From April 2nd, 2018, the NIFTY will look different. Bajaj Finserv, Grasim Industries, and Titan Company will be included and Ambuja Cements, Aurobindo Pharma, and Bosch will be excluded. [link]
This is not a rare or ad-hoc phenomenon, but a semi-annual ritual conducted by the organization that owns the NIFTY index, the National Stock Exchange, effective every April and September.
Here’s what the index methodology document says about index reconstitution:
The index is reconstituted semi-annually considering 6 months data ending January and July respectively. The replacement of stocks in NIFTY 50 (if any) is generally implemented from the first working day after F&O expiry of March and September. In case of any replacement in the index, a four weeks’ prior notice is given to the market participants.
Additional index reconstitution may be undertaken in case any of the index constituent undergoes a scheme of arrangement for corporate events such as merger, spin-off, compulsory delisting or suspension etc. The equity shareholders’ approval to a scheme of arrangement is considered as a trigger to initiate the exclusion of such stock from the index through additional index reconstitution.
As part of the semi-annual reconstitution of the index, a maximum of 10% of the index size (number of companies in the index) may be changed in a calendar year. However, the limit of maximum 10% change shall not be applicable for any exclusion of a company on account of scheme of arrangement as stated above.
NIFTY constituents over the years
Graphic below shows changes in the NIFTY over the last nine years (click on the image to show fullscreen)Green cells are new additions in that cycle, Red cells were removed in the subsequent cycle. So the number of reds in a column should correspond to the number of greens in the next column except when the NIFTY went from 50 to 51 stocks in April 2016.
Price momentum drives changes to the NIFTY
I looked at price movements of inclusions and exclusions to the NIFTY from Sep 2008 to Sep 2017 to spot discernible patterns.
Being added or dropped from the NIFTY looks a lot like selection for most team sports. The average inclusion to the NIFTY delivers 44% returns in the year leading up to being included which is 4x the NIFTY itself. Compared to that, the candidates chosen not only underperform the NIFTY, but show negative one-year returns.
Reinforcing the “momentum is the key-driver in NIFTY changes” message, 27 (out of 35) inclusions had delivered positive and NIFTY-beating returns. Among the exclusions, only 10 (out of 32) had positive one-year returns and only 6 had beaten the NIFTY before being “dropped”.
Over the short-term, Inclusions barely outperform exclusions
The difference in returns between inclusions and exclusions narrows significantly once index changes are made. Average returns of both are comparable to the NIFTY. Median returns show exclusions struggle for at least a year after they are dropped from the index.
Once added or dropped from the index, the percentage of stocks delivering positive returns and those beating the index converge, to almost 50-50.
Also consider, 53% of exclusions deliver higher one-year returns than the year before after being excluded while only 31% of inclusions surpass the returns that got them included in the first place, suggesting classic reversion to the mean in play.
What if we take a longer-term view
Chart shows median annualized returns from the time stocks are added or removed from the NIFTY.
The exclusions narrow the gap over a three-year period but the delta again widens over five years. Note that since longer-term returns are only available for a subset of the stocks considered for 1-year returns.
At the start of this exercise, the contrarian in me would have loved to conclude that stocks that fall out of favour and get excluded from the NIFTY then proceed to beat the pants off those that get added. To that end, it is tempting to eliminate a lot of the underperformance of the exclusions that come from a small group of stocks that saw massive erosion due to corporate governance or leverage issues; Suzlon (5 yr: -72%), Unitech (-92%), JP Associates (3yr: -75%), RCOM (-55%), RPOWER (-60%). But then that would just be cherry picking and against the spirit of good data analysis.
- Stocks get added or excluded from the NIFTY based on near-term price momentum
- Stocks that rise sharply and get added to the index, then proceed to deliver relatively moderate returns over the next year
- Stocks that perform poorly and get dropped from the NIFTY, improve performance relatively in the short term but overall still deliver inferior returns over the next year
- This outperformance of inclusions over exclusions narrows over the longer term of three and five years but tends to persist
Overall, selling stocks being dropped from the NIFTY to replace them with stocks being added makes sense as an investing strategy.