Asymmetric Information in Wage Contracts: Experimental Evidence and Welfare Implications

TEXT START: Compared to self-employment or performance-based pay, fixed wages reduce earnings uncertainty but are prone to market distortions through moral hazard and adverse selection.

The Dissection

This is a competent microeconomic autopsy of a labor market that still requires human data-entry workers. It measures two frictions: fixed wages reduce productivity by 8.74%, while higher wage offers attract more productive workers. It then converts those estimates into welfare calculations and a tax proposal.

The paper’s contribution is real but tightly bounded: it improves the allocation of human labor inside an intact wage system.

The Core Fallacy

The paper treats contract design as the binding problem. Under the Discontinuity Thesis, the binding problem is whether human data-entry labor remains economically necessary at all.

If AI achieves durable cost and performance superiority, moral hazard and adverse selection become secondary debris. The primary event is substitution. A tax on piece rates may rearrange incentives among remaining human workers and firms, but it cannot restore mass demand for human data entry, preserve the wage-consumption circuit, or prevent productive participation from collapsing.

The reported $0.88 social cost per dollar is a local efficiency result, not a defense against systemic obsolescence. It optimizes the contract attached to a job whose underlying production function is being deleted.

Hidden Assumptions

• Human data-entry labor remains necessary for long enough for contract reform to matter.
• The measured productivity effects survive widespread AI substitution.
• Employers do not respond by automating, outsourcing, or abandoning the affected contract forms.
• Fixed wages and piece rates remain the relevant choice set.
• Taxation can correct selection without accelerating displacement or reducing labor demand.
• Recovering welfare lost to asymmetric information is more important than addressing the larger welfare shock from the disappearance of the job itself.
• Human productivity remains the central economic variable rather than a temporary benchmark against machine output.

These assumptions are not irrational within the experiment. They are simply inadequate for a transition in which the human labor market itself loses strategic importance.

Social Function

Classification: partial truth and transition management, with a mild ideological-anesthetic effect.

The study does not fabricate its findings. Moral hazard and adverse selection are genuine contract distortions. But its policy frame gives decision-makers a finer steering wheel on a vehicle whose engine is being removed. It makes labor-market optimization appear like labor-market preservation.

The Verdict

Technically useful, strategically myopic. The paper can improve welfare allocation while humans remain indispensable. It does not challenge P1, P2, or P3; it operates entirely inside the pre-collapse wage regime. Under the Discontinuity Thesis, this is not a solution to the future of work. It is better bookkeeping for one of the jobs most exposed to extinction.