Employment trends are analyzed from ES-202 data
By Edward T. Doukas Jr., Research Analyst

For each calendar quarter, the Covered Employment and Wages Program, referred to as the ES-202, provides a compilation of employment, total wages, taxable wages, and contribution data from employers subject to State unemployment insurance laws. The ES-202 furnishes a distribution of this data according to the area, ownership, and industrial classifications of the reporting employers.

As part of the Social Security Act of 1935, the ES-202 was initially developed to collect and analyze the employment, wages, and contributions data from quarterly Unemployment Insurance (UI) reports filed by employers subject to State Unemployment Insurance laws. This action was taken to ensure the State programs were in compliance with the provisions of the Act. In addition, the data were used to determine the extent of Unemployment Insurance coverage in the labor force.

Since that time, the ES-202 has evolved into a timely and accurate indicator of Connecticut's labor market. ES-202 data represent the largest available universe of monthly employment and quarterly wage information by industry and area. However, there are several sectors that are not included in the ES-202's coverage. Among noncovered workers are: some domestic employees, most workers in agriculture, railroad, churches, and nonprofit establishments; and state and local government elected officials. In spite of these exclusions, approximately 96 percent of all Connecticut's employment is represented in ES-202 data.

In addition to its continued use for federal and state Unemployment Insurance program administration, information from the ES-202 program has many other applications in statistical analysis. The ES-202 serves as a foundation for some of the key activities of the Bureau of Labor Statistics (BLS) of the United States Department of Labor. With its comprehensive coverage of employment, it provides an excellent frame from which to draw samples for use in survey-based statistical programs and special studies. The Current Employment Statistics program (CES) uses the ES-202 both as a sample frame for producing current monthly estimates and for the annual benchmarking of the industry employment estimation process. The Current Employment Statistics differs from the ES-202 in that it also includes estimates for jobs in industries not covered by Unemployment Insurance laws.

Other Bureau of Labor Statistics activities which rely on ES-202 data include the Industry and Area Wage Surveys, Occupational Employment Statistics, and Occupational Safety and Health Statistics Programs. The Bureau of Economic Analysis of the United States Department of Commerce uses the aggregated ES-202 employment and wage data to estimate personal income at the county, state and national levels. ES-202 data are also used in a variety of econometric and economic forecasting models, area and industry analyses and impact studies. The following analysis focuses on employment data from the ES-202.

First Quarter 1996 Data

Connecticut's Unemployment Insurance covered employment averaged 1,529,595 during the first quarter of 1996, an increase of 0.9 percent when compared to the 1,515,545 average posted during the first quarter of 1995. The first quarter 1996 increase marked the second consecutive rise in average first quarter employment, and was the highest figure posted for the period since 1991 when total employment averaged 1,536,052. First quarter average employment reached its pinnacle in 1989 at 1,648,725.

Comparing first quarter 1996 data to that of the peak first quarter, first quarter 1989, shows that only three industry divisions had a quarterly average employment greater in first quarter 1996 than in first quarter 1989. The divisions with greater employment were: services, up 14.4 percent; agriculture, forestry and fishing, up 10.9 percent; and transportation and public utilities, up 1.1 percent. The mining division showed the greatest decline from first quarter 1989, down 48 percent. Construction followed, down 38.3 percent, and manufacturing, down 25 percent.

Employment data for first quarter 1996 displayed the continuation of two significant trends in Connecticut's economy: the decline in manufacturing employment and the increase in service sector employment. first quarter 1996 figures confirmed that manufacturing employment continued to decrease as a percentage of total covered employment. During first quarter 1996 manufacturing employment accounted for 18.0 percent of Connecticut's total covered employment compared to 18.5 percent in first quarter 1995, 21.5 percent in first quarter 1990, 27.4 percent in first quarter 1985, and 32.0 percent in first quarter 1980. In contrast, service sector employment has continued to increase as a percent of total covered employment rising to 29.7 percent in first quarter 1996. This compares to 28.7 percent in first quarter 1995, 25.1 percent in first quarter 1990, 21.4 percent in first quarter 1985, and 18.9 percent in first quarter 1980.

A look at major industry group data also highlights the decline in manufacturing employment and the increase in services sector employment. Among major groups with quarterly average employment of 1,000 or more, five of the ten major groups with the largest percentage decline from first quarter 1995 were in the manufacturing division. These included: textile mill products, 24.8 percent; apparel and other textile products, 9.1 percent; food and kindred products, 6.6 percent; transportation equipment, 6.3 percent; and rubber and miscellaneous products, 3.5 percent.

Conversely, six of the major groups which ranked in the top ten with the largest percentage growth over first quarter 1995 were in the services division. These included: services, not elsewhere classified, 23.2 percent; business services, 8.6 percent; museums and art galleries, 7.9 percent; amusement and recreational services, 7.0 percent; private households, 6.9 percent; and auto repair, services & parking, 6.8 percent.

Comparing first quarter 1996 major group level data to that of first quarter 1989 further exhibits the contracting nature of manufacturing employment and the expanding significance of service sector employment in Connecticut. Of the twenty major groups that constitute the manufacturing division, only three have rebounded significantly enough from the recession of the early 1990s to have an average quarterly employment greater in first quarter 1996 than the average posted in first quarter 1989. Oppositely, of the sixteen major groups that comprise the services division, eleven had an average quarterly employment greater in first quarter 1996 than they had in first quarter 1989.

Table 1:

Return to Top

Housing Update
October housing permits increase

The Connecticut Department of Economic and Community Development announced that Connecticut communities authorized 746 new housing units in October 1996, a 1.2% increase compared to September 1996 when 737 were authorized.

The Department further indicated that the 746 units permitted in October 1996 represent an increase of 3.8% from the 719 units permitted in October 1995, and that the year-to-date permits are down 9%, from 7,076 in 1995 to 6,438 in 1996.

Reports from municipal officials throughout the state indicate that Tolland County showed the greatest percentage increase in October compared to the same month a year ago: 29.7%. Windham County reported the greatest percentage decline: 27.8% for the same period.

Fairfield County documented the largest number of new, authorized units in October with 225. Hartford County followed with 169 units and New Haven County had 113 units. Stamford led all Connecticut communities with 122 units, followed by Avon with 21 and Manchester with 20.

Year-to-date totals indicate that Hartford County has issued the most building permits through October of 1996 with 1,460, followed by Fairfield county with 1,336, and New Haven County with 1,286. Stamford authorized 248 new units during this period, followed by Southington with 162, Shelton with 146, and Glastonbury with 145.

Return to Top

A brief look at the Consumer Price Index
By Salvatore A. DiPillo, Labor Statistics Supervisor

The Consumer Price Index (CPI), produced by the United States Department of Labor's Bureau of Labor Statistics (BLS), determines the average change over time in the prices paid by urban consumers for a fixed market basket of consumer goods and services. The Consumer Price Index is one measure of inflation and is an indicator of the effectiveness of government policy. Indeed, changes in the Consumer Price Index and the Producer Price Index (PPI) are determinants in the Federal Reserve Board's monetary policy. The Consumer Price Index is also used to deflate other economic series for price change, and to translate these series into constant, or inflation-free dollars. It is often used in escalation agreements to adjust payments for changes in prices. Common uses are in collective bargaining agreements, rental contracts, alimony and child support payments, retirement benefits and Social Security payments.

Represented by the Consumer Price Index are all goods and services purchased for consumption by urban households. Expenditure items are arranged into seven major groups: food and beverages, housing, apparel and its upkeep, transportation, medical care, entertainment, and other goods and services (haircuts, college tuition and bank fees are some examples). The Consumer Price Index reflects spending patterns for each of two population groups: All Urban Consumers (Consumer Price Index-U), and Urban Wage Earners and Clerical Workers (Consumer Price Index-W). The Consumer Price Index-U, United States City Average for All Items represents about 80 percent of the total United States population and is the index most commonly used in reports of inflation.

Bureau of Labor Statistics usually updates the reference base period every ten years or so. The Bureau most recently used the 1980 Census of Population to select from which urban areas to collect prices and to determine the number of consumers to be represented in each area. Data from the Consumer Expenditure Survey conducted from 1982 and 1984 provided details on consumers' spending habits, which Bureau of Labor Statistics used to construct the Consumer Price Index market basket of goods and services, and to assign a weight to each item based on total family expenditure. Each month Bureau of Labor Statistics field representatives call upon thousands of retail stores, services establishments, rental units, doctors' offices and other business places to obtain price information on specific goods and services in the Consumer Price Index market basket. The recorded data is sent to the national Bureau of Labor Statistics office where commodity specialists check the data for accuracy and consistency, striving to keep changes in the quality of items from affecting the Consumer Price Index's measure of price change.

Each month, indexes are published along with short-term changes, the latest 12-month change and, at the national level, unadjusted and seasonally adjusted percent changes, as well as annualized rates of change. As with any survey-based data, the Consumer Price Index is subject to some amount of sampling and non-sampling error. Since the Consumer Price Index measures price change based on only a sample of items, the published indexes differ from what would be if actual records of all purchases by everyone in the population could be used to compile the index. Non-sampling errors are caused by problems of price data collection, logistical lags in conducting surveys and difficulties in handling the problems of quality change. Bureau of Labor Statistics expends much effort to minimize these.

In addition to the publication of the national (United States City Average) Consumer Price Index-U and Consumer Price Index-W indexes, Bureau of Labor Statistics publishes monthly indexes for the four regions: Northeast, North Central, South and West. Bureau of Labor Statistics also publishes five major metropolitan areas monthly; among these is the New York-Northern New Jersey-Long Island, New York-New Jersey-Connecticut metropolitan area. (There is no Consumer Price Index developed separately for Connecticut.) Other areas are published bi-monthly, semi-annually and annually. These local area indexes are by-products of the national Consumer Price Index program, have much smaller sample sizes than the national or regional indexes and are therefore, subject to more sampling and other measurement error. As a result, local area indexes are more volatile than the national or regional indexes, even though their long-term trends are similar. It is not possible to make comparisons between the cost of living in different subnational areas using the Consumer Price Index for these areas. The Consumer Price Index measures only time-to-time changes in each place. In general, both the market basket and relative prices of goods and services in the base period vary substantially across areas.

How the Consumer Price Index is used to measure changes in prices can be seen in the following example. The New York-New Jersey-Connecticut Consumer Price Index for October 1996 was 168.2. A year earlier, the index number was 163.6. The difference between the index numbers, 4.6, taken as a percentage of the index number for the base period (October 1995), determines the change in prices over the year. This calculation shows that consumer prices rose by 2.8 percent between October 1995 and October 1996 in the New York-New Jersey-Connecticut metropolitan area. Table 2 below shows this calculation and compares the change in cost of living to that determined for the United States city average.

Table 2:

(Compiled from Bureau of Labor Statistics reports)

Return to Top

Leading & Coincident Indicators
Coincident and leading indexes signal continued recovery

Connecticut's coincident employment index moved, once again, to its highest level in the current recovery with the release of the (preliminary) September data, having not fallen on a month-to-month basis since December 1995. Connecticut's leading employment index also moved to its second highest level in the current recovery, falling just a fraction short of its previous peak in June 1996.

The coincident index, a gauge of current employment activity, continued its strong upward momentum. This recent momentum contrasts sharply with the weak upward movement in the coincident index in the early phases of the current recovery, as shown in the chart. Note that the coincident index experienced little growth between 1991 and 1995, but a stronger upward movement in 1996. Unfortunately, it is only the upward movement in the coincident index in 1996 that approaches the trend movement in this index over the entire 1980s expansion.

The September data signal that the economy continues to climb out of the Great Recession, and at an accelerated pace in 1996. With the current recovery already two-thirds the length of the 1980's expansion, can the coincident index continue to emit such strong signals? Connecticut watchers can only hope that this trend continues in the near term and that no national events transpire, causing the Federal Reserve to apply the policy brakes and to halt the current recovery.

The leading index, a barometer of future employment activity, has bounced around considerably since reaching a peak in December 1994. It has since then exceeded that peak on four separate occasions -- September 1995, May 1996, June 1996, and, most recently, September 1996. We shall continue to monitor closely the leading index to see if, and when, it starts a strong downward move. No such signal is yet observable in the data.

The coincident employment index rose from 83.3 in September 1995 to 89.0 in September 1996. All four index components continued to point in a positive direction on a year-over-year 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 September 1995 to 89.7 in September 1996, or just below its previous peak of June 1996. Three of the five index components sent positive signals on a year-over-year basis with lower initial claims for unemployment insurance, a lower shortduration (less than 15 weeks) unemployment rate, and higher total housing permits. The final two components sent negative signals on a year-over-year basis with a lower average work week of manufacturing production workers and lower Hartford help wanted advertising.

Source: Connecticut Center for Economic Analysis, University of Connecticut. Developed by Pami Dua [(203) 322-3466, Stamford Campus (on leave)] and Stephen M. Miller [(860) 486-3853, Storrs Campus]. Tara Blois [(860) 486-4752, Storrs Campus] provided research support.

Return to Top