| Summary: | The article presents comments from the author on the change in the U.S. import demand function between 1950 and 1960. Estimates shown by economists James Joy and J. D. Stolen indicate a shift in the income elasticity of U.S. import demand from about 2.0 in the 1950s to about zero in the 1960s. They gave explanations for this shift is the change in the composition of imports that took place over the two periods; the share of food and crude materials declined and that of finished manufactures increased. However, a shift in import composition toward more consumer luxury items and investment goods and less consumer necessities and material inputs would seem to be consistent with an increase in income elasticity rather than the decrease that they have found. Much of the problem with Joy and Stolen's analysis lies in the use of a time-trend variable in their equations. However, a time trend variable is a less than perfect substitute for the various nonincome, nonprice factors affecting import demand. Special problems arise during periods of strong trend growth in income as noted above in reference to observed multicollinearity.
|
|---|