government failure
Suppose a market is failing and the government steps in to fix it — say, with a price cap to make housing affordable. Years later, builders have stopped building, apartments sit shabby because landlords have no reason to maintain them, and a black market thrives. The cure made things worse. Government failure is exactly this: when state intervention, meant to correct a market failure or improve outcomes, ends up producing a worse result than leaving things alone.
Government failure is the deliberate counterpoint to market failure, and it has recognizable causes. Information problems: officials can't know everyone's true costs and preferences, so they set the wrong price or quota. Incentive problems: politicians chase votes and the next election, bureaucracies seek bigger budgets, and special interests lobby for rules that benefit the few (this study is called public choice theory). Unintended consequences: people adapt to rules in ways planners didn't foresee. And plain administrative cost and delay. The result can be a deadweight loss as large as, or larger than, the one intervention was meant to remove.
This concept matters because it dismantles a tempting but fallacious argument: 'the market is imperfect, therefore the government should act.' That skips a step — you must also show the government can do better in practice, not just in theory. Honest economists treat the choice as a comparison between two imperfect institutions, neither idealized. Acknowledging government failure is not an argument against all intervention (that would be its own error); it is a reminder to weigh the real, flawed cure against the real, flawed disease — case by case, with evidence.
A strict rent control law is imposed to help tenants. Over time landlords stop maintaining units and few new ones get built, so the housing shortage it was meant to ease can deepen instead — a market failure 'fix' that creates its own deadweight loss.
When the cure is worse than the disease.
Government failure cuts both ways: 'the market is imperfect, so intervene' is invalid without showing government can do better — but 'government can fail, so never intervene' is equally invalid. The real test is comparing two flawed options case by case.