Santa Rally 2017

Posted on 5 January 2018 by Almanacist

The Santa Rally describes the tendency of the market to rise in the last two weeks of the year.

In 2017 the FTSE 100 Index had a return of +2.6% in the last two weeks of the year. So the Santa Rally effect held in 2017.

As can be seen in the following chart, the Santa Rally has only failed to deliver in two years since 2000.

Trading around Christmas and New Year

Posted on 19 December 2017 by Almanacist

Does the equity market display any particular pattern in the days around Christmas and New Year?

Mean returns

The following chart plots the average daily returns of the FTSE 100 Index for nine days around Christmas and New Year for the periods 1984-2017 and also 2000-2017.

The nine days studied were-

Days 1-3: the three trading days leading up to Christmas.
Days 4-6: the three trading days between Christmas and New Year.
Days 7-9: the first three trading days of the year.

For example, since 1984 the average return of the index on the day before Christmas has been 0.24%

Observations

Market strength increases to the fourth day (the trading day immediately after Christmas). Since 1984 the fourth day has been the strongest day of the whole period, with an average daily return of 0.49% (albeit the volatility of returns on this day is high).
Generally the profile of returns for the shorter time range (2000-2017) is similar to that for the whole period from 1984. The one significant difference is that since 2000 the strongest day of the period has been the first trading day of the new year. The new year generally starts strongly on the first day, with performance trailing off the following two days.
The weakest day in the period is the third day of the New Year, followed by the last trading day of the year.

Let’s now see if the pattern of positive returns confirms the above findings.

Positive returns

The following chart plots the proportion of daily returns for the FTSE 100 Index that were positive on the nine days around Christmas and New for the period 1984-2017.

For example, for 84% of the years since 1984 the returns on the day after Christmas were positive.

The profile of behaviour demonstrated by the positive returns is similar to that for the mean returns above.

So, how did equities perform last year around Christmas compared to the average behaviour seen above?

Last year

The following chart replicates the first chart above with the average day returns for the period 2000-2017, and also plots the actual day returns for the nine days around Christmas in 2016.

As can be seen, the actual returns last year roughly followed the average pattern since 2000: the strongest days were the days after Christmas and New Year, with performance quickly trailing off after New Year.

The above is an extract from the newly published UK Stock Market Almanac 2018.

Order your copy now!

A very average start to 2017

Posted on 5 January 2017 by Almanacist

The following chart plots the daily returns of the FTSE 100 Index for the nine days around Christmas and New Year.

The blue bars plot the average daily returns of these days for the period 2000-2016. The orange bars plot the daily returns for the last nine days.

As can be seen the actual daily returns for the last nine days have been on the whole pretty close to the average daily returns seen for the last 16 years..

Strong returns have been seen on the trading days following Christmas and New Year.
After the first day after New year, returns have trailed off (days 8 and 9 in the chart).

Market returns around Christmas and New Year (update)

Posted on 18 December 2015 by Almanacist

This updates the previous analysis of the historical behaviour of the FTSE 100 Index since 1984 for the nine days around Christmas and New Year. The days studied were:

Days 1-3: the three trading days leading up to Christmas.
Days 4-6: the three trading days between Christmas and New Year.
Days 7-9: the first three trading days of the year.

The following chart shows the average returns for these nine days around Christmas and New Year.

Notes:

The average daily change of the FTSE100 index from 1984 for all days is 0.03%, so it can be seen that all nine days around Christmas and New Year are stronger than the average daily returns for the rest of the year.
Generally, the market strength increases to the fourth day (the trading day immediately after Christmas) – this is the strongest day of the whole period, when the markets increases 83% of years since 1984 with an average return on this day of 0.47%. Although it should be noted that the standard deviation is the second highest on this day, meaning that the volatility of returns is greatest (the index actually fell 3% on this day in 1987 and 2002).
The weakest day in the period is the third day of the New Year.
The new year generally starts strongly on the first day, with performance trailing off the following two days.

The following chart shows the proportion of returns that are positive for each of the nine days.

The profile is similar to that for the average returns: the market is increasingly strong to the first day after Christmas, and then drops off after that.

To check the persistency of these results, the following chart compares the average daily returns for each day for the period 1984-2015 (i.e. as above) with the shorter period 2000-2015.

Broadly, the behaviour in the last 15 years has been similar to that for the longer period since 1984. The one obvious difference has been the extraordinarily strong average returns on the first trading day of the year seen since 2000.

Extract taken from the newly published UK Stock Market Almanac 2016.

Order your copy now!

Easter holidays and the stock market

Posted on 14 April 2014 by Almanacist

What impact, if any, does the Easter holiday have on the market?

A previous post looked at the behaviour of the market around holidays (sometimes referred to as the holiday effect). In this post we will narrow the focus to look at the behaviour of share prices around the Easter holiday.

The following chart shows the average daily returns for the FTSE 100 Index for the four days before, and three days after, the Easter holiday over the period 1984-2013.

The general profile of behaviour around Easter is similar to that seen before for all holidays.

The main differences are that H(-4) is significantly weak, and the average returns for the two days immediately before and after Easter are significantly higher than for all holidays. For example, the average return for H(-1) is 0.4% (13 times greater than the average return for all days in the year); for all holidays the figure is 0.2%. The standard deviation for the Easter H(-1) average return is also significantly low.

The following chart is similar to the above, but this time the period studied is 2000-2013.

This second chart suggests that the behaviour of the market around Easter has not changed significantly in recent years.

US holidays and European markets

Posted on 14 February 2014 by Almanacist

Summary

European stock markets tend to have positive and high returns on days when the NYSE is closed. The effect is significant when the previous day’s return on the NYSE has been positive.

Brief overview

The US has six holidays every year that are not holidays in Europe:

Martin Luther King Day (the third Monday in January)
President’s Day (the third Monday in February)
Memorial Day (last Monday in May)
Independence Day (the fourth of July)
Labour Day (the ﬁrst Monday in September)
Thanksgiving day (the fourth Thursday in November)

How do the European equity markets behave on these days when the US markets are closed?

This was the question asked by the authors of an academic paper (Casado, Muga and Santamaria, 2011).

The authors analysed open and close values for the CAC40, DAX, FTSE 100, IBEX35 and EUROSTOXX50 (for the euro-zone stock market) for the period 1991-2008. Their results were remarkable.

Their research found the average daily returns for the European markets when the US market was closed was 0.32%, which was 15 times greater than the daily returns on all days.The greatest (NYSE-closed) daily returns were 0.42% for the German market.

Interestingly they found similar results for the open to close data on the NYSE-closed days. Meaning that the information from the previous day’s US market had been fully absorbed at the market open, and the effect is attributable to European trading.

Because the effect is so great it has a clear economic signiﬁcance as it is possible to obtain signiﬁcant returns after deducting trading costs by trading index futures.

The following figure from the paper shows the result of systematically buying FTSE 100 futures at the open on a NYSE-closed day and where the previous day’s NYSE return was positive, and then closing the position at the close on the same day. The red line is the equity chart for the strategy (left axis), and blue line the FTSE 100 Index (right axis).

The strategy had positive returns in both bull and bear periods for the market.

Reference

Casado, Jorge and Muga, Luis and Santamaria, Rafael, The Effect of US Holidays on the European Markets: When the Cat’s Away (2011)

Holidays and the market

Posted on 21 January 2014 by Almanacist

An academic paper (Ariel, 1990) was published with the finding that the trading day prior to holidays in the US market had an average return 14 times greater than the average for the other days in the year. This, and other papers, found that the day immediately before holidays had the highest returns (in the period around holidays), with the third day before the holidays having the next highest return and the day following the holiday having negative returns.

More recently this was updated in a paper (Ziemba and Dzahabarov, 2011) that found that the holiday effect had diminished in the 1990s and 2000s and that the out-performance was occurring largely in just the third day before holidays.

Does such a holiday effect exist in the UK market?

The following charts show the results of analysis of the daily returns of the FTSE 100 Index around holidays. The days studied were the four trading days immediately prior to holidays, H(-4) to H(-1), and the three trading days after holidays, H(+1) to H(+3). A holiday was defined as a 3-day (or longer) period with no trading.

1984-2013

The following chart shows the average returns for the seven trading days around holidays for the period 1984-2013.

We can see that, as with the US studies, H(-3) and H(-1) were strong during the holiday periods. Although, unlike the US studies, the day after a holiday, H(+1), was also found to be strong – this day has an average return of 0.17% (six times greater than the average return for all days in the year).

The following chart shows the proportion of positive returns for the same period.

The profile is broadly similar to that for the average returns: H(-1) and H(-3) are strong, as is H(+1). The weakest day around holidays has been H(+3) – the only day with a negative return and proportion of positive returns under 50%.

2000-2013

To look at the persistency of these results, the following charts restrict the study to the more recent period 2000-2013.

It can be seen that the UK holiday effect has changed slightly in recent years.H(-1) is still relatively strong, H(-3) less so, but the main change has been the relative strength of H(+1).

References for the holiday effect

Ariel, R. A., High stock returns before holidays: existence and evidence on possible causes (1990)
Arsad, Zainudin and Coutts, Andrew J., Security price anomalies in the London International Stock Exchange: a 60 year perspective (1997)
Brockman, Paul and Michayluk, David, The persistent holiday effect: additional evidence (1998)
Cadsby, C. B. and M. Ratner, Turn-of-the-month and pre-holiday effects on stock returns: some international evidence (1992)
Dumitriu, Ramona and Stefanescu, Razvan and Nistor, Costel, Holiday Effects During Quiet and Turbulent Times (2012)
Dzhabarov, C. and W. T. Ziemba, Do seasonal anomalies still work? (2010)
KIM, Chan-Wung and Jinwoo PARK, Holiday Effects and Stock Returns: Further Evidence (1994)
Lakonishok, J. and S. Smidt, Are seasonal anomalies real? a ninety-year perspective (1988)
Pardo Tornero, Ángel and Meneu, Vicente, Pre-Holiday Effect, Large Trades and Small Investor Behaviour (2002)
Swinkels, Laurens A. P. and van Vliet, Pim, An Anatomy of Calendar Effects (2010)
Thaler, Richard, Anomalies – Seasonal anomalies in security prices II:Weekend, Holiday, Turn of the Month, and Intraday effects (1987)
Vergin, Roger and McGinnis, John, Revisiting the Holiday Effect: is it on holiday? (1999)
Dzhabarov, C. and W. T. Ziemba, Do seasonal anomalies still work? (2010)

Market returns around Christmas and New Year

Posted on 18 December 2013 by Almanacist

This analysis looks at the historic behaviour of the FTSE 100 Index since 1984 on the days around Christmas and New Year. The days studied were:

Days 1-3: the three trading days leading up to Christmas.
Days 4-6: the three trading days between Christmas and New Year.
Days 7-9: the first three trading days of the new year.

The following chart shows the average returns for these nine days around Christmas and New Year.

And the following chart shows the proportion of positive returns on the nine days.

Analysis

The average daily change of the FTSE100 index from 1984 for all days is 0.03%; it can be seen that all nine days around Christmas and New Year have higher average returns than the average for all days.
The strongest days are historically the two days leading up to Christmas and the two immediately following it.
Generally, the market strength increase to the fourth day (the trading day immediately after Christmas) – this is the strongest day of the whole period, when the markets increases 81% of years since 1984 with an average rise of 0.34%. Although it should be noted that the standard deviation is the highest on this day, meaning that the volatility of returns is greatest (the index actually fell 3% on this day in 1987 and 2002).
The weakest day in the period is the final trading day of the year – this is perhaps not surprising with traders closing positions for the year end.
The new year generally starts strongly, but not as strong as those days of the week before.

The UK Stock Market in March

Posted on 1 March 2013 by Almanacist

March is a middling month for the market, ranking sixth of all months in performance. In the last 43 years, since 1970, the market has risen 28 years in March (a rate of 65%), with an average return of 0.7% in the month.

Strong sectors this month tend to be Industrial Engineering, Aerospace & Defense and General Retailers; while weak is the Gas, Water & Multiutilities sector. Historically, medium-sized companies have performed marginally better than large companies in March.

The following chart plots the percentage returns of the FTSE All Share Index for each March since 1980. For example, last year in 2012 the market fell 1.4% in March.

As can be seen, although on average the market increases in March, there have been the occasional large falls. The general trend for the market in March is to rise for the first three weeks and then fall back in the final week (in fact, the final last week of March has historically been one of the weakest in the whole year).

The year’s first quarterly review of the FTSE 100 Index will take place on 6 March. This is when the index is re-balanced and companies can be added to the index (following which their share price often falls) or ejected from the index (following which their share price tends to rise). It is difficult to predict the outcome of the review a few weeks beforehand, but just at the moment companies looking possible candidates for ejection are Serco, Melrose and John Wood, to be replaced by easyJet, London Stock Exchange and Direct Line Insurance.

The first Triple Witching (when the expiry of futures and options converge on the same day) of the year takes place on 15 March. Volatility can increase around this day, while in recent years the return on the FTSE 100 Index on the Triple Witching day itself has been significantly greater than other days.

Easter falls at the end of the month (the LSE will be closed 29 March and 1 April), when we might see an interesting market anomaly called the holiday effect. Academic studies have found that the market tends to be abnormally strong on the days immediately before and after holidays.

The big economics event in March will be the chancellor’s Budget, scheduled for 20 March. The other significant date of the month will be the 7 March MPC interest rate announcement at 12 noon.

Article first appeared in Money Observer.

The UK Stock Market Almanac

Seasonality analysis and studies of market anomalies to give you an edge in the year ahead

Tag Archives: HOL