Abstract
This article goes beyond the principle of relative constancy (PRC) by testing 2 new models of consumer mass media spending using 1953-1991 Belgian data. Unlike traditional PRC models, which have focused exclusively on the long-term relationship between income and mass media spending, these 2 models contain additional regressors (price, population, unemployment, and interest rate) in current (Model 1) and lagged (Model 2) form. Time-series regression analyses were performed to determine which variables significantly predict changes in consumer mass media expenditures. Model 1 regressions revealed that price and population were better predictors of mass media expenditures than income, stressing the importance of developing models of consumer mass media spending that go beyond a simple mass media expenditures-income relation. Model 2 regressions showed that lagged variables played an important role in explaining changes in mass media expenditures, indicating the need for incorporating lags in future mass media spending work.
