Presentation on theme: "Global distributive justice Milanovic, “Global inequality and its implications” Lectures 10-12."— Presentation transcript:
Global distributive justice Milanovic, “Global inequality and its implications” Lectures 10-12
Rodrik’s trilema Economic integrationNational sovereignty Welfare state If sovereignty and welfare state (like now), no integration If integration and sovereignty, reduce spending and cut welfare state If welfare state and integration, global organizations to decide on policies (thus no national sovereignty)
Rodrik’s view The most important thing is not trade but greater policy autonomy Growth = fct (govt quality; appropriateness of policies) If policies (=Washington Consensus) are not appropriate for the existing institutions, bad outcomes (example of transition countries) Finding appropriate policies is key=>that’s why policy autonomy is essential
Rodrik’s recommendations More labor mobility Tobin tax International agreement on ending subsidizing DFI and repelling “odious debt” Soften intellectual property rights protection (to enhgance tewchnology transfer) Reform IFIs (focus on “knowledge”)
Rawls’ Law of Peoples Types of peoples (nations) –Liberal} –Decent (consultative hierarchy)} –“Burdened” –Outlaw states –Benevolent absolutism Transfers only from well-ordered to “burdened” peoples Well- ordered
Transfers (1) limited to type of society (‘burdened’) and (2) limited in time (until it becomes a ‘decent society’) “Peoples have a duty to assist other peoples living under unfavorable conditions that prevent their having a just or decent political and social regime” (LoP, p. 37) Explicit rejection of a global difference principle (among other reasons because it is unlimited in time) No discussion of responsibility toward outlaw or hierarchical societies Limits to immigration
Principles of justice (national level) 1) Each person to have equal right to most extensive liberty compatible with a similar scheme for others 2) Social and economic inequalities to be arranged so that (a) they are expected to be to everyone’s advantage and (b) attached to offices open to all. Difference princoiple applies not only to income & wealth but to positions of authority Lexicographic ordering: first principle comes before second Definition of injustice: inequality that is not to the benefit of all TJ, p.53-55
Lexicographical ordering of principles “Imagine…that people seem willing to forego certain political rights when the economic returns are significant. It is this kind of exchange which the two principles rule out; being arranged in serial order they do not permit exchanges between basic liberties and economic and social gains…”(TJ, p. 55).
Interpretation of the difference principle Equally open To everyone’s advantage Careers open to talents 1. System of natural liberty (accepts inheritance) Natural aristocracy (greater natural ability) (noblesse oblige) Equality of fair opportunity 2. Liberal equality corrects for social inequality implicit in 1 (fair equality of opportunity: limits inheritance + free education) 3. Democratic equality Corrects for inequality of talent and family implicit in 2 In all cases (except natural aristocracy) everyone has formally the same rightsa (so the first principle ofr justice is satisfied)
Why no global difference principle It would lead to open-ended transfers Real income per capita (wealth) is not important once societies become ‘decent’ (general proposition re. unimportance of pursuit of wealth) Once a people is ‘decent’ there is no point in comparing wealth/income of the two peoples: the differences are the outcome of voluntary societal decisions on savings vs. consumption and leisure vs. work
Legitimacy: Why Rawls is not a cosmopolitan (Wenar) Peoples are different from individuals: legitimacy is the building block on which a pact between peoples is created Different peoples’ legitimate governments are grounded in different political cultures Because cultures are different, the exact shape of legitimacy in different societies will be different Peoples can cooperate only if they view each other as legitimate
Legitimacy (cont.) Since the pact is made between peoples (not individuals), there cannot be global difference principle The bottom line: (1) difference in political cultures leads to differences in the ways legitimacy is defined; (2) people to people relations are based on legitimacy; (3) individuals are not involved in this ‘pact’; (4) there cannot be global difference principle
Cosmopolitan position (Pogge, Singer) No major difference between Rawlsian original position within a single nation-state (people) and the world The same principles should apply globally: an increase in inequality is acceptable only if it leads to a higher absolute income of the poorest “Monism”: all ethically meaningful relationships are between individuals not mediated by the state (people) Pogge: we are required not to harm others (and some decisions by IO may have harmful consequences)
Rejection of cosmopolitanism: political theory of justice (Nagel) Strong statism: Redistribution (and responsibility for poverty) possible only if there is shared government For concerns of justice to kick in, you need “associative relation” (shared sovereignty, common endeavor) We redistribute because we have a contractarian relationship with people with whom we share the same institutions Could be also based on our expectation to be in need of similar transfers in the future; or affinity that we feel for co-citizens; shared culture or historical memories (J.S. Mill)
Statism (cont.) Only under world government can we have a global difference principle Accepts humanitarian duties only Existence of IO does not introduce new obligations because these are govt-to-govt relations (similar to Wenar’s point) Pluralism (rather than monism) in our relations with others: different normative priciples depending on the position in which we stand with respect to them; but pluralism may introduce a sliding scale & an intermediate position =>
Intermediate position: meaningfully consequential relationships are sufficient (Beitz, Cohen & Sabel) Responsibility stems from having consequential relationships with others not only from sharing the same polity (government) This happens not only directly through trade and communications, but through the role of IO like World Bank and IMF Sliding scale of inter-relationships: from within the same people, to “proximate” peoples to peoples with whom there are few relationships (“density of the relationship”)
Intermediate position (cont.) We are required to give more than implied by humanitarian considerations alone but less than inmplied by the global difdference principle Sliding scale of responsibility Critique of statism: why are newer forms of international governance not norm-generative and only state is? There are forms of connection that do not involve the state & trigger norms beyond mere humanitarianism Direct rule-making relationship between the global bodies and citizens of different states
Intermediate position (cont.) Aristotle: within each community there is philia (affection; goodwill) but the philia spreads (diminishes) as in concentric circles as we move further from a very narrow community To each philia corresponds adequate reciprocity (that is, redistribution) Thus the sliding scale of philia and reciprocity
What is a “consequential relationship”? Obviously, a political relationship is consequential (Nagel) Economic relationships reflected in trade, investment of capital etc (Julian: “economistic” definition of consequential relationship) Beitz: (1) interrelationship must reach a certain threshold, (2) there are global non-voluntary institutions in which different peoples belong ▬► institutional conditions under which considerations of global justice kick in
Decisions made by international organizations (even if only states are signatories) and by global networks => imply inclusion of all and duty of wider assistance (Cohen & Sobel) Institutional explanation applies not only to glaobl institutions but to “institutional clubs” like Commonwealth, Europrean Unhion, Communaute Francaise etc. Sliding scale of responsibility (within institutional explanation)
Discussion Do we do nothing until global government comes? Economistic requirement for global justic is easier to measure But “density “ of economic relations is greater among rich countries. Should then justice conditions start selectively (club-like) among the rich countries first? Institutional requirement brings in global justice considerations already now It would embrace even peoples whose density of relations is small (say, United States and Mauritania)
Among whom does duty of assistance exist? PoliticalRawlsEconomisticInstitutionalCosmopolitan Among people who share a polity Political + burdended societies Among people who have dense economic relations Among people who share global governance institutions Among all people in the world
Discussion (cont.) This is why we need some rules re. global redistribution Go back to the three rules: Progressivity 1; global progressivity and reduced inequality in both donor and beneficiary country
Rawls on Concept 1 and Concept 3 inequality Neither of them matters Concept 1 (divergence) is irrelevant if countries have liberal institutions; it may be relevant for liberal vs. burdened societies Irrelevance rooted in two key assumptions: (i) political institutions of liberalism are what matters; (ii) acquisition of wealth immaterial Concept 3 is similarly irrelevant once the background conditions of justice exist in all societies But Concept 0 (within-national) inequality matters because the difference principle applies within each people
“once we accept the value of collective self-government, there is no reason to hope for convergence in living standards— the absence oif convergence is not a defect awaiting correction” (Joshua Cohen)
In Gini terms: Go back to our definition of global inequality Rawls would insist of the minimization of each individual Gini (Gi) so that Term 1 (within-inequality) would be minimized. But differences in mean incomes between the countries can take any value. Term 2 (between inequality) could be very high. And this is exactly what we observe in real life: Term 1 Term 2
Rawls’ global “original position” Assume Rawls’-like veil of ignorance for all citizens of the world where citizenship and social class are “allocated” to each individual 60% of one’s income position in the world will be determined by one’s location Major difference from the situation two centuries ago (Marx would have been surprised)
Explaining person’s income position in the world Circumstan ce Ln(GDI per capita) of his/her country +22.2*** Gini of his/her country -0.34***-0.33*** Circumstan ce + effort His/her social class within country +2.78** R2R2 0.610.91
Citizenship premium. If mean income of country where you live increases by 10%, your position in the world goes up by 2.2 percentiles Trade-off. If through effort and luck you jump ahead 5 social classes (e.g. in the US, going from the median household per capita income of $14,000 to $22,000) this is equivalent to a citizenship premium of about 60% (e.g. being born in Mexico rather than in China * ) * China is at the median (unweighted) world income
Composition of global inequality changed: from being mostly due to “class” (within-national), today it is mostly due to “location” (where people live; between-national) 1870 2000 Source: Bourguignon and Morrisson (2002) and Milanovic (2005)
Global Redistribution of Income (Bourguignon, Levin & Rosenblatt): The context Much of global inequality is determined by international inequality rather than within country. How much redistribution takes place via international flows and implicitly through international policies? Data availability limits accounting exercise: models required (e.g., trade). Based on slides provided by D. Rosenblatt.
Forms of international redistribution Aid Remittances (not covered here). –Profit –Worker Implicit redistribution due to policy restrictions: e.g. trade.
Redistribution through aid OECD/DAC Database. Track donor to recipient flows. Accounting of share of donor and recipient incomes. Construct counterfactual of what international distribution would be without aid (in pure accounting, not GE sense). Measure impact on international distribution of income.
Some data/measurement complications OECD-DAC data Treatment of grant element of debt flows. –Tried various measures. Treatment of debt relief (included) PPP values versus dollars. -Former implies non-zero sum redistribution. -Tried various approaches.
Redistribution through aid results Case (all income inequality measures population – weighted---Concept 2) GiniTheil Entropy Mean Log Dev. Atkinson e=0.5e=2e=5 Base (after aid)0.53800.53050.53160.23780.61870.8291 “Maximum Scenario” (aid netted out) -0.8 grant equivalent of imputed multilateral -All debt relief included -All technical cooperation included 0.54240.53910.55070.24270.64680.8805 “Preferred Scenario” (aid netted out) -0.5 grant equivalent of imputed multilateral -96 percent of debt relief deducted -All technical cooperation deducted 0.54040.53510.54150.24040.63190.8498 Maximum distributional gain from aid (in Gini points) 0.44
“Redistribution” through trade restrictions Unavoidably need a model. Borrowed results from van der Mensbrugghe’s simulations using World Bank trade model. Measure potential lost income from high income country protection. Simulation of counterfactual from model: what would 27 country groupings’ incomes be in the absence of this protection?
“Redistribution” via trade restrictions: results Case (all inequality measures are population weighted—Concept 2)Gini Theil entrpy Mean log deviationAtkinson e=0.5e=2e=5 Base (with existing protection of merchandise trade by high-income countries) 0.54260.54080.53960.24180.61900.8266 With high-income countries merchandise trade reform 0.54250.54020.53940.24160.61890.8263 Distributional gain from trade reform (in Gini points) 0.01
“Redistribution” via trade protection: impact by decile Largest gains Average gain