Academic journal article Federal Reserve Bank of New York Economic Policy Review

Sources of New York Employment Fluctuations

Academic journal article Federal Reserve Bank of New York Economic Policy Review

Sources of New York Employment Fluctuations

Article excerpt

New York's economy depends heavily on developments elsewhere in the United States, usually contracting when the rest of the nation is in a recession and expanding when the nation is growing rapidly. It is far from a lockstep relationship, however. In some episodes, such as the 1970s, the region fared considerably worse than the United States. In other periods, such as the early 1980s, it performed better than the nation.

This paper investigates employment fluctuations in the New York metropolitan area with the goal of understanding the similarities and differences between the region and the rest of the nation. The investigation has two parts. The first part describes cyclical movements and long-run shifts in regional employment and compares them with employment fluctuations in the nation as a whole. The second part quantifies the relative importance of aggregate, industry-specific, and region-specific factors in explaining the region's fluctuations.

The investigation focuses on two key industries: manufacturing, and the finance, insurance, and real estate (FIRE) sector. Much of the persistent job loss in the region has been in these two industries--first, with the exodus of manufacturing jobs in the 1970s, and more recently, with the restructuring of financial services in the late 1980.(1) One potentially important implication of the evolution of employment shares is a change in the region's response to aggregate factors. As New York's employment base shifts from highly cyclical manufacturing jobs to relatively acyclical financial services, one would expect changes in the relationship between the region and the nation like those documented by McCarthy and Steindel (1996).

To assess the importance of these factors, we use a statistical model that can, by virtue of its factor structure, attribute New York employment fluctuations to readily interpretable aggregate, industry-specific, and regional factors. Our approach also relates the regions response to aggregate and industry shocks to irs industry mix, allowing us to characterize changes in the behavior of regional employment resulting from changes in its employment base.

Our results reveal some significant changes in the region's relationship to the rest of the nation. While New York employment shares a strong cyclical component with U.S. employment, the region has experienced major shifts in its trend growth rate: the largest are associated with negative shocks in the mid-1970s and the late 1980s. Some of these can be traced to specific industries, such as the FIRE-related weakness in the late 1980s. Others, such as the stagnation in the mid-1970s, seem to be due primarily to region-specific factors. At the same time, the region's declining reliance on cyclical industries has made the region's fortunes less closely tied to those of the nation.


The quarterly growth (at quarterly rates) of national and regional employment and their decomposition into trend and cyclical components appear in Chart 1. The regional payroll employment figures used here and elsewhere in the paper are taken from the data set compiled by McCarthy and Steindel (1996). As in their paper, the New York metropolitan area refers to the New York City, Nassau-Suffolk, Duchess County, Jersey City, Bergen-Passaic, Newark, Middlesex-Somerset-Hunterdon, Monmouth-Ocean, Trenton, and New Haven-Bridgeport-Stamford-Danbury-Waterbury metropolitan statistical areas. Further details on the data set construction appear in their paper. U.S. employment data by industry are taken from the payroll employment survey. All data are seasonally adjusted.

[Chart 1 OMITTED]

These decompositions utilize a classification of economic fluctuations dating back to Burns and Mitchell (1946): fluctuations lasting between six and thirty-two quarters are defined as "cyclical," while those lasting more than thirty-two quarters are defined as "trend" components. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.