October has an extraordinary history of sudden financial crashes.
Six of the eight largest falls in Dow Jones history happened in October, including 1929, 1987, 1997 and 2008.
Couple that fact with Bank of England Governor Andrew Bailey warning this week that markets face a high risk of a disorderly correction.
His concern is not abstract. Valuations are stretched, meaning markets are priced far above what fundamentals justify, and that correction is overdue.
The danger is amplified by debt. Bailey thinks too much borrowed money is being used to buy shares through Exchange Traded Funds and hedge funds, and so when sentiment turns, losses might intensify.
Concentration makes this risk worse. Seven giant tech companies, above all those tied to frontier AI, now dominate valuations while cross-investment between AI models and data centres ties their fates together.
Bond markets are already in panic and stock markets are swinging sharply, and the upside for staying exposed is shrinking.
Bailey's message, echoed here, is to head for safety while you can. Choosing the safest option for your finances is not pessimism but prudence when stretched valuations, debt and concentration collide in the most dangerous month.
History shows October does not forgive stretched markets when debt and concentration have already made them fragile.
This is the audio version:
This is the transcript:
Could stock markets crash this October? October has an extraordinary history of sudden financial crashes. The crashes of 1929 and 1987 reached their decisive moments in October of those years, and some of the most violent falls of the 2008 crisis also happened in October. Now this does not mean that October must mean we'll have a financial crash, but the historical pattern is striking enough to take seriously.
October has earned its reputation as the month of market crashes. Wall Street crashed decisively on the 28th and 29th of October 1929.
Black Monday happened on 19th of October 1987. I remember it well.
The Asian financial crisis produced a major sell-off on 27th October, 1997.
And the 1937 crash in the USA, which we tend to ignore now, also produced one of its sharpest falls in October.
Whilst in 2008 there were exceptionally large falls in stock market valuation during the course of October, although the crisis did, in fairness, start in September.
The numbers behind the October pattern are also remarkable. Six of the eight exceptionally large Dow Jones Index falls in the period we are looking at happened in October. The Dow fell 22.61% on 19th October, 1987. It fell 12.82% on 28th October 1929. It fell another 11.73% the following day.
October has repeatedly produced sudden and exceptionally violent market moves, and we are living at a moment when such moves are expected again, not least by people like the governor of the Bank of England.
But let's be clear, major market crashes do not always happen in October. The COVID crash reached its most violent point in March 2020. The Dot-com collapse in the UK happened in January, and reached its peak in March, and the 2025 tariff selloff came in April. So there is no rule that financial collapses must happen in October, but there is something else to note as well.
Sudden crashes are different from prolonged bear markets. Bear markets can develop gradually over months or years. The 1973 to 74 collapse ran from January 1973 to December 1974. October had nothing to do with that, but this was not a one-off moment of crisis. It was a bear market, which means that markets are in a continual sale mode and decline steadily. The Dot-com collapse unfolded over roughly three years. October looks exceptional, mainly when we consider sudden market collapses.
And overall, the worst month for stock markets is, in fact, September. September has historically been worse on average for markets, but October has sometimes been catastrophic, and let's note in that context that September 2026 has started badly. Bond markets are in a state of panic, and the stock market is fluctuating. We do look as though we are in a period of market uncertainty. October could be catastrophic as a result, and there are those who are warning that this could be the case.
Andrew Bailey at the Bank of England has said in recent days that the risk of a disorderly correction within financial markets is high. He told the G20 that vulnerabilities in the financial system remain ‘elevated', to use his precise words. And he has suggested that asset valuations remain high despite increasing financing costs. The result is markets are vulnerable to a potentially disorderly correction spreading right across international borders.
He highlighted the risk in government bond markets.
He also stressed that there are risks in the shadow banking system.
And he has suggested that there are stretched asset valuations, as he put it. And all of these things create fragility.
And let's be clear what he means by stretched asset valuations. He means that things look overpriced, to be blunt: that stock markets are too high and that a correction is inevitable. And in this context, Bailey specifically highlighted stretched valuations in AI-related investments and the risk within that technology itself.
He stressed that he sees a risk from what he called frontier AI: the risk that AI could go rogue, in other words. Those are my words and not his, but his conclusion was very straightforward: we therefore cannot be complacent.
And I think he was sending a signal to the world. He thinks the world is being complacent. And for once I find myself in complete agreement with Andrew Bailey, which is a position I find uncomfortable because he's right.
He's right to warn that markets are overvalued. He's also right to warn about another issue. He is warning that too much money is being borrowed to buy shares. This is happening in ETF funds, as they're called. It's also happening in other types of savings vehicles. And hedge funds are also borrowing money to buy shares. And this creates significant connections between risk in equity markets and banks.
The fear is that this approach can reinforce market profits at this moment where markets might still be rising. But where money is borrowed to buy shares, the risk is that losses are intensified when sentiment turns because everyone panics to offload their shares at that moment because they have to repay their debt, which they won't be able to do if they don't get out at the peak of the market, and so the risk of a major downturn is exacerbated as a consequence. And this is the risk that Andrew Bailey is talking about.
The concern now is that high valuations, high levels of debt and high levels of market concentration in the shares of a few companies are now interacting with each other, and add into that the confusing cross-investment between AI companies and data centre providers, and there is a risk that these too could amplify a market correction.
In other words, if one of the big seven tech companies does have a genuine fall in value for a particular reason, other companies who have invested in it, but who themselves are otherwise still trading profitably, could see their share prices marked down heavily. We could therefore get this disastrous consequence of a collapsing market.
So let's be clear, October does have an extraordinary history of sudden financial crashes, and Andrew Bailey is right to identify that conditions exist that could make another correction possible right now. Asset valuations, artificial intelligence and debt funding of share-based investment could all magnify any initial fall in the value of shares and turn it into a run. The result could be a large shock, or several shocks added together, and this could trigger multiple vulnerabilities simultaneously, knocking over into the banking system requiring government bailouts.
October does not predict a crash, but this October will arrive with some deeply worrying conditions already in place, if we haven't seen the crisis in September. It might be time to find your financial tin helmet in that case, or in other words, head for safety. That's what I think Andrew Bailey was saying to people, and for once I agree with him. At this moment, if you've got a choice with regard to your finances, go for the safest option you can find. Andrew Bailey is saying the upside risk is no longer worth taking. The downside risk is real. We have to take that seriously. October may be interesting.
That's what I think. What do you think? You might not agree. You might still believe that markets will weather this thing out. If that's the case, let us have your opinion. Please do let us know your opinion in the poll. Please do like and share this video. Please do subscribe to the channel and ring that bell so you get told when we make new videos. And if you'd like to buy Tom and me a coffee, thank you very much. That would be much appreciated, and there is a link to enable you to do that just down below.
Poll
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
Has the current US/UK stock market obsession with AI and LLM overlooked the potential impact on investments of the increasing release by China of free to use AI products? Could all these vastly overvalued Silicon Valley tech companies be severely devalued as cost pressures lead to an uptake in free to use Chinese versions? There may be security/privacy issues but are the Chinese versions anymore invasive than those produced by the likes of Thiel, Musk and Altman?
I share your view on this. I see no greater, and maybe less, risk in using Chinese AI than in US-based versions.
At the moment stock market buyers don’t seem worried, though they should be. If they were worried they would be heading for safety, which means they would be buying gilts. But gilts are relatively low priced. As Chuck Prince, the former CEO of Citigroup, famously said said “as long as the music is playing, you’ve got to get up and dance. We’re still dancing”.
If I really knew where to put my money I would be on my yacht in the South of France…………….
But the idea of borrowing to buy shares is to put it bluntly madness
I know markets and politics don’t necessarily work in tandem but the midterms might play a part in market volatility. The orange vegetable will have his militia out intimidating voters and Iran may well decide to ratchet up the pressure to turn the screw. I’ve half my funds already in the money market, the rest will follow before October.
The question is … what will western governments do if the orange baboon simply cancels the mid-terms or declares the results as caused by foreign interference and invalid ? Is there anything he could do which would wake them up and actually mean sanctions being imposed on the US ? Or will they just cower with some stronger words perhaps from Canada and that’s all ?
Good questions.
There are no answers.
There are probably no preparations either.
Might this pressing market+debt market matter be validly described as an example of recent and current [Western?] goverments accepting/promoting that problems resulting from private “big money” speculation/gambling are “paid for” by the rest of the “not big money” community?
I’m curious how this affects people who have been forced into DC pension schemes, which is now a good proportion of the working public. Of course there are “less risky” funds one can diversify into but a lot of people’s retirements are potentially going to be exposed. One wonders whether the UK government will sit idly by and watch many people tipped into a much more parsimonious situation because of this mania? Are we in too big to fail territory now?
Many people could lose a lot of money unless they play it safe now.
Worth also noting the demographic decline. Fertility rates are drastically reducing world wide. In Singapore the fertility rate in 2025 was 0.87, a further decline on the previous year. It’s falling off a cliff. This means that there will be fewer than half the children born in this generation than the previous one. While Singapore is an outlier, the same pattern applies worldwide.
With fewer children born over time there will be fewer people buying in to DC pensions (and more people taking out). With fewer purchases and more sales prices will inevitably fall, perhaps dramatically.
I’m afraid DC pension schemes are effectively a Ponzi scheme. Those who got in, and out, early, the Boomers, may have done well. Those who come later, Gen Z, not so well.
I agree with you about DC pensions.
We live in a political and economic system that uses deception, propaganda, and financial manipulation to maintain the illusion of prosperity rather than solving the root problems. This in addition to war and the impacts of climate change. Not going to end well, is it?
I’ve been pondering AI over investment and a possible crash. In similar previous episodes of over investment, e.g. railway mania in the 1840’s and the dot com boom, there was physical infrastructure left, rail tracks and fibre optic cables. These could be bought cheaply and productively reused. This is unlikely to be the case with AI.
Although the physical chips, CPUs, GPUs etc will not be destroyed they cannot be reused. They have a very limited lifetime and a lot of built in obsolescence. That’s because, after a couple of years, it is cheaper to simply discard old chips because new ones are so much better. And AI companies know this.
If an AI crash starts, which seems likely, the companies know that their investments have a limited lifetime. They will not be able to simply start them up again a couple of years later when demand may return. If they don’t use it now it becomes worthless. This different to previous episodes. So, when a crash starts, and knowing this, companies will be desperate to unload their investments quickly, before they become worthless. This may create an intense fire sale as companies unload. Because of the specific characteristics of semiconductor development this is likely to be much more intense than previous crashes.
Stand by for a very rapid crash.
Much to agree with
As has been alluded to on this blog many times, the stock markets are no longer traders waving pieces of paper on the floor. They are heavily computerised, meaning a crash can take place in a matter of seconds, much quicker than any human can react. Reallocating to cash seems sensible at this time.
Richard, an idea for you to consider. It cannot be a simple post from me since the replies might be too much of an invasion of privacy, but the following idea for an anonymous poll came to me after watching the video on YT.
A poll of your readers covering the following questions, or a version thereof:
With regard to the possibility of a market crash in the near future. Which of the following answers best describes you.
A) To the best of my knowledge, my current and future income is likely to be seriously affected by a crash and I am very worried about my future financial security.
B) To the best of my knowledge my current and future income is largely protected against the effects of a crash and I expect my future financial security will be similarly unaffected to any great extent.
C) My answer to (B) is ‘YES” but I also have access to significant funds which I expect to be largely unaffected by a crash which could allow me to buy shares at ‘bargain basement ‘ prices and thus make a significant financial gain during the recovery from any imminent market crash (answer ‘yes’ to this even if you have such funds available but think you may not use them in this way).
Whether such a poll is relevant for your blog is for you to decide, but it might provide an interesting insight into the kind of people who are your regular readers.
Regards,
Kit.
Thoughts, anyone?
I think it would be helpful to break (C) down a bit further to get a feel of the extent to which people with funds to invest have had enough of the casino and would like to see their money used to do some good.
“The purpose of making an investment is the purchase of an income” is an old saying and it’s too often forgotten in a “dash for capital growth” which is entirely ephemeral.
I often use ancient investment trust companies (F&C, Alliance Witan, Merchants, et.al) with clients because they do this for a low fee. Several have increased their dividends each and every year for 50 years and more. Many IFA’s simply ignore these funds as they have never paid commission, unlike unit trusts used to do. You don’t get that income from a leveraged ETF, but a steady stream of dividends, interest payments from bonds and cash, even guaranteed insured annuities, are what most people want later in life, in my long experience. DC pensions can do this if organised properly, but not the Ponzi-like “workplace schemes” prevalent today.
Many are looking through the wrong end of the telescope.
I so agree with that conclusion