Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Applic Business cycles and the RBC model Advanced Macroeconomics Micha l Brzoza-Brzezina & Jacek Suda SGH Warsaw School of Economics 1 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Business cycles and the RBC modelAdvanced Macroeconomics
Micha l Brzoza-Brzezina & Jacek Suda
SGH Warsaw School of Economics
1 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
2 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Introduction
In the previous lectures we anaylzed growth models
They describe long-run macroeconomic processes...
...but have nothing to say about short & medium termfluctuations
This lecture:
facts about business cyclesthe Real Business Cycle (RBC) modelsome papers around RBC modelssome codinglots of fun :-)
3 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Motivation
Growth economists often claim that business cycles are not very importantLucas:
”once you have begun to think about economic growth, it is hard
to think about anything else”Indeed fluctuations seem small compared to growth
Figure: US GDP, trend and cycle (HP Filter)
-.06
-.04
-.02
.00
.02
.04
7.5
8.0
8.5
9.0
9.5
10.0
50 55 60 65 70 75 80 85 90 95 00 05 10 15
LOG_GDP Trend Cycle
Hodrick-Prescott Filter (lambda=1600)
Source: Own calculations based on Fred
4 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Motivation cont’d
But from a”here and now” perspective cyclical fluctuations
matter a lot
Figure: GDP growth (yoy) and unemployment rate in Spain
-10.0
-5.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
20
00
Q1
20
00
Q4
20
01
Q3
20
02
Q2
20
03
Q1
20
03
Q4
20
04
Q3
20
05
Q2
20
06
Q1
20
06
Q4
20
07
Q3
20
08
Q2
20
09
Q1
20
09
Q4
20
10
Q3
20
11
Q2
20
12
Q1
20
12
Q4
20
13
Q3
20
14
Q2
20
15
Q1
20
15
Q4
20
16
Q3
20
17
Q2
20
18
Q1
20
18
Q4
20
19
Q3
Unemployment rate GDP (yoy)
Source: Eurostat
5 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Motivation cont’d
And if rates are too anonymous, here is the number of people
Figure: Unemployed in Spain
0
1 000
2 000
3 000
4 000
5 000
6 000
7 0002
00
0Q
1
20
00
Q4
20
01
Q3
20
02
Q2
20
03
Q1
20
03
Q4
20
04
Q3
20
05
Q2
20
06
Q1
20
06
Q4
20
07
Q3
20
08
Q2
20
09
Q1
20
09
Q4
20
10
Q3
20
11
Q2
20
12
Q1
20
12
Q4
20
13
Q3
20
14
Q2
20
15
Q1
20
15
Q4
20
16
Q3
20
17
Q2
20
18
Q1
20
18
Q4
20
19
Q3
Unemployed ('000)
Source: Eurostat
6 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Evolution of business cycle models
A long time ago in a galaxy far, far away: IS-LM (Keynes1936, Hicks 1937)
Then comes Lucas (1974) and says:”No, no, no! It wont’t
work unless its microfounded”
And then come Kydland & Prescott (1982) and say:”Hey,
but we already have a nice microfounded model! Let’s make itstochastic”
And then come the New Keynesians (1990s) and say:”Nice
job, but you forgot about important frictions”
7 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Evolution of thinking about sources of cyclical fluctuations
In the earlier days, no systematic thinking about fluctuations,but rather about particular events:
Keynes: animal spirits responsible for the Great DepressionFriedman & Schwarz: monetary tightening was the culprit
After the microfoundation revolution:
Kydland & Prescott (1982):”Business cycles can be to a large
extent explained by technology shocks”New Keynesians (since 1990s):
”Other shocks (fiscal,
monetary, preference etc.) matter as well”After the financial crisis:
”OMG, financial shocks are so
important!”
8 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
9 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
How to measure business cycle features?
1 Isolate the cyclical component from the data2 Calculate various moments:
1 standard deviations2 correlations3 autocorrelations
10 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
The cyclical component
Popular (but not unique) definition: data component with aperiod from 6 to 32 quarters
How can this be extracted from the data?
Many options. Most popular:
Band-pass filterHoddrick-Prescott filter
11 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
The Band-pass filter
Able to (approximately) extract only selected frequencies fromthe data
US GDP detrended with Baxter-King (1999) filter (6-32 quarters)
-.06
-.04
-.02
.00
.02
.04
7.5
8.0
8.5
9.0
9.5
10.0
50 60 70 80 90 00 10 20
Y Non-cyclical Cycle
Fixed Length Sym m etric (Baxter-King) Filter
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
.00 .05 .10 .15 .20 .25 .30 .35 .40 .45 .50
Actual Ideal
Frequency Response Function
cycles/period
Source: Own calculations
12 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
The Hoddrick-Prescott filter
Hodrick-Prescott (HP) filter very popular in applications(though also criticised, see e.g. Hamilton (2017)
Let yt denote the log of a time series variable, τt its trendcomponent and ct its cyclical component, so that yt ≡ τt + ct .
Then there exists a τt that solves:
min
{T∑t=1
(yt − τt)2 + λ
T−1∑t=2
[(τt+1 − τt)− (τt − τt−1)]2}
The first term penalizes the deviations from trend, the secondthe variability of the trend growth rate.
λ determines the smoothness of the trend
For quarterly data λ = 1600 extracts preriods below/above 32quarters
13 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Exercise: calculate business cycle moments (1)
Calculate basic business cycle moments from US data
Go to https://fred.stlouisfed.org/
Download at quarterly frequency (if data is monthly: Editgraph → Modify frequency) since 1947:
Real Gross Domestic Product: Billions of Chained 2012Dollars, Quarterly, Seasonally Adjusted Annual RateReal Gross Private Domestic Investment: Billions of Chained2012 Dollars, Quarterly, Seasonally Adjusted Annual RateReal Personal Consumption Expenditures: Billions of Chained2012 Dollars, Quarterly, Seasonally Adjusted Annual RateUnemployment Rate: Seasonally AdjustedConsumer Price Index for All Urban Consumers: All Items inU.S. City Average, Seasonally Adjusted
Never forget about seasonal adjustment!!! (Junk in,junk out)
14 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Exercise: calculate business cycle moments (2)
Save in Excell, import to Matlab
Take logs of all series except unemployment rate (it is alreadyin percent, just divide it by 100)
Apply the Hoddrick-Prescott filter (hpfilter.m) or theBand-pass filter (bpass.m) to all series except unemploymentrate (it has no trend)
For the cyclical components and the unemployment ratecalculate:
standard deviationscorrelations with GDP1st order autocorrelations
15 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Moments in the US economy
Main business cycle moments in the US economy
Source: Kydland & Prescott (1982)
16 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Business cycle properties: main takeaways
Consumption in less volatile than GDP
Investments are more volatile than GDP
Both are strongly procyclical
So are hours
Unemployment is countercyclical
Inflation is mildly procyclical
These properties characterize business cycles in mostdeveloped economies
17 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
18 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
The history and idea of RBC
Lucas (1976) critique:
macro models should be microfoundedand should give an explicit role to expectations
Economists start looking for microfounded business cyclemodels
Kydland and Prescott (1982) come up with stochastic variantof the Ramsey economy
The only force to drive the business cycle are stochastictechnology shocks
Expectations are rational
They claim that these shocks are able to explain a domoniantshare of cyclical fluctuations
19 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Why study RBC?
RBC is the cornerstone of modern business cycle analysis
Has all the basic ingredients of a dynamic stochastic generalequilibrium model
And is simple
20 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Is RBC too simple?
The basic RBC model assumes an economy featuring perfectlyfunctioning competitive markets and rational expectations.
It gives primacy to technology shocks as the source ofeconomic fluctuations.
It exhibits complete monetary neutrality and Ricardianequivalence (and sets role for monetary and fiscal policy).
21 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
RBC as benchmark
The RBC model should be seen as a benchmark.
If a model with optimizing agents and instantaneous marketclearing can explain the business cycle, no need forimperfections such as sticky prices to explain macroeconomicfluctuations.
22 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Households - the problem
A representative household maximises lifetime utility
max{lt+i},{ct+i},{kt+i},{it+i}
Et
∞∑i=0
βt+i
[c1−σt+i − 1
1− σ− lt+i
1+ϕ
1 + ϕ
]
subject to the budget constraint
ct + it = wt lt + rtkt−1 + Divt
and the capital accumulation rule
kt = (1− δ)kt−1 + it
The household rents labour and capital to firms and receivesas compensation the real wage wt and the rental rate rt .
23 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Lagrangean
First, to simplify substitute for investment
ct + kt = wt lt + (1 + rt − δ)kt−1 + Divt
where rt − δ is the real interest rate.
The household chooses ct , lt and kt . Write down theLagrangean:
Lt = Et
∞∑i=0
βt+i
[(c1−σt+i − 1
1− σ−
l1+ϕt+i
1 + ϕ
)−λt+i
(ct+i + kt+i − wt+i lt+i − (1 + rt+i − δ)kt−1+i − Divt+i
)]
24 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
First order conditions
ct : ∂L∂ct
= c−σt − λt = 0
lt : do it yourself :-)
kt : do it yourself :-)
Note that additionally we have a transversality condition (TVC)limt→∞λtkt = 0
25 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Equilibrium conditions - Euler equation
λt = Et [λt+1(1 + rt+1 − δ)] substitute for λt to get
c−σt = βEt [c−σt+1(1 + rt+1 − δ)]
This is the Euler equation. It determines the household’sintertemporal choice (how much to consume today, how muchto save).
In equilibrium the disutility from one unit less consumed todayequals expected discounted utility of consuming (1 + rt+1 − δ)units tomorrow.
26 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Equilibrium conditions - consumption vs. leisure
lϕtc−σt
= wt
This equation determines the household’s intratemporalchoice (how much to consume, how much to work).
In equilibrium the utility of one unit more of work should beequal to the utility from consuming the compensation (realwage).
27 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Firms - the problem
A representative firm uses capital and labour hired fromhouseholds to produce a unique good yt .
Its objective is to maximise profits:
Divt = yt − wt lt − rtkt−1
subject to technologyyt = ztk
αt−1l
1−αt
28 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Firms - equilibrium condition for labour
Substitute for production to get:
Divt = ztkαt−1l
1−αt − wt lt − rtkt−1
First order condition for labour is:
lt : δDivtδlt
= (1− α)ztkαt−1l
−αt − wt = 0
In equilibrium the marginal product of labour equals the realwage
To simplify things substitute from production funtion:
(1− α) ytlt = wt
29 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Firms - equilibrium condition for capital
kt−1 : do it yourself
In equilibrium the marginal product of capital equals the realrental rate of capital.
To simplify things substitute from production funtion:
α ytkt−1
= rt
In the RBC setting profits are zero (perfect competition). Tosee this substitute rt and wt into the profit function.
30 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Productivity and market clearing
It is assumed that productivity zt follows an AR(1) process:
zt = exp(εt)zρt−1
where εt is a productivity shock
This is the only stochastic process in the basic RBC model
Together with the internal persistence of the model itgenerates the business cycle
The goods market clears:
ct + it = yt
31 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Equilibrium conditions - summary
Now we have a system of 8 equations with 8 endogeneousvariables (c, r , l , w , k, i , y , z):
c−σt = βEt [c−σt+1(1 + rt+1 − δ)] (1)
lϕtc−σt
= wt (2)
kt = (1− δ)kt−1 + it (3)
ct + it = yt (4)
32 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Equilibrium conditions - cont’d
yt = ztkαt−1l
1−αt (5)
(1− α)ytlt
= wt (6)
αyt
kt−1= rt (7)
zt = exp(εt)zρt−1 (8)
This can be solved
But two problems arise:
equations are non-linearthese are expectational difference equations
33 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
34 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Interest rate and productivity
In the deterministic steady state uncertainty disappears andall variables are constant, for example ct = ct+1 = css .
We assume that in the steady state z is constant and
zss = 1 (9)
From (1) we get
c−σt = βEt [c−σt+1(1 + rt+1 − δ)]
(css)−σ = β(css)−σ(1 + r ss − δ)
Simplifyingr ss = β−1 − (1− δ) (10)
35 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Capital-labour and investment-capital ratios
Substituting form the production function (5) for yt into (7)we get
αztk
αt−1l
1−αt
kt−1= rt
In the steady state
r ss = α(kss)α−1(l ss)1−α = α(kssl ss
)α−1
Rearranging we can obtain the formula for the capital-labourratio
kss
l ss=( r ssα
) 1α−1
(11)
where r ss is given by (10).
36 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Wage and investment-capital ratio
Substituting form the production function (5) for yt into (6)we get
wt = (1− α)ztk
αt−1l
1−αt
lt= (1− α)zt
kαt−1
lαt
which in the steady state becomes
w ss = (1− α)(kssl ss
)α(12)
where kss/l ss is given by (11).
From (3) in the steady state we have
kss = (1− δ)kss + i ss
which simplifies toi ss
kss= δ (13)
37 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Consumption-labour ratio
Substituting from (5) for yt into (4) we get
ct + it = ztkαt−1l
1−αt
which in the steady state becomes
css + i ss = (kss)α(l ss)1−α =(kssl ss
)αl ss
Dividing by l ss
css
l ss+
i ss
l ss=(kssl ss
)αSubstituting for i ss from (13) and rearranging
css
l ss=(kssl ss
)α− δ k
ss
l ss(14)
where kss/l ss is given by (11).38 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Labour
From (2) we havelϕtc−σt
= wt
which in the steady state it becomes
(l ss)ϕ(css)σ = w ss
Substituting for w ss from (12) and for css from (14)
(l ss)ϕ[(kss
l ss
)α− δ k
ss
l ss
]σ(l ss)σ = (1− α)
(kssl ss
)αSolving with respect to l ss we get
l ss =
[(1− α)
(kss
l ss
)α[(kss
l ss
)α− δ kss
l ss
]σ] 1ϕ+σ
(15)
where kss/l ss is given by (11).39 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Remaining variables
To obtain css , note
css =css
l ssl ss (16)
where css/l ss is given by (14) and l ss is given by (15).
We can obtain capital kss from
kss =kss
l ssl ss (17)
where kss/l ss is given by (11) and l ss is given by (15).
To obtain output we use the production function
y ss = (kss)α(l ss)1−α
where kss is given by (17) and l ss is given by (15).
and we can obtain investment form (13)
i ss = δkss
where kss is given by (17).40 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
41 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Non-linear vs. linear models
We have a system of non-linear difference equations
Two options
1 take it as it is and let Dynare linearise it
2 linearise by hand
Linearising on your own is tedious but has some advantages:
1 some parameters may disappear
2 the system is easier to understand
3 solution in Dynare easier (steady state is known)
42 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearisation
Log-linearisation allows to change non-linear equations intolinear equations
This is a valid approximation in the viscinity of a given point(usualy the steady state)!!!
Two steps:
Express variables as log deviation from steady state using theidentity:
xt = x ssxtx ss
= x ss exp(ln xt − ln x ss) = x ss exp xt
43 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Useful tricks
Apply first-order series expansion w.r.t. xt around steadystate, i.e. around xt = 0
Recall: f (xt) w f (xt | x0) + f′(xt | x0) (xt − x0)
This yields (derive #1 and #3 yourself)
xt = x ssexp(xt) ≈ x ss(1 + xt)xtyt = x ssexp(xt)y
ssexp(yt) ≈ x ssy ss(1 + xt + yt)
xat ≈ (x ss)a (1 + ˆaxt)
xat ybt ≈ (x ss)a (y ss)b (1 + axt + byt)
44 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise labour - consumption choice
lϕtc−σt
= wt
In the steady state:
(l ss)ϕ
(csst )−σ= w ss
Let’s log-linearise:
(l ss)ϕ
(css)−σ(1 + ϕlt + σct) = w ss(1 + wt)
Divide by steady state:
ϕlt + σct = wt
45 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise Euler
c−σt = βEt [c−σt+1(1 + rt+1 − δ)]
In the steady state:
1 = β(1 + r ss − δ)
So let’s linearise:
(css)−σ (1− σct) =β (css)−σ Et [(1− σct+1)(1− δ + r ss(1 + rt+1))]
46 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise Euler cont’d
Substitute for r ss :
1− σct = βEt
[(1− σct+1)(1− δ + ( 1
β − 1 + δ)(1 + rt+1))]
1− σct = Et [(1− σct+1)(β − βδ + (1− β + βδ)(1 + rt+1))]
1− σct = Et [(1− σct+1)(1 + (1− β(1− δ))rt+1)] Multiply and
drop higher order terms:
1− σct = 1− σEt ct+1 + (1− β(1− δ))Et rt+1
Rearange terms:
σ(Et ct+1 − ct) = (1− β(1− δ))Et rt+1
47 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise market clearing condition
ct + it = yt
In the steady state:
css + i ss = y ss
Let’s linearise:
css(1 + ct) + i ss(1 + it) = y ss(1 + yt)
Substract steady state equation:
(1− i ss
y ss )ct + i ss
y ss it = yt
48 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise firm’s equilibrium conditions
For labour
(1− α) ytlt = wt
yt − lt = wt
For capital
α ytkt−1
= rt
yt − kt−1 = rt
49 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise production function
yt = ztkαt−1l
1−αt
Do it yourself :-)
50 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise capital accumulation equation
kt = (1− δ)kt−1 + it
Do it yourself :-)
51 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearise shock process
zt = exp(εt)zρt−1
zt = ρzt−1 + εt
52 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Steady state values
Our equations contain steady state ratio i ss
y ss .
These are determined by our parameters
From the Euler equation:
r ss = β−1 − (1− δ)
and from the equilibrium condition for capital:
r sskss = αy ss
kss
y ss=
α
r ss=
α
β−1 − (1− δ)
53 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Steady state values cont’d
From the capital accumulation equation:
δkss = i ss
thusi ss
y ss= δ
kss
y ss=
αδ
β−1 − (1− δ)
54 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Log-linearised system
We now have a system of 8 linear (difference) equations and 8variables
Have to solve it
55 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
56 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Solving linear DSGE models
Solving a DSGE model means changing the system of forwardlooking diference equations that we have ...
... into a VAR system
Dynare will solve it for you
If you want to learn it, there are several techniques for solvingsuch systems
E.g. Blanchard & Kahn (1980), Sims (2002)
57 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Blanchard & Kahn condition
One very important thing is the stability condition
Write the system in state space form:
A1
[Xt+1
EtPt+1
]= A0
[Xt
Pt
]+ γZt+1
where:
Xt : vector n × 1 of state variables (backward-looking)
Pt : vector m × 1 of jumpers (forward-looking)
Zt : vector k × 1 of shocks (with mean equal to 0 every period)
A1, A0: (n + m)× (n + m) matrices
γ: (n + m)× k matrix
58 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Blanchard & Kahn condition cont’d
Assume that A1 is invertible. Then:[Xt+1
EtPt+1
]= A
[Xt
Pt
]+ A−1
1 γZt+1
where A = A−11 A0
The Blanchard-Kahn condition says that for the model tohave a unique solution, the number of eigenvalues of A lyingoutside the unit circle (unstable roots) must equal the numberof forward looking variables (jumpers)
Dynare solves the model and checks fulfilment of theBlanchard-Kahn condition
59 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Calibration
α = α ykky = rk
y
This is the share of capital remuneration in GDP
On which we have statistics: α ≈ 0.33 in the US economy
δ = ik =
iyky
≈ 0.22 = 0.10 in annual terms. Hence, in quarterly
terms δ ≈ 0.025
If we added bonds to the model we could show that β = 1rb
isthe inverse of the real interest rate.
A usual assumption from the RBC literature is 4% annual realinterest rate. In quarterly terms this means β ≈ 0.99.
Calibration of technology to match moments for GDP
60 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Simulations
Go to Matlab, set path to Dynare, open RBC.mod,
Run stochastic simulation (asymptotic) and analyse moments,
How does the model behave compared to the data moments?
Generate a 200-period long stochastic simulation. Plotoutput. Does it resemble business cycles?
Generate impulse response and check how (and understandwhy) technology affects all model variables.
61 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
RBC - impulse response to technology shock
Figure: Effects of a technology shock in the RBC model
5 10 15 20 25 30 35 400
2
4
6
810-3 y
5 10 15 20 25 30 35 400
0.2
0.4
0.6
0.8
1
1.210-3 c
5 10 15 20 25 30 35 40
0
0.005
0.01
0.015
0.02
0.025
0.03i
5 10 15 20 25 30 35 400
0.5
1
1.5
210-3 k
5 10 15 20 25 30 35 40-1
0
1
2
3
410-3 l
5 10 15 20 25 30 35 400
1
2
3
4
5
610-3 z
Source: Own calculations
62 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
RBC - first conclusions
Looks like we have a (very simple) model that reflects mainfeatures of the business cycle
At the same time the model constitutes a usefull benchmark
the model is microfoundedone shock explains all fluctuations
It can be extended in various directions
But even the simple RBC can be used for various purposes
63 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
64 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
The role of Governments
Governments are big players in modern economies
Source: Eurostat
65 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Functions of governments
Main functions of governemnts:
provision of public goodsincome redistributionstabilization policies
66 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Provision of public goods
Public good is something freely available to everyone, e.g.
law and orderstreetlightsdefenceeducationenvironmental protection
These goods could be provided by the private sector, but in anunsatisfactory quality/ quantity/ widespread availability, e.g.:
externalities are not priced-in by the private sectorfree rider problem (e.g. streetlights)
These are so called”market failures”
67 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Externalities
Informally: Effects of an activity that affect other people
Formally (in modeling sense): variables that agents influence,but do not take it into account in the optimization process(e.g. because of very small impact of individual)
Negative externalities:
pollution, noise
Positive externalities:
synergic effects of education
Pricing in externalities: e.g. selling (and trading) rights topollute requires government intervention
Given formal definition DSGE models are well designed todiscuss externalities
68 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Income redistribution
Factors of production are paid their marginal product (?)
But this results in very unequal distribution of wealth andincome ...
... which need not be preferred from the social point of view.
Standard measure: Gini coefficient
69 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Gini coefficient
Standard (but not unique) measure of inequalityEquals (twice) the field under Lorenz curve
Source: http://www.urbansim.org/docs/tutorials/lorenz-curve.pdf
G=0 complete income inequality (one person holdseverything)G=1 complete income equality (everybody has the same)
70 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Income distribution
Source: United Nations Human Development Report
71 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Role of redistribution
Source: Forster and Pearson, “Income Distribution and Poverty in the OECD Area: Trends and Driving Forces”OECD Economic Studies. No. 34, 2001.
72 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Stabilization policies
There are business cycle fluctuations
Which may be perceived as inefficient
The government has policies that can smooth the businesscycle:
fiscal policy (expenditure, taxes)monetary policy
In the context of the RBC model we will speak about fiscalpolicy:
effects of government spending shocksdesireability
73 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Macroeconomic effects of stabilization policies
Important question in macro (at least since Keynes): what isthe effect of a government expenditure shock on the economy
Key concept - fiscal multiplier m = ∆y∆g
In”hardcore” keynesian models multiplier can be very high.
m = 11−MPC . If MPC = 0.8 then m = 5. (see Chapter 5.1 of
Romer textbook)
But this is”ceteris paribus”. In the real world there are many
frictions and adjustments and the multiplier is probably muchlower:
capacity constraintstax (or debt) adjustmentshousehold adjustment of labor supply and savingsmonetary policy reaction
The RBC model can take (some of) them into account
74 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Empirical evidence
Before we check what the RBC model has to say, how aboutempirical evidence?
Not so simple - endogeneity and expectations!!!
One needs identifying restrictions in VARs or exogenousinstruments
Blanchard & Perotti (2002) use institutional informationabout tax, transfer and spending programs in the US
Next they estimate a SVAR model on quarterly US data
75 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Empirical evidence - Blanchard & Perotti (2002)
Tax shocks crowd out consumption and investments
Spending shocks crowd out investment, crowd in consumption
Spending multiplier higher than tax multiplier
Spending multiplier around one
76 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Empirical evidence cont’d
But identifying exogenous spending/ tax shocks is difficult
As well as controling for the effect of expectations
E.g. Ramey (2011) finds:
most components of consumption fall after a positive shock togovernment spendinggovernment spending multiplier is 0.6-1.1
All in all, empirical evidence is far from conclusive
77 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Application of RBC models: government policy
Baxter & King (1993) check what the RBC model has to sayabout effects of government spending
We (they) ask the folowing questions:
what is the effect of a temporary change in governmentexpenditure?what is the effect of a permanent change in governmentexpenditure?how does the financing decission affect the results?
The RBC model is a good laboratory to analyse these issues(provided that we add a government to the model)
78 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Adding government expenditure to the model (lump-sumtaxes)
Assume that there exists a government, which taxeshouseholds (lump-sum) and buys goods
In the simplest variant (used here) they are waisted. But theycan also affect utility or form public capital
Compared with the baseline model, four modifications aremade:
Household budget constraint becomes:ct + it + tt = wt lt + rtkt−1 + DivtResource constraint becomes: ct + it + gt = ytFiscal rule (balanced budget): gt = ttGovernment spending rule (AR(1) process):
gt = gρgt−1 (g y ss)1−ρg exp(εgt )
where tt are lump-sum taxes and gt is governmentexpenditure.
79 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Modified equilibrium conditions
No changes to household or firm FOCs
This is the nature of lump-sum taxes, they do not affectagent’s behavior
Resource constraint becomes: ct + it + gt = yt
Fiscal rule (balanced budget): gt = tt
Government spending rule (AR(1) process):gt = g
ρgt−1 (g y ss)1−ρg exp(εgt )
80 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Simulations -temporary rise
Temporary increase in govermnet spending (four quarter rise)financed by lump-sum taxes
Figure: Effects of a temporary governemmnt spending shock in the RBCmodel
0 5 10 15 20 25 30 35 40-9
-8
-7
-6
-5
-4
-3
-2
-110-4
cons
0 5 10 15 20 25 30 35 40-4
-2
0
2
4
6
810-4
y
0 5 10 15 20 25 30 35 400
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
gov
0 5 10 15 20 25 30 35 40-6
-5
-4
-3
-2
-1
0
110-3
inv
0 5 10 15 20 25 30 35 400
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
labor (%)
0 5 10 15 20 25 30 35 40-0.03
-0.025
-0.02
-0.015
-0.01
-0.005
real wage (%)
Source: Own calculations
81 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Simulations - permanent rise
Permanent increase in govermnet spending financed bylump-sum taxes
Figure: Effects of a permanent governemmnt spending shock in the RBCmodel
0 50 100 150 200-4.6
-4.4
-4.2
-4
-3.8
-3.6
-3.4
-3.2
-3
-2.810-3
cons
0 50 100 150 2004
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
510-3
y
0 50 100 150 2000
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
gov
0 50 100 150 2001.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2
2.110-3
inv
0 50 100 150 2000.15
0.155
0.16
0.165
0.17
0.175
0.18
0.185
labor (%)
0 50 100 150 200-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
real wage
Source: Own calculations
82 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
A nice, little exercise :-)
Code rbc g lumpsum.mod is the RBC model with governmentexpenditure financed by lump-sum taxation
Calculate the i periods ahead government spending multipliermt,i ≡ yt+i−yt
gt+i−gton impact (i.e. one period ahead) in the temporary exerciseon impact and in the long-run (say after 200 periods) in thepermanent exercise
And explain the difference
Now, that you understand why the multiplier is positive, howcan you raise it even further?
83 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Adding government expenditure to the model(distortionary taxes)
Assume now that the government taxes household income andbuys goods
Again, they are waisted.
Compared with the baseline model, three modifications aremade:
Household budget constraint becomes:ct + it = (1− τt) (wt lt + rtkt−1) + DivtResource constraint becomes: ct + it + gt = ytFiscal rule (balanced budget): τt (wt lt + rtkt−1) = gtGovernment spending rule (AR(1) process):
gt = gρgt−1 (g y ss)1−ρg exp(εgt )
where τt is the income tax rate and gt is governmentexpenditure.
84 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Modified equilibrium conditions
This time household FOCs are affected
This is the nature of distortionary taxes, they affect agent’sbehavior
Euler: c−σt = βEt [c−σt+1(1 + (1− τt+1)rt+1 − δ)]
Labor choice: lϕtc−σt
= (1− τt)wt
Resource constraint becomes: ct + it + gt = yt
Fiscal rule (balanced budget): gt = τt (wt lt + rtkt−1)
Government spending rule (AR(1) process):gt = g
ρgt−1 (g y ss)1−ρg exp(εgt )
85 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
And another nice exercise :-)
Code rbc g distortion.mod is the RBC model with governmentexpenditure financed by income taxation
Unfortunately someone deleted the dynamic equation
Restore the code so that the model assumes distortionarytaxation
Run the simulations (temporary and permanent increase ingovernment spending financed with distortionary taxes)
What is the impact on the economy? How does it differ fromthe previous one? Why?
This exercise is decribed in Baxter & King (1993), you cancheck if you get sth similar.
86 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Or a homework :-)
Code rbc g lumpsum.mod is the RBC model with governmentexpenditure financed by lump-sum taxation
And derivations in RBC gov derivations.pdf derive the RBCwith distortionary taxation as well (or maybe you want toderive yourself?)
Modify the code so that the model assumes distortionarytaxation now
Run the simulations (temporary and permanent increase ingovernment spending financed with distortionary taxes)
What is the impact on the economy? How does it differ fromthe previous one? Why?
This exercise is decribed in Baxter & King (1993), you cancheck if you get sth similar.
87 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
88 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Some problems with RBC models
RBC models have some welcome features
But also some unwelcome
possibly overstated role of TFP shocksno role for stabilization policysmall cost of business cycle fluctuationsno role for monetary policyno role for demand shocksvoluntary unemployment
89 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Role of TFP shocks
Our calibration of technology shocks was done to match thes.d. and auroreggression of output with GDP
Let us instead construct a TFP series from the data andcheck its properties
The exercise follows (approximately) Solow (1957) andPrescott (1986)
File data Solow residual.mat contains three series (US:1Q1950 - 3Q2019, seasonally adjusted):
GDPCapital (adjusted for utilization as in Solow 1957)Labor hours in nonfarm business activity
90 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Exercise: role of TFP shocks (1)
Write Matlab code that:
calculates the Solow residual (assume CD production functionwith α = 0.33)calculate its cyclical component (use hpfilter.m)estimate regression: zt = ρzt−1 + εtcalculate standard deviation σ (εt)see next slide for some formulae
Use the estimated ρ and σ (εt) to recalibrate the RBC model
Compare again the moments for output in the model and forGDP (cyclical component) in the data
What fraction of fluctuations can the model explain now?
91 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
92 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Exercise: role of TFP shocks - useful formulae
Some useful formulae:
Solow residual:
Zt = Yt/(Kαt L
1−αt
)OLS estimator:
β =(XTX
)−1XT y , where β is the vector of estimated
parameters, X is the matrix of explanatory and y the vector ofdependent variables
Standard deviation of residuals:
σ(εt) =√
εTt εtn−k−1 where n is sample size and k the number of
explanatory variables
92 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
93 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Exercise: role of TFP shocks (2)
Go back to the Matlab code:
calculate the growth rate of the Solow residual (original, beforedetrending)plot ithow often is it negative?what does this imply in economic terms?
93 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Role of TFP shocks - some conclusions
TFP shocks calculated as Solow residuals do not explain100% of cyclical fluctuations of output (just about 75%)Solow residual declines almost 25% of sampleProbably even Solow residual overstates variability of TFPOther shocks might be necessary to explain business cycles toa satisfactory degreeSolow residual in the US (qoq)
0 50 100 150 200 250 300-0.015
-0.01
-0.005
0
0.005
0.01
0.015
0.02
0.025
Source: Own calculations
94 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Role for stabilization policy
The RBC economy fluctuates around steady state (orbalanced growth path)
Are these fluctuations efficient?
In other words: should policymakers react to them and imposea different allocation than the market does?
Introduce”social planer”: an imaginary agent who maximizes
to welfare instead of agents (decentralize equilibrium)
”Social planer” can for instance help overcome the negative
impact of externalities on welfare
95 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Social planer: problem
The social planer seeks to maximize
max{lt+i},{ct+i},{kt+i},{it+i}
Et
∞∑i=0
βt+i
[c1−σt+i − 1
1− σ− lt+i
1+ϕ
1 + ϕ
]subject to the resource constraint of the economy
yt = ct + it
the capital accumulation rule
kt = (1− δ)kt−1 + it
and technology
yt = ztkαt−1l
1−αt
96 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Social planer: Lagranean
After substituting for investments and output the Lagrangeanis
Lt = Et
∞∑i=0
[βt+i
(c1−σt+i − 1
1− σ−
l1+ϕt+i
1 + ϕ
)−µt+i
(ct+i + kt+i − (1− δ)kt−1+i − zt+ik
αt−1+i l
1−αt+i
)]
97 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Social planer: first order conditions
ct : ∂L∂ct
= βtc−σt − µt = 0
lt : ∂L∂lt
= −βt lϕt + µtzt(1− α)kαt−1l−αt = 0
kt : ∂L∂kt
= −µt + (1− δ)Etµt+1 + Etµt+1zt+1αkα−1t l1−αt+1 = 0
Additionally we have a transversality condition (TVC)limt→∞µtkt = 0
98 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Social planer: equilibrium conditions
Substituting for µt and using production function we get:
lϕt = c−σt (1− α) ytlt
c−σ = Etc−σt+1
(1− δ + α yt+1
kt
)Note that these imply the same allocation as the decentralizedequilibrium.
To see this substitute
(6) into (2)(7) into (1)
Except for prices wt and rt , which are absent in the socialplaner’s problem.
On the top we have the same production function, capitalacumulation law, resource constraint and TFP process.
99 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Social planer: consequences
The decentralized allocation coincides with the socialoptimum:
The government cannot improve upon the market solutionIn particular countercyclical policy can only reduce welfare
This is what the 1st Fundamental Welfare Theorem tells us: acompetitive equilibrium leads to an efficient allocation.
If we study a competitive economy we do not need to considereach agent’s first order conditions. Instead we can use thefirst order conditions of the social planner.
This is a formal statement of Adam Smith’s”invisible hand”
concept.
Of course assumptions need not always be fulfilled:monopolistic power, externalities etc. allow for improvementsupon decentralized equilibrium.
100 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Cost of business cycle fluctuations
Another important topic widely discussed in the literature:how important are BC fluctuations?
Microfounded models have a clear metric to evaluate thisquestion
How much do fluctuations lower a representative household’swelfare?
In other words, how much would the HH like to pay to live ina world w/o fluctuations?
Lucas (1987; 2003) offers a calculation
Simplified variant - allows for analytical solution.
101 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Cost of business cycle fluctuations: intro
Assume consumption (endowment) follows ct = e−σ2
2 εt
where εt ∼ Λ(0, σ2) (i.e. εt follows log-normal distribution)
Preferences are given by CRRA utility funcion:
Et∑∞
t=0 βt c
1−γt
1−γHow much more consumption would agents like to have (in aworld with fluctuations) to be indifferent to a world w/ofluctuations?
Solve for λ:
Et
∞∑t=0
βt[(1 + λ)ct ]
1−γ
1− γ=∞∑t=0
βt[css ]1−γ
1− γ
102 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Cost of business cycle fluctuations: solution (1)
Note that css = 1
And recall :-) that for a log-normal distributionE (xa) = eaµ+a2σ2/2 . Then we have
∞∑t=0
βt(1 + λ)1−γEtc
1−γt
1− γ=∞∑t=0
βt1
1− γ
∞∑t=0
βt(1 + λ)1−γEt
(e−
σ2
2 εt
)1−γ
1− γ=
1
(1− γ) (1− β)
103 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Cost of business cycle fluctuations: solution (2)
∞∑t=0
βt(1 + λ)1−γe−
σ2
2(1−γ)e(1−γ)2σ2/2
1− γ=
1
(1− γ) (1− β)
(1 + λ)1−γe−σ2
2(1−γ)e(1−γ)2σ2/2
(1− γ) (1− β)=
1
(1− γ) (1− β)
(1 + λ)1−γeσ2
2 (1−2γ+γ2−(1−γ)) = 1
(1 + λ)1−γeσ2
2γ(γ−1) = 1
ln(1 + λ)1−γ + ln eσ2
2γ(γ−1) = ln 1
λ ' σ2
2γ
104 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Cost of business cycle fluctuations: result and way forward
So (quite intuitively) compensation λ depends on risk σ2 andrisk aversion γ
US consumption volatility (quarterly, HP detrended) for1959-2020 is σ = 0.0061
λ ' σ2
2 γ = (0.0061)2/2 = 0.000018
Lucas provides somewhat different number based on annualdata & linear trend: λ = 0.0005
Both are negligibly small
Do we miss something?
unemploymentcredit constraints
105 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Other points of critique
So far we have:
TFP shocks are not able to explain 100% of fluctuationsGoverment policy cannot improve upon the decentralized(market) allocationCost of fluctuations is negligible
Some extra points (see Rebelo 2005 for a nice overview andmore details):
What caused the Great Depression? RBC model suggestsnegative technology shocks. Really???What are the effects of monetary policy on the economy? RBCmodel (if MP is added) suggests that it affects only prices.Really?What causes large cyclical swings of labour hours? RBC modelsuggests voluntary decissions (and high elasticity of laborsuppply). Really?
106 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Consequences of the critique
This critique should be taken seriously but not panically
The RBC model should be (and usually is) considered as ausufull benchmark of a frictionless, competitive economy
Some questions can be asked in this framework
Many others require a richer specification (which often buildson the RBC core)
Think of economic modeling (research) in the spirit ofOkham’s razor:
if a small (simple) model can explain the problem then do notlook for a a bigger (more complicated) oneif it cannot, look for the simplest (though realistic)modification that solves the problem (e.g. explains empiricalfacts)
107 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Plan of the Presentation
1 Motivation
2 Business cycles
3 Model
4 Steady state
5 Linearisation
6 Solution, calibration & simulation
7 Application to fiscal policy
8 Critique
9 Conclusions
108 / 109
Motivation Business cycles Model Steady state Linearisation Solution, calibration & simulation Application to fiscal policy Critique Conclusions
Conclusions
The RBC model is the simplest microfounded business cyclemodel
It explains cyclical fluctuations as resulting from technologyshocks
Quite succesful in some areas (e.g. main moments) but hasseveral problematic features
We think of RBC as a useful frictionless benchmark andbuilding block of more complicated models
109 / 109