Modern Portfolio Theory is Dead…?

At a recent luncheon, I watched the Chief Investment Officer of Northern Trust make arguments that seem all too common this year.

His main point was that using Modern Portfolio Theory (MPT) is “the old way of doing things” because in 2008, assets that should have zig-zagged independently all dropped last year.  The implication, of course, is that diversification failed when it was needed most, and that asset allocation doesn’t work. 

I find this to be a typical gimmick to lure people (new customers) who got slaughtered by investment markets last year and are searching for better ways to invest their money.  But allow me to pick apart their main arguments:

Their Point: Correlations are not stable and go to one (1) during a crisis; so regardless of whether you have real estate, international stocks, or small cap stocks, it will not do you any good. 

My Counterpoint: Many studies show that this conclusion is not true.  Work by Loretan & English as well as Kim & Finger and Ragea show that higher correlations are simply a result of higher volatility, not because of some “contagion” spreading across equity investments; that correlations apparently increase, but don’t actually change.

If they really believed this claim, their portfolios would only include two assets, the market index (S&P 500) and US treasury bills; however, their sample portfolio included allocations to real estate and hedge funds.

Their Point:  You should create a cash flow immunized portfolio to minimize the potential shortfall.  In other words, treat your portfolio like a pension fund.

My Counterpoint:  This strategy may be useful in many situations, but equally inappropriate in just as many others.  If you knew with a high degree of confidence that the objectives, timing, and risk tolerance would not change over time, or with changes in market conditions, then you have a much better case with the immunization strategy.  The reality is that life happens, goals shift, tax laws change, and among other things, people change.

Their Point:  Diversificaton failed, MPT is dead. 

My Counterpoint:  Modern Portfolio Theory has long been abused by many practitioners; particularly by those who rely on mean-variance optimization software – as if mean returns and variance are the only important factors?!  What does it say on all investment literature? Past performance is no indication or guarantee of future performance.  Yet, many practitioners plug in historical return information into their software to finalize an asset allocation model:  Garbage in, Garbage out.

In the end it’s just a theory; so what happens in practice?  When the father of MPT, Harry Markowitz, was asked how he would invest among stocks and bonds, instead of making intricate calculations, he replied, “My intention was to minimize my future regret… So I split my contributions 50/50 between bonds and equities.” 

Markowitz chose to minimize his regret instead of minimizing portfolio variance!

Money Fund breaks the buck!

The real headline investors should pay attention to is slipped into a side column in today’s WSJ.  A large New York based money market fund manager “broke the buck” on a few funds including the Reserve Primary Fund (RFIXX).  This is the first time this has happened in 14 years!  Virtually anyone with a brokerage account has some kind of money market fund in their account where uninvested cash is held.

The mandate of money market funds is to maintain a $1 price (net asset value) every day and earn some interest buying short term corporate debt, CDs, treasuries, etc.  Funds run by the above mentioned manager are now worth less than $1 — thanks in part to Lehman’s collapse yesterday.

In recent times, many of these so called “conservative” funds have had to dip into less secure debt in order to pump up their yields and attract more investors.  In the past, a fund or two have had the parent company (like the big institutions that are failing at the moment) pump in some cash to avoid breaking the $1 threshold.  There was no bailout for the folks at the Reserve and their investors.

The lesson here is that not all money funds are alike — if you’re really worried about this happening to you, try a money market fund that only invests in US treasury securities.  You may get a lower yield, but at least you’re going to be in a default risk free asset.

Some comments on P/E ratios and valuation

A recent article in the WSJ asserts that if you use the Price/Earnings (P/E) ratio as a general guide, stocks have gotten expensive. Although the author points out that that looking solely at P/E ratios might have some pitfalls, he doesn’t seem to really back up that statement.

He points out that the 60 year average P/E ratio for the S&P 500 is 16x. When you look at the earnings for the last 12 months, the P/E ratio jumps to 21x. Thus, stocks must be “getting pricey”.

I think most savvy investors take this assertion lightly; however, others should be conscious of the fact that one line generalizations about the valuation of the Stock Market are dangerous.

A different way to look at the P/E ratios referenced in the article is one attributed to Nicholas Molodovsky called the Molodovsky effect. He posits that in mature, cyclical industries, the P/E ratios will actually go up in bad times… and P/E ratios will go down in good times. It seems counterintuitive because most people agree that low P/E stocks are “cheap” and high P/E stocks are “expensive”. Once you think P/E ratios trending up or down over time instead of a snapshot of value, the effect makes some sense, but here’s a hypothetical example:

MG Motor Company sells much fewer cars during a recession than when times are good (all things being equal), so it’s certainly in a cyclical industry since company earnings are tied closely with the economy. Let’s assume their stock is priced at $30 in 2006, and $23 in 2007. Earnings go from $1.50/share in 2006 to $.90/share in 2007. The 2006 P/E is 20x and jumps up to 26x in 2007.

In this example, the earnings decline boosts MG stock into “expensive” territory, even though the share price dropped $7. This is because the percentage decrease in the denominator (earnings) is greater than the decrease in the numerator (price) from 2006 to 2007.

The point here is that one line arguments of valuation should at least be narrowed to a particular industry and not applied market wide. Very often there are opposing interpretations of commonly accepted measures of value such as P/E ratios — depending on the assumptions made, you could make a case that stocks are expensive or cheap!

Beware the returns of March (and beyond)

As the first quarter of 2008 comes to a close, it is now a common time for people to check their investment performance and investigate new investment opportunities. If you look at the investment performance of various mutual funds (e.g. US large cap), many of them had terrible performance in 2001 and 2002 (even when compared to what we’ve experienced these past few months).

Beginning with 2008, advertised 5-year performance figures will not include these “spurious” old returns. Caveat emptor…

The Dow is up, down, sideways?!

Time and again you hear on the radio or TV, the market is up 100 points, down 250, “hitting a ceiling at 13000″, or some other jargon — what does all this mean and should it mean anything to you?

Often the media refers to the Dow Jones Industrials Index when they refer to the stock market. I think this is very misleading, and serves their interests better than it serves the general public. Here’s why:

The “Dow” is a price weighted list of 30 arbitrarily chosen stocks that are supposed to represent the stock market. This price weighting means that McDonald’s has more influence on the index movements (recent price $56.05) than Microsoft (recent price $27.84) — even though Microsoft is a $265 billion dollar company and McDonald’s is a $67 billion dollar company by market cap. Strange isn’t it? There are several other problems with the index cited in this paper.

The index itself is very easy to calculate at various times throughout the day. This is why radio and TV loves it so much – it’s a catchy story to tell about the minute by minute changes in the “market”.

When the Dow Industrials Index was created over a hundred years ago it may have provided a good representation of market performance. Now that many investors have globally diversified portfolios in investments of various weights and market capitalizations — and are interested in total returns, the Dow should no longer be the benchmark of choice.

Rates of Return: Who’s counting what?

Many of you may remember the Beardstown ladies, a 14-member investment club from Beardstown, Illinois, who wrote books about how their common sense investing style made them amazing returns (see an old Time article on the subject).  It turns out their published 23.4% returns were actually 9.1% after an accounting firm audited their records!

Better yet, what about those cable shows called something like “Flip that house” where people claimed ten’s or hundreds of thousands of dollars in gains. 

Why aren’t people like us making that kind of money?!

I think there are many good take-aways from these examples but here are two: 1. Don’t trust rates of return people brag about; and 2. Most people don’t know how to calculate their actual rate of return. 

Let’s take a hypothetical house flipper who buys a run-down home for $200k, fixes it up with $60k over three months, then lists and sells the place in 90 days for $360k.  Sounds like an easy $100k doesn’t it?  You might say that’s a 38% return ($100k profit / $260k investment), but that’s not the entire picture.

Lets assume they put 20% down, get a great interest only mortgage rate pegged at 5%, pay closing costs of 7%, and pay cash for the upgrades.  Lets do the math and look at the cash flows:

1. In Jan, they pay $40k cash down payment, $20k house upgrades and interest of $667 ($160k loan * 5% / 12 months).
2. In Feb, they pay $20k house upgrades and interest of $667.
3. In Mar, they pay $20k house upgrades and interest of $667.
4. In Apr, they list the home for sale and pay the interest of $667.
5. In May, they pay the interest of $667.
6. In Jun, they sell the home for $360k, payoff the $160k mortgage, pay closing costs of $25k, and interest of $667.

That’s a rate of return of 12.63% (IRR), and if they did the same project for the next six months their annual rate of return is 25.26%.   A far cry from the 38% above however!