Monetary and Fiscal Policy under Imperfect Knowledge

Size: px
Start display at page:

Download "Monetary and Fiscal Policy under Imperfect Knowledge"

Transcription

1 Monetary and Fiscal Policy under Imperfect Knowledge Monash University Mini-course Part III Bruce Preston July 16, 2012

2 Motivation Macroeconomic models depend on household and Örm expectations Policy design ñ Not enough to observe some history of beliefs ñ Need a theory of the determination of beliefs: rational expectations Rational expectations policy design ñ Heavy reliance on managing expectations through announced commitments ñ What are the consequences of imprecise control of beliefs?

3 Motivation II Reasonable to suppose agents have limited information about policy regime ñ US Financial Crisis witnessed dramatic change in the nature and conduct of stabilization policy ñ Does imperfect knowledge limit the e cacy of policy? For example, Bernanke (2004): ì[...] most private-sector borrowing and investment decisions depend not on the funds rate but on longer-term yields, such as mortgage rates and corporate bond rates, and on the prices of long-lived assets, such as housing and equities. Moreover, the link between these longer-term yields and asset prices and the current setting of the federal funds rate can be quite loose at times.î

4 The Agenda Part 3 of this mini-course proceeds as follows ñ Lecture 1: Tools and conceptual issues ñ Lecture 2: Consequences of learning for the choice of monetary policy rule ñ Lecture 3: Monetary and Öscal interactions

5 Introducing Learning Dynamics Two foundations of rational expectations equilibrium analysis ñ Optimization ñ Mutual consistency of beliefs Strong knowledge assumptions ñ Agents know more than the modeler

6 ScientiÖc Method Principle components ñ Maintained theories ñ Procedures for collecting data ñ Procedures for confronting theory with data ñ Procedures for updating theory given discrepancy between theory and data

7 Bounded Rationality: Learning Dynamics Attempts to place the agent and modeler on equal footing Agents might proceed as: ñ Classical econometrician: know model but unsure about parameters ñ Bayesian econometrician: unsure about model and parameters but understand how they are unsure

8 Learning Dynamics Analyze broad class of model that relaxes the second pillar of REE analysis ñ Retain optimization ñ Permit non-rational expectations Beware of theorists bearing free parameters! ñ Constrain the analysis by requiring the procedure to nest rational expectations

9 Why learning? Permits analysis of a broader class of problems May resolve/elucidate ñ Questions of indeterminacy ñ Questions of robustness ñ Questions of dynamics ñ Questions of regime change/regime uncertainty

10 Today Mathematical foundations ñ Learning renders model dynamics self-referential ñ Requires new set of tools to analyze model properties Conceptual issues in modeling multi-period decision problems

11 A Simple Model Consider a variant of the Cobweb model p t = + E t1 p t + w t1 + t (1) where ; a; > 0; w t and exogenous process; and t a bounded iid disturbance Unique bounded solution p t =a + bwt1 + t where a =(1) 1 and b =(1) 1

12 Statistical Learning Suppose agents forecast p t as a linear function where fa t1 ;b t1 g are currently maintained beliefs ^E t1 p t = a t1 + b t1 w t1 (2) Price dynamics then p t =( + a t1 ) +( + b t1 ) w t1 + t ñ Self-referential system ñ Data generating process is non-stationary

13 Least Squares Learning Beliefs fa t1 ;b t1 g estimates using data fp i ;w i g i=t1 i=0 Ordinary least squares implies 0 a t1 t1 X b t1 i=1 z i1 z 0 i1 where z i =[1 w i ] ñ Fully speciöed system: relations (1) - (3) 1 A1 0 t1 i=1 1 z i1 p i A (3)

14 Question Under what conditions do a t! a and b t! b as t!1?

15 Theorem Consider the model (1) - (3). If <1 then (a t ;b t )! a; b with probability one. If >1 then convergence occurs with probability zero. ñ Property <1 referred to as expectational stability ñ E-Stability principle governed by mapping between agent beliefs and true model coe cients

16 E-Stability principle Recall ^E t1 p t = a t1 + b t1 w t1 (4) and p t =( + a t1 ) +( + b t1 ) w t1 + t Latter expression implies optimal rational forecast E t1 p t =( + a t1 )+( + b t1 ) w t1

17 Mapping DeÖned Agentsí beliefs and optimal forecast deöne the mapping ñ An REE is a Öxed point of this mapping T a b! = + a + b ñ Application of method of undetermined coe cients!

18 E-Stability principle II DeÖne the associated ordinary di erential equation where is notional time d d a b! = T a b! a b! ñ Local stability properties govern E-Stability principle

19 E-Stability principle III In particular, the REE is E-Stable if and only if the ODE is local stable at a; b Hence requires <1 da d db d = + ( 1) a = + ( 1) b

20 Recursive Least Squares Ordinary least squares regression can be written as where t =[a t b t ] 0 t = t1 + t 1 R t z t1 pt 0 t1 z t1 R t = R t1 + t 1 z t1 zt1 0 R t1

21 Recursive Least Squares II Since p t = ( + a t1 )+( + b t1 ) w t1 + t = T t1 0 zt1 + t we have t = t1 + t 1 Rt 1 z t1 z 0 t1 T t1 t1 + t R t = R t1 + t 1 z t1 zt1 0 R t1 ñ Question: does this system converge??

22 Stochastic Approximation Methods Provide results characterizing convergence of such systems ñ Ljung (1977), Marcet and Sargent (1989) Consider stochastic recursive algorithm t = t1 + t Q (t; t1 ;X t ) where t - parameter estimates (a t ;b t ;R t ) X t - state vector (e ects of p t, z t and t ) t - deterministic sequence of gains (t 1 )

23 Stochastic approximation approach associates the ODE where d d = h ( ()) h () lim t!1 EQ(t; ; X t) ñ E denotes the expectation of Q (t; ; X t ) with respect to the invariant distribution of X t for Öxed

24 Stochastic Approximation Results Under suitable assumptions: ñ If is a locally stable equilibrium point of the ODE, then is a possible point of convergence of the SRA ñ If is not a locally stable equilibrium point of the ODE, then is not a possible point of convergence of the SRA. That is t! with probability zero.

25 Technical Assumptions Stochastic approximation results rely on assumptions regarding ñ Regularity conditions on Q ñ Conditions on the rate at which t! 0 ñ Assumptions on the properties of X t Evans and Honkapohja (2001, Chapter 6 & 7)

26 Example Revisited To write our RLS algorithm in SRA form deöne S t1 R t and write the system as t = t1 + t 1 St1 1 z t1 z 0 t1 T t1 t1 + t so that S t = S t1 + t 1 t zt z 0 t S t1 t +1 t = vec ( t ;S t ) X t = [1 w t w t1 t ] 0 t = t 1

27 Example Revisited II Hence Q (t; t1 ;X t ) = St1 1 z t1 z 0 t1 T t1 t1 + t Q S (t; t1 ;X t ) = vec t zt zt 0 S t1 t +1 ñ To do: compute the associated ODE! ñ Fix t and compute expectation over X t

28 Example Revisited III Fixing (; S) implies h (; S) = lim t!1 ES1 z t1 z 0 t1 (T () )+ t h S (; S) = t zt lim t!1 E zt 0 S t +1 Since " Ez t zt 0 = Ez t1 zt = 0 where =E w t wt 0 ; Ezt1 t =0and lim t!1 # M t t+1 =1 h (; S) = S 1 M (T () ) h S (; S) = M S

29 The Associated ODE is then d d ds d = S 1 M (T () ) = M S ñ The latter is globally stable for any initial S: S! M ñ Therefore S 1 M! I Hence d d = T () ñ Intuition?

30 Summary Have motivated the concept of E-Stability ñ Governed by eigenvalues of the associated ODE Remainder ñ Microfoundations and modeling issues ñ Subsequent lectures: Applications in policy design

31 Modeling Learning Dynamics: Some Issues Intertemporal optimization implies multi-period forecasts ñ Fundamental to macroeconomic analysis Transversality conditions central to this theory ñ Partial Equilibrium: Marcet and Sargent (1989) ñ General Equilibrium: Preston (2005) Yet most of the macro learning literature ignores this requirement of optimality

32 Beliefs Under rational expectations: 1. Agents optimize given beliefs 2. The probabilities they assign to future state variables coincide with the predictions of the model This paper retains (1) and replaces (2) with 2 0. Future state variables outside agentís control are forecasted using an econometric model.

33 Knowledge Know own preferences and constraints Do not know true economic model of determination of variables outside their control. ñ Uncertain about the equilibrium mechanisms determining prices and policy variables ñ E.g. even if Y i t = Y j t = Y t in EQ ; agents know Y i t = Y j t Observe aggregate variables and disturbances Forecast variables outside their control using an atheoretical VAR ñ Takes the minimum state variable form

34 Townsend Investment Problem Consider a representative Örm solving max fk T g ^E t 1 X T =t T t p T k T 2 kt k T 1 2 where 0 < < 1 and ; > 0: The price of output, k t, is determined on competitive markets according to p t = K t + u t u t = u t1 + " t where 0 <<1 and K t the aggregate capital stock. Equilibrium requires: k t = K t = Z1 k t (i) di ñ Assumption: Örms do not know market mechanism determining prices 0

35 Townsend Investment Problem II To solve this problem a Örm requires forecasts of future capital prices and state variables ñ Prices are exogenous to their decision problem ñ As are the aggregate capital stock and stochastic disturbance Denoting the set of variable beyond the Örmís control as z t fp t ;K t ;u t g assume forecasting model of the form z t = a t + b t z t1 + t ñ Use data available until period t 1

36 Forecasts are then constructed according to ^E t z T =(I 3 ) 1 I 3 b T t t1 at1 + b T t t1 z t ñ Note that it is assumed that when making decisions over the forecast horizon beliefs are held Öxed ñ Formally we solve an anticipated utility problem in the sense of Kreps and Sargent ñ That agents do not account for subsequent revisions in their beliefs when making decisions in the current period represents the non-rationality of agents decisions

37 Townsend Investment Problem III The Örst-order conditions of the Örmís problem are k t k t1 = ^E t (k t+1 k t )+ p t and lim E t T k T =0 T!1 Solving the Örst condition forward and applying the second gives the optimal investment rule k t = k t1 + ^E t 1 X T =t T t p T ñ This is the optimal rule conditional on arbitrary beliefs: holds under rational expectations

38 Rational Expectations Solution and Learning Straightforward to show capital is determined by where 0 <<1 k t = K t1 + ( ) u t ñ Under learning dynamics, optimal decisions converge to this allocation with probability 1

39 Alternative Approaches The predominant approach to modeling learning in macroeconomics is to commence with Euler equations implied by a rational expectations analysis. In the context of the Townsend problem, commence with as the primitive behavioral assumption k t k t1 = ^E t (k t+1 k t ) + p t Requires a forecasting model k t =~a t + ~ bt K t1 +~c t u t1 + t so that ^E t k t+1 =~a t1 + ~ bt1 K t +~c t1 u t

40 Alternative Approaches II This second approach is fundamentally distinct from the Örst. transversality condition agents make suboptimal decisions By ignoring the Represents a basic confusion of the economics: Örms forecast their own endogenous decisions variable using an arbitrary model ñ Represents a contradiction of what agents know to be true about their decision problem ñ Forecasts of endogenous variable induces by optimal decision rules and primitive assumptions about forecasting variables beyond the control of the decision maker

41 It is not that the Euler equation is irrelevant. optimal decision rule yields Note that quasi di erencing the k t = k t1 + ^E X 1 t T t p T T =t = k t1 + p t + X 1 ^E t ^E t+1 T =t+1 = k t1 + p t + ^E t (k t+1 k t ) which is the Euler equation. T t p T Optimality requires capital to be forecast in a particular way. SpeciÖcally ^E t k t+1 = k t + ^E t 1 X T =t+1 T t p T ñ In general this will not be equal to ^E t k t+1 =~a t + ~ bt K t +~c t u t

42 Model Agents Households Firms Monetary authority Fiscal authority

43 Model Features Money or cashless limit Monopolistic competition/nominal rigidities Incomplete asset markets No capital Non-rational expectations

44 Beliefs Under rational expectations: 1. Agents optimize given beliefs 2. The probabilities they assign to future state variables coincide with the predictions of the model This paper retains (1) and replaces (2) with 2 0. Future state variables outside agentís control are forecasted using an econometric model.

45 Knowledge Know own preferences and constraints Do not know true economic model of determination of variables outside their control. ñ E.g. even if Y i t = Y j t = Y t in EQ ; agents know Y i t = Y j t Observe aggregate variables and disturbances Forecast variables outside their control using an atheoretical VAR ñ Takes the minimum state variable form

46 Household Problem Households seek to maximize subject to ^E t i 1X T =t T t h U C i T ; T v h i T ; T i Z 1 Bt i = (1 + i t1 ) Bt1 i + w th i t + t (j) dj T t P t Ct i 0 = (1 + i t1 ) Bt1 i + PYi t T t P t Ct i ñ Beliefs: a-theoretical VAR of exogenous variables ó nests MSV REE ñ Assume agents forecast period income

47 First-order conditions Log-linear approximation yields ^C i t = ^E i t ^C i t+1 i t ^E i t t+1 g t + ^E i tg t+1 and 1X 1X where ^E i t T =t T t ^C i T = Ai t + ^E i t T =t T t ^Y i T A i t = B i t=p t Y ; g t = t u c =u c ; 1 = u cc Y =u c ñ Assume government debt in zero net supply

48 Deriving the Consumption Decision Rule Solving the Euler equation backwards recursively from date T to date t and taking expectations at that time gives ^E i t ^C i T = ^C i t g t + ^E i t 2 4g T + TX 1 T =t (^{ t ^ t+1 ) ñ Use this to eliminate future expected consumption in the intertemporal budget constraint 3 5

49 Optimal decision rule First-order conditions imply ^C i t = (1 ) A i t + ^E i t 1X T =t T t h (1 ) ^Y i T ^{ T T +1 + gt g T +1 i ñ This is an example of permanent income theory ñ Forecasts about prices into the indeönite future matter! ñ A particularly important source of uncertainty are anticipated future interest rates

50 Comparison to Existing Approaches Are decisions implied by the optimal decision equivalent to those implied by ^C i t = ^E i t ^C i t+1 i t ^E i t t+1 g t + ^E t g t+1 ñ Where forecasts determined by the model z t = a t1 + b t1 z t1 + " t with z t = h ^C i t t i t g t i 0 and zt = h ^Y t t i t g t i 0

51 Comparison to Existing Approaches II Are decisions implied by the optimal decision equivalent to those implied by ^C i t = ^E i t ^C i t+1 i t ^E i t t+1 g t + ^E t g t+1 ñ Where forecasts determined by the model z t = a t1 + b t1 z t1 + " t with z t = h ^C i t t i t g t i 0 and zt = h ^Y t t i t g t i 0 In general - No! Optimal decision rule if expectations treated correctly ñ i.e. the conditional expectation of consumption induced by model primitives and optimality

52 Forecasting Consumption The optimal forecast ^E i t ^C i t+1 = (1 ) ^E i ta i t+1 + ^E t i 1X T =t+1 T t h (1 ) ^Y i T ^{ T T +1 + gt g T +1 i Not the same as ^E i tz t+1 = a t1 + b t1 z t ñ Conditional expectations of endogenous variables induced by policy function and forecasts about variable the are exogenous to the agentís decision problem

53 Commentary Market clearing conditions are part of the REE One-period-expectations approach does not deliver optimal decision rule ñ Confuses endogenous decision variables and exogenous state variables ñ The correct distribution with respect to which expectations are taken is induced from the decision problem and properties of exogenous variables ñ In general intertemporal budget constraint will be violated in this formulation Expected future consumption plans are inconsistent with what agents know to be true about their own future behavior given their own decision problem

54 Aggregate Consumption Dynamics Aggregating over the continuum provides X 1 ^C t = ^E t T t h (1 ) ^Y T ^{ T T +1 + gt g i T +1 T =t where ^C t Z ^C i tdi; ^E t Z ^E i tdi; 0 Z A i tdi ñ For any variable X t, ^E t ^E t+1 X t+1 6= ^E t X t+1

55 Summary Under rational expectations equilibrium probability laws by construction satisfy all relevant constraints ñ In particular: Transversality conditions Under learning dynamics expectations about future endogenous decision variables are taken with respected to a distribution ñ Induced by beliefs about exogenous states ñ Induced by optimal decisions

56 Firms Continuum of Örms maximize ^E j t 1X T =t T t Q t;t h j T (p t(j)) i where j T (p) = Y T P T p1 w i T f 1 Y T P T p =A T y i t = A t f h i t

57 Optimal Price Setting Log-linear approximation implies p j t = ^E j t 1X T =t () T t 2 4 (1 )! + 1 x T + T 1+! 3 5 ñ InÖnite horizon forecasts matter ñ x t = Y t ^Y n t is the output gap

58 Model Summary Aggregate demand and supply: x t = X 1 ^E t T t (1 ) x T +1 i T T +1 r T T =t t = X 1 x t + ^E t () T t x T +1 +(1) T +1 + u t T =t where r t = r t1 + " t and 0 <<1 u t = u t1 + t and 0 <<1

59 Rational Expectations Under this assumption ^E t ^E t+1 X t+1 = ^E t X t+1. Hence x t = E t [(1 ) x t+1 (i t t+1 r t )] + ^E t 1 X T =t+1 T t (1 ) x T +1 i T T +1 r T = E t (1 ) xt +1 (i t t+1 r t ) + E t x t+1 = E t x t+1 (i t E t t+1 r t )

60 Rational expectations analysis gives x t = E t x t+1 (i t E t t+1 r t ) t = kx t + E t t+1 Bullard and Mitra (2002, JME) and Evans and Honkapohja (2003, ReStud) x t = ^E t x t+1 i t ^E t t+1 r t t = kx t + ^E t t+1

61 Does any of this matter?

62 Objectives Implications of Learning for the choice of monetary policy rule The role of communication in policy design

63 Benchmark Model: Optimal Decisions Aggregate demand and supply: X 1 x t = ^E t T t (1 ) x T +1 i T T +1 r T T =t X 1 t = x t + ^E t () T t x T +1 +(1) T +1 + u t T =t where r t = r t1 + " t u t = u t1 + t are known exogenous processes with 0 < ; < 1 and " t, t iid disturbances Implicitly assumes Ricardian Öscal policy ñ Agents need not forecast debt or taxes

64 Alternative Model: ìone-period-ahead Forecastsî Interested in comparing policy emerging under the anticipated utility approach to the more familiar approach given by x t = ^E t x t+1 i t ^E t t+1 r t t = kx t + ^E t t+1 ñ Bullard and Mitra (JME, 2002), Evans and Honkapohja (ReStud, 2003)

65 Issues in Policy Design Early literature concerned with policy robustness and equilibrium selection ñ Given that there are a multiplicity of rules consistent with implementing a given equilibrium, can impose additional desiderata ñ Robustness to small expectational errors Models with learning also permit addressing critiques of policy ñ Friedman (1968) ó Monetary policy procedures based on interest-rate rules prone to Wicksellian cumulative process

66 An Interest-Rate Peg: Optimal Decisions Consider the policy i t = r r t where r t is iid and no u t ñ Indeterminacy of rational expectations equilibrium: Sargent and Wallace (1975) Agent forecasts given by ^E t z T = a t1 for z t = [ t x t i t ] 0 : Combined with structural model gives mapping da d db d = T (a) a = T (b) b ñ E-Stability determined by the eigenvalues of the Jacobian of these ordinary di erential equation

67 Conditions on Eigenvalues Consider the matrix A with dimension (3 3) : From ja Ij =0the characteristic equation is 3 c c 2 c 3 =0 where c 1 =Trace(A), c 2 is the sum of all second-order principal minors of A and c 3 = jaj. The following restrictions on the coe cients c i must be satisöed for all eigenvalues to have negative real parts: c 1 < 0 c 3 c 1 c 2 > 0 c 3 < 0:

68 For a matrix A with dimension (2 2) ; ja Ij =0implies the characteristic equation is 2 c 1 + c 2 =0 where c 1 =Trace(A) and c 2 = jaj. For both eigenvalues to have negative real parts c 1 < 0 and c 2 > 0 must be satisöed

69 Stability Analysis Straightforward to show and T 0 (a) = det(t 0 (a) I) = (1) ñ Expectations driven áuctuations (1 ) (1 ) > 0 ñ What about the Euler equation approach: Evans and Honkapohja (2003)

70 An Interest Rate Peg: One-Period-Ahead Expectations Consider the policy i t = r r t with r t i.i.d. Agent forecasts given by ^E t z t+1 = a t1 for z t =[ t x t ] 0 : Combined with structural model gives mapping da d db d = T (a) a = T (b) b

71 An Interest Rate Peg II The true data generating process is " # " xt 1 = ( + ) t which imply forecasts E t " xt+1 t+1 # = #" ax;t1 a ;t1 # " 1 ( + ) " (r 1) ( r 1) #" ax;t1 a ;t1 # # r t

72 DeÖning the T-map T a x a! = " 1 ( + ) #" ax a # Expectations stability requires the real parts of the eigenvalues of T 0 (a) I to be negative ñ That is: det T 0 (a) I > 0 and trace T 0 (a) I < 0 The determinant is det T 0 (a) I = < 0 ñ The minimum-state-variable solution is not learnable

73 Taylor Rule McCallum (1983): conditioning on endogenous variables resolves indeterminacy Consider the rule i t = t + x x t ñ Requires ( 1) + (1 ) x > 1 for determinacy under REE and stability under learning ñ Bullard and Mitra (2002), Preston (2005)

74 Emerging di erences Policy actions that depend on misspeciöed beliefs can lead to self-fulölling expectations Investigate a range of policies that are consistent with implementing the optimal commitment equilibrium under rational expectations ñ Implementation of such policy involve a dependence of interest-rate policy on expectations ñ This will reveal some sharp di erences across the anticipated utility approach and the one-period-ahead expectations approach

75 Optimal Commitment Policy under Rational Expectations Central bank is assumed to minimize the loss function E t0 X 1 tt 0 t=t 0 2 t + x 2 t where 0 <<1 for some weight >0 subject to (5) t = kx t + E t t+1 + u t t0 = t0 =(1 ) x t1 + 1 u t ñ Optimality from the ìtimeless perspectiveî: Woodford (2003, chap. Giannoni and Woodford (2002, 2012) 6) and

76 Optimal Commitment Solution Standard methods show that the optimal state-contingent paths of f t ;x t ;i t g are given by the following relations for all t t: t = (1 ) x t1 + 1 u t and x t = x t1 1 u t i t = (1 ) x t1 + ( + 1) 1 u t + 1 rn t where 0 <<1 is the modelís only eigenvalue within the unit circle.

77 Optimal Commitment Solution II Useful to represent this solution in terms of a particular exogenous state variable Solving the output gap relation backwards recursively gives x t = 1 1X j=0 j u tj : DeÖning the exogenous state variable t 1X j=0 j u tj ; which satisöes the process t = t1 + u t

78 Allows the optimal solution to be written as t = 1 [u t (1 ) t1 ] x t = 1 [ t1 + u t ] i t = 1 [u t (1 )( t1 + u t )] + 1 rn t where denotes the optimal solution when expressed as a linear function of f t1 ;u t ;r t g : ñ Note the question of what policy implements the optimal commitment equilibrium is yet to be answered ñ We have only described what that equilibrium looks like

79 Candidate Policy Rules Consider the following instrument rule as a means to implement the optimal equilibrium: i t = i t + x (E cb t x t+1 E t x t+1 )+ (E cb t t+1 E t t+1 ) where E cb t = { f t + xe cb t x t+1 + E cb t t+1 denotes the forecasts that the central bank intends to respond to and { l t = i t xe t x t+1 E t t+1 collects exogenous terms and E t x t+1 and E t t+1 ñ Recall f t ;x t ;i t g denote the optimal paths for each endogenous variable in the optimal commitment equilibrium when written as a linear function of the exogenous state variables f t1 ;u t ;r t g.

80 Candidate Policy Rules II Expectations based rules argued to be desirable and given empirical support ñ Batini and Haldane (1999), Levin, Wieland and Williams (2003), Clarida, Gali and Gertler (1998, 2000) Assume E cb t = ^E t where the forecasts are described below The policy parameters ; x determine the response to deviations of the actual path of the ináation rate and output gap expectations from that path consistent with the optimal equilibrium ñ In the desired equilibrium it is clear that Et cb t+1 = E t t+1, Ecb t x t+1 = E t x t+1 and i t = i t

81 Forecasts Agents therefore estimate the linear model z t = a t + b t z t1 + c t u t + d t r t + t (6) where z t =( t ;x t ;i t ; t ) 0, t is the usual error-vector term, fa t ;b t ;c t ;d t g are parameters to be estimated of the form and a t = a ;t a x;t a i;t a ;t ; c t = b t c ;t c x;t c i;t c ;t ; d t = b ;t b x;t b i;t b ;t d ;t d x;t d i;t d ;t 3 7 5

82 Forecasts II Believes revised according to recursive least squares formulation t = t1 + t 1 Rt 1 w t1 (z t1 wt1 0 t1) R t = R t1 + t 1 (w t1 wt1 0 R t1) where t =(a t ;b 4t ;c t ;d t )and w t f1; t1 ;u t ;r t g 0. Given homogeneity of beliefs, average forecasts can then be constructed by solving (9) backward and taking expectations to give ^E t z T = (I 4 b t1 ) 1 (I 4 b T t t1 )a t1 + b T t t1 z t +u t (I 4 b t1 ) 1 T t I 4 b T t t1 ct +r t (I 4 b t1 ) 1 T t I 4 b T t t1 dt for T t, where I 4 is a (4 4) identity matrix:

83 Properties A unique bounded rational expectations equilibrium will obtain i 0 <( 1) + (1 ) x < 2 (1 + ) Under learning a necessary condition for E-Stability is + x > (1 ) : ñ Clearly violates the Taylor principle; consider the limit case of! 1 ñ Di culty: by having central bank behavior depend on ìmisspeciöedî private beliefs makes self-fulölling expectations more likely Note central regulator of how shifting expectations impact current outcomes

84 Properties II What about the one-period-ahead expectations models? Can show that under the same rule expectation stability obtains if and only if ñ That is ó the Taylor principle holds ( 1) + (1 ) x > 1 ñ Fundamentally di erent predictions: arises from the one-period-ahead expectations approach gutting the model of economics. For example, expectations of future interest are irrelevant to dynamics in that approach

85 Raises important questions What is the appropriate use of forecasts in policy design? Can instability be mitigated? ñ Optimal policy/targeting rules imply a particular kind of dependency of interest rate decisions on forecasts ñ The role of communication ñ The role of debt policy Di erent statistical properties ñ Forecasts, dynamics and policy evaluation

86 Optimal Internal Forecasts Consider a sophisticated central bank that understands private-sector expectations formation Substituting the private forecasts into the structural relations implies the output gap and ináation to be given by t = a + b t1 +c u t + d r t +e i t (7) x t = a x + bx t1 +c x u t + dx r t +e x i t (8) where the coe cients a i ; bi ; c i ; di ; e i for i 2f; xg are functions of the coe cients (a t ;b t ;c t ;d t ) These equations give the period t ináation rate and output gap realizations, conditional on the current choice of the interest rate.

87 Can then compute ìoptimalî forecasts as Et cb t+1 = a + b t +c u t + d r t +e Et cb i t+1 Et cb x t+1 = a x + bx t +c x u t + dx r t +e x Et cb i t+1 : Substitution of these relations into yields an equation of the form i t ={ f t + xe cb t x t+1 + E cb t t+1 i t = ( a + x a x )+( b + x bx ) t +( c + x c x )u t +( d + xdx )r t +( e + x e x )Et cb i t+1 ñ Has unique bounded solution if j e + x e x j < 1 ñ Determines an interest-rate decision that is a function of agentsí learning behavior

88 Optimal Internal Forecasts II If the monetary authority constructs optimal internal forecasts then the Taylor principle ( 1) + (1 ) x > 0 is necessary and su cient for stability under learning dynamics. ñ Underscores an important point: understanding the details/mechanics of expectations formation critical ñ Policy needs to depend on expectations in a very speciöc way to ensure stability

89 Implementing Optimal Policy using Target Criteria Recent monetary economics literature has proposed implementing policy using target criteria ñ See, for example, Svensson and Woodford (2005), Svensson (2003), Giannoni and Woodford (2002, 2012) and Woodford (2003, chp. 7) The approach speciöes a criterion that must be checked each time an interest-rate decision is made ñ For example: Bank of Englandís ináation forecast targeting ñ Requires assumption about what model the central bank has when implementing policy ñ The approach gives very speciöc guidance on how forecasts should be incorporated in policy decisions

90 Deriving the Target Criterion Recall the optimal commitment equilibrium implies the dynamics t = (1 ) x t1 + 1 u t x t = x t1 1 u t i t = (1 ) x t1 + ( + 1) u t rn t

91 Deriving the Target Criterion II Giannoni and Woodford (2002) and Woodford (2003, chap 7) show t = (x t x t1 ) fully characterizes the optimal equilibrium from the timeless perspective in the sense that the previous dynamics hold for all t t 0 if and only if the above target criterion holds ñ Take this as the desired target criterion: under rational expectations implies determinacy for all parameter values ñ It is a restriction on endogenous variables ó central bank requires a model to implement ñ History dependence of optimal commitment captured by the lagged output gap

92 Forecasts Agents therefore estimate the linear model z t = a t + b t z t1 + c t u t + d t r t + t (9) where z t =( t ;x t ;i t ) 0, t is the usual error-vector term, fa t ;b t ;c t ;d t g are parameters to be estimated of the form and a t = a ;t a x;t a i;t ; c t = b t c ;t c x;t c i;t 3 0 b ;t 0 0 b x;t 0 0 b i;t ; d t = d ;t d x;t d i;t 3 7 5

93 Forecasts II Believes revised according to recursive least squares formulation t = t1 + t 1 Rt 1 w t1 (z t1 wt1 0 t1) R t = R t1 + t 1 (w t1 wt1 0 R t1) where t =(a t ;b 2t ;c t ;d t )and w t f1;x t1 ;u t ;r t g 0. Given homogeneity of beliefs, average forecasts can then be constructed by solving (9) backward and taking expectations to give ^E t z T = (I 3 b t1 ) 1 (I 3 b T t t1 )a t1 + b T t t1 z t +u t (I 3 b t1 ) 1 T t I 3 b T t t1 ct +r t (I 3 b t1 ) 1 T t I 3 b T t t1 dt for T t, where I 3 is a (3 3) identity matrix:

94 The Central Bankís Model A number of assumptions could be made. The central bank could: ñ Make projections based on minimum-state-variable rational expectations equilibrium ñ Make projections based on the structural equations implied by the learning model ñ Make projections based on the assumption that the Euler equations implied by rational expectations are valid.

95 Projecting under Rational Expectations Suppose the central bank mistakenly assumes agents have rational expectations If it projects the evolution of the economy assuming the minimum-state-variable equilibrium the target criterion will be satisöed under the interest-rate policy i t = (1 ) x t1 + ( + 1) 1 u t + 1 rn t ñ This is not an exogenous instrument rule ó depends on lagged output gap ñ Concerns of Sargent and Wallace (1975) need not be valid ñ Result: >su cient for instability under learning

96 Projecting under the True Model Suppose the central bank understands the true structural model ó that is, the aggregate demand and supply equations, and the precise details of agents belief formation The central bank can then guarantee satisfaction of the target criterion by setting interest rates according to i t = 1 ^x t + 1 X 1 ^E t T t (1 )x T +1 ( i T +1 T +1 )+r T T =t where ^x t = + 2x t1 + 2 ^E t 1X T =t () T t x T +1 +(1 ) T +1 + u T ñ Second relation is the value of the output gap that jointly satisöed the aggregate supply curve and the target criterion

97 This approach to implementing policy guarantees satisfaction of the target criterion regardless of the nature of expectations formation ñ Requires an interest-rate setting that depends upon expectations into the indefinite future ñ Is consistent with implementing the optimal commitment equilibrium if agents have rational expectations Under learning dynamics implies the underlying rational expectations equilibrium is E-Stable for all parameter values

98 Projecting using Euler Equations Suppose the central bank thinks the true model is x t = ^E t x t+1 i t ^E t t+1 r t t = kx t + ^E t t+1 Then the target criterion implies the instrument rule i t = 1 ^E t x t+1 + 2x t ^E t t u t + r n t What are the stability properties?

99 Final Remark on Target Criteria Under rational expectations the target criterion p t = x t implies the same state-contingent responses to disturbances as t = (x t x t1 ) Under learning they imply di erent outcomes ó that is, out of rational expectations equilibrium they imply di erent policy responses when the central bank Önds its target criterion is not met because expectations evolved in a way that was not anticipated ñ It turns out the price-level targeting is more conducive to stability

100 Communication as Stabilization Policy The discussion so far emphasizes that policy must be conditioned on expectations in certain ways An important source of instability is that agents do not know how interest rates are determined ñ This is one of the many equilibrium restrictions that they are attempting to learn ñ Agents beliefs need not be consistent with monetary policy strategy; coherent statement of ìunanchoredî expectations ñ But what if agents understood some details of policy? Does this improve the stability properties of forecast based instrument rules?

101 Communication as Stabilization Policy II Suppose the central bank declares that it interest rate decisions are determined by i t = E t t+1 Agents can use this information to restrict their forecasting model. SpeciÖcally, policy-consistent forecasts of the nominal interest rate can be determined as ^E t i T = ^E t T +1 ñ Such ináation and interest-rate forecasts are consistent with monetary policy strategy ñ They ensure projections satisfy the Taylor principle

102 Communication as Stabilization Policy III Policy consistent forecasts imply aggregate demand is X 1 x t = ^E t T t h (1 ) x T +1 ^E i t T +1 T +1 r T T =t ñ Agents need only forecast ináation and the output gap ñ Can show that E-Stability obtains so long as the Taylor principle holds ñ Communication about future policy intentions can resolve important uncertainty Taken up in detail by Eusepi and Preston (2010, AEJM) Now consider simpler analysis by Orphanides and Williams (2005)

103 Part III: Monetary and Fiscal Interactions The analysis so far has pushed Öscal policy to the back ground The remainder of the lecture will consider the consequences of debt management policy for the implementation of monetary policy ñ Interested in the role of scale and composition of the public debt ñ What kind of constraints do Öscal policy impose on monetary policy?

104 Motivation US Financial Crisis witnessed dramatic change in the nature and conduct of stabilization policy ñ Large-scale asset purchases intended to ináuence long interest rates ñ Under what conditions can this be e ective? Standard models satisfy an irrelevance proposition ñ Ricardian equivalence; more generally state-contingent consumption independent of government liabilities

105 What we do Simple model of output gap and ináation determination One informational friction: ñ Agents have an incomplete knowledge about the economy ñ Implication: departures from Ricardian Equivalence Explore constraints imposed on stabilization policy by choice of Öscal policy ñ SpeciÖcally: Scale and composition of Debt ñ Are there additional consequences of shortening the maturity structure of debt, over and above the presumed stimulus of lower interest rates

106 Literature Monetary and Öscal interactions: ñ Unpleasant monetary arithmetic: Sargent and Wallace (1980) ñ FTPL: Leeper, Sims and Woodford (early 90s) Learning: Marcet and Sargent (1989), Evans and Honkapohja (2001) Extensive literature on stability of expectations in NK models ñ Howitt (1992), Bullard and Mitra (2002) ñ Evans and Honkapohja (2011): survey of recent literature

107 Model Agents Households Firms Monetary authority Fiscal authority

108 Beliefs Under rational expectations: 1. Agents optimize given beliefs 2. The probabilities they assign to future state variables coincide with the predictions of the model This paper retains (1) and replaces (2) with 2 0. Future state variables outside agentís control are forecasted using an econometric model.

109 Knowledge Know own preferences and constraints Do not know true economic model determining variables outside their control. ñ E.g. even if Y i t = Y j t = Y t in EQ ; agents know Y i t = Y j t Observe aggregate variables and disturbances Forecast variables outside their control using an atheoretical VAR ñ Takes the minimum state variable form

110 Asset Structure and the Fiscal Authority Exogenous purchases of G t per period Issue two kinds of debt ñ B s t : One period debt in zero net supply with price P s t = (1 + i t) 1 ñ B m t : An asset in positive supply that has the payo structure T (t+1) for T t +1 Let P m t denote the price of this second asset. Asset has the properties: ñ Price in period t +1of debt issued in period t is P m t+1 ñ Average maturity of the debt is (1 ) 1

111 Monetary and Fiscal Authorities Flow budget constraint P m t B m t = B m t1 (1 + P m t ) P t S t Fiscal policy maintains intertemporal solvency (ëpassiveí) i t = i Bt1 m! i l ; i B m l >;i l Monetary policy controls ináation (ëactiveí) 1+i t 1+{ = t ; > 1 Under rational expectations: standard view of monetary policy

112 Household Problem Saving and work decision. No capital in the model, only gov. bonds. Householdsí preferences! U(C t ;H t ) = (1 ) 1 C t C 1+ tht 1+ 1, >0 where ñ =0! GHH; =1! KPR preferences ñ 1! IES 1; C t ;H t complements

113 Household Problem Household i maximizes (1 ) 1 ^E (i) t 1X T =t! 1 T t C t (i) C 1+ t(i)ht 1+ (i) subject to P s t B s t (i)+p m t B m t (i) (1 + P m t ) B m t1 (i)+bs t1 (i)+ 1 W t Wt H t (i) and No-Ponzi condition +P t t LS t P t C t (i) lim T!1 ^E (i) t Q i t;t h (1 + P m T ) B m T 1 (i)+bs T 1 (i)i =P T 0

114 Key Equation 1: Consumption Decision Combining Euler equations, labor supply, budget constraint to log-linear approximation provides ^C (i) t (1 ) ^H (i) t =s 1 P C (; ) m B m! P Y 1 1 t1 ^ t + ^P t m ^E (i) t 1X T =t T t h T LS^ LS T + T W ^ W T ^{ T ^ i 1 A T +1 {z } Gov. Debt and Taxes +s 1 C (; ) (1 ) ^E (i) t 1X T =t T t I W ^w T + I ^T + 1 ^E (i) t 1X T =t T t ^{ T ^ T +1

115 Key Equation 1: Consumption Decision Combining Euler equations, labor supply, budget constraint to log-linear approx. provides ^C (i) t (1 ) ^H (i) t = s 1 P C (; ) m B m! P Y 1 1 t1 ^ t + ^P t m ^E (i) t 1X T =t T t h T LS^ LS T + T W ^ W T ^{ T ^ i 1 A T +1 {z } Gov. Debt and Taxes +s 1 C (; ) (1 ) ^E (i) t 1X T =t T t I W ^w T + I ^T + 1 ^E (i) t 1X T =t T t ^{ T ^ T +1

116 Key Equation 1: Consumption Decision Combining Euler equations., labor supply, budget constraint to log-linear approx. provides ^C (i) t (1 ) ^H (i) t =s 1 P C (; ) m B m! P Y 1 1 t1 ^ t + ^P t m ^E (i) t 1X T =t T t h T LS^ LS T + T W ^ W T ^{ T ^ i 1 A T +1 {z } Gov. Debt and Taxes +s 1 C (; ) (1 ) ^E (i) t 1X T =t T t I W ^w T + I ^T + 1 ^E (i) t 1X T =t T t ^{ T ^ T +1

117 Key Equation 1: Consumption Decision Combining Euler equations., labor supply, budget constraint to log-linear approx. provides ^C (i) t (1 ) ^H (i) t =s 1 P C (; ) m B m! P Y 1 1 t1 ^ t + ^P t m ^E (i) t 1X T =t T t h T LS^ LS T + T W ^ W T ^{ T ^ i 1 A T +1 {z } Gov. Debt and Taxes +s 1 C (; ) (1 ) ^E (i) t 1X T =t T t I W ^w T + I ^T + 1 ^E (i) t 1X T =t T t ^{ T ^ T +1

118 Key Equation 2: Public Debt Price of government debt X ^P t m 1 = ^E t () T t ^{ T T =t {z } Expectation Hypothesis Evolution of public Debt ^bm t = 1 ^bm t1 ^ t + (1 )^{t 1 1 ^s t + (1 ) ^E t 1 X T =t () T t ^{ T +1

119 Firms Monopolistic competition + nominal rigidities (Rotemberg) Firm j has produces according to Y t (j) =H t (j) =(p t (j) =P t ) Y t p t (j) chosen to maximize expected proöts, where t (j) =[p t (j) W t ](p t (j) =P t ) Y t [p t (j) =p t1 (j) 1] 2

120 Firms Monopolistic competition + nominal rigidities (Rotemberg) In aggregate, the Phillips curve Y ^ t = +! C ^Y t + X 1 + ^E t () T t T =t ^w T +1 +(1 ) ^ T +1

121 Learning and Expectations (use ináation as an example) Rational Expectations (RE) ^E RE t T +1 = 0 + RE b ^bm t1, where T>t Learning ^E t T +1 = L c;t1 + L b;t1^b m t1 + lagged variables, Convergence to RE? E-Stability ñ Use methods of Marcet and Sargent (1989), Evans and Honkapohja (2001)

122 E-Stability and Learning Actual Law of Motion: t = T ( L c;t1 ; L b;t1 ) " 1 ^bm t1 # + ::: Convergence under least-squares learning IFF the associated ODE: d d ( c; b )= T ( L c ; L {z b ) } Actual Law of Motion is locally stable at the REE of interest: T ( RE c ( L c ; L b ) {z } Perceived Law of Motion ; RE b ) = ( RE c ; RE b ). Source of instability: economy as a self-referential system

123 E-Stability Numerical study: model with LUMP SUM taxation Key parameters: 1 =1=4,and P m B m 4 P Y =1:5 Others: ñ Preferences and technology: =0:99; =0:8 (Price stickiness) ñ Fiscal Policy: LS l =1:5 ñ Steady State: C=Y =0:78

124 1.6 E-stability: KPR vs. GHH preferences KPR preferences GHH preferences φ π *(1- βρ) - 1 Figure 1: E-stability properties in monetary policy ñ debt maturity space. Instability regiones lie above the contours.

125 Aggregate Demand: Increase in Ináation Expectations Ricardian Households ^C t (1 ) ^H t = 1 ^E t 1 X T =t T t ^ T ^ T s 1 C (; ) ^E t 1 X T =t T t I T ñ we have substituted for the bond price equation and the monetary policy rule, ^{ t = ^ t

126 Aggregate. Demand: Increase in Ináation Expectations Baseline ^C t (1 ) ^H t = 1 s 1 C (; ) S! Y ^E t 1 X T =t T t ^ T ^ T +1 + s 1 C (; ) S X 1 Y ^E t () T t ( ^ T )+ T =t +s 1 C (; ) S Y ^b m t1 +s1 C (; ) (1 ) ^E t 1 X T =t T t (I T T T ) :

127 Aggregate. Demand: Increase in Ináation Expectations Baseline ^C t (1 ) ^H t = 1 s 1 C (; ) S! Y ^E t 1 X T =t T t ^ T ^ T +1 + s 1 C (; ) S X 1 Y ^E t () T t ( ^ T )+ T =t +s 1 C (; ) S Y ^b m t1 +s1 C (; ) (1 ) ^E t 1 X T =t T t (I T T T ) :

128 Aggregate. Demand: Increase in Ináation Expectations Baseline ^C t (1 ) ^H t = 1 s 1 C (; ) S! Y ^E t 1 X T =t T t ^ T ^ T +1 + s 1 C (; ) S X 1 Y ^E t () T t ( ^ T )+ T =t +s 1 C (; ) S Y ^b m t1 +s1 C (; ) (1 ) ^E t 1 X T =t T t (I T T T ) :

129 Government Debt: Increase in Ináation Expectations Using the bond price equation and the monetary policy rule, ^{ t = ^ t, permits the evolution of real debt to be written as ^bm t = 1 ^bm t1 ^ t +(1 ) ^ t 1 1 ^s t + (1 ) ^E t 1 X T =t () T t ^ T +1 which depends on the maturity structure

130 Government Debt: Increase in Ináation Expectations Using the bond price equation and the monetary policy rule, ^{ t = ^ t, permits the evolution of real debt to be written as ^bm t = 1 ^bm t1 ^ t +(1 ) ^ t 1 1 ^s t + (1 ) ^E t 1 X T =t () T t ^ T +1 which depends on the maturity structure through (with exp. constant ináation) X 1 (1 ) ^E t () T t ^ T +1 = T =t (1 ) 1 ^E t ñ For =0:99 the right hand side peaks ' 0:9: average maturity of 2 years ñ For =1: dynamics independent of its own price.

131 1.8 E-stability: KPR preferences with σ = B/4Y =.7 B/4Y = 1.5 B/4Y = 2.2 φ π *(1- βρ) - 1 Figure 2: E-stability properties in monetary policy ñ debt maturity space. Instability regiones lie above the contours which are indexed by average level of indebtedness..

132 4.5 E-stability: KPR preferences B/4Y =.7 B/4Y = 1.5 B/4Y = 2.2 φ π σ Figure 3: E-stability properties in monetary policy ñ elasticity of substitution space. Instability regiones lie above the contours which are indexed by average level of indebtedness.

133 /s KPR C 1/s GHH C 1/σ- δ/s KPR C 1/σ- δ/s GHH C σ Figure 4: Decomposing wealth and subtitution e ects. The black lines plot s 1 C, which indexes the scale of wealth e ects from holdings of the public debt, for KPR and GHH preferences. The blue lines give the elasiticy of substitution of consumption deemand with respect to the real interest rate. Both are plotted as function of.

134 Forward-looking policy rules Alternative policy rules that responds to expectations ^{ t = ^E t^ t+1 Here we assume that agents know the policy rule

135 E-stability: forward-looking vs. baseline monetary policy rules Baseline rule Forward-looking rule φ π *(1- βρ) - 1 Figure 5: E-stability properties in monetary policy ñ debt maturity space for a standard Taylor rule and an expectations-based Taylor rule. Instability regions lie above the contours.

136 Intuition: Debt Dynamics Using the bond price equation and the monetary policy rule, ^{ t = ^E t^ t+1, permits the evolution of real debt to be written as ^bm t = 1 ^bm t1 ^ t + (1 ) ^E t^ t ^s t +(1 ) ^E t 1 X T =t () T t ^ T +2 ñ Debt depends on expectations regardless of ñ Destabilizing e ects on consumption and debt decline monotonically with

137 Distortionary taxation Consider the model labor taxes ( w t )......and assume now: 1 =1=2

138 1.9 E-stability: distortionary taxes B/4Y =.7 B/4Y = 1.5 B/4Y = 2.2 φ π *(1- βρ) - 1 Figure 6: E-stability properties in monetary policy debt maturity space. Instability regions lie above the contours indexed by average indebtedness.

139 Intuition: Distortionary taxation Same mechanism as before but with a ëtwistí Phillips curve has an extra term ^ t = " + Y C! ^Y t + w (1 w )^ w t # + ^E t 1 X T =t () T t ^w T +1 +(1 ) ^ T +1 Changes in tax rates as cost-push ìshocksî, reinforcing the drift in expectations

140 Conclusion Uncertainty about policy regime can induce drift in expectations High debt levels and short to medium maturity debt induce instability ñ Instability generated through wealth e ects Instability of expectations as ëöscal limití Fundamentally changes the nature of household and Örm responses to shocks ó even if expectations stable in the long-run

141 Motivation Revisited Macroeconomic models depend on household and Örm expectations Policy design ñ Not enough to observe some history of beliefs ñ Need a theory of the determination of beliefs: rational expectations Rational expectations policy design ñ Heavy reliance on managing expectations through announced commitments ñ What are the consequences of imprecise control of beliefs?

142 Motivation II For example, Bernanke (2004): ì[...] most private-sector borrowing and investment decisions depend not on the funds rate but on longer-term yields, such as mortgage rates and corporate bond rates, and on the prices of long-lived assets, such as housing and equities. Moreover, the link between these longer-term yields and asset prices and the current setting of the federal funds rate can be quite loose at times.î

143 The Agenda Simple model of output gap and ináation determination One informational friction: ñ Agents have an incomplete knowledge about the economy Explore constraints imposed on stabilization policy by Önancial market expectations ñ What if the expectations hypothesis of the yield curve does not hold? ñ Is the zero lower bound on nominal interest rates an important constraint?

144 Asset Structure and the Fiscal Authority Exogenous purchases of G t per period Issue two kinds of debt ñ B s t : One period debt in zero net supply with price P s t = (1 + i t) 1 ñ B m t : An asset in positive supply that has the payo structure T (t+1) for T t +1 Let P m t denote the price of this second asset. Asset has the properties: ñ Price in period t +1of debt issued in period t is P m t+1 ñ Average maturity of the debt is (1 ) 1

145 Asset Pricing: Theory I Log-linear approximation of Euler equations for each kind of bond holdings implies ^{ t = ^P t s = ^E t i ^P t m ^P t+1 m ñ Restriction on asset price movements ñ Combined with transversality X ^P t m 1 = ^E t () T t ^{ T T =t ñ Asset price is a function of fundamentals: the expectations hypothesis of the term structure holds; does not imply interest-rate expectations consistent with monetary policy strategy ñ Call: Anchored Financial Market Expectations

146 Optimal Spending Plans I Assume agents understand Öscal policy is Ricardian When the expectations hypothesis of the term structure holds ó aggregate demand relation of the form ^C i t = ^E i t +s 1 C 1X T =t T t i T T +1 (1 ) ^E i t " 1X T t T =t ñ Example of permanent income theory! # ^w T + 1^ T (10) ñ Independent of the average maturity structure of debt; Isomorphic to an economy with one-period debt ó i.e. =0 ñ Independent on long-debt-price expectations

147 Asset Pricing: Theory II Under non-rational expectations and incomplete markets: exist alternative ways to impose no-arbitrage Consider pricing the asset directly in all periods t ^{ t = ^E i t ^P m t ^P m t+1 ñ Given monetary policy and expectations, asset price determined. No-arbitrage consistent forecasts of the short-rate can then be determined by ^E t^{ i T = ^E t i ^P T m ^P T m +1 ñ Because interest rate projections are not directly related to current interest rates ó expectations hypothesis need not hold

148 Demand then determined by ^C i t = ^E i t +s 1 C 1X T =t T t h ^P m T +1 ^P m T (1 ) ^E i t " 1X T t T =t T +1 i! # ^w T + 1^T (11) ñ Nest anchored expectations model when =0;since ^E i t^{ T = ^E i t ^P T ñ In this case forecasting bond prices and forecasting interest rates equivalent As maturity increases this divergence increases

149 Beliefs Formation Agents construct forecasts according to ^E tx i t+t = a X t1 where X = n ; ^w; ^;^{; ^P mo for any T>0. ñ In period t forecasts are predetermined. ñ Beliefs are updated according to the constant gain algorithm a X t =(1 g) a X t1 + gx t where g>0 ñ With i.i.d. shocks and zero debt nests the REE Learning only about the constant

150 Beliefs Formation II Under anchored expectations, based on interest-rate data: ^E i t^{ T = a i t1 Under unanchored expectations, based on bond-price data: ^E t^{ T = ^E t ^P m T ^P m T +1 ñ In general they are not equal ñ Equivalent when =0 = (1 ) a P m t1

151 Calibration Discount factor is =0:99 Labor supply elasticity 1 =2 Nominal rigidities =0:75 Elasticity of demand across di erentiated goods =8

152 Unanchored Expectations and Simple Rules Consider the case of a simple Taylor rule ^{ t = ^ t + y x t Use notion of ìrobust stabilityî ñ Dynamics converge for a given gain coe cient ñ Distinct from E-Stability

153 φ π (policy resp. to inflation) φ x = 0 φ x =.5/4 φ x =.5 Stability with a Taylor Rule ρ (debt duration) Figure 7: Robust stability regions for di erent maturity structures. The three contours correspond to di erent Taylor distinguished by their respond to the current output gap.

154 Intuition Aggregate demand can be written x t = ^{ t + ^E t ^P m t+1 + ^E t 1 X T =t T t h (1 ) ^P m T +1 + T +1i ^A t +s 1 1 " X C (1 ) ^E t T t T =t! ^w T ^T +1# : ñ For inönite-period debt =1: depends only on ^E t ^P m t+1 ñ Requires aggressive adjustment of current nominal interest rates Key mechanism: increasing debt maturity implies arbitrage restrictions in the expectations hypothesis are weaker

Adaptive Learning and Applications in Monetary Policy. Noah Williams

Adaptive Learning and Applications in Monetary Policy. Noah Williams Adaptive Learning and Applications in Monetary Policy Noah University of Wisconsin - Madison Econ 899 Motivations J. C. Trichet: Understanding expectations formation as a process underscores the strategic

More information

Seoul National University Mini-Course: Monetary & Fiscal Policy Interactions II

Seoul National University Mini-Course: Monetary & Fiscal Policy Interactions II Seoul National University Mini-Course: Monetary & Fiscal Policy Interactions II Eric M. Leeper Indiana University July/August 2013 Linear Analysis Generalize policy behavior with a conventional parametric

More information

The Basic New Keynesian Model. Jordi Galí. June 2008

The Basic New Keynesian Model. Jordi Galí. June 2008 The Basic New Keynesian Model by Jordi Galí June 28 Motivation and Outline Evidence on Money, Output, and Prices: Short Run E ects of Monetary Policy Shocks (i) persistent e ects on real variables (ii)

More information

Learning and Global Dynamics

Learning and Global Dynamics Learning and Global Dynamics James Bullard 10 February 2007 Learning and global dynamics The paper for this lecture is Liquidity Traps, Learning and Stagnation, by George Evans, Eran Guse, and Seppo Honkapohja.

More information

Lars Svensson 2/16/06. Y t = Y. (1) Assume exogenous constant government consumption (determined by government), G t = G<Y. (2)

Lars Svensson 2/16/06. Y t = Y. (1) Assume exogenous constant government consumption (determined by government), G t = G<Y. (2) Eco 504, part 1, Spring 2006 504_L3_S06.tex Lars Svensson 2/16/06 Specify equilibrium under perfect foresight in model in L2 Assume M 0 and B 0 given. Determine {C t,g t,y t,m t,b t,t t,r t,i t,p t } that

More information

Expectations, Learning and Macroeconomic Policy

Expectations, Learning and Macroeconomic Policy Expectations, Learning and Macroeconomic Policy George W. Evans (Univ. of Oregon and Univ. of St. Andrews) Lecture 4 Liquidity traps, learning and stagnation Evans, Guse & Honkapohja (EER, 2008), Evans

More information

Learning to Optimize: Theory and Applications

Learning to Optimize: Theory and Applications Learning to Optimize: Theory and Applications George W. Evans University of Oregon and University of St Andrews Bruce McGough University of Oregon WAMS, December 12, 2015 Outline Introduction Shadow-price

More information

Learning about Monetary Policy Rules when Long-Horizon Expectations Matter

Learning about Monetary Policy Rules when Long-Horizon Expectations Matter Learning about Monetary Policy Rules when Long-Horizon Expectations Matter Bruce Preston Princeton University May 1, 2003 Abstract This paper considers the implications of an important source of model

More information

Learning and Monetary Policy

Learning and Monetary Policy Learning and Monetary Policy Lecture 1 Introduction to Expectations and Adaptive Learning George W. Evans (University of Oregon) University of Paris X -Nanterre (September 2007) J. C. Trichet: Understanding

More information

Federal Reserve Bank of New York Staff Reports

Federal Reserve Bank of New York Staff Reports Federal Reserve Bank of New York Staff Reports Central Bank Transparency under Model Uncertainty Stefano Eusepi Staff Report no. 99 January 2005 This paper presents preliminary findings and is being distributed

More information

The Basic New Keynesian Model. Jordi Galí. November 2010

The Basic New Keynesian Model. Jordi Galí. November 2010 The Basic New Keynesian Model by Jordi Galí November 2 Motivation and Outline Evidence on Money, Output, and Prices: Short Run E ects of Monetary Policy Shocks (i) persistent e ects on real variables (ii)

More information

Lecture 3, November 30: The Basic New Keynesian Model (Galí, Chapter 3)

Lecture 3, November 30: The Basic New Keynesian Model (Galí, Chapter 3) MakØk3, Fall 2 (blok 2) Business cycles and monetary stabilization policies Henrik Jensen Department of Economics University of Copenhagen Lecture 3, November 3: The Basic New Keynesian Model (Galí, Chapter

More information

Volume 29, Issue 4. Stability under learning: the neo-classical growth problem

Volume 29, Issue 4. Stability under learning: the neo-classical growth problem Volume 29, Issue 4 Stability under learning: the neo-classical growth problem Orlando Gomes ISCAL - IPL; Economics Research Center [UNIDE/ISCTE - ERC] Abstract A local stability condition for the standard

More information

Fiscal Multipliers in a Nonlinear World

Fiscal Multipliers in a Nonlinear World Fiscal Multipliers in a Nonlinear World Jesper Lindé Sveriges Riksbank Mathias Trabandt Freie Universität Berlin November 28, 2016 Lindé and Trabandt Multipliers () in Nonlinear Models November 28, 2016

More information

Taylor Rules and Technology Shocks

Taylor Rules and Technology Shocks Taylor Rules and Technology Shocks Eric R. Sims University of Notre Dame and NBER January 17, 2012 Abstract In a standard New Keynesian model, a Taylor-type interest rate rule moves the equilibrium real

More information

Policy Inertia and Equilibrium Determinacy in a New. Keynesian Model with Investment

Policy Inertia and Equilibrium Determinacy in a New. Keynesian Model with Investment Policy Inertia and Equilibrium Determinacy in a New Keynesian Model with Investment Wei Xiao State University of New York at Binghamton June, 2007 Abstract Carlstrom and Fuerst (2005) demonstrate that

More information

Simple New Keynesian Model without Capital

Simple New Keynesian Model without Capital Simple New Keynesian Model without Capital Lawrence J. Christiano March, 28 Objective Review the foundations of the basic New Keynesian model without capital. Clarify the role of money supply/demand. Derive

More information

Dynamic stochastic general equilibrium models. December 4, 2007

Dynamic stochastic general equilibrium models. December 4, 2007 Dynamic stochastic general equilibrium models December 4, 2007 Dynamic stochastic general equilibrium models Random shocks to generate trajectories that look like the observed national accounts. Rational

More information

Chapter 6. Maximum Likelihood Analysis of Dynamic Stochastic General Equilibrium (DSGE) Models

Chapter 6. Maximum Likelihood Analysis of Dynamic Stochastic General Equilibrium (DSGE) Models Chapter 6. Maximum Likelihood Analysis of Dynamic Stochastic General Equilibrium (DSGE) Models Fall 22 Contents Introduction 2. An illustrative example........................... 2.2 Discussion...................................

More information

Optimal Simple And Implementable Monetary and Fiscal Rules

Optimal Simple And Implementable Monetary and Fiscal Rules Optimal Simple And Implementable Monetary and Fiscal Rules Stephanie Schmitt-Grohé Martín Uribe Duke University September 2007 1 Welfare-Based Policy Evaluation: Related Literature (ex: Rotemberg and Woodford,

More information

Simple New Keynesian Model without Capital

Simple New Keynesian Model without Capital Simple New Keynesian Model without Capital Lawrence J. Christiano January 5, 2018 Objective Review the foundations of the basic New Keynesian model without capital. Clarify the role of money supply/demand.

More information

ECON 5118 Macroeconomic Theory

ECON 5118 Macroeconomic Theory ECON 5118 Macroeconomic Theory Winter 013 Test 1 February 1, 013 Answer ALL Questions Time Allowed: 1 hour 0 min Attention: Please write your answers on the answer book provided Use the right-side pages

More information

Learning about Monetary Policy Rules when Long-Horizon Expectations Matter

Learning about Monetary Policy Rules when Long-Horizon Expectations Matter Learning about Monetary Policy Rules when Long-Horizon Expectations Matter Bruce Preston Columbia University This paper considers the implications of an important source of model misspecification for the

More information

Central Bank Communication and the Liquidity Trap

Central Bank Communication and the Liquidity Trap Central Bank Communication and the Liquidity Trap Stefano Eusepi y Federal Reserve Bank of New York October 8, 2009 Abstract Central bank communication plays an important role in shaping market participants

More information

Advanced Macroeconomics

Advanced Macroeconomics Advanced Macroeconomics The Ramsey Model Marcin Kolasa Warsaw School of Economics Marcin Kolasa (WSE) Ad. Macro - Ramsey model 1 / 30 Introduction Authors: Frank Ramsey (1928), David Cass (1965) and Tjalling

More information

Government The government faces an exogenous sequence {g t } t=0

Government The government faces an exogenous sequence {g t } t=0 Part 6 1. Borrowing Constraints II 1.1. Borrowing Constraints and the Ricardian Equivalence Equivalence between current taxes and current deficits? Basic paper on the Ricardian Equivalence: Barro, JPE,

More information

The Design of Monetary and Fiscal Policy: A Global Perspective

The Design of Monetary and Fiscal Policy: A Global Perspective The Design of Monetary and Fiscal Policy: A Global Perspective Jess Benhabib New York University Stefano Eusepi Federal Reseve Bank of New York Very Preliminary-Comments Welcome November 6, 004 Abstract

More information

Problem 1 (30 points)

Problem 1 (30 points) Problem (30 points) Prof. Robert King Consider an economy in which there is one period and there are many, identical households. Each household derives utility from consumption (c), leisure (l) and a public

More information

problem. max Both k (0) and h (0) are given at time 0. (a) Write down the Hamilton-Jacobi-Bellman (HJB) Equation in the dynamic programming

problem. max Both k (0) and h (0) are given at time 0. (a) Write down the Hamilton-Jacobi-Bellman (HJB) Equation in the dynamic programming 1. Endogenous Growth with Human Capital Consider the following endogenous growth model with both physical capital (k (t)) and human capital (h (t)) in continuous time. The representative household solves

More information

Indeterminacy and Sunspots in Macroeconomics

Indeterminacy and Sunspots in Macroeconomics Indeterminacy and Sunspots in Macroeconomics Friday September 8 th : Lecture 10 Gerzensee, September 2017 Roger E. A. Farmer Warwick University and NIESR Topics for Lecture 10 Tying together the pieces

More information

Resolving the Missing Deflation Puzzle. June 7, 2018

Resolving the Missing Deflation Puzzle. June 7, 2018 Resolving the Missing Deflation Puzzle Jesper Lindé Sveriges Riksbank Mathias Trabandt Freie Universität Berlin June 7, 218 Motivation Key observations during the Great Recession: Extraordinary contraction

More information

Signaling Effects of Monetary Policy

Signaling Effects of Monetary Policy Signaling Effects of Monetary Policy Leonardo Melosi London Business School 24 May 2012 Motivation Disperse information about aggregate fundamentals Morris and Shin (2003), Sims (2003), and Woodford (2002)

More information

Lecture 6, January 7 and 15: Sticky Wages and Prices (Galí, Chapter 6)

Lecture 6, January 7 and 15: Sticky Wages and Prices (Galí, Chapter 6) MakØk3, Fall 2012/2013 (Blok 2) Business cycles and monetary stabilization policies Henrik Jensen Department of Economics University of Copenhagen Lecture 6, January 7 and 15: Sticky Wages and Prices (Galí,

More information

The New Keynesian Model: Introduction

The New Keynesian Model: Introduction The New Keynesian Model: Introduction Vivaldo M. Mendes ISCTE Lisbon University Institute 13 November 2017 (Vivaldo M. Mendes) The New Keynesian Model: Introduction 13 November 2013 1 / 39 Summary 1 What

More information

Monetary Policy Design in the Basic New Keynesian Model. Jordi Galí. October 2015

Monetary Policy Design in the Basic New Keynesian Model. Jordi Galí. October 2015 Monetary Policy Design in the Basic New Keynesian Model by Jordi Galí October 2015 The E cient Allocation where C t R 1 0 C t(i) 1 1 1 di Optimality conditions: max U (C t ; N t ; Z t ) subject to: C t

More information

Lecture 4 The Centralized Economy: Extensions

Lecture 4 The Centralized Economy: Extensions Lecture 4 The Centralized Economy: Extensions Leopold von Thadden University of Mainz and ECB (on leave) Advanced Macroeconomics, Winter Term 2013 1 / 36 I Motivation This Lecture considers some applications

More information

Advanced Macroeconomics

Advanced Macroeconomics Advanced Macroeconomics The Ramsey Model Micha l Brzoza-Brzezina/Marcin Kolasa Warsaw School of Economics Micha l Brzoza-Brzezina/Marcin Kolasa (WSE) Ad. Macro - Ramsey model 1 / 47 Introduction Authors:

More information

Capital Structure and Investment Dynamics with Fire Sales

Capital Structure and Investment Dynamics with Fire Sales Capital Structure and Investment Dynamics with Fire Sales Douglas Gale Piero Gottardi NYU April 23, 2013 Douglas Gale, Piero Gottardi (NYU) Capital Structure April 23, 2013 1 / 55 Introduction Corporate

More information

Solutions to Problem Set 4 Macro II (14.452)

Solutions to Problem Set 4 Macro II (14.452) Solutions to Problem Set 4 Macro II (14.452) Francisco A. Gallego 05/11 1 Money as a Factor of Production (Dornbusch and Frenkel, 1973) The shortcut used by Dornbusch and Frenkel to introduce money in

More information

Stagnation Traps. Gianluca Benigno and Luca Fornaro

Stagnation Traps. Gianluca Benigno and Luca Fornaro Stagnation Traps Gianluca Benigno and Luca Fornaro May 2015 Research question and motivation Can insu cient aggregate demand lead to economic stagnation? This question goes back, at least, to the Great

More information

MAGYAR NEMZETI BANK MINI-COURSE

MAGYAR NEMZETI BANK MINI-COURSE MAGYAR NEMZETI BANK MINI-COURSE LECTURE 3. POLICY INTERACTIONS WITH TAX DISTORTIONS Eric M. Leeper Indiana University September 2008 THE MESSAGES Will study three models with distorting taxes First draws

More information

Two-sided Learning in New Keynesian Models: Dynamics, (Lack of) Convergence and the Value of Information

Two-sided Learning in New Keynesian Models: Dynamics, (Lack of) Convergence and the Value of Information Dynare Working Papers Series http://www.dynare.org/wp/ Two-sided Learning in New Keynesian Models: Dynamics, (Lack of) Convergence and the Value of Information Christian Matthes Francesca Rondina Working

More information

Adaptive Learning in Macroeconomics: some methodological issues

Adaptive Learning in Macroeconomics: some methodological issues Adaptive Learning in Macroeconomics: some methodological issues George W. Evans University of Oregon and University of St. Andrews INEXC Paris Conference June 27-29, 2012 J. C. Trichet: Understanding expectations

More information

Session 4: Money. Jean Imbs. November 2010

Session 4: Money. Jean Imbs. November 2010 Session 4: Jean November 2010 I So far, focused on real economy. Real quantities consumed, produced, invested. No money, no nominal in uences. I Now, introduce nominal dimension in the economy. First and

More information

The Design of Monetary and Fiscal Policy: A Global Perspective

The Design of Monetary and Fiscal Policy: A Global Perspective The Design of Monetary and Fiscal Policy: A Global Perspective Jess Benhabib New York University Stefano Eusepi Federal Reseve Bank of New York January 4, 005 Abstract We study the the emergence of multiple

More information

1 The Basic RBC Model

1 The Basic RBC Model IHS 2016, Macroeconomics III Michael Reiter Ch. 1: Notes on RBC Model 1 1 The Basic RBC Model 1.1 Description of Model Variables y z k L c I w r output level of technology (exogenous) capital at end of

More information

Forward Guidance without Common Knowledge

Forward Guidance without Common Knowledge Forward Guidance without Common Knowledge George-Marios Angeletos 1 Chen Lian 2 1 MIT and NBER 2 MIT November 17, 2017 Outline 1 Introduction 2 Environment 3 GE Attenuation and Horizon Effects 4 Forward

More information

Advanced Macroeconomics II. Monetary Models with Nominal Rigidities. Jordi Galí Universitat Pompeu Fabra April 2018

Advanced Macroeconomics II. Monetary Models with Nominal Rigidities. Jordi Galí Universitat Pompeu Fabra April 2018 Advanced Macroeconomics II Monetary Models with Nominal Rigidities Jordi Galí Universitat Pompeu Fabra April 208 Motivation Empirical Evidence Macro evidence on the e ects of monetary policy shocks (i)

More information

Demand Shocks, Monetary Policy, and the Optimal Use of Dispersed Information

Demand Shocks, Monetary Policy, and the Optimal Use of Dispersed Information Demand Shocks, Monetary Policy, and the Optimal Use of Dispersed Information Guido Lorenzoni (MIT) WEL-MIT-Central Banks, December 2006 Motivation Central bank observes an increase in spending Is it driven

More information

Deviant Behavior in Monetary Economics

Deviant Behavior in Monetary Economics Deviant Behavior in Monetary Economics Lawrence Christiano and Yuta Takahashi July 26, 2018 Multiple Equilibria Standard NK Model Standard, New Keynesian (NK) Monetary Model: Taylor rule satisfying Taylor

More information

Open Market Operations and Money Supply. at Zero Nominal Interest Rates

Open Market Operations and Money Supply. at Zero Nominal Interest Rates Open Market Operations and Money Supply at Zero Nominal Interest Rates Roberto Robatto March 12, 2014 Abstract I present an irrelevance proposition for some open market operations that exchange money and

More information

Simple New Keynesian Model without Capital

Simple New Keynesian Model without Capital Simple New Keynesian Model without Capital Lawrence J. Christiano Gerzensee, August 27 Objective Review the foundations of the basic New Keynesian model without capital. Clarify the role of money supply/demand.

More information

Taylor rules with delays in continuous time

Taylor rules with delays in continuous time Taylor rules with delays in continuous time ECON 101 (2008) Taylor rules with delays in continuous time 1 / 38 The model We follow the flexible price model introduced in Benhabib, Schmitt-Grohe and Uribe

More information

Advanced Economic Growth: Lecture 8, Technology Di usion, Trade and Interdependencies: Di usion of Technology

Advanced Economic Growth: Lecture 8, Technology Di usion, Trade and Interdependencies: Di usion of Technology Advanced Economic Growth: Lecture 8, Technology Di usion, Trade and Interdependencies: Di usion of Technology Daron Acemoglu MIT October 3, 2007 Daron Acemoglu (MIT) Advanced Growth Lecture 8 October 3,

More information

Chapter 1. GMM: Basic Concepts

Chapter 1. GMM: Basic Concepts Chapter 1. GMM: Basic Concepts Contents 1 Motivating Examples 1 1.1 Instrumental variable estimator....................... 1 1.2 Estimating parameters in monetary policy rules.............. 2 1.3 Estimating

More information

Monetary Policy and Equilibrium Indeterminacy in a Cash in Advance Economy with Investment. Abstract

Monetary Policy and Equilibrium Indeterminacy in a Cash in Advance Economy with Investment. Abstract Monetary Policy and Equilibrium Indeterminacy in a Cash in Advance Economy with Investment Chong Kee Yip Department of Economics, The Chinese University of Hong Kong Ka Fai Li Department of Economics,

More information

Chapter 4. Applications/Variations

Chapter 4. Applications/Variations Chapter 4 Applications/Variations 149 4.1 Consumption Smoothing 4.1.1 The Intertemporal Budget Economic Growth: Lecture Notes For any given sequence of interest rates {R t } t=0, pick an arbitrary q 0

More information

PANEL DISCUSSION: THE ROLE OF POTENTIAL OUTPUT IN POLICYMAKING

PANEL DISCUSSION: THE ROLE OF POTENTIAL OUTPUT IN POLICYMAKING PANEL DISCUSSION: THE ROLE OF POTENTIAL OUTPUT IN POLICYMAKING James Bullard* Federal Reserve Bank of St. Louis 33rd Annual Economic Policy Conference St. Louis, MO October 17, 2008 Views expressed are

More information

Learning and Monetary Policy. Lecture 4 Recent Applications of Learning to Monetary Policy

Learning and Monetary Policy. Lecture 4 Recent Applications of Learning to Monetary Policy Learning and Monetary Policy Lecture 4 Recent Applications of Learning to Monetary Policy George W. Evans (University of Oregon) University of Paris X -Nanterre (September 2007) Lecture 4 Outline This

More information

Discussion of. "Indeterminacy with Inflation-Forecast- Based Rules in a Two-Bloc Model" N. Batini, P. Levine and J. Pearlman

Discussion of. Indeterminacy with Inflation-Forecast- Based Rules in a Two-Bloc Model N. Batini, P. Levine and J. Pearlman Discussion of "Indeterminacy with Inflation-Forecast- Based Rules in a Two-Bloc Model" N. Batini, P. Levine and J. Pearlman Marc Giannoni Columbia University International Research Forum on Monetary Policy

More information

Comprehensive Exam. Macro Spring 2014 Retake. August 22, 2014

Comprehensive Exam. Macro Spring 2014 Retake. August 22, 2014 Comprehensive Exam Macro Spring 2014 Retake August 22, 2014 You have a total of 180 minutes to complete the exam. If a question seems ambiguous, state why, sharpen it up and answer the sharpened-up question.

More information

Lecture 9: The monetary theory of the exchange rate

Lecture 9: The monetary theory of the exchange rate Lecture 9: The monetary theory of the exchange rate Open Economy Macroeconomics, Fall 2006 Ida Wolden Bache October 24, 2006 Macroeconomic models of exchange rate determination Useful reference: Chapter

More information

Foundations of Modern Macroeconomics Second Edition

Foundations of Modern Macroeconomics Second Edition Foundations of Modern Macroeconomics Second Edition Chapter 5: The government budget deficit Ben J. Heijdra Department of Economics & Econometrics University of Groningen 1 September 2009 Foundations of

More information

Research Division Federal Reserve Bank of St. Louis Working Paper Series

Research Division Federal Reserve Bank of St. Louis Working Paper Series Research Division Federal Reserve Bank of St. Louis Working Paper Series The Stability of Macroeconomic Systems with Bayesian Learners James Bullard and Jacek Suda Working Paper 2008-043B http://research.stlouisfed.org/wp/2008/2008-043.pdf

More information

Y t = log (employment t )

Y t = log (employment t ) Advanced Macroeconomics, Christiano Econ 416 Homework #7 Due: November 21 1. Consider the linearized equilibrium conditions of the New Keynesian model, on the slide, The Equilibrium Conditions in the handout,

More information

Foundations of Modern Macroeconomics B. J. Heijdra & F. van der Ploeg Chapter 6: The Government Budget Deficit

Foundations of Modern Macroeconomics B. J. Heijdra & F. van der Ploeg Chapter 6: The Government Budget Deficit Foundations of Modern Macroeconomics: Chapter 6 1 Foundations of Modern Macroeconomics B. J. Heijdra & F. van der Ploeg Chapter 6: The Government Budget Deficit Foundations of Modern Macroeconomics: Chapter

More information

Time-varying Consumption Tax, Productive Government Spending, and Aggregate Instability.

Time-varying Consumption Tax, Productive Government Spending, and Aggregate Instability. Time-varying Consumption Tax, Productive Government Spending, and Aggregate Instability. Literature Schmitt-Grohe and Uribe (JPE 1997): Ramsey model with endogenous labor income tax + balanced budget (fiscal)

More information

Foundations for the New Keynesian Model. Lawrence J. Christiano

Foundations for the New Keynesian Model. Lawrence J. Christiano Foundations for the New Keynesian Model Lawrence J. Christiano Objective Describe a very simple model economy with no monetary frictions. Describe its properties. markets work well Modify the model to

More information

Sophisticated Monetary Policies

Sophisticated Monetary Policies Federal Reserve Bank of Minneapolis Research Department Sta Report 419 January 2008 Sophisticated Monetary Policies Andrew Atkeson University of California, Los Angeles, Federal Reserve Bank of Minneapolis,

More information

Demand Shocks with Dispersed Information

Demand Shocks with Dispersed Information Demand Shocks with Dispersed Information Guido Lorenzoni (MIT) Class notes, 06 March 2007 Nominal rigidities: imperfect information How to model demand shocks in a baseline environment with imperfect info?

More information

DSGE-Models. Calibration and Introduction to Dynare. Institute of Econometrics and Economic Statistics

DSGE-Models. Calibration and Introduction to Dynare. Institute of Econometrics and Economic Statistics DSGE-Models Calibration and Introduction to Dynare Dr. Andrea Beccarini Willi Mutschler, M.Sc. Institute of Econometrics and Economic Statistics willi.mutschler@uni-muenster.de Summer 2012 Willi Mutschler

More information

STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics

STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics Ph. D. Comprehensive Examination: Macroeconomics Fall, 202 Answer Key to Section 2 Questions Section. (Suggested Time: 45 Minutes) For 3 of

More information

Expectations and Monetary Policy

Expectations and Monetary Policy Expectations and Monetary Policy Elena Gerko London Business School March 24, 2017 Abstract This paper uncovers a new channel of monetary policy in a New Keynesian DSGE model under the assumption of internal

More information

Deep Habits, Nominal Rigidities and Interest Rate Rules

Deep Habits, Nominal Rigidities and Interest Rate Rules Deep Habits, Nominal Rigidities and Interest Rate Rules Sarah Zubairy August 18, 21 Abstract This paper explores how the introduction of deep habits in a standard new-keynesian model affects the properties

More information

Monetary Economics Notes

Monetary Economics Notes Monetary Economics Notes Nicola Viegi 2 University of Pretoria - School of Economics Contents New Keynesian Models. Readings...............................2 Basic New Keynesian Model...................

More information

Discussion Papers in Economics

Discussion Papers in Economics Discussion Papers in Economics No. 2000/62 2000/60 Dynamics Desirability of of Output Nominal Growth, GDP Targeting Consumption Under and Adaptive Physical Learning Capital in Two-Sector Models of Endogenous

More information

On Determinacy and Learnability in a New Keynesian Model with Unemployment

On Determinacy and Learnability in a New Keynesian Model with Unemployment On Determinacy and Learnability in a New Keynesian Model with Unemployment Mewael F. Tesfaselassie Eric Schaling This Version February 009 Abstract We analyze determinacy and stability under learning (E-stability)

More information

Monetary Policy, Expectations and Commitment

Monetary Policy, Expectations and Commitment Monetary Policy, Expectations and Commitment George W. Evans University of Oregon Seppo Honkapohja University of Helsinki April 30, 2003; with corrections 2/-04 Abstract Commitment in monetary policy leads

More information

Getting to page 31 in Galí (2008)

Getting to page 31 in Galí (2008) Getting to page 31 in Galí 2008) H J Department of Economics University of Copenhagen December 4 2012 Abstract This note shows in detail how to compute the solutions for output inflation and the nominal

More information

Optimal Inflation Stabilization in a Medium-Scale Macroeconomic Model

Optimal Inflation Stabilization in a Medium-Scale Macroeconomic Model Optimal Inflation Stabilization in a Medium-Scale Macroeconomic Model Stephanie Schmitt-Grohé Martín Uribe Duke University 1 Objective of the Paper: Within a mediumscale estimated model of the macroeconomy

More information

Fiscal Multipliers in a Nonlinear World

Fiscal Multipliers in a Nonlinear World Fiscal Multipliers in a Nonlinear World Jesper Lindé and Mathias Trabandt ECB-EABCN-Atlanta Nonlinearities Conference, December 15-16, 2014 Sveriges Riksbank and Federal Reserve Board December 16, 2014

More information

1 The social planner problem

1 The social planner problem The social planner problem max C t;k t+ U t = E t X t C t () that can be written as: s.t.: Y t = A t K t (2) Y t = C t + I t (3) I t = K t+ (4) A t = A A t (5) et t i:i:d: 0; 2 max C t;k t+ U t = E t "

More information

Dynamic Optimization: An Introduction

Dynamic Optimization: An Introduction Dynamic Optimization An Introduction M. C. Sunny Wong University of San Francisco University of Houston, June 20, 2014 Outline 1 Background What is Optimization? EITM: The Importance of Optimization 2

More information

The Basic New Keynesian Model, the Labor Market and Sticky Wages

The Basic New Keynesian Model, the Labor Market and Sticky Wages The Basic New Keynesian Model, the Labor Market and Sticky Wages Lawrence J. Christiano August 25, 203 Baseline NK model with no capital and with a competitive labor market. private sector equilibrium

More information

Monetary policy at the zero lower bound: Theory

Monetary policy at the zero lower bound: Theory Monetary policy at the zero lower bound: Theory A. Theoretical channels 1. Conditions for complete neutrality (Eggertsson and Woodford, 2003) 2. Market frictions 3. Preferred habitat and risk-bearing (Hamilton

More information

Macroeconomics Theory II

Macroeconomics Theory II Macroeconomics Theory II Francesco Franco FEUNL February 2016 Francesco Franco (FEUNL) Macroeconomics Theory II February 2016 1 / 18 Road Map Research question: we want to understand businesses cycles.

More information

Learning in Macroeconomic Models

Learning in Macroeconomic Models Learning in Macroeconomic Models Wouter J. Den Haan London School of Economics c by Wouter J. Den Haan Overview A bit of history of economic thought How expectations are formed can matter in the long run

More information

(a) Write down the Hamilton-Jacobi-Bellman (HJB) Equation in the dynamic programming

(a) Write down the Hamilton-Jacobi-Bellman (HJB) Equation in the dynamic programming 1. Government Purchases and Endogenous Growth Consider the following endogenous growth model with government purchases (G) in continuous time. Government purchases enhance production, and the production

More information

Lecture 2 The Centralized Economy: Basic features

Lecture 2 The Centralized Economy: Basic features Lecture 2 The Centralized Economy: Basic features Leopold von Thadden University of Mainz and ECB (on leave) Advanced Macroeconomics, Winter Term 2013 1 / 41 I Motivation This Lecture introduces the basic

More information

In the Ramsey model we maximized the utility U = u[c(t)]e nt e t dt. Now

In the Ramsey model we maximized the utility U = u[c(t)]e nt e t dt. Now PERMANENT INCOME AND OPTIMAL CONSUMPTION On the previous notes we saw how permanent income hypothesis can solve the Consumption Puzzle. Now we use this hypothesis, together with assumption of rational

More information

1. Money in the utility function (start)

1. Money in the utility function (start) 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

More information

Markov Perfect Equilibria in the Ramsey Model

Markov Perfect Equilibria in the Ramsey Model Markov Perfect Equilibria in the Ramsey Model Paul Pichler and Gerhard Sorger This Version: February 2006 Abstract We study the Ramsey (1928) model under the assumption that households act strategically.

More information

Ramsey Cass Koopmans Model (1): Setup of the Model and Competitive Equilibrium Path

Ramsey Cass Koopmans Model (1): Setup of the Model and Competitive Equilibrium Path Ramsey Cass Koopmans Model (1): Setup of the Model and Competitive Equilibrium Path Ryoji Ohdoi Dept. of Industrial Engineering and Economics, Tokyo Tech This lecture note is mainly based on Ch. 8 of Acemoglu

More information

Lecture 15. Dynamic Stochastic General Equilibrium Model. Randall Romero Aguilar, PhD I Semestre 2017 Last updated: July 3, 2017

Lecture 15. Dynamic Stochastic General Equilibrium Model. Randall Romero Aguilar, PhD I Semestre 2017 Last updated: July 3, 2017 Lecture 15 Dynamic Stochastic General Equilibrium Model Randall Romero Aguilar, PhD I Semestre 2017 Last updated: July 3, 2017 Universidad de Costa Rica EC3201 - Teoría Macroeconómica 2 Table of contents

More information

Permanent Income Hypothesis Intro to the Ramsey Model

Permanent Income Hypothesis Intro to the Ramsey Model Consumption and Savings Permanent Income Hypothesis Intro to the Ramsey Model Lecture 10 Topics in Macroeconomics November 6, 2007 Lecture 10 1/18 Topics in Macroeconomics Consumption and Savings Outline

More information

ADVANCED MACROECONOMICS I

ADVANCED MACROECONOMICS I Name: Students ID: ADVANCED MACROECONOMICS I I. Short Questions (21/2 points each) Mark the following statements as True (T) or False (F) and give a brief explanation of your answer in each case. 1. 2.

More information

Financial Factors in Economic Fluctuations. Lawrence Christiano Roberto Motto Massimo Rostagno

Financial Factors in Economic Fluctuations. Lawrence Christiano Roberto Motto Massimo Rostagno Financial Factors in Economic Fluctuations Lawrence Christiano Roberto Motto Massimo Rostagno Background Much progress made on constructing and estimating models that fit quarterly data well (Smets-Wouters,

More information

Eco504 Spring 2009 C. Sims MID-TERM EXAM

Eco504 Spring 2009 C. Sims MID-TERM EXAM Eco504 Spring 2009 C. Sims MID-TERM EXAM This is a 90-minute exam. Answer all three questions, each of which is worth 30 points. You can get partial credit for partial answers. Do not spend disproportionate

More information

A Modern Equilibrium Model. Jesús Fernández-Villaverde University of Pennsylvania

A Modern Equilibrium Model. Jesús Fernández-Villaverde University of Pennsylvania A Modern Equilibrium Model Jesús Fernández-Villaverde University of Pennsylvania 1 Household Problem Preferences: max E X β t t=0 c 1 σ t 1 σ ψ l1+γ t 1+γ Budget constraint: c t + k t+1 = w t l t + r t

More information

General Examination in Macroeconomic Theory SPRING 2013

General Examination in Macroeconomic Theory SPRING 2013 HARVARD UNIVERSITY DEPARTMENT OF ECONOMICS General Examination in Macroeconomic Theory SPRING 203 You have FOUR hours. Answer all questions Part A (Prof. Laibson): 48 minutes Part B (Prof. Aghion): 48

More information