What caught my eye this week.
I read hundreds of investing-related articles to compile these Weekend Reading links each week. Far more than when I was just doing my own active investing research.
I enjoy it. But I also wonder how much it skews my perception of the markets and investor behaviour.
Hot takes and weird observations are what spreads and commands attention, after all. Nobody is very motivated to write “same old, same old” – except of course my staunchly passive co-blogger.
And reading all this kerfuffle every week has led me to wonder whether the stock market really has become as ‘degenerate’ as the Millennials commentators say?
Or is that just how it appears from inside this snow globe of opinion?
Funding the fanaticism
Some things have clearly changed a lot over the past decade. Mostly driven I’d suggest by free share trading and vast social media platforms, but also by the influence of crypto – especially the mega-bagging returns from Bitcoin and Ethereum that have underwritten this shift towards investo-gambling.
How many twenty-somethings would be YOLO-ing their life savings if Bitcoin had fallen back to $10 and stayed there?
Exactly.
It didn’t though – it minted millionaires – and the lingo of the resultant crypto movement has leaked into how punters wielding free share dealing apps talk about stocks, and how at least some trade them.
There’s lots of reports with data showing that retail traders are an ever-bigger driver of stock volatility. But are these just the same people who were punting on tinpot resource stocks 25 years ago, and hyping their trades on ADVFN and The Motley Fool?
Or is it all a sign of some deeper structural malaise?
Asymmetric investing warfare
Over the past few years a narrative has developed that explains this apparent embrace of reckless speculation not through the technological drivers I see – zero-commission apps, mass-broadcast platforms, and blockchain – but through an almost Marxist lens.
In a compelling piece this week, a crypto-focused blogger called Jez presented what he dubs ‘hypergambling’ as a logical response to asset inequality:
the core issue here is the cost of owning a house, and the expected timeline on an average salary.
with this core social contract broken, people look for shortcuts. crypto, memestocks, and the rise of option and leverage trading are examples of the public’s increasing desire for volatility and asymmetric upside when linear can’t buy a house.
It’s interesting that my fellow curator-in-arms Tadas Viskanta also believes these forces are real:
For a long time it seemed the arc of financial markets was bending towards the interests of the individual investor. One could easy argue that arc has shot off in another more degenerate direction.
But then Tadas reads even more from the opinion hosepipe than I do. Pehaps he’s suffering from the same narrative overload?
Either way, there’s also the bigger, bigger picture.
If you’re someone like me who believes the current US administration is wildly overstepping multiple lines of legality, cultural norms, and decency, then it becomes even easier to fear the wider world “turning and turning on the widening gyre”, as Yeats once put it:
“Things fall apart, the centre cannot hold. Mere anarchy is loosed upon the world.”
Why play by the old rules when even the ostensible leader of the free world is trying to bend the data to his will?
As the longstanding economics blogger at Bonddad put it this week:
Now we have the additional wrench in the works in the form of a mafia-style blowout being the operative behavior from the US Administration.
If sowing chaos were a winning economic move, banana republics everywhere would be wealthy.
There’s a good reason why they’re not, and that’s because chaos and corruption make it impossible for producers to foresee the results of their economic actions.
With the first family having their hands all over crypto even as legislation is rewritten by their guys at the top, the stage is arguably set for what Bloomberg’s Joe Weisenthal has dubbed ‘The Golden Age of Grift’ [paywalled link].
Investment manager Cullen Roche quotes official statistics to show a trend that isn’t all in our heads:

Will this chart now go ‘to the moon’ like a heavily-pumped memecoin? Or will the US government stop collecting the data before it gets the chance?
Unfazed while Rome burns
This dispiriting landscape is a long way from the core Monevator message of sensible passive investing.
Heck, even my active investing antics are snoozy and long-term by comparison.
And in contrast to the flashmob stock punters who gather at Reddit’s Wall Street Bets, I’ve stressed you should take what I and anyone else writes with a large dose of salt.
Moreover there’s plenty of evidence that ever more people are investing in index funds.
Fund giant Vanguard has produced data too that shows very few of its customers are trading in and out of their funds based on the latest news headlines, or other tumult in the markets.
So which way are we really going?
Perhaps like everything else these days we’re polarising into two camps. Shut-out degenerate gamblers looking for a quick leg-up into money-baller society on the one hand, and steady Eddie millionaires next door – eventually – plodding towards financial freedom on the traditional path on the other?
Or perhaps it’s all just light and mirrors and it’s the same as it ever was?
Tell us what you think in the comments, and have a great weekend!
From Monevator
How to construct your own asset allocation – Monevator
Stoozing: why borrow money on a credit card just to save it? – Monevator
From the archive-ator: When to buy insurance – Monevator
News
Government considers replacing stamp duty with a new property tax – Guardian
The lowdown on London’s new ‘Pisces’ market for private companies – Yahoo Finance
Borrowing dip offers some respite for Reeves, but tax rises still loom – This Is Money
The average retiree spends £22,140 a year [And other retirement data] – Quilter
DeepFake of Anthony Bolton drives latest ‘pump and dump’ shares scam – This Is Money
The ONS is overhauling how it calculates house price statistics – ONS
UK housing has slightly outpaced population growth over the past decade – Property Industry Eye
Post-Brexit industrial resurgence latest: UK’s third-largest steelworks collapses – BBC
Denmark to end letter deliveries in sign of the digital times – BBC

Built to letdown: housing supply up, rents…up? – FT
Products and services
Where can you earn inflation-busting interest rates on cash? – Which
The pros and cons of fixing your mortgage for ten years – This Is Money
Get up to £1,500 cashback when you transfer your cash and/or investments to Charles Stanley Direct through this affiliate link. Terms apply – Charles Stanley
Lloyds Bank launches new way to deposit cash in shops – Which
Freetrade’s shares ISA will be free from 1 September – T.I.M. [Sign-up for a free share worth up to £100 via our affiliate link]
Most affordable commuter hotspots revealed – Yahoo Finance
Get up to £100 as a welcome bonus when you open a new account with InvestEngine via our link. (Minimum deposit of £100, T&Cs apply, affiliate link. Capital at risk) – InvestEngine
Where are the cheapest places to buy a cottage…? – Which
…and more characterful cottages for sale, in pictures – Guardian
Yet another long-term government bonds mini-special

Why it’s worth watching long-term gilt yields [Paywall] – Bloomberg
Long-term rates are rising with no compelling explanation… [Video] – Sky News
…though inflation came in at a hotter-than-expected 3.8% in July – CNBC
…and the US curve is steepening, too – Apollo Academy
Are long gilts at 5.5% a no-brainer? – Interactive Investor
Fiscal dominance and the unexpected rise of emerging markets [Paywall] – FT
Comment and opinion
“I’m still working at 70. I love my job so much, I commute three hours a day” – The Times
Tax policy prevarication comes for the property market – Propegator
The extremely frugal might be on the right side of history – Guardian
Stop wasting time worrying about safe withdrawal rates – Purpose Code
Gold is shiny enough for a strategic portfolio allocation – Carson Group
Playing the ultra-long game – Novel Investor
Crypto and your portfolio – The Uncertainty of it All
How to eliminate that intense financial FOMO – Financial Samurai
Why these 75-year-olds love working – Next Avenue
Larry Swedroe returns explanations mini-special
Price predicts future equity returns, not future earnings growth – Morningstar
The key drivers of corporate bond returns – Larry’s Substack
Naughty corner: Active antics
The calculus of value – Howard Marks
Where to invest when nothing looks cheap – Morningstar
GLP-1s are booming. Shares in their producers, not so much – Sherwood
Things are hotting up in the UK REIT sector – CNBC
Harvourvest CEO on private equity’s great jumble sale – Semafor
Kindle book bargains
What They Don’t Teach You About Money by Claer Barrett – £0.99 on Kindle
Too Big to Fail by Andrew Ross Sorkin – £0.99 on Kindle
50 Economics Ideas by Edmund Conway – £0.99 on Kindle
Mastering the Business Cycle by Howard Marks – £0.99 on Kindle
Or read one of the best investing books of all time – Monevator shop
Environmental factors
Are we on our way to Earth’s sixth major mass extinction? – Guardian
The climate crisis will blow up via the insurance sector [From July] – How Things Work
Salmon breed in Yorkshire’s River Don for first time in 200 years – BBC
Alphabet [Google] is the latest tech giant to fund a nuclear reactor – Semafor
For heat stressed trees, autumn is coming early – BBC
Wildlife is thriving in Korea’s demilitarised zone – Guardian
The Thames has dried up just a few miles from its Cotswolds source – BBC
Robot overlord roundup
MIT reckons 95% of generative AI pilots are failing – Fortune
AI is a mass delusion event… – The Atlantic [via Abnormal Returns]
…with even Microsoft boss troubled by reports of ‘AI psychosis’… – BBC
…but should we really embrace a world of many AI personalities? – Noema
Evidence investors use ChatGPT in their trading – Marginal Revolution
What if AI doesn’t get much better than this? – Cal Newport
The puzzle of AI facial recognition – Harpers
Not at the dinner table
What’s with the thousands of Union Jacks and St George’s flags? – BBC
Charity workers being targeted by far-right anti-asylum activists – Guardian
The hidden costs of trade protection – Larry’s Substack
ICEing the US economy – Paul Krugman
Off our beat
The violinist problem – Seth Godin
Mapping the battle for online grocery delivery – Platform Aeronaut
A veteran’s guide to self-publishing [Exhaustive!] – Kevin Kelly
The hypersonic missile race is hotting up, and the West is far behind – BBC
Materialists: a true reflection of today’s dating market – Guardian
Over-tourism is hitting Europe’s hotspots, and some locals are fed up – CNN
Rotten Tomatoes is rotten – Stat Significant
And finally…
“…0.01% of your net worth is actually a great proxy for what constitutes a trivial amount of spending for you. For example if you have a net worth of $10,000, then paying $1 more (or 0.01% more) for something shouldn’t have any long-term impact on your finances. Similarly, if you have a net worth of $100,000, you should be able to pay $10 more for an item without skipping a beat. I call this the 0.01% rule.”
– Nick Maggiulli, The Wealth Ladder
Like these links? Subscribe to get them every Saturday. Note this article includes affiliate links, such as from Amazon and Interactive Investor.
It is as it always was-life is a crapshoot but there are certain well tried and proved behavioural rules that give an investor a fighting chance but that’s all it is -a fighting chance
Environmental factors like wars and tsunamis etc can make fools of us all
I think there are two camps but I would divide them into “extremists” and the rest
Extremists ( a small percentage of investors) move the game on but mostly crash and burn-active investors ?
The rest (most investors) put safety over performance-index investors ?
It’s a complementary synergistic system that seems to work
As a paid up member of the rest I constantly follow the antics of the extremists with great interest and deep fascination-there are nuggets of gold in amongst their wild antics
Its sobering to remember that John Bogle was classed as an extremist in his early years but his particular golden nuggets moved all of us investors many notches up the successful investing scales
xxd09
Can you share a link to the vanguard survey you mentioned about index fund investors not moving in and out of markets?
It would interesting to see the picture across wider index providers. I expect vanguard’s “boglehead” following may be more disciplined than others.
I think speculation is very much a “same as it always was” situation, it’s just more visible now with social media.
Ps. That investment scam chart is a real eye opener!
The reference to Howard Marks memo on value is thought provoking.
We know equities are expensive, AI will change things, will AI equities provide exceptional returns…
We have seen the TMT market exuberance of the late 1990’s , we know how it played out and it may well provide pointers for what’s happening now. Back then we saw that playing safe early was the same as being wrong , the art is riding the up wave and getting out toward the end of the run. On that occasion it went on for five years, I wonder whether that previous experience will guide what will happen this time round ? Do people try and jump slightly earlier ?
We have crypto doing its thing at the same time, personally I see zero underlying value and expect it to end badly but I don’t ‘get’ gold either , so it’s probably me !!!
I wonder how potentially two simultaneous crashes could play out ?
Emphasis on value/yield/ rafi indexes for equites ?
I read Distortion by Nomi Prins recently. The thesis was that the money pumped into the financial system in the GFC has distorted the world of finance by bloating it far beyond any correlation with the ‘real’ economy of goods and services. This has, of course, made those with assets, you and I at the very bottom end, feel richer than perhaps we really are.
So I’m with Jez. hypergambling is perhaps a logical response to asset inequality due to asset inflation. As an analogy, I was that asset-lite muppet in the early 1990s, getting on for half underwater on a house. I did seriously consider letting it all go, walking away and going to work in Germany – the world was less interconnected in those days, as an EU citizen I had the right to work there, those were the days my friend, we thought they’d never end… I would probably have been able to shuck the ‘with recourse’ issue of the mortgage. Abbey National would have been able to claim on the mortgage indemnity guarantee they effing well charged me £1000 on top for even though it did me no good at all, and hello world, start again.
As it was I concluded that the future value of a decent job and the use value of friends and local knowledge was worth more. But I can sure understand the temptation of the disaffected to lay it all on red, because I saw it once in different circumstances.
If the thesis of the book is correct, and it was at least made cogently, then for those without assets life as a lottery ticket is a rational response in aggregate. A lottery with its egregious promise of it could be you is a financial system where people’s dreams are far greater than the underlying reality, because everyone thinks it’s more likely to be them than it really is.
The pumped up financial system of today is not the homely farms and productive assets that Warren Buffet cites as the reason to be in the stock market. There’s more of a lottery in it than old-stagers give credit for, this ain’t your father’s stock market. We do not know how the story will play out.
Some of this asset inflation is because virtualisation has dematerialised a lot of things, but they still theoretically have value – ChatGPT is patterns of code running on patterned stones. But some of the distortion is inflated asset prices, houses people can’t buy because all that created money has to be lent out, and various other pathologies.
Even as an old git I have some element of Naseem Taleb’s barbell in the mix – I buy stuff that I can’t see as other than going to zero (AI, biotech) as a small part of my holdings. Because of the intense belief in this trash – well, not so much biotech, but AI still. Even though I think it’s all tulip bulbs in the end, the world will never have enough power to run the wet dreams of AI fanatics and AI will never do what it’s supposed to IMO. We are already seeing hints of the impending AI crash in the Zuck doing an AI hiring freeze, presumably because even as a tech bro perhaps he fears this is the great circling drain.
I’m not a hypergambler because if the risk end barbell goes to zero I can let it go and say that’s not a big part of my networth. But if you start with zero and can walk away from any downside, well, leveraging up may make sense. Somebody has got to destroy that dreamed up capital the GFC created, and some mix of hypergambling GenZ, inflation and financial repression will probably do it. That will probably also cost passive investors something. Not all of the presumption that equities go up in real terms in the long run is due to greater efficiency and productivity. Some unknown part of it is GFC flab.
@Hariseldon #3 > Back then we saw that playing safe early was the same as being wrong ,
I believe Warren Buffet didn’t do so much tech, because he didn’t understand it, so he says. He seemed to do OK. Nobody has to invest in the tulip bulbs du jour, as I think WB also said, you don’t have to swing for every ball
If La Reeves is worried about the cost of long term borrowing I suggest that she announce that the IHT-raid on pensions will not apply to any gilts held within pensions as long as the gilts’ maturities are greater than twenty years away.
Well, if all sorts of other mad advice is offered to her, why not this bit?
Obviously there’s a world of difference to meme stonks and s**t coins OTOH, and market-cap-weighted index funds and ETFs on the other.
But… isn’t there just a teeny weeny bit of the Perpetual Motion machine in both, if truth be told here (or, as Terry Smith puts it, “Passive Investing: The Self-Reinforcing Momentum Strategy”)?:
A company buys back its own shares, so fewer shares are left in circulation.
This makes its earnings look bigger on a per-share basis, which makes the stock look cheap.
As the stock price climbs, the company becomes a bigger part of major indexes like the S&P 500.
Since index funds have to buy more of the biggest companies, they keep adding to that stock.
This extra buying pushes the price up even further.
After all, @TI, you invest exclusively actively because, I presume, you see opportunity and risk not captured by the ‘passive’ approach.
There’s many ways of winning (getting to your personal goal) and many types of winning (what that goal is).
Some want all weather
Some want low volatility
Some want near certainty
Some want highest risk adjusted returns (Sharpe and Sortino)
Some want to shoot the lights out
Some want diversification
Some want concentration
Some want value
Some want growth
Who are any of us to judge? And there’s no need for exclusivity or permanence. Goals change and people change. A mix of approaches can bring its own benefits (and challenges).
I don’t like crypto but I’d never say no to considering a dabble in it just because of that (and I have some minimal MSTR exposure for that very reason).
Each to their own. Live and let live. It takes all sorts of investors and speculators to make a market.
Good post – perhaps the only rational thing to do for young people given their situation is to try and shoot the lights out, with at least a portion of their portfolio, if they are lucky enough to have one. In fact taking a punt with part of your portfolio may be the only way for anyone to move to a higher level without jeopardising their future too much.
Crypto, and altcoins in particular are ways of doing this. In the past, it would have been micro/ nano cap stocks – “penny stocks” – which are still useful to “turbocharge” a portfolio as the returns can be quite extreme.
Interestingly we are coming up to the possible end stage of the boom year of the Bitcoin/ crypto four year cycle which usually ends in a large boom in Bitcoin, followed by altcoins. If this plays out as it has done in the past then some large returns are still ahead of us, particularly in altcoins!
Of course, this is not a recommendation, or advice!
I don’t think it’s as binary as described with many pursuing 2 or 3 strategies. For example, of my portfolio:
75% is largely passive long term holdings
20% is more active trading but still would be willing to hold for medium to long term if needed
5% is my fun fund which I use to gamble with including crypto
@Indyinv 3.0 #7: alt season?…KR1? 😉
OTOH 28.11.2012 (halving 1) supply 10 mn, reduction in coins mined p.a. 1.3 mn (from 2.6 mn); 20.04.2024 (halving 4) supply 20 mn, reduction 325k (from 650k) – so should have only about an eighth of the effect on stock to flow now of that 1st halving.
And SBF’s empty boxes and all that….
I sense a return and increase in the view of mainstream financial bloggers (not so much here) that index funds are causing increasing distortions and problems, and that such investors may experience losses at some undetermined point in the future. Indeed, big losses may be coming for investors, including those mostly in indexes. The only thing that matters (and what anti-index reporting ignore) is whether most active funds are still underperforming, as surely if the market is being distorted with all the dumb money going to indexes then all the “star fund managers” will have a much easier time picking stocks and navigating the general bloat happening, to beat the respective index?
Alas, the SPIVA research continues to show gross underperformance on active fund managers and more humerously, continues to show that those few who did manage to outperform in recent times then typically have even worse returns in the years that follow.
I personally think a big reset is probably due and everyone takes a big equity loss, but I doubt we will see active management beating index returns after fees in any reasonable proportions. Some unknown fund managers may emerge from the ashes of an AI crash due to an anti-AI play/strategy as the next star fund managers, but on the whole, until active managers start delivering better returns after fees, it is hard to support them.
Active managers who carry out thematic investing and run such funds with such objectives are fine, as arguably, beating market returns etc is not their aim, but for your run of the mill active manager that is simply trying to beat the market – the grift will continue.
I may sound like Eeyore but I have been worrying about a government debt crisis for years now and about a crash in equities for over a year. So I hope I am as well prepared as I can be.
Over the last year, I have driven my equities level down very significantly. I also hold largely equity income funds which sometimes suffer less in a crash. Those equity income funds are mostly quoted and those have stop-losses on all of them. I don’t know when the crash is coming, obviously, but I sense it will be within 12 months.
As for a government debt crisis, I have tried to diversify the countries held and have switched the balance towards inflation-linked in case of attempts at financial repression.
What do retail investors, largely I think in the US, believe they are doing in keeping buying in large volumes at these market highs? I have no idea. Americans can be quite delightful – but equally surely they are a very peculiar breed. I have not joined in, with no net purchase of equities really for ages. I think a lot of folk are about to take a serious bath.
@xxd09 — Good point about Bogle, and I agree with you that it’s good to keep an open mind. (I wish I’d had my mind a bit more open in the first year when I heard about Bitcoin, in fact… 😉 )
@Edward — It’s come up a few times, I wasn’t thinking of any particular study. Here’s an article from FT Adviser though:
https://www.ftadviser.com/platforms/2023/05/15/majority-of-vanguard-investors-did-not-trade-in-2022/
Admittedly this isn’t the study I’m thinking of re: bear markets etc (I think the one I’m thinking of looked at the Covid 2020 volatility) but it’s out there somewhere! 🙂
@Hariseldon — I think Bitcoin is apart from crypto, in that it is gold-like now, but yes if you don’t get gold… 😉 Agree with you and @delta hedge that it’s impossible to know whether we’re in year one or year five, except that for all the sophistry about start and end dates this US bull market really seems to have been going on for a long time.
@ermine — Yes, I think there’s a near-zero chance that near-zero interest rates for so long didn’t effect *something*. Turning off ‘gravity’ is sure to do that. I just don’t believe it was a malign conspiracy, not that you’re saying that and it’s an article for another day anyway. (Lots more food for thought in your message! 🙂 )
@dearieme — Hmm, what happens when my 20-year gilt because a 19-year gilt because I had the temerity to live another year? 😉
@Random Coder — I differ a bit from my co-blogger, in that I think index flows probably are having some kind of distortive effect on markets, though I don’t think it’s likely to be as extreme as some make out. The best companies are broadly getting the most money and getting bigger, and their valuations are not wildly egregious versus some smaller companies that are being bid up the old fashioned way, but I’m prepared to believe they’re a bit bigger than they would have been in a passive-flow-less world. Either way, where me and my co-blogger do agree is that there’s not much the average person can do about it except tilt away from the US to some extent, and even that has been a mostly losing trade so far since we started mooting it 1/2 years ago.
@all — Thanks for all the other comments and thoughts. Hopefully this thoughtful community can be a bit of a vaccination against the more extreme ones out there, even if some readers choose to walk both lands… 😉
Some interesting links this week. Especially enjoyed the increasingly jaundiced view of AI.
I also found the How Things Work article on the role that the insurance sector will play in coming environmental transitions. I have long thought that the insurance industry is one of the most clear-eyed assessors of what will happen given what we know. Swiss Re has produced some excellent work on climate threats and the built environment. Up there with the IPCC and our own CCC. Whether anyone, particularly politicians, is listening is another issue. But the insurance industry has great power to affect what will or will not be done in the future.
On the subject of this week’s discussion, yes, it all feels febrile and dangerous, but what are you going to do? I am a firm believer that optimum strategies are only discovered in hindsight and are never predicted. Yes, we all know that a bubble will burst, but when, and how can we identify a bubble from one of the rare radical changes in the global economy? So having set out a strategy that seems to make some kind of sense based on history and your understanding of risk, following the irrational twists of market sentiment seems pointless. No amount of reading the entrails will protect you from the chaos. Yes, some people will come out of it smelling of roses, and some will claim it was their brilliant strategy, but it was not. Blind luck has much more to do with it.
So we earnestly discuss the signals and the behaviours of people individually and in aggregate, and we are all worried by it, but point in different directions for our salvation. Let’s face it, we know a lot and nothing at the same time.
The Bloomberg article on yield spreads between 30 year and 10 year gilts (and also seen in other countries too) is interesting and set me to wondering what have typical spreads been.
A look at the Ellison and Scott database of historic gilts (a version of the paper is at https://www.lse.ac.uk/CFM/assets/pdf/CFM-Discussion-Papers-2017/CFMDP2017-27-Paper.pdf and the data are available online) shows that between 1915 and 2018 the yield spread (or term premium) between gilts of 20* and 10 years has ranged from -2.5 percentage points to 9.7pp (the latter in 1975 during another period of uncertainty for the UK and globally following the oil crisis) with a median of just under 0.1pp and 10th and 90th percentiles of -0.7 and 1pp, respectively. Similar results (but limited to the period from 1970 onwards) were found using the BoE database.
In other words, the current position is not yet too unusual in an historical context (e.g., the decade of QE after 2010 saw spreads in the range of 0.5 to 1.0 pp).
*I chose 20 years because, although the database includes 30 year yields, historically there were long periods where there were no dated 30 year gilts and the 30 year yield was then interpolated from those of undated gilts (e.g., consols).
100% a real phenomenon: find the alternative path.
I started work in 2000, got made redundant in 2001, then watched house prices more or less doubled by 2003 and thought: what is this? I dug in, learned about self-cert mortgages and how the money system really works, and waited for the reckoning. Wages didn’t move much while houses kept rocketing.
The financial crisis came and, yes, I messed up (financially) by not buying in 2009 because I thought it wasn’t over. It isn’t – but we know now who gets protection and to what extent they’ll beggar the masses to keep the party going. Around then I dropped to part-time to claim my time back as they can’t tax what isn’t there. I waited again: prices flatlined… then Osborne rolled out Help to Buy in 2013.
At that point, apoplectic, I decided no more than the legal minimum goes into this system. I salary-sacrificed everything above minimum wage into my pension (SIPP). Eventually, a bit late but still early, I bought a small amount of BTC.
Since then my wage has accelerated, to the point I can bung the full annual allowance via salary-sacrifice into the pension on a three-day week. I’ve got enough of a buffer above minimum wage that I’m eyeing a £6k e-MTB to complement the £7k gravel bike I picked up on Cycle to Work a few years back, another neat way to reduce the tax take.
With a stuffed SIPP, some BTC, and a bunch of other bits and pieces, I’m more or less FI. Most of that is simply the tax saving. If I weren’t a degenerate gambler on the odd stock punt, I’d be even better off!
And yes – they drove me to it by rigging housing. I would have been a good citizen if they hadn’t!
‘The climate crisis will blow up via the insurance sector‘
In the UK the industry and government came up with Flood Re as a way to spread risk, and allow homeowners to continue to receive flood cover. It was only meant to last 25 years, but it’s no secret it will almost certainly be extended.
The fact that retail are an increasing proportion of equity volatility should be welcomed. Equity volatility is far too low and needs to increase.
What is an issue is retail’s BTFD mentality. They are so used to equities only going up in the last 15 years, that they don’t consider the possible downside. Add to that the issues around index investing, which reduces market elasticity and exacerbates momentum. Everything goes up forever. Till it goes down forever …
Retail trading was driven, foremost, by tech lowering the bar to entry. The second element is the attention economy. It was MSNBC first but social media really shoved the tech stocks and the S&P500 in everyone’s face. I’m not sure the majority, though, are “hypergambling”. Even Gen Z. The numbers suggest most are actually relatively cautious: looking at holdings of cash vs. stocks.
I’d say it’s a significant minority, mostly those who’ve been most captured by the attention economy, that are vulnerable. If you spend too much time on social media, you will end up deciding you might aswell gamble everything. You’ll be told repeatedly that the world is stacked against you. Fiat currencies are worthless, everything is corrupt. The global world order is looking to replace you with someone more ‘brown’ or that superintelligent AI will first take your job, then exterminate you as an after thought.
This is all being now being driven hyperbolic. We are in a world where the US president and his adminstration is totally corrupt. They tell us that there are no facts, just opinions. That Silicon valley has evolved from genuine tech industry increasingly into snakeoil marketing. ChatGPT-5 is PhD level at everything … except it can’t even win a game of chess against a 1970s era games console.
#18/@ZX: if 80% of ‘active’ managers aren’t really active, and FAPP shadow the index with some smaller outlier ‘contrarian’ positions; and with >50% of the US large caps now formally held by trackers (S&P 500 passive share ~53% with the passive share going up currently by ~3% p.a); then. without retail directly investing into single names, there’s not much else going on by way of meaningful sources for price discovery.
#10/@Random Coder’s of course right that ‘active’ managers, as a group are pretty hopeless (SPIVA etc), and that it’s a zero sum game – but, tbf, most active managers aren’t even really trying to be truly different.
They just can’t risk it when the maths of passive flow meets market inelasticity (plus lower fees/costs for trackers) means truly active has odds stacked against it.
On the question of ‘is now equivalent to year 1 (1995) or 5 (1999) of the current tech set up’; there’s no magic rule here that we’ve got exactly 5 years for this cycle. Dot.com lasted about 5 years, but Nifty 50 was just 3 (1970-72).
And Dot.com was embedded in a larger bull cycle, either beginning in 1974 or in 1982, and lasting either over 25 or over 17 years, right through until March 2000 (during which, when measured from an August 1982 start, the Nasdaq did 3,000% overall).
Maybe there is a pattern here, maybe there isn’t (colour me sceptical though: my thoughts are here):
https://monevator.com/what-to-do-if-youre-queasy-about-the-us-stock-market-members/#comment-1904581
@ Delta Hedge 9 – KR1 is certainly an option but there are a few others too!
As far as whether we are in a bubble or not – I think it’s pretty clear that we are, or at least at the beginning/ middle of one – but that is not that important. The thing to realize about bubbles is that they can go on for a lot longer than you might imagine and if you don’t take part you might miss out on some great returns. However, I do think we are getting in to the frothy stage (similar to a late 1990s meltup). My base case, is that before this is all over, the S&p 500 will have hit 10,000, the Dow 100,000 and the Nasdaq 50,000. But after the boom there will be the inevitable bust so buying the S & P 500 now may mean that when the bust arrives the S & P 500 would have doubled and then come right back down to the level it is at right now, if not lower. So there is certainly risk in buying at this level.
The trick, of course, is to ride the bubble until it peaks and then either sell or hedge! My plan is to start hedging the downside around 10,000 on the S & P 500.
I think the Powell Jackson Hole speech is indicating a change which will only add fuel to the fire of the “bubble” – lower interest rates.
Not financial advice of course.
@Delta Hedge # 6
Is this actually the case? Sure, if I divide Nvidia by 10 the stock price goes up by 10, but that’s because it represents more of the company? Index funds can chill the heck out because they pay 10x more but need to buy 10x fewer new shares.
@ZXSpectrum48k
> If you spend too much time on social media, you will end up deciding you might as well gamble everything.
the #1 problem is switch that shit off, it craps in your mind, But addictive stuff is like that. OTOH
If you got nothing to lose, why the hell not? I asked myself that question once, it was a knife edge yay or nay. The tragedy is that the world is stacked against the young. I was that young pup once, and after sinking God knows how many pints of ESB in BBC Broadcasting house bar listening to yuppies chuntering about how much they made on property I cycled back along the A40 to my rented shithole, put salt down to keep the black slugs out and determined to GTFO London because I was obviously too poor to live there. I was a lousy broadcast engineer making stuff where I should have been flipping houses on BOMAD’s dime. Trouble is, I had a working class BOMAD so it wasn’t good for ‘owt. They did try to tell me don’t do what I was doing in fairness to them 😉
Only to torch 2x gross salary in the ensuing 1989 housing market clusterfuck, the one good thing was I didn’t do that in London. TI has a property blog that I haven’t looked at because I paid good money for my dinner and would prefer to keep it in my gut but hey, SOSDD 😉
I am old, but I’m with these Gen Zers. The system is stacked against you. For the love of God I’m not saying it’s a conspiracy @TI #12, capital always conspires against the impecunious, I think that Adam Smith geezer had something to say about that back in the day. I was suckered two effing years gross salary by the housing market. On the upside, I have a capital gains tax problem of several years’ worth of the high water mark of my earning capacity, what the housing market took away the stock market giveth, now that I am an old git. Tragedy is, of course, that the sands of time are running out, would I swap my CGT problem with not eating that shit in my twenties. yup, in a heartbeat. Still working on the time machine in my basement.
But the stock market apparent gain of the last 16 years ain’t all real. Above all else, owe now’t. The grapes of wrath are being trodden, and the vintage is sour and you can taste it in the wind. Don’t owe money. However clever you think you are.
Yeah. All you need is a world tracker. Hell, it’s my biggest stake, I’m going down with you guys. But I’ve made sure I got options if that falls by half. It’s been known. Oranges are not the only fruit…
An interesting read. I’ve been noticing for a while the disconnect between my circle of friends and “normal” online groups when it comes to cryptocurrency with my friends going nowhere near it and seemingly everyone online having massive holdings. With the rise of AI it’s hard not to start seeing it as a concerted effort to normalise it to feed the few whales with huge holdings in crypto. But perhaps this is just paranoia to justify me missing out.
@ermine #20: the short simplified summary version is: active investors as a group in aggregate focus upon earnings per share, or EPS growth, as an investor buys or sells shares in a business and (this is very important) not the business as a whole (unless it’s an institution doing a private equity deal to take the whole business private, or a corporate takeover of a competitor).
Buybacks result in the shares acquired by the business being cancelled.
This necessarily causes EPS and EPS growth to rise faster than the rate of growth (if any) in profits of the business as a whole.
This makes the shares more attractive on a per share basis, and so active investors then bid them up.
A more attractive asset will have more competition to buy it and more demand for a reducing supply of shares will make the price go up.
As the price per share rises momentum traders step in and buy bidding up the price yet more.
Without passive trackers at some point value investors would likely counteract this (at least somewhat) by selling the shares before they got overvalued, bringing down the price per share.
But, before that happens, or fully happens, the value of the company in capitalisation terms has already inflated and now constitutes a bigger share of its index.
Passive trackers now have to buy up shares to avoid being underweight the stock.
They must do this at their next rebalance. There’s no discretion not to. It’s their legal mandate.
Passive buying rebids up the stock.
Repeat and rinse.
The start of the process is led by active, the end by passive.
The shift in US capital markets here had two drivers. First, in 1982, Rule 10b-18 gave companies legal cover to repurchase their own shares, turning buybacks into the leading method of returning capital, surpassing dividends by 1997 and remaining dominant. Second, passive investing through index funds and ETFs surged, fueling the trend.
A longer version is that the mechanics of share buybacks should not be incorrectly analogised to a reverse stock split (where shares are consolidated, such as dividing the share count by 10 to increase the price by a factor of 10).
In a reverse split, nothing fundamental changes about the company—market capitalisation remains the same because the reduction in shares is offset exactly by the proportional increase in price per share. Each remaining share represents a larger portion of the company, but the overall value doesn’t grow.
Index funds tracking market capitalisation weighted indexes like the S&P 500 wouldn’t need to buy or sell net new shares; they simply hold fewer shares at the higher price, and their portfolio allocation stays balanced without additional transactions or price pressure.
Share buybacks, however, are fundamentally different.
When a company repurchases its own shares, it uses actual cash (often from profits or debt) to permanently retire those shares, reducing the outstanding share count.
This isn’t a neutral restructuring like a split—it’s a capital allocation decision that can enhance shareholder value.
Earnings are now spread over fewer shares, boosting EPS.
If the market views this positively (e.g., as a signal of undervaluation or strong cash flow), the stock’s valuation multiple (like price-to-earnings ratio) may expand, driving the price up more than enough to offset the cash outflow.
The result can be a net increase in market capitalisation, making the company a larger component of market-cap-weighted indexes.
At that point, passive index funds cannot “chill out” by simply holding fewer shares at a higher price.
Their mandate is to replicate the index’s weights, so if the company’s relative market cap grows (due to the price climb outpacing the broader market), funds simply must buy additional shares of that stock to rebalance—often funded by selling shares of underperforming stocks or using new inflows.
This creates a self-reinforcing cycle: buybacks drive EPS and price higher, index weight increases, passive buying adds more upward pressure, and the cycle continues.
This dynamic has been observed in companies like Nvidia, where buybacks (combined with strong growth) have amplified their index dominance, forcing trillions in passive assets to allocate more capital there.
You can’t treat buybacks as a cosmetic change, like a split, ignoring the real economic effects that can inflate index weights and compel forced buying.
BTW this isn’t the biggest cause of passive increasing market capitalisations.
That probably a combination of both:
a). index optimisation which results in over ‘sampling’ the largest index weights and vice versa, such that index trackers using optimisation deliberately over buy the largest index weights causing their share of the index to ride; and,
b). the relative to their capitalisation inelasticity of the largest stocks because market makers whilst holding more shares of the largest stocks don’t hold as many more as inventory as the cap weight would require because they earn such a smaller spread on them (basis points, or hundredths of a percentage) compared with thr smallest stocks (the spread on nano caps can be multiple percentage). With less liquidity relative to their ballooning market caps there’s increased inelasticity in those shares meaning buying and selling has a larger impact than it should otherwise have. Net inflow to passive tracker equity products therefore causes the value of the largest companies to rise. This then in turn has a feedback loop into the optimisation approach to tracking.
@Delta Hedge (#22)
“Passive trackers now have to buy up shares to avoid being underweight the stock.”
I’ve not really previously thought about this much, but I think the outcome will depend on how much the cap weight of the Company with the buyback has increased.
Consider a simple example where two companies (A and B) each have a total of 100 shares in issue at a price of $100 and therefore have equal market caps of $10k. A passive index that just tracks A and B would then hold equal amounts of each.
Assume company A buys back 10 shares. If the price goes higher than $111.11 (i.e., 10000/90) then the market cap of company A will be higher than that of company B, while if the price is less than that, the market cap will be lower.
So, let’s take the case where the price goes to exactly $111.11 and overall Company A and B still have equal market cap (A=90*111.11=$10k and B=100*100=$10k). If the index initially held 10 shares of each, its holding of Company A would now be worth $1111 and that of Company B $1000. It would have to sell shares in Company A and buy those in Company B to rebalance. That transaction would presumably push down the price of A and increase that of B.
However, there will be a threshold price, that I’ve not yet calculated, above which the rebalancing will go from B to A instead.
Note that my simple calculation assumes that there are no contributions to or withdrawals from the index fund.
@Delta Hedge #22 Thanks. Never really thought of it like that. Just what we all need, eh, another thing to worry about trackers 😉
@Alan S #23: I don’t want to overstate my case, but an idealised example ignores the valuation and signaling effects of buy backs.
The example makes an assumption of static market cap equivalence.
It assumes that after Company A buys back 10 shares, its stock price adjusts exactly to $111.11, maintaining an identical market cap ($10,000) to Company B.
While this is a useful hypothetical to illustrate rebalancing, it ignores the real world impact of buybacks.
Buybacks typically signal positive information (e.g., management believes the stock is undervalued or the company has excess cash), which often leads to a price increase greater than the mechanical effect of reduced shares (i.e., above $111.11 in this case).
If Company A’s price rises significantly higher—say, to $120 due to market enthusiasm—its market cap becomes $10,800 (90 × $120), increasing its weight in a market-cap-weighted index.
The example doesn’t account for this dynamic, which is central to why buybacks can amplify a company’s index weight and force passive funds to buy more shares, not sell them.
Then there’s an issue of the characterisation of the rebalancing direction.
The example suggests that if Company A’s price reaches exactly $111.11, the passive index fund would sell shares of A (worth $1,111) and buy shares of B (worth $1,000) to rebalance back to equal weighting.
This is incorrect for a market cap weighted index like the S&P 500.
In a market cap weighted index, the fund aims to hold shares proportional to each company’s market cap, not equal dollar amounts.
If A and B have equal market caps ($10,000 each), the fund should already be balanced at 50/50, and no rebalancing is needed at $111.11 for A.
The description only holds for an equal weighted index, which is not typical for major indexes like the S&P 500.
The mention of a “threshold price” where rebalancing flips from B to A hints at this but doesn’t clarify that in a market cap weighted index, a higher price for A (above $111.11) increases its index weight, prompting funds to buy more A, not sell it.
There’s arguably also an issue with the example around neglecting broader market dynamics.
The example assumes rebalancing occurs in isolation, with the fund’s selling of A pushing its price down and buying of B pushing its price up.
In reality, passive index funds operate in a broader market where buybacks can trigger significant demand.
If Company A’s buyback pushes its price higher (e.g., to $120), its increased market cap raises its index weight, forcing passive funds to buy more A shares to match.
This demand can further elevate A’s price, creating a feedback loop.
The example description ignores this self reinforcing effect, which is critical to why buybacks can disproportionately boost a company’s index presence (as seen with companies like Nvidia)..
If the market perceives the buyback as value accretive (e.g., by boosting EPS or signaling confidence), the price could rise beyond what maintains equal market cap, skewing the index weight further.
The example’s description’s focus on a precise $111.11 price ignores how buybacks often lead to valuation multiple expansion, which drives the need for passive funds to buy more shares, not sell them.
The example mentions a “threshold price” above which rebalancing shifts from buying B to buying A.
In a market cap weighted index, this threshold is simply the point where A’s market cap exceeds B’s, requiring funds to increase A’s weight.
For example, if A’s price exceeds $111.11 (e.g., $120), its market cap ($10,800) surpasses B’s ($10,000), and funds must buy A to align with the new 51.9%/48.1% weighting.
@ermine #24: I wouldn’t worry yet. Even the advocates of the Inelastic Market Hypothesis (which is what we’re really talking about, the absence of perfect efficiency, price discovery, rational decision making under certainty and elasticity) say stay invested in passive.
At some point net passive flow may reverse.
There are two sources AFAICT.
First, and possibly largest, active to passive market share transfer. Second, US led auto enrollment contributions to passive large cap (e.g. S&P 500) trackers.
If both these dry up, then it’s probably “Huston we have a problem” time.
Until them it should be a tailwind to returns.
As the demographic pyramid inverts and if companies do recruit fewer new intake due to “AI”/automation whatnot then the first source of net passive inflows will thereafter but only thereafter dry up.
That won’t be great of itself but it’s no biggie unless the other source also has or is about to dry up.
According to Mike Green’s calculations every $1 flow going into passive trackers immediately increases the $ cap value of the S&P 500, at around a 50% passive share (roughly where we are now), by about $17 to $20, of which half sticks.
For genuine active fund in flow (not just active in name only) the figure is about a $2 immediate capitalisation rise for each $1 of flow, with a similar capitalisation effect stickiness
So as long as more money flows into passive from active then, even if there are net outflows from the stock market overall because of retiree’s drawing down more than those in work are investing, the ‘passive effect’, for want of a better expression, can still keep the market going up.
I can’t see passive share ever getting to 100% though.
Most likely it’ll be an “S” curve with passive share topping out below 100%, but well above where it is now.
The S&P 500 leads the way on passive share. It’s around 48%, 50%, 52% or 53% depending on which source you look at and from when.
It seems to be rising now by about 3% annually.
If it were to top out at not less than say for example 68%-72% and not more than say 85% provided the 3% p.a. stays the same we’re looking Ok out to the end of 2030 at the earliest or end of 2035 at the latest.
Of course, as the ratio between the share of assets held OTOH by genuinely active investors engaged in price discovery and OTOH price insensitive passive investors changes in passive’s favour so the dynamics become ever more extreme and the distortions of net flow into passive greater.
But as long as the music’s playing you’ve got to keep dancing even if it starts as a waltz and ends up like Belgian techno 😉
Should have also mentioned that
a). the biggest firms are the most successful (the most scalable and the most monopolistic) and, therefore, have greater capacity for buy backs. This creates another feed back loop.
b). as multiples for the biggest firms expand, their weighted average cost of capital (WACC) falls, as they’re the safest bet to lend to (at one point Microsoft could borrow for less than the US Government which controls the currency). A lower WACC means that the biggest firms are likely to be even more successful still, setting up yet another flywheel.
In practice, I’d guess that these effects are more subtle and subdued than I’ve perhaps made them out to be above (hence my comment about not wanting to overstate my case), but I’ve spelt out the effects more starkly than I suspect that they actually manifest IRL in order to aid explanation.
While I don’t disagree with the core themes of the above, the bit that is missing is my point from earlier. These potentially runaway valuations and bloating of the bigger companies by “dumb money” in index funds should be making life easy for the active fund managers as the market distortions and misallocations of capital create great opportunities for the hard working, deserving of their money, active fund managers doing all this original research to aid with their stock picking. Such well read and good value buys would tilt the market caps back towards the ‘correct’ proportions, and index funds would then need to rebalance accordingly (away from the dumb money bloat). But alas, active fund managers are doing just as bad as ever.
This is the problem here – index money is bulking up the total market but if the argument is that the market is becoming more and more inefficient and money is flowing to the wrong places, then this is exactly what active fund managers are meant to be navigating and getting paid well for.
A generally bloated market with money slushing around fairly aimlessly (how I view index investors, at worst) will cause problems, but this is the problem active managers are paid for, and the problem of identifying undervalued (relatively speaking) stocks should only get easier as a greater proportion of investors go index. But then, why wouldn’t you, when active management continues to be so bad?
Does anyone think actively managed funds etc as a whole will survive a huge equity correction better than a respective index? I don’t, except for themed funds etc that are working to a different aim than beating the market.
Until active management demonstrates it is competent, why would anyone use it if the aim is simply to invest and grow your wealth?
@Random Coder #27:
Active can’t outperform as it’s net zero sum.
One active investor wins relative to another. Wins and loses net out less frictions like fees. And as a group they charge more fees. So as a group they underperform passive.
They have to as long as their fees etc are more than passives’ are.
However, the extent to which they actually underperform – but not the fact of it – is a bit surprising, unless you also account for the effect of passive flow.
Efficiency or inefficiency doesn’t make markets (or the shares making them up) go up or down in price.
Net fund flows relative to supply of shares for sale do that.
Passive (so far) is resulting in a net constant bid for shares which is strongest where passive share is highest – US large caps, the most heavily indexed area of the equity market.
Whilst on q day to day basis this bid is typically drowned out by all the other factors making share prices move up or down; it’s still always there in the background.
And over time it changes market dynamics and structure.
Largest caps become larger than they would otherwise.
The gap between them and the rest on valuation increases.
The gap between the US, as the most indexed market, and the rest of world rises.
Large caps increasingly outperform small and mid caps.
The growth style increasingly outperforms value.
This is neither good or bad. It is a feature not a bug.
It does, however, fit in neatly with just about every trend of the recent era since index tracking became highly prevalent (and now in the S&P 500 dominant).
An active investor choosing to do the opposite of that trend in their investments (i.e. investing in non-US Small Cap Value, and shunning US Large Cap Growth) will underperform.
And those that have gone against that trend have underperformed.
Very badly unperformed.
Passive flow makes such a contrarian active investor’s lot much harder, not any easier.
And that’s why, as a group, active has underperformed even more (much more) than you’d expect them to underperform just on account of their higher fees (and typically higher trading costs due, for many active funds, to their higher turnover of holdings compared to passive).
Of course, those ‘active in name only’ funds, which really just shadow the index, and which are, to one extent or another, really closet trackers, are not going to under perform as much, so they’ll probably still keep their investors money invested in the fund (fund withdrawals being every active managers’ nightmare, as keeping AUM is all); and all whilst charging a higher fee than an index tracker for doing essentially the same job.
Ok, I can just about accept for some companies, share price has nothing to do with business fundamentals etc, at least until the floor falls in.
The end result is the same though, why should anyone pay an individual for a service when there is another almost identical one readily available that costs less and almost certainly gives similar or better outcomes?
Finally, what are the solutions? A few years back I used to see active management proponents suggest index funds should be outlawed, I haven’t seen much of this recently to be honest but I did think back then it was quite funny how the financial industry can be less free market when they are the ones losing out (think evidence based investor had a lot of tales about this at conferences if I recall correctly). What is the “solution” if active can’t win and passive is potentially dangerous? Because right now, even if passive is bad or dangerous, the only reason to go active is if you genuinely believe a contrarian position will survive a crash or bust scenario better, as like you point out, you certainly aren’t winning with the active fund in normal times.
@Random Coder #29: Noone should pay for an actively managed fund where there’s a cheaper passively managed ETF tracker covering the same sector (with the small caveat that the active fund in question is not an Investment Trust on a clearly unjustified and unusual discount to its NAV).
I don’t see a problem about passive investing requiring a solution.
I just see a situation. It is what it is.
Not every situation needs something doing about it.
But I do like to know what the full situation is, at least where the information to do so is available.
Open eyes and all that.
I could be very wrong on all of this, of course.
If I had more conviction here then I’d just invest in the passive SP20 ETF, which just holds the 20 largest constituents of the S&P 500.
But the whole inflation of US Large Cap valuations compared to the Rest of (developed) World (RoW), Small caps, Value firms and EM stocks is such now that in actuality I’m tilted away from the US towards all those other RoW choices, above all European and EM Small Cap Value.
I own some Palantir directly and Nvidia through trackers, but I just find it more attractive, for example, to own EM deep value (like Columbian EcoPetrol) on 5x PE rather than over 50x for Nvidia and over 500x for Palantir.
The gap now between US Large Growth and RoW Small Cap Value is the largest ever, even more than in 1999/2000.
Deep down, I still suspect that passive flow will cause US mega caps to overall outperform notwithstanding the valuation gap, which is arguably unjustified from a fundamental point of view.
OK, fair enough, I just wondered if you thought passive was causing a problem that needed solving because it could leads to bad outcomes that active investors could avoid. I can believe passive has consequences we don’t appreciate yet but at the same time I struggle to see active funds somehow managing to avoid any such yet to be appreciated issues.
I didn’t invest much at all outside pensions until a few years back and am still too cash-heavy, but nowhere near where I was as recently as 5 years ago, and I probably could quit work now if I wanted. By sheer risk aversion I chose (back before AI exploded) to cap my allocation to ANY geographic region to a maximum percentage I was comfortable with, so have always been underweight US and overweight everywhere else in the relative rescaled region proportions. So yes, I have been underperforming compared to being in a single world index, but that is fine, and will continue into the future by my choices. If world equities crashed to zero tomorrow I would still be in a much better place than most. I viewed a region concentration of 60%+ back then far too high, and still do now, but not because I expect to do better than the world index. I just wanted to gradually increase my equity exposure from near 0% outside of pensions (look, I started investing late after a decade of saving!) and minimise any obvious paths to huge losses. The US concentration in world indexes was viewed as one such path then, and I am reluctant to deviate from my strategy now as I have no need to.
@DH (#26)
The simple model I introduced above (#22) does indeed properly represent a market weighted index. In the, admittedly special case, example I gave where the price of Company A had risen to $111, the market caps were then equal (as you also say in your post) and therefore the index fund would hold them in equal dollar amounts. However, as you said, if the price for Company A exceeded $111, it would have a larger share of the market.
So, lets look at what happens at other prices.
The market weight of Company A, MA after buyback is given by (NA*PA)/(NA*PA+NB*PB) where NA and PA are the number of shares in issue and price for Company A (and similarly NB and PB for Company B).
The proportion of Company A in the fund after buyback, but before rebalancing, FA is given by (nA*PA)/(nA*PA+nB*PB), where nA is the number of shares of Company A held in the fund (which in my example was nA=nB=10, since before buyback the two companies had equal market caps).
So for PA=$120, we have MA=(90*120)/(90*120+100*100)=0.5192 and FA=(10*120)/(10*120+10*100)=0.5455
In other words, Company A is overweight in the fund and would have to be sold to rebalance.
For PA=$200, MA=0.6429 and FA=0.6667 and therefore Company A is still overweight in the fund compared to the market. I’ve calculated the values with higher prices than this and FA always exceeds MA although the difference reduces at higher prices (i.e., an asymptote). It would appear that the statement I made in my previous post (#24) “However, there will be a threshold price, that I’ve not yet calculated, above which the rebalancing will go from B to A instead.” was incorrect – there is no threshold price.
Of course, in this I’ve assumed that the price of Company B, PB remains $100 after buyback, so just for completeness, for PA=$120, and PB=$90, MA=0.5455 and FA=0.5714 (i.e., Company A is overweight in the fund).
My conclusion from this simple model is that for money already in the fund at buyback, Company A subsequently becomes overweight. However, in this example, provided PA exceeds $111, a larger fraction of new money entering the fund will be placed with Company A than Company B since MA has increased.
@DH are new issues expected to have the opposite effect? And not just rights issues. I worked for a company that, as I understand it, funded staff share options by issuing new shares, and also did share buybacks. The net result was sometimes positive sometimes negative. News flow only ever mentioned buybacks, of course.
It’s also interesting that splits should have no effect, but at least in the 90’s and noughties there often appeared to be an otherwise unexpected lift in market capitalisation after a split – which would be the active participants bidding it up irrationally.
Blimey 🙂
I didn’t expect my original post to get this much engagement! 😉
@Alan S #33: I’m not really in a position today to get into the minutiae of the maths, and I anticipate that you may well be right on that (e.g. the threshold issue etc) within the context of the narrow assumptions which the simplified model operates under; but I do think that the simplified model is missing broader market dynamics: It underplays empirical evidence that buybacks often boost prices long-term via EPS growth or signaling, increasing index weights and passive buying through inflows, and also ignores factors like debt-financed buybacks affecting buyback scale.
There seem to me to be 4 main issues with the simplified model, which (if I may be so bold) it either misses/under-acknowledges or (I’d respectfully suggest) gets wrong, namely:
1. Ignoring Net Fund Flows: The model separates rebalancing for existing assets (selling Company A to correct an overweight) from new inflows (buying more A due to its higher index weight). While correct in isolation, this overlooks how massive inflows into passive funds—often billions monthly for S&P 500 ETFs—can dwarf one-time rebalancing sales. For instance, at a $120 share price for A, the model shows a small overweight (~2.63%), requiring minor sales. But modest inflows (e.g., 10% of AUM) could necessitate buying far more A shares, flipping the net effect to purchases. This aligns with real-world trends where buybacks amplify passive demand for outperforming stocks.
2. Static Price Assumptions: The model assumes fixed post-buyback prices and instant rebalancing, ignoring how funds’ trades (selling A, buying B) might influence prices. In reality, buybacks unfold gradually, and passive funds rebalance periodically, allowing price momentum from buyback signaling to alter weights before trades occur, potentially reducing the overweight.
3. Limited Two-Company Model: Using only two companies exaggerates rebalancing impacts on B. In a 500-stock index, sales of A distribute buys across many stocks, diluting effects on any single B. The model also ignores small buybacks (1-5% vs. 10%) and funds’ tolerance for minor deviations to avoid costs, which real funds prioritise to minimise transaction expenses.
4. Overlooking Float and Participation: The model assumes passive funds don’t participate in buybacks, creating the overweight. However, share lending by ETFs could lead to some shares being repurchased, slightly reducing the overweight. It also skips float-adjusted index nuances, like liquidity or eligibility impacts, which matter in practice.
Points 1 and 2 seem to me to be more important.
Point 3 could cut both ways in its effects.
Now, I’m no economist, mathematician etc, and I could well have this wrong, but, notwithstanding that caveat, I do strongly suspect that a simplified model is not the real world landscape, and, personally, I’d not rely on simplified models other than for their explanatory power in relation to an idealised example.
It’s like a Three Body Problem. With two elements it all looks immaculate and predictably linear, but go out to real world constraints of thousands of elements and actors and the system shows non linear, and sometimes chaotic or exaggerated, behaviour.
So I’m pretty confident that market dynamics are typically non-linear and that, at this time (and really since 1982), have favoured – in the market price reaction – EPS growth over dividends, even though $1 bn used on a buy back and $1 bn used on a dividend both reduce cash balances on the balance sheet by the same amount.
@Nearly There #34: Crikey, that’s an interesting point and a bit of exam question in terms of working it out…. but on splits they *should* be economically and, therefore, in principle, market neutral but, in practice, the market rewards splits, as it does buy backs.
Passive net inflow would then exacerbate that both in terms of scale of price response and its stickiness.
Another point’s just occurred to me in relation to my Point 4 (and perhaps my Point 2) above (although, possibly it’s actually two new points, depending upon how you think about it):
In open market buybacks, companies buy from willing sellers (active managers or retail), while passive funds inly rarely sell (as long term holders, passive investors being B&H accumulators in net terms).
This creates a one sided dynamic (as compared to a market with only active buyers and sellers), with a price forcing character.
A regime of buy backs also then reduces the share float, perhaps significantly over time, in turn reducing liquidity, which in a net passive inflow situation – such as we have IRL – further exacerbates the upward price forcing.
This all goes to my earlier points about needing to take account of feedback loops / flywheels, and the more general point about non-linearity in the effect of fund flows.
@DH (#35)
If you make interesting comments, then expect some replies 😉
I agree that this simple model is limited, but it does provide some insight. To your points:
1) Inflows. Annual fund inflows were around 3-4% of AUM in 2024 and 2025, so I agree that in the long-term, there will be more of Company A bought than sold by indices.
2) Timescales. AFAIK, some equity indices are updated quarterly, so that will be the timescale over which the index fund reacts to changes.
3a) Two company. Company B can represent an aggregate of the rest of the index. Where the model results are limited is that I assumed that Company A was 50% of the index. On a quick test, changing that to 10% of the market, increases the size of the overweight if using a ratio and decreases if using subtraction (for $120 and 10% market share, MA=0.1071, FA=0.1176).
3b) Smaller buybacks (e.g., 1% instead of 10%) reduces the overweight.
4) Agreed that none of these effects are included.
@Alan S #37: Many thanks 🙂
IRL example, perhaps, of changing market structure / increasing market inelasticity this evening.
Nvidia Quarterlies just in (last hour or so): with current gross margin slightly up (perhaps the most significant data point, IMO), at 72.7%, from 72.1% (i.e. the competition’s not yet catching up); with next Quarter’s gross margin guidance above expectations (of 73%) at 73.5%. $60 bn buyback (1.35% market cap) announced. Revenues overall a beat.
Overall a solid beat, but a bit of nothing burger.
However, instead of no reaction, a tiny ‘miss’ of $240 mn ($41.1 bn v $41.34 bn, so just over 0.5%) in Quarterly Data Centre revenue (versus quite bullish buy side expectations) sends the stock down 3% immediately in after hours, or over $135 bn of cap weight change for just $240 mn/quarter (i.e. less than $1 bn annualised) difference (v those expectations) in one line of revenue – So, Nvidia, the largest stock, and supposedly the most liquid, is manifesting about 2.5x to 3x the rate of change in cap weight response which one might expect from Nvidia’s trailing PE of 58x and forward PE of 50x alone.