The Recession the Unemployment Rate Can't See | Eric Pachman on the Data Beneath the Jobs Report

Episode Summary

Executive Summary: Eric Pachman argues that headline labor and inflation metrics obscure a deteriorating real economy: labor-force participation is weakening, especially among older men, job growth is concentrated in low-paid care work funded by Medicaid, and inflation pressure remains embedded through shelter, gasoline, diesel, and broad service categories. He urges investors to drill into raw BLS/Census data rather than trust top-line releases.

Main Topics: Why headline labor data misleads (Priority: 5/5): Pachman contrasts the CES and CPS surveys, arguing the unemployment rate understates labor-market weakness because participation and demographic composition matter more than the headline rate. Labor-force participation and demographic decline (Priority: 5/5): He highlights falling participation among 45+ workers, especially men, and notes revisions/model changes that remove working-age people from the data, affecting taxes, Social Security, and growth forecasts. Medicaid-driven care economy as job-growth engine (Priority: 5/5): He argues much of recent employment growth comes from social assistance/home health care, funded largely by Medicaid, making job growth vulnerable to policy cuts and exposing low-wage labor dependence. Wage quality and single-income stress (Priority: 4/5): Using the wage ledger, he shows many high-employment jobs pay too little for a family to live on, driving multiple-job households and revealing the fragility of middle-class living standards. Inflation measurement flaws and broad price pressure (Priority: 5/5): He criticizes core inflation and OER-heavy CPI methodology, arguing many small categories are simultaneously spiking and that official measures understate the consumer experience of inflation. Oil, refining bottlenecks, and inflation tail risk (Priority: 5/5): He explains how crude quality, refinery design, inventories, and Strait-of-Hormuz disruptions can create persistent price shocks that feed into CPI/PCE through gasoline, diesel, freight, and goods prices. Data tooling, AI, and Data for the People mission (Priority: 3/5): He describes building open-access interactive dashboards and an AI-assisted research workflow to make public data usable for investors, journalists, and citizens, with a nonprofit structure and no paywall.

Key Arguments: The unemployment rate is too crude to capture labor weakness; participation, age mix, and not-in-labor-force trends provide a better signal. BLS/Census models and revisions can materially change how many working-age people appear in the labor force, which affects long-term economic and fiscal planning. Recent job growth is heavily concentrated in healthcare/social assistance, especially home health and personal care, which are low-paid and often publicly funded. Medicaid cuts represent a major labor-market tail risk because they could remove a key source of employment growth in many states. Median wages in many cities do not provide enough room for a single-income family of four to live above the poverty line once modest shocks are added. Core inflation is an arbitrary construct; consumers feel gasoline, groceries, and shelter far more than volatile academic exclusions. CPI shelter effects are dominated by owner’s equivalent rent, a proxy that can mask changes in the broader cost structure. Oil is not one commodity but a blend of products; refinery mismatches and shipping bottlenecks can push crack spreads and consumer prices higher. Diesel matters more than gasoline for broad inflation because it flows through trucking, rail, freight, and everyday goods. Investors should use granular, public datasets—especially QCEW and wage data—rather than rely solely on top-line government releases or consensus commentary.

Data Points: Labor force participation rate (45+ years): 50.7% 12-month average; 49.6% current reading - Eric cites a sharp drop in participation for workers age 45 and over after recent model revisions. Male labor force participation rate: 54.7% - He says this is the lowest level on record for men. Prior male participation minimum: 57.5% - He notes the previous low was in the 1990s. Civilian labor force participation rate: 61.5% - Referenced as the latest BLS headline release, which Eric says underplays the real story. Home health/personal care aides employment: 4,305,810 - The largest-employment occupation in his wage ledger example. Median hourly wage for health and personal care aides: $17.21 - Used to illustrate how low-paying the highest-employment job category is. Median annual wage for health and personal care aides: $35,800 - He argues this income is not sufficient for robust disposable income or family support. Pediatric surgeons median salary: $559,000 - Presented as the top-paying occupation in the wage ledger. Pediatric surgeons employment: 1,190 - Shows high pay but tiny employment base relative to care jobs. Home-care worker share of job growth affected by Medicaid cuts: About 32-33 states would have had negative job growth without Medicaid care economy (end of 2025 data) - He uses state-level QCEW analysis to show dependence on Medicaid-funded employment growth. Medicaid cuts: $1 trillion over 10 years - He frames this as a major risk to care-economy employment. CPI shelter weight: About 26-27% of CPI - He says owner’s equivalent rent is the biggest single influence on CPI. CPI anomaly categories: 42-45% of items - He says this share of categories has breached one standard deviation over historical mean in his visualization. AAA gasoline vs CPI correlation: ~0.99 R-squared - He claims gasoline CPI can be predicted almost exactly from monthly AAA retail gasoline prices. Inventory comparison horizon criticized: 5 years - He says official comparisons against only five years can hide how severe current petroleum inventory declines are. QCEW coverage: 98% of all W-2 employees - He cites QCEW as the most valuable non-modeled job dataset. Single-income stress test: $15,000 shock makes the whole country red - He says adding a $15k unexpected expense pushes median-income households below the poverty threshold in his city-level model. Data production pace: About one new original study per day - He attributes the pace to AI-assisted workflows and editorial oversight.

Pivotal Quotes: "You will not see the recession coming with the unemployment rate." — Eric Pachman: On why labor-market headlines fail to capture the real deterioration in participation and underlying employment health. "All models are wrong, some are useful." — Eric Pachman: Used to explain revisions to population and labor-force estimates and why model updates can materially alter the data picture. "We created COVID for oil markets from a supply chain standpoint." — Eric Pachman: Describing the effect of repeated Strait-of-Hormuz disruptions on oil flows, refinery economics, and inventory stability.

Implications: Listeners should treat headline labor and inflation prints as incomplete. The deeper risk is a weak, low-wage, care-dependent labor market plus supply-chain/oil shocks that could keep inflation sticky and hit households, margins, and policy expectations.