The paper shows that a comparison of the appropriately-weighted sum of households' marginal willingness to pay for a public good with the net effect of the increased supply of the public good on shadow, as distinct from actual, government revenue is a generally valid rule for public good...