In a comment on my recent post on why investor’s should not try to out-perform, Robin Sole brought up the important issue of how to choose tracker funds, and the lack of a good totally passive approach to investment.
A global tracker fund replicates the performance of the market portfolio of the largest possible market. It therefore sits on both the efficient frontier, and by combining it with debt or government bonds it is possible to reach any point on the securities market line (in theory: in practice the spread between borrowing and lending, together with trading costs, pushes reachable points a little below the line). This answers one of Robin’s questions: a tracker fund does not need to explicitly implement portfolio theory.
At this point it is already becoming clear that a single investment in a tracker fund is not likely to be the only investment you need:
As far as I am aware there are no passive funds that invest globally in equity and debt. To replicate the global market portfolio you will need both a global equities tracker and a bond tracker. You also need to make additional investments, or to borrow, to gear the fund up or down to suit risk preferences. In practice, most people are likely to buy bonds to gear down.
In addition, most people have individual needs that are not necessarily met by a tracker that replicates and index. For example, if you live in Britain and are saving for retirement, you probably want to protect the value of your savings relative to British incomes. You therefore want a heavy UK weighting. On the other hand, you should probably diversify as much as possible. So you need both UK and global trackers.
Next comes the question of investment styles. There are value, growth, income, small cap, large cap, and sector indices, and trackers that follow them.
Given the strong evidence for the value effect, and evidence for the size effect, I would be inclined to put a disproportionate amount in small cap value shares. I do not know if there is such a thing as a UK, or global, small cap value index tracker, so it may be nessesary to compromise on a US fund or a UK value tracker.
Finally the most difficult question: what proportions, and rebalanced how often. I do not have anything like a scientific answer to that. For the typical British investor I would put half in UK value (preferable small cap value), and split the rest between global (or separate US, European and Asian) value (again, small cap when possible), and bonds. The UK weighting may appear high, but with so many British companies (even small ones) having global businesses, the true economic exposure will be lower).
Obviously the proportion in bonds will vary with appetite for risk. The usual advice about being more cautious as you grow older also applies.
I would also advocate a laid back approach to rebalancing. Simply rebalance when the proportions change significantly. Look at it once a year and rebalance when it seems necessary. Keeping the proportions exact hardly matters unless you have been more exact than I suggest in deciding them: and even then the difference it makes is likely to be tiny.
The answer to Robin’s question is, yes, you cannot completely avoid decisions. However, you can restrict your decisions to a few top down choices that can be made on the grounds of reasonably solid evidence, and your own needs and preferences. For most people this may result in holding a several different tracker funds and occasionally rebalancing them. With a sensible choice of style you are likely to do better than most active investors, at less risk and doing less work.
Many thanks Graeme, an excellent response. It is one of the best I have read on the subject, although one which leaves me to think that, given I have 20 to 30 years until retirement; I should forget bonds and take a fairly leveraged global equities income tracker (dividends reinvested).
Is there any reason at all to hold bonds over the long term? Obviously you want to hold them as you near retirement but all they will do early on is drag down your final returns.
What do you think?
Thanks for the good questions Robin!
Yes, the odds are very much in favour of equities outperforming bonds over the long term, although perhaps by not as much as in the past. If you are sure you will not need the money before then (or if it is locked in a pension scheme), putting it all in equities makes sense.
Why an income tracker rather than a value tracker? It may well be the right decision (it is not one I have thought much about), I just want to know your reasoning.
Below is a link to an interesting paper questioning the common wisdom of rebalancing from equity to bonds as people grow older (Basu and Drew, “Portfolio Size and Lifecycle Asset Allocation in Pension Funds”). The central argument is that people’s contributions into their pension savings tend to be largest when they get closer to retirement, which means that by following the common strategy you will have equity exposure on your relatively small initial contributions only, and once you start putting significant money to work, you switch to bonds, and forego growth potential. http://centerforpbbefr.rutgers.edu/2007/Papers/119-Basu%20%20Drew_LIFECYCLE.doc
Hi Graeme,
I should have said a Growth and Income tracker. The growth will keep the portfolio growing over time and the income will prop up the the bad years a little. Note that I would want the yield to be no more than about 1% to 2% in general.
I do not really have an opinion on value investing. I certainly would not regard value investing as passive and therefore I would be suspicious of any index tracker that followed a value investing concept. I am not even sure if there is such a thing for UK investors.
In order to gear up the investment I would want to look at an ETF.
Note that this is purely hypothetical as I currently use a different investment style. However if I start to lose interest and get lazy then I would probably try this one.
Regards
There is strong evidence for the value effect.
There are purely mechanical value strategies. All the research on the value effect necessarily uses a mechanical strategy, tracking a value index is little different from this.
There is a lack of choice in the UK: I know of one European value index tracker ETF (large cap, unfortunately), and no UK ones: the nearest you can get is probably an income index tracker. There is plenty of choice in the US….
Hi ijontichy
I just read the paper in the link above (LIFECYCLE.DOC) and it seemed quite interesting at first, but at the end I do not really see what was so great about their conclusions.
As far as I can tell, they are simulating investing in bonds and stocks for the long term using historical data to work out the best ratio.
We know that historically bonds will return about 6% over the long term and stocks will return about 12%. Therefore it is hardly rocket science to conclude that investing in stocks for the longer term will generate better results and the more money you invest, the better off you will be. You could proably use simple algebra to work out similar results.
The most notable thing about this paper is that they highlight the problem of switching to bonds as you approach retirement. You will lose out on some great growth potential with only a small amount of downside risk (assuming you have a well diversified equities based portfolio).
I have to say, from all the evidence they gave, I would be most happy to choose the option
Lifecycle (35,5) assuming I had reached my retirement goal after 35 years. Otherwise I would choose Lifecycle (40,0).
The assumption behind life-cycle strategies is that as you get older, avoiding the worst outcomes becomes more important. This is not unreasonable: if you have “enough” saved, why take a risk, even if the odds are good?
As Robin says, it is fairly obvious that equities are likely to out-perform. However, this does not mean it is always a good idea to put yourself in a position where a major market crash at the wrong time will wreck your life.
Hi Graeme,
I agree with you completely. Unless you really understand what you are doing, it would probably be unwise to move into just equities in the latter half of your investment life.
The only way to attain average market returns is to be very familiar with asset allocation techniques which is not easy to understand at all.
I have been actively using asset allocation concepts for a year now so I feel I understand it a little. From the start I have been looking at adjusting the target allocations periodically. The only question now is whether I should adjust them every 3 months or every 40 years :-)
In 10 years time I may be confident enough to do this all the way to retirement!
One final question, do you know of any links which demonstrate, through research and empirical data, the benefits of mechanical value investing over other types of investment strategies?
Many thanks
Robin, studies of the Dogs of the Dow style strategies have been done, but I do not have links or references for them.
Studies of the effectiveness of different strategies often work by reducing them to mechanical strategies that are back tested, so you might find useful results there as well.
If you find anything interesting please let me know.
I realize you’re arguing for passive investment here, but the lack of UK small cap tracker shouldn’t perhaps mean taking on currency risk via a US or Euro small cap tracker unless that’s what you’re after, too.
I think a couple of UK small cap investment trusts offer acceptable returns at reasonably low fees for UK investors – try Rights and Issues Trust or Aberforth, which have both been hammered in the bear market and are also on discounts.