So far in this series I have shared three ways in which asset allocation as typically executed by investment professionals can lead to undesirable outcomes. In Part 1, I shared that asset allocation frequently leads to too much risk being taken on by investors, as well as deworsification. While in Part 2, I pointed out treating frontier and emerging markets as an asset class can starve frontier and emerging markets of much needed capital, thus impoverishing them. Most recently, in Part 3 I pointed out that asset allocation strategies are rarely submitted to the independent benchmarking that active managers are usually subjected.

Now I switch from complaining and explaining, to put my own skin in the game. Specifically, what is it that I believe can be done to rectify the problems that I identified in the first three parts of the series? This suggestion, by the way, comes from a reader (thank you, Liju Jose Kurian, CFA, FRM) – let that be inspiration for you to join the conversation : )

Possible Solutions

Unintentionality Fixes

Recall that in Part 1 of The Dark Side of Asset Allocation I outlined the idea that investors frequently lose site of the actual risks they are taking on due to how they go about portfolio construction. One example is that an incredibly risky asset’s purchase can be justified because of a belief on the part of the purchaser that ‘every portfolio needs risk capital.’ Another variation is a belief that markets are unpredictable so adopting a ‘who the heck knows?’ philosophy. While still another version is the ‘well its beta is 0.65 and its covariance is low’ and adding it in because of the past performance statistics.

So how do you fix this issue? Very simply, thank you very much. Instead of fixating on the relative level of risk as asset allocation would argue you do, you instead come to grips with the absolute level of risk embedded in the security. Not to do this is comparable to letting a repeated violent offender become one of your kid’s school teachers because ‘recent performance indicates he isn’t violent.’ Meaning he has been in prison and didn’t commit any crimes. Never mind the fact that if you took the time to do his pscyh-evaluation you would learn that he suffers from homicidal fantasies. Bottom line: absolute risk is absolute risk no matter how your elegant, sophisticated, quantitatively driven portfolio construction pretends it isn’t and whose techniques actually serve to cover it up.

Deworsification Fixes

I also argued in Part 1 that asset allocation can frequently lead to alpha-sucking deworsification (parenthetically, I just saw this spelled diworsification this week…I like it). Here the myopia of asset allocation leads to security after alpha-gobbling security after beta-worshipping security being added to a portfolio until an imagined ‘efficient frontier’ and its utility maximization is attained.

How to fix this is not as hard as you might think. Use some other metric beyond beta and covariance for determining whether or not to add a security to a portfolio. What might that be? Hmmmm. How about valuation, for instance? How about securities that are trading at a discount to an estimate of fair value/with a margin of safety? How about a security that is estimated to meet and possibly exceed a minimum return on invested capital figure?

If you want to get very clever and even sophisticated about it, you could create a list of possible business and investment risks that are nearly universal to all businesses and securities. Then you could ensure that not too many of your securities are sensitive to the same risks. Instead you can look for as little concentration in the various risks you have identified. I am being obscure about this because if you know my investment success over the years, a part of it was due to this very approach. Sorry to be oblique about it. If you would like to bid for my services then I am happy to share some secrets with you : )

Asset Classless Fixes

I must confess I am not quite sure how to fix asset allocation unintentionally impoverishing nations as I suggested in Part 2 of The Dark Side of Asset Allocation. Why? In truth I do know how to fix it: don’t you do it! There, that was easy. Here I am imploring you to not trade emerging markets and frontier markets as an asset class. But heck I am not unwise (sorry for the double negative), I know that most people and investors are going to continue to buy emerging market and frontier passive funds and ETFs, beggar thy poor nation.

You could correct this by ensuring that you have an independent view of an entire country and its economy. That would require a bit of macro economic work on your part. You could also identify industries within nations that might perform well. Again, some fundamental analysis is necessary to pull this off. Also, there are certainly not products currently that really allow you to pick and choose industries within emerging and frontier markets. But maybe someday Blackrock or Vanguard will log some frequent flyer miles and do this.

Last, some clever derivatives fellow could develop a currency neutral ETF/passive fund that would eliminate or significantly mitigate the risk of getting the currency and exchange rate wrong. One of the commenters to this series asked about what to do to adjust the asset allocation models and mentioned some Black-Litterman stuff. This kind of mind could branch out to create a way of smoothing the choppy waters of emerging/frontier markets. But, not me. My sophistication stops at the Sortino ratio, bootstrapping yield curves, and having a unique view of convexity.

Benchmarking Fixes

Part 3 of The Dark Side of Asset Allocation had me wailing about the fact that asset allocation strategies are never subjected to the same benchmarking standards as are active managers. Ironic isn’t it? I say this because so often active managers are picked based on their homogeneity via a deep preference for low tracking error and style drift. Why? Asset allocators need predictable returns so they can plug you into their (presumably) more smart/more good asset allocation [yes, I meant more smart, not smarter]. Yet, no one knows if this works in the aggregate. Argh!

[By the way, Part 3 was the least popular in the series. Perhaps this point cut too close to home?]

How do we fix this given how wild and crazy and diverse are asset allocation strategies? First, make asset allocators define a strategy and a process rather than willy-nilly, finger-to-the-wind winging the asset allocation. Second, have the asset allocator define their time horizon. Third, have the asset allocator define their opportunity set, including asset classes, securities lists, and so on. Fourth, using a faster computer than my laptop, calculate the possible portfolios that could have been constructed in adherence to their stated strategy and universe. Fifth, figure out how well they did relative to these optimal portfolios. Sixth, we call this a ‘batting average.’

This method also has the advantage of allowing them to be compared to the component parts of their asset allocation strategy. Say grace. Amen!

Universal Fixes

Say none of the above fixes that I proposed works for you, either do to your disagreeing with me, the cost of implementation, or that altering the way you do business is fraught with rug-burn-like frictions, then what? Essentially, all of my complaints about asset allocation boil down to the same complaint: the unconscious application of a tool.

Consciousness

At the heart of this complaint are several things to note and that provide a hint as to a possible universal fix. First, is that word ‘unconscious.’ Here the universal fix is for consciousness to be brought to bear on your asset allocation. This means slowing down for a little while and questioning the assumptions that underlie your use of asset allocation. For starters, ask yourself when was the last time your approach to asset allocation was subjected to renewed scrutinization? When was the last time your approach was changed in any significant way?

If the answer to these questions ranged from unknown to a long time ago, then it probably means you need to place some attention on to your asset allocation approach again. Another possible response to a lack of consciousness is to insert a pause between your decision to act and your actual trade. That pause is a perfect time to ask how your decisions intentionally and unintentionally affect you, your client, your firm, and others. What are the underlying emotional factors, if any, as to why you are making your decision? Use this pause to ensure cause and effect are in harmony.

Tools

Asset allocation is a tool, just like a screwdriver, or discounted cash flow analysis for that matter. It is not a god, or demi-god. Said more directly, and less politely, sacred cows make for tasty hamburgers. I have been in too many heated discussions over the years with people who think that questioning asset allocation and trying to make it better is tantamount to heresy.

If your response to my series has been a gut-check, or an urge to fire off a heated comment to me, then chances are that asset allocation has ceased being a tool, and has morphed into being a limb. Better for it to be a tool that can be dropped out of your limb, so that you can use others tools, too. Tools that are actually designed for the jobs under consideration.

Jason A. Voss, CFA – Your Next Excellent Hire

I would love it if you would receive notification of my new articles – sign up here 🙂