Comment 29A: Addendum to Comment 29: How to bridge the gap between household income survey and personal income per capita. (May 5, 2006)

In Comment 29 the divergence between the real household income survey conducted by U. S. Census Bureau ( and the real personal income per capita from The Bureau of Economic Analysis ( since the year of 2000 was noted, but the reason of this divergence was not pursued vigorously. The real personal income per capita has risen strongly in recent years, and has surpassed the previous peak of 2000 by a substantial amount, pointing toward a robust economic expansion. However, household income survey shows a steadily dropping income for middle class families until 2004, and does not show any sign of income surpassing the 2000 peak anytime soon; the household income survey paints the picture of a stagnated economy. A detailed study reveals that this divergence is due to the skyrocketing medical costs. In the statistics of personal income, the cost of medical insurance paid by the employers for their employees is included as a component of personal income. The costs of Medicare and Medicaid incurred by governments are also included as a component of the personal income. On the other hand in the household income survey, only money received by households is counted as income, and thus the above-mentioned medical expenditures are not counted. Due to the much faster expansion of medical costs than wages and salaries in recent years, the personal income per capita is boosted up rapidly by those medical costs, whereas the household income that does not include those medical expanses incurred by businesses and governments is lagging behind. To make both kinds of statistics consistent medical expenses incurred by businesses and governments are stripped from the personal income. The resulting adjusted real personal income per capita becomes reasonably consistent with the household income survey.

The costs of health insurances of businesses for their employees are included in a component called "supplement to wage and salary" in the tabulation of personal income. For example, from 2003 to 2004, this component rose by 7.3%, whereas the component of "wage and salary" only rose by 5.5%. The component of supplement to wage and salary also includes the corporate contributions to pensions, but pension costs certainly will not increase substantially faster than wage and salary itself. This means that the health insurance part has risen faster than 7.3% of the component as a whole. Since the component of supplement to wage and salary is not counted as income in the household survey, we remove this component from personal income to perform our adjustment. The costs of Medicare and Medicaid are grouped together with the social security pension payment and others into a component called "personal current transfer receipt". Then contributions for government social security insurance must be subtracted from the component to get the true cost of government for those social services. This social service component rose by 6.7% from 2003 to 2004, compared to 5.5% rise of wage and salary. Since the social security pension per retiree is indexed to the Consumer Price Index (CPI), the real pension payment per retiree is always constant, that is, no increase. This means that health care costs of government have been rising much faster than the overall increase of 6.7% of the component. Since health care costs are not separately listed in the component of personal current transfer receipt nor the tax paid for Medicare, we are forced to remove the whole component of personal current transfer receipt and add back the whole component of contribution to social security program. It should be noted that the pension part of social security is included in the household income survey. The removal of the slow growing part of the transfer receipt, that is, the pension portion along with the super fast growing component of health care cost will still make the adjusted real personal income per capita rise somewhat faster than the household income survey, but the difference in the rate of rise is probably quite small. The adjusted real personal income per capita and the adjusted real disposable income per capita are plotted as black and red curves respectively in the graph at the right. In the graph one-half of the real income of 60%-household (middle class household) from the household income survey is plotted as the green curve. We can see that the behaviors of the green curve and the black curve are more or less synchronized, a substantial improvement over the unadjusted real personal income per capita as shown in Comment 29. The adjusted real disposable personal income, the red curve, displays much stronger growth than both the household income of middle class families and the adjusted real personal income per capita since 2000. The difference is due to the tax cuts of 2001 and 2003. Since those tax cuts benefit very rich households most, the effect on the income of middle class households is minimal. The slight divergence between the green curve and the black curve in 2004 probably can also be attributed to this effect of wealth gap. The household income survey of 2005 is keenly awaited to see whether there is any rebound from 2004 income.

Though some economists complaint about the exclusion of incomes like the medical care costs discussed here from the household income survey, we believe that the household income survey and the adjusted real personal income per capita as used here represent a more realistic picture of the actual economic conditions than the unadjusted personal income data that is artificially boosted by the skyrocketing medical costs. Our belief can be summarized in the following exaggerated example:

A surveyor visits a middle class household. The head of the household tells the surveyor gloomily that his household's income has been falling 20% a year for the past five years, and pretty soon he will have the difficulty of providing enough foods for his children. The surveyor says, "Your household's take home income was $50,000 last year. Our government's cost for the national health program is skyrocketing. Last year it came out to be $450,000 per household. Some economists insist that the medical cost incurred by our government is actually a part of personal income. Thus your household's last year income was actually $500,000, according to economists. This year the cost of national health system has had another quantum jump. It now averages to $960,000 per household. Since your household's take home income is $40,000 this year, this year's actual income of your household becomes 1 million dollars, according to economists. Not long ago our government had a huge party to celebrate the epoch event that an average household like yours is boosted into the millionaire's club." Hearing that news, how will the head of the household react? If he is an economist, he will probably open a bottle of champagne to celebrate his newly acquired "millionaire" status. If he is not an economist, he will probably be depressed more from the thought that pretty soon his whole family need to move into a hospital to ride on the run away medical spending, otherwise his family will surely be starved to death gradually.