Valuation and Profit Growth
Asset returns can be considered a function of the differential between expected real earnings growth, and realised real earnings growth. This can be borne out by solving for future S&P 500 alpha over short-term corporate bonds, using starting valuation (LERP), realised inflation, and realised EPS growth:
The R^2 of the model is 0.7. This leaves about 30% of the variation of equity returns explainable by other factors, which we will generalize as the variable expected Risk Premium. Sometimes investors are willing to take on more risk to get the same expected returns. But these animal spirits are demonstrably less dominant in explaining valuation than expectations of earnings growth.
Using some basic algebra, we can rearrange the equation, solving for expected EPS growth by using LERP, and substituting realised inflation for the Cleveland Fed’s inflation expectations series:
Using the most recent 5y inflation expectation (1.56%) and LERP (0.75%), we can impute market expectations for EPS growth at 6.34% compounded annually. (The historical average is very similar — 6.39% compounded annually.) This would put S&P 500 EPS at $149 by the end of 2021.
Let’s shift to macro-level profits. We can describe the growth in profits as the growth in nominal national income, combined with the shift in profit share of national income.
The equation works out to:
y = -0.33 + 1.05(x) + 339.88122(z)
…where y is the expected 5y nominal profit growth (CAGR), x is 5y NGDP growth, and z is the 5y shift in the profit:NGDP ratio.
We immediately are struck by how much more significantly change in profit capture explains variability in profit growth than NGDP. The baseline of profit growth is NGDP growth (or very nearly, with a coefficient of 1.05), but the swings are dominated by shift in wage capture. 1sd of NGDP variation works out to 2.3% change in profits, compared to 6% from a shift in profit capture.
Thus, in judging the veracity of valuation-implied EPS growth, our time is far more productively spent analysing the aggregate relationship between labour and the corporate sector than economic growth.
The past 5 years of NGDP growth have compounded annually to 3.65%. If we extrapolate that, to get to our 6.34% EPS growth, we need profit share of GDP to rise around 84bps.
There are a few problems with this scenario. The most immediate is that the profit share of GDP is falling, and precipitously — from its peak of nearly 11% in 2012, it has fallen to almost 9%.
Buying equities here is essentially a bet that on the reversal of this trend toward increased wage capture of GDP.
Rising wage capture is mid- to late-cycle phenomenon. It tends to start when the economy reaches full employment, and last until the end of the economic cycle.
There is in fact no post-war precedence for a reversal of wage capture after it has eclipsed its present expansion before the beginning of the subsequent economic cycle.
As cycles age, the unemployment level shifts lower, and the wage capture rises:
The cycle likely won’t end until the fed ends it. Until then, at the present level of unemployment, the wage share of GDP is gaining about half a percentage point per year — and this appears to be accelerating as the employment market becomes tighter.
As Guillermo Roditi pointed out, profit growth has so far been saved by the rapid decline in interest rates:
With LERP back around 0.75%, the corporate sector can simply borrow to buy back to maintain nominal EPS, in an environment which has seen aggressive clawbacks of wage income capture.
This is one of the reasons that change in labour share and change in profit share are not precise mirror images.
But domestic labour costs are the dominant factor to profitability, and they will continue to pressure margins through at least the end of this cycle.
It is true: there are possibilities where 6.34% compounded EPS growth can occur. Corporations could lobby for more favourable corporate tax treatments. Or they could take advantage of the strong dollar to buy cheap foreign output to lower their costs somehow without reducing export profitability. But, at this point in the labour market cycle, you have to get creative to get there, while little upward optionality seems available in the purchase price of the S&P 500.
What we’re left with is betting on the declining risk premium investors demand for taking equity risk. Put more plainly, a bubble. And, while the monetary policy and economic conditions are eerily similar to the Asian financial crisis (and subsequent .com bubble), speculating on less economical prices is never something we’re comfortable doing.