Monday, September 30, 2013

'Til death do us part? Portfolio edition

In response to my post A Rorschach test for bulls and bears at Seeking Alpha, I saw a couple of important comments. In the article, I made a reference to the analysis of John Hussman, to which a reader wrote that Hussman had a terrible track record, why should anyone listen to him. Another responded that "HSGFX since 11/21/00 is up 61.29%....SPY is up 58.47% (dividends reinvested).....Hussman has had 2 huge bear markets to work with to demonstrate his out performance and yet...."

The more important underlying question at hand is: Hiring a manager, or implementing a model, is like getting married, largely because our lifespans are measured in decades. So when should you get a divorce by getting rid of a manager or model?

Why did you buy it in the first place?
We have all seen the studies about how investors chase performance, over-trade and lose returns. The same is true of fund manager and investment models. Proper portfolio construction calls for proper asset class, style, or model diversification. At any one time, a well-constructed portfolio will have some components that are well performing and others that lag because of uncertainty about the future (see my recent post My Lehman lesson: Model diversification and flexibility).

The proper question to evaluate a strategy or manager who is lagging is, "Is his underlying model broken, or is the evaluation time horizon too short?"

In the case of John Hussman, I agree with his comment:
The problem is that investors constantly abandon effective full-cycle strategies when those strategies encounter their own bear markets, and then pile into other effective (and ineffective) full-cycle strategies at the peaks of their own bull markets. This amounts to investing with no discipline at all.
The value of Hussman's analysis is his long-term forecast of equity returns. For instance, consider his chart of 10-year SP 500 returns compared to the Price to Revenue ratio as shown in his recent commentary entitled Psychological Ether:

Don't just look at the returns and say that a fund manager or an investment model is terrible and therefore it's time for a divorce. Just remember the tenets of diversification. There will always be outperforming managers and models and underperforming managers and models. The time to get rid of one is because you believe that a model is broken, irrespective of recent returns.

In the above example, a good reason to terminate Hussman is to take the position that the Price to Revenue ratio is an inappropriate way to measure equity valuation. The wrong reason to terminate a manager or discard a model is for pure performance reasons.

Contrarian investing is painful
One of the most painful parts of contrarian investing is the psychological pain of going against the crowd and of your instincts to chase recent performance. These principles apply to managers and models as well. David Merkel recently wrote an important post, where he said that there are many valid strategies, but the catch is they don't work all the time:
There are valid theories on investing, and they work on average.  If you pursue them consistently, you will do well. If you pursue them after failure, you can do better still.

How many times have you seen articles on investing entitled “The Death of ____.” (fill in the blank)Strategies trend. There is an underlying kernel of validity; it makes economic sense, and has worked in the past. But any strategy can be overplayed, even my favorite strategy, value investing. My style of value investing tries to adjust for that, but it is not perfect there. (And to tell the truth, September has been a bad month for me, though 2013 has been a very good year.)
To underline Merkel's point about out of favor strategies and managers, Morningstar published an article about how manager performance can turn around:
Would you consider a 10.5% annualized gain a satisfactory investment return for the five years ended Aug. 31, 2013?

Would you believe it if you were told you could have gotten that result from one of the worst-performing and most unpopular large-growth funds of that time period: Fidelity Magellan (FMAGX), which posted an anemic 3% annualized, time-weighted gain and watched investors pull nearly $20 billion out during that span?

Would it further shock you to learn that you could have squeezed that performance from the much maligned Magellan by buying it the same month Lehman Brothers imploded, committing to an automatic investment plan, and then doing next to nothing for the next five years?
With that in mind, I can cite a couple of examples of the kinds of models and strategies that are out of favor and may qualify for the title of the death of _____. As usual, you need to do your own analysis and due diligence before investing.

Managed futures
Consider, for example, this article about managed futures programs:
How bad is this Managed Futures Malaise that we’re talking rain dances and Wilson Philips songs? Pretty bad. The main CTA indices are all at 3.5 year lows and sitting on their worst drawdown levels in the past 15 years – at 28 months and -12% from their past all time highs. And with the year to date numbers negative through August (albeit a small amount), there’s the very real possibility of managed futures postings its 4th losing year out of the past 5 in 2013. What’s more – a comparison of the 10 years prior to 2009 and the period from March 2009 until now shows the asset class at levels not seen decades, in terms of the amount and duration of the losses, reminding us of the ‘generational low’ terminology used by the stock folks circa 2009).

Most managed futures managers use systematic trend following models to derive their returns. When the trend is choppy, these models fail. Are you willing to take the position that the character of markets have changed and they will not be trending anymore? If not, maybe it's time to take the painful contrarian position of putting money in while others are fleeing.

Tactical asset allocation
Another "death of ..." candidate is tactical asset allocation. Recently, the WSJ published an article questioning the value of tactical funds:
There are 41 mutual funds with "tactical" in their names that are tracked by investment-research firm Morningstar Inc., including 18 that were launched since the beginning of 2012. Total assets in this fund category have risen to $4.92 billion from $1.10 billion in December 2007, according to Morningstar.

Just as the so-called market-timing funds did years ago, the tactical funds are underperforming balanced portfolios: On average, they gained 6.9% over 12 months and 7.7% annually over three years through Aug. 31. A classically balanced, passively managed portfolio--with 60% invested in the SP 500 index and 40% invested in the Barclays U.S. Aggregate Bond Index--would have gained 10.2% and 12.1%, respectively, in the same periods, Morningstar said.
Josh Brown chimed in [emphasis added]:
Unfortunately, investors (and their advisors) are now finding out that there is a price to be paid, a cost for this tactical "insurance" being added to the mix: In a bull market, it can easily be a drag. And if you pick a really poor manager, or a quantitative strategy with signals that no longer work, it can actually lose you money. And, even if they work great, they can be expensive (turnover, taxes, internal fees, sales loads) - more expensive, sometimes, than the drawdowns they claim to be protecting you from.
Tactical asset allocation strategies are of little value if the stock market just goes up, as it has largely done since the Lehman Crisis shown in the chart below.

Here's a history lesson: A lot of money got allocated to these kinds of strategies in the wake of the Crash of 1987, but evaporated in the secular bull market of the 1990's. Brown added that he expects that as funds get withdrawn from this class of strategy, the protection that they offer won't be there when it's needed the most:
With tactical funds now adding more drag than performance over the three year period relative to the most vanilla benchmarks available, I fully expect to see the trend toward tactical strategies on the wane in the coming year. And then, just when they're needed most, no one will own them.

Rinse and repeat.
The key issue for investors invested in tactical asset allocation strategies is whether you believe that stocks are embarked in a secular bull market like the 1990's, or more likely to remain range-bound, as stocks did during the 1966-1982 period. I've made my position quite clear on this point (see A new secular bull? Don't count on it).

The moral of this story is, know why you bought into the strategy, fund manager, or investment model in the first place. Just remember the Morningstar cautionary tale about Fidelity Magellan - managers and strategies can turn around unexpectedly. Unless you believe the premise underlying the investment model is broken, arguably you should be a good contrarian investors with a rebalancing strategy - take funds away from top performers and add them to lagging performers.

Cam Hui is a portfolio manager at Qwest Investment Fund Management Ltd. (“Qwest”). The opinions and any recommendations expressed in the blog are those of the author and do not reflect the opinions and recommendations of Qwest. Qwest reviews Mr. Hui’s blog to ensure it is connected with Mr. Hui’s obligation to deal fairly, honestly and in good faith with the blog’s readers.”

None of the information or opinions expressed in this blog constitutes a solicitation for the purchase or sale of any security or other instrument. Nothing in this blog constitutes investment advice and any recommendations that may be contained herein have not been based upon a consideration of the investment objectives, financial situation or particular needs of any specific recipient. Any purchase or sale activity in any securities or other instrument should be based upon your own analysis and conclusions. Past performance is not indicative of future results. Either Qwest or I may hold or control long or short positions in the securities or instruments mentioned.

No comments: