Investment strategy is a little like religion in
the financial advisor community. There are few situations that would
get emotions boiling, fists flying, and require police action faster
than putting a buy-and-hold advocate and a market timing zealot in a
room and asking them to resolve their differences. The truth is that
most strategies work some of the time, a few work most of the time, and
only Bernie Madoff figured out how to make one work all the time, right
up until he got caught. Investment strategies have two major parts: 1)
what investments to buy, and 2) when to buy and sell. Because I'm an
investment advisor and human, I have some built-in biases, but
following is an attempt to objectively look at several common
strategies with a minimum of sarcasm.
Allocation Strategies (what to buy)
Strategic Asset Class Allocation
Traditional
asset classes include stocks, bonds and cash. These classes are then
divided into subcategories based on geographic location (U.S.,
developed foreign countries, emerging markets), company size
(small-cap, mid-cap, large-cap), and bond style (treasuries,
mortgage-backed, high-yield, etc). Real estate, commodities, and hedge
funds are sometimes added as additional asset classes. The idea behind
Strategic Asset Class Allocation is to come up with a portfolio of
non-correlated assets that meets an acceptable risk profile, and then
stick with that allocation as the market goes up and down. The
portfolio is typically rebalanced periodically to maintain the
percentages of each asset class, but mostly the portfolio is left alone.
Most Common Supporting Arguments:
- Easy to set up with mutual funds, which are typically aligned with asset classes.
- Mutual funds provide diversification by owning many stocks with professional management.
My Rebuttal:
- Many
mutual fund managers tend to favor certain stock sectors at the same
time, making the portfolio less diversified than it appears (e.g.
overweighted in Energy or Financials).
- Most stock asset classes are highly correlated when looked at over the last decade.
Semi-Objective Opinion:
Dividing
the stock world by geographic location (U.S. & foreign) or by
company size no longer results in a diversified portfolio. This has
been a long-term trend developing and getting worse over the last 20 or
so years. As an intuitive example, when oil drops from $150/barrel to
$35/barrel, all energy companies get hurt, whether they are large or
small, based in the U.S. or based in Brazil. However, it is true that
an asset class allocation model is easy to implement with mutual funds,
and the addition of non-correlated alternative investments can improve
overall diversification.
Balanced Sector Allocation
As
stated above, a major problem with Asset Class Allocation is that the
major equity classes do not behave differently enough to do an
effective job of diversification. Balanced Sector Allocation gets
around this by diversifying across low-correlated sectors (Technology,
Energy, Financials, Healthcare, etc). This is not a new concept. Just
about any portfolio that uses individual stocks diversifies this way,
and the strategy can be implemented using either individual stocks or
sector-based Exchange Traded Funds (ETFs).
Most Common Supporting Arguments:
- Spreading
investments across non-correlated sectors does a much better job of
diversification than dividing investments by company size or where
their headquarters happens to be located.
- Individual stocks and ETFs typically have significantly lower expenses than mutual funds.
- Sector allocation can be precisely controlled.
My Rebuttal:
- If
Sector Allocation is implemented with a few individual stocks for each
sector, there is a significant amount of company-specific risk added to
the portfolio.
Semi-Objective Opinion:
In
addition to showing a significant performance improvement over the last
10-20 years, Sector Allocation passes the "this just makes sense" test.
Intuitively, a Healthcare stock and an Energy stock will do a better
job at diversification than a large-cap Energy stock and small-cap
Energy stock. The manager of an actively-managed mutual fund is
typically doing sector allocation within a particular Asset Class (e.g.
Large Cap Value), but if you own several mutual funds, there is
obviously no coordination between the managers.
Tactical Asset Allocation/Tactical Sector Allocation
These
strategies are similar, with the difference being that one uses
traditional asset classes and the other uses stock sectors. In both
cases, the objective is to predict which stock class or area of the
market will perform better in the near future, and overweight the
portfolio to take advantage of that market segment or segments. The
basis for determining which asset class or sector to invest in or stay
out of can be based on a computer model, economic indicators, or (more
commonly) an advisor's opinion or gut feel.
Most Common Supporting Arguments (some with questionable accuracy):
- The advisor has a track record of picking the winning sectors.
- When in a bear market, it's better to be in bonds, cash, or defensive sectors (e.g. healthcare).
- It is possible to time the market, it's just that most people do it wrong.
My Rebuttal:
- There
are enough advisors trying new things that, statistically, some will be
right on their predictions. When this happens, they get their own radio
show. When they're wrong, you never hear about them.
- Unpredictable events or government intervention can make any prediction completely worthless.
- Overweighting some sectors and ignoring others adds risk.
Semi-Objective Opinion:
In
order to significantly beat the market, you have to take some
additional risk, and this strategy does that. When called correctly,
this strategy can make huge gains. It can also lose a significant
amount of money while everyone else is making money. By picking the
right sectors or asset classes at the right time, it is possible to
make money in practically any environment. However, similar to flipping
a coin and trying to get "heads", I'm not sure past success is a great
predictor of future success.
Buy and Sell Strategies
Buy-and-Hold
A
pure buy-and-hold strategy involves buying a high-quality investment
such as stocks or a mutual fund, and then holding the investment
through highs and lows until either your investment objectives change
or you find out the investment is not as high-quality as you thought it
was. The rationale is that the overall market goes up over time, and
you don't want to miss a big up day in the market by holding cash.
Most Common Supporting Arguments (some with questionable accuracy):
- The
majority of market gains occur on a relatively few number of days, so
if you miss one of these days, your returns will be significantly less.
- "Time in the market" is more important than "timing the market".
- Warren Buffet is a buy-and-hold advocate.
My Rebuttal:
- Missing
the worst days of the market is far better than catching all of the
best days. However, since no timing system exists that misses only the
best days or misses only the worst days, both situations are ridiculous
and using them as arguments stretches the definition of integrity.
- Warren
Buffet does not "buy-and-hold" like you and I would, unless you have
the resources to buy a company, install the management, hold the
management accountable for performance, etc.
When It Works/When It Doesn't Work:
Buy-and-hold
makes money when investments go up, and loses money when they go down.
Therefore, it works well during bull markets and works poorly during
bear markets. For this strategy to continue to work for the next 30
years like it did the last 30 years, you have to assume that
investments will continue to go up like they have during a period of
economic growth that was fueled by the Baby Boom generation, an Energy
bubble, a Technology bubble, and a Real Estate bubble.
Market Timing (prediction-based)
Market
Timing is one of the most loosely-defined terms in the financial
industry. There are many advisors who deride market timing, and yet
routinely practice market timing themselves. Broadly-defined, market
timing is a strategy that makes changes to a portfolio based on
predicted market performance. These changes may involve selling all
investments and moving to cash, or simply adjusting the percentage of
stocks and bonds because of economic conditions or anticipated market
behavior. Prediction-based market timing bases decisions on an
advisor's assessment of future conditions. If high-inflation is
anticipated, investments that hedge against inflation would be added.
If economic contraction is anticipated, an advisor might move to a
heavier cash position.
Most Common Supporting Arguments:
- By
using indicators such as inflation, unemployment, factory usage, etc,
it is possible to anticipate which sectors have a higher chance of
outperforming in the future.
My Rebuttal:
- Economic
indicators work when nothing interferes with them, but unexpected
events such as government action or national conflict override any
statistical probability used for predictions.
- Overweighting some sectors and ignoring others adds significant risk to a portfolio.
When It Works/When It Doesn't Work:
This
method is highly dependent on the person or statistical model making
the prediction. If the predictions are accurate, this strategy has a
good chance of significantly outperforming other methods. If the
predictions are wrong, the opposite is true. Because of the large
number of advisors who make predictions, a certain number will get it
right several times in a row, but statistically this will not indicate
any greater likelihood that they will continue to be right in the
future. As mentioned above, unanticipated news events or government
action will instantly derail most statistical models.
Market Timing (momentum-based)
Momentum-based
market timing uses technical indicators (stock charts and current
market behavior) to determine whether the market is in a downtrend or
an uptrend. Downtrends occur when more people want to sell than want to
buy, and uptrends occur when more people want to buy than want to sell.
Price movement and trading volume can determine whether there is more
buying pressure or more selling pressure at any given time, and the
theory behind momentum is that once a trend is in place, it tends to
stay in place. For how long? Until it stops.
Most Common Supporting Arguments:
- Price
movement and trading volume offer strong clues about buying pressure
and selling pressure, and whether large institutional traders are
buying or selling.
- Institutional traders do not establish
or eliminate entire positions in a single trade, and typically spread
trading over several days or weeks. Therefore, trends tend to stay in
place for some period of time once they are established.
My Rebuttal:
- This makes a lot of sense to me, so I don't typically argue against it. However, it has some weak points (see below).
- Some
advisors can go over-board on technical patterns (head and shoulders,
cup and handle, shallow birdbath with a floating stick...I made that
one up). These advisors are traders looking for short-term movements.
Trends, on the other hand, are determined more by a pattern of
higher-highs or lower-lows, and it doesn't need to be very complicated.
When It Works/When It Doesn't Work:
There are some key components required for this system to work.
1)
The method for determining trends must not be too early or too late.
Stocks seldom move in a straight line. They typically make a strong
move, and then rest or pullback. Assuming too early that a trend is
being established or ending will result in jumping in or out during
pullbacks or corrections. Waiting too long or for too many confirmation
signals will result in missing a good portion of the trend.
2) Investments must be liquid. You must be able to act when your system tells you to buy or sell.
3)
Whether you use Moving Averages, charting, or any other system to
determine a trend, the trend will not always hold. Each system will
break down under certain conditions, so the objective is to use a
system that works under the widest set of conditions and/or breaks down
under the narrowest set.
Market Timing (emotion-based)
This
is not a strategy that is typically planned for or entered into
intentionally, and is the form of market timing most often practiced by
those who swear they hate market timing. Many practitioners of this
strategy consider themselves to be buy-and-hold investors, but they end
up moving to cash when the pain gets too great or the market is too
scary. Typically, this happens after a significant loss is already on
the books, which actually makes this a form of momentum. The rationale
is that if my investments have already lost money, they may continue to
lose money. The problem is that if emotion or fear drives the sell
decision, then the decision to get back in is typically based on
"feeling better", which almost always happens at a higher price than
the sell price.
Most Common Supporting Arguments:
- Not too many people are active proponents of this strategy, but a lot of people practice it.
My Rebuttal:
- Not
much to rebut, other than pointing out that you can't call yourself a
buy-and-hold investor if you move to cash or change your stock
allocation when the market gets scary, and no one should use this
method as an example to "prove" that all market timing systems are
doomed to failure.
When It Works/When It Doesn't Work:
This
strategy seldom works, and is the reason that the vast majority of
investors buy when the market is high and sell when the market is low.
It doesn't matter which strategy you use; just about anything is better
than basing investment decisions on emotion.
Disclosure (my bias)
I
use a Balanced Sector Allocation strategy using low-correlated ETFs,
and momentum-based market timing. The objective is to participate as
much as possible in uptrends, and avoid as much of the downtrends as
possible. This requires a set of rules that makes the decision points
unemotional. A Balanced Sector Allocation guarantees participation in
the hottest trending sector at any given time, but with a mechanism to
get out of a sector when it starts heading back down.
Weak Points:
Because
it takes a little while for a downtrend to show itself, sell decisions
will never happen right at the top of a trend. The same holds true for
uptrends and buy decisions. If the market gets indecisive and swings
far enough that it keeps looking like uptrends and downtrends are
forming but no follow-through happens, a condition could occur where
losses are exaggerated. This would be a very specific and narrow set of
conditions, and I have other checks that attempt to minimize this
condition, but it still exists.