[PPT] - The Great Lockdown and the Big Stimulus: Tracing the Pandemic PowerPoint Presentation

SLIDE 1

The Great Lockdown and the Big Stimulus:

Tracing the Pandemic Possibility Frontier for the U.S.

Greg Kaplan Ben Moll Gianluca Violante Slides at http://violante.mycpanel.princeton.edu/Slides/PPF_slides.pdf

JHU

October 22, 2020

SLIDE 2

What We Do

✎ Focus: policy response to COVID-19 in the United States

1. Lockdown: business closure and stay-at-home orders
2. Stimulus: CARES Act

✎ Goal: quantify trade-offs ✎ Aggregate: lives versus livelihoods ✎ Distributional: who bears the economic costs? ✎ Approach: distributional Pandemic Possibility Frontier (PPF)

✦ PPF

✎ Separate economic costs from fatalities ✎ Menu of choices, independent of policymakers’ preferences ✎ Seek policies that shift the frontier inward

KMV - Great Lockdown

SLIDE 3

How We Do It

1. SIR model

✎ Two-way behavioral feedback: between virus & economic activity

2. Heterogeneous Agent model

✎ Sectors: regular, social, and home production ✎ Types of market-labor: workplace and remote ✎ Occupations: flexibility and sectoral intensity ✎ Economic exposure to pandemic correlated with financial vulnerability ✎ Calibrate model to U.S. economy and examine counterfactuals

1. Laissez-faire
2. Lockdown & Lockdown + Fiscal support
3. ‘Smarter’ policies: Pigouvian taxation with targeted redistribution

KMV - Great Lockdown

SLIDE 4

How We Do It

1. SIR model

✎ Two-way behavioral feedback: between virus & economic activity

2. Heterogeneous Agent model

✎ Sectors: regular, social, and home production ✎ Types of market-labor: workplace and remote ✎ Occupations: flexibility and sectoral intensity ✎ Economic exposure to pandemic correlated with financial vulnerability ✎ Calibrate model to U.S. economy and examine counterfactuals

1. Laissez-faire
2. Lockdown & Lockdown + Fiscal support
3. ‘Smarter’ policies: Pigouvian taxation with targeted redistribution

KMV - Great Lockdown

SLIDE 5

How We Do It

1. SIR model

✎ Two-way behavioral feedback: between virus & economic activity

2. Heterogeneous Agent model

✎ Sectors: regular, social, and home production ✎ Types of market-labor: workplace and remote ✎ Occupations: flexibility and sectoral intensity ✎ Economic exposure to pandemic correlated with financial vulnerability ✎ Calibrate model to U.S. economy and examine counterfactuals

1. Laissez-faire
2. Lockdown & Lockdown + Fiscal support
3. ‘Smarter’ policies: Pigouvian taxation with targeted redistribution

KMV - Great Lockdown

SLIDE 6

How We Do It

1. SIR model

✎ Two-way behavioral feedback: between virus & economic activity

2. Heterogeneous Agent model

✎ Sectors: regular, social, and home production ✎ Types of market-labor: workplace and remote ✎ Occupations: flexibility and sectoral intensity ✎ Economic exposure to pandemic correlated with financial vulnerability ✎ Calibrate model to U.S. economy and examine counterfactuals

1. Laissez-faire
2. Lockdown & Lockdown + Fiscal support
3. ‘Smarter’ policies: Pigouvian taxation with targeted redistribution

KMV - Great Lockdown

SLIDE 7

What We Find

1. Economic welfare costs of the pandemic are very heterogeneous

✎ Regardless of the policy response ✎ Laissez-faire vs lockdown: who bears the cost differs ✎ Largest welfare costs in the middle of earnings distribution

2. Slope of PPF varies with length of the lockdown

✎ Reconcile conflicting views on extent of health-wealth trade-off

3. CARES Act reduced average economic cost by 20%

✎ Highly redistributive toward bottom ✎ Explains rapid recovery in consumption of poor households

4. Taxation-based alternatives to lockdown flatten the PPF

✎ Mean trade-off improved, but even more dispersion in outcomes

✦ dimensions not considering today KMV - Great Lockdown

SLIDE 8

What We Find

1. Economic welfare costs of the pandemic are very heterogeneous

✎ Regardless of the policy response ✎ Laissez-faire vs lockdown: who bears the cost differs ✎ Largest welfare costs in the middle of earnings distribution

2. Slope of PPF varies with length of the lockdown

✎ Reconcile conflicting views on extent of health-wealth trade-off

3. CARES Act reduced average economic cost by 20%

✎ Highly redistributive toward bottom ✎ Explains rapid recovery in consumption of poor households

4. Taxation-based alternatives to lockdown flatten the PPF

✎ Mean trade-off improved, but even more dispersion in outcomes

✦ dimensions not considering today KMV - Great Lockdown

SLIDE 9

What We Find

1. Economic welfare costs of the pandemic are very heterogeneous

✎ Regardless of the policy response ✎ Laissez-faire vs lockdown: who bears the cost differs ✎ Largest welfare costs in the middle of earnings distribution

2. Slope of PPF varies with length of the lockdown

✎ Reconcile conflicting views on extent of health-wealth trade-off

3. CARES Act reduced average economic cost by 20%

✎ Highly redistributive toward bottom ✎ Explains rapid recovery in consumption of poor households

4. Taxation-based alternatives to lockdown flatten the PPF

✎ Mean trade-off improved, but even more dispersion in outcomes

✦ dimensions not considering today KMV - Great Lockdown

SLIDE 10

What We Find

1. Economic welfare costs of the pandemic are very heterogeneous

✎ Regardless of the policy response ✎ Laissez-faire vs lockdown: who bears the cost differs ✎ Largest welfare costs in the middle of earnings distribution

2. Slope of PPF varies with length of the lockdown

✎ Reconcile conflicting views on extent of health-wealth trade-off

3. CARES Act reduced average economic cost by 20%

✎ Highly redistributive toward bottom ✎ Explains rapid recovery in consumption of poor households

4. Taxation-based alternatives to lockdown flatten the PPF

✎ Mean trade-off improved, but even more dispersion in outcomes

✦ dimensions not considering today KMV - Great Lockdown

SLIDE 11

Outline I

KMV - Great Lockdown

SLIDE 12

Epidemiological Model

✎ ❙t: susceptible ✎ ■t: infectious ✎ ❘t: recovered ✎ ❊t: exposed = latent virus, not yet infectious ✎ ❈t: critical = in ICU, may ultimately die ✎ ◆t: population = ❙t ✰ ❊t ✰ ■t ✰ ❈t ✰ ❘t ✎ Death probability of ❈t’s depends on ❈t ❄ max ICU capacity ❈♠❛① ✎ Effective reproduction number: ❘t ❂ ☞t ✶

✕■ ❙t ◆t

✎ Feedback from economic activity ✦ infections ☞t ❂ ☞✵ ❈st ✖ ❈s

✗☞

▲✇t ✖ ▲✇

✗☞

✎ Parameterization: ✗☞ ❂ ✵✿✽ ✎ ❘t drops from ✷✿✺ to ✵✿✽ after the lockdown ✎ Google Community Indexes of Workplace and Retail drop ✺✵✪ ✎ Assumption: vaccine arrives after 24 months

✦ Google Indexes ✦ Parameters Epi model KMV - Great Lockdown

SLIDE 13

Epidemiological Model

✎ ❙t: susceptible ✎ ■t: infectious ✎ ❘t: recovered ✎ ❊t: exposed = latent virus, not yet infectious ✎ ❈t: critical = in ICU, may ultimately die ✎ ◆t: population = ❙t ✰ ❊t ✰ ■t ✰ ❈t ✰ ❘t ✎ Death probability of ❈t’s depends on ❈t ❄ max ICU capacity ❈♠❛① ✎ Effective reproduction number: ❘t ❂ ☞t ✶

✕■ ❙t ◆t

✎ Feedback from economic activity ✦ infections ☞t ❂ ☞✵ ❈st ✖ ❈s ✗☞ ▲✇t ✖ ▲✇ ✗☞ ✎ Parameterization: ✗☞ ❂ ✵✿✽ ✎ ❘t drops from ✷✿✺ to ✵✿✽ after the lockdown ✎ Google Community Indexes of Workplace and Retail drop ✺✵✪ ✎ Assumption: vaccine arrives after 24 months

✦ Google Indexes ✦ Parameters Epi model KMV - Great Lockdown

SLIDE 14

Epidemiological Model

✎ ❙t: susceptible ✎ ■t: infectious ✎ ❘t: recovered ✎ ❊t: exposed = latent virus, not yet infectious ✎ ❈t: critical = in ICU, may ultimately die ✎ ◆t: population = ❙t ✰ ❊t ✰ ■t ✰ ❈t ✰ ❘t ✎ Death probability of ❈t’s depends on ❈t ❄ max ICU capacity ❈♠❛① ✎ Effective reproduction number: ❘t ❂ ☞t ✶

✕■ ❙t ◆t

✎ Feedback from economic activity ✦ infections ☞t ❂ ☞✵ ❈st ✖ ❈s ✗☞ ▲✇t ✖ ▲✇ ✗☞ ✎ Parameterization: ✗☞ ❂ ✵✿✽ ✎ ❘t drops from ✷✿✺ to ✵✿✽ after the lockdown ✎ Google Community Indexes of Workplace and Retail drop ✺✵✪ ✎ Assumption: vaccine arrives after 24 months

✦ Google Indexes ✦ Parameters Epi model KMV - Great Lockdown

SLIDE 15

Epidemiological Model

✎ ❙t: susceptible ✎ ■t: infectious ✎ ❘t: recovered ✎ ❊t: exposed = latent virus, not yet infectious ✎ ❈t: critical = in ICU, may ultimately die ✎ ◆t: population = ❙t ✰ ❊t ✰ ■t ✰ ❈t ✰ ❘t ✎ Death probability of ❈t’s depends on ❈t ❄ max ICU capacity ❈♠❛① ✎ Effective reproduction number: ❘t ❂ ☞t ✶

✕■ ❙t ◆t

✎ Feedback from economic activity ✦ infections ☞t ❂ ☞✵ ❈st ✖ ❈s ✗☞ ▲✇t ✖ ▲✇ ✗☞ ✎ Parameterization: ✗☞ ❂ ✵✿✽ ✎ ❘t drops from ✷✿✺ to ✵✿✽ after the lockdown ✎ Google Community Indexes of Workplace and Retail drop ✺✵✪ ✎ Assumption: vaccine arrives after 24 months

✦ Google Indexes ✦ Parameters Epi model KMV - Great Lockdown

SLIDE 16

Occupations and Sectors

Flexible Rigid ❈-intensive Software engineer, architect Car mechanic, miner ❙-intensive Event planner, social scientist Waiter, shop assistant

1. Flexibility: substitutability between remote and workplace hours

✎ Total labor supply = ▲❥

✇ ✰ ✣❥▲❥ r

2. Sectoral intensity: share of employment in social vc regular sector, ✭✘❥

s❀ ✘❥ ❝✮

❨✐ ❂ ❩✐◆☛✐

✐ ❑✶☛✐ ✐

❀ ◆✐ ❂

❏ ❥❂✶

✘❥

✐

✶ ✛

◆❥

✐

✛✶ ✛ ✛ ✛✶

❀ ✐ ✷ ❢s❀ ❝❣

KMV - Great Lockdown

SLIDE 17

Occupations and Sectors

Flexible Rigid ❈-intensive Software engineer, architect Car mechanic, miner ❙-intensive Event planner, social scientist Waiter, shop assistant

1. Flexibility: substitutability between remote and workplace hours

✎ Total labor supply = ▲❥

✇ ✰ ✣❥▲❥ r

2. Sectoral intensity: share of employment in social vc regular sector, ✭✘❥

s❀ ✘❥ ❝✮

❨✐ ❂ ❩✐◆☛✐

✐ ❑✶☛✐ ✐

❀ ◆✐ ❂  

❏

❥❂✶
✘❥

✐

✶

✛

◆❥

✐

✛✶

✛

 

✛ ✛✶

❀ ✐ ✷ ❢s❀ ❝❣

KMV - Great Lockdown

SLIDE 18

Occupations and Sectors

Flexible Rigid ❈-intensive Software engineer, architect Car mechanic, miner Supermarket clerk ❙-intensive Event planner, social scientist Waiter, shop assistant Nurse

1. Flexibility: substitutability between remote and workplace hours

✎ Total labor supply = ▲❥

✇ ✰ ✣❥▲❥ r

2. Sectoral intensity: share of employment in social vc regular sector, ✭✘❥

s❀ ✘❥ ❝✮

❨✐ ❂ ❩✐◆☛✐

✐ ❑✶☛✐ ✐

❀ ◆✐ ❂  

❏

❥❂✶
✘❥

✐

✶

✛

◆❥

✐

✛✶

✛

 

✛ ✛✶

❀ ✐ ✷ ❢s❀ ❝❣

3. Essential occupations: exempt from lockdown

KMV - Great Lockdown

SLIDE 19

Households

✎ Forward looking and discount at time preference rate ✭✚✮ + death rate ✎ Period utility: ❯❬❝❀ s❀ ❤❪ ❱ ❬❵✇❀ ❵r❀ ❤❪ ✎ ❝: regular consumption ✎ s: social consumption ✎ ❤: home production ✎ ❵✇: workplace hours ✎ ❵r: remote hours ✎ Budget constraint of healthy household working in occupation ❥ ❴ ❜ ❂ ✭✶ ✜✮✇ ❥③ ❵✇ ✰ ✣❥❵r ✰ r ❜❜ ✰ ❚ ❝ ♣ss ❞ ✤✭❞❀ ❛✮ ❴ ❛ ❂ r ❛❛ ✰ ❞ ✎ ❜: liquid assets ✎ ❛: illiquid assets ✎ ✣❥ ✷ ❬✵❀ ✶❪: flexibility of occupation ❥ ✎ ✤: transaction cost ✎ Sick households (= ❈, in ICU): cannot produce, gov’t feeds them

KMV - Great Lockdown

SLIDE 20

Households

✎ Forward looking and discount at time preference rate ✭✚✮ + death rate ✎ Period utility: ❯❬❝❀ s❀ ❤❪ ❱ ❬❵✇❀ ❵r❀ ❤❪ ✎ ❝: regular consumption ✎ s: social consumption ✎ ❤: home production ✎ ❵✇: workplace hours ✎ ❵r: remote hours ✎ Budget constraint of healthy household working in occupation ❥ ❴ ❜ ❂ ✭✶ ✜✮✇ ❥③

❵✇ ✰ ✣❥❵r
✰ r ❜❜ ✰ ❚ ❝ ♣ss ❞ ✤✭❞❀ ❛✮

❴ ❛ ❂ r ❛❛ ✰ ❞ ✎ ❜: liquid assets ✎ ❛: illiquid assets ✎ ✣❥ ✷ ❬✵❀ ✶❪: flexibility of occupation ❥ ✎ ✤: transaction cost ✎ Sick households (= ❈, in ICU): cannot produce, gov’t feeds them

KMV - Great Lockdown

SLIDE 21

Feedback from Virus ✦ Economic Activity

✎ Period utility: ❯❬❝❀ ✢s✭■✮s❀ ❤❪ ❱ ❬✢❵✭■✮❵✇❀ ❵r❀ ❤❪ ✎ Circulation of virus affects (dis)utility from working and consuming s ✎ Not connected to ‘value of life’/fear of death, but to inability to lead a normal life ✎ Calibration: match drop in activity before executive orders (Google Index) ✎ Probably a lower bound for behavioral response at the moment

KMV - Great Lockdown

SLIDE 22

Feedback from Virus ✦ Economic Activity

✎ Period utility: ❯❬❝❀ ✢s✭■✮s❀ ❤❪ ❱ ❬✢❵✭■✮❵✇❀ ❵r❀ ❤❪ ✎ Circulation of virus affects (dis)utility from working and consuming s ✎ Not connected to ‘value of life’/fear of death, but to inability to lead a normal life ✎ Calibration: match drop in activity before executive orders (Google Index) ✎ Probably a lower bound for behavioral response at the moment

KMV - Great Lockdown

SLIDE 23

Lockdowns

1. Workplace lockdown: Mandated maximum individual workplace hours

❵✇ ✔ ✔❵✭❵✇ ✰ ❵r✮❀ ✔❵ ✷ ❬✵❀ ✶❪

2. Social sector lockdown: Mandated decrease in ❑ utilization in s sector

❈s ❂ ❩s✭✔s❑s✮☛s◆✶☛s

s

❀ ✔s ✷ ❬✵❀ ✶❪ ✎ 1. ‘stay-at-home orders’ and 2. ‘non-essential business closures’ ✎ Lockdowns reduce infections because reduce ☞t ❂ ☞✭❈st❀ ▲✇t✮ ✎ Lockdowns affect same behavioral margins as the pandemic ✎ Calibration: (i) 2 months, (ii) match decline in activity (Google Index) ✎ Assumption: no future lockdown in case of 2nd wave

KMV - Great Lockdown

SLIDE 24

Lockdowns

1. Workplace lockdown: Mandated maximum individual workplace hours

❵✇ ✔ ✔❵✭❵✇ ✰ ❵r✮❀ ✔❵ ✷ ❬✵❀ ✶❪

2. Social sector lockdown: Mandated decrease in ❑ utilization in s sector

❈s ❂ ❩s✭✔s❑s✮☛s◆✶☛s

s

❀ ✔s ✷ ❬✵❀ ✶❪ ✎ 1. ‘stay-at-home orders’ and 2. ‘non-essential business closures’ ✎ Lockdowns reduce infections because reduce ☞t ❂ ☞✭❈st❀ ▲✇t✮ ✎ Lockdowns affect same behavioral margins as the pandemic ✎ Calibration: (i) 2 months, (ii) match decline in activity (Google Index) ✎ Assumption: no future lockdown in case of 2nd wave

KMV - Great Lockdown

SLIDE 25

Lockdowns

1. Workplace lockdown: Mandated maximum individual workplace hours

❵✇ ✔ ✔❵✭❵✇ ✰ ❵r✮❀ ✔❵ ✷ ❬✵❀ ✶❪

2. Social sector lockdown: Mandated decrease in ❑ utilization in s sector

❈s ❂ ❩s✭✔s❑s✮☛s◆✶☛s

s

❀ ✔s ✷ ❬✵❀ ✶❪ ✎ 1. ‘stay-at-home orders’ and 2. ‘non-essential business closures’ ✎ Lockdowns reduce infections because reduce ☞t ❂ ☞✭❈st❀ ▲✇t✮ ✎ Lockdowns affect same behavioral margins as the pandemic ✎ Calibration: (i) 2 months, (ii) match decline in activity (Google Index) ✎ Assumption: no future lockdown in case of 2nd wave

KMV - Great Lockdown

SLIDE 26

Lockdowns

1. Workplace lockdown: Mandated maximum individual workplace hours

❵✇ ✔ ✔❵✭❵✇ ✰ ❵r✮❀ ✔❵ ✷ ❬✵❀ ✶❪

2. Social sector lockdown: Mandated decrease in ❑ utilization in s sector

❈s ❂ ❩s✭✔s❑s✮☛s◆✶☛s

s

❀ ✔s ✷ ❬✵❀ ✶❪ ✎ 1. ‘stay-at-home orders’ and 2. ‘non-essential business closures’ ✎ Lockdowns reduce infections because reduce ☞t ❂ ☞✭❈st❀ ▲✇t✮ ✎ Lockdowns affect same behavioral margins as the pandemic ✎ Calibration: (i) 2 months, (ii) match decline in activity (Google Index) ✎ Assumption: no future lockdown in case of 2nd wave

KMV - Great Lockdown

SLIDE 27

Lockdowns

1. Workplace lockdown: Mandated maximum individual workplace hours

❵✇ ✔ ✔❵✭❵✇ ✰ ❵r✮❀ ✔❵ ✷ ❬✵❀ ✶❪

2. Social sector lockdown: Mandated decrease in ❑ utilization in s sector

❈s ❂ ❩s✭✔s❑s✮☛s◆✶☛s

s

❀ ✔s ✷ ❬✵❀ ✶❪ ✎ 1. ‘stay-at-home orders’ and 2. ‘non-essential business closures’ ✎ Lockdowns reduce infections because reduce ☞t ❂ ☞✭❈st❀ ▲✇t✮ ✎ Lockdowns affect same behavioral margins as the pandemic ✎ Calibration: (i) 2 months, (ii) match decline in activity (Google Index) ✎ Assumption: no future lockdown in case of 2nd wave

KMV - Great Lockdown

SLIDE 28

Remaining Model Ingredients

Investment Fund ✎ Illiquid asset = shares of a risk-neutral investment fund ✎ The fund owns ❑ and makes investment decisions Fiscal Authority ✎ Issues liquid debt ✭❇❣✮, spends ✭●✮, taxes and transfers ✭❚✮ Monetary Authority ✎ Central bank absorbs the additional debt needed to finance CARES Act

✦ market clearing conditions ✦ occupation flexibility details ✦ occupation sectoral details KMV - Great Lockdown

SLIDE 29

Economic Exposure and Financial Vulnerability

1000

1000 2000 3000 4000 5000

Median liquid wealth ($)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

ONET (share of teleworkable jobs) Correlation between Flexibility and Median Liquid Wealth Across Occupations

Weighted Correlation: 0.51

✎ Model reproduces median liquid wealth holdings by occupation

KMV - Great Lockdown

SLIDE 30

Outline I

KMV - Great Lockdown

SLIDE 31

Laissez-faire vs Lockdown: Aggregates

Apr Jul Oct 2021 Apr Jul

Month

0.00 0.02 0.04 0.06 0.08 0.10 0.12

Monthly Death Rate (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul

Month

25 20 15 10 5

Output (%)

Laissez-faire Full Lockdown

✎ Lockdown ✦ second wave, but fewer cumulative deaths ✎ Lockdown ✦ longer, deeper contraction and ❲-shaped recovery

✦ laissez-faire dynamics ✦ lockdown dynamics ✦ lockdown decomposition KMV - Great Lockdown

SLIDE 32

Laissez-faire vs Lockdown: Occupations

Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

Labor Income (CF: C-intensive, Flexible) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

Labor Income (CR: C-intensive, Rigid) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul

Month

50 40 30 20 10

Labor Income (SF: S-intensive, Flexible) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul

Month

50 40 30 20 10

Labor Income (SR: S-intensive, Rigid) (%)

Laissez-faire Full Lockdown

✎ Laissez-faire: large drop in income mainly for S-intensive occupations ✎ Lockdown: severe drop in income also for C-intensive occupations ✎ Second wave looks like a laissez-faire recession

KMV - Great Lockdown

SLIDE 33

Laissez-faire vs Lockdown: Occupations

Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

Labor Income (CF: C-intensive, Flexible) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

Labor Income (CR: C-intensive, Rigid) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul

Month

50 40 30 20 10

Labor Income (SF: S-intensive, Flexible) (%)

Laissez-faire Full Lockdown

Apr Jul Oct 2021 Apr Jul

Month

50 40 30 20 10

Labor Income (SR: S-intensive, Rigid) (%)

Laissez-faire Full Lockdown

✎ Laissez-faire: large drop in income mainly for S-intensive occupations ✎ Lockdown: severe drop in income also for C-intensive occupations ✎ Second wave looks like a laissez-faire recession

KMV - Great Lockdown

SLIDE 34

Pandemic Possibility Frontier (PPF)

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Deaths (% of Population)

2 4 6 8 10

Economic Welfare Cost (Multiples of Monthly Income)

Laissez-faire US Lockdown 12-Month Lockdown 17-Month Lockdown

Mean p25-p75 p10-p90 p5-p95

✎ Large average economic costs and big dispersion ✎ Heterogeneity in economic costs exacerbated with longer lockdowns ✎ Very non-linear trade-off: role of ICU constraint and vaccine

✦ ppf by occupation KMV - Great Lockdown

SLIDE 35

Pandemic Possibility Frontier (PPF)

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Deaths (% of Population)

2 4 6 8 10

Economic Welfare Cost (Multiples of Monthly Income)

Laissez-faire US Lockdown 12-Month Lockdown 17-Month Lockdown

Mean p25-p75 p10-p90 p5-p95

✎ Large average economic costs and big dispersion ✎ Heterogeneity in economic costs exacerbated with longer lockdowns ✎ Very non-linear trade-off: role of ICU constraint and vaccine

✦ ppf by occupation KMV - Great Lockdown

SLIDE 36

Distribution of Economic Welfare Costs

1 2 3 4 5 6 7 8 9 10

Earnings Deciles

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

(a) Economic Welfare Cost

Laissez-faire US lockdown 0.6 0.8 1.0 1.2 1.4 1.6

Economic Welfare Cost

0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200

(b) Lockdown - Laissez-faire

Earnings Decile ✎ Largest economic costs in middle of distribution ✎ Transfers (bottom) vs Rigid labor (middle) vs Flexible labor (top)

✦ welfare cost distribution KMV - Great Lockdown

SLIDE 37

CARES Act Shifts Down the PPF

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Deaths (% of Population)

2 4 6 8 10

Economic Welfare Cost (Multiples of Monthly Income)

US Lockdown L a i s s e z

f

a i r e US Policy 1 7

M
n

t h L

c

k d

w

n 12-Month Lockdown Mean w/o Fiscal Support p10-p90 w/o Fiscal Support Mean w/ Fiscal Support p10-p90 w/ Fiscal Support

✎ CARES Act: stimulus checks, pandemic UI, PPP

✦ CARES Act details ✦ CARES Act dynamics ✦ components of CARES Act ✦ CARES Act by income quartile ✦ components by income quartile KMV - Great Lockdown

SLIDE 38

Distribution of Economic Welfare Costs

Earnings Decile ✎ Big impact of CARES Act on households below the median ✎ On economic grounds, bottom 1/3 prefers US policy to laissez-faire

✦ welfare cost distribution KMV - Great Lockdown

SLIDE 39

Consumption Dynamics

✎ US Data: biggest ② drops, but fastest ❝ recovery at the bottom of the income distribution

✦ US data Apr Jul Oct 2021 Apr Jul Month 30 20 10

(a) Labor Income (%)

Bottom Quartile Top Quartile

Apr Jul Oct 2021 Apr Jul

Month

20 15 10 5

(b) Consumption (%)

Bottom Quartile Top Quartile

✎ CARES Act redistributed heavily toward low-income hh with high MPC

✦ components of CARES Act by income quartile KMV - Great Lockdown

SLIDE 40

Consumption Dynamics

✎ US Data: biggest ② drops, but fastest ❝ recovery at the bottom of the income distribution

✦ US data Apr Jul Oct 2021 Apr Jul Month 30 20 10

(a) Labor Income (%)

Bottom Quartile Top Quartile

Apr Jul Oct 2021 Apr Jul

Month

20 15 10 5

(b) Consumption (%)

Bottom Quartile Top Quartile

✎ CARES Act redistributed heavily toward low-income hh with high MPC

✦ components of CARES Act by income quartile KMV - Great Lockdown

SLIDE 41

Alternative Smarter Polices

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 2 4 6 8 10

(a) Comparison

Mean: US Lockdown Mean: Social Consumption Tax Mean: Workplace Hours Tax

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 2 4 6 8 10

Laissez-faire 12-Month Lockdown 21-Month US Lockdown 17-Month Lockdown

(b) US Lockdown

Mean p10-p90 p25-p75

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Deaths (% of Population)

2 4 6 8 10

L a i s s e z

f

a i r e 17-Month 12-Month 21-Month 2-Month

(c) Social Consumption Tax

Mean p10-p90 p25-p75

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Deaths (% of Population)

2 4 6 8 10

L a i s s e z

f

a i r e 17-Month 12-Month 21-Month 2-Month

(d) Workplace Hours Tax

Mean p10-p90 p25-p75

KMV - Great Lockdown

SLIDE 42

Outline I

KMV - Great Lockdown

SLIDE 43

Messages

1. Economic cost of pandemic: large & diverse, regardless of lockdown
2. Distributional PPF as a tool for quantifying trade-offs:

✎ Aggregate: between lives vs livelihoods ✎ Distributional: over who bears economic burden

3. Non-linear PPF: reconciles conflicting views on aggregate tradeoff
4. Exposure correlated with vulnerability ✮ scope for fiscal policy
5. US CARES Act:

✎ Shifts PPF inward: reduces economic costs w/o increasing deaths ✎ Faster recovery of spending for low income households

6. Pigouvian schemes alternative to lockdowns improve aggregate trade-off

KMV - Great Lockdown

SLIDE 44

Thanks and Stay Safe!

KMV - Great Lockdown

SLIDE 45

Outline I

KMV - Great Lockdown

SLIDE 46

Pandemic Possibility Frontier

Deaths Economic Welfare Cost

Laissez-faire US Lockdown 12-Month Lockdown 17-Month Lockdown

Mean ✦ back to intro KMV - Great Lockdown

SLIDE 47

Pandemic Possibility Frontier

Deaths Economic Welfare Cost

Laissez-faire US Lockdown 12-Month Lockdown 17-Month Lockdown

Mean p25-p75 p10-p90 p5-p95 ✦ back to intro KMV - Great Lockdown

SLIDE 48

Some Dimensions we Abstract From

1. Differential impact of the epidemic across age groups

(Glover-Heathcote-Krueger-RiosRull, Bairoliya-Imrohoroglu, Acemoglu et al., Brotherhood-Kircher-Santos-Tertilt, ...)

2. Differential impacts of the epidemic across gender

(Alon-Doepke-Olmstead Rumsey-Tertilt, ...)

3. Impact of the epidemic on deaths from other causes
4. Input-output linkages in production

(Baqaee-Farhi, ...)

5. Firm balance sheets, liquidity provision to firms

(Buera-Fattal Jaef-Neumeyer-Shin, Elenev-Landvoigt-VanNieuwerburgh, ...)

6. Costly destruction of viable employment relationships
7. ...

✦ model ingredients KMV - Great Lockdown

SLIDE 49

Google COVID-19 Community Mobility Data

Feb 18 Mar 03 Mar 17 Mar 31 Apr 14 Apr 28 Date 2020

60
50
40
30
20
10

10 20 Workplace Mobility Index Over 90% of population in lockdown Feb 18 Mar 03 Mar 17 Mar 31 Apr 14 Apr 28 Date 2020

60
50
40
30
20
10

10 20 Retail Mobility Index Over 90% of population in lockdown

✦ back to Epi model KMV - Great Lockdown

SLIDE 50

Background on Lockdowns in SIR Models

✎ Some vocabulary:

1. Basic reproduction number:

❘✵ ✿❂ ☞✵❂✕■

2. Effective reproduction number:

❘❡

t ✿❂ ❘✵ ✂ ❙t❂◆t

3. Herd immunity threshold:

❙✄❂◆ ✿❂ ✶❂❘✵

r

❘✄❂◆ ❂ ✶ ❙✄❂◆ ❂ ✶ ✶❂❘✵

4. Final size of disease:

❙✶ ❂ ❡❘✵✭✶❙✶✮ ✎ Two key features of SIR models:

1. Infections ✧ if ❘❡

t ❃ ✶ or ❙ ❃ ❙✄ and ★ otherwise

2. Epidemic “overshoot”: total infections > herd immunity, ❙✶ ❃ ❙✄

✎ Results on lockdowns := ❘✵ ★ ✎ Even temporary lockdowns reduce total number of infections ✎ But total number of infections ✕ herd immunity threshold ✎ Best lockdowns-only can do is eliminate epidemic “overshoot” ✎ If lockdown too short or too tight, get 2nd wave

✦ back to Epi model KMV - Great Lockdown

SLIDE 51

Background on Lockdowns in SIR Models

✎ Some vocabulary:

1. Basic reproduction number:

❘✵ ✿❂ ☞✵❂✕■

2. Effective reproduction number:

❘❡

t ✿❂ ❘✵ ✂ ❙t❂◆t

3. Herd immunity threshold:

❙✄❂◆ ✿❂ ✶❂❘✵

r

❘✄❂◆ ❂ ✶ ❙✄❂◆ ❂ ✶ ✶❂❘✵

4. Final size of disease:

❙✶ ❂ ❡❘✵✭✶❙✶✮ ✎ Two key features of SIR models:

1. Infections ✧ if ❘❡

t ❃ ✶ or ❙ ❃ ❙✄ and ★ otherwise

2. Epidemic “overshoot”: total infections > herd immunity, ❙✶ ❃ ❙✄

✎ Results on lockdowns := ❘✵ ★ ✎ Even temporary lockdowns reduce total number of infections ✎ But total number of infections ✕ herd immunity threshold ✎ Best lockdowns-only can do is eliminate epidemic “overshoot” ✎ If lockdown too short or too tight, get 2nd wave

✦ back to Epi model KMV - Great Lockdown

SLIDE 52

Background on Lockdowns in SIR Models

✎ Some vocabulary:

1. Basic reproduction number:

❘✵ ✿❂ ☞✵❂✕■

2. Effective reproduction number:

❘❡

t ✿❂ ❘✵ ✂ ❙t❂◆t

3. Herd immunity threshold:

❙✄❂◆ ✿❂ ✶❂❘✵

r

❘✄❂◆ ❂ ✶ ❙✄❂◆ ❂ ✶ ✶❂❘✵

4. Final size of disease:

❙✶ ❂ ❡❘✵✭✶❙✶✮ ✎ Two key features of SIR models:

1. Infections ✧ if ❘❡

t ❃ ✶ or ❙ ❃ ❙✄ and ★ otherwise

2. Epidemic “overshoot”: total infections > herd immunity, ❙✶ ❃ ❙✄

✎ Results on lockdowns := ❘✵ ★ ✎ Even temporary lockdowns reduce total number of infections ✎ But total number of infections ✕ herd immunity threshold ✎ Best lockdowns-only can do is eliminate epidemic “overshoot” ✎ If lockdown too short or too tight, get 2nd wave

✦ back to Epi model KMV - Great Lockdown

SLIDE 53

Market Clearing Conditions

✎ Regular goods market ❨❝ ❂ ❈❝ ✰ ■ ✰ ● ✰ ✤ ✎ Social goods market ❨s ❂ ❈s ✎ Labor market for each occupation ◆❥

❝ ✰ ◆❥ s ❂

③✭❵❥

✇✭h❀ ❛❀ ❜❀ ③✮ ✰ ✣❥❵❥ r✭h❀ ❛❀ ❜❀ ③✮✮❞✖❀

❥ ❂ ✶❀ ✿✿✿❀ ✺ ✎ Liquid asset market ❇❤ ❂ ❇❣ ✎ Illiquid asset market ❆ ❂ ❱fund✭❑❀ ✂❝❀ ✂s✮❀ ❑ ❂ ❑❝ ✰ ❑s

✦ model ingredients KMV - Great Lockdown

SLIDE 54

Epidemiological Parameters

Description Parameter Value Initial basic reproduction number ❘init

✵

❂ ☞init

✵ ❂✕■

✷✿✺ Final basic reproduction number ❘end

✵

❂ ☞end

✵

❂✕■ ✷

Avg. duration of ■nfectious

❚■ ✮ ✕■ ❂ ✶❂❚■ ✹✿✸ days

Avg. duration of ❊xposure (latency)

❚❊ ✮ ✕❊ ❂ ✶❂❚❊ ✺✿✵ days Infection fatality rate IFR = ✤✍❈ ✵✿✵✷ ✂ ✵✿✸✸ ❂ ✵✿✵✻✻ ✎ Time trend in transmissions (masks,...): ⑦ ❘t ❂ ✭✶ ✦t✮❘✵ ✰ ✦t ✖ ❘❀ ✖ ❘ ❂ ✶✿✺❀ ✦t = logistic ✎ Herd immunity threshold: ✶ ✶❂❘init

✵

❂ ✻✵✪ ✮ ✶ ✶❂❘end

✵

❂ ✺✵✪ ✎ Vaccine arrival after 18 months

✦ back to Epi model KMV - Great Lockdown

SLIDE 55

Occupations: Flexibility

✎ O*NET: Share of tasks that can be performed at home (Dingel-Neiman) ✎ ATUS Q: As part of your (main) job, can you work at home? ✎ Systematic variation across 3-digit SOC occupations

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 ATUS (share of individuals who can work from home) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 O*NET (share of teleworkable jobs)

✎ Two flexibility levels: high flexibility occupation if O*NET share > 0.5.

✦ back to model KMV - Great Lockdown

SLIDE 56

Occupations: Social vs Regular Intensity

✦ back to model KMV - Great Lockdown

SLIDE 57

Occupations: Exposure vs Vulnerability

Occupation ✣❥ ✘❥

❝

✘❥

s

Empl Share Earnings Liq Wealth Essential ✵✿✶ ✵✿✶✾ ✵✿✸✺ ✵✿✸✶ ✩✹✺❑ ✩✶❀ ✸✵✵ CF: C-intensive, Flexible ✶ ✵✿✺✼ ✵✿✶✷ ✵✿✷✶ ✩✼✾❑ ✩✶✽❀ ✹✵✵ SF: S-intensive, Flexible ✶ ✵✿✵✸ ✵✿✶✾ ✵✿✶✵ ✩✺✶❑ ✩✽❀ ✾✵✵ CR: C-intensive, Rigid ✵✿✶ ✵✿✶✾ ✵✿✵✹ ✵✿✶✸ ✩✹✺❑ ✩✶❀ ✵✵✵ SR: S-intensive, Rigid ✵✿✶ ✵✿✵✹ ✵✿✷✾ ✵✿✷✹ ✩✸✷❑ ✩✾✵✵ Source: O*NET, OES, SIPP

✎ Estimate stochastic processes for household wage dynamics by

ccupation from PSID

✎ To match liquid wealth we add occupational-specific wedge on liquid rate

✦ back to model KMV - Great Lockdown

SLIDE 58

Aggregates Dynamics: Laissez-Faire

Apr Jul Oct 2021 Apr Jul Month 0.0 0.5 1.0 1.5 2.0 2.5

(a) Reproduction Number

R0 Effective R Apr Jul Oct 2021 Apr Jul Month 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

(b) Infectious (% of Population)

Apr Jul Oct 2021 Apr Jul Month 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200

(c) Monthly Death Rate (%)

Actual No ICU Constraint Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10

(d) Output (%)

Y Regular Y Social Y Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10 10

(e) Consumption (%)

C Regular C Social C Apr Jul Oct 2021 Apr Jul Month 20 15 10 5 5 10

(f) Investment and Share Price (%)

Investment Share Price Apr Jul Oct 2021 Apr Jul Month 100 80 60 40 20 20 40 60

(g) Hours (%)

Workplace Hrs. Home Hrs. Apr Jul Oct 2021 Apr Jul Month 30 20 10 10 20

(h) Labor (%)

Labor Productivity Labor Income Apr Jul Oct 2021 Apr Jul Month 2.35 2.40 2.45 2.50 2.55 2.60 2.65 2.70

(i) Government Debt

Quarterly GDP 600 400 200 200 400 Remote Hrs.

✦ back KMV - Great Lockdown

SLIDE 59

Aggregates Dynamics: Lockdown

Apr Jul Oct 2021 Apr Jul Month 0.0 0.5 1.0 1.5 2.0 2.5

(a) Reproduction Number

R0 Effective R Apr Jul Oct 2021 Apr Jul Month 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

(b) Infectious (% of Population)

Apr Jul Oct 2021 Apr Jul Month 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200

(c) Monthly Death Rate (%)

Actual No ICU Constraint Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10

(d) Output (%)

Y Regular Y Social Y Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10 10

(e) Consumption (%)

C Regular C Social C Apr Jul Oct 2021 Apr Jul Month 20 15 10 5 5 10

(f) Investment and Share Price (%)

Investment Share Price Apr Jul Oct 2021 Apr Jul Month 100 80 60 40 20 20 40 60

(g) Hours (%)

Workplace Hrs. Home Hrs. Apr Jul Oct 2021 Apr Jul Month 30 20 10 10 20

(h) Labor (%)

Labor Productivity Labor Income Apr Jul Oct 2021 Apr Jul Month 2.35 2.40 2.45 2.50 2.55 2.60 2.65 2.70

(i) Government Debt

Quarterly GDP 600 400 200 200 400 Remote Hrs.

✦ back KMV - Great Lockdown

SLIDE 60

Lockdown Decomposition

Apr Jul Oct 2021 Apr Jul 0.00 0.02 0.04 0.06 0.08 0.10 0.12

(a) Monthly Death Rate (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

(b) Labor Income (CF: C-intensive, Flexible) (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown Apr Jul Oct 2021 Apr Jul 50 40 30 20 10

(c) Labor Income (SF: S-intensive, Flexible) (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown Apr Jul Oct 2021 Apr Jul Month 25 20 15 10 5

(d) Output (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10

(e) Labor Income (CR: C-intensive, Rigid) (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10

(f) Labor Income (SR: S-intensive, Rigid) (%)

Laissez-faire Full Lockdown Social Sector Lockdown Workplace Lockdown

✦ back KMV - Great Lockdown

SLIDE 61

Economic Welfare Cost Distribution

1 2 3 4 5 6 7 8 9 10

Earnings Deciles

1 2 3 4

(a) Economic Welfare Cost

Laissez-faire US Policy US lockdown 4 3 2 1 1 2

Economic Welfare Cost

0.00 0.05 0.10 0.15 0.20

(b) Lockdown - Laissez-faire

4 3 2 1 1 2

Economic Welfare Cost

0.0 0.1 0.2 0.3 0.4

(c) Fiscal - Lockdown

4 3 2 1 1 2

Economic Welfare Cost

0.0 0.1 0.2 0.3

(d) Fiscal - Laissez-faire

✦ back KMV - Great Lockdown

SLIDE 62

Production Possibility Frontier by Occupation

✎ ❈-intensive, rigid occupations (green line) hurt most by longer lockdowns

✦ back KMV - Great Lockdown

SLIDE 63

Modeling CARES Act

Apr Jul Oct 2021 Apr Jul

Month

0.0 0.5 1.0 1.5 2.0 2.5 3.0

(a) Transfers (% of Quarterly GDP)

Check UI

Apr Jul Oct 2021 Apr Jul

Month

10 20 30 40 50

(b) PPP Subsidies (%)

Wage Profit

Apr Jul Oct 2021 Apr Jul

Month

25 20 15 10 5

(c) Withdrawal Cost (%)

✎ Stimulus checks: unconditional transfer of $1,900 to everyone ✎ Pandemic UI: replacement earnings loss by decile (Ganong-Vavra) ✎ Paycheck Protection Program: part wage & profit subsidies (half each) ✎ 401(k) withdrawals up to $100,000: reduction in withdrawal cost

✦ back KMV - Great Lockdown

SLIDE 64

Aggregates Dynamics: Lockdown + CARES Act

Apr Jul Oct 2021 Apr Jul Month 0.0 0.5 1.0 1.5 2.0 2.5

(a) Reproduction Number

R0 Effective R Apr Jul Oct 2021 Apr Jul Month 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

(b) Infectious (% of Population)

Apr Jul Oct 2021 Apr Jul Month 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 0.200

(c) Monthly Death Rate (%)

Actual No ICU Constraint Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10

(d) Output (%)

Y Regular Y Social Y Apr Jul Oct 2021 Apr Jul Month 50 40 30 20 10 10

(e) Consumption (%)

C Regular C Social C Apr Jul Oct 2021 Apr Jul Month 20 15 10 5 5 10

(f) Investment and Share Price (%)

Investment Share Price Apr Jul Oct 2021 Apr Jul Month 100 80 60 40 20 20 40 60

(g) Hours (%)

Workplace Hrs. Home Hrs. Apr Jul Oct 2021 Apr Jul Month 30 20 10 10 20

(h) Labor (%)

Labor Productivity Labor Income Apr Jul Oct 2021 Apr Jul Month 2.35 2.40 2.45 2.50 2.55 2.60 2.65 2.70

(i) Government Debt

Quarterly GDP 600 400 200 200 400 Remote Hrs.

✦ back KMV - Great Lockdown

SLIDE 65