Last week I ran a post about the median stock trading at an all-time high valuation that included this chart from “Millennial Investor” Patrick O’Shaughnessy showing historical EBITDA yields for all stocks in the universe greater than $200 million market capitalization from the period 1971 to date:
I backtested the returns for all three portfolios during the most extreme period in the data (from 1999 to date marked in the red square above). I limited my universe to the stocks in the S&P 500 and lagged the fundamental data by 6 months (so portfolios formed 6/30 in year t use data from 12/31 in year t-1). All portfolios are equally weighted (the 50 stocks in each for the Cheap and Expensive portfolios hold 2 percent of the theoretical portfolio capital at inception and the 500 stocks in the Market portfolio hold 0.2 percent of portfolio capital at inception). Here’s the return chart for the three portfolios:
Here are the return statistics for each of the three portfolios (the cells are conditionally formatted such that green indicates a low yield or a high return, both “good” things, and red indicates a high yield or a low return (both “bad” things):
There are a few things to note in the chart above.
- First, the median stock is the most expensive it has been (in the data) at an EBITDA yield of 8.2 percent (here’s my overview of the research on the enterprise multiple — the inverse of the EBITDA yield). The previous peak was 9.2 percent in 2007, and before that 9.7 percent in 1999. We see something similar in the Cheap portfolio too. It is the much more expensive than average, and is exceeded only by 2007 (15.1 percent), and 2002 (14.7 percent–the all-time high).
- Second, it’s no accident that the worst returns are associated with the lowest EBITDA yields. If we compare the performance of the portfolios we can see that the Expensive portfolio always has the lowest yields and has consistently underperformed the Market and Cheap portfolios (with the only exception being the last year of the dot com bubble in 1999). The Cheap portfolio always has the highest yields and has consistently outperformed the Market and Expensive portfolios (again, with the only exception being the last year of the dot com bubble in 1999).
- If we look within portfolios and compare one year to the next, we can see that, though the relationship isn’t perfect, there too low yields produce low returns. (The relationship persists perhaps because it isn’t perfect.) The worst returns don’t necessarily occur in the year of overvaluation, but they follow closely. For example, the worst yearly performances for the Market occurred in 2008 (-31.1 percent), 2007 (-11 percent) and 2001 (-6.6 percent) and 1999 (-1.9 percent), and those dates roughly conform with the lowest yields.
- Finally, the worst years for the Cheap portfolios were in 2008 (-32.1 percent), 1999 (15.7 percent), 2011 (-8.4 percent), 2002 (-5.4 percent), and 2007 (-5.3 percent), and those dates also roughly conform with the dates that the portfolios held stocks with the lowest yields.
There’s no magic to value investment. Low yields produce low returns. High yields produce high returns. The relationship isn’t perfect. Outlier years like 1999, and 2011 will occur occasionally, but, on average, you’re better served buying Cheap stocks, and remaining cautious during periods when the median stock in the market offers a historically low yield, like right now.
Click here if you’d like to read more on Quantitative Value, or connect with me on LinkedIn.
Tobias,
Do you include financials and utilities in your calculations? Do you use an alternative measure of value for them?
LikeLike
The portfolios include utilities and financials. And no, I don’t use an alternative measure.
LikeLike
[…] Many value investors regard the period since the beginning of 2008 as a difficult one for value strategies. After the bonanza for value in the early 2000s, when a number of value guys made their name, the going has been much tougher in the latter part of the 2000s, and early 2010s. One theory for the lower performance of value–the one to which I subscribe–is that value stocks were much cheaper in the early 2000s than they have been since because the dot com boom was mostly restricted to big, “new economy” stocks. Smaller “old economy” stocks were neglected, and unusually good value. (For more on this, see for example, this post: Implications of All-Time-High Median Valuations). […]
LikeLike
Do these returns include dividends or are they just price returns?
LikeLike
Both include dividends.
LikeLike
Cool.
I have a few other questions if you don’t mind me bothering you. You seem to have been able to pump out this analysis pretty quickly and it looks like you have done a good job.
– How are you able to construct back test portfolios this fast? What software do you use?
– How often do you rebalance these portfolios? or are they simply determined in 1971 carried forward?
– How do you deal with dividend reinvestment? Do you reinvest the dividends into individual stocks or do you allocate to under weighted positions?
LikeLike
[…] « Implications of All-Time-High Median Valuations (Backtest 1999 to date) […]
LikeLike
[…] Tobias Carlisle, “There’s no magic to value investment. Low yields produce low returns. High yields produce high returns.“ (Greenbackd) […]
LikeLike
Hi Tobias,
Another excellent post. Thank you very much!
I wanted to ask how you define the cheap/market/expensive portfolio. Could you share the criteria?
Thanks!
Joris
LikeLike
The Cheap and Expensive portfolios each contain the decile of stocks with the highest and lowest ranking on EBITDA yield, and Market is the S&P 500 equal weight including dividends.
LikeLike
I have read your posts on the best price ratios (and your book), and noted your disdain for ROIC. Have you looked at the best ratios for a longer holding period – say 5 years (+) instead of 1 year? Is ROIC more significant then? Personally I would like to select a long term holding portfolio, may using shareholder yield. Is it better worse that yearly holding of best EBITDA/EV?
LikeLike
ROIC seems to be noise. My recollection is that the results don’t change, or don’t change much, over longer holding periods. Shareholder yield is a good metric. EBITDA / EV has delivered better returns over the period tested, but that doesn’t mean that it always will. Meb Faber’s Cambria offers an ETF holding the best stocks on a shareholder yield basis: http://www.cambriafunds.com/syld/
LikeLike
[…] ………… […]
LikeLike