Diversification’s Role as a Risk-Reduction Tool

by Charles Rotblut, CFA and Richard Bernstein

Diversification’s Role As A Risk Reduction Tool Splash image

Richard Bernstein is the chief executive officer of Richard Bernstein Advisors LLC. We spoke recently about diversification and asset allocation.

—Charles Rotblut

Charles Rotblut (CR): You sub-advise a mutual fund that can vary widely in terms of its allocation to bonds and stocks. How do you determine what the right asset allocation is?

Share this article

About the

Charles Rotblut is a vice president at AAII and editor of the AAII Journal. Follow him on Twitter at twitter.com/CharlesRAAII.
Charles Rotblut Profile
All Articles by Charles Rotblut

Richard Bernstein is the chief executive officer of Richard Bernstein Advisors LLC.
Richard Bernstein Profile
All Articles by Richard Bernstein

Richard Bernstein (BR): There are two mutual funds we run. Eaton Vance Richard Bernstein Equity Strategy fund (ERBAX) is a global equities fund, and we have great latitude in that we can go anywhere around the world. Eaton Vance Richard Bernstein All Asset Strategy fund (EARAX) is probably the one you are referring to: It is a global, go anywhere, all-asset fund.

How do we set the asset allocation? Well, I was at Merrill Lynch for over 20 years and one of the things we used to talk a lot about at Merrill was that people never understood diversification. They see charts and they are told that if they diversify their portfolio, they will get less risk and higher returns. What we always used to tell people is that diversification is not a return-enhancement tool; it is a risk-reduction tool. If you happen to get higher returns, then that’s gravy; that’s the icing on the cake. But really what diversification is all about is reducing the volatility of your portfolio. So that’s one thing.

SPECIAL OFFER: Get AAII membership FREE for 30 days!
Get full access to AAII.com, including our market-beating Model Stock Portfolio, currently outperforming the S&P 500 by 2-to-1. Plus 60 stock screens based on the winning strategies of legendary investors like Warren Start your trial now and get immediate access to our market-beating Model Stock Portfolio (beating the S&P 500 2-to-1) plus 60 stock screens based on the strategies of legendary investors like Warren Buffett and Benjamin Graham. PLUS get unbiased investor education with our award-winning AAII Journal, our comprehensive ETF Guide and more – FREE for 30 days

Number two is if that’s the goal, then how do you reduce the volatility of your portfolio? Well, you have to look at the correlation among the asset classes. And it’s not the number of asset classes that determines the diversification; it’s the correlation among those asset classes that determines diversification [Figure 1]. Somebody right now could say, well, I have U.S. stocks, non-U.S. stocks, corporate bonds, high-yield bonds, non-U.S. bonds, commodities, gold, private equity and hedge funds, but there is no diversification in the portfolio because those asset classes named are very highly correlated. So it’s not the number of asset classes. You could technically form a very well-diversified portfolio with just two asset classes if their correlations are negative.

What we like to do is to think of diversification as a seesaw. Some assets go up and some assets go down; the fulcrum of the seesaw is the volatility of your portfolio. So you have to try to figure out which side of the seesaw your assets are on. For a long time, you could actually balance out a portfolio by kind of an even number of asset classes. The problem today—and for the last five or six years—is that the seesaw has become very skewed. It is completely imbalanced. What is happening is that the correlation of all of these asset classes is going up. Really, the only asset class sitting on the other side of the seesaw is basically the Treasuries. What we do is try to find out the correlations in all of these different asset classes, how much risk and how we want to tilt that seesaw. That’s kind of what we are doing.

The final point is: How do you measure the correlation? Honestly, you can make correlation say anything you want, depending on the time period you use. “I want to use daily data for three weeks,” or “I want to use monthly data for five years.” Basically you can make correlation say anything you want. If you are a commodity manager, you can show that commodities are negatively correlated—all you have to do is find the right time period. We look at what we call structural correlations. Structural correlations are the long-term trend in the correlation. So we are not looking at all the ups and downs. It’s literally like watching paint dry. We can measure true underlying longer-term correlations, and they don’t change very frequently. That’s what we use and that’s how seven years ago we began to notice that the seesaw was becoming very unstable. And I think people have learned that it has become very hard to diversify portfolios. That’ll change, obviously. As those correlations change, we will change our portfolio.

Charles Rotblut, CFA is a vice president at AAII and editor of the AAII Journal. Follow him on Twitter at twitter.com/CharlesRAAII.
Richard Bernstein is the chief executive officer of Richard Bernstein Advisors LLC.


Donald Marburger from TX posted over 3 years ago:

The chart showing correlations is quite revealing. I would like to have seen more thoughts and insight into why there has been such a shift.

I personally believe we have been living in an artificial market ever since QE started and will continue to do so until there is no more QE.

Marvin Menzin from MA posted over 3 years ago:

I dont think 5 year covers a full market cycle-- eg in july 13 it leaves out the huge fall in last half of 08
i prefer a full cycle- peak to peak-- so oct 2007 to oct 2013 if its still do high is a good measuring time.
also chart did not include US
stocks -- small mid, large, etc - article too long and rambling--nature of interviews I guess - thx

Fernando Robles from FL posted over 3 years ago:

Spot on investment advice, grounded in financial theory. Excellent!

James Jennings from VA posted over 3 years ago:

What the Lord don't taketh away, he don't giveth either.

Steve Daniels from CT posted over 3 years ago:

I read the interview with Richard Bernstein on Diversification's Role as a Risk-Reduction Tool and thought it was quite good for the most part. However, I felt the article fell short on two issues:

--it failed to point out the fact that correlations across asset classes tend to increase in declining markets which makes diversification harder to accomplish.

--it also tended to ignore correlations of the S&P 500 with alternative investment vehicles (e.g. managed futures, long/short strategies,real estate) which in many cases show lower correlations and, therefore, provide greater diversification benefits.

Sam Wilson from TX posted over 3 years ago:

It certainly is difficult to know the answers to all of these questions. Many of us have our own criteria for finding value stocks. To balance that out on the see saw, one can simply go with the three month CD.

Mark Marotta from NJ posted over 3 years ago:

Great article. Rich Bernstein, again, does a terrific job. Thanks Rich and thanks AAII.


Thomas Scheller from MN posted over 3 years ago:

This articles is really good in many respects, emphasizing that diversification is not about about quantity but quality of correlations, and total return being key, not yield.
There is one disturbing misconception:
That negative correlation equals well diversified. It is easy to set up 2 assets that are perfectly negatively correlated with each other - buy the S&P500 and simultaneously short the S&P500. They have (almost) exact negative correlations, and the net effect will be that you have zero volatility and ZERO return.
The goal should be to find assets that have low (ideally zero) correlation, not negative correlation. (They also need to individually have positive return, otherwise keeping cash is better.)
Taking figure 1 alone (and without important info like returns), it would be better to pair gold (correlation=0.3) or long-term treasuries (correlation=-0.4) with the S&P500 than 3-month T-bills (correlation=-0.7).

Mercere Collins from TN posted over 3 years ago:

What am I missing here? Seems like the positive correlation items went up because the Fed lowered rates, and vice-versa.

If we just knew in advance what the Fed was going to do we could all have perfect correlations with growth as well.

As he said, back testing can produce whatever the back tester wants.

L Baer from MD posted over 2 years ago:

Very logical and interesting. Please give us a more comprehensive AAII article on this use of negative correlation to balance my portfolio.

You need to log in as a registered AAII user before commenting.
Create an account

Log In