The resources on InvestorCraft.com provide information
that can help you to evaluate if a portfolio of investments is well diversified, but why is it that
diversification is so important? The
answer can be captured in one word: risk.
We, as investors, are risk averse.
We don’t like to take undue risks.
Diversification, quite simply, is a technique for reducing risk when it
comes to investing.
We have all heard that it is un-wise to put all our eggs in
one basket. That’s because if something
goes wrong with that one basket (i.e. stock), then we are left with
nothing. By diversifying, we are taking
this concept and extending it. The point
of diversification is to build a portfolio that has multiple assets that are
not highly correlated with each other. Assets
that are not highly correlated do not track each other closely, whether it is
during up or down market periods. By
building a portfolio that has a mix of assets that are dissimilar, we can
protect the total portfolio by reducing its volatility and risk. So correlation,
or the lack of it, is the key to building a diversified portfolio that reduces
risk.
To better understand how correlation plays such an important
role in diversification, let’s take an example.
First, let’s assume we have 2 securities named A and B. Assume the expected return and standard
deviation ( a measure of risk/volatility) of these two
securities are as follows:
|
Security
|
A
|
B
|
|
Expected Return
|
10%
|
15%
|
|
Expected Standard deviation
|
15%
|
25%
|
Now, let’s consider two scenarios. In Scenario 1, let’s assume that A and
B are perfectly correlated, meaning they are so similar that they always move precisely
together in the same direction. If we have a 50/50 mix of A
and B in a portfolio, the expected return is simply 12.5%,
and the expected standard deviation is 20%.
If we plot Expected Return vs Expected
Standard Deviation
on the graph below, the various possible portfolio weightings of A and B fall
on a straight line connecting the A and B points above. Because of the perfect correlation between A
and B, mixing assets A and B does not effectively diversify the
portfolio.
In Scenario 2, let’s assume that assets A and B have a
non-perfect correlation. Correlation can
be measured by a statistic called the correlation factor. In this case, we will assume the correlation
factor between A and B is 0.5. This
indicates the two assets are still somewhat correlated, but not highly
correlated. So sometimes when the price
of A goes up or down, the price of B may not move in
the same
direction or by the same percentage. If
you have a 50/50 mix of A and B, in this case, we can show mathematically that
the overall portfolio expected return is still 12.5%. However, the standard deviation of the
overall portfolio is 17.5 % versus the 20% of the first scenario. Mixing assets A and B actually adds value by
reducing the volatility, or risk, associated with the portfolio. As investors, we like this. By mixing the assets in a 50/50 mix, we have
reduced the expected volatility of the portfolio, thereby improving our
expected reward/risk ratio. Since we are
risk averse, this is a good thing! In
the graph below, the green curved line represents the various possible portfolio
weightings of A and B, and the expected returns and standard deviations of each at a correlation
factor of 0.5. The lower the correlation factor,
the more convex is the bowed line, as demonstrated by the curve for the
correlation factor of 0.25. The more convex the line, the better the reward/risk ratio. That is
because for a given level of expected return (y), the expected volatility (x) is
lower.
This concept can be further extended to portfolios with multiple
assets. The number of possible portfolio
combinations is increased, and the math is more complex, but the concept is the same. The more diversified the portfolios, as
indicated by weaker correlation factors, the more convex is the curved line
that represents the best expected return for each possible value of standard deviation.
So, we can extend this concept to a portfolio with many
more assets than just our securities A and B.
Our goal is to build a portfolio with a mix of assets that each
individually have strong long-term expected
performance, yet that are not highly correlated with each other. Usually, such securities full into different
asset classes (stocks vs bonds, domestic vs international)
or different
industries (energy vs technology).
InvestorCraft has two key tools
that you can use to help you to (1) understand how well a portfolio is
diversified and (2) understand what
can be done to achieve a higher level of diversification. These two
tools are the
Correlation Analysis
and the Efficient Frontier. The Correlation Analysis is very simple in
nature. It looks at historical price
performance and reports a correlation factor for each pair of assets in the
portfolio. Beware of portfolios that
have too many correlation factors close to 1.0, because such a portfolio has
assets that
have not historically added the value of diversification. The next tool, the Efficient Frontier, is an
advanced optimizer that calculates expected return and expected risk for all
possible asset weightings in a portfolio of selected securities. It does this based on historical price behavior. It plots the Efficient Frontier (curved line)
representing those portfolios that have the best expected return for a given
level of volatility (standard deviation), and selects the one that maximizes
the return/risk ratio.
You can use these two tools to analyze a variety of possible
portfolios, studying the Correlation Factors and Efficient Frontier curves for
each. Doing so will help you to better
understand the level of diversification of your portfolio.