MONETARY POLICY AND THE FINANCING OF FIRMS

Monetary Policy and the Financing of Firms Fiorella De Fiorey Pedro Telesz Oreste Tristaniy y European Central Bank z Banco de Portugal, U. Catolica Portuguesa and CEPR

Helsinki, 15-16 October 2009

De Fiore, Teles, Tristani ()

MP and the Financing of Firms


1 / 29

The questions

How should monetary policy react to …nancial shocks? And to other shocks, when …nancial conditions are relevant? Are the risks of debt-de‡ation a reason for monetary policy to induce some in‡ation during recessions? Do …nancial factors provide a reason for monetary policy to deviate from zero nominal interest rates?




2 / 29

The model

A model where monetary policy a¤ects …rms’…nancing conditions. Distinguishing features: Firms’internal and external sources of …nance are imperfect substitutes. Firms’internal and external funds are nominal assets. Those funds, as well as the interest rate on bank loans, are predetermined when aggregate shocks occur.




3 / 29

Main results

1. Maintaning price stability at all times is not optimal. Because …rms’funds are nominal and predetermined, after setting interest rates policy can choose the price level so as to adjust the real value of total funds. 2. The distortions introduced by …nancial factors do not justify deviating from a zero nominal interest rate.




4 / 29

Main results

3. The optimal response to a negative productivity shock is to maintain the nominal interest rate …xed and to engineer a short period of in‡ation. This policy stabilizes default rates, credit spreads and the …nancial markup. 4. The optimal response to a reduction in internal funds is to reduce the nominal interest rate or, if it is at the zero bound, to engineer a short period of in‡ation. This mitigates the adverse consequences on bankruptcy rates and allows …rms to de-leverage more quickly.




5 / 29

Outline of the presentation

1 2 3 4 5

Related literature The model Optimal monetary policy Numerical results Conclusions




6 / 29

Related Literature Financial factors and the transmission of shocks Bernanke, Gertler and Gilchrist (1999) Calstrom and Fuerst (1997, 1998) Financial factors and optimal monetary policy Ravenna and Walsh (2006) Curdia and Woodford (2008) De Fiore and Tristani (2008) Faia (2008)




7 / 29

The environment Agents: Households, …rms facing credit constraints and iid productivity shocks, competitive banks, and a central bank. Timing: There is a goods market at the beginning of the period and an assets market at the end, when …rms’internal funds, external funds and interest rates are decided for the following period. Financial intermediation: Firms need to pay wages in advance of production. They have nominal internal funds but also need external …nance. Banks raise deposits from households and lend to entrepreneurs on the basis of a nominal debt contract. De Fiore, Teles, Tristani ()



8 / 29

The model

Households Maximize E0

(

∞

∑β

t

[u (ct ) + κ (mt )

0

αnt ]

)

subject to Mt + Et Qt,t +1 At +1 + Dt


At + Rtd 1 Dt

1 + Mt 1


Pt ct + Wt nt


Tt

9 / 29

The model Firm i, i 2 (0, 1) produces a homogeneous good with the technology yi ,t = ω i ,t At Ni ,t ; decides in the assets market at t 1 the amount of internal funds, Bi ,t 1 , and total funds, Xi ,t 1 , to be available in t; pays wages before production, so decisions are restricted by Wt Ni ,t


Xi ,t


1.


10 / 29

The model The …nancial contract For given Bi ,t 1 , the optimal contract sets an amount Ril,t 1 (Xi ,t 1 Bi ,t 1 ) to be repaid when ω i ,t ω i ,t , where ω i ,t is the minimum productivity level such that the …rm is able to repay. If the …rm defaults, it hands out production to banks but a constant fraction µt is destroyed in monitoring. De…ne bt 1 = XBtt 11 . The optimal contract is a vector bt 1 , Rtl 1 , ω t that solves a standard costly state veri…cation problem. It is the same across …rms.




11 / 29

The model

Useful notation f (ω t ) is the average share of production accruing to …rms, when the threshold that triggers default is ω t . µt G (ω t ) is aggregate output lost in monitoring φ (ω ) and Φ (ω ) are the density and cdf of the log-normal iid shock.




12 / 29

The model Entrepreneurs Entrepreneurs are in…nitely lived and risk neutral. Their discount rate is su¢ ciently low that entrepreneurs keep postponing consumption and only accumulate internal funds. There is a proportional tax γt that prevents their wealth from growing to the point where there is no need for external …nance. The accumulation of internal funds is given by Bt = (1


γt ) f (ω t ) Pt At Nt



13 / 29

Some properties of the model Since Xt and Bt are predetermined, bt = XBtt does not change on impact, neither does leverage (equal to b1t 1). bt = Bt 1 /Pt is the real value of internal funds. It changes according to b t = (1

γt

1) f

(ω t

1)

vt bt

bt 1 . 2 πt

1

At vt = w is the …nancial markup. Larger vt increase …rms’pro…ts, t because …rms pay a lower real wage, for given productivity At .




14 / 29

The …nancial murkup

vt re‡ects three …nancial distortions: the predetermination of …nancial decisions the credit constraint faced by …rms’ the presence of asymmetric information and monitoring costs

In equilibrium, Et

1

vt 1


µt G ( ω t )

f (ω t )

Et Et


[µt ω t φ (ω t )] Φ (ω t )] 1 [1

1

= Rtd 1 .


15 / 29

The …nancial murkup If µt = 0, Et Rtd 1

1

fvt g = Rtd

1

and = 1 would minimize the average distortion. If µt = 0 and decisions were not predetermined, vt = Rtd and Rtd = 1 would achieve the …rst best. If µt = 0, decisions were not predetermined, and …rms did not have to borrow, vt = 1 and the economy would be in the …rst best irrespective of Rtd . De Fiore, Teles, Tristani ()



16 / 29

Optimal interest rate policy Since ω is independent of R d , we can consider a social planner that max utility, for given ω, s.t. the resource constraint only c = An [1

µG (ω )] .

Optimality requires that v=

1

1 . µG (ω )

In equilibrium, Rd

v= 1

µG (ω )

ωφ(ω ) Φ(ω )

.

µf (ω ) 1

When µ 6= 0, it is optimal to set R d = 1. De Fiore, Teles, Tristani ()



17 / 29

Intuition for the optimality of the Friedman rule The …rst best response to the restriction on the accumulation of internal funds is b = 1, then ωt = Et

Rtl

et 1v

1

(1 bt vt

= Rtd

1

1)

< Et

= 0. 1 vt

As a second best, it is still optimal to subsidize, but subsidizing also increases production and the amount of resources lost in bankruptcy when ω t > 0; overall, it is optimal to subsidize, but not possible all the way to the point where vt = 1. Production subsidy can take the form of a negative deposit rate; since R d 1, R d = 1 is optimal. De Fiore, Teles, Tristani ()



18 / 29

Monetary policy instruments

Because …rms’funds are nominal and predetermined, after setting interest rates policy can a¤ect the price level. This can be seen from the set of implementability conditions. The solution to the planner’s problem is unique if Pt is set exogenously




19 / 29

Numerical analysis

Calibration: µ = .15; σω and γ are set such that 1% of …rms go bankrupt each quarter and the credit spread is 2% per year. bt Taylor rule: bit = 1.5 π




20 / 29

Technology shock (OMP black, TR blue) x 10

r

-3

y

t+1

1 0 -1 -2 -3

x 10

x 10

4 -3

6

8 10 12

-0.01

2 4 6 8 10 12 b

Φ(ϖ ) t

1 2

4

x 10

x 10

t

-4

6 ∆

8 10 12

2

4 6 π

t+1

0 -2 -4 -6

8 10 12

n

t

2

4 6 -3

ν

4 6 b

t

8 10 12 t

8 10 12

t+1

0.01

2

-0.01 0

1 0

2

0 -0.01 -0.02 -0.03

2

-3

6 4 2 0

-0.005

2

0

t

0

2

4


6

8 10 12

-0.01

-0.02 -0.03 2

4 6

8 10 12


2

4 6

8 10 12


21 / 29

Shock to internal funds (OMP black, TR blue) x 10

-6

r

t+1

1

0 -5 -10 -15

x 10

-5

y

t

1

0

x 10

4 -5

6

8 10 12

-2

Φ(ϖ ) t

2 x 10

-2 -4

0

-6 2 x 10

4

6

-6

∆

8 10 12

4 -4

6 b

8 10 12

0

2

4


6

8 10 12

-2

t

2

2

4

x 10

-5

6 ν

8 10 12 t

0

x 10 4 2 0 -2

t

1

2

t+1

5

c

-1

0 2

-5

0

-1 2

x 10

4 -5

6 π

8 10 12

-1

t

0

2 x 10

4

6

-4

b

4

6

8 10 12

t+1

-5 2

4

6

8 10 12


2

8 10 12


22 / 29

Policy shock r

y

t+1

0

c

t

0.01

0.01

0.005

0.005

-0.005

t

-0.01 2 4 6 8 10 12 x 10

-3

Φ(ϖ ) t

-0.005

0

x 10

-4

∆

π

2 4 6 8 10 12 b

t

t+1

0

2

-0.005

1

-0.01 2 4 6 8 10 12


-0.01 2 4 6 8 10 12

t+1

t

0

-0.01 2 4 6 8 10 12

2 4 6 8 10 12 ν

t

0.01

1

0

0

2 4 6 8 10 12 b

2

0

0

-0.01 -0.02 -0.03 2 4 6 8 10 12


2 4 6 8 10 12


23 / 29

Role of the zero lower bound We compare the response to a negative shock to internal funds under the optimal policy when R d = 0 and when R d > 0. Technology shock: Despite the ability of monetary policy to move the nominal interest rate, it is optimal not to do so; a policy that keeps it …xed and creates some in‡ation on impact is able to generate the same response as in the …rst best. Financial shock: The ability to lower the nominal interest rate enables to a¤ect credit conditions directly, by reducing ceteris paribus - the loan rates. This generates a much faster adjustment in response to the shock.




24 / 29

Zero lower bound: …nancial shock 0

x 10

r

-5

t+1

0

x 10

y

-5

t

0

-0.5

-0.5 -1 -1.5 2

4

x 10

4

6

8

10

12

-1

-1.5

-1.5

-2

t

2

x 10

4

6

8 b

-4

10

12

-2

t

2

-1

3

c

-5

t

-0.5

-1

Φ(ϖ )

-5

x 10

2

x 10

4

6 ν

-5

8

10

12

8

10

12

8

10

12

t

1.5

-2

2 1 0

-3

1

-4

0.5

-5 2

x 10

4

6 ∆

-6

8

10

12

2

t+1

4

x 10

4

6

8 π

-5

10

12

t

0

2

4

0

2

x 10

4

6 b

-4

t+1

-2

0 2

-4

-2 0

2

4

6


8

10

12

-4

2

4

6

8

10

12


-6

2

4

6


25 / 29

Zero lower bound: …nancial shock x 10

r

-6

γ shock t+1

x 10

-6

t

y

t

5

5 0

-20

2

4 -5

6

8 10 12

0

2

t

2

2 x 10

4 -6

6

8 10 12

5

x 10

4 -4

6 b

8 10 12

2

4

x 10

t

-6

6 ν

8 10 12 t

10 5 0 -5 2

∆ t+1

t

-10

Φ(ϖ ) 0 -2 -4 -6

c

-5

-10

x 10

-6

0

0 -5

-10

x 10

x 10

4 -5

6

8 10 12

2

πt 0

4

x 10

-4

6 8 10 12 bt+1

2 0

0

-5

-2 2

4


6

8 10 12

2

4

6

8 10 12


2

4

6

8 10 12


26 / 29

No asymmetric information: technology shock x 10

-3

r

t+1

0

1 0 -1 -2 -3

x 10

-3

t

x 10

4

6

8 10 12

2

4

6 b

Φ(ϖ ) t

0.01

n

t

8 10 12

2 x 10

t

4 -3

6 ν

8 10 12 t

0 0 -2 -4

-0.02

0.005

-0.04 2 x 10

4

6

-4

∆

8 10 12

2

t+1

x 10

4 -3

6 π

8 10 12

2

b

4


6

8 10 12

8 10 12

t+1

-0.02

-1 -2 2

6

0

0

5

4

t

1

0

-3

6 4 2 0

-5 2

0

y

-0.04 2

4

6

8 10 12


2

4

6

8 10 12


27 / 29

Conclusions

When …rms’…nancial positions are denominated in nominal terms and debt contracts are not state-contingent, policy can induce some in‡ation during recessions in order to reduce the real value of funds according to current production needs. The optimal response to a negative productivity shock is to maintain the nominal interest rate …xed and to engineer a short period of in‡ation. This way, nominal wages and labor can be kept constant and the predetermined value of total funds is ex-post optimal. This policy stabilizes default rates, credit spreads and the …nancial markup.




28 / 29

Conclusions

The optimal response to a reduction in internal funds is to reduce the nominal interest rate or, if it is at the zero bound, to engineer a short period of in‡ation. This mitigates the adverse consequences on bankruptcy rates and allows …rms to de-leverage more quickly.




29 / 29

MONETARY POLICY AND THE FINANCING OF FIRMS

Recommend Documents