The chaotic consensus on Goods and Services Tax
To refer to India’s soon to be unleashed indirect tax as a goods and services tax (GST) would be a stretch. Conceptually, one of the main advantages of the GST is that it is a tax on value added at each stage of the supply chain. Not only does this avoid a “tax on tax”, but also encourages compliance by the information technology-enabled mechanism of crediting input tax credit from one stage to another, thus incentivizing firms to obtain a receipt for their input purchases.
However, states have been successful in keeping close to 45% of their indirect taxes, including sales taxes on alcohol and specific petroleum products, outside the ambit of the tax.
This is only one among the many concessions that states have been able to wrest even though, as per the 122nd amendment to the Constitution, the Central government seemingly enjoys overweening power on the GST council. Why then did the Prime Minister surrender those powers in favour of a consensus-based approach to give us a watered down version of GST?
Let us examine the true import of the voting rules of the council.
The council consists of representatives of the Central government as well as of each state. Every decision requires a majority of not less than three-fourths of the weighted votes of the members. The vote of the Central government has a weight of one-third of the total votes cast. The votes of all the state governments taken together have a weight of two-thirds of the total votes cast, with each state having equal weight.
The Centre enjoys the power of veto in the sense that no resolution can achieve a three-fourth majority without its assent. But does a veto necessarily translate into a large measure of legislative power?
In the GST council, the Centre must get 18 of the 29 states on its side in order to cross the mark. Assuming the states are independent, this is not an easy task.
The Shapley-Shubik power index, that assigns a measure of power in a legislature based on the ability of an entity to convert a randomly chosen coalition from a losing to a winning coalition, assigns a power index of only 40% to the Centre in the extant GST council. In other words, in a randomly chosen coalition, the Centre is able to cast a decisive vote only 40% of the time.
But the states are not independent, you might rightly contend. Even before the recent round of assembly election, the National Democratic Alliance (NDA) was in power in 14 states, with the Bharatiya Janata Party (BJP) itself in the driving seat in 10 of these states. If NDA partners or even just BJP states acted in concert, the Centre would be firmly in the driver’s seat. Can one not see the current compromise as an attempt by the Centre to preserve fiscal federalism?
Not quite. The GST council carries out its deliberations as part of a larger system of political democracy, comprising the Central legislature and state legislatures. The passing of the constitutional amendment bill has to be followed by the passage of several bills, including the Central GST and inter-state GST in Parliament and the state GST in the individual state legislatures.
It is unlikely that any decision by the council that involved the ceding of significant fiscal power by the states would pass muster.
Even without the external constraints, models of negotiation in game theory developed by Ariel Rubinstein and Martin Osborne identify “patience” as an additional factor determining the trajectory of decisions in the GST council.
Imagine a situation in which a dollar has to be split between two parties. At each stage, one of the parties proposes a split. If the other party rejects the split offered, then it has to make a counter offer. With a finite number of stages, the player that gets to make the proposal in the last stage seems to hold a decisive advantage as the other player will have no option but to accept the offer. However, this advantage is significantly curtailed if this player is more impatient than the other one.
The Centre too enjoys an advantage on account of the voting rules of the GST council. But both the Centre and the states know that protracted bargaining is to the disadvantage of the Centre. Therefore, as in the alternating offers game, the Centre makes a generous offer right in the first stage and terminates the negotiation.
As per this interpretation, the decision to go for consensus reflects not the desire to uphold fiscal federalism as much as an olive branch extended to the states in order to achieve an early closure.
While the patience of the states can indicate a desire to understand issues, and seek common ground, it could also reflect the ability to stonewall merely for the sake of it. For instance, states that refused to accept the harmonized international system of nomenclature in which goods are bucketed in commonly accepted categories reflect an unconstructive obduracy. With Parliament having passed the Central and inter-state GST bills, the boot is on the other foot. States that do not pass the state GST bills by 16 September will lose out on tax credit for their industries selling in other states. Hence, they will be impatient to pass the bill.
Finally, the storm of activity in the business world in the wake of GST indicates that many states will end up losing business activity. Negotiations in the council are only going to get more contentious. Having bent backwards already, will the Centre be prepared to bend some more?
Rohit Prasad is a professor at MDI, Gurgaon, and author of Blood Red River. Game Sutra is a fortnightly column based on game theory.
With inputs from Chandrahas Mathur, a former Indian Revenue Service officer.
Comments are welcome at firstname.lastname@example.org