United States presidential inauguration day

The United States presidential inauguration day used to be on 4th March, but in 1937 the Twentieth Amendment changed the date of inauguration day to 20 January. If that day is a Sunday, inauguration day is moved to 21 January.

Has this day had any significant effect on the stock market?

Let’s see.

The following chart plots the daily returns for the S&P 500 Index for inauguration day (ID) in the years from 1953 to 2009. Note: the chart only includes inauguration days for first terms (on the grounds that the market most likely knows what to expect with second-term presidents).

US president inauguration days (first term) [1953-2009] 1

As can be seen shares have been weak on inauguration days. Since the 1963 inauguration of Lyndon B. Johnson the S&P 500 has been down on every inauguration day.

The following chart plots the average daily returns for the S&P 500 Index for the trading day before inauguration day, the day itself and the day after.

US president inauguration days (first term) [1953-2009] 3

Since 1953 the average daily return for the S&P 500 on inauguration day has been -1.1%. For the day after ID the average daily return is 0.7%, so there does seem to be a partial relief rally afterwards.

Social Share Toolbar

Do European stocks follow the US on a daily basis?

Do European stocks follow the lead of the US market from the previous day? In other words if, say, the US market is down one day are European stocks more likely to fall in their trading session the following day?

To test this the following chart plots the daily returns of the S&P 500 Index against the corresponding daily return of the EuroSTOXX 50 Index for the following day.

Europe v US stocks_EuroSTOXX 50 v S&P 500 (n-1) [2000-2016]

There is a positive correlation here, but as can be easily seen it is a very weak correlation. And this observation is supported by the very low R2 of 0.05.

So the immediate answer to the question of whether European stocks follow the US is: only very slightly.

However, the following chart is interesting. This next chart plots the daily returns for the two indices as above, but this time it is the daily returns for the same day. In other words, this time the US market movements come after those in Europe.

Europe v US stocks_EuroSTOXX 50 v S&P 500 [2000-2016]

As can be seen, here the correlation is higher than in the above first case. The R2 = 0.3; which while not statistically very significant is quite a bit higher than in the first case.

So, this might suggest that it is the US market that follows Europe.

Is this the case?

Probably not. Rather it is likely to be a feature of the trading hours of the respective markets. The illustration below shows the trading hours for five exchanges.

NB. Strictly, UK and Swiss stocks are not in the EuroSTOXX 50 Index but the exchanges are included here for reference.

Europe v US stocks_exchange hours

The times referenced here are UTC – which are accurate at the time of writing (in May), but will be shifted one hour when countries switch to Daylight Savings Time. However, for the purposes of the discussion here the times are fine, because what we are interested in is the overlap of trading hours at the end of trading in Europe and the beginning of trading in New York each day.

As can be seen, each day there is an overlap of a couple of hours between the Paris and New York exchanges, and longer for Frankfurt and New York. Each day European markets can be active at their open in the morning (reacting to overnight developments – including US stock movements), then often these markets can tread water for a while waiting for the US market to open in the afternoon. The European markets can then take their lead from the US for the rest of their trading day.

The higher correlation seen in the second chart above is therefore probably reflecting this overlap period when European stocks are influenced by what is happening in the US that same day.

Social Share Toolbar

US stock market average returns by calendar day of the month

The following chart plots the average daily returns for the 31 calendar days in a month for the S&P 500 index over the period 1950-2015. (NB. This is looking at calendar – not trading – days of the month.) For example, since 1950, the S&P 500 index has on average increased 0.22% on the 1st day of each month.

S&P 500 average returns by calendar day of the month [1950-2015]

Observations

  1. The first day of each month has the highest average daily return for the S&P 500 index. Followed by the last day of the month.
  2. The worst average daily return has been on the 9th of the month
  3. As can be seen in the chart, the periods of strongest daily returns occur in the first and last weeks of months.
  4. Three particular phases of the month can be highlighted:
  • Phase 1 (1st-6th): the index sees positive daily returns
  • Phase 2 (18th-22nd): the index sees negative daily returns
  • Phase 3 (26th-31st): the index sees positive daily returns

The chart below replicates that above, but highlights these three phases of months.

S&P 500 average returns by calendar day of the month [1950-2015] b

 

Social Share Toolbar

The FTSE 100/S&P 500 monthly switching strategy

Although since 1984 the S&P 500 has overall greatly out-performed the FTSE 100 (+1021% against +575%), there are months in the year when the FTSE 100 fairly consistently out-performs the S&P 500.

The following chart shows the monthly out-performance of the FTSE 100 Index over the S&P 500 Index since 1984.

Comparative average monthly returns of FTSE 100 v S&P 500 [1984-2014]Looking first at the orange grey bars in the chart, this shows, for example, that on average in January the FTSE 100 has out-performed the S&P 500 by -0.6 percentage points (i.e. the UK index has under-performed the US index in that month). From the chart we can see that the five months that are relatively strong for the FTSE 100 are: February, April, July, August and December. For example, the FTSE 100 has out-performed the S&P 500 in February in 13 of the past 15 years.

Now, turning to the bown bars, these display the same average monthly out-performance of the FTSE 100 over the S&P 500, except this time the S&P 500 Index has been sterling-adjusted. One effect of adjusting for currency moves is to amplify the out-performance of the FTSE 100 index in certain months (April, July, and December). Conversely, the FTSE 100 under-performance is amplified in January, May and November.

Whereas, before, the relatively strong FTSE 100 months were February, April, July, August and December, we can see that the currency-adjusted strong months are just April, July, and December.

The FTSE 100/S&P 500 monthly switching strategy (FSMSP)

The above results suggest a strategy of investing in the U.K. market (i.e. the FTSE 100 Index) in the months April, July and December and in the U.S. market (i.e. the S&P 500 Index) for the rest of the year. In other words, the portfolio would be invested in the S&P 500 from January to March, at the end of March it switches out of the S&P500 into the FTSE 100 for one month, then back into the S&P 500 for two months, into the FTSE 100 for July, back into the S&P 500 for four months, then back into the FTSE 100 for December, and finally back into the S&P 500 to start the next year.

The following chart shows the result of operating such a strategy from 2000. For comparison, the chart also includes the portfolio returns from continuous investments in the FTSE 100 and S&P 500.

FTSE 100-S&P 500 monthly switching strategy [2000-2014]The final result: the FTSE 100 portfolio would have grown 7%, the S&P 500(£) risen 32%, but the FTSE 100/S&P 500 monthly switching portfolio (FSMSP) would have increased 114%. Switching six times a year would have incurred some commission costs, but these would not have dented performance significantly.


UK Stock Market Almanac cover [160 x 240]The above is an extract from the newly published UK Stock Market Almanac 2015.

Order your copy now!

 

Social Share Toolbar

US mid-term elections

Multiple ballots are held at the time of the US mid-term elections, including those at the municipal and state level, and also all the seats are up for election in the House of Representatives and a third of the seats in the Senate.

They are called “mid-term” as they take place in the middle of the four-year presidential term; in other words they take place two years after the presidential election. As such they are often regarded as a referendum on the performance of the prevailing president and his party.

In a recent article in the Financial Times, Ken Fisher described a market anomaly that he calls the 86.4 per cent miracle. According to Fisher, since 1925 returns on the S&P 500 have been positive for 67.4% of all calendar quarters, but for the 4th quarter of a mid-term election year and the two following quarters returns have been positive on average 86.4%. Fisher summarises-

midterm elections mean three straight quarters where the market rises 28 per cent more of the time than average.

What is the reason for this? According to Fisher: legislative gridlock. During electioneering campaigns politicians promise lots of radical legislation (that investors invariably dislike) to buy votes. But the reality of most mid-term elections is that the president’s party loses seats resulting in gridlock in Washington. In other words, while there is much sound and fury in the lead up to an election, it is followed by relative political calm – which investors like.

Given the high correlation of the US and UK equity markets, might this anomaly also apply to the UK?

The following chart plots the proportion of positive returns for the FTSE All Share Index for all quarters (grey bars) and those for the 4th quarter of a mid-term election (MTE) year (purple bars) and following 1st and 2nd quarters. To analyse the consistency of the anomaly over time, results are given for four different time periods.

For example, for the period 1910-2014, the FTSE All Share Index has had positive returns in 61% of all quarters, 62% of 4th quarters of a mid-term election year, 77% of the following 1st quarters, and 81% of the following 2nd quarters.

US mid-term elections and positive returns for FTAS [2014]Looking at the above chart the first observation to make is that the UK market experienced a greater proportion of positive returns in the 4th quarter of mid-term election years and the following two quarters than the average for all quarters – and this applied for all four of the different time periods tested. So this was consistent with the US results quoted by Fisher.

Regarding the period 1925-2014 (the period referred to by Ken Fisher), returns have been positive in 62.1% of all quarters (this compares a figure of 67.4% for the S&P 500 quoted by Fisher), and the average for the three (MTE) quarters has been 75.8% (compared with 86.4% for the S&P 500). So, where Fisher found that the three (MTE) quarters rose 28% more of the time than the average, in the UK the equivalent figure has been 22%.

A second observation to make is that the out-performance of the 4th and 1st (MTE) quarters over the average for all quarters has markedly increased in the most recent period from 1980. And that since 1980 the (MTE) quarter with the highest proportion of positive returns has been the 4th – in fact the UK market has risen in every 4th (MTE) quarter since 1980.

The following chart is similar to the above, except that it plots the average returns instead of the proportion of positive returns. For example, since 1910, the average return of the FTSE All Share Index for all quarters has been 1.5%, for the 4th (MTE) quarter it has been 2.4%, for the 1st (MTE) quarter 6.3%, and for the 2nd (MTE) quarter 4.2%.

US mid-term elections and average quarterly returns for FTAS [2014]Generally, the same profile of performance seen above is repeated here – all three (MTE) quarters out-perform the average. Since 1925 the average return for all quarters has been 1.7%, whereas the average return for the three MTE quarters has been 5.0%.

In 2014 the US mid-term elections will be held on 4 November, while the 4th (MTE) quarter starts 1 October. Fisher predicts “glorious gridlock” and a consequent “magical melt-up” for the market.

 

Social Share Toolbar

Index changes (S&P 500) – paper review

The S&P 500 Index Effect describes the tendency for companies joining the S&P 500 Index to experience a positive and permanent impact on their share prices and betas.

Academic research on the topic has addressed:

  1. Whether the effect actually exists and, if it does, if the effect is symmetric (i.e. do companies leaving the index experience a fall in price and beta), and whether the effect is permanent.
  2. What causes the effect
  3. Are these S&P 500  index changes information-free events
  4. The effect on index funds

This article presents a brief review and listing of academic papers on the S&P 500 Index Effect.


1. Form of the effect

The following papers found that the share prices of companies joining the S&P 500 Index experienced positive abnormal returns and that this effect was permanent: Beneish and Whaley (1997), Beneish and Whaley (2002), Chen, Noronha and Singal (2004), Cai (2007), Kappou, Brooks and Ward (2008) and Hrazdil and Scott (2009).

While some found the effect on price only temporary: Harris and Gurel (1986), Lynch and Mendenhall (1997) and Pruitt and Wei (1989).

Some found the effect asymmetric, whereby prices did not fall for companies leaving the S&P 500: Chen, Noronha and Singal (2004) and Zhou (2011).

After inclusion in the index these papers found that comovement (beta) increased: Barberis, Shleifer and Wurgler (2005), Kasch and Sarkar (2011) and Kasch and Sarkar (2012).

2. Causes of the effect

Several possible causes for the effect have been proposed-

The excess demand is due to indexing in the presence of downward sloping demand curves: Shleifer (1986), Beneish and Whaley (1996), Lynch and Mendenhall (1997) and Wurgler and Zhuravskaya (2002).

The bid-ask spread decreases which results in improved liquidity: Hegde and McDermott (2003) and Erwin and Miller (1998) .

Investor awareness increases: Dhillon and Johnson (1991), Chen, Noronha and Singal (2004), Elliott, Van Ness, Walker and Wan (2006) and Xie (2013).

Analyst coverage increases: Kalok Chan and Hung Wan Kot and Gordon Y.N. Tang (2013).

Operating performance of the companies improves: Denis, McConnell, Ovtchinnikov and Yu (2003), Kalok Chan and Hung Wan Kot and Gordon Y.N. Tang (2013), Jain (1987) and Dhillon and Johnson (1991).

3. An information-free event?

The following found that inclusion in the S&P 500 Index was not an information-free event: Geppert, Ivanov and Karels (2011), Gygax and Otchere (2010), Cai (2007) and Denis, McConnell, Ovtchinnikov and Yu (2003).

4. The effect on index funds

The following papers looked at the effect of index changes on index funds: Madhavan and Ming (2002), Chen, Noronha and Singal (2006), Dunham and Simpson (2010), Kappou, Brooks and Ward (2010) and Green and Jame (2011).

INDEX (of papers listed below)

[Papers listed in reverse date order; indicates major paper.]

  1. Stock price response to S&P 500 index inclusions: Do options listings and options trading volume matter? [2013]
  2. A comprehensive long-term analysis of S&P 500 index additions and deletions [2013]
  3. Media coverage, analyst recommendation upgrade and information content of inclusions into S&P indexes [2013]
  4. Is There an S&P 500 Index Effect? [2012]
  5. An examination of the information content of S&P 500 index changes: Analysis of systematic risk [2011]
  6. Strategic trading by index funds and liquidity provision around S&P 500 index additions [2011]
  7. Comovement Revisited [2011]
  8. Asymmetric Changes in Stock Prices and Investor Recognition Around Revisions to the S&P 500 Index [2011]
  9. Do Index Fund Managers Trade Opportunistically Around Index Changes? An Empirical Examination of S&P 500 Index Funds [2010]
  10. Analysis of the probability of deletion of S&P 500 companies: Survival analysis and neural networks approach [2010]
  11. S&P 500 Index Inclusions and Analysts’ Forecast Optimism [2010]
  12. The S&P500 index effect reconsidered: Evidence from overnight and intraday stock price performance and volume [2010]
  13. Index composition changes and the cost of incumbency [2010]
  14. S&P 500 Index Revisited: Do Index Inclusion Announcements Convey Information about Firms’ Future Performance? [2009]
  15. The Effect of Demand on Stock Prices: Evidence from the S&P Index Float Adjustment [2008]
  16. A re-examination of the index effect: Gambling on additions to and deletions from the S&P 500′s ‘gold seal’ [2008]
  17. What’s in the News? Information Content of S&P 500 Additions [2007]
  18. What Drives the S&P 500 Inclusion Effect? An Analytical Survey [2006]
  19. Index Changes and Losses to Index Fund Investors [2006]
  20. The addition and deletion effects of the standard & poor’s 500 index and its dynamic evolvement from 1990 to 2002: demand curves, market efficiency, information, volume and return [2006]
  21. Comovement [2005]
  22. The Price Response to S&P 500 Index Additions and Deletions: Evidence of Asymmetry and a New Explanation [2004]
  23. The liquidity effects of revisions to the S&P 500 index: an empirical analysis [2003]
  24. Price Pressure on the NYSE and Nasdaq: Evidence from S&P 500 Index Changes [2003]
  25. S&P 500 Index Additions and Earnings Expectations [2003]
  26. S&P 500 Index Replacements [2002]
  27. The Hidden Costs of Index Rebalancing: A Case Study of the S&P 500 Composition Changes of July 19, 2002 [2002]
  28. Does Arbitrage Flatten Demand Curves for Stocks? [2002]
  29. Price Effects of Addition or Deletion from the Standard & Poor’s 500 Index: Evidence of Increasing Market Efficiency [2001]
  30. The liquidity effects associated with addition of a stock to the S&P 500 index: evidence from bid/ask spreads [1998]
  31. A Scorecard from the S&P Game [1997]
  32. New Evidence on Stock Price Effects Associated with Charges in the S&P 500 Index [1997]
  33. An Anatomy of the “S&P Game”: The Effects of Changing the Rules [1996]
  34. Changes in the Standard and Poor’s 500 List [1991]
  35. Institutional Ownership and Changes in the S&P 500 [1989]
  36. The Effect on Stock Price of Inclusion in or Exclusion from the S&P 500 [1987]
  37. Price and Volume Effects Associated with Changes in the S&P 500 List: New Evidence for the Existence of Price Pressures [1986]
  38. Do Demand Curves for Stocks Slope Down? [1986]
  39. Does Delisting from the S&P 500 Affect Stock Price? [1986]


Stock price response to S&P 500 index inclusions: Do options listings and options trading volume matter?
Authors [Year]: Yangyang Chen and Constantine Koutsantony and Cameron Truong and Madhu Veeraraghavan [2013]
Journal [Citations]: Journal of International Financial Markets, Institutions and Money, 23, pp379–401
Abstract: This study investigates the stock price response to Standard & Poor’s (S&P) 500 index inclusions during the period 1996–2010 and the role of options listings and options trading volume with regard to the information content of index inclusion announcements. Specifically, we address the following questions: (1) Is the magnitude of abnormal returns from the announcements of S&P 500 inclusions significantly lower for stocks with options listings? and (2) Is the magnitude of abnormal returns from the announcements of S&P 500 inclusions significantly lower for stocks with a high level of options trading volume? Our findings indicate that options listings themselves are not related to the magnitude of abnormal returns from the announcements of S&P 500 inclusions. We also find that greater levels of options trading volume do not convey private information about the S&P 500 index changes. We document that any measurable impact of options trading on the stock price response to S&P 500 inclusion announcements lies primarily in the level of abnormal options trading volume in the period immediately preceding the announcements.
Ref: AA645

Continue reading

Social Share Toolbar

Triple Witching – paper review

This article presents a brief review and listing of academic papers on triple witching.

Summary

The expiry of stock index futures, stock index options and stock options happens in a programmed calendar throughout the year. On four days a year these three different types of derivative all expire on the same day – the third Friday of the months of March, June, September and December. This day is sometimes referred to as triple witching day, and is associated with increased trading volumes and volatility.

Review

Following the 1987 stock market crash there was great interest in program trading (a term not so commonly used today) and its impact on volatility. This led to a minor flurry of academic interest in associated topics such as triple witching.

The earliest mention of triple witching we can find in an academic paper is Feinstein and Goetzmann (1988), which looked at the increased volatility caused by the coincident expirations. A couple of years later Stoll and Whaley (1990) found greatly increased trading volume in the last half-hour on expiration days. However, they did not find any significant difference between stocks subject to program trading and other stocks.

In June 1987 the settlement of S&P 500 and NYSE index futures was changed (to settle at the open and not the close) in an attempt to decrease the impact of expiration. The effect of this was the topic of the most cited paper on triple witching, Stoll and Whaley (1991). Not surprisingly, perhaps, they found that volume and volatility of the S&P 500 and NYSE contracts was lower at the close and higher at the open for the period after June 1987 compared to the period before. The impact on price at the open was slightly smaller post-June 1987 than it had been at the close pre-June 1987.

Academic interest in triple witching then waned, albeit articles continued to appear on the more general topic of option expiration.

A rare, recent article (Stratmann and Welborn, 2012)  found a positive relationship between ETF settlement failures and ETF short sale volume, the cost to borrow ETFs, and triple witching days.

Continue reading

Social Share Toolbar

Super Bowl Indicator

This coming Sunday is Super Bowl XLVIII.

One of the most famous market predictors in the U.S. is the Super Bowl Indicator. This holds that if the Super Bowl is won by a team from the old National Football League the stock market will end the year higher than it began, and if a team from the old American Football League wins then the market will end lower.

Unlikely?

Well, it certainly sounds far-fetched that a game of mutant rugby could affect the economy and stock market. However, in 1990 two academics published a paper (Krueger and Kennedy, 1990) finding that the indicator was accurate 91% of the time.

And then in 2010 George Kester, a finance professor at Washington and Lee University, published a paper (Kester, 2010) with new research that found that the Super Bowl Indicator still worked (although its accuracy had fallen to 79%). Kester also calculated that a portfolio that switched between stocks and treasury bills governed by the Super Bowl Indicator would be worth twice that of a simple portfolio invested continuously in the S&P 500.

And the connection between American football and the UK stock market is…?

Seeing how closely correlated the U.S. and U.K. stock markets are, it might be interesting to see how the Super Bowl Indicator applies to the U.K. market.

The following chart shows the annual returns of the FTSE All Share index since 1967 (when the Super Bowl started). The Y-axis has been capped at +/- 50%, which truncates the bars for the years 1974 (-55%) and 1975 (+136%). The years for which the Super Bowl Indicator failed to accurately predict the direction of the market has been indicated with white bars in the chart.

FTSE All-Share annual returns as predicted by the super bowl [1967-2013]As one can see, the indicator got off to a great start in the years following 1967, but recently its record has been patchy. Overall, the indicator was accurate in 72% of years (only slightly less than its accuracy rate in the US).

Unfortunately a paper (Born and Acherqui, 2013) published last year has rather spoilt the fun. The authors found that the ability of the Super Bowl Indicator to forecast the market had reduced to almost zero in the years since publication of the Krueger and Kennedy paper in 1990.

Market around the time of the Super Bowl

The chart below shows the market behavior around the time of the Super Bowl; the bars represent the average daily returns in the FTSE All Share Index since 1967 for the three days before, and three days following, the Super Bowl (which always takes place on a Sunday).

FTSE All-Share average daily returns for the six days around the Super Bowl [1967-2013]The average daily returns in the index for all days since 1967 was 0.03%; we can see therefore that the market is abnormally weak two days before a Super Bowl and abnormally strong one day before it.

References

 

Social Share Toolbar

Market Music

The S&P 500 in 2013 set to music.

Social Share Toolbar

Comparative performance of FTSE 100 and S&P 500

The following chart shows the average monthly out-performance of the FTSE 100 Index over the S&P 500 Index for the period 1984-2013. For example, the FTSE 100 Index has out-performed the S&P 500 Index by an average of -0.6 percentage points in January.

COMPARATIVE MONTHLY RETURNS OF FTSE 100 V S&P 500 [1984-2013]NOTES

  1. Although since 1984 the S&P 500 has overall greatly out-performed the FTSE 100 (+1021% against +575%), there are months in the year when the FTSE 100 fairly consistently out-performs the S&P 500.
  2. The five months that are relatively strong for the FTSE 100 are: February, April, July, August and December. For example, the FTSE 100 has out-performed the S&P 500 in February in 13 of the past 15 years.
  3. The greatest monthly difference in performance is seen in May, when the S&P 500 on average beats the FTSE100 by 1.3 percentage points each year.

Data

The table below shows the monthly out-performance of the FTSE 100 Index over the S&P 500 Index since 1984. For example, in January 1984, the FTSE100 increased 6.3%, while the S&P 500 fell -0.9%; the out-performance of the former over the latter was therefore 7.2 percentage points.

The cells are highlighted if the number is negative (i.e. the S&P 500 out-performed the FTSE 100).

 FTSE 100 v S&P 500(£)

The following chart is similar to the above except the S&P 500 monthly returns have been currency-adjusted into sterling returns.

COMPARATIVE MONTHLY RETURNS OF FTSE 100 V S&P 500(£) [1984-2013]An effect of adjusting for currency moves is to amplify the out-performance of the FTSE 100 index in certain months (April, July, and December). Conversely, the FTSE 100 under-performance is amplified in January, May and November.

Whereas before the relatively strong FTSE 100 months were February, April, July, August and December, we can see that the currency-adjusted strong months are just April, July, and December.

 

Social Share Toolbar