The IIP-Diwali-market connection

The Index of Industrial Production (IIP) numbers are depressing. September 2011 wasn't a boom month but September 2012 suffers in comparison, falling 0.4 per cent point-to-point. Given large error margins and corrections whenever IIP data is revised, I tend to ignore quick estimate changes of less than plus/minus 1.5 per cent.

The data-gathering process is inefficient and the ministry of statistics and programme implementation often leaves numbers unchanged for months. The best guess is that there wasn't a discernible pick up in activity in September 2012 versus September 2011.

IIP is never a smooth series due to seasonal effects and “reporting effects”. The series always has a March peak because a lot of activity is reported at the end of the financial year. Apart from that, monsoon and agro-activity also show up in changing patterns.

There is a Diwali effect to be considered along with seasonality. The Q3 activity is, in part, influenced by festive season consumption. Since Diwali shifts around (due to being a lunar calendar event), that creates an unusual effect.

The Diwali effect is lagged. Instead of a peak occurring before Diwali, which is when businesses are gearing up, the index usually surges post-Diwali. This may be some sort of reporting lag but it's consistent over the past seven years.

For example, last year October 26 was Diwali, and the IIP fell through September 2011 (index value 164) to October 2011 (158) picking up in November (167) and December (180). Similarly in 2010, when Diwali occurred on November 5, 2010, the IIP fell through October 2010 (167) and November 2010 (158) , jumping in December 2010 (175).

If this behaviour holds in 2012, we may see a further fall in October 2012 IIP and maybe a muted November 2012, coupled with a jump in December 2012. What sort of effect will that have on the stock market?

There is little apparent correlation between corporate sales, profitability and the IIP index numbers. This is probably due to poor data-gathering methods. IIP can be safely ignored when it comes to a connection with quarterly results.

However, there is a positive correlation – albeit not a very strong one, between point to-point (P2P) changes in IIP quick estimates and between share prices. As noted above, quick estimates are prone to large corrections and these are lagged by two months.

But, the market tends to move in the same direction as the changes in the P2P. If this pattern holds and if the IIP conforms to its normal Diwali effect, we could see this November correction continue for a couple of sessions. We may also expect a short correction in mid-December, if weak October IIP quick estimates are released.

The author is a technical and equity analyst