Read any decent book
Modifying your portfolio like this unlocks the potential for greater performance in exchange for extra helpings of risk.
But how big are these premiums, and are they worth playing for?
Return premium size and volatility
Return premiums have historically been handsome. And they’ve delivered over prolonged periods of time, too, although we can’t know if their pep will stretch into the future.
The table below shows an example of US return premium pay-offs. It’s taken from a paper published by Robeco Asset Management and is available in full from the Journal of Index Investing:
The excess return column on the left shows the result delivered by each return premium portfolio over and above the return offered by ‘risk-free’ government bonds, between 1963 and 2009.
If you swoosh your eye to the right-hand column, CAPM Alpha, you can see what each return delivered on top of the broad US equity market, after adjusting for risk.
Particularly eye-catching is the extra 4.6% annualised excess return from the value and momentum portfolios – albeit at the expense of extra volatility.
Very tempting.
Risk it for a biscuit
That volatility column is not to be dismissed lightly. It shows that only the returns of low volatility stocks are subject to less violent swings than the equity market. (The clue is in the name, I guess).
Investors in return premium strategies therefore have to be able to tough it out when the market gives your cunning plan a right booting. Individual premiums have lagged the market for 10-20 years during the roughest patches.
But what the Sharpe ratio tells us is that time spent adrift in the choppy seas of risk has proved worth it for the higher overall returns.
The higher the Sharpe ratio, the better the risk-reward trade-off. And every return premium portfolio beats the market in this respect.
Indeed, the momentum and value portfolios deliver a risk-adjusted reward that’s almost double the market index.
Correlation
It gets better. There’s plenty of evidence that the pitch and yaw of multiple return premiums has diversification advantages when combined in a portfolio.
The Robeco paper illustrates this nicely by showing the relative lack of correlation between the return premiums:
- A correlation score of 1 means that two assets move up and down together.
- -1 means they move in opposite directions.
- 0 means that the relationship is random.
Any score from around 0.3 to -0.3 implies a lack of correlation, which accounts for most of the relationships above, bar value with low volatility and small cap stocks.
However, value has been known to decouple from low volatility when recession strikes and value stocks come under pressure.
The lack of correlation between the different premiums implies that some may wax while others wane, rather than all plunge together like climbers on a rope.
Diversification
Here’s another study from the Journal of Indexes (Europe) that shows the diversification benefits gained by melding return premiums into a joint portfolio, between 1990 and 2011:
(Click to enlarge)
- The S&P 500 stands in for the market portfolio.
- Fundamentally Weighted is a value premium strategy.
- Equal Weighted is a small cap strategy.
- Alternative Beta Composite is a portfolio that is divided one quarter each into the value, small cap, low vol and momentum strategies.
Note that the combined portfolio beats the market by 2% a year and with much better risk adjusted returns (see the Sharpe Ratio).
Even the overall risk is slightly lower – 14.2% as compared to 15.1% – indicating that the different equity strategies mesh together to create a stronger, more stable package.
We can see that the low volatility strategy has the best risk-adjusted returns and lowest risk overall, but it would be a courageous investor who hangs their hat on that bravura performance continuing. As always it’s best to spread your bets.
To illustrate the point, investment strategist and Larry Swedroe’s co-author, Jared Kizer, has analysed the likelihood of return premiums triumphing over longer time frames.
Kizer looked at US stock market data between 1927 and 2011 and came up with this table:
Over monthly periods, you can see that the chances of positive returns for size and value are only a smidgeon over 50-50.
Happily, performance is more satisfactory over longer horizons, though you still have to be prepared for size and value to lag the market 25% of the time over five-year periods.
More interesting still is the table below from Kizer showing that one of the premiums will bring home the bacon 96% of the time:
If diversification is a free lunch then let’s make it bacon sandwiches!
Caveat City
The strength of the return premiums change over time, by country and indeed according to how each premium is defined. So the numbers above may well differ from the results of other studies, and from the results in the UK.
Moreover, strategies that attempt to capture return premiums in reality tend to have bigger holes in their butterfly nets than is allowed for by the academic studies that originally pinpointed the opportunities.
There is though plenty of evidence to show that the premiums have persisted across international markets and the historical record.
I’ve seen enough to make me believe that it’s worth tilting my portfolio to collect at least some of the extra return juice. We’ll delve deeper into the possibilities in forthcoming posts.
Take it steady,
The Accumulator
“much like a car modder might bolt a giant spoiler onto his Vauxhall Astra”
Looks ridiculous, accomplishes nothing? 😉
Just kidding…
Excellent article! I wish the combining studies went back further than 1990. I look forward to the follow up articles on applying the return premiums.
Thanks for a really great post. I have been inclining to the same view but have held back for three or four reasons:
1 concern that ongoing premium might be less (partly back- testing bias, partly market efficiency closing premium)
2 concern that the added costs / complexity aren’t worth the candle
3 concern about lack of company diversification
4 concern that psychologically I might find it difficult when say the low-volatility tracker under-performs the broad market for long-periods.
Your thoughts as always welcome!
Hi Accumulator, great post.
I would like to tilt my portfolio towards these premiums, but problem is in the UK we simply don’t have the index funds to replicate these, unless I’m missing these funds. I use Hargreaves as my platform.
Any info on funds would be great.
http://indexinvestor.com/Free/fundamentalDividend.php
This is a bit out of date but it makes the case for erosion of any return premium through market efficiency. (Which is not to say that the market is perfectly efficient just that it is efficient enough)
There are funds inthe UK that offer these premiums but some of them, like DFA, are really aimed at the large or institutional investors.
If you google for funds that exploit these factors you should come up with a couple of others that are retail friendly.
Thanks accumulator, interesting stuff as always, I’ll be keen to see what options you find for these things, as in the UK we are limited by choice, especially if you want to avoid dealing cost of ETF’s/IT’s etc although hopefully over the next year the dust will settle with RDR and there will be some better options.
If anyone is interested this is how I tilt my SIPP portfolio on HL:
Commodities (JPMorgan Nat resources) 6.0% :Enhancer
Emerging Markets (Black Rock Tracker) 7.0%
Asia Pacific (Black Rock Tracker) 7.0%
EU (Black Rock Tracker) 4.0%
Vanguard Mix 60% EQ 53.0%
Value (Vanguard UK Div) 6.0% :Enhancer
Hedge (Troy Total) 5.0% :Smoother
Small Cap (Investec UK) 6.0% :Enhancer
Property (Black Rock Tracker) 6.0%
I’m not 100% happy with this, it has many flaws and having to use managed funds to get what i want. I’m not sure Commodities will have much benefit and know Troy isn’t really a Hedge fund. I may end up selling these and also converting my small cap to the HSBC FTSE 250 fund. Maybe when the funds build up I will convert to ETF’s a bit, plus move away from Blackrock but at the moment they aren’t big enough to warrant the platform fees. My TER is about 0.7 in total.
Anyway just though this might interest some people, I’d be keen to see how other people tilt theirs.
Cheers
Geo
Hi Geo – I have the same prob with small cap and use Marlborough Special Situations (my only active fund and it wrecks my weighted TER). I do this because I read somewhere that the small cap premium is really about true small caps not the medium caps you get in FTSE 250.
There may still be a diversification reason for tilting to FTSE 250 though as it is not fully correlated with FTSE 100.
@All — The Accumulator plans to offer up some ideas for how to tilt in the UK (or possibly by selective investments in the US) in a future article. These ideas are great fodder for that — cheers.
Regarding small caps, I previously put The Accumulator onto the Aberforth Smaller Companies Trust, which I’m very familiar with for my active investing sins.
It is definitely small-cap orientated, and the manager has a value bent, too (often comparing the P/E and yield of its holdings with the wider market, for example) so you might get a kick there, too. Obviously it’s expensive to own compared to passive investing . It’s also had a bit of a run which has narrowed the discount a little, which is good if you bought when it was wider but arguably increases the risk a little now.
Clearly not for committed passive investors, though if memory serves The Accumulator held his nose and bought some in despair at the lack of small cap passive options. (Or perhaps he was just considering it?)
I have held in the past but don’t currently, so this is doubly *not* advice to buy, just an idea. As ever, please do your own research.
Hi Geo, Passive Investor
I based my asset allocation after reading Tim Hales excelent book – Smarter Investing. My current SIPP portfolio with H & L is:
SWIP UK Equity – 12%
Vanguard Dev World, ex-UK – 28%
BlackRock Emerging Markets – 8%
BlackRock Property – 6%
ETFS Physical GOLD – 6%
L & G Index Linked UK Gilts- 20%
Cash – 20% (will eventually convert to HSBC conventional UK Gilts when panic over Gilt prices/yields settles)
Note this is classsic 60:40 Equities/Bonds.
Total TER for this portfolio: 0.38%.
Part of me wants to just run with this portfolio without changes and annual rebalancing to retirement in 15 years, another part (the dark active side) wants to add some premium tilts and make it more complicated.
Regards, Jon
@ john. Not too different from me (though I aim for 50:50). Just interested whether the TER includes the platform charge ?
Superb article.
I think careful diversification is the only free lunch going & I’m a big fan of non-cap-weighted funds, even if I don’t currently hold any just yet. (See my comments in the previous ‘inefficiencies’ article, which pointed out the Ossiam & iShares minimum volatility ETFs, as well as some others.)
One article I’d like to see is how the correlations change over time, which is very significant. A quick google throws this up http://www.cboe.com/Institutional/JPMCrossAssetCorrelations.pdf which I haven’t had time to read yet, but I would like to see the essence extracted through the Monevator distillery…
(I’ll stick the answers to last weekend’s numeracy questions up in the coming weekend. 🙂 )
Greg
Another quick article thrown up is this, which is US-centric but nicely shows a key problem of assuming static correlation co-efficients.
http://seekingalpha.com/article/245781-contemplating-asset-class-correlations-stocks-bonds-commodities-reits
Take a look at the last chart ( http://www.capitalspectator.com/011011c.html ). See how everything apart from bonds get spanked together. Without Government bonds, even though they smoothed out returns during normal circumstances, the maximum drawdown is pretty much the same in a crisis.
Hmm, this has got me thinking. Perhaps portfolio can be put together that tries to capitalise on all of this and has no cap-weighting… (Even if it needs some active parts…)
Off the top of my head I’d probably have the following mix:
MINV (iShares World min volatility ETF. 0.3%. Mostly US & Japan.)
FCS (F&C Smaller companies IT. 0.81%. Mostly UK & US)
Cazenove UK Opportunities. OEIC. 1.07%. UK
If emerging markets were desired there are a number of options which can be added, e.g.
EMV (iShares EM Min vol ETF. 0.4%)
AAS / SST (Asian Smaller Companies, TER: 1.51/1.01)
or SEMS (EM smallcap 0.74%)
Bung in a couple of other diversifiers (e.g. Bonds / commodities / psuedo-hedge) and I’d say that would be a pretty good portfolio! (It would need rebalancing though…) I say Pseudohedge as I mean things like SL GARS, Cazenove Absolute UK Dynamic & CF Eclectica absolute macro rather than proper hedge funds due to the fees! (I’ll admit to a 2% holding in BABS though…)
I think iShares are doing well at novel ETFs – they even have a value ETF (IDJV) too, though I haven’t considered it as I think small + MV is sufficient. However, if Vanguard offer up something similar, I’d certainly check them out!
Greg
@ McTrex – heh, heh, yes, the analogy is definitely open to interpretation.
@ rjack and Passive Investor – I share your concerns. But even if we had evidence for a strong premium going back 200 years there’s no guarantee that it would continue to deliver during our investing lifetime. It’s a risk. That’s the inescapable truth. So is investing in the total market even though we’ve all read that it’s always delivered over 20 year cycles. Without the risk, there’s no premium. No reason to reward. I keep hoping to read the article that puts my doubts to bed, but it’s not out there. I have to live with those doubts as the price of investing in addition to the TER. Continued research does make me comfortable that tilting is the right thing to do for me, though. Even though I’m also now convinced that the products available to retail investors will not capture the full premium as defined by the academics.
PI, most of the recommendations for passive portfolios on tilts only allocate say 10% of the equity portion to small caps and 10% to value etc. That seems a reasonable way to mitigate against market tracking error. Re: diversification, a tilted fund may be more concentrated than a broad market fund, but it’s adding to the diversification of my entire portfolio as I’m exposing myself to a greater number of return factors than just Beta.
@ Rob – care to throw in one or two suggestions for funds?
I am planning an article or two outlining the passive options for UK investors. There aren’t many but they are out there. The Investor is right, I have invested in Aberforth to gain UK small value exposure.
No hedge funds for me, though. What was it Buffet said: “a compensation scheme dressed up as an asset class.”
Very interesting post and it seems to make a lot of sense but the devil as I see it is in the detail.
In the first table, unless you’re tracking named indices – or even better, ETF’s that follow them and including charges – then it depends a lot on your definition of ‘value stocks’ and how you choose your ‘momentum’ stocks.
As the article was published by an asset management company, they obviously want to magnify the difference between these asset classes and the baseline in the hope of selling relevant funds.
In the third table we have defined funds but a shorter timescale and this looks more promising. I wonder how much of the extra return is simply from the diversification itself rather than the constituent indices. And what about the effect of charges? I’m not clear if and how this was rebalanced.
What would be really interesting would be to do this for the UK market using known indices (FTSE100, FTSE high yield, FTSE350 etc.) as the UK market appears to differ from the US – I do remember reading that the FTSE equal-weighted index did no better for example.
@Geo
Troy Trojan might not officially be a hedge fund (which is a good thing, because it has no performance fee and an AMC of only 1%), but it is better than most hedge funds at delivering the sort of performance hedge funds claim to provide.
@ Accumulator
I too spurn hedge funds, because of the high level of fees and poor record of performance. But I do dabble in a few of the funds that have strategies similar to hedge funds but much more acceptable fees, such as Troy Trojan, Newton Real Return, Standard Life GARS. They dampen the volatility of my portfolio somewhat, but without accepting the low returns from bonds.
An interesting point arises from the discussion about the difficulty of finding passive funds available to UK investors that target the sort of strategies highlighted in this article. As a smart investor, I want to invest in passive funds, but I also want to invest in funds that target value stocks and momentum stocks. If I can’t combine the two, am I better picking passive funds that have a more general strategy or picking active funds that do target something like a value strategy all momentum strategy? In the post RDR world, most active funds have an AMC of 0.75%, which is not that much higher than you would be likely to pay for a passive fund with the same strategy. So, I don’t feel too bad about picking active funds if they really do have the strategy that I am looking for.
@ Mike – I can (and will) write a fairly lengthy post about why it’s highly unlikely a small-time passive investor (that’s me, btw) will capture the kind of numbers seen in any paper. But, I do think there’s a good chance of capturing some of the effect and that’s good enough for me.
The size of each effect absolutely depends on the definition of each portfolio of equities. Those definitions will vary between different ETFs and indexes too. Not all Value funds are equal.
While you’re right to be wary of any single source and potential conflicts of interest, I chose this paper because it was the clearest example I could find of 5 factors at work. There are plenty of other papers and posts out there that show numbers in the same ballpark, and often over longer periods of time.
Here’s a piece by Larry Swedroe that shows something similar:
http://www.cbsnews.com/8301-505123_162-57413875/how-the-four-stock-premiums-work/?tag=mncol;lst;3
Sure, he’s a financial advisor who uses Dimensional Fund Advisor products but he also spends an inordinate amount of his time trying to help and educate investors not just through his books and blog but unsung on public forums.
Charges will lower your return (that’s one of the reasons we won’t get the numbers that appear in the academic studies) and rebalancing is neither here nor there for me. Obviously it will effect the numbers outputted by any one study, but, I’m not relying on the data from any one study. Rebalancing will also effect the returns we see in practice but we can’t know in advance what the optimal rebalancing strategy will be. For me, rebalancing is an act of risk control that ensures I stay diversified.
Here’s a link to a paper that shows the value and momentum premiums stand up across international markets including the UK:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1363476
I’ve read again and again that historical evidence for the premiums has been found across the board, but that’s not saying anything about the FTSE Equal Weighted index. The FTSE 100 Equal Weighted isn’t going to be much of a small cap index so I’d rather look for the small cap premium in something like the Hoare-Govett Smaller Companies.
@ Ivan – Passive Investor has listed above most of the reasons why you might want to stick to a vanilla passive strategy, but I must admit that even pre-RDR I’ve invested in Investment Trusts that I judge to be operating in small value territory.
I agree with you that the drop in AMCs widens the opportunity set, but I haven’t looked into the options yet. Obviously the devil is in picking funds with the right strategy and then monitoring for style drift. I’d be very interested in hearing about any funds that have caught your eye in this regard.
@ivanopinion – I too have a few of the Absolute Return-style funds which you mention in my portolio which is otherwise passive except for components where I cannot find a suitable UK-domiciled passive fund.
My original plan was to see how well the Absolute Return-style funds performed compared to a passive portfolio which has a fairly conventional split between bonds (Global non-hedged to cash in on sterling weakness), Corporate Strategic Bond and Index-linked Gilt) and Vanguard Equity trackers. I created this structure just over a year ago soon after I was able to exploit the Bestinvest Select platform and located the range of Vanguard low-cost index trackers. Obviously, over the last year, because of the very good returns on equities, the Abs. Ret. funds have lagged but I’m interested to see how well they hold up when there is a major market correction. They do individually and jointly have a low volatility and held up well in 2008/9 but past performance etc..
Prior to my use of Bestinvest Select and Vanguard, I tried to maintain a
similar asset allocation but mainly using active funds (except for the FTSE All Share index tracker. I’ve used Trustnet to compare how well my old portfolio and new are performing and, it is interesting and pleasing to note that the new portfolio has outperformed the old by 2% over the last 14 months with a similar volatility. Lower charges account for part of that plus the performance of the Vanguard FTSE Equity Income fund.
Once there is a market correction, I’ll rebalance probably retaining the Abs. Ret. funds as part of the low-volatility portion of the portfolio. If, however, any of the Abs. Ret. funds don’t hold up reasonably well, I’ll ditch ’em.
@Accumulator – isn’t the FTSE All Share indicator a less than perfect proxy for momentum?
New Vanguard on HL!
Looks like HL have finally got round to adding new Vanguard funds including Global Small cap. SO there’s an option at last 😉
Groovy 😉