Connecticut Industry Employment Outlook to Fourth Quarter 2003
By Daniel W. Kennedy, Ph.D., Senior Economist, and Noreen P. Passardi, M.A., Economist, DOL

Introduction of Two-Year Employment Outlook

This is the first of what will be semi-annual outlooks on Connecticut industry employment covering a two-year forecast horizon (eight quarters). Each outlook published in September or October will present the forecasts from fourth quarter to fourth quarter. The current outlook covers the forecast period fourth quarter 2001 (2001:Q4), to 2003:Q4. The outlooks published in May or June will cover the midyear forecast periods. For instance the next outlook will present forecasts for the 2002:Q2 - 2004:Q2 period. In addition, the next outlook will present the forecasts in the North American Industry Classification System (NAICS). This will be the only outlook presented under the 1987 Standard Industrial Classification (SIC) System, as the SIC-to-NAICS conversion is currently underway.

The next section briefly reviews recent U.S. and Connecticut labor market conditions. The following section presents the industry employment outlook for Connecticut over the forecast period. The detailed outlook for the major SIC divisions is then discussed. Finally, the risks to the forecast are briefly outlined. Curr Current ent Conditions I: U.S. Labor Markets So far, in U.S. labor markets, there has been a repeat of the jobless recovery of the early nineties. However, the August employment data, released by the U.S. Bureau of Labor Statistics (BLS), did suggest brighter prospects for labor markets. BLS revised payroll employment gains for the two previous months by 29,000. Since its recent low in April, employment has edged up by 162,000. Surprisingly, the unemployment rate fell 0.2 percentage points to 5.7 percent. Nevertheless, layoffs may be on the rise again. In the week ending August 31, the advance figure for seasonally adjusted initial unemployment claims was 403,000, a decrease of 8,000 from the previous week's revised figure of 411,000. However, the four-week moving average returned to the threshold 400,000 level. This is an increase of 5,250 from the previous week's revised average of 394,750. Further, initial claims jumped by 19,000 to 426,000 in the first week of September, bringing the four-week moving average up to 409,500. This, coupled with other data, indicates that the current U.S. economic recovery is rapidly decelerating.

Current Conditions II: Connecticut Labor Markets

Connecticut labor markets, like their national counterparts, are also not experiencing robust job growth. The State's unemployment rate increased to 4.0 percent in August, up 0.2 percentage points from July. The seasonally adjusted labor force and employment-topopulation ratio remained unchanged from July to August. Nevertheless, like its national counterpart, Connecticut employment appeared to fare better in August. Jobs in August increased by 1,200 from July to 1,675,400, the first gain since May. However, they were down 4,600 from August 2001. Since January 1,400 jobs were lost, which is certainly an improvement over the loss of 17,600 jobs between January and August 2001.

The State economy, like the national economy, appears to be experiencing a slowdown in what, overall, is a jobless recovery. Like the national labor markets, Connecticut's labor markets, as well as other sectors, seem to be sending mixed signals about the current state of the economy. However, it does appear that Connecticut is not experiencing the dramatic increase in the intensity of layoffs that has suddenly gripped the U.S. economy in August and early September. Nevertheless, Connecticut does seem to be sharing in the slowdown from the recovery that seemed to be underway in the first half of 2002.

Recent Historical Employment Trends

The following discussion is based on the unadjusted employment series. Seasonally adjusted series are not available at the industry detail level that is used to forecast short-term employment. This should be kept in mind when interpreting the forecasts. The short-term forecasts have a twoyear forecast horizon. Therefore, the following analysis will look at the historical and forecast periods in terms of two-year analysis periods.

Graph 1 on the front page depicts the Connecticut employment time series covering the historical period 1997:Q4 - 2001:Q4, and the forecast period 2001:Q4 - 2003:Q4. For the unadjusted nonfarm series, the peak of the last expansion was 2000:Q4, or one quarter later than the peak observed in the seasonally adjusted series (and, it is the seasonally adjusted series that is used to identify turning points in the State's business cycle). The rise in employment during the 1997:Q4 - 1999:Q4 boom period is quite apparent. Employment rose to 1.695 million in that period. More than 50,000 jobs were added on an unadjusted basis as shown in the table next page. Reflecting the long-run shift to a service economy, coupled with the boost from the late nineties boom, a little more than 57 percent, or 29,000 of the newly created jobs over this period were in the services division. Financial deregulation and innovation and the boom in the stock market, among other factors, fueled growth in employment in the finance, insurance, and real estate (FIRE) division, especially in finance. Of the more than 8,000 jobs created in the FIRE division over 6,000 of those were in the finance sector. Retail trade added 8,000 jobs over this period, driven by the rapid expansion of suburban retail centers such as the Buckland Hills Mall. Construction added 6,000 jobs and government (which includes Indian tribal government), added over 11,000 jobs. The growth in both of these divisions was, to a significant degree, the result of the expansions of the casinos by the New London County tribal nations. Gains in government were also the result of local government employment growth, due largely to the growth in local education employment. Manufacturing continued its long-run trend, shedding 12,000 jobs over the two-year period. This reflects continued gains in productivity brought about by replacement of labor with high-technology capital and the increasing pace of outsourcing many of the remaining labor-intensive functions.

The next two-year period, 1999:Q4 - 2001:Q4 is punctuated by the current recession. In the aggregate, a little over 4,000 jobs were lost over this period. The only bright spot was the government division. Again, fueled by casino expansions by the tribal nations and the "Echo Baby-Boom" in education, government added 10,000 jobs. Two critical divisions contributed to the net loss of jobs. Between 18,000 and 19,000 jobs were lost in the manufacturing division over this two-year period. In addition to the long-run trend job decline in manufacturing, the current recession was led by the sudden collapse of business investment-demand due to the Dot.Com bust and over-investment in some sectors, such as telecommunications. This resulted in massive layoffs to shore up bottom lines and maintain productivity levels. Though not as heavily exposed to the semiconductor and durable goods sectors as other regions, the Connecticut economy still experienced significant job losses in manufacturing. The job losses in durable goods in Connecticut were about the same as those during the previous 1997:Q4 - 1999:Q4 boom period. Thus, durable goods still accounted for the largest number of lost manufacturing jobs. However, in relative terms, it was the nondurable goods sector that was hit hardest with job losses over the 1999:Q4 - 2001:Q4 period, where employment losses were four times those of the losses over the previous two-year period. The retail trade division shed 1,500 jobs as a result of the recession and industry trends that resulted in the closing of traditional "big boxes" such as Caldor and Lechmere. Wholesale trade lost 3,000 jobs over the 1999:Q4 - 2001:Q4 period.

Though the services division did not actually lose employment, job gains slowed to only one-fourth of the gains in the previous twoyear period. This eliminated an engine of growth that could have mitigated job losses in other divisions. Over 15,000 jobs were lost in the travel, tourism, and leisure sector, especially in transportation and advertising, due to the recession and then the September 11th terrorist attacks. In addition, 1,200 jobs were lost in the business services sector due to the Dot.Com bust and the manufacturing- led recession. The net result was a reduction in the gain in services from 29,000 during the 1997:Q4 - 1999:Q4 period to just under 7,000 jobs over the 1999:Q4 - 2001:Q4 period.

Forecast for the Major Industry Divisions: 2001:Q4 - 2003:Q4

Referring again to Graph 1, Connecticut employment is expected to decline from its 2001:Q4 level of 1.691 million, to a level of 1.683 million in 2002:Q4, a loss of over 8,000 jobs over the first half of the forecast period. The forecast for jobs in 2003:Q4 is 1.693 million, a gain of approximately 10,000 over the second half of the forecast period. The net change over the entire two-year forecast period is an increase of just over 1,700 jobs. This amounts to recouping nearly 43 percent of the 4,173 jobs lost over the recent historical period, but nowhere near the previous gain of over 50,000 jobs experienced over the 1997:Q4 to 1999:Q4 period.

Graph 2 above presents the year-to-year (YTY) percent change in Connecticut employment over the sixteen quarter historical period, the first two quarters of 2002 for which preliminary data are available, and the forecasts for remaining six quarters of the forecast period. As depicted in Graph 2, the employment losses are expected to decelerate for the last half of 2002, and into 2003. Positive job-growth is expected to return at the end of 2003. The rebound can also be observed in the forecast region of the employment series presented in Graph 1.

Job gains over the 2001:Q4 - 2003:Q4 forecast period are confined to two divisions: services and retail trade, which together will account for 50 percent of all jobs. The services division is expected to add between 8,000 and 9,000 jobs. These gains will be concentrated in other services (5,000), and travel, tourism, and leisure (just under 2,900), which will experience a slight rebound in jobs, but still leaves the level of employment far below its 1999:Q4 level and only slightly above its 2001:Q4 level. Also, health services is projected to see a gain of 1,300 jobs. This modest growth is predicated on the expected net result of two countervailing trends that are exerting opposing forces on employment growth in health services. Clearly, the aging of the baby boomers, and the increased longevity of the population in general, is increasing the need for health services and therefore increased demand for workers in the health services sector. However, putting downward pressure on that job-growth is the return of rapidly increasing health-care costs, especially driven by increases in the cost of prescription drugs, the current Federal cap on Medicare payments, and the dropping of elderly subscribers by HMO's. This should mute what would otherwise be robust growth in health services employment over the forecast period. The only other division forecasted to add jobs over the 2001:Q4 - 2003:Q4 forecast period is the retail trade division. Just under 3,000 retail jobs are expected to be added. This will be largely due to the growth of employment in eating and drinking places, especially the chain establishments like Outback Steak House and Applebee's. In addition, "mega-boxes" such as Wal-Mart, and big-box specialty stores such as Target and Kohl's should continue to add jobs. Forecasted growth in retail trade could be optimistic if the saturation point is reached, especially in the Hartford market. Growth in the retail trade sectors is also expected to come from building materials and garden supply adding 1,167 jobs.

Substantial gains in the services and retail trade sectors will be somewhat offset as manufacturing is expected to return to its long-run trend of employment losses as firms continue to replace labor with new-technology-based investment and contract out remaining labor-intensive functions. However, the losses are expected to be somewhat smaller than those of even the 1997:Q4 - 1999:Q4 boom period. The current and planned increases in expenditures on defense and homeland security should particularly benefit Connecticut. Thus, manufacturing job losses over the forecast period should be under 10,000, compared to over 12,000 during the boom period. Durable goods manufacturing is expected to level at 165,554 after an estimated loss of 7,521 jobs. An optimistic outlook for manufacturing is that restructuring is slowing; hopefully the bulk of firms have completed the process of implementing new technology and/or the outsourcing of accounting functions, thus leading to a decreasing rate of job loss and eventual stability within the industry.

Projected employment changes for the other major industry sectors are expected to be small. Construction will have a small gain (+312) over the forecast period, but will essentially be flat. Some large projects, like Adriaen's Landing in Hartford, will increase construction jobs over the forecast period. However, the large-scale casino projects that fueled the boom in construction employment in the nineties will not be there, unless the new tribal petitions make it past current State challenges. Even then, such projects would not break ground well beyond the forecast period. Government jobs are expected to remain stable. FIRE is expected to experience a slight decline in employment (-87), particularly in the finance sector. Industries with fewer projected jobs are wholesale trade (-747); transportation, communications, and utilities (- 133). Mining is expected to decline by 13 jobs.

Risks to the Forecast

There are a number of events that would make the preceding forecast pessimistic. These include the following: GDP grows by 3.0% or more in 2002:Q3 and 2002:Q4; the stock market begins to recover and consumer confidence rebounds; war with Iraq is averted and oil prices fall back to around $25/bbl by 2003:Q1; U.S. export growth accelerates; the unemployment rate drops by another 0.25 percentage points or more; or U.S. employment experiences back-toback increases of 50,000 or more.

A number of domestic economic events could result in the forecasts being optimistic. They include: the economy continues its August slowdown and sinks into another recession; the stock market continues its slide and further drags down U.S. consumer confidence, as well as world markets along with it; the national unemployment rate increases 0.25 to 0.50 percentage points and the level rises above 6.5 percent; or a return to backto- back declines of 50,000 or more in employment.

Some economic events on the international scene that could render the forecasts optimistic include an economic crisis in Latin America originating in Brazil and Argentina; or the dollar continues to slide, and is exacerbated as speculation moves from shorting stocks to shorting the dollar. And, in response, the Fed raises interest rates to defend the dollar in an effort to prevent further capital flight from U.S. markets.

Specific to Connecticut, the State budget is once again sliding into deficit. This could seriously and negatively impact the State's economy. Further, effects of the stock market's decline and rising house prices could severely constrain Connecticut household spending.

The biggest and most likely noneconomic event that could make the forecast overly optimistic is the growing likelihood that the U.S. and Britain will invade Iraq and fight a protracted ground war involving a large and long commitment of ground troops. This could potentially push oil prices up toward $40/bbl or more. This especially seems likely in light of OPEC's backing off its original promises to boost production in the event of a U.S. war with Iraq, and Russia's recent newly signed agreements with Iraq. An invasion of Iraq would also result in a further plunge in consumer confidence. Further, with former coalition members not kicking in to help defray costs, as in the Gulf War, the effects on the Federal Budget deficit could be severe. Other non-economic factors that present a risk to the forecast include another domestic terrorist attack, the re-igniting of Indian- Pakistani tensions, and a continued deterioration of the situation in the Israeli-Palestinian conflict.

For the complete paper with forecast methodology contact Daniel Kennedy at 860-263-6268, or email daniel.kennedy@po.state.ct.us

Return to Top

Connecticut Economy Treaded Water in July

Alan Greenspan's appearance before the Senate Budget Committee on September 12 was decidedly uninspiring. While acknowledging that the U.S. economy was weaker than expected, he offered nothing that would instill confidence in the consumers. Rather, he warned against the mounting Federal budget deficit, and suggested that it could cause interest rates to rise. This is rather puzzling because the link between budget deficits and interest rates is tenuous at best. This could only serve to further erode confidence in the economic recovery. Adding to the uncertainty is the mounting tension between U.S. and Iraq. Should hostilities break out between U.S. and Iraq, oil prices are certain to rise. We may be facing the unpleasant prospect of a weak economy and rising inflation rate.

In what is becoming a pattern, the CCEA-ECRI coincident and leading employment indexes turned in a mixed performance for July 2002. The CCEA-ECRI coincident employment index fell on a year-to-year basis from 109.3 in July 2001 to 107.9 in July 2002. All four components are negative contributors to the index, with a higher insured unemployment rate, a higher total unemployment rate, lower total nonfarm employment, and lower total employment. On a sequential month-to-month basis, the CCEAECRI Connecticut coincident employment index remained stable at 107.9 for June and July 2002. Total employment is the only positive contributor, while the insured unemployment rate, total unemployment rate, and total nonfarm employment are negative contributors. Year-to-date, the CCEA-ECRI Connecticut coincident employment index has risen twice - in January and May 2002.

The CCEA-ECRI Connecticut leading employment index, rose from 113.0 in July 2001 to 114.1 in July 2002. Three components of this index are negative contributors, with higher initial claims for unemployment insurance, a higher short duration (less than 15 weeks) unemployment rate, and a lower Hartford help-wanted advertising index. The three positive contributors are a lower Moody's Baa corporate bond yield, higher total housing permits, and higher average weekly hours worked in manufacturing and construction. On a sequential month-to-month basis, the leading employment index also rose from 113.0 in June to 114.1 in July 2002. On this basis, three components are positive contributors, with a lower Moody's Baa corporate bond yield, higher total housing permits, and a lower short duration (less than 15 weeks) unemployment rate. Two components are negative contributors, with higher initial claims for unemployment insurance, and lower average weekly hours worked in manufacturing and construction, while the Hartford help-wanted index remained the same from a month earlier. Year-to-date, the CCEA-ECRI Connecticut leading employment index has risen five months out of seven, in January, February, April, June and July.

Although the situation in July in Connecticut is not much changed from the previous month, I am encouraged by the second consecutive rise in July in the CCEA-ECRI Connecticut leading employment index. Finally, as we observe the first anniversary of September 11, I am reminded of how fortunate we are to be living in a free country like the U.S. and the price that we have to pay to ensure our freedom.

Francis W. Ahking, Department of Economics, University of Connecticut, Storrs, CT 06269. Phone: (860) 486-3026. Stan McMillen [(860) 486-0485, Storrs Campus], Connecticut Center for Economic Analysis, University of Connecticut, provided research support. Leading and coincident employment indexes were developed by Pami Dua and Stephen M. Miller, in cooperation with Anirvan Banerji at the Economic Cycle Research Institute.

Return to Top

HOUSING UPDATE

Year-to-Date August Permits Up 0.6 Percent

Commissioner James F. Abromaitis of the Connecticut Department of Economic and Community Development announced that Connecticut communities authorized 811 new housing units in August 2002, a 23.1 percent decrease compared to August of 2001 when 1,055 units were authorized.

The Department further indicated that the 811 units permitted in August 2002 represent a 12.5 percent decrease from the 927 units permitted in July 2002. The year-to-date permits are up 0.6 percent, from 6,494 through August 2001, to 6,534 through August 2002.

Only the Danielson and Waterbury Labor Market Areas (LMAs) showed a slight gain in permits while the remaining eight LMAs had decreases compared to a year ago. Farmington led all Connecticut communities with 43 units, followed by Southington with 30 and Middletown with 18. From a county perspective, Hartford County had the largest gain (19.2 percent) compared to a year ago.

Return to Top

Industry Clusters

Grant Establishes Greater Valley Manufacturing Training Network

The industry cluster initiative has from its inception focused on industry collaboration, most importantly to meet the need for workforce training and development. That collaboration was extended late this summer with the announcement of the 11th Connecticut Business Training Network (CBTN). A $10,000 grant was awarded to establish the Greater Valley Manufacturing Training Network among Naugatuck River Valley manufacturers.

An assessment report prepared for the Community Foundation for Greater New Haven and its Valley Advisory Committee cited the need for the development of such a demanddriven training program for Valley manufacturers. Provided by the Connecticut Department of Economic and Community Development (DECD), the grant will be used to assess the employee training needs and identify training sources to meet them on a collaborative basis.

The network includes: Spectrum Plastics Molding, Inc.; Shaw Mudge & Company; OEM Controls, Inc.; Inline Plastics Corp.; H. George Caspari, Inc.; Peabody Engineering; Perkin Elmer Instruments; Fluidyne Ansonia; Hasler, Inc.; and Firing Circuits. The Greater Valley Chamber of Commerce will serve as the organizational center for the network.

The CBTN program is a direct result of a partnership among the DECD, the Connecticut Department of Labor, and the Governor's Council on Economic Competitiveness and Technology. The Connecticut Business and Industry Association (CBIA) administers the program. Since its establishment in 2000, the program has helped 87 businesses employing over 12,800 workers.

Return to Top