RBI data isn’t enough: It’s premature to argue if demonetisation was a success or failure

Demonetisation of Rs 500 and Rs 1000 notes resulted in long queues to exchange them.
This composite picture shows the demonetised Rs 500 note as well as people standing in a queue to exchange them. Source: Miran Rijavec, Scott Edmunds/Flickr

Other factors need to be taken into consideration like the number of people who now are part of the formal economy.    

Demonetisation debates, which had lain dormant for some time, sprang to new life this past week, with the publication on 30 August 2017 of the Reserve Bank of India (RBI) annual report. Tucked away innocuously near the back is a statement that, of the total stock of demonetised currency, which was Rs. 15.4 trillion, fully Rs. 15.28 trillion had come back to the RBI by 30 June 2017, the end of its operating year. This amounts to a whopping 99 percent of the total. What is left unreturned, amounts, essentially, to spare change.

The quantum of unreturned currency had been one measurable criterion set up by both defenders and critics of demonetisation. For instance, shortly after the demonetisation announcement, Minister of Finance Arun Jaitley had opined that some of the demonetized currency would be “extinguished” as people “who had used cash for crime purposes were not foolhardy enough to try and risk and bring the cash back into the system”. Meanwhile, in an appearance on NDTV, Niti Aayog Member Bibek Debroy had anticipated that some “10 percent” of notes in circulation would not return. Notwithstanding the obvious challenges involved in estimating the quantum of currency returned, especially in a continuously evolving regulatory environment, the recent news that almost all demonetised currency returned to the system has led many to conclude that the exercise was a failure.

But was it? First off, it must be conceded that if demonetisation is to be judged narrowly on the basis of the triple rationale originally advanced (see for instance in this publication), it would at best be unclear if it could be accounted a success. For, little black money was literally “destroyed” and there is scant evidence that the policy had much if any impact on counterfeiting or terror finance.

Having conceded as much, one may further ask, is the quantum of unreturned money the most relevant test for judging the success of the pursuit of black money? It is our contention that this rush to judgement — declaring demonetisation a failure on the strength of the RBI news — is premature, at best, and requires a sober second look. Indeed, we draw here upon, revisit, and refine the argument that we had originally made on 27 December 2016 (“Demonetisation fallacies and demonetisation math”, Mint), and which remains relevant to the controversies today.

The Rs 500 notes that were demonetised by the Reserve Bank of India
Source: Miran Rijavec/ Flickr

Fundamentally, as we had argued even in December, it is a fallacy to argue that the return of much or most cash (including presumed black money) into the banking system represents a failure of the policy to flush out black money, for this is only one of two possible avenues through black money may be “taxed”. Indeed, it is true that the cleanest and simplest form of gain likely occurs with unreturned notes, as these liabilities are extinguished, thereby creating pure and simple seigniorage revenue for the central bank which may, in principle, be turned over to the government, so generating a “fiscal gain”. Importantly, however, unextinguished liabilities, i.e., money deposited into bank accounts can also generate fiscal gain, as these will invite the scrutiny of tax officials. These are the two key paragraphs of our original analysis:

Clearly, at least from the perspective of its effectiveness in dealing with the black money issue, success has to be measured by the sum of tax revenue generated and black money destroyed. Suppose we accept the estimate that one-third of the approximately Rs 15 trillion in demonetised notes is black money. Roughly speaking, the revenue that would have been generated had that income been taxed in the first place is 30 percent of that (so, Rs 5 trillion times 0.3 = Rs 1.5 trillion). Perfect detection of black money should now yield 50 percent as tax revenue (so Rs5 trillion times 0.5 = Rs 2.5 trillion), if all black money is returned and identified as such.

Of course, perfection is rarely achieved in practice. Allowing for slippages, we should neither expect all the black money to be returned, nor that all the black money that is returned will be perfectly identified and taxed. If we assume that by 30 December, Rs 1 trillion is unreturned, as is believed, and we further assume that only half of the remaining Rs 4 trillion of black money that is returned falls within the tax net, the net gain works out to Rs1 trillion of black money destroyed and 50 percent times 2 trillion = Rs 1 trillion in tax revenue. When compared with the approximately Rs 2.5 trillion in income taxes collected annually, the government could reasonably claim this as a successful outcome.

The reader will note that in our preceding analysis we had indeed expected that most of the black money would be deposited back into bank accounts — an outcome whose likelihood had increased substantially given the less punitive rules for black money deposits that the government had announced by late November. What is more, the nation’s banks, with the encouragement of the RBI, made a Herculean effort to open new counters to allow the unbanked to create new bank accounts and deposit their currency with minimum fuss and no cost. In other words, it is evident that by this stage, the RBI and most observers had assumed most of the money would return: even if many, including ourselves and including critics, overestimated the quantum of black money that would ultimately be unreturned.

That, then, leaves us to revisit the second half of our demonetisation maths, which concern the fiscal gain to be had from the taxing of money deposited into bank accounts after November 8. Clearly, in the event of perfect laundering, such as by spreading black money into small “benami” deposits each under the threshold for scrutiny (Rs 2.5 lakh) announced by the government, such revenue to be gained would be zero. But the figures announced by the government reveal a very significant number of large deposits. Information placed before the nation by Minister of Finance Arun Jaitley in his Budget speech on 1 February 2017 indicated that: “During the period 8th November to 30th December 2016, deposits between Rs 2 lakh and Rs 80 lakh were made in about 1.09 crore accounts with an average deposit size of Rs 5.03 lakh. Deposits of more than 80 lakh were made in 1.48 lakh accounts with average deposit size of Rs3.31 crores.” These deposits, which are mostly open to scrutiny by tax officials, amount to over Rs. 10 trillion, or around two-thirds of the value of the demonetised currency. This is not peanuts.

It is evident that the prospect of fiscal gain via the conventional taxation channel, while admittedly less pristine than the pure seigniorage revenue associated with extinguished liabilities, keeps open the possibility that demonetisation may yet yield a sizeable fiscal gain. The complete return of demonetised money into the banking system does not indicate, in itself, a failure of the policy, and a fair observer should be obliged to keep an open mind. Should, however, the government fail in identifying and taxing black money deposits in any significant quantity, we can all conclude that demonetisation will have failed in achieving its primary goal.

We make three additional points:

A queue of people stand outside a bank to exchange old notes in 2016 after the government announced the demonetisation scheme.
A queue of people outside a bank in Mumbai to exchange demonetised notes | Source: Scott Edmunds/Flickr

First, several observers have suggested that an “amnesty” approach allowing for black money to be deposited into bank accounts, to be subject to taxation, but not other punitive action, would have been a reasonable alternative to demonetisation. This does not hold true. In the absence of the Damoclean sword of demonetisation that un-deposited money would soon become worthless, there would have been no incentive for black money holders to use the amnesty window and subject themselves to taxation. As a singular policy instrument, demonetisation carries a unique combination of carrot and stick which a policy such as an amnesty alone cannot match in its incentive effects.

Second, it is worth remembering that much of the furore over the ill-effects of demonetisation only arose due to the time-consuming process whereby remonetisation occurred, something rendered unavoidable by the fact that planning for demonetisation and its aftermath were conducted in secrecy. This said, in principle, had demonetisation occurred without transition costs — for instance, if old notes could have been seamlessly converted or deposited within a few days after 8 November, or if demonetisation had been pre-announced to occur with a lag, allowing time for an orderly remonetisation — there could only have been largely upside gain without any downside cost. The reason that demonetisation continues to remain controversial and emotive for its critics is the images we have in our minds of throngs of desperate folks lining up in snaking queues to exchange their old notes.

Despite the daunting challenges in implementation, the RBI and Governor Urjit Patel are nevertheless to be credited for having successfully managed this remonetisation — surely the largest operation of its type in modern monetary history — with only modest economic disruptions: a point conceded even by early critics of demonetisation, such as economists Paul Krugman and Kenneth Rogoff, who have recently acknowledged as much. Indeed, the fact that GDP growth witnessed only a modest drop in the third quarter of the fiscal year — when the putative impact of demonetisation ought to have been greatest — speaks to the RBI’s remarkably successful remonetisation effort.

Finally, while many may have succeeded in laundering their holdings of black money, it is apparent that a large implicit tax had to be paid during the money laundering process. Thus, anecdotal evidence suggests that, starting on the night of 8 November itself, a secondary black market had already developed in the old, soon to be demonetised notes which fetched in the first instance a discount of 60-70 percent, a discount that was to stabilise around 30 percent in the coming weeks (interestingly, approximately equivalent to what might be expected were one to deposit and face taxation, suggesting the market found an equilibrium). They perhaps paid somewhere between 30 and 70 percent as an in-effect-punitive intermediation “fee” (even if this is of small consolation to fiscal authorities and if we remain agnostic on the fact that the exercise presumably enriched some shady intermediaries).

Demonetisation resulted in a lot of short term costs but may have long-term benefits writes economist Jagdish Bhagwati
People stand in a bank in Kerala to exchange demonetised notes | Source: Wikimedia Commons

In short, it is premature to declare the demonetisation exercise a success or failure — using its fiscal calculus as the primary metric, as we have been doing all along since December. The only basis on which to assert unequivocally today that the policy was a failure is to stand on the narrow ground that no pure seigniorage revenue was earned, which as we have argued is an incomplete analysis.

It remains to be added that when a policy which amounts to the greatest natural experiment in modern monetary history is undertaken, it will take many more years for its full impacts and possible upside gains, quite apart from its short-term costs, which are evident, to be realised and quantified, especially when assessed against a broader range of criteria. The widely-reported disruptions to existing supply chains and the failure of some small firms without access to non-cash credit are a short and perhaps even a long-lasting negative consequence. On the positive side, evidence suggests a jump in digitisation and formalisation of the economy and banking system due to demonetisation, but it remains to be seen if this is a transitory effect due to a period of liquidity crunch or represents a permanent change in behaviour, for which demonetisation may take credit. More broadly, as we have previously argued (see here), by creating a disruption, demonetisation has revealed the urgency of completing the unfinished reforms agenda, the essence of which involves moving firms and jobs from the informal to the formal: this is at the heart of creating productive and well-paying jobs which will power growth and continue the attack on poverty.

At a minimum, whether or not tax is collected in the end, much previously unaccounted cash is now in the banking system, which in itself gives a fillip to formalisation. In the same vein, there appears to have been a jump in the number of tax assessees, although one must be careful to distinguish the impact of demonetisation from an overall rising trend. Finally, research by Rupa Subramanya (available here) on various metrics of digital banking clearly shows a structural break in the data after 8 November, strongly suggesting that demonetisation may indeed have a beneficial effect in this area. We would note that while this was not the original intended goal of demonetisation, it may yet prove the most important long-lasting benefit.

Jagdish Bhagwati, Vivek Dehejia and Pravin Krishna are, respectively, university professor, law and economics, at Columbia University; resident senior fellow at IDFC Institute; and Chung Ju Yung distinguished professor of international economics at Johns Hopkins University.


  1. All I can say is that Professor Dehejia, Bhagwati and Krishna dig deeper into the hole with this article. I will remind readers of this execrable piece talking about the demonetization boom http://blogs.timesofindia.indiatimes.com/toi-edit-page/looking-back-at-demonetisation-the-concerns-of-its-legion-of-critics-have-all-been-proven-plain-wrong/. They said “Indeed, we were perhaps among a small minority of economists who had argued explicitly that, contrary to this emerging consensus, demonetisation could, paradoxically, be expansionary rather than contractionary..” No mea culpa for that.
    Now, as per usual, they shift the goalposts. What is really disappointing is to see Professor Bhagwati, once a mighty hand against state excess, now arguing for it. I guess you become what you fight against in the end.

  2. Jagdish mama, we love you… But please give up. The misery in the “hinterland” is real. The jobs are nonexistent. The hate attacks are real. The combination of overweening authority and incompetence is stifling.

  3. This is truly a brilliant analysis. The sad part is that most of the opposition parties will hardly be concerned about reading the line of logic laid out in this article and understanding it because the exercise of demonetization had been criticised so strongly by Mr Manmohan Singh in his speech in the Rajya Sabha. Among the things he mentioned was that the scheme would be a total failure in the short term and in the long run we will all be dead. This, coming from the Finance Minister said to be the father of liberalisation of the Indian economy in 1991, ( in which he had quoted the words and thoughts of John Maynard Keynes) was enough for the Anand Sharmas and Surjewalas of the Congress party (whose knowledge of Economics is too superficial to be commented upon ) to perform their cameos in Parliament.

  4. I sometimes wonder that if the entire population of the country, including the opposition and their handful of supporters concede the 2019 election to the NDA today, they will stop trying to sell us these piles of horse manure whenever someone questions them about results. I’d be willing to make that compromise.


Please enter your comment!
Please enter your name here