**Over the last 20 years, the performance gap between the FTSE 250 and FTSE 100 has been enormous.**

While the FTSE 100 has spent most of the last 20 years failing to beat its dot-com peak, the FTSE 250 has raced ahead. In fact it’s now almost three times higher than it was during the tech bubble.

So what does that say about the FTSE 250’s valuation today? Is the FTSE 250 in some kind of runaway bubble fuelled by quantitative easing and super-low interest rates?

## The FTSE 250 is booming, but what about its earnings?

The FTSE 250’s price is booming, but that price doesn’t exist in a vacuum. It is intimately tied, in one way or another, to the earnings and dividends of the 250 companies which make up the index.

That’s why the price to earnings ratio is a key metric for most investors, although I much prefer the cyclically adjusted PE ratio, or CAPE ratio for short.

The chart above shows that the FTSE 250’s price has risen at an impressive rate, tripling over the last decade. At first glance this does look like another bubble, one of many according to some.

But if the FTSE 250’s earnings have kept pace with its price, then perhaps the massive run-up is justified.

To get the ball rolling, here’s a chart of the FTSE 250’s earnings and dividends over the last 25 years or so:

The chart above clearly shows a long-term trend of earnings and dividend growth for the FTSE 250, driven by inflation and real economic growth.

But can those increasing earnings and dividends justify the index’s massive price increase? Let’s have a look.

At the end of the dot-com boom in 1999 the FTSE 250 generated earnings equal to 284 index points. That figure is based on its price at the time of 6,387 and its PE ratio of 22.5.

By the end of 2017 its earnings had grown to 1,076 points, based on its price of 20,064 and PE ratio of 18.65.

**So although the FTSE 250 has gone up more than 200% since the peak of the dot-com boom, its earnings have gone up by almost 300%.**

According to the PE ratio, the FTSE 250 is less expensive today than it was in 1999, despite its much higher price. In fact, a PE ratio of 18.7 is not unusually high and is lower than the index’s average PE over the last 25 years.

**The FTSE 250’s PE ratio is 18.7 which is below the average of the last 25 years, suggesting the mid-cap index is slightly cheap**

That’s a pretty big chink in the idea that the FTSE 250’s in a bubble, but a bigger chink comes from its CAPE ratio.

## The FTSE 250’s CAPE ratio shows no signs of a bubble

Earnings can be volatile from one year to the next, even for an in index like the FTSE 250. CAPE gets around this by using ten-year average earnings, adjusted for inflation.

These smoothed earnings give the CAPE ratio a much more stable base, making it a much more reliable indicator of value.

As for using the CAPE ratio to value market indices, **history suggests that CAPE can swing between about half and double its long-term average value.**

The further CAPE goes above average the more expensive the market is, and further CAPE falls below average the cheaper the market is.

This applies to the FTSE 250 too, so all we need to do is look at its average CAPE ratio to see whether the market is closer to being expensive or cheap.

**Over the past 25 years the FTSE 250’s CAPE ratio has ranged between 13 and 32, with an average value of 23.**

However, to err on the side of caution I use a more conservative average figure of 20 because the last 25 years included two major booms (dot-com and housing/banking).

With an average CAPE of 20, the likely range for values falls between 10 at the cheap end and 40 at the expensive end.

The chart below shows this range of likely CAPE ratios for the FTSE 250 over the past 25 years, as well as where the index sits today within that range:

In the rainbow zone, red indicates expensive valuations, above which lies the land of extreme bubbles, and green represents cheap valuations, below which lies the land of extreme depressions.

Interestingly enough, the late 2000s credit boom was a more expensive period for the FTSE 250 than the dot-com boom.

But enough of the past. What can the FTSE 250’s current CAPE tell us?

**With a current CAPE ratio of 25.8, the FTSE 250 is very close to where it was in the dot-com boom in terms of valuation.**

It’s about 30% above my estimate of its “fair value” CAPE of 20, so by this measure the mid-cap index is “slightly expensive”. It isn’t dangerously hot, as it was in the credit boom, but it is definitely getting warm.

**The FTSE 250’s CAPE ratio of 25.8 is 30% above “fair value” which makes it slightly expensive, but it isn’t in bubble territory**

So far we have two mildly conflicting bits of information. The below average PE ratio suggests the mid-cap index is slightly cheap while its above average CAPE ratio (a much better measure in my opinion) suggests it is slightly expensive.

To get one more perspective we can look at its dividend yield.

## The FTSE 250’s dividend yield shows no sign of a bubble

Over the last 25 years the FTSE 250’s dividend yield has mostly stayed in a range between 2.5% and 3.5%, reaching highs during recessions and market crashes and lows during economic and market booms (lending yet more evidence to the idea that it’s a good idea to be a contrarian investor):

As the chart above shows, today’s yield of 2.7% is slightly below the long-term average of 3%. Low yields are a sign of high prices, so:

**The FTSE 250’s dividend yield of 2.7% is slightly below average, suggesting the index’s valuation is slightly above average**

We now have three valuation metrics, two of which agree:

**The PE ratio**(which personally I don’t use as a valuation tool) suggests the FTSE 250 is slightly cheap**The CAPE ratio**suggests the FTSE 250 is slightly expensive**The dividend yield**suggests the FTSE 250 is slightly expensive

On that basis I’m comfortable with the idea that the FTSE 250 is probably slightly expensive.

If that’s true then the index is slightly more risky than usual because valuations are like mountaineers: The risks increase the higher up they go.

And with that dramatic conclusion in the bag, we can move on to the credibility-shredding practice of making a forecast for the future.

## FTSE 250 forecast for 2018: You may be disappointed

This forecast is based on the CAPE ratio and takes into account two main factors:

**Cyclically adjusted earnings grow**from one year to the next, most of the time, and**CAPE is mean reverting**, which means there is a gravity-like force which pulls it back towards average

In terms of earnings growth, cyclically adjusted earnings have grown by an average of 5% per year over the last 25 years, so I’ll assume that applies for 2018 as well.

Today the FTSE 250’s is at 20,064 and its CAPE ratio is 25.8, which means its cyclically adjusted earnings are equal to 776 index points.

**So 5% growth would see the FTSE 250’s cyclically adjusted earnings increase to 815 points.**

As for mean reversion of the CAPE ratio, without the aid of a crystal ball I have no idea what it will do.

However, a reasonable assumption is that the gap between the current value of CAPE and its average value will be halved over the next year.

In other words, **I’ll assume that CAPE declines from 25.8 today to 22.9 by the end of 2018.**

With cyclically adjusted earnings of 815 and a CAPE ratio of 22.9, here’s the forecast for 2018:

**My “expected value” for the FTSE 250 at the end of 2018 is 18,700**

That’s a decline of about 7% which isn’t even technically a correction, let alone a bear market.

If I was to be more pessimistic then a decline to “fair value” would mean a decline to 16,300. That’s a decline of almost 19%, which very nearly be a technical bear market (i.e. a decline of 20% or more).

**A more negative outlook from investors could easily lead to a much more serious decline than that.**

On a more cheery note, although many investors are seeing bubbles everywhere at the moment, I don’t think the FTSE 250 is anywhere near a bubble.

However, if it keeps going up at double digits each year then it won’t take long to get there.

So that’s forecast for 2018, now here’s a brief run-through of the same calculations for “beyond 2018”.

## FTSE 250 forecast beyond 2018: The future still looks bright

In the long-run we’re all dead, but in the next decade I hope most of us will still be around so a forecast to 2027 seems like a good idea.

I’ll use the same assumptions as above, i.e. that cyclically adjusted earnings grow by 5% each year and that CAPE closes the gap to “fair value” by half each year.

In that scenario, by 2027 we’d end up with FTSE 250 cyclically adjusted earnings of 1,261 and an “expected” CAPE ratio which has reverted all the way back to its fair value of 20.

Plugging in those numbers gives the following 2027 forecast:

**My “expected value” for the FTSE 250 at the end of 2027 is 25,200**

That’s a 26% increase from where the index sits today. That might sound okay, but it’s an annualised growth rate of of just 2% per year.

**If you include dividends at 3% per year then the total expected return from the FTSE 250 over the next decade is just 5% per year.**

So unless there’s a boom or bubble in 2027 there’s a good chance the FTSE 250 (and any passive funds that track it) will produce disappointing returns over the next decade.

This doesn’t mean I’m avoiding FTSE 250 stocks though.

In fact, over the last year or so I’ve increased my portfolio’s mid-cap weighting to 53%, ahead of large-caps which make up 40% of the portfolio.

And FTSE 250 stocks also make up the bulk of my stock screen’s top 50 stocks, so despite the index’s slight overvaluation, there are still attractively valued mid-cap companies out there.

**FTSE 100 valuation and forecast for 2018**

If you’re interested in a valuation and forecast for the FTSE 100, take a look at the one I did last week.

Steve Kirk says

Hi John

Thanks for sharing your analysis. In the dodgy world of share market forecasting, I think this is as accurate and useful as is practically possible.

I’m curious to know where you get your FT250 earnings and dividend data from to calculate the CAPE and yield? My data provider (Digital Look) only gives this for each constituent company so have you used a spreadsheet to calculate it or does one of your providers give this data at index level?

Also, have you done the same analysis at FTSE350 level? I’m assuming it lies somewhere between FTSE 100 and FTSE 250 in terms of value but a more exact analysis would be interesting. (I have a vested interest as I only choose my portfolio from FTSE350 companies)

Regards

John Kingham says

Hi Steve,

The data comes from a variety of sources which I’ve collected over the years.

Recent data comes from the FT Data Archive “World Markets at a glance” report, which is sadly now behind a paywall:

https://markets.ft.com/data/archive

Older data comes, amazingly enough, from a couple of charts (I tried to find the underlying data but failed):

https://www.ukvalueinvestor.com/2015/07/ftse-250-valuation-and-forecast-for-2015-q3.html/

A bit of squinting and drawing software with crosshairs allows the PE ratio and yield to be extracted going back to 1993, and they can be used to get the earnings and dividend numbers. Some sanity checking and comparison to things like inflation suggest that the numbers I have are probably pretty close to the official figures. Having said that, it is very annoying that nobody publishes official long-term FTSE PE and yield figures for free.

As for the FTSE 350, no, I haven’t done any analysis on that index. The FTSE 100 dominates the 350, probably making up about 80% of it so the correlation with the FTSE 100 is probably quite high, both in terms of price and valuation.

If you only look at FTSE 350 companies then this 250 review and my FTSE 100 review should do the trick, seeing as the 350 is just the combination of the 100 and the 250.

Walter Hin says

I used to pay attention to market multiples but don’t solely rely on it because you can miss out on investment opportunities.

Although earnings multiple is slightly above fair value, earnings are making new highs and the question is: “Can it continue?”

With the bond yield below dividend yield, along with the weaker pound attracting foreign capital, then the answer is probably yes.

A more interesting observation is the dividend cycle. The rise in dividends payout has more than doubled from their trough and is looking like a bubble.

Anyway, the missing link is the total borrowing cycle. If total borrowings are at all-time highs, then we need to watch for interest rate changes.

That affects dividend.

John Kingham says

Hi Walter, I definitely agree with not relying 100% on market multiples. No matter how high the market goes, there will always be pockets of value and sectors that are out of favour.

Your point about the borrowing or credit cycle is interesting. I have noticed that more and more of the attractively valued stocks in the market also have worryingly high levels of debt. I might do some number crunching to look at averages, quartiles, that sort of thing, to see if there really has been a shift to higher levels of leverage and therefore higher levels of risk.

Another useful measure is the market’s average price to book, which is a way to “look through” cyclical factors like profit margins to see how the market price compares to the “replacement cost” of its constituent companies.

And specifically in terms of dividends, the payout ratio for the FTSE 250 isn’t that bad. Currently about 50% of aggregate earnings are paid out as dividends, i.e the dividend cover is about 2. So I don’t see any serious risk to the FTSE 250’s dividend, at least not without a major UK or global recession.

Bret says

I initially thought that it’d be interesting to do the analysis for all other asset classes to see where the funds would flow to in a rational market. But then I started struggle with the philosophy behind the FTSE250 prediction.

Given that the CAPE ratio is effectively an measure of how much an investor is prepared to pay for a given level of earnings, if earnings are constant (say), then the CAPE can only drop as predicted if there is a net outflow of investors from the FTSE250. But the money must move to another asset class, typically because another asset class offers a more attractive valuation.

But the analysis appears to rely on regression the mean by solely examining the FTSE250 metrics. Thus it assumes that the FTSE250 valuation is isolated from all other asset classes, i.e. that it is uncorrelated with them, and that variations in its valuation are just short-term noise, with no other underlying cause. At the very least, an assumption has been made that other asset classes will simultaneously regress to their means, AND that the relative attractiveness of these other classes has remained unchanged over time. As an example of this last point, I’m unconvinced that the relative attractiveness of the US and Chinese stock markets will ever revert to their value of 20 years ago (or whatever CAPE timespan is used), because there are clearly factors other than statistical noise involved.

TLDR: the analysis appears to assume that the FTSE 250 valuation is uncorrelated with other asset classes, when obviously it isn’t! Thus regression to the mean seems a misleading method to use.

John Kingham says

Hi Bret, lots of interesting points in there.

“CAPE can only drop […] if there is a net outflow […] from the FTSE250 […] typically because another asset class offers a more attractive valuation.”

I agree with that.

“it assumes that the FTSE250 valuation is isolated from all other asset classes”

No it doesn’t. Because…

“an assumption has been made that other asset classes will simultaneously regress to their means”

I’m not sure about simultaneously, but yes, the assumption is that reversion to mean valuations is a widespread phenomenon across major asset classes.

“AND that the relative attractiveness of these other classes has remained unchanged over time.”

Not necessarily. This implies an efficient market and efficient flows of capital, neither of which I think exist in the real world. Instead we have very efficient markets and flows of capital, but far from perfectly efficient.

So the FTSE 250 CAPE can go up and down and it doesn’t have to keep the same relative attractiveness as other markets. E.g. the S&P 500 looks fairly expensive using CAPE while the FTSE 100 looks about average. So over the last decade the S&P 500 has become less attractively valued relative to the FTSE 100. However, market’s aren’t efficient so this isn’t a problem.

“I’m unconvinced that the relative attractiveness of the US and Chinese stock markets will ever revert to their value of 20 years ago (or whatever CAPE timespan is used), because there are clearly factors other than statistical noise involved”

I agree that other factors are involved, but I disagree that it’s obvious why the US stock market won’t revert to mean CAPE valuations, i.e. mid-teens rather than the current high 20s. History suggests that most people are invariably wrong about future market valuations, and the stronger their opinions the more wrong they are likely to be. E.g. the late 1920’s boom (everyone was optimistic), early 1930’s crash (everyone was pessimistic), mid-70s slump (pessimistic), late-90s boom (optimistic) etc. The actual direction and future valuation of the market turned out to be the complete opposite to mainstream opinion.

However, I don’t want to be too negative. Your points are all excellent but they seem to rely the efficient market. Without the efficient market, valuations can pretty much do what they like with no regard for rationality! Even mean reversion isn’t written in stone, but it is based on the natural discount rate of humans, which probably hasn’t changed much for millions of years.