EDITIONS
Opinion

Is size destiny?

Shailesh Jha
29th Jul 2016
  • Share Icon
  • Facebook Icon
  • Twitter Icon
  • LinkedIn Icon
  • Reddit Icon
  • WhatsApp Icon
Share on
Switzerland is way small, they don’t even have this kind of diversity; that’s what makes it so easy to get things done there.

A stock refrain I hear when I compare the ease-of-doing business and public infrastructure in India and Switzerland.

Pop economic statements have a factual basis though. The top ranks for a range of human development indicators—whether the HDI Index or World Bank Development indicators—are dominated by small states (population-wise) –such as Iceland, Norway, Switzerland, Sweden and New Zealand. That naturally poses a question- if development of human capacity and easy growth of economic activity is a metric, do small countries have a natural advantage over large diverse ones?

Big-Opportunity

When lack of variety helps

The most cited empirical exercise on this topic comes from Alesina, Baqir and Hoxby (2005), who created an index of heterogeneity that ranks countries on the basis of linguistic, ethnic and religious diversity. Seeing countries with high HDI ranking (over 60 percent of them being small) through this prism, we find the ones in the top 10 percentile tend to have 22 percent lower levels of diversity than those in the bottom 10 percentile, and about eight percent lower than the median. Further, despite lower resources to tax, smaller states tend to have lower fiscal deficits and levels of public debt. As per data from the World Bank, such states also spend a relatively higher share of their GDP on education and healthcare services.

One way to understand this would be that small homogeneous states avoid the inefficiencies of scale- public services like policing, healthcare and education can be provided in a standardised template without customising them for diversity. The other could be that homogeneity of population reduces the costs of political transactions and replication of governance models, allowing consensus over public policies to be arrived at quickly.

The success of city-states

While lower diversity does explain the successes of small Nordic and Alpine states, it doesn’t explain why small city states flourished economically and socially in the past (Venice, Genoa) or in the present (Singapore, Hong Kong). Not only have these states been ethnically diverse, they have actively courted outsiders as a part of their survival strategy. Perhaps this riddle could be explained by population density. The advantage of lower costs of politics and public services helps small states where population density is low and where similar communities are geographically separated.

However, when diverse communities are squeezed into a small space—as in city states of present and the past—the aspiration to integrate develops its own cosmopolitan identity among its diverse inhabitants. This combined with the benefits of proximity on the productivity of firms and people offsets the higher costs of political transaction. In fact, the autonomous city state model seems to be the recipe for success even for large countries with a strong record of human development- over two-thirds of output for large Japan is produced by locally governed city-states of Tokyo and Osaka; the same could be said of Amsterdam and Rotterdam in relation to the Netherlands.

Size isn’t destiny, nor is diversity a curse. That said, compactness does offer a greater flexibility to change and lower political costs. For countries like India, which are not as small as Switzerland, a good start on the road to human development would be to create a cluster of efficiently-managed Singapore within.


Note: The classification of states as small and large population wise has been done by the author. I have classified states with population less than 10 million as of 2014 as 'small' and ones with population larger than 30 million as 'large'. The relationship between population density, diversity and level of human development in small states was also explored by me independently as a part of this exercise. I have done some data analysis to arrive at the conjectures stated in this article. Further details can be provided on request.


References: Alesina A, Baqir R and Hoxby C (2004). “Political jurisdictions in heterogeneous communities”. Journal of Political Economy 112 (2), pp 348-56

  • Share Icon
  • Facebook Icon
  • Twitter Icon
  • LinkedIn Icon
  • Reddit Icon
  • WhatsApp Icon
Share on
Report an issue
Authors

Related Tags