Monevator reader Anna sent us a common question about investing in particular sectors of the stock market:
How should one behave if I would like to be overweight in a specific sector or theme, like clean energy or AI for instance, or perhaps being overweight in a specific country like Russia? Would it be possible to implement these themes in the global portfolio?
Our world is changing rapidly. The ground rules of our lives constantly shift beneath our feet as society is reshaped by geopolitical, socioeconomic, technological, and environmental transformation.
Surely we need to reposition our portfolios, then? Tweak them like wind turbines, pivoting to harness these powerful forces?
There’s an exchange-traded fund for that
The ETF industry certainly wants to help you feel like you’re in control.
For example a dozen climate change ETFs launched in Europe in 2020 alone, according to the ETF data service justETF.
Other ETFs enable you to invest in everything from cannabis to robotics, through esports, gender equality, genomics, and immunology.
Like fashion houses, fund providers know their job is to stay on trend.
Those options come on top of existing ETFs targeting sectors and industries, smart beta strategies, countries, and regions.
And that is but a cherry balanced on a giant fruitcake of investments of every candy-stripe – slicing and dicing the global economy into a pick ‘n’ mix cascade.
All you have to do is predict which themes, megatrends, rising powers, and techno-tipping points will profitably rewrite the future.
(Ahem. “All”…)
Spread your bets
There was a similar explosion of ETFs in 2018 that enabled us to invest in exciting themes such as:
- Blockchain
- Cybersecurity
- Artificial intelligence and robotics
- Bio-tech
- Battery tech
I picked a selection at random to see how they did.
It turned out that all these ETFs beat a plain old MSCI World ETF in the short-term1:
Source: justETF [Click to enlarge.]
- The MSCI World ETF benchmark (orange line) falls flat into last place with a 28% cumulative return over two years.
- The battery ETF (grey line) went from last place to third in just six months, with a 83% cumulative return. Hmm, volatile!
- The blockchain ETF beat the World ETF by less than 2% despite Bitcoin going berserk.
- Try tearing your eyes away from the towering 121% return of the artificial intelligence ETF in the no.1 spot.
First AI beat me at chess, next it’s having my job, and now it’s a better investor, too? Humans are yesterday’s news, baby.
Maybe, but it’s easy to get overexcited about new toys.
Let’s take a longer term view
Here’s some megatrends that I could get medium range data for:2
- Big tech
- Rise of China
- Global water resources
- Clean energy
- Healthcare in the face of aging demographics
- Global infrastructure fueled by urbanisation
Source: justETF [Click to enlarge.]
This time our MSCI World ‘accept you have no edge’ benchmark ETF (orange line) sits comfortably mid-table.
Three of the selected big trend ETFs beat the world market. Three under-performed.
Here are the scores (returns are cumulative):
- Big tech – 931% – Win
- Healthcare – 268% – Win
- Global water – 267% – Win
- World tracker – 213% – Acceptable
- Global infrastructure – 114% – Fail
- China – 101% – Fail
- Clean energy – 14% – Fail
Are you surprised? Could you have predicted these outcomes in advance?
It’s not that the narrative behind the underperforming themes was wrong.
- The world has continued to urbanise since 2007.
- China has continued to rise.
- Clean energy is a growing part of the world economy.
What then?
- Was the importance of the theme overblown?
- Were the companies overpriced?
- Did the lion’s share of returns fail to accrue to shareholders – diverted instead to management, employees, customers, future investment, or private equity?
- Was the industry hit by an ill-wind? Loss of subsidies, social backlash, constrained by regulation or the technology failing to fulfill its potential?
- Was there something wrong with the index? Perhaps it wasn’t truly representative of the industry or country or was too concentrated in less competitive companies? Or did it include firms that jumped aboard a sexy new trend train but were mere passengers, not the real engines of growth?
Identifying a trend seems to be easier than profiting from it.
A much longer term view
To understand how the roulette wheel of progress affects market performance, I recommend reading Industries: Their Rise & Fall by the finance academics Dimson, Staunton, and Marsh.
Here’s their breakdown by industry sector of the US stock market in 1900 as compared to 2015:
Source: Credit Suisse Global Investment Returns Yearbook 2015
You can see that the market has been radically remixed over a century.
According to the professors:
Of the US firms listed in 1900, more than 80% of their value was in industries that are today small or extinct.
For instance it’s hard not to notice the huge blue wedge devoted to the rail industry in 1900. That snake-jawed Pacman was worth 63% of the market at the dawn of the Twentieth Century.
But by 2015 rail made up less than 1% of the stock market.
Meanwhile 62% of the 2015 US stock market’s value lay in industries that were negligible or nonexistent in 1900.
Back on this side of the pond, 65% of the 1900 UK market value resided in industries that were marginal or had disappeared by 2015, while 47% lay in industries that were small or not yet invented in 1900. (More evidence that US capitalism has been more dynamic than its UK counterpart?)
Here’s how those 1900-era industries fared against the market during the upheaval of the subsequent 115 years:
Source: Credit Suisse Global Investment Returns Yearbook 2015
The wider market is in dark red and is beaten by five industries.
Notice the market is beaten by rail (light green) despite its huge slice of the pie being reduced to crumbs by 2015.
But the runaway winner is tobacco.
What’s the story here?
Tobacco – addictive, cool, seductive as a ‘50s film star – so of course it won?
Given the number of kids in my school who couldn’t wait to drag on a fag, that was a plausible narrative up until the 1990s.
But few industries have taken such a cultural and regulatory battering in the developed world as tobacco since then.
The equivalent now would be placing all your chips on coal as the fuel of the future. Would you make that bet?
The Investor will pop a long in a moment to explain that pumping out cigarettes required little investment in innovation over the subsequent 100-plus years, and so it was wildly cash generative. Reinvesting that cash drove up returns. [Confession: I just popped along and did – @TI]
But multiplying the difficulty of picking a winning sector by the uncertain strength of its business models does not lead to odds I’m willing to take.
Here’s another counter-intuitive finding. The US rail industry was laid low by the rise of road and air, but rail is the only one of those industries that had beaten the market by 2015:
Source: Credit Suisse Global Investment Returns Yearbook 2015
Rail had a torrid time for 80 years, especially over the 1960s to 1980s. But it emerged fit as a super-productive flea to overtake the market in 2013.
The road industry never beat the market. Air had a few golden ages but always returned to earth with a bump.
Think of how unbelievably exciting the air industry must have looked after the role it played in World War 2. You can see the value of aviation equities spike in the mid-1940s in the graph above.
Would you have predicted that the air industry’s promise would not be fulfilled for shareholders, as opposed to the rest of society?
Could you credit that you’d have been better off just staying invested in the broad market?
The cycle of rise and fall
The merry-go-round of modern capitalism hastens our desire to predict new dawns and to declare existing hierarchies dead.
Survival in the knowledge economy makes us desperate to stay ahead of the curve, after all.
Dimson, Staunton, and Marsh highlight market historian Jeremy Siegel’s explanation of this anxiety from an investing perspective:
Investors have a propensity to overpay for the ‘new’ while ignoring the ‘old’ …
Growth is so avidly sought after that it lures investors into overpriced stocks in fast-changing and competitive industries, where the few big winners cannot compensate for the myriad of losers.
The academics observe how technology cycles through:
- Scepticism
- Over-enthusiasm, which can lead to a bubble
- Sober reassessment once everyone cools off
Could the explanation for the outperformance of AI, robotics, and battery tech in our first example be that investors are caught up in the over-enthusiasm phase?
Dimson, Staunton, and Marsh advise:
Investors should shun neither new nor old industries.
There can be times when stock prices in new industries reflect over-enthusiasm about growth, and times when investors become too pessimistic about declining industries.
However, it is dangerous to assume that investors persistently make errors in the same direction: they may at times underestimate the value of new technologies and overestimate the survival prospects of moribund industries.
It’s clear the market can be beaten. If you’d backed tobacco equities in 1900 then you’d have smashed the market over the long-term.
But can you beat the market?
Remember, as a passive investor you don’t have to get involved in making these explicit bets.
You already gain exposure to AI, robotics, cybersecurity, blockchain, China, and every other plausible investment narrative through a broadly diversified global tracker.
By over-weighting any theme, strategy, or country you’re claiming greater insight than the aggregate of all investing decisions that make up the market.
A market that’s dominated by huge financial players rather than greater fools. A market even the world’s best investors struggle to consistently beat.
Your prospects rest on this key insight from our academics:
It all depends on whether stock prices correctly embed expectations.
Expectations amount to the market’s best guess, given current information.
The smart money has better information than you or I. It reacts to new developments with frightening speed.
- Did you see a global pandemic coming in 2020?
- Did you see shares rebounding right after the fastest crash in history?
- Can you predict the winners and losers as Covid-19 reorders society?
Or, to highlight another sweeping change, will big tech be more or less profitable by the time the antitrust lawsuits have played out?
As ex-hedge fund manager Lars Kroijer wrote about overweighting:
What is it that you know that the wider market doesn’t?
The best strategy is to be humble
Plan A is to stick with a global tracker fund.
Can’t resist a wee flutter?
Plan B is to limit the damage of being wrong.
Just like you wouldn’t bet your house on the outcome of the Grand National, keep your stock market bets survivable.
Overweight 10% of your equity allocation, tops.
Enough to enjoy your win should you back the right horse. But not so much that the pain will unbearable if you invest in the next air industry or Argentina.
Take it steady,
The Accumulator
- The time frame chosen was simply the maximum span over which I could compare these ETFs. It was no more scientific than that, and I don’t think it tells us anything except that the short-term is no guide to the quality of investing decisions. [↩]
- The time frame is the longest period I can compare the ETF selection against each other. [↩]
Comments on this entry are closed.
That comparison of the rail industry returns since 1900, a century in which (As you say) it was humbled as an industry, is one of the best factoids I’ve seen for a long time. Completely counter-intuitive, brilliantly illustrates the enormous difficulty of beating the market.
So much to enjoy here, fascinating data and graphs.
In spite of your conclusion (which i was relieved to see, as ever, bringing us neatly back to the MO of monevator) it got me thinking about those tobacco stocks. Presumably big dividends were a huge driver of that return – are they included in the data that makes up the graph?
Makes me wonder when big tech will stop refusing to pay dividends. Could Netflix be the tobacco of the future – a subscription model opium-of-the-people to replace the outlawed vices of old?
Financial advice gold:
“Plan A is to stick with a global tracker fund.
Can’t resist a wee flutter?
Plan B is to limit the damage of being wrong.”
What an excellent article, as per e17jack totally fascinating.
I have shared with my inner circle, who enjoy and follow investing and its kicked off interesting discussions already.
Look forward to the discussion sparked here.
Okay i have a question:
Chart says:
1900-2015, rail up x62,000, whole market up x38,000.
Which it think takes rail’s share from 63% up a bit higher, to about 73%. Not down to 1%.
Assuming the chart is showing total returns, ie with dividends reinvested.
Or is there something big that i am missing?
Or… is there a bug in someone’s spreadsheet? Hope not…
Timely article… all I hear these days is “INRG”.
Excellent piece, and the railroad example really drove the point home!
How about adjusting for bigger sectors/regions? For example, I believe that as European investors and for tax reasons (higher tax on dividends if I’ve understood right) we should be slightly underweight on the US. For an opposite example, personally I’m a little bit overweight on the Developing Markets where I believe that both risk and expected return are higher (and have the conviction that they are underrepresented in the World index). I also have some home bias because that’s where my current and future consumption happens, so some overweight on Europe.
I agree it may be hard to beat the market by choosing equity sectors but I disagree with the corollary that implies a market-cap weighted tracker is optimal. I think the difference stems from the fact how you define ‘passive’. I define passive as a benchmark portfolio of assets whose performance should best meet my long-term target objectives/risks/liabilities, without requiring me to formulate views on the assets’ short-term performance.
So I see a market-cap weighted tracker as risky since it’s not necessarily a good replicating hedge. I’m comfortable to tilt allocations if that tilt is risk-reducing or better meets my long-term liabilities. I don’t see that as active investing, more prudent risk management. Basically, I tilt overweight scenarios that “f**k me” and tilt underweight those that benefit me!
How accurate is the classification of companies by sectors? Are Tesla and Netflix really tech companies? If Disney or Discovery take back a big share of streaming market, would you add them to tech sector? If VW does well with their electric cars, does that mean Tesla drags tech performance down, while VW pushes auto up?
And what about conglomerates? GE, Hutchison, Swire pacific, Philips? And what about companies who have changed their business models? IBM, HP, Disney, Sony etc.
All these add to the reasons why sector investing has to be avoided long term.
@David R
I guess the answer is that many more companies have listed on the stock market over 115 years. Their listings wouldn’t increase the return of “The Market” but do dilute the share of Rail.
Absolutely superb article. So many people don’t differentiate between companies with seemingly superb prospects and good investment opportunities!
@ZXSpectrum48k Care to provide examples as to what you are currently overweight and underweight? And how large are your tilts relative to global market cap weighted index?
For me those graphs aren’t click to enlarge – very vanilla Edge on Windows 10 – am I doing something wrong?
This did make me think. Why is a equity cap tracker better than say investing the same % in every sector? In the first, if Tesla takes off I buy more putting me in a buy high situation. In the second case if Tesla takes off I sell to bring the sectors back into balance putting me in a sell high situation. Has this ever been looked at to see which performs better?
Great thought Provoking article.
@Richard – From memory there is (or was, it could be out of date now) research showing equal weight indices have historically beaten market cap weighted indices, which goes someway towards your suggestion.
However I believe the explanation was the small cap effect — equal weight means a greater allocation to small cap shares which have (had) historically outperformed, away from mega-cap laggards.
As I say that data may have changed now after the last period of large cap growth running rampant.
Lars would say the market knows best; if your proposal was going to lead to higher returns then the market would have sniffed it out and already weighted allocations accordingly.
(I wouldn’t necessarily agree, but I am a naughty active investor for my sins. Which for most people will eventually likely lead mean underperformance).
The main reason for market-weighted funds over equal-weighted is the lower transaction costs. It automatically rebalances as the market moves. I can’t see any logic to invest 10x more in Apple than in Pepsi. The market “thinks” the current price of both represents equal risk-adjusted returns over the future. Therefore investing in either should be equally profitable (from my point of view that I know no better than the market). But I’m not prepared to pay more to put an equal amount in each. Therefore I invest in a market-weighted fund and accept I am more exposed to Apple.
I guess there might be a point that I think this exposure to Apple is a risk that it is worth paying to avoid (without saying I know something the market doesn’t). But personally I think the lowest-cost market-weighted approach is good enough for me and most people.
@ Richard – intriguing point. There are a few equal weighted ETFs out there. I’ve just compared an equal weighted S&P 500 ETF vs its vanilla counterpart over the max 7 years I can get data for – vanilla S&P won.
Then equal weighted STOXX Europe 600 ETF vs vanilla version over 11 years. Equal weighted won!
So I guess you can buy the argument but you’ve gotta be able to withstand the broad market mocking you year in, year out – which is what I’ve experienced with my multi-factor holdings. Goddamn them!
The most interesting thing in this article is that Railways were the FANGs of the ragtime age.
Then … cars.
Everything changes and everything remains the same.
It’s interesting to think about, thanks for the replies. I guess something like the slow and steady portfolio has a similar mechanism but on country weighting. You are not changing the country allocation based on how it’s market cap changes over time, only the shift into bonds right. So if Tesla pushes up your developed world ex UK you would potentially sell it to buy say emerging markets. Something a world tracker wouldn’t do, it would buy more Tesla. I guess the main difference is your large caps will always remain a major part of the portfolio unlike with the sector approach where you can end up overweighted in small caps.
This is very timely stuff.
I’m trying hard not to be a tinkerer but I’ve been failing in recent times. Only last week, having spent the Christmas period getting on the straight-and-narrow, true, passive path, I started to invest in some other bits and pieces. And then this article pops up. Result? I sold some of those naughty active shares and recycled the proceeds back into some good, ol’ VWRL. 🙂
Thanks, TA! It’s almost religious this. Praise be.
Another way of reading this is that if things are in theory perfectly priced, then being overweight in a sector probably doesn’t matter – the graph shown of long run performance of different industries were broadly similar and better than cash – perfection is the enemy of good, and really as long as you’re diversified and hodl, you’re ok. However passive is the best diversification you’re going to get, but you have to accept that the market is skewed by US home bias, and that although assets might be perfectly priced for their risk level, your risk level might be different to mine, and assets might have a different value to you than to me (although mostly handled by tweaking broader asset allocation)
Tax treatment is an edge, age is an edge, risk tolerance is edge, and humility is an edge
Essentially anyone who owns their own business is an active investor very biased to their own business. Anyone who buys their own house is also active – buying one house rather than the generic market of all assets (although there are efficiency reasons to own your own house like being a lower risk tenant to yourself and greater tax efficiency)
> Are Tesla and Netflix really tech companies?
@BuildTheWall I would say they have been exceptional in ability to leverage technology, even if you don’t consider them to be in the technology sector.
I have spent more than a decade in “definitely technology sector” companies and have pondered the same questions. Is Google is an advertising company? Is Amazon a bazaar?
Fabulous article -full of good stuff!
Absolutely reinforces my Tracker only investment policy in equities and bonds
Without this detailed and clear exposition of active investing and its ups and downs I might relapse!
Crystal Ball put away again for another year or two or till the next article like this
xxd09
I’m trying to define a passive ESG and/or SRI-flavoured portfolio, or else identify a good one to substitute for a Vanguard LifeStrategy 80% Equity fund (serving the same purpose, spreading out risk and pursuing gains over a long time period). This is incredibly difficult to do. This article didn’t dive into SRI or ESG, and I’m not sure it’s right to call them trends, but is there any pointer in the right direction for putting together a tidy little portfolio of (say) 2-5 ESG/SRI-flavoured ETFs, to do what the LifeStrategy 80% Equity does? I can keep it balanced myself…
L&G have some ESG tilted trackers that would replicate Vanguard Life Strategies. They wanted to move my workplace pension from simple world index tracker into ESG tilted fund, which I would have been broadly ok with (although ESG factors are still too subjective for my liking), but in addition this fund was 50% GBP hedged and equal weighted so taking it too far away from a true market cap passive approach for my liking. There are only 4 funds to choose from in my workplace pension but if you are free to choose from L&G’s full range you should find what you need with the in built bond / gilts allocation too.
@Stephen. Generally I attempt to take account of my objectives/risks/liabilities at the asset allocation level, rather than playing around with things I don’t understand like equity sectors (or equities at all …).
Nonetheless, I think there can be reasons to take sector equity tilts. For example, when I started investing, I decided to take an overweight in US tech vs. underweighting US financials. One reason, obviously, was that tech had been badly beaten up by 2002. It was also clear, however, that tech was in many ways the “competition” to my job in finance. I’d made the decision to stay in finance rather than move into tech in 2001 (short term good decision, now looking dumbass). So I had an opportunity cost argument. Tech could undermine the finance sector or even make my job obsolete. I naturally got long of bank equity every year through my job. It was just a good hedge.
Even now, tech is really my competition. My net worth increased over 50% last year, but the guys I know at Amazon, well there net worth increased by just as much given the 60% rise in AMZN. So I lost in 2020 and I’m competing with that guy for resources, whether it be houses, schools, holidays etc. So I maintain that overweight. If the NASDAQ dumps 50% then yes I take a hit but relative to guys at the tech companies, I’d be sitting pretty.
Similarly, I’ve been overweight healthcare for two decades. There are good demographic arguments for it but also I financially support my parents. Healthcare costs have increasingly become a significant part of that liability. It’s not a perfect hedge but it’s a relevant proxy. Healthcare inflation and healthcare sector equity returns are well correlated.
@ZX:
Your tilting approach sounds perfectly reasonable – assuming you can identify appropriate hedges. I guess it is fundamentally personal and will be subject to change e.g. when you retire, as the kids leave home, etc. Does using such an approach require a fair bit of your time?
I will constanly need to be reminded of this fact. And I’m sure the rest of the readers too. We know it makes sense, but that shiny new toy is so shiny!
A bit like a meditation practice, once you realise you are lost in thought, come back to the beginning. Thanks for being a constant voice of reason in my investing life.
A timely post to keep me from straying too far…
The bulk of my passive investing is in VWRL but over the last year or so, I’ve invested small amounts in ‘themed’ ETFs, the shiny new toys you mention, including clean energy, global water, ageing population and esports. All have done well for me in 2020 but who knows in the future? I’ll keep my bets small so I can sleep at night!
10% max. for tilts seems a good guide.
I recently started a few weeks ago by buying Blackrock resources IT (BERI), which has done marvellously already (volatility I know).
Was also wondering how to get into private equity, and then I saw the iShares Private Equity ETF (IPRV) and purchased a dollop…. just seems to trend with VWRL at the moment.
@ Weenie & Algernond – I am shocked! Shocked I tell you 😉
Hi, great article, thanks.
I am guilty of a few tilts, one was taking a small position iShares Clean Energy on 30 July 2020, and it has increased around 115%. So, a bit perplexed by your summary of a fail at only 14% growth?
It’s worth remembering that the Global tracker has done very well of late, mainly because US large caps form such a large part of the index and its done great for the last 10 years. It’s never going to be a bad option.
I have some nice data courtesy of the Association of Investment Trusts, that compared individual trusts with a wide range of data indices going back to 1997. They stopped publishing that comparison data in 2017, but its very interesting..
A typical generalist global Investment Trust after costs( Foreign and Colonial is one I used, I have it held from 1990) does slightly better than the world index before costs (F&C IT was competently managed, costs were fairly low, often at a discount and the gearing offsets the effect of the charges) I used the data from the 16 snapshots of 1,3,5 and 10 year performance in each of July 2007,2009,2012 and 2017 the FT250 was equal to or much better that the FTSE All World Index…using the same data.
( the S&P500 was often a laggard of the World index too with this data…)
Sure thats looking after the events, but it’s important to remember that trends change over time, past performance is not a great guide to the future.
However the advice that the Global tracker is a sensible investment is good but there are plenty of other rational portfolios that are likely to be at least as good.
Good piece here that sheds more light on why thematic ETFs can run out of momentum after the initial burst of enthusiasm:
https://abnormalreturns.com/2021/01/24/when-the-hype-fades-the-case-of-thematic-etfs/
Excellent piece. Thank you.
Obviously there’s sector specific reasons at play too, but might the outperform for consumer staples type stocks like tobacco be due at least in part to the low volatility anomaly? A leverage low volatility equity ETF could be a useful product especially if the leverage was of the capital efficient/ return stacked variety with physical replication of the low volatility stocks with 90% of the capital and the balance of 10% used for futures based exposure to 90% intermediate bonds (1.8x leverage); rather than the daily reset 3x LETF type.