Different Industries and the Profitability of Momentum Strategies in

Document Sample
Different Industries and the Profitability of Momentum Strategies in Powered By Docstoc
					    Momentum and Industry-Dependence: An Analysis of the Swiss
                         Stock Market

                     Tim A. Herberger, Daniel M. Kohlert and Andreas Oehler*




                                            This draft: September 2009



                                                        Abstract

Various studies have shown that future stock returns are predictable based on past returns in
many international security markets. Developing strategies to benefit from these autocorrelations
of security returns and finding reasons for the abnormal positive returns resulting from trading
strategies are major objectives of investment research. Following Jegadeesh and Titman (1993)
and (2001), we analyze the profitability of momentum strategies in the Swiss stock market over
the period from December 1979 until February 2009. Controlling for market returns and for
transaction costs we find that investors using momentum strategies could have indeed generated
superior returns during that time period. This applies for several combinations of ranking and
holding periods. In addition, we analyze for the first time whether the returns of momentum
strategies involving Swiss stocks are industry-dependent. Our findings suggest that momentum in
the Swiss stock market is clearly driven by high-technology stocks. The financial sector, in
contrast, performs worst over that period of time.




JEL Classification: G11, G14, G15
Key Words: Behavioral finance, Underreaction, Momentum strategies, Industry effects




* Authors are members of the Department of Management, Business Administration and Economics, University of Bamberg, Bamberg,
  Germany. We thank Hannes Frey, Henrik Schalkowski and Stefan Wendt for helpful comments and Johannes Rettig and Silvia
  Eichhorn for technical support. The usual disclaimer applies. Please address correspondence to Daniel M. Kohlert, University of
  Bamberg, Department of Management, Business Administration and Economics, Kirschaeckerstr. 39, 96045 Bamberg, Germany,
  phone: +49 (0) 951 863-2717, fax: +49 (0) 951 863-2538; email: kohlert.finanz@sowi.uni-bamberg.de.
   Momentum and Industry-Dependence: An Analysis of the Swiss
                        Stock Market




                                            Abstract

Various studies have found that future stock returns are predictable based on past returns in many
international security markets. Developing strategies to benefit from these autocorrelations in
security returns and finding reasons for the abnormal positive returns resulting from trading
strategies are major objectives of investment research. Following Jegadeesh and Titman (1993)
and (2001), we analyze the profitability of momentum strategies in the Swiss stock market over
the period from December 1979 until February 2009. Controlling for market returns and for
transaction costs we find that investors using momentum strategies could have indeed generated
superior returns during that time period. This applies for several combinations of ranking and
holding periods. In addition, we analyze for the first time whether the returns of momentum
strategies involving Swiss stocks are industry-dependent. Our findings suggest that momentum in
the Swiss stock market is clearly driven by high-technology stocks. The financial sector, in
contrast, performs worst over that period of time.




JEL Classification: G11, G14, G15
Key Words: Behavioral finance, Underreaction, Momentum strategies, Industry effects




                                                                                                1
1.     Introduction

On the foundation of “Relative Strength” by Levy (1967), Jegadeesh and Titman (1993) show

that contrary to the neoclassic efficient market hypothesis proposed by Fama et al. (1969) and

Fama (1970), stock prices are auto-correlated and momentum exists in the U.S. stock market.

Based on 3, 6, 9 and 12 months ranking and holding periods involving U.S. stocks they find that

buying stocks which have performed well in the past and selling stocks that have performed

poorly in the past can achieve positive market-adjusted returns. The profitability of momentum

investment strategies in international stock markets has been confirmed, for example, by

Rouwenhorst (1998), who shows that momentum can be found in a sample of 12 European stock

markets (including Switzerland), Chui, Titman and Wei (2000) who confirm momentum for

eight Asian stock markets (without Japan), and Glaser and Weber (2003) who show that

investment strategies based on momentum can be profitable in the German stock market. Rey

and Schmid (2007) show that it is possible to implement successful momentum strategies in the

Swiss stock market. In an experimental setting, Oehler et al. (2003) also identify momentum

traders and disposition investors. In particular during the last few years, researchers modified the

momentum strategy as it was used by Jegadeesh and Titman (1993). Chan, Hameed and Tong

(2000), for instance, examine the profitability of momentum strategies applied to international

stock market indices. Their results indicate significant positive abnormal returns. George and

Hwang (2004) show that the 52-week high of the stock price can also be used as a criterion to

predict future stock returns. They construct portfolios with long (short) positions in stocks whose

current price is close to (far from) the 52-week high. Using the sample of Jegadeesh and Titman

(1993), they show that this modified momentum strategy yields returns that are about twice as

high as the “classic” momentum strategy.



                                                                                                  2
Using Swiss stock data from December 1979 until February 2009, we examine whether

momentum effects can still be found on the Swiss stock market. This is particularly interesting

after the financial crisis as turbulent market phases characterized by extreme volatility can

equally enhance or reduce momentum profits. We control for both market returns and transaction

costs. For the first time we also analyze whether returns from momentum strategies involving

Swiss stocks are industry-dependent. Controlling for market returns and for transaction costs we

find that investors using momentum strategies could have indeed generated superior returns

during that time period. This applies for several combinations of ranking and holding periods. In

addition, we analyze for the first time whether the returns of momentum strategies involving

Swiss stocks are industry-dependent. Our findings suggest that momentum in the Swiss stock

market is clearly driven by high-technology stocks. The financial sector, in contrast, performs

worst over that period of time.

       The article is organized as follows. Section 2 provides a detailed review of explanations

for the momentum effect. In Section 3 we explain the dataset and introduce the methodology

adopted. The presentation and the discussion of the empirical results follow in Section 4. Section

5 concludes.



2.     Related Literature

Although the profitability of momentum strategies is rarely disputed, there is still a great deal of

controversy over the reasons for such abnormal returns. Neoclassical economists attempt to

explain momentum as a rational compensation for risk, a liquidity premium and/or an illusion

induced by market frictions (Fuertes, Miffre and Tan, 2009). Lesmond, Schill and Zhou (2004),

for instance, argue that momentum cannot be profitably exploited when transaction costs are



                                                                                                  3
considered. Korajczyk and Sadka (2004), however, find that transaction costs can only partially

explain the profits of momentum strategies. Conrad and Kaul (1998) argue that cross-sectional

variation can potentially explain the profitability of momentum strategies, and Moskowitz and

Grinblatt (1999) find that it is for the most part attributable to momentum in different industries.

After controlling for momentum across industries, the authors hardly indentify profits in

individual stock returns.

       Another explanation is offered by Wu (2002) and Wang (2003) who assume that the

profitability of momentum trading is a result of time-variation in expected returns. Grundy and

Martin (2001) and Karolyi and Kho (2004), however, show that industry effects, time-variation

in expected returns and cross-sectional disparities explain the momentum effect only partially.

Chordia and Shivakumar (2002) indicate that macroeconomic variables (e.g., the yield on three-

month T-bill) predict the momentum phenomenon and the associated payoffs whereas Griffin, Ji

and Martin (2003) show that this approach only provides an incomplete explanation. Behavioral

economists follow another line of reasoning. They argue that momentum is the consequence of

cognitive biases and/or limits to arbitrage. In the model of Barberis, Shleifer and Vishny (1998)

momentum results from conservatism while representativeness leads to overvaluation and

consequently to price correction. Daniel, Hirshleifer and Subrahmanyam (1998) argue that

biased self-attribution and overconfidence are the reasons for the predictability of equity returns.

Chan, Jegadeesh and Lakonishok (1996) as well as Hong and Stein (1999) attribute momentum

to an only gradual distribution of new information on the market. In addition, Hong and Stein

(2000) show that momentum strategies are more profitable among stocks with low analyst

coverage and relate this effect to gradual distribution of new information on the market. In an

analysis of 55 countries Chui, Titman and Wei (2009) find indications that cultural differences



                                                                                                  4
affect stock return patterns and the extent of the success of momentum trading. They argue that

individuals from different countries respond differently to risk and relate this phenomenon to

different cultures. Hwang and Rubesam (2008), however, note that at least for the U.S. stock

market momentum profits have slowly eroded since the early 1990s. Steiner (2009) finds similar

results in an analysis of the US and Swiss stock market. In addition, he argues that the

predictability of returns seems not to be robust to the choice of methodology.



3.        Data and Methodology

In contrast to Rey and Schmid (2007) we do not center our analysis only on Swiss large-cap

stocks. Instead, we use monthly stock prices over the period from December 31st, 1979 until

February 28th, 2009 that are obtained from Datastream for all common stocks listed on the SIX

Swiss Exchange 1 , excluding American Depository Receipts (ADRs), Real Estate Investment

Trusts (REITs), closed-end funds. In order to optimize the quality of our data sample we

followed the recommendations of Ince and Porter (2006). In consequence, our sample includes

469 stocks, which are / were listed on the SIX Swiss Exchange. For our analysis we use returns

adjusted for capital gains and dividends. An equal weighted SIX Swiss Exchange Index that

consists of all active shares of the dataset over the respective time period is used as a market

proxy. The momentum strategy is intended as a zero-cost portfolio strategy that buys stocks that

have performed well in the past and sells stocks that have performed poorly in the past. Our

methodology mainly follows Jegadeesh and Titman (1993). We couch everything in terms of raw

returns, and we equal-weight these returns. Like Jegadeesh and Titman (1993) and Rouwenhorst

(1998), we analyze 16 trading strategies on the Swiss stock market: We use ranking periods J of


1
    Since August 2008 the name of the Swiss stock exchange is SIX Swiss Exchange. Before the name was SWX
    Swiss Exchange.

                                                                                                       5
3, 6, 9 or 12 months and holding periods K of 3, 6, 9 or 12 months. To increase the power of our

tests we construct overlapping test runs like Moskowitz and Grinblatt (1999), Jegadeesh and

Titman (2001), and Fuertes, Miffre and Tan (2009). As Jagadeesh and Titman (1993) and

Rouwenhorst (1998) we sort stocks into 10 deciles according to past performance, and then

measure the return differential of the most extreme deciles which they denote by P10 – P1. One

month after the ranking takes place we sort our sample into only two parts based on past

performance: P1, which includes the worst-performing 10 percent, and P2 which includes the

best-performing 10 percent of SIX Swiss Exchange stocks in the ranking period. Our basic

measure of momentum is then P2 – P1. The top 10 percent portfolio is the winners’ portfolio and

the bottom 10 percent portfolio is the losers’ portfolio. The stocks in the individual portfolios are

equally weighted. In order to avoid biases because of low-priced stocks, which are documented

for example in Conrad and Kaul (1993), particularly for January, we follow the method of

Jegadeesh and Titman (2001) and Fuertes, Miffre and Tan (2009) by (temporally) excluding

stocks which are priced below CHF 5 at the end of the ranking period. By skipping a month

between the end of the ranking period and the beginning of the holding period like Rouwenhorst

(1998), Grundy and Martin (2001) and Rey and Schmid (2007), we avoid some of the bid-ask

spread, price pressure and lagged effects, which could skew our results. These effects are

documented for example by Jegadeesh (1990) and Lehmann (1990). The gross return of a test

run (buy and hold return of a holding period of T months in test run i, BHRi ,T ) is defined as

follows:


        BHRi ,T   RW ,t  RL ,t 
                    T
(1)
                   t 0




                                                                                                   6
where RW ,t is the monthly return of the winners’ portfolio and R L ,t represents the monthly return

of the losers’ portfolio.



The market-adjusted return of a test run (buy and hold abnormal return of a holding period of T

months in test run i, ABHRi ,T ) is defined as follows:


           ABHRi ,T   RW ,t  RL ,t   RM ,t 
                         T
(2)
                        t 0



where RW ,t is the monthly return of the winners’ portfolio and R L ,t represents the monthly return

of the losers’ portfolio. RM ,t is the monthly return of the equal weighted Six Swiss Exchange

Index.

          For evaluating the success of a trading strategy it is essential to consider transaction costs,

which are incurred when the portfolios are assembled and when the strategy closes out the

positions at the end of a test run. 2 In our study we consider 4 different rates of transaction costs.

Three depend on the type of investor employing the momentum strategy. We assume that

institutional investors face transaction cost of .2% for every purchase or sale of extreme

portfolios, wealthy private clients .5%, and private clients 1.0%. The fourth one is taken from

Grundy and Martin (2001) who show that abnormal positive returns are only statistically

significant till a transaction costs level of .375% for every purchase or sale of extreme portfolios.

The market-adjusted return after transaction costs of a test run (buy and hold abnormal return

after transaction costs with a holding period of T months in test run i, ABHRTi ,T ) is defined as

follows:



2
    For an overview of the components of trading costs see Korajczyk and Sadka (2004).

                                                                                                       7
           ABHRTi ,T   RW ,t  RL ,t   RM ,t   Ti
                           T
(3)
                          t 0



where RW ,t is the monthly return of the winners’ portfolio and R L ,t represents the monthly return

of the losers’ portfolio. RM ,t is the monthly return of the equal weighted Six Swiss Exchange

Index and Ti is the rate of transaction costs.

           In order to analyze whether the returns of momentum strategies involving Swiss stocks

are industry-dependent, we use Datastream’s standard industry sector classification to identify

the industry of each stock. As the Swiss market data sample is only comprised of 469 stocks,

however, we do not use all 47 sectors. Instead, we aggregate those sectors into the 4 sector-

dimensions high-technology, services, financials and production and use these specific samples

to calculate momentum returns. 3




3
      The High-Tech sector encompasses Datastream’s Aero Space & Defense, Alternative Energy,
      Pharmaceuticals & Biotech, Software & Computer Services, and Technology Hardware & Equipment
      sectors. The Services sector is comprised of the Electricity Fixed Line, Telecommunications, Food & Drug
      Retailers, General Retailers, Media Mobile Telecommunication, Support Services, and Travel and Leisure
      sectors. The Financials sector is comprised of the Banks, Financial Services, Life Insurance, and Non-Life
      Insurance sectors. The Production sector encompasses the Automobiles & Parts, Beverages, Chemicals,
      Construction Materials, Electronic & Electrical Equipment, Food Producers, Foresting & Paper, Gas Water
      Multiutilities, General Industrials, Health Care Equipment & Services, Household Goods, Industrial
      Engineering, Industrial Metals & Mining, Leisure Goods, Mining, Oil & Gas, Oil Equipment, Personal
      Goods and Tabacco sectors.



                                                                                                                   8
4.     Results

Table I presents the average monthly returns on the composite portfolio strategies between

December 1979 and February 2009. The returns have been obtained based on the methodology

discussed in the previous section. The table shows that all strategies yield highly significant

returns. For the strategy with a three month ranking and a three month holding period (J = 3 / K

= 3) an equally-weighted portfolio formed from the stocks in the bottom percent of previous

three-month performance returns 0.005 percent per month, 1.13 percent less than the top percent

portfolio which returns 1.18 percent. When we expand the holding period to 6, 9 and 12 months,

the momentum portfolios’ returns decrease slightly to 1.13 percent, 1.01 percent, 1.08 percent

and 1.04 percent, respectively. Although both the returns of the loser portfolios (0.21 percent,

0.22 percent and 0.3 percent, respectively) and the returns of the winner portfolios (1.22 percent,

1.30 percent and 1.04 percent) increase, the loser portfolios’ returns increase slightly more. In the

case of the strategy with a ranking period of six months and a holding period of six months (J = 6

/ K = 6), an equally-weighted loser portfolio returns 0.04 percent per month, 1.32 percent less

than the winner portfolio which returns 1.36 percent. The highest momentum portfolio return of

1.39 percent is obtained by combining a six-month ranking period and a three-month holding

period. Increasing the holding period from 6 to 9 and to 12 months, respectively, does not

increase performance as returns are identical with 1.32 percent for the 6/9-strategy and slightly

lower for the 6/12-strategy. Nevertheless, the absolute differences among loser portfolio returns

on the one hand and winner portfolios on the other are relatively low. While the minimum return

of the loser portfolios is -0.007 percent in the case of the 6/3-strategy and highest for the 6/12-

strategy with .029 percent, the highest winner portfolio return is obtained with the 6/9-strategy




                                                                                                   9
yielding 1.32 percent and the lowest winner portfolio return with the 6-3-strategy yielding 1.32

percent.

       Using a strategy based on a ranking period of six months and an equally long holding

period (J = 6 / K = 6), we obtain a momentum portfolio return of 1.18 percent which results from

an equally-weighted loser portfolio returning 0.19 percent per month and a winner portfolio

which returns 1.37 percent. The shorter the time horizon of the holding period, the higher is the

momentum return in this case. While the 9/3-strategy returns 1.43 percent per month, the 9-6-

strategy returns 1.34 percent and the 9/12-strategy .99 percent. Only once there is a negative

return, resulting from the loser portfolio of the 9/3-strategy. Finally, a ranking period of twelve

months and a holding period of twelve months (J = 12 / K = 12) leads to a return of .93 percent.

In this case, an equally-weighted portfolio formed from the stocks in the bottom percent of

previous three-month performance returns .42 percent per month, while the top percent portfolio

returns 1.35 percent. When we shorten the holding period to 6, 9, and 12 months, the momentum

portfolio’s returns increase considerably to 1.05 percent, 1.20 percent, and 1.49 percent,

respectively. While the returns of the loser portfolios increase with the length of the holding

period (-0.003 percent, 0.17 percent and 0.28 percent), the returns of the winner portfolios

decrease (1.46 percent, 1.20 percent and 1.05 percent). The highest returns based on a ranking

period of 12 months are therefore obtained by the 12/3-strategy. Average overall returns are

highest for the medium-term 6-months strategies. This supports the approach in the literature to

mainly focus on a strategy based on six-month ranking and holding periods (e.g., Jegadeesh and

Titman, 1993; Rouwenhorst (1998); Hong, Lim and Stein, 2000). The winners of all strategies

clearly outperform the losers. All of the returns are significant at the 1 percent level. Our findings

are well in line with previous findings in the literature. Jegadeesh and Titman (1993), Jegadeesh



                                                                                                   10
and Titman (2001), Rouwenhorst (1998), and Hong, Lim and Stein (2000), for example, do not

find negative returns for their loser portfolios at all, and we only find them in two instances.

These authors also report considerably higher average returns of their winner portfolios as

compared to their loser portfolios. Moreover, the magnitude of the momentum returns found is

comparable to what is reported in the literature. As Rouwenhorst (1998) aptly states, stocks with

higher standard deviations, all else equal, are more likely to show unusual performance. Looking

at the risk of the momentum portfolios, however, we find that the average standard deviations

increase with the time horizon of the strategies, no matter whether they are based on a 3, 6, 9 or

12 month ranking period. This explains the rising t values even in cases with lower momentum

portfolio returns. While the monthly standard deviations of the portfolios based on the short-term

horizon of three months lie in a range from .2 percent to 3 percent, portfolios based on a six-

months ranking period have standard deviations from .8 to 3 percent. Momentum strategies

based on a ranking period of 9 months show standard deviations from .8 to about 2 percent and

12-month ranking period strategies have monthly standard deviations between 1.6 and 4 percent.

Although the higher standard deviations of the longer-term strategies comes along with a higher

return, the relationship between risk and return is best for the 6-9-strategy. Compared to

Rouwenhorst (1998) who reports monthly standard deviations of 5.62 percent for his loser

(decile) portfolio, and of 5.27 percent for his winner (decile) portfolio for the momentum

strategy based on six-month ranking and holding periods, the standard deviations of our sample

seem relatively low. This, however, is mainly a result of the long time horizon that our data set

covers. Volatility was low in the 70’s and 80’s of the past century compared to the more recent

past. Even the effects on risk and return of the dot-com bubble at the beginning of the 20th

century as well as the financial crisis are partly equalized by the long time series.



                                                                                               11
                                        Table I
       Average Monthly Returns of Momentum Portfolios in the Period from 1979 to 2009
     This table presents average equal-weighted monthly returns in percentages for price momentum
     portfolio strategies involving SIX Swiss Exchange stocks from December 1979 to February 2009. At
     the end of each month, all stocks are ranked in ascending order based on the actual t and the past J - 1
     months’ Buy and Hold returns. One month after the ranking takes place, ten equal weighted monthly-
     rebalanced portfolios are constructed. The two extreme portfolios consist of the ten percent of SIX
     Swiss Exchange stocks with the highest returns and the ten percent with the lowest returns in the
     ranking period. P1 represents the low-return loser portfolio, and P2 represents the high-return winner
     portfolio. Overlapping portfolios are constructed to increase the power of the tests. K represents
     holding periods where K = three, six, nine or twelve months. Returns of the winner and loser
     portfolios in t are simply the average of J portfolio returns. The momentum portfolio (P2 - P1) is the
     zero-cost, winner minus loser portfolio. T-statistics for monthly returns are shown in parentheses.

      Ranking
                                                                   Holding Period (K)
       Period
                                                                             
         (J)              Portfolio                 3               6              9               12


         3             Loser (P1)                 0.0005          0.0021         0.0022          0.0030
                      Winner (P2)                 0.0118          0.0122         0.0130          0.0134
                 Winner – Loser (P2 - P1)         0.0113          0.0101         0.0108          0.0104
                         (t-stat)                 (6.90)          (8.44)         (9.61)          (10.64)

         6             Loser (P1)                -0.0007          0.0004         0.0012          0.0029
                      Winner (P2)                0.0132           0.0136         0.0144          0.0140
                 Winner – Loser (P2 - P1)        0.0139           0.0132         0.0132          0.0111
                         (t-stat)                 (7.91)          (9.47)         (10.73)         (10.54)

         9             Loser (P1)                -0.0011          0.0005         0.0019          0.0034
                      Winner (P2)                0.0132           0.0139         0.0137          0.0133
                 Winner – Loser (P2 - P1)        0.0143           0.0134         0.0118          0.0099
                         (t-stat)                 (7.76)          (9.15)         (9.60)          (9.25)

         12            Loser (P1)                -0.0003          0.0017         0.0028          0.0042
                      Winner (P2)                0.0146           0.0137         0.0133          0.0135
                 Winner – Loser (P2 - P1)        0.0149           0.0120         0.0105          0.0093
                         (t-stat)                 (7.37)          (8.15)         (8.67)          (9.04)




Table II presents the average monthly abnormal returns of the momentum portfolios on a market-

adjusted base using the average monthly SIX Swiss Exchange index return as the market proxy.

The returns have been obtained based on the methodology discussed in section two. It is clearly



                                                                                                                12
visible that the performance of all strategies decreases with increasing holding period horizon.

For the strategies based on a 3-month holding period, market-adjusted returns are highest for the

short-term 3/3-strategy with .46 percent and decrease with time to .28, .27 and .16 percent,

respectively. All returns are significant at the 10 percent level, those of the 3/3- and 3/9-

strategies even at the 5 percent level. In comparison to the returns obtained by the 3-month

ranking period strategies, the strategies based on 6-month ranking periods yield much higher

market-adjusted returns with .74, .59, .51 and .23 percent. While the returns of the 6/3-, the 6/6-

and the 6/9-strategy are significant at the 1 percent level, the return of the 6/12-strategy is still

significant at the 5 percent level. The average market-adjusted return is highest for these

strategies even if the single highest return is obtained by a 9/3-strategy with .77 percent.

Although the 9/6-strategy also yields a comparatively high return of .60 percent, returns drop

considerably thereafter. While the 9/9-strategy returns only .36 percent, the 9/12 strategy yields

.09 percent. The highest overall return, however, has the 12/3-strategy with a significant .82

percent. With increasing holding-period-horizon, however, returns drop to .46, .22, and .00

percent, respectively.




                                                                                                  13
                                            Table II
                 Average Monthly Returns After Market-adjustment over the Period
                            from December 1979 until February 2009
      This table presents average equal-weighted market-adjusted monthly returns in percentages for price
      momentum portfolio strategies involving SIX Swiss Exchange stocks from December 1979 to
      February 2009 for a given J/K-strategy. J represents ranking periods, where J = three, six, nine or
      twelve months, and K represents holding periods where K = three, six, nine or twelve months. An
      equal weighted SIX Swiss Exchange Index that consists of all active shares of the dataset at time is
      used as a market proxy. T-statistics for monthly returns are shown in parentheses.

             Ranking Period                                  Holding Period (K)                   
                   (J)                     3                 6                 9               12


                    3                   0.0046            0.0028            0.0027           0.0016
                                        (1.84)            (1.54)            (1.81)           (1.29)

                    6                   0.0074            0.0059            0.0051           0.0023
                                        (2.82)            (3.04)            (3.21)           (1.65)

                    9                   0.0077            0.0060            0.0036           0.0009
                                        (2.82)            (2.89)            (2.16)           (0.59)

                   12                   0.0082            0.0046            0.0022           0.0001
                                        (2.91)            (2.19)            (1.36)           (0.094)



Table III presents market-adjusted abnormal returns of the momentum portfolios adjusted to

different levels of transaction costs as described in section 3. In Panel A, transaction costs of .8

percent are considered in calculating the return of the momentum portfolio. In Panel B,

transaction costs are 2 percent. Panel C considers transaction costs of 4 percent, Panel D of 1.5

percent. For all 16 strategies that are based on a three-month ranking period, positive significant

abnormal returns cease to exist, no matter which level of transaction costs is considered. At a

transaction cost level of 4 percent, returns become significantly negative. Some of the strategies

based on a ranking period of 6 months, however, can still deliver positive abnormal market-

adjusted returns till a transaction cost level of 2 percent. Here, the 6/9-strategy still yields a

monthly abnormal return of .029 percent. At the .8 percent transaction cost level, all the 6/3-, the

6/6-, and the 6/9-strategies lead to highly significant positive abnormal returns after market-

                                                                                                             14
adjustment. Although some of the returns of the strategies based on ranking periods of 9 and of

12 months are even higher than in the case of a 6-month ranking period, none of the abnormal

returns is significant at the 2 percent transaction cost level anymore. At the .8 percent transaction

cost level, the 9/3-, 9/6-, 9/9- and 12/3-strategies still deliver positive abnormal returns.




                                                                                                  15
                                                                  Table III
                        Market-adjusted Monthly Returns of Momentum Portfolios Using Different Transaction Costs Levels
                                           over the Period from December 1979 until February 2009
This table presents average monthly abnormal returns in percentages for price momentum portfolio strategies involving SIX Swiss Exchange stocks from December
1979 to February 2009 for a given J/K-strategy. J represents ranking periods, where J = three, six, nine or twelve months, and K represents holding periods where K
  = three, six, nine or twelve months. In Panel A, transaction costs of .08 percent are considered in calculating the return of the momentum portfolio. In Panel B,
  transaction costs are .2 percent. Panel C considers transaction costs of 4 percent, Panel D of 1.5 percent. T-statistics for monthly return differences are shown in
                                                                              parentheses.

                        Panel A                                       Panel B                                     Panel C                                       Panel D
Ranking
 Period            Holding Period (K)                         Holding Period (K)                             Holding Period (K)                            Holding Period (K)
   (J)                                                                                                                                                               
             3         6        9        12            3          6             9     12             3           6          9        12            3           6          9       12
                                                                                                                                              
   3      0.0019 0.0009       0.0018    0.0001       -0.0021 -0.0006       0.0005   -0.0001        -0.0087    -0.0039    -0.0017   -0.0017       -0.0004     0.0003   0.0011    0.0004
          (0.78) (0.47)       (1.22)    (0.77)       (-0.82) (-0.32)       (0.33)   (-0.02)        (-3.49)    (-2.17)    (-1.15)   (-1.34)       (-0.15)     (0.14)   (0.70)    (0.31)
                                                                                                                                              
                                                                                                                                              
                                                                                                                                              
   6      0.0047     0.0046   0.0043    0.0016       0.0007     0.0026    0.0029    0.0006         -0.0060     -0.0007   0.0007    -0.0010       0.0024      0.0034   0.0035    0.0011
          (1.80)     (2.35)   (2.66)    (1.17)       (0.26)     (1.33)    (1.82)    (0.46)         (-2.29)     (-0.38)   (0.44)    (-0.73)       (0.90)      (1.76)   (2.17)    (0.75)
                                                                                                                                              
                                                                                                                                              

   9      0.0050     0.0046   0.0027   0.0002        0.0010    0.0026     0.0014    -0.0008        -0.0057     -0.0007 -0.0008     -0.0025       0.0027      0.0035   0.0019    -0.0004
          (1.84)     (2.24)   (1.63)   (0.13)        (0.37)    (1.27)     (0.83)    (-0.56)        (-2.07)     (-0.35) (-0.50)     (-1.70)       (0.98)      (1.68)   (1.17)    (-0.27)



                                                                                                                                              
   12     0.0056     0.0032   0.0013 -0.0005         0.0016     0.0012     0.0001   -0.0015        -0.0051    -0.0021    -0.0022   -0.0032       0.0032      0.0021   0.0006    -0.0011
          (1.97)     (1.55)   (0.82) (-0.38)         (0.56)     (0.59)     (0.01)   (-1.10)        (-1.80)    (-1.02)    (-1.34)   (-2.30)       (1.15)      (0.99)   (0.35)    (-0.80)




                                                                                                                                                                                    16
Based on these results we can conclude that momentum existed for the Swiss stock market over

the time period from 1979 until 2009. In order to obtain more detailed information, however, we

additionally investigate whether momentum is industry dependent, e.g., whether it works better

for some industries than for others. We do this using the 6/6-strategy which has become a

standard approach in the literature and also yields good results for our general data sample. Table

4 presents average monthly abnormal returns in percentages for such price industry momentum

portfolio strategies. There are clear differences among the sector returns. The monthly returns of

the momentum portfolios range between 1.82 percent for the High-Technology sector and .09

percent for the Financials sector. The service sector monthly return is .17 percent while the

Production Sector’s is .12 percent. All are highly significant. Looking at the market-adjusted

abnormal monthly returns for the 6/6-strategy, the results for the high-technology sector, the

service sector and the production sector are still positively significant with .10 percent, .04

percent and .05 percent, respectively. The financial sector only returns .01 percent. 4 Overall

performance is clearly dominated by the high-technology sector which produces significant

positive abnormal monthly returns after market-adjustment even at a transaction cost level of 4

percent. In general, the returns are much higher than for the whole data sample in some cases.

Only the production sector, however, also yields positive significant returns at a transaction cost

level of 2 percent. By far the lowest returns are obtained in the financial sector. Not even the

market-adjusted return without transaction costs is significant in this case.




4
    The distribution of industries and companies is as follows: high-technology 59 of 469, financials 118 of 469,
    services 90 of 469, production 202 of 469.


                                                                                                                    17
                                                               Table IV
          Market-adjusted Monthly Returns of Industry Momentum Portfolios and Returns Using Different Transaction Costs Levels
                                        over the Period from December 1979 until February 2009
This table presents average monthly returns in percentages for price industry momentum portfolio strategies involving SIX Swiss Exchange stocks from
December 1994 to May 2009. R represents the raw return of the momentum portfolio (winner minus loser portfolio) for a given 6/6-strategy. M represents the
market-adjusted returns. T .2% represents market-adjusted returns with a transaction costs level of .2 percent and so on. In Panel A, average returns of high-
technology sector stocks are presented. Panel B presents average returns of financial sector stocks. In Panel C, average returns of service sector stocks are
presented and Panel D shows average returns of the production sector. T-statistics for monthly return differences are shown in parentheses.


                                                                                                                        




                     Panel A                             Panel B                               Panel C                                 Panel D


Returns         High-Technology                         Financial                              Service                               Production  
                                                                                                                        

                      0.0182                              0.0096                                0.0117                                  0.0120
   R
                      (6.38)                              (4.03)                                (6.13)                                  (8.07)

                      0.0109                              0.0023                                0.0044                                  0.0047
  M
                      (3.65)                              (0.86)                                (1.89)                                  (2.25)

                      0.0096                              0.0010                                0.0031                                  0.0034
  T .8
                      (3.21)                              (0.36)                                (1.32)                                  (1.61)

                      0.0076                             -0.0010                                0.0011                                  0.0014
  T2
                      (2.54)                             (-0.39)                                (0.46)                                  (0.66)


  T4                  0.0043                             -0.0044                                -0.0023                                 -0.0019
                      (1.43)                             (-1.65)                                (-0.97)                                 (-0.93)

 T 1.5                0.0084                             -0.0002                                0.0019                                  0.0022
                      (2.82)                             (-0.08)                                (0.82)                                  (1.06)




                                                                                                                                                           18
5.     Conclusions

Using SIX Swiss Exchange data, we analyze the profitability of momentum strategies in the

Swiss stock market over the period from December 1979 until February 2009. Our findings

suggest that even after years of thorough analyses and considerable awareness of momentum

effects investors can still generate superior returns using momentum portfolio strategies. This

applies even after market-adjustment and the consideration of transaction costs as well as for

several combinations of ranking and holding periods. Returns are mainly driven by the

winners. We also find that the returns of momentum strategies involving Swiss stocks are

industry-dependent. Momentum in the Swiss stock market is clearly driven by high-

technology stocks. The financial sector, in contrast, performs worst over that period of time.



References

Barberis, N., Shleifer, A. and Vishny, R. (1998) “A Model of Investor Sentiment.” Journal of

       Financial Economics, 49(3): 307-343.

Chan, K., Hameed, A. and Tong, W. (2000) “Profitability of Momentum Strategies in the

       International Equity Markets, Journal of Financial and Quantitative Analysis, 35(2):

       153-172.

Chan, L., Jegadeesh, N. and Lakonishok, J. (1996) “Momentum Strategies.” Journal of

       Finance, 51(5): 1681-1713.

Chordia, T. and Shivakumar, L. (2002) “Momentum, Business Cycle, and Time-varying

       Expected Returns.” Journal of Finance, 57(2): 985-1019.

Chui, A., Titman, S. and Wei, J. (2000) “Momentum, Legal Systems and Ownership

       Structure: An Analysis of Asian Stock Markets.” Working Paper, University of Texas,

       Austin, Texas.

Chui, A., Titman, S. and Wei, J. (2009) “Individualism and Momentum Around the World.”

       SSRN, 1150076, fourthcoming in Journal of Finance.
                                                                                             19
Conrad, J. and Kaul, G. (1993) “Long-term Overreaction or Biases in Computed Returns?.”

       Journal of Finance, 48(1): 39-63.

Conrad, J. and Kaul, G. (1998) “An Anatomy of Trading Strategies.” Review of Financial

       Studies, 11(3): 489-519.

Daniel, K., Hirshleifer, D. and Subrahmanyam, A. (1998) “Investor Psychology and Security

       Market Under- and Overreaction.” Journal of Finance, 53(6): 1839-1885.

Fama, E., Fisher, L., Jensen, M. and Roll, R. (1969) “The Adjustment of Stock Prices to New

       Information” International Economic Review, 10(1): 1-21.

Fama, E. (1970) “Efficient Capital Markets: A Review of Theory and Empirical Works,

       Journal of Finance, 25(2), 383-417.

Fuertes, A.-M., Miffre, J. and Tan, W.-H (2009) “Momentum Profits, Nonnormality Risks

       and the Business Cycle.” Applied Financial Economics, 19(12): 935-353.

George, T. and Hwang, C.-Y. (2004) “The 52-Week High and Momentum Investing.”

       Journal of Finance, 59(5): 2145-2176.

Glaser, M. and Weber, M. (2003) “Momentum and Turnover: Evidence From the German

       Stock Market.” Schmalenbach Business Review, 55(2): 108-135.

Griffin, J., Ji, X. and Martin, S. (2003) “Momentum Investing and Business Cycle Risk:

       Evidence From Pole to Pole.” Journal of Finance, 58(6): 2515-2547.

Grundy, B. and Martin, S. (2001) “Understanding the Nature of the Risks and the Source of

       the Rewards to Momentum Investing.” Review of Financial Studies, 14(1): 29-78.

Hong, H., Lim, T. and Stein, J. (2000) “Bad News Travels Slowly: Size, Analyst Coverage,

       and the Profitability of Momentum Strategies.” Journal of Finance, 55(1): 265-295.

Hong, H. and Stein, J. (1999) “A Unified Theory of Underreaction, Momentum Trading, and

       Overreaction in Asset Markets.” Journal of Finance, 54(6): 2143-2184.

Hwang, S. and Rubesam, A. (2008) “The Disappearance of Momentum.” SSRN, 968176


                                                                                            20
Ince, O. And Porter, R.B. (2006), “Individual Equity Return Data From Thomson Datastream:

       Handle with Care!” Journal of Financial Research, 29(4): 463-479.

Jegadeesh, N. (1990) “Evidence of Predictable Behaviour of Security Returns.” Journal of

       Finance, 45(3): 881-898.

Jegadeesh, N. and Titman, S. (1993) “Returns to Buying Winners and Selling Losers:

       Implication for Stock Market Efficiency.” Journal of Finance, 48(1): 65-91.

Jegadeesh, N. and Titman, S. (2001) “Profitability of Momentum Strategies: An Evaluation of

       Alternative Explanations.” Journal of Finance, 56(2): 699-720.

Karolyi, G. and Kho, B. (2004) “Momentum Strategies: Some Bootstrap Tests.” Journal of

       Empirical Finance, 11(1): 509-536.

Koijen, R., Rodríguez, J. and Sbuelz, A. (2009) “Momentum and Mean Reversion in Strategic

       Asset Allocation.” Management Science, 55(7): 1199-1213.

Korajczyk, R. and Sadka, R. (2004) “Are Momentum Profits Robust to Trading Costs?.”

       Journal of Finance, 59(3): 1039-1082.

Lehmann, B. (1990) “Fads, Martingales and Market Efficiency.” Quarterly Journal of

       Economics, 105(1): 1-28.

Lesmond, D., Schill, M. and Zhou, C. (2004) “The Illusory Nature of Momentum Profits.”

       Journal of Financial Economics, 71(2): 349-380.

Levy, R. (1967) “Relative Strength as a Criterion for Investment Selection.” Journal of

       Finance, 22(4): 595-610.

Moskowitz, T. and Grinblatt, M (1999) “Do Industries Explain Momentum?.” Journal of

       Finance, 54(4): 1249-1290.

Oehler, A., Heilmann, K., Läger, V. and Oberländer, M. (2003). “Coexistence of Disposition

       Investors and Momentum Traders in Stock Markets: Experimental Evidence.” Journal

       of International Financial Markets, Institutions & Money, 13(5): 503-524.


                                                                                        21
Rey, D. and Schmid, M. (2007) “Feasible momentum strategies: Evidence from the Swiss

       stock market.” Financial Markets and Portfolio Management, 21(3): 325-352.

Rouwenhorst, G. (1998) “International Momentum Strategies.” Journal of Finance, 53(1):

       267-284.

Steiner, M. (2009) “Predicting premiums for the market, size, value and momentum factors.”

       Financial Markets and Portfolio Management, 23(2):137-155.

Wang, K. (2003) “Asset Pricing with Conditioning Information: A New Test.” Journal of

       Finance, 58(1): 161-195.

Wu, X. (2002) “A Conditional Multifactor Analysis of Return Momentum.” Journal of

       Banking and Finance, 26(8): 1675-1696.




                                                                                       22