Fiscal Policy Under Inflation Targeting

Røisland, Øistein; Torvik, Ragnar Working Paper Fiscal Policy Under Inflation Targeting Arbeidsnotat, No. 2000/15 Provided in Cooperation with: Norges Bank, Oslo Suggested Citation: Røisland, Øistein; Torvik, Ragnar (2000) : Fiscal Policy Under Inflation Targeting, Arbeidsnotat, No. 2000/15, ISBN 82-7553-174-8, Norges Bank, Oslo, https://hdl.handle.net/11250/2500390 This Version is available at: https://hdl.handle.net/10419/209788 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-nd/4.0/deed.no 2000/15 Oslo December 15, 2000 Arbeidsnotat Economics Department Financial Infrastructure and Payment Systems Department Fiscal Policy under Inflation Targeting by Øystein Røisland and Ragnar Torvik Arbeidsnotater fra Norges Bank kan bestilles over Internett: www.norges-bank.no/publikasjoner eller ved henvendelse til Norges Bank, Abonnementsservice, PB 1179 Sentrum, 0107 Oslo Telefon 22 31 63 83, Telefaks 22 41 31 05 Norges Banks arbeidsnotater inneholder forskningsarbeider og utredninger som vanligvis ikke har fått sin endelige form. Hensikten er blant annet at forfatteren kan motta kommentarer fra kolleger og andre interesserte. Synspunkter og konklusjoner i arbeidsnotatene står for forfatternes regning. Arbeidsnotater/working papers from Norges Bank can be ordered via the Internet: www.norges-bank.no/english/publications or from Norges Bank, Subscription service, P.O.Box. 1179 Sentrum, 0107 Oslo, Norway. Tel. +47 22 31 63 83, Fax. +47 22 41 31 05 Norges Bank's Working papers present research projects and reports (not usually in their final form), and are intended inter alia to enable the author to benefit from the comments of colleagues and other interested parties. Views and conclusions expressed in Working Papers are the responsibility of the authors alone. ISSN 0801-2504 ISBN 82-7553-174-8 Fiscal Policy under Inflation Targeting By Øistein Røisland Norges Bank and Ragnar Torvik Norwegian University of Science and Technology and Norges Bank December 2000 Abstract The paper discusses the role of fiscal policy as an instrument for macroeconomic stabilisation when monetary policy pursues inflation targeting. Within a theoretical model of an open economy with a traded and non-traded sector, we show that inflation targeting may lead to large sectoral imbalances when fiscal policy is passive. There is a potentially large welfare gain from an active fiscal policy, in particular when sectoral stability is included in the welfare function. However, with reasonable parameter values, a small cost of adjusting fiscal policy reduces the optimal degree of activism considerably. The reason is that a given change in output requires a large change in the fiscal policy stance under inflation targeting, because of the monetary policy reaction function. JEL Classification: E52, E62, F41 The views expressed are those of the authors, and not necessarily those of the Central Bank of Norway. We wish to thank Hans Genberg and Bent Vale for useful comments and suggestions. 2 1. Introduction In the last few decades monetary policy has taken over much of the role fiscal policy used to have as a tool of macroeconomic stabilisation. One might, for several reasons, argue that this development is advantageous. Experience shows that using fiscal policy to manage aggregate demand tends to lead to excess government debt, which limits the room for manoeuvre in fiscal policy over time and places large tax burdens on future generations. One reason for this is that it tends to be easier to obtain political support for an expansionary fiscal policy during recessions than for a tight fiscal policy during peaks. Moreover, even though outside lags until fiscal policy affects the economy are relatively short – arguably shorter than for monetary policy – inside lags within the decision process tend to be long, and it is often difficult to reverse fiscal policy decisions. The long inside lags imply a risk that fiscal policy measures come too late and might even become counterproductive. Several countries have therefore adopted fiscal policy rules, which are meant to provide budgetary discipline with less focus on short-term stabilisation. Although there are arguments against using fiscal policy for aggregate demand management, there are also arguments in favour of an active fiscal policy. The general argument is put forward by Ball (1997, p. 2) in the following words: “A carpenter is more successful at his job if he uses both a hammer and a saw. So it is, I will argue, with macroeconomic policy: it is best to use both fiscal and monetary tools.” In aggregated one-sector models, as most models of inflation targeting are, the advantage of using both monetary and fiscal policy for stabilisation purposes is not clear. The social loss in most of these models typically depends on the variability of inflation and the variability of output. Since an expansionary monetary policy and an expansionary fiscal policy both produce higher inflation and higher output (in the short run), the policy tools appear alternative, as opposed to complementary, for stabilisation purposes. The advantage of using both tools comes out clearer when considering a disaggregated model. Specifically, we assume that the aggregate economy consists of two sectors; a traded sector and a non-traded sector. The different stabilisation properties of the two tools are illustrated in Table 1. Table 1 Expansionary monetary policy Expansionary fiscal policy Non-traded sector + + Traded sector + - 3 Monetary expansion leads to lower interest rates and exchange rate depreciation. Output in the non-traded sector increases due to the lower interest rate, and output in the traded sector increases due to the weaker exchange rate. Since output in both sectors is affected in the same direction, monetary policy has a strong effect on aggregate output. A fiscal expansion, however, gives higher output in the non-traded sector since demand increases, while output in the traded sector decreases, because an expansionary fiscal policy generates a real exchange rate appreciation. Since an expansionary fiscal policy generates an expansion in the nontraded sector and a contraction in the traded sector, fiscal policy is less suitable for stabilising aggregate output, but has a strong effect on the balance between traded and non-traded sectors. Thus, monetary policy has a comparative advantage in aggregate stabilisation, while fiscal policy has a comparative advantage in disaggregated stabilisation. In the earlier literature, policy mix issues have in particular been discussed in connection with external balance considerations. Boughton (1989) argued, within a one-sector model, that monetary policy has a comparative advantage in internal stabilisation, since both a lower interest rate and a weaker exchange rate give rise to higher demand for domestic goods. The improved current account following the exchange rate depreciation is, however, partly offset by higher imports due to increased expenditure, so that monetary policy has a comparative disadvantage in external stabilisation. For fiscal policy, the opposite is true. A fiscal tightening leads to an improvement of the current account, both because it leads to real exchange rate depreciation and because lower expenditure gives lower imports. To the extent that the level of national savings is suboptimal, Boughton argued that fiscal policy should be concerned with current account imbalances, while monetary policy should pursue price stability. This view was shared by Genberg and Swoboda (1989), who emphasised the need to focus on the appropriate setting of instruments, as opposed to targets. Tirelli and Vines (1995) argued, within an extended Mundell-Fleming model, that the question of which policy instrument – fiscal versus monetary policy – should be used for the internal objective is a subordinate issue. However, they showed that fiscal policy must play an active role in the policy package in order to prevent dynamic instability. By focusing on disaggregated stability issues, we show that there is a case for an active fiscal policy that does not hinge on an assumption of suboptimal national saving. Concerns about sectoral stability also seem to have received more attention in the public debate in recent 4 years.1 By using both monetary and fiscal tools under inflation targeting, one is able to achieve a better sectoral balance in the economy. Although it may not be clear from a theoretical point of view why sectoral stability should matter, most people seem to have an intuitive understanding of its importance. The intuitive understanding is perhaps best illustrated by considering the following example: Suppose two hypothetical policy regimes provide perfect aggregate output stability. One regime also stabilises each sector perfectly, while the other gives high variability in each sector through contraction in one sector and expansion in the other, so that aggregate stability is achieved. Most people would probably agree that the former regime is superior. The potential importance of sectoral stability is also shared by policy makers. Eddie George (1999, p. 4), the Governor of Bank of England, notes that “We are concerned – as you are – with the health of every sector of the economy, we fully appreciate the interdependence of the different sectors, and we well understand the part that greater real exchange rate stability can play in promoting more balanced economic growth.” The importance of sectoral stability may also exceed that of short term stabilisation, which we focus on in the present paper. In models with dynamic increasing returns to scale, stability may also affect investment, productivity, and long-term growth. van Wijnbergen (1984), Krugman (1987), Sachs and Warner (1995) and Torvik (2001) point out different channels through which the balance between traded and non-traded sectors in the short run affect longterm productivity and growth. Gylfason et al. (1999) show how an unstable real exchange rate that generates instability in the traded sector reduces investment, and how this translates into lower growth because of dynamic increasing returns to scale. Although it is generally accepted that sectoral stability is important and that the policy mix affects sectoral stability, this issue has not, to our knowledge, been studied in the literature on inflation targeting. Our aim, therefore, is to develop a simple model of inflation targeting which may be used to analyse these issues. We derive the optimal fiscal policy under inflation targeting in a model with a traded and non-traded sector. The optimal fiscal policy is defined 1 The following citation from Reuters (1998), which dealt with the recent economic development in the UK, provides an example of the public concern for the balance between traded and non-traded sectors and its implications for economic policy: “Buiter [one of the members of the Monetary Policy Committee of the Bank of England] said the macro-economic policy mix could, in theory, be more balanced, and highlighted the twospeed nature of the economy: ‘Internationally shielded parts of the economy are doing fairly well, some very well, but the internationally exposed parts of the economy are getting hammered pretty hard’.” 5 as the fiscal policy that minimises a given loss function where aggregate stability as well as sectoral stability are included. The paper is organised as follows: Section 2 presents the model and the solution of the model assuming an exogenous fiscal policy. Section 3 discusses the effects of various shocks and derives the optimal fiscal policy responses to these. Concluding remarks are presented in Section 4, while an appendix discusses time-inconsistency problems in fiscal policy and possible solutions to these problems. 2. The model We keep the model as simple as possible in order to focus on sectoral stability considerations. In doing so, we disregard important dynamic features. Specifically, we assume that monetary policy affects prices and output in the same period as it is implemented. The advantage of the simple dynamic structure is that an analytical solution to the model is tractable even with rational expectations in the foreign exchange rate market. This solution provides our benchmark from which to discuss fiscal policy under inflation targeting. An additional advantage of our simple dynamic structure is that it allows us to know with certainty that the results follow from the assumptions in the theoretical model and not from calibrated parameter values. Bharucha and Kent (1998) and Leitemo and Røisland (2000) provide calibrated inflation targeting dynamic models with a traded and non-traded sector, although these models are not intended for studying fiscal policy. Our model extends the model in Røisland and Torvik (1999), which is a two-sector version of the one-sector models by Genberg (1989) and Rødseth (1996). The country is assumed to be small compared with the rest of the world, so that the world market price of traded goods and the world market interest rate can be taken as exogenous. All variables are in logs, except for the interest rate. Real variables are measured as deviations from a given steady state equilibrium and all shocks have zero expectation and are uncorrelated over time. We assume rational expectations where monetary and fiscal policy do not affect the real economy in the long run. In the short run, unexpected monetary or fiscal policy affects real variables due to predetermined wages. Wages are determined one period in advance so that expected unemployment equals its natural rate (given that the expectation of all shocks are zero and that the fiscal and monetary policy stance thus are not expected to deviate from their longterm equilibrium values). 6 The supply of non-traded goods is given by where yN is the (log of the) supply of non-traded goods, pN is the (log of the) price of nontraded goods and w is the nominal wage level. λN is a positive constant, which measures the supply elasticity with respect to the producer real wage, and uN is a supply shock to the nontraded sector. The supply of traded goods is given by where yT is the supply of non-traded goods, p* is the world market price and s is the nominal exchange rate. λ T is the supply elasticity of the traded sector and uT is a supply shock to the traded sector. Since production is measured in logs and as deviations from steady state, aggregate production is a weighted average of production in the two sectors: where θ is the share of traded sector production in steady state GDP. The real exchange rate e is given by The foreign exchange market is represented by uncovered interest rate parity and rational expectations, i.e.2 where r* is the world real interest rate and z is a shock to the foreign exchange market, which can be interpreted as a stochastic risk premium. The second equality follows from the 2 Note that real UIP follows from its nominal counterpart by defining r* = i* - (Ep* - p*) and using the definition of e. TTT uwspy +−+= ∗)()2.2( λ NT yyy )1()3.2( θθ −+= NNNN uwpy +−= )()1.2( λ * (2.5) ( )rr Eee zr ez ∗ =+ −+=−+ N pspe −+= ∗ )4.2( 13 The optimal fiscal policy is then given by 1)6.3( −= v x δ δ The demand curve thus shifts back to the initial position, and the new equilibrium is in point D, which is equal to the initial position. Since fiscal policy neutralises the demand shock completely, the interest rate and thus the exchange rate can remain unchanged. If it is costly to change the fiscal policy stance, the demand shock should not be completely neutralised by fiscal policy. Then, the interest rate must be increased in order to reach the inflation target. The result will then be somewhere in between point C and point D in Figure 2, where nontraded output has increased somewhat and traded output has decreased. The optimal fiscal policy modifies the result that inflation targeting destabilises the traded sector when demand shocks occur. Only in the case where there are no costs involved in using fiscal policy for stabilisation purposes is it optimal to fully prevent the destabilisation. 3.2 Foreign exchange shocks; z A positive shock to the risk premium (or the world real interest rate) leads, for a given interest rate, to an exchange rate depreciation. The depreciation increases traded goods prices directly and non-traded goods prices indirectly through the substitution effect in demand, so that y N A, D pT y T Figure 2.a Figure 2.b C B A, B C D 14 output in both sectors increase. The result without monetary or fiscal response is illustrated by point B in Figure 3.a and 3.b. Since prices on both goods have increased, the central bank must raise the interest rate in order to achieve the inflation target. For a given fiscal policy, the rise in the interest rate should not offset the initial depreciation completely. To see this, assume that the rise is sufficiently large to keep the exchange rate constant. Then, traded goods prices would be constant, and demand would thus not shift towards non-traded goods. However, since the interest rate would be higher, private sector demand for non-traded goods would decrease, leading to lower prices in this sector. Unchanged traded goods prices and lower non-traded goods prices imply that the rise in the interest rate would be excessive. The appropriate monetary policy response, given the fiscal policy, should let the exchange rate partly depreciate, but offset higher prices and production in the traded sector by lower prices and production in the non-traded sector. The equilibrium under inflation targeting and a neutral fiscal policy is represented by point C in the figures. The optimal fiscal policy response to a foreign exchange shock is found from equations (3.3) and (3.4): 0)7.3( 4 2 313 2 212 2 111 2 313 2 212 2 111 2> +++ ++ = ββββ βββ α δ δ aaa aaa v x Under inflation targeting, depreciation pressure in the foreign exchange market should thus be met by an expansionary fiscal policy. As compared to the case with a passive fiscal policy, the expansionary fiscal policy leads to higher output and prices in the non-traded sector, which necessitates a further rise in the interest rate in order to achieve the inflation target. This gives an exchange rate appreciation, which reduces production in the traded sector, compared to y N A, D pT y T Figure 3.a Figure 3.b C B A, D C B 15 point C in the figure. In the case where there are no costs of changing the fiscal policy stance ( β 4 = 0), the increase in the interest rate must be sufficient to maintain a constant exchange rate and thus constant output in the traded sector. In the non-traded sector, lower demand due to the rise in the interest rate is offset by the expansionary fiscal policy. Equilibrium with an optimal fiscal policy and no fiscal adjustment costs is represented by point D in Figure 3.a and 3.b. The effects of the foreign exchange shock is then completely neutralised. When there are adjustment costs, the effects of the shock are not fully neutralised, and the result of the optimal fiscal policy response will be somewhere in between point C and D. Compared to a passive fiscal policy, stability in both the non-traded and traded sector is higher. We have assumed that the shocks to the economy are truly exogenous. However, shocks, in particular foreign exchange shocks, may not be independent of economic policy. If the source of the depreciation pressure is low credibility of the fiscal authorities, the optimal response is obviously not to ease up fiscal policy. The combination of tight monetary policy and loose fiscal policy might then exacerbate the depreciation pressure. If, however, the source of the shock is e.g. increased focus on exchange rate risk among international investors, as experienced in the period after the crises in South-East Asia and Russia in 1997-98, the appropriate policy measure would be a higher interest rate and an expansionary fiscal policy. 3.3 Cost-push shocks; w The stabilisation properties of inflation targeting (at least when strictly conducted) are often accused of being poor when supply shocks occur, since these push prices and output in opposite directions. Higher wages shift supply curves in Figure 4.a and 4.b up. With no macroeconomic policy response, the shock brings the economy to point B in the figure, with lower production in both sectors. Prices on non-traded goods increase, and under inflation targeting, the central bank must raise the interest rate. The tight monetary policy leads to an exchange rate appreciation, shifting the price line in Figure 4.b down, further decreasing output in the traded sector. The appreciation and the higher interest rate result in a negative shift in the demand curve in the non-traded sector, decreasing output further also here. As confirmed by equation (2.12), a positive shock to wages leads to reduced output in both sectors under inflation targeting, represented by point C in the figures. In the traded sector, both higher labour costs and an exchange rate appreciation induced by the rise in the interest rate leads to lower production. In the non-traded sector, higher labour costs, a higher interest rate and an exchange rate appreciation lead to reduced output. 16 To what extent can an active use of fiscal policy counteract the destabilising effects of the monetary policy response? Suppose that the fiscal authorities respond to falling production by an expansionary fiscal policy. This shifts the non-traded sector demand curve outwards. To fulfil the inflation target, the central bank must respond with a further increase in the interest rate. The result is a further decline in traded sector output, while output in the non-traded sector increases. In the realistic case where λN > λT, the increase in non-traded output is larger than the decrease in traded output, so that the net effect on aggregate output is positive. In the special case where only aggregate output stability matters, i.e. β 1 > 0, β 2 = β 3 = β 4 = 0, the optimal fiscal policy response to a wage shock is found by setting y = 0 in equation (2.12): Complete aggregate output stabilisation is represented by point D in Figure 4.a and 4.b. As seen from the figure, a positive wage shock and an expansionary fiscal policy results in a deep recession in the traded sector and a boom in the non-traded sector. Stabilising aggregate output by fiscal policy under inflation targeting thus generates large sectoral imbalances. Suppose instead that only traded output stability matters, i.e. β 2 > 0, β 1 = β 3 = β 4 = 0. The optimal fiscal policy response is found by setting y = 0 from equation (2.12): 0 1 )9.3( 21 0,0 4312 < − ++ −= ===> θ λαα δ δ ββββ N w x 0 ))(1( ))1()(( )8.3( 21 0,0 4321 > −− +−++ = ===> TN NTNT w x λλθθ λθλλθθλαα δ δ ββββ y N A pT y T Figure 4.a Figure 4.b C B A C B D D 17 Thus, if only traded sector stability matters, fiscal policy should be tight when a positive wage shock occurs. The reason is that higher wages lead to lower production in the traded sector. In order to offset the adverse cost-push effect, a depreciation in the exchange rate is required. Since monetary policy is committed to the inflation target, fiscal policy must be tight if the central bank shall lower the interest rate and thus induce an exchange rate depreciation. For completeness, we consider the case where only stability in the non-traded sector matters, i.e. β 3 >0, β 1 = β 2 = β 4 = 0. From equation (2.12) we see that the optimal fiscal policy response is given by 0)10.3( 21 0,0 4213 > + = ===> θ α α δ δ ββββ w x Note that the optimal fiscal policy response is lower in the case where only non-traded output stability matters compared to the case where only aggregate output stability matters. The reason is that unchanged production in the non-traded sector implies that aggregate output is lower, since production in the traded sector has decreased. In order to stabilise aggregate output, fiscal policy must be even looser, so that the non-traded sector goes into a boom, while the traded sector goes into an even deeper recession. The burden on the traded sector is thus smaller if the fiscal policy authorities aim at stabilising the non-traded sector rather than aggregate output. 3.4 Productivity shocks; uN, uT If both sectors are hit by the same productivity shock, i.e. uN= uT, the effects of a positive shock are equivalent with a negative wage shock, w. Thus, the same conflicting objectives between stability in the traded versus the non-traded sector apply. The sign of the optimal fiscal policy response to asymmetric supply shocks depends on the weight attached to traded sector stabilisation. Suppose first that the fiscal authorities only care about aggregate stability, i.e. β 1 > 0, β 2 = β 3 = β 4 = 0. The optimal fiscal policy responses are then given by 18 0 ))(1( )( )12.3( 21 0,0 4321 < −− ++ −= ===> TN N T u x λλθθ θλααθ δ δ ββββ A positive shock to non-traded supply, uN, leads to higher production in both sectors. The reason is that in addition to the direct effect of the shock, the central bank must lower the interest rate in order to achieve the inflation target. This produces an exchange rate depreciation, which also increases production in the traded sector. In order to stabilise aggregate output, fiscal policy must be tight. This leads to lower non-traded output, but higher traded output, since the interest rate is lowered even more. In order to stabilise aggregate output when λN > λT, it can be seen by inserting equation (3.11) in equation (2.12) that fiscal policy must be adjusted so that the traded sector enters a boom, while the non-traded falls into a recession. Thus, under inflation targeting and optimal fiscal policy, the non-traded sector that experiences the positive supply shock in the first place ends in a contraction when only aggregate output stability matters. The traded sector is destabilised more than under a passive fiscal policy. A positive shock to traded output, uT, does not trigger any direct monetary policy response, since it does not affect inflation (since prices on traded goods are given on the world market). Since the shock results in higher traded output, it also leads to higher aggregate output. To neutralise the effect of the shock on aggregate output, fiscal policy must be tight. A tight fiscal policy is met by a reduction in interest rates, and thereby exchange rate depreciation, under inflation targeting. Traded output consequently increases further, while non-traded output decreases so that aggregate output is constant. Thus, when only aggregate stability matters, the fiscal policy response destabilises the traded sector further, irrespective of which sector experiences a supply shock. Consider then the case where the fiscal authorities only care about stability in the traded sector. The optimal fiscal policy responses are then given by 1)13.3( 0,0 4312 = ===> ββββ δ δ N u x 0 ))(1( ))(1( )11.3( 21 0,0 4321 < −− ++− −= ===> TN T N u x λλθθ θλααθ δ δ ββββ 19 0 )1( )14.3( 21 0,0 4312 > − ++ = ===> T N T u x λθ θλαα δ δ ββββ A positive shock to non-traded supply leads to higher production and lower prices in the nontraded sector. The central bank must then lower the interest rate to achieve the inflation target. This produces an exchange rate depreciation, which increases production in the traded sector. In order to stabilise traded output, fiscal policy must be sufficiently expansionary to prevent the central bank from lowering the interest rate. A positive shock to traded production does not trigger a monetary policy response. In order to prevent higher production in the traded sector, an exchange rate appreciation is required. In order to have an exchange rate appreciation, fiscal policy must be expansionary, so that the central bank can raise the interest rate. When stability in the non-traded sector is the only thing that is important, the optimal responses to supply shocks are given by 0)15.3( 21 0,0 4213 < + −= ===> NN u x θλ α α δ δ ββββ 0)16.3( 0,0 4213 = ===> ββββ δ δ T u x A positive shock to non-traded output should give rise to a tight fiscal policy in order to neutralise the positive direct effect of the shock and the indirect effect from a lower interest rate. A shock to traded output, on the other hand, should neither trigger any fiscal policy response, nor any monetary policy response, as long as stability in the non-traded sector is the only thing that matters. 3.5 Flexible inflation targeting We have seen that inflation targeting may give rise to large sectoral imbalances. At first sight, one may believe that this is partly the result of the strict inflation targeting regime studied. In the more realistic case of flexible inflation targeting, the central bank also takes into account aggregate output stability. This may, however, exacerbate the problem. To see this, reconsider the case of a demand shock studied above. With strict inflation targeting (and passive fiscal policy), this leads to a higher interest rate, lower production in the traded sector, higher production in the non-traded sector, and higher aggregate production provided λN > λT. If the central bank has output stability, in addition to inflation, in its objective function, the interest 20 rate will increase even further in response to a demand shock, and stabilise aggregate production more than in the case with strict inflation targeting. This being the case, traded sector production becomes even more unstable than with strict inflation targeting. Therefore, flexible inflation targeting does not remove the motive for an active fiscal policy under inflation targeting, given that sectoral and not only aggregate instability is considered costly. 3.6 A numerical example Since the analytical solution for optimal fiscal policy becomes less transparent in the general case, we will consider a numerical example. The following structural parameters need to be calibrated: θ, α 1, α 2, λ N and λ T. We use the manufacturing sector as a proxy for the traded sector and manufacturing production as a share of GDP as a base value for θ. This gives a value of θ close to 1/3 in most countries. For α 1 and α 2, we use the aggregate elasticities from Batini and Haldane (1998) and scale them by 3/2 in order to make the elasticities in the nontraded sector consistent with the aggregate elasticities. This gives α 1 = .75 and α 2 = .2. These are also broadly consistent with estimates used in other central banks. We choose λ T = .5, which is close to the estimate of the elasticity with respect to wages in the manufacturing sector in Nickell and Wadhwani (1991). There are few empirical studies that cast light on plausible values for the supply elasticity in the non-traded sector. Disney and Ho (1990) estimated both λ N and λ T for Singapore, but the estimate for λ N seems unreasonably small, which the authors explain by measurement errors and simultaneity in the determination of domestic demand, employment and real wage levels. However, Burgess (1988) reports estimates of the economy-wide elasticity of around 1.0, which we use as a base value. λ N is then found from (1 ) 1.0 TN θλ + −θ λ = , which, with the above parameter values, gives λ N = 1.25. First consider the solutions for aggregate, traded and non-traded output when fiscal policy is neutral (passive), i.e. x = 0. From (2.12) and the above parameter values, we find that (3.17) .11 .09 .86 .56 .33 NT yvzwu u=+− + + (3.18) .23 .17 .79 .23 1.0 TNT yvzwuu=− + − + + (3.19) .30 .23 .96 .77 NN yvzwu=−−+ According to conventional wisdom, inflation targeting stabilises output when demand shocks occur, since these drive prices and output in the same direction. The numerical exercise conducted here illustrates that aggregate output is quite stable when demand shocks occur. 21 However, the monetary policy response results in relatively larger fluctuations in each sector. Thus, if sectoral stability is considered important, there seems to be a potential welfare gain from an active fiscal policy that is larger than what the number for aggregate output fluctuation suggests. The same picture emerges from shocks in the foreign exchange market. As regards supply shocks, however, the two sectors move in tandem and the output fluctuations are quite large compared with the fluctuations with demand shocks. Thus, in this case also, it should clearly be of interest to investigate the optimal fiscal policy responses. In the numerical simulations, we restrict the analysis to the case where aggregate output, traded output and non-traded output have the same weight in the loss function. The optimal fiscal policy and the solutions for aggregate, traded and non-traded output under the optimal fiscal policy are summarised in Table 2. For each type of shock, we investigate two different cases; one without any cost attached to changes in the fiscal policy stance ( β 4 = 0) and one where there is a direct cost associated with using fiscal policy for stabilisation purposes ( β 4 = .2). For each type of shock, we show the optimal fiscal policy response as well as effects on aggregate and disaggregated output. Table 2. β 1 = 1, β 2 = 1, β 3 = 1 vzwu NuT β 4 = 0 β 4 = .2 β 4 = 0 β 4 = .2 β 4 = 0 β 4 = .2 β 4 = 0 β 4 = .2 β 4 = 0 β 4 = .2 x-1.0 -.44 .75 .33 1.33 .59 -1.55 -.68 1.22 .54 y0 .06 0 -.05 -.71 -.79 .38 .48 .47 .39 yT0 -.13 0 .10 -1.09 -.92 .59 .39 .72 .88 yN0 .17 0 -.13 -.56 -.79 .30 .56 .37 .16 Consider first the case where there are no costs of adjusting the fiscal policy. Fiscal policy is then sufficiently flexible to completely neutralise demand shocks and shocks in the foreign exchange market. The fiscal policy coefficient on the wage shock is positive. Thus, while the absolute value of the output response in the traded sector equals .79 with a passive fiscal policy, it rises to 1.09 when fiscal policy is used for stabilisation purposes. The traded sector output response is also higher under an active fiscal policy when there are supply shocks to the non-traded sector. The reason for these results is that in the above set of weights and parameter values, the relative weight attached to traded sector stability is not large enough to induce a fiscal policy that decreases the effect of the shocks to the traded sector. On the contrary, the optimal fiscal policy response is a policy that reduces the effects on aggregate and non-traded output, but 22 enhances the effect on the traded sector. In the case of a wage shock, the optimal response of an expansionary fiscal policy induces an even higher interest rate and thereby a further appreciation. With a supply shock to the non-traded sector, the fiscal policy is tight, meaning an even lower interest rate and a larger depreciation than in the case with a passive fiscal policy. With the exception of supply shocks in the traded sector, all shocks produce more stable output in the non-traded sector when fiscal policy is adjusted to maximize the welfare function. The fiscal policy response to a positive supply shock in the traded sector should be an expansionary policy, which, under inflation targeting, requires a rise in the interest rate and thereby an appreciation, which stabilises traded output. Consider then the case where it is costly to adjust fiscal policy. Specifically, we have chosen β 4 = .2, which might be considered as a case with positive, but moderate, fiscal adjustment costs. As seen from Table 2, even relatively small fiscal adjustment costs reduce the degree of optimal fiscal activism considerably. The reason is that the optimal degree of fiscal activism depends on the size of fiscal adjustment costs relative to the effect fiscal of policy on target variables. Under inflation targeting, an expansionary (contractionary) fiscal policy stance is met by a rise (fall) in the interest rate. For this reason, fiscal policy is less effective than under monetary policy regimes where the interest rate does not automatically respond to increased demand. Since fiscal policy is relatively uneffective, it must not be used frequently when it is associated with costs. Figure 5 0 0.2 0.4 0.6 0.8 1 0.5 1 1.5 2 β 4 |b| 29 References Agell, J., L. Calmfors and G. Jonsson (1996). “Fiscal policy when monetary policy is tied to the mast.” European Economic Review 40, 1413-1440. Ball, L. (1997). “A proposal for the next macroeconomic reform.” Victoria Economic Commentaries, March. Batini, N. and A.D. Haldane (1998). “Forward-looking rules for monetary policy.” Bank of England Working Papers No. 91. Bharucha, N. and D. Kent (1998). “Inflation targeting in a small open economy.” Reserve Bank of Australia Research Discussion Paper No. 9807. Boughton, J. (1989). “Policy assignment strategies with somewhat flexible exchange rates.” In R. Portes (eds), Blueprints for Exchange Rate Management, CEPR and Academic Press, London. Burgess, S.M. (1988). “Employment adjustment in UK manufacturing.” Economic Journal 98, 81-103. Cecchetti, S.G. and M. Ehrmann (1999). “Does inflation targeting increase output volatility? An international comparison of policymakers’ preferences and outcomes.” NBER Working Paper No. 7426. Disney, R. and S.K. Ho (1990). “Do real wages matter in an open economy? The case of Singapore: 1966-1987.” Oxford Economic Papers, vol. 42, pp. 635-57. Dornbusch, R. (1983). “Real interest rates, home goods, and optimal external borrowing.” Journal of Political Economy, vol. 91, pp. 141-53. Frankel, J., and M. Chinn (1995). ’The Stabilizing Properties of a Nominal GNP Rule.’ Journal of Money, Credit, and Banking 27, 318-334. Genberg, H. (1989). ‘Exchange rate management and macroeconomic policy: A national perspective.’ Scandinavian Journal of Economics, vol. 91, pp. 439-69. 30 Genberg, H. and A. Swoboda (1989). “Policy and current account determination under floating exchange rates.” IMF Staff Papers 36, 1-30. George, E. (1999). “Mr. George offers an overview of the UK economy.” Speech given by Eddie George to a dinner hosted by Birmingham City 2000 in Birmingham, England, on 18 May 1999, BIS ReviewNo 56, http://www.bis.org/review/ Gylfason, T., Herbertson, T.T., Zoega, G. (1999). “A mixed blessing: Natural resources and economic growth.” Macroeconomic Dynamics 3, 204-225. King, M. (1997), "Changes in UK monetary policy: Rules and discretion in practice", Journal of Monetary Economics 39, 81-97. King, M. (1999). “Mr. King speaks about interest rates, policy and the UK economy.” Speech given by Mervyn King to the Scottish Council Development and Industry, held in Edinburgh on 11 October 1999. BIS Review No. 111, http://www.bis.org/review/ Krugman, P. (1987). “The narrow moving band, the Dutch disease, and the competitive consequences of Mrs. Thatcher: Notes on trade in the presence of dynamic scale economies.” Journal of Development Economics 27, 41-55. Leitemo, K. and Ø. Røisland (2000). “Alternative Monetary Policy Regimes for Small Open Economies.” Working Paper No. 5, Norges Bank. Nickell, S. and S. Wadhwani (1991). “Employment determination in the British industry: investigations using micro-data.” Review of Economic Studies 58, 955-69. Persson, T. and G. Tabellini (1996), ”Monetary Cohabitation in Europe.” NBER Working Paper No. 5532. Persson, T. and G. Tabellini (1997). “Political economics and macroeconomic policy.” NBER Working Paper No. 6329. Reuters (1998). “UK: BOE’s Buiter – government econ policy ‘not unhelpful’ to MPC.” Reuters Business Briefing, 02.04.98. 31 Rogoff, K. (1985). "The optimal degree of commitment to a monetary target." Quarterly Journal of Economics 100, 1169-1190. Rødseth, A. (1996). “Exchange rate versus price level targets and output stability.” Scandinavian Journal of Economics, vol. 98, pp. 559-77. Røisland, Ø. (2001). “Institutional arrangements for monetary policy when output is persistent.” Journal of Money, Credit and Banking, forthcoming. Røisland, Ø. and R. Torvik (1999). “Exchange rate versus inflation targeting: A theory of output fluctuations in traded and non-traded sectors.” Working Paper No. 1, Norges Bank. Sachs, J.D., Warner, A.M. (1995). ”Natural resource abundance and economic growth.” NBER Working Paper No. 5398. Svensson, L.E.O. (1997). "Optimal inflation targets, 'conservative' central banks, and linear inflation contracts." American Economic Review 87, 1111-46. Tirelli, P. and D. Vines (1995). “Kohl, Reagan, and open economy macroeconomics: Revisiting rules for fiscal and monetary policy.” Oxford Economic Papers 47, 561-583. Torvik, R. (2001). “Learning by doing and the Dutch disease.” European Economic Review, forthcoming. Walsh, C. (1995). "Optimal contracts for independent central bankers." American Economic Review 85, 150-167. van Wijnbergen, S. (1984). ”The ‘Dutch disease’: A disease after all?” Economic Journal 94, 41-55. Øistein Røisland and Ragnar Torvik: Fiscal Policy under Inflation Targeting Arbeidsnotat/Working Paper 2000/15 KEYWORDS: Inflation Targeting Fiscal policy Monetary policy Small open economy 21127