Introduction
Macroeconomics IV
Ricardo J. Caballero
MIT
Spring 2011
R.J. Caballero (MIT)
Introduction
Spring 2011
1 / 43
References
1
2
Caballero, R.J.,“Macroeconomics after the Crisis: Time to deal with the
Pretense-of-Knowledge Syndrome." JEP Fall 2010, 85-102.
Caballero, R.J. "A Caricature (Model) of the World Economy," MIT WP
10-17, November 2010.
R.J. Caballero (MIT)
Introduction
Spring 2011
2 / 43
The Pretense of Knowledge
“Of course, compared with the precise predictions we have learnt to expect in the
physical sciences, this sort of mere pattern predictions is a second best with which
one does not like to have to be content. Yet the danger of which I want to warn is
precisely the belief that in order to have a claim to be accepted as scientific it is
necessary to achieve more. This way lies charlatanism and worse. To act on the
belief that we possess the knowledge and the power which enable us to shape the
process of society entirely to our liking, knowledge which in fact we do not
possess, is likely to make us do much harm.” Hayek (1974)
R.J. Caballero (MIT)
Introduction
Spring 2011
3 / 43
The Pretense of Knowledge
The subject of macroeconmics is inmensily complex (both for researchers and
economic agents!)
The core of the field (DSGE) has become so mesmerized with its own
internal logic that it has began to confuse the precision it has achieved about
its own world with that it has about the real one
The periphery (intersection of macroeconomics and corporate finance) has
been chasing many of the issues that played a central role during the current
crisis...
We have a tension between a type of answer we aspire to but that has limited
connection with reality (the core), and more sensible but incomplete answers
(the periphery).
Solution? Do NOT abandon models... but understand their proper place...
and incorporate complexity (in the sense of extremely complicated) into the
analysis and its limitations
R.J. Caballero (MIT)
Introduction
Spring 2011
4 / 43
Introduction
The aftermath of a huge global financial crisis has left us with a very
confusing economic environment
I want to discuss one organizing framework that has helped me to navigate
through some of this complexity
I will do this by presenting an extremely stylized model of the workings of a
global economy, which I will integrate with a discussion of current events and
facts
The basic perspective is one of the macroeconomics of asset shortages
Much of what I will say has a formal "micro-founded" model in the
background but I do not make any major effort to draw those connections
here. That’s what the rest of the course is for
R.J. Caballero (MIT)
Introduction
Spring 2011
5 / 43
Introduction
The key driving factor in this perspective is economic agents’continuous
struggle to find assets to park financial resources
This struggle comes with euphoria and disappointments, as many of the
"parking lots" are built too quickly or are not of the desired size
There are also global asymmetries, as some countries are endowed with more
empty “land” and growth potential than others
I use this caricature of the world economy to describe several of the main
driving forces behind recent global macroeconomic events
For now, I will focus on financial markets and fiows, and downplay the
important impact of these on the real side.
R.J. Caballero (MIT)
Introduction
Spring 2011
6 / 43
A "Model"
Time is continuous. At each instant:
Ct = θWt .
(1)
There are two assets, A and B, in which to store wealth. There are β of the
former and 1 − β of the latter
Agents want to hold a share α of their wealth in asset A and, (1 − α) in asset
B
Differences between assets stem from agent’s perception or tastes. This is
just a "catch all" reduced form for the many factors that determine portfolio
decisions in reality, that are not purely return-driven
Asset A is in relatively short supply: α > β
R.J. Caballero (MIT)
Introduction
Spring 2011
7 / 43
A "Model"
Aggregate financial wealth is:
Wt = ptA xtA + ptB xtB .
(2)
In equilibrium:
xA
x
R.J. Caballero (MIT)
B
=
β
= 1−β
Introduction
Spring 2011
8 / 43
A "Model"
Aggregate output is exogenous and grows at rate g :
Yt = Y0 e gt .
(3)
A fraction δ of this output is pledgable (i.e., its present value can be used to
produce assets) and the rest is not
Equilibrium in goods and financial markets require that:
Ct (
ptA
ptB
R.J. Caballero (MIT)
β
1−β
= θWt ) = Yt
=
Introduction
α
.
1−α
Spring 2011
9 / 43
A "Model"
Solving out endogenous variables, we have:
ptA
=
ptB
=
α Yt
,
β θ
1 − α Yt
.
1−β θ
(4)
The interest rates, r A and r B , that are consistent with these asset prices and
the standard arbitrage condition are (from r i = (δY + ṗ )/p)
β
α
rtA
= g + δθ
rtB
= rtA + λt
(5)
And the scarcity premium of asset A over asset B is:
λt =
R.J. Caballero (MIT)
δθ
(α − β) > 0
α (1 − α )
Introduction
Spring 2011
10 / 43
Discussion and apology
The distinction between type A and type B assets captures the fact that
there are assets that seem scarcer than others
The reasons for these scarcities are complex and change over time.
During the years between the Nasdaq crash and the recent financial crisis, type
A assets were almost any AAA bond or tranche
When the crisis hit, suddenly only AAA-bonds issued by sovereigns, especially
the U.S., made the type A cut.
On the other extreme, at times, commodities, real estate, or the NASDAQ
may become type A assets
(All the examples in this section have some concept of risk in the background
while the model has none! I apologize but move on...)
R.J. Caballero (MIT)
Introduction
Spring 2011
11 / 43
Disgression: Solow’s insight...
R.J. Caballero (MIT)
Introduction
Spring 2011
12 / 43
Global Force 1: Gradual decline in delta*theta
There has been a sharp rise in the relative income of the largest EM
economies and commodity producers, coupled with their enormous desire to
save for a “rainy day.”
These are economies that have limited capacity to produce financial
instruments (low δ) and most of them have a higher propensity to save than
developed economies (low θ).
This pattern amounts to a decline in the global (income-weighted) δ and θ.
Recall that:
= g + δθ
rtB
=
rtA + λt
δθ
=
(α − β) > 0
α (1 − α )
λt
R.J. Caballero (MIT)
β
α
rtA
Introduction
(6)
(7)
Spring 2011
13 / 43
Global Force 1: Gradual decline in delta*theta
Three effects (present before the crisis):
Very low equilibrium interest rates, rtA and rtB
Risk compression (λ drops), as there is a proportional shift in demand for all
assets which dilutes the relative scarcity of assets type A.
Rise in asset prices ("bubbles," etc.), ptA and ptB
R.J. Caballero (MIT)
Introduction
Spring 2011
14 / 43
Global Force 1: Gradual decline in delta*theta
Asian crisis
Subprime crisis
5
Percent
4
US long real
3
2
1
World-short real
0
-1
08
20
06
20
04
20
02
20
00
20
98
19
96
19
94
19
92
19
19
90
-2
Worlds and US real interest rates, 1990:1 - 2008:2. Data source: IFS, WDI,
OECD & SPF. Authors' calculations. The world short real rate is a
GDP-weighted of the 4-quarter average ex-post 3-months treasury
bill real rates for the G-7 countries.
R.J. Caballero (MIT)
Introduction
Image by MIT OpenCourseWare.
Spring 2011
15 / 43
Global Force 2: Gradual rise in alpha
Not only has the net demand for assets risen over time but also this rise has
been concentrated on AAA-assets (the type A assets of this episode)
In the model, the direct effect of a rise in α is an increase in the price of
assets type A and a decrease in the price of assets of type B, and the
opposite behavior for rates
The combination of forces 1 and 2 through this period led to a generalized
rise in the cap value of both fixed (type A in that period) and variable income
(type B in that period), but with a much stronger rise in the former
R.J. Caballero (MIT)
Introduction
Spring 2011
16 / 43
Global Force 2: Gradual rise in alpha
U.S. Financial Assets
Percentage of GDP
600
500
400
300
200
100
0
Jan- Jan- Jan- Jan- Jan- Jan- Jan- Jan- Jan- Jan- Jan- Jan52
57
62
67
72 77
82
87
92
97
02
07
Debt
R.J. Caballero (MIT)
Introduction
Equity
Spring 2011
17 / 43
Global Force 3: Temporary (artificial?) rise in beta
Forces 1 and 2 led to attempts to "arbitrage" λ by transforming B into A
assets (and partly by transforming non-pledgeable assets into pledgable ones).
During much of the 1990s artificial assets were created in emerging markets
until the sequence of crises starting with the Asian crisis destroyed a large
share of these assets.
The pressure then moved to U.S. assets, and the Nasdaq in particular, which
also culminated with a crash; to then be followed by the financial system’s
rapid rise in the production of AAA tranches from the securitization of lower
quality loans. This also came to an abrupt end during the so called
"subprime" crisis
R.J. Caballero (MIT)
Introduction
Spring 2011
18 / 43
Global Force 3: Temporary (artificial?) rise in beta
Assets
Liabilities
AAA tranche
AA tranche
A tranche
BBB tranche
BBB- tranche
Toxic waste
R.J. Caballero (MIT)
Introduction
Spring 2011
19 / 43
Global Force 3: Temporary (artificial?) rise in beta
A shift toward CDOs
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009*
CDO issuance as a share of corporate debt issuance
R.J. Caballero (MIT)
Introduction
Spring 2011
20 / 43
Global Force 4: Spikes in alpha over beta
From a systemic point of view, this new-found source of triple-A assets was
much riskier than the traditional single-name highly rated bond
The triggering event was the crash in the real estate “bubble” and the rise in
subprime mortgage defaults that followed it. Confidence vanished and
perceived complexity spiked. Eventually, even super-senior tranches were no
longer perceived as invulnerable (A assets turned into B assets).
The underlying structural deficit of safe assets worsened as the newly found
source of triple-A assets from the securitization industry dried up (β
declined), and the spike in perceived uncertainty further increased demand for
these assets (α increased).
During this episode safe interest rates plummeted to record low levels and all
forms of risk-premia (λs) skyrocketed
R.J. Caballero (MIT)
Introduction
Spring 2011
21 / 43
Global Force 4: Spikes in alpha over beta
3-Month Treasury Bill rate
1
5.00
Lehman bankrupcy
Money funds break the buck
AIG bailout
WaMu seized
TARP
US buys bank equity
Fed rescues money
market funds
4.00
3.00
0.9
0.8
0.7
0.6
0.5
Countrywide
rescue
0.4
2.00
Monolines
downgrade
0.3
0.2
1.00
R.J. Caballero (MIT)
Oct-09
Sep-09
Jul-09
Aug-09
Jun-09
Apr-09
May-09
Mar-09
Jan-09
Feb-09
Oct-08
Dec-08
Nov-08
Sep-08
Jul-08
Introduction
Aug-08
Jun-08
Apr-08
May-08
0.1
Mar-08
Jan-08
Feb-08
Dec-07
Oct-07
Nov-07
Sep-07
Jul-07
Aug-07
Jun-07
Apr-07
May-07
Jan-07
Feb-07
0.00
Mar-07
Bear Stearns
collapse
0
Spring 2011
22 / 43
Jan-07
0
R.J. Caballero (MIT)
Introduction
Sep-09
Aug-09
Jul-09
Jun-09
May-09
Apr-09
Mar-09
Feb-09
Jan-09
Dec-08
Nov-08
Oct-08
Sep-08
Aug-08
Jul-08
Jun-08
May-08
Apr-08
Mar-08
Feb-08
Jan-08
Dec-07
Nov-07
Oct-07
Sep-07
Aug-07
Jul-07
Jun-07
May-07
Apr-07
Mar-07
Feb-07
basis points
Global Force 4: Spikes in alpha over beta
450
TED Spread
400
350
300
250
200
150
100
50
Spring 2011
23 / 43
Global Force 4: Spikes in alpha over beta
Initially, the fiight to quality was a boon for money market funds, which
suddenly found themselves facing a herd of new clients. In order to capture a
large share of this expansion in demand from these new clients, some money
market funds began to invest in short-term commercial paper issued by the
investment banks in distress (that is, they found their own temporary
mechanism to transform B into A assets).
This strategy backfired after Lehman’s collapse, when the Reserve Primary
Fund “broke-the-buck” as a result of its losses associated with Lehman’s
bankruptcy. Perceived complexity reached a new level as even the supposedly
safest private funds were no longer immune to contagion. Widespread panic
ensued and led to an even more extreme rise in α/β.
R.J. Caballero (MIT)
Introduction
Spring 2011
24 / 43
Regions
There are limits to how far we can go without referring to the heterogeneity,
both ex-crisis and post-crisis, in the world economy
Next I highlight some of these differences, pointing to their broad
implications rather than focusing on the mechanics of global equilibrium
R.J. Caballero (MIT)
Introduction
Spring 2011
25 / 43
Force 5: Asymmetric delta*theta
One of the key differences between emerging and developed economies is the
institutional development supporting financial markets and contracts. That
is, δ is higher in developed economies than in emerging markets
The acute asset shortage in EMs is reinforced by the high propensity to save
(low θ) of some EMs, in particular from Asia and some commodity producing
economies
As is apparent in the model, these forces lead to lower pledgable return
relative to asset demand in the "South" than in the "North", and hence
justify the seemingly paradoxical direction of net capital fiows from emerging
markets to developed economies in recent years
R.J. Caballero (MIT)
Introduction
Spring 2011
26 / 43
Force 5: Asymmetric delta*theta
Current Account
1.5
% of World GDP
1.1
0.7
0.3
-0.1
-0.5
-0.9
-1.3
-1.7
-2.1
USA
EU
Rest of World
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
R.J. Caballero (MIT)
Introduction
Spring 2011
27 / 43
Force 6: Asymmetric alpha over beta
The relative weakness in financial development of emerging market
economies is particularly severe in the production of type A assets
Other things equal, this asymmetry in α/β means that r A is higher in
developed economies while r B is higher in emerging markets
Given net fiows, this mechanism helps to explain why the typical gross capital
fiows pattern is one in which emerging markets buy "safe" assets from
developed economies, while the latter buy "risky" assets from emerging
markets
The U.S. as a venture capitalist (Gourinchas and Rey)
R.J. Caballero (MIT)
Introduction
Spring 2011
28 / 43
Force 6: Asymmetric alpha over beta
Ratio of debt to equity and FDI
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
19
90
0.00
Gross liabilities of US
R.J. Caballero (MIT)
Introduction
Gross assets of US
Spring 2011
29 / 43
Force 7: Asymmetric growth (decoupling)
While emerging markets typically grow at a faster pace than developed
economies, this gap has become very pronounced in the post-crisis phase
This effect raises the expected return of all emerging market assets over
those in developed economies
This is probably a key factor behind the surge in capital fiows to emerging
markets until very recently (i.e., until the US began to look a lot better)
(Reinforcing this effect, the many adjustments EMs made in response to their
own crisis in previous decades paid off. Developed economy investors took
notice and have began to upgrade type B emerging market assets into type A
assets)
R.J. Caballero (MIT)
Introduction
Spring 2011
30 / 43
Force 7: Asymmetric growth (decoupling)
Equity Prices
(January 4, 2007 = 100)
250
200
FTSE Latin America
150
S&P 500
S&P 500 Financials
100
50
0
2007
2008
2009
2010
Image by MIT OpenCourseWare.
R.J. Caballero (MIT)
Introduction
Spring 2011
31 / 43
Force 8: Transitions
In this new environment of safe assets scarcity, it makes a great difference to
countries, and their sovereigns in particular, whether they are perceived as
primarily type A or type B asset producers
The PIIGS have seen the consequence of the bad transition, from A to B ,
while many emerging markets, such as Indonesia or Chile, are on the other side
of the spectrum
These transitions, when involving a large group of countries, have global
equilibrium consequences:
The CHF and JPY appreciations owe much to the fall in expected return in
other developed economies
The surge in capital fiows to many EMs owes much to the relative weakness of
the developed world
These general equilibrium sources of capital fiows are important to keep in
mind when assessing strengths and weaknesses of particular recoveries
R.J. Caballero (MIT)
Introduction
Spring 2011
32 / 43
Force 8: Transitions
10 - Year Spreads with German Bund
(Basis Points)
180
1000
160
900
140
800
120
700
600
100
500
400
80
60
300
40
200
20
0
Jan 06
Feb 07
Mar 08
Italy
Belgium
Apr 09
May 10
100
0
Jan 06
-100
Feb 07
Mar 08
Spain
Ireland
France
Austria
Apr 09
May 10
Portugal
Greece
Image by MIT OpenCourseWare.
The Irish guarantee.... and now the EFSF guarantee...
R.J. Caballero (MIT)
Introduction
Spring 2011
33 / 43
QE
The ultimate financial goal of such a policy is to reduce r B , as most private
sector produced assets (borrowing by corporations and households) have a
large component of type B assets
In the early stages of QE, r B was targeted directly through the purchase of
MBS and other distressed assets. This "credit-easing" policy was
instrumental in stabilizing the economy, but as the recovery took hold, a
series of political constraints and concerns brought that unorthodox strategy
to an end
The faltering in the recovery during 2010 (post-Euro crisis) was not severe
enough to make it politically feasible to go back to credit-easing policies,
which left the FED and other central banks with the second best policy of
lowering r A (Treasury rates) and hoping that this would indirectly reduce r B
R.J. Caballero (MIT)
Introduction
Spring 2011
34 / 43
QE
Let βΔ denote the purchases of type A assets by the Fed, which reduces the
net supply of these assets faced by the private sector from β to β(1 − Δ)
It is easy to see in the stark model that QE targeted at assets type A have no
effect on r B : Since the share of income invested in assets type B is constant
and the net supply of assets type B is not changed by QE, there is no effect
of the policy on the price and return of this asset. Instead, all that happens is
that p A rises by (approximately) Δ percent, and r A drops correspondingly:
ptA,QE =
α
Yt
≈ (1 + Δ)ptA ,
β (1 − Δ ) θ
(8)
β
rtA,QE − rtA = −δθ Δ.
α
R.J. Caballero (MIT)
Introduction
(9)
Spring 2011
35 / 43
QE
Thus, in order for QE to have an effect on r B , there needs to be a leak out of
demand for assets type A.
As r A drops to extremely low levels, it triggers a search for yield process that
lowers r B . (The fear that this leak would lead to massive capital fiows to EM
led to heated debates!).
Assume that the minimum return investors are willing to accept for assets
type A is r A,min .
Then, there is a maximum QE, Δmax , such that any further increase in QE:
ptA,QE =
α
Yt
;
β(1 − Δmax ) θ
rtA,QE = r A,min
and
ptB ,QE =
1 − α + β(Δ − Δmax ) Yt
;
1−β
θ
R.J. Caballero (MIT)
rtB ,QE = g + δθ
Introduction
1−β
.
1 − α + β(Δ − Δmax )
Spring 2011
36 / 43
QE
Some of this search for yield is concerning, as agents that should not be
holding certain risks begin to do it (this is what caused the demise of Reserve
Primary Fund at the worst point of the subprime crisis).
Initially the search goes to marginally riskier assets, but as the progression
continues the private sector loads increasing amounts of risks into its balance
sheet
In fact, this pattern is already building up, as some pension funds that
traditionally have invested in type A assets are now being forced to move into
type B assets since r A is too low for them to honor their future contingent
liabilities
On net, a good policy, but it has risks... mostly because it pushes the private
sector into taking more risks
R.J. Caballero (MIT)
Introduction
Spring 2011
37 / 43
The near future and some financial policy considerations
At the world level, g , δ, θ, β have declined while α has risen.
The most immediate consequence is an extremely low r A for the few assets
that are considered type A (a few sovereigns and corporations), and an
enormous reluctance to hold macroeconomic risk (a sharp rise in λ).
It is not that interest rates are low in developed economies because central
banks have decided to keep them there. The causality runs the other way
around: they have to set low policy rates because the equilibrium rates are so
low that if they didn’t, the economy would experience strong defiationary
forces (this is by Walras’Law, since an excess demand for assets must mean
an excess supply of goods)
R.J. Caballero (MIT)
Introduction
Spring 2011
38 / 43
The near future and some financial policy considerations
An extreme example of how tough it has gotten for type B assets is in the
market for long-dated macro volatility
Currently, the price for insurance against "Black Swan" type events is so
high, that is pricing in the possibility of an event worst than the great
depression in the next few years
In the model, this price of insurance is exactly λ, once we think of B assets
as those that are particularly exposed to a large systemic event.
R.J. Caballero (MIT)
Introduction
Spring 2011
39 / 43
The near future and some financial policy considerations
50%
Dec 20 SPX variance swap bid
40%
30%
Dec 15 SPX variance swap bid
5-Year SPX realized vol.
20%
10%
10-Year SPX realized vol.
43
Ja
n
48
Ja
n
53
Ja
n
58
Ja
n
63
Ja
n
68
Ja
n
73
Ja
n
78
Ja
n
83
Ja
n
88
Ja
n
93
Ja
n
98
Ja
n
03
Ja
n
08
38
Ja
n
Ja
n
Ja
n
33
0%
Source: Morgan Stanley Quantitative and Derivative Strategies
Image by MIT OpenCourseWare.
This situation is worrisome not only because it refiects a major dislocation,
but also because it provides potentially dangerous incentives for the
distribution of aggregate risk holding (a sort of AIG on steroids).
R.J. Caballero (MIT)
Introduction
Spring 2011
40 / 43
The near future and some financial policy considerations
The counterpart of the tough environment for B assets is the enormous
reward from being a type A asset producer, which is precisely what has
maintained very low deficit funding costs for prime sovereigns.
The knee-jerk policy reaction after a crisis takes place is to attempt to
prevent further private transformations from B to A
For example, by increasing capital requirements
R.J. Caballero (MIT)
Introduction
Spring 2011
41 / 43
The near future and some financial policy considerations
Unfortunately that kind of policies ignores the structural deficit of type A
assets that drives the perverse incentives...
There is a complementary policy reaction which is for the government to
provide (paid) insurance against the panic component of systemic crises
This has the potential to expand the effective supply of A assets (i.e., β) and
thereby reduce the λ that is behind much of the incentives to create
macroeoconomically fragile instruments. (TIC proposal)
When governments cannot offer any credible guarantee, the private sector
has to build up a buffer to prevent panic-driven asset perception swings. But
this is not an objective in itself, rather it is one of the many costs of chronic
fiscal misbehavior.
R.J. Caballero (MIT)
Introduction
Spring 2011
42 / 43
Final remarks
The model has no nominal assets. Adding these would show that defiation is
a complementary mechanism to restore equilibrium in financial markets when
real interest rates do not converge quickly enough (although this Pigou
mechanism may not be strong enough to pull the economy out of a recession)
What makes an asset type A or type B? This varies from time to time.
There are "super"-A assets that remain immutable though crises (such as
U.S., German or Swiss treasuries). On the opposite extreme there are those
that are type B par-excellence, such as the illiquid equity of small caps
The great majority of assets lies in between. "Bond vigilantes" and their
relatives live from and in this region; while hawkish policy reactions are often
motivated in terms of the distance from it. The soundest reforms are those
that shrink the region of instability while maximizing the availability of A type
assets. I’m not sure policymakers are giving much weight to the latter goal...
R.J. Caballero (MIT)
Introduction
Spring 2011
43 / 43
MIT OpenCourseWare
http://ocw.mit.edu
14.454 Economic Crises
Spring 2011
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.