What is consumption puzzle
Keynes’ assertion that the APC falls as income rises led some Keynesians to formulate the secular stagnation thesis around 1940. According to these economists, as incomes grew in the economy, households would save more and consume less.
As a result, aggregate demand would fall short of output. If the government spending was not increased at a faster rate than income, the economy would lapse into stagnation. But after World War II, the American economy experienced inflation rather than stagnation even when the government expenditures were reduced below 1941 level in constant dollars.
The Keynesian consumption function had been proved wrong. This was due to the conversion of government bonds into liquid assets after the War by the households in order to meet their pent up demand for consumer goods.
In 1946, Kuznets studied the consumption and income data for the United States during the period 1869-1938 and estimated the consumption function for this period as 0.9. Further, he arrived at two conclusions: one, over the long-run, on the average, the APC did not show any downward trend so that the MPC equaled the APC as income increased along a long run trend.
This means that the consumption function is a straight line through the origin, as shown by the C, line in Fig. 10, and two, the years in which the APC was below the long-run average were boom periods and the years in which the APC was above the long-run average were of slump periods. This implies that in the short-run as income changes over the business cycle, the MPC is less than the APC, as shown by the C, curve.
These findings were later verified by Goldsmith in 1955 who found the long-run consumption function to be stable at 0.87. Thus these two studies revealed that for the short-run time series, the consumption function is non-proportional because APC > MPC and for the long-run time series, the consumption function is proportional, APC = MPC.
The failure of the secular stagnation hypothesis and the findings of Kuznets and Goldsmith were a puzzle to the economists which is knows as the consumption puzzle. Figure 10 illustrates this puzzle where there are two consumption functions. C is the Keynesian consumption function which is non-proportional (APC> MPC) and based on the short-run time series data.
C, is the long-run proportional consumption function (APC = MPC) based on long-run time series data. Over the years, economists have been engaged in solving this puzzle by reconciling the two consumption functions.
