There are many articles on Seeking Alpha and other news websites that state the market is overvalued (NYSEARCA:SPY). Here is a sampling of a few titles which can be found:
"Yes; Stocks Really Are Overvalued"
"I estimate that the S&P 500 is 16-32% overvalued in the base case."
"A popular valuation metric pioneered by Nobel Prize-winning economist Robert Shiller says that stocks are dangerously expensive."
In contrast, there are very few articles that state the market is undervalued. Despite this, the market has risen 28% since the start of 2014.
This piece will set to prove the opposite that the market is still undervalued. The article will examine two things that affect valuation.
- U.S. Treasury yields are at historic lows.
- Payout ratios are statistically different from their average.
U.S. Treasury yields are at lows
U.S. Treasury yields represent the market risk free rate. Conversely, the equity risk premium is the added return that you would need to pick an investment over buying a U.S. Treasury.
If we look at a 5,000-year chart of human history, we can see that yields on the risk free rate have never been so low.
More seriously, the U.S. 10-year Treasury yield closed at a record low on July 5 of 2016. Many European countries have negative yields on their bonds, which means that you pay the country for them to borrow your money. The German 2-year bond is nearly negative 1%.
Given the historically low Treasury yields, one would expect historically low equity risk premiums, but this does not appear to be the case.
A couple of graphs are striking:
The first graph shows the S&P 500 yield against that of the risk free rate for a 10-year U.S. Treasury. A second graph shows the risk premium that the market is demanding over the risk free rate:
The market is currently demanding one of the larger risk premiums. The average equity risk premium over a 10-year bond is 3.84% since 1960. Even if the middle of the 1980s are removed as outlier years, due to super high inflation, the equity risk premium is still 4.12%. The current risk premium based on the market ending 2016 was 5.5%.
In order to build the forward-looking Internal Rate of Return (IRR) model, the calculation of the market's growth potential had to be done.
Since 1960, GDP growth, non-inflation adjusted, has been compounded at 6.547% vs. 6.561% of the S&P 500 earnings. Therefore, the long-term S&P 500 growth rate for the model used an estimate of American GDP. Future GDP growth rate was determined by the Congressional Budget Office (CBO) to determine GDP going forward - remember these GDP figures include inflation.
The IRR of the market was determined to be 7.8% and the 10-year Treasury bond was 2.3% at the end of 2016. A quick explanation of the model and IRR is below. Furthermore, the model projects 2017 earnings to only be $113.65 and 2018 earnings to be $118 while the Standard and Poor's forecast's earnings of $130.21 next year and $148.22 in 2018 (this is an excel sheet from S&P).
If the model is adjusted to the S&P earnings for 2017 and 2018 followed by the U.S. Government's forecasted GDP growth then the IRR jumps to 8.5%, which would show the market is even more undervalued. This may make sense given the large losses the energy sector experienced over the last two years. Below is a chart of the Standard & Poor's earnings breakdown - note 2017 and 2018 are S&P's forecasted numbers and the earnings before 2017 are the actual.
As can be seen, depressed energy earnings held down the EPS in 2015 and 2016.
* * * * *
A calculation of the IRRs can be seen in my attached sheet. IRR is a number that brings net present value to $0. Succinctly put, it is the discount rate where a future dollar is worth a present dollar. The IRR from 1960 to today uses actual earnings; therefore, the model assumes people in 1960 had perfect visibility of the future until the year 2017. The calculated equity discount rate demanded by the market in 1960 based on actual numbers is 9.8%. However, this amount may have been higher had market participants forecasted higher growth than actually happened and lower had the market been forecasting lower growth.
A quick sample calculation can be shown on how the model determined an IRR/Equity discount rate of 9.8% for 1960. If you were to buy the S&P 500 index in 1960, it would cost $58.11 (which was the price of the S&P at the time) and then one would receive a cash flow of its earnings, which was $1.98 in 1960. $2.04 in 1961, etc., up to the present including the forecasted cash flows after 2017.
* * * * *
Since Treasury rates are rising in 2017 and are expected to rise, it is more appropriate to use a long run 10-year Treasury yield of 3% as a risk free rate. The IRR should then be the expected risk free rate + the average equity risk premium. If the long-term rate is expected to be 3% and the market used the average risk premium of 3.8% (calculated above), then the market should have an IRR of 6.8%, but it is currently 7.8%, indicating the market may be undervalued.
A lowering of the market discount rate to 7.3% and 6.8% would produce an S&P 500 at 2,500 and 2,900 respectively, which is a 5% to 22% increase from current levels (2372 as of March 11, 2017).
The 10-year Treasury is currently yielding 2.57% which translates into an equity risk premium of 5.23% (7.8% calculated IRR at the end of 2016 less current U.S. Treasury yield of 2.57%). The argument is that the equity risk premium should be a function of Treasuries. Therefore, if the return on treasuries drops so should the total required equity return.
A market-based example can be illustrated by comparing yields on preferred stock compared to the price of the market. Preferred stock, in most cases, is similar to equity. And equity would likely only demand a slightly higher risk premium than Preferred. There are very few preferred stocks selling around 7.8% yields; in fact, most trade at much lower yields.
Payout ratios are significantly different from the average
Productivity in the United States since 1950:
U.S. businesses have been able to get more out of a laborer over time. The historical return to shareholders from 1960 was estimated at 62% - I had to estimate for share buybacks from 1960 to 2006.
Most people value the market on a constant P/E or ROE basis and assume no ability to improve the metrics over time, which is contrary to the facts. The chart below illustrates this fallacy. A number 1 means the payout ratio was above the average. One can quickly see that payout ratios consistently beat the average since 1987.
Furthermore, an even more consistent long-term trend has been observed since 2006.
A standard deviation on the payout ratio was calculated at 19%. A quick statistical rule would show a normal distribution should fall within 1 standard deviation 68% of the time. The last 4 years would have a 95 to 1 chance of occurring. The chart below again illustrates higher payout ratios over time.
Last year the payout ratio was 98.5% but for modeling purposes, it was assumed that the future payout ratio will be 80% going forward - which is 1 standard above the average but below the 10-year average payout ratio. Many readers may believe that a higher payout ratio will mean less room for increased cash flows over time.
Quite the contrary, in the past, with more manufacturing companies, U.S. companies could support a lower payout ratio. More cash needed to be used to grow at a near equal rate to today. With a different base of companies, the S&P 500 can now afford a higher payout ratio. Buybacks and dividends are included in the payout ratio.
Buybacks are the same as dividends at fairly valued prices; better than dividends at discounted price and worse than dividends at overvalued prices. Buybacks are more tax efficient all things being equal. Therefore, buybacks are included as part of the calculation of the payout ratio.
There are a lot of people saying the market is overvalued but cash flow/earnings has changed over time and yields on the risk free rate are likely to remain at a low level indefinitely.
Given these two factors, most people are not understanding why their basic P/E metrics are not working. This is because the market always used cash flow and cash flow is now a higher percentage of earnings. Cash flow is and has always been the correct metric to use. The risk free rate is also lower than it has been throughout human history, once again indicating that a lower return is expected in the market going forward.
Disclaimer: Do not base any investment decision upon any materials found in my article. I am not registered as a securities broker-dealer or an investment adviser either with the U.S. Securities and Exchange Commission (the "SEC") or with any state securities regulatory authority. I am neither licensed nor qualified to provide investment advice. The information contained in our report is not an offer to buy or sell securities.
Disclosure: I am/we are long XSP.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I do happen to be long XSP (a hedged version of the S&P 500) but I do not consider it to be a core position. The position was supposed to be a bet on the Canadian Dollar (Sold the USD At 1.44 and purchased XSP at $21.94) But that trade has not added that much value. I may exit The position in the year. I typically consider ETF's as a very lazy way to play the market and Bill Ackman (last year) did write a very good piece on the dangers of too many people holding ETF's. However, this position made a bit more sense given that I felt more bullish on the CAD last year. I no longer feel that bullish about the CAD prospects and therefore holding an ETF is not in my core positions.