The focus of today’s lecture will be on redistribution as discussed in Chapter 3(Mueller, 2003). Additionally, we will discuss papers quantitatively assessing the situation (De Haan & Sturm , 2017; Sturm & de Haan, 2015).

A justification for the state can be redistribution. But redistribution itself can be argued for based on different reasons. In this post, we will illuminate the main arguments. First three voluntary redistribution arguments will be covered, then we will have a look at involuntarily redistribution.

*Redistribution as insurance*

If one assumes Rawls’ veil of ignorance (Rawls, 2009), redistribution can be seen as an insurance against the uncertainties of what kind of role one will assume in society. Insurance can be covered privately, so at first state intervention may seem inadequate. However, since people can assess their risk, high-risk individuals would select the insurance whereas low-risk individuals would shun the insurance. To overcome the issue of adverse selection, public insurance is introduced. The issue of adverse selection has been introduced by Akerlof (Akerlof, 1970) and shows that information asymmetry can break markets. The public insurance overcomes this issue by forcing a pareto-optimum on a societal level. Typical cases for this are health care insurance, unemployment insurance, and retirement insurance.

*Redistribution as public good*

Another justification comes from altruism or empathy (“warm glow”). The utility equation is expanded to [latex]max U_m + \alphaU_o[/latex] where [latex]0\leq\alpha\leq1[/latex].

*Redistribution as fairness norm*

The assumption that fairness is an important norm, is the basis for this redistribution argument. The classical example is the *dictator game*, where anonymous individuals are paired and one gets an amount of money and may share it with the other. Usually, any individual share around 30% with the other despite being able to keep everything and not knowing anything about the other. So far, the assumption is that the random element of the game let people share their gain because they also could have ended up on the other side.

*Redistribution as allocative efficiency*

If two individuals ([latex]P[/latex] and [latex]U[/latex]) work a fixed amount of land. The productivity of [latex]P[/latex] is 100 whereas [latex]U[/latex]’ productivity is 50. The connecting curve describes the production possibility frontier. Any initial allocation (e.g. [latex]A[/latex] may not be optimal on a societal level (i.e. [latex]A[/latex] is not tangential on a [latex]45°[/latex] line), the societal optimum would be in [latex]B[/latex], which is however unacceptable for [latex]U[/latex]. The inefficient allocation would end up at [latex]A'[/latex]. The state could either redistribute land to reach [latex]B[/latex] or production to reach [latex]C[/latex]. Note that C in the graph should amount to a value above 100. Alternatively, private contracting could reach the same result given that the state enforces property rights and contracts.

The example is based on (Bös & Kolmar, 2003).

*Redistribution as taking*

Groups can lobby to increase their utility [latex]U[/latex] by increasing their income [latex]Y[/latex] based on their political resources [latex]R[/latex] available. However, if two antagonistic groups lobby their policies may cancel each other leaving them only with the additional cost of lobbying without any gains.

**Measuring redistribution**

To measure redistribution, inequality needs to be measured first. A typical measure of inequality is done via the Lorenz curves and the Gini coefficient (Gini, 1912). The Gini coeffcient is the ratio of areas under two curves. The Gini market coefficient (before taxes) and the Gini net coefficient (after taxes and subsidies) are subdivisions that taken at ratio help to assess redistribution.

The causation of inequality is difficult to assess. Some argue for politics (Stiglitz, 2014), whereas others argue for the market-based economies (Muller, 2013). A new line of inquiry attributes inequality to ethno-linguistic fractionalisation reducing the interest in redistribution (Desmet, Ortuño-Ortín, & Wacziarg, 2012).

Sturm and de Haan (Sturm & de Haan, 2015) follow up on the argument and examine the relationship between capitalism and income inequality. A large sample of countries is analysed using an adjusted economic freedom (EF) index as proxy for capitalism and Gini coefficients as proxy for income inequality. Additionally, they analyse the relation between income inequality and fractionalistion given similar capitalist systems. For the first analysis, there is no conclusive evidence that capitalism and income inequality are linked. However, if fractionalisation is taken into account, than inequality can be explained based on the level of fractionalisation. The more fractionalised a society is, the less redistribution takes place and consequently inequality remains high.

In a second paper de Haan and Sturm (De Haan & Sturm , 2017) analyse how the financial development impacts income inequality. Previous research on financial development, financial liberalisation and banking crises (theoretical and empirical) has been ambiguous. TBC.

## References

*The Quarterly Journal of Economics*, 488–500.

*Journal of Public Economics*,

*87*(11), 2431–2457.

*European Journal of Political Economy, Forthcoming*.

*Journal of Development Economics*,

*97*(2), 322–338.

*Memorie di metodologica statistica*(p. 1). Rome: Libreria Eredi Virgilio Veschi.

*Public Choice III*. Cambridge, UK: Cambridge University Press.

*Foreign Affairs*,

*92*(2), 30–51.

*A theory of justice*. Harvard university press.

*New York Times*, pp. 1–2.

*American Economic Review: Papers and Proceedings*,

*105*(5), 593–597.