Paradigm Shift Chronicles

April 12, 2012

Mean-variance optimization, fathered by Harry Markowitz in 1952, is still a widely used approach to asset allocation today despite the fact that the underlying assumptions of the theory have some serious flaws.  This approach to asset allocation is derived by using statistical methods to estimate expected returns, volatilities, and covariances.  All of this information gets plugged into a piece of software called an optimizer.  The optimizer then sifts through every possible combination of assets and produces a graph showing a curve called an efficient frontier.  Ranged along it are a series of optimal portfolios, from the lowest risk and return to the highest.  

This is an elegant theory, but the theory can only approach reality if the inputs (expected returns, volatilities, and covariances) are accurate going forward.  One approach to come up with those inputs is to forecast them, but forecasts, even expert forecasts, are notoriously inaccurate.  A second way to come up with those inputs is to just use the long-term averages.  An examination of the chart below should illustrate why the latter approach is fraught with danger.

Source: Rex Macey, Investments & Wealth Monitor, March/April 2012 Issue.

This chart demonstrates that historical relationships can change.  The five-year correlation between domestic large stocks (Russell 1000) and the MSCI EAFE index varied but never exceeded 0.6 from the start of the dataset until the late 1990s.  Consultants used this data to argue for international diversification.  Who would have expected based on historical data that the correlation would rise to the 0.9 level matching the correlation of large U.S. stocks with small U.S. stocks?  I suspect those relying on international diversification were quite disappointed.

Rather than relying on an approach to asset allocation that makes enormous assumptions about how the future should look, why not embrace a tactical approach to asset allocation that is designed to adapt?  Correlations can change, variances can change, and returns can change and tactical asset allocation still has the potential to produce excellent returns over time.

Click here to view a video presentation on our Global Macro portfolio to learn about our approach to tactical asset allocation.

Click here and here for disclosures.  Past performance is no guarantee of future returns.

Posted by:

Fund Flows

April 12, 2012

The Investment Company Institute is the national association of U.S. investment companies, including mutual funds, closed-end funds, exchange-traded funds (ETFs), and unit investment trusts (UITs).  Members of ICI manage total assets of $11.82 trillion and serve nearly 90 million shareholders.  Flow estimates are derived from data collected covering more than 95 percent of industry assets and are adjusted to represent industry totals.

Posted by:

Clients Demanding Tactical Allocation

April 12, 2012

Advisor One had an article reprising the findings of the most recent Curian Capital survey of advisors.  Their results will not surprise anyone.

• Nearly two-thirds of the advisors say that they have begun using more tactical asset allocation strategies to mitigate economic volatility, and more than half of respondents report they are using more alternative investing strategies.

• As a result of market volatility, nearly 4 out of 5 advisors report an increase in their clients’ demands for more conservative investments; in addition, 72% say their clients have an increased demand for guaranteed income features, 55% report an increase in demand for more tactical asset allocation, and 47% report an increase in demand for alternative investments.

Clients want tactical allocation strategies.  This may be for diversification, but it may also be for risk mitigation, since strategic asset allocation didn’t help them much last time around.  (I added the emphasis above.)

I find the demand for more tactical allocation interesting for a couple of reasons.  Even ten years ago, tactical allocation was derided as market timing by hard core strategic allocation advocates.  Now clients are objecting to holding asset classes during a prolonged nosedive, whether it is in service of diversification or not.  There’s much more awareness of the risk inherent in strategic asset allocation given that we have gone through two bear markets in the last decade or so.

Now that everyone is a tactical asset allocator, the question really boils down to methodology.  What process are you going to employ to make your allocation decisions?  I will suggest that gut feel will get you in trouble almost immediately.  Relying on emotion is not a good way to go.  I think either valuation or relative strength methodologies will work in the long run, but they put different analytic demands on the allocator.

To run a valuation-based process, you need to have a reliable way of generating reasonably accurate expected returns for asset classes.  (Simply using past history will not work, as many strategic allocators discovered over the past decade.)  That’s not an easy task.  It requires a ton a relevant data and a lot of testing to make sure your forecasting process has some validity.  You’re still going to have a large margin of error, so your portfolios will never be optimal.  Paradigm shifts are still going to create major problems.

Relative strength offers a reasonable alternative.  You need to have a reliable ranking method for the assets included in your universe, but we like it because you don’t have to forecast.  Instead, you are relying on the ability of the process to cast out losers and adapt to new trends.

Valuation and relative strength don’t have to be mutually exclusive.  In fact, excess returns are typically negatively correlated.  This is just a fancy way of saying that the two strategies tend to perform well at different times.  Combining two tactical allocators, one using relative strength and one using valuation, is also a very good way to go.

Posted by: