Warren Buffett and Charlie Munger’s Best Advice

November 26, 2013

…talk about the best advice they have even gotten in a short piece from Fortune.  I think it clarifies the difference between a blind value investor and an investor who is looking for good companies (not coincidentally, many of those good companies have good relative strength).  Warren Buffett and Charlie Munger have made a fortune implementing this advice.

Buffett: I had been oriented toward cheap securities. Charlie said that was the wrong way to look at it. I had learned it from Ben Graham, a hero of mine. [Charlie] said that the way to make really big money over time is to invest in a good business and stick to it and then maybe add more good businesses to it. That was a big, big, big change for me. I didn’t make it immediately and would lapse back. But it had a huge effect on my results. He was dead right.

Munger: I have a habit in life. I observe what works and what doesn’t and why.

I highlighted the fun parts.  Buffett started out as a Ben Graham value investor.  Then Charlie wised him up.

Valuation has its place, obviously.  All things being equal, it’s better to buy cheaply than to pay up.  But Charlie Munger had observed that good businesses tended to keep on going.  The same thing is typically true of strong stocks—and most often those are the stocks of strong businesses.

Buy strong businesses and stick with them as long as they remain strong.

Source: CNN/Money (click on image to enlarge)

Posted by:

A Reminder About Real Return

November 20, 2013

The main thing that should matter to a long-term investor is real return.  Real return is return after inflation is factored in.  When your real return is positive, you are actually increasing your purchasing power— and purchasing goods and services is the point of having a medium of exchange (money) in the first place.

A recent article in The New York Times serves as a useful reminder about real return.

The Dow Jones industrial average broke through 16,000 on Monday for the first time on record — well, at least in nominal terms. If you adjust for inflation, technically the highest level was on Jan. 14, 2000.

Adjusting for price changes, the Dow’s high today was still about 1.3 percent below its close on Jan. 14, 2000 (and about 1.6 percent below its intraday high from that date).

There’s a handy graphic as well, of the Dow Jones Industrial Average adjusted for inflation.

Source: New York Times/Bloomberg

(click on image to enlarge)

This chart, I think, is a good reminder that buy-and-hold (known in our office as “sit-and-take-it”) is not always a good idea.  In most market environments there are asset classes that are providing real return, but that asset class is not always the broad stock market.  There is value in tactical asset allocation, market segmentation, strategy diversification, and other ways to expose yourself to assets that are appreciating fast enough to augment your purchasing power.

I’ve read a number of pieces recently that contend that “risk-adjusted” returns are the most important investment outcome.  Really?  This would be awesome if I could buy a risk-adjusted basket of groceries at my local supermarket, but strangely, they seem to prefer the actual dollars.  Your client could have wonderful risk-adjusted returns rolling Treasury bills, but would then also get to have a lovely risk-adjusted retirement in a mud hut.  If those dollars are growing more slowly than inflation, you’re just moving in reverse.

Real returns are where it’s at.

Posted by:

Dumb Talk About Smart Beta?

October 7, 2013

John Rekenthaler at Morningstar, who usually has some pretty smart stuff to say, took on the topic of smart beta in a recent article.  Specifically, he examined a variety of smart beta factors with an eye to determining which ones were real and might persist.  He also thought some factors might be fool’s gold.

Here’s what he had to say about value:

The value premium has long been known and continues to persist.

And here’s what he had to say about relative strength (momentum):

I have trouble seeing how momentum can succeed now that its existence is well documented.

The italics are mine.  I didn’t take logic in college, but it seems disingenuous to argue that one factor will continue to work after it is well-known, while becoming well-known will cause the other factor to fail!  (If you are biased in favor of value, just say so, but don’t use the same argument to reach two opposite conclusions.)

There are a variety of explanations about why momentum works, but just because academics can’t agree on which one is correct doesn’t mean it won’t continue to work.  It is certainly possible that any anomaly could be arbitraged away, but Robert Levy’s relative strength work has been known since the 1960s and our 2005 paper in Technical Analysis of Stocks & Commodities showed it continued to work just fine just the way he published it.  Academics under the spell of efficient markets trashed his work at the time too, but 40 years of subsequent returns shows the professors got it wrong.

However, I do have a background in psychology and I can hazard a guess as to why both the value and momentum factors will continue to persistthey are both uncomfortable to implement.  It is very uncomfortable to buy deep value.  There is a terrific fear that you are buying a value trap and that the impairment that created the value will continue or get worse.  It also goes against human nature to buy momentum stocks after they have already outperformed significantly.  There is a great fear that the stock will top and collapse right after you add it to your portfolio.  Investors and clients are quite resistant to buying stocks after they have already doubled, for example, because there is a possibility of looking really dumb.

Here’s the reason I think both factors are psychological in origin: it is absurdly easy to screen for either value or momentum.  Any idiot can implement either strategy with any free screener on the web.  Pick your value metric or your momentum lookback period and away you go.  In fact, this is pretty much exactly what James O’Shaughnessy did in What Works on Wall Street.  Both factors worked well—and continue to work despite plenty of publicity.  So the barrier is not that there is some secret formula, it’s just that investors are unwilling to implement either strategy in a systematic way–because of the psychological discomfort.

If I were to make an argument—the behavioral finance version—about which smart beta factor could potentially be arbitraged away over time, I would have to guess low volatility.  If you ask clients whether they would prefer to buy stocks that a) had already dropped 50%, b) had already gone up 50%, or c) had low volatility, I think most of them would go with “c!”  (Although I think it’s also possible that aversion to leverage will keep this factor going.)

Value and momentum also happen to work very well together.  Value is a mean reversion factor, while momentum is a trend continuation factor.  As AQR has shown, the excess returns of these two factors (unsurprisingly, once you understand how they are philosophical opposites) are uncorrelated.  Combining them may have the potential to smooth out an equity return stream a little bit.  Regardless, two good return factors are better than one!

Posted by:

Stocks for the Long Run

September 20, 2013

Unlike certain authors, I am not promoting some agenda about where stocks will be at some future date!  Instead, I am just including a couple of excerpts from a paper by luminaries David Blanchett, Michael Finke, and Wade Pfau that suggests that stocks are the right investment for the long run—based on historical research.  Their findings are actually fairly broad and call market efficiency into question.

We find strong historical evidence to support the notion that a higher allocation to equities is optimal for investors with longer time horizons, and that the time diversification effect is relatively consistent across countries and that it persists for different levels of risk aversion.

When they examine optimal equity weightings in a portfolio by time horizon, the findings are rather striking.  Here’s a reproduction of one of their figures from the paper:

Source: SSRN/Blanchett, Finke, Pfau  (click to enlarge)

They describe the findings very simply:

Figure 1 also demonstrates how to interpret the results we include later in Tables 2 and 3. In Figure 1 we note an intercept (α) of 45.02% (which we will assume is 45% for simplicity purposes) and a slope (β) of .0299 (which for simplicity purposes we will assume is .03). Therefore the optimal historical allocation to equities for an investor with a 5 year holding period would be 60% stocks, which would be determined by: 45% + 5(3%) = 60%.

In other words, if your holding period is 15-20 years or longer, the optimal portfolio is 100% stocks!

Reality, of course, can be different from statistical probability, but their point is that it makes sense to own a greater percentage of stocks the longer your time horizon is.  The equity risk premium—the little extra boost in returns you tend to get from owning stocks—is both persistent and decently high, enough to make owning stocks a good long-term bet.

Posted by:

From the Archives: Inherently Unstable Correlations

September 19, 2013

No, this is not a post on personality disorders.

Rather, it is a post on the inherently unstable nature of correlations between securities and between asset classes.  This is important because the success of many of the approaches to portfolio management make the erroneous assumption that correlations are fairly stable over time.  I was reminded just how false this belief is while reading The Leuthold Group‘s April Green Book in which they highlighted the rolling 10-year correlations in monthly percentage changes between the S&P 500 and the 10-year bond yield.  Does this look stable to you?  Chart is shown by permission from The Leuthold Group.

Correlation Inherently Unstable

(Click to Enlarge)

If you are trying to use this data, would you conclude that higher bond yields are good for the stock market or bad?  The answer is that the correlations are all over the map.  In 2006, William J. Coaker II published The Volatility of Correlations in the FPA Journal.  That paper details the changes in correlations between 15 different asset classes and the S&P 500 over a 34-year time horizon.  To give you a flavor for his conclusions, he pointed out that Real Estate’s rolling 5-year correlations to the S&P 500 ranged from 0.17 to 0.75, and for Natural Resources the range was -0.34 to 0.49.   History is conclusive – correlations are unstable.

This becomes a big problem for strategic asset allocation models that use historical data to calculate an average correlation between securities or asset classes over time.  Those models use that stationary correlation as one of the key inputs into determining how the model should currently be allocated.  That may well be of no help to you over the next five to ten years.  Unstable correlations are also a major problem for “financial engineers” who use their impressive physics and computer programming abilities to identify historical relationships between securities.   They may find patterns in the historical data that lead them to seek to exploit those same patterns in the future (i.e. LTCM in the 1990′s.) The problem is that the future is under no obligation to behave like the past.

Many of the quants are smart enough to recognize that unstable correlations are a major problem.  The solution, which I have heard from several well-known quants, is to constantly be willing to reexamine your assumptions and to change the model on an ongoing basis.  That logic may sound intelligent, but the reality is that many, if not most, of these quants will end up chasing their tail. Ultimately, they end up in the forecasting game.  These quants are rightly worried about when their current model is going to blow up.

Relative strength relies on a different premise.  The only historical pattern that must hold true for relative strength to be effective in the future is for long-term trends to exist. That is it.  Real estate (insert any other asset class) and commodities (insert any other asset class) can be positively or negatively correlated in the future and relative strength models can do just fine either way.  Relative strength models make zero assumptions about what the future should look like.  Again, the only assumption that we make is that there will be longer-term trends in the future to capitalize on.  Relative strength keeps the portfolio fresh with those securities that have been strong relative performers.  It makes no assumptions about the length of time that a given security will remain in the portfolio.  Sure, there will be choppy periods here and there where relative strength models do poorly, but there is no need (and it is counterproductive) to constantly tweak the model.

Ultimately, the difference between an adaptive relative strength model and most quant models is as different as a mule is from a horse.  Both have four legs, but they are very different animals.  One has a high probability of being an excellent performer in the future, while the other’s performance is a big unknown.

—-this article originally appeared 4/16/2010.  It’s important to understand the difference between a model that relies on historical correlations and a model that just adapts to current trends.

Posted by:

Bucket Portfolio Stress Test

September 4, 2013

I’ve long been a fan of portfolio buckets or sleeves, for two reasons.  The first reason is that it facilitates good diversification, which I define as diversification by volatility, by asset class, and by strategy.  (We happen to like relative strength as one of these primary strategies, but there are several offsetting strategies that might make sense.)  A bucket portfolio makes this kind of diversification easy to implement.

The second benefit is largely psychological—but not to be underestimated.  Investors with bucket portfolios had better performance in real life during the financial crisis because they didn’t panic.  While the lack of panic is a psychological benefit, the performance benefit was very real.

Another champion of bucketed portfolios is Christine Benz at Morningstar.  She recently wrote a series of article in which she stress-tested bucketed portfolios, first through the 2007-2012 period (one big bear market) and then through the 2000-2012 period (two bear markets).  She describes her methodology for rebalancing and the results.

If you have any interest in portfolio construction for actual living, breathing human beings who are prone to all kinds of cognitive biases and emotional volatility, these articles are mandatory reading.  Better yet for fans of portfolio sleeves, the results kept clients afloat.  I’ve included the links below.  (Some may require a free Morningstar registration to read.)

Article:  A Bucket Portfolio Stress Test  http://news.morningstar.com/articlenet/article.aspx?id=605387&part=1

Article:  We Put the Bucket System Through Additional Stress Tests  http://news.morningstar.com/articlenet/article.aspx?id=607086

Article:  We Put the Bucket System Through a Longer Stress Test  http://news.morningstar.com/articlenet/article.aspx?id=608619



Posted by:

Quote of the Week

August 23, 2013

Avoiding danger is no safer in the long run than outright exposure.  The fearful are caught as often as the bold.—Helen Keller


I doubt that Helen Keller was thinking about bond investors when she wrote this, but she may as well have been.  The safe haven trade hasn’t worked out too well since May.  Bond investors sometimes think they have an extra measure of security versus stock investors.  And it is true that most bonds are less volatile than stocks.  Volatility, however, is a pretty poor way to measure risk.  An alternative way to measure risk is to look at drawdown—and measured that way, bonds have had drawdowns in real returns that rival drawdowns in stocks.

In truth, bonds are securities just like stocks.  They are subject to the same, sometimes irrational, swings in investor emotion.  And given that bonds are priced based on the income they produce, they are very vulnerable to increases in interest rates and increases in inflation.

So I think that Helen Keller’s point is well taken—instead of pretending that you are safe, make sure you understand the exposures you have and make sure you take them on intentionally.

Source: Wikipedia  (click on image to enlarge)


Posted by:

Correlation and Expected Returns

July 31, 2013

Modern portfolio theory imagines that you can construct an optimal portfolio, especially if you can find investments that are uncorrelated.  There’s a problem from the correlation standpoint, though.  As James Picerno of The Capital Spectator points out, correlations are rising:

A new study from the Bank of International Settlements (BIS) raises doubts about the value of commodities as a tool for enhancing portfolio diversification. The paper’s smoking gun, so to speak, is that “the correlation between commodity and equity returns has substantially increased after the onset of the recent financial crisis.”

Correlations are a key factor in the design and management of asset allocation, but they’re not the only factor. And even if we can find assets and strategies with reliably low/negative correlations with, say, equities, that alone isn’t enough, as I discussed last week. You also need to consider other factors, starting with expected return. It may be tempting to focus on one pair of assets and consider how the trailing correlation stacks up today. But that’s hardly the last word on making intelligent decisions on how to build a diversified portfolio.

As more investors pile into commodities, REITs, hedge funds, and other formerly obscure corners, the historical diversification benefits will likely fade. Granted, the outlook for expected diversification benefits fluctuates through time, and so what looks unattractive today may look considerably more compelling tomorrow (and vice versa). But as a general proposition, it’s reasonable to assume that correlations generally will inch closer to 1.0. That doesn’t mean that diversifying across asset classes is destined to become worthless, but the expected payoff is likely to dim with the passage of time.

Mathematically, any two items that are not 100% correlated will reduce volatility when combined.  But that doesn’t necessarily mean it’s a good addition to your portfolio—or that modern portfolio theory is a very good way to construct a portfolio.  (We will set aside for now the MPT idea that volatility is necessarily a bad thing.)  The article includes a nice graphic, reproduced below, that shows how highly correlated many asset classes are with the US market, especially if you keep in mind that these are 36-month rolling correlations.  Many asset classes may not reduce portfolio volatility much at all.

Source: The Capital Spectator  (click on image to enlarge)

As Mr. Picerno points out, optimal allocations are far more sensitive to returns than to correlations or volatility.  So even if you find a wonderfully uncorrelated investment, if it has a lousy return it may not help the overall portfolio much.  It would reduce volatility, but quite possibly at a big cost to overall returns.  The biggest determinant of your returns, of course, is what assets you actually hold and when.  The author puts this a slightly different way:

Your investment results also rely heavily on how and when you rebalance the mix.

Indeed they do.  If you hold equities when they are doing well and switch to other assets when equities tail off, your returns will be quite different than an investor holding a static mix.  And your returns will be way different than a scared investor that holds cash when stocks or other assets are doing well.

In other words, the return of your asset mix is what impacts your performance, not correlations or volatility.  This seems obvious, but in the fog of equations about optimal portfolio construction, this simple fact is often overlooked.  Since momentum (relative strength) is generally one of the best-performing and most reliable return factors, that’s what we use to drive our global tactical allocation process.  The idea is to own asset classes as long as they are strong—and to replace them with a stronger asset class when they begin to weaken.  In this context, diversification can be useful for reducing volatility, if you are comfortable with the potential reduction in return that it might entail.  (We  generally advocate diversifying by volatility, by asset class, and by strategy, although the specific portfolio mix might change with the preference of the individual investor.)  If volatility is well-tolerated, maybe the only issue is trying to generate the strongest returns.

Portfolio construction can’t really be reduced to some “optimal” set of tradeoffs.  It’s complicated because correlations change over time, and because investor preferences between return and volatility are in constant flux.  There is nothing stable about the portfolio construction process because none of the variables can be definitively known; it’s always an educated approximation.  Every investor gets to decide—on an ongoing basis—what is truly important: returns (real money you can spend) or volatility (potential emotional turmoil).  I always figure I can afford Maalox with the extra returns, but you can easily see why portfolio management is overwhelming to so many individual investors.  It can be torture.

Portfolio reality, with all of its messy approximations, bears little resemblance to the seeming exactitude of Modern Portfolio Theory.

Posted by:

From the Archives: If You Miss the 10 Best Days

June 7, 2013

We’ve all seen numerous studies that purport to show how passive investing is the way to go because you don’t want to be out of the market for the 10 best days.  No one ever mentions that the “best days” most often occur during the declines!

It turns out that the majority of the best days and the worst days occur near one another, during the declines.  Why?  Because the market is more volatile during declines.  It is true that the market goes down 2-3x as fast as it goes up.  (World Beta has a nice post on this topic of volatility clustering, which is where this handy-dandy table comes from.)

 If You Miss the 10 Best Days

from World Beta

You can see how volatility increases and the number of days with daily moves greater than 2.5% really spikes when the market is in a downward trend.  It would seem to be a very straightforward proposition to improve your returns simply by avoiding the market when it is in a downtrend.

However, not every strategy can be improved by going to cash.  Think about the math: if your investing methodology makes enough extra money on the good days to offset the bad days, or if it can make money during a significant number of the declines, you might be better off just gritting your teeth during the declines and banking the higher returns.  Although the table above suggests it should help, a simple strategy of exiting the market (i.e., going to cash) when it is below its 200-day moving average may not always live up to its theoretical billing.

 If You Miss the 10 Best Days

 If You Miss the 10 Best Days

click to enlarge

Consider the graphs above.  (The first graph uses linear scaling; the second uses logarithmic scaling for the exact same data.)  This test uses Ken French’s database to get a long time horizon and shows the returns of two portfolios constructed with market cap above the NYSE median and in the top 1/3 for relative strength.  In other words, the two portfolios are composed of mid- and large-cap stocks with good relative strength.  The only difference between the two portfolios is that one (red line) goes to cash when it is below its 200-day moving average.  One portfolio (blue line) stays fully invested.  The fully invested portfolio turns $100 into $49,577, while the cash-raising portfolio yields only $26,550.

If you would rather forego the extra money in return for less volatility, go right ahead and make that choice.  But first stack up 93 boxes of  Diamond matches so that you can burn 23,027 $1 bills, one at a time, to represent the difference–and then make your decision.

 If You Miss the 10 Best Days

The drawdowns are less with the 200-day moving average, but it’s not like they are tame–equities will be an inherently volatile asset class as long as human emotions are involved.  There are still a couple of drawdowns that are greater than 20%.  If an investor is willing to sit through that, they might as well go for the gusto.

As surprising as it may seem, the annualized return over a long period of time is significantly higher if you just stay in the market and bite the bullet during train wrecks–and even two severe bear markets in the last decade have not allowed the 200-day moving average timer to catch up.

At the bottom of every bear market, of course, it certainly feels like it would have been a good idea (in hindsight) to have used the 200-day moving average to get out.  In the long run, though, going to cash with a high-performing, high relative strength strategy might be counterproductive.  When we looked at 10-year rolling returns, the fully invested high relative strength model has maintained an edge in returns for the last 30 years running.

 If You Miss the 10 Best Days

click to enlarge

Surprising, isn’t it?  Counterintuitive results like this are one of the reasons that we find testing so critical.  It’s  easy to fall in line with the accepted wisdom, but when it is actually put to the test, the accepted wisdom is often wrong.  (We often find that even when shown the test data, many people refuse, on principle, to believe it!  It is not in their worldview to accept that one of their cherished beliefs could be false.)  Every managed portfolio in our Systematic RS lineup has been subjected to heavy testing, both for returns and–and more importantly–for robustness.  We have a high degree of confidence that these portfolios will do well in the long run.

—-this article originally appeared 3/5/2010.  We find that many investors continue to refuse, on principle, to believe the data!  If you have a robust investment method, the idea that you can improve your returns by getting out of the market during downturns appears to be false.  (Although it could certainly look true for small specific samples.  And, to be clear, 100% invested in a volatile strategy is not the appropriate allocation for most investors.)  Volatility can generally be reduced somewhat, but returns suffer.  One of our most controversial posts ever—but the data is tough to dispute.

In more recent data, the effect can be seen in this comparison of an S&P 500 ETF and an ETN that switches between the S&P 500 and Treasury bills based on a 200-day moving average system.  The volatility has been muted a little bit, but so have the returns.

(click on image to enlarge)

Posted by:

Investment Risk Re-imagined

May 8, 2013

Risk is fundamental to investing, but no one can agree what it is.  Modern Portfolio Theory defines it as standard deviation.  Tom Howard of AthenaInvest thinks investment risk is something completely different.  In an article at Advisor Perspectives, he explains how he believes investment risk should be defined, and why the MPT definition is completely wrong.  I think his point is a strong one.  I don’t know how investment risk should be defined—there’s a lot of disagreement within the industry—but I think he makes, at the very least, a very clear case for why volatility is not the correct definition.

Here’s how he lays out his argument:

The measures currently used within the investment industry to capture investment risk are really mostly measures of emotion. In order to deal with what is really important, let’s redefine investment risk as the chance of underperformance. As Buffett  suggests, focus on the final outcome and not on the path travelled to get there.

The suggestion that investment risk be measured as the chance of underperformance is intuitively  appealing to many investors. In fact, this measure of risk is widely used in a number of industries. For example, in industrial applications, the risk of  underperformance is measured by the probability that a component, unit or service will fail. Natural and manmade disasters use such a measure of risk. In each situation, the focus is on the chances that various final outcomes might occur. In general, the path to the outcome is less important and has little influence on the measure of risk.

I added the bold to highlight his preferred definition.  Next he takes on the common MPT measurement of risk as volatility and spells out why he thinks it is incorrect:

In an earlier article I reviewed the evidence regarding stock market volatility and showed that most volatility stems from crowds overreacting to information. Indeed, almost no volatility can be explained by changes in underlying economic fundamentals at the market and individual stock levels.  Volatility measures emotions, not necessarily investment risk. This is also true of other measures of risk, such as downside standard deviation, maximum drawdown and downside capture.

But unfortunately, the investment industry has adopted this same volatility as a risk measure that, rather than focusing on the final outcome, focuses on the bumpiness of the ride. A less bumpy ride is thought to be less risky, regardless of the final outcome. This leads to the unintended consequence of building portfolios that result in lower terminal wealth and, surprisingly, higher risk.

This happens because the industry mistakenly builds portfolios that minimize short-term volatility relative to long-term returns, placing emotion at the very heart of the long-horizon portfolio  construction process. This approach is popular because it legitimizes the  emotional reaction of investors to short-term volatility.

Thus risk and volatility are frequently thought of as being interchangeable. However, focusing on short-term volatility when building long horizon portfolios can have the unintended consequence of actually increasing investment risk. Since risk is the chance of underperformance, focusing on short-term volatility will often lead to investing in lower expected return markets with little impact on long-term volatility.1 Lowering expected portfolio return in an effort to reduce short-term volatility actually increases the chance of underperformance, which means increasing risk.

A clear example of this is the comparison of long-term stock and bond returns. Stocks dramatically outperform bonds over the long run. By investing in bonds rather than stocks, short-term volatility is reduced at the expense of decreasing long-term wealth. Equating short-term volatility with risk leads to inferior long-horizon portfolios.

The cost of equating risk and emotional volatility can be seen in other areas as well. Many investors pull out of the stock market when faced with heightened volatility. But research shows this is exactly when they should remain in the market and even increase their stock holdings, as subsequent returns are higher on average.2  It is also the case that many investors exit after market declines only to miss the subsequent rebounds. Following the 2008 market crash, investors withdrew billions of dollars from equity mutual funds during a period in which the stock market more than doubled.

The end result is that investors frequently suffer the pain of losses without capturing the subsequent gains. Several studies confirm that the typical equity mutual fund investor earns a return substantially less than the fund return because of poorly timed movements in and out of the fund. Again, these are the dangers of not carefully distinguishing emotions from risk and thus allowing emotions to drive investment decisions.

I added the bold here as well.  I apologize for such a big excerpt, but I think it’s important to get the full flavor here.  The implication, which he makes explicit later in the article, is that current risk measures are largely an agency issue.  The advisor is the “agent” for the client, and thus the advisor is likely to pander to the client’s emotions—because it results in less business risk (i.e., the client leaving) for the advisor.  Of course, as he points out in the excerpt above, letting emotions drive the bus results in poor investment results.

Tom Howard has hit the nail on the head.  Advisors often have the choice of a) pandering to the client’s emotions at the cost of substantial long-term return or b) losing the client.  Since investment firms are businesses, the normal decision is to retain the client—which, paradoxically, leads to more risk for the client.  While “the customer is always right” may be a fine motto for a retail business, it’s usually the other way around in the investment business!

There’s another wrinkle to investment risk too.  Regardless of how investment risk is defined, it’s unlikely that human nature is going to change.  No matter how much data and logic are thrown at clients, their emotions are still going to be prone to overwhelm them at inopportune times.   It’s here, I think, that advisors can really earn their keep, in two important ways, through both behavior and portfolio construction.

  1. Advisor Behavior: The advisor can stay calm under pressure.  Hand-holding, as it is called in the industry, is really, really important.  Almost no one gets good training on this subject.  They learn on the job, for better or worse.  If the advisor is calm, the client will usually calm down too.  A panicked advisor is unlikely to promote the mental stability of clients.
  2. Portfolio Construction: The portfolio can explicitly be built with volatility buckets.  The size of the low-volatility bucket may turn out to be more a function of the client’s level of emotional volatility than anything else.  A client with a long-horizon and a thick skin may not need that portfolio piece, but high-beta Nervous Nellies might require a bigger percentage than their actual portfolio objectives or balance sheet necessitate—because it’s their emotional balance sheet we’re dealing with, not their financial one.  Yes, this is sub-optimal from a return perspective, but not as sub-optimal as exceeding their emotional tolerance and having the client pull out at the bottom.  Emotional blowouts are financially expensive at the time they occur, but usually have big financial costs in the future as well in terms of client reluctance to re-engage.  Psychic damage can impact financial returns for multiple market cycles.

Tom Howard has laid out a very useful framework for thinking about investment risk.  He’s clearly right that volatility isn’t risk, but advisors still have to figure out a way to deal with the volatility that drives client emotions.  The better we deal with client emotions, the more we reduce their long-term risk.

Note: This argument and others are found in full form in Tom Howard’s paper on Behavioral Portfolio Management.  Of course I’m coming at things from a background in psychology, but I  think his framework is excellent.  Behavioral finance has been crying out for an underlying theory for years.  Maybe this is it.  It’s required reading for all advisors, in my opinion.

Posted by:

Smart Beta vs. Monkey Beta

April 9, 2013

Andy wrote a recent article entitled Smart Beta Gains Momentum.  It’s gaining momentum for a good reason!  A recent study at Cass Business School in London found that cap-weighting was not a very good way to construct an index.  Lots of methods to get exposure to smart beta do better.  The results were discussed in an article at Index Universe.  Some excerpts:

Researchers have found that equity indices constructed randomly by ‘monkeys’ would produce higher risk-adjusted returns than an equivalent market capitalisation-weighted index over the last 40 years…

The findings come from a recent study by Cass Business School (CBS), which was based on monthly US share data from 1968 to 2011. The authors of the study found that  a variety of alternative index weighting schemes all delivered superior returns to the market cap approach.

According to Dr. Nick Motson of CBS, co-author of the study, “all of the 13 alternative indices we studied produced better risk-adjusted returns than a passive exposure to a market-cap weighted index.”

The study included an experiment that saw a computer randomly pick and weight each of the 1,000 stocks in the sample. The process was then repeated 10 million times over each of the 43 years.   Clare describes this as “effectively simulating the stock-picking abilities of a monkey”.

…perhaps most shockingly, we found that nearly every one of the 10 million monkey fund managers beat the performance of the market cap-weighted index,” said Clare.

The findings will be a boost to investors already looking at alternative indexing.  Last year a number of European pension funds started reviewing their passive investment strategies, switching from capitalisation-weighting to alternative index methodologies.

Relative strength is one of the prominent smart beta methodologies.  Of course, cap-weighting has its uses—the turnover is low and rebalancing is minimized.  But purely in terms of performance, the researchers at Cass found that there are better ways to do things.  Now that ETFs have given investors a way to implement some of these smart beta methods in a tax-efficient, low-cost manner, I suspect we will see more movement toward smart beta in the future.

Posted by:

The Ridiculous Efficient Frontier

March 22, 2013

It’s hard to believe this paper was not written ironically.  Perhaps I am missing the author’s sense of dry humor?  In a paper entitled Principal Component Analysis of Time Variations in the Mean-Variance Efficient Frontier, author Andreas Steiner subjects mean-variance optimization to principal component analysis, a mathematical way to determine the relative importance of factors.  He extracted three important factors that determine the efficient frontier.  The three factors together explained 99% of the shape of the efficient frontier.

In fact—and this is the funny part to me–one factor explained 95% of the shape of the efficient frontier.  And what was that magic factor?

It was the level of returns.  In other words, the shape of the efficient frontier depends on the returns of the various assets.  If you can predict the returns, you will (mostly) know the shape of the efficient frontier.  And, in case you were wondering, the shape of the efficient frontiers varies enormously depending on the time period.  Below, for example, is a clip from the paper showing efficient frontiers calculated from trailing data at different times.  I’m sure you can see the slight problem—the curves look nothing alike.

The Ridiculous Frontier

Source: Andreas Steiner/SSRN  (click on image to enlarge)

The author writes:

We find that the level factor is highly correlated with average asset returns.

We interpret this result as evidence that successful investment management is mainly driven by return estimates and not “risk management” as has been in the spotlight since the Financial Crisis.

Here’s the immediate question that occurs to my feeble brain, although I’m guessing most 5th graders would be right there with me: If I could predict the return of each asset, why would I need an efficient frontier?  Wouldn’t I just buy the best-performing asset?

Indeed, risk management is no big deal if I simply predict all of the asset returns.  We’ve discussed many times before that mean-variance optimization is highly dependent upon returns, although correlations and standard deviation play a supporting role.  All of these factors are moving targets, none more so than returns.  Mean variance optimization, in practice, is a complete bust because obviously no one can reliably and consistently predict returns.

This kind of study—although mathematically rigorous—is silliness of the first degree.  It reminds me certain academic follies, like the professors who wondered if monkeys at typewriters really could reproduce the works of Shakespeare.  (The short answer is “no.”)

Modern portfolio theory would be relatively harmless if it remained in academia.  However, when investors try to use it to build portfolios, it has the potential to cause a lot of damage.  Although it is simply another theory that does not work in practice, it is enshrined in many finance textbooks and still taught to budding practitioners.  Is it any wonder that we prefer tactical asset allocation driven by relative strength to guessing at future returns?

HT to CXO Advisory

Posted by:

Target Date Fund Follies

March 7, 2013

Target date and lifecycle funds have taken off since 2006, when they were deemed qualified default investment alternatives in the Pension Protection Act.  I’m sure it seemed like a good idea at the time.  Unfortunately, it planted the idea that a glidepath that moved toward bonds as the investor moved toward retirement was a good idea.  Assets in target date funds were nearly $400 billion at the end of 2011—and they have continued to grow rapidly.

In fact, bonds will prove to be a good idea if they perform well and a lousy idea if they perform poorly.  Since 10-year future returns correlate closely with the current coupon yield, prospects for bonds going forward aren’t particularly promising at the moment.  I’ve argued before that tactical asset allocation may provide an alternative method of accumulating capital, as opposed to a restrictive target-date glidepath.

A new research paper by Javier Estrada, The Glidepath Illusion: An International Perspective, makes a much broader claim.  He looks at typical glidepaths that move toward bonds over time, and then at a wide variety of alternatives, ranging from inverse glidepaths that move toward stocks over time to balanced funds.  His findings are stunning.

This lifecycle strategy implies that investors are aggressive with little capital and conservative with much more capital, which may not be optimal in terms of wealth accumulation. This article evaluates three alternative types of strategies, including contrarian strategies that follow a glidepath opposite to that of target-date funds; that is, they become more aggressive as retirement approaches. The results from a comprehensive sample that spans over 19 countries, two regions, and 110 years suggest that, relative to lifecycle strategies, the alternative strategies considered here provide investors with higher expected terminal wealth, higher upside potential, more limited downside potential, and higher uncertainty but limited to how much better, not how much worse, investors are expected to do with these strategies.

In other words, the only real question was how much better the alternative strategies performed.  (I added the bold.)

Every strategy option they considered performed better than the traditional glidepath!  True, if they were more focused on equities, they were more volatile.  But, for the cost of the volatility, you ended up with more money—sometimes appreciably more money.  This data sample was worldwide and extended over 110 years, so it wasn’t a fluke.  Staying equity-focused didn’t work occasionally in some markets—it worked consistently in every time frame in every region.  Certainly the future won’t be exactly like the past, so there is no way to know if these results will hold going forward.  However, bonds have had terrific performance over the last 30 years and the glidepath favoring them still didn’t beat alternative strategies over an investing lifetime.

Bonds, to me, make sense to reduce volatility.  Some clients simply must have a reduced-volatility portfolio to sleep at night, and I get that.  But Mr. Estrada’s study shows that the typical glidepath is outperformed even by a 60/40-type balanced fund.  (Balanced funds, by the way, are also designated QDIAs in the Pension Protection Act.)   Tactical asset allocation, where bonds are held temporarily for defensive purposes, might also allow clients to sleep at night while retaining a growth orientation.  The bottom line is that it makes sense to reduce volatility just enough to keep the client comfortable, but no more.

I’d urge you to read this paper carefully.  Maybe your conclusions will be different than mine.  But my take-away is this: Over the course of an investing lifetime, it is very important to stay focused on growth.

Posted by:

Relative Strength Everywhere

December 14, 2012

Eric Falkenstein has an interesting argument in his paper Risk and Return in General: Theory and Evidence.  He proposes what is essentially a relative strength argument about risk and return.  He contends that investors care only about relative wealth and that risk is really about deviating from the social norm.  Here is the summary of his draft from the excellent CXO Advisory:

Directly measured risk seldom relates positively to average returns.  In fact, there is no measure of risk that produces a consistently linear scatter plot with returns across a variety of investments (stocks, banks, stock options, yield spread, corporate bonds, mutual funds, commodities, small businesses, movies, lottery tickets and bets on horse  races).

  • Humans are social animals, and processing of social  information (status within group) is built into our brains. People care only about relative wealth.
  • Risk is a deviation from what everyone else is doing (the market portfolio) and is therefore avoidable and unpriced. There is no risk premium.

The whole paper is a 150-page deconstruction of the flaws in the standard model of risk and return as promulgated by academics.  The two startling conclusions are that 1) people care only about relative wealth and that 2) risk is simply a deviation from what everyone else is doing.

This is a much more behavioral interpretation of how markets operate than the standard risk-and-return tradeoff assumptions.  After many years in the investment management industry dealing with real clients, I’ve got to say that Mr. Falkenstein re-interpretation has a lot going for it.  It explains many of the anomalies that the standard model cannot, and it comports well with how real clients often act in relation to the market.

In terms of practical implications for client management, a few things occur to me.

  • Psychologists will tell you that clients respond more visually and emotionally than mathematically.  Therefore, it may be more useful to motivate clients emotionally by showing them how saving money and managing their portfolio intelligently is allowing them to climb in wealth and status relative to their peers, especially if this information is presented visually.
  • Eliminating market-related benchmarks from client reports (i.e., the reference to what everyone else is doing) might allow the client to focus just on the growth of their relative wealth, rather than worrying about risk in Falkenstein’s sense of deviation from the norm.  (In fact, the further one gets from the market benchmark, the better performance is likely to be, according to studies on active share.)  If any benchmark is used at all, maybe it should be related to the wealth levels of the peer group to motivate the client to strive for higher status and greater wealth.

I’m sure there is a lot more to be gleaned from this paper and I’m looking forward to having time to read it again.

Posted by:

Yet Another Blow to Modern Portfolio Theory

November 19, 2012

Modern Portfolio Theory is predicated on the ability to construct an efficient frontier based on returns, correlations, and volatility.  Each of these parameters needs to be accurate for the efficient frontier to be accurate.  Since forecasting is tough, often historical averages are used.  Since the next five or ten years is never exactly like the last 50 years, that method has significant problems.  Apologists for modern portfolio theory claim that better efficient frontiers can be generated by estimating the inputs.  Let’s imagine, for a moment, that this can actually be done with some accuracy.

There’s still a big problem.  Volatility bumps up during adverse market conditions, as reported by Research Affiliates.  And correlations change during declines—and not in a good way.

From the abstract of a recent paper, Quantifying the Behavior of Stock Correlations Under Market Stress:

Understanding correlations in complex systems is crucial in the face of turbulence, such as the ongoing financial crisis. However, in complex systems, such as financial systems, correlations are not constant but instead vary in time. Here we address the question of quantifying state-dependent correlations in stock markets. Reliable estimates of correlations are absolutely necessary to protect a portfolio. We analyze 72 years of daily closing prices of the 30 stocks forming the Dow Jones Industrial Average (DJIA). We find the striking result that the average correlation among these stocks scales linearly with market stress reflected by normalized DJIA index returns on various time scales. Consequently, the diversification effect which should protect a portfolio melts away in times of market losses, just when it would most urgently be needed.

I bolded the part that is most inconvenient for modern portfolio theory.  By the way, this isn’t really cutting edge.  The rising correlation problem isn’t new, but I find it interesting that academic papers are still being written on it in 2012.

The quest for the magical efficient portfolio should probably be ended, especially since there are a number of useful ways to build durable portfolios.  We’re just never going to get to some kind of optimal portfolio.  Mean variance optimization, in fact, turns out to be one of the worst methods in real life.  We’ll have to make do with durable portfolio construction.  It may be messy, but a broadly diversified portfolio should be serviceable under a broad range of market conditions.

Posted by:

Durable Portfolio Construction

November 14, 2012

Durable portfolio construction comes from diversification, but diversification can mean a lot of different things.  Most investors, unfortunately, give portfolio construction very little thought.  As a result, their portfolios are not durable.  In fact, they tend to come unglued during every downturn.  Why does that happen?

I think there are a couple of inter-related problems.

  • Volatility tends to increase during downturns
  • Certain correlations tend to increase during declines

Volatility is an artifact of uncertainty.  Once a downturn starts, no one is sure where the bottom is.  That uncertainty often creates selling, which may cause the market to decline, which in turn may create more selling.  We’ve all seen this happen.  Eventually there is capitulation and the market bottoms, but it can be quite frightening in the middle of the move when no one knows where the bottom will be.

Research Affiliates had a recent article on diversification, and included in it was a table that showed the change in volatility that accompanied recessions.  The bump is typically pretty large.

Source: Research Affiliates, via RealClearMarkets  (click to enlarge image)

In general, riskier assets had the biggest jumps in volatility when the economy was under pressure.  Thus, it makes perfect sense how a relatively sedate portfolio under typical conditions becomes much more volatile when conditions are tough.

Correlations are also observed to rise during declines.  “Risk on” assets, especially, often have rising correlations among themselves as risk is shunned.  Similarly, “risk off” assets may see their internal correlations rise.  However, it may be the case that correlations between dissimilar asset classes don’t change nearly as much.  In other words, risk-on and risk-off assets might not have rising correlations during a period of market stress.  In fact, it wouldn’t be surprising to see those correlations actually fall.  So, one way to make portfolios more durable is to diversify by volatility.

There are probably multiple ways to do this.  You could use volatility buckets for low-volatility assets like bonds and high-volatility assets like stocks.  Or, you could just make sure that your portfolios have exposure to a broad range of asset classes, including asset classes with different responses to market stress.

Within an individual asset class, you are likely to see rising correlations between members of your investment universe.  For example, during a sharp market decline, you’re likely to see increasing correlations among stocks.  However, it’s possible to think about diversifying by return factor within an asset class.

AQR and others have shown, for example, that the excess returns of value and relative strength stocks are uncorrelated.  That means that years where relative strength outperforms the market are likely to be years when value lags, and vice versa.  Both types of stocks might go up in a rising market or fall in a declining market, but they will likely have different performance profiles.  Diversifying by using complementary strategies is another way to make portfolios more durable.

As Research Affiliates points out, simple diversification is not a panacea.  As their table shows, almost every asset class (possible exception: short-term bonds) has higher volatility in a bad economy.

Durable portfolio construction, then, might consist of multiple forms of diversification:

  • diversification by volatility
  • diversification by asset class
  • diversification by strategy

While there might be rising correlations between some types of assets, you are also likely to see falling correlations between others.  Although the entire portfolio might have an elevated level of volatility, an absence of surging cross-correlations might make tail events a little more manageable.  Good portfolio construction obviously won’t eliminate market risk, but it might make regular market volatility a little more palatable for a broad range of investors.

Posted by:

Complementary Strategies: One Key to Diversification

October 18, 2012

We use relative strength (known as “momentum” to academics) in our investment process.  We’ve written extensively how complementary strategies like low volatility and value can be used alongside relative strength in a portfolio.  S&P is now on board the train, as they show in this research paper how alternative beta strategies are often negatively correlated.  In fact, here’s the correlation matrix from the paper:

Source: Standard & Poors  (click to enlarge image)

You can see that relative strength/momentum is negatively correlated with both value and low volatility.  This is why we prefer diversification through complementary strategies.

They conclude:

…combining alternative beta strategies that are driven by distinct sets of risk factors may help to reduce the active risk and improve the information ratio.

Diversification is important for portfolios, but it’s not easily achieved.  For example, if you decide to segment the market by style box rather than by return factors, you will find that the style boxes are all fairly correlated.  Although it’s a mathematical truism that anything that isn’t 100% correlated will help diversification, diversification is far more efficient when correlations are low or negative.

We think using factor returns to identify complementary strategies is one of the more effective keys to diversification.

Posted by:

From the Archives: Is Modern Portfolio Theory Obsolete?

May 29, 2012

It all depends on who you ask.  Apologists for MPT will say that diversification worked, but that it just didn’t work very well last go round.  That’s a judgment call, I suppose.  Correlations between assets are notoriously unstable and nearly went to 1.0 during the last decline, but not quite.  So I guess you could say that diversification “worked,” although it certainly didn’t deliver the kind of results that investors were expecting.

Now even Ibbotson Associates is saying that certain aspects of modern portfolio theory are flawed, in particular using standard deviation as a measurement of risk.  In a recent Morningstar interview, Peng Chen, the president of Ibbotsen Associates, addresses the problem.

It’s one thing to say modern portfolio theory, the principle, remained to work. It’s another thing to examine the measures. So when we started looking at the measures, we realized, and this has been documented by many academics and practitioners, we also realized that one of the traditional measures in modern portfolio theory, in particular on the risk side, standard deviation, does not work very well to measure and present the tail risks in the return distribution.

Meaning that, when you have really, really bad market outcomes, modern portfolio theory purely using standard deviation underestimates the probability and severity of those tail risks, especially in short frequency time periods, such as monthly or quarterly.

Leaving aside the issue of how the theory could work if the components do not, this is a pretty surprising admission.  Ibbotson is finally getting around to dealing with the “fat tails” problem.  It’s a known problem but it makes the math much less tractable.  Essentially, however, Mr. Chen is arguing that market risk is actually much higher than modern portfolio theory would have you believe.

In my view, the debate about modern portfolio theory is pretty much done.  Stick a fork in it.  Rather than grasping about for a new theory, why not look at tactical asset allocation, which has been in plain view the entire time?

Tactical asset allocation, when executed systematically, can generate good returns and acceptable volatility without regard to any of the tenets of modern portfolio theory.  It does not require standard deviation as the measure of risk, and it makes no assumptions regarding the correlations between assets.  Instead it makes realistic assumptions: some assets will perform better than others, and you ought to consider owning the good assets and ditching the bad ones.  It’s the ultimate pragmatic solution.

—-this article originally appeared 1/21/2010.  As we gain distance from the 2008 meltdown, investors are beginning to forget how badly their optimized portfolios performed and are beginning to climb back on the MPT bandwagon.  Combining uncorrelated strategies always makes for a better portfolio, but the problem of understated risk remains.  The tails are still fat.  Let’s hope that we don’t get another chance to experience fat tails with the Eurozone crisis.  Tactical asset allocation, I think, may still be the most viable solution to the problem.

Posted by: