the economic impact of covid on the us economy bloom slides.pdf · nottingham and stanford...
TRANSCRIPT
-
The Economic Impact of COVID onthe US economy
Nick Bloom (Stanford)
SIEPR, May 2020
-
Warning – depressing economic forecasts. Take precautions
2
-
Key Points
3
1. Massive uncertainty - medical progress, policy response, industrial impact etc
2. Short-run - unemployment rising to about 20% and GDP down about 20% in 2020Q2
3. Long-run – likely U-shaped (very slow recovery) or W-shaped (double-dip)
4. Recovery headwinds – bankruptcies, skills erosion, trade, immigration, taxes, politics
-
I asked a friend who is a senior forecaster about his views:
“Since the beginning of March we basically quit trying – if youask me, it is a stupid exercise and the only reason anyonewould even attempt a forecast at any horizon is because theyare getting paid, have no shame, are drunk, or likely all three.”
-
My PhD was on uncertainty – these are incredible figures
5
Notes: Weekly US economic policy uncertainty index, monthly average data 1/2/1990 to 5/7/2020. This reports the normalized share of US daily newspaper articles discussing economic policy uncertainty. Source: https://fred.stlouisfed.org/series/USEPUINDXD
Notes: Weekly implied volatility on the S&P500 index from the Chicago Board of Options Exchange 1/2/1990 to 5/7/2020. This is the 1 month VIX (implied volatility over the next 30 days) in annualized units. Source: https://fred.stlouisfed.org/series/VIXCLS
S&P500 Implied Volatility (VIX) Economic Policy Uncertainty (EPU)
020
4060
80V
IX
1990 2000 2010 2020Year
020
040
060
0D
aily
US
Eco
nom
ic P
olic
y U
ncer
tain
ty In
dex
1990 1995 2000 2005 2010 2015 2020Year
https://fred.stlouisfed.org/series/USEPUINDXDhttps://fred.stlouisfed.org/series/VIXCLS
-
COVID now the largest source of uncertainty for 90% of firms
6
0102030405060708090100
6 March(29%)
7-11 March(20%)
12-16 March(30%)
17-20 March(20%)
3 April (29%) 4-8 April(15%)
9-15 April(28%)
16-17 April(28%)
Survey submission date
March survey April survey
Source: Decision Maker Panel Survey conducted by the Bank of England, Nottingham University and Stanford University, Details in Bloom, Bunn,Chen, Mizen, Smietanka and Thwaites (2019) and on www.decisionmakerpanel.com .
% U.K. firms reporting Covid-19 as their top source of uncertainty
3 April
http://www.decisionmakerpanel.com/
-
This uncertainty is leading firms to abandon earnings guidance
Source: https://www.wsj.com/articles/stocks-keep-rallying-despite-lack-of-visibility-on-economy-11588498201?mod=hp_lead_pos1
https://www.wsj.com/articles/stocks-keep-rallying-despite-lack-of-visibility-on-economy-11588498201?mod=hp_lead_pos1
-
Key Points
8
1. Massive uncertainty - medical progress, policy response, industrial impact etc
2. Short-run - unemployment rising to about 20% and GDP down about 20% in 2020Q2
3. Long-run – likely U-shaped (very slow recovery) or W-shaped (double-dip)
4. Recovery headwinds – bankruptcies, skills erosion, trade, immigration, taxes, politics
-
One way to see the damage is the share of daily lost GDP, which has risen to about 30%
9Source: https://www.wsj.com/articles/state-coronavirus-shutdowns-have-taken-29-of-u-s-economy-offline-11586079001
https://www.wsj.com/articles/state-coronavirus-shutdowns-have-taken-29-of-u-s-economy-offline-11586079001
-
Happened incredibly fast - went from February 2020, a 60 year low rate of unemployment, to April 2020, an 80 year high
10Source: Survey of Business Uncertainty (March 9-20 & April 13-24, 2020), (Atlanta Fed, Chicago University and Stanford University). Details in Altig, Barrero,Bloom, Davis, Meyer, Mihaylov and Parker (2020) and https://www.frbatlanta.org/blogs/macroblog
What is your best guess for the impact of coronavirus developments on your firm’s sales revenue in 2020?
https://www.frbatlanta.org/blogs/macroblog
-
COVID has a hugely varied impact on firms and industries
11Source: Stanford-Stripe survey of US 2,000 firms forecast for impact of COVID on 2020Q2 sales. Details in Bloom, Fletcher and Yeh (2020) and on (https://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurship).
https://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurship
-
Also variation across firms by founder types (driven by industry mix)
12Source: Stanford-Stripe survey of US 2,000 firms forecast for impact of COVID on 2020Q2 sales. Details in Bloom, Fletcher and Yeh (2020) and on (https://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurship).
https://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurship
-
COVID → “Frankenstein recession”: combines the size of the Great Depression, the speed of Katrina and the reallocation of WWII
13
Great Depression
COVID (so far)
Hurricane Katrina COVID (so far)
“The coronavirus pandemic is forcing the fastest reallocation of labor since World War II, with companies and governments mobilizing an army of idled workers into new activities that are urgently needed.” (Wall Street Journal, 3/29/2020
https://bfi.uchicago.edu/working-paper/covid-19-is-also-a-reallocation-shock/
https://bfi.uchicago.edu/working-paper/covid-19-is-also-a-reallocation-shock/
-
Hence, I am less optimistic than the market
14
-
Key Points
15
1. Massive uncertainty - medical progress, policy response, industrial impact etc
2. Short-run - unemployment rising to about 20% and GDP down about 20% in 2020Q2
3. Long-run – likely U-shaped (very slow recovery) or W-shaped (double-dip)
4. Recovery headwinds – bankruptcies, skills erosion, trade, immigration, taxes, politics
-
faster
16Source: https://www.wsj.com/articles/why-the-economic-recovery-will-be-more-of-a-swoosh-than-v-shaped-11589203608?mod=hp_lead_pos1
https://www.wsj.com/articles/why-the-economic-recovery-will-be-more-of-a-swoosh-than-v-shaped-11589203608?mod=hp_lead_pos1
-
Key Points
17
1. Massive uncertainty - medical progress, policy response, industrial impact etc
2. Short-run - unemployment rising to about 20% and GDP down about 20% in 2020Q2
3. Long-run – likely U-shaped (very slow recovery) or W-shaped (double-dip)
4. Recovery headwinds – bankruptcies, skills erosion, trade, immigration, taxes, politics
-
18
Taxes: another long run concern as they eventually will need to rise
PresenterPresentation NotesTotal US Government Debt is about $25tn. Of that about $22tn is federal debt, the remaining $3tn is about equally split between state and local debt.Of this debt around 75% is held by the public – mainly investors (domestic and foreign), banks, mutual funds etc. Fed holds about 11% although rising fast. Of the remaining 25% of debt mainly Social Security, Disability Insurance Trust Fund, Military Retirement Fund and other retirement funds�Total US GDP is about $22tn in 2019
-
Deficits from Great Depression and World War II led to sharp rises in income taxes that only came fully back down in the 1980s
19Source: Bradford Tax Institute https://bradfordtaxinstitute.com/Free_Resources/Federal-Income-Tax-Rates.aspx
PresenterPresentation NotesPlease use the sharing tools found via the share button at the top or side of articles. Copying articles to share with others is a breach of FT.com T&Cs and Copyright Policy. Email [email protected] to buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found here. �https://www.ft.com/content/f2b79b3a-8ae5-11ea-9dcb-fe6871f4145a��Thomas Huray at Fulcrum estimates that the direct fiscal measures introduced in the advanced economies so far is equivalent to 6 per cent of 2019 nominal gross domestic product, roughly double the equivalent stimulus from 2008 in the financial crisis. Furthermore, JPMorgan estimates that the expansion in central bank balance sheets may exceed 18 per cent of GDP by 2021, about three times the scale in the financial shock.
https://bradfordtaxinstitute.com/Free_Resources/Federal-Income-Tax-Rates.aspx
-
20
Political extremism: 50% of Americans are facing real hardship, which risks political disasters (think 1930s politics)
-
21
I’m guessing you finished your drink…
-
The Promise of Working From Home Post COVID
Nick Bloom (Stanford)
SIEPR, May 2020
-
Working from home traditionally had a bad image (a 2017 search)
-
24
In 2012 I ran a randomized control trial of WFH on Ctrip, China’s largest travel agency ($15bn on NASDAQ, 40,000 employees)
PresenterPresentation NotesAverage age is 23, 68% are women, and all have high school but only 39% have tertiary education. Only 15% are married, 9% have children.
-
25
CTrip – inside like a typical US office
PresenterPresentation NotesAverage age is 23, 68% are women, and all have high school but only 39% have tertiary education. Only 15% are married, 9% have children.
-
The experiment was public: Chairman James Liang pulls the ball
PresenterPresentation NotesJames is a SIEPR advisory board member
-
27
Individuals randomized home (even birthdays)
Working at home
Working at home Working at home
Working at home
-
First, massive improvement in performance – 13% more output
-20-10
010
20
-20 -15 -10 -5 0 5 10 15 20 25 30 35
% Im
prov
emen
t in
perf
orm
ance
Weeks after the start of the experiment
-
29
Second, quit rates drop by 50%
05
1015
2025
3035
4045
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
Home (♦)
Office (●)
Cum
ulat
ive
quit
rate
(per
cent
)
Weeks after the start of the experiment
-
Third, choice doubled the impact – after the experiment the firm let all employees choose
-100
1020
3040
50
-20 -12 -4 4 12 20 28 36 44 52 60 68 76 84 92 100
% Im
prov
emen
t in
perf
orm
ance During the
experimentCompanyroll-out
Before theexperiment
Weeks after the start of the experiment
-
CTrip increased profits by $2,000 per employee working from home, and actively rolled this out
PresenterPresentation NotesOutput up by 12% as we find, labor (as measured in a Census) would be constant as shift length not changes (in reality workers may work more but less comute so labor time including commute probably less) and capital would fall (as need less office space while equipment about flat). While call centers are not used for standard TFP analysis it is clear any type of productivity measure of this would tend to rise if WFH was rolled-out.
-
Firms with more employees WFH are now less negative on COVID
Source: Decision Maker Panel Survey conducted by the Bank of England, Nottingham University and Stanford University and Bloom, Bunn, Chen,Mizen, Smietanka and Thwaites (2019) and www.decisionmakerpanel.com
-80
-60
-40
-20
0Ex
pect
ed im
pact
of C
ovid
-19
on s
ales
in Q
2 (%
)
Percentage of employees working from home0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-100
http://www.decisionmakerpanel.com/
-
But working from home under COVID is hard for four reasons
-
1) Kids
-
2) Job match – only a third of jobs can (and are) working from home
0102030405060708090100
Accom & FoodRecreational ServicesConstructionWholesale & RetailManufacturingTransport & StorageReal EstateAdmin & SupportOther ServicesProf & ScientificHealthFinance & InsuranceOther ProductionInfo & Comms
All firms
Furloughed
Unable to work (egsick, self isolating)
Working onbusiness premises
Working from home
Percentage of employees
Source: Headline: How many jobs can be done at home” Dingel and Nieman (2020), Covid Journal of Economicshttps://cepr.org/sites/default/files/news/CovidEcon1%20final.pdf, Graph: Decision Maker Panel Survey conducted by the Bank of England,Nottingham and Stanford Universities, and Bloom, Bunn, Chen, Mizen, Smietanka and Thwaites (2019) and www.decisionmakerpanel.com
https://cepr.org/sites/default/files/news/CovidEcon1%20final.pdfhttp://www.decisionmakerpanel.com/
-
3) Space
If you have any photos to share of WFH challenges please send them to me at [email protected] thanks
mailto:[email protected]
-
85.3
8.12.6 1.9 2.1
020
4060
80m
ean
of s
hare
Nev
er
Occ
asio
nally
(e.g
. mon
thly
)
1 to
2 d
ays
per w
eek
3 to
4 d
ays
per w
eek
5+ d
ays
per w
eek
4) Full-time – pre-COVID only 2% of people were full time
Source: BLS data https://www.bls.gov/news.release/flex2.htm
Why? Office time important for1) Creativity2) Motivation3) Loyalty
https://www.bls.gov/news.release/flex2.htm
-
Three key tips of WFH post COVID
1) Part-time – regular 2 days per week at home (e.g. T,Th)
2) Optional – only about 50% of employees want to WFH
3) Privilege – under-performers warned, recalled to the office
PresenterPresentation NotesSource: “How many jobs can be done at home” Jonathan Dingel and Brent Nieman (2020), Covid Journal of Economics https://cepr.org/sites/default/files/news/CovidEcon1%20final.pdf
-
COVID will shrink the modern office, not eliminate it
Pre-COVID 5% of working days from home
During COVID 35% of working days at home
Post COVID I predict about 15% to 20% of working days at home
PresenterPresentation NotesSource: “How many jobs can be done at home” Jonathan Dingel and Brent Nieman (2020), Covid Journal of Economics https://cepr.org/sites/default/files/news/CovidEcon1%20final.pdf
-
Further Reading References
40
COVID related blogs:“COVID induced economic uncertainty”, Scott Baker, Nicholas Bloom, Steve Davis and Stephen Terry, April 2020, https://nbloom.people.stanford.edu/sites/g/files/sbiybj4746/f/covid-induced_economic_uncertainty_4_april_2020.pdf
“The productivity pitfalls of working from home in the age of COVID-19”, Nick Bloom, March 2020https://news.stanford.edu/2020/03/30/productivity-pitfalls-working-home-age-covid-19/
“The bright future of working from home”, Nick Bloom, May 2020https://siepr.stanford.edu/research/publications/bright-future-working-home
“COVID is also a reallocation shock” Jose Barrero, Nicholas Bloom and Steve Davis, May 2020https://bfi.uchicago.edu/working-paper/covid-19-is-also-a-reallocation-shock/
Background:Recent background of slowing US productivity growthhttps://blogs.wsj.com/cio/2018/07/13/in-an-era-of-tech-innovation-whispers-of-declining-research-productivity/
Three Surveys feeding into this briefingwww.decisionmakerpanel.comhttps://www.frbatlanta.org/research/surveys/business-uncertaintyhttps://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurship
Three uncertainty databaseswww.economicuncertainty.comwww.worlduncertaintyindex.comwww.stockmarketjumps.com
https://nbloom.people.stanford.edu/sites/g/files/sbiybj4746/f/covid-induced_economic_uncertainty_4_april_2020.pdfhttps://news.stanford.edu/2020/03/30/productivity-pitfalls-working-home-age-covid-19/https://siepr.stanford.edu/research/publications/bright-future-working-homehttps://bfi.uchicago.edu/working-paper/covid-19-is-also-a-reallocation-shock/https://blogs.wsj.com/cio/2018/07/13/in-an-era-of-tech-innovation-whispers-of-declining-research-productivity/http://www.decisionmakerpanel.com/https://www.frbatlanta.org/research/surveys/business-uncertaintyhttps://siepr.stanford.edu/research/projects/stanford-stripe-study-internet-entrepreneurshiphttp://www.economicuncertainty.com/http://www.worlduncertaintyindex.com/http://www.stockmarketjumps.com/
The Economic Impact of COVID on�the US economy��Nick Bloom (Stanford)��SIEPR, May 2020Warning – depressing economic forecasts. Take precautionsKey PointsSlide Number 4My PhD was on uncertainty – these are incredible figuresCOVID now the largest source of uncertainty for 90% of firmsThis uncertainty is leading firms to abandon earnings guidanceKey PointsOne way to see the damage is the share of daily lost GDP, which has risen to about 30%Happened incredibly fast - went from February 2020, a 60 year low rate of unemployment, to April 2020, an 80 year highCOVID has a hugely varied impact on firms and industriesAlso variation across firms by founder types (driven by industry mix)COVID → “Frankenstein recession”: combines the size of the Great Depression, the speed of Katrina and the reallocation of WWIIHence, I am less optimistic than the marketKey PointsSlide Number 16Key PointsSlide Number 18Deficits from Great Depression and World War II led to sharp rises in income taxes that only came fully back down in the 1980sSlide Number 20Slide Number 21The Promise of Working From Home Post COVID��Nick Bloom (Stanford)��SIEPR, May 2020Working from home traditionally had a bad image (a 2017 search)In 2012 I ran a randomized control trial of WFH on Ctrip, China’s largest travel agency ($15bn on NASDAQ, 40,000 employees)CTrip – inside like a typical US officeThe experiment was public: Chairman James Liang pulls the ballSlide Number 27First, massive improvement in performance – 13% more outputSecond, quit rates drop by 50%Third, choice doubled the impact – after the experiment the firm let all employees chooseCTrip increased profits by $2,000 per employee working from home, and actively rolled this outFirms with more employees WFH are now less negative on COVIDBut working from home under COVID is hard for four reasons1) Kids2) Job match – only a third of jobs can (and are) working from home3) Space4) Full-time – pre-COVID only 2% of people were full timeThree key tips of WFH post COVID���1) Part-time – regular 2 days per week at home (e.g. T,Th)��2) Optional – only about 50% of employees want to WFH��3) Privilege – under-performers warned, recalled to the office���COVID will shrink the modern office, not eliminate it���Pre-COVID 5% of working days from home��During COVID 35% of working days at home ��Post COVID I predict about 15% to 20% of working days at home���Further Reading References