Rational Addiction (Becker-Murphy)

In the AI Story

The model identifies the failure mode with clinical precision. The rational addict discounts future costs too heavily relative to present returns. The discount rate — the rate at which future consequences are weighted against present satisfactions — is the parameter that separates the addict from the moderate consumer. A high discount rate means the future matters less. The present benefit looms large. The future cost recedes into a haze of discounted insignificance.

This model, developed to explain heroin and cigarettes and alcohol, is the most precise analytical framework available for understanding what happened to knowledge workers in the winter of 2025. The viral Gridley post — 'Help! My Husband Is Addicted to Claude Code' — was not a metaphor. It was a textbook case of Becker-Murphy rational addiction. The husband was not wasting time. He was producing real output. The consumption was not destructive in the way heroin is destructive. It was productive. And the productivity was the trap.

Adjacent complementarity operates in productive addiction through the same mechanism as in substance addiction, but the reinforcement pathway is more insidious because it is socially validated. Each building session with AI produces tangible output — a feature, a prototype, a solution. The output generates satisfaction, professional recognition, and the specific pleasure of seeing an idea become real. This satisfaction increases the marginal utility of the next session. The ordinary working day, the pre-AI rhythm of meetings and documentation and slow iteration, now feels intolerably slow — not because it was ever inefficient in absolute terms, but because the complementarity has recalibrated the builder's internal standard.

Two forces push the discount rate higher in productive addiction that do not operate in substance addiction. First, social validation: the substance addict's consumption is stigmatized; the productive addict's is celebrated. Second, the visibility asymmetry between present returns (immediately observable code, features, progress) and future costs (invisible erosion of judgment, relationships, attentional capacity). The builder cannot see the costs until they have compounded beyond easy reversal.

Origin

Becker and Murphy's paper generated predictions that the data confirmed, ending the argument in Becker's characteristic style: not through persuasion but through accuracy. Addicts respond to expected future price increases by reducing current consumption — the announcement of a future cigarette tax reduces smoking today, before the tax takes effect. The finding vindicated the forward-looking assumption and established rational addiction as a serious analytical framework despite its initial reception as an offense against moral common sense.

Key Ideas

Adjacent complementarity. Current consumption increases the marginal utility of future consumption, creating the feedback loop that drives the escalation.

The discount rate is the critical parameter. A high enough rate makes present satisfaction dominate future cost. The addict is not irrational — the addict is maximizing on a path the addiction itself has shaped.

Discount rates are not stable. They are affected by the consumption. Each high-intensity session trains the nervous system to weight immediate returns more heavily, pushing the rate upward with use.

Binge-crash cycles. Systems governed by adjacent complementarity with high discount rates and delayed costs tend toward unstable equilibria — periods of intense use followed by collapse when the cost threshold is finally breached.

Debates & Critiques

Critics argued the model licenses cruelty — if addicts are rational, we need not intervene. Becker's response was empirical rather than moral: the model predicts how consumption responds to price changes, taxes, and expected future costs. Those predictions support intervention precisely because they identify the points at which intervention can change behavior. The model does not say addiction is good. It says addiction responds to incentives — which is the precondition for addressing it through policy.