John S. Greenlees

Is the Consumer Expenditure Survey Representative by Income

Aggregate under-reporting of household spending in the Consumer Expenditure Survey (CE) can result from two fundamental types of measurement errors: higher-income households (who presumably spend more than average) are under-represented in the CE estimation sample, or there is systematic under-reporting of spending by at least some CE survey respondents. Using a new data set linking CE units to zip-code level average Adjusted Gross Income (AGI), we show that the very highest-income households are less likely to respond to the survey when they are sampled, but unit non-response rates are not associated with income over most of the income distribution. Although increasing representation at the high end of the income distribution could in principle significantly raise aggregate CE spending, the low reported average propensity to spend for higher-income respondent households could account for at least as much of the aggregate shortfall in total spending.

NEW EVIDENCE ON OUTLET SUBSTITUTION EFFECTS IN CONSUMER PRICE INDEX DATA

In this paper we provide new evidence on the impact on the U.S. CPI of the appearance and growth of new types of product outlets. Our CPI food microdata permit a more detailed categorization of outlet types than in previous studies, and we can adjust for numerous differences in item characteristics. We also examine the effects of changes in outlet mix not only across outlet categories but also within those categories. In our sample, we find that the upward impact on price from increased item quality has offset most, but not all, of the downward impact of lower-priced outlets.

The Impact of Direct Taxes on the Cost of Living

In this paper, the authors define a cost-of-living index including direct taxes. They show its relationship to the traditional index and demonstrate how nonconsumption costs are properly treated. They t hen define a fixed-weight approximation, a tax and price index (TPI). Using federal, state, local, and social security tax rates for 1967- 85, the authors construct annual TPI series based on household data. They find that inclusion of direct taxes has sizable impacts on the e stimated rate of inflation. Partitioning their household sample, they find that recognition of taxes significantly alters inflation rate d ifferentials estimated using consumption prices alone. Copyright 1987 by University of Chicago Press.

Improving the Preliminary Values of the Chained CPI-U

This paper employs a constant-elasticity of substitution (CES) index formula to improve the accuracy of the preliminary values of the Chained Consumer Price Index for All Urban Consumers (C-CPI-U). Using the CES behavioural model, I present estimates of the overall extent of consumer response to relative price changes exhibited in Consumer Expenditure Survey data for 1999-2008. The associated parameter estimates are then used to develop CES forecasts of the final C-CPI-U index values. Simulations demonstrate that use of the CES approach over the last several years would have resulted in smaller index revisions between the preliminary and final C-CPI-U releases. Looking to the future, CES-based preliminary estimates could increase the usefulness of the C-CPI-U to government programs and other users.

Reconsideration of Weighting and Updating Procedures in the US CPI

Summary In 2002, the US Bureau of Labor Statistics (BLS) introduced a supplemental C-CPI-U employing a superlative formula to provide a closer approximation to a cost-of-living index (COLI). This paper focuses on whether the BLS can improve upon the headline CPI-U’s current biennial weight update process, thereby reducing the CPI-U’s growth rate and bringing the index closer to the C-CPI-U. We begin by estimating superlative price indexes for 1999 through 2007 along with indexes based on the constant-elasticity-of-substitution demand model. Our analyses confirm that the consumer expenditure data underlying the CPI imply substantial consumer substitution, implying that the CPI-U’s Lowe index formula yields higher inflation estimates than would a true COLI. Simulating feasible weight update processes, we find that a Lowe index with two-year weight reference periods but annual updating rises by about 0.03 percentage points less per year than the CPI-U. Another 0.01 percentage point on average is subtracted by imposing annual revision with one-year base periods. Thus, indexes with more timely weights may offer improved representation of current price change, as well as closer approximations to a COLI.

Grouping Tests for Misspecification: An Application to Housing Demand

We consider several grouping tests for regression misspecification, with reference to housing-demand function estimation. We compare three existing test procedures, demonstrate modifications necessary in most applications, and propose a fourth test to distinguish between two categories of potential specification error. The test procedures are evaluated in artificial simulations of alternative errors. Finally, we apply the tests to FHA home purchase data. We reject the hypothesis that household and grouped regressions differ only by sampling error or random mismeasurement of household income or price. Our results have implications for choices among test procedures and interpretations of previous housing-demand analysis.

Estimating Inter-city Differences in the Price of Housing Services: Further Evidence 1

This paper reconsiders an approach to inter-area house price index construction applied by Harvey Rosen in a recent article in Urban Studies. Rosens cost function-based approach is shown to rely upon certain implicit assumptions regarding consumer preferences and housing production technology. Using FHA home purchase data, we estimate a more general production model and generate alternative inter-area indexes. Based on our results, Rosens production technology assumptions are rejected. Further, the production-theoretic indexes are shown to differ only negligibly from a fixed-weight index using FHA data. Thus the difference which Rosen finds between the production-theoretic FHA and published BLS indexes does not result from Rosens recognition of substitution in production, but because of very different data sources and scopes of coverage.