ChapterPDF Available

“Sell in May and Go Away” on the Russian Stock Market

Authors:

Abstract

There is an ongoing discussion if the ‘sell in May’ strategy, that prescribes stock investments only during the months from September to May, offers a higher profit than a buy-and-hold strategy throughout the whole year. In the literature, some empirical evidence is particularly found for several emerging markets. In comparison to the US stock market, we examine the profitability of the ‘sell in May’ strategy at the Russian stock market and analyze its risk-return trade-off. Additional emphasis is laid on the persistence of the corresponding returns.
Sell in May and Go Away–
Summer Break also at the Russian Stock Market?
Peter Reichling and Elena Moskalenko*
January 2007
For a revised version of this paper see:
Gischer, H.; Reichling, P.; Spengler, T.; Wenig, A. (ed.) (2008):
Transformation in der Ökonomie, Wiesbaden, pp 257–267
* Prof. Dr. Peter Reichling (corresponding author–Otto-von-Guericke-University Magdeburg, Chair of Banking
and Finance, Postfach 41 20, 39016 Magdeburg, Germany, phone +49 391 67 18412, e-mail Peter.Reichling@
ww.uni-magdeburg.de) holds the chair of Banking and Finance at the Otto-von-Guericke-University Mag-
deburg, Germany. Elena Moskalenko holds a master’s degree from the German MBA program Moscow.
2
Abstract
There is an ongoing discussion if the ‘sell in May’ strategy, that prescribes stock investments
only during the months from September to May, offers a higher profit than a buy-and-hold
strategy throughout the whole year. In the literature, some empirical evidence is particularly
found for several emerging markets. In comparison to the US stock market, we examine the
profitability of the ‘sell in May’ strategy at the Russian stock market and analyse its risk-
return trade-off. Additional emphasis is laid on the persistence of the corresponding returns.
Keywords
Capital Market Anomaly, Halloween Effect, Russian Stock Market
JEL Classification
G11, G14, G15
3
Sell in May and Go Away–
Summer Break also at the Russian Stock Market?
October.
This is one of the peculiarly dangerous months to speculate in stocks in.
The others are July, January, September, April, November,
May, March, June, December, August, and February.
Mark Twain–The Tragedy of Pudd’nhead Wilson (1894)
1 Introduction
The seesaw changes of stock markets are frequently explained with certain sayings. These
rules are like country sayings: They are correct–at least every now and then. Probably, the
most well-known stock market rules is: Sell in May and go away. For example, by googling
‘sell in May’ one finds more than 80,000 entries in the web. The advise is to part in the
spring with stock market investments. The ‘sell in May’ rule is based on the observation that
share prices apparently tend to decrease during the summer months. This is one of the so-
called calendar effects in connection with numerous capital market anomalies. A capital mar-
ket anomaly contradicts the hypothesis of a weak-form efficient capital market.
In the literature there is an ongoing discussion if the ‘sell in May’ strategy offers a significant-
ly higher profit than a buy-and-hold strategy throughout the whole year. For example,
Bouman/Jacobson (2002) found that a strong ‘sell in May’ effect is present in particular in
European countries and emerging markets. However, Sullivan/Timmermann/White (2001)
attributed test results of this effect to data mining. Maberly/Pierce (2004) re-examined the
Bouman-Jacobson result and found that the effect disappeared in the US data after an adjust-
ment for outliers. Lucey/Whelan (2002) in turn obtained significant results for the Irish equi-
ty market.
4
Actually, in May 2006 the US S&P 500 index declined by three per cent. At the same time
the Japanese Nikkei 225 index lost nearly nine per cent, the European Stoxx 50 dropped by
five per cent, and the Russian RTS index decreased by 12 per cent. Accordingly, Forbes stat-
ed: “The axiom ‘sell in May and go away’ worked like a charm” (June 6th, 2006). The Econ-
omist regarded the ‘sell in May’ rule “as an explanation of why investors the world over have
been selling shares since May 11th” (May 25th, 2006). The Financial Times reasoned that
“this year, ‘sell in May and go away’ would have been a great strategy” (July 14th, 2006).
Much is and has been presumed about the reasons for this seasonality. The range of rationales
put forth reaches from the weather, according to which human beings are devoted to idleness
in the summer and recover from the exchange hustle and bustle,1 up to the race of institutional
investors, whose portfolio managers perform their asset allocation at the beginning of the year
and compete for the best starting positions. Profits are preferably realized in the middle of the
year and, at the end of the year, only small upturns are needed in order to induce the managers
to purchase, so that they do not chase the market.2 We note that in both cases it is ignored to
some extent that all buyers face sellers who likely have contrary opinions regarding the future
trend of stock prices.
In order to examine if the ‘sell in May’ rule is useful for investing at the Russian stock market
we proceed as follows. In comparison to the US stock market, in section 2 we analyse if an
investment strategy, which omitted the summer months, was particularly favorable at the Rus-
sian stock market. Subsequently, we fix May as the exit month and look for the best month to
enter the stock market. In section 3, we regard single months in order to determine the strate-
gy which optimally took seasonalities into consideration. Section 4 analyses the risk-return
trade-off of exit-in-May investments as well as the persistency of ‘sell in May’ returns. Sec-
tion 5 concludes with a brief summary.
2 Optimal Annual Entry and Exit Months
The ‘sell in May’ strategy initially leaves us alone with the question about the most beneficial
entry time. But the British know: ...and come back on St. Leger's Day. This is in the middle
1 For rationales in this regard see, e.g., Bouman/Jacobsen (2002) and Garrett/Kamstra/Kramer (2005).
2 In this line of argumentation see, e.g., Wermers (2003) and Doeswijk (2005).
5
of September. Therefore, one version of our stock exchange rule reads: sell in May and go
away, but remember to come back in September.3 And indeed, relying on monthly S&P 500
returns from 1960 to 2006 (figure 1, left axis) the investment strategy to enter the stock mar-
ket at the end of September and to exit end of May performs best amongst the ones with a
duration of eight months.
Sep-May Best of 8-Months
Duration Stock Investments?
0%
3%
6%
9%
Jan-Sep
Feb-Oct
Mar-Nov
Apr-Dec
May-Jun
Jun-Feb
Jul-Mar
Aug-Apr
Sep-May
Oct-Jun
Nov-Jul
Dec-Aug
0%
15%
30%
45%
S&P 500
RTS (USD)
Fig. 1: Average Returns of Stock Investments with 8-Months Duration
If this observation was a universal rule which could then be applied to the Russian stock mar-
ket, figure 1 (right axis) would answer as follows: from 1995 to 2006 we find the highest
average return of the RTS index on US dollar basis for the October to June strategy. 4 ‘Sell in
May’ ranks second highest. This strategy promised nearly eight per cent at the US stock mar-
ket on the 47 years average and even nearly 40 per cent per year at the Russian stock market
during the past 12 years.
If we insist on the ‘sell in May’ rule, which appears more or less valid, what would then actu-
ally be the best time for a re-entry? Again, our stock market rule seems to prove true, because
for both, the US stock market (figure 2, left axis, database as above) and the Russian stock
market (right axis), it was optimal to enter at the end of September, given one wanted to exit
end of May. Note, that intermediate investments remained unconsidered. Thus, we simply
3 Some studies found similar results for an entry at the end of October. Therefore, the supposed profitabil-
ity of the ‘sell in May’ strategy is also known as the Halloween effect, see, e.g., Jacobson/Visaltanachoti
(2006).
4 The RTS time series only goes back to 1995.
6
assumed that investors spent their money or put it into the piggy bank during the intermediate
months.5
Optimal Stock Marke t Entry in Sep?
-2%
0%
2%
4%
6%
8%
Apr
Mar
Feb
Jan
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Month of Entry
-10%
0%
10%
20%
30%
40%
S&P 500
RTS (USD)
Fig. 2: Average Returns of ‘Sell in May’ Strategies
3 Monthly Returns
A view at the monthly returns shown in figure 3 gives comprehensive information on the most
beneficial entry and exit times. Only in September the S&P 500 return was negative on aver-
age (left axis). Further, it is worth noting that on average the return was below the money
market rate in February and also from May to August. Therefore, the optimal investment
strategy would be to enter at the end of September and to exit end of April with a break in
February. The pattern is different for the RTS index (right axis). Here, the return in May and
from July to September was negative on average during the past twelve years. In November,
it was, on average again, better to hold bonds. Hence, the recommendation would read: enter
at the end of September and exit end of June with stopovers in November and May.
5 Our findings are not strongly affected if we, after the moratorium for foreign debt of the Russian govern-
ment in 1998, take interbank interest rates as a proxy for risk-free returns into consideration. Without any
fixed-income investments in the meantime, during-the-period returns in our figures correspond to annual-
ized returns.
7
Decreasing Share Prices
in Summer?
-1,0%
-0,5%
0,0%
0,5%
1,0%
1,5%
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
-8%
-4%
0%
4%
8%
12%
S&P 500
RTS (USD)
Fig. 3: Average Monthly Returns
In fact, significant differences in monthly returns arise in particular cases. Table 1 merely
contains combinations, where such significant differences occurred. It is worth noting that,
for the S&P 500 index as well as for the RTS index, the average September return is statisti-
cally significantly different from the mean returns of five and four other months, respectively.
In this respect, the advantage of the ‘sell in May’ strategy is not primarily based on the exit
time at the end of May, but on the re-entry time end of September.
P-values S&P 500 RTS (USD)
Sep Nov Dec Sep Dec
Jan 2.3 %
Feb 3.8 % 1.6 %
Mar 3.7 % 2.9 %
Apr 2.7 % 2.5 %
Jun 3.3 %
Sep 1.0 % 0.3 % 2.0 %
Tab. 1: Significant Differences in Monthly Returns
4 Risk-Return Trade-off
Today even private investors are familiar with the no risk, no fun principle, according to
which a higher return cannot be expected without bearing additional risk. The range of histor-
ical September to May returns is accordingly large. Figure 4 shows that, based on the S&P
500, returns range from minus 20 to above 40 per cent during the 47 years we covered in our
8
analysis. The range for the RTS is clearly wider, ranging from below minus 60 to almost 130
per cent during the past 12 years.
Low-risk Sep-May Investments?
0%
15%
30%
45%
-60%
-40%
-20%
0%
20%
40%
60%
80%
100%
120%
Return
Frequenc
y
RTS (USD)
S&P 500
Fig. 4: Risk of September-to-May Investments
The broad distribution of ‘sell in May‘ returns leads us to the question if the September-to-
May strategy dominates other exit-in-May strategies with different re-entry months. Figure 5
illustrates the risk-return trade-offs of these strategies. Both, mean return and volatility, tend
to increase with the duration of the stock investment until the September-to-May duration of
eight months is reached. Afterwards, mean returns tend to decline with roughly stable volatil-
ities. This observation holds for both, the US and the Russian stock market. Again, the ‘sell
in May’ strategy appears to be more beneficial than the buy-and-hold strategy of a stock in-
vestment during the whole year. Nevertheless, the return of the September-to-May strategy
on average is for both, the US and the Russian stock market, statistically not significantly
above the average return of an exit-in-May strategy with an earlier entry month.6
6 P-values for one-tailed tests are above 26 and 37 per cent for the S&P 500 and the RTS time series, re-
spectively.
9
Exit-in-May Investments‘ Performance
1Month
12 Months
Sep-May
0%
5%
10%
0% 5% 10% 15%
Risk
Return
1Month
12 Months
Sep-May
-10%
0%
10%
20%
30%
40%
0% 20% 40% 60%
S&P 500 RTS (USD)
Fig. 5: Risk and Return of Exit-in-May Investments
A trading rule is only helpful for an investor if she or he may rely on its constancy. There-
fore, in the following we analyse if from a high September-to-May return in one year we can
refer a similar return next year. Figure 6 visualizes that no such persistency of ‘sell in May’
returns exists. Rather, the September-to-May returns are negatively correlated,7 so that high
returns of our regarded strategy in one year tend to be followed by low returns in the next
year. However, this result is again not statistically significant. All in all, this result brings our
stock market rule back to the level of a country saying.
Persistence of Sep-May Returns
-25%
0%
25%
50%
May 65
May 70
May 75
May 80
May 85
May 90
May 95
May 00
May 05
-70%
0%
70%
140%
S&P 500
RTS (USD)
Fig. 6: Persistency of September-to-May Returns
7 The coefficient of autocorrelation of the ‘sell in May’ returns amounts to 0.15 and 0.33 for the S&P
500 (figure 6, left axis) and the RTS (right axis) time series, respectively.
10
5 Summary
‘Sell in May and go away’ likely represents the most well-known stock market saw. Surpris-
ingly, the saying seemed to prove true for both, the US and the Russian stock market. Based
on the S&P 500 index from 1960 to 2006 and the RTS index on US dollar basis from 1995 to
2006, the September-to-May strategy performed best and ranked second-best, respectively,
amongst stock investments with a duration of eight months. The analysis of single month
returns showed that the advantage of this strategy is predominantly due to the entry time at the
end of September. The exit time end of May comes second.
In addition, the ‘sell in May’ strategy offered a higher risk-return trade-off than exit-in-May
strategies with a longer investment period. However, the ‘sell in May’ strategy unfortunately
turns out to be risky, in particular as soon as an investors from the observation that ‘sell in
May’ worked one year refers that high returns are assured also next year. There is utterly no
evidence for such persistency.
References
Bouman, S.; Jacobsen, B. (2002): The Halloween Indicator, “Sell in May and Go Away”:
Another Puzzle, American Economic Review 92, 1618–1635.
Doeswijk, R.Q. (2005): The Optimism Cycle: Sell in May, IRIS (Rabobank/Robeco) Working
Paper.
Garrett, I.; Kamstra, M.J.; Kramer, L.A. (2005): Winter Blues and Time Variation in the Price
of Risk, Journal of Empirical Finance 12, 291–316.
Jacobson, B.; Visaltanachoti, N. (2006): The Halloween Effect in US Sectors, Working Paper.
Lucey, B.M.; Whelan, S.F. (2002): A Promising Timing Strategy in Equity Markets, Journal
of the Statistical and Social Inquiry Society of Ireland 31, 74–110.
Maberly, E.D.; Pierce, R.M. (2004): Stock Market Efficiency Withstands another Challenge:
Solving the “Sell in May/Buy after Halloween” Puzzle, Econ Journal Watch 1, 29–46.
Sullivan, R.; Timmermann, A.; White, H. (2001): Dangers of Data Mining: The Case of Cal-
endar Effects in Stock Returns, Journal of Econometrics 105, 249–286.
11
Wermers, R. (2003): Is Money Really “Smart”? New Evidence on the Relation Between Mu-
tual Fund Flows, Manager Behavior, and Performance Persistence, Working Paper.
Article
We present economically and statistically empirical evidence that the Halloween effect is significant. A trading strategy based on this anomaly works persistently and outperforms the buy and hold strategy in 8 out of 10 indices in our sample. We present evidence that the Halloween strategy works two out of every three calendar years and if an investor followed it "blindly", it would yield an annual average excess of return of approximately 2.4%, compared to the buy and hold strategy and further ensure a significant reduction in risk in all indices (around 7.5% on an annual basis). We have considered several possible explanations for the anomaly, however, none was able to fully justify the seasonal effect. We suggest that a possible explanation may be related to negative average returns during the May-October period, rather than superior performance during the November-April period.
Full-text available
Article
In a working paper, Jacobsen and Bouman (2001) claim that that the old stock market saying of "sell in May and go away but buy back by St. Leger Day" produces statistically significant profit when tested on a large database of equity market returns over the last decade, three decades, and even longer periods. In a recently published paper, Sullivan, Timmerman and White (2001) dismissed the statistical significance of this or any other calendar-based trading rule, attributing the reported test results to a large data mining exercise of the academic and financial communities. In this paper, we provide an out-of-sample test on the Bouman and Jacobsen strategy and conclude that the reported results are indeed statistically significant. In doing so we reintroduce a reliable index of capital returns on the Irish equity market maintained contemporaneously by the Irish Central Statistical Office (and its forerunner) since January 1934 which, in its early decades, displays markedly different statistical properties to both the US and UK equity markets of that time and equity market returns generally in recent decades. As a subsidiary exercise we reconsider the extensive literature on monthly seasonality in equity markets with this novel index. It is contended that the abnormally high returns frequently reported in January and April and occasionally in February and other months are perhaps more accurately and certainly more parsimoniously ascribed to the half-year effect captured in the old stock market adage.
Full-text available
Article
Mutual fund returns strongly persist over multi-year periods—that is the central finding of this paper. Further, consumer and fund manager behavior both play a large role in explaining these long-term continuation patterns—consumers invest heavily in last-year's winning funds, and managers of these winners invest these inflows in momentum stocks to continue to outperform other funds for at least two years following the ranking year. By contrast, managers of losing funds appear reluctant to sell their losing stocks to finance the purchase of new momentum stocks, perhaps due to a disposition effect. Thus, momentum continues to separate winning from losing managers for a much longer period than indicated by prior studies.
Article
Mutual fund returns strongly persist over multi-year periods - that is the central finding of this paper. Further, consumer and fund manager behavior both play a large role in explaining these long-term continuation patterns - consumers invest heavily in last-year's winning funds, and managers of these winners invest these inflows in momentum stocks to continue to outperform other funds for at least two years following the ranking year. By contrast, managers of losing funds appear reluctant to sell their losing stocks to finance the purchase of new momentum stocks, perhaps due to a disposition effect. Thus, momentum continues to separate winning from losing managers for a much longer period than indicated by prior studies. Even more surprising is that persistence in winning fund returns is not entirely explained by momentum - we find strong evidence that flow-related buying, especially among growth-oriented funds, pushes up stock prices. Specifically, stocks that winning funds purchase in response to persistent flows have returns that beat their size, book-to-market, and momentum benchmarks by two to three percent per year over a four-year period. Cross-sectional regressions indicate that these abnormal returns are strongly related to fund inflows, but not to the past performance of the funds - thus, casting some doubt on prior findings of persistent manager talent in picking stocks. Finally, at the style-adjusted net returns level, we find no persistence, consistent with the results of prior studies. On balance, we confirm that money is smart in chasing winning managers, but that a "copycat" strategy of mimicking winning fund stock trades to take advantage of flow-related returns appears to be the smartest strategy.
Article
Economics is primarily a non-experimental science. Typically, we cannot generate new data sets on which to test hypotheses independently of the data that may have led to a particular theory. The common practice of using the same data set to formulate and test hypotheses introduces data-mining biases that, if not accounted for, invalidate the assumptions underlying classical statistical inference. A striking example of a data-driven discovery is the presence of calendar effects in stock returns. There appears to be very substantial evidence of systematic abnormal stock returns related to the day of the week, the week of the month, the month of the year, the turn of the month, holidays, and so forth. However, this evidence has largely been considered without accounting for the intensive search preceding it. In this paper we use 100 years of daily data and a new bootstrap procedure that allows us to explicitly measure the distortions in statistical inference induced by data mining. We find that although nominal p-values for individual calendar rules are extremely significant, once evaluated in the context of the full universe from which such rules were drawn, calendar effects no longer remain significant.
Article
Previous research has documented robust links between seasonal variation in length of day, seasonal depression (known as seasonal affective disorder, or SAD), risk aversion, and stock market returns. The influence of SAD on market returns, known as the SAD effect, is large. We study the SAD effect in the context of an equilibrium asset pricing model to determine whether the seasonality can be explained using a conditional version of the CAPM that allows the price of risk to vary over time. Using daily and monthly data for the US, Sweden, New Zealand, the UK, Japan, and Australia, we find that a conditional CAPM that allows the price of risk to vary in relation to seasonal variation in the length of day fully captures the SAD effect. This is consistent with the notion that the SAD effect arises due to the heightened risk aversion that comes with seasonal depression, reflected by a changing risk premium.
Article
This paper investigates stock market mean returns and volatility spillover between stock markets of political and friendly countries. The potential foes and friends are selected according to the political situations in the past ten years. The three pairs of foes tested are Israel-Jordan, India-Pakistan, and Greece-Turkey. The United States has been historically and traditionally friendly toward these six countries. Spillover between the United States and these countries is also investigated. The empirical tests are conducted by means of a nonlinear GARCH-t model. Results indicate bidirectional mean and volatility spillover between two countries not on friendly terms. Results also provide ample evidence that mean and volatility spillover takes place from a larger distant friendly country (the United States) to these smaller emerging markets, but not much the other way around.
Article
On average, stocks deliver close to zero returns from May through October. We hypothesize that this seasonal pattern is caused by an optimism cycle. With year end approaching, investors start to look towards next year, often with overly optimistic expectations. Several months into the year, the initial optimism becomes hard to maintain and the stock market experiences a summer lull. A global sector-rotation strategy based on this theory appears to be highly profitable. Global earnings growth revisions follow a seasonal pattern parallel to that of the stock market. Investors' optimism as measured by the initial returns on IPOs almost completely captures the results of the sector-rotation strategy in a separate analysis for the US stock market. All these findings support the optimism-cycle hypothesis.
Article
Examining the years 1970 to 1998, Bouman and Jacobsen (2002) document unusually high monthly returns during the November-April periods for both United States (U.S.) and foreign stock markets and label this phenomenon the Halloween effect. Their research suggests that the Halloween effect represents an exploitable anomaly and has negative implications for claims of stock market efficiency. Re-examining Bouman and Jacobsen’s empirical results for the U.S. reveals that their results are driven by two outliers, the “Crash” of October 1987 and the collapse of the hedge fund Long-Term Capital Management in August 1998. After inserting a dummy variable to account for the impact of the two identified outliers, the Halloween effect becomes statistically insignificant. This anomaly is not economically exploitable for U.S. equity markets. We extend the research to the S&P 500 futures contract and find no evidence of an exploitable Halloween effect over the period April 1982-April 2003.
Article
We test a trading strategy of (national) tactical asset allocation based on the popular market wisdom 'Sell in May and go away' also known as the 'Halloween indicator'. With this simple market timing strategy one holds a market portfolio of stocks during November through April and short term government bonds during May through October. We find that this simple market timing rule outperforms riskier market portfolios in almost all 17 countries in our study. Annualized returns in excess of returns on market indices in these countries are substantial. We show that this trading strategy has statistically significant market timing ability and we are able to reject mean variance efficiency of stock market indices for many countries in our study. Key Words: Stock returns, Sell in May, Market timing, Return predictability, Halloween indicator. Acknowledgments: The authors wish to thank Arnoud Boot, Dennis Dannenburg, Angelien Kemna, Theo Nijman and Enrico Perotti for detailed comments. The usual disclaimer applies. The views expressed in this paper are not necessarily shared by the ING-Group. 2 The Halloween Indicator: Sell in May and go away Abstract We test a trading strategy of (national) tactical asset allocation based on the popular market wisdom 'Sell in May and go away' also known as the 'Halloween indicator'. With this simple market timing strategy one holds a market portfolio of stocks during November through April and short term government bonds during May through October. We find that this simple market timing rule outperforms riskier market portfolios in almost all 17 countries in our study. Annualized returns in excess of returns on market indices in these countries are substantial. We show that this trading strategy has statistically significant market timi...
The Halloween Effect in US Sectors
  • B Jacobson
  • N Visaltanachoti
Jacobson, B.; Visaltanachoti, N. (2006): The Halloween Effect in US Sectors, Working Paper.