One year and nearly four million Google search results later, the “great Indian demonetisation” of November 8, 2016, continues to be much discussed and debated.
The sheer scale of this economic experiment is unparalleled in contemporary economic history. It attracted wide global attention and curiosity. No other economic policy decision in India has provided as much fodder for people across the entire spectrum — from scholars to satirists — as this decision to extinguish nearly 90 percent of all currency in circulation overnight. Reactions spanned from awe to bewilderment.
A year later, tempers have cooled and the Lakshman Rekha separating demonetisation supporters and opponents has thinned amid a tacit consensual acknowledgement that perhaps the grand demonetisation experiment did not quite work out the way it was envisaged.
I was an active participant in the debates and discussions on demonetisation. Within a month of the announcement, I had written four articles in various publications critiquing demonetisation using data and analogies.
- To me, the introduction of a new Rs 2,000 note was a sure sign that this experiment was not about retiring high-value currencyas the classical definition of demonetisation would posit.
- I then pointed out through an analysis of the Prime Minister’s speeches how the goalposts for demonetisation had changed dramaticallyfrom black money to cashless/digital economy within three weeks of the announcement.
- I followed up with an article arguing how if indeed a digital economy was the goal, then achieving it through harsh coercive meansis unbecoming of a mature liberal society.
- Finally, using empirical evidence I showed how the very reasons cited by the Government of India for demonetisation — inordinately large share of high-value currency in circulation, a large share of counterfeit notes and high-value currencies’ role in terrorism financing — were all plain wrong and that the Governor of the Reserve Bank of India had failed in his dutiesto right these wrong beliefs of the government.
A year after the exercise, there is a lot of commentary on the impact of demonetisation on GDP, monetary variables, financial inclusion, formalisation and digitisation of the economy and so on.
In this piece, I have restricted myself to attempting to quantify the impact of demonetisation on economic output. To be sure, any counterfactual methodology to estimate ‘what could have been’ is not a perfect science and hence it will always be impossible to accurately estimate any impact of demonetisation. With that caveat, one can always contextually quantify the impact of a policy decision through some reasonable assumptions.
The impact of demonetisation on headline GDP numbers can be very misleading.
GDP is at best a crude estimate of economic activity, which in itself is a complex interplay of various factors. Disturbing one economic variable can have an attendant impact on various others which is impossible to model quantitatively, which is why economists rely on what is called the ‘general equilibrium theory’, i.e. assume a certain state of equilibrium to measure economic activity.
I am reminded of a story (apocryphal?) of India’s first Prime Minister Jawaharlal Nehru’s visit to a coal mine in erstwhile Bihar. Nehru was told that coal mining is down significantly. When he asked why, he was told that it was because there were not enough train carriages to carry the coal from the mine to the factory. Upon inquiring why there were not enough carriages, he was told there was not enough steel to manufacture more carriages. He was then told that there was not enough steel because steel plants did not have enough power to operate their factories at full capacity. Exasperated, when he asked why there was not enough power, he was told rather cheekily, it was because there was not enough coal to feed the power plants!
Now imagine trying to quantify the impact of a singular policy intervention on the broader economy amidst this complex interplay of economic activities. This is why analysing the impact of demonetisation using headline GDP numbers is too simplistic and misleading.
Demonetisation was first and foremost a liquidity shock that sucked out cash from the economy.
Such a liquidity shock impacts certain sectors of the economy much more than the rest.
It is possible that government spending which is a critical variable in the calculation of GDP was not impacted nearly as heavily as say personal consumption. Or one could posit that perhaps imports were not impacted as severely as say exports. The structure of India’s economy is one that is dominated by small and medium producers who play a critical role in the value chain of the few large producers. Perhaps the three sectors that are most prone to liquidity impacts are agriculture, manufacturing and construction. These sectors tend to be heavily cash dependent and have fragile working capital cycles. Together these three sectors have typically accounted for nearly half of nominal gross value added (GVA) in the economy.
More importantly, these three sectors account for more than three-quarters of all employment in the country.
The Central Statistics Office (CSO) provides sectoral GVA data every quarter for eight different sectors. Since these are the three most important sectors from the perspective of a liquidity shock as well as employment, it is possible to analyse quarterly GVA growth rates of these three sectors combined, to come up with a rough ‘before and after’ analysis of the economic impact of demonetisation. For the purposes of this analysis, we use nominal GVA data and not real, since comparisons of real GVA will also have to account for the deflator impact.
Between Q4FY16 to Q3FY17, the sectors of agriculture, manufacturing, and construction grew at 8 percent (year-on-year). In the quarter immediately after demonetisation (Q4 FY17), the nominal GVA growth in these sectors combined fell to 6.7 percent. In the subsequent quarter (Q1FY18) growth fell even further to 2.5 percent.
It is thus quite evident that demonetisation has disrupted the steady 8 percent nominal growth rates of these sectors combined. That is, ceteris paribus (other things equal), perhaps these sectors could have sustained their 8 percent growth rates.
Put another way, the total nominal GVA of these three sectors combined in Q4FY16 and Q1FY17 was roughly Rs 28 lakh crore. If we assume this would have grown at a steady state of 8 percent annually, then the combined GVA of these sectors in Q4FY17 and Q1FY18 would have been roughly Rs 30 lakh crore. Instead, the total GVA of these sectors was only around Rs 29 lakh crore. So, a fall in nominal GVA growth rates in these sectors from 8 percent to 6.7 percent and further to 2.5 percent is the quantum of loss of economic output, measured in GVA terms. This translates to an absolute amount of Rs 89,000 crore.
In other words, a very rough ballpark estimate of the impact of demonetisation in agriculture, manufacturing and construction in the two quarters post-demonetisation is roughly $15 billion.
This then squares up with other narratives on job losses post demonetisation since these sectors account for almost all of India’s employment.
Of course one can quibble and argue that it is erroneous to presume that these sectors would have continued their 8 percent growth rates, had there been no demonetisation. But purely for the sake of a contextual quantification of the impact of demonetisation in economic output, this methodology is still better than merely imputing from headline GDP growth rates.
“It is not my case that there have been no unintended positive consequences of demonetisation.
The recent announcement of capitalisation of public sector banks through recapitalization bonds is a clear example of one benefit of demonetisation since the recapitalisation just converts excess deposits in banks received from demonetisation into equity for banks.
But if one were to move away from analysing the economic impact of demonetisation through simplistic headline GDP numbers, this methodology of sectoral nominal GVA impact gives a more nuanced estimate. And that estimate is a $15 billion cost of demonetisation in lost output in three sectors that employ the most number of people.
Video url: https://youtu.be/tMG3Rcbxqlc
Published on 08 November, 2017 in Bloomberg Quint