1. Of course, these criticisms are not universal. One clear, though expensive, counterexample is the work of the Regional Research Institute at West Virginia University under the direction of William H. Miernyk.
2. One example is Leontief's early input-requirements model which is based on a division of locally produced commodities into two classes, those with nationally balanced production and those with regionally balanced production. Isard's empirical implementation of this model met with little success, requiring a substantial amount of primary local data. See Wassily Leontief and others,Studies in the Structure of the American Economy (New York: Oxford University Press, 1953), chaps. 4 and 5. A more recent attempt to use this scheme is reported in Wassily Leontief, Alison Morgan, Karen Polenske, David Simpson, and Edward Tower, “The Economic Impact—Industrial and Regional—of an Arms Cut,”The Review of Economics and Statistics, XLVII (August, 1965), pp. 217–241. A second example is a procedure using equations based on the gravity or potential model to estimate interregional commodity flows. Both data and computational needs are well beyond our limited resources. See Wassily Leontief and Alan Strout, “Multiregional Input-Output Analysis,” in Tibor Barna, ed.,Structural Interdependence and Economic Development (New York: St. Martin's Press, 1963), chap. 7.
3. Other measures than output may be used, e.g., participation income, value added, population, employment, etc. For a convenient summary discussion of location quotients, see Walter Isard,Methods of Regional Analysis (Cambridge: The M.I.T. Press, 1960), pp. 123–126. The advantages in computing location quotients are offset by several limitations: the method assumes that taste and expenditure patterns, household-income levels, production practices, and industry mixes are similar across the nation.
4. CONSAD Research Corporation,Regional Federal Procurement Study (prepared for Office of Economic Research, United States Department of Commerce Contract 7-35211, October, 1967), pp. 3.7–3.13. CONSAD used this technique to construct input-output models for each state in the nation but extended their calculations to insure that the sums of state-industry outputs did not exceed national-industry outputs.
5. This approach was attributed to Charles Leven by Tiebout inIntroduction and Section I: Techniques of Regional Forecasting, Proceedings of the Summer Institute on Regional Economic Development, Williamstown, Massachusetts, July-August, 1966 (Pittsburgh: Regional Economic Development Institute, Inc., 1966).