Does Paying Money Actually Help Break Habits? What the Evidence Says

Short answer: yes, with an asterisk. Putting your own money at risk is one of the few habit-change techniques with consistent support in controlled trials, and the effect size is respectable rather than miraculous. The asterisk is design: money helps when the consequence is automatic and verifiable, and does very little when you can quietly excuse yourself from it.

Here's the evidence, and the design rules it implies.

The core mechanism: loss aversion

Kahneman and Tversky's prospect theory established what's now one of the most replicated findings in behavioral science: losses loom larger than gains, roughly by a factor of two. A $50 loss feels about as bad as a $100 gain feels good. That asymmetry is why deposit-based schemes (risk your own money) reliably outperform reward-based schemes (earn someone else's) even when the dollar amounts match.

Habit relapse is a fast, emotional cost-benefit run: does this cost anything, and will anyone know? Motivation tries to argue with the calculation. A pledge changes its inputs.

The trial evidence

• Smoking, deposit contracts: In a large employer trial published in the New England Journal of Medicine (Halpern et al., 2015), smokers assigned to deposit programs, where their own $150 was at risk, quit at substantially higher rates than those offered equivalent rewards. Fewer people accepted the deposit condition, but those who did quit far more often: commitment filters for seriousness and then amplifies it.
• The Philippines CARES study (Giné, Karlan, Zinman): smokers offered a savings-deposit contract that forfeited to charity if they failed a urine test were about 3 percentage points more likely to be smoke-free at 12 months, a large relative gain for the population.
• Weight loss and gym attendance: multiple studies (including Royer et al. on gym habits) find commitment contracts increase follow-through during the contract period, with effects that fade when the contract ends unless the habit has taken root.
• stickK's founding research at Yale grew from exactly this literature; commitment contracts as consumer products are the applied version.

Honest caveats: effects are strongest while the contract is active, some studies show decay after it ends, and uptake is self-selected. Financial stakes are a scaffold for building the habit, not a personality transplant. That's fine; scaffolds are what you need during the first months, which is when most relapses happen.

The design rules the research implies

1. Your own money, at risk, beats rewards. Deposit beats bonus in nearly every head-to-head.
2. Verification must not rely on your testimony. The Philippines study used urine tests, not diaries. A contract you self-report defects with you. This is the single most common failure in DIY attempts; see the Beeminder for porn breakdown.
3. The consequence must be automatic. Any step where a human, including you, decides whether to enforce is a step where enforcement dies.
4. Price it for your worst moment. The amount must beat the urge, not impress your calm self.

Applying it to porn

Porn is an awkward fit for generic contracts because relapse happens alone and leaves no verifiable trace, which breaks rule 2. Pledgely resolves it by charging the enabling act instead of the act itself: it blocks porn system-wide on Android via a local VPN, you set a daily pledge of $1 to $100, and the pledge is charged only if you deactivate the blocker (pause it, disable the VPN, or remove device admin in Hard Mode). Keep it on and every daily hold is released. Verification is the blocker's own state: automatic, objective, and immune to 2am storytelling. Nothing you browse is logged.

That satisfies all four rules, which is rarer than it sounds; the comparison across stickK, Beeminder, Forfeit, and Pledgely in commitment contract apps shows how each one trades off rule 2.

So: does paying money help? The literature says yes, when you can't excuse yourself from paying. Build the contract so that the only way out is through your card, and the research is firmly on your side.

Next: Apps that charge you money when you relapse