Q: On demonetisation, its costs and benefits a year on:
A: I am going to make an objective assessment of where we are. Whenever you have large interventions in the economy, there is a trade-off: there is always an immediate, short-term cost, and there are medium-term benefits. Firstly, we must appreciate that GDP growth was slowing well before demonetisation, so this notion that the entire slowdown is only because of demonetisation or GST is not borne out by the data. As we have written extensively, India benefited from the huge terms-of-trade shock from lower oil prices — that was a one-time windfall for the economy in 2015-16, so despite it being a drought year, growth picked up from 7% to 8%. But when oil prices stabilised, it was inevitable that GDP growth would begin to slow. That, in conjunction with the fact that policymakers were pushing hard on the deleveraging, meant that between March 2016 and September 2016, GDP growth had already slowed by 200 basis points. It was a pretty sharp slowing. So, in a way, one has to separate how much of that slowdown would have continued, and how much was incrementally added by demonetisation.
Looking at the government’s own numbers, the CSO every January puts out its advance forecast for growth in that year, and last year, they actually said that we are basing our advance estimate on data that stops in October. So they were using all the pre-demonetisation data to say what GDP growth in 2016-17 would actually be. For us economists, this is fantastic — this is the counterfactual, this is what GDP growth would have been, absent demonetisation. You update that for the eventual strength of agriculture and exports, which are largely exogenous, and you get the counterfactual. You compare that to the actual printed number, and that’s your delta. Our own sense was, somewhere between 0.8% and 1% of full year GDP was the slowdown because of demonetisation. That was one near-term cost.
There was a second, less obvious cost. See, the whole idea of these reforms is to push small and medium enterprises into the formal economy, and that process is necessarily disruptive. So, one puzzle over the last year has been that even as GDP growth has slowed, imports have been surging. This is not oil, this is not gold, these are manufacturing imports — growing at 18% a year for the last 8 or 9 months. These imports were contracting the year before, how do you explain strong, broadbased manufacturing import growth when domestic demand is slowing?
The way we see it is, both demonetisation and GST imparted an adverse supply shock, where domestic supply chains, which involved SMEs, got disrupted temporarily, and therefore, rather than those domestic inputs, larger firms were relying on imports. So the second manifestation of demonetisation was the fact that the current account deficit quadrupled — in the quarter in which growth slowed, the current account deficit went from 0.6% of GDP to 2.4%, 80% of which was manufacturing imports. For me, those are the two near-term costs.
Gems and jewellery, electronic items, some capital goods, leather, textiles — the hypothesis is that a lot of these would have been manufactured by SMEs and the informal sector. And because those supply chains got disrupted, you had to temporarily rely on imports. That’s gone on for three quarters now, the hope is that this is a temporary supply shock, and therefore, at some point, imports will come down. But till the last data point in September, imports were still very strong.
But to be fair, there were benefits as well.
As we proceed to the second or third anniversary (of demonetisation), there are three objective benchmarks to measure: cash in the economy, the quantum of digitisation, and the tax base. On cash, the right comparison is, what would the outstanding stock of cash in the economy have been today absent demonetisation? Every year, it was growing by between 10% and 15%, the day before demonetisation, this was about Rs 17.5 lakh crore. Given that the typical increase every year is Rs 1.8 lakh crore to Rs 2 lakh crore, if demonetisation hadn’t happened, by this October the cash in the economy should have been Rs 19.3 lakh crore — but today it is Rs 16 lakh crore. That is almost a 20% reduction in cash and, I think, a big benefit.
On taxes, you are seeing that new tax filers grew 26% in 2016-17 over the previous year, much higher than the previous year. So let’s be patient, there was a short-term cost and there will be medium-term benefits; just because we can’t quantify those benefits does not mean that they don’t exist. But they are already beginning to show up.
Q: On the possibility of growth numbers not accounting for small manufacturers, who were hit badly by demonetisation:
A: A famous governor once said that in India not only is it hard to predict the future, it is hard to predict the past as well, because of sharp data revisions. I think when the annual survey of industry comes out, the last year’s data will be revised, and it is very likely that GDP growth will be revised down. This is because we essentially use formal sector indicators to extrapolate to the informal economy. That relationship clearly broke down during demonetisation, when the formal economy was a very bad proxy for the informal economy. So it is very likely that about 2 or 3 quarters from now, the 2016-17 numbers will be revised down retrospectively. There will be multiple revisions, but in the first pass we will have a good sense. Again, the key is to separate what slowing would have happened even without demonetisation, because to the extent that we were already on a slowing path, it is the delta that matters. [Link]