“The market is overvalued” warning occurs every time there’s a recession recovery, like now. It is created by two flaws: bad math and looking backward.
The real damage is done by investors heeding that message and waiting to buy stocks. With that perspective, they view a rising stock market skeptically, feeling the higher prices are making the overvaluation worse.
This stock market’s steady 12-month 40% rise is a perfect example. It has failed to disprove the overvalued notion to investors. Instead of proving that the overvalued proposition is wrong, the higher prices have ratcheted up the belief.
This months-long good performance has had one effect. It has lessened the “Crash is coming!” omens. But still investors wait, only now their focus is the hope for a sizable correction that will provide a buying opportunity. However, this hope for bargain prices is unlikely to be fulfilled without some unexpected negative news, as happened last October. Instead, the pattern of short-term mini-drops indicates a strong Wall Street willingness to jump in and buy quickly on any minor sign of weakness (like the so-called “plummet” on Friday, June 18).
So, why the entrenched view of overvaluation?
At the heart of the overvaluation argument are two flaws: bad math and looking backward.
The first flaw is the arithmetic used to calculate the stock market’s key valuation measure: the price/earnings ratio. Compared to the dividend yield (D/P), the price earnings ratio (P/E) is inverted, putting price as the numerator. As a result, it follows a curved path that skews results. Fixing the skewed path is simple: use the same calculation as dividend yield – i.e., earnings yield (E/P).
The second flaw is looking backwards at reported earnings instead of focusing on the forecast future ones. Those “trailing-12-month” earnings are not how Wall Street determines valuations. Instead, professional investors are always looking ahead. In 2021 that has been especially important because this year is a return-to-normal period – a mixture of Covid-19 recession hangover and new growth. Therefore, the 2022 forecasts, as uncertain as they may be, are the best measure of “normalized” valuations.
How fixing these two flaws removes the overvalued view
The Wall Street Journal’s “Market data” site has the Dow Jones Industrial Average price/earnings ratio at 29.0x. Fix up the first flaw (bad math) and it drops to 25.1x. Correct the second flaw (past earnings) and it drops further to 18.6x. [Forecast earnings source: Financial Visualizations FinViz.com]
(Note: The “bad math” correction is straightforward, but the example I created is a bit long. Therefore, it’s at the end of this write-up.)
Now, focus on that 18.6x P/E for the DJIA. It is less than 2/3 of the 29.0x investors have in their minds. Clearly, that means the market is not overvalued. It should be viewed as reasonably valued – or perhaps even undervalued.
A better view of valuation: Earnings yield (E/P)
While the P/E ratio is a commonly used for valuation, a more useful measure is earnings yield (E/P). It ties directly to other yields and even the payout ratio, a critical calculation for analyzing dividend income safety and potential growth.
Start with changing the P/E ratios above to earnings yields:
- 29.0x = 3.4%
- 25.1x = 4.0%
- 18.6x = 5.4%.
Notice how the picture’s clarity improves. Instead of a P/E decrease from 29.0x to 18.6x, the change becomes an earnings yield increase from 3.4% to 5.4%.
Add in dividend yield and payout ratio for more fulsome valuation
Now, let’s bring in the 1.8% dividend yield. (Remember, that yield has a bit of a forward look because it is based on the current payment extended out for a year.)
Here are the payout ratios based on the yields. It is simply the ratio of the dividend yield to the earnings yield – in other words, (D/P) / (E/P), which equals the payout ratio D/E.
Here are calculations:
- 1.8% / 3.4% = 53% (bad math, looking backward)
- 1.8% / 4.0% = 45% (correct math, looking backward)
- 1.8% / 5.4% = 34% (correct math, looking forward)
So, now the view of the stock market is not only that it is not overvalued, but the dividend safety/growth picture is significantly better with a payout ratio of only one-third of forecast earnings.
Reminder: The 10-year US Treasury note is often referred to when discussing the stock market. Okay, here’s the contrast. The 10-year UST note currently yields about 1.5%. As we’ve seen, the DJIA’s earnings yield is 5.4% and its dividend yield is 1.8%. Looks good. But wait… there’s more. That note’s yield is fixed. Held to maturity, the investor will receive 1.5% (ignoring taxes and inflation). For the DJIA, however, those earnings and dividends will change over time. That’s risk, but also opportunity. Chances are, investing in the DJIA for ten years will result in earnings and dividend growth providing a higher yield.
The bottom line: This stock market is not overvalued – it may even be undervalued
I know. Undervalued seems a silly notion. Except think about that knee-jerk, mental reaction. It’s a possible contrarian sign that puts us on the right analytical path. In the stock market, when many (most) investors are looking one way, it can pay to look the other.
If that contrarian sign is correct, the 1.8% dividend yield, 5.4% earnings yield and 34% dividend payout ratio provide good fundamental support for the idea that today’s market is a buy.
Getting rid of the bad math – An example using a 3-stock portfolio
This 3-stock example below illustrates why the change for P/E to E/P is necessary to analyze portfolios. (All numbers are per share; dividends and earnings are annual; the portfolio is equal-weighted.)
- Stock #1: Price = $100, Dividend = $1, Earnings = $3
- Stock #2: Price = $100, Dividend = $3, Earnings = $8
- Stock # 3: Price = $100, Dividend = $5, Earnings = $10
Start with the dividend yield (D/P):
- #1: 1/100 = 1% #2: 3/100 = 3% #3: 5/100 = 5%
- 3-stock average: (1% + 3% + 5%)/3 = 9%/3 = 3%
- And 3% is exactly what would be received from that 3-stock portfolio:(1+3+5)/(100+100+100) = 9/300 = 3%
Now look at the price/earnings ratio (P/E):
- #1: 100/3 = 33.3x; #2: 100/8 = 12.5x; #3: 100/10 = 10x
- 3-stock average = (33.3 + 12.5 + 10)/3 = 55.8/3 = 18.6x
But that verification math fails. In the 3-stock portfolio, the earnings “received” are 3 + 8 + 10 = 21. The combined investment is 300. Therefore, the actual P/E = 300/21 = 14.3x, significantly below the 18.6x from above.
The problem is the inverted division. It changes a straightforward yield slope into a parabolic curve. That creates especially serious portfolio/index calculation problems when stocks have low earnings. (For example, 100/1.00 = 100x; 100/0.10 = 1,000x; 100/0.01 = 10,000x). Worse is that companies with zero or negative earnings must be excluded because dividing by zero cannot be done and dividing by earnings below zero produces a nonsensical negative P/E ratio that improves (reduces) the average.
The cure? Simple. Invert the division to create the earnings yield calculation. This isn’t an odd adjustment. “Earnings yield” is well-known. It’s just that it’s rarely used to calculate the average P/E ratio. So, back to the example…
Earnings yield (E/P):
- #1: 3/100 = 3% #2: 8/100 = 8% #3: 10/100 = 10%
- 3-stock average: (3% + 8% + 10%)/3 = 21%/3 = 7%
- And 7% is exactly what would be earned from that 3-stock portfolio:(3+8+10)/(100+100+100) = 21/300 = 7%
And now the accurate P/E for the 3-stock portfolio can be calculated: 300/21 = 14.3x
About those problematic P/E numbers when earnings are low or negative…
The earnings yield slowly shrinks in line with earnings, reaching zero when earnings are nil. What to do with negative earnings? Don’t exclude the stocks, as is done with the P/E averaging. Instead, since the companies are not producing positive earnings, use zero as their E. That keeps them in the calculation, correctly reducing the earnings yield (raising the P/E), but not subtracting losses.
Now to the market index calculations
As reported in the market data section of The Wall Street Journal’s website, the current dividend yield and P/E for the DJIA are 1.8% and 29.0x.
The DJIA is price-weighted which makes for an interesting mathematical formula for calculating the average P/E.
First, each stock’s P/E is multiplied by its price = (P/E) x P = P2/E
That squaring of the price is where the skewness comes in. The final step is those individual measures are summed and then divided by the sum of the prices.
Now look at the average earnings yield formula:
Each stock’s E/P is multiplied by its price = (E/P) x P = E
How about that? The first calculation simply removes the price per share.
Thus, to get the DJIA earnings yield average, all we need to do is add up the earnings per share numbers (converting the negatives to zero), then divide by the sum of the current prices.
Still want an average P/E? Then invert the calculated E/P and there you have it.
Note: “Price-weighted” simply means that one share of each stock is held in the index. Therefore, each stock’s weight in the index is determined by its price. “Market capitalization-weighted” indexes (the common calculation) also weight by price, which is multiplied by shares outstanding. The P/E skewness exists for market cap-weighted indexes also.