Our model implies the probability of a recession starting within the next 12 months is 25%-30% (see chart 1). The model is based on key forward-looking financial market indicators and the Organization for Economic Co-operation and Development's leading economic index through May. While the probability of a recession in the near term has fallen since last fall due to easing financial conditions, it remains elevated at twice the unconditional probability of recession (baseline comparison) since World War II. The still-high probability of a recession compared with normal times largely reflects the inverted yield curve (see The Mighty Yield Curve Is Still Inverted).

Chart 1

Chart 2

Risks to real GDP growth based on financial conditions deteriorated slightly since our January publication (after improving in the second half of last year) but remains near historically normal levels. Given the leverage, credit, and risk components of the Chicago Fed's National Financial Conditions Index up to the first quarter, the near-worst possible outcome (the bottom fifth percentile of the distribution) for annual quarterly GDP growth in the next 12 months turned negative once again, albeit only slightly (see chart 3).

Chart 3

First-quarter 2024 real GDP growth slowed from the quarter before to below-trend at 1.3% (quarter-over-quarter annualized). On a year-over-year basis, growth was 2.9% (versus 3.1% in the fourth quarter). Peak growth is likely behind us in this post-pandemic recovery. We anticipate year-over-year growth will soften more as the year progresses to about 1.8% by year-end. The underlying potential of the economy limits the sustainability of further high growth.

Risks could increase considering the ongoing conflict in the Middle East (which currently appears contained but could escalate). Real interest rates are still restrictive by most estimates, and it's possible that inflation could surprise on the upside for longer. (May inflation out last week was a step in the right direction toward the Fed's inflation target after several months of not so encouraging data.) If disinflation progress stalls or inflation even starts rising, the Fed may reverse its easing, which would tighten financial conditions and dampen growth. Our dashboard of leading indicators doesn't quite indicate that the coast is clear, either. Coincident indicators are either in a late cycle or approaching one, which generally means there is limited scope for a short-run cyclical boost to growth.

Mixed Signals From The Leading Indicators

Our comprehensive dashboard comprising nine leading indicators presents us a more nuanced perspective. As of May, four indicators flashed negative signals, while five exhibited neutral or positive indications. Term spread, manufacturing new orders, consumers sentiment, and bank lending conditions remain in the red. Notably, consumer sentiment moved back to negative (after remaining positive since December last year) as consumers were wary about the 12 months ahead of business conditions as interest rates continued to remain higher for longer.

Credit spread

Credit spreads narrowed since our last business cycle barometer report in January despite persistent higher interest rates and elevated inflation (though softened quite a bit, but not enough for the Fed to cut rates soon). Speculative-grade credit spreads narrowed by 44 basis points (bps) in May from January and by 125 bps from October (when the 10-year bond yield surged close to 5%). The investment-grade spread inched down by 16 bps in May from January and by 49 bps since October (see chart 4). A further drop in spreads for both speculative-grade and investment-grade bonds indicates optimism and risk appetite among investors on the back of healthy economic growth and corporate profits.

Chart 4

Consumer sentiment

The University of Michigan's preliminary consumer sentiment index (MSCI) of May dropped by 8.1 points to a six-month low of 69.1 from 77.2 in April. A similarly intense drop was last seen almost two years ago in June 2022 when the sentiment fell to 50 from 58.4. The index declined further to a seven-month low of 65.6 in the first half of June, suggesting that households are now more pessimistic about current, as well as expected, welfare under the weight of higher interest rates and still-elevated consumer prices. However, the Conference Board's consumer confidence index (CCI) increased for the first time in the last three months to 102 from 97.5 in April (see chart 5). The divergence in the two indexes could be attributed to various factors, including differences in survey timing, variations in sample size, and nuances in question detail. The drop in MSCI--a leading indicator of future consumer spending--suggests that continued higher interest rates are eroding consumers' purchasing power. On the other hand, CCI--a leading indicator of labor market conditions--showed that fewer consumers expected deterioration in future business conditions, job availability, and income.

Further, both short-term and long-term inflation expectations remain above the pre-pandemic range of 2.4%-2.6%. According to the University of Michigan's survey, consumers expect prices to rise by 3.3% over the coming year from 3.2% last month. Long-term inflation expectations held steady at 3.0% for the second month in a row.

Chart 5

Chart 6

Personal income and outlays

U.S. goods-related consumption growth has slowed quite notably in the first half of 2024. In May, core retail sales (excluding autos, gas stations sales, and building material) was up 0.2% month over month, although this came after a downwardly revised April (-4%). In inflation adjusted terms, core retail sales were up 0.5% in May following a decline of 0.6%. From a recent peak in December, the real value of core retail sales is up just 0.6% annualized through May--much lower track in comparison with a 5.2% rise in full-year 2023. An even weaker consumption pattern is evident from restaurant spending, a discretionary item. Year to date, real restaurant spending is down 5.8% annualized versus up 5.7% in full-year 2023. We don't expect a full-blown retrenchment in consumption, but, at the margin, even a modest slowdown in consumption growth would slow GDP growth.

Meanwhile, real disposable income dipped 0.1% month over month and was up 1.1% year over year in April. Real personal consumption also fell 0.1% month over month and was up 2.6% year over year (chart 7). That said, the labor market continued to display resilience, rebounding strongly with 272,000 job gains in May after adding 165,000 jobs in April. Average monthly job gains through May this year was 248,000, a tad down from an average 251,000 in 2023. This suggests that despite higher interest rates, businesses in the U.S. continued to display buoyancy in hiring.

Initial jobless claims, a leading indicator of the labor market, drifted higher in June from near historical lows earlier in the year (see chart 8). The four-week average of initial claims was 222,250 as of June 8. The uptick in past few weeks was more than expected given seasonal factors (but still below last year at the same time), which suggests that the U.S. labor market was losing momentum in June, leaving some job seekers to look for work longer.

Chart 7

Chart 8

Manufacturing activities

The Institute of Supply Management (ISM) manufacturing index has been in contraction since November 2022--the only exception was in March this year when it rose just above 50. The contraction seems broad based. Since the 1970s, a manufacturing purchasing managers' index (PMI) below 45 has aligned with a recession, but it has not reached that level yet.

The ISM's new orders index--a leading indicator of manufacturing activity--also slipped further into contraction territory. The index came in at 45.4 in May, down 3.7 points from April. Besides, the S&P Global Market Intelligence' manufacturing new orders index for May remained in contraction territory for the second straight month following a three-month expansionary phase. Both surveys depict the weakness in overall operating conditions of the manufacturing sector. In contrast, manufacturing nondefense new orders (excluding aircrafts), which goes into the GDP accounting directly, picked up again in April on a year-over-year basis, though the pace of increase has softened even as it remained in positive territory. The divergence between soft data (PMIs) and the actual Census Bureau data in the past several quarters makes us cautious in concluding that the downswing in manufacturing has bottomed out (charts 9-10).

Chart 9

Chart 10

Building permits

Building permits, a leading indicator for housing activity, fell for the second straight month, down 3% month over month in April, below market expectations, to 1.44 million units on seasonally adjusted annualized rate (saar) (chart 11). It was the lowest monthly permits since January last year. Multifamily unit (5+ units) permits remained weak for the past several months, suggesting higher interest rates are dragging permits for multifamily units more than single-family units. On a year-over-year basis, building permits decreased by 6%. Single-family housing starts, which account for the bulk of homebuilding, were down by almost 1% in April to 977,000 units (saar) compared with the previous month, but down 5% on a year-to-date basis.

Meanwhile, the National Association of Home Builders housing market confidence index inched down by 5 points to 45 in May (below market expectations) amid a pickup in mortgage rates that drove home builder confidence down. The average rate of popular 30-year fixed mortgages increased further to 7.06% through May, up by 7 bps.

Chart 11

Federal Reserve's Lending Conditions Survey

According to the Federal Reserve's Senior Loan Officer Opinion Survey (SLOOS), a smaller net percentage of banks reported tightened lending standards across most loan categories over the first quarter (with the exception of auto loans with a higher share reported having tightened), signaling the strongest headwinds to growth momentum may be behind us (chart 12). Still, lending standards remain relatively tight, which shows that economic uncertainty remains somewhat elevated. Concerns around commercial real estate valuations remain particularly heightened. Demand has weakened for most consumer loans.

That said, a broader indicator of financial stress, which incorporates a wide variety of financial variables--the Chicago Fed's nonfinancial corporate index (NFCI)--remains negative as it continues to signal that the U.S. financial system is operating at below average levels of risk, credit, and leverage. A zero value for the NFCI can be thought of as the U.S. financial system operating at a historically average level of risk, credit, and leverage (chart 13).

Chart 12

Chart 13

Financial conditions impulse

In addition, we also like to track the Financials Conditions Impulse on Growth (FCI-G), which assesses the extent to which financial conditions pose headwinds or tailwinds to economic activity. As of April, the FCI-G increased by 0.16 percentage points to 0.24% (chart 14). A reading of 1% above (or below) the zero line means that financial conditions are a notable headwind (tailwind) to economic activity that is equivalent to a 100-bps drag (boost) on GDP growth over the following year.

Chart 14

Where Do The Coincident Indicators Stand?

Monthly data on the key coincident indicators suggest that the economy is in "late cycle." It is likely to slow below potential growth--as we have been reporting--but that growth will remain positive. Headline payroll job gains remain solid, but there are mixed signals in other metrics that suggest we are about at pre-pandemic conditions. The three-month average pace of monthly jobs gain is running at 249,000 in May (chart 15). Most major sectors added jobs in May, but it has been the case since 2023 that hiring has been concentrated in less cyclically sensitive industries such as health care and public sectors. Given the immigration flow of the last 12 months, we think near-term inflation-neutral monthly pace of jobs gains is near 200,000 (longer run is 100,000). Furthermore, the employment-to-population ratio of prime-age workers (25-54)--a demographic unaffected by voluntary retirement and schooling--is above the pre-pandemic high (chart 16).

Despite the upside surprise in hiring in May, the data elsewhere appear less strong and showing signs of moderation. Besides, job openings fell further to 8.1 million, a steady decline of 3.4 million from 12 million in May 2022, suggesting a cooling labor market. There are now just 1.2 job openings per unemployed worker, which is in line with the pre-pandemic ratio. Quit rate is back to the 2017-2019 average. Temporary help employment, which has historically led overall hiring, has continued to see outright layoffs.

Chart 15

Chart 16

Production at factories in April remained steady from the previous month, but it fell by 0.4% on a year-over-year basis. The manufacturing sector continues to face headwinds from higher borrowing costs and tighter credit conditions. The industrial capacity utilization over the last three months hovered around 78.5%--closer to the pre-pandemic average (chart 18). Overall, the manufacturing sector has been on a broad-based slowdown from its peak in 2022.

Chart 17

Chart 18

The inventory-to-sales ratio for wholesalers dipped modestly in the first quarter of 2024 to 1.35% compared to the first half of 2023. But it stayed above its long-term average and the 2021 level, when we saw deterioration in inventory management due to supply disruptions. Retailers, however, have shown better inventory management as the ratio broadly remained stable at the end of the first quarter compared to the 2023 average and continued to improve significantly from 2021 and early 2022 levels when we saw the ratio dip sharply. Improvement in inventory management also helped ease price pressure in the country (chart 19) from a multidecade high. Similarly, the manufacturing ratio is steady and has been in the tight range of 1.45-1.48 since 2021.

Chart 19

Appendix

Definitions of positive, neutral, and negative signals

• Positive: Overall economic activity will continue to expand in the near term, without an obvious slowdown.
• Neutral: Overall economic activity will continue to expand but may be slower. Right after a recession, a neutral signal indicates that the recovery may have just started.
• Negative: Overall economic activities will start to contract shortly afterward, roughly within a year.

The signals refer to the near-term growth perspectives suggested by the changes in a certain leading indicator during the past three months (since our previous publication). To determine whether the changes have been statistically significant (at a 5% significance level) to change the signal, we carry out a simple t-test against the historical sample. To capture the expansion and recession periods of business cycles without bias, we select the series between December 1982 and June 2009 as the historical representation, which includes three complete business cycles. For shorter series, we use the entire series until the end of 2017. Note that a negative growth signal is not equivalent to a recession in the near term.

Table 2