Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


The lunch may be free, but how much is it worth?

by Mark Dow
July 9, 2009

This is Mark Dow. Brad is still away.

Diversification is the only free lunch in investing. But the way many investment managers sing its praises, one can’t help but wonder if its value is overstated (much the way all portfolio managers are programed to say in interviews that “it’s a stock picker’s market”, no matter what the underlying market conditions). Here’s a graph below that speaks a thousand words.

Brazilian Real vs Turkish Lira, over the past year

This is a chart of the Brazialian Real (BRL, in white) vs. the Turkish Lira (TRY, red), over the last year. They have been nomalized to facilitate comparison. Brazil is well known as a commodity exporter, and Turkey as a commodity importer (oil is a huge component of Turkey’s import bill). Keep in mind that over this same period, the front contract on West Texas crude oil fell from USD 136 per barrel to today’s price of 60, and other commodities–both soft and hard–fell by similar amounts.

You will note the one period of divergence in the performance of these two fundamentally distinct countries came in December of last year. This is due to the unwinding of TARKO (Target Knock-Out) contracts that were sold to Brazilian companies ostensibly as pre-hedging vehicles for future dollar export proceeds. However, once they started “working”, these contracts proliferated and quickly became vehicles for specualtion, often by companies that lacked the necessary sophistication in complex derivatives. And had it not been for the coordination of the hedging of these contracts by the Brazilian Central Bank, the December divergence would have been greater. Once this demand for dollars cleared, however, BRL returned to correlate very closely to TRY.

This notwithstanding, the overall similarity of performance over the period would, I think, come as a surprise to most people–given the fundamental backdrop. I know it took the veteran traders I work with by surprise. And while it is true that all correlations “go to one” when markets are under stress, one would still expect to have seen a greater degree of differentiation over the period shown.

A lot of it has to do with fund flows. Funds flows tend to be trendy, with investors getting into and out of strategies–such as emerging market currencies–roughly at the same time. This makes it tough for investors who do their homework and are counting on markets to be rational and efficient. It also makes risk management at least as important as the investment theses themselves.

Once the fund flows come in, the stories that justify the investment follow. Turkey was a great example. Investors at first expressed reluctance to buy TRY with oil going up through USD 100 per bbl. But the story soon followed that made it okay. Strategists and traders started whispering “with oil at these levels there will be large investment outflows from oil producing states, many of which are Muslim. Muslims will feel more comfortable investing their windfalls in other Muslim states, from which Turkey will be a main beneficiary. This will be good for the country and lead to appreciation of TRY”. And thus it was so. An investment thesis was born. Later, after markets crashed and oil had fallen to $40 per bbl., TRY started appreciating again. The story then reverted to the old one we all know and love: lower oil is good for Turkey and for TRY. The Muslim investment thesis had disappeared. In short, when prices go up, a good story will follow.

The bottom line on diversification: some is good, but there is an optimal degree. Too much diversification and you may lull yourself into a false degree of security. If you know your corrleations under both normal and stressful states of the world you can keep things simple and still make the same portfolio bets.


  • Posted by jonathan

    I enjoy your cynicism.

    Disagree there’s an optimal level of diversification. Models which justify that work only as long as their assumptions hold. So one can, at best, IMHO, say that for each market period there is an optimal degree of diversification with a large variety and weighting.

  • Posted by Mark Dow

    I think you may have misunderstood me. I was trying to get the point across that a lot of people–professional investors included–overdiversify, precisely because they are mislead by using models and a rearview mirror. So, I think we actually agree. By saying optimal–a crutch word for many economists, I confess–I am really saying more is not always better.

  • Posted by Brian Shriver

    Your point is well taken. All asset class returns are correlated by their exposure to the same global pool of investors.

    So how do you manage risk without sacrificing returns?

  • Posted by Abnormal Returns

    Another take on the role of changing correlations when both the global economy and the financial system are under stress.

  • Posted by Michael

    “This makes it tough for investors who do their homework and are counting on markets to be rational and efficient.”

    The one and only conclusion that applies to all investors in all markets at all times.

  • Posted by MarcoPolo

    Mark, I’m still on the last chapeter struggling to understand.

    THIS from QB Partners via The Big Picture doesn’t seem to jive with what you were telling us a couple of days ago.

    Care to comment?

  • Posted by mark folsom

    Bob Prechter wrote an article in Barrons 5 years ago describing his “all the same market” theory and your post is a little late but supports his theory. Basically all markets… commodities, bonds, stocks are just responding to the global ebb and flow of liquidity. Massive deflation is working its way through the system and the markets will one day all come down together again just like the end of 2008. Diversification is another media/ Wall Street fraud in this environment. Look at the dismal performance of junk bonds during the bull market of the nineties. Now they track the stock indexes.

  • Posted by Mark Dow

    MarcoPolo – I took a look at the article you referenced. You’re right: It doesn’t jibe with what I wrote at all.

    Their analysis of the modern global transmission mechanism is precisely why I wanted to address this topic. As I said, stories that are simple and intuitively appealing are hard to shoot down–irrespective of how untrue they may be.

    I think the article reflects an anachronistic and superficial view of monetary economics. But, honest men can differ…

  • Posted by Rien Huizer

    The only optimal degree of diversification is ex post..

  • Posted by Rien Huizer


    Pse tell us more about investing. It would be nice to learn how we can get rich (which is Glorious of course).

    Re diversification: portfolio theory does not work in the real world because we do not know with what data to feed the models. Past data work only if we are lucky. So we are just making gambling for someone else’s account a little less transparent and the gambling agent less suable.

    If you have a theory that justifies a given set of covariances and have proof that that theory is valid during a given segment of the future, then it would make sense to see if an optimal portfolio according to that theory and data would beat the risk free rate over that segment of the future. If it did, it would be interesting to understand what market failure was responsible for that anomaly, harvest it and tell no one else…

  • Posted by Aditya

    Great article Mark. I completely agree with you. I think USD and Crude Oil moves have been a lead indicator or trend setters for a large section of equity and commodity markets due to changes in risk aversion. How do you keep track of correlations and which financial indicators do you put an emphasis on?

  • Posted by MarcoPolo

    Mark, (using my spellchecker) transmission mechanism? I thought you were going to tell me that it was about base money not money supply.

    Frankly, I’ve been in the camp with those others. Intuitively, I feel we use the term deflation to mean contraction. Investing for the long term (what is thought of as dumb money) requires understanding this macro – though I suspect I don’t. There is already too much money in the S&P 500, etc. Diversification among those already oversold traditional instruments is my own idea of dumb money. Real diversification is to break the bonds of feudal industrialism and to own/control once again the means of production.

  • Posted by q
  • Posted by Twofish

    Diversification is not a free lunch. Your expected return goes up, but your maximum return goes down. If your bonus is based on maximum returns, then you don’t want to diversify.

    Also real diversification is a *LOT* harder than it sounds. Globalization and the internet have made diversification even harder because now everything is connected to everything else so that when one thing falls, everything falls.

    What’s really funny is if when you hear a thousand people talking about diversification, and then they all end up diversifying to the same thing.