Between 2004 and 2014, productivity of American workers in the nonfinancial corporate sector averaged 1.2% annually, but real wages grew only by 0.25%. That unprecedented “productivity-wage gap” went straight into the bottom lines of companies, creating historically high profit margins and historically low real wages.

real wages profit margins

This gap has been closing over the past three years and may have just three years left before the pendulum could swing the other way, according to an analysis by Goldman Sachs.

In the second quarter of 2014, the 10-year average annualized growth rate of labor productivity fell to 1.09%, its lowest level since World War II. It has since fallen further to 0.85% at the end of 2016. On the other hand, real wages are rising 3% year-on-year, up from a post-crisis low of roughly 1.5% in 2011.

Currently, the ratio of real wages to productivity is closing at a rate of 1.5%. At this pace, corporate profit margins would cumulatively decline by as much as 25% — to 8.6% from the current 12.5%. Even if the gap closed at a slower 0.5%-1.0% per year pace – assuming real wages won’t outpace productivity growth – profit margins would decline to between 9.9-11.2%.

“With productivity stagnant, inflation contained, wages rising, and the US economy already past full employment, the trends are all wrong-way for corporate profits margins,” Goldman Sachs in its Economics Research newsletter.

Based on National Income and Product Accounts (NIPA), nonfinancial corporate profit margins have already declined to 12.5% in the first quarter of 2017, from a recent peak of 15.5% in the third quarter of 2014, authors Charles P. Himmelberg and James Weldon pointed out.

“The current level of margins is still well-above our “counter-factual” estimate of 8.6%, which is where we estimate margins would be today if real wages had not fallen behind productivity in the first place. Even if only half of this margin gap closes, it implies that margins fall further to 10.6%. Put differently, if the growth rate of real wages exceeds the annual growth rate of productivity by 0.5%-1.0% over the next three years, NIPA profit margins will see a further decline to between 9.9% and 11.2%. The only interesting question in our view is, how much lower?”

Goldman Sachs’ analysis is based on NIPA numbers but the bank believes its conclusion would equally apply to the universe of S&P 500 companies because the two are closely correlated, and “both measures are similarly exposed to the business cycle forces driving productivity, prices, and wages.” Consequently, the historical behavior of profit margins in NIPA data provides a solid indication of the macro forces confronting large public companies as well, it said.

At the heart of Goldman’s analysis is the following equation:

%?(Margin) = a[%?(Productivity) + %?(Price) ? %?(Wage)]

Here, “price” means output price, “productivity” means average labor productivity, “wage” means compensation per hour, and “margin” means the markup over cost as a fraction of price (that is, price minus unit cost over price).

“This equation lays bare the intimate relationship among real productivity, nominal prices, nominal wages, and profit margins. It shows, for example, that an increase in nominal wages cuts into margins only if it is not offset by either an increase in price inflation or an increase in real labor productivity,” the bank said. “Conversely, in what is likely a more accurate description of the near-term outlook, it also says that if productivity growth is stagnant, and inflation pressures are subdued, then an increase in nominal wage pressures will likely translate into downward pressure on margins.”

Is there visibility on which sectors might be more affected than others?

Real wages have kept pace with productivity growth in retail, financials, services and transportation (which together represent a third of the market cap in the S&P 500). But the divergence is most visible in manufacturing (both durables and nondurables), information and mining, which collectively represent roughly 60% of S&P market cap, the bank said.

“And while we resist the temptation to speculate much further on the various reasons why wages may have deviated from productivity in the first place, there may be a clue in the fact that manufacturing, information and mining were all exposed to the China ‘trade shock,’ whereas retail, financials, services and transportation were not,” the authors said.