Redistribution Efficiency Fairness 1 Consider a Possibility Frontier

Redistribution, Efficiency, Fairness 1

• • • Consider a Possibility Frontier Most government action we have thought about is getting you from inside the frontier to the frontier, or at least towards it. This is about efficiency. E. g. , if there is an externality, you can make everyone happier by taxing the negative externality and giving away the proceeds appropriately. If there is no insurance market because insurers are plagued by information problems, you can make everyone better off by forcing them to buy an insurance package. But, governments also move you along the frontier. This is about equity. For example, governments take money from people who have money and give it to people who don� ’t have money. This is redistribution. Sometimes, the frontier responds to redistribution. For example, if there is no redistribution, then the frontier on money in a pie‐sharing game is linear (with a slope of � ‐ 1). Either I have the pie or you have it. If there is redistribution, then we might lose some pie as we move it from you to me (or vice versa). Thus, the frontier may be pulled inwards as we try to move more pie from one person to the other. Now, think about a world with production. If I know that my earnings will be taken away, I may work less hard and produce less. This is like the pie� ‐sharing problem, because moving money around is costly (in terms of total production), just like moving pie around is costly (in terms of lost pie). 2

• Why is there redistribution? • Positive explanations: • People are altruistic. That is, their well� ‐being depends in part on the well‐being of others. After all, people donate to charities. • Thus, they vote to redistribute. • But, why do they redistribute through government, rather than by giving to charities? • There is an positive externality on giving to charities: if everyone is altruistic, then if one person donates to the food bank and improves the lives of those without food, then it makes all those hungry ones better off, which in turn implies that everyone gets a little better off (because of the altruism). • Thus, governments should subsidize charitable giving. • In addition, government may have an advantage in redistributing: they can more easily get information on who is poor than can small private agencies. • E. g. , government has Canada Revenue Agency. 3

• Normative explanations: • Redistribution is good. • But, what defines goodness? All such arguments are necessarily ethical arguments. • Veil of Ignorance (Rawls). If you didn’t know who you were going to be, what kind of society would you want to live in? – This is like an insurance, but it is � insurance at birth�. • (Some) Ideas for Redistribution and/or Fairness • Could try to get the total maximized (like Marshall efficiency suggests) – Someone extremely greedy => give much to that guy • Could try to maximize the well‐being of the worst off member – Implies even distribution 4

• But distribution of what? – Money? – Goods? – Utility? • Everybody gets same piece of a (every) pie • If people want different things, this will be Pareto inefficient • Consider a (voluntary) trade • But this gives an idea: – Everybody gets everything same – Then people are allowed to trade – The result must be a � no envy� equilibrium 5

• The Second Fundamental Theorem of Welfare Economics: • Any feasible efficient allocation can be achieved starting from appropriate endowments through free markets, provided – – Perfect Competition No Externalities No Public Goods No Information Problems • What it says is that under certain conditions, the distribution and efficiency can be treated as completely separate tasks. • � No envy� equilibrium could be achieved this way • But endowments may be very difficult to split equally • No� Envy + Efficient is what is usually called � FAIR allocation in Welfare Economics 6

• Progressive Taxation • Commonly used way to redistribute from rich to poor • One way to define progressive taxation is o Average tax rate increases as income increases • Another common way o Net tax (= tax. transfer) increases as income increases o Think increasing tax rate with equal consumption of public goods o Or think flat tax rate with greater consumption of public goods by poor • Redistribution means taxing • Taxes create deadweight loss • Deadweight loss from taxes = cost of redistribution 7

• Poverty may be defined as a lack of material resources, the state of not having enough. This is hard to operationalise, but if you want to think seriously about the distribution of economic outcomes, it is the very first thing you have to figure out. • The Low‐Income Cut. Offs are a set of income thresholds defined by Statistics Canada to capture the idea of ‘not enough money’. • If your household has less money than the LICO for your household type in your city size, then you are called a ‘low‐income household’. • The low‐income, or poverty, rate is equal to P/N where P is the number of people living in low‐income households and N is the size of the population. • Statistics Canada always says ‘these numbers are not poverty counts; they are low‐income counts’, but the press always reports them as poverty counts. After all, what is the difference? • Income is not wealth • But it is highly correlated (wealth is the sum of incomes) 8

• The low� ]income cutoffs are computed via the following algorithm: o Define necessities as food, clothing, and shelter. o From the Family Expenditure Surveys, or the Surveys of Household Spending, estimate the average share of all expenditure commanded by necessities. o In 1992, this share was 43. 6%. o Declare that spending 20 percentage points more than this on necessities is � ”a lot more�. ” o Figure the income at which the necessities share exceeds 63. 6%. This is the low� ‐income cutoff. o Go through this algorithm for each household size and city size. o LICO discussion o LICO tables 9

Inequality Economic growth may be very unequal across people or groups of people Should equality of outcome (income) or equal opportunities matter? People may be better off sacrificing some income for nonmeasurable things (leisure, status, etc) • Most common measure is Gini coefficient • • o G = 0 for complete equality o G = 1 for complete inequality • Economic growth usually increases equality • There’s more stuff, so it is easier to convince the winners to support the losers • But it also may be that redistribution slows down economic growth • Weaker incentives for poor to work • Weaker incentives for rich to work • Extreme poverty (little redistribution) is likely to slow economic growth • Poor cannot invest much in human capital for themselves but also for their kids 10