 "Image via Shutterstock")
The Diwali effect regularly impacts October and November data. Consumption rises during the Dussehra-Diwali period due to bonuses, the traditional annual buying, etc. And, with holidays, manufacturing falls. It is hard to deseasonalise this effect, since the festival season "wanders" across two months. When year-on-year comparisons are made, these effects lead to exaggerated trends. Take any three-year period. Let's say Dussehra (October) and Diwali (November) fall in different months in year zero. Then, in year one, both festivals fall in October. In year two, Dussehra (October) and Diwali (November) again fall in different months.
In this case, manufacturing will shrink in October of year one, compared to October of year zero because October in, year one has more holidays. However, manufacturing will jump in November of year one, as there will be fewer holidays compared to November of year zero.
There will be an inverse effect in year two. Manufacturing will rise in October of year two, since there are fewer holidays compared to October of year two. Manufacturing will also show a fall in November of year two, since that has more holidays in comparison to November of year one.
This pattern will repeat regularly. The Chinese have similar issues with their New Year celebrations. They handle the deseasonalisation by clubbing two Western months. Maybe India's statisticians should adopt a similar system.
As it happened, 2013-2015 has seen this effect play out. In 2013 (year zero), Dussehra and Diwali were split across October and November. In 2014 (year one), October hosted both Dussehra and Diwali, while Dussehra was in October 2015 (year two) and Diwali was in November 2015 (year two).
The usual indicator for manufacturing activity is the Index of Industrial Production (IIP). The Index is anyway volatile. A classic Diwali effect was seen across 2013-2015. The year-on-year (y-o-y) change for the IIP of October 2014 (year one) over October 2013 was minus 2.7 per cent, the first apparent contraction in six months. October 2014 had more holidays.
After this, the y-o-y change for IIP of October 2015 (year two) over October 2014 (year one) was 9.8 per cent - and that was a five-year high. Of course, October 2015 had fewer holidays. The y-o-y change for IIP of November 2015 (year two) over November 2014 was minus 3.2 per cent. November 2015 had more holidays than November 2014.
By clubbing October and November 2015, we can average the IIP values and compare those to a similarly averaged period of October-November 2014. This averaging seems to lead to a small positive change in IIP in October-November 2015 combined, over October-November 2014 combined. This seems to indicate growth in manufacturing slowed but it was still growing in October-November 2015.
There are further worries, in that the Manufacturing PMI (Purchase Managers' Index) for December has a negative value, at under 50. This suggests contraction in December. Again, we would have to allow for the Chennai floods of December, which crippled a major industrial hub.
The market reaction to the data doesn't seem to have adjusted for these nuances. There was a temporary surge when the October 2015 IIP data was released in December 2015. There was a crash when the November 2015 was released last week. Admittedly, the November IIP came through in a week when there was a crash in energy contracts across global commodity markets. But, sentiment got noticeably worse when the IIP data was released.
IIP data don't correlate much with corporate results. However, PMI tends to correlate quite well. Anecdotally, most corporate news flow and analyst projections suggest there has indeed been a slowdown in manufacturing. This also gels with consensus estimates that India Inc has slashed investment spending. This might lead to collapse in the capital goods segment. Let's assume, for argument's sake, that the third quarter results will indeed confirm that manufacturing took a hit in October-December 2015. If that is so, share prices could fall across manufacturing sectors. We have seen signs of such an impending sell-off before results have come through.
The key question: Is this the bottom of the cycle or near the bottom of the cycle for manufacturing? If it is, lower share prices will be an opportunity for bargain hunters. The most beaten-down segment is probably capital goods -and, by contrarian logic, capital goods could have the biggest upside. There seems to be some consensus that 2016-17 will see a pick-up in growth.
