Investments

***********************************************************
                Short Reviews of Academic Articles
***********************************************************
>Appeared in August 27th, 2003 newsletter

Market efficiency is a tough thing to beat. Go ahead, find an anomaly
(note to Matrix fans: I am not talking about Neo) and then have it torn
to bits in future papers. Lesmond, Schill (I’m telling you he is
everywhere!), and Zhou come to the defense of market efficiency and find
that the reported profits from momentum investing are minimally
overstated and possibly non-existent because of the higher than normal
transactions costs involved with the necessary trading. For example
while previous papers (example Jegadeesh and Titman 1991) account for
transactions costs, they only use the average cost of trading. This
paper reports that where momentum investing seems to be profitable is
concentrated in stocks with higher than average transactions costs.
Thus after they make adjustments for transaction costs, the abnormal
returns disappear.
http://papers.ssrn.com/abstract_id=256926

Socially Responsible Investing (SRI) means different things to different
people, but essentially is investing in firms that treat their employees
well, care for the environment, and make products or perform services
that are aligned with the goals and desires of the investors (for
example, many investors may refuse to buy tobacco stocks). For as long
as I can remember there has been a debate as to whether investors give
up pecuniary returns when they choose to invest in a socially
responsible fashion (SRI). Theoretically limiting the choice of firms
you can invest must reduce the efficient frontier. Dupre, Girerd-Potin,
and Kassoua investigate the actual cost of socially responsible
investing from a different angle. Using the ARESE ratings (a rating
based on firms’ social responsibility) for 173 European firms from
1999-200, they construct efficient frontiers with and without social
concerns. Predictably, the social concerns push the efficient frontier
to down (lower return) and to the right (more risk). This is NOT to say
do not invest ethically, merely it points out that utility maximization
(and not merely maximization of financial returns), is the reason for
the increased popularity of socially responsible investing.
http://papers.ssrn.com/abstract_id=394101

Empirical studies to date have largely failed to show convincingly that
SRI funds have lagged other funds. This may be because the studies have
focused on actual returns (ignoring risk) or because of the makeup of
socially responsible portfolios (for example, SRI funds tend to be
heavily weighted towards tech stocks which out performed other stocks
for much of the 1990s). Most academic studies of the topic find that
when risk is accounted for, there is a SRI penalty (for example, Geczy,
Stambaugh, and Levin ) but for the die hard believers (get it? ;)),
there are always problems with these studies. For example, recent
papers by the Geczy et al paper came under attack for using comparison
funds that were closed to new investors. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=416380

Maybe volume and noise are synonymous in more ways than one. Well,
maybe. Using a sample of trades from the Taiwan Stock Exchange, Hu and
Chan conclude that more frequent trading leads to more noise. Their
specific test looks at the time between trades and find that the longer
period between trades (90 seconds vs. 4 seconds) is associated with
lower levels of noise. Ok, let me first clarify things. Noise here
means the transitory errors in asset pricing and not the volume of the
trading floor. Thus, neither volume or noise should be measured in
decibels.
http://papers.ssrn.com/abstract_id=325340

>Appeared in June 5th, 2003 Newsletter

The turn of the year effect is the CAPM anomaly that finds that stock
returns are influenced by the calendar. IN particular that small firms
due better in the period right after the tax year end. (this has been
used to explain portions of the small firm effect and is based on the
idea that investors sell their losers for tax reasons.) Dai looks at
this using
the market-adjusted returns and tax code information for the 1984-1999
period. He finds that “the size of previous capital loss, the tax rate,
the interest rate [as well…]; limitations on the capital loss
write-offs, the length of the holding period during which the capital
loss is tax-deductible, the size of the listed firm.” All help to
explain the return behavior.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=392034
 
Rigobon and Sack looked at the US financial market responses to war.
They found stocks fell, there was a move towards Treasuries (the
so-called flight to quality), spreads between hi and low grade bonds
increased, and oil rose. None of which should come as any surprise but
it is always nice to get academic confirmation of what seems, but is not
always, obvious.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=390980

One of the really cool and also really frustrating things about finance
is that just when you think you get a handle on it, a new paper comes
out and makes you think again. In recent years there have been a slew
of papers that draw market rationality (and by association efficiency)
into question. For example, in 1998 Dichev (1998) found that distressed
firms earn less than would otherwise be predicted for firms of
equivalent risk (Three factor model) This is a problem for market
efficiency believers since the distressed state of the firms should
already be priced in such a manner as to result in the “correct” level
of return for the level of risk. So for the past 5 years, EMHers (my
term for believers of those who believe in the Efficient Market
Hypothesis) have been on the defensive. Aggrawal and Taffler have come
to the rescue! They include a more complete model of the risk-return
relationship and once “the impact of changes in GDP growth rate and the
impact of stock market movements” are considered, “bankruptcy is, in
fact, being appropriately priced by the market.” So market efficiency
adherents can breathe a bit easier. :-)
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=392587
(BTW sorry that was so long, but there were several key concepts to
include, probably worth a read, even if it is too long.)

>Appeared in April 23rd, 2003 newsletter

With apologies to Semisonic, Aitken, Comerto-Forde, and Frino look at
closing time. Not the closing time of bars, but the closing at the
Australian Stock Exchange. One way of classifying markets is whether
there is continuous trading or if periodic auctions are held. Quite a
few markets have adopted some version of an auction market for either
the open or the close. This trading mechanism allows orders to
accumulate. The rationale is that by gathering a larger number of
participants you can get better liquidity and lower transaction costs.
In 1997 the Australian Stock Exchange adopted a closing call auction.
Aitken, Comerto-Forde, and Frino examine this and find that about 2.5%
of trades take place at the close. Interestingly, these trades are not
new trades (that is, volume has not increased significantly), but trades
that were being done earlier are being pushed back to the close. That
is rather than trade over the previous two hours, when the market is
still open, the traders wait to trade at the end of the day when there
is greater liquidity. (Bar joke deleted)
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=391564

>Appeared in March 24th, 2003 newsletter

It is always interesting to see what happened when two theories clash.
For instance, I am largely a backer of the EMH (efficient market
hypothesis) which would suggest that mutual fund managers can not do
better than that the index and that overall all fund managers are about
as good as any other. However, more than I believe EMH, I am sure that
incentives are critical and that if you pay people to do something, it
usually will happen. So it was with great anticipation that I read the
forthcoming article by Elton, Gruber, and Blake in the Journal of
Finance (JF). The authors study whether mutual funds where managers
have an incentive pay system do better than those without. And they
find that yes incentives do seem to lead to better performance.
(Prediction, within the year, many more mutual funds will change their
incentive structure, as of now less than 2% of funds reward their
mangers with incentive pay plans.
http://www.afajof.org/Pdf/forthcoming/april23.pdf

Another chink in the EMH hypothesis. A paper by Coval, Hishleifer, and
Shumway finds that a significant proportion of investors did beat the
market for the 1990 to 1996. Moreover, the top 10% did so by am amazing
15 basis points per DAY! (There are 100 basis points per percent). READ
IT!
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=364000

Is the US market overvalued? (See the next story for the perspective of
others). That is what Reinker and Tower try to determine by comparing
expected returns for US firms with those of other nations. Their
finding, which rests on the Gordon dividend growth model, suggests that
it is and that international investments may earn higher returns going
forward. Of course, this is no guarantee, but it is an interesting
paper none-the-less.
http://www.econ.duke.edu/Papers/Other/Tower/Equity_Returns.pdf

>Appeared in March 3rd, 2003 newsletter

Ok, but come on? A JF article? If your friends invest in the stock
market, you are more likely to do so as well. That is the finding of a
new paper by Harrison Hong, Jeffrey D. Kubik and Jeremy C. Stein . To
which I can only ask one question: “Are you going to jump off a cliff if
your friends do?”
http://www.afajof.org/Pdf/forthcoming/Social-jf-revision.pdf

>Appeared in February 7th, 2003 newsletter

In a related article, Lynch and Musto examine the puzzling finding that
investors seem to disregard really bad performance. That is if a fund
does well in one period, more money tends to come to the fund in the
next period and conversely less money come in following bad
performances. But what has been puzzling is why really bad performers
are seemingly judged roughly the same by investors as only marginally
bad performers.   One hypothesis that has been floated (Ipploito 1992)
is that any return below a threshold is seen as being the same. Now in
a forthcoming JF article, Lynch and Musto suggest the reason may be that
really bad performers are likely to change strategies and/or managers so
that their previous bad performance may carry less weight. (I’ll
suggest another possibility---some investors may be looking for volatile
funds and banking on for reversion—see above article).
http://www.afajof.org/Pdf/forthcoming/howinv.pdf

After reading the forthcoming JF paper by Hischleifer and Shumay, I
propose that the financial center of the US be moved to Arizona. Why?
For the second time in recent months there is academic evidence that the
stock market does better when it is sunny out. This current paper finds
that the transaction costs PROBABLY make it unprofitable to try to take
advantage of it, but after looking at 26 markets for 15 years, a Z stat
of -3.96 for returns as a function of cloud cover suggest that the sun
plays an important role. How is unclear, but a leading hypothesis
suggests that it may be psychological. (Yes this was included before,
but it is now forthcoming in JF, so I figured you might like to see the
changes.)
http://www.afajof.org/Pdf/forthcoming/shumway.pdf

Speaking of short sellers, The Financial Review has an interesting
article on the apparent role that short sellers can play in REDUCING
volatility. Why? They flatten the peaks and valleys of stock
movements. Additionally the article looks at where the short sellers
were (on the sidelines because of limited float of many shares) during
the dot.com run-up of the late 1990s. (note to my students: stop and
internalize this one!)
http://afr.com/financialservices/2003/01/31/FFXYXCLJJBD.html

The Journal of Portfolio Management has two interesting articles
(Ilmanen and Hoisington/Hunt) that suggest the expected return
differential between stocks and bonds is not as great as we have been
lead to believe. Now this is coming after a bear market in stocks, so
it is not completely unexpected, but still it definitely should get us
thinking more about diversification and the usefulness of using
historical risk premiums that are shown to vary significantly over
various time horizons.
http://www.iijpm.com/issue.asp

>Appeared in January 22nd, 2003 newsletter

Jung and Shiller have an interesting paper that looks at Samuelson’s
dictum: that is that the market is more efficient pricing individual
stocks than getting the overall price level (i.e. the market) correct.
In English it means that while we can price stocks relative to one
another reasonably well, we can not price the overall market as well.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=348180

>Appeared in January 7th, 2003 newsletter

Mmm? Data mining or more evidence that things are not as simple as we
would like to believe. Kramstra, Kramer, and Levi have found a fairly
large and statistically significant relationship between stock returns
and the amount of sunlight that is available at the location of the
market.   (This is a fun article even if I am not totally convinced.)
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=361940

>Appeared in December 3rd, 2002 newsletter

Rule#1. Market efficiency does not mean market perfection. Take the
case of event studies. An event study can be seen as a test of how
quickly and how accurately markets incorporate new information. (Fama
1991). By and large we have seen event studies show that the market
responds with great alacrity and usually gets things correct. However,
it is not perfect.

In an upcoming JFE article, Chan reports that event studies are not the
nice right angles that we would like to see. Using an incredible data
set (all news announcements for from 1980-2000) he finds that after the
announcement, stocks continue to exhibit abnormal behavior (If markets
were PERFECT there would be no drift after the initial price reaction).
This drift is particularly pronounced after bad news where the stock
continues to fall relative to various control “groups.” He also finds
that this is most concentrated in smaller stocks. This is somewhat
comforting as most agree small stocks are less efficient than large
(widely followed stocks).
http://jfe.rochester.edu/02207.pdf

>Appeared in September 3rd, 2002 newsletter

How well can investors forecast growth? Apparently not well at all.
That is one conclusion coming from a forthcoming Journal of Finance
paper by Chan, Karceski, and Lakonishok. This finding should be more a
bit troubling given the importance of growth forecasts in most valuation
models (for example, in the constant growth model Price = Dividend in
year one /(Required return –growth rate)). Further, the paper
empirically finds little evidence of firms that can consistently
maintain the same rate of growth.
http://www.afajof.org/Pdf/forthcoming/chan.pdf

>Appeared in August 1st, 2002 newsletter

Well this may force people to reconsider some trading strategies. In
recent years there has been a well-advertised trading strategy of buying
stocks that split. Of course this stems from the initial event study by
Fama, French, Jensen, and Roll back in 1969. In an upcoming Journal of
Finance article,   Byun and Rozeff find little evidence of long-term
abnormal performance following most splits (they do find some positive
abnormal performance after 2:1 splits during certain periods and with
certain restrictions.)
http://www.afajof.org/Pdf/forthcoming/ROZEFF.pdf

Bradley, Jordan, and Ritter have found that IPO’s that had more analyst
coverage during the 1996-2000 period, did much better than those that
had no analyst coverage. But wait, this is not on the initial day of
trading, but at the end of the quiet period. Very interesting!
http://www.afajof.org/Pdf/February_03/QuietPeriod.pdf

Ok, this makes sense. Lawrence Kryzanowski and Arturo Rubalcava looked
at Canadian shares that trade in both the US and Canada,    Finds that
investors who buy and sell in the US (where trading costs are lower)
tend to have shorter holding periods. This is consistent with Ahmed and
Mendelson’s famous 1986 paper that suggests those with longer holding
periods will trade in less liquid markets.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=314385


>Appeared in June 19th, 2002 newsletter

Back in 1997 Pontiff found that US closed-end funds were more volatile
than the underlying assets that make-up the portfolio. This drew market
efficiency back into question. Now Agyei-Ampomah and Davies find that
similarly, UK investments trusts exhibit "excess volatility." What
makes the study all the more interesting are the institutional
differences between investments trusts and closed-end funds. For
example, if you try to explain the excessive volatility of US closed end
funds on the grounds of irrational individual investors (who own the
bulk of closed-end funds), the argument falls apart when looking at
investment trusts since they are largely held by institutions.
Interesting work!
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=313923

>Appeared in April 30th, 2002 newsletter
The equity risk premium has a long and storied history. This is the
premium that investors require to assume the risks associated with
holding equity investments. However, what has puzzled investors and
academics is why the premium has been so high for so long. (In other
words, dividends do not seem risky enough to lead to investors demanding
such a high return.) This has been termed the equity risk premium
puzzle and has been studied with varying degrees of success by many
researchers. Now Nicholas Barberis has used Behavioral Finance to
successfully describe the premium. This is based on the idea that
investors are not as concerned with their losses if their losses if the
money lost was "won" in an earlier round of the stock market game.
However, losses of invested money for some reason seems to hurt more and
thus lead to higher discount rates. So when this idea is applied it
leads to lower discount rates (and accordingly higher stock prices) in
good times and higher discount rates (and accordingly lower stock
prices) when the stock market has fallen. This is a recipe for
volatility and a time-varying discount rate, something simple models
such as CAPM do not account for.
http://gsbwww.uchicago.edu/news/capideas/fall01/stockmarket.html
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=214388

>Appeared in March 14th, 2002 newsletter

Gee, this is a different story! It asks what makes shareholders so
important? Well, beside they are the owners, the residual claimant (and
therefore not only in the best monitoring position but also by
satisfying shareholders they are at least hopefully satisfying all other
claimants as well. While I don’t like the idea of the paper, I do like
some of the story’s cool facts such as the size of the issuance relative
to buybacks in recent years. (hint: firms have been negative net
issuers.
http://web.uccs.edu/rweigand/The%20Incredibly%20Unproductive%20Shareholder.pdf