Quarterly Workforce Indicators: Case Studies and Examples C2ER Training Workshop June 4, 2012 Erika McEntarfer LEHD Program US Census Bureau
Feb 24, 2016
Quarterly Workforce Indicators:Case Studies and Examples
C2ER Training WorkshopJune 4, 2012
Erika McEntarferLEHD Program
US Census Bureau
2
In this section:
• Apply knowledge about basic employment and wage concepts in QWI to specific questions you may encounter in your work
• Specifically:– Smoothing seasonal data– Calculating rates– Producing custom aggregates
3
Smoothing seasonal indicatorsHiring in California, 1993-2011: Not Seasonally Adjusted
1993.11994.1
1995.11996.1
1997.11998.1
1999.12000.1
2001.12002.1
2003.12004.1
2005.12006.1
2007.12008.1
2009.12010.1
2011.11500000
2000000
2500000
3000000
3500000
4000000
4500000
Hires - NSA
Source: Quarterly Workforce Indicators, US Census Bureau
• QWI currently doesn’t generate a seasonally adjusted series.
• Hard to see cyclical trends with all the seasonality.
4
Seasonal adjustment: Options
• Annualize the data– Easier for some indicators than others
• Take rolling averages – Easy, but crude (available in QWI online)
• Do your own seasonal adjustment– Best option– X12 (SAS, others)– Excel seasonal adjustment module
5
Smoothing seasonal indicatorsHiring in California, 1993-2011: Seasonally Adjusted
Source: Quarterly Workforce Indicators, US Census Bureau
• This series is adjusted using X12 in SAS.
• Much easier now to see cyclical trends and graphs look much cleaner.
1993.11994.1
1995.11996.1
1997.11998.1
1999.12000.1
2001.12002.1
2003.12004.1
2005.12006.1
2007.12008.1
2009.12010.1
2011.11500000
2000000
2500000
3000000
3500000
4000000
4500000
Hires - NSA Hires - SA
6
Constructing rates
• Separations, Accessions, Job Creation, etc. all very useful statistics, – but often more meaningful expressed as rates
• Because there are several types of hires, separations, and employment indictors, it’s not always clear how to construct simple rates.
7
Constructing an accession rate
• Hires: several options– Hires– New Hires– Recalls– Stable Hires
• Employment: several options– Beginning of Quarter
Employment – End of Quarter
Employment– Flow Employment– Stable Employment
8
Constructing an accession rate
Date Accessions Beginning of Qtr Emp End of Qtr Emp Flow Employment1993.1 2541912 11708810 11738008 143017581993.2 2691719 11717995 11790650 143914281993.3 2680001 11719405 11781341 143942721993.4 2720066 11752412 11795646 144626761994.1 2719362 11786936 11841965 145375011994.2 2856576 11792267 11890286 146373251994.3 2923723 11881980 11949629 14812717
Hiring and Employment in CA, Seasonally Adjusted Accessions (A): -- all hires in a quarter, regardless of length of employment spellFlow employment (M): -- all persons who had positive wages during the quarter, typically much larger than point in time employment estimatesB & E employment: -- point in time estimates of employment at start and end of quarter.
Accessions, particularly in small, high turnover firms, can exceed point in time employment -- so A/(B+E)*1/2 can be greater than 100% -- A/M is bounded by 100%
9
Constructing a separation rate
1993.11994.1
1995.11996.1
1997.11998.1
1999.12000.1
2001.12002.1
2003.12004.1
2005.12006.1
2007.12008.1
2009.12010.1
2011.10.100
0.120
0.140
0.160
0.180
0.200
0.220
0.240
0.260
Separations/Ave Emp Separations/Flow Employment
Job separation rate in California, 1993-2011: Seasonally Adjusted
Source: Authors calculations from the Quarterly Workforce Indicators, US Census Bureau
• What is true for accessions is also true for separations
• While either choice is valid, using flow employment does benchmark better to other series such as JOLTS.
Recommended
10
Hiring vs. Job Creation
• Often, we are interested in both hiring, job creation, and net job flows:– Hires: growth hires and replacement hires– Job Creation: growth hires only– Net job flows: Job Creation – Job
Destruction, or net employment change
11
Hiring vs. Job Creation
1993.11994.2
1995.31996.4
1998.11999.2
2000.32001.4
2003.12004.2
2005.32006.4
2008.12009.2
2010.30
500000
1000000
1500000
2000000
2500000
3000000
3500000
4000000
4500000
0.6
0.62
0.64
0.66
0.68
0.7
0.72
0.74
0.76
0.78
Job Creation AccessionsShare that are Replacement Hires
Source: Authors calculations from the Quarterly Workforce Indicators, US Census Bureau
Hiring and Expansionary Hiring in California, 1993-2011: Seasonally Adjusted
Note: All Hires are more cyclical than expansionary hiring – employment churn is procyclical
Can calculate the share of all hires that are replacement hires (A/JC).Note replacement hiring falls much more steeply in the Great Recession -- workers either not separating from jobs -- or employers leaving vacancies unfilled
12
Exercise 1: Examine Hiring Patterns in California (10 minutes)
• Calculate hires as a share of employment– Use both flow employment and average
employment, why are they different?• Calculate the share of hires in CA that are
expansionary
13
Exercise 1:
1993.11993.4
1994.31995.2
1996.11996.4
1997.31998.2
1999.11999.4
2000.32001.2
2002.12002.4
2003.32004.2
2005.12005.4
2006.32007.2
2008.12008.4
2009.32010.2
2011.10.100
0.120
0.140
0.160
0.180
0.200
0.220
0.240
0.260
0.280
0.300
Job Hiring Rate - A/M Job Hiring Rate - A/average emp
Job hiring rate in California, 1993-2011: Seasonally Adjusted
14
Exercise 1:
1993.11993.4
1994.31995.2
1996.11996.4
1997.31998.2
1999.11999.4
2000.32001.2
2002.12002.4
2003.32004.2
2005.12005.4
2006.32007.2
2008.12008.4
2009.32010.2
2011.10.15
0.17
0.19
0.21
0.23
0.25
0.27
0.29
0.31
0.33
JC/A
Share of Hires in California that are Expansions in Firm Employment, 1993-2011: Seasonally Adjusted
15
Exercise 2: Comparing Separation Rates Within a sector (10 minutes)• Health Care is often thought of as a high turnover sector,
but there’s quite a bit of heterogeneity in turnover within health care
• Calculate worker separation rates using your preferred measure for:
• Ambulatory Health Care (Physicians offices, clinics)
• Hospitals• Nursing Facilities
16
Exercise 2: Comparing Separation Rates Within a sector
1993.11994.1
1995.11996.1
1997.11998.1
1999.12000.1
2001.12002.1
2003.12004.1
2005.12006.1
2007.12008.1
2009.12010.1
2011.10.00
0.05
0.10
0.15
0.20
0.25
NAICS 623 -- Nursing and Residential Care Facil-itiesNAICS 621 - Ambulatory Health Care ServicesNAICS 622 -- Hospitals
17
Creating custom aggregations
• QWI are available by: – Worker age, education, gender, race– Detailed Industry– Detailed Geography
• But often want to create custom aggregations of available categories– Older workers– Industry Clusters– Etc.
18
Be careful when aggregating
• Employment and net job flows fairly straight-forward– Simply aggregate them
across categories• However:
– Because of noise infusion and suppressions, be cautious when aggregating small cells
– Always use tabulated aggregation if available
• Earnings and nonemployment more complicated– Should compute weighted
averages using the appropriate employment number (stable for stable wages, etc)
Job Creation and Destruction:Most Common Aggregation Error
• Note that for categories like age and sex, the published net job flows for the subcategories will sum to the margin
• But for gross Job Creation and gross Job Destruction this is not true
• (Job Creation for men) + (Job Creation for women) does not equal (total Job Creation)– For example, a job could be created at a firm and filled by a woman,
while another job at the same firm is destroyed, previously filled by a man Men Women Total
Job Creation 0 1 0
Job Destruction 1 0 0
Net Job Flows -1 +1 0
19
20
Exercise 3: Younger Workers (10 minutes)
• Graph the share of workers under 25 in California over the time series.
• Calculate and graph growth trends in average nominal earnings for workers under 25 in California, relative to those for all workers in California.
21
Exercise 3: Younger Workers
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 731000000
1200000
1400000
1600000
1800000
2000000
2200000
2400000
2600000
0.12
0.125
0.13
0.135
0.14
0.145
0.15
0.155
0.16
0.165
Workforce < 25 years of age Share of workforce < 25 years of age
Workers under 25 as a share of the California workforce, 1993-2011: Not Seasonally Adjusted
Great Recession impacted share of young workers in market quite severely, is at almost 20 year low.
22
Exercise 3: Younger Workers
1993.11994.1
1995.11996.1
1997.11998.1
1999.12000.1
2001.12002.1
2003.12004.1
2005.12006.1
2007.12008.1
2009.12010.1
2011.11
1.2
1.4
1.6
1.8
2
2.2
Growth in average monthly wages, all workers Growth in average wages, workers < 25
Growth Average Nominal Monthly Wages, Workers < 25, California workforce, 1993-2011: Not Seasonally Adjusted (1993:1=1)
Around 2007, wage growth for young workers stalls out, even falls
23
To sum up
• While many QWI indicators can be used as is, frequently they require manipulation to produce the information needed
• These exercises show how to:– Handle seasonality– Construct rates– Create custom aggregates