On Wednesday the FTSE 100 closed just above 6301 points, continuing a near-20 year tradition of failing to get much above that level.
For some investors that’s depressing news; after all an investment going nowhere for 20 years is not exactly a stunning success.
But for investors who are thinking about returns over the next 20 years rather than the last 20 years, this lengthy sideways market is in fact better news than it might at first appear.
Sideways markets usually lead to cheaper markets
The chart below shows the basic story:
The last 20 years can be split into two main parts:
1) Party Time – Where investors thought the economy was going to grow massively, thanks to the internet and the World Wide Web.
2) The Hangover – Where investors realised that the FTSE 100’s profits and dividends were in no way sufficient to justify its dot-com peak of 7,000 and so the market collapsed. An unwillingness to relinquish the dream of 20% per annum growth led to the 2003-2009 credit-driven boom and bust and then the 2009-2016 stimulation-driven boom and bust.
If humans were rational (which they’re not, either individually or in groups) then the FTSE 100’s price history would look something like the blue “Rational Path”.
That Rational Path shows the market price increasing gradually, in line with the gradual increase in average profits and dividends generated by the 100 companies which make up the FTSE 100.
But humans are not rational and so rather than a smooth increase in market value we saw multiple greed-driven bubbles followed by multiple fear-driven crashes.
However, after 20 years of steady growth the FTSE 100’s long-term average profits and dividends have increased to the point where today’s FTSE 100 value of 6,300 is not only not expensive, but is instead positively cheap.
Growth, mean reversion and fair value
Trees do not grow to the sky and neither do stock market prices (or house prices for that matter).
In the long-run trees, stock markets and house prices vary around some long-term average. In the case of stock markets and house prices that average is the price to earnings ratio.
In the specific case of equity indices like the FTSE 100, my ratio of choice is the cyclically adjusted PE ratio (CAPE), which is the ratio between the current index price and its ten-year inflation-adjusted average earnings.
The long-term average CAPE ratio for the FTSE 100 and most other major indices is somewhere in the mid-teens (16 is a commonly used figure).
Today the FTSE 100’s CAPE ratio is 12.2, which is of course well below that average of 16.
That’s fairly normal though because most of the time the FTSE 100’s CAPE is above or below average rather than exactly on it, largely because investors are usually either somewhat optimistic or somewhat pessimistic.
So where are we today in terms of historic norms?
Have a look at the chart below, which shows how the FTSE 100’s CAPE ratio has varied relative to its long-run average over the last 30 years or so.
It covers the lead up to the dot-com bubble, through to the near-depression of the great financial crisis and finally on to where we are today.
You can think of the yellow line in the middle of the rainbow as a more accurate version of the Rational Path shown in the previous FTSE 100 price chart.
As the index’s price moves into the red or green zones it departs from rational pricing and moves further into the world of emotionally driven pricing.
In other words, in a rational world:
- In 1999 the FTSE 100 “should” have been at 3,300 rather than 7,000
- In 2007 the FTSE 100 “should” have been at 5,500 rather than 6,500
- Today, the FTSE 100 “should” be at 8,300 rather than 6,300
If the FTSE 100 were “rationally” priced today at 8,300 it would have a dividend yield of 3% rather than its current 4%, and 3% is an entirely normal yield and not excessively low by any stretch of the imagination (unlike in 1999 when the yield was about 2%).
Since the market is currently below fair value I expect returns over the next five, ten or 20 years to be somewhat better than usual, i.e. better than about 7% per year, which is the long-run average rate of return on UK equities.
However, if future returns are likely to be better than usual, why don’t investors pile into the FTSE 100 and drive the price up to where the current price and future expected returns are “normal”?
There is always a reason, and the reason is usually either fear or greed. In this case it’s fear.
An uncovered dividend is holding the FTSE 100 back
There are a million and one reasons why investors would be fearful; the impending Brexit referendum chief among them.
But there is another problem: For the first time in at least 30 years the FTSE 100’s dividend is not covered by its earnings.
As you might have guessed, an uncovered dividend is not good. Unless those earnings recover fairly quickly there is a very good chance that the FTSE 100’s dividend will decline over the next year or so.
I suspect that this risk of a falling dividend, combined with a high PE ratio which comes from having very low current earnings (the standard one-year PE ratio for the FTSE 100 is currently above 30), is keeping the market closer to 6,300 than 8,300.
As always, how this situation will be resolved is unknowable, but:
One way or another the FTSE 100’s CAPE ratio will revert back to its historic average of 16; of that I have no doubt.
It will happen either because the FTSE 100’s price goes up or because its earnings stay low for several years, dragging down its cyclically adjusted earnings.
History suggests the former, and if that suggestion is right then the FTSE 100 today is much better value than it has been throughout most of the past 30 years.
I absolutely agree that the CAPE ratio will most probably revert to the norm. What is worrying, however, is that the shape of the ‘rainbow’ has gradually changed over the last few years. We can only hope that it doesn’t begin to take on a true rainbow shape!
Hi Chris, yes I thought that the first time somebody called it a rainbow! (and then I subsequently borrowed the name).
If you look at the longer data for the S&P500 there are periods in which the cyclically adjusted earnings go nowhere, or even down, over a decade or so, which means that an upward trajectory always and forever isn’t a given.
To me that just makes valuation even more important; if cyclically adjusted earnings aren’t going anywhere then a larger bulk of returns will come from dividends, which is fine by me as I tend to be in high yield stocks anyway.
Couple of points :-
Cannot get my head around the y scale in the charts !
This investor has often complained bitterly about the FTSE100 post-millenium nowhere market, having never really held much above the 6930 of Dec 1999, esp when compared to the S&P500.
Another investor with access to better data than I, has pointed out that the total return figures UK v US since the turn of the millenium are quite similar, due to higher UK Dividend Yield compensating for lack of UK indices growth. Found this surprising.
Do not share complete confidence in CAPE, due inter-alia the reservations expressed in the article re dividend cover.
Whereas US CAPE figures are today distorted upwards by 2008 ish decline in earnings still being in the sample set; we have the reverse situation in UK with today’s ‘below trend’ earnings not yet playing a major part in sample set. It may be that all will be well and Banks, Miners, Oil, etc, earnings will recover. Who knows?
All Best
Hi Magneto, your points about the fallibility of CAPE are spot on. Anything we say about the future is an educated guess, to some extent, and CAPE falls a long way short of being an accurate crystal ball. But it is one of the better tools out there, along with Tobin’s q, which measures the current value of the market against the replacement value of the underlying companies.
As a Dividend Investor have you given any thought to using Market Dividend Yield(s) as a guide?
Yes, I do look at the market yield as well, although I don’t explicitly use it for valuation or forecasting purposes. I would say the current yield of about 4% backs up the idea that the market is currently somewhat cheap, although that does of course depend on whether or not that dividend will grow, or even be sustained, over the coming years.
Thank you ver much for sharing your insight on FTSE 100..The analysis was very interesting..FTSE 100’s uncovered dividend seems to be a big problem and with EU refendrum this month, I think it will be a tough time for FTSE 100
Hi Sukanya, we shall soon see, but personally I have absolutely no idea what will happen (and, very probably, neither does anybody else).
This is a very interesting outlook for the UK stocks. I had been looking for data on FTSE 100, and your post summarizes the developments over the past 20 – 30 years quite nicely ( plus I enjoyed your FTSE 100 spreadsheet under “Free Resources”). I really liked the fact that the FTSE 100 had such a high starting yield.
But now, I am a little surprised to see that FTSE 100 has not been able to collectively grow EPS for quite some time. Without earnings growth, the best an investor can do is collect that 4% starting dividend ( and hope they don’t cut it too much).
Perhaps this decreasing EPS is due to a high allocation to cyclical commodity companies which currently are down on their luck ( BP, RD, BHP etc)? Or perhaps during the years 2005 – 2006 the earnings from the UK Financial sector were much higher than today due to the financial boom going on?
Either way, if earnings per share can grow over time, and investment in FTSE 100 could deliver good returns to shareholders in the future.
Thank you for your posting!
Dividend Growth Investor
Hi DGI, yes the FTSE 100 does have a very nice yield, especially when compared to US indices.
Earnings for the FTSE 100 has been very volatile this past decade thanks to the credit boom/bust and subsequent stimulus boom/bust. Part of that was the impact of boom/bust in banks and commodity stocks, but both of those sectors are now a smaller part of the index so hopefully their future influence is decreased too.
Also I hope that from here on out things might settle back to the historically normal earnings and dividend growth rate of inflation +2% or thereabouts, but time will tell as usual.
Thanks for your comment
– John
According to the Price/book ratio the FSE 100 is not expensive (1.7) plus the ROE is good (22,8%).”It’s far better to buy a wonderful company at a fair price than a fair company at a wonderful price.” – Warren Buffett
And Aswath Damodaran thinks similarly http://gregspeicher.com/?p=770
Hi, in that case the price/book ratio agrees more or less with what I’m saying, although I don’t use that particular ratio for valuing either companies or indices. As for ROE, I don’t use that either but I would doubt it’s as high as 22.8%. That does seem rather high, unless the 100 companies in the FTSE 100 were especially highly geared, which I don’t think they are. Would be interesting to know the source of your data.
See Vanguard FTSE 100 UCITS ETF (VUKE): Characteristics (As at date 31 May 2016)
https://www.vanguard.co.uk/adviser/adv/detail/etf/overview?portId=9509&assetCode=EQUITY##overview
Thanks. So we both agree: the FTSE 100 has decent profitability and isn’t expensive, although it does face a lot of uncertainty at the moment.
See Vanguard FTSE100 ETF https://www.institutional.vanguard.co.uk/portal/site/institutional/uk/en/detail/etf/overview?portId=9509&assetCode=EQUITY##overview
It also demonstrates the importance of reinvested dividends. Anyone invested over that period and putting dividends back in will be sitting on a substantial gain.
Hi John, Now that UK politics and economics is unstable due to brexit, do you still believe that this pattern will be followed or will it takes years to return to this pattern.
Hi Mudra, generally I don’t like to speculate about future profits, but it seems likely to me that the cyclically adjusted earnings upon which CAPE is based are going to come down over the next few years. That’s because we had two profit bubbles during that period; the first one from banks and the credit boom in general and the second one from commodity stocks and the post credit-bust global central bank stimulus boom.
Both of those booms have now petered out and we are left, perhaps, with a period of normal post-boom depressed profits. So even if the FTSE 100 stays at say 6000, CAPE could well head back towards its mean mid-teen value over the next few years, but I don’t think it will get all the way back to 16 or thereabouts without an increase in price as well.
Now, Price/earnings ratio is 41.5 x.
VERY EXPENSIVE?
Hi Gustavo, that’s a good point. One of the problems with the standard PE ratio is that it only looks at earnings from the last year. In the case of the FTSE 100, earnings over the last year have been very poor thanks to losses at banks and commodity companies. Those weak earnings are why the index’s PE ratio is currently very high.
However, it is unlikely that the earnings from banks and commodity companies will continue to be as bad as they are today, and overall I expect the FTSE 100’s earnings to move back towards more normal levels over the next year or two (although I can’t be certain as I don’t have a crystal ball).
If earnings do normalise then what looks like a very expensive PE ratio today will be a normal PE ratio in a year or two, even if the market stays at its current level.
That’s why the CAPE ratio is far superior to the PE ratio, as it uses ten-year average earnings which smooths out the ups and downs that earnings go through from one year to the next.
So the short answer is no, I don’t think the FTSE 100 is expensive.
There’s another obvious sense in which the market is cheap and that is that people are largely not seeing the real value because of the endemic use of nominal data like the FTSE index series in commentary and media coverage which have the effect of obscuring the relative dearness or affordability of equities. An index re-calculated in real terms to reflect the diminishing value of money over time would allow people to see that when the market hits 6,900 today it is nowhere near as expensive for the investor as when it also hit 6,900 back in 1999 or 2007. The persistent use of the nominal figures is deeply misleading because it doesn’t show just how much more over-priced the market has been at certain points in the past than it is right now.
HI Nick, I would largely agree with that. However, most people naturally think in nominal terms and so I don’t think the dominance of nominal data is going to change any time soon.
I have written about the FTSE 100 before in real terms, but not for a few years I think (I can’t recall exactly when it was so it must have been a while ago!). Perhaps with the spectre of inflation rearing its head again (notably in the recent Marmite price war between Tesco and Unilever) I should do another market review focusing on inflation-adjusted earnings, dividends and prices.
Thank you John. I wonder if you could clarify your graph a little? What is the basis of the graph bands trending inexoribly upwards? What would happen if you flattened the trajectory? I note the bands have flattened out after 2012 ish, so there is obviously some calculation involved here that I am not aware of.
And I would love to see a graph of FTSE100 P/Es and Yields going back to 2000 (I know you prefer CAPE but….). Can’t seem to find them anywhere.
I’m delighted to say that my Portfolio has gained 75% this year.. currently yielding 5%….good calls on Commodities and Brexit behind this. So not worried about a pull back.
Hi Murai, those bands represent fixed multiples of the index’s cyclically adjusted earnings (inflation adjusted ten-year average). They trend upwards because cyclically adjusted earnings trend upwards and they have flattened out recently because inflation adjusted earnings have flattened out. Long-term PE and yield charts would be interesting, so I will include those perhaps next time if I remember.
Well done with your returns and good luck for the future.
Unfortunately there’s an apples&oranges variable in all this that you’re not taking into account: companies have been going into debt in order to buy back their own stocks, which obviously has the effect of artificially inflating earnings per share. If this were factored out then the earnings chart would flatline. If you think about it, it’s highly unlikely that earnings genuinely spiked up like that across the 2000s, let alone in 2010/11.
Remember this latter was the period in which the big FTSE100 retailers were all posting losses, as were the banks suffering from the effects of ZIRP.
If I had to bet, I’d say the CAPE’s just outright invalid and we’re better off looking at a non-adjusted P/E. Given that earnings have been grossly goosed, these sky-high raw P/Es are indicative of EXTREME overpricing.
FTSE 3000 in the next crash? Not impossible.
I always thought it was a bit odd that aggregate earnings jumped up so much after the financial crisis. The CAPE approach to dealing with potentially misleading earnings in any given year is to look at the average of earnings over decade (or whatever long-term period appeals to you), which are far more likely to be representative of “true” earnings and perhaps even cash flows.
I would be very reluctant to toss CAPE into the bin just yet. It has over a century of pretty solid data to back it up as one of the better metrics, so I’d have to see it fail repeatedly over a multi-decade period before I’d label it as invalid.
Having said that, I agree with you that FTSE 3000 is not impossible, but barring nuclear war it would probably turn out to be the buying opportunity of the century.
Thanks for your response John.
Let’s hope 2017 isn’t *that* year.
It worries me that earnings fell off a cliff in Q3/Q4 in spite of all the stock buybacks, but we shall see. The prospect of firms like Sainsbury’s, Vodafone, even Barclays becoming genuine value investments is obviously appealing.
Of course all three of those low price-to-book, lowish P/E, dividend-paying stocks have done horribly over the past two years: the FTSE is after all being buoyed by all the dollar-delimited multinationals.
CAPE has its detractors, but I think it’s probably as good a measure as any.
If dividends are uncovered as you say, then it is not necessarily bearish. I could argue that instead of dividends going down, it means that earnings will go up. I read an interesting article some time ago: “Surprise! Higher Dividends = Higer Earnings Growth”
https://www.researchaffiliates.com/documents/FAJ_Jan_Feb_2003_Surprise_Higher_Dividends_Higher_Earnings_Growth.pdf
Also, a high PE does not necessarily imply an over-valuation. I remember that it was in about 1993 that Seth Klarman said that the US market was at a bonkers valuation because the PE was too high. The Dow Jones was at about 3300 at the time. In retrospect, not a great call by Klarman, to put it politely.
What I think he failed to consider is that was a year of recession, when reported earnings bombed, but subsequently recovered.
Personally, I think UK markets are at fair value. Growth is not necessarily inevitable, and we may still be paying the price for 2008 for many years to come.
I certainly prefer your assessment a lot more, though 🙂
US markets look frothy to me.
Thanks for a thought-proviking article, John.
Hi Mark, I think FTSE 100 earnings are are likely to rebound next year, but to what I have no idea. Perhaps something like the 10yr average; who knows.
When I wrote this post the FTSE 100 was at 6,300, so it was cheap according to CAPE, but at today’s 7,200 it’s close enough to the long-run average of developed markets that I’d call it normal or fair value now.
The question now is whether this currency-based bull run is enough to set off a real bull market. I would be surprised if it was because we aren’t exactly bathed in positive economic news, which is what bull markets usually require. But you never know, perhaps the FTSE 100 will be at 10,000 by the end of the decade?
I will certainly be glad to see the back of the 7,000 barrier which it’s been stuck below these past 17 years!