, we say that the utility function is
“unidentified” From observed demands and
supplies alone when people differ in their welfare-
relevant characteristics.
The Challenge of Inferring Utility from
Behavior in Markets
Given that heterogeneity in how command over
commodities translates into welfare, we will inevitably
need to broaden the information base for assessing
welfare, beyond observed behavior in markets.
This calls for information relevant to people’s welfare
that economists have not traditionally favored, such as
data on capabilities and also subjective well-being
data.
Thank You
