US Recession and Deflation Risk Analysis

The Great Divergence: A Forensic Economic Analysis of the US "Rolling Recession," Deflationary Risks, and Structural Resilience (2022–2025)

1. Introduction: The Anatomy of an Economic Anomaly

The macroeconomic trajectory of the United States economy from mid-2022 through the conclusion of 2025 presents a theoretical paradox that challenges foundational economic models. This period, characterized by the most aggressive monetary tightening cycle in forty years colliding with an unprecedented fiscal expansion, defied the binary classifications of "recession" versus "expansion" that have historically defined business cycle analysis. Instead, the economy entered a complex state of disequilibrium, a phenomenon increasingly categorized by scholars and market analysts as a "rolling recession." In this scenario, aggregate economic activity—measured by Gross Domestic Product (GDP)—masked profound, sequential contractions in specific sectors, ranging from housing and technology to manufacturing and regional banking.1

The central research inquiry of this report evaluates the validity of the deflationary and recessionary risks that dominated economic discourse during this timeframe. Specifically, it investigates the divergence between flashing "red" leading indicators—such as the historic contraction of the M2 money supply and the inversion of the Treasury yield curve—and the realized economic data which showed stubborn inflation and delayed labor market deterioration. While the consensus forecast in 2022 and 2023 anticipated a "hard landing" akin to the 2008 Global Financial Crisis or the roughly synchronized downturns of the 1970s and 1980s, the outcome was a fragmented deceleration. This report argues that the "recession" was not absent but rather distributed temporally and sectorally, effectively acting as a slow-motion correction that prevented a systemic collapse while inflicting localized depressions across the economic landscape.

To understand this dynamic, one must analyze the interplay of opposing forces: the deflationary pressure of the Federal Reserve’s quantitative tightening and the inflationary support of "fiscal dominance," where persistent federal deficits averaging nearly 6% of GDP counteracted monetary restriction. This analysis synthesizes data from the Federal Reserve, the Bureau of Economic Analysis (BEA), and independent market research to construct a definitive account of this unique economic cycle. It explores how the US economy navigated the expiration of the "free money" era, the bursting of pandemic-induced asset bubbles, and the emergence of structural headwinds in commercial real estate and labor markets, ultimately determining that the US experienced a "growth-recession"—a period defined by high nominal activity but deteriorating underlying health.

2. The Monetary Contraction: M2 Mechanics and the Deflation Mirage

The most technically alarming signal for economists during the 2022–2025 period was the behavior of the US money supply. For the first time since the Great Depression of the 1930s, the M2 money supply—a broad aggregate of liquidity that includes physical currency, demand deposits, savings accounts, and money market funds—experienced a sustained and significant contraction. This development reignited a century-old debate regarding the Quantity Theory of Money and its predictive power over price levels (deflation) and economic output (depression).

2.1 The Historic M2 Decline: Magnitude and Precedent

Between the peak in April 2022 and October 2023, the M2 money supply contracted by approximately 4.76%, a reduction of over $1 trillion in nominal terms.4 By mid-2024, the aggregate decline stood at roughly 3.21% from the all-time high, settling at roughly $21.025 trillion.5 This contraction was not merely a statistical blip; it represented a fundamental reversal of the monetary regime that had prevailed for decades. Historically, M2 has grown in lockstep with the economy, fueled by credit expansion and population growth. A decline of this magnitude is structurally rare.

Analysts tracking long-term monetary trends, such as Nick Gerli of Reventure Consulting, emphasized the dire historical precedents associated with such a move. In the last 155 years of US economic history, year-over-year declines in M2 of at least 2% have occurred only five times: 1878, 1893, 1921, 1931-1933, and the current 2023-2024 period.4 In every previous instance, the monetary contraction was a precursor to a depression or a severe panic characterized by double-digit unemployment rates.5

The mechanism of the 2022–2024 contraction differed from historical episodes. In the 1930s, the money supply collapsed due to widespread bank failures and the destruction of deposits—a solvency crisis. In contrast, the recent contraction was an engineered policy outcome driven by the Federal Reserve's Quantitative Tightening (QT) program and the rapid rise in interest rates. As the Fed stopped reinvesting the proceeds of maturing Treasuries and Mortgage-Backed Securities (MBS), liquidity was drained from the system. Simultaneously, depositors moved funds from low-yielding bank deposits to higher-yielding non-M2 instruments or engaged in reverse repurchase agreements (Reverse Repos), effectively sterilizing a portion of the money stock.

2.2 The Deviation from Historical Determinism

Despite the terrifying historical correlation between M2 collapse and economic depression, the 2023–2025 cycle broke the pattern. While the money supply contracted, the US economy did not spiral into a deflationary depression with mass unemployment. Understanding this "broken" correlation requires analyzing the starting conditions of the cycle.

The 2022 peak in M2 was not a "normal" cyclical high; it was the summit of a mountain of liquidity created during the COVID-19 pandemic. In 2020 and 2021 alone, M2 surged by over 26%, an unprecedented injection of stimulus fueled by fiscal transfers (stimulus checks, PPP loans) and monetary accommodation (zero interest rates, QE).8 Consequently, the subsequent contraction of ~4.7% did not cut into the "muscle" of the transactional economy required for daily operations. Instead, it merely drained the "froth"—the excess liquidity that had accumulated in savings accounts and investment portfolios. Even after the decline, M2 levels in 2025 remained trillions of dollars above the pre-pandemic trend line.7

Furthermore, the velocity of money—the frequency with which a unit of currency is used for transactions—remained historically depressed.8 In a high-velocity environment, a reduction in the money stock leads to an immediate and sharp drop in nominal GDP. However, in the post-pandemic environment, the velocity of M2 was low enough that the reduction in supply was absorbed by idle balances rather than forcing a reduction in spending. The "money multiplier" effect was dormant; banks were not lending out every available dollar, so the reduction in reserves did not trigger a cascading credit crunch for the broader economy, although it did stress specific banking cohorts.

2.3 The "Cathie Wood" Deflation Thesis vs. Sticky Reality

A prominent voice predicting deflation during this period was Cathie Wood of ARK Invest. Wood argued that the combination of the M2 contraction and the rapid technological deflation driven by Artificial Intelligence (AI) and robotics would lead to a collapse in the Consumer Price Index (CPI).11 Her thesis rested on the idea that innovation is inherently deflationary—reducing costs and increasing efficiency—and that the Federal Reserve was making a policy error by tightening into this deflationary wave.13

Wood’s forecast suggested that headline inflation would not only return to 2% but potentially turn negative. She pointed to commodity price corrections and inventory gluts in retail as early evidence.12 However, the data through 2024 and 2025 largely refuted the "imminent deflation" narrative in the aggregate. While goods prices did disinflate significantly—with durable goods prices falling and sectors like used cars seeing price drops—services inflation remained remarkably sticky.

The "sticky" components of the CPI basket, particularly shelter and insurance, defied the monetary signal. Shelter inflation, which accounts for a massive portion of the CPI, continued to rise at monthly rates of 0.2–0.3% well into 2025.14 This persistence was driven by the lag in housing data and the structural shortage of housing units, which kept rents elevated despite higher interest rates. Furthermore, wage growth in the service sector, while cooling, remained above the level consistent with 2% inflation, fueled by a labor market that remained structurally tight due to demographic shifts (retirements) and lower immigration levels compared to pre-pandemic trends.

Thus, the US economy experienced a bifurcation: a "goods recession" with deflationary characteristics, consistent with the M2 signal, and a "services boom" with inflationary persistence, consistent with the fiscal impulse. The result was not the aggregate deflation predicted by monetarists and techno-optimists, but a grinding "stag-flationary" persistence where inflation stabilized around 3%—too high for the Fed to cut rates aggressively, but too low to destroy consumption completely.

3. Fiscal Dominance: The Counter-Cyclical Leviathan

While monetary policy slammed on the brakes, fiscal policy kept a heavy foot on the accelerator, creating a dissonance that defined the 2023–2025 economy. This phenomenon is best described through the lens of "Fiscal Dominance," a theoretical framework where government debt and deficit levels become so large that they overwhelm the central bank's ability to control inflation through interest rates alone.16

3.1 The Structural Deficit Anomaly

In Fiscal Year 2024, the US federal deficit reached 6.3% of GDP, decreasing only marginally to 5.9% in Fiscal Year 2025.18 In absolute terms, the government ran a deficit of approximately $1.8 trillion in FY2025.20 To place this in context, deficits of this magnitude relative to GDP have historically been reserved for periods of existential crisis—such as the World Wars, the Great Financial Crisis of 2008, or the COVID-19 pandemic. To maintain a deficit near 6% of GDP during a time of relative peace and with unemployment rates hovering near historic lows (approximately 4%) is an economic anomaly without modern precedent.18

This fiscal expansion acted as a massive counter-cyclical force. While private sector activity in interest-rate-sensitive sectors contracted, the public sector continued to expand. Government consumption and investment contributed directly to GDP growth, while transfer payments (Social Security, Medicare, and various subsidies) supported household incomes. The bipartisan nature of this spending—driven by defense commitments, infrastructure bills (IIJA), and semiconductor incentives (CHIPS Act)—ensured that the fiscal tap remained open regardless of the political gridlock in Washington or threats of government shutdowns.22

3.2 The Interest Rate Feedback Loop

The interaction between the Federal Reserve's rate hikes and the Treasury's deficit spending created a perverse feedback loop that exacerbated the "Fiscal Dominance" dynamic. As the Fed raised rates to fight inflation, the cost of servicing the federal debt exploded. By 2025, gross interest payments on the national debt surpassed $1 trillion annually, eclipsing the entire defense budget.19

This dynamic suggests that the interest rate hikes themselves became inflationary through the fiscal channel. The $1 trillion in interest payments represented a massive transfer of wealth from the government to the private sector (bondholders), effectively acting as a stimulus check to the investor class. Theoretical literature on fiscal dominance supports this observation: when debt levels are high, raising rates increases debt service costs, expanding the deficit further, which in turn can stimulate demand and inflation, negating the original intent of the rate hike.17 This "unpleasant monetarist arithmetic," as termed by economists Sargent and Wallace, appeared to be playing out in real-time, explaining why the economy remained resilient despite restrictive monetary policy.

4. The Yield Curve: The "Broken" Barometer?

The inversion of the US Treasury yield curve, specifically the spread between the 10-year and 2-year Treasury notes, has been the single most reliable predictor of recessions since the 1950s. An inversion (where short-term rates are higher than long-term rates) signals that the market expects tighter policy in the near term to break the economy, leading to lower rates in the future.

4.1 The Great Inversion of 2022–2024

In July 2022, the 10-year minus 2-year spread turned negative and remained inverted for a record duration, surpassing the streaks seen before the 2008 and 2000 recessions.24 At its depth, the inversion reached levels not seen since the Volcker era of the early 1980s.25 Standard recession probability models, such as those maintained by the New York Fed, utilized this spread to forecast a recession probability exceeding 70% for the 12 months ahead.26

Yet, month after month, the predicted recession failed to materialize in the headline GDP data. This led to a widespread debate in 2024 about whether the indicator was "broken" or a "false positive".28 Critics of the indicator argued that the inversion was driven by the "term premium" being artificially suppressed and by the market's faith in the Fed's ability to lower inflation without crushing growth (the "immaculate disinflation" thesis).

4.2 The Bear Steepening and the "Term Premium" Return

By late 2024 and throughout 2025, the yield curve began to "un-invert" or normalize. However, this normalization did not occur through the typical "bull steepening" mechanism, where the Fed aggressively cuts short-term rates to rescue a faltering economy. Instead, the market witnessed a "bear steepening".29 Long-term yields (the 10-year and 30-year) rose faster than short-term yields, driven by investors demanding a higher premium to hold long-term US debt.

This shift reflected a growing recognition of the structural risks facing the US fiscal position. With deficits compounding and inflation settling above 2%, the "term premium"—the extra compensation investors demand for the risk of holding long-term bonds—returned with a vengeance. The un-inversion was thus a signal of fiscal concern rather than economic optimism.

While the yield curve may have appeared to be a "false positive" regarding a synchronized NBER recession, it accurately predicted the severe stress in the banking sector (which holds these inverted assets) and the transactional freeze in the housing market. The "lag" between inversion and recession is variable, and historical analysis suggests that the recession often begins after the curve un-inverts, as the full weight of tight credit conditions finally breaks the labor market—a pattern that began to emerge in late 2025.

5. Sectoral Analysis: The Reality of the Rolling Recession

The aggregate stability of US GDP concealed violent rotations of capital and activity beneath the surface. This phenomenon, termed a "Rolling Recession," implies that while the economy as a whole never contracted, individual sectors took turns experiencing recession-level downturns. This sequential processing of pain allowed the broader economy to absorb shocks that, if synchronized, would have been catastrophic.1

5.1 Phase I: The Interest Rate Shock (Housing and Tech) – 2022

The first phase of the rolling recession was a direct consequence of the Federal Reserve's initial rate hikes. The housing market, the most interest-rate-sensitive sector of the economy, entered a deep freeze. Existing home sales collapsed by nearly 40% in 2022 as 30-year mortgage rates surged from historical lows of 3% to over 7%.30 This freeze devastated mortgage originators, title companies, and real estate brokerages. However, unlike the 2008 crisis, home prices did not collapse. The "lock-in effect"—where homeowners with 3% mortgages refused to sell—restricted supply so severely that it put a floor under prices, preventing the negative wealth effect that typically accompanies a housing downturn.

Simultaneously, the technology sector experienced a valuation reset. The "Tech Wreck" of 2022 decimated the share prices of non-profitable growth companies, typified by the 67% decline in the ARK Innovation ETF.12 This valuation collapse triggered a wave of "white-collar" layoffs in Silicon Valley, affecting tens of thousands of high-income workers.2 Yet, because this labor weakness was geographically concentrated (San Francisco, Seattle) and sector-specific, it did not bleed into the national unemployment rate.

5.2 Phase II: The Industrial Contraction (Manufacturing and Freight) – 2023–2024

As the economy moved into 2023, the recession rolled into the goods-producing economy. The ISM Manufacturing Purchasing Managers' Index (PMI) dropped below the neutral 50.0 level in late 2022 and remained in contraction territory for a historic stretch of consecutive months, signaling a recession in American factories.31

By late 2025, the manufacturing sector was still struggling to emerge from this slump. The ISM PMI reading of 48.7 in October 2025 highlighted persistent weakness in new orders and export demand.31 This "goods recession" was exacerbated by the "bullwhip effect" reversal: retailers who had over-ordered inventory during the supply chain crisis of 2021 spent 2023 and 2024 destocking, slashing orders to factories. The freight sector mirrored this decline, with shipping volumes and trucking rates collapsing from pandemic highs, creating a "freight recession" that squeezed logistics providers.2

5.3 Phase III: The Financial Squeeze (Regional Banks and CRE) – 2023–2025

The most dangerous phase of the rolling recession struck the financial system. The rapid rise in interest rates devalued the bond portfolios held by banks, creating unrealized losses that threatened solvency. This crystallized in the Banking Crisis of March 2023, with the failures of Silicon Valley Bank (SVB), Signature Bank, and First Republic Bank.33

Crucially, the banking crisis did not end in 2023; it merely slowed down. The "long tail" of the crisis extended through 2024 and 2025, driven by the deterioration of Commercial Real Estate (CRE) loans. Regional banks, which hold a disproportionate share of CRE debt, faced a slow-motion disaster as office vacancy rates in major cities stabilized near 20%—a structural shift caused by remote work.34

Table 1 illustrates the persistent nature of bank failures well past the initial 2023 shock, refuting the narrative that the crisis was "contained" to a few weeks in March 2023.

Bank Name

Location

Closing Date

Total Assets (Approx.)

Silicon Valley Bank

Santa Clara, CA

March 10, 2023

~$209 Billion

First Republic Bank

San Francisco, CA

May 1, 2023

~$229 Billion

Citizens Bank

Sac City, IA

Nov 3, 2023

$66 Million

Republic First Bank

Philadelphia, PA

April 26, 2024

$6 Billion

First National Bank of Lindsay

Lindsay, OK

Oct 18, 2024

Small Cap

Pulaski Savings Bank

Chicago, IL

Jan 17, 2025

$49.5 Million

The Santa Anna National Bank

Santa Anna, TX

June 27, 2025

$63.8 Million

33

These failures in 2024 and 2025, though smaller in asset size than SVB, signaled deep systemic rot in the community banking layer. The collapse of Republic First Bank in 2024 and others in 2025 demonstrated that the combination of "higher for longer" rates and toxic CRE assets continued to claim victims long after the headlines had moved on. The SPDR S&P Regional Banking ETF (KRE) reflected this stress, struggling to regain its pre-crisis highs and trading with elevated volatility as investors priced in the risk of further consolidation.39

6. The Labor Market: Cracks in the Armor

Throughout 2022 and 2023, the US labor market was the primary argument against the recession thesis. With the unemployment rate hovering near 50-year lows (3.4%–3.7%), proponents of the "soft landing" argued that a recession was impossible with such robust employment. However, leading indicators painted a darker picture that eventually converged with official data in 2025.

6.1 WARN Notices vs. Official Data: The Lag

In 2023 and early 2024, a notable divergence emerged between the official U-3 unemployment rate and Worker Adjustment and Retraining Notification (WARN) Act notices. WARN notices, which legally mandate that large employers provide advance notice of mass layoffs, surged in key states like California, Texas, and New York.41 Historically, a rise in WARN notices leads the unemployment rate.

The disconnect in this cycle—where WARN notices rose but the unemployment rate did not—can be attributed to "labor hoarding" and severance buffers. Companies, scarred by the post-pandemic labor shortages, were reluctant to fire workers even as demand slowed, opting instead to cut hours or freeze hiring. Additionally, generous severance packages in the tech and finance sectors delayed the entry of laid-off workers into the unemployment insurance system, masking the real-time destruction of jobs.

6.2 The 2025 Deterioration and the Sahm Rule

By late 2025, the labor market shield began to crack. The official unemployment rate drifted upward, reaching 4.6% in November 2025, a significant increase from the cycle lows.43 This rise of nearly a full percentage point triggered the "Sahm Rule"—a recession indicator that signals a downturn when the three-month moving average of the unemployment rate rises by 0.50 percentage points or more relative to its low during the previous 12 months.44

The composition of this unemployment rise was qualitative as well as quantitative. The U-6 unemployment rate, which includes discouraged workers and those working part-time for economic reasons, rose more sharply than the headline number.43 This indicated that while massive layoffs were not universal, the quality of employment was degrading. Full-time jobs were being replaced by part-time roles, and the "quit rate" plummeted as workers lost the confidence to switch jobs—a classic sign of a cooling labor market. The transportation sector, a bellwether for the physical economy, saw its unemployment rate tick up to 4.8% 45, confirming the weakness in the goods-producing sectors.

7. The AI Factor: The Engine Under the Hood

Any analysis of the 2023–2025 economy is incomplete without accounting for the role of Artificial Intelligence (AI). The public release of ChatGPT in late 2022 ignited a capital expenditure (CapEx) boom that acted as a powerful stimulant, masking weakness in other areas.

The race to build AI infrastructure spurred hundreds of billions of dollars in investment from "hyperscalers" like Microsoft, Google, and Amazon.46 This massive spending on data centers, energy infrastructure, and semiconductors (benefiting Nvidia and the broader semiconductor supply chain) created a localized boom. It effectively substituted for the decline in traditional industrial CapEx. Moreover, the stock market rally driven by the "Magnificent 7" created a wealth effect that supported high-end consumer spending, further bifurcating the economy between the "AI haves" and the "industrial have-nots".46

However, this AI-driven growth raised questions about sustainability. By 2025, debates intensified regarding the "productivity paradox"—whether the massive investment in AI was yielding commensurate economic returns or merely creating a financial bubble.46 While the long-term potential of AI to boost GDP is widely accepted (with estimates of adding trillions to global GDP by 2030), the short-term impact was primarily an investment impulse rather than a broad-based productivity miracle.

8. Conclusion: The "Soft Landing" Was a Rolling Crisis

Reviewing the data from July 2022 through the end of 2025, the verdict on the US economy is nuanced. The "Hard Landing" predicted by the inverted yield curve and M2 contraction did not manifest as a singular, catastrophic event like 2008. However, the "Soft Landing" narrative of painless disinflation is equally flawed.

The reality was a Rolling Recession. The US economy traded a sharp, acute crisis for a prolonged period of disjointed, sector-specific depressions.

* Monetary Policy successfully deflated the goods and housing bubbles but failed to quickly crush services inflation due to fiscal interference.

* Fiscal Policy prevented a demand collapse but at the cost of structurally higher deficits and interest rates.

* The Labor Market bent slowly, absorbing the shock through reduced churn and hours before finally breaking in late 2025.

As the US enters 2026, the buffers that protected the economy—excess pandemic savings and labor hoarding—are largely exhausted. With the unemployment rate at 4.6% and the term premium returning to bond markets, the risks have shifted. The danger is no longer overheating, but rather that the "rolling" recession finally coalesces into a unified downturn as the cumulative weight of high rates and debt service costs suppresses growth. The economic cycle of 2022–2025 will be remembered not for a singular crash, but for the "Great Divergence" between mechanical economic signals and a reality distorted by fiscal intervention and structural shifts.