Record inflows of $236bn into Vanguard funds in 2015

Glen_Clark

Sam_J12 wrote: »

can choose to avoid under-performing funds and sectors.

How do you know whether they are going to under-perform or not?

TheTracker

Sam_J12 wrote: »

By the way, ditching "poorly performing" funds/stocks to buy those doing well is also known as "momentum investing" and has enormous amounts of academic literature and practical application supporting it as a way to predict future short term performance dating back to the 1930s. It is one of the few enduring exploitable anomalies in the stock markets.

Not really. You can track momentum with a beta tracker, like the ETF Vanguard just released. Literature says its worth less than a percentage point per year over the very long term.

Chasing returns based on previous experience is dumb. Here is a link from just yesterday
http://news.morningstar.com/articlenet/article.aspx?id=735070

AlwaysLearnin

Sam_J12 wrote: »

By the way, ditching "poorly performing" funds/stocks to buy those doing well is also known as "momentum investing" and has enormous amounts of academic literature and practical application supporting it as a way to predict future short term performance dating back to the 1930s. It is one of the few enduring exploitable anomalies in the stock markets.

That sounds more like "sell low, buy high" to me, which is preferably avoided for obvious reasons...

darkidoe

Linton wrote: »

I believe you should be basing your investing following a top-down strategy. Not on choosing a set of the best performing funds.....

1) Decide on a high level allocation based on the general characteristics of each type of asset and the need for a high level of diversification. This could be 100% equity for very longterm investing, or perhaps 50% equity, 40% bonds 10% property for some more medium term growth perhaps with income. Or whatever is approprpriate for your circumstances.

2) Now you can divide up each asset type giving a % for each. eg by geography, company size, and sector for equity. Government versus corporate, time til maturity, risk for bonds.

3) Finally you need to identify the optimum funds to meet your %s. So you only need to compare funds in very narrow bands. You never need to see the whole universe. It is here where the active/passive question comes in, right at the end. For some requirements passive may be as good as anything else and cheaper. For others passive may not be a viable option. The key point is that the choice is not a bet on active outperforming passive nor on lucky managers, but primarily one of which funds can actually do the job.

My long term growth portfolio ROI of 4.5% for last year seems to be one of the highest reported here, if one neglects small high risk portfolios. The portfolio is based on the above strategy and is 100% equity and 100% managed. Average fund fee 0.78%, the highest fees > 1% are all ITs which I found interesting. The relatively good performance was helped considerably by the assignment of a higher than average % to small companies funds, an area where passive investing is difficult.

Can I ask what's the % of small cap you include in your portfolio compared to large cap?

redux

Glen_Clark wrote: »

How do you know whether they are going to under-perform or not?

Easy. You decide, then you're either right or wrong.

Then, as we can see in some comments here, if things go well great, congratulate yourself, and if they don't that's also great as it's an opportunity to buy more at a good price before they go up.

What I don't understand on here is denying or denigrating the results that other people had. Apparently I'm a mug. It doesn't look like it to me, going by last year, though I shouldn't claim too much credit as my approach was nearer supine than just passive.

Sam_J12

TheTracker wrote: »

Not really. You can track momentum with a beta tracker, like the ETF Vanguard just released. Literature says its worth less than a percentage point per year over the very long term.

Chasing returns based on previous experience is dumb. Here is a link from just yesterday
http://news.morningstar.com/articlenet/article.aspx?id=735070

I don't want to sound condescending, but I probably know a lot more about it than you having read extensively. It certainly is not dumb to believe that recent performance predicts future performance, and exploiting this idea can be worth far more than 1% per year. One seminal paper you might like to read is Jegadeesh and Titman (1993) ( http://www.e-m-h.org/JeTi93.pdf ) where they show that exploiting the momentum effect led to a 1% excess performance per month in the US market during the sixties, seventies and eighties. In a follow-up paper in 2000 they showed that this strategy continued to outperform to a similar level throughout the the 1990s. Their strategy was extremely simple - buy the best performing stocks from the last six months and hold them for six months. Performance in the last 3-12 months categorically gives information about likely performance in the next 3-12 months. This really is unequivocal.

And people who say things like "that is just buying high and selling low" simply don't know what they are talking about. "Buying high and selling higher" would be much more accurate.

TheTracker

Sam_J12 wrote: »

I don't want to sound condescending, but I probably know a lot more about it than you having read extensively. It certainly is not dumb to believe that recent performance predicts future performance, and exploiting this idea can be worth far more than 1% per year. One seminal paper you might like to read is Jegadeesh and Titman (1993) ( http://www.e-m-h.org/JeTi93.pdf ) where they show that exploiting the momentum effect led to a 1% excess performance per month in the US market during the sixties, seventies and eighties. In a follow-up paper in 2000 they showed that this strategy continued to outperform to a similar level throughout the the 1990s. Their strategy was extremely simple - buy the best performing stocks from the last six months and hold them for six months. Performance in the last 3-12 months categorically gives information about likely performance in the next 3-12 months. This really is unequivocal

Yes it does sound condescending. I read the titman paper just last week as it happens, as it was cited 3 pages into Swedroe's "Incredible Shrinking Alpha" book that I read over the new year. Informative book, particularly the discussion of the q-factor model which successfully does away with momentum as a factor since momentum is a result, not a reason.

The punchline of the book is summarised by fama and french's conclusion that "Unexplained average returns for individual portfolios" after adjusting to factor/beta factors "are almost all close to zero".

So, build a portfolio based on well researched asset allocation science and well researched factor/beta science. Don't bother with pin striped fund managers who use your fees to buy yachts, they don't add anything greater than the fees they charge.

Linton

darkidoe wrote: »

Can I ask what's the % of small cap you include in your portfolio compared to large cap?

Using the Morningstar Xray tool the overall allocations in my long term portfolio are:

Large cap:45%
Mid Range:27%
Small Cap:24%

The missing 4% is probably cash held within funds.

My funds are rather less weighted to small/medium cap, but some of the mainly largecap funds hold some smaller companies.

For comparison VLS100 is
Large Cap: 81%
Mid Range: 15%
Small Cap: 3%

bowlhead99

TheTracker wrote: »

The punchline of the book is summarised by fama and french's conclusion that "Unexplained average returns for individual portfolios" after adjusting to factor/beta factors "are almost all close to zero".

If i were looking for an opportunity to be critical of the classic market-cap-weighted passive funds, this sort of research would be gold.

Effectively, another way of looking at that paragraph is to say that every time a dumb regional or world tracker is beaten by an active fund that *doesn't* simply allocate capital to every single equity opportunity based on size and skew your portfolio to a small number of large companies in certain sectors... then there will always be a reason for it.

" well, it's not surprising you got better returns than my tracker because

a) you didn't concentrate all your money into the big companies but spread your money better among the opportunities to make sure you used medium and small companies too, which generate better returns in the long term, so beating me doesn't count, because you cheated by using a "size factor"

b) you didn't pile your money haphazardly into the market letting most of it land on the companies which had more shares in issue, but you thought about what you were doing and bought companies more competitively priced relative to their assets or profitability, so beating me doesn't count, because you used a "value factor"

c) you didn't just buy every company out there throwing money into the dartboard and hitting ABC Co most because it's biggest, but instead did proprietary research into companies and sectors in detail and bought the ones that appeared to be investing in themselves for the future and had management teams that were successfully delivering share price growth over time. So beating me doesn't count, because you used investment factor or momentum factor.

d) you only won because the portfolio you selected had a different set of risks because it wasn't the complete theoretical market basket in the full market weights because when investing across 3000 companies for diversification you didn't end up putting over 3% of all your money into Apple and Exxon. I'll define "taking risk" as "any kind of departure from my market basket". Consequently, you only got your 10% return compared to my 8% return by taking extra risk. If I control for risk, and take out the extra returns you got by not using my portfolio allocation, we got the same.

Therefore, everyone not using a UK all share tracker for their UK equity exposure needs to know that they are taking risks, and any out performance against the tracker, is only because of reasons. The unexpected variance of returns they get against the tracker, after adjusting for all the reasons, is close to zero."

In other words, it is not worth paying money to have a manager make investment decisions, because he is only going to go and use research to identify ways to do better or worse than a tracker, and once he delivers a result more suitable for your needs than a tracker for the various reasons, we can nit pick all those reasons and control for them and explain that the out performance was due to the reasons.

"In other news, I compared the sports performance of Usain Bolt and Michael Phelps with the average UK resident. Once I controlled for genetic factors, and for gender, training, protein intake, lifestyle and determination factors, they were not much better at running or swimming than anyone else, so I don't know why you'd pay money for someone to give you a tip that you should put them in your team."

So, build a portfolio based on well researched asset allocation science and well researched factor/beta science. Don't bother with pin striped fund managers who use your fees to buy yachts, they don't add anything greater than the fees they charge.

I would agree of course you should have a portfolio based on sound decisions and research. If an advisor or manager constructs you a suitable portfolio based on their good research, it is not unreasonable to pay them and it doesn't really devalue their product if they use the income from selling the product (and from investing in their own products and their friends' products) to buy striped trousers or a car, boat or plane.

Of course as I said earlier, I would prefer to pay lower fees where I can. Some of my broad exposure in my pension comes from things like iShares size factor ETFs which are relatively cheap though inherently more expensive than a cap weighted equivalent.

There is no harm in following a reasoned and researched approach to portfolio construction using all the tools you can find, though a theoretical "best tool for the job" may be hard to identify in some areas - or, once identified, not widely available or liquid or easy to access. The more enlightened posters on "both sides of the debate" will do just that, and probably not identify themselves as even "taking a side".

FWIW, it's certainly good to see a tracker fan explain that he continues to research, understands the merits of allocating using factor analysis etc, and recognises that different people have different needs and might not be able to use the same approach for every sector. Rather than just religiously parroting that based on the theoretical research in the 1970s, everyone should just get a cap weighted equity tracker and a total bond market tracker and be done with it.

TheTracker

(trimmed to read)

bowlhead99 wrote: »

If i were looking for an opportunity to be critical of the classic market-cap-weighted passive funds, this sort of research would be gold.

Effectively, another way of looking at that paragraph is to say that every time a dumb regional or world tracker is beaten by an active fund that *doesn't* simply allocate capital to every single equity opportunity based on size and skew your portfolio to a small number of large companies in certain sectors... then there will always be a reason for it.

"In other news, I compared the sports performance of Usain Bolt and Michael Phelps with the average UK resident. Once I controlled for genetic factors, and for gender, training, protein intake, lifestyle and determination factors, they were not much better at running or swimming than anyone else, so I don't know why you'd pay money for someone to give you a tip that you should put them in your team."

What the research shows, and much of it in the last 20 years, is that many successful fund managers of the last century were successful because they chose a style and stuck with it before everyone fully realised that style would reap dividends. Plenty of others who chose other styles then fell away in disappointment. Its a good thing we have the warren buffets of the world, because their real world experiments helped theory.

The models that have been subsequently developed are able to explain nearly all of those variations in return in terms of a number of factors. Different models give 3, 4, and 5 terms. No doubt future models will refine the factors again. Recently more and more instruments to expose yourself to these factors have arisen and become cheap enough they offer benefit beyond their cost of employment. Some people think these historical factors are of little utility moving forward, but that's not a majority belief.

So I see active fund management in 2015 as a bit of a mug's game in many ways. The field has been farmed so deep that we know what works and what doesn't. I simply don't believe in star fund managers with superhero intuition. Although I do believe there are enough funds out there that through random chance one will happen upon a strategy that exposes a new factor, though good luck spotting that fund in advance and distinguishing it from chance.

So once you've allocated a portfolio to studied sectors, the question remaining is how does one pick an active fund that would outperform an index/etf that tracks that sector? Studies have shown, for instance, that most funds that beat a value index do so because they are "more valuey", after accounting for randomness. Same for the Small cap sector because they chase funds that are "more smally". We read about "closet index trackers" in the news with distaste, but really nearly all successful funds are "closet factor trackers".

That's what Swedroe's book is about, the "incredible shrinking alpha" where "alpha" is a fund manager's skill is progressively explained away through the "beta" of factor exposure.

In some areas there are currently no or few indexes or cheap passive funds to track. For instance, UK Value or UK Small Cap. Then we get into false benchmark territory when we look at the active funds in those spaces versus the poor proxies for passively tracking them. So some people opt for an actively managed fund because the cost of providing the fund may outweigh the exposure focus through the poor tracker. We make personal choices/stabs here: I wouldn't pay 0.78% for a small cap tracker, but I might pay 0.22% for an active value tracker.

Future gazing also brings up the spectre/hope of robo-portfolios and even the demise of index trackers. A passively managed index tracking fund is still a middleman, albeit with a very low cost. Some say there will be a day where trading costs are so low that one can directly buy the logical index constituents cheaper than the cost buying into an index fund.

By the way, Swedroe talks about sports in his book. Usain Bolt and Michael Phelps are great examples. He talks about how we still see outliers in individual sports like Tennis (Federer) and Swimming, but far few in team sports like Baseball. It's a numbers game of who is competing against how many. Active managers aren't competing against passive indexes, they are competing against each other. And like in Baseball, the best players are distinguishing themselves less and less. And its getting more and more possible to measure a baseball team's chance of success by its characteristics rather than with a "star coach"(baseball statistics is enormous). Passive vehicles will continue to just subsume their increasingly better performance.

Sorry if that sounds unclear, maybe Sam_J12 can explain it more clearly, since Sam has read extensively and knows more about it.

Finally and back to the thread, I don't think the point has been made well enough about assets under management. Virgin may have a 1% fee versus Vanguards 0.1% to track a similar, but Virgin may actually be spending less money on administration. It's all about how much money is under management. Fees will get lower and lower and that goes for active management too.