Over the recent years, the link between stock returns and the economic news flow (epitomized by ISM for instance) has weakened significantly (see chart below). On explanation is that once the recovery is well under way, the ISM stabilizes above 50 and valuation growth provides the bulk of equity returns.
Here I explore the relationship between EPS (Earnings Per Share) and the expected - not observed - changes in economic growth. I use the variation in the consensus forecast of the nominal GDP to gauge the impact of changes in the perception of economic activity on EPS growth. Given that profitability depends on the volume of sales and the pricing power, I compile real GDP and CPI forecasts (there is no forecast available for the GDP deflator in the database that we use) to get a proxy of expected GDP growth.
The consensus forecast is based on calendar year data while EPS are generally published and analyzed on a 12-month forward basis. Hence, to avoid any comparison of apples with oranges, I use a combination of current and 1-year ahead growth forecasts to get a rolling estimate of the average nominal GDP growth expected for the next 12 months. The chart below compares this series to the expected year-over-year (yoy) growth rate of 12-month forward EPS.
The relationship has been very strong (orange area) during the financial crisis. The relation is more noisy for the rest of the time but we can identify long lasting sub-periods of positive links between both series.
To dig deeper into the relationship I then compare the 1-year change in the average growth expected for the next 12 months to, once again, the year-over-year (yoy) growth rate of 12-month forward EPS. The result is given in the chart below. The link appears much more robust. It suggests that an upward revision in the Consensus for GDP growth leads to an upward revision in 12-month forward EPS growth. The level of the year-over-year (yoy) growth rate of 12-month forward EPS is dependent on the acceleration / deceleration of expected GDP growth.
Given the strong concentration of points in the [-1;+1] area of the scatter plot above, I would rely on this link essentially for significant changes in the consensus forecast. As can be seen below, there is a strong non-linear relationship between the revision of GDP growth and the adjustment (1-year change) in year-over-year (yoy) growth rate of 12-month forward EPS.
Over the last months, the average nominal GDP growth expected for the next 12 months has fallen from 5.25% to 3.63%. Given that the amplitude of the adjustment is out of the "cluster of insignificance" of the chart above, it would call for a downward adjustment in the yoy growth rate of 12-month EPS. Interestingly enough, the current 12-month forward EPS yoy expected growth is 6.2%, which is close to what the chart below suggests.
What's next? Given that my views on the U.S. economy are distant from those of the consensus (2.2% vs. 2.7% for real GDP growth in 2016), the odds for a sharp rebound in the expected growth rate for 12-months forward EPS are clearly limited. This should be enough to justify a downward revision of the expected return for U.S. stock indexes in the next few quarter. If we add the downward pressure on the buyback yield that may accompany the fall in the self-financing ratio, there is clearly another reason to be less exposed to U.S. Stocks.
Bottom Line: The relationship between EPS growth and GDP growth expectations is not always robust. It tends to be reinforced in periods of turmoil or sharp reversal in economic expectations. I have shown that the recent revision in U.S nominal growth prospect is significant enough to call for lower expected returns for U.S. stock indices.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.