The study of the movement of persons into and out of the labor force and between employment and unemployment, or "labor force dynamics,” has had a resurgence in recent years. Much of this renewed interest stems from a joint Bureau of the Census and Bureau of Labor Statistics conference in 1984 relating to the gross labor force flows derivable from the Current Population Survey (CPS).

Most research in this area in the 1980's has focused on ways of overcoming statistical problems associated with the CPS flows. Few researchers realize, however, that gross labor force flows can a1so be derived from a re1atively new household survey called the Survey of Income and Program Participation (SIPP). It is possible that this data source may add something to our knowledge of labor force dynamics.

As a beginning, this paper presents a comparison of the gross labor force flows as recorded in the SIPP and CPS during 1984. In that year, employment was growing rapidly and unemployment was dropping sharply in response to vigorous economic growth. In other words, we are examining flows occurring in a very strong phase of the business cycle. Our comparison shows that 1) the SIPP flows are smaller in general than the CPS flows and, 2) when the SIPP flows are balanced out, net changes in them are more consistent with the net changes in SIPP stocks than is the case with the CPS flows and stocks.

We reserve judgement, however, with respect to the quality of the SIPP labor force flows at this point. The survey designs in SIPP and CPS are very different. As is well known, because of CPS's survey design, month-to-month flows are frequently found to be inconsistent with their stocks. As will be shown in the paper, because of SIPP's survey design, month-to-month flows are bound to be more consistent with its stocks, by definition. But this particular survey design may a1so create other statistical problems that are embodied in the flows. This aspect of the comparison, therefore, awaits further investigation.

The paper begins with a review of the CPS and SIPP survey designs, with emphasis placed on sample rotation differences and reference period differences. The second section examines the monthly 1abor force status estimates, or the stocks, from both surveys for each month between December 1983 and December 1984. The following section presents the labor force flows from both surveys, while the last section discusses some of the statistical problems that might affect the quality of the SIPP gross change data.