Loading...
In 2019, the newly elected governor of Verdana Province announced the Green Canopy Initiative — an ambitious plan to increase forest cover by 40% within five years. The centerpiece: a bounty of $12 per verified tree planted, paid directly to landowners. The first year looked like a triumph. Satellite data showed 2.3 million new saplings across the province. Governor Mira held press conferences beside rows of freshly planted eucalyptus. International media praised the program as a model for climate action. But by year two, ecologist Kai noticed something disturbing. Old-growth forest patches — some over 200 years old — were disappearing. Farmers like Ren had done the math: a hectare of ancient forest earned nothing, but clearing it and replanting could yield $14,400 in bounties. Ren's ne...
Popular framing: Corrupt farmers exploited a well-meaning environmental program.
Structural analysis: A per-tree bounty made the measurable proxy diverge from the actual goal — Goodhart's Law converted 'forest cover' into 'tree-counts'. The incentive structure made clearing old-growth and replanting monoculture more profitable than preserving the original forest, producing cobra-effect dynamics where the cure drove the disease faster. Second-order effects (lost topsoil, dried wells, flooding) and iatrogenic damage from misaligned incentives were predictable from the policy's measurement geometry; replace the farmers with anyone and the bounty geometry produces the same destruction.
The popular frame locates agency in bad actors and solutions in enforcement, preserving the assumption that the incentive mechanism was fundamentally sound. The structural frame reveals that the mechanism itself was the pathogen — the intervention caused the harm it sought to prevent (iatrogenics), the economic logic of clearing-and-replanting was a predictable second-order effect of the bounty design, and the program rewarded the metric so effectively it destroyed the goal (cobra effect). Closing this gap matters because the popular fix — better monitoring and stricter rules — leaves the core substitution error intact and will reproduce the same failure in the next program.
