Monetary and Fiscal Policy Interactions: the Impact on the Term Structure of Interest Rates

Marzo, Massimiliano Working Paper Monetary and Fiscal Policy Interactions: the Impact on the Term Structure of Interest Rates Quaderni - Working Paper DSE, No. 409 Provided in Cooperation with: University of Bologna, Department of Economics Suggested Citation: Marzo, Massimiliano (2001) : Monetary and Fiscal Policy Interactions: the Impact on the Term Structure of Interest Rates, Quaderni - Working Paper DSE, No. 409, Alma Mater Studiorum - Università di Bologna, Dipartimento di Scienze Economiche (DSE), Bologna, https://doi.org/10.6092/unibo/amsacta/4892 This Version is available at: https://hdl.handle.net/10419/159250 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/3.0/ Monetary and Fiscal Policy Interactions: the Impact on the Term Structure of Interest Rates. Massimiliano Marzo¤ Yale University Università di Bolognay July 19, 2001 Abstract In this paper I consider a General Equilibrium framework to evaluate the role and the importance of the interactions between Monetary and Fiscal policies in the sense outlined by Leeper (1991) and the Fiscal Theory of the Price Level, in the determination of nominal and real term structure. The main results show that the term structure will depend not only upon monetary and technological factors - as in the traditional models - but also upon the …scal policy reaction function parameters. This aspect is very useful in explaining the recent experience of some countries, like Italy, whose term structure shifted down after the …scal retrenchement imposed by Fiscal Rules of Maastricht Treaty. In this paper the term structure is derived by using a very general approach, starting from implicit assumptions on the driving stochastic processes of the model, rather than imposing a speci…c structure right from the beginning. The results are consistent with the more recent results outlined by Fiscal Theory of the Price Level and by monetary models of the term structure. JEL Classi…cation Number: G12 Keywords: Fiscal Policy Rule, Price Level, In‡ation, Nominal Term Structure. ¤Many thanks go to Silvia Romagnoli for the technical part and Gabriella Chiesa for clarifying many doubts on the interpretation of the results. All responsibilities are mine. yCorrespondence Address: Dipartimento di Scienze Economiche, 2, Piazza Scaravilli, 40126 Bologna Bo ITALY. Voice: ++39-051-2098019; Fax: ++39-051-2098040. e-mail: [email protected] 1 1Introduction Recently, the growing importance of public debt has forced many governments to adopt severe …scal restrictions through cutting of expenses and/or increase of taxes. At the same time, we observed a marked reduction in nominal interest rates paid on the outstanding level of public debt. The reduction of interest rates can be observed for several maturities, implying a downward shift of the entire term structure equation. Starting from these simple observations, it becomes natural to ask if nominal term structure of interest rates could depend upon …scal policy parameters. The traditional models of the Term Structure of interest rate, like for example, Cox, Ingersoll and Ross (1985), Longsta¤ and Schwartz (1992), makes the nominal structure as dependent upon only two essential factors: monetary and technological factors. In particular, the monetary factors - generally modelled as relating with the pattern of the stochastic processes of the price level, determines the structure of nominal rates, while the technological factors will determine the structure of real rates. Usually, the technological factors are represented through a simple stochastic process for the rate of technological progress, while the nominal structure is inserted by considering a stochastic process for the in‡ation rate or the price level (as in Cox, Ingersoll and Ross (1985a,b), for example), or by specifying a stochastic process for the growth rate of nominal money supply. The limit of this type of modelling is evident because it does not take into consideration any new progress made on the literature on monetary policy rules and especially on the role of …scal policy in the stabilization of the in‡ation rate, as discussed by the recent contributions on the Fiscal Theory of the Price Level (FTPL, henceforth). The present paper tries to …ll in this gap, by presenting a framework where through the explicit modelling of the Government Budget constraint, it is possible to introduce a …scal policy reaction function making the total …scal revenue as reacting to the outstanding level of public debt, in order to ensure the solvability of government and of preserve the explosion of the price level, as proposed by Leeper (1991), Sims (1994), Woodford (1996, 2000) and more recently, by Cochrane (1998, 1999, 2000). This type of policy is often de…ned as ‘Passive’ (see Leeper (1991)) or as ‘Ricardian’ …scal policy (see, in particular Woodford (1996, 2000)), even if this terms does not have very much to do with the traditional Ricardian equivalence term, or. Apart from …scal policy, the model builds a general equilibrium framework similar to what has been proposed by Bakshi and Chen (1996), where a representative agent optimally chooses among shares, nominal and real bonds. The economy is an endowment-economy type, where the output follows an exogenous stochastic process. To keep the model simple and at the same time comparable with most of the existing literature on the term structure of interest rate, monetary policy is assumed to follow an exogenous stochastic process for the growth rate of money supply. The choice problem for the representative agent is represented in discrete time, while the main equilibrium relationships are obtained in continuous time. The general equilibrium price of a zero coupon bond is the found together with the …nal equation for the nominal and real term structure equation. All the main results are obtained in closed-form solution. From the equilibrium relationships of the model, it is possible to check that the position of the nominal term structure depends upon both monetary and …scal policy parameters. Thus, the mean 2 of the stochastic process of the expected money growth supply and the parameter which indicates the stance of …scal policy. With a ‘tight’ …scal policy, i.e. a …scal policy where taxes react to the stock of real debt, the price of bonds are high and the nominal interest rates are low. With a loose …scal policy - a …scal policy which sets a level of taxes independent upon the stock of public debt (nominal or real) - the price of a zero coupon bond will be lower than in the case with a tight …scal policy and the nominal interest rates will be much higher. The key channel for these results is the determination of the stochastic process for the price level, whose drift is de…ned as the in‡ation rate. Given the structure of the model, it is shown that the stochastic process of the price level entirely depends upon both elements a¤ecting the nominal equilibrium of the economy: the drift of the money supply process and the parameter describing the …scal policy stance. An innovative aspect of the present paper is given by the solution method adopted to obtain the term structure. Di¤erently, from the approach followed by Bakshi and Chen (1996) and Balduzzi (2000), the motion law for both nominal and real interest rates are derived following a fairly general approach where the coe¢cients of the main stochastic processes are time-variant, by directly applying the results from Feyman-Kaµ c Theorem . The closed form solution is then obtained from the general case, by imposing a set of time invariant coe¢cients, as considered by the current literature. To show the empirical performance of the model, I report the results from simulations of the term structure equations for given values of the parameters derived from Balduzzi (2000). The pictures con…rm the presumption by which the position of the term structure in the plane depends upon both monetary, …scal and technological factor. Fiscal policy appears to be important as much as monetary policy in determining the position of the term structure of interest rates. This will explain the performance of the term structure characterizing the experience of the industrialized countries which experienced episodes of …scal retrenchment. The rest of the paper is organized as follows. The next section introduces the reader to framework employed in the derivation of the results, including the description of monetary and …scal policy aspects. Section derives the equilibrium in discrete time, while its continuous time limit is reported in the following section. The analysis of an example is conducted in Section 5. Then the following sections contain a description of the steps to obtain the real spot interest rate, the price of a real and nominal zero coupon bond, the term structure of real and nominal interest rate. Finally, a section with comments to the simulation results together with a brief set of concluding remarks will close the paper. 2TheModel 2.1 The Representative Agent The economy is populated by a representative agent who optimally decides on the allocation of his (her) portfolio across a wide number of assets, including nominal and real bonds and a set of shares of stocks describing a set of property rights on private companies. The choice setup is described in discrete time intervals of length ¢t. 3 The representative investor allocates his (her) wealth by maximizing the following utility function: 1 X t=0 e¡¯tE0½uµCt;Mt Pt¶¾¢t(1) The utility under (1) depends upon consumption Ctover the interval [t; t +¢t].Moreover, Mtindicates the nominal money stock delivering utility to the representative agent over the interval [t¡¢t; t]and Pt is the price of the consumption good, here considered as the Consumer Price Index (CPI). The discount factor is ¯, so that the instantaneous discount rate is given by e¡¯t.In (1) real money balances Mt=Pt bring directly utility to the representative agent, as in Bakshi and Chen (1996). Alternative formulation would consider either the transaction cost approach, like in Sims (1994) and Leeper and Sims (1994), or the analysis of Cash in Advance constraints, as has been done by Balduzzi (2000) in a framework similar to the present one, in order to simplify algebra when we are going to derive the term structure equation. Feenstra (1990) showed the functional equivalence between the various approaches followed to include money in a general equilibrium model. I assume also that the utility function is twice continuously di¤erentiable and concave in both consumption and real balances. Formally, this means that: uc>0;u m>0;u cc <0;u mm <0;u cm <0,uccumm ¡(ucm)2>0(2) where mt=Mt Ptis the demand for real money balances. In (2) the subscript to uindicates the partial derivative. The optimal choice problem of the representative agent consists in maximizing the utility function (1) subjected to the following intertemporal budget constraint: Mt+¡Pa e;t +PtYt¢t¢et+Pta1;t +a2;t + N X i=3 Pa i;tai;t =PtCt¢t+Mt+¢t+Pa e;tet+¢t+ +Pt a1;t+¢t 1+Rt¢t+a2;t+¢t 1+it¢t+ N X i=3 Pa i;tai;t+¢t(3) From equation (3) we see that each investor can choose among a wide range of assets traded on the market, as described by Grossmann and Shiller (1983), Marshall (1992), Sun (1992), Bakshi and Chen (1996) and Balduzzi (2000). At each time t, the agent demands Mtfor cash, Ct, for consumption (in real terms), and equity holdings et(shares). The …nancial holdings are represented by vector at=(a1;t,a2;t,¢¢¢,aN;t)0where ai;t,fori=1;::: ;N indicates the number of units of …nancial assets iheld from (t¡¢t)to t.Inparticular, a1;t is the number of units of risk-free real bonds paying a real interest rate Rt,issuedattimetand maturing at t+¢t. Similarly, a2;t is the number of units of risk-free nominal bonds paying a nominal interest rate it,issuedattimetand maturing at t+¢t: Moreover, each agent can invest ai;t number of units in risky …nancial activities (stocks) whose nominal price (including dividend payments) is given by Pi;t,fori=3;::: ;N. Also, each representative agent is allowed to invest in one (single) equity share etwhich gives to the holder the property right on all the output produced through a single technology. In what follows I assume a stochastic endowments economy so that the stochastic process governing 4 the technology output is given by: ¢Yt Yt =Yt+¢t¡Yt Yt =¹y;t¢t+¾y;ty;tp¢t(4) where ¹y;t and ¾y;t are, respectively, the conditional expected value and the standard deviation of the output growth per unit of time, and fy;t :t=0;¢t;:::gis an i.i.d. standard normal random process. From (4) ¹y;t and ¾y;t can be time variants, as in Cox, Ingersoll and Ross (1995a,b) and Sun (1992). De…ne now the real price in terms of the consumption goods of asset i at time t as pa i;t =Pa i;t Pt.Thus, as it is customary in the literature on asset prices (Merton (1971), Grossmann and Shiller (1983)), I assume that the real asset prices follow a vector di¤usion process of the following type: ¢pa i;t pa i;t =¹a i;t¢t+¾a i;ta i;tp¢t(5) where ¹a i;t and ¾a i;t are, respectively, the conditional expected value and the standard deviation of real return on asset iper unit of time and ©a i;t :t=0;¢t;:::ªis an i.i.d. standard normal random process. For explanatory reasons, let us de…ne the following stochastic process for the price level (CPI), as follows: ¢Pt Pt =¹p;t¢t+¾p;tp;tp¢t(6) For the moment, the drift term ¹p;t and the standard deviation ¾p;t are taken as given, and will be derived later as functions of the core parameter of the economy. In (6) p;t is an i.i.d. standard normal random process. Given these assumptions, the representative agent maximizes his (her) utility function (1) subjected to the constraint (3). The …rst order conditions for Ct,a1;t,a2;t,Mtand ai;t are respectively given by: uc(Ct;m t)=¸tPt(7) Et£e¡¯¢t¸t+¢tPt+¢t(1 + Rt¢t)¤=¸tPt(8) Et£e¡¯¢t¸t+¢t(1 + it¢t)¤=¸t(9) Et·e¡¯¢t¸t+¢t+um(Ct+¢t;m t+¢t) Pt+¢t¸=¸t(10) Et£e¡¯¢t¸t+¢tPi;t+¢t¤=¸tPi;t (11) where ¸tis the stochastic Lagrange multiplier. The …rst order conditions (7) ¡(11) are derived from a dynamic portfolio choice and they form the basis of the equilibrium characterization of this economy. 2.2 Monetary Policy The crucial novelty of this paper is represented by the joint role of monetary and …scal policy in the determination of the price level and, by consequence of that, of the explicit risk structure of the 5 economy. In what follows, the interactions between monetary and …scal policy are designed according to the literature on the ‘Fiscal Theory of the Price Level’ (FTPL), elaborated by Leeper (1991), Sims (1994), Woodford (1996, 2000) and Cochrane (1998, 1999, 2000). Basically, FTPL states that monetary policy, without any additional constraints on …scal policy cannot achieve the goal of price stability by himself. The …scal policy constraints are essentially solvency constraints on the government budget constraint. These type of constraints are often described as ‘Ricardian’ …scal policy. In practice, this implies that in order to ensure solvency and to guarantee the existence and the stability of a price level path, …scal policy must be set such that taxes should be react proportionally to the existing stock of real (or nominal) debt1. The inclusion of …scal policy variables in the determination of the term structure of interest rates, is done by exploiting the role of the Government Budget Constraint. In this section I will introduce the role of monetary policy. In the economy here described suppose that there exists a Central Bank which keeps under control the aggregate amount of Money Supply Ms t, de…ned as follows: Ms t=Ht+Ft(12) From (12) we see that money supply is composed by two distinct components: monetary base (or high powered money), given by Ht,andFt, which is the amount of money needed by the Government to budget its balance. Apart from …scal revenue and bonds, Ftrepresents an additional source of …nancing for the Government. To avoid confusion, it is worth to stress that in this framework, money demand is not divided, but only money supply is notational divided into two components. As pointed out by Walsh (1998), monetary aggregate Ftis not often well detected by national accounts. Basically, it represents a sort of money …nancing of the government de…cits, a monetization of the de…cit up to a small amount. It is true, indeed, that in the most part of advanced economy this fraction of government de…cit is very small and in many cases is also forbidden to allow monetary …nancing of the public debt. In practice, however, there exists always a margin - though small - where Central Bank transfers money to the Government under some forms. In US, for example, the Federal Reserve rebates to the Government the interest rate proceedings derived from bond holdings in her portfolio2. It is clear that the set up here described considers a Central Bank separated from the Government sector. This represents a case where there might be possibility for the born of con‡icts between the two authorities. However, in what follows we abstract from the coordination debate, by assuming that Central Bank and Government are perfectly correlated among them, even if they are functionally separated. This also means that the Government Budget Constraint we are going to consider is not the consolidated public sector balance. We can now specify the stochastic processes for both aggregate money supply Htand Ft.Tothis 1To be precise, Leeper (1991) de…nes this type of policy as ‘passive’, while Woodford (1996, 2000) calls such policy as Ricardian, even if this does not have very much to do with the notion of Ricardian Equivalence, as clearly pointed out by Woodford (1996, 2000) and Sims (1999). 2The amount of F in U.S. is $20.1 billion in 1996, as documented by Walsh (1998, page 132). For a careful discussion on the role of Fin the construction of a monetary model (though di¤erent from the present one) see Walsh (1998). 6 goal I assume that the growth rate for Htis given by: ¢Ht Ht =Ht+¢t¡Ht Ht =¹H;t¢t(13) where ¹H;t is the mean of money growth rate. For simplicity, from (13) we have that monetary base contains only a deterministic component: this does not a¤ect dramatically the main conclusion of the model. From the practical point of view, the assumption under (13) is equivalent to imagine a steady growth rate for money supply, àlaFriedman. From the terminology proposed by FTPL, the above assumption is equivalent to say that monetary authority sets a ‘passive’ monetary policy rule3.The second part of the paper will include a generalization of the process considered in (13), by including a stochastic trend into (13) in order to make the results more comparable with the existing empirical literature. The only stochastic source to the money supply comes from the motion law speci…ed for Ft,given by: ¢Ft Ft =Ft+¢t¡Ft Ft =¹F;t¢t+¾F;tF;tp¢t(14) where ¹F;t and ¾F;t are, respectively, the mean and the standard deviation of the stochastic process leading the growth rate of the money supply term Ftand fF;t :t=0;¢t; : : : gis again an i.i.d. standard normal random process. Combining (13) and (14) with the de…nition given to money supply (12) we get the stochastic process leading money supply given by: ¢Ms t Ms t =Ms t+¢t¡Ms t Ms t =¹M;t¢t+¾M;tM;tp¢t(15) where: ¹M;t =¹H;t +¹F;t (16) ¾M;tM;t =¾F;tF;t (17) Monetary policy rule (15) is widely employed in the literature on the term structure of interest rates. In this framework it is clear that the trend component given by (16) is the sum of the two components of the monetary policy rule, and the stochastic component is exclusively given by the stochastic ‡uctuations in the transfers from Central Bank to the Government. For convenience it will be instructive to express to above relationships in real terms. Thus, let me de…ne the real (aggregate) money supply as ms t=Ms t Pt,wherePtis the price level (or CPI) whose stochastic process is de…ned according to (6). The other component of money supply Ht,andFtare de…ned by ht=Ht=Ptand ft=Ft=Pt. Thus, by taking the limit for ¢t!0and by applying Ito’s Lemma to the stochastic processes of nominal money supply (13) and (14) we get that the real money 3It will be not di¢cult to include a di¤erent monetary policy rule, like, for example, an interest-rate rule in the present setting. This is an extention that would change the drift of the monetary base speci…ciaction in (13). This is a subject of investigation. 7 supply processes are given by: dms t ms t =¹m;tdt +¾m;tm;tpdt (18) dht ht =¹h;tdt +¾h;th;tpdt (19) df t ft =¹f;tdt +¾f;tf;tpdt (20) where the drift terms are given by (after having dropped time subscript, in order to save notation): ¹m=¹M¡¹p+2¾2 p¡¾pM (21) ¹h=¹H¡¹p+2¾2 p(22) ¹f=¹F¡¹p+2¾2 p¡¾pF (23) where the stochastic terms are given by: ¾m=¾M¡¾p(24) ¾h=¡¾p(25) ¾f=¾F¡¾p(26) Equipped with the above relationships, we can now describe the role of …scal policy, in order to fully appreciate the importance of the interactions between monetary and …scal policy in the present model. 2.3 Fiscal Policy Since the work by Sargent and Wallace (1981) it has been recognized the importance of a joint determination of monetary and …scal policy. This ins not a new idea, however, as witnessed by Cochrane (1999). Old writings, especially Lerner (1948), document the fact that the price level is jointly determined by monetary and …scal policy. This is beaches the monetary base together with nominal debt is a residual claimant to government surpluses. In a crude way, FTPL says that if market participants are afraid that future surpluses will not be su¢cient to pay back the outstanding level of public debt, the government is then expected to default or to in‡ate away the debt. Therefore, the price level can be simply determined through the simple valuation equation for government debt: B P=DPV of future gov. surpluses where B is the total values of nominal government debt, P is the price level and DPV is the Discounted Present Values of future government surpluses. In a truly idealized world, it could be possible to imagine a framework in which monetary policy is useless in the determination of the price level, which is entirely determined by the role of …scal policy4. Even this is not a new idea: the famous economist Knut Wiksell a¢rmed that as long as the monetary authority is able - through an interest rate pegging regime - to modify the ‘natural rate of interest’ then, the control of a monetary aggregate is a redundant tool to ensure the control of the price level. In any case, in the following paper I do not take this extremal view 4A very good description of the presence in old economic writings of reasoning similar to what has been formalized by FTPL theorists is on Cochrane (1999). 8 this framework the riskiness on bonds given by nit¡¼t+vartndPt Pto¡Rtobecomes a function of the overall riskiness structure of the economy. Therefore, nominal risk free assets - like nominal bonds - become ’risky’ assets, through the riskiness role of the price level. The stochastic process of the price level is a function of the overall riskiness structure of the economy, including the role of monetary and …scal policy. The de…nition of the real interest rate is in the following proposition: Proposition 4 The equilibrium real interest rate is given by: Rt=¯¡uccYt uc 1 dtEt½dYt Yt¾¡1 2 uccY2 t uc vart½dYt Yt¾¡ucmmt uc 1 dtEt½dmt mt¾+ ¡1 2 ucmmm2 t uc vart½dmt mt¾¡Ytmtuccm uc covtµdYt Yt ;dmt mt¶(56) Proof. see appendix A. See Appendix. ¥ Proposition 5 The equilibrium nominal interest rate is given by: it=um(Yt;m t) uc(Yt;m t)(57) Proof. see appendix A. See Appendix A. Few additional comments are in order. If we look at equation (56) we observe that the stochastic process of money supply a¤ects the real interest rate, if the utility function is non-separable in money and output. However, if the utility function is logarithmic in both money and output - like f.e. in Stulz (1986) - (this is equivalent to say that ucm =uccm =ucmm =0 ) there is no way by which monetary policy - and in the present particular case also …scal policy - can a¤ect real interest rate. The right hand side of equation (57) is the marginal rate of substitution between consumption and real cash holdings and it can be viewed as the marginal bene…t of holding one additional unit of cash balance. In equilibrium, this cost must be equal to the bene…t. Let us examine the following results on the term structure. Proposition 6 (a) The nominal term structure equation is de…ned as: N(t; ¿) Pt =e¡¯¿Et·uc(Yt+¿;m t+¿) uc(Yt;m t) 1 Pt+¿¸(58) where N(t; ¿)is the time tnominal price of a discount bond paying a dollar in ¿periods. (b) The real term structure is: b(t; ¿)=e¡¯¿ Et·uc(Yt+¿;m t+¿) uc(Yt;m t)¸(59) where b(t; ¿)is the time tnominal price of a discount bond paying a unit of consumption good in ¿ periods. 15 Proof. The result follows directly from the Euler equation (41) Equations (58) and (59) will be particularly useful in the following examples to establish the behavior of the term structure. 5 A speci…c example In what follows we will provide a speci…c model from which we will derive the stochastic processes for the price level (CPI) and the other various elements of the model. The stochastic processes speci…ed here describe a more realistic economic setting. The setup is closed to what has already been proposed in the literature by Bakshi and Chen (1996) and Balduzzi (2000). The novelty of the following approach lies in the approach taken in the solution method. Di¤erently from the other approaches undertaken in the literature I explicitly consider a solution for the term structure of interest rates which is directly derived from the assumptions here considered by following the method described in Cox, Ingersoll and Ross (1985), instead of imposing a structure to the movement law of the nominal interest rate and later verifying the model after this imposition. Basically, the methodology followed here builds on an arbitrage-free model. In order to simplify algebra, let us assume the following strongly separable utility function given by: uµCt;Mt Pt¶=Álog Ct+(1¡Á)logmt(60) where mt=Mt=Pt. Assume that the stochastic process leading output is given by: dYt=(¹Y+´Yxt)dt +¾YpxtdWx;t (61) where the process for the technology factor xtis given by: dxt=a(b¡xt)dt +¾xpxtdWx;t (62) where (Wx;t)tis a unidimensional Q-Brownian motion, ¹Y,´Y,¾Y,a,b,and¾xare …xed real numbers. Let us assume the following two stochastic processes for the two monetary aggregates Htand Ft: dln Ht=¹¤ Hdt +dln (qt)(63) dln Ft=¹Fdt +dln (qt)(64) where qtis the detrended money supply process. Basically from (63)-(64) we see that the stochastic process for the two-types of money supply has two components: a drift term and a stochastic component qtgiven by the detrended component. In particular, ¹¤ His assumed to be constant and positive, while ¹Fis the drift term depending upon the …scal policy parameters as discussed in a previous section and is still de…ned by (??)Following Bakshi and Chen [?],qtis assumed to follow the stochastic process: dqt qt =kq¡¹q¡qt¢dt +¾qpqtdWi;t;i=H; F (65) 16 where (Wi;t)tis a unidimensional Q-Brownianmotionindependentupon(Wx;t)t. Therefore, by using the de…nition of money supply given in (12) and the above assumptions, we get that the stochastic process leading money supply is given by: dMt Mt =¹M;tdt +¾qpqtdWM;t (66) where (WM;t)tis a unidimensional Q-Brownian motion independent upon (Wx;t)tand (Wi;t)tand where: ¹M=¹¤ M+2kq¡¹q¡qt¢(67) with: ¹¤ M=¹¤ H+¹F(68) and dM;t =dH;t +dF;t:Followingthesamestepswedidinthesectionon…scalpolicy 10,wehave that ¹Fis given by: ¹F=¾pH +¡¹i¡Á1+þ2 q¢ 1+Ã(69) where ¾pH the correlation between the price level and the monetary aggregate H. Given our set of assumptions, above considered, we can now derive the main results related with the commodity price level and the in‡ation process, as reported and discussed in the following Theorem: Proposition 7 Given the utility function of the representative agent as described by equation (60),then the equilibrium price level is given by: Pc t=Á 1¡Á q2 t(¯+¹¤ M)¡¯+¹¤ M+2kq¹q¢ (¯+¹¤ M)+¡kq+3¾2 q¢qt2kq¹q Mt Yt (70) The stochastic process of the CPI is given by: dPc t Pc t =¼tdt +¾qpqt·1+(¢qª¡¢ªq) ¢ª qt¸dWM;t ¡¾ypxtdWx;t (71) where the in‡ation rate is given by: ¼t=¹¤ M¡¹y+¡¾2 y¡´y¢xt+(¢qª¡¢ªq) ¢ª qtÃkq¡¹q¡qt¢+¾2 qqt 2!+ +[2 (¢qqª¡¢ªqq)¡¢qª+¢ª q]¾2 qq3=2 t 2ª2¢(72) with: ¢(q)= Á 1¡Á£q2 t(¯+¹¤ M)¡¯+¹¤ M+2kq¹q¢¤ (73) ª(q)=(¯+¹¤ M)+¡kq+3¾2 q¢qt2kq¹q(74) and ¢q=@¢(q) @q ;ª q=@ª(q) @q ;¢qq =@¢(q) @q . 10Recall that in order to derive the expression of ¹F(nominal). We need to consider the government budget constraint written in real terms, and exploting the de…nition of Fand Hweitteninrealterms. 17 Proof. See Appendix A. The price level (equation (70)) and the in‡ation rate given in equation (72) are function of both …scal and monetary parameters in a very complicated way. In the section dedicated to the simulation analysis I will provide an intuitive explanation of the pattern behavior of such variables. In the following pages I will describe all the necessary steps to solve out the present model. 6 The real spot interest rate In this section I am going to derive the dynamic of real spot interest rate driven by the exogenous dynamic of technology. Since the technology is the solution of the EDS given by (62),xtis Markov and satisfy the necessary technical conditions to apply the Representation Theorem of Feyman-Kaµ c11. We can now follow the PDE approach to derive the dynamic of real spot rate as a function of time and technology. First, let us consider how to derive the real spot rate dynamics. Proposition 8 Given the dynamic stochastic process for the technology as in (62) ;the real spot rate Rtis a function Á(t; xt)which is the unique solution of the Kolmogorov PDE: (1 2¾2 xxt@2Á(t;xt) @x2 t+a(b¡xt)@Á(t;xt) @xt+@Á(t;xt) @t ¡xtÁ(t; xt)=0 Á(T;xt)= 2ab °+a where the …nal condition is the long time spot rate determined as in Cox, Ingersoll and Ross (1985). So we have: Rt=A(µ)eC(µ)xt(75) where µ=T¡tand C(µ)= ¾2 x(2 + a)¡1¡e°µ°¢ 2a[¾2 x¡a+e°µ (a¡¾2 x)¡(1 + e°µ)°](76) A(µ)= 1 °+an2Âabeº(µ)°³£a+°¡¾2 x+e°µ ¡°¡a+¾2 x¢¤¡³o(77) and where Â=a2¡°2¡2¾2 x(a+b+ab)+¾4(1 + b) (a¡¾2 x)2¡°2 º(µ)=bµ¾2 x(2 + a¡°) 2(a+°¡¾2 x) ³=b¾2 x¡¾2 x¡2a¡2¢ (a¡¾2)2¡°2 °=pa2+2¾2 x 11In particular the drift and volatility term in (62) must be Lipschitz and bounded on <. 18 Proof. See Appendix A. Now we can derive the dynamic of the real spot interest rate in the following Lemma. Proposition 9 The dynamic of real spot interest rate is: dRt=a¤(µ)[b¤(µ)¡Rt]dt +C(µ)Rt¾xpxtdWx;t (78) where (Wx;t)tis a unidimensional Q¡Brownian motion, and where: a¤(µ)=¡·a(b¡xt)C(µ)+C0(µ)xt+1 2¾2 xxtC2(µ)¸ b¤(µ)=A0(µ)eC(µ)xt a¤(µ) Proof. See Appendix A. Few comments are in order. One might naturally ask whether the real interest rate here derived is an equilibrium real rate, given the solution method here adopted not orthodox to the traditional equilibrium term structure theory. In truth, the real rate derived in the above theorem is e¤ectively an equilibrium real rate, even if the solution method adopted is very much close to the arbitrage-free models. In fact, according to Cox, Ingersoll and Ross (1985), the choice made for the functional form of Á(t; xt)is the only possible choice which makes the solution here derived exactly equal to the equilibrium solution. Now we have all the elements to derive the term structure of interest rates, in real terms. 7 Zero coupon bonds and term structure of real interest rates In this section we derive the price of zero coupon bonds as a function of time, technology and real spot interest rate whose dynamics was determined previously. We follow again a PDE approach because xtand rtboth follow Markov processes and satisfy the necessary technical conditions to apply the Representation Theorem of Feyman-Kaµ c12. The solution is not explicit but it is necessary to follow a numerical method. The approach taken here can be de…ned as a ‘two step’ procedure, because I …rst derive the zero coupon bonds price and only afterward we …nd the term structure of real interest rates. This approach is slightly di¤erent from what has been proposed by Bakshi and Chen (1996). Here the stochastic process of both real and nominal interest rate is directly derived from the assumptions made on the stochastic processes for the technology and policy processes. Instead Bakshi and Chen (1996) impose the stochastic process of the nominal interest rate in less orthodox way. In some sense, the structure they use considers some ad hoc assumptions, that I removed here. Proposition 10 Given the dynamic for technology and real spot rate, as in (62) and (107), respectively, then the zero coupon bond B(t; T)is a function #(t; xt;R t)that is the unique solution of the following 12In particular, the drift and volatility term in (107) must be Lipschitz and bounded on <. 19 Kolmogorov PDE: 8 > > > > < > > > > : 1 2¾2 xxt@2#(t;xt;Rt) @x2 t+a[b¡xt]@#(t;xt;Rt) @xt+1 2C2(µ)C¤2(µ)R2 t¾2 xxt@2#(t;xt;Rt) @R2 t+ +a¤(µ)[b¤(µ)¡Rt]@#(t;xt;Rt) @Rt+RtC(µ)¾2 xxt@2#(t;xt;Rt) @rt@xt+@#(t;xt;Rt) @t + ¡Rt#(t; xt;R t)=0 #(T;xt;Rt)=1 (79) Proof. See Appendix A. It is now straightforward to derive the dynamic for the real term structure. Lemma 11 The term structure of real interest rates is: r(t; T)=¡1 µ[ln A¤(µ)¡RtC¤(µ)] (80) where A¤(µ)and C¤(µ)are the solutions of system (110). Proof. The term structure of real interest rates is immediately derived by the relation between the price of zero coupon bond and the continuous real interest rate: B(t; T)=e¡µr(t;T) With this machinery at hand we can now derive the nominal term structure of interest rates. 8 The Nominal term structure of interest rates In what follows we will derive the equilibrium nominal spot interest rate and the analytical expression of the nominal zero coupon bond and the nominal term structure of interest rates (in implicit form). To start with, consider together the expression for the real interest rate and the expression for the price level given, respectively, by (75),and(95). Lemma 12 The nominal spot interest rate is given by: it=A(µ)eC(µ)xt"µÁ 1¡Á¶q2 t(¯+¹¤ M)¡¯+¹¤ M+2kq¹q¢ (¯+¹¤ M)+¡kq+3¾2 q¢qt2kq¹q Mt Yt#(81) where A(µ),C(µ)are given by (76) and (77), respectively. Proof. To get (81) it is enough to multiply (75) by (70). Lemma 13 The nominal zero coupon bond is given by N(t; T)=B(t; T)"µÁ 1¡Á¶q2 t(¯+¹¤ M)¡¯+¹¤ M+2kq¹q¢ (¯+¹¤ M)+¡kq+3¾2 q¢qt2kq¹q Mt Yt#(82) where B(t; T)is the solution of (108). 20 Proof. To get (82) it is enough to multiply the solution of (79) by (70). Lemma 14 The nominal term structure of interest rate is given by I(t; T)=r(t; T)"µÁ 1¡Á¶q2 t(¯+¹¤ M)¡¯+¹¤ M+2kq¹q¢ (¯+¹¤ M)+¡kq+3¾2 q¢qt2kq¹q Mt Yt#(83) where R(t; T)is given by (80) Proof. To get (83) it is enough to multiply (80) by (70). By looking at the expressions for (81),(82) and (83) we observe that the nominal variables are bigger than the corresponding real variable if (70) is bigger than one. According to that, we can easily …nd a condition on Á1(contained in ¹¤ M) which veri…es this assertion. Therefore, the condition on ¹¤ M is ¹¤ M<¹ ¤a Mand ¹¤ M>¹ ¤b M(84) where ¹¤a M,¹¤b Mare given by: ¹¤a;b M=¡¡§p¡2+4MtÁq2 t¤ 2MtÁq2 t where: ¡=2MtÁq2 t¡¯+kq¹q¢¡Yt(1 ¡Á) ¤=¡Yt(1 ¡Á)¯¡2Ytkq¹qqt¡kq+3¾2 q¢(1 ¡Á)+MtÁq2 t¯¡¯+2kq¹q¢ These equations establish a bound on the drift of the stochastic process leading the money growth rate. By virtue of the de…nition of ¹¤ Mgiven in (68) we get that the bound for Á1is given by: Áb 1<Á 1<Á a 1 where: Áa;b 1=¡¹i+¾2 q¢¡(1 + Ã) ó¹¤a;b M¡¹¤ H´(85) with ¹¤ Hde…ned as in (68). 9SimulationResults The results outlined in these pages are not easily interpretable without the help of a set of simulations. In what follows, I will construct a set of simulation results that will highlights the main properties of the present model. To do so, we need to specify some parameters. For the choice of the core parameters I follow the work by Balduzzi (2000) who presented an estimated version of the stochastic processes, through the methods of moments estimation method. The parameter chosen are collected in the following table: 21 ¯Ák q¹qab¹ Y´Y¾Y¾x¾q 0.998 0.01 0.56 0.11 0.86 0.06 1.3 0.7 0.06 0.03 0.02 Table 1 The parameter ¯is set equal to 0.998, as in the traditional Real Business Cycle literature. Also the preference parameter Áis chosen in accord to Balduzzi (2000), who derived it according to an estimate of the key equations of the model through methods of moments. Moreover, all the other parameters are taken from the very few empirical studies existing in the literature on this subject. This is done especially for comparisons maters. However, it should be remembered that the main goal of the present paper is to show the joint role of monetary and especially …scal policy parameters in the determination of the term structure of interest rates. The search for the model which best could …t the data is still ongoing. A …nal remarks before the discussion of results regarding the treatment of the stochastic processes for money supply, output growth, technology and the stochastic process describing the stochastic trend in money supply. To run simulations I integrated all these di¤usion processes13 and I initialized them by setting an initial condition equal to one, i.e. x0=Y0=M0=q0=1 . The …rst step is to consider the simulation analysis for both the price level and the in‡ation rate for di¤erent values of both the …scal policy parameter Á1and the monetary parameter ¹H. In …gures 1-2 I reported the pictures to detect the e¤ect of changes in these two parameters. It is worth to recall that the simulations reported in Figures 1 and 2 are conducted within an a-temporal context, without any reference to time elapsing. This is done in order to show the impact e¤ect on both the price level and the in‡ation rate induced by changes in parameters Á1and ¹H:In Figure 1 we observe that a progressive increase in Á1creates a proportional reduction of the level of the price index and, at the same time of the in‡ation rate. This means that a tighter …scal policy has a de‡ationary impact. These pictures con…rms the intuition provided in a simpler model in the previous sections. These e¤ects are interpretable along the lines of the Fiscal Theory of the Price Level and , in particular, under the new Wiksellian monetary policy framework highlighted by Woodford (2000). In Figure 2, I show the e¤ects on the same variables after changing the monetary policy parameter. In this case, we observe an almost reversed e¤ect: In the simulation here considered I have set the …scal policy parameter equal to 0.55. This is a value coherent with the bounds here assumed and with the literature on FTPL. It should be noted that the monetary policy parameter of interest here is given by the drift term of the stochastic process of the money supply growth rate which goes into the public’s hands, i.e. the monetary base held by the public. The in‡ationary e¤ect after and increase of money supply is normal and expected, according to more than one hundred monetary history. Probably the more interesting set of simulations is represented by Figures 3-5, where I reported the e¤ects of changes of the …scal policy parameter for the pattern behavior of the nominal and real spot rate, the zero coupon bond (nominal and real) and the term structure of interest rates (nominal and real), respectively. All the pictures here represented show dynamical e¤ects of the various policies. In fact, the simulations are run over a set of 10 periods. Each simulation step is 0.01, in the interval [0, 10]. All the curve showed in this set of picture refers to particular values for Á1. 13The analytical expressions for the itegrated processes are reported in Appendix B. 22 In Figure 3, I have considered di¤erent plots, according to di¤erent values for Á1. The picture shows that if …scal policy becomes tighter (Á1raises) the spot rate curve shifts down. However, even if for very small values of Á1(implied by a very high taxation policy) the curve appears to loose its sensitiveness to this parameter. The starred curve represents the real term structure. We observe that the starred curve lies below all the nominal curves. This is because the nominal curve incorporate the in‡ationary e¤ect (expected and/or actual). Since the choice of the parameter is done punctual and kept constant over time horizon chosen for the simulation period, it might happen that the curve of the real term structure overcome the nominal term structure, showing an increase of the in‡ation rate. This e¤ect would disappear if …scal authority will keep on raising the coe¢cient on the …scal policy reaction function. Note that the type of pathology is evident especially for very long periods far away the simulation staring point. Let us look now at Figure 4, where, in order to simulate the model and have a closed form solution useful for the simulation analysis, I have made a simplifying assumption by supposing that the coe¢- cients of equations (110) are time independent. We are aware of the limitation of this choice but this is motivated by the need to present some results in an intuitive way which can be readable according to an handy economic sense. In any case, in our view this represents a remarkable step ahead in the current literature, where all the results about the term structure are almost imposed and not explicitly derived as considered here. The parameters chosen for the simulation are exactly the same of those considered for simulating the nominal and real spot rate curve. The picture shows that the position in the plane of the term structure strictly depends upon the value assumed for the tax rate Á1. In particular, it is evident that if the …scal pressure decreases (i.e. if Á1decreases), the curve of the nominal term structure zero coupon bond shifts down, even if for very small values of Á1(implied by a very high taxation policy) the curve appears to loose its sensitiveness to this parameter. The behavior of the curve is coherent with both the results from Fiscal Theory and the empirical evidence. In fact, if the tax rate get reduced, this means, that the price asked for newly issued debt must be reduced in order to convince new subscribers to buy additional debt. In fact, newly issued debt, without a backing of more taxes implies an in‡ationary risk for the future (causing a reduction of the nominal value of debt right by now). This will even more evident when we will discuss the Term Structure for both Nominal and Real Zero Coupon Bonds in the next picture. With respect to the punctual choice for Á1can be repeated here the considerations of the same sort done before for the nominal and real spot rate curve. Finally, consider the simulation for the term structure of nominal and real rates reported in Figure 5. The picture shows that the position in the plane of the term structure strictly depends upon the value assumed for Á1. In particular, it is evident that if the …scal parameter Á1tax rate decreases, the curve of the nominal term structure shifts up, even if for very high values of Á1the curve appears to loose its sensitiveness to this parameter. The behavior of the curve is coherent with both the results from Fiscal Theory and the empirical evidence. In fact, if the tax rate raises, this means that it will be possible to reduce the number of issue of new public debt in order to …nance the current position of the government. This will call for a lower interest rate, and for a bigger government credibility. This shows not only monetary policy is important for the term structure, but also …scal policy is crucial in order to determine the position of the curve. Empirically, this has been veri…ed as true in the experience 23 of Italy after 1996-1997 episode of …scal retrenchment which shifted down the position of the nominal term structure after a crucial …scal consolidation, necessary to get into the EU. Here too the starred curve represents the real term structure and we can replicate here the same kind of considerations done for the punctual choice of Á1. We observe that even in this case the starred curve lies below all the nominal curves, in accord with economic theory. From the economic point of view, the case where the nominal curve stays above the real, can be thought as the result of a very tight …scal policy which produces a de‡ationary equilibrium, such that the nominal term structure is below (or at) the real term curve. Such de‡ationary equilibria are not impossible within the Fiscal Theory framework, as discussed by Sims (1994): this can be interpreted also by thinking about a contraction of expenses (and of aggregate demand) determined by a very high level of …scal pressure. 10 Concluding Remarks In this paper I have considered a simple intertemporal model for the determination of the nominal and real term structure where …scal and monetary policies play a joint key role in the determination of the position of the term structure and the zero coupon bond curve in the plane. This shows to be an important development in a literature where only monetary and technological factors were seen as unique determinants of the term structure of interest rates. The novelty of the present paper is represented by a di¤erent way employed to determine the term structure of interest rates in the monetary models, relying on an arbitrage-free model, rather than relying on equilibrium arguments. The results show that the introduction of both monetary and …scal policy rules adds an important element in the determination of the position in the plane of the term structure of interest rates. In particular, it is shown that as showed by important contributions in the Fiscal Theory of the Price Level (FTPL), in‡ation is not necessarily a monetary phenomena. An expected …scal retrenchment shifts downwards the entire term structure of nominal interest rates and the nominal spot rate curve. The e¤ect is entirely due to the reaction of expected in‡ation: if government adopts a tighter …scal policy, bonds become more ’secure’ in the sense, that an expected higher level of taxes will make the government able to pay back taxes. Thus, bond prices will increase, while their return will decrease. The nominal terms structure equation depends upon the technological parameters. However, the real curve does not seem to be a¤ected by the same set of parameters a¤ecting the nominal curve. From this point of view, this model presents similar conclusion with respect to Bakshi and Chen (1996). The approach taken here, is however, general enough which will allow to design a more complex set of stochastic processes able to create a set of interrelations between nominal and real term structure of interest rates which has not been explored in the literature. The results here presented generalize two distinct ‡ow of literature: the FTPL and the term structure models in a promising way. Further generalization could go, for example, towards a more careful modelling of the monetary policy function. The pattern of the nominal term structure here considered captures quite well the dynamic of the the term structure pattern of many countries after having experienced a …scal retrenchment, as discussed by Corsetti and Pesenti (1999), in the case of European Countries. 24 References [1] Bakshi, Gurdip S. and Chen Zhiwu,(1996), “In‡ation, Asset Prices, and the Term Structure of Interest Rates in Monetary Economies”, The Review of Financial Studies, Spring, Vol.9, No.1, pp.241-275. [2] Balduzzi, Pierluigi, (2000), “Money, Market Equilibrium, and Stock-Return Dynamics”, mimeo, Carrol School of Management, Boston College. [3] Breeden, Douglas, (1979), “An Intertemporal Asset Pricing Model with Stochastic Consumption and Investment Opportunities”, Journal of Financial Economics,7,265-296. [4] Breeden, Douglas, (1986), “Consumption, Production, In‡ation and Interest Rates: A Synthesis”, Journal of Financial Economics, 16, 3-39. [5] Canzoneri, Matthew, B., Robert E. Cumby, and Behzad T. Diba, (1998), “Fiscal Discipline and Exchange-Rate Regimes”, CEPR Discussion Paper No. 1899, May. [6] Cochrane, John, (1998), “A Frictionless view of U.S. In‡ation”, NBER Macroeconomic Annual, edited by J. Rotemberg and M. Woodford, MIT Press, Cambridge, MA. [7] Cochrane, John, (1999), “Long Term Debt and Optimal Policy in the Fiscal Theory of the Price Level”, unpublished, Chicago Business School, Univ. of Chicago. [8] Cochrane, John, (2000), “Money as Stock: Price Level Determination with no Money Demand”, NBER Working Paper No. 7498, January. [9] Cochrane, J., (1999), “Money as Stock: Price Level Determination with no Money Demand”, mimeo, Chicago Business School, August. [10] Cooley, Thomas, F., and Edward Prescott, (1995), “Economic Growt and Business Cycles”, in Frontiers of Business Cycle Research, edited by Thomas, F. Cooley, Princeton University Press. [11] Corsetti, Giancarlo, and Paolo Pesenti, (1999), “Stability, Asymmetry and Discontinuity: The Launch of European Monetary Union”, Brookings Papers on Economic Activity,Vol.2,295-358. [12] Costantinides, George, (1992), “A Theory of the Nominal Term Structure of Interest Rates”, Review of Financial Studies, Vol 5, 531-552. [13] Cox, John C., Jonathan E. Ingersoll, and Stephen A.Ross, (1985a), “An Intertemporal General Equilibrium Model of Asset Prices”, Econometrica, Vol. 53, No.2, pp. 363-384. [14] Cox, John C., Jonathan E. Ingersoll, and Stephen A.Ross, (1985b), “A Theory of the Term Structure of Interest Rates”, Econometrica, Vol. 53, No.2, pp. 385-407. [15] Danthine, Jean Pierre, and John Donaldson, (1989), “In‡ation and Asset Prices in an Exchange Economy”, Econometrica, Vol. 54, 585-605. [16] Feenstra, Robert, (1986), “Functional Equivalence Between Liquidity Costs and the Utility of Money”, Journal of Monetary Economics, Vol.17, 271-291. 31 [17] Foresi, Silverio, (1990), “Valuation of Nominal Securities”, mimeo, New York University. [18] Geske, Robert, and Richard Roll, (1983), “The Monetary and Fiscal Linkages between Stock Returns and In‡ation”, Journal of Finance, Vol. 38, 1-33. [19] Grossmann, S., and R. Shiller, (1982), “Consumption Correlatedness and Risk Measurement in Economics with Non-Traded Assets and Heterogeneous Information”, Journal of Financial Economics, 10, 195-210. [20] Leeper, Eric, M., (1991), “Equilibria under Active and Passive Monetary and Fiscal Policy”, Journal of Monetary Economics, 27, pp.129-147. [21] Longsta¤, Francis, and Eduardo Schwartz, (1992), “Interest Rate Volatility and the Term Structure: A Two-Factor General Equilibrium Model”, Journal of Finance, 47, 1259-1282. [22] Marshall, David, (1992), “In‡ation and Asset Returns in a Monetary Economy”, Journal of Finance, Vol. 47, 1315-1342. [23] Marzo, Massimiliano, (2001), “Monetary and Fiscal Policy Interactions: the Impact on the Term Structure of Interest Rates”, unpublished, University of Bologna. [24] Merton, Robert, (1971), “Optimal Consumption and Portfolio Rules in a Continous Time Model”, Journal of Economic Theory,3,373-413. [25] Sims, Christopher, A., (1994), “A Simple Model to Study the Determination of the Price Level and the Interaction of Monetary and Fiscal Policy”, Economic Theory, 4, pp. 381-99. [26] Sims, Christopher, A., (1999), “The Precarious Fiscal Foundations of EMU”, unpublished, Princeton University. [27] Stulz, R., (1986), “Asset Pricing and Expected In‡ation”, Journal of Finance, 41, pp. 209-223. [28] Sun, Tin, (1992), “Real and Nominal Interest Rates: A Discrete Time Model and its Continous Time Limit”, Review of Financial Studies,Vol.5,581-611. [29] Walsh, Carl, (1998), Monetary Theory and Policy, MIT Press, Cambridge, Mass. [30] Woodford, Michael, (1996), “Control of the Public Debt: A Requirement for Price Stability ?”, NBER Working Paper No. 5684, July. [31] Woodford, Michael, (2000), “Fiscal Requirements for Price Stability”, O¢cial Lecture of the 2000 Money, Credit and Banking Lecture, presented at the Ohio State University, May 1, 2000. 32 0 0.5 1 1.5 133.6 133.8 134 134.2 134.4 134.6 φ1 Price Effects of changes in φ1 on P and π 0 0.5 1 1.5 6.85 6.86 6.87 6.88 6.89 6.9 φ1 π Figure 1: 33 1 1.5 2 2.5 3 3.5 4 4.5 5 30 40 50 60 70 80 90 µH P Effects of changes in µH on P and π 1 1.5 2 2.5 3 3.5 4 4.5 5 1 2 3 4 5 6 µH π Figure 2: 34 0 1 2 3 4 5 6 7 8 9 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 t i t , R t Nominal and Real Spot Rate Curve Figure 3: 35 0 1 2 3 4 5 6 7 8 9 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 θ N(t,T), B(t,T) Term Structure of Nominal and Real Zero Coupon Bond Figure 4: 36 0 1 2 3 4 5 6 7 8 9 -1 0 1 2 3 4 5 6 7 8 9 θ I(t,T), r(t,T) Term Structure of Nominal and Real Interest Rates Figure 5: 37