Forecasting unemployment: a small business-survey model; Small business plans and expectations provide accurate and timely forecasts of national employment

Business Economics, Oct, 2004 by William C. Dunkelberg, Jonathan A. Scott, William J. Dennis, Jr.

The National Federation of Independent Business (NFIB) has surveyed its membership (numbering over 600,000 member firms) since 1973 about economic decisions and expectations. In this paper, responses to questions about plans to create new jobs and hard-to-fill job openings are used to predict the national unemployment rate. Small firms account for the bulk of job creation in the economy, and the NFIB measures explain 80 percent of the variation in the unemployment rate. The NFIB data also indicate a substantial "overshoot" in hiring at the end of the 1990s expansion that helps to explain the period of employment loss and subsequent period of slow employment growth.

**********

Small firms make up a large fraction of the total economy, producing half the private sector GDP and employing an even larger fraction of the private workforce. Recently, the Census reported that firms with fewer than 20 employees created 1.1 million net new jobs from March 2001 to March 2002 (latest data available). Firms with employment between 20 and 500 added 40,000 new jobs and firms with 500 or more employees lost 150,000 jobs. (1) Because of their collective size and importance, indicators of small firm economic behavior should be reliable predictors of changes in macroeconomic indicators such as the unemployment rate. (2)

The same economic forces impact small and large firms. Federal Reserve policy, tax-based fiscal policy, and shifts in consumer spending, for example, affect businesses of every size. (3) The distribution of large and small firms differs by sector, with manufacturing dominated by larger firms and construction along with many services dominated by small firms. Large firms are heavily involved in international trade while small firms are domestically focused. So, from time to time, the economic fortunes of large and small firms may collectively diverge. But, since the same fundamental economic forces affect all firms, these differences should not seriously compromise the usefulness of small business-based indicators for economic analysis. Small business owners with flatter organizations may even be quicker to sense changes in the economy than their large company counterparts.

The Data

This paper examines the predictive power of the employment variables included in the small business surveys conducted by the National Federation of Independent Business (NFIB). NFIB began quarterly economic surveys of its membership in 1973. (4) Since that time, it has mailed a virtually identical three-page questionnaire to a sample of its small-business owner members on a regular basis. From October 1973 through 1985, it mailed the survey to a random sample of the NFIB membership on the first day of every quarter. About ten days later, a second mailing followed. Since January 1986, the same procedure has been followed monthly. Responses are collected for about 25 days, after which duplicate responses are eliminated. The yield is 1,300 to 1,800 responses in the first month of each quarter and 500 to 700 responses in the following two months. A monthly report, Small Business Economic Trends, based on the findings from the survey is available from NFIB (nfib.com/research).

The NFIB membership reasonably reflects the small-business population (Dunkelberg and Scott 1983). Members tend to be somewhat older and are over-represented in the Midwestern, Plains, and Mountain states. Although the industry composition of the NFIB membership has shifted toward services (following the overall U.S. business sector trend), the size distribution of employment of the membership has not changed since the surveys were started in 1973. As long as sample frame and response biases are relatively stable over time, thereby not disturbing the relationship between changes in the indicators and changes in the economic activity measures of interest, the NFIB indicators will be reliable predictors of macroeconomic statistics such as the unemployment rate. The practical value of data from these surveys is that: (1) the data are not subject to revision because they are taken at one point in time, (2) they can be empirically related to changes in macroeconomic indicators that are historically correct (after all revisions are in), and (3) the sample of units included need not be representative of all units in the population nor do all units in the sample need to respond to the survey. (5)

Predicting the Unemployment Rate

Two monthly NFIB survey measures are directly related to the unemployment measure that is computed by the Bureau of Labor Statistics: the net percent of owners who report plans to expand total employment at their firms, HIREPLAN, (6) and the percent of owners who report at least one hard-to-fill job opening, JOBOPEN. (7) The larger the percent of owners planning to expand total employment (a job creation frame of reference, many hires are just replacements) in the months following the survey, the larger the expected employment growth in current and future periods and thus the lower the unemployment rate. (8) The higher the percent of owners reporting a hard-to-fill job opening, the tighter the labor market and the lower the unemployment rate will be. Expectations about growth should also affect hiring activity and, consequently, the unemployment rate. ERSALE, (9) the net percent of owners expecting their own real sales volume to increase in the next three months, should reflect the firm's need to increase employment to support growth and be inversely correlated with the unemployment rate--see Howrey (2001), Bram and Ludvigson (1998), and Cashnell (2003) for examples of empirical analysis using survey data.