The Dynamic Variable Input-Output Model : An Advancement from the Leontief Dynamic Input-Output Model

The Dynamic Variable Input-Output (VIO) model extends the static single regional version of the Multiregional Variable Input-Output (MRVIO) model which is a general equilibrium model applied to the Leontief Input-Output model. The Dynamic VIO model which incorporates time dimensions can describe the real situation more accurately while maintaining the computational simplicity. Under the model, the investment expenditures are directly linked with the profit maximizing behavior of firms. Both technical coefficients and capital stock requirement coefficients include price terms, and they become variable instead of being fixed. By gathering investment terms together instead of separating them as is done in Leontief's Dynamic output equation, we not only preserve the consistency between time-specific dynamic multipliers and total dynamic multipliers, but also get over the shortcoming of the negative occurences of the Leontief's multiplier matrices, particularly the impact (initial period) multiplier matrix. Using the 15 sector interindustry transaction table derived from the 1987 U.S. benchmark input-output table, we estimate dynamic output and income multipliers for all industries. Empirical results show that, over all industries, dynamic total multipliers of the Dynamic VIO model are larger than static multipliers of the static Household Interactive VIO (HIVIO) model. Static multipliers of the static Household Interactive VIO (HIVIO) model which lie between Leontief's static type I and type II multipliers are exactly the same as the impact (initial period) multiplier of the Dynamic VIO model. Similarly, dynamic total multipliers of the Dynamic VIO model are larger than those of the Leontief static type I and type II multipliers, and multipliers of the Leontief dynamic (in nested form) IO model. Thereby, the study demonstrates that the static HIVIO model and the Leontief IO models (both static and dynamic (in nested form) underestimate actual impacts. The study also shows that the sum of all time-specific multipliers of the dynamic VIO model equals to dynamic total multiplier; thus, consistency is ensured in the dynamic multipliers of the Dynamic VIO model. The multipliers of the dynamic VIO model vary among different industries, and they decrease as the number of time lags increase with the initial period impact as the largest. The percent distributions of multipliers over time periods reveal that ripple effects of spending are mostly recognized during the first four periods. The dynamic multipliers of the Dynamic VIO model are useful information in evaluating the long-term economic effects of spending