PMP Summit 2014 Replacements

Report
Proposed Changes to BLS
Replacement Rate
Methodology
Michael Wolf
Employment Projections Program
PMP Summit
June 3, 2014
Overview
Proposed Methodology
 Advantages of New Method
 Analysis of Results
 Issues and Future Considerations

2
Proposed Methodology
3
Proposed Method
Survey data are used to measure labor
force exits and occupational transfers
 Regression models are used to estimate
projected rates of separations for each
occupation
 Projected rates are used to calculate
projections of separations

4
Labor Force Exits
CPS measures individual household
members for 4 months, then for 4
months again after an 8 month break
 If a matched worker was in the labor
force for the first 4 months, and out of
the labor force for the last 4, they are
counted as having exited the labor
force

5
Labor Force Exits

Records where individuals moved in
and out of the labor force within either
4 month period are excluded from
analysis
Intended to remove temporary leavers
and those marginally attached to the labor
force
 Also removes noise from the data

6
Occupational Transfers
CPS ASEC (March) supplement asks
about work activity in prior calendar
year
 Respondents are asked whether their
•Iflongest job in the prior year was the
same as their current job
 If not, the occupation of their longest
job from the prior year is coded

7
Occupational Transfers
Why ASEC supplement?
 Monthly CPS occupation data is only
dependently coded for months 2,3,4
and 6,7,8 – values a year apart are
•If independently coded
 January supplement also asks about
occupation from prior year, but is
more seasonally affected, and also is
only conducted biennially

8
Occupational Transfers
If respondents were employed in a
different SOC major group in the prior
year, they are considered to have
transferred from their prior occupation
•If
 Major group transfers used because

Major group coding is more reliable
 Eliminates transfers between comparable
occupations, like retail salespersons and
parts salespersons

9
Comparison – data use
Cohort method takes difference
between estimates of employment for a
specific occupation by age cohort at
two points in time
•If
 Alternate method uses longitudinal
aspects of CPS to directly measure
workers who separate from the labor
force or transfer occupations

10
Comparison – concepts
Cohort method measures separations
net of entrants for each cohort – some
separations are not measured
 Labor force method measures leavers
•If
who are out of the labor force for a full
4 months
 Occupational transfers method
measures workers who change SOC
major occupational groups

11
Advantages of New
Method
12
Direct Measure

Longitudinal survey data identifies
individual workers who leave an
occupation

Old method uses panel data and
demographics to estimate the number
of workers who leave
13
Differentiation of leavers
Occupation
Postal Service Clerks
Management Analysts
Occupational Transfer Labor Force
Rate
Exit Rate
35.9%
44.3%
47.3%
27.0%
Clearer for users what is being
measured
 Still cannot directly estimate
retirements

14
Robust methods
Regression uses more information
 Explanatory variables include:

Age
Sex
Educational attainment
Economic conditions
15
Robust methods

Projections incorporated consistently
Old:
Employment Replacement
Job Openings due to Growth
Change
Needs
and Replacement Needs
2,600
3,800
6,400
-5,800
5,000
5,000
New:
Openings due to Employment
Employment
Occupational
Change and Occupational
Change
Separations
Separations
2,400
17,200
19,600
-5,800
19,400
13,600 16
More reliable results

Method works for new SOC occupations
Occupation Title
Occupational
Replacement Rate Separation Rate
Computer systems analysts
15.7%
54.5%
Information security analysts
15.7%
54.1%
Web developers
15.7%
63.7%
Fewer proxies for small occupations
 All together, data available for 118
additional occupations

17
More reliable results

Consistency year to year
2010-20
2012-22
Replacement Rate Replacement Rate
60.4
28.4
Animal Trainers
9.7
44.2
Budget Analysts
21.2
40.1
Dredge Operators
29.6
8.0
Actuaries

130 occupations had rates that changed
by at least 10 percentage points
18
State-based results
Demographics of occupations vary
state-to-state
 Regression coefficients must be based
on national data
 Can be applied to state-based
demographics

19
Analysis of Results
20
Total Job Openings
2012 employment of 145 million
projected to increase to 161 million by
2022.
 Projected job openings, 2012-22:

16 million openings due to growth
155 million separations
– 65 million labor force leavers
– 90 million occupational transfers
171 million openings in total
21
Total Job Openings

What does 171 million openings mean?
Equivalent to every current worker either
leaving the labor force or changing
occupations once, on average, over the
next 10 years
JOLTS shows 558 million hires, 553 million
separations, from 2003-2012
BLS projects 35 million labor force
entrants from 2012-2022
22
Example - Teachers
BLS Projections
Preschool, primary,
secondary, and
special education
school teachers
NCES Projections
Elementary and
Secondary teachers
Job
Openings due
Openings
to
due to
Employment
Growth and Change and
Employment Replacement Occupational
2012
2022
Change,
Needs,
Separations,
Employment Employment 2012-22
2012-22
2012-22
4,115,300
2012
Employment
3,511,000
4,544,700
429,400
1,401,200
2022
Employment
Employment Change, 2012-22
3,931,000
420,000
3,537,600
New Hires,
2013-22
4,201,000
23
Example - Lawyers
Openings due to
Job Openings due to Employment Change
Growth and
and Occupational
Replacement Needs, Separations, 20122012-22
22
Lawyers

196,500
414,600
ABA reports 29,000 law school graduates
find employment in positions requiring bar
passage on average from 2011-2013
24
Example – Registered Nurses
Openings due to
Job Openings due to Employment Change
Growth and
and Occupational
Replacement Needs, Separations, 20122012-22
22
Registered nurses

1,052,600
1,853,700
NCSBN reports an average of 211,000
newly licensed registered nurses per
year, 2009-2011
25
Issues and Future
Considerations
26
Issues

Who will fill these openings?
Old/young
New/experienced
Education level/degree
Foreign/domestic
27
Issues

Communications with data users
New methods
New terminology
Significantly different results
28
Next Steps
Gather feedback through Summer 2014
 Review and modify methods, if
necessary
 Final decision by end of calendar 2014
 Next projections published December
2015

29
Contact Information
Michael Wolf
Branch Chief
Employment Projections Program
www.bls.gov/emp
202-691-5714
[email protected]

similar documents