Investment Management. Clarity and Transparency.

Clarity and Transparency

Static allocation from MPT to AW

NAAIM says 95% of advisor funds remain in Static Allocation Strategies

Adaptive Selective Allocation

Cycle Dynamics

Macro models, Bayesian models, and combinations with static allocation

Human judgment still necessary

Always learn and adapt

Investment Research Sources

————————————————

“An investment in knowledge always pays the best interest”  – Ben Franklin

Clarity and Transparency.

Investors constantly face a bewildering cloud of investment advice. How does anyone filter through it all to find out what really makes sense or at least resonates with their own thinking?

I don’t have a simple answer because, as Howard Marks says, “it’s not supposed to be easy”.  Investment involves assessing a multi-factor dynamic process. It also involves the unknowable future and how to prepare and react.

The only way I know how to shed some light on my own approachimages is to provide as much clarity and transparency as possible in describing my own process, at least at the macro level. This is necessarily my own journey through many investment strategies and has to take some short cuts to be readable and understandable.

 

 

I have attempted to make this travelogue through investment management thinking as straightforward as possible. Whether my analysis of each approach is accurate or the justification for the shifts in thinking are valid can be left for the observer to examine and reach their own conclusions. This is what works for me.

Static allocation from MPT to AW

The starting point for investment strategy is always static allocation. It is easy to understand and simple to execute. Choosing a fixed allocation between different asset classes diversifies risk, while at the same time enables investors to participate in the long term returns of multiple asset classes. It is also easy to manage as periodic rebalancing to maintain the weightings can be done as infrequently as annually, and ensures that over the long term the investor is buying low and selling high at the margin.

Given this starting point the searchReal Results is then on for the “best” static allocation, but what does “best” mean? Different investors may prefer different allocations, and there are many different style factors that come into play.  Any portfolio needs to take account of income, tax, and risk and there may be other elements to consider.

Once these personal choices have been determined then static allocation theory will determine how to go about constructing the most effective portfolio to fit the clients preferences.

For static allocation theory I think there are two main approaches worth analyzing, as perhaps they describe the range of options. First, and by far the most widely used, is modern portfolio theory (MPT).

https://en.wikipedia.org/wiki/Modern_portfolio_theory

The benefit to MPT is that it provides1015MPT a first approximation methodology for portfolio diversification which optimizes for the best return relative to risk. It provides a workable solution, but it does make multiple assumptions.

The issues surrounding this approach concern whether all the assumptions that go into these calculations are realistic or valid premises. The Wikipedia article above, lists some of the assumptions and thinking behind this theory. More on this in other analysis below.

One other approach worth understanding is what is commonly termed All Weather (AW). This approach introduces thinking about the whole economic cycle and how to offset risks throughout the whole cycle. Here is how this thinking evolved.

http://www.bwater.com/Uploads/FileManager/research/All-Weather/All-Weather-Story.pdf

Many studies have been done to compare the effectiveness of different methodologies used to construct static allocation portfolios.  The longer the testing period the more valid are the results. In terms of return relative to risk the evidence in favor of a simplistic version of the AW portfolio measured over a 42 year period, shown in the link below, is very compelling.

On all measures of volatility, in particular1015bridgewater_gr_big the sharpe ratio and drawdowns, the AW portfolio shows a huge reduction in risk over all time periods. It also demonstrates much reduced correlation. Yet after 42 years it loses only around 1% pa in performance relative to the much higher risk of more conventional portfolios. Put another way AW lowers risk substantially, by around 50%, and yet produces far higher risk adjusted alpha, multiples higher than other static models over the very long term.

There are many other portfolios that could be tested against the AW portfolio, however, none in my experience has produced a fund that has such a durably low level of risk with such a high risk-adjusted alpha over a 42 year period.

This may be a first clue to the power of how introducing thinking about the whole economic cycle can significantly improve investment strategy.

NAAIM says 95% of advisor funds remain in Static Allocation Strategies

It is worth fully understanding the limitations of static allocation approaches because at a recent National Association of Active Investment Managers (NAAIM) it was quoted that 95% of advisor funds are focused on static allocation theory. The 5% that is actively managed may well be growing, but it still represents only a fraction of the total.

There are two main problems with static allocation approaches. First of all, how stable is the recommended allocation? If the underlying assumptions are unstable then just a 6 month period could produce a very different static allocation solution. If so then how can this really be said to be an appropriate methodology for a long term static allocation strategy?

The second main problem with static Commodity092615allocation is that there is very little inbuilt risk management. Some investment downtrends can last for years. Commodities, generally, have recently been in a 4 year downtrend. Static allocation stays in the allocation no matter how long or significant the downtrend with only a very long term solution through reweighting. This should work out in the very long term, assuming that the asset class is genuinely in a long term cycle, but it is clearly far from optimal and can produce very long periods of underperformance.

Beyond these two main issues, static allocation theory raises a whole range of questions, some of which become almost philosophical in nature. Are static equilibrium models based on valid assumptions or are other models more realistic? Are simple solutions good enough or sub-optimal? Can some additional complexity really produce reliably better performance over time? These may be the most important questions, and they can only be resolved by engaging in what is well reasoned, tested and effective. Even then, value is subjective, not everyone will reach the same conclusion.

These are very significant problems to resolve, but there are clearly many inefficiencies with static allocation, but they also represent a considerable opportunity for improvement from active asset management.

Adaptive Selective Allocation

In order to move up the return to risk curve and solve for the multiple shortcomings of static allocation, it is necessary to shift to active management methodologies.

The most promising research on how to take the first steps in this direction is shown in the link below. This powerfully demonstrates not only the shortcomings of static allocation models, and MPT in particular, but also the very productive direction that active management can take to develop systematically better results. These techniques not only show a significant shift upwards in return for risk, they also produce far more consistent returns, by responding automatically to trends.

http://www.advisorperspectives.com/dshort/guest/BP-120514-Adaptive-Asset-Allocation.php

Cycle Dynamics

As was discovered from the AW portfolioAssetSelector092615 above, cycle thinking can add a great deal to asset allocation. So now that we are armed with the superior results from selective adaptive allocation models, additional improvements can be made by adding in cycle thinking as well.

Cycles contain knowledge that heighten detection and selection within the adaptive selection process. These benefits show up clearly in the Cycle Dynamic models by demonstrating a very substantial improvement, across all asset classes, in the return for risk equation in all four models shown in the link below.

http://chrisbelchamber.com/introduction-to-cycle-dynamics-active-investment-management/

Now it is possible to develop a comprehensive framework for a complete portfolio solution with a dramatic improvement in return for risk results as well as consistency compared to static allocation approaches.

http://chrisbelchamber.com/the-cycle-dynamics-portfolio/

Macro models, Bayesian models, and combinations with static allocation

Finally, there are now some very effective1015AW4quad macro economic models which have further improved cycle detection, even to the level of almost daily marginal adjustment. Some of these models can also improve execution, and separate out the time period for different levels of decision making. Hedgeye have produced such a model with a very good track record.

Combining different complementary models can add great value for investors. There is no reason why these more complex systems could not be used as additions to a partial static allocation. The knowledge derived from these more complex systems may even inform the choice of which static allocation to choose over the next year or so.

Human judgment still necessary.

Computers can almost fly planes these days, but would you get on a plane without a pilot who understands the computers and could shift to manual if necessary?

Much of the investment system business1015MM wants to do away with human judgement in applying active management systems. System acceptance very often requires no human discretionary involvement. Compliance to system status means that systems become officially invalid if human judgement is permitted or exercised, but could that possibly be a mistake? These systems are incredibly valuable, but what if combinations are better and not necessarily programmable. What if the systems themselves have strengths and weaknesses. Wouldn’t you want a pilot that respects and understands the systems but knows when a judgement call needs to be made?

To do this well demands understanding, and so also some human involvement!  It is important to consider not only the benefit of the systems but also the ability of  an operator to use some judgment, and problem solve for new developments and solutions.

Always learn and adapt

I have been using systems and rule based approaches for over 30 years., but I don’t use some of the systems I used 20, 10, or even 5 years ago. Always learn and adapt. Keep time tested approaches for what they provide, but never stop innovating.

If you want to be successful over the very long term, this is not a choice. I hope to be updating and improving this note repeatedly.

Investment Research Sources

Finally, sources of information make a huge contribution to investment results. Time is precious, so it is very important to filter through to find the best information available. Comprehensive but not overwhelming. My best subscription information sources are listed below. Fortunately, I became a life time subscriber to the top 3 at an early stage, so the current cost is very low.

Stansberry & Associates                                          17 daily, weekly or monthly newsletters

The Oxford Club                                                          20 daily, weekly or monthly newsletters

Agora Financial                                                           21 daily, weekly or monthly newsletters

Palm Beach Research                                                   7 daily, weekly or monthly newsletters

Tradestops                                                                    Lifetime member

Sovereign Man                                                             Lifetime member

Hedgeye

Macro Watch – Richard Duncan

Chris Weber

Churchouse Letter (Hong Kong)

Private Investor

Price Value International

Four Pillars

 

 

 

 

 

 

 

 

 

 

 

This entry was posted in Market Notes. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *