Ken Fisher: Why CAPE is useless

If Zero for Four, Simply Ignore

What is hugely overvalued? Valuation tools! Particularly popular ones like the most popular PE voodooism, CAPE, which has been bearish the last four years of this bull market. My rule: Zero for four, simply ignore. CAPE is worth nada.

CAPE, the cyclically adjusted price-to-earnings ratio, is also known as the Shiller PE, named after its co-creator (Yale's Robert J. Shiller). CAPE divides the market's price by a rolling average of the last 10 years' earnings, adjusted for inflation.

Disciples argue this beats traditional PEs (price divided by the last 12 months' earnings) because it smooths the economic cycle's occasionally huge skew, better reflecting stocks' true value. Today's S&P 500 CAPE is 29.3, a level seen just twice in history: before the 1929 crash and the Tech Bubble's 2000 implosion. Pundits warn this means trouble is imminent.

Hogwash! They've been saying high CAPE signals doom since 2013, when it first approached pre-2008 levels. The bull market was four years old then. It's eight now, and still going strong.

If I had to choose between a broken clock and CAPE, I'd pick the busted timepiece - at least it's right twice a day. CAPE isn't. My 2007 and 2015 books (The Only Three Questions That Count and Beat the Crowd) discuss this at length. When former Fed head Alan Greenspan warned of "irrational exuberance" in December 1996, high CAPE was his inspiration. Stocks disagreed, and, including dividends, the S&P 500 rose 116% over the next 39 months.

CAPE's flaws are manifold. Averaging 10 years of earnings might make it smoother, but it also makes CAPE extra backward-looking. Today's CAPE still includes earnings from 2008-2009's recession. How are those relevant now? Markets weigh future profitability, not what happened a decade ago. Past performance doesn't predict future returns. Heck, two years from now, CAPE could fall just because those recessionary earnings are out of the 10-year calculation. So would it suddenly be bullish?

The calculations are also problematic. Shiller uses GAAP earnings on his website, but GAAP standards have changed over time and the dataset doesn't account for shifts, rendering historical comparison useless.

Some argue this is why CAPE has signaled below-average valuations in only 17 of 326 months since January 1990. More importantly, normal PEs aren't adjusted for inflation, and why would they be? Investors earn nominal stock returns, and companies reap nominal profits. It's like with like. Comparing nominal stock prices to inflation-adjusted earnings, as CAPE does, is bad math. You'd flunk Econ 1 for that.

Then, too, the definition of inflation isn't consistent over time. The presentation Shiller gave Greenspan back in 1996 deflated earnings using an odd producer price index. His website now shows the consumer price index. Why the change? What would a comparison of the two series look like?

CAPE's creators never meant it to be a market-timing tool. They sought to forecast the next decade. The media misinterprets it as a tool to pinpoint peaks and troughs, largely because Shiller's book citing the data published in early 2000, overlooking the preceding four years.

Then again, even used "properly," CAPE is useless. Stock prices are determined by supply and demand. Hence to forecast long-term stock returns, you must be able to predict far-future stock supply. No one knows how to address this. I surely don't! CAPE doesn't even try, instead relying on past information, which markets already discounted.

Even if CAPE proves correct and the next decade's returns are below-average, that isn't actionable. Investors don't (or shouldn't) position for 10 years from now. They want to position for this year and the next.

Would you have cared about the next 10 years' return in 1996 if you knew the market would be up massively through March 2000, only for (what was then) the biggest bear market since 1929-1933 to strike? Would you have been OK with skipping those huge up years? Would you have held on through the down? Every study of human behavior says no and no.

Similarly, everyone calls the 2000s a lost decade because the S&P 500 fell -9.1% from December 31, 1999 – December 31, 2009. But there was a nice bull market from October 2002 through October 2007, and a fantastic 10 months from March 9, 2009 through year end. Would you have wanted to sit those out?

Get ahead now by interpreting CAPE differently: When folks fear CAPE and overvalued markets, skepticism lingers. Meltdowns happen when investors are too euphoric. Melt-ups happen as animal spirits awaken and confidence grows. They're stirring now, and the melt-up is coming. Own stocks like these to be ready:

Mastercard is up 41.7% over the last three years. Buy it. It's confused for a credit card company instead of a moderate growth payments-processing IT firm. At 23 times my 2017 earnings estimate it's priced A-OK. Its 23 PE equals an E/P of 4.4% (1/23), the after-tax return we would get forever if earnings never changed. But earnings will rise fast.

Late in bull markets Telecom acts like Tech - if earnings positively surprise! 2017 PE estimates averaging 12 for Verizon are tepid based on meager growth prospects tied to high smartphone penetration. I'm way more optimistic tied to rising mobile data usage and VZ's great reputation (why its market share is high). It should blow away the top forecasts by 10%. The surprise should move the stock. Meanwhile, it has a 4.6% dividend yield.

This article is for information and discussion purposes only and does not form a recommendation to invest or otherwise. The value of an investment may fall. The investments referred to in this article may not be suitable for all investors, and if in doubt, an investor should seek advice from a qualified investment adviser.