The Point: Attanasio and Pistaferri introduce a new imputation method for analyzing consumption inequality in the US. Most studies of consumption inequality in the U.S. rely on data from the consumer expenditure survey (CEX), which provides comprehensive micro-level data starting in the 1980s, but the CEX data is not without problems. The imputation in this paper is based on data from the Panel Study of Income Dynamics (PSID), which stretches back to the 1960s. The imputations provide an alternative reference point in the discussion around consumption inequality.
Results: Some earlier papers have suggested that consumption inequality grew minimally between 1980-2000 (see Krueger and Perri 2006), but the enhanced PSID data supports claims that the rise in consumption inequality in the period stretching from the early eighties until the Financial Crisis was more pronounced than previously expected.
The Data: The authors use PSID data from 1968-2011. Their imputation relies on the fact that beginning in 1999, PSID began collecting more comprehensive consumption data. Prior to 1999, PSID had consistent and reliable data on food expenditure, the value of food stamps, and a few other items, but PSID data now captures 70% of non-durable spending from national accounts. The data includes spending on items such as health, utilities, car maintenance, education and child care. Attanasio and Pistaferri use the more comprehensive data of recent years to infer total expenditures on consumption in the years prior to 1999. The imputation gives them a series of data that extends more than a decade further back in time than data from the CEX.
The Quote (refer to Panel D in the image below): "How does the growth in consumption inequality compare to the growth in income inequality?... During the 1970s the two series are synchronized and relatively stable, a well-known fact. Between 1978 and 1993, both series grow, but income inequality rises faster than consumption inequality (15 versus 6 points). However, income inequality after 1993 slows down, while consumption inequality keeps rising. In the last 10 years of the sample period, however, income inequality again rises faster than consumption inequality, and during the Great Recession income inequality keeps rising while the rise in consumption inequality comes to a complete halt." (125-126)
Limited Access: There is limited access to the paper through the AEA web. A copy of the paper can be openly accessed here.
Citation: Attanasio, Orazio, and Luigi Pistaferri. 2014. "Consumption Inequality over the Last Half Century: Some Evidence Using the New PSID Consumption Measure." American Economic Review, 104(5): 122-26.