With Laveesh Bhandari, I have sometimes attempted to rank Indian states. Every such ranking exercise has been controversial. None more than the present one, which is an economic freedom rating for states, sponsored by Friedrich Naumann Stiftung and published by the Rajiv Gandhi Institute for Contemporary Studies in March 2005. Media reports on what we have done are nothing short of the bizarre, linking it even with the denial of a visa to Narendra Modi. Whenever one wishes to rank states, there has to be an objective. In this case, the objective was to capture economic freedom ' human freedom consisting of economic, social and political freedom. Explicitly and deliberately, we were quantifying and measuring economic freedom, not social and political freedom, a distinction that people who react to Gujarat's rank don't seem to appreciate.
What is economic freedom' The Heritage Foundation, which brings out a cross-country index of economic freedom, defines economic freedom as 'the absence of government coercion or constraint on the production, distribution, or consumption of goods and services beyond the extent necessary for citizens to protect and maintain liberty itself'. Fair enough. But while capturing the negative that the government should not interfere in the working of the market, one should also capture the positive that the government should fulfil its duty of protecting life and property and enforcing legal contracts and law. Such variables are included in the Fraser Institute's cross-country 'Economic Freedom of the World' exercise.
Most economic freedom ratings are cross-country. Sub-national attempts have been made in the United States of America, Canada and China. Ours is the first sub-national or inter-state attempt for India, and being the first attempt, is a first cut and no more. Whenever our state rankings (and we have done some others) are criticized, the criticism is usually because we haven't included some variables. While this is a valid point, our rankings are based on objective data. There is no attempt to collect data on the basis of subjective responses to questionnaires. Consequently, the inclusion of variables is data-constrained and this is the reason why the present exercise only covers 20 states, not 35 states and Union Territories, as we would have liked to.
The 26 variables for which we had data were divided into three heads of size of government, legal structure and security of property rights, and regulation of credit, labour and business. For instance, government revenue expenditure as a share of gross state domestic product is a variable under the first head. The ratio of total value of property recovered to total value of property stolen is a variable under the second head. And the number of man-days lost in strikes and lockouts is a variable under the third head. After making data comparable and normalizing them for size of the states, each state obtains a score under each head. These scores are absolute indicators of how a state has performed.
Consider the first head, size of government. Chhattisgarh heads this list with a score of 0.47 and Orissa is at the bottom with a score of 0.17. A point needs to be made about scores and ranks. Attention is usually focused on ranks, but what is important are the scores. For instance, a state may be ranked above another, but the difference in scores may be minimal. Gujarat is ranked second in size of government with a score of 0.46, but the difference between Chhattisgarh and Gujarat is negligible. But clearly, there is a robust and significant difference between Chhattisgarh and Gujarat, on the one hand, and Orissa, on the other.
Incidentally, West Bengal is ranked 5th in size of government with a score of 0.36, but there is little to choose between the scores for Bihar, West Bengal, Jharkhand, Karnataka, Tamil Nadu and Maharashtra. One should also make another obvious point. One can't correlate a state's performance with the success or failure of an incumbent government. A state's performance usually reflects legacies of earlier governments. We have ranked Gujarat. We haven't ranked the Narendra Modi government. That would have been impossible to do in our framework and that wasn't our intention.
Moving on to the second head of legal structure and security of property rights, Tamil Nadu heads the table with a score of 0.55, followed by Madhya Pradesh with a score of 0.54. West Bengal is 17th with a score of 0.17, followed by Assam, Bihar and Jharkhand, in that order. Gujarat is 8th with a score of 0.39. Finally, on regulation of credit, labour and business, Maharashtra tops the rankings with a score of 0.50. Madhya Pradesh is at the bottom with a score of 0.13 and West Bengal 5th with a score of 0.37.
Gujarat is 6th with a score of 0.37, the difference in rank between West Bengal and Gujarat occurring at the third decimal place. Thus, Gujarat doesn't do uniformly well in our rankings. It does relatively well on size of government, but not on legal structure and security of property rights or on regulation of credit, labour and business. We could have left our ratings at that, with three separate scores for states on the three individual heads. However, an overall composite index is always useful. But that requires a process of aggregation across heads, and aggregation is a function of weights used and of the aggregation method.
Alternative weights and alternative aggregation can lead to differing results. For instance, in some other state exercises, we have used principal components. But these are difficult to explain to ordinary readers. In this case, we therefore used an additive arithmetic index to arrive at the composite or overall values. And this shows a Gujarat score of 0.40 and a Gujarat rank of first, followed by Andhra Pradesh. West Bengal is 14th with a score of 0.30 and Assam is at the bottom with a score of 0.22. This overall top ranking for Gujarat occurs because size of government pulls Gujarat up by far more than legal structure and security of property rights or regulation of credit, labour and business pulls Gujarat down.
Media reports and advertisements have focused on Gujarat's top rank, to the exclusion of other interesting findings the study throws up. For instance, there is the expected positive correlation between per-capita state income and the overall economic freedom index. There is also an interesting correlation between the economic freedom index and out-migration from a state. There is an inverse relation between economic freedom and out-migration, but only when a threshold level of economic freedom has been crossed. Below that, transaction costs of out-migration are presumably difficult to handle.
There is a weak positive correlation between inflation and economic freedom. Most economically free states have repealed the Urban Land Ceiling and Regulation Act. And most economically free states have passed right to information acts, Gujarat being an exception. As a first attempt in capturing sub-regional economic freedom in India, we believe our exercise deserves attention. But not in the sense attention has been attracted. First, we measured economic freedom, not social or political freedom. Second, Gujarat does not come out with flying colours everywhere. Third, any ranking is subject to variables and aggregation used. Fourth, there is nothing in the study that links Gujarat's overall top score to the performance of the present Gujarat government. Such an interpretation is both suppressio veri and suggestio falsi. Fifth, this is a study undertaken by RGICS, not by the Rajiv Gandhi Foundation. And in all RGICS publications, the usual caveat about views of authors not necessarily representing the views of the organization applies. Unfortunately, people sometimes tend to stretch the truth.