Are You Really Diversified?

Photo credit: steelo

Sometimes we fool ourselves. Sometimes we think we’re doing the right thing, when in fact the result is that we’re not at all doing what we think we are.

I’m talking about your investment diversification. Within your 401(k) you have certain options available for you to choose from: a large cap stock fund, a mid cap stock fund, an international stock fund, and a bond fund. Recalling an article you read somewhere… you know you need to split up your investments among many allocation options. So, wanting to do this diversification thing right, you split up your 401(k) contributions with 25% in each of the funds available. You’re well-diversified, right?

Wrong city, bucko.

Correlation

Welcome to correlation. Investopedia defines correlation as:

a statistical measure of how two securities move in relation to each other.

It’s pretty complicated, but the gist is this – if two securities are perfectly correlated, when one moves up or down, the other moves up or down in perfect relation to the other. Such securities are said to have a correlation coefficient of +1.

On the other hand, if one security moves up and the other moves down (and vice versa, by the same proportions), they are said to be negatively correlated, with a correlation coefficient of -1.

Lastly, if one security’s movement has no relationship whatsoever to the other security – that is, any particular movement by one of the securities may or may not result in a movement in the same direction, the opposite direction or no movement at all. These two securities have a correlation coefficient of 0 (zero).

Most pairs of common securities fit somewhere along the spectrum between +1 and 0, since very few are perfectly correlated. Negative correlation is typically found in hedge funds – which are a costly, complex sort of asset to hold, being designed to work opposite of the general market movements. Since long-term stock market movement is in a positive direction, many hedge funds are “hedging” that the opposite will occur.

With the above explanation, hopefully it becomes clearer to you why we want securities in our portfolio that are not correlated closely to one another… having such pairs of securities spreads out our risk of any single market event having adverse impact on everything in our portfolio.

Examples of Correlation

Back to our example portfolio, here are the hypothetical correlation coefficients* for your four choices, shown in a matrix:

*Note: These hypothetical correlation coefficients are for illustration only, but were accurate at one time, but they are subject to change over time. In addition, the specific makeup of each fund will produce a different correlation coefficient versus the other funds in the real world. Use one of the many tools available on the internet to get a handle on the coefficients for your chosen funds.

As you can see, the large cap, mid cap, and international stock choices are very closely related to one another. That’s why, even though you thought you were well-diversified during the market slump a couple of years ago, everything you had took a dive.

Note how the Bond Fund is far less correlated to the to the stock funds. Each of the correlation coefficients is less than 0.5. The large- and mid-cap are nearing 0.25, coming very close to the 0 of perfect non-correlation.

This is why the first, most important allocation choice you can make is between stocks and bonds (we’ll get to some other allocation options later). These two, of the choices you have, are the least correlated, so it’s very important to include these non-correlated assets together in your allocation scheme. And then within your chosen split into stocks, you can choose some of the other asset options – large cap, mid cap, small cap, international – since those assets aren’t perfectly correlated, it can be beneficial to include diversification among these options as well.

The same goes for bonds – other types of bonds, such as Treasury Inflation-Protected bonds, are not perfectly correlated with the total bond market, so it might make sense to include some of these as allocation options as well.

What about other types of assets?

We’ve talked about some very basic allocations – but what about other types of assets? There’s real estate (both domestic and international), emerging markets stocks, commodities, and others. How does the correlation of these assets look?

The table below details the hypothetical correlation matrix for these additional assets in relation to domestic stocks, international stocks, and bonds.

As you can see in the matrix, adding these additional asset classes gives you even more diversification (per the correlation). Commodities show up as the next most non-correlated to stocks (after bonds), which explains why this is a popular asset class to consider. Not only are commodities not well correlated with stocks, they are even less correlated to bonds.

Real estate, both domestic and international, gives you additional diversification, but not nearly as much as bonds and commodities – turns out that real estate, while not a perfect match for stocks, does follow the movement of stocks somewhat closely.

How?

You might be saying “but I don’t have those kinds of options available in my 401(k)” – what can you do? This is part of why it can be useful to have other savings plans in your scheme, such as a Roth IRA or a taxable account. With these other accounts, you can have the flexibility to invest in whatever asset classes you like.

Investing in multiple asset classes has the effect of lowering the overall risk in your portfolio, while (potentially) enhancing the return, more than just averaging the returns together. Since your well-diversified assets do not move in direct relation to one another, when the domestic (US) stock market has a downturn, your other assets don’t necessarily reflect that same downturn, buoying your overall return.