Transcript Global Macroeconomic Consequences of Pandemic Influenza

Global Macroeconomic
Consequences of Pandemic
Influenza
Warwick J McKibbin
CAMA, Australian National University &
The Brookings Institution, Washington, DC &
The Lowy Institute for International Policy, Sydney
Prepared for the Elsevier conference on Corporate Pandemic Influenza
Preparedness in the Asia Pacific , Hong Kong 1 December 2008
Based on
McKibbin W. and A. Sidorenko
(2006) “Global Macroeconomic
Consequences of Pandemic
Influenza” Lowy Institute
Analysis.
Structure of the Paper
•
•
•
•
•
•
Introduction
Background on Influenza Pandemics
Macroeconomic Costs of Disease
A Global Economic Model
Modeling the Pandemic shocks
Results
– Core scenarios
– Sensitivity Analysis
• Summary and Conclusions
Overview
• H5N1 Avian influenza might be the source
of the next human influenza pandemic
• What are the economic consequences
globally?
– ADB study (Asia focus)
– CBO study (US focus)
• Enormous Uncertainty but can use
historical experience adjusted for current
conditions
Overview
• Explore 4 possible scenarios based on
pandemics of the 20th century
– Mild (1968 Hong Kong Flu)
– Moderate (1957 Asian Flu)
– Severe (lower estimates of 1918/19 Spanish
Flu)
– Ultra (higher estimates of 1918/19 Spanish
Flu)
Approach
• Following the new literature from Lee and
McKibbin (2003) on SARS
• Translate pandemic scenarios into a series
of shocks and implement them in a global
economic model
– The Asia Pacific G-Cubed multi-country multisector model
Countries
•
•
•
•
•
•
•
•
•
•
•
•
United States
Japan
United Kingdom
Europe
Canada
Australia
New Zealand
China
India
Korea
Taiwan
Singapore
Hong Kong
Malaysia
Thailand
Indonesia
Philippines
Oil Exporting Developing Countries
Eastern Europe and the former Soviet Union
Other Developing Countries
Sectors
•
•
•
•
•
•
Energy
Mining
Agriculture
Durable Manufacturing
Non-Durable Manufacturing
Services
G-Cubed (Asia Pacific) Model
• Estimated dynamic intertemporal model
with Keynesian short-run rigidities
– Adjustment costs in capital accumulation
– Financial capital mobile given risk premia
– Wages adjust slowly given labour market
rigidities
– Financial markets for equity, bonds, money
– Mix of intertemporal optimizing and rule of
thumb decision rules
– Imposition of intertemporal budget constraints
Modeling the
economics of a
Pandemic
Creating the Shocks
• Major shocks:
– Reduction in labour force (due to mortality
and illness, includes carers)
– Increase in business costs (differentiated by
sector);
– Shift in consumer demand (away from
affected sectors);
– Re-evaluation of country risks
Critical assumptions
• Start with US assumptions on
epidemiological outcomes
(study by Meltzer, Cox and Fukuda(1999))
– Attack rate
– Case mortality rate
• Scale for other countries based on a
series of relative indicators
Critical assumptions
• Country specific
– Geography indicator
• Ease of entry, capacity to spread internally
– Factors (air transport data, location in northern or southern
hem, pop density, share of urban pop)
– Health policy indicator
• Resources available
– Per capita health spending, antiviral doses available
– Governance Quality indicator – govt effectiveness,
regulatory quality, corruption control
– Index of financial risk
– Sectoral exposure in services
Epidemiological Assumptions:
Attack Rate
• 1918/19 (USA) - 10% to 40%
• 1957/58 (USA) - 15% to 40%
• 1968/69 (Hong Kong) 10% to 30%
• Metzler US estimate 15% to 35%
• G-Cubed assumption 30%
Epidemiological Assumptions:
Case Fatalities
• 1918/19 (USA) – 0.2% to 4%
• 1957/58 (USA) – 0.04% to 0.27%
• 1968/69 (Hong Kong) 0.01% to 0.07%
• G-Cubed assumption
– Mild
– Moderate
– Severe
– Ultra
0.0233%
0.2333%
1.1667%
2.3333%
Figure 1: Geography Indicator
1.8
1.6
1.4
1.2
Index
1
Index_geo_US
0.8
0.6
0.4
0.2
0
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
Figure 2: Health Policy Indicator
1
0.9
0.8
0.7
Index
0.6
0.5
index_health_policy
0.4
0.3
0.2
0.1
0
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
Figure 3: Mortality Rate Under each scenario
6.0
5.0
4.0
index_mild
index_mod
index_severe
index_ultra
3.0
2.0
1.0
0.0
AUS LDC EEB CAN CHI EUR GBR HON IND
INO JPN KOR OPC MAL NZL
PHI SNG TAI
THA USA
Figure 4: Additional labor force shock due to Absenteeism
1.45
1.40
% adjustment to labor shock
1.35
1.30
index_sickness
1.25
1.20
1.15
1.10
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
Figure 5: Service Sector Exposure in Production
40%
sensitive output relative to total service output
35%
30%
25%
20%
15%
10%
5%
0%
AUS
LDC
EEB
CAN
CHI
EUR GBR HON
IND
INO
JPN
KOR OPC MAL
NZL
PHI
SNG
TAI
THA
USA
Figure 6: Governance Indicator
0.7
0.6
0.5
Index
0.4
index_governance
0.3
0.2
0.1
0
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
Figure 9: Cost shocks - Moderate Scenario
4.500
4.000
3.500
3.000
% change
energy
2.500
mining
agriculture
durable man
2.000
non-durable manufacturing
services
1.500
1.000
0.500
0.000
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
Figure 13: Demand shocks - Moderate Scenario
AUS LDC EEB CAN CHI EUR GBR HON IND INO JPN KOR OPC MAL NZL PHI SNG TAI THA USA
0.000
-0.500
% change
-1.000
-1.500
-2.000
-2.500
-3.000
-3.500
energy
mining
agriculture
durable man
non-durable manufacturing
services
Results
Number of Deaths
(thousands)
35000.0
30000.0
Thousands
25000.0
Mild
Moderate
Severe
Ultra
20000.0
15000.0
10000.0
5000.0
0.0
SA
U
n
pa
a
J
K
U
a
e
ia
ia
lia
es
nd
nd
pe
or
ad
ys
in
es
la
la
ra
ro
p
i
t
n
a
p
n
a
l
a
u
s
a
a
a
g
E
lip
C
do
Ze
Th
Au
in
M
hi
In
S
w
P
e
N
C
na
hi
a
di
In
an
iw
Ta
a
g
re
on
o
K
K
g
on
H
s
C
LD
Global Deaths in millions: Mild (1.4); Moderate (14); Severe (71); Ultra (142)
SU PEC
F
O
EE
Deaths as a proportion of Population
6.000
5.000
4.000
Mild
Moderate
Severe
Ultra
3.000
2.000
1.000
0.000
SA
U
n
pa
a
J
K
U
a
e
ia
ia
lia
es
nd
nd
pe
or
ad
ys
in
es
la
la
ra
ro
p
i
t
n
a
p
n
a
l
a
u
s
a
a
a
g
E
lip
C
do
Ze
Th
Au
in
M
hi
In
S
w
P
e
N
C
na
hi
a
di
In
an
T
w
ai
a
g
re
on
o
K
K
g
on
H
s
C
LD
SU
F
EE
O
C
PE
Table 7: 2006 percentage GDP loss by region
Mild
Moderate
Severe
USA
-0.6
-1.4
-3.0
Japan
-1.0
-3.3
-8.3
UK
-0.7
-2.4
-5.8
Europe
-0.7
-1.9
-4.3
Canada
-0.7
-1.5
-3.1
Australia
-0.8
-2.4
-5.6
New Zealand
-1.4
-4.0
-9.4
Indonesia
-0.9
-3.6
-9.2
Malaysia
-0.8
-3.4
-8.4
Philippines
-1.5
-7.3
-19.3
Singapore
-0.9
-4.4
-11.1
Thailand
-0.4
-2.1
-5.3
China
-0.7
-2.1
-4.8
India
-0.6
-2.1
-4.9
Taiwan
-0.8
-2.9
-7.1
Korea
-0.8
-3.2
-7.8
Hong Kong
-1.2
-9.3
-26.8
LDCs
-0.6
-2.4
-6.3
EEFSU
-0.6
-1.4
-2.9
OPEC
-0.7
-2.8
-7.0
Source: APG-Cubed model version 63A
Ultra
-5.5
-15.8
-11.1
-8.0
-5.7
-10.6
-17.7
-18.0
-16.3
-37.8
-21.7
-10.3
-9.1
-9.3
-13.8
-15.1
-53.5
-12.2
-5.4
-13.6
GDP Change in the Moderate Scenario
2
0
% deviation from base
-2
USA
Japan
Europe
Malaysia
Philippines
Hong Kong
-4
-6
-8
-10
2005
2006
2007
2008
2009
2010
2011
2012
2013
Sensitivity of GDP and Inflation to Demand assumptions
5.0
0.0
-5.0
GDP
GDP2
Inflation
Inflation2
-10.0
-15.0
-20.0
-25.0
SA
U
n
pa
a
J
K
U
a
e
ia
ia
lia
es
nd
nd
pe
or
ad
ys
in
es
la
la
ra
ro
p
i
t
n
a
p
n
a
l
a
u
s
a
a
a
g
E
lip
C
do
Ze
Th
Au
in
M
hi
In
S
w
P
e
N
C
na
hi
a
di
In
an
iw
Ta
a
g
re
on
o
K
K
g
on
H
s
C
LD
SU
F
EE
O
C
PE
Sensitivity of GDP and Inflation to Cost assumptions
2.0
0.0
-2.0
GDP
GDP2
Inflation
Inflation2
-4.0
-6.0
-8.0
-10.0
SA
U
n
pa
a
J
K
U
a
e
ia
ia
lia
es
nd
nd
pe
or
ad
ys
in
es
la
la
ra
ro
p
i
t
n
a
p
n
a
l
a
u
s
a
a
a
g
E
lip
C
do
Ze
Th
Au
in
M
hi
In
S
w
P
e
N
C
na
hi
a
di
In
an
iw
Ta
a
g
re
on
o
K
K
g
on
H
s
C
LD
SU
F
EE
O
C
PE
Summary
• Even a mild pandemic has significant
costs (0.8% of global GDP or $330 billion)
• A repeat of the 1918/19 Spanish flu could
cost up to $4.4 trillion
• The impacts are larger on developing
countries because of larger shocks and
the relocation of international capital flows
to the relative safe havens of the US and
Europe
Summary
• Equity markets fall and bond markets rally
• Inflation may rise or fall depending on the
relative scale of cost increases versus
demand switches
Summary
• Monetary responses matter
– Countries which peg the exchange rate tend
to suffer even more because of a tightening of
policy to maintain the peg
Conclusion
• Predicting the impacts of pandemic
influenza is difficult but the range of
estimates found in this paper suggest that
costs of any outbreak is potentially large
and much larger than the resources
currently being spent globally to tackle the
likely sources of an outbreak
New Research
• New research is under way on evaluating
alternative intervention strategies by
integrating the approach in this paper with
epidemiological models.
• Joint with Josh Epstein at Brookings and a
team headed by Georgiy Bobashev at RTI
International
Background Papers
www.CAMA.ANU.edu.au
www.SENSIBLEPOLICY.com