Demonetisation: What We Know Three Years On...

Three years after demonetisation, what does research and data tell us about the impact of the shock decision?

People wait in line outside a bank branch in Delhi, India. (Photographer: Anindito Mukherjee/Bloomberg)
People wait in line outside a bank branch in Delhi, India. (Photographer: Anindito Mukherjee/Bloomberg)

Three years after the Indian government decided to scrap nearly 86 percent of the country’s currency overnight, researchers and data scientists continue to study the impact of the unparalleled ‘natural experiment’.

These are all questions that continue to be studied, with some answers starting to emerge with time.

Impact On Economic Activity

Among the studies initiated to assess the impact of demonetisation on India’s economy is a December 2018 working paper authored by economists Gita Gopinath, Prachi Mishra, Abhinav Narayanan and Gabriel Chodorow-Reich for the National Bureau of Economic Research.

Using RBI data, the authors constructed a metric called the local area demonetisation shock—the ratio of post-demonetisation to pre-demonetisation currency in an area. The top 10 percentile of the districts were those that had cash equal to 70 percent of their pre-demonetisation level. The bottom 10 percentile had currency equal to only 33 percent of its pre-demonetisation level. Districts which were relatively ranked in between, had currency equal to 42 percent of the pre-demonetisation level.

The authors then went on to assess the adverse effect on real economic activity using a new household survey of employment and satellite data on human-generated nightlight activity to measure demonetisation’s effects at the district level.

The study found a difference in output of roughly 4 percentage points between districts at the 10th and 90th percentile of the demonetization shock in the period immediately after the announcement.

  • Districts that experienced more severe demonetisation shocks had much larger contractions in ATM withdrawals.
  • Economic activity, as measured by employment rates and nightlights, fell in these areas relative to areas experiencing smaller shocks.
  • These areas adopted alternative forms of payment such as cards, e-wallets, informal lines of credit or use of old notes. That is evident from the fact that the difference in output is far less than the difference in the amount of currency replaced.
  • Deposits increased more and credit fell in areas with higher shocks as households could not withdraw money from their bank accounts.
More broadly, the study pointed to a contraction in employment and nightlights-based output of 2 percentage points due to demonetisation, the effects of which dissipate over the next few months.
Demonetisation: What We Know Three Years On...

Impact On Financial Profile Of Household Sector

A key question following the demonetisation experience was whether people’s financial savings habits would change. Would they see the need to hold on to more cash, given the deposit withdrawal restrictions they faced during the phase when new currency was in short supply?

The latest reading of National Account Statistics throw up some interesting, if inconclusive, findings.

First, the data shows that in the year immediately after demonetisation, people did appear to be holding on to more cash. As such, the share of currency in gross financial savings of households jumped to 25.2 percent, a sharp increase from the levels seen in years prior to demonetisation. Hindustan Times first reported the data last week.

Did that preference for cash holdings continue even after 2017-18? Data for subsequent years will tell.

Another indicator, though, which gives a share of the share of cash in the economy’s broader money supply throws up a different conclusion.

The share of currency with the public as a percentage of M3, or broad money, has remained close to the range seen before demonetisation. At present levels, currency with public accounts for about 13.54 percent of broad money, shows RBI data.

The National Accounts Statistics also throw up another interesting finding.

The years 2016-17 and 2017-18 saw financial liabilities of the household sector rise sharply. The household sector, as defined by the Ministry of Statistics, comprises individual households, agricultural households, unincorporated enterprises and private trusts.

Financial liabilities as a percentage of gross financial savings rose to 32.6 percent in the year of demonetisation compared to 25.8 percent a year ago. This further rose to 39.6 percent a year later. To be sure, this increase in leverage could have also very likely been linked to the increased consumption seen in the economy over the last few years.

Did It Push Digital Payments?

A often stated objective for demonetisation was to encourage digital payments and reduce the use of cash in everyday transactions.

However, currency in circulation as a percentage of GDP moved back towards its long term trend quite quickly after demonetisation. Prior to demonetisation, currency in circulation as a percentage of GDP was just about 12 percent. Over a longer period of time, this ratio has been between 11-12 percent.

As soon as reprinting of currency was completed, this share of currency in GDP started to move back up. For the full year FY19, it settled at 11.257 percent.

However, there are other indicators that do suggest an increased usage of non-cash methods of payments, particularly as a share of consumption expenditure, perhaps across urban consumers.

Spends on credit and debit cards at point-of-sale terminals as a percentage of personal final consumption expenditure have nearly doubled since before demonetisation, shows data from the RBI when compared with National Accounts data.

Did It Reduce Counterfeit Currency?

Another reason cited for demonetisation, albeit a small one, was that counterfeit currency in circulation had risen. Judging the extent of counterfeit currency is difficult. The only data available is the number of counterfeit notes detected, put out by the RBI but this only tells the story of what has actually been detected as opposed to what exists.

Still, going by that data, the extent of counterfeit notes detected have gone down following a spike in the year of demonetisation when banks were screening all the notes being deposited.

In 2018-19, 3.17 lakh counterfeit notes were detected shows the RBI’s latest annual report.