Estimation of ability to raise tax revenues of localities is a policy relevant problem of extreme importance. Revenue sharing funds, education aid, and other grants from the federal and state governments use, or will in the future use, formulas in which either tax effort or fiscal capacity is an important variable. A perfect measure would have to come from the context of a general equilibrium model, taking account of such factors as capitalization of tax and service rates into property values, relocation of tax bases as a result of taxation levels, and competition among local units for tax bases. Lack of local data precludes the attempt to build such ambitious models for practical policy use. For many local units the maximum amount of feasible data available (given state tax agency cooperation) would be income, property values, and population data. Most past efforts to estimate local fiscal capacity have relied on either property values or income. The methodology presented in this paper creates an index which effectively adds to the above-mentioned two another factor: ability to shift taxes to non-residents. Because of the omnipresent data difficulties, proxies for ability to tax outsiders—in the form of values of commercial, industrial and resort property—are used in the actual estimation. The resultant practical estimation method is, as the title implies, an improved, not a conceptually perfect method. It is, however, a method which bases its estimates on norms derived from actual taxing behavior, and on the hypothesis that income, wealth, and the ability to tax outsiders are the important determinants of ability to raise local tax revenues. Implicit in the acceptance of such norms is the value judgement that for policy purposes it is reasonable to expect local units to tax themselves at a statewide average level.