The U.S. economy has outperformed expectations following consecutive quarters of contraction in the first half of 2022. Real GDP grew at an about 2.9% annual rate in four quarters leading up to September of this year. For the entire 2023, the economy is on track for a 2.6% growth on a fourth-quarter-over-fourth-quarter basis (Q4/Q4), which translates to a 2.4% annual average growth--just about the average of previous expansion period (2010-2019).

The continued solid growth comes even amid sharp monetary policy tightening during the past year and a half. We remain of the view that the ongoing resiliency will be tested going forward, as real interest rates stay relatively high in the coming year (relative to the last monetary cycle as well as the longer-run equilibrium rate) and the lags of monetary policy tightening feed through the economy. Businesses are facing higher costs of capital, the outcome of which will lower capital expenditure and hiring. Consumer spending is poised to be more in line with real income growth, as the firepower from excess cumulative savings has dwindled.

S&P Global Ratings now expects the U.S. economy to expand 1.5% in 2024 on an annual average basis (up from 1.3% in our September forecast) and 1.4% in 2025 (unchanged from the September forecast), before converging to the longer-run sustainable growth of 1.8% in 2026 (see chart 1). Although we revised up our 2024 growth forecast, it had partly to do with the statistical "carry-over" effect. In fact, on a Q4/Q4 basis, growth is expected to come in little over 0.8% in 2024, with the second half of the year half as strong as the first half (see table 1 for forecast highlights and table 2 for extended baseline forecasts).

Chart 1

Admittedly, getting the timing of cyclical slowdown has been a trickier endeavor during this cycle than in the past. A higher share of fixed-rate debt has sheltered households and businesses from higher interest rates during the current rate-hike episode, and contribution of public-sector policies to growth has leaned pro-cyclical: both unique features of the current cycle that may very well keep recession at bay for the time being. But it doesn't mean we won't see a business cycle adjustment to growth below its long-run potential in the next 12-18 months.

The weaker growth will cause demand for labor to slacken further and the unemployment rate to rise in the next two years--from 3.9% currently to 4.6% in 2025--slightly above the longer-run steady state (consensus estimates clustered on 4.1%-4.4%). As normalization in the product and labor markets continues, disinflation will endure, albeit on a bumpy path. We continue to forecast core inflation (excluding volatile energy and food prices) finally falling closer to levels consistent with the Fed's target of 2.0% on a sustained basis by the middle of next year (from the current 4.0% year over year as measured by the consumer price index [CPI] and 3.7% year over year as measured by the personal consumption expenditure index).

Table 1

As for monetary policy, we are not convinced yet that the Fed is done hiking. Tighter-than-anticipated financial conditions led the Fed to hold policy rate steady at 5.25%-5.50% on Nov. 1. Since then, financial conditions have eased somewhat (paradoxically increasing the chances of another rate hike), seemingly because of the following factors:

• The Treasury announced that its issuance of debt in the coming months will be weighed more towards shorter duration (as opposed to longer duration in the third quarter), providing relief on the bench-mark 10-year yield; and
• October CPI inflation came in slightly below consensus expectations, which led markets to take further rate hikes completely off the table and pull forward first rate cut in mid-spring of next year.

A closer look at the inflation data reveals that the easing in core inflation pressures appears somewhat exaggerated. With the economic activity and labor demand still carrying a fair bit of momentum, we think the upcoming policy meetings are live (toss up), until inflation data shows it is near 2% on a sustained basis (at least month over month should be averaging 2% annualized on a consistent basis).

We continue to think there will be another 25-bp rate hike (likely December)) before rate cuts start in mid-2024. At that point, if past cycle is any guidance, the Fed is likely to pivot quickly from "pause" to "insurance cuts" to minimize damage to the labor market as the real economy would have weakened considerably with increasing downside risks and inflation is closer to target. We pencil in policy rates to land at 4.6% and 2.9% by the end of 2024 and 2025, respectively. This would still be higher than the previous cycle's peak of 2.4% (in the second quarter of 2019).

In real (inflation adjusted) terms, the peak restrictive stance would occur in the second and third quarters of 2024. Afterward, the restrictive stance will gradually diminish to a neutral stance by 2026, when real rates reach the assumed longer-run equilibrium of 0.9%, which is within the Fed's central tendency for the longer-run neutral estimate of 0.5%-1.3%, higher than 0.5% in 2019.

The interest rate on 10-year Treasury yield should reflects the underlying neutral real rate of interest plus expected inflation over the holding period. Consistent with our forecasts for growth, inflation, and policy rate, we anticipate the yield on benchmark risk-free rate to peak in the current quarter and decline gradually in the next two years to a 3.0%-3.5% range.

Reasons To Doubt Sustainability Of Current Growth Momentum

The U.S. economy had a blockbuster third quarter with real GDP expanding 4.9% (annualized Q/Q), according to the U.S. Bureau of Economic Analysis' advance estimate (see chart 2). Although part of the gain was due to a surge in public spending (17% of GDP) and faster pace of inventory build-up, private domestic demand, excluding inventories (as measured by final sales to U.S. private purchasers), was strong at 3.5%. Consumer spending (69% of GDP) led the way by increasing at a 4.0% annualized pace in the third quarter, with spending on services (two-thirds of total consumer spending) rising at the strongest rate since the 2021 pandemic recovery.

Chart 2

Residential fixed investment (3.5% of GDP) saw its first gain in 10 quarters, while investment in non-residential structures (2.4% of GDP) gained for a fourth consecutive quarter, led yet again by construction in the manufacturing sector (the ongoing resurgence of manufacturing in electronics and semi-conductor sectors that are enjoying subsidies and tax exemptions under CHIPS and Science Act [CHIPS] and Inflation Reduction Act [IRA]). One negative factor in the report was the third consecutive decline in four quarters in spending on machinery and equipment (two-fifths of private non-residential fixed investment), with which rising interest rates now seem to have caught up.

Data released since the advance estimate was published suggests third-quarter GDP growth will be revised up in its second estimate to 5.3%, according to tracking estimates by S&P Global Market Intelligence. This potential revision is normal--near the 0.5 percentage point average revision (without regard to sign) seen from advance to the second estimate based on data from 1996 through 2022. That said, there are many reasons to believe the growth momentum will slow down meaningfully from the fourth quarter onwards.

Take consumer spending, for instance. First, spending on services is now back to the pre-pandemic trend, leaving little in the way of catch-up demand (see chart 3). Second, last quarter's spending largely came from savings, as real disposable income has now actually declined four months in a row with the saving rate falling to a very low 3.4% in September (well below the pre-pandemic average) from the 5.1% average in the second quarter (see chart 4). Third, there are signs that consumers are starting to feel the squeeze in their debt servicing capacity.

Chart 3

Chart 4

According to the New York Fed/Equifax credit data, there was a significant increase in the percentage of consumer loans that tipped over into serious delinquency (unpaid for 90 days or more) in the third quarter. The delinquency ratio is already at its highest level outside recessionary periods for auto loans (driven primarily by the sub-prime segment which is about one-seventh of total outstanding auto-loan balances), and delinquency rates on credit card loans also rose in the third quarter at the fastest pace since the Great Recession (see chart 5). And as interest rates continued to climb in the fourth quarter, a further rise in delinquency in the same quarter seems inevitable, especially as student loan repayments have now resumed (starting October), which is likely to push up debt servicing costs further and weigh on consumption at the margin.

Chart 5

On private investment, the deterioration in surveys of capex intentions from surveys (soft-data) as well as new- and unfilled-orders (hard-data) point to renewed weakness in equipment investment soon (see chart 6). While a strong contributor in the third quarter, the outsized inventory investment gain (1.3 percentage point contribution to overall GDP, reflecting in part restocking efforts amid resilient consumer spending) is not likely to be repeated in the fourth quarter. The idling of several auto assembly plants from mid-September through most of October will also contribute to the slowdown in inventory investment in the fourth quarter. This resulted from United Auto Workers' (UAW) strike against the Big Three automakers, which now appears to be resolved.

Chart 6

Chart 7

Challenging affordability backdrop likely will spur another reduction in home sales (see chart 7), and motivated sellers may revise downward price expectations. Housing starts is likely to stall at current levels, reflective of retrenchment of multifamily (rental vacancy rates have moved up in the last four quarters from its recent trough) while single-family starts grind up.

Government Policies Partially Counter The Monetary Headwind To Growth

The contribution of recent public-sector policies to growth has increasingly leaned toward net neutral to pro-cyclical, which has provided not only a buffer for demand against monetary policy headwinds in the current cycle, but also potentially added to aggregate supply (capacity to produce). Since the passage of the CHIPS and IRA in 2022, U.S. manufacturing investment has surged more than we anticipated (see chart 8). To be sure, the consequent change in investment-to-GDP ratio is still small because the share of the manufacturing sector is small. While direct impacts of fiscal policy on demand (including both discretionary fiscal policy and automatic stabilizers) will turn slightly negative in our forecast horizon, the multiplier impact from policies such as the CHIPS, IRA, and Infrastructure Investment and Jobs Act is hard to pin down and could provide a larger-than-expected impulse to growth.

Chart 8

Another important factor that has kept growth resilient amid higher interest rates, especially holding off the consumer spending slowdown, is the household sector's aggregate accumulated excess savings (as a result of "fiscal transfers" intended to minimize dislocation in household balance sheets during the pandemic). Data revisions to historical data since our last forecast show that U.S. households still have approximately $430 billion in excess savings (1.6% of GDP) relative to the pre-pandemic level (using the San Francisco Fed's method, comparing the quarterly dollar amount of saving with its pre-pandemic trend). It is quite likely that part of those savings was set aside for the resumption of student loan payments in October, a down payment for buying a house (now delayed for some), and to meet obligations ahead of California's delayed tax deadline this month.

Chart 9

At the current pace of spending above income, this cushion may last until the spring of next year (versus until September 2023 in pre-revision data heading into our previous forecast update; see chart 9). The additional savings on households' balance sheets also means the consumer spending growth slowdown to about 1% in our current forecast will start in the second quarter of next year (versus the fourth quarter of 2023 in our September publication). That said, there is a huge amount of uncertainty underlying these estimates. Households' desired savings as a percent of disposable income could now be far different from the pre-pandemic trend, for example, because the rise in household net worth since 2019 reduces the need for households to save from incomes. Or if inflation expectations have risen, it is only rational that households may not have wanted to save today due to concerns about erosion of purchasing power tomorrow. If so, the saving rate could fall even further.

Easing Labor Market Conditions

The supply and demand imbalance in the labor market continues to narrow further as payroll growth slowed on a trend basis and the employment rate of prime aged workers hovered around a 22-year high. While establishment survey showed 150,000 (plus 30,000 of temporary laid-off UAW workers) net jobs added in October--which is still well above 100,000 jobs per month that is consistent with population growth--there were certain leading indicators that suggest hiring could slow further in coming months.

The diffusion index, for example, slipped to its lowest level in October since the pandemic, a sign that job creation is increasingly dependent on a small number of sectors (see chart 10). Employment in the temporary help services sector, another tried-and-tested leading indicator of the labor market, was down 6.1% from a year ago, a decline whose magnitude has in the past been a harbinger of recession (see chart 11).

Chart 10

Chart 11

Quits rate are back down to 2019 level, and given its leading correlation with wage growth, we look for wage growth (currently 4.1% year over year) to likely continue decelerate in coming quarters towards 3.5%, consistent with inflation target of 2% plus 1.5% labor productivity growth average of the last 30 years (see chart 12).

Chart 12

With the unemployment rate rose to 3.9% in October from 3.4% in January, the Sahm rule (named after its originator, Claudia Sahm) will probably be triggered soon, which will prompt claims a recession has started. According to the Sahm rule--when the three-month average of the unemployment rate increases by more than 0.5 percentage points above the minimum of that average of the preceding 12 months, a recession has already begun. Since the current rise is partly also due to increased labor supply as much as it is to weaker demand, we would caution against relying on that indicator/rule. We like to look in tandem at both initial claims and the employment-to-population ratio of prime-aged workers, which suggests the economy is not in imminent danger (although continuing claims have risen above the 2019 average, suggesting it has become more difficult to find jobs for people once they're laid off). But given the economic growth is about to slow down materially (below trend) in the coming quarters, we expect payroll employment to contract starting sometime in late spring next year and the unemployment rate to rise through 2024 before peaking at 4.6% in 2025 (see chart 13).

Chart 13

Still Elevated Core Inflation But Headed In The Right Direction

Without much damage to the labor market, the CPI data showed a continued deceleration in inflation in 2023, with the annual growth rate on the headline index slipping by a half a percentage point to 3.2% in October after a 3.7% year-ago increase in September. The softening in the core index (excluding food and energy) was a bit more modest at a 4% annual rate in October, which remains elevated for the Fed's comfort (see chart 14). On a momentum basis, monthly sequential core CPI (three-month average annualized) accelerated in the last three months to 3.3% in October (see chart 15). Fed Chairman Jerome Powell's favored metric, the super core services excluding housing, slowed to 3.7% year over year.

Chart 14

Chart 15

We raised slightly our core CPI forecast for this year and the next. The lagged ripple effect of shelter prices (the largest component in CPI) into the CPI calculation was less than we had anticipated in the last couple of months. Still, we anticipate shelter price inflation to plummet in coming quarters, reflecting a recent pause in rent increases (see chart 16). Additionally, the Bureau of Labor Statistics now tweaked the methodology on how it incorporates health insurance in medical care component of CPI. This component jumped by 1.1% in October after 12 consecutive months of decline (down 37% in health insurance CPI in 12 months through September). Our understanding is that the increases may get steeper because of the tweaked version of the index now includes "smoothing", which delays the impact of the positive values on the current index.

Chart 16

Near The End Of Rate Hikes?

Given still elevated inflation, we anticipate another 25-bp rate hike by the Fed in December and hold afterward. In real terms, policy stance would be most restrictive in the middle quarters of 2024 as the Fed will allow passive tightening through disinflation and "hawkish" guidance, even as nominal rates are left unchanged.

Chart 17

We expect core CPI inflation to fall below 3% in the second quarter of next year and get comfortably near the Fed's inflation target by the fall of next year. We anticipate inflation on a monthly annualized basis (so-called "momentum") to consistently come in at 2% by the second quarter, which will give the Fed enough confidence to pivot away from "hold" to its first "cut" in June 2024. When all is said and done, rates will end up at 2.9% by the end of 2025 (versus the 2.4% high of the previous rate hike cycle).

Table 2

Risks To Our Baseline Forecast

The risks to our baseline growth forecast are nearly balanced. Based on ordinary risks for our baseline forecast, our pessimistic scenario has mild recession-like dynamics in 2024, while in our optimistic scenario, the slowdown is shallower and the unemployment rise starts later in 2024. This is similar to our September forecast publication.

Pessimistic scenario

In a possible pessimistic scenario, the current downturn in energy prices is short-lived, with energy prices returning to higher levels and persisting long enough to slow down the disinflationary momentum in the near term. In turn, the Fed feels compelled to keep policy rates on hold through the summer (versus June in our baseline scenario), and markets are forced to adjust to a more restrictive monetary outlook. Such a tighter monetary stance could cause a deeper drop in interest rate-sensitive sectors and private investment more broadly. Erosion of purchasing power leads to a lower confidence and spending among households than in our baseline.

Growth would stall in the first half of 2024 and pick up only gradually afterward, even as the Fed rushes to cut rates once it sees accelerating layoffs and weaker demand reigniting disinflationary forces. In this scenario, GDP would still end up eking out a 1.1% growth for the full year, deceptively sugar-coating a near flat 0.6% growth 4Q/4Q.

Both residential and nonresidential investment would be weaker for the forecast horizon, as would payrolls, raising the unemployment rate (peaking at 5.2% versus 4.6% in the baseline), and consequently, weakening demand growth versus our baseline. Inflation, although higher in the first half of next year, would drop sharply as near-recessionary winds ensue. A broad-based slowdown of this nature, combined with weak employment, would likely resemble the 2001 recession, as classified by the National Bureau of Economic Research.

Table 3

It is also possible that the economy will fare better next year than in our baseline. In one such scenario, the current strength of the U.S. economy dissipates slower than our baseline case. The slowdown comes only late next year and nowhere near as drawn out as in the baseline case.

Even though pandemic-related excess savings would have been depleted by the second quarter of next year as conventional estimates suggest, the money spent is still income to others, moving around the economy (if it hasn't been extinguished as debt repayment). That continues to grease the wheels of consumers and businesses alike for most of next year. In this scenario, consumers still show some fire-power left in their wallets in 2024, as demand for labor keeps pace with labor force supply. The unemployment rate edges down to 3.7%. World oil prices are steady at current levels (which is $5 below our baseline assumption) and continuing disinflation makes households richer in real terms. The Fed raises rates by 25 bps in December, as in our baseline scenario, but holds through much of 2024 (consistent with the resiliency of domestic demand) before starting rate cuts late in the fourth quarter of next year.

Annual average GDP growth comes in at 1.8%. The cumulative effect of rate hikes starts taking effect in earnest in late 2024 and in 2025, as some business cycle purging in the private sector starts to take place, but it is far from anything stringent like in the past.

The lack of excess leverage in the household sector that historically sets the economy into a tailspin during high interest-rate periods puts a floor on any kind of slowdown, together with the past government initiatives via the CHIPS and Science Act and the Inflation Reduction Act, still leaving their mark on business investments for green energy and semiconductor facilities. In addition, new infrastructure funds from Washington and healthy state finances leave combined government investment spending growth positive.

Table 4