Transcript model - Amazon Web Services

Debt-financed demand percent of aggregate demand
25
20
15
Percent
10
5
0
0
5
Great Depression
including Government
Great Recession
including Government
 10
 15
 20
 25
0
1
2
3
4
5
6
7
8
9
10
11
12
Years since peak rate of growth of debt (mid-1928 & Dec. 2007 resp.)
Modelling Minskys Financial Instability
Hypothesis: From Implicit to Explicit Money
Steve Keen
University of Western Sydney
Debunking Economics
www.debtdeflation.com/blogs
www.debunkingeconomics.com
www.cfesi.org
13
Typical neoclassical forecast in 2007
– OECD Chief Economist Jean-Philippe Cotis 2007
– “the current economic situation is in many ways
better than what we have experienced in years…
– Our central forecast remains indeed quite benign:
• a soft landing in the United States,
• a strong and sustained recovery in Europe,…
• In line with recent trends, sustained growth in
OECD economies would be underpinned by strong
job creation and falling unemployment.” (p. 9)
• Based on OECD “small macroeconomic model”
And all’s well… until 2008
• “the past two decades has seen not only significant
improvements in economic growth and productivity but also
a marked reduction in economic volatility… dubbed "the
Great Moderation.” (Bernanke 2004)
Great
Moderation
to Great since
Recession
US Inflation
and Unemployment
1970
16
15
Inflation
Inflation
Unemployment
Unemployment
14
U-6 Measure
12
Percent
Percent p.a.
10
10
8
5
6
4
02
0
Inflation
Unemployment
U-6 Measure
 52
• Whoa!
• Did anybody
get the
number of
that Black
Swan?…
• Nope:
0 • “Nobody could
have seen it
coming…”
70
74 7620
78 803082 84
92 94 70
96 9880
100 102
104 106
108 110
0 72 10
4086 88
50 90 60
90
100
110
Year
Year
“Nobody could have seen it coming…”
Inflation
Unemployment
• “The Queen asked me: ‘If these
things were so large, how come
everyone missed them? Why did
nobody notice it?’.”
• When Garicano explained that at
“every stage, someone was relying
on somebody else and everyone
thought they were doing the right
thing”, she commented: “Awful.”
Debt to GDP
• As obvious as the nose on a swan’s face to some of us…
• A debt-driven boom and collapse
Why WE did see “It” coming!
• At least 12 anticipated & warned of Great Recession
(Bezemer 2009, 2010, 2011)
• Common themes Bezemer 2009
Analyst
Dean Baker
Wynne Godley
Fred Harrison
Michael Hudson
Eric Janszen
Stephen Keen
Jakob Brøchner Madsen
Kurt Richebächer
Nouriel Roubini
Peter Schiff
Robert Shiller
• “Distinction between financial wealth
and real assets…
• Concern with debt as the
counterpart of financial wealth…
• Growth in financial wealth and the
attendant growth in debt can
become a determinant (instead of an
outcome) of economic growth …
• Recessionary impact of the bursting
of asset bubbles…
• Emphasis on the role of credit
cycles in the business cycle…”
Credit in a Boom/Depression Pair
• Expanding debt in Boom, deleveraging in Depression
US Debt to GDP Ratios
320
300
280
Private Debt
Government Debt
260
240
220
200
180
160
140
120
100
80
60
40
20
0
1920
1930
1940
1950
1960
1970
1980
1990
2000
2010
2020
Change in private debt now
• $4 trillion boost (+28%) 2008; $2.5 trillion cut (-18%) 2010
7
10
7
10
GDP
7
10
GDP plus Change in Debt
7
10
7
including Government Debt
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
7
10
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
+28%
2
1.95
1.9
1.85
1.8
1.75
1.7
1.65
1.6
1.55
1.5
1.45
1.4
1.35
1.3
1.25
1.2
1.15
1.1
1.05
1
-18%
US $ billion
Aggregate Demand in the USA, 1986-2011
Change in private debt “Then”
• $8.5 billion boost (+9%) 1928; $9.2 billion cut (-18%) 1933
US Aggregate Demand GDP 1920-1940
+9%
110000
100000
$ million
90000
80000
70000
60000
40000
1920
1922
1924
1926
1928
1930
Year
-18%
50000
GDP alone
+ Change in Private Debt
+ Change in Public Debt
1932
1934
1936
1938
1940
Change in private debt now
• Slowdown in debt growth explains the inexplicable now…
Great Moderation to Great Recession
15
2008
12.5
10
7.5
• Define debt5
financed
2.5
aggregate demand 0
 2.5
as ChangeInDebt
5
1980
1985
1990
1995
2000
2005
2010
2015
Great Moderation to Great Recession
0
1
2
3
2008
U-3 Unemployment
Debt % Agg. Demand
0 30
25
20
15
4
5
10
5
6 0
0
7
8
5
 10
9
 15
10
11
 20
 25
1975
1980
1985
1990
1995
2000
2005
2010
 30
2015
Debt-financed Percent Demand
• Unemployment
falls when debt
rises…
Unemployment Rate (inverted)
GDP  ChangeInDebt
1975
0
Inflation
Unemployment
Change in private debt “Then”
Roaring Twenties to Great Depression
30
25
20
1930
Inflation
Unemployment
15
10
5
0
0
5
 10
1920
0
2
4
6
8
10
12
14
0
16
18
20
22
24
26
28
1920
1922
1924
1926
1928
1930
1932
1934
1936
Roaring Twenties to Great Depression
1938
1940
1938
0 30
26
22
18
14
10
6
2
2
6
 10
 14
 18
 22
 26
 30
1940
1930
Unemployment
Debt % Agg. Demand
1922
1924
1926
1928
1930
1932
1934
1936
Debt-financed Percent Demand
Debt to GDP
• Same
process
applied in
Roaring
Twenties and
Great
Depression…
The Credit Accelerator
• Since AD = GDP + DDebt
DDDebt
CreditAccelerator 
• Then DAD = DGDP + DDDebt
GDP
– “Credit Impulse” (Biggs et al. 2010)
• Change in GDP dominant factor in change in employment
• But Credit Impulse made this recession “Great”
Acceleration of private debt & change in employment, USA
10
5
0
Percent p.a.
0
5
 10
 15
 20
 25
 30
1955
Acceleration of private debt
Change in Private Employment
1960
1965
1970
1975
1980
1985
Year
1990
1995
2000
2005
2010
2015
The Credit Impulse
• Negative impulse this time worse than Great Depression
 100
10
 50
5
0 0
00
50
5
100
 10
150
-15%
 15
 20
200
-25%
250
 25
Acceleration/Deceleration of Debt
Change in Unemployment
Annual Change in Unemployment & Debt Acceleration
300
 30
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
Year •
Unemployment
Unemployment
Credit Impulse
Credir Impulse
That’s why the US is in a crisis
• Driving asset prices as well as
GDP…
Mortgage debt dynamics in the USA
Mortgage Acceleration & Real House Price Change
20
VB End
2
10
0
0
2
0
4
 10
6
8
1992
Mortgage Acceleration
House Price Change
1994
1996
1998
2000
2002
2004
2006
www.debtdeflation.com/blogs
2008
2010
 20
2012
CPI-adjusted House Price Change percent p.a.
Mortgage Acceleration (percent of GDP) p.a.
4
Why Neoclassical economists didn’t see It coming
• DSGE the dominant modelling framework today
• Like IS-LM before it, non-monetary model
• Seen as superior to IS-LM because it is based on micro:
– “Dynamic Stochastic General Equilibrium” model…
• “simple, analytically convenient, and has largely
replaced the IS-LM model as the basic model of
fluctuations in graduate courses…
• Unlike the IS-LM model, it is formally, rather
than informally, derived from optimization by
firms and consumers.” (Blanchard 2009)
• This is not an advantage: it is instead a fallacy
– Robert Solow on DSGE modelling…
Solow rejects DSGE
• “The prototypical real-business-cycle model goes like
this. There is a single, immortal household—a
representative consumer—that earns wages from
supplying labor. It also owns the single price-taking firm…
• This is nothing but the neoclassical growth model…
• [When I built it] … It was clear … what I thought it did
not apply to, namely short-run fluctuations ... the
business cycle...
• Now ... an article today [on the] 'business cycle' … will be
... a slightly dressed up version of the neoclassical growth
model.
• The question I want to circle around is: how did that
happen?”
Solow: SMD conditions invalidate DSGE
• “Suppose you wanted to defend the use of the Ramsey
model as the basis for a descriptive macroeconomics.
What could you say? ...
• You could claim that … there is no other tractable way to
meet the claims of economic theory.
• I think this claim is a delusion.
• We know from the Sonnenschein-Mantel-Debreu
theorems that…” (Solow 2008)
• Sonnenschein-Mantel-Debreu: demand curve for a single
market can have any (polynomial) shape at all
– Even study of a single market can’t be reduced to
study of a single utility-maximizing agent
– Yet DSGE macro models the whole economy as a
single utility-maximizing agent
SMD: “Anything goes” for market demand curves
• SMD Conditions (Sonnenschein 1973):
– Market demand curves do not obey the „Law of Demand“
– Even if summing „well behaved“ individual demand curves
P
Crusoe
P
q
Friday
P
q
Market
Q
• Proof by contradiction:
– Assume market demand curves do obey Law of Demand
– Derive conditions under which this is true
– Contradict initial assumption
• Therefore they don‘t obey the „Law“ of Demand
Neoclassical reaction
• A very few reacted rationally:
– Alan Kirman 1989
• “If we are to progress further we may well be
forced to theories in terms of groups who have
collectively coherent behavior.
• Thus demand and expenditure functions if they are
to be set against reality must be defined at some
reasonably high level of aggregation.
• The idea that we should start at the level of the
isolated individual is one which we may well have
to abandon.”
• But most didn’t know about these conditions at all…
Mark Thoma in 2010
• Mark Thoma said... “One thing I learned from it is that I
need to read the old papers by Sonnenschein (1972),
Mantel (1974), and Debreu (1974) since these papers
appear to undermine representative agent models…
• I need to learn the full extent to which this work
undermines the whole microfoundations approach
• I didn't understand that extent to which representative
agent models are an analytical convenience to work
around this problem (the DSGE theorists who
understood this kept quiet about it).”
• Some that did—even those that discovered them—
reacted irrationally…
Representative agent madness instead
• Gorman 1953
– “we will show that there is just one community
indifference locus through each point if, and only if,
the Engel curves for different individuals at the same
prices are parallel straight lines…
– The necessary and sufficient condition quoted above is
intuitively reasonable.
• It says, in effect, that an extra unit of purchasing
power should be spent in the same way no matter
to whom it is given.”
• Intuitively reasonable?
– No, it’s intuitively false!
• Real consequence: even behavior of a single market an
emergent phenomenon…
Macro an “emergent property”
• Real meaning of SMD conditions
– Macroeconomic behavior an “emergent property” of
interaction of agents in a complex system
• Cannot deduce behavior of macroeconomy from
behavior of utility-maximizing individuals
• Cannot reduce macroeconomics to “applied
microeconomics”
• But that is what DSGE models do!
• Fallacy of “Strong Reductionism”
– Believe “macroeconomics is applied microeconomics”
– But SMD conditions prove otherwise
• “macroeconomics cannot be applied microeconomics”
Fallacy of Strong Reductionism
• Can’t deduce even market behavior from model of
individual behavior
– Let alone deduce macro behavior from individual
• Common knowledge in real sciences: Anderson, “More is
Different”, Science (1972)
– The behavior of large and complex aggregates of
elementary particles, it turns out, is not to be
understood in terms of a simple extrapolation of the
properties of a few particles.
– Instead, at each level of complexity entirely new
properties appear, and the understanding of the new
behaviors requires research which I think is as
fundamental in its nature as any other.”
Fallacy of Strong Reductionism
• “one may array the sciences … “The elementary entities
of science X obey the laws of science Y”
X
Solid state or many-body physics
Chemistry
Molecular biology
Cell biology
…
Psychology
Social sciences
Y
Elementary particle physics
Many-body physics
Chemistry
Molecular biology
…
Physiology
Psychology
• But this hierarchy does not imply that science X is “just
applied Y”. At each stage entirely new laws, concepts, and
generalizations are necessary, requiring inspiration and
creativity to just as great a degree as in the previous
one. Psychology is not applied biology, nor is biology
applied chemistry.” (Anderson 1972)
• And “macroeconomics is not applied microeconomics”
Macro model must be able to generate Depression
• Minsky 1982
– “Can “It”—a Great Depression—happen again…?
– To answer these questions it is necessary to have an
economic theory which makes great depressions one of
the possible states in which our type of capitalist
economy can find itself…
– The abstract model of the neoclassical synthesis
cannot generate instability.
– When the neoclassical synthesis is constructed, capital
assets, financing arrangements that center around
banks and money creation, constraints imposed by
liabilities, and the problems associated with knowledge
about uncertain futures are all assumed away.
– For economists and policy-makers to do better we
have to abandon the neoclassical synthesis.”
A tentative, but not-bankrupt, alternative
• A Minskian macroeconomics must:
– Treat the economy as inherently monetary;
– Model it dynamically;
– Consider social classes rather than isolated agents;
– Consider rational but not prophetic behavior;
– Have endogenous creation of money by banking sector;
– Give credit and debt have pivotal roles; and
– Be able to generate a Great Depression as a feasible
state of the core model
• First, Minsky’s verbal model…
Minsky’s FIH: dynamic-disequilibrium-debt model
•
•
•
•
Economy in historical time
Debt-induced recession in recent past
Firms and banks conservative re debt/equity, assets
Only conservative projects are funded
– Recovery means most projects succeed
• Firms and banks revise risk premiums
– Accepted debt/equity ratio rises
– Assets revalued upwards…
• “Stability is destabilising”
– Period of tranquility causes expectations to rise…
• Self-fulfilling expectations
– Decline in risk aversion causes increase in investment
– Investment expansion causes economy to grow faster
The Euphoric Economy
• Asset prices rise: speculation on assets profitable
• Increased willingness to lend increases money supply
– Money supply endogenous, not controlled by CB
• Riskier investments enabled, asset speculation rises
• The emergence of “Ponzi” financiers
– Cash flow less than debt servicing costs
– Profit by selling assets on rising market
– Interest-rate insensitive demand for finance
• Rising debt levels & interest rates lead to crisis
– Ponzi “investments” inherently loss-making
– Rising rates make conservative projects speculative
– Non-Ponzi investors sell assets to service debts
– Entry of new sellers floods asset markets
– Rising trend of asset prices falters or reverses
The Assets Boom and Bust
• Ponzi financiers go bankrupt:
– Can no longer sell assets for a profit
– Debt servicing on assets far exceeds cash flows
• Asset prices collapse, increasing debt/equity ratios
• Endogenous expansion of money supply reverses
• Investment evaporates; economic growth slows
• Economy enters a debt-induced recession
– Back where we started...
• Process repeats once debt levels fall
– But starts from higher debt to GDP level
• Final crisis where debt burden overwhelms economy
• My work: converting this from verbal description to
mathematical model…
Theoretical dynamics of debt: Minsky + Circuit
• Monetary model of capitalism built from combination of:
– Goodwin’s growth cycle
– Minsky’s Financial Instability Hypothesis
– Circuit theory of endogenous money creation
• Product: “Monetary Circuit Theory”—MCT
• Physical side: Goodwin put into mathematical form Marx’s
“growth cycle” model in Capital I, Ch. 25:
– “The mechanism of the process of capitalist
production removes the very obstacles that it
temporarily creates. The price of labor falls again to a
level corresponding with the needs of the selfexpansion of capital, whether the level be below, the
same as, or above the one which was normal before the
rise of wages took place…”
Keen 1995 Model Foundations: Nonlinear dynamics
• Inherently cyclical growth (Goodwin 1967, Blatt 1983)
• Capital K determines output Y via the accelerator:
K
1/3
Accelerator
K
1/3
Y
Y
Goodwin's cyclical growth model
Accelerator
1.50
• Y determines employment L via productivity a:
l
/
1
a
r
Y
l
Labour Productivity
/
1
a
r
l
1
/
r
LabourPopulation
Productivity N
.96
"NAIRU"
+
10
*
L
l
WageResponse
/
100
N
r
1
Population
+
Initial Wage
1/S
+
Integrator
L
Employment
Wages
1.25
L
l
1.00
• L determines employment rate l via population N:
.75
PhillipsCurve
dw/dt
l
.50
0
2
4
6
Time (Years)
8
10
• l determines rate of change of wages w via Phillips Curve
+
.96
"NAIRU" 10
WageResponse
*
Pi
*
W
Goodwin's cyclical growth model
1.3
PhillipsCurve
I
dK/dt
dw/dt
1.2
1.1
Wages
+
- Y
w
L
• Integral
of w determines W (given initial value)
1
3
Initial Capital
Initial Wage
dw/dt
+
1/S
+
Integrator
+
1.0
.9
w
1/S
+
Integrator
L
*
W
.8
.7
.9
• Y-W determines profits P and thus Investment I…
Y
W
+
-
Pi
I
dK/dt
• Closes the loop:
.95
1
Employment
1
Initial Capital
dK /dt
1/S
+
+
1.05
Modelling Minsky with Implicit Money…
• Debt essential to introduce Minsky
– “Debt seems to be the residual variable in financing
decisions. Investment increases debt, and higher
earnings tend to reduce debt.” (Fama & French 1997)
– “The source of financing most correlated with
investment is long-term debt… These correlations
confirm the impression that debt plays a key role in
accommodating year-by-year variation in investment.”
(Fama & French 1998)
• Nonlinear investment function of rate of profit:
– Low—invest nothing;
– Medium—invest as much as earn;
– High—invest more than earn
Modelling Minsky with Implicit Money…
• Important (normal) feature of dynamic modelling:
increasing generality of model makes it more realistic
– No need for absurd assumptions to maintain fiction of
equilibrium, coherent micro/macro behaviour, etc.
• Exponential form:
– Investment=Profit at profit rate of 3%
– Investment>Profit at profit rate > 3%
– Investment<Profit at profit rate < 3%
– Slope of change at 3%=2
– Minimum investment –1% output (depreciation)
Modelling Minsky with Implicit Money…
• Investment increases debt; profit decreases it
• Debt rises if investment exceeds profits
• Debt also increases due to interest on outstanding debt…
0.03
Initial Debt
0
Investment
Profit
+
-
1/S
r
+
+
*
/
l
r
Debt
Output
• Profit net of both wages and interest payments:
Profit
Output
• And the whole model is:
+
+
Wages
Interest
Modelling Minsky with Implicit Money…
• Notice
debt
becomes
negative
• Capitalists
accumulate
• Equilibrium
is stable in
Fisher’s
sense…
Capital
Output
Productivity
l
/
r
Employment
Population
l
/
r
Employment Rate
Graphs
Output
Profit
0.03
Initial Debt
0
Investment
Profit
1
+
-
+
+
Wages
Interest
r
+
+
1/S
*
/
l
r
Debt
Output
Debt/Output
Debt
0
0
-25000000000
-1
-2
0
-50000000000
100
200
Time (Years)
300
400
0
200
Time (Years)
400
Modelling Minsky with Implicit Money…
• “we may tentatively assume that, ordinarily and within
wide limits, all, or almost all, economic variables tend, in a
general way, towards a stable equilibrium” (Fisher 1933)
• But this stability is…
– “so delicately poised that, after departure from it
beyond certain limits, instability ensues” (Fisher 1933:
339).
• Start further from equilibrium, system is unstable:
Modelling Minsky with Implicit Money…
• Higher initial
Capital
level of
unemployment
leads to
disaster…
• Inverse tangent
route to chaos
• Existence of
equilibrium
depends on
7.5
initial
5.0
conditions
• Higher initial 2.5
level, apparent 0
0
stability then
collapse…
Output
Productivity
l
/
r
Employment
Population
l
/
r
Employment Rate
Graphs
Investment
Profit
Output
+
+
Wages
Interest
0.03
r
Profit
Debt
*
134
Output
l
/
r
Initial_Population
Debt
Debt/Output
750000
500000
250000
0
50
100
Time (Years)
150
0
50
100
Time (Years)
150
Modelling Minsky with Implicit Money…
• Nonlinear model can be
– Locally stable around equilibrium
• “linear” component of system dominates) but
– Globally unstable
• past a certain range, nonlinear overwhelm linear
• Below one, a^3 is less than a^2 is less than a…
• Above 1, a^3 is bigger than a^2 is bigger than a…
– Start too far from equilibrium, a debt-induced
collapse
• Inspired a “rhetorical flourish” in 1995 paper
Modelling Minsky with Implicit Money…
Output
Output
A Great Moderation?
Followed by a breakdown
• Keen, 1995:
9
1 10
500
• “This
vision of a capitalist economy
with finance requires
8 108
400
us to go beyond that habit of mind
which Keynes
8
6
10
300
described so well,
4 108
200
• the excessive reliance on the (stable)
recent past as a
2 108
100
guide0 to5 the
10 future.
15 20 25 30
250 260 270 280 290 300
Year
Year should warn
• The chaotic dynamics
explored in this paper
Employment
Cycle with Debt a period of relative
Wage Share Cycle
with Debt
us against
accepting
tranquility
in a
1.0
capitalist economy
0.9
0.9
• as anything other than a lull before
the storm.”
0.8
W ages Share of Output
EmploymentRate
O u t[4 4 9 ]=
0.8
0.7
0.6
0
50
100 150 200 250 300
Year
0.7
0.6
0
50
100 150 200 250 300
Year
Modelling Minsky with Implicit Money…
• Finally, government:
– Minsky: government spending works by
• Giving firms a cash flow during slump, thus letting
them pay off their debts;
• Restraining cash flow during boom, thus attenuating
euphoric expectations
– Model: government pays subsidy (can be negative) to
firms, where change in subsidy is a function of the
rate of employment…
– Constant parameters means model government
“resolute” against unemployment
• Actual governments have clearly shifted on this…
dG
L
Y g 
dt
N
Modelling Minsky with Implicit Money…
• Government Subsidy:
– Constant if Unemployment = 5%
– Increasing if Unemployment > 5%
– Reducing if Unemployment < 5%
E_rate
0.95
0
-0.5
Output
Exponential:
x,
y,
slope at (x,y),
min.
0
*
+
+
1/S
• Profit now net of wages, interest, & government subsidy…
Profit
Output
+
+
-
Wages
Interest
G
Modelling Minsky with Implicit Money…
• Cyclical instability
– depending on slope of government reaction function
1.05
Limit Cycle
1.1
1.00
1.0
.95
.9
.90
.8
.85
.7
.80
.6
.725
.85
.975
1.1
.6
0
Wage Share
Employment rate
2.0
1.5
1.0
.5
100
200
300
0
0
400
100
Time (Years)
Debt/Output
4
.80
200
300
400
Time (Years)
Government spending to output
.55
2
.30
0
-2
0
.05
100
200
Time (Years)
300
400
-.20
0
100
200
Time (years)
300
400
Modelling Minsky with Endogenous Money…
• Monetary Foundation Graziani “Circuit Theory” (1989)
– “The starting point of the theory of the circuit, is
that a true monetary economy is inconsistent with the
presence of a commodity money.
– A commodity money is by definition a kind of money
that any producer can produce for himself. But an
economy using as money a commodity coming out of a
regular process of production, cannot be distinguished
from a barter economy.
– A true monetary economy must therefore be using a
token money, which is nowadays a paper currency”
• Endogeneity of money supply well established
– But ignored by neoclassical modellers
Skip Neoclassical attitude to banks
Neoclassical Theory wrong from first principles
• Neoclassical vision of money & debt:
– “Patient agent” lends to “Impatient agent”
– Bank as intermediary
– No change in aggregate demand
– E.g., Krugman trying to explain why distribution of
debt matters—while assuming aggregate level doesn’t:
• “we begin by setting out a flexible-price endowment
model in which “impatient” agents borrow from
“patient” agents, but are subject to a debt limit.”
– A crisis?
• “If this debt limit is, for some reason, suddenly
reduced, the impatient agents are forced to cut
spending” (2010, p. 3)
Neoclassical Theory wrong from first principles
• Patient lends to Impatient
•
•
•
•
•
Patient’s spending power goes down
Impatient’s spending power goes up
No change in aggregate demand
Banks mere intermediaries (ignored in analysis)
Versus reality: new spending power endogenously created
Monetary Circuit Theory
• Basic process of endogenous money creation
• Entrepreneur approaches bank for loan
• Bank grants loan & creates
deposit simultaneously
• Alan Holmes, Senior
Vice-President New
York Fed, 1969:
• “In the real world,
banks extend credit,
creating deposits in
the process, and look
for the reserves
later.” (1969, p. 73)
• New loan puts additional spending power into circulation
• Modeling this using strictly monetary framework:
Explicitly Monetary Minsky Model
• Input financial relations in Table:
Assets
Liabilities
Equity
d
ReservesReserve
  A Loan
F
Firm Deposit Worker Deposit Bank Equity
dt
Lend
-AB
A
FL
d
V
BV 

d
Record
Loan
dt Loan
 L  rA
  VFr G A
Interest
B
dt
BT
d
B

r

F

r

F

H



Pay Interest
B
ddt T L L D D D  B -B
Record
FirmDeposit  A -BB  C  D  E  F  G
BV
FL
d
dt
Wages
C
FL 

 Y  Inv  r  -C
 V  r   L  r 
Consumption
D+E
-D
-E
ddt
WorkerDeposit
 C  D B -F
Repay
Loan
F
d
dt FD   rD  FD  rL  FL   W  L   V  FL    B T  H D   Y  Inv  r 
Record
-F
dt
 V  r   L  r     B  H 

d Money
New
G
DE
d BankEquity  B H
H D  rD  H D  W  L 
dt
Record
dt
 G
H
• System of dynamic equations derived automatically:
• Placeholders replaced by behavioural functions:
Explicitly Monetary Minsky Model
• Coupled with physical output model via
– Price equation (derivation in Keen 2010)
d
1
P
dt
P

1
W
  P 

1    a




• Monetary “Phillips curve” including all 3 factors in Phillips
– Unemployment
– Rate of change of unemployment
– Cost of living adjustment
d
1 d
1 d 

W  W   Ph      

P
dt
 dt
P dt 

Explicitly Monetary Minsky Model
• Full system of 14 coupled differential equations
Financial Sector
FL( t)
BV( t)
d
BV( t)
dt
 RL  r( t)
d
BT( t)
dt
BT( t)
rL FL( t)  rD FD( t)  rD HD( t) 
B

BV( t)



 LC  r( t)
FL( t)



BV( 0)
BV0
BT( 0)
BT0
FL( 0)
FL0
d
FL( t)
dt
 LC  r( t)
d
FD( t)
dt
BV( t )
FL( t)
BT( t)
HD( t)
W ( t)  Yr( t)
rD FD( t )  rL FL( t) 



 P( t )  Yr( t)  Inv  r( t) 
 LC  r( t)
 RL  r( t)
B
W
a( t )
FD( 0)
FD0
d
HD( t)
dt
HD( t)
W ( t)  Yr( t)
rD HD( t) 

W
a( t )
HD( 0)
HD0




 RL  r( t)

 P( t)  Yr( t)  Inv  r( t )






Physical output, labour and price systems
Level of output
Employment
Rate of Profit
Rate of employment
Rate of real economic growth
Kr( t)
Yr( t)
L( t)
Yr0
v
Yr( t)
L( 0)
a( t )

P( t)  Yr( t)  W ( t)  L( t)  rL FL( t )  rD FD( t)
 r( t)
v  P( t)  Yr( t)
d
 ( t)
dt
g( t)
Yr( 0)
 ( t)  [ g( t)  (    ) ]

Inv  r( t)
v



 r( 0)
L0
 r0
 ( 0)
0
g ( 0)
g0
Explicitly Monetary Minsky Model
• Generates both “Great Moderation” & “Great Depression”
Inflation, Unemployment and Debt
25
500
Inflation
Unemployment
Debt to GDP
15
400
10
5
300
0
0
5
200
 10
 15
100
 20
 25
0
10
20
30
40
50
0
60
Debt to GDP Ratio Percent
Inflation & Unemployment Percent
20
Explicitly Monetary Minsky Model
• Fits stylized facts of crisis
Unemployment, Inflation & Debt (smoothed)
15
3
12.5
2.5
7.5
5
2
2.5
0
 2.5
5
1980
0 1.5
Unemployment
Inflation
Debt to GDP
1985
1990
1995
2000
Year
2005
2010
1
2015
Ratio to GDP
Percent
10
Explicitly Monetary Minsky Model
• Approach extensible to multiple commodity model
FinancialSystem
FLA1( t)
d
BR( t)
dt

 RL prA ( t)
d
FLK1( t)
dt
d
FLK2( t)
dt
BR( t)

 RR prK( t)
BR( t)

 RR prK( t)
BR( t)
d
FLC1( t)
dt

 RR prC( t)
BR( t)
d
FLC2( t)
dt

 RR prC( t)
d
FLA1( t)
dt
d
FLA2( t)
dt

 RR prA ( t)

 RR prE( t)

 RL prK( t)
FLC1( t)

 RL prC( t)
FLC2( t)



FLA2( t)
 RL prA ( t)


 RL prE( t)



FLK2( t)

 NM prK( t)
FLC1( t)

 NM prC( t )
FLC2( t)




 NM prC( t )


2 BR( t)

FLK1( t)



 NM prK( t)


 RL prA ( t)
FLE2( t)




2 BR( t)
 RR prE( t)

FLA1( t)
 RL prE( t)



 RL prC( t)
FLE1( t)


 RR prK( t)

 NM prA ( t)

 NM prE( t)
FLA2( t)

 RL prA ( t)


FLC1( t)

 RL prC( t)


FLC2( t)

 RL prC( t)

FLE1( t)


 RL prE( t)

FLE2( t)


 RL prE( t)


FLK1( t)

 RL prK( t)

FLK2( t)


 RL prK( t)







FLA2( t)
FLE2( t)


 NM prA ( t)
 NM prE( t)
2 BR( t)
 RR prA ( t)

FLA1( t)
FLE1( t)






FDA1( t )
FDC1( t)
FDE1( t)
FDK1( t)
FDK2( t)
FLK1( t)
FLK1( t)
BI( t)
FDA1( t)
FDC1( t)
FDE1( t)
FDK1( t)
FDK1( t)
FDK1( t)
FDK1( t)
FDK2( t)
HD( t)
 rL FLK1( t)  LK1( t)  W M ( t ) 
















 FDK1( t)  rD FDK1( t)  KA  LK1( t)  W M ( t)  KC LK1( t)  W M ( t)  KE LK1( t)  W M ( t)
 RR prK( t)
 pr prA ( t)
 pr prC( t)
 pr prE( t)
 pr prK( t)
 pr prK( t)
 RL prK( t)
 NM prK( t)
2  KBC
 KAC
 KCC
 KEC
 CKA
 CKC
 CKE
 KKC
 KKC
2  KWC


BR( t)
 RR prK( t)
d
FDC1( t)
dt
 RR prC( t)
d
FDC2( t)
dt
 RR prC( t)
d
FDA1( t)
dt
 RR prA ( t)

BR( t)

BR( t)


 RR prA ( t)
d
FDE1( t)
dt
 RR prE( t)
d
FDE2( t)
dt
 RR prE( t)





FDE2( t)


FDK1( t)


FDK2( t)


FLK2( t)


FLK2( t)
BI( t)
FDA2( t)
FDC2( t)
FDE2( t)
FDK2( t)
FDK2( t)
FDK2( t)
FDK1( t)
FDK2( t)
HD( t)




FDC1( t)
FLC1( t )
FLC1( t)
BI( t)
FDA1( t)
FDC1( t)
FDC1( t)
FDC2( t)
FDC1( t)
FDE1( t)
FDC1( t)
FDK1( t)
HD( t)
 rL FLC1( t)  LC1( t)  W M ( t) 












 FDC1( t)  rD FDC1( t )  AC LA1( t)  W M ( t)  CA  LC1( t)  W M ( t)  CC LC1( t)  W M ( t)  CC LC2( t)  W M ( t)  CE LC1( t)  W M ( t)  EC LE1( t)  W M ( t)  KC LK1( t)  W M ( t)
 pr prC( t)
 RL prC( t)
 NM prC( t)
2  CBC
 CAC
 CCA
 CCC
 CCC
 CCE
 CEC
 KCC
 CKC
2  CWC
BR( t)
BR( t)

FDC2( t)




 rL FLK2( t)  LK2( t)  W M ( t ) 
















 FDK2( t)  rD FDK2( t)  KA  LK2( t)  W M ( t)  KC LK2( t)  W M ( t)  KE LK2( t)  W M ( t)
 pr prA ( t)
 pr prC( t)
 pr prE( t)
 pr prK( t)
 pr prK( t)
 RL prK( t)
 NM prK( t)
2  KBC
 KAC
 KCC
 KEC
 CKA
 CKC
 CKE
 KKC
 KKC
2  KWC
BR( t)
BR( t)

FDA2( t )

d
FDA2( t)
dt
d
BI( t)
dt
FLK2( t)


d
FDK2( t)
dt
d
HD( t)
dt




BR( t)
FLK1( t)




 RL prK( t)




d
FDK1( t)
dt


BR( t)
BR( t)
d
FLE2( t)
dt

 RR prA ( t)
 RR prE( t)

 RR prC( t)

BR( t)
BR( t)
d
FLE1( t)
dt
2 BR( t)






 rL FLC2( t)  LC2( t)  W M ( t) 



FDC2( t)

 pr prC( t)






FLC2( t )

 RL prC( t)













FLC2( t)

 NM prC( t)


BI( t)
2  CBC

FDA2( t)
 CAC

FDC2( t)
 CCA

FDC1( t)
 CCC
FDC2( t)

 CCC

FDC2( t)
 CCE

FDE2( t)
 CEC

FDC2( t)
 KCC

FDK2( t)
 CKC

HD( t)
2  CWC




 FDC2( t)  rD FDC2( t )  AC LA2( t)  W M ( t)  CA  LC2( t)  W M ( t)  CC LC1( t)  W M ( t)  CC LC2( t)  W M ( t)  CE LC2( t)  W M ( t)  EC LE2( t)  W M ( t)  KC LK2( t)  W M ( t)
FDA1( t)
FLA1( t)
FLA1( t)
BI( t)
FDA1( t)
FDA2( t)
FDA1( t)
FDC1( t)
FDA1( t )
FDE1( t)
FDA1( t)
FDK1( t)
HD( t)
 rL FLA1( t)  LA1( t)  W M ( t) 












 FDA1( t)  rD FDA1( t)  AA  LA1( t)  W M ( t)  AA  LA2( t)  W M ( t )  AC LA1( t)  W M ( t)  CA  LC1( t)  W M ( t)  AE LA1( t)  W M ( t)  EA  LE1( t)  W M ( t)  KA  LK1( t)  W M ( t)
 pr prA ( t)
 RL prA ( t)
 NM prA ( t)
2  CBA
 CAA
 CAA
 CAC
 CCA
 CAE
 CEA
 KAC
 CKA
2  CWA


FDA2( t)


FLA2( t)


FLA2( t)
BI( t)
FDA1( t)
FDA2( t)
FDA2( t)
FDC2( t)
FDA2( t )
FDE2( t)
FDA2( t)
FDK2( t)
HD( t)




 rL FLA2( t)  LA2( t)  W M ( t) 












 FDA2( t)  rD FDA2( t)  AA  LA1( t)  W M ( t)  AA  LA2( t)  W M ( t )  AC LA2( t)  W M ( t)  CA  LC2( t)  W M ( t)  AE LA2( t)  W M ( t)  EA  LE2( t)  W M ( t)  KA  LK2( t)  W M ( t)
 pr prA ( t)
 RL prA ( t)
 NM prA ( t)
2  CBA
 CAA
 CAA
 CAC
 CCA
 CAE
 CEA
 KAC
 CKA
2  CWA


 rL FLE1( t)  LE1( t)  W M ( t) 

 rL FLE2( t)  LE2( t)  W M ( t) 
FDE1( t)

 pr prE( t)
FDE2( t)

 pr prE( t)






FLE1( t)

 RL prE( t)
FLE2( t)

 RL prE( t)






FLE1( t)

 NM prE( t)
FLE2( t)

 NM prE( t)





BI( t)
2  CBE
BI( t)
2  CBE


FDA1( t)
 CAE
FDA2( t)
 CAE


FDE1( t)
 CEA
FDE2( t)
 CEA


FDC1( t)
 CCE
FDC2( t)
 CCE


FDE1( t)
 CEC
FDE2( t)
 CEC


FDE1( t)
 CEE
FDE1( t)
 CEE


FDE2( t)
 CEE
FDE2( t)
 CEE


FDE1( t)
 KEC
FDE2( t)
 KEC


FDK1( t)
 CKE
FDK2( t)
 CKE
HD( t)
HD( t)
HD( t)
HD( t)
LA1( t)  W M ( t)  LA2( t)  W M ( t)  LC1( t)  W M ( t)  LC2( t)  W M ( t)  LE1( t)  W M ( t)  LE2( t)  W M ( t)  LK1( t)  W M ( t)  LK2( t)  W M ( t) 



 HD( t)  rD HD( t)
 CWA
 CWC  CWE  KWC



HD( t)
2  CWE
HD( t)
2  CWE




 FDE1( t)  rD FDE1( t)  AE LA1( t)  W M ( t)  EA  LE1( t)  W M ( t)  CE LC1( t)  W M ( t)  EC LE1( t)  W M ( t)  EE LE1( t)  W M ( t)  EE LE2( t)  W M ( t)  KE LK1( t)  W M ( t)
 FDE2( t)  rD FDE2( t)  AE LA2( t)  W M ( t)  EA  LE2( t)  W M ( t)  CE LC2( t)  W M ( t)  EC LE2( t)  W M ( t)  EE LE1( t)  W M ( t)  EE LE2( t)  W M ( t)  KE LK2( t)  W M ( t)

BI( t )
BI( t)
BI( t)
BI( t)
rL FLA1( t)  rL FLA2( t)  rL FLC1( t)  rL FLC2( t)  rL FLE1( t)  rL FLE2( t)  rL FLK1( t)  rL FLK2( t) 



 FDA1( t)  rD FDA1( t)  FDA2( t)  rD FDA2( t)  FDC1( t)  rD FDC1( t)  FDC2( t)  rD FDC2( t)  FDE1( t)  rD FDE1( t)  FDE2( t)  rD FDE2( t)  FDK1( t)  rD FDK1( t)  FDK2( t)  rD FDK2( t)  HD( t)  rD HD( t)
 CBA
 CBC  CBE  KBC

Production system
Capital 1
KK1( 0)
Capital Stock
d
KK1( t)
Capital 2
KK10
KK2( 0)
FDK1( t)
   KK1( t)
d
KK2( t)
KK20
FDK2( t)
   KK2( t)

















0
Explicitly Monetary Minsky Model
• Basic system generates multisectoral limit cycle
2
20
The Rate of Profit in a Monetary Multisectoral Model of Production
25
30
35
t
Change in Nominal Credit and Nominal GDP
15
50
40
GDP
Debt
Percent change p.a.
Profit/Capita (Percent)
10
100 prK( t )
100 prC( t )
100 prA( t ) 5
100 prE( t )
0
Capital Goods
Consumer Goods
Agriculture
Energy
5
0
20
40
60
t
Years
80
100
40
30
30
20
20
10
10
20
25
30
35
0
40
Explicitly Monetary Minsky Model
• Monetary and income distribution dynamics
Income Distribution Limit Cycles
Bank Assets & Liabilities
8
100
7
110
Loans
Deposits
Bank Reserves (RHS)
110
110
7
6
110
6
5
110
5
20
25
30
35
Wages Share of Output
110
95
30
Wages
Profit
Interest
25
90
20
85
15
80
10
75
5
70
0
65
5
60
 10
10000
55
94
96
98
100
Employment Rate
• Minsky modelling approach clearly “works”
– How to make it more accessible?
• INET funding to develop GUI program “Minsky”
102
 15
104
Capitalist & Banker Shares
110
Making Monetary Dynamics Accessible
• Prototype “QED” already available
– http://www.debtdeflation.com/blogs/qed/
Minskian Prognosis
• Deleveraging till Ponzi debt overhang eliminated
– (USA, 50-100% of GDP from 300% peak)
• Continued shortfall of aggregate demand
• Government deficits attenuate decline
• But less effective given private sector deleveraging
– However austerity will make it worse
• Private debt abolition a better policy (Hudson, Graeber)
– Long term decline from honouring debts that were
dishonourably created
• We are in a Great Depression
• And bad economic thinking helped us get here…
Minskian Prognosis
• “To conclude, evidence-based macro research needs to
replace faith-based models.
• Theory needs to be applied in a less heavy handed and
exclusionary manner, and data should be used to
discriminate between theories.” (Muellbauer 2010, p. 27)
• Amen!
References
•
•
•
•
•
•
•
•
•
•
•
•
•
Anderson, P. W. (1972). "More Is Different." Science 177(4047): 393-396.
Bezemer, D. J. (2009). ““No One Saw This Coming”: Understanding Financial Crisis Through
Accounting Models.” Groningen, The Netherlands, Faculty of Economics University of Groningen.
Blatt, J. M. (1983). Dynamic economic systems : a post-Keynesian approach. Armonk, N.Y, M.E.
Sharpe.
Bezemer, D. J. (2010). "Understanding financial crisis through accounting models." Accounting,
Organizations and Society 35(7): 676-688.
Clark, J. B. (1898). "The Future of Economic Theory." The Quarterly Journal of Economics 13(1): 114.
Fama, E. F. and K. R. French (1999). "The Corporate Cost of Capital and the Return on Corporate
Investment." Journal of Finance 54(6): 1939-1967.
Fama, E. F. and K. R. French (2002). "Testing Trade-Off and Pecking Order Predictions about
Dividends and Debt." Review of Financial Studies 15(1): 1-33.
Friedman, M. (1969). The Optimum Quantity of Money. The Optimum Quantity of Money and Other
Essays. Chicago, MacMillan: 1-50.
Goodwin, R. (1967). A growth cycle. Socialism, Capitalism and Economic Growth. C. H. Feinstein.
Cambridge, Cambridge University Press: 54-58.
Keen, S. (1995). "Finance and Economic Breakdown: Modeling Minsky's 'Financial Instability
Hypothesis.'." Journal of Post Keynesian Economics 17(4): 607-635.
Keen, S. (2010). "Solving the Paradox of Monetary Profits." Economics: The Open-Access, Open
Assessment E-Journal, 4 (2010-31)
Keen, S. (2011). "A monetary Minsky model of the Great Moderation and the Great Recession."
Journal of Economic Behavior & Organization In Press, Corrected Proof.
Kirman, A. (1989). "The Intrinsic Limits of Modern Economic Theory: The Emperor Has No
Clothes." Economic Journal 99(395): 126-139.
References
•
•
•
•
•
•
•
•
•
•
•
Lucas, R. E., Jr. (1972). Econometric Testing of the Natural Rate Hypothesis. The Econometrics of
Price Determination Conference, October 30-31 1970. O. Eckstein. Washington, D.C., Board of
Governors of the Federal Reserve System and Social Science Research Council: 50-59.
Lucas, R. E., Jr. (2004). "Keynote Address to the 2003 HOPE Conference: My Keynesian
Education." History of Political Economy 36: 12-24.
Kydland, F. E. and E. C. Prescott (1990). "Business Cycles: Real Facts and a Monetary Myth."
Federal Reserve Bank of Minneapolis Quarterly Review 14(2): 3-18.
Marx, K. and F. Engels (1885). Capital II. Moscow, Progress Publishers.
Muellbauer, J., (2010) “Household decisions, credit markets and the macroeconomy: implications
for the design of central bank models”, BIS Working Papers No 306
Minsky, H. P. (1982). Can "it" happen again? : essays on instability and finance. Armonk, N.Y., M.E.
Sharpe.
Samuelson, P. A. and W. D. Nordhaus (2010). Microeconomics. New York, McGraw- Hill Irwin.
Schumpeter, J. A. (1934). The theory of economic development : an inquiry into profits, capital,
credit, interest and the business cycle. Cambridge, Massachusetts, Harvard University Press.
Shafer, W. and H. Sonnenschein (1993). “Market demand and excess demand functions”. Handbook
of Mathematical Economics. K. J. Arrow and M. D. Intriligator, Elsevier. 2: 671-693.
Sonnenschein, H. (1973). "Do Walras' Identity and Continuity Characterize the Class of Community
Excess Demand Functions?" Journal of Economic Theory 6(4): 345-354.
Varian, H. R. (1984, 1992). Microeconomic analysis. New York, W.W. Norton.