Mean reversion of market valuations will be one of the primary headwinds or tailwinds affecting your portfolio over the long-term. In this article I look at the FTSE 250’s cyclically adjusted PE ratio (CAPE), and whether it’s likely to help or hinder your portfolio over the coming decade.
Updating the FTSE 250’s CAPE ratio for 2020
Following on from my recent update of the FTSE 100’s CAPE ratio, it’s now time to look at the FTSE 250.
The process is the same as it was for the FTSE 100:
- Go to the FT World Markets at a Glance report (PDF)
- Note the index’s price and PE ratio
- Go to the Office for National Statistics (ONS) and get CPI inflation for last year
- Add that to your archive of FTSE 250 and CPI data
- Calculate CAPE
I did that on January 1st and got the following results:
- FTSE 250 price = 21,883
- FTSE 250 PE ratio = 22.57
- Average CPI index for 2019 = 107.8
We can get the FTSE 250’s 2019 earnings like so:
FTSE 250 earnings = price / PE = 21883 / 22.57 = 970
To calculate CAPE, we also need the FTSE 250’s inflation-adjusted (real) earnings over the last ten years.
|Year||Nominal earnings||CPI index||Real earnings|
Table 1. FTSE 250 real earnings over ten years
The FTSE 250’s ten-year average real earnings to 2019 is 901 index points, and with that we can calculate CAPE:
CAPE = 21883 / 901 = 24.3
As always, this tells us nothing unless we compare the current CAPE ratio to its long-term average.
If CAPE is below average then its cheap, and if it’s above average then it’s expensive. But exactly how cheap or expensive depends on how far the ratio is above or below its long-term average.
Is the FTSE 250 expensive, cheap or somewhere in between?
To help answer that, here’s the rainbow chart from the top of this article:
The chart shows where the FTSE 250 would have been over the last few decades if its CAPE ratio had been at various levels.
For example, if the FTSE 250’s CAPE ratio had been above 39 for the last 30 years, the index would have travelled along the red band at the top of the chart and would currently sit somewhere in the high 30,000s.
Alternatively, if the FTSE 250’s CAPE ratio had stayed below 14 for the last 30 or so years, then the index would have stayed in the dark green zone at the bottom of the rainbow and would have a value today of less than 12,000.
The black line shows where the index actually was in the past.
The central yellow band on that chart (what I call the normal or fair value band) covers a range of CAPE values from 19 to 28, and the actual long-term average CAPE value for the FTSE 250 since 1993 is 23.6.
However, a period of almost 30 years is actually quite short in terms of stock market history, so this value of 23.6 is unlikely to be equal to CAPE’s “true” long-term average.
So using a bit of judgement, I think an average CAPE of 22 is a better fit for the data and is the value I’ll use in the rest of my calculations (I could have easily used 24 and in reality it probably doesn’t make much difference).
Of course, CAPE isn’t always 22, or 23.6 or 24.3. It moves around with the stock market and history shows that it tends to stay within a range of about double to half its long-term average.
With the average set at 22, that gives us an expected range of anywhere between 11 and 44, and that’s precisely the range covered by the rainbow chart.
Turning back to the FTSE 250’s current valuation, we know that:
- The FTSE 250’s CAPE is currently 24.3
- The FTSE 250’s estimated long-term average CAPE is 22
So according to CAPE the FTSE 250 is slightly expensive, although well within the normal or fair value range of 19 to 28.
This assessment is backed up somewhat by the FTSE 250’s dividend yield. On January 1st it was 2.95% which is surprisingly close to its long-term average of 2.96%, suggesting that the index’s valuation is also close to average.
This means that valuation mean reversion (i.e. CAPE’s tendency to periodically return to its average value) is unlikely to provide much of a headwind or tailwind for the FTSE 250 over the coming decade, because the index’s valuation is already very close to that long-term mean.
And speaking of the coming decade, it’s time for a stock market “forecast”.
Ten-year FTSE 250 forecast
I’ll base this “forecast” on the same assumptions I used for my recent FTSE 100 forecast:
- Over the next ten years inflation stays close to the post-2000 average of 2%
- Over the next ten years the FTSE 250’s average real earnings growth stays close to its post-2000 average of 5%
- Dividends grow at the same rate as earnings
- Ten years from now in 2030 the FTSE 250’s CAPE ratio will be at its long-term average of 22
Under those assumptions, by 2030 we’ll have:
- CPI inflation index at 131.6
- FTSE 250 average real earnings of 1,904
- A fair value FTSE 250 price of 41,900
That’s a “forecast” increase in the FTSE 250 of 91% over the next decade and, assuming dividend growth in line with earnings growth, a total return of 9.9% annualised.
That is a very attractive forecast rate of return, although of course reality will differ to a greater or lesser extent.
As I said earlier, the FTSE 250’s CAPE ratio can reasonably be expected to range from 11 to 44, or thereabouts, in the decades ahead.
So if we reach those extreme highs or lows in 2030 then we could see:
- 2030 Depression valuation (CAPE at 11):
- FTSE 250 at 20,900 in 2030 (4% capital loss from today)
- 2030 Bubble valuation (CAPE at 44):
- FTSE 250 at 83,800 in 2030 (283% capital gain from today)
Those outcomes are unlikely, but not impossible.
On balance though, and given that we have no idea what will actually happen, the sensible thing to do is to expect average earnings and dividend growth and an average CAPE value of 22.
That leaves us with a fair value estimate for the FTSE 250 of almost 42,000 in 2030, and that leaves me mildly optimistic for returns over the next decade.
Could you please elaborate on why you suggest growth rate + dividend yield should be above 10%. Are you suggesting the share price should increase by this amount each year for you to invest?
I believe you mentioned something similar in a recent post of yours by Master investor.
John Kingham says
Hi Liam, I don’t think I mentioned a 10% target in this FTSE 250 article, although I do use 10% as a very ballpark target rate of return.
The yield plus growth model is a very simple way to estimate the return from a given stock. It assumes that:
a) the dividend continues to grow at its historic rate
b) the dividend yield doesn’t change
If those hold true then your return will equal the dividend yield plus the dividend growth rate.
For example, if a stock has a 4% yield and grows its dividend by 6% each year, then you’ll get a 4% return from the dividend each year and a 6% boost from capital gains (because if the dividend increases by 6% and if the yield stays the same, then the share price must also have gone up 6%).
So your return is 4% + 6% which is 10%.
Some people get confused with percentages, so here’s another example.
A company pays a dividend of 10p and can grow that dividend by 10% each year. The current share price is 100p.
In year 1 you’ll get a 10p dividend, which is 10% of 100p so your dividend income return is 10%. The dividend goes up by 10% next year to 11p. The yield stays the same so the share price goes up to 110p, leaving the yield at 10%. So you now have a 10p capital gain as well as a 10p dividend income, so your total return for the year is 20p, or 20% of your starting investment.
This 20% comes from a 10% yield and a 10% dividend growth rate.
Now, obviously none of this will hold true in the real world which is full of uncertainty, but then again that’s true of any valuation method, no matter how complex.
Personally I value simplicity, so I think the yield plus growth model is a useful if very rough valuation tool which can be applied (with a pinch of salt) to relatively mature and stable businesses.
Thank you John! Apologies for writing on the incorrect blog post.
That was well explained. Do you have any further reference material I can read up on? Is it a simple “Yield + Growth Model” Google search?
John Kingham says
Your best bet is to search for the Gordon Growth Model, because yield plus growth is basically derived from that.
Gordon’s Formula = D / (R – G)
Where D = next year’s dividend, R = the required rate of return and G = a constant dividend growth rate.
So using the example I gave before:
D = 10p
G = 6%
R = 10% (this is my required rate of return, although higher is obviously better)
So the price required to give me my 10% return is:
price = 10p / (10% – 6%) = 250p
So for the expected return to be 10% on this stock, the price would need to be 250p.
And lo and behold, at 250p the dividend of 10p gives a yield of 4%, which also matches the yield plus growth answer (4% yield plus 6% growth = 10% expected return).
Wonderful; great explanation. Thank you very much.
Thanks for the analysis John, I love geeking out on this sort of thing.
I wondered if your analysis would change your investment approach if the outcome was that the CAPE was at a more extreme point? For example, if the assumption of mean reversion is correct would you rethink what constitutes a bargain at either extreme?
John Kingham says
Hi MR, that’s an interesting question. I don’t think extreme CAPE levels would change my definition of a bargain, but it might change my cash weighting if there are very few bargains to be had.
For me, a bargain stock is one that has an estimated expected return which is better than the market’s. However, if the market’s CAPE is extremely high then the market’s expected return could be low single digit or even negative, even over periods as long as ten years.
Obviously I don’t want low single digit returns, even if that is better than the market, so at extreme valuations I might hold more cash if bargains are few and far between.
However, I’m not 100% sure what I’ll do if we ever see CAPE at extreme highs again. I might hold more cash, but I might just keep my cash levels below 5% which is my usual target. And even if I do hold more cash, it wouldn’t be an extreme allocation of say 50% or more. I would be surprised if I ever went above 25% cash, regardless of the market’s valuation.
Remember, even in the dot-com bubble there were lots of “old economy” businesses trading at bargain prices, so my assumption is that there will always be some corner of that market that is out of favour.
Investing Worldwide, Naturally says
Hi again John,
another superfine article. I was wondering, would you like to do a similar evaluation for the S&P, so people can see exactly how overvalued it is right now, and how poor the prospects for ongoing returns are? I’ve seen examples suggesting the next decade in the US could see zero (or worse!) average annual returns from figures as luminous as Jack Bogle, and John Hussman, but nothing as succinct and well explained as your stuff.
John Kingham says
That was exactly my plan. I’m hoping to get an S&P valuation article up next week, and that’s still likely as long as the purchase review I’m working on goes reasonably smoothly.
Investing Worldwide, Naturally says
Fantastic! Looking forward to it. We’ve had a pretty wild ride the last week!