Short-Cuts and Short-Termism
Look, the basic premise of asking “are stocks overpriced?” or “is the market overpriced?” is a short-term trading mentality. If you’re a long-term investor, transitory swings in Mr. Market’s emotional pendulum shouldn’t make any difference. You’re better off riding this emotional wave through its ups and downs and coming out on top in the long-term. It’s impossible to time the markets successfully over time. The greatest investors in the world – with verifiable long-term track records – will tell you that.
The overpriced-market phenomenon is relevant if you’re a) short-term trader and b) you constantly trade asset classes via index ETFs. There are some Global Macro hedge funds that do this kind of thing. But chances are that you’re not a Global Macro fund manager. If you do buy asset classes (odds are that this is how your financial advisor thinks), the question of market-overpricing is almost irrelevant because your strategy is likely to be a) long-term (5 years or more) and b) you keep putting in money to work every month or every few months.
So, short-termism is the source of that pesky question. The other source is generalizations. We humans tend to generalize far too often. We take short-cuts in making judgments or making sense of a chaotic world. This tendency to make blanket statements, to have an answer for everything, leads to questions like “is the market overpriced?”
However, as irrelevant as the question might be to most long-term investors, it is comforting to “know it all”. And it is comforting to know the environment in which we operate. You might ask, “doesn’t the overall environment affect valuations of individual assets?” Yes, but in the short-term. You could argue that Mr. Market’s overall exuberance pushes up prices of ALL assets – the rising tide that lifts all boats. But it’s impossible to quantify the effect of that tide on each individual asset. Academics and even some practitioners have fallen for scientific-sounding concepts like Beta (a measure of the effect of market swings on to individual stocks). But in our experience, and in the experience of the legends of investing, Beta is just an academic concept born out of a tendency to slap on a formula to everything. At best, its usefulness is short-term.
“Beta and Modern Portfolio Theory and the like – none of it makes any sense to me…how can professors spread this? I’ve been waiting for this craziness to end for decades. It’s been dented and it’s still out there.” – Charlie Munger
History has taught us repeatedly that bouts of optimism are followed by bouts of pessimism. Playing this game of musical chairs is highly risky because nobody really knows what they’re doing.
We find it much more useful to judge overprice-y-ness or underprice-y-ness down at the “micro” level – individual stocks (and bonds if we ever get to it). We have devised an efficient way to estimate overprice-y-ness for individual companies and stocks over the last few months. You would have seen this in action in our Watch List, in which we categorize companies and their stocks using our own Investability Rating. This helps us narrow down our hunting ground. But, we wondered, if our bottom-up method answers that pesky top-down question. This exercise forced us to make an adjustment in the way we operate our Watch List.
In Investing, one must always practice Kaizen. There is no “perfect” process to strive for, but there are always ways to refine our process through experience and observation. Our attempt to answer the most popular question in investing is another step on our Kaizen path to becoming the best we can be in our craft.
The Industry Toolkit
If you watch a lot of TV or read popular financial newspapers, you’ll already know some of the usual suspects in measurements of overprice-y-ness. These types of stories are aired/printed almost daily because they catch a lot of eyeballs. Media is in the eyeballs business, not in the performance business. Sadly, some of these questionable measures are also used by investment managers, who are supposed to be in the performance business.
Here are some of the “industry” measurements used to answer our pesky little question about market overprice-y-ness:
- Valuation Ratios like P/E or Price/Book
- Historical Prices
- Relative Value against other asset classes
- The so-called Buffett Method
It’s tempting to use a “market-wide” valuation metric like P/E Ratios. But we should remind you that these ratios are, in fact, bottom-up metrics that are aggregated company by company, stock by stock. So, when someone says that the S&P 500 is trading at a P/E of 21X and is therefore overpriced, that 21X P/E ratio is a weighted average aggregation of all the 500-odd stocks in the index. This is where “average” or “weighted average” doesn’t make much sense to us. How does one use that number if they’re not buying index ETFs to make money in the short term? To quote an old statistics joke, “A man with his head in a furnace and his legs in a freezer will be quite comfortable, on average”.
When they say, “it’s overpriced at 21X”, they have a rough marker in mind. Conventional wisdom suggests that the threshold should be 15X. This threshold is usually based on some sort of long-term average. So, what they’re saying is that compared to the past (the last couple of decades or more), stocks are overpriced because they’re trading at 21X earnings. At The Buylyst, we don’t take any of this seriously. This type of measure is riddled with flaws. Here are 2 complaints we have:
- The long-term average may be irrelevant. Think about what made up the S&P 500 in, say, 1990. It’s a totally different set of companies, stocks, economic conditions etc. Why 1990? Even more recent comparisons will reveal that the S&P 500 was a different beast in 2012 or even 2014. The FANMAG (Facebook, Amazon, Netflix, Microsoft, Apple and Google) phenomenon has been the tail that wags the dog over the last couple of years. They were never so dominant even 5 or 6 years ago.
- Earnings are a terrible measure of profit. This sounds counterintuitive but it’s true. Earnings can be wildly different from cash profit because accounting rules allow it. Moreover, the “E” in P/E is usually backward looking. If a “forward 1 year” E is used, it’s based on very precise (but usually not accurate) forecasts of earnings by a handful of Wall Street analysts (some with questionable track records of returns).
Thankfully, methods of estimating overprice-y-ness get a bit more sophisticated. One such method is called Relative Value. It is what it sounds like – we can measure certain equity market metrics relative to metrics for other asset classes. One of the most popular relative value measures is Earnings Yield vs. the 10yr US Treasury rate. This measure is quite intuitive. Earnings Yield is the reciprocal of the Price/Earnings ratio. It measures a theoretical yield on the “principal” value invested in the stock (your purchase amount). They’re essentially measuring “would we rather earning a yield of 5% in stocks or 1% in bonds?” They compare this over time to make calls about overprice-y-ness.
We have the same complaint with that measure: how relevant is the long-term average “spread” or premium in today’s times when Central Banks have a do-whatever-it-takes approach to grease the economic wheels? It hasn’t even been a decade since the US Federal Reserve has officially specified an Inflation Target! We can’t take a long-term Equity Premium average seriously. Returns in equities are driven by the earning power of underlying companies, not by quantitative measures of equity risk premia.
Then there is a so-called Buffett Method of measuring the total market-capitalization of stocks vs. the GDP of the country. We say “so-called” because we’re not sure Buffett still uses this. We suspect that his measures have adapted to the changing times. This measure is a sort of P/E ratio, but on a very macro level. Market Capitalization represents the total value of firms in the country. GDP measures the annual income of the country. Again, some long-term average threshold must be assumed to judge overprice-y-ness of the market at any given point in time. That has its pitfalls, not to mention the inherent flaws of GDP as a measure of national income.
Overall, we take all these measure with a (big) pinch of salt. We certainly don’t rely on them to accurately describe any level of irrational exuberance or despondency in stocks. So, we had to come up with our own method that is consistent with our style of long-term, focused, sleep-well-at-night-investing.
The Buylyst Investability Ratings
Over the past few weeks, we’ve delineated the method underlying our Watch List. We estimate 2 main attributes of each of the companies/stocks in our Watch List:
Ultimately, we want to generate high returns (double-digit annualized) while sleeping well at night. To do that, we hope to find stocks that are priced modestly compared to the earnings (free cash flow) growth power that the underlying company possesses. How do we measure this earnings growth power?
Well, revenue needs to grow but it needs to be believable; and there needs to be significant operating leverage in the portfolio. Operating leverage means that much of that incremental revenue growth will trickle down to free cash flow. So, free cash flow growth should outpace revenue growth. This is the scenario we need to believe if we invest in a company.
All this is built upon the principles of Expectations Investing. We found a simple, numerical way to answer the question: “what do we need to believe to buy this stock?” and “is it believable?” We just reverse-engineered the process. Our Watch List streamlines our search for investment ideas. It’s a numerical filter. The real work begins when we start digging into the subjective reasons why an “implied revenue growth rate” is believable or not.
There is a subtle underlying concept embedded in our Watch List. Our Watch List results are relative – meaning that stocks/companies are rated High or Low depending on the rest of the universe. Until yesterday, our final Investability Scores were rated on a scale – stocks in the top 25th percentile were rated as “High” investability. Stocks in the top 50th percentile (up to 25th) were rated “OK”. Everything else was rated “Low”. This was our brand of “relative value analysis”. The implicit assumptions behind this approach are:
- It’s always a good idea to find the BEST deals available, regardless of the market environment because…
- …it’s always sunny somewhere – good deals can be found somewhere in any market environment.
But when someone asks us the question, “are stocks overpriced?”, a purely relative method doesn’t work. To answer that question, we need some absolute parameters.
Lines in the Sand
Let’s put some absolute markers down to help us in our quest. As a reminder, we rate our investment universe (our Watch List) based on 2 attributes: Believability and Scalability.
Our Believability Factor compares our estimate of “revenue growth we need to believe to buy the stock”, with the company’s past revenue growth performance. It’s imperfect but it does the job. Our Scalability Factor enhances the Believability Factor. It simply breaks down the company’s cost structure into Fixed and Variable. A scalable business has a higher percentage of its operating costs in Fixed. That way, beyond a certain point, each dollar of incremental revenue flows down almost entirely to free cash flow. We love these kinds of businesses. Software companies generally have this trait – most of their operational costs are Fixed (mostly R&D and Cost of Labor). So, IF revenue grows fast, there comes a point when most of those extra dollars flow right through to Free Cash Flow.
We like that sort of business because the market tends to underestimate this sort of operating leverage. Of course, we’re assuming that a company’s cost structure will remain the same in the future. In some cases, upon digging in, we find that the cost structure is likely to change. Etsy, which we analyzed recently, is a good example – high Believability and Scalability, but our expectation of increase future fixed costs (marketing) dampens both those factors.
Let’s put some lines in the sand. To get closer to an answer to that pesky question, let’s have fixed thresholds for both the Believability and Scalability factors. Now that we’ve been maintaining our Watch List for a couple of months, and have analyzed hundreds of companies/stocks over many years, we have the confidence to stick our necks out and can put down these simple markers:
- Believability Factor > 0.8 AND
- Scalability Factor > 0.4
This may become clear with an example. Let’s say Company ABC’s 3-year Annualized Growth Rate was 20%. And let’s assume that our estimate of “future revenue growth (annualized) we need to believe to buy the stock” is just 10%. The Believability Factor would be 2 (20% divided by 10%).
The Scalability Factor is simple. Let’s say that Company ABC’s revenue over the last 12 months was $100 million. Let’s assume it’s Variable Operating Costs were $20 million, and its Fixed Operating Costs were $40 million. Total Operating Costs, therefore, are $60 million. The Scalability Factor, then, would be 0.67 – Fixed Costs as a percentage of Total Operating Costs is 40/60 or 67% or 0.67.
Company ABC would carry a High Investability Rating. Believability Factor = 2 and Scalability Factor = 0.67. Both comfortably surpass our lines in the sand.
This more “absolutist” method is a shade better than our current Watch List method in 3 ways:
- It eliminates the need for us to invent a formula that combines Believability and Scalability into one single Investability Rating.
- It doesn’t require us to make distributional assumptions about our Watch List. We’re not required to use Mean, Median or Percentiles in categorizing stocks has high or low ratings.
- It doesn’t force us to rank stocks in a deterministic, formulaic way. If we do this, we tend to limit our thinking. Investing is an art, not a science.
In fact, keeping believability and scalability separate gives us comfort that we’re not disproportionately rewarding or penalizing any companies just because they excel or lack in one measure. We’d rather be approximately right rather than being precisely wrong.
So, are stocks overpriced?
Let’s work with the S&P 500 – as a proxy for stocks in general. With our new lines in the sand, let’s see how the S&P 500 (ex Financials) looks on these two factors:
Now, we’re getting somewhere. The stocks in the top-right quadrant are the highest rated – their Believability Factor is more than 0.8 AND their Scalability Factor is more than 0.4.
The bottom-left quadrant is the area to avoid. These stocks have low Believability AND low Scalability factors. The other two quadrants satisfy one criterion or the other. They’re rated as “Medium”.
Instead of calculating average or median investability ratings, we will now display the investability of the market in this way:
Distribution of The Buylyst Investability Ratings:
The important number here is the Percent of stocks rated “High”. In simple English, this pie chart tells us: 10% of the companies have believable or reasonable revenue growth expectations priced into their stocks. Investing in stocks rated “Medium” will need a big leap of faith that the future will be much brighter than the present – they are low on the Believability Factor or the Scalability Factor. Stocks rated “Low” are low on both the Believability Factor and the Scalability Factor – they’re just not consistent with our brand of sleep-well-at-night investing.
This is a more intuitive measure of Investability than our previous system. Can we say if the market is overpriced now? We can’t say for sure. But if we track this pie chart over time, we can say whether the market is overpriced or not.
Here’s a fun exercise: How will this pie chart look if the market suddenly crashes by 20% next month? Assuming the same fundamental company data about revenue and costs, here’s how the pie chart would shift.
The number of stocks with High Investability Ratings will increase by about 4%. That's about 16 more stocks in our hunting ground. Not bad. The weekly, monthly, yearly shifts in this pie-chart will depend on:
- Changes in stock prices – happens daily
- Changes in company fundamentals every quarter – revenue, costs, cash flow etc.
- Changes in each company’s Believability and Scalability factors.
You could argue that the market is too rich now with only 10% of stocks with High Investability Ratings. In a month from now, once more quarterly earnings reports flow through our database, this number could dip to 5%. Then your case for “the market is overpriced” gets much stronger.
We will begin tracking this pie chart over time, specifically the number of companies in the S&P 500 with High Investability Ratings. If the number dips, the market is much more expensive. If the number increases, it means there are many more deals worth a look.
One last thing: Remember that our Watch List methodology is based on the principles of Expectations Investing. This means we’re not making any forecasts – we’re simply backing into revenue growth expectations that are already built into the stock price. This, in our opinion, is miles better than forecast-driven measures of market overprice-y-ness such as Price to Forward Earnings ratios measure through time. Forecasts tend to be very precise but very inaccurate. In our method, we’re literally tracking the growth expectations built into a stock price. We’re not forecasting anything. Whether these expectations are too high, however, can only be judged through time.
If we’re ever in a situation when there are almost no stocks left with High Investability Ratings, we can confidently say that the market is too rich. Until then, we can only say the market is too rich or not – compared to last week or month or year. This the best measure of overprice-y-ness that we know.
This is still a relative measure. It is impossible – for anyone – to come up with an absolute measure of overprice-y-ness because of the inherent nature of investing, which is ever-changing with no immutable governing formulas or laws. We hope this puts your mind at ease. Remember, it always sunny somewhere. At The Buylyst, we help you find those spots.
Many Happy Returns.