Income Redistribution:Conceptual Issues

1
Income RedistributionConceptual Issues

• Tonight, well consider a framework for thinking
  about the normative and positive aspects of
  government income redistribution policy.

2
Introduction

• A Question to Begin Should economists be
  concerned with distributional issues at all?
• Some say NO.
• Value judgments embodied in the right income
  distribution.
• No scientific basis for the right distribution.
• Economists should focus exclusively on efficiency
  and leave equity considerations alone.

3
Introduction

• A Question to Begin Should economists be
  concerned with distributional issues at all?
• Others say YES.
• A Focus on efficiency alone also has problems.
• Efficiency by itself is an inadequate normative
  standard. To focus on it exclusively, is itself
  a value judgment.
• Policymakers certainly care about income
  distribution. If economists ignore distributional
  issues, policy makers will ignore economists and
  focus exclusively on distribution and ignore
  efficiency.

4
Introduction

• Related Question Should government try to
  change the existing income distribution?
• Arguments for NO. Reasons similar to why
  economists should not consider distribution in
  first place.
• Arguments for YES. Almost anything the
  government does will affect income distribution.
• Example Tax the public to build public works.
• Losers Taxpayers who do not (on net) benefit
  from projects.
• Winner Taxpayers who do and contractors who
  receive lucrative contracts.

5
Distribution of Income

• Can analyze household income, and see how equally
  or unequally the pie is distributed.
• Table 7.1 shows the percentage of money income
  among households for more than 30 years.

6
Table 7.1
7
Distribution of Income

• Richest 20 receives about 50 of total income.
• Poorest 20 receives about 4 of total income.
• Inequality has increased over time.

8
Distribution of IncomePoverty

• The poverty line is a fixed level of real income
  which is considered enough to provide a minimally
  adequate standard of living.
• Inherently arbitrary, but still a useful
  benchmark when making comparisons.
• Trends over time
• Differences across groups

9
Distribution of IncomePoverty

• Poverty line for a family of four was 19,157 in
  2004.
• Median household income more than double that,
  44,389 in 2004.
• Table 7.2 shows poverty rates for selected groups
  in 2001.

10
Table 7.2
11
Distribution of IncomePoverty

• Poverty rates in U.S. in 2004 might be considered
  surprisingly high 12.7 for population as
  whole.
• Concentrated among certain groups, such as female
  headed households, children, and minorities.
• Elderly have lower poverty rates than the U.S.
  average.

12
Distribution of IncomePoverty

• Can also look at trends over time.
• See Table 7.3.
• Poverty considerably lower than in 1960s, but not
  much progress since 1970.

13
Table 7.3
14
Interpretation Problems

• Poverty line (and poverty rate) is subject to a
  number of criticisms.
• When interpreting the numbers, it is useful to
  know the conventions and limitations.

15
Interpretation Problem 1

• Income consists only of cash receipts.
• Excludes in-kind transfers like health insurance,
  food stamps, and housing.
• Excludes non-market work such as childcare or
  housework.
• Ignores income flow from durable goods.

16
Interpretation Problem 2

• Income is before tax.
• It ignores cash refunds from the Earned Income
  Tax Credit, which has grown dramatically in the
  last decade, and now amounts to more than 31
  billion annually.
• Ignoring this overstates poverty rates and also
  affects the trends over time.

17
Interpretation Problem 3

• Income is measured annually.
• Not obvious what the correct time frame should
  be.
• Income does fluctuate from year to year.
• Lifetime income considerations seem relevant.
• Consider a starving college student, for
  example. Not really poor in a lifetime sense.

18
Interpretation Problem 4

• Unit of observation
• Person, family, household?
• People often make decisions as an economic unit,
  and there are economies of scale in household
  production.
• Classifications can matter for poverty numbers
• Bauman (1997) calculates that including the
  income of nonfamily members (such as nonmarried
  cohabitors) would reclassify 55 of people who
  are poor out of official definition.
• His methodology controversial and strongly
  criticized by other researchers

19
Interpretation Problem 5

• A Nutritional Standard is totally unrelated to
  actual expenses required to not be poor.
• In 1950s, average family spent 1/3 on income on
  food.
• Today, average family spends less than 1/5.
• No reason to believe 2/3 (or 4/5) of remaining
  income is adequate to pay for rent, clothing,
  heathcare, child-related expenses,
  transportation, incidentials.
• In fact, empirical evidence suggests 2/3 is not
  adequate.

20
Interpretation Problem 6

• No attempt is made to control for regional
  variations in costs-of-living
• It will clearly require a higher income to not be
  poor in California or New York than it would to
  not be poor in Iowa or Mississippi because
  housing and other costs are so much higher.
• The official poverty measure creates a
  one-size-fits-all poverty line.

21
An Alternative to the Official Poverty Measure

• In 1995, a panel from the National Academy of
  Sciences proposed a whole new poverty measure.
• Their alternative measure addresses all the
  interpretation problems cited above.
• Census Bureau has since refined this alternative.
  Still not adopted officially.
• Alternative measure shows slightly higher poverty
  overall than official measure.
• Alternative measure shows increases in poverty
  for some groups and regions, decreases in poverty
  for other groups and regions.

22
Rationales for Income Redistribution

• Different kinds of social welfare functions
• Utilitarian
• Maximin criterion (Rawlsian)
• Pareto efficient
• Non-individualistic

23
Simple Utilitarianism

• The utilitarian social welfare function is
• W F(U1, U2, , Un)

• Which depends on all n members of society. One
  specific function form is
• W U1 U2 Un
• This special case is referred to as an additive
  social welfare function.

24
Simple Utilitarianism

• With the additive SWF that was given, also
  assume
• Identical utility functions that depend only on
  income.
• Diminishing marginal utility of income.
• Societys total income is fixed.
• Implication government should redistribute to
  obtain complete equality.

25
Simple Utilitarianism

• This can be illustrated with two people.
• See Figure 7.1.
• Any income level other than I does not maximize
  the SWF.
• I entails equal incomes.

26
Figure 7.1
27
Simple Utilitarianism

• So, should Full Income Equality be pursued?
• Assumptions are too limiting to make such a
  conclusion reliable.
• Assumes identical utilities.
• Assumes decreasing marginal utility.
• Assumes total income fixed
• e.g., no disincentives from this kind of
  redistributive policy.

28
The Maximin Criterion

• The Rawlsian social welfare function is
• W Minimum(U1, U2, , Un)
• Social welfare in this case depends only on the
  utility of the person who has the lowest utility.
• Rawls (1971) asserts it has ethical validity
  because of the notion of original position.
• Notion that ex-ante individuals do not know where
  in the income distribution they will be.

29
The Maximin Criterion

• These ethical claims are controversial
• Still selfish view in original position
• Individuals extremely risk averse here
• All that is relevant is the welfare of the
  worst-off person, even if a policy is extremely
  detrimental to everyone else.

30
Pareto Efficient Income Redistribution

• Under both Utilitarian and Maximin social welfare
  functions, by making some better off,
  redistribution made others worse off.
• Given these approaches, redistribution would
  never lead to a Pareto improvement.
• Each individuals utility depends only upon their
  own income.

31
Pareto Efficient Income Redistribution

• Suppose that an individuals utility depends not
  only upon their own income, but the income of
  others.
• For example, suppose Peter is rich, and Paul is
  poor but that Peters utility function takes the
  following form
• UPETER U(IPETER , U(IPAUL))
• Now it would be efficient to transfer income from
  Peter to Paul so long as Peters utility gain
  from improving Pauls utility is greater than
  Peters utility loss from lower consumption.

32
Pareto Efficient Income Redistribution

• Government redistribution in this case might
  improve efficiency. It may be difficult for
  private actions to do this, if, for example, the
  rich lack information on just who really is poor,
  or if there are free rider aspects.
• Formally, this amounts to an externality problem
  Pauls consumption (or lack thereof when he is
  poor) affects Peters welfare in a way that is
  external to the market.

33
Pareto Efficient Income Redistribution

• This argument assumes people would not give in
  absence of governmental coercion. But there is,
  in fact, a great deal of private charitable
  giving.
• Additional Self-Interest Arguments
• Social insurance aspects reflect self-interest,
  not altruism. Pay in when you are rich to protect
  yourself in case you need a pay-out if you become
  poor.
• Social peace aspects also reflect self interest.
  Pay in when you are rich to protect yourself by
  preventing poor from rioting and staging
  revolution.

34
Nonindividualistic views

• In previous cases, social welfare derived from
  individuals utilities.
• Some theorists specify a social welfare function
  independent of individual preferences or
  utilities.
• Plato the richest person should have no more
  than 4 times the income of the poorest.
• A less extreme example commodity egalitarianism.
  
• Right to vote, food, shelter, education, perhaps
  health insurance.

35
Processes versus Outcomes

• Some argue that a just distribution of income is
  defined in terms of the process that generated
  it.
• For example, equal opportunity in U.S.
• Ensuing outcome would be considered fair,
  regardless of the income distribution it happened
  to entail.
• Gottschalk finds a substantial amount of income
  mobility.
• Does raise problem of how to evaluate social
  processes.

36
Expenditure Incidence

• Relative Price Effects
• Public Goods
• Valuing In-Kind Transfers

37
Relative Price Effects

• Example suppose government subsidizes housing
  for the poor.
• As a first pass, redistribution strictly from
  rich to poor.
• But what if subsidy increases the poors demand
  for housing and pushes up housing prices?
• Landlords may reap part of the gain.
• Contractors and construction workers may also
  gain.
• Generally, any government program sets off a
  complicated chain of price changes, and the
  complete effects on winners and losers is unclear.

38
Public Goods

• Do rich and poor benefit similarly from the
  provision of public goods?
• Example with national defense, almost 400
  billion spent annually. Do rich and poor receive
  same level of benefit?
• These kinds of questions are difficult to answer,
  because they are sensitive to assumptions made
  concerning the utility functions of rich and poor.

39
Valuing In-kind Transfers

• Government provides many benefits to the poor
  in-kind that is, direct provision of goods
  rather than cash.
• Food stamps
• Medicaid
• Public Housing
• Estimating value is difficult. Not always valued
  at dollar-for-dollar (if resale is difficult).

40
Valuing In-kind Transfers

• Consider how the provision of an in-kind benefit
  changes the budget constraint in Figure 7.2.
• In this case, giving an in-kind benefit lowers
  utility relative to an equally costly cash
  transfer.
• Although the person is better off by having the
  in-kind transfer than not having it, she would be
  even happier with the cash transfer.

41
Valuing In-kind Transfers

• Assumptions
• Poor person has 300 in cash income.
• Price of cheese 2.00 per pound.
• Price of other goods normalized to 1.00.

42
Figure 7.2
43
Valuing In-kind Transfers

• A person can never be made better off with an
  in-kind transfer that is equal in cost to a cash
  transfer.
• There are instances, however, when a person is
  indifferent between the two transfer schemes.
• See Figure 7.3.

44
Figure 7.3
45
Valuing In-kind Transfers

• Why give in-kind transfers if they tend to be
  inefficient?
• Iggy Pop Syndrome Public concerned that poor
  would spend cash transfers on Liquor and Drugs.
• Commodity egalitarianism.
• May reduce welfare fraud (especially if the
  in-kind transfer is an inferior good). Non-poor
  might cheat to get cash, but not food stamps or
  public housing.
• Politically viable because they help the producer
  of the in-kind good. Food stamps supported by
  Agricultural

46
Recap of Income Redistribution Conceptual Issues

• Distribution of income
• Poverty line
• Social welfare functions
• Valuing In-kind transfers