I have written a great deal on the return premiums – styling them as the superheroes of passive investing with the potential to kick the ass of the market.
The problem is, like all superheroes the return premiums have a dark side – tragic flaws that cause them to underperform or even fail to show up just when you need them most.
Consider this an anti-hero article designed to show just how much Kryptonite there is out there that can transform the passive investing equivalent of a comic book fantasy into just another chump in pants.
Looks good on paper
On paper, the best of the return premiums (also known as risk factors) have been shown to deliver market-beating returns of 4% per year.
That’s Mr Fantastic. But it’s not possible in the real world that you and I live in.
Why?
Because those returns are generated in academic simulations that:
- Short equities
- Ignore costs
- Ignore taxes
The academics aren’t really cheating – at least not in the way that car makers cook up amazing fuel efficiency stats by testing their cars in labs rather than on roads and by removing wing mirrors and sealing up the door cracks.
The disconnect occurs because academic interest lies in understanding financial theory rather than producing practical products. (They can get hired for that later!)
The long and the short of it
The return premiums identified by academics are delivered by long-short portfolios. These theoretical portfolios buy equities with positive characteristics and short-sell those with negative features.
For example, the value premium can be defined as the annual average return on equities with high book-market ratios minus the annual average return on equities with low book-market ratios.
The problem is that UK investors can only buy trackers that invest in long-only portfolios.
As a general rule of thumb, you can assume that such a long-only approach will only capture 50% of the premium1.
So our hoped-for premium is cut in half at a stroke:
4% x 0.5 = 2%
Your real world fund also comes with costs that don’t trouble Ivory Tower boffins. A factor-based tracker might charge you anywhere from 0.3% to 0.8% a year, and this too must be deducted from a premium’s theoretical returns.
2% – 0.5% = 1.5% left of our premium.
Higher costs are common to factor-based trackers because capturing a premium may require a high turnover of holdings (for example in the case of momentum) or investing in small and illiquid securities (in the case of the size premium) that are saddled with higher spreads.
Worst still, the bulk of a premium may exist in equities that are ill-served by commercially available products.
For example, there’s evidence that large value US equities only beat the S&P 500 by 0.5% a year, whereas small value trounced it by 3.8% a year. Yet many value funds are concentrated in large value equities meaning that your factor fund may only collect the merest smidgeon of the premium.
0.5% x 0.5 (long only loss) – 0.2 (cost of large value fund over a regular large cap fund) = a miserable 0.05% left of the premium.
It’s scarcely worth the bother.
The size premium is particularly susceptible to hollowing out, given the Wild West that exists in small cap definitions.
MorningStar’s fund compare tool can help you spot the difference and find out what you’re actually buying into.
No more heroes anymore
The final return premium hazard to watch out for is essentially one of self-harm.
The very emergence of mass-market funds – enabling us to buy into a premium – can flood the space with cash, raise valuations, and lead to low future returns as assets become overpriced.
Rick Ferri, for example, now advises investors to expect a future return on the small value premium of just 1%.
That’s a note of caution you’d do well to extend to other return premiums like profitability and momentum.
Indeed, the paper Does Academic Research Destroy Stock Return Predictability? estimates that returns decline by 35% on average once a market anomaly is ‘discovered’.
So by all means pursue the premium path if you fancy a crack at outsized returns, but be aware that their superpowers might have waned.
Expect George Clooney Batman as opposed to Christian Bale Batman and you stand less chance of disappointment.
Take it steady,
The Accumulator
- The reality is more nuanced, depending on the premium, as this paper explains. [↩]
Add in on top of that the uncertainty factor. Investors (including myself at one time) see a figure like 4% outperformance for small cap value and thing that they WILL outperform by 4% a year.
Unfortunately the 4% refers to past performance and says nothing about the future, and even if it did, outperformance would just be more likely rather than guaranteed.
Unlike my active investing activities my passive investment approach is (deliberately) less interesting than watching paint dry or grass grow. I just split it between ETFs that track the FTSE 100, 250, global stocks, UK Gilts and corporate investment grade bonds. Then I rebalance each to 20% once a year (as long as trading costs aren’t more than a few percent of any trade), which takes about 15 minutes, and of course add new funds when transaction costs aren’t prohibitive. Boring, yes, effective, hopefully (or at least my son hopes so since it’s his JISA).
Very sobering but GOOD article!
In Your example the excess return is 0,6%. This means for me it covers the costs of a plain vanilla market cap weighted „traditional” portfolio. Due to this You have the result of a plain vanilla portfolio without costs. Otherwise 0,6% in the long run increases Your return due to the compound interest. In ten years with this extra 0,6% 100.000 grows to 106.165.
That 50% long-only hit looks tough. This is presumably specific to people trying to be smart and focusing on individual sectors of the market with risk premiums?
It doesn’t quite seem to square with the ‘track the overall market and capture the long-term updrift due to reflecting the gradual increase of goods and services in the world as time goes on’ passive axiomatic assumption which would inherently favour long-only.
Thanks for an excellent article. Despite being a long-standing and fairly disciplined passive investor I am occasionally tempted by the factor ETFs which are newish to the UK market. This article was a timely reminder that they probably aren’t a great idea.
You provide a great description of how the excess returns going forward are likely to be minimal. If you also consider the fact that there will certainly be long periods when the factor ETFs under-perform the market portfolio and that by their nature value ETFs can be quite undiversified they become even less attractive.
(Note: The iShares MSCI world value factor has 10% in only four companies Pfizer, Toyota, Intel and Cisco. It is also more than 25% in Japan stocks. Although the TER is a not too bad 0.3% the portfolio turnover rate is likely to be significantly higher than a market-cap weighted fund).
I’ll re-read your article whenever I get tempted……
@Ermine — I think you might be getting confused? It’s a 50% hit to the *excess return* from the premium factors. Academic research essentially inverts the premiums to get an extra return by shorting the losers on the same factor metrics. Some stocks will always do worse than others, and if the factors can capture a genuine differential (a big if!) then shorting the laggards should hold equally true, juicing returns further.
@TI I was getting confused. Thanks! And I can’t even blame it on the festive cheer yet what with the sun still not being above the yardarm 😉
I guess even a residual 0.6% on top of the usual 4-5% real return assumption is worth getting out of bed for as a decent uplift.
http://www.msci.com/resources/factsheets/MSCI_Factor_Indices.pdf.
@ Gregory “The chart above, for example, illustrates that over the period 1988–2013, a combination of a risk weighted with a quality strategy delivered a 3.1% annualized
return increase over the MSCI World Index with a significantly lower level of risk.”
One of the simpliest and best known strategies is the Dogs Of The Dow.
It buys the ten stocks in the Dow that have the highest dividend yield as of year end, equally weighted. It beats the Dow itself 12 of 14 times from 2001. Very, very simple strategy and still works. http://seekingalpha.com/article/2777285-dogs-of-the-dow-2015-firming-up
“A half a percent that can change your retirement” http://www.marketwatch.com/story/a-half-a-percent-that-can-change-your-retirement-2014-10-15?link=MW_home_latest_news
Sucessful high risk investors do not run high risks. They have an edge which allows them to recognise when something is percieved by the majority as high risk, when, in reality there is low to no risk.
Then they buy lots and make a killing whne the majority realise their mistake.
“Some maintain that once these successful value-based strategies are well known , stock prices will adjust and nullify their advantage. But I disagree. Warren Buffett had it right when he said in 1985, “I have seen no trend toward value investing in the 35 years I’ve practiced it. There seems to be some perverse human characteristic that likes to make easy things difficult.””
Siegel, Jeremy J. The Future for Investors: Why the Tried and the True Kindle Locations 4000-4002). Crown Publishing Group. Kindle Edition.
Wow, very interesting…
I guess based on this a lot of people will be wondering whether they need to change their portfolios.
Looks like Vanguard Life Strategy in the lowest cost platform is a way to go – just got to get my grain into gear to seriously think about dumping the extra stuff!
This blows-up lots of model portfolios!
Best of luck out there consolidating in 2015!
TA, great balanced article; thought provoking 🙂 Thank you
@Geo, think I am right in saying that Vanguard Life Strategy funds should be seen as “multi fund” products. The investments held under the umbrella of any of these Life Strategy products have their own ongoing management charges. Though in comparison with “active” funds the charges are negligible the total is never-the-less higher than what you see on the tin.
I didn’t think that Life Strategy had hidden pricing as its very against Vanguards policy. The TER should include all underlying costs, so the costs of the funds within. I know you can get the equivalent cost down if you made your own mix of funds but the beauty of Life Strategy is that it controls the demon investor in you better.
Anyone know any different if there is a dual layer of pricing though?
@ Geo – you’re correct. There isn’t a second tier of charges for the LifeStrategy funds. You pay the initial charge and TER as published. Kean, you’re right that famously multi-funds to cake on an extra layer of charges but Vanguard don’t behave this way.
@ TA, pleasantly surprised – thanks for the conformation.
@ Geo, thanks for clearing up this misapprehension.