[Accessibility Information]

August 2005, Vol. 128, No.8

Précis

Immigrants of New York
Productivity and business cycles

Download the PDF (28K)
Précis from past issues

Immigrants of New York

Although Ellis Island is today just a national park, New York is still a city very much affected by immigration. In "New York City Immigrants: The 1990s Wave," the June 2005 title in the Federal Reserve Bank of New York series of Current Issues in Economics and Finance, Rae Rosen, Susan Wieler, and Joseph Pereira outline the impact immigration has had on the City’s population and labor force in the 2000 census.

In the decade just preceding the decennial census, 1.2 million foreign immigrants very nearly replaced the 1.3 million residents who left New York for nearby counties or other States. Over the years, that "cycling" of migration resulted in foreign-born persons making up fully 45 percent of New York City’s adult population. Obviously, such a large group has a significant impact on the characteristics of the population and labor force; that impact reflects a remarkable diversity in the characteristics of recent immigrants.

Rosen, Wieler, and Pereira find, for example, that "although the 1990s adult immigrants are on the whole better educated than foreign-born city residents who arrived in earlier decades, they tend to cluster at opposite ends of the education spectrum." New arrivals from Latin America, the Caribbean, or Mexico may often have limited English or be without a high school diploma. At the other end of the scale, recent immigrants from many parts of Asia have a higher proportion of college graduates among them than the proportion of degree holders among native-born residents. Within the Asian immigrants, new arrivals were the exception with relatively low rates of both college graduation and English fluency.

Labor force participation and labor market outcomes also vary widely among recent immigrant groups. Recent immigrants from China have relatively high labor force participation rates, while those from various segments of the former Soviet Union have a participation rate lower than 50 percent.

"Contrary to what one might expect, however," write Rosen, Wieler, and Pereira, "it is not always the least educated or least English fluent groups that have the highest unemployment or public assistance rates. As noted earlier, immigrants from China have relatively low levels of education, English fluency, and income, yet their public assistance rate (1.8 percent) is one of the lowest reported. Their unemployment rate (5.9 percent) is also among the lowest reported, and their labor force participation rate (62.6 percent) one of the highest."

Productivity and business cycles

An often-noted common feature of the business cycle recoveries of 1991 and 2001 has been relatively slow employment growth in the early years of the upturn. In the July/August issue of the Federal Reserve Bank of St. Louis Review, Kathryn Koenders and Richard Rogerson examine this phenomenon from the perspective of organizational dynamics. Their article, "Organizational Dynamics Over the Business Cycle: A View on Jobless Recoveries," starts by noting that the two most recent recoveries share another, less-frequently noted, common feature: both followed the recession that ended an unusually long expansion.

Using that feature as a starting point, Koenders and Rogerson developed a model by which the dynamics of reorganizing production to eliminate unneeded labor could be the link between the speed of net job growth during recovery and the duration of the previous expansion. In their model, labor utilization inefficiencies emerge over time, but the effort to reorganize might be postponed during a long period of expansion. "Because," say Koenders and Rogerson, "reorganization leads to the shedding of unnecessary labor and takes time, this gives rise to an extended period in which the economy sheds labor, thereby delaying the date at which aggregate employment begins to increase during the recovery."

One very useful aspect of Koenders and Rogerson’s research is an extension of the long-expansion-delayed-employment-growth observation beyond the past two business cycles. In their analysis of the eight post-1950 recessions, they found not two but three that followed exceptionally long expansions: the recovery from the 1969-70 recession followed an expansion that ranked second in duration between the two most recent. Using the perspective of their model, Koenders and Rogerson found that "the behavior of employment in the 1970 recovery is in fact very similar to the behavior of employment in the recoveries of 1991 and 2001 and is qualitatively different from the behavior of employment in the other post-World War II recoveries."

Specifically, Koenders and Rogerson examined the change in employment relative to trend after the 1970 trough and found that the "cyclical" component of employment continued to fall for a year after the business cycle trough, as did cyclical employment in 1991 and 2001. Although trend-adjusted employment in the earlier episode started to recover after only a year’s delay, Koenders and Rogerson suggest that the period of reorganization should be dated from four quarters before the 1969 peak, rather than the typical two quarters. Koenders and Rogerson suggest that, with these adjustments, one could consider that all three recoveries from recessions ending long expansions were characterized by delayed recoveries in employment.

Top