Spendthrift in America? On two decades of decline in the U.S. saving rate
in Ben Bernanke and Julio Rotemberg eds, NBER Macroeconomics Annual 1999, MIT Press, 317-70.
During the past two decades, the personal saving rate in the United States has fallen from eight percent to below zero. This paper demonstrates that this change represents a major shift in the allocation of newly produced goods. The share of GDP that households consume rose by 6 percentage points since 1980. This increase occurred concurrently with a reduction in the growth rate of real consumption spending per person, high real rates of return, and an increasing ratio of aggregate wealth to income. Despite this last fact, wealth changes can explain little of the boom in consumption spending. The largest increases in national wealth post-date the consumption boom and households with different wealth levels have similar increases in consumption. The paper also finds that the changing age distribution of the U.S. population does not explain the consumption boom. While it may be that new wealthier cohorts are driving this boom, the preponderance of evidence suggest rather that the rising consumption to income ratio is due to a common time effect. The main findings of the paper are consistent with either an increase in the discount rate or with a general belief in better economic times in the future. Alternatively, the low rates of saving could be due to a combination of factors such as the increase in intergenerational transfers from the Social Security system raising the consumption of the elderly and an increase in access to credit and expanded financial instruments raising the consumption of the young.
NBER WP 7238 (link requires access to NBER WP site)
Graph based on data revised Oct, 1999.
Since the research was completed, the Bureau of Economic Analysis has released a major revision of the national accounts designed in part to fix some measurement issues in the construction of saving rates and output. Most notably, this revision reclassifies expenditures on software as investment, treats government pension plans in the same manner as private pension plans, and removes some asset transfers from disposable income. These changes increase personal saving throughout the sample period, but the decline in the personal saving rate studied here remains. The two-decade long increase in the consumption share of output also remains, although the magnitude is reduced by one percent of GDP (about 17 percent of the rise). The graph at the top of the page and the one below demonstrate these claims.
Graph based on data revised Oct, 1999. Click here to see pdf version of the paper's Figure 1 but based on the new data.