As the U.S. economy begins what looks to be a long, difficult journey to recovery from its pandemic-induced slump, its first steps offer some promise. With consumer spending proving largely resilient through the summer (helped by federal fiscal stimulus) and unemployment--while still notably high--softening a bit more than we had forecast, third-quarter GDP is poised for a steeper rebound than many market participants expected.

This is not to say that the world's biggest economy is out of the woods yet. Real-time economic data has started to show signs of weakening momentum. The New York Fed's Weekly Economic Index--a composite of 10 daily and weekly indicators of real economic activity, scaled to align with the four-quarter GDP growth rate--has stalled in its recovery path since the end of August after having recovered about halfway. Meanwhile, the Conference Board's Leading Economic Index, a measure of 10 leading indicators, decelerated to 1.2% in August, from 2% in July and 3.1% in June.

The recovery from the steepest downturn since World War II may face more hurdles now that extended federal unemployment benefits have expired, state and local governments' budgets remain in limbo, and job gains have slowed after the initial burst out of the gates (see table 1).

Against this backdrop, consumer spending will likely slow down, as early indications from the Johnson Redbook Index of weekly same-store sales haven fallen for two straight weeks in September. If Congress fails to agree on another package of aid to American households and businesses, the recovery could suffer from a cutback in consumer outlays--especially from the lowest-income families, who tend to spend a higher percentage of their incomes than wealthier individuals do.

Table 1

S&P Global Ratings now expects U.S. GDP to grow 6.7%--or annualized 29.5%--in the third quarter (following a 10.25% peak-to-trough contraction during the first half), and the contraction in full-year 2020 to be 4%--significantly better than the 5% drop we forecasted in June. But to put this into perspective, a decline of 4% for the U.S. economy is still historically bad, and the projected gain in July to September barely makes up three-fifths of what was lost this year after the worst-ever annualized contraction of 31.7% in the second quarter.

Moreover, we expect the recovery to enter a slower growth phase heading into 2021, and we forecast the economy won't get back to its pre-pandemic levels (real GDP of fourth-quarter 2019) until late 2021, with an annual GDP growth rate for next year at 3.9% (was 5.2% in June).

While the nearly 6.4 percentage point drop in the unemployment rate to 8.4% in August from its post-World War II record peak of 14.75% in April is a relief, we suspect that was the easy part. Indeed, at 8.4% (4.9 percentage points higher than its precrisis level), the unemployment rate remains at the top of the range, compared with past cycles. We don't expect the unemployment rate to reach precrisis levels until third-quarter 2024. In our baseline scenario, we assume a COVID-19 vaccine will be widely available in mid-2021, with some modest upside potential that it could be earlier.

Table 2

Risks To Recovery: Lack Of Both A Vaccine And New Stimulus

Perhaps because of lockdown fatigue, combined with the savings cushion, stimulus, and decisions by policymakers to end quarantines, the hit to the economy in July was relatively modest even as virus spread peaked in the southern and western parts of the country. COVID-19 infection rates have moderated from July, and measures of mobility have stabilized (except for transit, which remains well below precrisis levels), as the warmer summer weather kept people outdoors, which likely reduced the spread.

However, there is no coronavirus vaccine, and now that the U.S. is moving into autumn and winter, social activities will likely be forced indoors--naturally creating restrictions on economic activities. With the start of the school season and the arrival of colder weather, the risk for a second wave is also rising, as we're now seeing in parts of Europe.

Early in the pandemic, federal stimulus was seen as a bridge to help American households and businesses. But the bridge seems to span only halfway, leaving many of the drivers of the economy stranded. And a second wave of coronavirus would almost certainly widen the recessionary ravine, requiring an even longer bridge.

But with the election season in full swing, as both the replacement of Supreme Court Justice Ruth Bader Ginsburg and a vote on a continuing resolution to keep the government open, legislators in Washington have been unable to strike a deal on new fiscal stimulus. This premature austerity, before private demand is able to recover, could lead to another plunge in economic activity.

At the same time, damage from the wildfires in western states, as well as from the hurricanes and tropical storms that have hit the Gulf Coast, may also weigh heavily on economic activity in those regions and eat into third-quarter GDP growth. Although some of the economic losses related to natural disasters are recouped in subsequent quarters as homes and business are rebuilt (and with work needed to extinguish wildfires, some of the economic recovery happens during the crisis), there's some measure of activity that is permanently lost. Still, with government resources strapped by COVID-19, the bandwidth to address these crises (both hurricanes and wildfires) is severely limited.

With no coronavirus vaccine available as the country heads into flu season, a lack of new fiscal stimulus and likely only limited stimulus in the coming quarters (we assume $500 billion additional stimulus spread out over the next fiscal year in our base case), and trade tensions with China on the rise, we continue to see the risk of a double-dip recession at a heightened 30%-35%. Although we think it's closer to 30% given new COVID-19 cases have slowed since the July peak, and there is evidence of more rational pandemic behavior, like mask wearing. Also, the health care system has collectively learned more about how to manage the disease.

Still, it is not a far-fetched possibility that we could get a scenario of no more fiscal stimulus and a COVID-19 resurgence that cripples growth in the fourth quarter (compared with baseline). In such a scenario, the economy is only likely to get back to its pre-pandemic level in the second half of 2022, with higher risk of leaving the economy with longer-term scarring.

Chart 1

Businesses Are Optimistic, But Investment Is Still Modest

The recent drop in business investment was a casualty of COVID-19. Final demand plunged as businesses repaired their balance sheets after the sudden stop of economic activity. Add to that, the drop in oil prices to decade lows earlier in the year (the break-even price is considered to be around $50 for West Texas Crude)--largely from a virus-related drop in world demand for energy, exacerbated by the Saudi-Russia standoff on oil production--dramatically weighted on investment spending in the energy sector.

Business sentiment readings for both services and manufacturing are now above 50, indicating expansion. But, despite this positive business outlook, business investment has been modest because of lingering fears that a second wave of coronavirus is on the horizon. The latest Institute for Supply Management (ISM) manufacturing index continued to edge up in August, to 56, which is healthy sentiment and should continue to help the recovery in production of goods. Lean inventory also means the inventory cycle should keep the machines humming at least in the near term. The ISM services index, on the other hand, ticked down in August to a still-healthy 56.9. Services have limited upside as long as the virus remains an issue, hence it is likely that growth in manufacturing will likely outperform growth in services.

Moreover, trade tensions between the U.S. and China have intensified, and we expect this to persist, no matter who wins the race for the White House in November. Our forecast accounts for the tariffs the countries have levied on each other since 2017 and are still in effect. The promise to buy U.S. goods by China under the Phase 1 trade deal between the two countries remains far short of its stated benchmark, even as exports to China have rebounded since April. Meanwhile, exports of services will continue to lag given the negative impact of virus fears on tourism and foreign exchange student education (which are counted as exports of the U.S.).

This leaves a cautious outlook for business investment for the remainder of this year and the next. All told, we expect real nonresidential (business) fixed investment to decline 5.4% this year before gaining 5.6% next year. That leaves fixed investment at the end of 2021 only 1.6% above that seen in precrisis fourth-quarter 2019.

A company's survival--or, rather, collapse--has also become a critical issue. A report from online directory Yelp suggests that both permanent and temporary business closures are rising across the country, with the pace of permanent closures much faster than temporary ones. As of Sept. 15, 60% of closed businesses (97,966 in all) had done so with no plan to reopen.

When a business fails, the jobs it supported disappear. And for businesses that do survive, it's likely that their capacity has been compromised, meaning less need to rehire the workers it let go.

Half Empty Or Half Full?

Job gains have slowed significantly since a May surge, and initial jobless claims remain historically high. The economy added back 1.37 million jobs in August, according to the Bureau of Labor Statistics (BLS)--but only half of those lost in March and April have been regained, and there are still 11.6 million fewer jobs than before the pandemic. Worryingly, this was likely the easier half of the jobs recovery. And with initial jobless claims currently at 860,000 and remaining stubbornly near the 1 million mark, the next half is going to take some time as we go through the "sticky" part of the unemployment pool (meaning long-term unemployed and permanent job losers).

Chart 2

Chart 3

While headline unemployment fell to 8.4% in August, a quicker-than-expected retracement from the 14.7% peak in April, the rate remains twice as high as in February, with 13.6 million people reporting that they are unemployed.

And, as Federal Reserve Chairman Jerome Powell highlighted in his press conference after the Sept. 15-16 Federal Open Market Committee (FOMC) meeting, job market conditions are likely worse than these headline numbers. The misidentification of people in the household survey as employed when they're actually jobless, as well as adding back some of the drop in labor force participation--i.e., holding the labor force at February levels--means that the headline figure would likely be 3 percentage points higher. (This also means the rate would have been above 20% in April.) We see the headline figure remaining at 7.9% in the fourth quarter and averaging a still-high 6.7% next year.

Worse yet are signs that what was once considered temporary unemployment is becoming permanent, a structural shift that could lead to deep economic scars.

Workers who were unemployed for "temporary" reasons are now at greater risk of losing their positions permanently, with some already falling into that bracket. The number of permanent job losers, in the BLS' term, jumped to 3.4 million in August (25% of total unemployed), around twice precrisis levels. Also, the number of people counted as unemployed for 15 weeks or longer increased to over 8 million in August, four times the number we saw last year.

One reason for temporary job losses becoming permanent is the length of the pandemic; the longer it runs, the higher the chances of business failures. On top of that, if state and local governments--which are some of the country's largest employers--don't get budget relief from the federal government, there will likely be more pressure to cut positions, rather than just furlough employees. This would mean longer (stickier) spells of unemployment.

We expect net job gains to remain well under 1 million for the foreseeable future. That will likely leave the total number of jobs lost still in the millions at the beginning of next year. We don't anticipate total employment will return to pre-pandemic levels before 2022. (And even then, the numbers would be short of "full" employment, given growth in the working-age population.)

Consumer Spending Is Subdued

This could weigh on consumer spending, which was robust through the summer despite the expiry of unemployment benefits, helped by housing-related spending and lockdown fatigue. In fact, consumers are increasingly cautious with their outlays, with spending momentum cooling for a third consecutive month in August, when retail sales rose just 0.6% and core sales slipped 0.1%. A solid advance in food services, along with sales of building materials and clothing, drove the headline gains. Spending through August points to real consumer spending rising an annualized 38% in the third quarter, which means that almost two-thirds of the pandemic-induced decline has been recovered.

But it's misguided to assume the consumer recovery is complete. Spending on services, which make up more than three-fourths of GDP, remains well below where it was before the crisis. And real-time data indicates consumer demand is struggling to sustain momentum, with the expiration of fiscal aid and a still uncertain health situation.

High-frequency data from our real-time economic data report says as much. Indoor restaurant dining, on a year-over-year basis, has weakened somewhat after seeing earlier gains. The meager boost in movie box office receipts during Labor Day didn't last, and hotel revenue per room rates declined to their lowest since Aug. 8. The Johnson Redbook Index of weekly same-store sales, on a month-over-month basis, fell for two straight weeks in September after surpassing breakeven in August.

And while consumer sentiment perked up a bit in early September, with consumers feeling better about current economic conditions and the outlook, sentiment remains below pre-pandemic levels, which has translated into slow consumer spending following the initial bounce.

Low-income households have been pivotal in driving the early phases of the recovery, helped by healthy unemployment benefits. But with job growth losing momentum, and jobless claims stubbornly high, further fiscal assistance to struggling households will be vital to prevent a decline in spending. We expect consumer spending to contract by 4.4% this year, and increase 4.9% in 2021.

On the bright side, the housing market has recovered from its collapse in April. Various indicators--sales of new and existing homes, pending sales, and building permits--all shot up to levels last seen 13 or more years ago as record-low interest rates and inventories, which have led to bidding wars, have played a key role.

We believe the recent strength may be temporary and housing starts will overshoot their long-run trend in the first three quarters of the forecast. We now see housing starts averaging 1.4 million (annualized) in the second half of this year, and 1.3 million annual average in 2021.

Stimulus: Cliffhanger

Congressional Democrats and Republicans have been unable to come together on a deal to deliver additional relief weeks after unemployment benefits and eviction protections lapsed for many Americans. The two sides are deadlocked over issues including unemployment insurance, funding for cash-strapped state and local governments, and liability protections for companies.

As Congress adjourned for August without passing a new package, we modified our fiscal assumptions to reflect a second round of stimulus checks disbursed starting in the beginning of the fourth quarter. While we recognize that the two parties are far apart, we assume they will finally agree to a $500 billion package that includes emergency unemployment benefits, another round of the Paycheck Protection Program, and funding for state and local governments.

However, there are growing concerns that an agreement won't be reached. With the death of Supreme Court Justice Ruth Bader Ginsburg, the political debate over her replacement will control Capitol Hill at least until Election Day or later. That, together with negotiations to approve a continuing resolution to avoid a government shutdown at the end of the month, leaves little bandwidth for Congress to negotiate an already difficult to pass large stimulus package aimed at getting more money to millions of unemployed workers, small businesses, and state and local governments.

In our downside scenario, we assume Congress is unable to reach an agreement to extend any emergency stimulus this year. However, even a smaller stimulus package, of just $500 billion that we assume in our baseline, is expected to only modestly lift the economy out of its hole. However, it does keep fourth-quarter GDP positive at 3.5% with the unemployment rate edging down a notch, to 7.9%.

Regarding the election, results will shape the country's economic trajectory in the coming years. Key differences between President Donald Trump and Democratic presidential nominee Joe Biden include tax and spending policies, health care, regulation, and trade policies--to name a few. And while we do not factor the outcome of the election into our September baseline forecast--and we acknowledge that any significant legislative change will depend on the combination of partisan control of the White House and Congress--we are closely watching these developments and their impact on the U.S. economy.

A Brave New Fed

August CPI (Consumer Price Index) inflation readings saw a bounce that lifted year-over-year inflation to 1.3% from 1.0% in July. Inflation is still a full percentage point below the 2.3% year-over-year increase in February. While virus-related demand for goods to upgrade homes, gardens, and remote-work spaces has pushed up prices in those items, we expect to see some of this price momentum abate heading into autumn. With a still-soft labor market leading to weak aggregate demand, businesses will likely be competing for fewer customers, which means upward pressure on prices will remain limited out a few years.

The FOMC committed to allow inflation to run above its target of 2% for "sometime," before raising benchmark interest rates again. The Fed revised its official policy statement to reflect its longer-run strategy shift to "a flexible form of average inflation targeting" and a maximum employment goal that is broad-based and inclusive. It also said it would continue its large-scale asset purchases of Treasuries and mortgage-backed securities at least at the current pace--of $120 billion per month--for the foreseeable future.

Even more interesting are the factors the Fed will consider before a policy move. At a press conference after the FOMC meeting, Chairman Powell laid out several--including wage growth, workforce participation, and disparities in minority joblessness relative to white workers--to consider before the Fed would view the economy at maximum employment, and start to consider raising interest rates.

We project inflation--as measured by the core CPI (CPI excluding energy and food)--will average 1.5% the rest of this year (on a year-over-year basis) before rising to 1.9% next year, and won't reach 2% until 2024. We expect the Fed to hold off on raising rates until mid-2024, when inflation moderately holds above 2%.

Table 3

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. Scenarios are based on ordinary risks to baseline growth, not extraordinary risks. Even risks surrounding our baseline forecasts are particularly wide this time around.

Upside: Come Together!

In the upside scenario, the U.S. government and Federal Reserve come together to offer support to households and businesses, bridging the gap caused by the shutdown. The July COVID-19 spike is contained, helped by more adherence to social distancing rules. America blots out remaining COVID-19 hotspots to contain the virus, leaving the U.S. clear of further outbreaks in October.

The U.S. economy would experience a V-shaped recovery (compared with a U-shaped recovery in the baseline and the double-dip in the downside scenario) in the second half of the year and one more year of solid above-trend growth, as well as significantly higher above-average GDP in 2020 than in our baseline case.

After stumbling in 2018 and 2019 and a COVID-19-related hit in the second quarter, housing rebounds in the second half, with real residential investment climbing 0.5% in 2020, as opposed to 0.2% in the baseline. As U.S. economic activity picks up and prices finally rise, the Fed would have room to raise rates gradually, by an additional 50 basis points in 2023, followed by a few more hikes in 2024. This is in sharp contrast to its "lower for even longer" approach, with no rate hikes until the fourth quarter of 2024 in the baseline.

While the world bounces back from the global recession, trade tensions would ease, at least over the near term, between the U.S. and China. The global recovery would help business sentiment rebound across both the industrial and service sectors, but the S&P 500 wouldn't be much different from the baseline given expectations for the Fed to raise rates sooner than later.

Higher-than-potential growth would require businesses to continue looking to hire more folks. The unemployment rate in this scenario would fall to 8.3% in 2020 (compared with 8.4% in the baseline) and then to 6.4% in 2021 (lower than 6.7% in the baseline scenario). Continued strength in the labor market would let workers enjoy healthier wage gains throughout next year. Together with the income effect, wealth from a rebound in stock market gains in 2021--helped by a cooling of trade tensions--would lead to higher consumer sentiment.

All of this would be sufficient for a virus-driven slowdown in consumer spending to be less severe, down 3.2% in 2020, compared with the baseline's 4.4% decline. Strength in the labor market would help push up household formation rates, providing a tailwind to the housing market (compared with the baseline). In this optimistic scenario, after a 2.8% drop in 2020 real GDP growth (much better than the 4.0% drop in the baseline), GDP would jump to 5.6% next year, larger than the 3.9% gain for the baseline and also from a much higher base.

Table 4

Downside: A Bridge Half-Built

In our pessimistic scenario, a second wave spreads through the U.S. during the fall flu season. With tensions flaring across the aisle, the government is unable to agree on a plan to extend stimulus for those in need. The bridge the government began building with stimulus could only make it halfway, leaving economic participants stranded.

In our pessimistic outlook, a second wave of COVID-19, together with a failure to provide fiscal support, leads to another economic downturn, with an even steeper contraction in consumer spending than in the baseline forecast and a much longer timeline to bring COVID-19 under control. This would delay the start of the recovery in spending, reduce its vigor, and delay laid-off workers getting back to work.

Consumer spending would plunge by 15% (annualized) in the fourth quarter, compared with a 3% gain in the baseline, with further disruption in the first quarter on fears that COVID-19 has returned. The contraction in consumer demand and output in the pessimistic forecast is much sharper. From both election tensions and stimulus fatigue, the government is slow to approve additional stimulus, delaying the recovery further.

Owing to the sharper contraction and slower recovery in consumer spending and related investment spending by businesses, GDP growth would plunge by 5.1% in 2020 in the pessimistic forecast, with barely a nod to recovery the following year, at just 2.8% growth in 2021.

In this scenario, the economy doesn't get back to the previous cycle peak before sometime in the second quarter of 2022 (compared with fourth-quarter 2021 in the current baseline). By fourth-quarter 2023, GDP would be $175 billion, or 3.4% smaller than in our December 2019 baseline forecast (compared with $115 billion, or 2.2% smaller, in the current baseline).

The larger hit to economic growth in our downside scenario translates into a corresponding higher unemployment rate. The initial degree of dislocation and uncertainty over reemergence of the virus would keep the unemployment rate elevated at close to 8.7% in the fourth quarter of 2020, on average, before it would finally start to move back down midway through 2021. The unemployment rate would average 8.6% in 2020 and 7.5% in 2021 (compared with 8.4% and 6.7% in the baseline).

With the world reeling from the global recession, trade tensions would flare, at least over the near term, between the U.S. and China, giving the private sector more reach to hoard savings rather than invest. Both business and residential investments would be hit hard, with business fixed investment declining both this year and the next. In such a scenario, stock markets would struggle to move up. With inflation stubbornly low, the Fed would be on an effective zero bound for several years to come.

Table 5