The employer of last resort (ELR) program is Hyman Minsky's specific policy proposal for addressing the structural problem that monetary policy cannot solve: the unemployment that persists when private firms will not hire at wages that workers can accept. The program, developed across Minsky's later career and refined by post-Keynesian economists including Pavlina Tcherneva, provides a federal guarantee of employment at a base wage for anyone willing to work, performing socially useful tasks that the private market does not adequately supply. The program operates as a permanent stabilizer: it absorbs workers during downturns, releases them to private employers during booms, maintains aggregate demand through all phases of the cycle, and establishes a floor beneath which private-sector wages cannot fall. The Opus 4.6 simulation identifies the program as a specific candidate for the institutional dam the AI economy needs but has not built.
There is a parallel reading that begins from the political economy of implementation rather than the elegance of theory. The ELR proposal assumes a benevolent state with genuine capacity to coordinate socially useful work at scale—an assumption that dissolves under empirical scrutiny. What actually happens when you create a permanent federal employment apparatus is not Minsky's stabilizer but a new terrain of capture: contractor ecosystems emerge to manage the "socially useful work," unions organize to protect the guaranteed positions, and the program becomes another site where concentrated interests extract rents while performing the theater of public benefit. The "base wage floor" becomes a new battlefield where political coalitions fight over what counts as adequate—not a technical parameter but a permanently contested political question that makes the program's stability the opposite of stabilizing.
The fiscal reality is more severe than defenders acknowledge. The program's defenders claim it "pays for itself" through maintained aggregate demand, but this accounting ignores the opportunity cost of capital deployed to administer a permanent employment bureaucracy rather than invested in productivity-enhancing infrastructure or research. During AI transitions specifically, the program risks becoming a holding pen—workers "employed" in make-work that neither builds skills nor produces value, indefinitely separated from the productivity frontier that continues advancing. The displaced software engineer doesn't need a caregiving job administered by a federal program; she needs exposure to the new production landscape. The ELR solves the immediate income problem while potentially calcifying the underlying displacement by removing the pressure that drives genuine adaptation.
The ELR concept responds to a specific failure of monetary policy: the central bank can inject liquidity into the banking system, but liquidity does not guarantee employment. Banks can receive liquidity and lend it to asset speculators rather than to firms that would hire workers. The injection may inflate asset prices without restoring aggregate demand. The ELR addresses this gap by placing the fiscal authority directly in the labor market — employing workers at a guaranteed wage without waiting for private demand to recover.
The program is fundamentally different from conventional public works or unemployment insurance. Public works programs are typically temporary, cyclically triggered, and targeted at specific infrastructure needs. Unemployment insurance provides income replacement but does not employ. The ELR is permanent, universally available to anyone willing to work, and designed to absorb whatever labor supply private employers decline to absorb at the prevailing wage structure.
Applied to the AI transition, the ELR addresses the specific problem of structural displacement that existing stabilizers cannot manage. The displaced software engineer whose fifteen years of framework expertise has become less valuable does not need temporary income replacement — she needs a bridge to new value-creating capability. The ELR provides that bridge: paid employment performing socially useful work (community health, environmental restoration, public infrastructure, caregiving, educational support) while retraining for the post-AI labor market. The program combines income security with productive engagement, avoiding the dignitary costs of pure transfer programs and the aggregate-demand failures of pure labor-market adjustment.
The program's political economy is unfavorable during booms. The private sector perceives it as labor-market competition. Fiscal conservatives perceive it as unfunded liability. Progressive critics sometimes perceive it as insufficient — preferring higher guaranteed wages or more ambitious redistribution. Minsky's response, consistent with his general framework, was that the program's function is stabilization rather than transformation — it moderates the cycle without attempting to eliminate it, and its value is most visible during corrections rather than during the booms when its construction must be undertaken.
Pavlina Tcherneva's 2020 The Case for a Job Guarantee provides the most developed contemporary articulation of the program, including operational details about wage levels, task selection, federal-local coordination, and integration with existing social insurance. The proposal has entered mainstream policy discussion in the United States through the work of the Levy Economics Institute and in the United Kingdom through the New Economics Foundation.
Minsky developed the concept across the 1970s and 1980s, initially as part of his critique of what he called the "War on Poverty" programs that sought to treat poverty without addressing the structural unemployment that produced it. The concept matured in Stabilizing an Unstable Economy (1986) and in later papers at the Levy Economics Institute.
The contemporary extension has been led primarily by post-Keynesian economists including L. Randall Wray, Pavlina Tcherneva, Stephanie Kelton, and Mathew Forstater — the "Modern Monetary Theory" school, which has adopted the ELR as a central policy proposal. The MMT treatment has been controversial; the ELR itself has been more widely accepted across economic traditions.
Universal access. Available to anyone willing to work, without means-testing or contribution requirements.
Base wage floor. Establishes a floor beneath which private-sector wages cannot fall, without imposing a ceiling.
Cyclical flexibility. Absorbs workers during downturns, releases them to private employers during booms.
Socially useful work. Tasks include community health, environmental restoration, infrastructure, caregiving, educational support — work the private market underprovides.
Permanent stabilizer. Operates continuously rather than being activated by recessions; the continuous operation is the stabilizing feature.
The program faces significant political and economic objections. Monetarists argue it would be inflationary; ELR defenders respond that the program sets a wage floor rather than bidding up existing wages. Progressives sometimes argue the base wage is too low; defenders respond that the program is a floor rather than a ceiling and complements rather than replaces other income-support programs. Conservatives argue the program expands government beyond its proper scope; defenders argue that unemployment imposes costs far larger than the program's fiscal burden, and that the comparison of costs must include the full social costs of unemployment rather than only the fiscal outlay of the program.
The right weighting depends entirely on which question you're asking. On the question of theoretical function—does a wage floor stabilizer address monetary policy's employment gap?—Edo's framing is fully correct (100%). Minsky identified a real gap in the policy toolkit, and the ELR addresses it directly. The contrarian critique doesn't contest this; it contests whether the elegant theory survives contact with implementation.
On implementation terrain, the weighting shifts dramatically toward the contrarian view (70/30). The historical record of large-scale federal employment programs—from New Deal work programs to contemporary job training initiatives—shows consistent patterns of capture, administrative bloat, and performance theater. The question isn't whether these problems are inevitable, but whether the ELR's structure contains mechanisms to resist them. The answer is unclear. Tcherneva's operational proposals address coordination and task selection but don't fully solve the political economy problem of a permanent program that creates its own constituency of administrators and beneficiaries resistant to the program's stated goal of releasing workers to private employment.
The synthetic frame the topic benefits from: think of ELR not as a single program but as a policy primitive—a basic building block that can be implemented well or poorly. The primitive itself (guaranteed employment at base wage performing useful work) is sound. The implementation challenge is preventing it from becoming an end rather than a bridge. This suggests the real design question isn't whether to build the ELR but how to build accountability mechanisms that keep it focused on transition rather than permanence—metrics tied to outflow rather than headcount, funding structures that reward skill-building rather than enrollment expansion, and sunset provisions that force periodic democratic re-authorization rather than bureaucratic autopilot.