July Joyride

The July jobs report had a strong showing, with most labor market measures posting solid gains. The Bureau of Labor Statistics' estimate of a huge 943,000 jobs added to the rosters in July (the fastest pace since last August) and sharp upward revisions of a cumulative 119,000 in prior months brushed aside market fears that the economy was losing steam.

We like to look at the three-month average pace of jobs growth, which is less affected by seasonal adjustments during the COVID-19 era. The three-month average (May-July) also shows a healthy pickup in the pace of job gains--at 832,000. If the current pace were to hold steady (an optimistic scenario), we would get back to the pre-pandemic trend path in the summer of next year (see chart 1). It is more likely that the low-hanging fruit will be picked in the next few months and the pace will start slowing closer to 250,000 by next year. (That would still be a healthy pace once temporary layoffs normalize and some semblance of normal operating capacity is sustainably in place.) Any effect from the latest increase in COVID-19 cases, stemming from the delta variant, is a risk to jobs growth in the coming months.

Chart 1

Payrolls have reclaimed about 75% of the jobs lost during the pandemic, with 5.7 million jobs, or 3.7%, still short of precrisis levels (of February 2020). Leisure and hospitality (up 380,000) and the government sector (up 240,000) did the heavy lifting, with almost all other sectors also contributing to added jobs, albeit at a smaller scale (see chart 2). Leisure and hospitality added, on average, 364,000 jobs in the last three months since the economy has been in full reopening mode. Seasonal factors affected job growth in the government sector, reflecting state and local education. There were fewer layoffs in state and local education this July than what normally occurs in a typical year given that last school year was disrupted by the pandemic and fewer employees were on the payroll to lay off (as shrewdly pointed out by the president's Council of Economic Advisers).

Chart 2

Chart 3

Not surprisingly, given supply constraints in the jobs market, businesses now rely on fewer workers to meet demand, with hours worked having recovered an even larger share of their pandemic drop, which bodes well for third-quarter GDP growth, following 6.5% annualized real GDP growth in the second quarter. The overall index of hours worked was up 0.6% month over month in July (up 6.5% annualized)--above the average of the second quarter--suggesting upside risk, for now, to our 5.6% quarter-over-quarter annualized GDP growth forecast for third quarter.

Demand for labor outstripping supply is reflected in wage gains as well. The average hourly wage increased 0.4% over the month, up 4.0% for the year. However, the July wage gains seemed to be concentrated in leisure and hospitality. With wage pressures not broad-based, we believe that it won't be enough to force the Fed to take away the punch bowl sooner than it had earlier signaled. The Atlanta Fed's wage tracker, a measure of wage growth that doesn't have the compositional issues the Bureau of Labor Statistics' (BLS) measure does, was 3.2% as of June, versus 3.8% a year ago and 3.9% two years ago (see chart 4).

The Fed also wants to see labor force participation regain the workers lost during the pandemic. The labor force participation rate rose 0.1 percentage point to 61.7% in July, though it's still 1.6 percentage points lower than in February 2020 and a concern for the Fed. While the plunge in the unemployment rate to 5.4% in July was music to our ears, when considering all the workers who dropped out of the labor force since February 2020, the adjusted unemployment rate stands at 7.5% (see chart 5).

Chart 4

Chart 5

Labor force participation has proven to be highly cyclical in the U.S., but with a significantly longer-lived response than the unemployment rate, according to new research from the Federal Reserve (Cajner, Coglianese, and Montes, 2021). That has led to one previously underappreciated measure of labor market slack--the employment-to-population ratio (EPOP) of prime-age workers (ages 25-54)--gaining prominence since last September, when the Fed stressed "maximum" employment (versus "full" employment) in its policy rate reaction function. Prime EPOP, which is less affected by changing population composition and participation rates of folks choosing to retire or stay in school, was 77.8% in June, below 80.5% when the COVID-19 downturn began (translating to approximately 3.4 million missing prime workers).

We expect the jobs market to heal further into the fall as work conditions return to normal, extended unemployment benefits expire, schools reopen, and people return to work, particularly in cities. Among the unemployed, the number of people on temporary layoff should fall further over the next couple of months.

The number of people on temporary layoff declined to 1.2 million in July from a high of 18 million in April 2020 but is near 500,000 above its February 2020 level (see chart 6). The number of permanent job losers also declined, by 257,000 to 2.9 million in July, but is 1.6 million higher than in February 2020. Folks who are unemployed for permanent reasons and have been unemployed long term (defined as 27 weeks or more) are the ones at greatest risk of being left behind for a longer time.

Chart 6

Indeed, we expected the jobs market dating game would finally see more matches in September, but the delta variant may have thrown a wrench in those plans. COVID-19 cases are already 4x higher than the week when the July BLS jobs survey was conducted. In contrast to the BLS, the LinkedIn workforce survey saw July hiring down 5.8% from June, perhaps capturing some of the delta effect missed by the BLS survey.

Nevertheless, the solid July BLS jobs report may have given just enough confidence that the jobs market has made "substantial further progress" toward the Federal Reserve's employment goals, increasing the possibility of a taper signal at the Fed's Jackson Hole meetings, followed by more specificity at its September policy-setting meeting. However, during his July Federal Open Market Committee press conference, Chair Jerome Powell highlighted that "there has tended to be less economic implications from each (virus) wave," which he believes is "not an unreasonable expectation for the delta variant," suggesting that it may have less influence on their policy decisions than previously.

We now see the tapering of quantitative easing starting around the end of 2021 and lasting for about one year, when the Fed would begin holding its asset holdings constant. (At that point, new bond purchases will hit zero, and reinvestment from maturing bonds and prepayments will hold the size constant.) That will clear the way for the Fed funds rate to lift off from the effective lower bound in early 2023, with an eventual long-run rate of around 2.50%.

Supply Bottlenecks Left Scars On Goods Production

Supply bottlenecks continued to weigh on goods production. The Institute for Supply Management (ISM) Manufacturing PMI edged down to 59.5% in July, from 60.6% last month, registering a decline for two consecutive months and falling further below its peak of 64.7% in March 2021. Despite some signs of relief, manufacturers still faced elevated raw material prices and slow delivery, with the ISM Prices Index edging down to 85.7% in July from 92.1% last month and the Supplier Deliveries Index falling slightly to 72.5% from 75.1%.

Accumulating unfilled orders of manufactured goods also reflect interruptions in goods production and, more broadly, the overall demand for durable and capital goods used in production as businesses strive to keep pace with rapid gains in product demand. According to the Census Bureau, unfilled orders for durable goods ticked up by 0.9% in June, the fifth consecutive month of increase. The rise in unfilled orders of durable goods in June and in the rest of the second quarter was mostly driven by iron and steel mill products, construction machinery, motor vehicle and parts, and furniture.

The difficulties manufacturers face indicate that although goods demand has retreated from the elevated levels we've seen in the second half of last year, suppliers still find it hard to satisfy demand at the current level. A good example is the situation in the auto sector (see chart 7). The motor vehicles industry has been constrained by supply chain bottlenecks while aircraft production is still recovering. Insufficient supply of semiconductors continues to weigh on orders and shipments of computers and electronic products.

Chart 7

Supply Constraints Play Out In Consumer Prices

Personal Consumption Expenditures (PCE) prices jumped by 4.0% in June on a year-over-year basis, with the price index increasing by another 0.5% from last month. Core PCE inflation came in at 3.5% in June, with the relevant price index rising by 0.4% from last month. Overall, year-over-year PCE inflation averaged 3.9% in second-quarter 2021 (our forecast was 3.7%), while core PCE inflation averaged 3.3% (our forecast was 3.4%).

The rise in inflation in the second quarter stemmed from higher prices of both goods and services. Prices for durables went up by 1.0% in June from the prior month, while prices for service jumped by 0.4%. On a year-over-year basis, in the second quarter, durable prices rose by 6.5% in the second quarter on average, while prices for service increased by 3.2%. Although goods inflation was higher than service inflation, contributions to the overall price increase from service inflation are greater, since service makes up 64% of personal consumption while durables make up only 13% of it.

Again, the historically high inflation reflects the ongoing supply constraints, heated goods demand, strong reopening in the service sector, and the base effect. The persistent supply bottlenecks may hold prices high in the coming months, but such pressures will ease as important input production, such as that of semiconductors, gradually catches up with goods demand, which is already declining from unsustainable levels.

Table 1

Table 2