Each month, I like to take more of a top-down view of the market by making an approximate projection of the value of the S&P 500 (NYSEARCA:SPY). I find it has value for couple different reasons - determining how the current market valuation might compare historically and to gauge what type of returns investors are currently expecting. These returns expectations can be viewed alongside your own. To someone expecting 4% real returns, a pricier market might not be frowned upon to the level of someone expecting an 8% return, and vice versa.
To do this, I use a discounted cash flow model with the following inputs:
- Market's current earnings yield
- Current dividend yield
- Cash payout ratio that aligns with historical norms (a cash payout ratio is equal to the sum of dividends and net share buybacks as a percentage of earnings)
- Earnings growth rate in the market (note that this is different from earnings yield; earnings growth is the annualized rate at which earnings will grow year over year while earnings yield is the annual earnings amount divided by the sum of the S&P 500 index)
- 10-year US Treasury yield (for use as a "risk-free rate")
- Equity risk premium (a large component of our discount rate: investors expected returns minus the 10-year Treasury)
- Terminal growth rate that accords with the long-term real growth expectations of the US economy
As of last Friday's close, the earnings yield in the S&P 500 is 3.97%. This figure is measured based on trailing 12-months' ("TTM") earnings relative to the current value of the index. This represents an 18-basis point contraction over this point last month due to the market's rise.
As can be inferred from the chart below, this is the lowest earnings yield observed since the market was recovering out of the financial crisis in 2009.
The current dividend yield comes to 2.05% using the same TTM approach, and will be the figure used in the discounted cash flow model.
With respect to cash payouts, companies cannot sustainably pay out more to their shareholders than the level of earnings they take in. With the cheapness of debt and lackadaisical business investment activity since the financial crisis, we have seen cash payout ratios elevate beyond their sustainable bounds. Dividends and buybacks as a percentage of earnings from 2011-2015 came to about 77%; over the 10-year period from 2006-2015, 88%; and over the past year have trended above 100% of earnings.
This can't persist indefinitely as it would entail a lack of re-investment, issuing larger quantities of debt or equity, or dipping into retained earnings should such a surplus exist. Payout percentage ratios above the high-80s% are generally only observed during recessions or notable macro shocks, which can be witnessed in the chart below during 1989 (Japan), 1998 (Russia), 2001 (dot-com), and 2007-09 (leading in and out of the financial crisis).
Nevertheless, 2016 represents an exceptional circumstance given the lack of organic earnings growth is forcing corporations to resort to financial engineering tactics to appease shareholders, often buying already overvalued stock to stretch its valuation even higher. The graph also shows a distinct divergence from the correlation between cash payout ratios and the high-yield bond spread, demonstrating the phenomenon is not crisis related.
From a historical perspective, corporations pay out around 75% of their earnings. With fixed investment in the US remaining low, I will use 80% in the model - sustainable yet above the long-run average.
Earnings growth tends to hover around 4-5% per year. Nailing this down is obviously difficult to do. But fortunately, the model isn't very sensitive to this input, with a 100-bp shift only generating a ~4.6% difference in the intrinsic value of the index. In the model, I'm using 4.5% as a reasonable figure.
The 10-year Treasury rate was 2.39% after Friday's market close. Last month, it stood at just 1.8% and rose the highest in any month since November 2009. A higher 10-year Treasury rate, holding all else equal, will lower the model's output due to the way it increases the discount rate at which the cash flows are valued. On a "flow of funds" basis, higher bond yields also lessen the attractiveness of equities.
I treat the equity risk premium as the sensitivity parameter in the model to determine a representative valuation range. Sensitizing the equity risk premium to the value of the index can essentially back out expected investors returns over the 10-year Treasury. The equity risk premium for the S&P 500 has historically been around 4.5-4.6%. Since 1960, the average equity risk premium, defined as the annual average sum of the S&P 500's earnings yield and dividend yield minus the year-end 10-year Treasury yield, has come to 4.1%.
In the past 10 years, the equity risk premium has expanded to 120 basis points as a consequence of large monetary easing regimes throughout the developed world. This has bid up bond prices, decreasing their yields, and pushed investors into equities, widening the gap between each asset class's relative degree of attraction.
The perpetual growth rate of the US economy is entered as 1.8% to reflect the Federal Reserve's projection that was lowered from 2.0% at its September policy meeting. This is a reflection of the fact that ultra-low interest rates and quantitative easing have had the effect of pulling forward growth into the future through the enticement of cheap debt. While this will ideally engineer some form of growth in the present, it comes at the cost of future consumption/investment.
Since this projection is done on the basis of cash flow, we need to multiply the expected earnings yield by the current value of the index to determine a cash flow for the given year. This figure is then multiplied by the expected payout ratio to derive a cash flow.
This is completed for each year of the projection period, comprising five years in total. These cash flows are discounted back to the present by dividing it by a denominator of the sum of the 10-year Treasury yield and equity risk premium (effectively investors' total return expectations in nominal terms) to the power of whatever year in the future we are projecting for. So, for example, if we are projecting a cash flow in three years' time, the denominator is taken to the power of three.
A terminal value is taken as the expected cash flow in the final year of the projection, divided by a discount rate calculated as: 10-year Treasury yield + equity risk premium - the expected long-term growth rate of the economy.
Given that at this point in time the 10-year Treasury yield is higher than the expected long-term growth rate of the economy, the discount rate is higher than the equity risk premium. This reflects that investors will normally expect higher returns on equities when bond yields increase without a concomitant uptick in growth expectations.
This terminal value is then discounted back to the present. We then add this value to the sum of the present value of the other cash flows, which creates the expected intrinsic value of the index.
Based on the inputs covered above, if the S&P 500 were trading off historical equity risk premiums in the 4.1-4.6% range, the S&P 500 would be valued in the 1,550-1,710 range (rounded to the nearest 10 points). This would suggest that the market is anywhere from 22-30% overvalued.
Even if we boost expected earnings growth up to 5% (from 4.3%), the long-run growth rate back up to 2.0% (from 1.8%), the payout ratio up to 85% (from 80%), and discount using an equity risk premium of 4.0% (down from 4.1-4.6%), this gets us to just 1,970, or a 10% overvaluation.
So even in a fairly bullish scenario, the S&P 500 remains more overvalued than at any point in which I've run this exercise in the past several months.
Investors Returns Expectations
If we are to use 4.3% as the annual earnings growth expectation and keep all standard assumptions constant as above, this would imply that the S&P 500 is trading at an equity risk premium of just 3.05%.
This is 73 basis points lower than the reading from one month ago, and 89 basis points lower than the figure from late September. This denotes that investors are continuing to place a high level of trust into stocks despite the increasing attraction of yields in the Treasury and corporate bond market. This would imply nominal annual returns expectations of 5.4-5.5%, only a slight drop from the 5.6% expectations from the days immediately preceding the election.
If we go by historical returns, from February 5, 1971, (the first day all three of the major US stock indices traded simultaneously) to today, the S&P 500 has yielded 7.05% annualized (adjusted for inflation with dividend reinvestment); the Dow Jones Industrial Average (NYSEARCA:DIA) has yielded 6.94%; and the NASDAQ (NASDAQ:QQQ) has averaged 9.03%. If we assume 5.5% annualized for the S&P 500 with 2.0% inflation, this comes to 3.5% in real terms or about a 350-bp drop from historical norms.
With the maturity of the US economy, it's probably unrealistic to expect US equities to return 7% per year going forward in real terms, but with forward returns estimated at sub-4%, it continues to demonstrate how crowded the US equities market is as a whole.
Going by 100-point increments, I believe 2,200 will be a difficult level for the market to hold. With the bloated liquidity levels and negative real interest rates in advanced economies, stretched valuations can be supported to a degree, but it's uncertain how low investors are willing to go in terms of thinking in term of the equity risk premium. Naturally, at such a low value, it would be expected that more investors would be willing to move into Treasuries and corporate bonds. But that hasn't happened on the back of increased inflation expectations due to the policy shifts that are expected under the Trump administration.
Earnings seem to have stabilized from a five-quarter recession (seven quarters in year-over-year terms). The value of the market remained resilient throughout the earnings decline, partially aided by aggressive corporate behavior to repurchase shares and appease stockholders.
Even if investors interpret the political scenario will be a tailwind to the market with lowered corporate tax rates, infrastructure spending, and repatriation of overseas cash, it is still difficult to model 2,200. It requires, for example, a combination of boosting earnings growth to 9.5% from 4.3%, a boost in the perpetual growth rate to 2.0% from 1.8%, and discounted at an equity risk premium of 4.1%.
Despite my interpretation that the market is overvalued, this should not be taken as a "short SPY" thesis, given there is no foreseeable catalyst to drop the index by a double-digit percentage. Valuation by itself is a lousy reason to long or short a security. However, it is a valid reason to stay away, and I do believe going long the index is certainly unappealing from a risk/reward perspective.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
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.