While still running hot, the U.S. economy has cooled as summer ends. Supply disruptions remain the leading suspect slowing the world's biggest economy, and the delta variant is now an additional drag.

Recent economic data indicates people are pulling back after the reopening earlier this year. Michigan consumer sentiment remains near August's 10-year low. Mobility has cooled, hotel occupancy rates fell below their 2019 average in late August, and TSA checkpoint numbers have fallen to 25% below the 2019 baseline in early September, from 11% below at the end of July.

Chart 1

In this light, we revised our forecasts of real GDP growth for 2021 and 2022 to 5.7% and 4.1%, respectively, from 6.7% and 3.7% in our June report, with our new 2021 GDP forecast down a whopping 1 percentage point from June. Still, the near-term health of the U.S. economy remains strong and our current GDP forecast, if correct, is still the highest reading since 1984. The number of new cases fell in the first week of September, for the first time since late June. The delta variant and FDA approval of the Pfizer vaccine are encouraging more people to get vaccinated, bringing the country closer to herd immunity with 55% of Americans fully vaccinated as of Sept. 20. Moreover, the U.S. economy has felt less impact with each wave of the virus and has been able to withstand the damage. We maintain our assessment of U.S. recession risk over the next 12 months at 10%-15%--our lowest assessment in six years.

Inflation: Transitory And Transitional

With a stimulus-driven surge in demand outpacing supply gains, inflation is running at breakneck speed. However, there are signs inflation is starting to ease (slightly) as supply bottlenecks unjam and base effects start to wear off.

Inflation, as measured by the Consumer Price Index (CPI), rose 5.25% in August from the prior year--slightly slower than the 5.37% increase in July. On a monthly basis, price gains moderated to a 0.3% increase, down from 0.5% in July and a sharper slowdown than many economists predicted, although we and the Fed agreed higher inflation would be transitory. Perhaps more telling, core CPI (which strips out volatile food and energy prices) rose only 0.1% in the month.

Table 1

We believe inflation peaked in second-quarter 2021. While we expect it to continue to soften, supply-side price pressures will remain somewhat "sticky" through 2022. Average core CPI inflation will likely slow from 5.5% in 2021 to 2.7% and 2.5% in 2022 and 2023, respectively. The core Personal Consumption Expenditure deflator (PCE deflator), the Fed's preferred inflation measure, will slow to around 2.3% and 2.1%, respectively, for 2022 and 2023. The PCE deflator will near the Fed's 2% target later in 2023.

A healthier economy and still-hot inflation will give the Fed reason to remove accommodation. We expect the Federal Open Market Committee (FOMC) to start tapering its quantitative easing (QE) program in December (announcement in November). We forecast the U.S. unemployment rate will dip below 4% in the fourth quarter of 2022, leading the Fed to raise the federal funds rate in December of 2022, followed by a couple of hikes in 2023, and two more in 2024.

The Big Question

The U.S. economy is resilient over the near term, but government bickering over infrastructure proposals and the threat of a government shutdown in October or, more ominously, a debt ceiling crisis, can't be ignored. Moreover, corporate debt levels are high, with a significant amount it speculative-grade ('BB+' and below). With a hot economy and interest rates at record lows, bond payments may be currently easy to manage. But when interest rates go up and the economic tide recedes, how many businesses will be left standing without a life preserver?

On The Job(s) Among The Missing

The Bureau of Labor Statistics (BLS) reported a lackluster 235,000 job gains in August--far below market expectations. However, with many business managers on vacation and therefore unable to fill out the forms, August is notorious for initially undercounting overall job gains. Given upbeat initial jobless claims, we expect a sharp upward adjustment to the August number. With the labor market still 5.3 million jobs short of the pre-pandemic peak, recovery could take a while.

While the August unemployment rate was down to 5.2%, 8.4 million people are still unemployed (2.4 million more than the 2019 average, and that is not accounting for the number of underemployed and working age population growth), with 37.4% long-term unemployed (unemployed for 27 weeks and over). The unemployment rate is even higher, over 7%, when adjusting for a smaller labor force since February 2020. On a positive note, the extremely tight jobs market may offset the usual roadblocks for the long-term unemployed, making it easier to find work once pandemic-driven constraints are lifted.

In fact, despite the delta variant, recent jobs data still indicates businesses are desperate to hire. The BLS Job Openings and Labor Turnover Survey (JOLTS) reached a new high of 10.9 million in July. Weekly Indeed data indicated that this continued through early September, with the number of job postings on the site increasing to 40.5% above pre-pandemic levels on Sept. 3. While all industries have recovered jobs lost during COVID-19 by varying degrees, only those industries tied to financial activities regained all jobs lost during the pandemic.

Chart 2

More folks are likely searching for jobs after federal pandemic unemployment benefits ended in early September. This may help alleviate the labor shortage, but we believe it will take some time for the end of unemployment benefits to be reflected in jobs data. As millions of Americans may have to quickly join the jobs market, a more complicated matching process may temporarily slow hiring. Given capacity constraints, we expect the economy to fully regain the 22.4 million jobs lost from the pandemic by fourth-quarter 2022 and the unemployment rate to reach its second-half of 2019 range of under 3.6% by midway through 2023.

Willing And Ready

Increased savings from the various COVID-19 relief packages through March helped lift personal income and savings rates dramatically. The personal saving rate, at 9.6% in July, is well over its precrisis historical average back to 1980 of 7.3%. Indeed, if we compare it to 2019 personal savings (monthly average), we saw significant excess savings to the tune of $2.5 trillion.

Households are putting that cash to good use. August total retail sales, up 0.7% from the previous month, were in line with our expectations of a 1.0% increase but well above consensus expectations for a 0.8% drop. The dramatic 3.6% drop in auto sales for the month (we expected a 3.0% drop), was largely due to semiconductor shortages. Excluding autos, retail sales were up 1.8%, in line with our forecast, indicating that consumers will buy--they just need something on the shelves. Supply-chain disruptions, not just semiconductors, will weigh on GDP growth in the fourth quarter and into the new year. That said, abundant spare capacity and solid household balance sheets will help support the U.S. expansion next year.

Chart 3

While U.S. households on average are flush with cash, supply disruptions will limit their ability to spend. Inventory constraints could impact this year's holiday shopping. Supply disruptions for U.S. output could have big impacts beyond the ones already seen for vehicles--such as the ongoing one in home building from out-of-stock materials, even as lumber prices have moderated. Furthermore, rising natural gas and propane prices will pinch household finances in the Northern states through the winter months, just as growing La Nina conditions may translate to a cold winter.

P.I. (Private Investment) On The Case

While it will take time for production and labor supply constraints to moderate, business investment will remain supported by robust profits from high demand for products sold at higher prices, low interest rates, and continued inventory rebuilding as businesses struggle to fill their shelves. We see equipment spending rising a dramatic 13.9% in 2021 and 5.2% in 2022 and intellectual property investment rising 9.4% and 3.6%, respectively. Nonresidential construction remains the laggard, but after declining in six of the last seven quarters, we expect it to start to turn the corner in the third quarter. Nonresidential construction will likely be down by 7.2% this year after falling by 12.5% in 2020, with a 1.8% gain in 2022. After this year, all three major categories of business fixed investment will see growth through 2024.

Residential investment, in the form of home building, has cooled as higher prices reduce the pool of potential buyers willing and able to purchase a new home. Home prices have surged since the second half of 2020, as fear of the pandemic and record low interest rates encouraged people to buy a home away from crowded cities. Strong demand and relatively modest supply have driven prices higher, with the June 2021 S&P CoreLogic Case-Shiller U.S. 20-City Home Price NSA Index surging by 19.08% over last June, almost 3x the 7% average seen over the past five years. Meanwhile, the number of mortgage applications for home purchases increased by 4% in the past month as the 30-year fixed mortgage rate declined from 3% in late June to around2.85% on average in September. That is still consistent with a softening in the market since early this year, with both new and existing home sales down 12% year-to-date. After housing starts likely reach 1.59 million units in 2021, a 15-year high. We expect starts to slow to a still-high 1.51 million units in 2022.

At the same time, stronger household balance sheets and lower interest rates during the pandemic led to a housing boom, which brought affordability in the spotlight. Despite the recent boom, the National Association of Realtors housing affordability index, at 150.4 in July, is well above 100, indicating that mortgage payments faced by a homebuyer with the median U.S. household income have remained under one-quarter of their income, a widely used measurement of home affordability. That said, this varies among income brackets, with those earning the least facing disproportionately heavier cost burdens.

Change Of Heart

Supply and labor bottlenecks are driving price gains. While we expect inflation to moderate from its second-quarter peak, prices will remain stubbornly high over the next two years. Given capacity constraints will only slowly be resolved, we expect CPI inflation to only slowly moderate toward the Fed's 2% target, with the core PCE price index on average holding above 2% until sometime in 2023. The Atlanta Fed wage tracker, an index not so easily swayed by the composition effect, also saw hourly paid worker wages climb 4.1% compared to the previous year, a 14-year high. It's something that we think may have made the Fed blink.

Chart 4

Chart 5

With that in mind, we expect the Fed to start tapering its monthly $120 billion asset purchases in December this year (announce in November FOMC meeting), reaching zero by mid-2022. Minutes from the July FOMC meeting earlier indicated a willingness to start reducing asset purchases before the end of the year. That changed at the September FOMC meeting, where policymakers indicated that the Fed will not only likely begin tapering at the end of next year, they may finish mid-year, a bit earlier than we had thought. The Fed also indicated a much steeper path on rate hikes that we earlier thought. The Fed's dot plot current median estimation of "lift off" was in line with our expectation of a move around end of the year. However, based on a slightly more upbeat economic outlook through 2023 relative to their June estimates, the median Fed member projection now expects three rate hikes by year-end 2023 and another three in 2024. Keep in mind, the dot plot includes estimates from 18 FOMC participants (one member seat remains vacant), including seven nonvoting members. Based on our more conservative economic outlook, we continue to expect a total of five rate hikes by year-end 2024, with the fed funds rate reaching 1.50% by year-end 2024. Still, the decidedly hawkish lean to overall Fed thinking in September adds upside risk to our rate hike forecast.

Infrastructure: Go Big Or Go Home

As stimulus checks work their way through the system, President Joe Biden's Administration is looking to further strengthen the U.S. economy with an infrastructure funding bill. The question now is: How big will the plan be? We recently updated our projections on the impact of a $1 trillion investment on the U.S. economy through 2030 (see "Economic Research: How U.S. Infrastructure Investment Would Boost Jobs, Productivity, And The Economy," Aug. 23, 2021).

On a much stronger outlook than in 2020, we currently estimate a 1.4x multiplier, meaning a $1 trillion infrastructure investment would add $1.4 trillion to the economy over the project's duration. Assuming the infrastructure project was smart, there would be an economic butterfly effect as the productivity boost and stronger economic activity would continue for years. We also found that the project would create 883,600 jobs by 2030, many of them middle class, and per capita income would be 10.5% larger than in the no infrastructure scenario. Private-sector productivity would also get a roughly 10 basis point boost from the infrastructure investment each year. That productivity boost would likely lift average GDP growth on an annual basis to 2.1% from around 2.0% over the project period and thereafter, assuming the project was wise. To give you an example, former President Dwight D. Eisenhower's highway system cost $500 billion in today's terms. With all the goods and people who travel on its highways every day, it has paid for itself.

Chart 6

Putting the boost in infrastructure spending aside for the moment, how government policy (not just infrastructure) is funded also matters. That said, the Congressional Budget Office noted in its August 2021 report on physical infrastructure that deficit financing would mitigate the positive effects on the economy of investments in infrastructure.

Shutdown And Debt Ceiling

Congress faces two enormous tasks over the next month: funding the U.S. government and increasing the debt ceiling. If no agreement is reached to fund the government by Sept. 30, the federal government will be closed until further notice. Even worse, Treasury Secretary Janet Yellen has warned that extraordinary measures to avoid broaching the debt limit are likely to run out in October.

The shutdown, if it's brief, wouldn't be a disaster, but would still reduce some of the economic gains the U.S. has felt from the reopening, and add more complications to a system already tangled up by supply chain disruptions. S&P Global Ratings Economics estimates the U.S. government shutdown may cost fourth-quarter growth around $1.8 billion (annualized) or 0.11 percentage points for every week the government is closed, on both direct and indirect costs. While some, but likely not all, indirect costs (e.g., canceled trips to closed national parks) may be regained once the government reopens, the productivity lost from direct costs (furloughed "nonessential" government workers) would reduce real GDP--since no "product" was created--and would never be regained. However, furloughed government employees do usually get paid (with taxpayer money, assuming Congress decides to compensate them afterward, as has been the case in the past). The shutdown would have no effect on nominal GDP and would add to fourth-quarter inflation. With inflation currently skyrocketing, the Fed will have another headache with which to contend.

Turning the government off and on comes with a cost. Indeed, even the January "weekender" shutdown during former President Donald Trump's Administration wasn't without costs. The most recent 35-day shutdown from fourth-quarter 2018 through first-quarter 2019 was so severe that it cut GDP by 0.1% and 0.2%, respectively, according to a January 2019 analysis from the nonpartisan Congressional Budget Office. Though risk of a shutdown is high, Congress may reach a new agreement on funding the government by the deadline. However, another short continuing-resolution reprieve, which kicks the "shutdown can" down the road a few feet, only brings a possible shutdown fight closer to the debt ceiling debates, complicating what could already be an ugly fight.

This is all in play after the debt ceiling issue came back to life on Aug. 1. The Treasury Department has since been using "extraordinary measures" to meet the government's obligations. On Sept. 8, Yellen stated in her official notification to Congress the need to address the debt ceiling as the extraordinary measures would be exhausted in October.

We believe Congress will raise or suspend the debt ceiling. A default by the U.S. government would be substantially worse than the collapse of Lehman Brothers in 2008, devastating global markets and the economy. Should a default occur, the resulting sudden, unplanned contraction of current spending would be staggering. The economy would fall back into a recession, wiping out much of the progress made by the recovery.

Table 2

Upside And Downside Scenarios

Each quarter, S&P Global economists project two scenarios in addition to their base case, one with faster growth than the baseline and one with slower. In this report, scenarios are based on risks to baseline growth coming from the timing of vaccine availability to the wider public, strength and duration of capacity restrictions, and amount of further fiscal support.

Chart 7

Upside: On A Rocket To The Moon

In our upside scenario, the $1.9 trillion stimulus-driven recovery together with a massive vaccine push allows the U.S. to reach herd immunity a little sooner than earlier thought. Households, itching to go out and celebrate with friends and family, dip into their savings accounts to splurge on the leisure activities they have neglected for most of the year. Consumer spending surges to a post-World War II high of 8.3% in 2021 and 4.9% the following year, above the 8.1% and 4.0% seen in the baseline.

As future economic conditions continue to suggest happy days and more revenue in store, businesses readily open up their bank accounts. Equipment spending surges by 14.9% and 3.4% this year and the next, with relative gains over the baseline increasing in 2022 as confidence improves. In the upside scenario, the U.S. economy rebounds 6.0% in 2021 and continues to grow at a higher pace of 4.5% in 2022 (versus 4.1% in our baseline). Higher growth rates also mean the unemployment rate falls under 4% by mid-2022. The labor force participation rate continues to rise, bringing back more folks from the sidelines. Core CPI inflation (actual and expectations) starts slowing down midway through next year, proving the current supply shock to be indeed transitory and giving the Fed more room to wait before they raise the policy rate. After letting the economy run hot for longer, the Fed raises its policy rate for the first time in this recovery in early-2023 (versus late 2022 in baseline).

Table 3

Downside: Houston, We Have A Problem

In our downside scenario, the U.S economy continues to suffer persistent shortages of labor and goods. Price inflation of 3% or more lasts longer and growth in real disposable income weakens, crimping real demand growth. As a result, producer prices are even higher and harder to pass on to consumers, thus corporate profits deteriorate, triggering a pullback in the S&P 500 that causes a decline in the wealth effect, weakening overall domestic demand. GDP growth clocks in at 5.4% and 3.4% in 2021 and 2022, respectively.

In this scenario, job growth slows (compared with the baseline) by early 2022, the ranks of the unemployed relying on benefits remain elevated, and there is a risk of long-term unemployment with expiring benefits. Still, the Fed blinks, in fear of inflation expectations unmooring, and thus raises rates in third-quarter 2022. Still, inflation remains elevated until the first half of 2023 since monetary policy can't do much about the supply side issues. To be sure, inflation does start to soften as the supply side wrinkle finally irons out by mid-2023, but demand has also weakened enough by that time that the Fed chooses to slow its rate hike cycle. Although the Fed hiked earlier in this scenario (versus the baseline), the interest rate increases slow enough that they lag the baseline by fourth-quarter 2023. The government's infrastructure package, spread out over eight years, helps, but it is not designed to prop up short-term demand (like transfer payments, boosters to unemployment benefit), so the overall cyclical expansion becomes weaker than in the baseline.

Table 4

Writer: Devon Reilly.