Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The Consumer Price Index (CPI) is an important social index number, central to monetary policy, well being measurement, optimal pricing, and tax and contract escalation. Shelter costs have a large weight in the CPI, so their movements receive much attention. The CPI incorporates two shelter indexes: Rent, covering renters, and Owners’ Equivalent Rent (OER), covering owners. Between 1999 and 2006, Rent and OER inflation twice diverged markedly; this occurred again recently. Because these indexes share a common data source—a large sample of market rents—such divergence often prompts questions about CPI methods, particularly the OER utilities adjustment. (This adjustment is necessary to produce an unbiased OER index, because many market rents include utilities, but OER is a rent-of-shelter concept.) The utilities adjustment procedure is no smoking gun. It was not the major cause of these divergences, and it generates no long-run inflation measurement bias. Nonetheless, it increases OER inflation volatility and can drive OER inflation far from its measurement goal in the short run. This article develops a theory of utilities adjustment and outlines a straightforward improvement of Bureau of Labor Statistics procedures that eliminates their undesirable properties. The short-run impact on inflation measurement can be very sizable.