Raghuram Rajan pitches tax credits as an antidote to AI job losses - CNBC TV18

TEXT START: Economist Raghuram Rajan has suggested that governments may consider offering tax credits to companies that provide additional training to employees as artificial intelligence transforms the workplace, highlighting the need to limit AI-driven job displacement.

The Dissection

This is a transition-management memo presented as an antidote. It acknowledges displacement, then routes the problem back through familiar institutions: firms train workers, governments subsidize retention, productivity lowers prices, demand expands, and jobs supposedly regenerate. Its real function is to preserve the employment-wage-consumption circuit without confronting whether humans remain economically necessary inputs.

The Core Fallacy

The article confuses retraining with restored indispensability. Under P1, if AI achieves durable cost and performance superiority across cognitive work, moving humans into adjacent tasks does not preserve mass employment; it moves them along the shrinking perimeter of useful labor.

A tax credit is a transfer and timing mechanism, not a source of human comparative advantage. An AI-token tax could slow deployment, correct fiscal asymmetry, or fund redistribution. It cannot make a human worker competitive when automation is structurally cheaper or better. The Jevons effect is also conditional: greater demand may increase total output without increasing human labor demand. AI-heavy firms can produce more while employing fewer people.

Hidden Assumptions

• Retraining will create durable roles at anything like the scale of displaced roles.
• New entrepreneurial activity and expanded demand will be human-intensive rather than AI-intensive.
• Firms will retain workers because subsidies outweigh the competitive pressure to automate.
• AI supervision will become a large, stable occupation rather than a temporary implementation niche.
• Governments can measure and enforce token taxes, including against foreign providers, without driving activity offshore or accelerating substitution.
• Human labor’s disadvantage is mainly a tax distortion rather than declining productive necessity.
• Corporate reputation and access to talent will remain powerful enough to preserve broad employment.
• AI will remain complementary to humans even after integration barriers fall.

These assumptions convert a structural problem into an administrative one. They assume away P2 and P3: human-only economic domains cannot be preserved at scale, and the majority eventually lose access to economically necessary labor.

Social Function

Primary classification: transition management. Secondary classifications: partial truth and ideological anesthetic.

The article is not pure propaganda. Integration costs, adoption lags, productivity gains, and some human-AI complementary roles are real. But it turns the harder question—what happens when humans are no longer needed in sufficient numbers—into the more manageable question of whether workers received enough training. It offers institutions a policy program and corporations a respectable retention narrative while leaving ownership and control of AI capital untouched.

The Verdict

Tax credits and AI taxes are hospice care, not an antidote. They may buy time, preserve selected occupations, and redistribute part of the gains. They cannot restore productive participation once P1, P2, and P3 mature. Rajan’s proposal manages the speed and political appearance of the decline; it does not reverse the mechanism killing the post-WWII order. The article mistakes a delayed autopsy for survival.