Log in or Register
You are here: Home  /  Books & Strategies  /  Is the S&P 500 Index Effect Gone for Good?

Is the S&P 500 Index Effect Gone for Good?

July 12, 2010  |  asues

I recently finished the excellent book, Beyond the Random Walk: A Guide to Stock Market Anomalies and Low-Risk Investing by Vijay Singal. The book details 10 pricing anomalies in the financial markets that have proven to generate abnormal (market-beating) returns.

I’ve always been very interested in the published academic papers that detail financial trading strategies, but often find myself getting bogged down in the dry ‘academic-speak’ and equations when I try to decipher them.

Singal does a great job of synthesizing these academic studies into real, actionable investment strategies.

Index Effect

One idea that caught my eye was the S&P 500 Index Effect.

Like many investors, I compare my financial returns to the S&P 500, a group of 500 leading companies in leading industries pulled from the U.S. stock market – many consider it to be the best proxy for the total market.

The individual stocks within the S&P 500 change throughout the year – corporate restructurings, market cap fluctuations, and M&A activity all affect the latest group of stocks in the index.

Standard & Poors usually announces the change five days before the stock is officially added to the index.

Historic Returns

For a variety of reasons – such as index funds needing to buy or sell to match the S&P – individual stocks added to the S&P 500 list have experienced an immediate price jump after the announcement, with additional gains to follow until the stock officially joins the index.

According to a study of 224 additions to the S&P 500 from 1989-2000, the abnormal return for this trading strategy was 3.1%.

With an average trade length of only 6 days, this trading strategy seems to offer very exciting annualized returns, so I decided to test the recent results.

Analysis

For this ‘S&P 500 Additions’ strategy, I went back and recorded every valid addition & deletion to the S&P 500 from 2009 to the present. I gathered the results separately for two types of events:

  • Brand new additions
  • Additions from another S&P index (i.e. S&P MidCap 400 or S&P SmallCap 600)

The results below:

It is a very small sample size, but the results are a bit disappointing.

Conclusions

New additions to the S&P 500 gained 0.13% vs. the market, while stocks that entered the S&P 500 by moving up from another S&P index list gained 0.05% relative to the market.

Has anyone traded on this strategy before? What was your experience?

Do you think the index effect will continue?

Part 2 – Coming Soon!

asues

By asues

July 12, 2010

Print or Share With Friends

The Discussion
July 12, 2010 at 4:01pm

I don’t see what the strategy would be. The outperformance for each is .13% and .05%, respectively.

To the extent there is any disproportionate performance, I would say that it is simply what has happened in the past. There is no guarantee or likelihood that history will repeat itself. Remember what they say: Past performance REALLY REALLY REALLY does not guarantee future results.

I would not base any type of trading strategy on this. Have you read Joe’s book?

July 12, 2010 at 7:38pm

Scott,

Thanks for the comment. I agree, my article shows no outperformance for firms added to the S&P 500 over the past year and a half.

However, historically, there HAS been a significant outperformance, to the tune of 3.1%, between the announcement and effective dates.

Singals’ book details a number of reasons why this abnormal return occurs, but the main reason lies with index funds. Index funds must closely replicate and hold the actual stocks in the index they are tracking.

With each addition, these funds must purchase the new security sometime between the AD and ED, to ensure the fund stays true to the index. Hence, the resulting price appreciation.

This effect has been documented in numerous academic studies over the years, and by Standard & Poors itself – I do not believe it is a random occurrence.

I’ve read Joe’s book and respect his writing and line of thinking immensely.The majority of my investments are in undervalued companies.

When markets are expensive, I supplement this strategy with special situation investing such as merger arbitrage, workouts, and yes, abnormal pricing anomalies.

July 12, 2010 at 10:12pm

Sues,

Thanks for your work, but I would be much more interested in seeing the performance of index deletions. Akin to spinoffs, investors typically sell when a company delists or gets kicked out of an index. Index funds and some large mutual funds are forced to sell, regardless of underlying value. I would think this would be more the area of value investors rather than a week long trading strategy for index additions. Maybe measuring the performance of index deletions beginning one week or one month after being deleted would be insightful. This is one area Seth Klarman likes.

Or perhaps this will be in part 2?

July 13, 2010 at 5:57pm
asues replied,

Hester,

In my view, value investors look for investment opportunities where the underlying security is mispriced. According to years of research and hundreds of additions, the S&P 500 index effect is permanent, with the higher price lasting well past the initial trading strategy.

By this logic, on announcement date, (before the huge influx of demand, recognition, higher liquidity, etc) I would argue that the security IS mispriced and presents an attractive opportunity.

However, recent data shows this effect has waned in recent years – I had hoped my article might provide material to debate on the reasons why.

P.S. – No spoilers on part 2 :)

asues
July 21, 2010 at 5:45am
Jason replied,

Hi asues,

I agree that in principle there may be a measurable effect, but it’s a bit like climate change, if we focus on one parameter we can start to convince ourselves that correlation is causation. No one disputes the fact that there is an effect, but depending too much on one parameter in any complex system with thousands of inputs and variables leaves huge question marks for me especially when we measure it over a very short sample period. I’m not saying the hypothesis in either case is wrong, I just require evidence that none of the other factors are having a significant impact.

For instance, if the money flow is generally out of the markets due to fear of sub prime debacles, European debt, or a BP oil spill can we still use an increase in buying volume from index funds as a rational for buying a newly listed stock. If fear really takes hold and we add to that a none primary impact, i.e. all the (none sophisticated) savers pulling their cash of index funds during fearful periods we risk becoming totally unstuck with the strategy as we’re now seeing volume go in the other direction.

Other than that what if a company does a radical share split on listing, massively increasing liquidity in the stock, or it issues a ton of new stock to fuel expansion at that time, or the CEO drops dead?

It did catch my eye when Berkshire was listed after the BNF purchase, as the effect seems to have supported their stock price since; that said, Berkshire is an exceptional beast.

All in all I’ve never seen as measurable an effect as an undervalued company being repriced over time to its value.

Jason
July 21, 2010 at 8:53pm
asues replied,

Jason,

There are certain unusual circumstances that could severely change the returns of this strategy. However, don’t you always assume this risk in ANY investment? i.e. the CEO dropping dead.

At least in the strategy above, your risk is limited to the 5 day window between the AD and ED.

There are certainly other possible explanations for the index effect including certification, investor recognition, increased liquidity, etc, that could lead to the abnormal price appreciation.

In any case, it seems to be there, although the effect appears to be weakening.

asues
July 22, 2010 at 7:08am
Jason replied,

asues,

Absolutely correct, that’s why I demand a very comfortable margin of safety on ANY investment and I just don’t see it here. If I believe the CEO is the sole reason that a company is worth $10 and not $5, and that CEO has a recent history of poor health, I’ll only buy with a margin of safety on the $5.

The five days figure is a bit of a misnomer, I’d prefer to think in terms of odds, I don’t know what they are that something might go wrong in those five days before listing, but I could guarantee you that if you asked me to spin plates on several poles for five days there is a 100% chance, I’ll drop the lot. So I need to know what the odds are.

As for the weakening effect you mention, it could be there are a lot of quant models that include just this kind of thing, and once they do, the effects start to get erased by the large volumes traded by the institutions that employ them.

Good luck with it anyway, looking forward to part 2.

Jason
Join The Discussion