RDP 9512: Consumption, Investment and International Linkages Appendix C: Estimation with National Account Consumption Measure

The table below gives results for the estimation of the consumption model when the constructed true consumption measure is replaced by the national accounts measure of consumption. Only the preferred model is considered, using all foreign output measures.

Table 7: Results Using the National Accounts Consumption Measure
  1973:Q2–1983:Q1 1983:Q1–1994:Q4 1973:Q2–1994:Q4
Point 0.38** 0.28** 0.27**
Estimate of Inline Equation (0.13) (0.09) (0.09)
Significance of Inline Equation 0.12 0.73 0.22
Point 0.35** 0.31** 0.28**
Estimate of Inline Equation (0.12) (0.09) (0.08)
Significance of Inline Equation 0.22 0.07 0.07
Point 0.29** 0.36** 0.26**
Estimate of Inline Equation (0.12) (0.10) (0.09)
Significance of Inline Equation 0.76 0.11 0.32

Note: Values marked ** (*) are significant at the 5% (10%) respectively.

Point estimates for domestic disposable income growth are similar in magnitude to those detailed in Table 2 though a little larger for the 1983:Q1–1994:Q4 period. Again, significance levels for the foreign growth term in the US and Japanese cases display a similar pattern to results for the true consumption measure. Interestingly, for the OECD case, the use of the national accounts consumption measure provides a significant result for the latter and full sample periods. This suggests that the actual consumption expenditure on durables rather than the flow of services generated from such expenditures is more highly correlated with OECD economic developments.