FTSE 100 and FTSE 250 mid-year valuations

We’re about halfway through 2019, so now is a good time to stop, smell the roses and see how attractively valued (or not) the FTSE 100 and FTSE 250 are.

I’ll be estimating fair value for these indices using their dividend yields and CAPE ratios (Cyclically Adjusted PE), because I think the standard PE ratio is basically useless as a valuation too.

Why I don’t use the PE ratio to value indices

If you’ve heard my anti-PE rant before then feel free to skip down to the next section. If not, here’s a quick summary of why the PE ratio is almost useless for long-term investors:

The reason is that the PE ratio compares an index or a company’s price to last year’s earnings, and earnings over a single year are just too volatile to be a useful measure of long-term value.

For example, in mid-2017 the FTSE 100’s price was 7,400 and its PE ratio was 25. That’s quite a high PE ratio, so a PE-focused investor might have thought the FTSE 100 was expensive at the time.

However, one year later in mid-2018 the FTSE 100’s price had risen very slightly to 7,700 but its PE ratio had almost halved to 13.4. So now a PE-focused investor might think the FTSE 100 was fairly valued or even cheap.

But how can that make sense? The FTSE 100 stayed roughly flat for a year and the PE ratio halved, so if we’re using the PE ratio as a measure of price to intrinsic, fair or long-term value then we’re saying that the fair value of the FTSE 100 doubled in just one year. Really?

Of course it doesn’t make sense to say that the FTSE 100’s intrinsic or fair value doubled in a single year.

What doubled was its earnings over one single solitary year, from 296 index points in mid-2017 to 575 in mid-2018. What didn’t double in a year was the ability of 100 very large companies to generate long-term profits, and that’s why the PE ratio is basically useless.

So rather than looking at the PE ratio, long-term investors need to value companies and market indices using something which is more stable and reliable, and my two preferred candidates are the dividend yield and the CAPE ratio.

Valuing the FTSE 100 using dividend yield

The dividend yield is a more useful measure than the PE ratio because dividends tend to be more stable than earnings from one year to the next, as the chart below shows:

This stability gives us something solid and consistent to compare volatile prices to, so let’s do that using the trusty old dividend yield:

• The FTSE 100’s price is 7,400 and its dividend yield is 4.4%

That’s comfortably above the index’s average dividend yield for the last 30 years, which is 3.3%:

An above average dividend yield is a reasonably good indicator that prices are below average and that expected future returns are above average. So today’s above average yield is good news for long-term investors.

Unfortunately though, most investors don’t see high yields as good news. They focus on the falling prices which create higher yields, and those falling prices scare them off.

A good example of this was the 2018 year-end ‘correction’. It left the FTSE 100 with a low price of 6,700 and a high dividend yield of 4.7%, but panicked investors were jumping ship left, right and centre.

If they’d focused instead on the FTSE 100’s increasing yield rather than its falling price, they would (or should) have realised that the index was in fact more attractively valued at the end of 2018 – according to its dividend yield – than at almost any time over the previous 30 years.

Returning to the FTSE 100’s current dividend yield of 4.4%, if we assume fair value is the price that gives the FTSE 100 its long-term average dividend yield (of 3.3%), then we get the following result:

• The FTSE 100’s yield-based estimate of fair value is 9,900

At 7,400 the FTSE 100 is about 25% below estimated fair value. Another way to think about this is that it would need to grow by 34% before reaching estimated fair value, which is quite a lot. In fact, as the chart above shows, the FTSE 100’s dividend yield is still close to its 30-year high, so the yield-based valuation is near its 30-year low.

The only time yields were much higher than this was during the depths of the financial crisis in March 2009, although that short-lived episode doesn’t show up on the chart because it only uses year-end data.

In summary then, dividend yield-based valuation are looking quite attractive. Next up is an estimate of fair value using Robert Shiller’s CAPE ratio.

Valuing the FTSE 100 using the CAPE ratio

The CAPE ratio (Cyclically Adjusted PE) is another useful measure of value because it smooths out the ‘E’ part of the ratio by using an inflation adjusted ten-year average of earnings rather than just a single year’s.

The chart below shows how dividends and CAPE earnings have both been relatively steady over the long-term, although a recent decline in CAPE earnings (caused by the 2009 financial crisis and the 2014 commodity price collapse) shows that no financial metric will be perfectly smooth and steady in what is a volatile, uncertain, complex and ambiguous world.

Since CAPE earnings are generally quite steady from year to year, changes in the CAPE ratio are usually driven by changes in price.

And as with dividend yield, when the ratio is equal to its long-term average we can assume (incorrectly but reasonably) that the index is at fair value, when it’s above average the index is above fair value and when it’s below average the index is below fair value.

Here’s a quick chart showing CAPE’s volatility over a typical investment lifetime:

The FTSE 100’s CAPE ratio has ranged from just over 30 at the peak of the dot-com boom down to about 10 at the bottom of the financial crisis in March 2009 (this doesn’t fully show up on the chart as it uses year-end data only).

The average CAPE for the period is 18.7, but I think that’s above the ‘true’ long-term average thanks to the wild excesses of the dot-com bubble and its wildly elevated CAPE ratio.

I generally assume the true long-term average is closer to 16, as that’s more in line with the long-term average CAPE of international markets. 16 also somewhat neatly sits in the middle of a range (actually a logarithmic scale) from 8 and 32, and that range pretty much encapsulates the range of CAPE value we’ve experienced over the last 30 years.

With the FTSE 100 at 7,400 and its cyclically adjusted earnings at 463 (index points), we get the following result:

• At a price of 7,400 the FTSE 100’s CAPE ratio is 16.0

Oddly enough, this is bang on my estimated long-term average for CAPE, which means:

• According to the CAPE ratio, the FTSE 100 is fairly valued at 7,400

Before I wrap up this FTSE 100 valuation, here’s a chart showing how the index price has moved around within its normal valuation range:

One thing you may have spotted is that the CAPE-based estimate of fair value is quite different to the yield-based estimate, so it’s probably time to say something about why that doesn’t really matter.

Fair value is only an estimate and that’s okay

The two estimates of fair value I’ve come up with so far are quite different, but that’s only half the story. In my 2018 year-end review, I said the FTSE 100’s dividend yield-based fair value was just over 10,100, and that’s north of the both the fair value estimates from this article.

So what’s going on? Was I wrong last time, am I wrong this time and can fair value really be this hard to pin down?

The answer is yes in all three cases, and this highlights two important facts:

1) Fair value can only ever be an estimate

Nobody knows what’s going to happen tomorrow, so nobody knows the true fair value of the FTSE 100. That’s because fair value is (in technical terms) the present value of all the FTSE 100’s future earnings, discounted at an appropriate rate.

We don’t know what those future earnings will be, so we don’t know what fair value is. But we can make sensible estimates.

2) Fair value is only really useful when an index is a long way from it

The relationship between index prices and fair value is the same as the relationship between a dog on a long elasticated lead and the dog’s owner. They are connected, but only loosely. The dog (prices) can travel a long way from its owner (fair value) and the only time there’s a strong pull back towards the owner is when the dog’s far away and the elasticated lead (investor sentiment) is stretched to breaking point.

This means the inevitable inaccuracy of our fair value estimate doesn’t matter when the market is close to fair value, because such a small difference is unlikely to affect the market’s future direction anyway.

And when an index’s price is a long way from fair value the gap between price and fair value will be so large that a ten or twenty percent error in our fair value estimate also won’t matter because the valuation gap will still be blindingly obvious.

The FTSE 100: Close to but likely below fair value

In summary then, the FTSE 100 is approximately zero to 25% below a reasonable estimate of fair value.

This isn’t a great surprise given the amount of negative sentiment surrounding the UK at the moment, largely thanks to ongoing Brexit uncertainties.

Given this starting point, I think the FTSE 100 is likely to produce annualised total returns at least in line with its historic norm of seven to nine percent over the next decade, and quite possibly even more if the index moves beyond the 9,900 estimate of fair value.

Okay, that’s enough crystal ball gazing. Now it’s time to estimate fair value for the FTSE 250 in a hopefully somewhat more brisk fashion.

Valuing the FTSE 250’s using dividend yield

The FTSE 250’s price is currently 19,290 and it’s dividend yield is 3.2%.

Over the last 27 years (that’s the period I have data for) the FTSE 250’s average yield has been 3.0%.

This gives the following result, if we again assume fair value is when the index’s yield is equal to its long-term average:

• The FTSE 250’s yield-based estimate of fair value is 20,900

The FTSE 250 is currently just 8% below this fair value estimate, which is nothing.

Valuing the FTSE 250’s using the CAPE ratio

The FTSE 250’s inflation adjusted ten-year average (CAPE) earnings are currently 823 index points.

With its price at 19,290 that gives the FTSE 250 a CAPE ratio of 23.4.

This seems high compared to the FTSE 100, but the FTSE 250 usually does have a higher CAPE ratio. That’s probably because FTSE 250 companies are smaller and often faster-growing than their large-cap cousins.

The average FTSE 250 CAPE ratio since 1993 is 23, but I usually round this down to 20 because a) it’s an easier number to work with and b) 23 is probably a little bit above the true long-term average because of high valuations during the dot-com and credit bubbles of the 1990s and 2000s.

If we assume that fair value occurs when the FTSE 250 has a CAPE ratio of 20, we get the following result:

• The FTSE250’s CAPE-based estimate of fair value is 16,500

As with the FTSE 100, we have a difference between the two estimates of fair value and that’s okay.

With one estimate of 20,900 and one of 16,500, we can see a reasonable range of possible values for fair value, and at 19,290 the FTSE 250 currently sits inside that reasonable range.

This means it’s reasonable to expect the FTSE 250 to produce historically average returns over the next decade because a) the dividend yield is close to average and b) price growth won’t be held back by an excessively high starting point or boosted by an excessively low starting point.

And in case you were wondering, average returns for the FTSE 250 means an annualised total return of something in the region of 10%, with much variation year-to-year.

To finish off, here’s the CAPE valuation rainbow chart for the FTSE 250, showing the index’s historical and current valuation gap:

Boring but useful

I will admit that sometimes it can seem a bit pointless to do these infrequent but regular market valuations, especially when valuations have been largely benign and close to fair value for several years.

On that final point I would say a couple of things:

1) If you’re a ship’s lookout, you don’t stop looking for icebergs just because you haven’t seen one in a couple of days, and

2) Regularly reminding ourselves about the basics of long-term fair value should help us to buy low when everyone else is panicking and to sell high when everyone else sees nothing but rainbows and unicorns.