Most existing models of political economy tend to be structured around the division of society into three distinct groupspoor, middle and rich. The median voter theorem which provides a coherent way to think about voter preferences for redistribution postulates that with neither of the groups commanding a majority, government formation and thus policy formulation is inherently dependent on a coalition of two of the three groups. In such a scenario, the middle income or median voter, by aligning with either the poor or the rich, ends up playing a crucial role in determining the direction of public policy. Thus, with the median voter casting the deciding ballot and playing a pivotal role in government formation, political formulations will ultimately cater to his needs. Consider for example an income distribution, skewed to the right, where mean income is higher than the median income (graph 1 shows the scenario in India). In such a scenario, the median voter will be more favourably inclined towards redistributive policies as he stands to gain from them.
Lupu and Pontusson (2011) however argue that rather than the level of inequality it is the structure of inequality, in other words the relative distance between the groups, that determines the median voters preference for redistribution. They argue that middle income voters are more likely to favour redistributive policies if the distance (income gap) between the middle and the poor (measured by lower tail inequality) is small relative to the distance between the middle and the rich (measured by upper tail inequality). However, Corneo and Grner (2002) argue that a smaller income gap doesnt necessarily imply greater affinity with the poor and support for redistribution. On the contrary, they hold that greater closeness will lead the middle class to oppose redistributive policies as it will threaten their own position in the status hierarchy.
In order to examine the changing structure of the distribution in India, graph 2 presents estimates of the 90/50 and 50/10 ratios over the past two decades. The former, a ratio of the monthly per capita consumption expenditure (MPCE) of the 90th (rich) and 50th (middle) percentile measures upper tail inequality, while the latter, a ratio of the MPCE of the 50th (middle) and 10th (poor) percentile measures lower tail inequality. Data from the NSS consumption expenditure surveys does point to a rise in upper tail inequality in the post liberalisation period. On the other hand, lower tail inequality is relatively stable, falling marginally in the initial period from 1993-94 to 2004-05 and rising marginally thereafter. What are the implications of the changing structure of the distribution for the direction of policy in India Will the median voter be more favourably inclined towards redistribution or will rivalry between groups limit its support to redistribution
One must point out that the theorem uses the terms median households and middle class interchangeably implying equivalence between them. That is not the case in India. The middle class, whichever definition one may use, lies in the 8th-10th decile. Thus the argument of rivalry between the middle and the poor made by Corneo and Grner (2002) may not apply in the Indian context. It is equally important to note that middle income or median households are significant beneficiaries of redistributive programmes. Data from the NSS surveys does confirm that households in the 3rd quintile receive subsidised food through the public distribution system (PDS) as well as income through MGNREGA. Furthermore, the Food Security Bill, which aims to cover roughly 60% of the country, will clearly encompass the median voter. Even the middle class in India is a significant beneficiary of subsidies. Thus, it can be argued that the combination of low per capita income and inequality in India implies that successful political platforms will pursue policies geared towards redistribution from the rich and middle classes to the poor, including the median voter. This raises a fundamental questioncan alternate political platforms, not solely revolving around redistributive policies, successfully cater to the median voter Is there a tipping point beyond which the demand for redistribution will taper off
Research does indicate that preferences for redistribution are closely tied to beliefs about equality of opportunity and social mobility (Alesina and La Ferrara 2001). Individuals, who believe that society offers equality of opportunity, are averse to redistribution if mobility is high. This leads to two conclusions. First, the allegiance of the median voter will shift depending on the distance between the income groups. Second, it will also depend on the extent of mobility between the groups.
Intuitively, mobility between groups is in part a function of the distance between them. An increase in the 90/50 ratio does imply that higher income growth is required for individuals to move up the income ladder. On the other hand, the smaller gap between the middle and the poor does imply that the middle has a higher probability of sliding back into poverty. Will such a scenario lead to greater affinity with the poor and thus tilt their preferences in favour of redistributive policies Or will high growth leading to greater mobility taper their preferences for redistribution Can a policy platform anchored in the idea of equality of opportunity, advocating greater social mobility, counter the appeal of greater redistribution at the hustings
In a recent post, Ajay Shah of NIPFP argued that many of them (recipients of welfare programmes) would be interested in a platform that shows a roadmap out of poverty instead of offering dole while perpetuating it. While a certain level of redistribution is undoubtedly desirable, is there a tipping point, in a country that has not yet reached middle income status, beyond which the appeal of greater social mobility will outweigh the demand for greater redistribution Put differently, will support for redistribution be less in a society that has higher levels of social mobility
The author works at the National Council of Applied Economic Research, New Delhi.
Views are personal