Notes on Alesina and Angeletos on Fairness and Redistribution

1
Notes on Alesina and Angeletos on Fairness and
Redistribution

• Econ 594ER
• October 29, 2007

2
Social spending and beliefs
Americans
Europeans
3
How are multiple equilibria sustained?

• Can we explain differences in beliefs w/o relying
  on differences in economic fundamentals or on
  persistent distortions in beliefs?
• Equilibrium 1
• taxes are higher, individuals invest and work
  less, and inequality is lower
• a large share of total income is due to luck
  which is why income redistribution is desirable
  and this sustains the equilibrium
• Equilibrium 2
• taxes are lower, individuals invest and work
  more, and inequality is higher
• a larger fraction of income is due to effort
  rather than luck and this sustains the
  equilibrium
• Authors do not try to explain how these
  equilibria are reached (probably through cultural
  evolution in belief), but argue that different
  beliefs are sustainable
• Previous literature
• Rawls (1971) and Mirrlees (1971) model fairness
  as demand for insurance
• Piketty (1995) agents form beliefs only based on
  their experience
• Benabou and Tirole (2005) multiple beliefs arise
  because agents bias their own perceptions to
  offset procrastination (presumably in
  investment?)

4
Experimental evidence of luck fairness

• Fehr and Schmidt (2003)
• Dictator games people give away some of
  endowment even though they could keep it all
• Ultimatum games players suffer monetary loss to
  punish unfair behavior
• Gift exchange games players reward fair
  actions
• Public good games players punish free riders
• Key Results are sensitive to how endowments are
  allocated
• Hoffman and Spitzer (1985) and Hoffman et. al.
  (1985) show players more likely to make unequal
  offers in ultimatum games when endowments are
  earned
• Clark (1998) finds that player in public goods
  games vote for less redistribution when
  endowments depend on performance rather than
  being randomly assigned

5
Framework of the model

• yi depends on talent, investment, effort, and
  random noise\luck (?i)
• Government imposes flat tax rate to generate
  funds for a lump sum redistribution transfer
• Individuals get common disutility from unfair
  social outcomes (O)
• O is basically the aggregate difference in
  utility from the income one gets (yi) and what
  one deserves (yi ), where yi yi ?i
• Definition An equilibrium is a tax rate and
  collection of investment and effort choices that
  maximize the utility of the median agent

6
Key assumptions and results

• Fairness and the signal-to-noise ratio
• RHS is signal to noise ratio in pre-tax income
  numerator is earned denominator is luck
• Optimal tax rate decreasing in signal to noise
  ratio
• E.g., if tax rate 1,100 of income is from luck
• Signal-to-noise ratio is endogeneous
• Heterogeneity in talent or willingness-to-work
  increases signal
• Luck increases noise
• Signal-to-noise ratio decreases in the tax rate,
  reflecting distortionary effects of taxation
• Optimal policy and multiple equilibria
• Demand for fairness leads to higher taxes and
  multiple equilibria
• The high-tax equilibrium is sustained by a
  self-fulfilling low signal-to-noise ratio

7
Comments

• Critical assumption is that, with less
  governmental intervention, luck has less of an
  influence on outcomes.
• Q Is the logic correct? Dont lucky results
  increase with effort and investment?
• What happens when we allow for intergenerational
  wealth transfers?
• Signal-to-noise ratio depends on policies chosen
  by past generations
• Authors say that a society that has history of
  high distortions will tend to have inherited
  wealth, which makes it more likely to have
  aggressive redistribution in present
• Q Is the logic correct? Is this consistent with
  empirical evidence?