The Engels Pause: Economists Identify the Most Serious Hidden Threat from AI Revolution
As artificial intelligence continues its rapid integration into workplaces around the world, economists are sounding alarms about a potentially devastating consequence that extends far beyond individual job losses. The mass displacement of human workers by AI systems could trigger a significant decline in personal income tax revenues, creating a fiscal crisis that governments are ill-prepared to address. This phenomenon, which some researchers are calling a modern iteration of the “Engels Pause,” represents perhaps the most underappreciated economic threat of the AI revolution.
The term “Engels Pause” originates from historical economic analysis, named after Friedrich Engels, who documented the conditions of the working class during the Industrial Revolution. During the early 19th century, despite unprecedented technological advancement and productivity gains, workers’ wages stagnated for nearly four decades while factory owners accumulated vast wealth. Economic historians have noted that between 1790 and 1840, British workers saw virtually no improvement in their living standards despite the nation’s economic output soaring. Today’s economists fear we may be entering a similar period, where AI-driven productivity gains flow primarily to capital owners while workers face displacement and wage stagnation.
The mechanics of this modern threat are straightforward but profound. Personal income taxes constitute a cornerstone of government revenue in virtually every developed economy. In the United States, individual income taxes account for approximately 50% of federal revenue, while in many European nations, the figure ranges between 25% and 40%. When AI systems replace human workers, those displaced individuals stop earning taxable wages. Even if they eventually find new employment, the transition period creates immediate revenue shortfalls. Moreover, many new positions may offer lower wages than the jobs that were automated, permanently reducing the tax base.
Corporate profits, meanwhile, may surge as companies reduce labor costs through automation. However, tax systems in most countries are structured to collect a larger share of revenue from individual earners than from corporations. Corporate tax rates have declined globally over the past four decades, and multinational technology companies have proven particularly adept at minimizing their tax obligations through complex international structures. This asymmetry means that even if AI generates substantial economic value, governments may capture a declining share of that wealth through existing tax mechanisms.
Historical precedent offers both warnings and limited comfort. During previous technological disruptions, from the mechanization of agriculture to the computerization of manufacturing, economies eventually adapted. New industries emerged, workers acquired new skills, and tax revenues recovered. However, these transitions often took decades and caused significant social upheaval along the way. The pace of AI advancement appears to be accelerating far more rapidly than previous technological shifts, potentially compressing the disruption timeline while expanding its scope across multiple industries simultaneously.
Some economists argue that the solution lies in fundamentally restructuring tax systems for an AI-driven economy. Proposals include implementing taxes on robots or automated systems, increasing corporate tax rates, introducing wealth taxes, or establishing universal basic income programs funded through new revenue mechanisms. Bill Gates notably suggested a “robot tax” in 2017, arguing that if a human worker earning $50,000 pays income and payroll taxes, a robot performing the same work should generate equivalent tax revenue. However, such proposals face significant political and practical obstacles, including questions about how to define and measure automation for tax purposes.
The international dimension adds further complexity to this challenge. Countries that move aggressively to tax AI and automation risk driving technology companies to more permissive jurisdictions. This creates a potential race to the bottom, where nations compete to attract AI investment by offering favorable tax treatment, ultimately reducing global revenue collection. International coordination on AI taxation remains minimal, and the rapid pace of technological change continues to outstrip regulators’ ability to respond effectively.
As governments worldwide grapple with mounting debt levels and aging populations demanding expanded social services, the potential erosion of income tax revenue from AI displacement could not come at a worse time. Policymakers face the challenge of supporting workers through a potentially wrenching economic transition while simultaneously finding new revenue sources to fund that support. The Engels Pause of the Industrial Revolution eventually ended, ushering in an era of broadly shared prosperity. Whether the AI revolution follows a similar trajectory, or produces a more prolonged period of fiscal and social disruption, may depend largely on how quickly and effectively governments adapt their economic policies to this new technological reality.