Active vs. Passive. When and Where?

Linton

masonic wrote: »

Unless I am misunderstanding what you are doing with your "high risk/high return portfolio", then it seems you are selling capital when appropriate and using profits from the growth of that portfolio to supplement your income stream (though it may be indirectly, by replenishing your cash reserves or topping up your other portfolios). I can completely understand cascading capital through one or more buffers that you can fall back on during times when you would otherwise be forced to sell capital at an unfavourable price.

Basically yes. Working in the way I do means that if conditions are inappropriate for selling volatile capital I dont need to for some time and also that I only need to think about what capital to sell once per year.

grey_gym_sock

bowlhead99 wrote: »

The average of all investors will give the result of the cap-weighted indexes.

Some of the 'active investors' are private investors who allocate their portfolios without rational thought as to fundamentals and overall allocation (e.g. holding employee sharesave shares, bonus windfall shares, shares they heard about from their mate down the pub, shares they inherited, shares they got in a Royal Mail or Direct Line or TSB IPO because it seemed like a good idea at the time even though they don't think of themselves as an 'investor').

yes, it is a bit jarring to label buy-and-forget investors as 'active' (i was effectively labelling everybody who isn't cap-weighted as 'active'). but do such investors actually do badly, on average? they probably have very unbalanced portfolios, so their risk-adjusted returns may be poor, but what of their raw returns? i imagine that they might have done OK, partly given that a lot of the privatizations were (at least with hindsight) bargains. and buy-and-hold is not that bad a strategy; a lot of investing mistakes involve excessive activity. (1 study found that the broker accounts which performed best were the 1s where the owner had forgotten they had the account )

the very active and very amateur private investors, who buy following every new tip and fad, would, i agree, be more likely to underperform.

in any case, overall, i can easily believe that the average professional active investor beats the average amateur active investor. the problem is that there are not enough private investors for the professionals to gain from. making up some figures: suppose that there is 10 times as much professional as amateur actively invested money; and that the amateur money trails the index by 2% a year; then the professional money will beat the index by 0.2% a year - not enough to overcome the higher costs of active management. (of course, you can make up different figures, and get a different answer .) i think some studies have indeed found that average active funds beat their index before costs, but not by enough to overcome the hurdle of their costs.

Some of the 'active investors' do not have free choice in what to invest in, because they are attempting to deploy large amounts of money and really do have to weight their portfolio towards larger or more liquid stocks. For example the institutions as a whole really do need to put 20x as much into HSBC as they do in Taylor Wimpey in the FTSE 100 indexes, or 20x as much in Taylor Wimpey as they do in 888 Holdings in the MSCI smallcap indexes. Because they can't allocate their portfolio the opposite way around even if they felt 888 was a better opportunity than HSBC.

So they end up with a lot of HSBC. It doesn't mean that having researched the opportunities of growth and profitability prospects and regulatory hurdles etc, you should also buy a lot of HSBC, unless you are aiming for the return of what the large pension fund or insurance company or government sovereign wealth fund is going to achieve from 'the market'. If you are managing a nimble £50m investment trust you can get away with investing each of those millions in pretty much whatever the heck you like.

well, a huge fund can't go equal-weighted on every share in the FTSE all-share index. but there is nothing to prevent them from overweighting (compared to its capitalization weight) 888, and underweighting HSBC. perhaps they won't bother if it's a mostly-big-cap fund; but they are probably also running small-cap funds, with less money, but significant amounts for small cap. my point is that the well-researched opinions of professional managers on the relative values of HSBC and 888 do get to have an influence on the relative market values of the companies.

there are possible reasons to want something other than cap-weighted, but i don't think anything else is obviously superior. for instance, perhaps holding more small-cap (and equal-weighted is 1 way to do that) will give higher returns after several decades, but it also seems to be more risky, and can underperform for a decade or more.

certainly, if you know that the manger is outstandingly skilled, you'd prefer them to be managing less money, so they can go heavily in small cap opportunities. i think there probably are genuinely skilled managers. but the trouble is that you don't usually know until it's too late. so yes, i'd rather put my money in a small fund run by warren buffett than in berkshire hathaway, but i'm a bit late for that ... (and sometimes, you might know the manager is skilled, but their fees are high enough to wipe out their outperformance, or higher still.)

Some of those 'active investors' are actually closet trackers, who are following an approach quite close to the index by weighting their portfolio similar to the index so they don't get criticised for losing to the index too much from one period to the next, with the result that their half a percent of fees is dead money so their performance is poor.

closet trackers don't affect the arithmetical argument. because closet trackers will (before costs) approximately match the index return. and therefore all active investors, excluding closet trackers, will still (very nearly) match the index returns before costs.

So, while you can easily see that mathematically, all investors will on average end up with the same as 'the index' and therefore it sounds foolish for them to pay active management fees to achieve that end - many of us don't aspire to end up with the same result as the index. Some people have specific needs (e.g. income, low volatility etc) so won't follow the index. But others won't follow the index because they are simply looking for more outright performance than the average of what the big institutions, closet trackers and idiot private investors are destined to achieve.

for specific needs, maybe.

for pure performance: ruling out closet trackers is necessary but doesn't get you ahead of the index. i agree there is probably something to be gained at the expense of idiot private investors, but i think it isn't much, because PIs don't manage enough money. i'm not convinced that a (small) active fund has any advantage over big institutional investors.

if you want to try for outperformance anyway, i've no objection. there's nothing wrong with trying to achieve an unlikely objective. or even (IMHO) with having a bit of a flutter. i do think you should be aware of how much money you're gambling with (e.g. that if you are spending 0.5% extra per year by using active management, and if it doesn't pay off at all, then after 30 years, you'll have 14% less than if you'd used passive investments).

If you can avoid the actively managed funds that turn out to be closet trackers or dogs - which is easier said than done but is not an impossible task - you can beat the indexes by enough to pay the fees. Quite what level of overall fees (annual fee, annual operating expenses, capitalized trading costs, performance fee etc) can be afforded and still beat the index net of fees without taking too much risk, is unknown, but there are plenty of managers who regularly outperform trackers in some sectors.

So my contention remains, that where the markets are imperfect, a good manager can reliably beat the index. Clearly that's harder to do in developed markets which are large and liquid. So large and developed markets would be the best place to use indexes - assuming you don't have a goal that precludes their use, e.g. higher income, lower volatility etc. Which comes back to smallcaps and emerging markets (and especially smallcap emerging markets, if your portfolio is broad enough to entertain them) as being the relatively better places to be an active investor or pay an active manager.

i still fail to see why small caps, or emerging markets, or generally "imperfect" markets, are different. unless that means markets in which most money is invested by (idiot) private investors, at whose expense the professionals can profit.

on the assumption that all active managers are just doing stuff at random (tossing coins to decide which companies to over- and under-weight), nearly half will beat the a tracker over 1 year, a smaller over 2 years (as their higher costs bite a little deeper), and an ever-diminishing number over longer periods. in fact, that is pretty much what is observed. this suggest that a very large majority of active fund managers are unskilled (this is a comment on the outcome, not the process, which doesn't look at all like "doing stuff at random"). i do believe some managers are skilled, but it is almost impossible to distinguish them from the "lucky !!!!!!".

grey_gym_sock

Linton wrote: »

According to the Vanguard website Vanguard's All World ETF has a yield of 3.13%. Roughly the same as the FTSE 100 or All Share. Hardly a must buy for an income investor.

well, in world terms, the FTSE all-share is high-yield ... i think 3.13% is quite good, for a more diversified index than the FTSE all-share.

Perhaps you can give an example of a diversified passive income portfolio.

i'd be inclined to combine that all-world ETF with a higher-yielding UK tracker. perhaps vanguard UK equity income index fund, which pays about 3.94%.

and with an investment-grade corporate bond tracker, such as ishares £ corporate bond ex-financials ETF (ISXF), which currently pays out 3.78% (though the YTM is only 3.17%, which implies that part of that payout will, other things being equal, be a trade-off for slowly reducing capital value).

those 3 holdings (without specifying what proportion to put into each) could give you a natural yield of 3.5% or so.

you can try for more than 3.5%, but would it be sustainable? i doubt if we know.

if i wanted to spend (say) 5% of my capital per year, i might invest about 80% of it in a portfolio generating 3.5% or so, and keep the other 20% in cash (or near-cash) and plan to spend the natural income + 2% of that 20% cash per year. so after 10 years, the cash would be gone and i'd need to sell investments to replenish the cash. (though i'd probably actually sell investments sooner. perhaps consider whether to sell any investments once a year. in an average year, sell a 2% slice, to replace what i've spent. but if investments have fallen, don't sell anything. and if they've shot up, sell more than 2%.)

having said all that, i do see the attractions of investment trusts for income. some pay more than 3.5%; and while some of those are perhaps overpaying and will struggle to increase the dividend, others may be fine.

jamesd

BLB53 wrote: »

You therefore have maybe a 5% or at best 10% chance of selecting a long term winner from the thousands on offer or a 100% of matching the index with a low cost tracker.

That is not true.

There is a 100% chance of failing to match the index with a tracker, unless it is using active management techniques like stock lending. Of course you may not get a low cot tracker if you follow the same random tracker selection as you are describing for active funds. You might instead get a tracker that charges 1.5% for the FTSE All Share Index, rather than one charging 0.1%.

You only have that 5% or 10% chance with the active funds if you use random selection. In practice you can eliminate the consistent poor performers and greatly improve your chance of success. Just as you can when picking a tracker to avoid the consistent dogs.

bowlhead99

grey_gym_sock wrote: »

i still fail to see why small caps, or emerging markets, or generally "imperfect" markets, are different. unless that means markets in which most money is invested by (idiot) private investors, at whose expense the professionals can profit.

on the assumption that all active managers are just doing stuff at random (tossing coins to decide which companies to over- and under-weight), nearly half will beat the a tracker over 1 year, a smaller over 2 years (as their higher costs bite a little deeper), and an ever-diminishing number over longer periods. in fact, that is pretty much what is observed. this suggest that a very large majority of active fund managers are unskilled (this is a comment on the outcome, not the process, which doesn't look at all like "doing stuff at random"). i do believe some managers are skilled, but it is almost impossible to distinguish them from the "lucky !!!!!!".

The efficient market hypothesis assumes that the market will always cause share prices (and bond prices for that matter) to instantly incorporate all relevant information. In some markets, or in some parts of markets, information is of a lower quality and is not instantly incorporated by the market. The mispricing of stocks at something other than their fair value provides opportunities for managers to evaluate information and make decisions.

In order for a manager to make a portfolio allocation decision which 'beats the market', selecting one stock over the other, someone else has to choose a different allocation which fares worse. There will be plenty of examples of people who want to take the other side of the trade, for the good fund manager to choose from.

For example if the fund manager seeks to hold 1 pound of Enterprise Inns for every pound of Royal Dutch Shell that he holds, instead of 1 for 185 like a Europe Tracker would hold, he needs to find someone else to hold his 'spare' 184 units of RDS.

There will be plenty of idiot private investors who are happy with holding Shell and not Enterprise. Maybe some of them work for Shell and can get a good deal on buying the shares and so are unconcerned with the fundamentals of any other opportunities. Maybe some heard about Shell in the papers and hadn't heard of Enterprise.

There are plenty of active fund managers who will hoover up the extra Shell shares because their investment remit is to only invest in companies with over $1bn market cap, or $5bn or $10bn or $20bn or whatever, so are happy with holding zero Enterprise shares, making their holding ratio 1:infinity rather than 1:185.

Likewise passive largecap fund managers who only take FTSE100 shares and not FTSE250. Or income funds which won't touch a company that doesn't pay a dividend so don't want any Enterprise, or energy funds which don't want anything in the consumer sector. And plenty of institutions (pension funds, insurance companies, sovereign wealth funds etc) who would be happy with just holding Shell and not the other company because of size / liquidity problems or because taking the 1 unit of the small company doesn't move the needle on their overall returns profile anyway.

So, it is quite easy for an active fund manager to select a winning trade (e.g. sell Shell buy Enterprise) without the other side of the trade being another active manager with the exact same remit. The other side of the trade could be another active manager with a quite different remit (such as investing in energy or in dividend payers) which exists as a function of what certain underlying investors would like to invest in. Or it could be a private investor who has not done their research, or an institution who is unable or can't be bothered to hold the smaller company.

The overall average blend of how the capital is allocated by 'the market' might be 1:185 which is a function of all the people playing in the market and reinforced by the trackers and closet trackers buying in those market proportions. But it could be that all the active unfettered managers are holding 1:15 and the big institutions are holding 0:170, and the active managers are all 'beating the market' no matter which one you pick.

Of course, if the shares in the two companies are both fairly priced in the first place, the active managers who are actively avoiding Shell or actively buying Enterprise will not necessarily be able to, after fees, beat 'the market', because both companies would have the same returns (allowing for risk - although perhaps the risk of favouring the smaller and less resilient company will be rewarded over time with greater returns, but that's not the thrust of the argument).

So for active management to routinely prevail, the market should be one which is not efficient: where companies do not behave as expected and where information is not instantly reaching all corners of the earth at the same pace automatically and only some people reach the conclusion that actually opportunity A is better for its price than opportunity B. Where people need to do research to uncover information to truly understand the fundamentals of risk and valuation of the company, they can have more of an advantage when coming up with their allocation strategies. Where they feel they have the pertinent information and others do not, they can prevail in the market.

From the example above, perhaps all the truly unfettered managers could prevail against the restricted active managers, institutions and private individuals, but what is more likely is that some will prevail and some will fail. The ones that fail often, or fail enough to not beat their fees will fall out of favour and may close, but even if they don't close they will have a track record that lags the market or at least lags the top active managers.

So, if you look at markets with less perfect information, you can see funds like Aberdeen Emerging Markets or investment trusts like Templeton Emerging Markets, outpace the average fund over long periods of time. In short periods anything can happen but there are plenty of EM specialists or smallcap specialists which have deserved good reputations because over time their performances have consistently exceeded what 'average' active funds or passive products have achieved.

If the efficient markets hypothesis stood true in all markets there would be no need for active management other than to cater for different risk profiles or return characteristics. In those markets where it doesn't stand true, then clearly there must be a greater case for paying active management fees for generalist investment from a performance perspective too, because of the greater opportunity to beat the market in a resounding way.

One could argue that you only get inefficient markets in emerging economies or in thinly traded smaller companies. However, the existence of periodic crashes (e.g. double digit losses in a day in 1987) means that the 'market' is likely not always displaying the 'fair' price of all companies at all times. And an investor like Warren Buffet has built a reputation over half a century of persistent outperformance, in mainstream sectors, which EMH would imply was impossible, though maybe it's just a run of luck like getting heads 45 times out of 50.

BLB53

That is not true.

I have no axe to grind either way but all the evidence I have seen points to the fact that, over the longer periods, the better returns will be provided by the low cost index trackers ( I would not class one that was charging 1.5% as low cost btw).

Here are a couple of links - one from monevator referring to research by Rick Ferri
http://monevator.com/weekend-reading-another-reminder-that-most-active-managers-fail/
and the other from diy investor which refers to a report from Vanguard
http://www.diyinvestoruk.blogspot.co.uk/2015/05/a-logical-investment-strategy.html
If most actively managed funds fail to consistently beat their chosen benchmark index over time, it seems logical to my mind to settle for a strategy of matching the index which is what the low cost trackers offer us.

Thats not to say, some managed funds and ITs cannot do a good job - it is clearly the case that some can and do consistently outperform their benchmark index and the savvy investor can probably identify these managers but, for the majority of ordinary investors, imho, they would be better served by the low cost, globally diversified index tracker.

Mistermeaner

Why not just hedge and do 50% passive global tracking and 50% globally managed

Or is that then an active portfolio

I'm confused

Linton

grey_gym_sock wrote: »

well, in world terms, the FTSE all-share is high-yield ... i think 3.13% is quite good, for a more diversified index than the FTSE all-share.

i'd be inclined to combine that all-world ETF with a higher-yielding UK tracker. perhaps vanguard UK equity income index fund, which pays about 3.94%.

and with an investment-grade corporate bond tracker, such as ishares £ corporate bond ex-financials ETF (ISXF), which currently pays out 3.78% (though the YTM is only 3.17%, which implies that part of that payout will, other things being equal, be a trade-off for slowly reducing capital value).

those 3 holdings (without specifying what proportion to put into each) could give you a natural yield of 3.5% or so.

you can try for more than 3.5%, but would it be sustainable? i doubt if we know.

....

I am taking a natural income of about 4.5% from my income portfolio. One problem with the Vanguard ETF is that it has a high % of US where dividends are less favoured because of taxation and a low % of Far East where local culture likes dividends. Also small company income is largely ignored. This is a pity as there are useful dividend payers there.

One of the advantages of using a tracker is that it represents the market. However the UK Dividend "tracker" uses a highly artificial bespoke index specially constructed by FTSE that prevents excess focus on particular sectors and companies that apparently pay very high dividends because their share price has fallen significantly. It is difficult to see what market, if any, this represents.

jamesd

BLB53 wrote: »

I have no axe to grind either way but all the evidence I have seen points to the fact that, over the longer periods, the better returns will be provided by the low cost index trackers ( I would not class one that was charging 1.5% as low cost btw).

Your assertion were:

1. "100% of matching the index with a low cost tracker"
2. 5% or at best 10% chance of selecting a long term winner" [from active funds].

The assertion is false because:

a. trackers that have any charges or costs naturally underperform the index due to the charges.
b. a tracker can use active management in the form of stock lending to try to overcome this but I don't know of any tracker that has successfully eliminated all of the effect of its charges and costs to allow it to actually match the index performance.
c. you asserted random selection for active managed funds but restricted the choices to only the low cost trackers for passive, thereby distorting your comparison between active and passive.

If you'd care to disagree kindly randomly select five trackers in different sectors and see whether any of them have met your claimed performance of matching their index. Then try to meet your claim of 100% of them doing it. To make it easier you might start by looking at the best tracker you know of and see whether it matches its index or not. Or one that you hold because you think its OK. If it doesn't then you might as well stop there and accept that it is not 100% sure that picking passives will match the index.

Note random: you're using random for active so to be fair you need to be doing the same for passive, so you have the option of buying the dogs of the passive world, just as you're suggesting those using actives must.

Of course I bet that you don't actually use passive selection of trackers: you undoubtedly actively select the best of the trackers. Just as someone using active managed funds should.

You also presumably knowingly loaded the odds against active managed funds by requiring a long term winner, presumably indicating that you would completely ignore known critical factors affecting active managed fund performance, like the human management team. As distinct from the manager - meaning management house - used in studies comparing active and passive when they use the word manager.

BLB53 wrote: »

Here are a couple of links - one from monevator referring to research by Rick Ferri

You might also mention that Monevator actually said "passive specialist Rick Ferri" who even founded a firm that makes money from selling passives. He's not neutral, he has promotion of passive funds as an objective. And surprise surprise, guess who produced the two "studies" he cited? Vanguard, best known as a passive fund vendor. I put "studies in quotes because both are really sales documents: The case for index-fund investing and SPIVA® U.S. Scorecard. Both of those papers exhibit the typical failings of actual academic studies that systematically bias them against active managed funds:

A. They hold regardless of manager changes even though a manager change is known to be a predictor of poor future performance.
B. They used all active managed funds of the types, ignoring rational selection that humans perform on both active and passive funds.
C. They held funds even though it was a time when they could be expected to do poorly, something those using active strategies would be expected not do do.

BLB53 wrote: »

If most actively managed funds fail to consistently beat their chosen benchmark index over time, it seems logical to my mind to settle for a strategy of matching the index which is what the low cost trackers offer us.

It is irrelevant to those using active funds what most do. What's relevant is what the ones they select do and those with a free choice will not include within their list of funds to select from the consistent dogs. No more than passive investors would include passives charging 1.5% with high tracking error on top when they could instead choose 0.1% with low tracking error.

BLB53 wrote: »

Thats not to say, some managed funds and ITs cannot do a good job - it is clearly the case that some can and do consistently outperform their benchmark index and the savvy investor can probably identify these managers

Some can do, indeed. And it's not that hard to greatly improve performance: ignore the dogs and sell when the manager changes.

BLB53 wrote: »

for the majority of ordinary investors, imho, they would be better served by the low cost, globally diversified index tracker.

But actually there we agree. Most investors just want to put money in and forget about it, so active managed funds are not a good match for their inclinations. You rally do need to be paying at least enough attention to read the manager changed email notifications, for example.

Personally I use both active and passive funds.

jamesd

ggb1979 wrote: »

Why not just hedge and do 50% passive global tracking and 50% globally managed ... Or is that then an active portfolio ... I'm confused

It's active of course. But so are passive funds. They select which index to track and when to change holdings and you find things like different MSCI and FTSE emerging markets country lists and weightings as a result. Then they can change index and holdings as a result, as when Vanguard switched from MSCI to FTSE indexes, presumably because FTSE offered them lower licensing fees. So picking a passive is also about picking which indexes and if you're doing it carefully you might want to be sure that you don't completely miss say Korea by selecting a developed market tracker that doesn't include it because it's emerging markets (MSCI) and an emerging markets tracker that doesn't include it because it's developed (FTSE).

Then you might decide whether you want to avoid active management techniques like share lending, where the active manager has to decide who and which collateral to trust when doing the lending, or contract out that work to someone else to do it for them.

Then you might want to try to decide whether implementing fixed rules about buying and selling instead of following an index is active or passive. The smart beta movement in "passives", say.