| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © The Author 2007. Published by Oxford University Press.
The evolution of institutions in India and its relationship with economic growth
* International Monetary Fund, Washington, DC, e-mail: asubramanian{at}petersoninstitute.org
 |
Abstract |
|---|
This paper examines the evolution in a few public institutions over time in India. It presents three types of evidence: on institutional outcomes (such as losses in power generation, backlogs in disposal of court cases); on perceptions-based measures of governance, some going back to the 1960s; and, finally, on customs administration and whether it has been more effective at detecting evasion over time. All the evidence suggests that institutional quality has not improved over time. It then addresses the two-way relationship between growth and institutions in terms of two apparent paradoxes. The first is why growth has turned around so dramatically in India despite the relatively limited nature of reforms, especially compared with other countries. The second paradox is why, despite nearly 30 years of rapid growth, institutions have not improved. The paper offers some explanations that might help explain these paradoxes.
Key Words: institutions • growth • India
I am grateful to Josh Felman, Ejaz Ghani, Ajay Chhibber, Shanta Devarajan, Nurul Islam, Simon Johnson, Devesh Kapur, Sanjay Kathuria, Deena Khatkhate, Kalpana Kochhar, Utsav Kumar, Pratap Bhanu Mehta, Nandan Nilekani, Raghuram Rajan, C. Rammanohar Reddy, Krishna Srinivasan, and S. Sriramachandran for useful discussions. This paper draws upon work with my colleagues, Prachi Mishra and Petia Topalova. An anonymous referee and the editors provided helpful comments. Manzoor Gill and Thoihen Singh provided excellent assistance with data.
1 The distinction between policies and institutions is not always easy to draw. In the schema described above, policies can be seen as actions that affect the scope of the state as owner and (over) regulator, while institutions affect the market-creating functions and the more necessary regulatory functions.
2 The Supreme Court recently even decided on the fate of 300 rhesus monkeys held in captivity.
3 This is not true of the state public-service commissions, which have been scandal-prone.
4 The changes in the mean and standard deviation of losses over time are as dramatic even when two apparent outliers—Delhi, and Jammu and Kashmir—are excluded.
5 Misreporting is a common problem for most crimes, but less so for those related to murder.
6 A similar decline in the conviction rate is also evident in cases handled by the Central Bureau of Investigation (CBI), from 17 per cent in 1972 to 9 per cent in 1999. Das (2005) characterizes this decline as arising from ‘wilfully incompetent investigation and prosecution by the CBI’.
7 The World Bank has begun computing more objective indicators of governance (World Bank, 2006), but these are available only since 2002.
8 Over the last decade, Indian performance on government effectiveness has shown a slight improvement, while that on corruption has remained flat. It is possible that perception-based measures, especially for a category as vague as ‘government effectiveness’, are more influenced by the overall performance of the economy, and hence should be discounted more heavily.
9 It is worth re-emphasizing that these are relative measures: at each point in time a country is assessed against other countries. If there has been a generalized upward drift in the quality of institutions, it is possible that the relative decline in Indian institutions portrayed might not represent an absolute decline.
10 A clarification might be helpful here. Even if the elasticity of evasion with respect to tariffs is constant (i.e. invariant with respect to tariffs), in theory, this evasion elasticity can vary with enforcement (institutions/administration etc.). What this means is that, if one draws a curve in evasion–tariffs space, the shift of the curve (when institutions change) is not necessarily parallel (Allingham and Sandmo, 1972; Slemrod and Kopczuk, 2002).
11 The basic specification involves regressing our measure of evasion which varies by time, partner country, and product on nominal tariffs, which vary by product and time. The specification is very general because it includes a full set of country–time, and country–product fixed effects, with standard errors clustered at the product level. Our core sample comprises more than 300,000 observations. Our key finding is that the coefficient on the tariff, which is significant and positive (as expected, because higher tariffs should increase the incentives for evasion) does not change over time. For details, see Mishra et al. (2007).
12 Of course, a host of other possible factors determine growth outcomes, including initial conditions, human capital, inequality, etc. But in a standard Barro-type cross-country growth regression, India proves to be an outlier even after controlling for some of these determinants (see Table 4 in Rodrik and Subramanian (2005)).
13 Latin America is, on average, richer than India, so convergence on its own would suggest higher growth rates in India. But the magnitude of the differential cannot all be explained by convergence dynamics alone.
14 As in Rodrik et al. (2004), institutions and openness have been instrumented.
15 Other controls include state and time fixed effects, the Besley–Burgess indicator of labour-market reform interacted with decadal dummies, and initial income plus its interactions with decadal dummies. The relationship between state-level growth and institutions is robust to alternative ways of measuring the quality of institutions (Kochhar et al., 2006).
16 Banerjee and Iyer (2005) present strong empirical evidence that institutions, not just contemporary ones, but those created nearly 200 years ago, have had lasting effects on contemporary economic performance in India. They show that variations in land tenure systems can explain the pattern of variation in agricultural investment and productivity, as well as in health and education indicators across districts in post-independence India. For example, the average yield of wheat is 23 per cent higher and infant mortality about 40 per cent lower in those districts that did not have a zamindari (landlord-based) tenure system compared with districts that did.
17 Acemoglu et al. (2005) argue, however, that this is a pure cross-sectional result and that the time-series evidence is less conclusive.
18 The relationship is positive but statistically weaker when we include the smaller states and union territories.
19 Student–teacher ratios are better in private schools, as is the problem of teacher absenteeism. Teacher salaries are about one-sixth those in public schools and teachers have greater contact with their students. Students in private schools have higher test scores and are taught English a grade earlier than in public schools. And accountability is much better, too: the dismissal rate of teachers is 35/600 in private schools compared with 1/3000 in public schools.
20 Note that efficient private-sector provision of essential services, such as education, does not eliminate an important role for the public sector: from an equity perspective, the fact that poor households have to pay for private education is a social cost. Luckily, this can be remedied by public financing of primary education, while allowing the private sector to continue to play a large role in provision.
21 Nandan Nilekani, Chairman of Infosys, the IT giant, argues that roads are likely to get built in India relatively quickly because the burgeoning middle class, having acquired cars—the iconic symbol of wealth and status—needs ‘somewhere to go’ to show them off.
22 In the study on customs evasion discussed earlier, we found that the average value of customs transactions handled by the typical customs officer in India is about Rs 29 m per month. The monthly salary, on the other hand, for a customs inspector is Rs 9,000 per month. In other words, even if, on average, corruption amounted to 0.1 per cent of the value of transactions, the customs official would make an amount that is more than three times his monthly salary.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  V. Bhaskar and B. Gupta India's development in the era of growth Oxf. Rev. Econ. Policy, June 1, 2007; 23(2): 135 - 142. [Abstract] [Full Text] [PDF]
|
|