This post contains the first of several planned recommended readings about “big picture” stock market investment frameworks. These papers challenge common assumptions and myths regarding stock market returns and deliver important conclusions. They are concise and the analysis is rigorous yet accessible to anyone with some background in finance and investing.
I think these papers are about as concise as they can be for anyone wanting to really understand the material.
That said, here’s the summary of this paper’s conclusions: Earnings per share grow slower than the overall economy. They grow slower than the economy by about 2 percentage points. This is because new capital is needed to fund the new companies and initiatives needed to generate economic growth. Total earnings growth is equal to economic growth, but gets diluted by the new capital requirements. It turns out that per share earnings growth is about equal to per capita economic growth.
This may come as a surprise.
Now it’s no surprise that charlatans and misled professionals alike have tried to convince people that earnings and stock prices can grow faster than the economy indefinitely.
For example, it seems every year we read double digit earnings growth forecasts while the nominal economy grows in the low-mid single digits. When you’ve got stock to sell, hope springs eternal.
The most common trick is to selectively use historical earnings data to depict high earnings growth. It’s a simple trick. Choose a starting year when earnings were depressed and choose an ending year when earnings were at a peak. Voilà!
Another trick is to show long-term (but not too long-term, just right for the point being made) earnings growth that includes past periods of high inflation. If an investor has been living in a 2% inflation world and somebody shows him a study showing “long-term” earnings growth of 8% it seems exciting even if much of that 8% was just inflation.
But you and I are too smart for that. We figured out long ago that earnings can’t grow faster than GDP. If they did then eventually earnings would be bigger than the economy. But it has been very easy to take that small step to per share earnings without even recognizing it. I hope this paper sheds some light on the matter.
I would encourage you to read the paper and reflect on it before reading this post further.
OK, I hope you enjoyed the read. Don’t feel bad if you need to read it again. I did.
Per share profits do not capture population growth! After you think about it for a while, that is the gist of the paper. Think about it some more and it really should be obvious. An increase in population cannot be served at current standards using (or employed with) the existing capital stock. There is one exception: If existing enterprises become more productive (and keep the split with labor constant), then earnings will increase at the rate of productivity growth which is equal to the economic growth they’ve generated. New capital to support either a growing population or innovative new products and services will generate the earnings associated with the additional economic growth.
A careful reader may notice the”Real Return” for the U.S. shown in Table 1 is 6.7% while Real GDP Growth was 3.3%. Isn’t the whole paper about how earnings per share trail GDP by 2%. In this table attention shifts to dividend growth, a paltry 0.6% real. But then how can the “Real Return” be more than double the Real GDP Growth?
The answer is that the “Real Return” reflects the earnings as a percentage of invested capital (as well as any change in stock valuation, the P/E multiple, from the beginning of the period to the end). If you initiate and maintain a capital investment of $100 and each year earn $5 then your return in 5%. Let’s say your investment becomes 2% more productive each year (indefinitely) resulting in earnings increases of 2%. Well, your return is 7%.
The point is this: Don’t confuse earnings growth with earnings yield. Your total return from stock market investing can exceed GDP growth. Stay tuned for future posts on this…
Some additional food for thought in no particular order…
Yet here we are. Despite the author’s conclusions ten years and a global financial crisis ago it would seem the strength of established companies and their share of profits has soared while total profits as a share of GDP are at record highs, well above historical norms. Earnings have been growing faster than the economy and I suspect per share earnings have too.
Preferential assignment of property rights, resource constraints, and the degree of competition (conversely, anti-competitive measures) may allow existing capital to extract profits from other creators of wealth.
It is land owners that benefit from population growth rather than existing owners of capital in a competitive market. Even here the benefit is only true to the extent the land becomes subject to constraints that were not incorporated into the original purchase price. Such constraints would include unanticipated development restrictions affecting other land or higher than expected population growth. Remember that land is a factor of production distinct from capital.
I think if there’s any confusion about “Real Return” vs. “Earnings Growth” it in part arises from the fact that all these numbers are expressed as percentages. Yet not all percentages are created equal. If you do dimensional analysis (it’s exciting) you’ll see the return percentage has units of profit-dollars per year (period) per capital-dollars. Since real dollars are fungible you can express it as per year (period). An interpretation is “how much of your capital you can extract in a year”. The earnings growth rate is an annual rate and is = year 2 period dollar earnings minus year 1 period dollar earnings over year 1 period dollar earnings = year 1 period minus year 1 period over year 2 period = unitless. That’s right, they’re both percentages, but one really has units “1/year” while the other is has no units. [Did you know that the P/E ratio has the dimension of time (years)?]