In brief

  • U.S. growth has reaccelerated this year, but the main drivers seem to be one-offs that many investors are overinterpreting. Fading headwinds from higher mortgage rates and negative real incomes — rather than a broader reacceleration — contributed to a pickup this year. In Q3, personal consumption did increase by more than can be explained by higher real incomes. But this unusual drop is more likely to reverse than continue, and it could even disappear when data is revised.
  • We expect further economic deceleration, owing to even tighter financial conditions, among other headwinds. We forecast a just-below-trend pace of 1.6% q4/q4 for 2024, with growth falling off more steeply in the coming months.
  • To be clear, slowing growth does not mean an inevitable slide into recession—we see just 20% odds over the next year. Underlying momentum is still reasonably healthy, and some headwinds have dissipated.
  • With GDP topside surprises set to moderate, our multi-asset portfolios remain cautiously positioned. We are overweight duration, given our view that soaring Treasury yields have over-discounted the growth reacceleration and underpriced cooling inflation. We also prefer high-quality carry in credit and remain neutral in equities, albeit with a preference for more defensive markets.

We believe the U.S. economy is still fundamentally decelerating and should revert to a modestly below-trend pace next year. How, then, do we make sense of the recent pickup in U.S. economic activity?

First, we think underlying growth might not be as strong as officially reported. Other activity data, such as business surveys, point to much less acceleration than hard data such as GDP, particularly in Q3. The average regional Fed survey has virtually flatlined since late last year and the ISM composite rebounded just a bit in Q3. The S&P PMI did rise more significantly, but that happened in Q1 and Q2; in Q3 it pulled back. Additionally, the trend in jobs gains is still clearly slowing, and it’s virtually impossible to imagine a sharp, broad reacceleration of economic activity without a pickup in payroll growth—who’d be producing all the extra output?

To the degree that growth has accelerated, though, we see two broad drivers: fading headwinds from 2022 and consumers’ more recent spending binge.

Fading headwinds in 2023: Homebuilding bottoms, real incomes normalize

Market participants have focused on Q3 real GDP growth of 4.9% q/q saar. But Q1 looks to be this year’s strongest quarter after excluding volatile components (for example, inventory changes) that reveal little about underlying demand.

To understand what happened in Q1, we must first consider the prior year. In 2022, sharply higher mortgage rates depressed demand for housing. Since peaking in the first quarter of 2021, U.S. residential fixed investment has contracted 19.8%. Most of that pullback came in the last three quarters of 2022, when it subtracted on average 1.1% q/q saar from overall real GDP growth (Exhibit 1). Additionally, surging inflation reduced real wages on a 3m/3m saar basis, to about -2% in the middle of last year, ultimately weighing on real consumption even as saving rates declined significantly. 

Around the start of 2023, though, homebuilding bottomed out. At the same time, real incomes normalized, helped by the fact that Social Security payments caught up to inflation with a belated cost-of-living adjustment (COLA). When these factors stopped detracting so significantly from activity, the simple arithmetic consequence was a growth reacceleration.

Another factor boosted growth in the first half of this year: a pickup in business investment in structures, presumably connected to the construction of new semiconductor fabricators following the passage of the CHIPS Act This growth driver, too, looks to be a one-off level shift that is no longer contributing extra growth to the economy (Exhibit 1). 

Q3's consumption binge

Even after we account for this year’s normalization in real wages, as well as interest rate expenses that stopped rising, consumption was still unexpectedly strong in Q3. The personal saving rate dropped sharply, from 5.3% in May to 3.4% in September. In isolation that change would contribute about 5% q/q saar to Q3 GDP.

Contrast that saving rate decline with the 2021 dip. While the earlier decline owed to a negative real income shock, we find no straightforward explanation for Q3’s persistent “Barbenheimer” consumer spending binge. In any case, the saving rate, already extremely low in historical terms, seems unlikely to decline further. On the contrary, a modest upward drift seems the more probable outcome, as households’ “excess” savings are gradually exhausted and as some consumers strain to keep funding their lifestyles with growing credit card debt.

The other possibility is that the puzzling drop in the personal saving rate may disappear altogether once the data are revised. That would be reminiscent of early 2015, when data initially showed consumers saving most of their real income windfall from falling gas prices. That would have been inconsistent with historical behavior. But later revisions revealed the opposite, a big pickup in real spending. We note as well that the recent spending surge is not corroborated by proprietary measures based on anonymized aggregate spending, including those from Visa and Chase. This disconnect would seem to make a downward data revision more likely. 

Further growth deceleration ahead

While other market participants have shifted their views, why do we continue to expect growth to revert to a sub-trend pace by early next year?

For starters, higher interest rates. The recent rise in interest rates will likely weigh further on both housing and consumption activity, as well as business capex and employment decisions. Businesses will also encounter increasingly binding supply constraints. While labor force participation has unexpectedly risen this year, it now looks stretched against the long-term (demographic-adjusted) trend. Less hiring should ultimately weigh on household incomes and ultimately aggregate demand.

Nearer term, Q4 faces myriad other discrete growth headwinds. These include the recent rise in energy prices, as well as the end of the student loan moratorium, the United Auto Workers strike, and a potential federal government shutdown in mid-November. Overall, we think these headwinds could subtract about 1.5% saar from Q4 growth. The proprietary card spending data we referenced earlier already signals some pullback in consumption in recent weeks.

Finally, we acknowledge that elevated uncertainties present meaningful upside and downside risks to our core view.

Bond yields are overpricing growth

Investors' perceptions of growth reacceleration have coincided with a sharp rise in government bond yields over the last several weeks. The 10-year U.S. Treasury yield has risen about 1.5 percentage points since May, briefly scraping 5.0%, the highest level since 2007. Breakeven inflation expectations proved remarkably stable through the sell-off, which was instead driven by real yields. That mix suggests stronger real growth expectations.

Meanwhile, the yield curve continued dis-inverting and now is nearly flat, indicating lowered recession risks and a market more decisively pricing in the higher-for-longer narrative. The curve may now reflect a belief that the economy is proving even more insensitive to higher policy rates than previously thought. We approach that view with skepticism.

To be sure, other drivers could help explain recent bond market moves. Using the alternative decomposition of nominal yields into “risk-neutral” policy rate expectations and a term premium of yield above this expectation, we note a significant jump higher in this term premium. Across various model estimates it is now positive at the 10-year horizon. This shift likely owes to heightened uncertainty, especially on inflation, which should persist.

Technical drivers also seem to be accelerating these moves, following higher issuance projections from the U.S. Treasury, position squeezes and investor outflows. CFTC positioning data suggest  asset managers have  sold  long duration  holdings to reduce the risk of further losses. Leverage funds stayed very short duration, having benefited from rising yields as well as positive carry. CTA momentum players even added to duration shorts, and negatively convex mortgage hedgers had to shorten duration as yields rose.

Still, fundamental factors such as resilient growth and robust inflation data undoubtedly helped feed the bond bears. Recently, the Fed and other central banks have acknowledged the tightening in financial conditions stemming from higher back-end yields, and this could stay their hand in hiking the policy rate further. We expect that slowing growth and inflation should ultimately help yields move lower.

Asset allocation implications

Our positive duration view is plainly challenged by recent market performance. However, we support a modest duration overweight with an expectation that growth moderates in the next quarter and monetary policy tightening starts to bite. With the yield curve close to flat and absolute yields offering attractive returns, we expect less technical pressure to drive yields higher. But we are conscious that for our position to perform we need growth expectations to shift lower. While we are not expecting a sharp rally in core government bonds, it is hard not to make the case for bonds in a multi-asset portfolio at near 5% yields.

Elsewhere, we also like high-quality carry in an anticipated weak but non-recessionary environment. We remain neutral in equities albeit with a preference for more defensive markets.

Multi-asset solutions

J.P. Morgan Multi-Asset Solutions manages over USD 242 billion in assets and draws upon the unparalleled breadth and depth of expertise and investment capabilities of the organization. Our asset allocation research and insights are the foundation of our investment process, which is supported by a global research team of 20-plus dedicated research professionals with decades of combined experience in a diverse range of disciplines.

Multi-Asset Solutions’ asset allocation views are the product of a rigorous and disciplined process that integrates:

  • Qualitative insights that encompass macro-thematic insights, business-cycle views and systematic and irregular market opportunities
  • Quantitative analysis that considers market inefficiencies, intra-and cross-asset class models, relative value and market directional strategies
  • Strategy Summits and ongoing dialogue in which research and investor teams debate, challenge and develop the firm’s asset allocation views

As of September 30, 2023.