The Persistence of Value & Growth
Growth and value stocks don't stay growth and value forever
This week we are revisiting one of our favorite themes, the persistence of growth, by examining the persistence of valuation multiples.
As we laid out in our previous writings, growth appears to be no more persistent than random chance would suggest. From 1997 to 2021, 58% of companies in the top quartile of sales growth and 49% of companies in the top quartile of EBITDA growth had above-median growth one year later. Random chance alone would suggest 50% of companies in the top quartile should exhibit above-median growth, not far off what the data shows.
As a follow-up to our previous posts, we decided to look at the relative persistence of value and growth and whether that persistence has changed over time. We created a composite value metric to rank companies with over $100M of market cap on a blend of price/cash flow (25%), price/earnings (25%), price/book (20%), EV/EBITDA (20%), and EV/sales (10%). Using our combined value metric, we then split the universe into 4 quartiles (Q1 = cheapest, Q4 = most expensive) each year to measure what percentage of companies moved out of their respective value quartile over the subsequent 12 months.
Consistent with our prior research, we found that roughly 50% of the most expensive stocks (those in the top quartile) remained in the top quartile one year later. Roughly 30% dropped into the next quartile, while roughly 10% fell to the second-cheapest quartile and the cheapest quartile respectively. Since these companies have the highest valuations for the current fundamentals, we can think of this most expensive quartile as a proxy for the most “growth-y” companies.
Figure 1: Persistence of “Expensive” Valuations in the US
Note: Q4 represents most expensive quartile, Q1 represents cheapest
Source: Capital IQ, Verdad research
Unsurprisingly, during periods of economic volatility (2008, 2020) we tend to see decreased persistence of expensive valuations or growth.
The picture overall looks slightly different for the lowest quartile of value, which represents the cheapest stocks. On average, 75% of the cheapest quartile of companies in the US remain in the cheapest quartile one year later, while 17% graduate to the second quartile, and 5% and 3% move to the third and fourth quartile respectively. Within the cheapest quartile, we find evidence that valuation multiples tend to mean revert, albeit gradually. Notably, the persistence of value has increased recently and is near the high end of the range since 1996, as shown in Figure 2 below.
Figure 2: Persistence of “Cheapest” Valuations in the US
Note: Q1 represents cheapest quartile, Q4 represents most expensive
Source: Capital IQ, Verdad research
The gradual mean reversion of multiples in the cheapest quartile is perhaps not surprising, in our opinion, for two reasons.
First, investors are compensated to wait for the mean reversion. The average dividend yield, as of November 2023, of the cheapest US quartile (where our value metrics explicitly did not include dividend yield) was 3%, versus 0.5% for the most expensive quartile.
Second, the cheapest quartile also has the most torque to an upward change in multiples. Companies in the cheapest US quartile have a median EV/EBITDA of 6x, versus -17x for the most expensive quartile. A 1x multiple increase has the biggest beneficial impact on equity returns for the cheapest quartile of companies, with a decreasing impact as the multiple increases, as shown in Figure 3 below. Note that the most expensive quartile has negative EBITDA.
Figure 3: Median EV/EBITDA by Value Quartile and +1x Multiple Turn Impact
Note: Q1 represents cheapest quartile, Q4 represents most expensive
Source: Capital IQ, Verdad research
For the sake of brevity we don’t show the same charts for Europe or Japan, but we found very similar trends to those in the US. Expensive valuation multiples tend to revert quickly, consistent with the lack of persistence in growth, while cheap multiples revert to the mean more slowly.
In aggregate, our findings here are largely consistent with our prior research on the persistence of growth and value. Growth is neither persistent nor predictable, as evidenced by the relatively high turnover among stocks in the most expensive quartile. Concurrently, the cheapest companies tend to see their multiples revert towards the mean gradually over time. Investors in the cheapest quartiles are compensated with significantly higher dividend yields and the most leverage to rising multiples as sentiment improves.