ICICI Bank chairman K.V. Kamath had made a perceptive point in the course of an interview with this newspaper in December. The question was on the tipping points that we would see in India, now that per capita income has crossed the $1,000 mark. Kamath pointed out that the end of the last century had seen a tipping point for aspirations and consumption, one reason why his bank had decided to refocus on retail lending. Millions of Indians have now moved into the consuming classes. There is also a growing class of people with average incomes of $1,000-2,000 “that are keen on a better quality of life, including better infrastructure".

Also Read Niranjan Rajadhyaksha’s previous columns

The demand for better roads, power supply, water and trains is bound to grow as median incomes rise in India. Indian business leaders and policy experts are all agreed on the fact that India needs far better infrastructure if it is to maintain high growth and international competitiveness. What is less discussed is the demand for such infrastructure from voters, because that is what will eventually drive the political class to action. Growing average incomes should ideally lead to a greater value being paid on the quality of life and infrastructure.

I also sense a greater readiness to pay for these goods and services. There was a huge debate about user charges around 10 years ago: Will citizens pay higher prices for better infrastructure? The implicit answer: no, they will not. At least in some areas such as roads, we see people now showing readiness to pay tolls, given the clear benefits, in terms of fuel costs and time saved, that new highways and bridges offer.

We usually assume that it is urban voters who are more keen on better infrastructure while rural voters prefer government spending on employment schemes and transfer payments. Political strategies have been built on this assumption. However, a recent World Bank paper authored by Stuti Khemani cites several surveys that show there is robust demand for village infrastructure from poor voters in various Indian states. Many of these surveys show that villagers value water and infrastructure higher than even education.

This is perhaps evidence of a curious divide between voter preferences and public policy. Few governments are as keen to spend on new roads or water schemes despite voter demand for them as they are to spend on entitlements. How does one explain this important paradox?

Khemani argues that “…this pattern is due to infrastructure projects being used at the margin for political rent-seeking, while spending on employment and welfare transfers are the preferred vehicles to win votes for re-election".

There is a deeper issue here. It is widely known that infrastructure projects are fountainheads of corruption. Look at the Commonwealth Games, for the latest example. Large projects are useful to bankroll political machines and bolster personal wealth. A senior bureaucrat in the Maharashtra government once told me that politicians love public-private partnerships because they are so lucrative. There is thus a clear incentive for politicians to spend taxpayer money on ambitious projects. So why don’t they spend more on infrastructure?

This is where there is an intriguing twist in the tale. The World Bank shows that much depends on the political climate and the strength of institutions to check corruption. States that have low levels of governance, relatively less political competition and low voter turnouts tend to show higher spending on infrastructure, because the attractiveness of rent seeking (or corruption) is higher and the threat of a voter backlash is lower. Paradoxically, better-run states tend to have a higher share of government spending on job schemes and welfare transfers.

“This evidence across the Indian states…is consistent with recent cross-country research, which finds greater public investment spending in countries with lower ratings for governance, political checks and balances, and electoral competition," says Khemani. These findings are definitely counter-intuitive. Most of us believe that the better-managed states are keener to spend on infrastructure while the mismanaged states are more likely to spend on populist schemes.

The reason these preliminary findings are important is that public policy will not be designed only on the basis of good ideas, other than in times of utter crisis when competing groups sink their difference in order to survive. The political class needs clear incentives to reorient its spending patterns away from populist schemes and towards infrastructure—to “improve the alignment of political interests with development interests", according to Khemani.

The neat divide between using infrastructure spending for personal benefit and welfare spending for getting votes is now standard practice in India. Disturbing this equilibrium is one of the most important tasks for reformers. The main reason for optimism is that voter preferences seem to be gradually changing in step with higher average incomes, especially among those who have benefited from two decades of economic reform and higher growth.

Niranjan Rajadhyaksha is managing editor of Mint. Your comments are welcome at cafeeconomics@livemint.com