When writing about a stock, it’s always easier to use a more conventional estimate so you don’t have to argue with people about your “model”. This is one way in which writing about investing – showing your work publicly – is bad for your own investment process.
Regarding macroeconomic variables – the expected rate of nominal GDP growth, the Fed Funds Rate, the price of oil, etc. – the conventional assumption is usually the most recent reading modified by the recent trend. So, if the price of oil was “x” over the last 3-5 years, some people think 0.75x is reasonable and some think 1.25x is reasonable. But, estimates of 0.5x and 1.5x and beyond are considered unreasonable.
This “recency” issue makes it hard to have macroeconomic discussions. The level and trend of the recent past becomes the conventionally accepted answer for future estimates. The problem for you as an investor is that any application of conventional wisdom is useless. If the conventional wisdom is that the Fed Funds Rate will be 1.5% fairly soon – and you also believe the Fed Funds Rate will be 1.5% fairly soon – you may be correct, but your correctness will do you no good. Stocks are already “handicapped” according to the conventional wisdom.
The only macro assumption that can do you any good as an investor is a belief that is both:
1) Correct and
2) Out of step with conventional wisdom
Sometimes, I do have such beliefs. But, they’re never any fun to write about. So, I try not to write about them. However, in terms of my actual investing behavior – I can’t help myself from buying something I believe to be incorrectly “handicapped”. So, you will sometimes see indications of macro assumptions in my actual investments even though they don’t appear in my writing.
Fed Funds Rate
When doing my own (private) work on Frost, I can assume a normal Fed Funds Rate of 3% to 4% without any problem. But, when writing about Frost for others – I have to spend a ton of time justifying something I think seems obvious. A lot of my Frost report was wasted on discussing the Fed Funds Rate instead of discussing Frost. In my own head: I spent very little time worrying about where the Fed Funds Rate would be and when.
In some cases, the mental toll writing and defending such justifications takes on you just isn’t worth it. So, you don’t write about that topic. For example, people who discussed stocks with me privately – like Quan – long knew that I was using assumptions of $30 to $70 a barrel for oil even when Brent was trading at $110 a barrel. In fact, since I started this blog 11 years ago, my assumptions for the price of oil have never changed. My process has always been “plug in $30 a barrel” and see what you get and then “plug in $70 a barrel” and see what you get. Assume the worst for the stock you’re looking at (so $70 for an oil consumer and $30 for an oil producer) and see if there’s any margin of safety left in the stock at that price per barrel for oil. So, now you know why I’ve never bought an oil producer and why I wrote about a cruise line – Carnival (CCL) – on this blog. I was only putting $30 a barrel into my assumptions for oil producers. And I was only charging Carnival $70 a barrel even in years when it was spending the equivalent of $110 a barrel for its fuel.
This is one where you may notice a difference between what I practice and what I preach. If you look at what I preach – I don’t talk much about inflation. And when I do, I say something like: “inflation may be 2% to 4% a year, let’s take the low-end of that for our assumptions about this stock’s growth rate.”
Now, what do I “practice” though?
Well, over 50% of my portfolio is in NACCO (NC). All of that stock’s earnings are indexed to inflation (they are cost plus coal supply contracts). I have about 15% of my portfolio in BWX Technologies (BWXT). A lot of that stock’s earnings are indexed to inflation (they are cost plus shipboard nuclear reactor contracts – among other things). So, that’s at least 65% of my portfolio that’s “cost plus”. Finally, I have 28% of my portfolio in Frost (CFR). Frost’s earnings are basically tied to nominal interest rates. So, that’s about 93% of my portfolio – all U.S. – that is in businesses that don’t put in much tangible capital up front and do insist their customers pay them more (as inflation happens). Finally, the remainder of my portfolio is in a company in Japan. So, although I never really write about inflation in the U.S. – you’ll notice that I’ve invested in things where I won’t be hurt by inflation in the U.S. Most U.S. stocks will do worse in periods of inflation than NACCO, BWXT, and Frost will.
Does this mean I’m predicting higher inflation in the future?
No. But, a lot of investors are putting 2% type inflation numbers into their assumptions about stocks when I’m not sure that – 10 years from now – 2% inflation is any more likely than 6% inflation in the U.S. Right now, the rate of inflation is not much different from where it was in the early 2000s or the mid-1960s. Ten years after the early 2000s, the inflation rate hadn’t really moved much. And ten years after the mid-1960s, it had gone from less than 2% to about 11%.
I’m not sure knowing what the inflation rate is today or what the Fed Funds Rate is today is helpful in predicting where either will be 10 years from now.
The same is true for P/E multiples. I’m not sure that knowing what the average P/E multiple – or Shiller P/E multiple – is today will help you predict what the average P/E multiple will be in 10 years. Maybe 50% of the time today’s average P/E multiple and the average P/E multiple 10 years from now are about the same. But, the other 50% of the time – today’s average P/E multiple and the average P/E multiple 10 years from now are completely different (just as a 2% inflation rate and an 11% inflation rate are completely different – yet in U.S. history those two readings were separated by just 10 years of time).
The only times I can think of where I’ve incorporated some sort of macro assumption into my investment decision making is where I’ve assumed that some macro variable will tend to be more like its long-term past (that is, the last 30-50 years) rather than its recent past (that is, the last 3-5 years). In other words: I can only make useful macro assumptions when some macro variable has been quite different over the last 3-5 years than the last 30-50 years and the conventional wisdom is that the recent past – rather than the long-term past – is right.
P.S. – China
Some of you who read this blog religiously know “I don’t invest in China”. That’s a rare hard and fast rule for me. Another hard and fast rule for me is: “I don’t write about why I don’t invest in China.”
I’m willing to be more controversial in my investment decisions than my writing.
I don’t see a point in writing about why I don’t invest in China. I’m not an expert on China. Anything I write on the topic would be controversial. And, more importantly: anything I write on the topic would be unlikely to change your mind about whether you should invest in China. So, it’s just not a topic I’m going to touch.