Goldman Warns: 30% Recession Probability as Construction Slumps and Energy Shocks Mount

The current economic expansion, which began in June 2009, has now lasted over 17 years. It officially became the longest in U.S. history in July 2019, surpassing the previous record of 120 months set by the expansion from 1991 to 2001. This longevity is a defining feature of the post–World War II era, where the average expansion has lasted about five years. The sheer length of this cycle invites a historical lens: past record-setting expansions have often ended with a recession, but their growth profiles and labor market dynamics offer a complex benchmark.

Comparing growth rates reveals a notable shift. Under the Trump administration, the annual GDP growth rate averaged 2.6 percent, outperforming the 2.2 percent rate seen during the Obama years. This acceleration, even after accounting for temporary headwinds like the General Motors strike, suggests a period of relatively robust expansion. Yet, the current profile is not without its vulnerabilities. The labor market, long seen as a key pillar of resilience, shows early signs of strain. The unemployment rate ticked up to 4.4% in February 2026, its highest level since late 2021. While still well below historical norms, this rise from a 50-year low signals a potential softening that could test the expansion's durability.

Historically, expansions often fade gradually, not abruptly. The record-setting run from 1991 to 2001, for instance, saw its final years marked by a slowdown in growth and rising unemployment before the dot-com bust. The current setup-a long cycle with a recent uptick in joblessness-echoes that pattern. The "extinction" thesis, which posits that a record expansion is due to end, finds some support in this historical precedent. However, the comparison is not straightforward. The post-2009 expansion has been characterized by a slower, more stable growth trajectory than the rapid booms of past eras. Its endurance may be less about explosive momentum and more about the absence of a major shock, a dynamic that could prolong its life but also make its eventual end more difficult to anticipate.

Testing the Resilience: Consumption, Construction, and the Q1 Growth Shock

The recent economic data reveals a clear sectoral divergence, testing the expansion's resilience. The Atlanta Fed's GDPNow model provides a stark snapshot of this volatility. In early 2025, the model estimated a sharp -3.7% annualized GDP decline for Q1, a significant quarterly contraction that raised immediate recession concerns. That shock has since receded. For the first quarter of 2026, the model has rebounded to a 2.7% growth estimate, up from 2.1% just a week prior, showing a recovery from the prior quarter's pace. This swing underscores the economy's sensitivity to specific data points and highlights the fragility beneath a steady headline.

The recovery is not uniform. A key driver of the recent uptick was a surge in private investment, which nowcasts have climbed. Yet, one critical sector shows clear weakness. January's construction spending fell 0.3%, a notable miss against expectations. This contraction in a major component of fixed investment is a red flag, echoing historical patterns where a downturn in building activity often signals broader economic softening. It suggests that while some parts of the economy are rebounding, others remain under pressure.

Consumer sentiment, the other pillar of growth, shows a similar mixed picture. The Conference Board's Consumer Confidence Index edged up in February, but the improvement is modest and the index remains well below the four-year peak seen in late 2024. This persistent gap indicates that underlying consumer optimism remains subdued, likely capping the potential for a robust consumption-led expansion. The recent energy price shock adds another layer of uncertainty, with Goldman Sachs warning it could lift recession probabilities and push unemployment higher.

Viewed through a historical lens, this pattern of sharp, sector-specific shocks is familiar. Record expansions often end not with a single catastrophic event, but with a series of contractions that gradually erode momentum. The Q1 2025 plunge to -3.7% was a classic example of such a shock, driven by a confluence of factors that can be difficult to predict. The rebound in 2026 suggests some resilience, but the concurrent weakness in construction and muted consumer confidence indicate the economy is still navigating a choppy path. The severity of the recent shock is now less about a single quarter's number and more about whether these sectoral pressures can be contained or if they will spread, ultimately challenging the expansion's longevity.

The Recession Threshold: What Constitutes a Contraction?

The debate over a looming recession hinges on a precise definition. A recession is not merely a slowdown; it requires an actual contraction in economic activity. The National Bureau of Economic Research (NBER), the official arbiter of U.S. business cycle turning points, defines it as "a significant decline in economic activity that is spread across the economy and lasts more than a few months." This is a higher bar than the common rule of thumb of two consecutive quarters of negative GDP growth. The NBER looks at a basket of monthly indicators, including payrolls, personal income, and industrial production, to assess whether the economy has entered a sustained downturn.

Applying this test to the current data reveals a clear distinction. The economy is undeniably slowing. We see slower job growth and sector-specific contractions, like the 0.3% drop in construction spending in January. Yet, there is no evidence of a broad-based contraction across the key pillars of the economy. Manufacturing, government, and residential sectors have not all turned negative simultaneously, which historical patterns show is typically necessary to trigger a recession. The recent GDPNow model swings-from a sharp quarterly decline to a rebound-highlight volatility, but not a sustained, multi-sector contraction.

Labor market data offers a similar picture of cooling, not collapse. The unemployment rate has risen to 4.4% in February 2026, its highest level in years, signaling a softening. However, flows data from March shows the market is not breaking down. The number of people losing jobs and becoming unemployed was very close to its 12-month average, as were the numbers finding work. The modest rise in the jobless rate appears driven more by a slight increase in people leaving the labor force than by a surge in layoffs. As one analysis notes, "the overall picture remains positive."

The bottom line is that the economy is in a period of transition, not a contraction. The pattern is one of gradual softening in some areas, not the simultaneous downturn in multiple sectors that defines a recession. Until that broader decline materializes, the expansion remains intact. The current setup is more akin to the gradual fade seen in the final years of past record expansions than a sudden collapse.

Catalysts and Risks: The Path from Slowdown to Contraction

The economy is navigating a narrow path between a sustained rebound and a deeper downturn. The key watchpoint is whether the recent uptick in growth is durable or a fleeting bounce. The Atlanta Fed's GDPNow model has shown remarkable volatility, swinging from a sharp -3.7% annualized GDP decline for Q1 2025 to a 2.7% growth estimate for Q1 2026 just weeks ago. This choppiness underscores the fragility of the current setup. The latest update, however, shows a slight retreat to 2.0% for Q1, driven by a weaker construction spending report. This pattern of sharp swings is a classic feature of record expansions nearing their end, where temporary shocks can mask or exaggerate the underlying trend.

Geopolitical tensions are now a primary catalyst for downside risk. Rising energy prices, fueled by Middle East disruptions, are beginning to reshape the outlook. Goldman Sachs warns this energy shock, combined with tighter financial conditions, is increasing recession probabilities and has raised its 12-month forecast to 30%. The bank projects unemployment will rise to 4.6% by year-end, a clear signal that a softening labor market could accelerate. This scenario echoes historical episodes where external shocks-like oil crises in the 1970s-combined with domestic policy missteps to create stagflationary pressures.

The risk of stagflation is a persistent shadow. Even as growth slows, inflation remains elevated. The potential for restrictive monetary policy to persist, despite weakening data, creates a dangerous dynamic. As one analysis notes, "if the Fed is unable to cut interest rates in the face of falling growth, the potential for stagflation would rise." This is the precarious position: the economy needs stimulus to support growth, but inflationary pressures may force a policy stance that could stifle it. The historical lens here is clear; the 1970s experience shows how such a mix can prolong economic distress.

The bottom line is that the expansion's longevity may be its own vulnerability. Its slow fade has been gradual, but the accumulation of headwinds-geopolitical, inflationary, and policy-related-could compress the timeline for a downturn. The path from slowdown to contraction often involves a series of sectoral shocks that spread. The recent weakness in construction and the energy price shock are early signs of this process. Whether the rebound in private investment can hold, or if these pressures will coalesce into a broad-based contraction, will determine if this record expansion finally meets its historical fate.