This paper presents preliminary comparisons of asset ownership estimates derived from wave 1 of the 2004 panel of the Survey of Income and Program Participation (SIPP) with estimates derived from the 2001 SIPP panel and from other sources. Of primary interest are the estimated ownership rates for “rare” asset types. In contrast to the procedures used in 2001, in the new questionnaire in use for 2004 some respondents, by design, do not receive the full battery of specific, individual questions about each of these asset types, but instead respond only to a single, global, screening question asking about “any other assets...” The primary goal of this instrument modification is to reduce burden – to increase efficiency generally, and in particular to reduce the number of unnecessary questions asked. Its primary risk, of course, is that removing specific cues for some types of assets might cause additional underreporting of asset income sources, beyond the baseline level of underreporting that is known to occur under standard interviewing practices. The paper also reviews evidence concerning the possible impact of other instrument changes on ownership estimates for retirement accounts and common income-producing assets, the other major categories of assets covered in SIPP.
