Insight : North Carolina's labor and economic outlook.

              1

0.0%

1.0%

2.0%

3.0%

S O N D J F M A M J J A

2000- 2001 1999- 2000

INSIGHT

Michael F. Easley, Governor Employment Security Commission of North Carolina Thomas Whitaker, Acting Chairman

Volume 2, Number 1 October 2001

North Carolina’s Labor and Economic Outlook

NC Quick Stats: August 2001

Labor Force 3,997,600

Employment 3,796,800

Unemployment 200,800

Unemployment Rate 5.0%

Note: Data are preliminary and are

seasonally adjusted.

INSIDE

Labor Market Abstract ............... 1

Economic Indicators in

North Carolina .............................. 1

Issues in North Carolina’s

Unemployment Insurance System:

Average Duration ......................... 3

Introduction .................................. 3

Economic and Demographic

Determinants of Duration ......... 4

Differences in UI Laws ................ 7

Comparison of North Carolina’s

and Georgia’s Reemployment

Programs ..................................... 7

Conclusion .................................... 9

The Reemployment Outcomes of

Dislocated Manufacturing

Workers ......................................... 11

Methodology: Tracking

Laid- Off Workers ..................... 11

Reemployment After a Layoff .. 13

Wages Earned After a Layoff ... 14

Differences by Age Group ........ 15

Differences by Educational

Level .......................................... 16

Differences by Industry ............ 16

Differences by Race

and Gender................................ 17

Conclusion .................................. 17

( Continued on Page 2)

Labor Market Abstract

During August 2001, the North Carolina seasonally adjusted unemployment

rate decreased to 5.0 percent from 5.3 percent the previous month. During

the same period, the civilian labor force grew by approximately 9,000. Em-ployment

in the service producing industries rose during the month with most

increases occurring in retail trade, hotels & lodging and health services. A

decrease occurred in the manufacturing industry with losses primarily in tex-tiles,

furniture and electronic equipment. Overall, the unemployment level

decreased from an estimated level of 210,800 in July 2001 to 200,800 in Au-gust

2001.

Economic Indicators in North Carolina

Economic indicators used to predict future economic activity are referred to

as leading indicators, while coincident indicators are used to help determine

changes in the economy that are concurrent with such indicators. All graphs

reflect the most recent monthly statewide data.

Insured Unemployment Rates*

Adjusted Unemployment Rates*

Total Nonagricultural Employment, in Thousands*

* Source: ESC, Labor Market Information Division

LMI Happenings ........................ 18

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

S O N D J F M A M J J A

3,700

3,800

3,900

4,000

4,100

S O N D J F M A M J J A

2

2000- 2001 1999- 2000

Sales and Use Tax Revenues, in Millions

Average Weekly Hours Worked in Manufacturing

Initial Claims

Economic Indicators in North Carolina ( Continued from Page 1)

Percent

Latest Previous Previous Change From

Month Month Year Last Year

Asheville 851 1,529 946 - 10.0

Charlotte 4,027 3,394 2,103 91.5

Durham 1,074 1,126 557 92.8

Fayetteville 1,347 1,661 1,007 33.8

Goldsboro 609 1,144 611 0.0

Greensboro 2,450 2,923 1,834 33.6

Greenville 1,274 1,781 794 60.5

Hickory/ Newton 5,683 10,669 2,099 170.7

Jacksonville 438 365 471 - 7.0

Raleigh 2,834 3,005 1,814 56.2

Wilmington 1,070 1,200 828 29.2

Winston- Salem 2,771 4,837 2,345 18.2

Source: Employment Security Commission

Percent

Latest Previous Previous Change From

Month Month Year Last Year

Asheville 39.8 39.9 41.7 - 4.6

Charlotte/ Gastonia 40.3 39.8 41.7 - 3.4

Greensboro/

Winston- Salem/

High Point 38.9 37.9 40.1 - 3.0

Raleigh/ Durham/

Chapel Hill 41.3 40.3 43.1 - 4.6

Source: Employment Security Commission

Percent

Latest Previous Previous Change From

Month Month Year Last Year

Asheville 211.9 191.3 207.0 2.4

Charlotte 1,024.5 1,124.5 1,062.2 - 3.5

Durham 252.1 253.1 258.4 - 2.4

Fayetteville 197.1 196.9 226.0 - 12.8

Greensboro 521.2 503.4 525.6 -. 01

Greenville 141.9 134.3 143.1 -. 01

Hickory 133.2 125.9 124.1 7.3

Raleigh 617.7 636.0 615.2 0.0

Wilmington 240.4 222.4 224.4 7.1

Winston- Salem 359.6 342.4 369.1 - 2.6

Source: N. C. Department of Revenue, Tax Research Division

Statewide In Selected Cities

In Selected Metropolitan Statistical Areas Statewide

Statewide, in Thousands By ESC Local Offices

( Continued on Page 19)

0

20

40

60

80

100

120

140

160

S O N D J F M A M J J A

37

38

39

40

41

42

43

S O N D J F M A M J J A

$ 0

$ 100

$ 200

$ 300

$ 400

S O N D J F M A M J J A

3

Why is North Carolina’s average du-ration

higher than Georgia’s? This

study looks at this issue from three

perspectives. First of all, there are

economic and demographic differ-ences

in the workforces between the

states. Factors such as the number

of manufacturing workers or the

overall unemployment rate may cause

a given state’s duration to differ from

another state’s, all else equal. The

next section, “ Economic and Demo-graphic

Determinants of Duration,”

attempts to show that some of these

differences imply that North

Carolina’s duration should be higher

than Georgia’s.

Issues in North Carolina’s Unemployment Insurance

System: Average Duration

Introduction

In the Unemployment Insurance ( UI) system, duration refers to the number

of weeks UI claimants receive benefits before returning to work. One of the

primary objectives of the Employment Security Commission ( ESC) is to as-sist

in job search, thereby reducing the average duration of filing for benefits.

A low average duration means UI recipients are returning to work quickly,

thus saving the UI system, and the employers in the state that fund it, consid-erable

sums of money. For example, if North Carolina’s average duration

had been reduced by one week over the last year, the UI Trust Fund would

have saved over $ 64 million. This value is obtained by multiplying the aver-age

weekly benefit amount in the state over the last year ($ 235.31) by the

number of “ first payments” in the same period ( 272,597).

North Carolina would benefit more from a reduction in average duration than

other states in the Southeast because both its average weekly benefit amount

and number of first payments are relatively high. For instance, Georgia and

Virginia, two comparably- sized states, would have saved only $ 44 million and

$ 22 million, respectively, if their durations had been reduced by one week.

Georgia’s average weekly benefit amount was $ 215.39 and its number of

first payments was 203,959 while the corresponding numbers for Virginia

were $ 210.22 and 106,018.

Fortunately, over the last few years North Carolina has had either the 2nd or

3rd lowest duration among the states, after Georgia ( and recently New Hamp-shire).

In the first quarter of 2001, North Carolina’s average duration for the

12- month period was 9.2 weeks, compared to 8.5 weeks for Georgia and 8.7

weeks for New Hampshire. As Figure 1 shows, the average duration in

North Carolina has been consistently above Georgia’s in the last few years. 1

A one- week reduction in North

Carolina’s average duration last

year would have saved its UI Trust

Fund over $ 64 million.

At 9.2 weeks, North Carolina’s

average duration is the third

lowest in the nation.

North Carolina’s low average

duration of filing for UI means

claimants are returning to work

faster than in other states.

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

1998.2 1998.3 1998.4 1999.1 1999.2 1999.3 1999.4 2000.1 2000.2 2000.3 2000.4 2001.1

Quarter

Weeks of Duration

Georgia North Carolina

Figure 1: Average Durations in North Carolina and Georgia

4

Secondly, there are differences in UI laws between North Carolina and Geor-gia.

Although the states must conform to general federal guidelines when

operating a UI system, each state has some flexibility in the procedures of its

UI system. For example, states have different maximum weekly benefit

amounts and they may establish different laws for allowing UI claimants to

refuse job offers. These differences, detailed in the section “ Differences in

UI Laws,” can have important implications for the length of average dura-tions.

Thirdly, reemployment programs in the respective states are likely to have an

important impact on duration. North Carolina and Georgia have both imple-mented

special reemployment initiatives in the last few years. We will

compare these initiatives in the section titled “ Comparisons of North Carolina’s

and Georgia’s Reemployment Programs.”

Economic and Demographic Determinants of Duration

In order to compare North

Carolina’s duration to other

states’, one should look at how

the characteristics of the

economy and workers in the

state affect average duration.

Using just some of the charac-teristics

of the workforce and

other economic factors that are

important to duration, we have

predicted the average duration

for the 50 states. Figure 2 shows

the predicted durations and the

actual durations for the seven

southeastern states. ( See Ap-pendix

1 for a graph of all 50

states.)

North Carolina’s predicted du-ration

is higher than several other southern states, particularly Alabama, South

Carolina, Georgia, Tennessee and Virginia. North Carolina’s actual average

duration was 2.9 weeks shorter than predicted by this model, which was the

largest difference of the states in the region.

The predicted durations were derived using a regression analysis on data

from the 50 states in 2000. With this regression, one can make some general

statements about how certain variables affect duration and how, when taken

together, these variables impact North Carolina’s duration relative to other

states.

Six variables were used to predict average duration. These variables were

chosen based on assumptions that they were important factors in determin-ing

duration, as explained in the following paragraphs. In addition, data on

these variables were readily available for all 50 states.

Differences in UI laws and

reemployment programs in the

respective states may explain the

differences in duration.

Regression analysis on data from

50 states in 2000 predicted North

Carolina’s duration to be higher

than several other southern states,

particularly Alabama, South

Carolina, Georgia, Tennessee and

Virginia.

0

2

4

6

8

10

12

14

16

AL SC GA TN VA NC FL

Weeks

Average Duration Predicted Duration

Figure 2: Average Duration in 2000 Compared with Predicted Duration

5

The first of these variables is the num-ber

of workers in the state who were

covered by unemployment insurance,

i. e. covered employment, in 2000. It

is expected that states with larger cov-ered

employment will have a higher

average duration. Employment of-fices

in larger states may often face

a greater number of job applicants.

North Carolina’s covered employment

is among the largest in the Southeast

region, as shown in Figure 3. How-ever,

it is less than Florida’s and not

significantly higher than either

Georgia’s or Virginia’s.

The second determinant of duration

used is the share of manufacturing

employment in the overall covered em-ployment

in the state. It is expected

that states with more manufacturing

will have higher durations, since many

manufacturing workers have more dif-ficulty

finding reemployment than

workers in other industries. In sup-port

of this, Current Population Survey

( CPS) data show that unemployed

manufacturing workers have longer in-dividual

durations than workers from

other industries. As Figure 4 shows,

among the Southeastern states, North

Carolina has the highest percentage

of its covered workers in the manu-facturing

sector.

The third variable is the maximum

weekly benefit amount for UI recipi-ents

in 2000. It is expected that a

higher maximum benefit amount will

increase duration. More generous

benefits may delay the need for

workers to find reemployment. Fig-ure

5 shows that, among the states in

the Southeast, North Carolina had the

highest maximum weekly benefit

amount as of July 2001. This rose to

$ 396 August 1, 2001.

Figure 3: Covered Employment in 2000

3,823 3,773 3,765

3,230

2,605

1,823 1,784

0

1,000

2,000

3,000

4,000

5,000

FL NC GA VA TN AL SC

In Thousands

Figure 4: Share of Manufacturing

Employment in Overall Employment in 1999

21.6%

20.4% 20.0% 19.7%

16.2%

12.6%

7.4%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

NC AL TN SC GA VA FL

$ 331

$ 275 $ 274 $ 259 $ 255

$ 190

$ 375

$ 0

$ 50

$ 100

$ 150

$ 200

$ 250

$ 300

$ 350

$ 400

NC VA FL GA SC TN AL

Dollars per week

Figure 5: Maximum Weekly Benefit Amounts in 2000

6

The average unemployment rate in the

state during fourth quarter 2000 is the

fourth variable. A higher unemployment

rate would imply a weaker labor market,

so that unemployed workers face a

harder time finding employment. This

would lengthen duration in states with high

unemployment. As Figure 6 shows,

North Carolina’s unemployment rate was

high relative to most of the other states

in the region during fourth quarter 2000.

The fifth factor considered important to

duration is whether the state has a one-week

waiting period for UI recipients.

Only twelve states do not have waiting

periods. Two of these states, Georgia

and Alabama, are in the Southeast. These

states may have a higher proportion of short- term unemployed filing for ben-efits

in that first week, which would reduce the state’s average duration.

The final determinant of duration is whether a state is located in the South.

This is used to isolate the idiosyncratic nature of the Southern labor market.

For instance, the lower unionization of the labor force in the South may in-crease

job availability and turnover. Given this, it is likely that duration will be

shorter in Southern states. All of the states in our region are expected to

have lower durations because of this.

The results of the regression are summarized in the following table. All of

the variables had the expected impact on duration, except the share of manu-facturing

in covered employment. The lower duration for states with higher

shares of manufacturing employment may be due to the high proportion of

attached claimants within the manufacturing sector. Attached claimants spend

a short time receiving benefits before returning to work with their company.

North Carolina has a high proportion of attached claims. The impact of

attached claimants cannot be directly obtained because the data on the other

states are unavailable.

Variables affecting duration . . . . . . and their impact estimated by the model

1. covered employment an additional one million workers raised duration by about

two- tenths of a week

2. share of manufacturing in employment a 10 percentage point increase ( from, say, 20% to 30%)

lowered duration by a little more than one week

3. maximum benefit amount a $ 10 increase raised duration by one- tenth of a week

4. average total unemployment rate a 1 percentage point increase raised duration by about

two- thirds of a week

5. 1- week waiting period increased duration by approximately .8 weeks

6. Southern state reduced duration by 1.2 weeks

A one- week waiting period for UI

recipients may affect a state’s

duration rate.

The proportion of attached claims

filed within the manufacturing

sector affects duration.

4.4%

3.8% 3.6% 3.5%

3.2%

2.9%

2.1%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

AL TN NC FL GA SC VA

Figure 6: Fourth Quarter 2000 Average Total Unemployment Rates

7

Although the regression explained over half the differences in the actual

durations of the 50 states, it did not predict every state’s duration exactly.

For instance, it overestimated North Carolina’s average duration by nearly

three weeks. There are other factors affecting duration that were not con-sidered

in the model. As mentioned earlier, the number of attached claimants

would be important. A second factor is the reemployment program in the

state, discussed later. Still other factors include the age and racial distribu-tion

of the state’s workforce, as well as, the amount of urbanization within

the state.

Differences in UI Laws

North Carolina and Georgia differ in eligibility requirements and benefits, as

established by their respective UI laws. As previously stated, North Caro-lina

has a higher maximum weekly benefit amount than Georgia, which

contributes to a higher expected duration. North Carolina’s maximum weekly

benefit was $ 375, compared to $ 274 in Georgia. This difference is a result of

the way the two states calculate the maximum weekly benefit: in North

Carolina, it is two- thirds of the average weekly wage in the state while in

Georgia it is less than one- half. Therefore, UI recipients whose high- quarter

earnings are relatively high will receive a larger proportion of their pre- layoff

wages in benefits in North Carolina than in Georgia. For example, a worker

with average weekly earnings of $ 800 would only receive 34% of this in UI

benefits in Georgia, but would have a wage replacement rate of 47% in

North Carolina. Thus, claimants in Georgia have a greater incentive to find

new jobs quickly.

Under the different state laws, it seems that claimants have an easier time

rejecting job offers in North Carolina than in Georgia. Georgia’s law speci-fies

that individuals who receive benefits for 10 or more weeks cannot reject

a job offer if the wages are at least 66 percent of their high- quarter base

period wages. North Carolina does not have such a provision. However, it

is a general practice in North Carolina’s local ESC offices to encourage

claimants who have been unemployed for many weeks to accept jobs which

offer lower wages. Also, North Carolina has a provision in its law that al-lows

individuals to refuse a job if they cannot obtain adequate childcare or

elder care.

Both North Carolina and Georgia determine the duration of benefits based

on wages earned in the base period. Most UI recipients in both states will be

eligible for 26 weeks of benefits. But if workers earned relatively little in the

entire base period compared to the high quarter, then the benefit period may

be reduced. In North Carolina, the minimum benefit period is 13 weeks,

while in Georgia the benefit period may be as low as eight weeks.

Comparison of North Carolina’s and Georgia’s Reemployment

Programs

Again, both North Carolina and Georgia have received state funds in order to

provide a reemployment program for eligible claimants that are receiving

unemployment benefits. The North Carolina program, the Reemployment

Initiative ( REI), was funded in January 2000 and was implemented in April

All factors which may affect

duration were not considered in

the model.

The difference in the way the two

states calculate the maximum

weekly benefit creates a

significant difference in what each

state pays.

There are also differences in

claimant eligibility requirements.

Maximum eligibility in both states

is 26 weeks of benefits.

8

2000. Georgia’s program, the Claimant Assistance Program ( CAP), began

with service to select areas of Georgia in 1987, but expanded to cover the

entire state eighteen months later. Georgia’s CAP was used by North Caro-lina

as a model for its REI; therefore, there are many similarities between the

programs. Some differences exist, as well.

The CAP went through a slow and evolving process over the years to be-come

what it is today. Initially, CAP provided one- on- one contact with

claimants and also offered workshops. Now, almost all efforts are in the

form of one- hour specialty workshops designed to meet the needs of the

claimants based on their input and suggestions. There is, however, one- on-one

time still available for claimants at the workshops. Until recently, CAP

participants included only claimants that were separated from work through

lack of work. Now the program includes those who are without work for

other reasons, as well, such as quitting due to child or elder care or other

cause, being fired, etc. North Carolina’s REI only includes those that have

been separated through lack of work.

The CAP local office staff undergoes training involving six consecutive courses

taught by consultants. Staff members of each district meet every six months

to discuss the program and ways they can improve their performance. There

is also an emphasis on trying to ensure that the most successful staff are

used in the program. Staff members participate in an information exchange

program that matches low performance workers with high performance ones

to improve overall quality. There is consistent monitoring of the performance

of staff members.

One of the reasons Georgia is able to provide so much training for its staff is

that it receives more appropriated funds for its program. Georgia has re-ceived

between $ 14- 19 million per year for the implementation of CAP. This

compares to approximately $ 9 million that was appropriated for the REI pro-gram

in North Carolina during its first year of implementation. CAP is funded

for a five- year time period while REI is funded for two years at a time. Also,

Georgia has received no indication from its legislature that it plans to termi-nate

CAP, while REI funding is not included in North Carolina’s budget

effective July 1, 2001.

Georgia and North Carolina both employ 160-

200 staff members in their programs. However,

Georgia has 53 offices statewide while North

Carolina has around 90 offices. This results in

more staff available at each office in Georgia.

The REI program served approximately 67,000

claimants last year, while CAP served 58,000.

However, the programs offered by the CAP

staff were also provided to 9,000 participants

in Georgia’s UI profiling program.

The CAP is a 17- week long program while REI

is 12 weeks long. Participants in CAP are re-quired

to meet with a staff person after the

first, fifth, ninth and fourteenth weeks of the

Georgia’s reemployment initiative

program was used by North

Carolina as a model for its REI.

In CAP, both methods of providing

claimants services and staff

training have evolved extensively

over time.

Georgia has received between

$ 14- 19 million yearly for CAP

while North Carolina received $ 9

million in the first year for REI.

67,000

58,000

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

North Carolina Georgia

Number of Claimants Served by a Reemployment Program in 2000

9

program. Contact is not required after the seventeenth week. REI participants are required to contact staff

either in person, by phone or by e- mail on a weekly basis for the first four weeks and biweekly for the remaining

eight weeks. CAP directs its participants into one of three tracks: self- serve, staff assisted and intensive. In both

programs, participants are subject to adjudication if they do not follow the expectations of the program. However,

this does not happen often.

For the year ending June 2001, Georgia received approximately 252,000 initial separated claims while North

Carolina received around 295,000. Of these, about 23 percent in each state participated in their respective

reemployment program, CAP or REI. These efforts resulted in an entered employment rate of 52.7 percent for

the 17- week CAP and a 44.4 percent rate for the 12- week REI, which in both cases amounts to roughly 30,000

people.

While the main goal of both CAP and REI

is to aid in UI claimants reentering the

workforce as soon as possible, another ben-efit

of both programs is to increase the

savings to each state’s UI Trust Fund. One

way this is obtained is by lowering the du-ration.

Although Georgia’s overall duration

is lower than North Carolina’s, the savings

are somewhat different. It is estimated that

CAP saved Georgia’s trust fund $ 38.9 mil-lion

for one year and North Carolina saved

its trust fund an estimated $ 42 million for

the same time period. Given that roughly

$ 5- 10 million more is spent on Georgia’s

CAP compared to REI, North Carolina’s

REI program is more cost effective.

Both the CAP and REI are beneficial to their participants and trust funds in their respective states. One might

argue that Georgia has been more successful in its reemployment efforts because it has been operating this

program for over a decade and funds it at a higher level than North Carolina. However, because these programs

are so similar and because they affect only a portion of UI claimants in each state, it is unlikely that the differ-ences

between the programs contribute much to the differences between average durations in these two states.

Conclusion

Shortening the length of time UI claimants receive benefits, or average duration, provides significant savings to a

state’s UI system. Currently, North Carolina has the third- lowest average duration among the states, although

Georgia’s duration is .7 weeks shorter. This article has attempted to cover some of the factors that explain why

one state’s duration is not as low as another’s.

Based on the regression analysis of data from the 50 states, North Carolina is expected to have a higher duration

of filing for UI benefits than many of the other states in our geographic area, including Georgia. One of the

factors which tends to push up North Carolina’s duration is its high maximum weekly benefit amount. North

Carolina has the highest maximum weekly benefit of all the states in the Southeast; it is approximately $ 100 more

than Georgia’s. The one- week waiting period for UI recipients is another factor increasing North Carolina’s

duration relative to Georgia’s. Also, North Carolina’s relatively high unemployment rate should make it harder for

UI claimants to find reemployment quickly.

In addition to the variables used in the regression, differences in UI laws between North Carolina and Georgia

would likely imply a higher duration in North Carolina. For instance, North Carolina’s laws permit claimants to

reject a wider range of unfavorable job offers.

$ 39

$ 42

$ 0

$ 5

$ 10

$ 15

$ 20

$ 25

$ 30

$ 35

$ 40

$ 45

North Carolina Georgia

Dollars in Millions

Trust Fund Savings from Reemployment Programs in 2000

10

However, North Carolina’s Employment Service ( ES), particularly its Reem-ployment

Initiative staff, has been very successful in job placement and

employment services. Over the past year, over 770,000 individuals regis-tered

for employment services and 213,005 entered employment after receiving

ES services. During the same period, intensive reemployment assistance

was offered to approximately 67,000 claimants. For these efforts, the aver-age

length of weeks claimants file for UI before obtaining work is currently

9.2 weeks, the third lowest in the nation.

North Carolina’s ESC administrative staff, as well as ES and REI staff, have

one goal in mind: assisting unemployed workers find employment quickly.

An important result of these efforts is a low duration of filing, which provides

savings to the UI Trust Fund and the state’s employers.

1 Average duration is calculated by dividing the weeks compensated in the previous

12 months by the number of first payments in the same period.

North Carolina has the third

lowest average duration for UI

claimants in the nation at 9.2

weeks.

0

2

4

6

8

10

12

14

16

18

AL SC NH MS GA VT AR WI IA SD TN IN VA MI NE DE KY NC MO CT OK ND AZ KS MD

Average Duration Predicted Duration

Appendix 1: Average Duration in 2000 Compared with Predicted Duration for All States

0

2

4

6

8

10

12

14

16

18

ME UT LA ID CO MN FL NV WY TX OR OH RI MT WV NM NJ HI IL PA WA AK CA MA NY

Average Duration Predicted Duration

11

The Reemployment Outcomes of Dislocated

Manufacturing Workers

Recent adverse economic conditions have par-ticularly

impacted the manufacturing sector in

North Carolina. The events of September 11th

will likely contribute to this, as the decline in the

airline and tourist- related industries further re-duces

demand for manufactured products.

Although most of the recent layoffs in manufac-turing

seem to be temporary, there have been

several permanent layoff events in the state in

the last few months. In addition, the volatility in

manufacturing employment may encourage some

workers to voluntarily leave this industry group

in search of a more stable career.

As in most states, the share of manufacturing

employment has been declining in North Caro-lina

for decades. This has led to an absolute drop

in employment in this industry over the last few

years. Between August 1995 and August of this year, employment in manu-facturing

has steadily declined from 862,500 to 731,900. The majority of this

decline has been in the textile industry, where employment has fallen by

approximately 67,000 over this time period. Total employment in the state,

however, continued to rise, with most of the new jobs being reported in the

services and retail trade sectors. Industrial employment projections suggest

that this trend will continue for the near future.

Do dislocated manufacturing workers find jobs with comparable wages?

Wages in the manufacturing sector are somewhat above the state average.

In 1998, the average weekly wage in manufacturing, according to the Em-ployment

and Wages program of the Employment Security Commission ( ESC),

was $ 629.84, 17 percent above the state average wage across all industries.

At the same time, the services industry’s average wage was $ 504.88. Of

course, this is a crude comparison because it does not take into account the

experience or education of the individual workers who make up the employ-ment

in these industries. Such data, as well as anecdotal evidence, do not

give a satisfactorily in- depth answer to how well the dislocated workers do

at maintaining their standards of living.

In this article, we take a different approach by following workers who were

dislocated from their jobs in 1997 and 1998 in order to see how their indi-vidual

wages two years after the layoffs compare to their pre- layoff wages.

This provides a view of a large number of workers during that period, to see

where they actually found a job and how much they actually earned. This

information is pertinent to dislocated workers today.

Methodology: Tracking Laid- Off Workers

The North Carolina ESC has developed a data series which tracks workers

who have lost their jobs due to business closures or permanent layoffs. These

workers, and the companies which laid them off, are identified by the state’s

Manufacturing employment in

North Carolina has been

declining for decades.

In 1998, the average weekly wage

in manufacturing was $ 629.84,

17 percent above the state

average wage across all

industries.

640

680

720

760

800

840

880

1995 1996 1997 1998 1999 2000 2001

In Thousands

Figure 1: Manufacturing Employment in North Carolina

August 1995 - August 2001

12

Mass Layoff Statistics ( MLS) program. This program only identifies busi-nesses

with at least 50 claimants for Unemployment Insurance ( UI) in a

consecutive 5- week period. The data show the employment history and

wage earnings of these workers for four quarters before their separation and

eight quarters afterwards.

The data series answers many questions about laid- off workers, among them:

· In what industries will the laid- off workers find new jobs?

· How much will the laid- off workers earn in their new jobs compared

with their old jobs?

· How long will it take for the laid- off workers to find new jobs?

UI taxes are collected by ESC on a quarterly basis, with no reference to a

person’s starting date within the quarter. Therefore, only quarterly wage

data were given in this series. This presents some limitations because the

hours a given person works during a quarter are not known. For example, if

a worker’s wages are lower in one quarter than another, we cannot deter-mine

if this is caused by a lower hourly wage as opposed to fewer hours of

work. Such information could be obtained by a survey of dislocated work-ers,

but this would be costly and time- consuming.

To calculate the pre- layoff quarterly wage, only the wage data on the four

quarters before the layoff event occurred were used. The maximum quar-terly

wages of the first three quarters of the year were used in this study.

( The 4th quarter [ Oct. – Dec.] was excluded because wages are typically

higher in this quarter due to seasonal reasons, such as end- of- year bonuses

and holiday- related employment. Although earnings in the fourth quarter

should be considered when estimating a worker’s yearly income, we are

only interested in quarterly earnings.)

Post- layoff wages were limited to employers covered by UI in North Caro-lina.

Data could not be tabulated for workers who are self- employed or who

earned wages in a state other than North Carolina.

Table 1 shows the total number of layoffs and manufacturing layoffs for

each quarter during the time period under study — first three quarters of

1997 and 1998. These two years were chosen because they were the most

recent years for which complete data were available. They are somewhat

atypical because the unemployment rate in North Carolina during this period,

and the following two years, was consistently below 4 percent. The majority

of layoffs occurred in the 1st and 2nd quarters of each year.

Table 1: Mass Layoffs by Quarter

Year. Quarter Total Layoffs Manufacturing Layoffs

1997.1 3,077 2,588

1997.2 2,860 1,833

1997.3 1,590 1,436

1998.1 4,158 2,930

1998.2 3,554 2,974

1998.3 1,447 1,110

Lower wages in one quarter may

be a result of fewer hours of work.

Workers involved in this study

were followed four quarters before

layoff and eight quarters after

layoff.

Pre- and post- layoff wage

information were limited to

covered wages in North Carolina.

13

Of the 16,686 workers laid off, 12,871 ( 77 percent) were employed by com-panies

in the manufacturing sector. In some cases, individuals were laid off

more than once during this period. When this occurred, the later observa-tions

for those individuals were deleted. Also, workers who were laid off by

a company in the “ tobacco products” industry were excluded because these

companies tend to have seasonal layoffs. This left 9,405 individuals in the

database. Four industries were responsible for over three- quarters of these

layoffs: textiles, apparel, industrial and commercial machinery, including com-puter

equipment, and furniture.

Reemployment After a Layoff

Approximately 67 percent of the laid- off manufacturing workers found re-employment

within one year and 74 percent were reemployed at the end of

two years. As previously stated, these percentages did not reflect the num-ber

of workers who became self- employed, found jobs outside the state or

retired.

Where did these workers find jobs? The steady decline in the overall num-ber

of manufacturing jobs in the state limits the possibility of reemployment in

these industries. The opportunities available for workers would be deter-mined

by the overall health of the economy and the availability of jobs in the

local area that pay enough to ensure a decent standard of living.

Table 2 shows the industries in which the laid off manufacturing workers

were employed two years after the lay- off event. Many of these workers

who returned to work were able to find jobs in the manufacturing sector. Of

the 44 percent reemployed within the manufacturing industries, 16 percent

were reemployed by their former employer and another 28 percent were

reemployed by a company within the same industry. As expected, a substan-tial

fraction were reemployed in either the services sector ( 24 percent) or the

Prior to analysis, adjustments

were made to the data to account

for anomilies.

Forty- four percent of those

workers who returned to work

found jobs in the manufacturing

industries.

retail trade sector ( 9 percent).

Table 2: Primary Industries of Reemployed Manufacturing Workers Two Years after Separation*

Industry Divisions Percent

Same industry** 19

Manufacturing ( other than same industry) 25

Services 24

Retail Trade 9

Wholesale Trade 4

Construction 3

Government 2

Transportation, Communications and Public Utilities 2

Finance, Insurance and Real Estate 1

Agriculture 1

Industry Code not available 10

* These numbers do not include quarters 1997.1 and 1997.2 due to a large proportion of missing industry codes.

* * Thirty- seven percent of these were reemployed by the same company that laid them off.

14

The reemployment opportunities in the manufacturing sector allowed many workers to find jobs where the skills

they had learned over many years could be used effectively. In the next section, we see that this helps those

workers who find manufacturing jobs retain their relatively high wages. But what happens to the wages of

workers who are unable to return to a manufacturing job?

Wages Earned After a Layoff

The median pre- layoff quarterly wages of

the workers was $ 4,895. One year after

the layoff event, the median wages of the

workers who were reemployed was $ 3,781,

a 23 percent decrease. By the end of two

years, median quarterly wages had risen to

$ 4,329, which was still 12 percent lower

than the wages of this group before the lay-off.

( The median pre- layoff wages of the

workers who were employed two years

after the layoff were slightly higher than

the median wages of all the workers who

were laid off.)

Figure 2 compares the median wages of

workers two years after the layoff to their

pre- layoff wages, based on the industry in which the workers were reemployed. It is important to note that

reemployment occurred mainly in the manufacturing, services and retail trade sectors. ( Refer to Table 2.)

The median wage of the workers who were reemployed in manufacturing was approximately the same as the

median wage of this group before the layoff. Those reemployed by the same company or in the same sub-industry

tended to do slightly better than those who found employment in a different manufacturing industry.

Those who found jobs in the services and retail trade industries faced significant reductions in their quarterly

wages. The median wage for the workers reemployed in the retail trade industry was only 63 percent of the pre-layoff

median wage for this group. It should be noted that the workers who went into the services or retail trade

sectors had lower pre- layoff median wages than the other workers. This may suggest that they had, on average,

lower skills or less tenure than other workers.

Another way of looking at the change in

wages before the layoff and afterwards

is to look at the distribution of each

individual’s change in wages. This

analysis is summarized in Figure 3. Two

years after the layoff event, 53 percent

of the workers who were working earned

less than 90 percent of their pre- layoff

wages; 18 percent earned within 10 per-cent

of their pre- layoff wages; and 29

percent earned more than 110 percent

of their pre- layoff wages. Furthermore,

20 percent earned less than half their

pre- layoff wages and 2 percent earned

more than twice as much.

87%

88%

87%

94%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

Less than 9 years 9 - 11 years 12 years 13 - 16 years

Median Pre- Layoff Wages Median Wages Two Years after Layoff

Figure 2: Quarterly Median wages of Reemployed Manufacturing

Workers Two Years After Layoff

53%

18%

29%

Earned Less Than 90%

Earned Within 10%

Earned More Than 110%

Figure 3: Wages Earned Two Years After Layoff as a Percentage of

Pre- Layoff Wages

15

One may assume that workers with higher wages before the layoff did better than lower- income workers.

Presumably, workers with higher incomes have more job skills, gained through a combination of education and

on- the- job training. Many of these skills can be transferred into new careers, allowing the workers to continue

earning relatively high wages. However, some of the skills may not be transferable. The returns to these job-specific

skills could be lost when workers move into other occupations.

As previously stated, workers with

higher quarterly wages before a lay-off

earned higher wages two years

afterwards than workers with lower

pre- layoff wages. However, Figure 4

shows that the higher- paid workers did

not do as well at replacing their wages.

Workers in the highest income group

had the lowest wage replacement rate

( 58 percent).

Also, only 67 percent of the workers

in the highest wage category were re-employed

by a company in the state

two years after the layoff, which is sig-nificantly

lower than the reemployment

rates of the other groups. It is likely

that a larger fraction of these workers

sought employment outside the state or became self- employed. However, as noted earlier, the data do not

contain information on the wages of these workers.

Differences by Age Group

One might expect that workers in different age brackets would have different experiences after a layoff. Young

workers were more likely to do well after being laid off than older workers because they had put less of an

investment into their jobs and had a greater incentive to move to a different job in order to develop new skills.

Older workers had invested a lot of their time developing skills which were specific to their jobs and, therefore,

would likely find it harder to get new jobs outside their field that paid as well.

Figure 5 shows the median pre- layoff

wages of workers compared to their

wages two years after the layoff event,

where workers were separated into age

groups. The workers under the age of

30 earned 90 percent of their pre- layoff

wage, while workers in the 55 and over

age group earned only 78 percent of their

pre- layoff wage. The median wage in

the middle two age groups was approxi-mately

87 percent of the pre- layoff

median wage. Older workers were also

less likely to re- enter the job market.

Only 52 percent of the workers aged 55

and over had returned to work after two

years, compared to approximately 77

percent of the workers in the other age

groups.

58%

73%

86%

93%

117%

$ 0

$ 2,000

$ 4,000

$ 6,000

$ 8,000

$ 10,000

$ 12,000

$ 0-

$ 2,999

$ 3,000 -

$ 4,999

$ 5,000 -

$ 6,999

$ 7,000 -

$ 9,999

$ 10,000 -

$ 14,999

Pre- Layoff Wages

Median Pre- Layoff Wage Median Wage Two Years after Layoff

Figure 4: Quarterly Median Wages of Manufacturing Workers by Pre- Layoff

Wage Categories

90% 87% 87%

78%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

Under Age 30 Age 30 to 44 Age 45 to 54 Age 55 and Over

Median Pre- Layoff Wage Median Wages Two Years after Layoff

Figure 5: Median Wages Two Years After Layoff Compared to Pre- Layoff

Wages - by Age Group

16

Differences by Educational Level

Education is an important indicator of earnings. More educated workers typically have greater opportunities for

employment and earn higher wages. Figure 6 compares the pre- layoff wages to wages earned two years after

the layoff by educational level. Both before and after the layoff, median wages increased as the level of educa-tion

rose. However, workers in manufacturing who have more years of education are challenged to find jobs

which allow them to sustain their relatively higher standard of living; many of them may possess firm- specific

skills which will not be compen-sated

by a different employer.

The group of workers with 12 years

of education made up 55 percent

of the sample. The median wage

for this group two years after the

layoff was 88 percent of the pre-layoff

wage. The median wage of

the workers with less than nine

years of education was 94 percent

of the pre- layoff median wage for

this group. However, only 58 per-cent

of this group were reemployed

two years after their layoff. This

might suggest that many of the large

number of older workers in this

group withdrew from the labor mar-ket

to retire or return to school for

further training.

Differences by Industry

As stated earlier, there were

four manufacturing industries

which contributed the majority

of the layoffs in our database:

textiles, apparel, industrial and

commercial machinery, includ-ing

computer equipment, and

furniture. These, notably, are

also the industries that seem to

have been particularly hard hit

during the current economic

downturn. Figure 7 shows the

median wages in each of these

industries before a layoff and

two years afterwards for laid-off

workers who found

reemployment in the state.

Median wages were highest in

the textile and industrial machin-ery

industries before the layoff,

but workers in the furniture in-dustry

did relatively better after

the layoff.

87%

88%

87%

94%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

Less than 9 years 9 - 11 years 12 years 13 - 16 years

Median Pre- Layoff Wages Median Wages Two Years after Layoff

Figure 6: Median Wages Two Years After Layoff Compared to Pre- Layoff

Wages - by Educational Levels

94%

106%

88%

84%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

Textiles Apparel Furniture Industrial and

Commercial

Machinery

Median Pre- Layoff Wages Median Wages Two Years after Layoff

Figure 7: Median Wages Two Years After Layoff Compared to Pre- Layoff

Wages - by Industry

17

Workers in the furniture industry were much more likely to find reemployment with other companies within the

same industry than workers in the other industries. This may be due to the geographic concentration of the

companies that manufacture furniture. Workers in the industrial machinery industry were most likely to be

reemployed in the services sector of the economy. However, most of these service jobs were in the “ help supply

services,” which includes many high- tech contract workers.

Differences by Race and Gender

Figure 8 shows the median wages

of the workers who were laid off

by gender. There are significant dif-ferences

between the median wages

of men and women before the lay-offs.

Much of this difference was

attributable to the fact that a larger

percentage of the women were em-ployed

in the low- paying apparel

industry ( 20 percent, compared to 8

percent of men). After two years,

the median wage of the women was

85 percent of the median pre- layoff

wage, while the comparable figure

for men was 89 percent. The re-employment

rates between men and

women were approximately equal.

As illustrated in Figure 9, there was not a significant difference in the median pre- layoff wage of workers among

the racial groups. The median wage for whites two years after the layoff was 86 percent of the pre- layoff

median wage for this group, compared to 88 percent for blacks and 92 percent for the other racial groups

combined ( Native Americans, Asians and Hispanics). The only significant difference was in the reemployment

rates: approximately 78 percent of blacks had returned to a job within the state, compared to 71 percent of

whites.

Conclusion

Recent dislocations in the manufac-turing

sector of the North Carolina

economy have led to great concern

about the future living standards of

those workers who have been laid

off. This study has looked at manu-facturing

workers who were laid off

in either 1997 or 1998, and who filed

for UI benefits, to identify their re-employment

experiences related to

where they achieved reemployment

and how much they earned as a per-centage

of their earnings before the

layoff ( that is, their wage replace-ment

rates).

85%

89%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

Men Women

Median Pre- Layoff Wage Median Wage Two Years after Layoff

Figure 8: A Comparison of Median Pre- Layoff Wage to Median Wage Two Years

After Layoff - by Gender

92%

88%

86%

$ 0

$ 1,000

$ 2,000

$ 3,000

$ 4,000

$ 5,000

$ 6,000

White Black Other

Median Pre- Layoff Wages Median Wages Two Years after Layoff

Figure 9: Median Wages Two Years After Layoff Compared to Pre- Layoff

Wages - by Race

18

Overall, 74 percent of the dislocated manufacturing workers were reem-ployed

in the state after two years. The median wage of these workers was

88 percent of the median pre- layoff wage. The workers who were reem-ployed

in the manufacturing sector were able to retain their pre- layoff

standards of living, based on the median wages of this group. The workers

who fared the worst were reemployed in either the services or retail trade

industries. Unfortunately, a large fraction of workers ( at least one- third) fell

into these categories.

This study also compared the wage replacement rates of workers in differ-ent

demographic groups. Among the findings: ( 1) lower- paid workers had

higher replacement rates than high- paid workers, ( 2) workers under the age

of 30 did better than older workers, ( 3) workers with less than nine years of

education had the highest replacement rate, ( 4) men did slightly better than

women, and ( 5) whites did slightly less well than blacks or other races. These

results are based on median wages; individual outcomes may vary.

The results of this study indicate that in an economy where workers are

expected to move from the manufacturing sector to service- related indus-tries,

more must be done to ensure that the workers’ standards of living are

maintained. Among the measures that might be appropriate are skill training

in service industry occupations, educational advancements and specialized

job placement services.

Those reemployed within the

manufacturing sector fared better

than those reemployed within the

services or retail trade industries.

In moving from manufacturing to

service industry employment,

workers may need specialized

assistance.

Internet Usage in Filing Claims

Offering those laid off from their jobs the ability to file initial

claims via the Internet is an example of ESC's commitment to

the convenience of its customers. Internet filing of initial claims

was first offered to claimants in October 2000. During that pe-riod,

43 initial claims were submitted. This has grown to 2,512 in

August 2001.

LMI Happenings: New Research and Products from the LMI Division

Changes in Occupational Employment

Statistics

In an effort to provide more timely occupational

employment by industry and occupational wage

information, the Occupational Employment

Statistical ( OES) survey will shift from being

annual to biannual starting with the 2001

survey. North Carolina is one of five states

that volunteered to test the two panel concept.

Those normally surveyed annually, 13,000

employers representing 1.1 million employed,

will be divided equally between the two panels.

Each panel will still include the basic

components and requirements of a yearly

survey. The Bureau of Labor Statistics, who

draws North Carolina’s sample and oversees

its management, will continue to receive

monthly reports as to the progress. The North

Carolina OES staff’s mission as a volunteer

state is to document, particularly logistically,

any problems and to troubleshoot any

unforeseen occurrences. Being a test state

enables North Carolina to “ wet its feet” before

the rest of the nation undertakes the new two

panel format.

0

500

1,000

1,500

2,000

2,500

3,000

Oct.

2000

Nov.

2000

Dec.

2000

Jan.

2001

Feb.

2001

Mar.

2001

Apr.

2001

May

2001

Jun.

2001

Jul.

2001

Aug.

2001

Submitted Internet Initial Claims

19

2000- 2001 1999- 2000

New Vehicle Registrations Housing Units Authorized by Building Permits

Unadjusted Construction Employment, in Thousands New Business Incorporations

Personal Income Tax Revenues, in Millions

Economic Indicators in North Carolina ( Continued from Page 2)

Total Tax Revenues, in Millions

Source: US Census Bureau

Source: NC Department of Revenue

Source: NC Secretary of State, Corporations Division

Source: NC Automobile Dealers Association

Source: ESC, Labor Market Information Division

Source: NC Department of Revenue

0

10,000

20,000

30,000

40,000

50,000

60,000

S O N D J F M A M J J A

0

2,000

4,000

6,000

8,000

10,000

S O N D J F M A M J J A

200

210

220

230

240

250

260

S O N D J F M A M J J A

0

1,000

2,000

3,000

4,000

5,000

6,000

S O N D J F M A M J J A

$ 0

$ 500

$ 1,000

$ 1,500

$ 2,000

S O N D J F M A M J J A

$ 0

$ 200

$ 400

$ 600

$ 800

$ 1,000

$ 1,200

$ 1,400

S O N D J F M A M J J A

20

750 copies of this publication were

produced at a cost of $ 225.00 or

$ 0.30 per copy.

EMPLOYMENT SECURITY COMMISSION OF NORTH CAROLINA

www. ncesc. com

NCESC 6059 October 2001

LABOR MARKET INFORMATION DIVISION

EMPLOYMENT SECURITY COMMISSION OF NORTH CAROLINA

POST OFFICE BOX 25903

RALEIGH, NORTH CAROLINA 27611- 5903

OFFICIAL BUSINESS

PENALTY FOR PRIVATE USE $ 300.00

FIRST CLASS MAIL

POSTAGE AND FEES PAID

U. S. DEPARTMENT OF LABOR

PERMIT NO. G- 12