Monetary Economics: Macro Aspects, 1/3 2012

Henrik Jensen
Department of Economics
University of Copenhagen
1. Money in the utility function (start)
a. The basic money-in-the-utility function model
b. Optimal behavior and steady-state equilibrium properties:
Long-run superneutrality of money
Literature: Walsh (2010, Chapter 2, pp. 3352)
c _ 2012 Henrik Jensen. This document may be reproduced for educational and research purposes, as long as the copies contain this notice and are retained for personal
use or distributed free.
Introductory remarks
The standard model for (exogenous) economic growth is the simple Solow model featuring
a xed savings rate
a law of motion for physical capital accumulation
When extended with optimizing savings behavior, we get the Ramsey model
Models have no role for money and, hence, monetary policy
Purpose of models/analyses in coming lectures is to introduce a role for money in these type of models
1
Money is introduced in various ways: Often short cuts
Short cuts are helpful for understanding simple features, and the more robust results are to a
particular short cut, of course, the better
Tobin model (1965, Econometrica) extends the Solow model by postulating a demand for money
(the short cut)
Highlights the implications of ination for the choice between investment in physical and nancial
assets
Higher ination leads to a substitution away from nancial assets towards physical capital
Higher ination leads to higher steady-state capital stock and output
Money-in-the-Utility-function models extend the Ramsey model by postulating that money
gives utility (the short cut)
Highlights the importance of micro foundations and optimizing private-sector behavior (absent
in the Tobin model)
2
Money in the utility function (start)
A standard Ramsey model with money (innite-horizon setting with optimizing households and
perfect competition in goods market)
The short-cut here: real money provides utility per se
One interpretation: Money facilitates transactions on the market and reduces shopping time
(money as such is an otherwise useless commodity ...)
The per-period utility function of households is
l
t
= n(c
t
. :
t
)
with n being increasing and concave in both arguments. (It is, of course, real money that enters in
n, i.e., `
t
s value relative to what it can buy at price 1
t
.)
Often, to ensure existence of an equilibrium where money is held (a monetary equilibrium), it is
assumed that for some :, n
:
(c
t
. :) = 0 and n
:
(c
t
. :
t
) < 0 for :
t
:
I.e., : is an optimal quantity of money
3
Aim of representative household is to maximize:
\ =

t=0
,
t
n(c
t
. :
t
) . 0 < , < 1. (2.1)
Households budget constraints are (ignoring for simplicity nancial assets 1
t
used in Walsh):
C
t
+ 1
t
+
`
t
1
t
= 1
t
+ t
t
`
t
+ (1 o) 1
t1
+
`
t1
1
t
. 0 < o < 1 (2.2)
Note, :
t
= `
t
, (1
t
`
t
) is end-of-period money balances; debatable in itself whether it should be
:
t1
or :
t
that should give utility
Output is produced by a CRS production function:
1
t
= 1 (1
t1
. `
t
) .
In intensive form:

t
= , (/
t1
) .
t
= 1
t
,`
t
. /
t1
= 1
t1
,`
t1
(population grows at a constant rate `
t
,`
t1
= 1 + :)
Government makes per capita real lump-sum transfers (or withdrawals) to households in the form of
money supply changes:
t
t
=
`
t
`
t1
1
t
`
t
(This is the simple government budget constraint.)
4
In per-capita version, assuming no population growth in contrast to Walsh (set : = 0, and `
t
= 1),
we get
c
t
+ /
t
+ :
t
=
t
+ t
t
+ (1 o) /
t1
+
1
1 + :
t
:
t1
Ination :
t
= (1
t
,1
t1
) 1, erodes initial real monetary resources
Rewritten:
c
t
+ /
t
+ :
t
= .
t
= , (/
t1
) + t
t
+ (1 o) /
t1
+
1
1 + :
t
:
t1
(2.4)
Hence, .
t
is the total available resources, treated as given at t by the households.
It is the relevant state variable when choosing the optimal paths of c, /, and : at date t
Households optimization problem is solved by dynamic programming
5
Dynamic programming involves use of the value functionthe maximal value of \.
given constraints
given optimal behavior of the household now and forever after
given the current state .
The value function \ therefore satises
\ (.
t
) = max n(c
t
. :
t
) + ,n (c
t+1
. :
t+1
) + ,
2
n(c
t+2
. :
t+2
) + . . .
\ (.
t
) = max n(c
t
. :
t
) + , [n(c
t+1
. :
t+1
) + ,n (c
t+2
. :
t+2
) + . . .]
compactly written as
\ (.
t
) = max n(c
t
. :
t
) + ,\ (.
t+1
)
Maximization is over c
t
, :
t
and /
t
subject to the budget constraint and the denition of .
t+1
(available resources one period ahead)
6
To make it simple here, one
substitutes out .
t+1
then substitutes out /
t
using /
t
= .
t
c
t
:
t
One then maximizes (unconstrained) just over c
t
and :
t
:
max
_

_
n(c
t
. :
t
) + ,\
_
_
_
_
, (/
t
) + t
t+1
+ (1 o) /
t
+
1
1 + :
t+1
:
t
. .
.
t+1
_
_
_
_
_

_
and thus
max
_
_
_
n(c
t
. :
t
) + ,\
_
_
, (.
t
c
t
:
t
) + t
t+1
+(1 o) (.
t
c
t
:
t
) +
1
1 + :
t+1
:
t
_
_
_
_
_
7
First-order condition concerning choice of c
t
:
n
c
(c
t
. :
t
) + ,\
.
(.
t+1
)
J.
t+1
Jc
t
= 0
n
c
(c
t
. :
t
) = ,\
.
(.
t+1
) [,
/
(/
t
) + 1 o] (2.6)
Marginal utility of period-t consumption equals its marginal loss (in terms of the discounted marginal
value of future capital)
First-order condition concerning choice of :
t
:
n
:
(c
t
. :
t
) + ,\
.
(.
t+1
)
J.
t+1
J:
t
= 0
n
:
(c
t
. :
t
) + ,\
.
(.
t+1
)
1
1 + :
t+1
= ,\
.
(.
t+1
) (,
/
(/
t
) + 1 o) (2.8)
Marginal utility period-t real money (in terms of direct utility plus discounted marginal value of more
future real money) equals marginal loss (in terms of the marginal value of less future capital)
Furthermore, transversality conditions must be satised:
lim
t
,
t
n
c
(c
t
. :
t
) /
t
= 0
lim
t
,
t
n
c
(c
t
. :
t
) :
t
= 0
(otherwise over-accumulation of / and : is taking placelifetime utility could be improved through
higher consumption by accumulating less)
8
In the rst-order conditions, \ is eliminated by use of the so-called Envelope Theorem
Note: optimal period t consumption and money holding choices will be functions of .
t
Dene these as c
t
= c (.
t
) and :
t
= :(.
t
), respectively
The value function is therefore by denition given by
\ (.
t
) = n(c (.
t
) . :(.
t
)) + ,\ (.
t+1
) . (*)
As (*) holds for any value of .
t
, it follows that
\
.
(.
t
) = n
c
(c (.
t
) . :(.
t
)) c
/
(.
t
) + n
:
(c (.
t
) . :(.
t
)) :
/
(.
t
) + ,\
.
(.
t+1
)
J.
t+1
J.
t
. (**)
Now, nd J.
t+1
,J.
t
when c
t
= c (.
t
) and :
t
= :(.
t
) applies:
Remember
.
t+1
= , (/
t
) + t
t+1
+ (1 o) /
t
+
1
1 + :
t+1
:
t
= , (.
t
c
t
:
t
) + t
t+1
+ (1 o) (.
t
c
t
:
t
) +
1
1 + :
t+1
:
t
One therefore gets
J.
t+1
J.
t
= [,
/
(/
t
) + 1 o] (1 c
/
(.
t
) :
/
(.
t
)) +
1
1 + :
t+1
:
/
(.
t
)
9
Combining this with (**):
\
.
(.
t
) = n
c
(c (.
t
) . :(.
t
)) c
/
(.
t
) + n
:
(c (.
t
) . :(.
t
)) :
/
(.
t
)
+,\
.
(.
t+1
)
_
_
_
[,
/
(/
t
) + 1 o] (1 c
/
(.
t
) :
/
(.
t
))
+
1
1 + :
t+1
:
/
(.
t
)
_
_
_
.
Collect the c
/
(.
t
) and :
/
(.
t
) terms:
\
.
(.
t
) = [n
c
(c (.
t
) . :(.
t
)) ,\
.
(.
t+1
) (,
/
(/
t
) + 1 o)] c
/
(.
t
)
+
_
_
n
:
(c (.
t
) . :(.
t
)) + ,\
.
(.
t+1
)
1
1 + :
t+1
,\
.
(.
t+1
) [,
/
(/
t
) + 1 o]
_
_
:
/
(.
t
)
+,\
.
(.
t+1
) [,
/
(/
t
) + 1 o]
... and get a pleasant surprise:
\
.
(.
t
) =
_
_
n
c
(c (.
t
) . :(.
t
)) ,\
.
(.
t+1
) (,
/
(/
t
) + 1 o)
. .
=0 by (2.6)
_
_
c
/
(.
t
)
+
_

_
n
:
(c (.
t
) . :(.
t
)) + ,\
.
(.
t+1
)
1
1 + :
t+1
,\
.
(.
t+1
) [,
/
(/
t
) + 1 o]
. .
=0 by (2.8)
_

_
:
/
(.
t
)
+,\
.
(.
t+1
) [,
/
(/
t
) + 1 o]
10
All terms in front of c
/
(.
t
) and :
/
(.
t
) are zero by the rst-order conditions
Note the terms capture the life-time utility eects of changing .
t
through the associated changes
in c
t
and :
t
As the value function is dened as the life-time utility where c
t
and :
t
are optimally chosen, their
eects are zero
Alternatively, use a contradiction argument: If a change in .
t
leads to a change in \ through the
changes in c
t
and :
t
, then c
t
and :
t
have not been optimally chosen, and \ is not the highest
life-time utility.
Hence, (**) reduces to
\
.
(.
t
) = ,\
.
(.
t+1
) [,
/
(/
t
) + 1 o]
Then use the rst-order condition for consumption choice,
n
c
(c
t
. :
t
) ,\
.
(.
t+1
) [,
/
(/
t
) + 1 o] = 0.
to obtain Walshs expression:
\
.
(.
t
) = n
c
(c
t
. :
t
) . (2.10)
Marginal utility of consumption (denoted `
t
in Walsh) equals marginal value of wealth
11
The rst-order conditions can then be rewritten as
n
c
(c
t
. :
t
) = ,n
c
(c
t+1
. :
t+1
) [,
/
(/
t
) + 1 o]
(a discrete-time, monetary version of the standard Keynes-Ramsey rule), and
n
:
(c
t
. :
t
) + ,
n
c
(c
t+1
. :
t+1
)
1 + :
t+1
= n
c
(c
t
. :
t
)
(marginal gain of :
t
equal to the marginal loss in terms of lower capital in period t + 1equal to
the marginal utility of c
t
by the Keynes-Ramsey rule)
These conditions, together with the budget constraint characterize the optimal paths of c, /, and :
NOTE: Only :
t
appears. Any change in `
t
is reected proportionally in 1
t
: Money neutrality
For now, concentrate on the long-run properties of the model; i.e., a steady state with
c
t
= /
t
= :
t
= 0
First, from Keynes-Ramsey rule it follows that in steady state
1 = , [,
/
(/
::
) + 1 o] .
or,
,
/
(/
::
) + 1 o =
1
,
(2.19)
Thisindependently of any monetary factorsdenes the steady-state capital per capita (and thus
output per capita).
12
Strong contrast with Tobin model mentioned in introduction
Dierence is because this model envisions micro-founded behavior.
If, e.g., /
t
< /
::
the current marginal product of capital is relatively high (as ,
//
< 0)
== optimal to postpone consumption to later
== capital is accumulated until ,
/
(/
::
) + 1 o = 1,, holds again
If one imagined that a Tobin eect was there; one would be self-contradictory:
+ Assume higher ination increases capital to a new, higher steady state
+ Marginal product of capital decreases, and households would want to consume now rather
than later
== they endogenously save less and capital decreases until /
::
is reached again!
+ Ination has no steady-state eect on capital. Only possible in Tobin model, where individuals
are modelled as having an exogenously xed savings rate
+ Example of the importance of considering changes in private sector behavior when examining
policy changes
13
What about steady-state consumption in the MIU model?
The budget constraint in steady state is
, (/
::
) + t
::
+ (1 o) /
::
+
1
1 + :
:
::
= c
::
+ /
::
+ :
::
Transfers are
t
t
= (`
t
`
t1
) ,1
t
.
= o
t
`
t1
,1
t
.
=
o
t
1 + :
t
:
t1
.
so in steady state:
t
::
=
o
1 + :
:
::
Since a constant : implies : = o, one gets
, (/
::
) o/
::
= c
::
(2.21)
The economys resource constraint (national account): Output less gross investment equals consump-
tion
Implication is that c
::
is determined exclusively by /
::
; and thus independent of nominal money
growth
Model exhibits superneutrality of money in steady state
14
Do money growth and ination not aect anything?
Yes, the opportunity cost of holding real money, and thus the steady-state value of real money balances
Relative demand for consumption versus real money is given by
[use rst-order condition for money holdings and divide through by n
c
(c
t
. :
t
)]
n
:
(c
t
. :
t
)
n
c
(c
t
. :
t
)
=
n
c
(c
t
. :
t
)
n
c
(c
t
. :
t
)

1
1 + :
t+1
,n
c
(c
t+1
. :
t+1
)
n
c
(c
t
. :
t
)
= 1
1
1 + :
t+1
1
(,
/
(/
t
) + 1 o)
= 1
1
(1 + :
t+1
)
1
(1 + :
t
)
with :
t
= ,
/
(/
t
) o being the real interest rate
Note that the real interest rate is the nominal rate net of expected ination:
1 + :
t
= (1 + i
t
) , (1 + :
t+1
) . (:
t
- i
t
:
t+1
)
Hence,
n
:
(c
t
. :
t
)
n
c
(c
t
. :
t
)
=
i
t
1 + i
t
= 1
t
(2.12)
So, as nominal interest rate is determined by the Fisher relationship, i
t
- :
t
+ :
t+1
, higher ination
leads to a higher nominal interest rate
Consistent with long-run data
15
For given c
t
, :
t
is likely to fall when i
t
increases (as n
::
< 0).
A micro foundation for conventional money-demand function
With n(c
t
. :
t
) =
_
cc
1/
t
+ (1 c) :
1/
t

1,(1/)
, 0 < c < 1, / 0 we get from (2.12)
n
:
(c
t
. :
t
)
n
c
(c
t
. :
t
)
=
_
cc
1/
t
+ (1 c) :
1/
t

/,(1/)
(1 c) :
/
t
_
cc
1/
t
+ (1 c) :
1/
t

/,(1/)
cc
/
t
=
1 c
c
_
c
t
:
t
_
/
=
i
t
1 + i
t
This gives a money demand function as
:
t
=
_
1 c
c
_1
b
_
i
t
1 + i
t
_

1
b
c
t
(2.31)
In logs (often used for estimation purposes):
log :
t
= const. + log c
t

1
/
log
_
i
t
1 + i
t
_
.
where 1,/ is the elasticity of money demand wrt. opportunity cost. I.e., the interest rate elasticity
of money demand (depending on money concept, empirically around 0.10.3)
16
Will a unique steady-state value for : exist? Must solve
n
:
(c
::
. :
::
) =
_
i
::
1 + i
::
_
n
c
(c
::
. :
::
)
n
:
(c
::
. :
::
) = 1
::
n
c
(c
::
. :
::
)
Not necessarily unique.....
Stability properties? For separable utility, n(c
t
. :
t
) = (c
t
) + c(:
t
), resulting dierence equation
(from rst-order condition) will imply a saddle-point stable :
::
0 (:
/
in Figure 2.1, page 45)
Problems:
One cannot necessarily rule out the paths with increasing : above steady state (speculative
hyperdeations)
One cannot necessarily rule out the paths with falling : below steady state (speculative hy-
perinations), leading to :
::
= 0
In the context of this model, we will not pursue the stability issue
17
Summary
MIU model has one for one relationship between ination and money growth
MIU model exhibits superneutrality
A model like the Tobin model is not superneutral; reason is the postulated and policy-invariant
private sector behavior. This dierence highlights the importance of micro foundations to avoid
Lucas critique
MIU is a structural model where the reactions of the private sector to changes in policy (money
growth) are taken into account
Potentially more appropriate for analyzing policy changes (even though the micro foundation for
money demand is a short cut)
(At least for steady-state analyses.)
18
Plan for next lectures
Tuesday, March 6
1. Money in the utility function (continued)
a. Welfare costs of ination
b. Potential non-superneutrality of money
c. Dynamics and calibration
Literature: Walsh (2010, Chapter 2, pp. 5286, so check the Appendix as well; i.e., get a grip on the
linearization technique)
19