An Uneven Recovery, 1993-97
By Daniel W. Kennedy, Ph.D., Senior Economist

In May 1999, the U.S. Bureau of Economic Analysis released the 1997 Local Area Personal Income for Counties, Metropolitan Areas, and Nonmetropolitan Areas. This article draws on that release for its analysis. For Connecticut, three metropolitan areas are designated, and defined by the BEA as New England County Metropolitan Areas. The largest is the New Haven-Bridgeport-Stamford-Waterbury-Danbury New England County Metropolitan Areas, which is also part of New York City's Consolidated Metropolitan Statistical Area. However, only the Connecticut portion will be discussed here. This New England County Metropolitan Areas is composed of Fairfield and New Haven Counties. The second New England County Metropolitan Areas analyzed is the Hartford New England County Metropolitan Areas, which is made up of Hartford, Tolland, and Middlesex Counties. The third New England County Metropolitan Areas in Connecticut is New London-Norwich, composed entirely of New London County.

Before continuing, there are a couple of definitions that must be discussed. The first is Earnings. Bureau of Economic Analysis defines earnings as wages and salary disbursements, other labor income, and proprietor's income. Thus, earnings are a broader measure of monetary compensation than wages and salaries. Employment here is also that defined by Bureau of Economic Analysis. It includes not only employment covered under the unemployment insurance program, but also the self-employed, those employed by religious organizations, and other employment not picked up by the Covered Employment and Wages (ES-202) program. There are also some methodological differences in Bureau of Economic Analysis' employment series compared to that of U.S. Bureau of Labor Statistics. Thus, the Bureau of Economic Analysis series will be much larger than that of BLS's ES-202 series.

Earnings Growth

The period of analysis covers the first four years of Connecticut's recovery from the "Great Recession," 1993 to 1997. Graph 1 on the front page shows the percent change in earnings and employment by major Standard Industrial Classification division for that period.

Connecticut's services division had the fastest growth in earnings over the 1993-97 period. In addition to services, earnings grew faster than average in agriculture, transportation and public utilities (TPU), wholesale trade, and finance, insurance, and real estate (FIRE). The slowest growth in earnings was in the mining and government divisions. Employment growth outpaced the overall industry average in agriculture, construction, TPU, wholesale and retail trade, and services. Mining, manufacturing, and FIRE lost jobs over the period, and government employment growth over the period was flat. Save agriculture, employment growth lagged behind the earnings growth for all of Connecticut's SIC divisions over the 1993-97 period.

The consequences of the growth rates of earnings and employment over the first four years of recovery are presented in Graph 2 at the bottom of this page. It shows the growth in the earnings-per-job, by SIC division, over the 1993-97 period. The growth in earnings-per-job for the FIRE division, at 40.9%, was double or more the growth in every other division, except mining. Mining, manufacturing, and wholesale trade, all had growth in the earnings-per-job that surpassed the average growth in all nonfarm earnings-per-job.

Table 1 above provides a summary of the growth in employment and earnings for Connecticut over the 1993-97 period, compared to the performance in the U.S. over the same period (keeping in mind that the U.S. came out of its recession in 1991).

Table 1 is set up as a contingency table. The column labeled "Earnings Growth" is divided between the top half, which indicates that earnings growth in Connecticut's SIC divisions exceeded its counterpart for the U.S., and the bottom half, indicating those divisions where growth lagged behind that of the U.S. Across the top is the employment growth comparison. There are two categories. "Greater than U.S." and "Less than U.S.," indicating that Connecticut employment in a given division grew faster or slower than its counterpart at the U.S. level. This type of table is also used to compare metropolitan area performance.

Table 1 shows that Connecticut's agriculture division had faster earnings growth and faster employment growth over the 1993-97 period than did the U.S. agriculture division. Though the TPU, services, and wholesale trade divisions had faster earnings growth over the period than their U.S. counterparts, they lagged behind in employment growth. Connecticut's construction, retail trade, and government divisions lagged behind their U.S. counterparts in both earnings growth and employment growth. Mining, FIRE, and manufacturing in Connecticut's also lagged behind their U.S. counterparts these areas, and experienced employment declines. In the case of mining, both the Connecticut and U.S. divisions experienced job losses; however, Connecticut's mining division suffered a greater decline.

Performance of the NECMAs

Following the form of Table 1, Tables 2, 3, and 4 compare the performance of earnings growth, employment growth, and earnings-per-job growth, respectively, in Connecticut's NECMAs with that of Connecticut and the U.S.

Table 2 on the next page shows that the New Haven - Bridgeport - Stamford - Waterbury - Danbury and the New London - Norwich NECMAs had faster growth in nonfarm earnings than either Connecticut or the U.S. Further, these two also had faster nonfarm earnings growth than either the Hartford NECMA, or the nonmetropolitan portion of the State. The Hartford NECMA and the nonmetropolitan portion lagged behind both Connecticut and the U.S. in nonfarm earnings growth over the 1993-97 period.

Table 3 below shows that employment growth exceeded that for Connecticut, but lagged behind that for the U.S., in the New Haven-Bridgeport-Stamford-Waterbury-Danbury and New London-Norwich NECMAs, and the balance of the State (non-metropolitan portion). The Hartford NECMA lagged behind both the U.S. and Connecticut in job growth.

The comparison of the growth in nonfarm earnings-per-job is depicted in Table 4 above. The New Haven-Bridgeport-Stamford-Waterbury-Danbury NECMA was the only sub-state region to outperform both the U.S. and Connecticut in the growth of nonfarm earnings-per-job over the 1993-97 period. Though the New London-Norwich NECMA outperformed the U.S in earnings-per-job growth, it lagged behind Connecticut's growth. Again, the Hartford NECMA lagged behind both the U.S. and Connecticut. The non-metropolitan portion of the State also lagged behind the U.S. and Connecticut nonfarm earnings-per-job growth.

Clearly, the New Haven-Bridgeport-Stamford-Waterbury-Danbury experienced the strongest performance of Connecticut's sub-state regions in terms of the growth in earnings and earnings-per-job. Its performance was less stellar in the job-creation department, though certainly not a laggard. This performance probably has something to do with this area's connection to the greater New York City economy. The Hartford NECMA experienced the most difficulty over the 1993-97 recovery period, at least based on its comparative growth in nonfarm earnings, employment, and earnings-per-job. The New London-Norwich NECMA did well in earnings growth, and had modest performance in employment and earnings-per-job growth. The non-metropolitan portion of the State, though it performed slightly better than the Hartford NECMA, was, like the Hartford NECMA, struggling to jump-start nonfarm earnings and earnings-per-job during the first four years of the State's economic recovery. The State itself lagged behind the U.S. in nonfarm earnings and employment growth over 1993-97 (again, given that the U.S. recession was shorter and less severe, and that its recovery began in 1991).

Earnings-Per-Job Growth by Industry for NECMAs

Why did the New Haven-Bridgeport-Stamford-Waterbury-Danbury NECMA have the State's best performance, and why did the Hartford NECMA lag behind the State? The following brief summary compares the two NECMA's industry mix and the role it played in their different experiences over the 1993-97 recovery period.

The New Haven-Bridgeport-Stamford-Waterbury-Danbury NECMA accounted for one-half to four-fifths of the shares of the State's earnings-per-job growth for the three fastest growing divisions: services, wholesale trade, and FIRE. The NECMA's growth in services earnings-per-job matched that of the State, and ranked second behind the New London-Norwich NECMA. FIRE earnings-per-job growth was faster than that for the State and the other two NECMA's. Though the New Haven-Bridgeport-Stamford-Waterbury-Danbury NECMA lost jobs in FIRE, its decline was only half that of the State's.

The Hartford NECMA experienced a decline in earnings-per-job for FIRE. Further, the Hartford NECMA's job losses in FIRE were much steeper than those for the State. The NECMA also lagged behind the State in earnings-per-job growth in services. Though the Hartford NECMA lost fewer manufacturing jobs than the other two NECMAs and had a decline that matched that of the State, earnings-per-job grew very slowly, and lagged behind that for the State. Similarly, Hartford's growth in construction earnings-per-job lagged behind the State's growth.

Return to Top

Industry Clusters
Cluster Activation

In the latest "Industry Clusters Progress Report," the DECD's Industry Cluster and International Division summarizes progress achieved with the various clusters organized to date. The report's ten chapters start with the issue of "Cluster Activation."

Cluster activation is a critical first step toward realizing the potential of a group of industries. Clusters are activated when companies formally organized to collaborate, cooperate, and become competitive. Corporate leaders drive the cluster's activities  and the public sector's role is support and facilitation. In Connecticut, BioScience, Aerospace Components Manufacturers, and Software/Information Technology clusters are up and running.

The DECD has provided seed funds totaling $425,000 to these three clusters and will leverage over $1 million in in-kind matches and private investment. A focused study of Connecticut Marine commerce is underway. Agriculture, Seafood, and Plastics are in the early stages, but progressing. Photonics is in the exploratory phase.

The ongoing challenges are to maintain each cluster's forward momentum while refining the program support elements appropriate to each one's developmental stage. This includes identifying outcomes to be achieved, meeting the State's goals by identifying and providing special attention required by key clusters, and simultaneously supporting Connecticut's major clusters such as finance/insurance/ reinsurance, medical devices, and health services.

Return to Top

HOUSING UPDATE
1999 Permit Activity, Second Highest in the Decade

Commissioner James F. Abromaitis of the Connecticut Department of Economic and Community Development announced that Connecticut communities authorized 648 new housing units in December 1999, a 42.6 percent decrease compared to December of 1998 when 1,129 units were authorized.

The Department further indicated that the 648 units permitted in December 1999 represent a decrease of 36.7 percent from the 1,023 units permitted in November 1999. The year-to-date permits are down 6.5 percent, from 11,541 through December 1998, to 10,794 through December 1999.

"Despite the December totals, 1999 was a very strong year for housing permit activity in Connecticut," said Commissioner Abromaitis. "As a matter of fact, the 10,794 permits authorized was the second highest total in the decade."

Hartford County documented the largest number of new, authorized units in December with 162. Fairfield County followed with 121 units and New Haven County had 117 units. Manchester led all Connecticut communities with 41 units, followed by Middletown with 22 and Tolland with 13.

Return to Top

Pop the Cork and Pass the Bubbly as the Party Rolls into 2000

The Connecticut coincident and leading employment indexes both increased with the release of (preliminary) November 1999 data. The rise in the coincident index reversed two consecutive monthly declines while the increase in the leading index was the second monthly rise since its large drop in September 1999.

The coincident index, a gauge of current employment activity, remains within hailing distance of its recent peak in August 1999. Focusing on its components, both total and nonfarm employment rose while both the total and insured unemployment rates fell on a month-over-month basis.

The leading index, a barometer of future employment activity, continues to dance along a plateau established in late 1996. See the accompanying chart. This month's release continued the leading index's rebound from its substantial month-to-month decline in September, primarily reflecting the large increase in total housing permits.

This month's release also represents a landmark among professional bean counters. Nonfarm employment finally, fully recovered all jobs lost during the Great Recession. That is, nonfarm employment rose to 1,678,600 jobs in November 1999, surpassing the previous peak 1,678,300 jobs recorded in February 1989. Now, this fact belies dramatic shifts in the job mix - manufacturing; wholesale and retail trade; and finance, insurance, and real estate jobs all dropped from their February 1989 counts, but the job increases in services (e.g., business and health) and government easily picked up the slack. (Remember that the increase in government jobs primarily reflects Native American tribal activity.)

As we go to press, a major unknown was what the Federal Reserve will do with interest rates at its February meeting. The decision not to raise rates in their last meeting before the new year for a smooth Y2K transition should not continue through the February meeting, as many analysts expect an interest rate increase. The secret will be divulged by the time this column appears in print.

In summary, the coincident employment index rose from 98.1 in November 1998 to 100.9 in November 1999. All four components of the index point in a positive direction on a year-overyear basis with higher nonfarm employment, higher total employment, a lower total unemployment rate, and a lower insured unemployment rate.

The leading employment index rose from 89.4 in November 1998 to 89.9 in November 1999. Three index components sent positive signals on a year-over-year basis with a lower short-duration (less than 15 weeks) unemployment rate, lower initial claims for unemployment insurance, and higher total housing permits. Two components sent negative signals on a your-over-year basis with a lower average workweek of manufacturing production workers and lower Hartford help wanted advertising.

SOURCE: Connecticut Center for Economic Analysis, University of Connecticut. Developed by Pami Dua [Economic Cycle Research Institute; NY, NY] and Stephen M. Miller [(860) 486-3853, Storrs Campus]. Stan McMillen, Kathryn Parr, and Hulya Varol [(860) 486-3022, Storrs Campus] provided research support.

Return to Top