Q3 Productivity & Wages: A Mixed Bag for the Fed
Labor productivity grew but remains below trend and the labor share keeps declining. More wage disinflation is needed to get wages aligned with 2% inflation.
Today's release of the BLS's preliminary "Productivity and Costs" report for Q3 suggests accelerating labor productivity growth. Of course, we all know this is a very noisy, volatile series so any interpretation should be done carefully. Similarly, the labor share continued to decline this quarter. What is more relevant is the medium-term trends in these series. And we can use the aforementioned trend estimates to interpret recent wage growth, such as, for example, for the Fed’s favorite wage growth gauge the Employment Cost Index (ECI). This note looks at the above in a bit more detail.
Key takeaways:
Labor productivity growth accelerated for the second consecutive quarter, but labor productivity still undershoots its trend estimate.
The pace of trend labor productivity growth slowed down during and after the pandemic, which read 1.3% year/year in Q3 compared to an average trend labor productivity growth rate of 1.9% year/year in 2019.
The labor share contracted again and remains below its trend estimate, which declined by 0.6% year/year in Q3 compared to an on average flat trend labor share estimate for 2019.
The Q3 ECI index for wages of private sector workers went up 4.5% year/year in Q3. This is about 1.8 percentage point above the wage growth rate consistent with 2% inflation given the trend growth rates of labor productivity and the labor share.
The Q3 overshoot in ECI wages of private sector workers was driven by elevated near-term inflation expectations of firms and households and above trend labor productivity growth, both of which will likely continue to push wage growth up over the next quarter.
Labor Productivity
Labor productivity for the non-farm business sector grew 2.2% year/year in Q3, after increasing 1.2% in Q2. This is a positive turnaround in labor productivity compared to the seven quarters that preceded Q2 when labor productivity dropped throughout most of that period. Of course, one swallow doesn’t mean a spring and we need to benchmark this data against its underlying longer-term trend.
Without wanting to resort to some structural model, I base my estimate of trend productivity on a number of different purely statistical approaches as outlined here and use an average of the estimates from these approaches as the trend labor productivity estimate. As such this average reflects the uncertainty with respect to the 'true' trend level of labor productivity.
This estimate of trend labor productivity is plotted in the chart above (orange line), and it suggests that during and after the pandemic trend labor productivity grew at a slower rate than during the pre-pandemic period. For example, trend labor productivity increased 1.3% and 1.4% year/year in Q2 and Q3, respectively, which trails the average trend labor productivity growth rate of 1.9% year/year for 2019.
Comparing actual labor productivity with my trend estimate suggests that after a long period of declines labor productivity turned a corner in 2023 and started to converge back to trend. Extrapolating this development suggests that we should expect to continue to see an above-trend labor productivity growth rate for at least another quarter, as labor productivity is converging back to trend.
The Labor Share
The labor share represents the compensation firms pay their workers as a share of the firms’ revenues. This measure is useful to have alongside labor productivity data, as higher labor productivity in principle should result in higher real wages but the extent to which this happens depends on this labor share. In Q3 the labor share for the non-farm business sector contracted 1.1% year/year, and it has been declining for eight of the ten preceding quarters. As was the case for labor productivity discussed earlier, I use an average across different purely statistical approaches (outlined here) to pin down the trend labor share.
The orange line in the chart above depicts the trend component of the labor share. Throughout 2016-2019 this trend labor share was stable, but it has been on a declining trajectory since 2020. For example, trend labor share growth in Q3 was -0.6% year/year but it was on average flat throughout 2019.
Q3 Wage Growth: Interpreting the Recent ECI Moves
The Employment Cost Index (ECI) is seen as the Fed’s favorite gauge of labor compensation growth, as it corrects for any compositional shifts across sectors (much like the Atlanta Fed Wage Growth Tracker). Yesterday, the Q3 ECI report was published. The most important measure from this report is the ECI for wages of private sector workers (ECIWP), stripping out non-wage labor compensation as well as public sector wages. ECIWP was up 4.5% year/year in Q3, only slightly less than the 4.6% that was observed in Q2.
Much in the same way I do for monthly wage measures, one can use the trend estimates for labor productivity and the labor share outlined above to get a trend wage growth estimate consistent with the Fed’s 2% inflation target. This simply equates to 2% plus the year/year growth rates implied by the earlier discussed trend estimates of labor productivity and the labor share. The chart above contrasts this trend wage growth measure with actual ECIWP wage growth, and we clearly see that in recent years actual wage growth overshot this trend value. More specifically, for Q3 ECIWP wage growth equal to 4.5% was about 1.8 percentage point above the rate of growth that would be consistent with 2% inflation.
So, what drove the recent overshooting of wage growth compared to the pace consistent with 2% inflation? To do that I decompose the gap between actual ECIWP wage growth and its 2% inflation consistent trend using the deviations of actual growth relative to estimated trend growth for both labor productivity and the labor share, the gap between year-ahead inflation expectations of firms and household compared to the 2% inflation target, and an unexplained residual. The “Main Street” inflation expectations are extracted from a number of surveys, as I described in an earlier post.
The publication of year-ahead expectations from the Conference Board’s Consumer Confidence survey earlier this week, the aforementioned year-ahead “Main Street” expectations for October now run at about 3.5% in terms of PCE inflation, up from about 3.2% in the previous month (chart above). Stepping back a little further, the cooling in near-term “Main Street” inflation expectations grinded to a halt by mid-2023 at around 3.2% and since then started to creep up again.
The chart above shows that, not surprisingly, during and right after the COVID lockdowns the ECIWP gap relative to the inflation target-based trend was heavily affected by (often offsetting) large swings in labor productivity and labor share gaps. Since mid-2021 the ECIWP wage growth gap has been persistently positive (above a pace consistent with 2% over the medium term), as the inflation expectations gap turned positive in a sizeable way.
The easing in the ECIWP gap in between 2023Q1 and 2023Q2 can be attributed to some easing in the inflation expectations gap. However, since 2023Q2 the overshoot of ECIWP wage growth stabilized at around 1.8 percentage point. Focusing on Q3, the inflation expectations gap did decline from +1.7 percentage point in Q2 to +1.2 percentage point, whereas the labor productivity growth gap contribution flipped from -0.2 percentage point to +0.9 percentage point in Q3. With a -0.5 percentage point labor share gap in Q3 this means that labor productivity effectively contributed about half a percentage point to the wage overshoot next to the push from elevated “Main Street” inflation expectations. So, essentially of the 1.8 percentage point wage growth gap observed for Q3, about 0.5 percentage point of it was owing to that part of above-trend labor productivity growth that actually accrued to workers.
As discussed above, though, there are tentative signs that beyond Q3 “Main Street” inflation expectations started to creep up again. Also, labor productivity growth beyond the next quarter is most likely to slow down again as labor productivity approaches its trend value. have eased further and rather remain stuck above inflation target levels. Thus, going into Q4 and beyond, we should not expect to get a lot more easing in the ECI wage growth index for private sector workers as long as inflation expectations of firms and household remain elevated relative to the Fed’s inflation target, which should be a worry for the Fed.