Two weeks after Prime Minister Narendra Modi announced that Rs 500 and Rs 1,000 notes would no longer be legal tender, the cash chaos continues.

This is not surprising, since 86% of currency in circulation and 55% of liquid money (currency plus demand deposits) were rendered useless, and the banking system has been over-stretched in disbursing or exchanging notes.

As of November 18, only Rs. 1.36 lakh crore had been exchanged or withdrawn – less than 10% of the cash that was made worthless.

After the compliments that were bestowed in the immediate, innocent aftermath of the announcement, questions are being asked about the entire project. From a one-way flow of praise, it has become a two-way exchange of arguments.

Since operational challenges in printing cash, reconfiguring ATMs, replenishing cash in remote areas, etc, have accentuated the costs of the decision, a common refrain has been: this was a good decision that was badly implemented by bureaucrats and bankers.

Disproportionate shock

Was this a good policy decision marred by poor implementation?

There are certain features of this decision that raise questions about the soundness of the policy itself, irrespective of the quality of its actual implementation.

This shock therapy is disproportionate to the scale of the problem.

In 2010, the World Bank published a study that estimated the size of shadow economy as a percentage of the official gross domestic product of countries. The study estimated India’s shadow economy to be 20.7% of the official GDP.

This is not a small number. But, to understand, we must compare. There is another country at 20.7%: Israel.

Countries like Spain, Portugal and Korea rank lower than India. Among the 98 developing countries included in the study, India is ranked 15th. This is much better than one would expect at India’s level of development. One would be hard-pressed to name a parameter on which we rank better.

So, while the black money problem is important, it is not as important as is often made out to be.

Further, earlier this year, a study by the National Investigation Agency and the Indian Statistical Institute, estimated that the total face value of fake currency in circulation is about Rs 400 crore.

Compare this with currency in circulation – more than Rs 17 lakh crore. This is not to say that we go easy on these problems, but that we do not panic and accept shock therapies to solve them.

The problem

The decision is not based on evidence and experience about the nature of the problem.

Black money is accumulated over a period of time, and stored in various instruments. Evidence shows that only a small portion of it is stored as cash.

A Ministry of Finance White Paper published in 2012 showed that, for the years between 2006 and 2012, cash seized during searches and seizures ranged from 3.75% to 7.3% of total undisclosed income for those cases. The Hindustan Times has reported that data from searches and seizures in recent years also attests to this fact.

Much more black money is likely to be stored in real estate, gold, foreign currency, offshore accounts, shares, etc. This means that banning notes would make those with black money lose small portions of their unaccounted wealth, and not lead to any significant punishment.

Black money creation depends on deterrence, which is enhanced when people think their likelihood of getting caught is higher. This decision does little to enhance deterrence, and is therefore not likely to affect black money creation, which is the source of the problem.

As far as fake currency is concerned, experience shows that elimination of fake currency does not require a sudden ban. It can be done through gradual replacement of old currency with new currency.

Image: PTI

Underestimating costs

The decision has grossly underestimated the expected costs.

Money is a store of value, but it is also a medium of exchange. Most people in India use cash for transactions, and not just to store wealth. This, coupled with the fact that 86% of all currency in circulation is in Rs 500 and Rs 1,000 notes, means that a sudden ban would disrupt legitimate economic activity. There are about 1.5 crore shops, but only 14.6 lakh point of sale devices that can handle credit or debit cards.

Since about 31% of those employed in India do casual labour (estimated to be more than 14.5 crore persons), mostly working on a daily wage for non-contractual employment, their livelihood is being hurt. There are many reports about manufacturing establishments and construction sites temporarily shutting down.

Further, this being the time when kharif harvest is brought to market and rabi sowing begins, the largely cash-based rural economy is paying a price.

This is also peak season for weddings, which are cash-intensive events.

Even before implementation began, one could have predicted the enormous costs it would impose. To make things worse, many of the costs are being imposed on people with low incomes. In any analysis of benefits and costs, such costs must be given a higher weight.

Expensive option

The government chose to adopt a very expensive course before exhausting much less disruptive options.

The prime minister has announced that the next policy step would be to go after unaccounted wealth stored in real estate. Given the cost-benefit considerations, perhaps that should have been the first step.

Often, we cannot just come up with a good solution; we discover it through a sound process of listing and evaluating alternatives in terms of their benefits and costs. I have blogged about this issue, and tentatively concluded that even before we knew about the delays in remonetisation, the costs were likely to outweigh the benefits, and less disruptive options should have been explored.

There are many far less expensive way of solving the problems of corruption. This article by Dr Vijay Kelkar and Dr Ajay Shah lists six of them.

Operational constraints

The decision did not take into account the constraints within which it would be implemented. Constraints are real, and no policymaker can afford to ignore them.

There are key operational constraints that government should have paid attention to – the capacity to print currency, the pace of ATM recalibration, the logistical limitations of high frequency replenishment of cash in remote areas, and the capacity of bank and postal offices.

For the first nine days after branches reopened, the exchanges and withdrawals amounted to Rs 15,000 crore per day. Even assuming that this pace improves by 50 percent, and if there is no constraint on supply of cash, it would take about a month more to re-monetise even half of the value of discontinued notes.

This constraint is most felt in small towns and rural areas, because of low density of branches and ATMs and bigger logistical problems.

However, the critical constraint is turning out to be the capacity to print notes. Due to the printing constraint, even if the branches and ATMs function perfectly, it may take more than three months to print enough cash to re-monetise half of the value of discontinued notes.

So, perhaps the banking system constraint may be smaller than the cash printing constraint.

There are other constraints as well. Even after the financial inclusion drives, crores of households are not banked, and many who have bank accounts do not have convenient and continuous access to banking establishments.

Since government has allowed exchange of old notes only once, the unbanked must take losses or try to open a bank account, which is very difficult under the present circumstances of capacity constraint in branches.

All these constraints were known well in advance, but were probably not given proper importance.

Image: PTI

Inflexible policy

A policy may face unforeseen problems and – especially one causing such large-scale disruption – must come with enough flexibility to allow government to change course.

On November 8, the prime minister said that ATMs would be functioning from November 11. It became almost immediately clear that this was not going to happen.

What was Plan B? None, it seems.

Since November 10, the day the branches opened, we have seen a series of inexplicable changes to limits and processes of exchange and withdrawal.

The policy is such that it inherently does not allow much flexibility, because its founding assumption is that any flexibility would lead to manipulation. The only escape clause is a rollback, which might be politically expensive.

So, the decision has placed the government in a tough spot.

Risks and uncertainty

No major policy decision comes with completely predictable benefits and costs. There are risks that can increase the costs or reduce the benefits. However, as long as the policymakers know the risks, they can put in place systems to manage them. For example, it was known that those with black money will try to launder their cash. This is a risk that the government is trying to manage.

While risks can be managed, uncertainty cannot be managed. This currency ban decision leads to a great deal of uncertainty. For example, conspiracy theories and frauds are thriving due to this large-scale disruption. Policy decisions should always try to minimise uncertainty, especially in decisions implemented on a large scale. This decision fails this test.

This is not hindsight. Even before implementation began, one could have foretold most of these problems. Poor implementation might be making things worse, but the policy itself is deeply flawed.

It appears that the policy was not informed by the kind of analysis that should go into such decisions. The real fight against black money creation or fake currency circulation involves slow, quiet and perhaps boring efforts to improve the administrative capacity of the government. And changes in policies that encourage people to create black money.

Suyash Rai is a senior consultant with National Institute of Public Finance and Policy. Views expressed here are personal.