Originally published by Cuffelinks
“We cannot suppress the powerful intuition that what makes sense in hindsight today was predictable yesterday. The illusion that we understand the past fosters overconfidence in our ability to predict the future.” Daniel Kahneman, Nobel Prize Winner, in Thinking, Fast and Slow.
“The only value of stock forecasters is to make fortune tellers look good.” Warren Buffett
“Far more money has been lost by investors preparing for corrections, or in trying to anticipate corrections, than has been lost in corrections themselves.” Peter Lynch
Like many people who manage their own portfolios, I actively engage in tactical asset allocation (TAA), despite evidence that it’s a waste of effort. TAA is market timing with a fancier name, and usually involves switching from equities to defensive assets and vice versa in anticipation of a major stockmarket move. Unfortunately, nobody rings a bell to tell when this is about to happen, so we all have our own techniques.
I think there are two main reasons why people engage in TAA:
- We think we’re good at it.
- The rewards for correct decisions are fantastic.
At an institutional level, many trustees and asset consultants prefer instead to set a static strategic asset allocation, and rebalance back to these set percentages as markets move. They may tolerate some small range either side of a target allocation, but relatively few public super funds in Australia tactically adjust their portfolios significantly the way an individual investor might.
Is TAA a waste of time?
Of course, academics have an expression to describe the belief that you can outguess the market: illusory superiority. People overestimate their positive qualities and abilities, and underestimate their weaknesses. Examples quoted in the research include:
- in surveys of driving safety and skills, 93% of people in the US put themselves in the top 50%
- at the University of Nebraska, 68% of faculty members rate themselves in the top 25% for teaching ability
- at Stanford University, 87% of MBA students rated their academic performance above median
- in a survey attached to the SAT exam in the US (taken by about one million students a year), 70% put themselves above the median on leadership ability, 85% on ‘ability to get on well with others’ and 25% rated themselves in the top 1%.
There have been many studies showing nobody really knows where markets are going (click this link for an excellent report on the perils of market timing). In 1994, some US academics analysed over 15,000 market timing calls in 237 investment newsletters over 13 years, and found that over 75% of the newsletters produced negative performance. In 2012, Morningstar compared 210 tactical asset allocation funds with a simple Vanguard 60/40 default fund, and with few exceptions, the tactical funds performed worse, were more volatile and ran as much downside risk as the simple fund.
Even those who have a well-researched and systematic approach to TAA often have a timing problem. It can take years after an asset allocation move to know if it is correct. Any fund manager with an open-ended fund who called the GFC downturn correctly but altered their asset allocation two years early probably lost a lot of funds as the market continued running until 2008.
This was the case for funds who used long term fundamental measures (like Nobel Prize laureate Robert Shiller’s ‘CAPE’ ratio model, and Vanguard founder Jack Bogle’s real dividend yield model) to under-weight equities as soon as they became expensive two or three years before the boom ended. They underperformed during the best part of the run up and they lost a lot of clients and funds. To make matters worse, they also lost when they re-entered the market a year too early, when the long term fundamental models said the market was ‘çheap’ in early 2008, right before the worst part of the crash in late 2008 with the collapse of Lehman.
And it’s difficult for institutions to do TAA with any conviction as they are obsessed with benchmark risk and competitor risk. Public super funds, for example, ‘force’ themselves to be in most asset classes all the time rather than avoiding some completely from time to time. Over the last year or so, I have heard many institutional investors say Aussie bonds look very expensive at present, yet most have an allocation to them.
Individual SMSFs are much better placed as benchmark and competitor risks are not such a worry, and an SMSF can also take a much longer timeframe to make decisions.
I personally believe it is still worth giving TAA a go and I think I do better than the market at least 51% of the time if I have an adequate long term time frame to work to.
If you could do it well, is TAA really worth it?
Let me show some simple examples of why so many of us strive for the elusive TAA heaven. I will be drawing my data from calendar year statistics in the Vanguard Australia Asset Class Tool, a publicly available source of returns across all asset classes since 1970.
Assume I have $1,000 to invest, and there are no transaction costs and no tax leakage (these are unrealistic assumptions for many people, but it is needed for simplicity and won’t change the message). The measurement period is 10 years from 1 January 2004 to 31 December 2013. The six asset classes chosen for the purposes of this article are:
- Australian shares
- International shares ($A unhedged)
- Australian property (listed REITs)
- Cash
- Australian bonds
- International bonds ($A hedged)
1. Unrestricted and perfect asset allocation
The Vanguard table shows which asset class performed best in each calendar year. If I could invest in any asset class without restriction (0% to 100%) and I had a perfect ability to asset allocate and did not care about diversification, then over the 10 year period from 1 January 2004 to 31 December 2013, I would have invested the initial $1,000 as follows:
1 January 2004, 100% Australian property, for a return over the year of 32.0%
1 January 2005, 100% Australian shares, for a return over the year of 21.0%
Etc, etc, switching between asset classes every year for 10 years.
This perfect approach turns $1,000 into $9,628 after 10 years, an annual compound growth rate of 25.4%. And for completeness, the worst allocation would have resulted in $1,000 turning into $445 after 10 years.
2. Set 70/30 asset allocation
Most institutions have a 70/30 asset allocation for their key balanced funds. Let’s assume, using our six asset classes, the average asset allocation was:
- Australian shares (35%)
- International shares (25%)
- Australian property (10%)
- Cash (5%)
- Australian bonds (15%)
- International bonds (10%)
With $1,000 invested at the beginning and the above set asset allocation, $1,000 becomes $1,962 after 10 years, a compound average growth of 6.98% per annum. This assumes the asset class allocation at the beginning was not rebalanced each year. If the standard asset allocation was rebalanced each year, then $1,000 turns into $2,094 after 10 years, a compound return of 7.6% per annum.
3. Allowing tactical asset allocation around the benchmark
Let’s now assume institutions are permitted to do tactical asset allocation around the standard asset allocation with ranges as follows:
- Australian shares (35%) – range 25% to 45%
- International shares (25%) – range 15% to 35%
- Australian property (10%) – range 5% to 20%
- Cash (5%) – range 0% to 10%
- Australian bonds (15%) – range 10% to 25%
- International bonds (10%) – range 5% to 20%
- Overall growth assets (70%) – range 50% to 80%
Using TAA discretion and allocating to the best returns, we have 80% growth (shares and property) and 20% (cash and bonds) from 2004 to 2006, then we switch to 50% in 2007 and 2008 (and thereby reduce the impact of the GFC), back up to 65% growth for the 2009 equity recovery, down to 50% growth in slow equity years of 2010 and 2011, and 80% growth since then. While this is far less extreme than the perfect results in example 1 with no diversification, these numbers are within the range assigned to many asset allocators within a balanced fund.
The results? The $1,000 invested at the beginning turns into $3,457 after 10 years, a compound annual return of 13.2% per annum. Compared with the above static 70/30 asset allocation that is rebalanced each year, this tactical asset allocation with perfection yields an additional $3,457 – $2094 = $1,363 after 10 years, a significant improvement.
We chase the Holy Grail
In a theoretical extreme, an asset allocator could have turned a $1 billion fund into $9.628 billion in 10 years with perfect vision, or a 70/30 balanced fund could have delivered 13.2% per annum instead of 7.6% per annum with excellent TAA. It’s little wonder the attraction of trying proves difficult to resist.
And, of course, there’s a third reason a lot of us do it. We enjoy it, it gives meaning to our role of managing money, and for many it’s a lot more fun than setting and forgetting 70/30 and going to the races. But I’ll leave the last word to John Bogle, the Founder of Vanguard:
“Sure it would be great to get out of the stock market at the high and back in at the low, but in 55 years in the business, I not only have never met anybody that knew how to do it, I’ve never met anybody who had met anybody that knew how to do it.”