سیاست های پولی فعال، سیاست های مالی غیر فعال و ارزش بدهی های عمومی: برخی از حساب های اضافی پول گرایان

We compare the properties of a `constant money growth rule' and a `strict inflation targeting rule' in an intertemporal equilibrium model with flexible prices in which monetary policy is `active', while fiscal policy is `passive'. The paper shows that dynamic properties of the economy may differ significantly between the two monetary policy rules if public debt is issued in nominal terms. Under a constant money growth rule which allows for temporary deviations of inflation from target in response to shocks there is scope for revaluations of public debt, acting as automatic stabilizers of government debt dynamics. By contrast, a policy of strict inflation targeting implements the target inflation rate also outside the steady state and precludes such stabilizing revaluations. Owing to this feature, additional fiscal restraint may be needed which is not required under a constant money growth rule.

As pointed out in the seminal contribution by Sargent and Wallace (1981), monetary policy by itself will not always be in a position to control the evolution of the price level, unless being appropriately supported by the fiscal agent. This insight reflects that in any macroeconomic model the government's budget constraint entails contributions of both monetary and fiscal policy. In terms of the widely used terminology introduced by Leeper (1991), the Sargent–Wallace result says that for the monetary agent to be able to control `actively' inflation the fiscal agent needs to behave `passively' in the sense that he accepts the residual role within the government's budget constraint, taking as given the behaviour of the active agent.1 But how to enforce a credible mix of active monetary and passive fiscal policy? This question is of particular importance in the context of the European Monetary Union which is characterized by potential coordination failures and additional incentive problems arising from `one money, but many fiscal policies' (Uhlig, 2003). Within this context, many economists have argued that it would be desirable to subject the fiscal agent to some kind of a rule which imposes certain limits on the government's borrowing behaviour (Chari and Kehoe, 1998; Sims, 1999). Evidently, the requirements of the Stability and Growth Pact prevailing in the European Monetary Union reflect such concerns. Woodford (2001) has recently reemphasized, however, that the interaction between monetary and fiscal policy can never be one-way only. Specifically, regarding the effects of monetary policy on fiscal policy, Woodford stresses the effects of monetary policy on the real value of government debt (and the real debt service associated with it) through its effects on the price level, given that public debt is largely issued in nominal terms. Moreover, these fiscal effects of monetary policy can be potentially large, even if the traditionally considered channel, the seigniorage contribution to the government's budget, is negligible. This paper takes the desirability of a combination of active monetary and passive fiscal policy rules for granted and investigates some of the implications of the revaluation channel of government debt for the design of such rules. The key idea is that because of this channel different monetary policy rules are likely to be associated with different government debt dynamics, restraining thereby the fiscal agent in different ways. To illustrate the potential strength of these differences, we present a small general equilibrium model with strong supply-side features in which the price level is fully flexible and determined according to simple `monetarist' principles. We compare two specifications of monetary policy, both of them being consistent with identical long-run equilibria. First, we consider a policy of a `constant money growth rule' which allows for temporary deviations of inflation from target and leaves room for revaluations of outstanding public debt as a response to supply shocks. Investigating the dynamic properties of equilibria, we show how this channel acts as an automatic stabilizer of government debt dynamics, relaxing thereby the constraint of the passive fiscal agent. Second, we look at a policy of `strict inflation targeting' which keeps inflation always on target and preserves the real value of outstanding debt in response to shocks. As a result, this policy lacks the stabilization properties regarding government debt dynamics and we derive how this feature leads to a tighter constraint of the fiscal agent, compared with a constant money growth rule.2 While the main insight of this paper is largely model-independent, we organize our analysis, inspired by Sargent and Wallace, around an overlapping generations economy of the Diamond-type with two interest-bearing assets (physical capital, government bonds) and return-dominated outside money. Specifically, following the two-stage modelling strategy of Sargent and Wallace, we start out with a simple, fully tractable benchmark economy with backward-looking dynamics in which the preferences of agents are specified in a highly monetarist way, yielding a demand for real balances which is strictly proportional to contemporaneous output and displays a constant velocity.3 We assume that monetary policy has a certain inflation target, while fiscal policy aims at a certain non-negative target value of the deficit ratio (corresponding in long-run equilibrium one-to-one to a certain debt ratio). Given such targets, the model gives rise to a unique steady state with positive levels of output and real balances as well as non-negative government debt, as long as the deficit ratio remains below a certain feasibility bound. Depending on the specific interaction between monetary and fiscal policy, however, steady states of this type are not necessarily stable. Defining our particular specification of a passive stance of fiscal policy, we restrict the fiscal agent to deliver stable dynamics, taking as given the specification of the monetary policy rule.4 In order to establish the different fiscal consequences associated with strict inflation targeting and a constant money growth rule we consider two scenarios. First, as a deliberately strong example, we require the fiscal agent to maintain a constant deficit ratio not only in steady state, but in all periods. Under this assumption, under a constant money growth rule all feasible steady states are always stable. By contrast, under a policy of strict inflation targeting feasible steady states become unstable beyond a certain threshold value of the deficit ratio, implying that the fiscal agent faces ex ante a narrower choice set than under a constant money growth rule. This different stability behaviour results from the fact that under strict inflation targeting, because of the absence of stabilizing revaluations of government debt, the economy is more vulnerable to adverse debt dynamics in response to shocks. Specifically, the severity of such dynamics depends on the initial steady-state level of the deficit ratio which, through its correspondence to the economy's debt ratio, is directly linked to the pre-shock level of the real interest rate. Hence, the lack of stabilizing revaluations of government debt is particularly harmful under a high deficit ratio, generating the possibility of unstable dynamics. Moreover, we also show that under strict inflation targeting the richer interaction between debt dynamics and crowding out effects implies that off-steady-state dynamics, even when being stable, can be associated with endogenous fluctuations––a feature which cannot occur under a constant money growth rule. Second, we discuss how the potential instability of steady states under strict inflation targeting can be removed by fiscal policy rules which maintain the same long-run target value of the deficit ratio, but allow for more flexibility outside the steady state. While upon appropriate changes in the fiscal rule all feasible steady-state deficit ratios can be stabilized, the main result remains nevertheless unaffected, i.e. the lack of stabilizing debt revaluations under strict inflation targeting may well require additional fiscal restraint which is not needed under a constant money growth rule.5 Following Sargent and Wallace, we then change preferences of agents in a way that the money demand specification becomes forward-looking and depends as well on future inflation. Everything else being equal, this modification leaves all the dynamic (in)stability properties of the benchmark economy under strict inflation targeting qualitatively unaffected, since the forward-looking component remains fully predictable. By contrast, under a constant money growth rule the forward-looking component is now genuinely expectation driven. As a result, real balances turn into a forward-looking jump variable and the overall dynamics now contain both backward-looking and forward-looking elements. Despite this change, however, the stability properties of the benchmark economy remain qualitatively unaffected. Specifically, we establish that for all feasible deficit ratios steady states are dynamically approachable from fixed initial conditions in a uniquely determined and smooth manner. The remainder of the paper is structured as follows. Section 2 introduces the benchmark model. Section 3 discusses the stability properties of this model under strict inflation targeting and a constant money growth rule. Section 4 extends the benchmark model by allowing for a forward-looking money demand specification. Finally, Section 5 offers some conclusions. Proofs not included in the main text and numerical simulation output are delegated to A.1, A.2 and A.3.

Following the seminal paper by Sargent and Wallace (1981), there is a large literature which stresses that the active control of inflation through the monetary agent requires that the fiscal agent accepts the residual role within the government's budget constraint, taking passively as given the behaviour of the monetary agent. For any particular inflation target, the fiscal implications of such a role assignment are likely to be quite different, however, depending on the details of the monetary policy rule. Specifically, as recently stressed by Woodford (2001), as long as government debt is issued largely in nominal terms, different monetary policy rules are likely to imply different government debt dynamics through rule-specific valuations of government debt, constraining thereby the fiscal agent in different ways. Against this background, this paper offers a simple analytical framework to study effects of monetary policy on the valuation of outstanding government debt from a dynamic general equilibrium perspective which takes the desirability of a mix of active monetary and passive fiscal policy as given. The main idea of this paper is to illustrate that monetary policy may indeed constrain fiscal policy in rather different ways, depending on whether monetary policy accepts stabilizing revaluations of government debt or not. More specifically, using a monetary growth model with flexible prices, we compare the properties of two stylized monetary policy rules which have identical steady-state properties but require different actions out of steady state. First, we consider a policy of a constant money growth rule which allows for temporary deviations of inflation from target. As a result, there is scope for revaluations of public debt in response to shocks, and these revaluations are shown to act as automatic stabilizers of government debt dynamics. Second, we consider a policy of strict inflation targeting which implements the target inflation rate also outside the steady state. Essentially, such a policy fixes the value of government debt in real terms and precludes thereby stabilizing revaluations. As we show, this feature implies that additional fiscal restraint may be needed under strict inflation targeting which is not required under a constant money growth rule. As it stands, our approach suffers from the fact that we do not establish optimal programs of monetary and fiscal policy. Similarly, our paper abstracts entirely from credibility issues. Extensions of the model along these dimensions would be important. Yet, given our particular assumption of overlapping generations with finite decision horizons, such extensions raise additional questions which are beyond the scope of this paper. Despite these shortcomings, the largely descriptive findings established in this paper do indicate, however, that the revaluation channel of government debt can be of considerable importance for the dynamic properties of an otherwise standard monetary growth model.