The Myth of Diversification or How to Really Diversify

Includes: BIV, EFA, GLD, LQD, SPY
by: Jeffrey Dow Jones

Those of you who follow us here at Seeking Alpha know that one of our biggest objectives is finding investments that move in different, unique ways. Getting true diversification is one of the most important things that a portfolio manager can do to increase his risk-adjusted performance.

When we were all getting started in finance, one of the things we'd do to get diversification was to look internationally. After all, everybody understands why most domestic stocks move together — they’re all largely driven by performance of the U.S. economy. Everything goes up together in bull markets and down when the economy hits a rough patch.

So the theory went that foreign markets were driven by a totally different set variables. The performance of Europe’s economy and stock market wasn’t so tightly linked to the performance of the United States. It was a different world over there. They were all doing their own European thing; going to Mallorca or buying tiny, weird-looking cars and eating smelly cheese.

But in the last few years, that hasn’t been the case. Nowadays, everything over there seems to move right alongside the stuff over here. As an active participant in the marketplace I’ve felt this correlation intuitively, but check out the following chart which brings this all into stark relief:

Click to enlarge:

Pretty incredible, huh? That's a rolling correlation chart between the U.S. market as represented by the S&P 500 and the rest of the developed world as represented by the MSCI Europe, Australasia, and Far East Index.

Since the financial crisis, international stocks have correlated almost perfectly with domestic stocks. That’s on top of a 15-year trend of increasing correlation. Ugh.

This chart also reminds us that it wasn’t always like that. International markets really used to be different! Before the last decade there wasn’t much correlation at all between the U.S. and foreign stock markets, fluctuating around a relatively low level of correlation.

And it's not just general correlation, it's the size of the movement too: (Click to enlarge)

What’s perhaps even more disturbing is the fact that these markets are correlating almost perfectly during market drawdowns. And drawdowns are exactly when you don’t want your investments to correlate! When one investment is going down, it’s really nice to have others that are going up or at the very least, aren’t going down as much. But with international stocks, not only are they all moving in the same direction, they are even moving in the same magnitude.

Back in the day it used to be that a bear market in another country didn’t necessarily translate to a bear market in the U.S. Nowadays a -40% drawdown in domestic stocks pretty much means a -40% drawdown in international stocks. Most investors are only interested in their final rates of return, and so what’s the difference between an investment in the S&P 500 (via SPY) or the MSCI Europe/Australasia/Far East Index (via EFA)? If you already own domestic stocks, what good is diversifying into international stocks if they do the same thing at the same time and to the same degree?

It gets worse

It's not just the U.S. and major international markets. Everything else is correlating too.

The really sophisticated portfolio managers used to venture beyond equities altogether and make investments in things like commodities, real estate, or even hedge funds. Yale’s David Swensen is the guy that most people credit with the popularization of this model. Many of us in the hedge fund space owe him a debt of gratitude for opening up the eyes of institutional investors everywhere to the world of alternative investments. Indeed, back at UCLA we even used his book as a textbook for one of my upper division economics classes.

Diversifying into totally different asset classes worked for Mr. Swensen for a while, but in the last few years, even these alternative investments have correlated with traditional ones.

Check out this correlation matrix. The darker green a cell is, the higher the correlation coefficient.

Click to enlarge:

If you're a portfolio manager, that matrix probably gives you the heebie jeebies.

Since 2008, the Dow, the S&P, technology stocks, small cap stocks, Japan, China, the BRICs(!), real estate, and hedge funds, have all correlated tightly. Even crude oil and commodities, the once “great diversifier,” are a little correlation of movement in this modern marketplace.

One thing, however, has done a pretty good job of not correlating. Gold.

Click to enlarge:

Gold doesn’t correlate with anything and never has.

Now, I'm already on record around these parts as believing that gold is in a bubble. That's not to say that it won't keep going up but that the current fair value of gold is very difficult to justify. Also, there are a lot of people playing around in gold right now who have never ridden through one of gold's epic corrections. All of you golden old-timers out there just chuckled knowingly. "Heh, rookies."

But still, gold is worth owning for one important reason: It's different. Last year I wrote about the six things that everybody needs to know about gold right here.

So the specific risks in gold and the gold ETFs (NYSEARCA:GLD) at this juncture make our quest for uncorrelated assets even more difficult. Perhaps if we understand what is causing this phenomenon, we might be able to work ourselves towards a solution.

Why is everything correlated?

I think a lot of it is attributable to the increasingly global economy. Our economy today is much more directly linked to China’s economy, to Japan’s economy, to Europe’s economy, and even emerging economies. If your company slows down and you lose your job, it hurts businesses like Target (NYSE:TGT) because you spend less money on the goods they sell. It means you don’t buy that new house, which means less new home construction is needed. That means that builders don’t buy the copper they’d need to build it, which hurts copper prices and copper exporters like Australia. You have to erase any previous dreams of buying a new TV from Japan or getting a new iPad that’s made in China. The Italian leather handbag that your wife wanted for Christmas? Forget about that, too.

When our economy slows down, we buy less stuff from other countries. That slows their economies down and decreases the amount of stuff they buy from other countries. It’s a chain reaction. On top of that, nearly every major corporation today has a global footprint, whether their labor is sourced overseas or their products are sold overseas. These are the fundamental side effects of globalization.

I think there’s another cause, and it’s one that is more difficult to pin down quantitatively. Over the last decade — and the last several years in particular – the market has become increasingly speculative. There are a lot of reasons for this.

To be sure, there are demographic trends at play, an increasing number of baby boomer retirees who are trying to engineer their portfolios in such a way that they deliver specific rates of return. One of the dominant investment themes to emerge from the legion of boomers (and their generational predecessors, the Silent) is an altered perspective on savings and retirement. Unlike the years before World War II where people basically saved up a big pile of money and then lived off of that money or moved in with their children after they were no longer capable of physical labor, modern generations have redefined what it means to be retired.

Nowadays, it’s about making specific investments, investments that grow over time and theoretically grow large enough such that at some magical point, they generate enough annual income to independently support an individual’s lifestyle. Ideally, that income will be further supplemented by things like a pension plan. Nevermind the fact that these are dependent on the exact same principle, this wonky notion of earning a reliable rate of return that throws off enough annual gains to pay all plan participants their entitled benefits.

A lot of it comes from speculation by institutional investors as well. These guys are also tasked with building portfolios that generate specific rates of return. Pension plan managers who manage trillions in assets are running around the world pulling out of the same markets at the same time and chasing identical other markets. By definition, it is speculative behavior. Active asset allocation is a necessarily speculative endeavor; it requires a manager to make a forecast about the most appropriate assets to allocate to at any point in time.

The problem is further exacerbated by the fact that these guys all are familiar with Markowitz’s Modern Portfolio Theory. It's like they're zombies that are starving for investments that don't correlate. When they see something moving in a different fashion, they all shuffle in for the kill. And that spoils all the wonderful lack of correlation that used to exist. The previously non-correlating asset gets zombified just like everything else.

All this co-movement has boiled down to pretty much one single trade: risk on or risk off. Investors of all shapes, sizes, and locales are either buying risky assets or they are selling them.

“Risk On / Risk Off” makes for difficult investing because it requires proper timing. I speak from experience when I say that timing is the most difficult thing to get right in this business. Some say it’s impossible.

Is there a way to build a portfolio with lower risk and higher returns that doesn’t require perfect timing?

The Solution

For me, the solution is relative value.

In a world where everything goes up and down together I think that being able to determine the degree to which one asset will outperform or underperform another asset will be a valuable skill. It’s tough for investors to make a buck or design a really good portfolio when very few assets move independently of everything else. It’s tougher still to differentiate yourself when the choice is simply whether or not to take risk. Most professionals that I know are interested in much more than this frustratingly abitrary yes/no question. They want to construct whole portfolios that are better than their individual parts.

I think you can do this with relative value strategies. The HFR Equity Market Neutral Index only has about a 33% correlation with the S&P 500 over the last decade. One of the major reasons why these don't correlate with the market is because any long/short relative value strategy has a certain degree of market neutrality built in.

I’m also of the belief that the yes/no question of whether to put risk on or take risk off is better avoided altogether. Assemble a big long/short portfolio like we discussed in The Trade of the Decade and then you won't have to worry so much about correlation and exactly when to buy and sell.

It won’t always be this way. I don’t think this condition of global correlation persists forever. At some point certain asset classes and investments will decouple from the rest of the train. You’ll start to see different assets behaving in different ways and you won’t see the generally high level of volatility and skittishness that dominates the current environment. And that’ll be a good time to cycle back towards directional strategies. I haven't a clue when that'll happen.

Other options

The bad news is that relative value strategies may be easy for a hedge fund like us to implement but it's difficult for the average investor at home. As an example, our market neutral strategy typically has between 120 and 180 different trades on at any given point in time. Even if the investor at home had the proper signals and algorithms to identify the right trades, a portfolio of that size and complexity requires a professional-grade infrastructure.

The good news is that there are a few more things you can do from the safety of your own living room. They will definitely help you lower your correlation. Though I'll warn up up front: They aren't very exciting.

Cash. One answer is to hold more cash. Cash never correlates. I always get two common objections to the suggestion to hold cash, objections which happen to contradict one another.

  1. “Cash pays me nothing!” It’s true. The Fed, in its efforts to get people to borrow and spend money, has slashed short-term rates to zero. Sorry, savers. I’m not sure why the Fed hates you so much. In the mean time try and find solace in the fact that the prices of many things are going down and your purchasing power remains strong.

  2. “I’m worried about inflation.” You might be worried about inflation, but the markets sure aren’t. Nobody is truly afraid of The Inflation Chupacabra right now. And it won’t sneak up on you in the middle of the night, either. The 30-year bond or the TIPS/Treasury spread will sound an alarm.

Investment grade corporate debt. With bonds, it’s important to emphasize the investment grade. It is junk bonds that are forming a bubble because it is their higher yields that everyone is starving for. Don’t fall for that trap. Junk bonds tend to correlate with stocks and they won’t help you during market dips. High-quality corporate debt, on the other hand, doesn’t correlate at all with stocks. Since 2008 the correlation coefficient with the S&P 500 is 0.05. Since 1996 the correlation coefficient is -0.01. The R-squared is 0.00! If you’re a professional portfolio manager, those kinds of statistics probably make you squeal with joy.

Again, the problem here is that this kind of stuff only yields about 3-4%. You can check out the iShares Investment Grade Corporate Bond Fund (NYSEARCA:LQD) or the Vanguard Intermediate Term Bond Fund (NYSEARCA:BIV). Those each yield around 4% and have stuff that's mostly A-rated or better. There are also about a billion mutual funds that are functionally identical.

Treasuries. Treasury bonds don't correlate. In fact, they usually exhibit a negative correlation with the stock market. The correlation coefficient between the S&P 500 (via SPY) and the iShares 7-10 Year Treasury Bond Fund is down around -0.5

The downside? Treasuries are booorrr-ing. But I'm boring and I love boring, predictable things and I understand that you may not feel the same.

A final word

It is possible to find stuff that doesn’t correlates but it means making peace with a lower yield. It means owning a bunch of the “risk off” stuff.

Right now we're in a cyclical bull market in the midst of a secular bear. I talk about it all the time: Bear markets are not about generating above-average returns. The risks are unbalanced. Bear markets are about protecting the gains that you made during the last bull market and they’re about staying intact to participate in the next bull market.

Do that, and you’ll have what it takes to go pro.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.