Model – Idiosyncratic Productivity Shocks
Model – Idiosyncratic Productivity Shocks • For married couples, earnings are given by + ν m , f ) w i f h i j exp ( η m , f l f + w i m ̟ ( i m , j ) exp ( η m , m + ν m , m ) ∗ l m j j � �� � � �� � labor endowment labor endowment
Model – Idiosyncratic Productivity Shocks • For married couples, earnings are given by + ν m , f ) w i f h i j exp ( η m , f l f + w i m ̟ ( i m , j ) exp ( η m , m + ν m , m ) ∗ l m j j � �� � � �� � labor endowment labor endowment • For j > 1, the bivariate AR(1) process is η m , m j + 1 = ρη m , m + ε m , m η m , f j + 1 = ρη m , f + ε m , f , j + 1 j + 1 j j with η m , m = η m , f = 0 and 1 1 � 0 � 0 , σ 2 σ ε f ε m ε m , m ( ε m , m j + 1 , ε m , f j + 1 ) ∼ N , σ 2 σ ε f ε m ε f , f
Model – Idiosyncratic Productivity Shocks • For married couples, earnings are given by + ν m , f ) w i f h i j exp ( η m , f l f + w i m ̟ ( i m , j ) exp ( η m , m + ν m , m ) ∗ l m j j � �� � � �� � labor endowment labor endowment • For j > 1, the bivariate AR(1) process is η m , m j + 1 = ρη m , m + ε m , m η m , f j + 1 = ρη m , f + ε m , f , j + 1 j + 1 j j with η m , m = η m , f = 0 and 1 1 � 0 � 0 , σ 2 σ ε f ε m ε m , m ( ε m , m j + 1 , ε m , f j + 1 ) ∼ N , σ 2 σ ε f ε m ε f , f • Permanent shocks: � � 0 , σ 2 σ ν m ν f 0 ( ν m , m , ν m , f ) ∼ N ν m , m 1 σ 2 σ ν m ν f ν f , f
Model – Idiosyncratic Productivity Shocks Comments:
Model – Idiosyncratic Productivity Shocks Comments: • Many parameters.
Model – Idiosyncratic Productivity Shocks Comments: • Many parameters. • Variances and covariances depend on gender and marital status.
Model – Idiosyncratic Productivity Shocks Comments: • Many parameters. • Variances and covariances depend on gender and marital status. • We infer these variances and covariances from data – inequality in wages and correlations in wages between spouses at different stages in life cycle.
Model – Idiosyncratic Productivity Shocks Comments: • Many parameters. • Variances and covariances depend on gender and marital status. • We infer these variances and covariances from data – inequality in wages and correlations in wages between spouses at different stages in life cycle. • Specification of shocks is a mixture of RIP and HIP.
Model – Demographics and Heterogeneity • Married households and single females differ in terms of the number of children attached to them. • Three possibilities: without , early , late ( b = 0, 1, 2 ) • If a female with children works, married or single, then the household has to pay for child care costs, that vary with the age of children • Joint market work for married couples also implies a utility cost. → Residual heterogeneity in labor force participation.
Model – Transfers
Model – Transfers • Welfare Programs: we use the Survey of Income and Program Participation (SIPP), 1995-2013. • Effective transfer functions from Rauh, Guner and Ventura (2019). • Include AFDC/TANF, SSI, Food Stamps/SNAP, WIC and Housing Assistance.
Model – Transfers • Welfare Programs: we use the Survey of Income and Program Participation (SIPP), 1995-2013. • Effective transfer functions from Rauh, Guner and Ventura (2019). • Include AFDC/TANF, SSI, Food Stamps/SNAP, WIC and Housing Assistance. • Child-related transfers: Child Tax Credit (CTC), Childcare Credit (CDCTC) and CCDF (childcare subsidies).
Model – Transfers • Welfare Programs: we use the Survey of Income and Program Participation (SIPP), 1995-2013. • Effective transfer functions from Rauh, Guner and Ventura (2019). • Include AFDC/TANF, SSI, Food Stamps/SNAP, WIC and Housing Assistance. • Child-related transfers: Child Tax Credit (CTC), Childcare Credit (CDCTC) and CCDF (childcare subsidies). • Earned Income Tax Credit (EITC)
Model – Transfers • Welfare Programs: we use the Survey of Income and Program Participation (SIPP), 1995-2013. • Effective transfer functions from Rauh, Guner and Ventura (2019). • Include AFDC/TANF, SSI, Food Stamps/SNAP, WIC and Housing Assistance. • Child-related transfers: Child Tax Credit (CTC), Childcare Credit (CDCTC) and CCDF (childcare subsidies). • Earned Income Tax Credit (EITC) • Total transfer functions: TR M ( I , b , j ) and TR S ( I , b , j )
Model – Taxation • Income tax functions : T M ( I , b ) and T S ( I , b ) • We estimate these functions from Internal Revenue Service (IRS) micro data – Guner, Kaygusuz and Ventura (2014) • There is a social security system financed by a flat payroll tax, τ p , plus additional flat capital income tax τ k .
Model – Preferences • Single males and single females: m ( c , l ) = log ( c ) − l 1 + 1 f ( c , l ) = log ( c ) − B f l 1 + 1 U S γ , U S γ .
Model – Preferences • Single males and single females: m ( c , l ) = log ( c ) − l 1 + 1 f ( c , l ) = log ( c ) − B f l 1 + 1 U S γ , U S γ . • Married couples 1 + 1 1 + 1 U M ( c , l f , l m , θ , q ) = 2 log ( c ) − l γ γ − θ B f l − χ { l f } q . m f
Model – Preferences • Single males and single females: m ( c , l ) = log ( c ) − l 1 + 1 f ( c , l ) = log ( c ) − B f l 1 + 1 U S γ , U S γ . • Married couples 1 + 1 1 + 1 U M ( c , l f , l m , θ , q ) = 2 log ( c ) − l γ γ − θ B f l − χ { l f } q . m f θ takes two values at start of life; θ ∈ { θ L , θ H }
Decisions – Big Picture
Decisions – Big Picture • Households have access to one-period, risk-free asset. They decide how much to consume, save and the work of their members.
Decisions – Big Picture • Households have access to one-period, risk-free asset. They decide how much to consume, save and the work of their members. • Given their state, married households decide whether the female member should work. • Costs of work: child care expenses, additional taxes. • Benefits: higher household income, future human capital.
Decisions – Big Picture • Households have access to one-period, risk-free asset. They decide how much to consume, save and the work of their members. • Given their state, married households decide whether the female member should work. • Costs of work: child care expenses, additional taxes. • Benefits: higher household income, future human capital. • Taxation plus presence and generosity of transfers affect the cost and benefits of work.
Model and Data Statistic Data Model Capital Output Ratio 2.93 2.93 LFP of Married Females (%), 25-54 Unskilled 68.2 67.7 Skilled 77.4 77.3 Total 71.8 71.5 Variance log-wages (Married Males, age 40) 0.37 0.37 Variance log-wages (Married Females, age 40) 0.33 0.35 Variance log-hours (Married Females, age 40) 0.13 0.14 Correlation Between Wages of Spouses (age 25) 0.27 0.27 Correlation Between Wages of Spouses (age 40) 0.31 0.31 Skill Premium 1.8 1.8 Variance log-consumption (Age 50-54 vs 25-29) 0.08 0.07
Model and Data Statistic Data Model Capital Output Ratio 2.93 2.93 LFP of Married Females (%), 25-54 Unskilled 68.2 67.7 Skilled 77.4 77.3 Total 71.8 71.5 Variance log-wages (Married Males, age 40) 0.37 0.37 Variance log-wages (Married Females, age 40) 0.33 0.35 Variance log-hours (Married Females, age 40) 0.13 0.14 Correlation Between Wages of Spouses (age 25) 0.27 0.27 Correlation Between Wages of Spouses (age 40) 0.31 0.31 Skill Premium 1.8 1.8 Variance log-consumption (Age 50-54 vs 25-29) 0.08 0.07
Var-Log Married Female Earnings 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age Model Data
Var-Log Married Female Hours 0.3 0.25 0.2 0.15 0.1 0.05 0 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age Data Model
Married Female Labor Force Participation 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 Age Model Data
Rethinking the Welfare State
Rethinking the Welfare State • What are the effects of abolishing the welfare state? Do households value the current scheme?
Rethinking the Welfare State • What are the effects of abolishing the welfare state? Do households value the current scheme? → Eliminate all transfers. Taxes reduced for all.
Rethinking the Welfare State • What are the effects of abolishing the welfare state? Do households value the current scheme? → Eliminate all transfers. Taxes reduced for all. • Replace all taxes and transfers with a Negative Income Tax (NIT) • Each household receives a transfer per member (including children) in all dates and states. • All households face same proportional income tax.
Rethinking the Welfare State • What are the effects of abolishing the welfare state? Do households value the current scheme? → Eliminate all transfers. Taxes reduced for all. • Replace all taxes and transfers with a Negative Income Tax (NIT) • Each household receives a transfer per member (including children) in all dates and states. • All households face same proportional income tax. • Replace all transfers with a Universal Basic Income (UBI) transfer. • Each adult receives a transfer in all dates and states. • Existing taxes unchanged. Additional resources via proportional income tax if needed.
Eliminating Welfare State No Transfer System Output (%) 1.7 Married Females LFP (%) 3.4 Married Females LFP (U, %) 4.8 Married Females LFP (S, %) 1.7 Aggregate Hours (MF, %) 3.8 Aggregate Hours (%) 2.8 Variance Log-Earnings (benchmark value: 0.524) 0.486 Welfare (CV, %) -2.9 Winning Households (%) 62.0
Eliminating Welfare State No Transfer System Output (%) 1.7 Married Females LFP (%) 3.4 Married Females LFP (U, %) 4.8 Married Females LFP (S, %) 1.7 Aggregate Hours (MF, %) 3.8 Aggregate Hours (%) 2.8 Variance Log-Earnings (benchmark value: 0.524) 0.486 Welfare (CV, %) -2.9 Winning Households (%) 62.0 • Asymmetric welfare effects. Large welfare losses – but majority support for eliminating current scheme.
Eliminating Welfare State No Transfer System Output (%) 1.7 Married Females LFP (%) 3.4 Married Females LFP (U, %) 4.8 Married Females LFP (S, %) 1.7 Aggregate Hours (MF, %) 3.8 Aggregate Hours (%) 2.8 Variance Log-Earnings (benchmark value: 0.524) 0.486 Welfare (CV, %) -2.9 Winning Households (%) 62.0 • Asymmetric welfare effects. Large welfare losses – but majority support for eliminating current scheme. • Elimination of welfare transfers (AFDC, etc) leads to largest losses.
Rethinking the Welfare State NIT NIT NIT (0%) (2%) (4%) Output (%) 2.9 1.6 0.1 Married Females LFP (U, %) 7.3 2.7 -2.3 Married Females LFP (S, %) 3.2 1.4 -0.8 Aggregate Hours (MF, %) 6.9 3.0 -1.3 Aggregate Hours (%) 4.4 2.3 0.0 Variance Log-Earnings (benchmark value: 0.524) 0.49 0.50 0.52 Tax Rate (%) 7.0 11.7 17.2 Welfare (CV, %) -4.0 -1.2 0.1 Winning Households (%) 50.4 63.4 73.8
Rethinking the Welfare State NIT NIT NIT (0%) (2%) (4%) Output (%) 2.9 1.6 0.1 Married Females LFP (U, %) 7.3 2.7 -2.3 Married Females LFP (S, %) 3.2 1.4 -0.8 Aggregate Hours (MF, %) 6.9 3.0 -1.3 Aggregate Hours (%) 4.4 2.3 0.0 Variance Log-Earnings (benchmark value: 0.524) 0.49 0.50 0.52 Tax Rate (%) 7.0 11.7 17.2 Welfare (CV, %) -4.0 -1.2 0.1 Winning Households (%) 50.4 63.4 73.8 → NIT can lead to welfare gains and majority support. But requires sizeable transfers.
NIT, Welfare (right) and Winners (left) 80 0.5 0 70 ‐ 0.5 60 ‐ 1 50 ‐ 1.5 Winners (%) Welfare (%) 40 ‐ 2 ‐ 2.5 30 ‐ 3 20 ‐ 3.5 10 ‐ 4 0 ‐ 4.5 0 0.5 1 1.5 2 3 4 5 6 Percentage of Mean Household Income Winners Welfare
NIT, Welfare (right) and Output (left) 4 0.5 0 3 ‐ 0.5 ‐ 1 2 ‐ 1.5 Welfare (%) Output (%) 1 ‐ 2 ‐ 2.5 0 ‐ 3 0 0.5 1 1.5 2 3 4 5 6 ‐ 3.5 ‐ 1 ‐ 4 ‐ 2 ‐ 4.5 Percentage of Mean Household Income Output Welfare
NIT, Output (right) and Var ‐ Log Earnings (left) 0.54 4 0.53 3 0.52 2 Var ‐ Log Earnings 0.51 Output (%) 0.5 1 0.49 0 0.48 ‐ 1 0.47 0.46 ‐ 2 0 0.5 1 1.5 2 3 4 5 6 Percentage of Mean Household Income Var ‐ Log Earnings Output
Rethinking the Welfare State NIT UBI No (Optimal) (Optimal) Transfers Output (%) -0.8 -0.9 1.7 Married Females LFP (U, %) -5.1 -3.2 4.8 Married Females LFP (S, %) 3.2 1.4 1.7 Welfare (CV, %) 0.2 -1.0 -2.9 Winning Households (%) 59.3 54.8 62.0 Transfer 5.0 5.15 0.0 (% Household Income) (per person) (per adult) – Tax Rate (%) 20.2 4.5 – (all income) (additional) –
Rethinking the Welfare State NIT UBI No (Optimal) (Optimal) Transfers Output (%) -0.8 -0.9 1.7 Married Females LFP (U, %) -5.1 -3.2 4.8 Married Females LFP (S, %) 3.2 1.4 1.7 Welfare (CV, %) 0.2 -1.0 -2.9 Winning Households (%) 59.3 54.8 62.0 Transfer 5.0 5.15 0.0 (% Household Income) (per person) (per adult) – Tax Rate (%) 20.2 4.5 – (all income) (additional) – → UBI does NOT lead to welfare gains. Dominated by NIT.
Rethinking the Welfare State NIT UBI No (Optimal) (Optimal) Transfers Output (%) -0.8 -0.9 1.7 Married Females LFP (U, %) -5.1 -3.2 4.8 Married Females LFP (S, %) 3.2 1.4 1.7 Welfare (CV, %) 0.2 -1.0 -2.9 Winning Households (%) 59.3 54.8 62.0 Transfer 5.0 5.15 0.0 (% Household Income) (per person) (per adult) – Tax Rate (%) 20.2 4.5 – (all income) (additional) – → UBI does NOT lead to welfare gains. Dominated by NIT. Optimal NIT transfer: $ 4,500 in current dollars.
Conclusions
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new).
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new). • Overall, it is hard to improve over the existing welfare system.
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new). • Overall, it is hard to improve over the existing welfare system. • Revenue-neutral elimination of all transfers leads to large welfare losses BUT is supported by a majority of newborn households.
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new). • Overall, it is hard to improve over the existing welfare system. • Revenue-neutral elimination of all transfers leads to large welfare losses BUT is supported by a majority of newborn households. • NIT arrangements can improve upon the status quo and be supported by a majority. However, ex-ante gains are not large.
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new). • Overall, it is hard to improve over the existing welfare system. • Revenue-neutral elimination of all transfers leads to large welfare losses BUT is supported by a majority of newborn households. • NIT arrangements can improve upon the status quo and be supported by a majority. However, ex-ante gains are not large. • NIT dominates UBI. KEY: larger redistribution is possible under NIT via lower distortions.
Conclusions • We develop life-cycle model suitable for policy analysis. It goes a long way towards reproducing patterns of life-cycle inequality (all and new). • Overall, it is hard to improve over the existing welfare system. • Revenue-neutral elimination of all transfers leads to large welfare losses BUT is supported by a majority of newborn households. • NIT arrangements can improve upon the status quo and be supported by a majority. However, ex-ante gains are not large. • NIT dominates UBI. KEY: larger redistribution is possible under NIT via lower distortions. • More to come...
EXTRA SLIDES
The Structure of Shocks
The Structure of Shocks • Permanent Shocks . Variance single males: σ 2 ν s , m = 0.255 Variance single females: σ 2 ν s , f = 0.226 Variance married males: σ 2 ν m , m = 0.220 Variance married females: σ 2 ν m , f = 0.216 Correlation (married males, married females): 0.216
The Structure of Shocks • Permanent Shocks . Variance single males: σ 2 ν s , m = 0.255 Variance single females: σ 2 ν s , f = 0.226 Variance married males: σ 2 ν m , m = 0.220 Variance married females: σ 2 ν m , f = 0.216 Correlation (married males, married females): 0.216 • Persistent Shocks .
The Structure of Shocks • Permanent Shocks . Variance single males: σ 2 ν s , m = 0.255 Variance single females: σ 2 ν s , f = 0.226 Variance married males: σ 2 ν m , m = 0.220 Variance married females: σ 2 ν m , f = 0.216 Correlation (married males, married females): 0.216 • Persistent Shocks . Common persistence: ρ = 0.958 – Kaplan (2012) Variance single males: σ 2 ǫ s , m = 0.005 Variance single females: σ 2 ǫ s , f ∼ 0 Variance married males: σ 2 ǫ m , m = 0.008 Variance married females: σ 2 ǫ m , f = 0.0006 Correlation (married males, married females): 0.44
Var-Log Male Earnings - Married 0.600 0.500 0.400 0.300 0.200 0.100 0.000 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age Model Data
Facts
Facts • Current Population Survey (CPS), 1980-2005. • Household heads and their spouses between ages 25 to 60; • For earnings and hours → drop all observations with (i) hourly wage lower than federal minimum wage; (ii) hours lower than than 520 hours per year; • Two groups: skilled (college educated and higher) and unskilled (less than college).
Facts • Current Population Survey (CPS), 1980-2005. • Household heads and their spouses between ages 25 to 60; • For earnings and hours → drop all observations with (i) hourly wage lower than federal minimum wage; (ii) hours lower than than 520 hours per year; • Two groups: skilled (college educated and higher) and unskilled (less than college). • Consumption Expenditure Survey (CEX) → non-durable consumption expenditure.
Facts • Current Population Survey (CPS), 1980-2005. • Household heads and their spouses between ages 25 to 60; • For earnings and hours → drop all observations with (i) hourly wage lower than federal minimum wage; (ii) hours lower than than 520 hours per year; • Two groups: skilled (college educated and higher) and unskilled (less than college). • Consumption Expenditure Survey (CEX) → non-durable consumption expenditure. • For all variables, we estimate age effects controlling for time (year) effects.
Hourly Wages, Males 2.00 1.75 1.50 1.25 1.00 0.75 ALL UnSkilled Skilled 0.50 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age
Variance of Log Earnings, Males 0.55 ALL Married 0.50 0.45 0.40 0.35 0.30 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age
Variance of Log Earnings, Females 0.54 0.52 0.50 0.48 0.46 0.44 0.42 ALL Married 0.40 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age
Variance of Log Earnings, Married Males and Females 0.55 0.50 0.45 0.40 0.35 Married Males Married Females 0.30 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age
Variance of Log Household Earnings 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 ALL Married 0.20 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Age
Variance of Log Yearly Hours, Females 0.18 ALL Married 0.17 0.16 0.15 0.14 0.13 0.12 0.11 0.10 0.09 0.08 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Ages
Recommend
More recommend
