What the Economic Survey tells us about interstate trade – and what it means for GST

As always, the Economic Survey offers a mix of interesting and useful data, some “quasi-official” opinion on hot-button topics, and some speculations. The last two could simply reflect the interests of the Chief Economic Adviser. On the other hand, they could also be preparing the ground for policy action somewhere down the line.

This edition of the Survey is more interesting than most, for several reasons. First, it lays out a quasi-official position on the demonetisation while admitting that at least one key question is outside the scope of the Survey.

Second, it has some interesting new data on interstate trade, which is an important subject that’s being explored for the first time. That data and its explicability, has a bearing on the Goods and Services tax. Third, the Survey has an entire chapter on Universal Basic Income, a topic that has suddenly entered mainstream discourse in the past few weeks.

The demonetisation chapter can be dismissed in a few lines. The Survey cannot comment on the implementation. There were multiple post-facto rationalisations for it, all of which are quoted. During the release of the Survey, Subramanian added another creative rationalisation – that demonetisation would help to drive down real estate prices which are unrealistically high.

The Survey also says that, while there has been economic damage caused by the demonetisation, the formal data will not pick up much of that damage. It is not known how much cash will be put back in circulation and how long the process will take. It is being assumed that Gross Domestic Product growth will bounce back quite soon, perhaps with a consumption surge as cash limits are removed (whenever that is).

Interstate trade

The interstate trade data estimates are really new. It suggests that there is a lot more in the way of trade between states than was previously assumed. Interstate trade is estimated to generate close to 54% of GDP. In fact, in many cases, inter-state trade between contiguous states exceeds the intra-state trade.

The data are surprising because one of the major roadblocks when it comes to the ease of doing business is the friction caused by differentials in state taxes and laws. There are permanent 5 km long jams caused by trucks being held up at every state border and every business operating in India alleges endemic corruption in every state excise department. In fact, the GST is supposed to ease this situation.

The Survey suggests that the surprisingly high levels of interstate trade despite the frictions is because the current system of taxation creates distortions where inter-state trade is more profitable than intra-state trade due to Value Added Tax offsets. Hence, goods are moved across borders (or shown to be moved across borders) since that is more tax-efficient.

If this is the case, GST implementation might, paradoxically, lead to a pullback in interstate trade. This premise could be true. The data are obviously important when it comes to assessing the impact of a potential GST implementation and indeed, when it comes to planning future economic policy, at both central and state levels. The data could also be cited as a convenient way to minimise the potential impact of missing the GST deadline (currently September 30, 2017).

Competitive federalism is, and will remain, a key element of India’s future growth and designing tax systems that foster trade within the Union is important. However, there is another data-set pertaining to income, which indicates that there are major anomalies in the Indian model of competitive federalism.

Internal migration

India has strong patterns of internal migration. The Survey has attempted to assess this for the first time. Labour moves to urban centres. Labour moves across states. We’ve always known this anecdotally. The free movement of labour is one sign that a free market is working. (I’d suspect that the Survey actually underestimates the volume of internal migration, or maybe the criteria it is using is too rigid). This free movement of labour should buttress a model of competitive federalism.

Everywhere in the world, (China and global data is cited in the Survey), competition results in a scenario where low-income regions (states, nations, continents) grow quicker in comparison to high-income regions. This leads to income convergence as low income areas start to catch up.

This form of convergence is also true where public health indicators such as Life Expectancy and Infant Mortality Rates are concerned. It’s easy to understand why. Incremental life expectancy and and infant mortality rate gains in high life expectancy/ low infant mortality rate countries ( such as Japan) tend to be lower than exponential gains in low life expectancy/ high infant mortality rate countries (Chad, Somalia).

The extreme example is Sub-Saharan Africa where the AIDS epidemic knocked life expectancy down to mid-30s. Africa is making very fast gains now. The life expectancy and infant mortality rate convergence is visible in India where low life expectancy/ high infant mortality rate states (Bihar, Chhatisgarh) have made more gains than the more advanced high life expectancy/ low infant mortality rate states (Kerala ,Tamil Nadu).

But the income convergence is not true internally, where India is concerned. Low income states grow slower than high income states widening the income disparities between states. The data points to some distortions though it is unclear what these are. The Survey speculates the divergence could be caused by barriers to growth in low-income states and internal migration patterns lead to labour moving to high-income states. Certainly this is an area which is worth future investigation.

Universal Basic Income

Finally, we come to the concept of Universal Basic Income. Suppose that everybody – and so the prefix Universal – in India was given a certain sum of money by direct transfer? Or what if a basic income was provided to just those below the poverty line (however that is defined)? Or specific sub-groups (women BPL, or pregnant women BPL, or the lowest decile of income) were targeted and given certain sums of money. This could help to pull them above the poverty line and have sufficient daily calories; help them to sustain sudden shocks (such as a health crisis, or job-loss due to demonetisation) etc. The idea is often discussed and generally dismissed because it would tend to be too expensive.

There are reasons why some version of this could actually be efficient in India. India already has multiple (950 central schemes and thousands more at state level) which target the poor. Those schemes cost huge sums, misidentify targets and misallocate resources, which disappear into the maw of government officials and middlemen. If most of these schemes were replaced by a version of UBI, the government might actually save money while improving the targeting and vastly reducing leakages.

On the other hand, such a UBI would, once implemented, be politically impossible to phase out. There would be vast amounts of lobbying (including rioting by caste-based groups) to be included in the definition of recipients. It is also likely that many key populist welfare schemes could not be phased out due to political reasons. The Survey builds a set of scenarios and guesstimates for the UBI thought experiment. The error margins for any of those estimates would be high.

As a paid-up member of the conspiracy party, I find it interesting that the concept of UBI was introduced into India’s public discourse only in the past couple of months. I also find it interesting that this could be a vehicle for some grand gesture that “gives away” the gains of demonetisation by putting the equivalent of “a chicken in every pot”. Still, conspiracy theories apart, it is an interesting idea and it could be useful if it’s implemented in certain ways.

In sum, the Economic Survey is an interesting document. You will note that I have avoided using any numbers in this review of the document. That’s deliberate.