Confessions Of An Asset Allocator

Several years ago when I made my first real attempts at managing my own assets the idea of a fixed asset allocation strategy made a lot of sense. Diversify by allocating broadly to a wide range of foreign and domestic securities using fixed income to control volatility. Rebalance regularly, limit transactions as much as possible, and always mind fees. When you run strategies such as these through a back-test the results come out to be fairly decent over many different time periods and market cycles. There is nothing wrong with these strategies, and the vast majority of individual retail investors out there are most likely well served through such investment policies. The difficulty is often finding a strategy that aligns with one’s personal preferences and tolerance for volatility.

As I have gained more and more experience with my own investments I have found myself struggling to blindly follow a fixed allocation strategy. Take the past several years for example. As the US exited its last recession domestic equity markets have risen at incredible rates while foreign markets haven’t faired as well. For those uninformed on the historic performance of financial markets it would appear that US stocks are the superior asset to buy as they have recently performed well. But the bargain-hunter inside of me says the opposite, that foreign stocks have done poorly relative to the US, and are therefore cheaper and priced more attractively (on a relative basis at least). If I’m looking to purchase an asset on the cheap it would seem reasonable to focus my attention on international stocks. But how much should I buy? Does it make sense to break a predetermined fixed allocation as a means to take advantage of this potential opportunity?

I previously examined the consequences of such action, which I termed overbalancing. To be more precise overbalancing is essentially allocating capital in excess of a predetermined policy to purchase an asset (international stocks) that have recently underperformed. The results in my examination were a mixed bag. In instances where the underperforming asset was down for a single period the strategy produced slightly higher returns. In situations where the underperforming asset continued to decline returns actually suffered, and in some cases were downright abysmal. Again, this gets at the idea that short-term asset performance can be difficult if not impossible to predict on a consistent basis.

What I was attempting to execute with this study was, to a certain extent, in conflict with the fundamental premise of asset allocation. Asset allocation strategies are meant to take advantage of the idea that the short-term performance of any given asset cannot be predicted. Thus, asset allocators diversify among many different assets with the knowledge that, in any given period of time, some will perform well and some will perform poorly. Over very long periods of time this strategy works primarily because the underlying assets tend to appreciate in value. Additionally, the action of rebalancing maintains diversification and helps reduce portfolio volatility over very long periods of time.

By overbalancing to an asset with recent poor performance I was indirectly making a claim that said asset was “cheaper” and expected to appreciate in value. Again, this sometimes works, and sometimes doesn’t. I was essentially engaged in a crude form of market timing–something that I generally frown upon. However, I don’t think I was being completely irrational either. I wasn’t, afterall, purchasing the more expensive asset because it had a recent bout of excellent performance. Perhaps my perspective and attempt were merely crude and unrefined.

One of the more famous and trite examples of successful tactical asset allocation comes from the Yale Endowment. Over the past 15 years it has produced an average (arithmetic) annual return of 11.4% with a standard deviation of only 12.8%. For comparison, the S&P 500 generated an average return of 6.8% with a volatility of 18.9%. The corresponding allocation targets are shown below through 2015

Yale Endowment Target Allocations
Source: Yale Investments Office [1]
If these tactical moves were providing little to no benefit I doubt the endowment would still be engaged in modifying their target allocations annually. But this alone is insufficient to justify that an individual should adopt such an investment process. There are massive differences in the assets, resources and expertise that Yale has access to compared to an individual investor. That being said I think there are some insightful takeaways:

  • Remain invested at all times. Holding cash or short-term fixed income carries with it a high opportunity cost. High rates of return can’t be realized without investing in assets that are likely to generate those returns. But…
  • Losses can and will happen. Remaining fully invested exposes a portfolio to a greater potential for loss. In 2009 the endowment posted a 24% loss while the S&P 500 returned over 26% coming out of a recession. Ouch!
  • Tactical movements are (and should be) small. Asset class turnover was the largest in 2009, when 13.5% of assets were re-allocated. On average, asset class turnover has run at an average of about 4.8% year-over-year. These data points are a tremendous lesson for individuals as any transaction carries with it some form of inefficiency or expense that will harm returns.
  • Use a variety of tools. According to endowment reports expected returns, mean-variance analysis and other forms of statistical probabilities are examined and tempered alongside qualitative assessments. [1] The implication being that there is no one magic bullet that will provide the ideal path forward. Experience and reasonable judgement are still required.

As I think about this more broadly, the idea that a target allocation can be successfully modified in a manner that improves returns and reduces volatility brings into question the idea of market efficiency–that assets are always priced correctly according to all available news and information. My opinion on the matter is somewhat frank. As a scientist I recognize that the efficient market hypothesis is, as of today, an idea and a theory–not a law. Gravity and conservation of energy are physical laws. They can be proven, observed and documented. As governing principles we use these laws to understand how real world systems function with a high degree of certainty. Markets on the other hand, have no such laws. Participants aren’t always rational. They are without a doubt emotional, and I too fall into this category. The so called wisdom of crowds, as it turns out, isn’t always very wise. Bubbles and mania occur every so often, causing assets to be mispriced, which may result in catastrophic crashes. Financial history has shown that such events have happened before and will happen again.

These thoughts considered, there would appear to be something to the idea of tactical asset allocation. But an evidence based emotion-free model is still required. Blindly purchasing an asset, regardless of whether it goes up or down, just doesn’t sit well with me. I fully realize these are emotions, and have no place in investment planning. Perhaps they can be proven wrong.

Post Script
Some well written arguments against market efficiency can be found here:
Losing My Religion by Sam Lee
The Great Divide Over Market Efficiency by Cliff Asness and John Liew

1. Yale Endowment Annual Reports.

What I’m Reading
The United States of Market History Amnesia (Tony Isola)
Seth Klarman (Mike Dariano)
Demystify Risk (Michael Carpenter)
Building That First Portfolio (Lowell Herr)
Indexing Is Capitalism at Its Best (Cliff Asness)
The Importance of Investing with ‘Appropriate Fear’ (Lawrence Hamtil)
Managed Futures: Understanding a Misunderstood Diversification Tool (Andrew Miller)
Pundit Or Professional? (Charlie Bilello)