Arun Kumar: The real cost of demonetisation is ₹9.4 lakh crore

Excerpts from economist Arun Kumar’s book Demonetisation and the Black Economy.

Impact on Economic Growth

The government’s argument that demonetisation has had no impact on the economy because growth has been robust belies its own earlier exhortations to the public to bear the pain for long-term gains. As discussed above, there was pain for major sectors of the economy, primarily in the unorganised sectors. The government is quoting the quarterly rates of growth to prove its point. There are problems with this. No data are available for the non-agriculture component of the unorganized sector on a quarterly basis. The numbers for the various parts of this component are based on data which comes from benchmark surveys conducted once every few years. In the meanwhile, some indices from the organised sector are used to project the data for these sectors.

Measuring the Quarterly Rate of Growth of the Economy

How is GDP growth calculated? It is based on data from the various sectors and sub-sectors of the economy. Data are needed from the sub-sectors, and because of the specificity of each of them a different method is used for calculating the growth rate for these sub-sectors. The methodology is changed from time to time as lacunae are found and fixed after consultation, and this practice is usually not questioned by analysts. However, the actual data are not usually available immediately so that some indicative data are used. Thus, immediately, only ‘estimates’ of growth can be made. These are revised from time to time and therefore they are called provisional, advance and revised estimates. One must ask whether the methodology devised for normal times can be applied when the economy has experienced a big shock. Surveys by manufacturers, business associations and other entities.

The real cost of demonetisation is ₹9.4 lakh crore

in December 2016 and January 2017 show that after November 8, 2016, employment, production and investment were hit hard in a whole range of sectors. Was this taken into account in the estimation of GDP growth? The government document, ‘Methodology For Estimating Quarterly GDP’, says, ‘The production approach used for compiling the QGVA [Quarterly Gross Value Added] estimates is broadly on the benchmark-indicator method.’ The document adds, ‘A key indicator or a set of key indicators for which data in volume or quantity terms is available on quarterly basis are used to extrapolate the value of output/ value added estimates of the previous year.’ It also says, ‘In general terms, quarterly estimates of Gross Value Added [GVA] are extrapolations of annual series of GVA.’ All these statements point to the use of the ‘benchmark-indicator’ and extrapolation of the ‘value of output/value added estimates of the previous year’. What the quotation from the government document means is that for the unorganised sectors for which data are not immediately available, some indicators are used for instance, the ratio of production in organised industry to that in the unorganised sector, along with how much employment is being generated. But when the economy is severely impacted, can the same benchmark indicators be used as in earlier years? Indeed not, because the ratio of the organised to unorganised sectors changed on 9 November 2016. So, how can the projection from pre-November 8 2016 be valid for the period immediately following? Projection from the previous year (2015-16) would be even less valid. As already argued, the impact of shortage of cash was different for the organised and unorganised sectors, and therefore the impact on their output was also different. That is why the ratio changed.

The government’s press note giving the growth figures for GVA says, ‘GVA from quasi corporate and unorganised segment has been estimated using IIP [Index of Industrial Production] of manufacturing. But, as argued above, the IIP reflects the growth of the organized sector. So, in the changed circumstances after November 9 2016, it cannot be used to estimate the unorganised-sector production. The incorrect assumption is that a sub-sector that is declining is taken to be growing at the same rate as the organized sector. The government’s press note adds, ‘IIP from manufacturing sector registered a growth rate of (-) 0.5% during April-December 2016-17.’ But the data show that the manufacturing sector grew at 7.7%. Is this an overstatement of the organised sector growth? If so, it is doubly wrong to use it to estimate the unorganized-sector growth. It is due to this series of mistakes in the methodology that the official data showed a 7% rate of growth. Should a note of caution not have been inserted? The head of the government’s statistical department had argued that at present the impact of demonetization on the economy cannot be estimated. A lot more data are needed. If that is so, the official estimates are premature. As such, instead of relying on official data, one should go by the reports of what is happening on the ground.

The Likely Rate of Growth

To estimate the real rate of growth in the economy, one would have to create alternative scenarios. This can be done in two ways. Overall, let us assume that for the first seven months till October 2016, the economy grew at the rate given by the government statistics, namely, 7.7%. In Scenario 1 in Annexure 19, it is assumed that there was some impact of demonetisation on the organized sectors, but not too drastic. So, in the period November to January when there was a cash crunch, this sector did not grow. As the cash crunch declined, this sector is assumed to have grown at 2 per cent in February-March 2017. Agriculture is assumed to have grown at the rate given by the government, namely, 4%. The final assumption is that the unorganized non-agriculture sectors collapsed at -60% in the November-to-January period and then at -30% in the February-March period.

When all this is factored in, the table shows that post-November 2016, the economy declined by 14 per cent, and for the year as a whole it declined at 1.2%. In Scenario 2, all the sectors were taken to be affected less than in Scenario 1. The organised sector was taken to grow at 2% in the November-to-January period, and by 5% in February-March. Agriculture was assumed to grow at the same rate as given by the government. The unorganised non-agriculture sector was taken to grow at -50% in the November-to-January period and at -20% in February-March. So, in this milder scenario, in the post-demonetization period, the economy declined at -9%, and for the year as a whole grew by 0.55%. These growth figures are far from the government-announced 7% rate of growth for 2016-17. The government takes refuge in the corroboration received from international agencies like, the IMF, the World Bank and the ADB, supporting its contention that the economy grew at around 7% in 2016-17. But, this is of little use since these agencies do not collect independent data and rely on what the government provides them. There is also an understanding that these agencies consult the government before releasing their results. So, what they put out cannot be considered as independent corroboration of the government’s position.

In brief, the shock to the economy was that up to October 2016, the economy grew at 7.7%, and after that, under the ‘better’ scenario, it declined at 9%. The overall rate of growth for the year fell by 7.1%. With a GDP of ₹135 lakh crore in 2015-16, this amounts to a reduction of GDP of ₹9.4 lakh crore. This is the real cost of demonetisation. And this does not include the decline in the production in the agriculture sector due to the shock.

Excerpts have been taken with permission from Penguin Random House