When constructing a portfolio, you need to focus on two things: return and risk.

Suppose you have a portfolio of mutual funds, and are considering adding another fund to the mix. The decision to add or not hinges on a single question; “What contribution will that additional fund make to the portfolio in terms of return and risk?”

The return side is relatively straightforward. If the fund being added has a higher return than what the portfolio is already making, then adding the fund should incrementally increase the overall portfolio return. In this sense, the new fund’s higher return should pull up the average.

Alternatively, if the new fund has a lower return than the portfolio, it will act as a drag on performance. So you wouldn’t buy it. Or would you? 

Within the context of a portfolio, you have to look beyond return. Equally important, is what if any incremental effect will the new fund have on the risk of the portfolio. If the new fund can reduce the risk within the portfolio to a greater extent than it dampens return, then maybe it is a good addition. 

When adding or subtracting securities in a portfolio, you need to understand what effect that decision will have on both return and risk. 

The Return Effect

The impact a new inclusion has on a portfolio’s return is proportional to the weighting that the asset class or fund represents in the portfolio as a whole. This return effect is the same whether it is equities, fixed income, cash, or any other asset class. In technical terms, the portfolio return is simply a weighted average of the returns of the securities in the portfolio, where the weights are the percentages each holding represents in the portfolio.

Suppose for example, a portfolio had 50% in fixed income securities and 50% in equities. The fixed income securities generated an 8% return, the equities a 12% return. The portfolio would have a return of 10%, calculated as (0.50 x 0.08) + (0.50 x 0.12) = 10%. 

If we assume for this exercise that historical performance is repeated, then if we increase the weighting of fixed income securities to say 60%, the portfolio return would be less than 10%. Calculated as (0.60 x 0.08) + (0.40 x 0.12) = 9.6%. 

The Risk Effect

The risk effect more complicated. It relates to how much diversification effect a new inclusion offers. For example, fixed income securities have a different sensitivity to various market forces than do equities, and equities respond differently to those forces than cash does. 

For the most part, interest rates drive fixed income values. Higher rates, lower bond prices, and vice versa. Equities are also affected by interest rates, but by a much smaller degree. Equally important for equity assets are company fundamentals such as earnings growth, revenue, debt to equity, etc. 

A portfolio that holds both equity and fixed income, benefits from the diversification effect. The equity holdings insulate the portfolio somewhat from interest rate shocks, and bond holdings will shore up the portfolio in the event that company profits fall. Cash holdings are relatively insensitive to both changes in profits and changes in interest rates, and so cash offers some diversification effect as well.

When the returns on different asset classes respond differently to the same influences, or respond to different influences altogether, that’s where the diversification effect comes into play. If your investments are spread over a number of different risks, then no one factor can take a large bite out of your portfolio. 

Taking that to an optimal limit, you can diversify a portfolio across such a wide array of risks that there is no great exposure to any one factor. In that scenario, the only way your portfolio could take a devastating hit is if all of the risk factors ganged up on your portfolio at the same time. Not likely, short of the world coming to an end. And if that happens, you’ll have bigger worries than what your portfolio is worth!

Measuring the Diversification Effect

Intuitively diversification reduces risk. But what about the specifics?

In technical terms, the diversification effect occurs when the returns on different asset classes are not perfectly positively correlated. This means that returns move at different speeds and sometimes in different directions, because they respond differently to the same risk factors, or to different risk factors altogether.

We measure that through a correlation coefficient, which effectively indicates the relationship between two securities. A correlation coefficient can have a value between +1.00 and -1.00. The former reflects perfect positive correlation, meaning both securities move to the same degree at the same time, which means no diversification effect whatsoever. 

The latter, indicating perfect negative correlation, where two securities move in exactly the opposite direction by exactly the same degree all the time. If you had a portfolio of two perfectly negatively correlated securities both exhibiting positive long term returns, the portfolio should produce a steady positive return with no risk (Risk being defined as the variability of return).

Obviously there is no such thing as perfect negative correlation across assets with long term positive returns. Hence we attempt to build the diversification effect within a portfolio by searching for assets with correlations between +1.00 and –1.00. The closer the coefficient within the portfolio gets to –1.00 the more diversification effect within the portfolio.