Why you should cheer PE pessimism

ARE stocks “too expensive” these days? The bears boldly believe so – claiming that elevated price/earnings ratios (PEs) and other global stock market valuations signal a bubble that poised to pop – threatening even relatively lower-PE markets such as Singapore’s.

The Monetary Authority of Singapore (MAS) echoes these fears, warning that lofty tech valuations risk future disappointment.

However, I am of the view that valuations do not predict stocks’ direction. Decades of history prove present fears false. And while this is not my forecast for 2026 just yet, valuations will not sway it either way.

It is normal and human to fear heights. When our ancient ancestors roamed the wilds, healthy heapings of acrophobia could mean the difference between life and death. Tiny tumbles did not injure much. But falling from higher heights usually results in tragedy, so the advice has been to avoid heights.

For most investors, valuations’ stock forecasting power is sacred. Pundits constantly tout metrics such as the PE, price-to-book and price-to-sales ratios as key timing tools.

The logic: Lower valuations imply “cheap” stocks – so “buy low”. Higher valuations supposedly signal frothy markets and meagre returns ahead, or worse. It is better to “sell high”, and hence, comes pessimism with world stocks’ PE around 23.

But take a deeper look. Consider a gold-standard quality statistic called R-squared, which shows how much of one recurring phenomenon is explained by another – showing possible causality on a scale of 0 to 1.0. Zero means no causality and 1.0 complete causality.

Since good data started in 1999, the MSCI Singapore’s annual starting PE (using trailing 12-month earnings) and forward one-year returns yield a 0.15 R-squared.

This means just 15 per cent of the city stocks’ return even possibly stems from starting PEs. Not so much. R-squared using three-year returns rises to 0.24 – higher, but still indicating that over three-quarters of the stocks’ performance rests on factors besides PEs. With five-year returns, R-squared falls to 0.12.

SEE ALSO

5 key issues confronting investors now

As for world stocks, starting PEs and forward one-, three- and five-year returns have R-squareds of 0.08, 0.18, and 0.26 since good data started in 1970. If PEs play a role in global stocks’ performance, it is not a major one.

Why is there so little power? First, valuation metrics are among the most widely known investing data. When everyone knows something, it is always already priced into markets.

Also, stock prices are forward looking, while earnings figures look backward. Even projected earnings used in some PEs derive from current forecasts (which include often volatile sentiment), which often miss badly – down and up.

Better times ahead

Back in 2009, stocks foresaw better times ahead while in recession. They soared, but recession-decimated earnings did not reflect it yet. The world PE jumped to nearly 30 that year – at a generational buying opportunity. Those who avoided stocks because of high-PE pessimism paid dearly.

Pundits’ current PE protests extend to less common metrics, too – such as the “Cyclically Adjusted PE (Cape) Ratio”, which Yale professor Robert Shiller created. Many tout it as a “better” PE, as it supposedly mutes cyclical factors by averaging a decade’s inflation-adjusted earnings.

It was not intended for market timing – hence, its smooths away markets’ inherent profit cycles. But use it for that anyway… and cherry pick intervals when it seems to “work”, regardless.

Using Shiller’s US stock data (stretching all the way back to 1881), annual starting PEs and forward 10-year returns’ R-squared approaches 0.30. Fully 70 per cent still must come from something else. And it fizzles completely, on both shorter and longer periods. Is a metric that offers a bit of value over only one timeframe worth building your investing strategy around?

Now, I am not saying bad returns cannot follow high valuations. They can and often do. Singapore stocks started 2000 at a 58 PE, then annualised -4 per cent over the next five years.

But abundant examples of the reverse exist also, like stocks starting 2004 with a 21 PE and still returning 18 per cent that year – and annualising 23 per cent over three years. From 2021’s starting PE of 30 (well above today’s 23) world stocks returned 73 per cent over the nearly five years till this October.

US and world stocks had elevated PEs continuously from 2009 to 2025, yet soared 765 per cent and 490 per cent, respectively, till October – despite two bear markets along the way.

Ignoring randomness by bias is just dumbness. Still, valuations can be useful at the category level. If you think value stocks will lead, you might consider cheaper values in stock selection. That was why I created the price-to-sales ratio. But as a broad market driver? No.

Recent valuation fretting is misguided, but good news. False fears add bricks to the “wall of worry” that bull markets legendarily climb. Let others believe the myth. And stay bullish.

The writer is the founder, executive chairman and co-chief investment officer of Fisher Investments, an independent investment adviser serving both individual and institutional investors globally