Post: 1 March 20th, 2015

Folklores of Finance – The Folklore of Time

According to a new study by State Street called “The Folklore of Finance: How Beliefs and Behaviors Sabotage Success in the Investment Management Industry” (which can be accessed here), investment professionals use easy to articulate – but often incorrect – guidance for making and justifying investment decisions.

We’d like to comment on a few of these folklores, and first up is the folklore of time.

The folklore of time describes beliefs related to what occurred in the past and what may occur in the future. For example, the paper says that investment professionals rely heavily on analysts’ expert forecasts (which are essentially a prediction of the future). Evidence shows, however, that when measured properly, accuracy rates of these forecasts over 24 month time frames are as low as 5% – 10%.

How much confidence would you put in forecasts with 10% accuracy over two year periods? Probably less than the investment management industry apparently does.

The folklore of time doesn’t merely affect attitudes about the future. The State Street study finds that investment professionals also show a baffling tendency to rely on past performance when making investment decisions. This is despite the truism that past performance is no indication of future results.

The State Street study corroborates this truism with a study of 715 US stick market funds which posted top quartile performance in March 2010. After four years, only two (yes, that was two!) of 715 had maintained that top performance throughout the subsequent four year period. Persistence in performance is a myth. Authors also point to another study that shows that managers fired because of poor past performance often subsequently outperform the managers hired to replace them.

The fact is that relying on extrapolations of the past or predictions for the future, even those provided by experts, is a very dangerous way to invest. Yet given the uncertainty we all face with investing, our brains are hardwired to do just that. The alternative, of course, is not to focus on predicting or extrapolating financial markets. Instead, focus on holding a low cost, widely diversified and carefully monitored portfolio, which can achieve your long term goals and is less likely to experience inevitable downturns that you can’t stomach.

And, having done that, do something completely novel (at least in the investment industry). Enjoy the present!

Post: 171 March 15th, 2015

Active versus passive – the debate is over

At Consilium, we believe a growing body of evidence has moved us past the traditional active vs. passive debate. The best evidence comes from the US where the research has been collected and mostly aptly documented. The traditional debate contrasts an index fund, representing the passive camp, and an active equity or hedge fund, representing the active camp.

A widely publicised wager between Warren Buffett and a hedge fund manager called Protégé Partners illustrates the point. Buffett bet that that the Vanguard S&P 500 (a passive index type fund) would beat a selected group of hedge funds over a 10 year time horizon. So far the Vanguard fund is up 43.8%, compared with the hedge fund’s gain of just 12.5% since the bet was made.

Buffett’s backing of the S&P 500 was primarily an argument about the significance of cost minimisation not philosophy. In his 2013 Berkshire Hathaway letter to shareholders, Buffett said he had instructed his estate to put 90% of his funds into the Vanguard S&P 500 and 10% in cash. Here’s his rationale:

“Both individuals and institutions will constantly be urged to be active by those who profit from giving advice or effecting transactions. The resulting frictional costs can be huge and, for investors in aggregate, devoid of benefit. So ignore the chatter, keep your costs minimal…”

This observation, from one of the greatest stock pickers in history, suggests that one does not necessarily need to agree with a passive investment philosophy to observe the weight of the evidence. The evidence shows, overwhelmingly, that investors are not rewarded by the high costs of active investment management.

When Russell Kinnel, director of research at Morningstar, attempted to identify the number one predictor of performance for investment managers, his findings also made a compelling argument in favour of low cost investing.

“If there’s anything in the whole world of mutual funds that you can take to the bank, it’s that expense ratios help you make a better decision. In every single time period and data point tested, low-cost funds beat high-cost funds… Investors should make expense ratios a primary test in fund selection. They are still the most dependable predictor of performance.” Russell Kinnel, Morningstar FundInvestor, vol. 18, no. 12 (August), p 1 – 3.

The second part of the debate focuses on persistence in performance. Every study of the persistence of active managers to deliver outperformance (and there have been plenty) shows that, once you control for risk, there is no statistically meaningful performance persistence.

To illustrate simply, Vanguard conducted a study in which they ranked all US equity funds in terms of excess return versus their stated benchmark over the five years ending 2008. They divided the funds into quintiles, separating out the top 20% of funds. They then tracked the performance of those top 20% of funds over the following five years through December 2013 to check for persistence. If the top quintile funds displayed any meaningful performance persistence, we would expect a significant majority to remain in the top 20% five years later. A random outcome would result in about 20% of the funds dispersed evenly across the five quintiles.

Active versus passive – the debate is over2

The results were close to random. Only 12% of funds repeated a top quintile performance, while 28% moved to the bottom quintile.

If costs really matter and performance persistence by active managers cannot be counted on, is there a reliable way outperform markets?

This is where we depart from the traditional passive vs. active discussion. Considerable research shows that low cost shares (based on a combination of fundamental ratios), small company shares and profitable company shares exhibit higher returns than the market over long periods of time. The figure below compares long term value, small cap, growth and total market indices for various markets.

Active versus passive – the debate is over3

Both value and small cap segments of the market outperform total market and growth. The implications of this are that not every share has the same expected return. This is probably because shares incorporate unique systematic risks, for which prices and market capitalisation are merely a proxy. And there are other proxies beyond just these two. The question is how to invest in shares with a higher expected return. Do we pay a manager to pick what he believes are the best of the bunch? Two problems arise with this approach – the manager’s fee (evidence shows this is not money well spent) and also that we lose some of the virtually costless benefits of diversification.

Passive investors can defeat both of these problems. They can use very low cost and highly diversified funds to access these sources of higher expected returns in a way that adds long term value over broad market benchmarks and peers.

In everything we do, the core of the argument is always evidence. We want to see evidence – evidence that has some statistical relevance; evidence that is persistent across time periods, pervasive across markets and has a sound economic rationale to it. We want to see evidence that ideas don’t work merely in theory or on paper, but can produce results after considering the costs of management, transactions and taxes. The problem with active management as it’s traditionally described is that the evidence simply doesn’t stack up.

As for the debate about active vs. passive… If it’s purely an argument based on conjecture and opinion, it will probably never end. However, if evidence counts for anything, we think this debate was settled a long time ago. Active versus passive – the debate is over4

Post: 167 March 15th, 2015

Tactical Asset Allocation

A consideration of the evidence

Short summary of our position on tactical asset allocation

According to Vanguard, “Tactical asset allocation (TAA) is a dynamic strategy that actively adjusts a portfolio’s strategic asset allocation (SAA) based on short term market forecasts. Its objective is to systematically exploit inefficiencies or temporary imbalances among different asset or sub-asset classes.”

Evidence shows that this approach has failed to deliver on its promised benefits, so we choose to take a different approach. Rather than time our moves out of expensive asset classes and sectors, we continuously tilt our portfolios towards relatively inexpensive assets, higher returning smaller companies, and profitable companies. In short, we believe in a consistent and robust process over the crystal ball approach.

We take the following common sense steps to achieve many of the promised benefits of TAA, without the unnecessary risk:

We provide continuous exposure to assets with higher expected return in a cost effective manner.
We invest in over 40 countries and 8,000 securities, from the outset reducing exposure to any one country, sector or company that may be problematic.
We work with advisers to ensure the overall portfolio can achieve the client’s objectives by a comfortable margin. We recommend reducing risk when it isn’t necessary and increasing it when it is, subject to achieving the financial objective.
We work with advisers to rebalance portfolios regularly, selling assets with recent good performance and buying assets with recent lower performance.

More detailed explanation of tactical asset allocation

In our view, TAA is simply market timing within a constrained or set portion of the whole portfolio.

For example, an institution may have an SAA of 60% equities and 40% fixed income, but utilisation of TAA may allow the equity allocation to vary between 50% and 70% and the fixed income allocation to vary between 30% and 50%. This is equivalent to saying that 80% of the portfolio will have a 60/40 allocation, and market timing will be performed with the remaining 20%. For institutional investors, TAA occurs at two levels – investment consultants allocating funds among different asset class managers and investment fund managers allocating funds among different asset classes and sub-asset classes.

It is our position that TAA is a futile exercise because market timing and style picking are unproductive at best, and potentially destructive at worst.

TAA is normally built on models that rely on ‘signals’ to determine when to go heavy on one asset class, at the expense of another. A well known example is the ‘Fed Model’, which compares the earnings yield on equities to the yield on 10 year treasury bonds. If the earnings yield (the inverse of the price-to-earnings ratio) on the S&P 500 were to fall below the 10 year treasury yield, a manager utilising TAA based on the Fed Model would interpret this difference as a clear signal to favour fixed income at the expense of US equities. The Fed Model, like all the other models used for market timing, has repeatedly been shown to be unreliable in both the short term and the long term. For further details, see Asness, Clifford, “Fight the Fed Model”, Journal of Portfolio Management, Fall 2003.

TAA at its heart involves predicting the returns of shares, called equities, and dividing a portfolio up between shares and fixed interest based on these predictions. In one of the most complete research papers of its type, called “A Comprehensive Look at The Empirical Performance of Equity Premium Prediction”, Drs Ivo Welch and Amit Goyal examine 22 common models of TAA and market timing. They make the following conclusion:

“Our article comprehensively re-examines the performance of variables that have been suggested by the academic literature to be good predictors of the equity premium. We find that by and large, these models have predicted poorly both in-sample (IS) and out-of-sample (OOS) for 30 years now; these models seem unstable, as diagnosed by their out-of-sample predictions and other statistics; and these models would not have helped an investor with access only to available information to profitably time the Market”

One problem with trying to read market sentiment for timing purposes is that sentiments can endure for years at a time. As John Maynard Keynes famously remarked, “The market can remain irrational longer than you can remain solvent.”

The reason why all these models are unreliable is so simple that it is easily overlooked. In three words, prices are fair. This means that every asset class has thousands of intelligent and informed people opining daily on the risks of the asset class in general, and all the securities contained in the asset class in particular. As a result of the discovery process, prices are set so that buyers can expect a return that is commensurate with the risk they assume. If an institution (or a consultant acting on its behalf) decides that an asset class is overvalued and thus should be pared back, the party taking the other side of the trade has the full expectation that they will be compensated appropriately for the risk they are taking. There is absolutely no reason for the institution or the consultant to assume that it has special proprietary knowledge that the rest of the market is lacking.

Now that our theoretical objections to TAA have been outlined, it would be useful to examine how mutual funds built on TAA have performed.

In a Wall Street Journal article from 2/11/2012, “How Practical is Tactical?“, Jason Zweig looked at 42 mutual funds and exchange traded funds with “tactical” in their name, based on Morningstar data. These funds were up 6.9% as of the date of his article – an average of five percentage points less than the various indexes they follow, according to Morningstar.

Over the past three years, these funds have gained an annual average of only 4.9%, or more than six points a year behind their benchmarks. A ‘tortoise portfolio’, which holds a steady 60% in US stocks and 40% in bonds, is up an annual average of 11% over the same period, more than twice the tactical funds’ average return.

Mr Zweig reminds us that two thirds of the tactical funds didn’t exist before the financial crisis erupted in 2008. The terror of watching stocks lose 60% in value during the crisis was the driving factor for these products, and Wall Street was happy to accommodate this demand.

Supporting this finding, long term research by Index Funds Advisers found that only four of 21 TAA funds with over 20 years of history (based on data collected by Morningstar) had outperformed a buy-hold and rebalance strategy of funds very similar to those used by us. Of course, this study didn’t include funds that had gone out of business during that time period.

As far as the ability of investment consultants to add value by moving funds from one manager to another, this myth was completely debunked in “Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors” (Financial Analysts Journal, Nov/Dec 2009). The authors evaluated over 80,000 investment decisions by trustees and consultants who oversee and advise on plans (typically large scale pools of retirement funds or endowments in the United States). As the chart below shows, in only two out of 18 years did plan sponsor decisions to move assets from one manager to another add value. The authors estimate that over $170 billion of investment value was destroyed over the period (1984 to 2007).

Tactical Asset Allocation

To summarise, TAA is simply one more attempt by big banks and others to package up luck and sell it as skill. Investors are better off with a sensible strategic allocation of low cost funds that reflects an appropriate risk level. Market timing and style picking should never be utilised for either a portion or the whole of a portfolio.

Post: 197 March 15th, 2015

Market timing: a picture is worth a thousand words

They say a picture is worth a thousand words.

We could easily spill a thousand words on market timing.

We could talk about how each trade involves a willing buyer and seller; each with equal interests, each with the same access to information.

We could talk about the fact that, over about 85 years, the S&P 500 has only gone up 51.02% of the days.

We could talk about the concentrated nature of returns. We could show that being out of the market, and missing the best month each year, drops returns by about 7% per year.

We could talk about a psychologist from Berkeley, named Philip Tetlock – who studied over 82,000 varied predictions, by 300 experts from different fields, over 25 years – and concluded that expert predictions barely beat random guesses. Ironically, the more famous the expert, the less accurate his or her prediction tended to be.

We could talk about Nobel Prize winner Bill Sharpe’s contention that timers need to be right 74% of the time, to overcome the frictions and costs of their moves.

We could talk about magazine covers, like the Death of Equities, that featured just before five years of 14.44% average compound returns for the S&P 500.

We could talk about how studies – ranging from the landmark paper by Brinson, Beebower and Hood, to the most recent study on New Zealand managed funds – have found that the average contribution of market timing to returns is negative.

We could talk about a study of 1,557 managed funds and 210 institutional funds, where the author concluded timing ability of managers is, on average, negative.

We could talk about how the majority of market timing newsletters underperform the market. We could talk about how, on average, market timing newsletters underperform the market by over 4.00%.

We could talk about how the market timing gurus whose calls are tracked have less than 50% accuracy.

We could talk about evidence that shows economists can’t time markets either.

We could talk about how the predictive power of last year’s return, to correctly forecast this year’s return, is 0.01%.

We could talk about the wise words of Warren Buffett, who said, “The only value of (share) forecasters is to make fortune tellers look good.”

We could talk about the simple logic that all market timing calls offset each other. If you buy, someone must sell. If you sell, someone must buy.

We could talk about a lot of things.

Or…

We could show you this picture of monthly returns and simply ask you to find the pattern.

Market timing- a picture is worth a thousand words

Wishing you the very best of luck…