Escaping the Last Malthusian Trap A talk by Eric Beinhocker INET Oxford Copyright © 2014 Eric Beinhocker All rights reserved London 6 February 2014 iied
Dec 05, 2014
Escaping the Last
Malthusian Trap
A talk by Eric Beinhocker INET Oxford
Copyright © 2014 Eric Beinhocker
All rights reserved
London 6 February 2014
iied
A complexity economics view
of growth
Why neoclassical economics is the wrong tool for climate change
Escaping the trap: creaAng a revoluAon in carbon
producAvity
The last Malthusian trap
Today’s discussion
A complexity economics view
of growth
Why neo-‐classical economics is the wrong tool for climate change
Escaping the trap: creaAng a revoluAon in carbon
producAvity
The last Malthusian
trap
Some 2.5 million years of economic history (in brief)
Source: US Census Bureau Historical EsHmates of World PopulaHon; Kremer (1993)
World populaAon Thousands
0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 7,000,000
-2,500,000 -2,000,000 -1,500,000 -1,000,000 -500,000 0 500,000 Year
World GDP per capita 1990 internaHonal dollars
Source: DeLong (2005); data 2.5 million to 1 million B.C. extrapolated
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000
-2,500,000 -2,000,000 -1,500,000 -1,000,000 -500,000 0 500,000 Year
The Malthusian trap (circa 1000 BC to 1800 AD)
Rising populaAon Thousands
Source: US Census Bureau Historical EsHmates of World PopulaHon; Kremer (1993)
900 800
600 500 400 300
700
200 100
0 500 1000 1500 1800 AD 1800 AD
Stagnant incomes Global income per person (1800 AD = 1)
Source: Clark (2007)
12
10
8
6
4
2
1000 BC 500 BC 0 500 1000 1500
Malthusian trap
0
PopulaHon
A
B C
Subsistence
Wages
Malthus in a nutshell
UnHl 1800 Malthus ruled…
12
10
8
6
4
2
0
1000 BC 500 BC 0 500 1000 1500 2000 AD
Malthusian trap
Source: Clark (2007)
Global income per person (indexed 1800 AD = 1)
The Great
Divergence
Industrial RevoluHon
…then a third of the world escaped
Source: IPCC AR4 WG1 (2007)
… but with an unsustainable growth model …
10,000 5000 0
Time (before 2005)
Changes in greenhouse gases from ice core and modern data
350
300
250
CO2 (ppm) RadiaHve forcing (Wm2)
400
Source: McKinsey Global InsHtute
Annual household disposable income Number of households (millions)
… and another third of the world are poised to escape
100-‐199
40-‐99
25-‐39
Less than 25
200 and above
Thousands RMB, real 2000 2005
1.0
8.8
71.4
107.5
1.6
1.2
10.9
91.3
101.1
2.4
2015
3.4
112.6
75.7
74.2
5.7
3.3
55.1
106.0
74.1
5.5
2025
8.2
214.1
54.1
57.8
19.0
9.5
94.9
93.1
49.9
33.1 500-‐999
200-‐499
90-‐199
Less than 90
1000 and above
CHINA
INDIA
Thousands RMB, real 2000
Expected temperature increase
3.0˚C
2.0˚C
1.8˚C
Probability of temperature increase under 2˚C
15-30%
40-60%
70-85%
Peak at 550 ppm, long-term stabilization 550 ppm Peak at 510 ppm, long-term stabilization 450 ppm Peak at 480 ppm, long-term stabilization 400 ppm
Source: “Taking Stock – Emissions Levels Implied by the Copenhagen Accord,” Project Catalyst, February 2010.
High range of pledges
Low range of pledges
We face our final Malthusian trap Annual emissions implied by Copenhagen Accord pledges (Gt CO2e)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
What we need to do
The world’s to-‐do list Re-‐do the Industrial RevoluHon, creaHng a sustainable economic system TransiHon to a low-‐carbon economy with minimal impact on welfare and growth, especially for the developing world Drive the above with policy – conduct global social engineering on an unprecedented scale
Unfortunately tradiAonal economics ill-‐equipped to answer these quesAons
QuesAons for economics How did the first Industrial RevoluHon occur? How might we create a new one? What are the interacHons between welfare, growth and de-‐carbonisaHon? How do we assess the trade-‐offs? What are the leverage points? How do we avoid unintended consequences? Preserve individual freedom?
and quesHons for economics
A complexity economics view
of growth Escaping the trap: creaAng a revoluAon in carbon
producAvity
The last Malthusian trap
Why neoclassical economics is the wrong tool for climate change
Neoclassical economics cannot explain key characterisHcs of the economy
The economy is viewed as an equilibrium system
The economy is viewed as an equilibrium system but such a system cannot grow explosively, create novelty, nor spontaneously self-‐organize
And such a system cannot just ‘crash’ – as ours has
The accidental history of equilibrium in economics
Neoclassical failure #1: Theory of growth
Source: Bolow (1956), Romer (1996), Nelson (1996), Daly (1999)
Cannot explain the Industrial RevoluHon
No connecHon with the physical world
Y (t) =
Output Capital Knowledge Labour
F (K (t) , A (t) L (t) ) *
Neoclassical failure #2: Human behaviour
Source: Axtel and McRae (2006a), (2006b)
Example Society spends $1 billion today to save 10 lives per year in perpetuity
Social cost of capital equals 5%
ExponenAal answer Cost = $4.76 million
per life saved
Hyperbolic answer Cost = $1 million to $4 million
per life saved
Theory doesn’t match real world behaviour
ExponenHal discounHng
Hyperbolic discounHng
Neoclassical failure #3: Cost-‐benefit analysis
Source: Stein (2006), Nordhaus (2007), Weitzman (2007), Barker (2008)
• Climate uncertainty has fat tails with power law scaling
• Cost-‐benefit analysis typically assumes away the tails
• Would pay a lot to avoid catastrophe, e.g. Weitzman’s ‘Dismal Theorem’
Prof. William Nordhaus Lord (Nicholas) Stern
“Discount rate!” ‘Discount rate!’
Neoclassical failure #4: Time symmetry
Source: Arrow and Fischer (1974), Frederich, Lowenstein, Donohue (2002), Dietz (2007)
Cost-‐benefit analysis and discounAng assume path independence and Ame symmetry
Samuelson : M R S (τ, τ’) independent of C τ’’
But climate effects are highly path dependent and largely irreversible on human Ame scales
Why neo-‐classical economics is the wrong tool for climate change
Escaping the trap: creaAng a revoluAon in carbon
producAvity
The last Malthusian trap
A complexity economics view of growth
AdapHve
Designs and strategies evolve over Hme
System
Macro parerns emerge from micro behavior
Complex
Many interacHng agents and organizaHons of agents
A different explanaHon – the economy is a ‘complex adapHve system’
A paradigm shis
TRADITIONAL ECONOMICS COMPLEXITY ECONOMICS
Economies are closed, staHc, linear systems in equilibrium
Economies are open, dynamic, non-‐linear systems far from equilibrium
Dynamics
Homogeneous agents • Only use raHonal deducHon • Make no mistakes, have no biases • No need to learn
Heterogeneous agents • Mix deducHve/inducHve decisions • Subject to errors and biases • Learn and adapt over Hme
Agents
Treats micro and macroeconomics as separate disciplines
Macro parerns emerge from micro behaviors and interacHons
Emergence
EvoluHonary process creates novelty and growing order and complexity over Hme
No endogenous mechanism for creaHng novelty or growth in order and complexity
EvoluHon
Explicitly accounts for agent-‐to-‐agent interacHons and relaHonships
Assume agents only interact indirectly through market mechanisms
Networks
Long history of evoluHon in economics (and vice versa)
Problems • Driven by a biological metaphor for the economy • Not built on a general computaHonal view of evoluHon
EvoluHon is a search algorithm for ‘fit order’
Create a variety of experiments
VARIATION Select designs that are ‘fit’
SELECTION Amplify fit designs,
de-‐amplify unfit designs
AMPLIFICATION
REPEAT
EvoluHonary search through ‘deducHve-‐Hnkering’
Technologies evolve
Economic evoluAon occurs in three ‘design spaces’
Physical technologies
Social technologies
Business plans
What would economic evoluHon look like?
Increasing variety and complexity
Non-linear wealth creation
Spontaneous self- organization
But we cannot avoid the Second Law of Thermodynamics – economic order does not come for free
Understanding the “mother of all complex systems”
???
A complexity economics view
of growth
Why neoclassical economics is the wrong tool for climate change
The last Malthusian trap
Escaping the trap: creaAng a revoluAon in carbon
producAvity
Industrial revoluAons are producAvity revoluAons
Physical technologies
Social technologies
Business plans
Rapid evoluAon (e.g. “Cambrian explosion”)
Rapid rise in producAvity
How do we evolve higher ‘carbon producHvity’?
Kaya idenAty
F p g e f = * * *
Anthropogenic (CO2 emissions)
PopulaHon GDP per capita
Energy intensity of GDP
Carbon intensity of energy
Carbon producHvity 1
Non-‐energy emissions and other GHGs
$GDP
CO2e e f * + ≈ = ~
Source: Beinhocker, et. al. (2008)
0255075
100125150
2000 2010 2020 2030 2040 2050
0102030405060
2000 2010 2020 2030 2040 2050
To grow the economy and reduce emissions, carbon producHvity must rise 10x to $7,300 per tonne by 2050
Global emissions, tCO2e
Carbon producHvity =
GDP Emissions
World GDP, US$ tn (real 2000)
0
2,000
4,000
6,000
8,000
2000 2010 2020 2030 2040 2050
10x
Carbon producAvity, US$ (real 2000)/tCO2e
7,300
55
20
146
41
+5.6% per year
+3.1% per year
/
-‐2.4% per year
740
Source: Beinhocker, et. al. (2008)
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
-‐2 -‐1 0 1 2 3 4 5
Source: Global Insight; IPCC; McKinsey analysis
Forecast GDP growth rate 2008-‐2050, percent
Carbon producAvity required to reach 20 Gt CO2e by 2050 US$ (real 2000)/tCO2e
Base case forecast
GDP growth required to hit 20Gt at BAU carbon producHvity growth
If emissions are capped, higher economic growth requires higher carbon producHvity
Annual real growth, %
Carbon producAvity required
-‐2 -‐1 0 1 2 3 4 5
870 1,300 2,000 3,100 4,700 7,000 10,500 15,800
Without carbon producHvity growth need to shrink economy by >-‐2% per annum
If we capped emissions and lived at today’s carbon producHvity, there is not much we could ‘afford’
* Emissions from land use change not included ** Based on 10Gt/year sustainable emissions and future populaHon of 10 billion people Source: McKinsey analysis
0
2
4
6
8
10
0 10 20 30 40 50 60 70 80 90 100 110 120 130Years
Index Year 0 = 1
A carbon producHvity revoluHon is required three Hmes faster than the industrial revoluHon
Carbon producHvity growth required 2008–50
US labor producHvity growth 1830–1955
Source: Beinhocker, et. al. (2008)
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
0 5 10 15 20 25 30 35 40 45 50
South Korea
Saudi Arabia Australia
Indonesia
Iran
Venezuela Nigeria
Liberia
Turkmenistan
Qatar
G8+5
Norway
France United Kingdom Japan
Italy
Germany Mexico United States
Canada Brazil India
Austria
South Africa Russia
Saint Kirs and Nevis Switzerland MauriHus
China
Sweden
Sri Lanka
Singapore
Turkey Pakistan
Bangladesh
But no-‐one today is close to required carbon producHvity
Carbon producHvity US$ 000 (PPP)/tCO2e
Average carbon producHvity
Source: WRI CAIT; UNFCCC; Global Insight; McKinsey analysis
Carbon producAvity 2007, 177 countries, all GHGs excluding LULUCF
Prosperity GDP per capita US$ 000 (PPP)
Adjusted for purchasing power parity, 2050 target = $13,300 GDP/tonne
Carbon producHvity has increased over Hme, but not nearly quickly enough
* 5-‐year running average. Emissions data includes CO2 from fossil fuels and cement, with projecHons for CO2 from land use changes and five non-‐CO2 gases (CH4, N2O, HFCs, PFCs, and SF6) Source: IEA, CDIAC, OECD, EPA, CEC, World Bank, US Bureau of Economic Analysis, McKinsey analysis
Technology will help – but we need to accelerate innovaHon and buy Hme
Source: Farmer, et. al. (2013)
Some hypotheses for climate policy
• Climate change is far riskier then convenHonal models lead us to believe – Fat tails, irreversibility, path dependence, etc.
• Carbon prices may be necessary but not sufficient – EffecHveness of price signals in noisy, complex markets – Industrial revoluHon not triggered by spike in labour
costs alone – broad socioeconomic phenomenon • Need to broadly change the “fitness funcHon” of the
economy – RegulaHon, standards (e.g. consumer and worker safety
laws early 20th c.) – Behaviour, social norms (e.g. slavery, smoking)
• Policy and poliHcs for homo realitus vs. homo economicus - The revenge of poliHcal economy and human behaviour
Some hypotheses for climate policy (cont.)
• Social technology innovaHon just as important as physical technology – InsHtuHons (e.g. green banks?) – Laws (e.g. carbon fiduciary responsibility?) – InformaHon (e.g. climate risk disclosure? GDP measures?)
• Must accelerate evoluHonary innovaHon process – VariaHon – dramaHcally increase shots on goal – SelecHon – bias fitness funcHon toward low carbon – AmplificaHon – capital and talent flows to low carbon – CreaHng green innovaHon clusters
• We need to buy Hme for tech progress - Role of natural gas as bridge?
• InternaHonal cooperaHon needs to emerge borom-‐up rather than top-‐down – EvoluHon of trade regime vs. “Rio Dream” and Copenhagen
Summary
Industrial RevoluHon enabled a third of the populaHon to escape the Malthusian trap of poverty, hardship and disease But it created our next, and possibly last, Malthusian trap – climate change Escaping that trap will require a low-‐carbon revoluHon on the scale of the Industrial RevoluHon, but at three Hmes the speed Economic revoluHons are profoundly disequilibrium phenomena – not explained well by neoclassical theory A complex systems view helps us understand the evoluHonary processes that drive disconHnuous innovaHon and growth Climate policies should acHvate and leverage economic evoluHonary processes – policymakers need new ideas, there is much work to do!
‘We cannot solve problems by using the same kind of thinking we used when we created them.’ ALBERT EINSTEIN
Unless we truly understand the system we are dealing with we will fail
We cannot afford to fail
But if we can more deeply understand that system, we just might succeed