37
The Decline of Men Working: Why It Is Happening, What It Means, and What to Do About It Jason Furman Senior Fellow, PIIE Peterson Institute for International Economics | 1750 Massachusetts Ave., NW | Washington, DC 20036 University of California, San Diego San Diego, CA May 2, 2017

Presentation: The Decline of Men Working: Why It Is ... · Other Labor Market Dysfunctions: Big Increase in Long-term Unemployment After the Recession. I. Labor Market Context. Source:

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

The Decline of Men Working: Why It Is Happening, What It Means, and What

to Do About ItJason FurmanSenior Fellow, PIIE

Peterson Institute for International Economics | 1750 Massachusetts Ave., NW | Washington, DC 20036

University of California, San DiegoSan Diego, CA

May 2, 2017

Outline of the Talk

I. Labor Market Context

II. Facts About Prime-age Men

III.Explanationsa) Supplyb) Demandc) Institutions

IV.Policy Response

Sharp Decline in Unemployment Rate…

I. Labor Market Context

Source: Bureau of Labor Statistics; author's calculations.

Average, 2001-2007

0

2

4

6

8

10

12

2007 2009 2011 2013 2015 2017

Unemployment RatePercent

Mar-17

…But Employment-Population Rate Has Not Recovered

I. Labor Market Context

Source: Bureau of Labor Statistics; author's calculations.

Average, 2001-2007

55

56

57

58

59

60

61

62

63

64

65

2007 2009 2011 2013 2015 2017

Employment-Population RatioPercent

Mar-17

Decline Exceeds Age Effects

I. Labor Market Context

Source: Bureau of Labor Statistics; author's calculations.

Average, 2001-2007

55

56

57

58

59

60

61

62

63

64

65

2007 2009 2011 2013 2015 2017

Employment-Population RatioPercent

Mar-17

Age Trend

Employment-Population Ratio

Other Labor Market Dysfunctions: Big Increase in Long-term Unemployment After the Recession

I. Labor Market Context

Source: Bureau of Labor Statistics; author's calculations.

Unemployment Rate

Long-term Unemployment Rate (27+ Weeks)Average,

2001-2007

Mar-17

0

2

4

6

8

10

12

1948 1958 1968 1978 1988 1998 2008

Unemployment Rate by DurationPercent

And Also People Working Part-time for Economic Reasons

I. Labor Market Context

Source: Bureau of Labor Statistics; author's calculations.

Average, 2001-2007

Mar-17

0

1

2

3

4

5

6

7

1955 1965 1975 1985 1995 2005 2015

Part-time for Economic ReasonsPercent of Labor Force

Men’s Participation Rate Has Been Declining Since the 1950s

II. Facts about Prime-age Men

Source: Bureau of Labor Statistics; author's calculations.

1948-2007 Trend

86

88

90

92

94

96

98

100

1948 1958 1968 1978 1988 1998 2008

Prime-Age Male Labor Force Participation RatePercent

Mar-17

Non-participation Is Not Simply Churn

II. Facts about Prime-age Men

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

2016

65

70

75

80

85

90

1988 1992 1996 2000 2004 2008 2012 2016

Share of Prime-Age Men Currently Not in the Labor Force Who Did Not Work Last Year

Percent

Participation Falling for Every Cohort

II. Facts about Prime-age Men

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

1933-19421943-1952

1953-1962

1963-1972

82

84

86

88

90

92

94

96

98

100

25 27 29 31 33 35 37 39 41 43 45 47 49 51 53

Prime-Age Male Labor Force Paricipation Over the Lifecycle by Birth Cohort

Percent

1973-1982

1983-1992

Age

Education an Increasingly Large Factor

II. Facts about Prime-age Men

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

High School or Less

Some College

Bachelor's or More 2016

80

82

84

86

88

90

92

94

96

98

100

1964 1974 1984 1994 2004 2014

Prime-Age Male Labor Force Participation by Educational Attainment

Percent

Share With Spouse in the Labor Force is Low and Declining

III. Explanations - Supply

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

Overall

2016

Not in Labor Force

0

5

10

15

20

25

30

35

40

45

50

1960 1970 1980 1990 2000 2010

Share of Prime-Age Men with Spouse in Labor ForcePercent

Time Spent on Responsibilities Not Up Much/At All; Television Up a Lot

III. Explanations - Supply

Note: Columns do not sum to 1,440 minutes per day because some individuals do not report all minutes of each day.Source: American Time Use Survey, ATUS-X; author's calculations.

Time Use Category Prime-Age MenPrime-Age Men,

Not in the Labor ForceCaring for Household Members 27 28Caring for Non-Household Members 5 12Education 7 27Household Activities and Services 80 113Socializing, Relaxing, Leisure 245 415

Watching Television 152 283Work 318 9Other (Including Sleep) 750 830

Time Spent (Minutes Per Day)

Time Use Breakdown in 2015 for Prime-Age Men Overall and Nonparticipating Prime-Age Men

Social Security (Disability) Is the Largest Government Income Source

III. Explanations - Supply

Note: Includes government income assigned to any household member. Other includes worker's comp, unemployment insurance, welfare, and survivor's benefits.Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

Prime-Age Men Prime-Age Men NILF

OtherVeteran's BenefitsDisability BenefitsSSISocial Security

Breakdown of Average Prime-Age Male Household Income from Government Sources in 2015

2015 Dollars

Disability Cannot Explain the Majority of the Decline in Prime-age Male LFPR

III. Explanations - Supply

The labor force participation rate for prime age men fell 8.4 percentage points from 1967 to 2015. At the same time, Social Security disability receipt increased by 1.5 percentage points for this group.

This is an upper bound for three reasons:1. Part of SSDI increase is shift towards older

population. This portion does not explain the age-adjusted decline in LFPR.

2. Many men receiving SSDI would not have participated anyway.

3. Reduction in LFPR can cause SSDI receipt.

Plausible Estimates are SSDI Accounts for 0.1 to 0.4 p.p. Reduction in LFPR

III. Explanations - Supply

Note: Participation rates for non-SSDI recipient disabled are age-adjusted sing a linear probability model. Gray counterfactual based on French and Song (2014).Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

Participation Rate (left axis)

Counterfactual 1: SSDI Recipients Participate Like

Non-SSDI Disabled (left axis)

Counterfactual 2: SSDI Recipients Participate at the Rate Predicted

by Causal Estimate (left axis)

2015

Share Receiving SSDI (right axis)0

5

10

15

20

80

85

90

95

100

1967 1972 1977 1982 1987 1992 1997 2002 2007 2012

Possible Effects of Social Security Disability Insurance (SSDI) on Prime-Age Male Labor Force Participation

Percent

Other Government Cash Benefits Have Fallen

III. Explanations - Supply

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

Overall

Not in Labor Force

2015

0

10

20

30

40

50

60

70

1965 1975 1985 1995 2005 2015

Share of Prime-Age Men Receiving Government Cash Transfers Other Than Social Security

Percent

As a Result Overall Government Cash Benefit Incidence Has Fallen

III. Explanations - Supply

Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

Overall

Not in Labor Force

2015

0

10

20

30

40

50

60

70

1965 1975 1985 1995 2005 2015

Share of Prime-Age Men Receiving Any Government Cash Transfers

Percent

Supply vs. Demand

III. Explanations - Demand

Relative Wages Have Fallen—Which Is Consistent With a Decline in Demand

III. Explanations - Demand

Note: The earnings ratio compares the median full-time, full-year worker age 25 years or older with just a high school degree to the same type of worker with at least a bachelor's degree. Prior to 1992, high school graduate is defined as completed 12th grade and bachelor's degree is defined as four years of college.Source: Bureau of Labor Statistics, Current Population Survey (Annual Social and Economic Supplement), IPUMS; author's calculations.

2015

50

55

60

65

70

75

1975 1980 1985 1990 1995 2000 2005 2010 2015

Ratio of High School Graduate Wages to College Graduate Wages

Percent

Explanation #1: Increase in Incarceration

III. Explanations – Institutions

Source: Bureau of Justice Statistics.

2015

0

200

400

600

800

1,000

1,200

1990 1995 2000 2005 2010 2015

Male Incarceration Rate, 1990-2015Inmates per 100,000 Residents

Effects of Incarceration on LFPR

III. Explanations – Institutions

Currently in prison: 1.1 million in Federal or State prison not in numerator or denominator. To the degree that people in prison would have participated at a lower rate, including them would lower the measured participation rate—but plausibly only by a few tenths of a percentage point.

Formerly in prison: ~5 million formerly incarcerated (~7% of prime age men). They appear to participate at a substantially lower rate, plausibly lowers prime-age male LFPR by ~1 p.p.

Explanation #2: Lack of Supportive Labor Markets: Comparatively Low LFPR

III. Explanations – Institutions

Source: Organisation for Economic Co-operation and Development.

70

75

80

85

90

95

100

Isra

elIta

lyU

nite

d St

ates

Nor

way

Finl

and

Irela

ndBe

lgiu

mAu

stra

liaTu

rkey

Pola

ndLa

tvia

Den

mar

kC

anad

aKo

rea

Aust

riaPo

rtuga

lU

nite

d Ki

ngdo

mC

hile

Hun

gary

Net

herla

nds

New

Zea

land

Esto

nia

Fran

ceG

erm

any

Spai

nSl

oven

iaG

reec

eSw

eden

Slov

ak R

epub

licIc

elan

dLu

xem

bour

gM

exic

oC

zech

Rep

ublic

Japa

nSw

itzer

land

Prime-Age Male Labor Force Participation Rates Across the OECDPercent

Large Decline in LFPR for Prime Age Men

III. Explanations – Institutions

Source: Organisation for Economic Co-operation and Development; author's calculations.

-8

-6

-4

-2

0

2

4

Italy

Uni

ted

Stat

esEs

toni

aTu

rkey

Den

mar

kKo

rea

Finl

and

Nor

way

Fran

ceU

nite

d Ki

ngdo

mAu

stra

liaBe

lgiu

mPo

rtuga

lC

anad

aJa

pan

Irela

ndIs

rael

Spai

nSw

eden

Net

herla

nds

Gre

ece

New

Zea

land

Luxe

mbo

urg

Ger

man

y

Change in Prime-Age Male Labor Force Participation RatesAcross the OECD, 1990-2015

Percentage Point Change

U.S. Labor Markets Are Very Flexible…

III. Explanations – Institutions

OECD Measure of Labor Market Flexibility U.S. Percentile Rank(100 = most flexible)

Overall Labor Market Regulation 100

Employment Protection for Regular Employment 100

Minimum Cost of Labor 96

Disability Compensation Index 96

Unemployment Insurance Benefit Duration 96

Scope of State Intervention 94

Coverage of Collective Bargaining Agreements 94

Source: Organisation for Economic Co-operation and Development; Council of Economic Advisers.

…But U.S. Labor Markets Not Very Supportive

III. Explanations – Institutions

OECD Measure of Labor Market Support U.S. Percentile Rank(100 = most supportive)

Nationwide Paid Leave Policy 0

Expenditure on Active Labor Market Policies 6

Net Childcare Costs (lone parent) 6

Implicit Tax on Returning to Work, Second Earner 10

Number of Weeks of Sick Leave 11

Source: Organisation for Economic Co-operation and Development; Stockholm University, Social Policy Indicators; Council of Economic Advisers; author’s calculations.

E.g., Consistent Declines in Spending on Active Labor Markets Policies…

III. Explanations – Institutions

Source: Organisation for Economic Co-operation and Development.

2014

0.00

0.05

0.10

0.15

0.20

0.25

0.30

1985 1990 1995 2000 2005 2010 2015

U.S. Expenditure of Active Labor Market Programs(% of GDP), 1985-2014Percent of GDP

…Which is Now Well Below Other Advanced Economies

III. Explanations – Institutions

Note: Data for Ireland, Poland, and Spain from 2013. Data for United Kingdom from 2011. Source: Organisation for Economic Co-operation and Development.

0.0

0.3

0.6

0.9

1.2

1.5

1.8

2.1

Mex

icoC

hile

Uni

ted

Stat

esIs

rael

Japa

nEs

toni

aSl

ovak

Rep

ublic

Can

ada

Uni

ted

King

dom

Aust

ralia

New

Zea

land

Italy

Cze

ch R

epub

licSl

oven

iaKo

rea

Pola

ndN

orw

aySp

ain

Portu

gal

Switz

erla

ndLu

xem

bour

gG

erm

any

Belg

ium

Aust

riaN

ethe

rland

sH

unga

ryIre

land

Fran

ceFi

nlan

dSw

eden

Den

mar

k

Public Expenditure on Active Labor Market Programs (% of GDP), 2014Percent of GDP

Explanation #3: Reduced Labor Market Fluidity and Propagation of Shocks

III. Explanations – Institutions

Note: Job creation and destruction expressed as share of total employment. Dashed lines indicate 1977-2014 linear trend.Source: Census Bureau, Business Dynamics Statistics.

Job Creation

2014

Job Destruction

10

12

14

16

18

20

22

1975 1980 1985 1990 1995 2000 2005 2010 2015

Labor Market Dynamism, 1977-2014Rate (Percent)

Increased Cyclical Sensitivity of Involuntary Part-time

III. Explanations – Institutions

Summary of the Key Facts & Explanation

III. Explanationsd Explanations

Key Facts

• Steady decline of the labor force participation rate for prime age men since the 1950s.

• The decline has been largest for people with a high school degree or less.

• The rate and decline are worse in the United States than in most other advanced economies.

Explanations

• Little evidence of reduced supply.

• Evidence for reduced demand, including lower relative wages for less skilled.

• But the demand shock is mediated by institutions. In particular: (1) mass incarceration; (2) less supportive labor markets; and (3) less fluid labor markets.

Impacts of Nonparticipation & Unemployment

IV. Policy Response

• Increased mortality, largely due to suicide and alcohol (Eliason and Storrie2009 and Gerdtham and Johannesson 2003)

• Higher rates of smoking initiation (Marcus 2014)

• Increased body weight (Black, Devereux, and Salvanes 2015)

• Lower reported happiness (Winkelmann and Winkelmann 1995; Knabe and Ratzel 2011; Lucas et al. 2001)

• Community effects, including crime (Raphael and Winter-Ebmer 2001; Gould et al. 2002; Lin 2008)

• Effects on children: school performance, future earnings and future benefits receipt (Rege, Telle and Votruba 2011; Oreopoulos, Page and Stevens 2008)

• Pain and pain medication (Krueger 2016)

Some Specific Policies

IV. Policy Response

1. Labor Demand

Infrastructure

Public employment

Work Opportunity Tax Credit for long-term unemployed

Some Specific Policies

IV. Policy Response

1. Labor Demand

2. Labor Supply

Minimum wage/unionization

EITC for workers without qualifying children

Tax treatment of secondary earners

Some Specific Policies

IV. Policy Response

1. Labor Demand

2. Labor Supply

3. Connective Tissue & Flexible Workplaces

Unemployment Insurance reform

Universal dislocated workers support

Disability insurance reform

Childcare subsidies, flexible workplaces, paid leave

Some Specific Policies

IV. Policy Response

1. Labor Demand

2. Labor Supply

3. Connective Tissue & Flexible Workplaces

4. Labor Market Fluidity

Wage insurance

Occupational licensing reform

Land use reform

The Decline of Men Working: Why It Is Happening, What It Means, and What

to Do About ItJason FurmanSenior Fellow, PIIE

Peterson Institute for International Economics | 1750 Massachusetts Ave., NW | Washington, DC 20036

University of California, San DiegoSan Diego, CA

May 2, 2017