Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

The macroeconomic effects of fiscal policy

The macroeconomic effects of fiscal policy Applied Economics, 2012, 44, 4439–4454 The macroeconomic effects of fiscal policy a,b, c,d Anto´ nio Afonso * and Ricardo M. Sousa Directorate General Economics, European Central Bank, Kaiserstraße 29, Frankfurt am Main D-60311, Germany Department of Economics, ISEG/TULisbon, Technical University of Lisbon, UECE – Research Unit on Complexity and Economics, R. Miguel Lupi 20, Lisbon 1249-078, Portugal Department of Economics and Economic Policies Research Unit (NIPE), University of Minho, Campus of Gualtar, Braga 4710-057, Portugal Financial Markets Group (FMG), London School of Economics, Houghton Street, London WC2 2AE, UK We investigate the macroeconomic effects of fiscal policy using a Bayesian Structural Vector Autoregression (B-SVAR) approach. We identify fiscal policy shocks via a partial identification scheme, but also: (i) include the feedback from government debt; (ii) look at the impact on the composition of output; (iii) assess the effects on asset markets; (iv) use quarterly data; and (v) analyse empirical evidence from the US, the UK, Germany and Italy. The results show that government spending shocks, in general, have a small effect on Gross Domestic Product (GDP); lead to important ‘crowding-out’ effects; have a varied impact on housing prices and generate a quick fall in stock prices. Government revenue shocks generate a mixed effect on housing prices and a small and positive effect on stock prices. The empirical evidence also suggests that it is important to explicitly consider the government debt dynamics in the model. Keywords: fiscal policy; Bayesian structural VAR; debt dynamics JEL Classification: C11; C32; E62; H62 I. Introduction fiscal policy committees, the influence of regulation in the structure of market incentives, and the Balanced Compared to the large empirical literature on the Budget Amendment in the US are based on the effects of monetary policy on economic activity, fiscal assumption that fiscal policy can be an effective tool policy has received less attention, a feature that for stabilizing business cycles. contrasts with the public debates on its role. The First, we consider the effects of fiscal policy on the government deficit and debt limits of the Stability composition of Gross Domestic Product (GDP), and Growth Pact in the context of the Economic and namely, by estimating the impact of government Monetary Union (EMU), the possibility of indepen- spending and government revenue shocks on pri- dent institutions running fiscal policy, the creation of vate consumption and private investment as in *Corresponding author. E-mail: [email protected]; [email protected] Applied Economics ISSN 0003–6846 print/ISSN 1466–4283 online  2012 The Author(s). Published by Taylor & Francis. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the 4439 named author(s) have been asserted. http://www.tandfonline.com http://dx.doi.org/10.1080/00036846.2011.591732 4440 A. Afonso and R. M. Sousa Galı et al. (2007). Consequently, we are able to results of Giavazzi and Pagano (1990, 1996), Giudice et al. (2004) and Afonso (2010). identify the potential ‘crowding-out’ effects of fiscal Interestingly, when one looks at the response of policy on the private sector. asset prices, the findings suggest that markets tend to Second, we also ask how asset markets (via not interpret expansionary fiscal policies as leading to a only stock prices but also housing prices) are affected deterioration of public finances. Therefore, stock by fiscal policy shocks. Sounder fiscal positions, in prices react negatively to a rise of government particular when associated with a permanent fall in spending and positively when there is fiscal consol- government debt, can have a strong impact on both idation. Nevertheless, while the reaction of stock stock market prices and long-term government bond prices is relatively quick, the effect on housing prices rates (Ardagna, 2009). In addition, fiscal policy can is, in general, persistent. have a major influence on housing markets, namely, When we explicitly take into account the feedback via specific subsidies, other tax measures and its from government debt in our framework, the effects effects on household’s disposable income. Therefore, of fiscal policy on (long-term) interest rates and GDP we look at the persistence of the effects of fiscal become more persistent and these variables are also shocks and assess whether the reaction of stock and more responsive to the shock. housing prices is quantitatively similar or exhibits The rest of this article is organized as follows. asymmetry instead. Section II reviews the related literature. Section III We identify fiscal policy shocks using a recursive explains the empirical strategy used to identify the partial identification based on the work of Christiano effects of fiscal policy shocks, and to take into et al. (2005), and estimate a Bayesian Structural account the uncertainty regarding the posterior Vector Autoregression (B-SVAR) model, therefore, distribution impulse-response functions. Section IV accounting for the posterior uncertainty of the provides the empirical analysis and discusses the impulse-response functions. results. Section V concludes this article with the main Another novelty of this article is that we explicitly findings and policy implications. include the feedback from government debt in our estimations. In fact, while equilibrium structural models are normally solved by imposing the govern- ment’s intertemporal budget constraint, this does not II. Literature happen with Vector Authogression (VAR)-based fiscal policy models. We deal with this limitation by For the US, different approaches have been used in following Favero and Giavazzi (2008), and, as a the identification of the fiscal policy shock. Ramey result, we consider the response of fiscal variables to and Shapiro (1998) use a ‘narrative approach’ to the level of the government debt. isolate political events, and find that, after a brief rise In addition, an important contribution of this in government spending, nondurable consumption article is the use of quarterly fiscal data, which allows displays a small decline while durables consumption us to identify more precisely the effects of fiscal falls. Following the same approach, Edelberg et al. policies. We analyse empirical evidence from the US, (1999) show that episodes of military build-ups have a the UK, Germany and Italy, respectively, for the significant and positive short-run effect on US output periods 1970:Q3 to 2007:Q4, 1964:Q2 to 2007:Q4, and consumption, and that the sign of the response 1980:Q3 to 2006:Q4 and 1986:Q2 to 2004:Q4. The set does not change when anticipation effects are taken of quarterly fiscal data, is taken from national into account. Fatas and Mihov (2001) and Favero accounts (in the case of the US and the UK) or (2002) use a Cholesky ordering to identify fiscal computed by drawing on the higher frequency shocks and show that increases in government (monthly) availability of fiscal cash data (for expenditures are expansionary, but lead to an Germany and Italy). To the best of our knowledge, increase in private investment that more than com- such fiscal data set, apart from not having been pensates for the fall in private consumption. compiled before, has not yet been used in this strand Blanchard and Perotti (2002) use information about of economic modelling. the elasticity of fiscal variables to identify the Broadly speaking, the results point to an expan- automatic response of fiscal policy, and find that sionary effect of fiscal policy in the case of the US and expansionary fiscal shocks increase output, have a the UK in line with the traditional ‘Keynesian’ model positive effect on private consumption and a negative and corroborated by Blanchard and Perotti (2002). impact on private investment. More recently, using As for Italy and Germany, there is some evidence of a sign restrictions on the impulse-response functions ‘Non-Keynesian’ multiplier in accordance with the and identifying the unexpected variation in The macroeconomic effects of fiscal policy 4441 government spending by a positive response of On the other hand, and as mentioned for Italy by expenditure for up to four quarters after the shock, Jappelli and Pistaferri (2007), tax deductibility of Mountford and Uhlig (2005) find a negative effect in interest rates may not have affected much the demand residential and nonresidential investment. for mortgage debt. In addition, sounder fiscal posi- Regarding other countries, Perotti (2004) investi- tions and lower sovereign financing needs allow for gates the effects of fiscal policy in Australia, Canada, lower interest and better financing conditions for West Germany, US and the UK, and finds a mortgage-loans, while higher government indebted- relatively large positive effect on private consumption ness can crowd-out resources available to would be and no response of private investment. Biau and home-owners (Maclennan et al., 1999). Girard (2005) find a cumulative multiplier of gov- In the case of stock prices the attention has been ernment spending larger than one, and positive normally targeted towards the role played by mon- reactions of private consumption and private invest- etary policy. Rigobon and Sack (2002, 2003) and ment in France. For Spain, Castro and Cos (2008) Craine and Martin (2003) use a heteroscedasticity- report that, while there is a positive relationship based estimator and find a significant response of the between government expenditure and output in the stock market to shocks in the interest. Bernanke and short-term, in the medium and long-term expansion- Kuttner (2005) show that a hypothetical unantici- ary spending shocks only lead to higher inflation and pated 25-basis-point cut in the Federal funds rate lower output. Heppke-Falk et al. (2006) use cash data target is associated with about a 1% increase in broad for Germany, and find that a positive shock in stock indexes. More recently, Ardagna (2009) reports government spending increases output and private that fiscal adjustments based on expenditure reduc- consumption, although the effect is relatively small. tion and signalling sounder fiscal behaviour are Giordano et al. (2007) show that, in Italy, govern- related with increases in stock market prices. Using ment expenditure has positive and persistent effects a panel of Organization for Economic Co-operation on output and on private consumption. and Development (OECD) countries, the author also As for the empirical importance of housing over shows that fiscal consolidation that lead to a perma- the business cycle, there are only a small number of nent and substantial fall in government debt are papers that discuss the empirical link between linked to a stronger increase in stock market prices. economic policy and housing prices, the focus has In terms of interest rates, according to Gale and mainly been on the effects of monetary policy. Some Orszag (2003) there are two important reasons as to examples are McCarthy and Peach (2002) and why budget deficits may raise nominal interest rates: Chirinko et al. (2004). Iacoviello and Minetti (2003) (i) they reduce aggregate savings when private savings emphasize the housing market as creating a credit do not increase by the same amount (no Ricardian channel for monetary policy. Iacoviello (2005) looks equivalence) and if there are no compensating foreign at the monetary policy-house price to consumption capital inflows, which leads to a decrease in the channel and finds a significant effect on house prices. supply of capital; and (ii) they increase the stock of Julliard et al. (2008) suggest that monetary policy government debt and, consequently, the outstanding contractions have a large and significantly negative amount of government bonds (relative to other impact on real housing prices, but the reaction is financial assets). In this case, there is a ‘portfolio extremely slow. On the other hand, monetary policy effect’, as a higher interest rate on government bonds shocks do not seem to cause a significant impact on would be required in order to incentive investors to stock markets. hold the additional bonds. In what concerns fiscal policy, it can impinge on While some studies find that interest rates tend to housing market developments notably via subsidies increase after a rise in the deficit, others do not (Engen and tax measures: taxation of the imputed rental and Hubbard, 2004). The empirical findings seem to value of the house, tax deductibility of interest depend on whether expected or current budget deficits payments, capital taxes on housing gains and Value are used as explanatory variables (Brook, 2003; Added Tax (VAT) on new houses. Given short-run Upper and Worms, 2003; Laubach, 2009), and also inelastic housing supply, fiscal subsidies for buying on whether yield differentials in Europe with houses may end up pushing up its demand and prices. respect to Germany (Codogno et al., 2003) or interest Giavazzi and Pagano (1990) and Alesina and Ardagna (1998) have uncovered the presence of ‘non-Keynesian effects’ (i.e. negative spending multipliers) during large fiscal consolidations, with output rising significantly despite large cuts in government spending. Perotti (1999) also obtains such findings, but only in circumstances of ‘fiscal stress’ (unusually high debt-to-GDP ratios). In addition, Afonso (2010) finds some evidence of expansionary fiscal consolidations, for a few budgetary items (general government final consumption, social transfers and taxes). 4442 A. Afonso and R. M. Sousa rate swap spreads are used as the dependent revenue and government spending need to adjust variable (Goodhart and Lemmen, 1999; Afonso and accordingly. Second, the debt dynamics may influ- Strauch, 2007). ence interest rates as they depend on future expected For Europe, the existing evidence points either to a monetary policy and the risk premium. Third, debt significant (although small) effect (Bernoth et al., may not be neutral and have an impact on inflation 2003; Codogno et al., 2003; Faini, 2006; Afonso and and output (Barro, 1974; Kormendi, 1983; Canzoneri Strauch, 2007), or to the absence of impact (Heppke- et al., 2001; Afonso, 2008). Therefore, it is important Falk and Hufner, 2004). For the US, the effect seems to allow for the fact that government revenues, to be substantially larger (Gale and Orszag, 2002). government spending, real GDP growth, inflation For OECD countries, Ardagna (2009) shows that and the interest rate are linked by the government long-term government bond rates fall in periods of intertemporal budget constraint. budget consolidation and rise when the fiscal position Fiscal policy is characterized as follows: deteriorates. G ¼ f ð Þþ " ð4Þ t t T ¼ gð Þþ " ð5Þ t t III. Modelling Strategy where G is the government spending, T the govern- t t ment revenue, f and g the linear functions,  the G T The modelling strategy adopted consists in the information set and " and " , respectively, the gov- t t estimation of the following Structural VAR (SVAR) ernment spending shock and the government revenue G T shock. The shocks " and " are orthogonal to the t t ðLÞ X þ  d ¼  X þ  X þ þ  d t i t1 0 t 1 t1 i t1 |ffl{zffl} |{z} elements in  . nn n1 We follow a recursive identification scheme and ¼ c þ " ð1Þ assume that the variables in X can be separated into three groups: (i) a subset of n variables, X , whose 1 1t ð1 þ i Þ contemporaneous values appear in the policy func- d ¼ d þð g  tÞð2Þ t t1 t t ð1 þ  Þð1 þ  Þ t t tion and do not respond contemporaneously to the fiscal policy shocks; (ii) a subset of n variables, X , 2 2t v ¼  " ð3Þ that respond contemporaneously to the fiscal policy t t shocks and whose values appear in the policy where " j X , s5t  N(0,), (L) is a matrix valued t s function only with a lag; and (iii) the policy variables polynomial in positive powers of the lag operator L, in the form of government expenditure, G , and/or n the number of variables in the system, " the government revenue, T . fundamental economic shocks that span the space of The recursive assumptions can be summarized by innovations to X and v the VAR innovation, t t X ¼ [X , G , T , X ] and t lt t t 2t d :¼ B /(P Y ), d :¼ B /(P Y ), g :¼ G /(P Y ) t t t t t1 t1 t1 t1 t t t t 2 3 0 0 and t :¼ T /(P Y ). t t t t 11 |{z} |{z} |{z} n 2 n n Equation 2 refers to the government’s intertem- 6 1 1 2 7 n n 1 1 6 7 poral budget constraint, and i , G ,T ,  , Y , P ,  6 7 t t t t t t t 21 22 6 |{z} 7 |{z} |{z} ¼ ð6Þ 6 7 and d represent, respectively, the interest rate (or the 2n 2n 22 6 1 7 4 5 average cost of debt refinancing), government pri- 31 32 33 |{z} |{z} |{z} mary expenditures and government revenues, infla- n n n n 2 1 n 2 2 2 tion, GDP, price level, real growth rate of GDP and The two upper blocks of zeros correspond, the debt-to-GDP ratio at the beginning of the period t. respectively, to the assumptions that the variables Following Favero and Giavazzi (2008), this speci- fication includes the feedback from government debt, in X do not respond to the fiscal policy shock 1t an assumption that is potentially important in the either directly or indirectly. This approach delivers a determination of the effects of fiscal policy shocks for correct identification of the fiscal policy shock but a number of reasons. First, when fiscal authorities not of the other shocks in the system. In practice, care about the stabilization of debt, government we include in our system the same variables as in Since the fiscal policy shock is identified regardless of the ordering restrictions among the nonpolicy variables, we can put arbitrary O in the nonpolicy blocks of  in order to obtain (n  1)n/2 linearly independent restrictions and to get consistent s 0 impulse-responses (Christiano et al., 1999). As a result, the Choleski decomposition emerges as a particular case, where the restrictions are imposed such that  becomes a lower triangular matrix. 0 The macroeconomic effects of fiscal policy 4443 Christiano et al. (2005), but also add housing price government revenue due to the sale of Universal among the X variables, that is, we allow the policy Mobile Telecommunications System (UMTS) 1t authority to react contemporaneously to changes in licenses. the housing market. We also include the stock Due to limitations of the data, housing starts are market index and the exchange rate in X . included only in the US, and among the set of 2t The identification procedure adopted in this article variables included in the SVAR, the average govern- is closer in spirit to Fata´ s and Mihov (2001), who use ment debt cost servicing deserves a special attention. a Cholesky ordering to capture the unexpected Therefore, we obtain the average implicit interest rate variation in fiscal policy. In contrast, Afonso and by dividing the net interest payments by the govern- Sousa (2011) identify the potential impact of fiscal ment debt at time t  1. policy on asset prices, by using a Fully Simultaneous The quarterly fiscal data refers to the Federal System approach in a Bayesian framework. Government spending and revenue in the case of the Nevertheless, the empirical findings deliver qualita- US, and the Public Sector spending and revenue in tively similar results, which allows one to be confident the case of the UK In both cases, quarterly fiscal data on their robustness. Finally, we assess the posterior is available directly from national accounts. As for uncertainty about the impulse-response functions Germany and Italy, we compute the quarterly series using a Monte Carlo Markov-Chain (MCMC) of government spending and revenue using budgetary algorithm. cash data, which is published monthly by the fiscal authorities of both countries. In this case, data for government spending and revenue refer to the Central Government and are available in a cash basis. IV. Empirical Analysis The data cover the following samples: 1970:Q3– 2007:Q4, in the case of the USA; 1971:Q2–2007:Q4, in Building the data set the case of the UK; 1979:Q2–2006:Q4, in the case of Germany; and 1986:Q2–2004:Q4, in the case of Italy. We use quarterly data for four countries: US, UK, Figures 1–4 provide a comparison of the annual Germany and Italy. All the variables are in natural values of such quarterly fiscal data with the annual logarithms unless stated otherwise. national accounts data provided by the European For the identification of the fiscal policy shocks, Commission (Ameco database). It is interesting to the variables in X – the ones predetermined with 1t observe that the patterns of both series are rather respect to fiscal policy innovations – are GDP, similar. Moreover, in most of the cases, the levels private consumption, GDP deflator and private themselves are also close. investment. To these variables, we add: the housing Finally, Fig. 5 plots the observed government debt- price index (or the median sales price of new houses to-GDP ratio and the implicit debt-to-GDP ratio, sold, in the case of the US), the housing starts (only that is, the one that would emerge by including the for the US), and the average cost of government debt feedback from government debt. It shows that, financing (or the yield to maturity of long-term despite some small discrepancies in the case of Italy, government bonds). The variables in X – the ones 2t the implicit series for the debt-to-GDP ratio tracks allowed to react contemporaneously to fiscal policy pretty well the actual series. shocks – are the S&P500 Index (for the US), the FTSE-A11 Shares Index (for the UK), and the MSCI index (for Germany and Italy). As measure of the Results fiscal policy instruments we use either the government The starting point is the estimation of a B-SVAR that expenditures (in which case, the government revenues does not include the feedback from government debt, are included in X ) or the government revenues (in 1t that is, where Equation 2 is not considered. Then, we which case, the government expenditures are included compare the results with the ones that emerge from in X ). We include a constant (or quarterly seasonal 1t estimating specifications (1–3). dummies), and the government debt-to-GDP ratio in Figures 6–9 show the impulse-response functions to the set of exogenous variables. For Germany, we also consider two dummies: (i) one dummy for 1991:Q1, a fiscal policy shock. The solid line refers to the corresponding to the German reunification; and (ii) median response when the VAR is estimated without another dummy for 2000:Q3, to capture the spike in including the debt feedback, and the dashed lines are, A detailed description is provided in Afonso and Sousa (2011). In this context, Perez (2007) argues that, for some EU countries’, intra-annual cash fiscal information can be used to improve the forecasting and monitoring of the annual General Government deficit in terms of ESA95. 4444 A. Afonso and R. M. Sousa Primary spending Total spending Total revenue 35 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 6 16 40 3 8 30 6 20 4 0 0 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1971 1976 1981 1986 1991 1996 2001 2006 From quarterly data AMECO Average Cost debt Government bond yield AMECO From quarterly data AMECO Fig. 1. Quarterly versus annual fiscal data, US Primary spending Total spending Total revenue 60 50 30 30 30 25 20 20 10 10 0 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 140 16 80 3 2 6 20 2 1963 1968 1973 1978 1983 1988 1993 1998 2003 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data EP3H BKTl AMECO Average Cost Debt Government Bond Yield AMECO From quarterly data AMECO Fig. 2. Quarterly versus annual fiscal data, UK Notes: EP3H – ‘Financial Intermediation Services Indirectly Measured (FISIM) adjusted GDP; BKTL – GDP at current prices. respectively, the median response and the 68% US. Figure 6(a) displays the impulse-response func- posterior probability intervals from the VAR esti- tions of all variables in X to a shock in government mated by imposing the government budget con- spending in the US straint. The confidence bands are constructed using a In the case we do not include the debt feedback, it MCMC algorithm based on 50 000 draws. can be seen that the effects on GDP are positive, % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP The macroeconomic effects of fiscal policy 4445 Primary spending Total spending Total revenue 50 52 42 44 40 43 38 42 42 41 1979 1984 1989 1994 1999 2004 1979 1984 1989 1994 1999 2004 1979 1984 1989 1994 1999 2004 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 80 4 70 3,5 60 3 50 2,5 2 6 30 1,5 0,5 0 0 1970 1975 1980 1985 1990 1995 2000 2005 1974 1979 1984 1989 1994 1999 2004 1974 1979 1984 1989 1994 1999 2004 From quarterly data AMECO From quarterly data AMECO Average Cost Debt Government Bond Yield AMECO Fig. 3. Quarterly versus annual fiscal data, Germany Primary spending Total spending Total revenue 50 50 40 40 30 30 20 20 10 10 1976 1981 1986 1991 1996 2001 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Long-term interest rate Government debt Interest payments on debt 140 18 100 10 80 8 60 6 40 4 20 2 0 0 0 1985 1990 1995 2000 1984 1989 1994 1999 2004 1982 1987 1992 1997 2002 From quarterly data AMECO From quarterly data AMECO Average Cost Debt Government Bond Yield AMECO Fig. 4. Quarterly versus annual fiscal data, Italy sizeable and reach a peak at after eight quarters. The ‘Keynesian’ effect in the economy and that there is impact on private consumption and private invest- no ‘crowding-out’. In addition, there is a positive ment is also positive, therefore, supporting the idea effect on the average cost of debt and the price level. that government spending has an expansionary In what concerns the reaction of asset markets, the Fata´ s and Mihov (2001), when looking at the response to changes in different components of government expenditures, find that increases in government consumption are always expansionary, while increases in public investment do not have a significant impact on output. Here, we cannot analyse some of these policy experiments because we have not made explicit the role of different components of government spending. % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % % % GDP % GDP 4446 A. Afonso and R. M. Sousa US UK Debt ratio Debt ratio 50 100 40 80 20 40 10 20 Implicit debt ratio Observed debt ratio Observed debt ratio Implicit debt ratio Germany Italy Debt ratio Debt ratio 0 0 Implicit debt ratio Observed debt ratio Implicit debt ratio Observed debt ratio Fig. 5. Implicit debt ratio and observed debt ratio, percentage of GDP Note: Implicit debt ratio computed via Equation 2. empirical evidence suggests that while there is a GDP are negative, very persistent and the trough is positive effect on housing prices that persists for reached at after 12 quarters. They also reveal a almost 20 quarters, the reaction of stock prices is change in the composition of GDP’s major compo- rather small and negative. nents: while private consumption is negatively When we include the debt dynamics in the model, impacted by a positive shock in government revenues, the effects of a government spending shock on GDP the effect on private investment is positive but quickly become somewhat smaller. The effects on private erodes after eight quarters. That is, in this case, the consumption and private investment also remain ‘crowding-out’ effect works mainly through the positive although smaller. In addition, the effect on consumption channel. In what concerns the reaction the average cost of refinancing the debt is negligible. of asset markets, the empirical evidence suggests that The reaction of asset markets is similar to the model the effects of revenue shocks tend to be negative for without the government constraint: there is a positive housing prices and slightly positive for stock prices. (although smaller) and persistent effect on housing UK. The impulse-response functions to a shock in prices, while the reaction of stock prices is still small due to the debt dynamics and the portfolio government spending in the UK are shown in reallocation. Fig. 7(a). Figure 6(b) shows the impulse-response functions The results show that an increase in government to a shock in government revenue. The results suggest spending has a positive effect on GDP in accordance that government revenue declines steadily following with the predictions of the standard IS-LM model. the shock which erodes after eight quarters. Contrary As for the US, there is evidence of a change in its to a shock in government spending, the effects on composition: private consumption increases but the 1966Q4 1970Q2 1968Q3 1972Q1 1970Q2 1973Q4 1972Q1 1975Q3 1973Q4 1977Q2 1975Q3 1979Q1 1977Q2 1980Q4 1979Q1 1982Q3 1980Q4 1984Q2 1982Q3 1986Q1 1984Q2 1987Q4 1986Q1 1989Q3 1987Q4 1989Q3 1991Q2 1991Q2 1993Q1 1993Q1 1994Q4 1994Q4 1996Q3 1996Q3 1998Q2 1998Q2 2000Q1 2000Q1 2001Q4 2001Q4 2003Q3 2003Q3 2005Q2 2005Q2 2007Q1 2007Q1 1985Q4 1963Q4 1965Q4 1986Q4 1967Q4 1987Q4 1969Q4 1988Q4 1971Q4 1989Q4 1973Q4 1990Q4 1975Q4 1991Q4 1977Q4 1979Q4 1992Q4 1981Q4 1993Q4 1983Q4 1994Q4 1985Q4 1995Q4 1987Q4 1996Q4 1989Q4 1997Q4 1991Q4 1998Q4 1993Q4 1995Q4 1999Q4 1997Q4 2000Q4 1999Q4 2001Q4 2001Q4 2002Q4 2003Q4 2003Q4 2005Q4 2004Q4 2007Q4 The macroeconomic effects of fiscal policy 4447 Fig. 6. Impulse-response functions, US: (a) Spending shock and (b) Revenue shock 4448 A. Afonso and R. M. Sousa Fig. 7. Impulse-response functions, UK: (a) Spending shock and (b) Revenue shock The macroeconomic effects of fiscal policy 4449 Fig. 8. Impulse-response functions, Germany: (a) Spending shock and (b) Revenue shock 4450 A. Afonso and R. M. Sousa Fig. 9. Impulse-response functions, Italy: (a) Spending shock and (b) Revenue shock The macroeconomic effects of fiscal policy 4451 effects on private investment are negative, suggesting These findings do not change significantly when the idea of a ‘crowding-out’ effect via investment (see, the debt dynamics is included in the model. e.g. Blanchard and Perotti (2002)). In what concerns Figure 8(b) plots the impulse-response functions to the reaction of asset markets, the empirical evidence a shock in government revenue. Similar to the US, the suggests that both housing and stock prices fall after results show that government revenue declines the spending shock but the effect on housing markets quickly after the shock, eroding after two quarters. is much more persistent. That is, markets seem to The effects on GDP are positive and support the idea interpret the expansion of government spending as of a ‘crowding-in’ effect as both private consumption signalling a deterioration of public finances and private investment react positively to the shock. (Ardagna, 2009). In contrast, the effects on the Revenue shocks tend to be significant and positive price level and on the average cost of financing the only for housing prices. debt are positive. When the debt dynamics is included in the model, the impulse-response functions do not Italy. We look at the effects of a fiscal policy shock change significantly. in Italy. Figure 9(a) displays the impulse-response Figure 7(b) shows the impulse-response functions functions to a shock in government spending. Despite to a shock in government revenue. They support the a very small positive effect in the first quarters, GDP, idea of a ‘crowding-in’ effect: private investment private consumption and private investment fall, reacts positively to the shock; private consumption suggesting a ‘crowding-out’ effect and a ‘Non- falls in the first eight quarters after the shock but then Keynesian’ pattern. Government spending shocks recovers and the effect becomes positive. The effects have a positive and persistent effect on both the price on housing prices are also positive, but the reaction level and the average cost of financing the debt, in takes place with a lag of around eight quarters. By accordance with the findings of Giordano et al. their turn, stock prices react positively to the shock (2007). Housing prices seem to be positively affected but the effect reaches a peak at around after eight by the shock, while stock prices fall immediately after quarters. Finally, the effects on the price level and the the shock but later start recovering. interest rate are negative and the impact is maximum When we take into account the feedback from at after 8 to 12 quarters. government debt, the results suggest that the effects When we include the feedback from the on GDP, private consumption, private investment, government debt in the model, the effects on the the price level and the interest rate become smaller. majority of the variables included in the model This gives rise to the importance of considering the become smaller. debt dynamics in the model. Figure 9(b) shows the impulse-response functions Germany. We now look at the impulse-response to a shock in government revenue. The effects on functions a shock in government spending for GDP, private consumption and private investment Germany (Fig. 8(a)). The effects on GDP are are negative in the first quarters, but the impacts are negative, reflecting the fall in both private consump- not persistent as they vanish after four to six tion and private investment, and providing evidence quarters and even become positive after that. of a ‘Non-Keynesian’ effect (see, for instance, Looking at the reaction of asset markets, one can Giavazzi and Pagano (1990, 1996), Giudice et al. see that the effects of government revenue shocks (2004) and Afonso (2010)). There is, therefore, tend to be positive for stock prices and negative for evidence of a ‘crowding-out’ effect. Nevertheless, housing prices. This suggests that while the credit- the magnitude of the impact is small and, in this channel (i.e. the fall in interest rates) impacts respect, it corroborates the findings of Heppke-Falk positively in stock markets, for housing markets et al. (2006). Government spending shocks also have that channel is annihilated by the ‘crowding-out’ a negative and persistent effect on the price level and effects. Moreover, it is in accordance with the the average cost of financing debt. Finally, housing argument that the effects of fiscal consolidation on prices seem to increase after the shock while stock stock prices are particularly strong in countries prices fall immediately after the shock, possibly, which typically run higher levels of government reflecting the deterioration of fiscal policy. deficit (Ardagna, 2009). Afonso and Claeys (2008) mention that large revenue reductions unmatched by expenditure cuts have pushed the deficit beyond the 3% threshold in Germany in 2002, putting the country in an excessive deficit situation within the EU fiscal framework. 4452 A. Afonso and R. M. Sousa Table 1. The effect of (d  d*) in a VAR t1 US T G Germany T G t t t t (d  d*) 1970:Q3 to 2007:Q4 0.187*** 0.180*** (d  d*) 1980:Q3 to 2006:Q4 0.211*** 0.021 t1 t1 (N ¼ 150) (0.068) (0.050) (N ¼ 106) (0.112) (0.064) 1970:Q3 to 1987:Q4 0.352 0.176 1980:Q3 to 1990:4 0.122 1.087*** (N ¼ 70) (0.261) (0.156) (N ¼ 42) (0.410) (0.271) 1988:1 to 2007:4 0.203 0.228 1991:Q1 –2006:Q4 0.017 0.075 (N ¼ 80) (0.158) (0.154) (N ¼ 64) (0.192) (0.060) UK T G Italy T G t t t t (d  d*) 1964:2 to 2007:4 0.120*** 0.065** (d  d*) 1986:Q2 to 2004:Q4 0.081 4.125** t1 t1 (N ¼ 175) (0.037) (0.030) (N ¼ 75) (0.476) (1.741) 1964:2 to 1984:4 0.013 0.102 1986:Q2 to 1994:Q4 1.344 3.963 (N ¼ 83) (0.109) (0.089) (N ¼ 35) (1.529) (8.288) 1985:1 to 2007:4 0.143** 0.043 1995:Q1 to 2004:Q4 4.353** 6.642 (N ¼ 92) (0.127) (0.104) (N ¼ 40) (1.838) (6.518) Notes: SEs in brackets. *, ** and *** denote statistical significanceat the 10, 5 and 1% levels, respectively. Fiscal shocks and government debt feedback debt-to-GDP ratio from its sample average only for the full sample. Indeed, one sees that when the debt- We now consider the potential debt feedback and to-GDP ratio is above its historical mean, govern- estimate the following structural VAR: ment primary spending decreases (the coefficient associated to (d  d*) in the government spending X þ  X þ  þ  ðd  d Þ¼ c þ " ð7Þ 0 t 1 t1 i t1 t t1 equation is negative and significant (0.180), while in the government revenue equation is positive and ð1 þ i Þ d ¼ d þð g  tÞð8Þ t t1 t t significant (0.187)). ð1 þ  Þð1 þ  Þ t t For the UK, the results are similar. They show a The specification of Equation 7 is suggested by the stabilizing effect of the debt level on the primary empirical findings in Bohn (1998), who estimates a budget balance that works through the response of fiscal reaction function in which d* is the uncondi- both government revenue and government spending tional mean of the debt ratio and allow us to take into to deviations of the debt from the target level. This is, account the debt feedback. Following Bohn (1998) particularly, the case of the full sample. It is also the and Favero and Giavazzi (2008), we model the target case of the second sub-sample (1985:Q1 to 2007:Q4) level of the debt as a constant on the basis of the but only for government revenue. Therefore, when evidence of stationarity of d. In addition, we impose the debt ratio is above its sample mean, it is possible the government’s intertemporal budget constraint as to observe an increase in government revenue (0.120, described by Equation 8. in the full sample; 0.143, for the second sub-sample). Table 1 reports the estimated coefficients on In the case of Germany, there is evidence of a small (d  d*) in the structural equations of the SVAR t1 stabilizing effect of the debt level on the primary (government spending and government revenue). We surplus that works through the government revenue report the coefficients (and the SEs in brackets) taken (0.211). In addition, the results of the estimation of from the estimation for the full sample and the sub- the model in the first sub-sample (1980:Q3 to samples. In the case of the US, we consider two sub- 1990:Q4) suggest that government spending tends to samples: 1970:Q3 to 1987:Q4, corresponding to the decrease (1.087) when the debt-to-GDP ratio is pre-Greenspan era; and 1988:Q1 to 2007:Q4, after above its sample mean. Greenspan. In the case of the Germany, we also split Finally, for Italy, one can see that when the debt the sample in two periods: 1980:Q3 to 1990:Q4, that ratio is above average, government spending strongly is, before the reunification; and 1991:Q1 to 2006:Q4, falls (4.125). Additionally, in the period 1995:Q1 to that is, after the reunification. For the UK and Italy, 2004:Q4, there is evidence supporting a stabilizing each sub-sample is built by splitting the entire sample effect of government debt on the primary surplus that into roughly two sub-samples of similar size. works through the government revenue: the coeffi- For the US, the results show a significant response cient associated to (d  d*) is positive and large in t1 of revenue and primary spending to deviations of the magnitude (4.353). This is in line with the increase of The macroeconomic effects of fiscal policy 4453 fiscal policy discipline imposed by the Maastricht comments and suggestions, to Boris Hofmann, Treaty. Bernhard Manzke and Sandro Momigliano, for data clarifications and to Silvia Albrizio and Matthijs Lof for research assistance. The opinions expressed herein are those of the authors and do not necessarily reflect those of the ECB or the V. Conclusion Eurosystem. UECE is supported by FCT (Fundac¸ a˜ o para a This article provides a detailed evaluation of the Cieˆ ncia e a Tecnologia, Portugal), financed by ERDF macroeconomic effects of fiscal policy. It, therefore, and Portuguese funds. also deals with a relevant policy question, in partic- Ricardo Sousa would like to thank the Fiscal ular, given the expansionary fiscal policies adopted by Policies Division of the ECB for its hospitality. governments in industrialized countries in the attempt to stabilize the economies after the burst of the financial crisis in 2008. We identify fiscal policy shocks using a recursive partial identification, and estimate a B-SVAR, there- References fore, accounting for the posterior uncertainty of the Afonso, A. (2008) Ricardian fiscal regimes in the European impulse-response functions. In addition, we explicitly Union, Empirica, 35, 313–34. include the feedback from government debt in our Afonso, A. (2010) Expansionary fiscal consolidations in framework. Europe: new evidence, Applied Economics Letters, 17, 105–9. The empirical evidence suggests that government Afonso, A. and Claeys, P. (2008) The dynamic behaviour of spending shocks: (i) have, in general, a small but budget components and output, Economic Modelling, positive effect on GDP; (ii) have a varied effect on 25, 93–117. private consumption and on private investment; (iii) Afonso, A. and Sousa, R. M. (2011) What are the effects of have a positive effect on housing prices; (iv) lead to a fiscal policy on asset markets?, Economic Modelling, quick fall in stock prices; and (v) in general, impact 28, 1871–90. Afonso, A. and Strauch, R. (2007) Fiscal policy events and positively on the price level and the average cost of interest rate swap spreads: some Evidence from the refinancing the debt. In addition, government reve- EU, Journal of International Financial Markets, nue shocks: (i) have a positive effect on GDP and Institutions and Money, 17, 261–76. private investment; (ii) have a varied effect on private Alesina, A. and Ardagna, S. (1998) Tales of fiscal adjust- consumption; (iii) a varied impact on housing prices ment, Economic Policy, 27, 489–545. Ardagna, S. (2009) Financial markets’ behavior around and a positive effect on stock prices; and (iv) in episodes of large changes in the fiscal stance, European general, do not impact on the price level, but have a Economic Review, 53, 37–55. mixed effect on the interest rate. Barro, R. J. (1974) Are government bonds net wealth?, When the debt dynamics is explicitly taken into Journal of Political Economy, 82, 1095–117. account, (long-term) interest rates and GDP become Bernanke, B. and Kuttner, K. (2005) What explains the stock market’s reaction to Federal Reserve policy?, more responsive and the effects of fiscal policy on Journal of Finance, 60, 1221–57. these variables also become more persistent. Bernoth, K. J., Von Hagen, J. and Schuknecht, L. (2003), Finally, the results support the existence of a The determinants of the yield differential in EU stabilizing effect of the debt level on the primary government bond market, Working Paper, ZEI – budget balance that works through both government Center for European Integration Studies. Biau, O. and Girard, E. (2005) Politique budgetaire et spending and revenue in the case of the US and the dynamique economique en France: l’approche VAR UK, through government revenue in the case of structurel, Economie et Pre´vision, 169–171, 1–24. Germany, and through government spending in the Blanchard, O. and Perotti, R. (2002) An empirical case of Italy. This signals that governments tend to characterization of the dynamic effects of changes in adjust their fiscal imbalances in response to govern- government spending and taxes on output, Quarterly ment debt developments. Journal of Economics, 117, 1329–68. Bohn, H. (1998) The Behaviour of US public debt and deficits, Quarterly Journal of Economics, 113, 949–63. Brook, A.-M. (2003) Recent and prospective trends in real long-term interest: fiscal policy and other drivers, OECD Working Paper No. 21, Organization for Acknowledgements Economic Co-operation and Development. We are grateful to Marten Blix, Carlo Favero, Jiirgen Canzoneri, M., Cumby, R. and Diba, B. (2001) Is the price Von Hagen, Ad van Riet and to participants to a level determined by the needs of fiscal solvency, European Central Bank (ECB) seminar for helpful American Economic Review, 91, 1221–38. 4454 A. Afonso and R. M. Sousa Castro, F. and Cos, P. H. (2008) The economic effects of Goodhart, C. and Lemmen, J. J. G. (1999) Credit risks and fiscal policy: the case of Spain, Journal of European government bond markets: a panel data Macroeconomics, 30, 1005–28. econometric analysis, Eastern Economic Journal, 25, Chirinko, R., de Haan, L. and Sterken, E. (2004) Asset 77–107. price shocks, real expenditures, and financial structure: Heppke-Falk, K.H. and Hu¨ fner, F. (2004) Expected budget a multi-country analysis, Emory University, Working deficits and interest rate swap spreads – evidence for Paper. France, Germany and Italy, Discussion Paper No. 40, Christiano, L. J., Eichenbaum, M. and Evans, C. L. (1999) Deutsche Bundesbank. Monetary policy shocks: what have we learned and to Heppke-Falk, K.H., Tenhofen, J. and Wolff, G. B. (2006) what end? in Handbook of Macroeconomics, 1A (Eds) The macroeconomic effects of exogenous fiscal J. Taylor and M. Woodford, Elsevier, Amsterdam, policy shocks in Germany: a disaggregated SVAR pp. 65–148. analysis, Discussion Paper No. 41, Deutsche Christiano, L. J., Eichenbaum, M. and Evans, C. L. (2005) Bundesbank. Nominal rigidities and the dynamic effects of a shock Iacoviello, M. (2005) House prices, borrowing constraints, to monetary policy, Journal of Political Economy, 113, and monetary policy in the business cycle, American 1–45. Economic Review, 95, 739–64. Codogno, L., Favero, C. and Missale, A. (2003) Yield Iacoviello, M. and Minetti, R. (2003) The credit channel of spreads on EMU government bonds, Economic Policy, monetary policy: evidence from the housing market, 37, 503–32. Working Paper, Boston College. Craine, R. and Martin, V. (2003) Monetary policy shocks Jappelli, T. and Pistaferri, L. (2007) Do people respond to and security market responses, Working Paper, tax incentives? An analysis of the Italian reform of the University of California at Berkeley. deductibility of home mortgage interests, European Edelberg, W., Eichenbaum, M. and Fisher, J. (1999) Economic Review, 51, 247–71. Understanding the effects of a shock to government Julliard, C., Michaelides, A. and Sousa, R. M. (2008) purchases, Review of Economics Dynamics, 2, 166–206. Housing prices and monetary policy, Working Engen, E. M. and Hubbard, R. G. (2004) Federal govern- Paper, London School of Economics and Political ment debt and interest rates, NBER Working Paper Science. No. 10681, National Bureau of Economic Research. Kormendi, R. C. (1983) Government debt, government Faini, R. (2006) Fiscal policy and interest rates in Europe, spending, and private sector behavior, American Economic Policy, 21, 443–89. Economic Review, 73, 994–1010. Fata´ s, A. and Mihov, I. (2001) The effects of fiscal policy Laubach, T. (2009) New evidence on the interest rate effects on consumption and employment: theory and evi- of budget deficits and debt, Journal oj the European dence, CEPR Discussion Paper No. 2760, Centre for Economic Association, 7, 1–28. Economic Policy Research. Maclennan, D., Muellbauer, J. and Stephens, M. (1999) Favero, C. (2002) How do European monetary and fiscal Asymmetries in Housing and Financial Market authorities behave?, CEPR Discussion Paper No. 3426, Institutions and EMU, CEPR Discussion Paper No. Centre for Economic Policy Research. 2062, Centre for Economic Policy Research. Favero, C. and Giavazzi, F. (2008) Debt and the effects of McCarthy, J. and Peach, R. (2002) Monetary policy fiscal policy, University of Bocconi, Working Paper. transmission to residential investment, Federal Gale, W. G. and Orszag, P. R. (2002) The economic effects Reserve Bank of ‘New York Economic Policy Review, of long-term fiscal discipline, Discussion Paper, 8, 139–58. Urban-Brookings Tax Policy Center. Mountford, A. and Uhlig, H. (2005) What are the effects of Gale, W. G. and Orszag, P. R. (2003) Economic effects of fiscal policy shocks?, Humboldt-Universita¨ t zu Berlin sustained budget deficits, National Tax Journal, 56, Working Paper SFB No. 649. 463–85. Pe´ rez, J. (2007) Leading indicators for euro area govern- Galı´, J., Lo´ pez-Salido, J. D. and Valle´ s, J. (2007) ment deficits, International Journal of Forecasting, 23, Understanding the effects of government spending on 259–75. consumption, Journal of the European Economic Perotti, R. (1999) Fiscal policy in good times and bad, Association, 5, 227–70. Quarterly Journal of Economics, 114, 1399–436. Giavazzi, F. and Pagano, M. (1990) Can severe fiscal Perotti, R. (2004) Estimating the effects of fiscal policy in contractions be expansionary? Tales of two small OECD countries, Working Paper, University of european countries, in NBER Macroeconomics Annual Bocconi. (Eds) O. J. Blanchard and S. Fischer, MIT Press, Ramey, V. and Shapiro, M. (1998) Costly capital realloca- Cambridge, MA, pp. 75–110. tion and the effects of government spending, Carnegie Giavazzi, F. and Pagano, M. (1996) Non-Keynesian effects Rochester Conference on Public Policy, 48, 145–94. of fiscal policy changes: international evidence and the Rigobon, R. and Sack, B. (2002) The impact of monetary Swedish experience, Swedish Economic Policy Review, policy on asset prices, Finance and Economics 3, 67–103. Discussion Series 2002–4, Board of Governors of the Giordano, R., Momigliano, S., Neri, S. and Perotti, R. Federal Reserve System. (2007) The effects of fiscal policy in Italy: evidence Rigobon, R. and Sack, B. (2003) Measuring the reaction of from a VAR model, European Journal of Political Economy, 23, 707–33. monetary policy to the stock market, Quarterly Journal Giudice, G., Turrini, A. and in’t Veld, J. (2004) Non- of Economics, 118, 639–670. Keynesian fiscal consolidation in the EU? Ex post Upper, C. and Worms, A. (2003) Real long-term interest evidence and ex ante analysis, CEPR Discussion Paper rates and monetary policy: a cross-country perspective, No. 4388, Centre for Economic Policy Research. BIS Papers, 19, 234–257. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Economics Taylor & Francis

The macroeconomic effects of fiscal policy

Afonso, António; Sousa, Ricardo M.
Applied Economics , Volume 44 (34): 16 – Dec 1, 2012
Download PDF
16 pages

Loading next page...
 
/lp/taylor-francis/the-macroeconomic-effects-of-fiscal-policy-i2q5BOUba0

References (131)

Publisher
Taylor & Francis
Copyright
Copyright 2012 The Author(s). Published by Taylor & Francis.
ISSN
1466-4283
eISSN
9999-7004
DOI
10.1080/00036846.2011.591732
Publisher site
See Article on Publisher Site

Abstract

Applied Economics, 2012, 44, 4439–4454 The macroeconomic effects of fiscal policy a,b, c,d Anto´ nio Afonso * and Ricardo M. Sousa Directorate General Economics, European Central Bank, Kaiserstraße 29, Frankfurt am Main D-60311, Germany Department of Economics, ISEG/TULisbon, Technical University of Lisbon, UECE – Research Unit on Complexity and Economics, R. Miguel Lupi 20, Lisbon 1249-078, Portugal Department of Economics and Economic Policies Research Unit (NIPE), University of Minho, Campus of Gualtar, Braga 4710-057, Portugal Financial Markets Group (FMG), London School of Economics, Houghton Street, London WC2 2AE, UK We investigate the macroeconomic effects of fiscal policy using a Bayesian Structural Vector Autoregression (B-SVAR) approach. We identify fiscal policy shocks via a partial identification scheme, but also: (i) include the feedback from government debt; (ii) look at the impact on the composition of output; (iii) assess the effects on asset markets; (iv) use quarterly data; and (v) analyse empirical evidence from the US, the UK, Germany and Italy. The results show that government spending shocks, in general, have a small effect on Gross Domestic Product (GDP); lead to important ‘crowding-out’ effects; have a varied impact on housing prices and generate a quick fall in stock prices. Government revenue shocks generate a mixed effect on housing prices and a small and positive effect on stock prices. The empirical evidence also suggests that it is important to explicitly consider the government debt dynamics in the model. Keywords: fiscal policy; Bayesian structural VAR; debt dynamics JEL Classification: C11; C32; E62; H62 I. Introduction fiscal policy committees, the influence of regulation in the structure of market incentives, and the Balanced Compared to the large empirical literature on the Budget Amendment in the US are based on the effects of monetary policy on economic activity, fiscal assumption that fiscal policy can be an effective tool policy has received less attention, a feature that for stabilizing business cycles. contrasts with the public debates on its role. The First, we consider the effects of fiscal policy on the government deficit and debt limits of the Stability composition of Gross Domestic Product (GDP), and Growth Pact in the context of the Economic and namely, by estimating the impact of government Monetary Union (EMU), the possibility of indepen- spending and government revenue shocks on pri- dent institutions running fiscal policy, the creation of vate consumption and private investment as in *Corresponding author. E-mail: [email protected]; [email protected] Applied Economics ISSN 0003–6846 print/ISSN 1466–4283 online  2012 The Author(s). Published by Taylor & Francis. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the 4439 named author(s) have been asserted. http://www.tandfonline.com http://dx.doi.org/10.1080/00036846.2011.591732 4440 A. Afonso and R. M. Sousa Galı et al. (2007). Consequently, we are able to results of Giavazzi and Pagano (1990, 1996), Giudice et al. (2004) and Afonso (2010). identify the potential ‘crowding-out’ effects of fiscal Interestingly, when one looks at the response of policy on the private sector. asset prices, the findings suggest that markets tend to Second, we also ask how asset markets (via not interpret expansionary fiscal policies as leading to a only stock prices but also housing prices) are affected deterioration of public finances. Therefore, stock by fiscal policy shocks. Sounder fiscal positions, in prices react negatively to a rise of government particular when associated with a permanent fall in spending and positively when there is fiscal consol- government debt, can have a strong impact on both idation. Nevertheless, while the reaction of stock stock market prices and long-term government bond prices is relatively quick, the effect on housing prices rates (Ardagna, 2009). In addition, fiscal policy can is, in general, persistent. have a major influence on housing markets, namely, When we explicitly take into account the feedback via specific subsidies, other tax measures and its from government debt in our framework, the effects effects on household’s disposable income. Therefore, of fiscal policy on (long-term) interest rates and GDP we look at the persistence of the effects of fiscal become more persistent and these variables are also shocks and assess whether the reaction of stock and more responsive to the shock. housing prices is quantitatively similar or exhibits The rest of this article is organized as follows. asymmetry instead. Section II reviews the related literature. Section III We identify fiscal policy shocks using a recursive explains the empirical strategy used to identify the partial identification based on the work of Christiano effects of fiscal policy shocks, and to take into et al. (2005), and estimate a Bayesian Structural account the uncertainty regarding the posterior Vector Autoregression (B-SVAR) model, therefore, distribution impulse-response functions. Section IV accounting for the posterior uncertainty of the provides the empirical analysis and discusses the impulse-response functions. results. Section V concludes this article with the main Another novelty of this article is that we explicitly findings and policy implications. include the feedback from government debt in our estimations. In fact, while equilibrium structural models are normally solved by imposing the govern- ment’s intertemporal budget constraint, this does not II. Literature happen with Vector Authogression (VAR)-based fiscal policy models. We deal with this limitation by For the US, different approaches have been used in following Favero and Giavazzi (2008), and, as a the identification of the fiscal policy shock. Ramey result, we consider the response of fiscal variables to and Shapiro (1998) use a ‘narrative approach’ to the level of the government debt. isolate political events, and find that, after a brief rise In addition, an important contribution of this in government spending, nondurable consumption article is the use of quarterly fiscal data, which allows displays a small decline while durables consumption us to identify more precisely the effects of fiscal falls. Following the same approach, Edelberg et al. policies. We analyse empirical evidence from the US, (1999) show that episodes of military build-ups have a the UK, Germany and Italy, respectively, for the significant and positive short-run effect on US output periods 1970:Q3 to 2007:Q4, 1964:Q2 to 2007:Q4, and consumption, and that the sign of the response 1980:Q3 to 2006:Q4 and 1986:Q2 to 2004:Q4. The set does not change when anticipation effects are taken of quarterly fiscal data, is taken from national into account. Fatas and Mihov (2001) and Favero accounts (in the case of the US and the UK) or (2002) use a Cholesky ordering to identify fiscal computed by drawing on the higher frequency shocks and show that increases in government (monthly) availability of fiscal cash data (for expenditures are expansionary, but lead to an Germany and Italy). To the best of our knowledge, increase in private investment that more than com- such fiscal data set, apart from not having been pensates for the fall in private consumption. compiled before, has not yet been used in this strand Blanchard and Perotti (2002) use information about of economic modelling. the elasticity of fiscal variables to identify the Broadly speaking, the results point to an expan- automatic response of fiscal policy, and find that sionary effect of fiscal policy in the case of the US and expansionary fiscal shocks increase output, have a the UK in line with the traditional ‘Keynesian’ model positive effect on private consumption and a negative and corroborated by Blanchard and Perotti (2002). impact on private investment. More recently, using As for Italy and Germany, there is some evidence of a sign restrictions on the impulse-response functions ‘Non-Keynesian’ multiplier in accordance with the and identifying the unexpected variation in The macroeconomic effects of fiscal policy 4441 government spending by a positive response of On the other hand, and as mentioned for Italy by expenditure for up to four quarters after the shock, Jappelli and Pistaferri (2007), tax deductibility of Mountford and Uhlig (2005) find a negative effect in interest rates may not have affected much the demand residential and nonresidential investment. for mortgage debt. In addition, sounder fiscal posi- Regarding other countries, Perotti (2004) investi- tions and lower sovereign financing needs allow for gates the effects of fiscal policy in Australia, Canada, lower interest and better financing conditions for West Germany, US and the UK, and finds a mortgage-loans, while higher government indebted- relatively large positive effect on private consumption ness can crowd-out resources available to would be and no response of private investment. Biau and home-owners (Maclennan et al., 1999). Girard (2005) find a cumulative multiplier of gov- In the case of stock prices the attention has been ernment spending larger than one, and positive normally targeted towards the role played by mon- reactions of private consumption and private invest- etary policy. Rigobon and Sack (2002, 2003) and ment in France. For Spain, Castro and Cos (2008) Craine and Martin (2003) use a heteroscedasticity- report that, while there is a positive relationship based estimator and find a significant response of the between government expenditure and output in the stock market to shocks in the interest. Bernanke and short-term, in the medium and long-term expansion- Kuttner (2005) show that a hypothetical unantici- ary spending shocks only lead to higher inflation and pated 25-basis-point cut in the Federal funds rate lower output. Heppke-Falk et al. (2006) use cash data target is associated with about a 1% increase in broad for Germany, and find that a positive shock in stock indexes. More recently, Ardagna (2009) reports government spending increases output and private that fiscal adjustments based on expenditure reduc- consumption, although the effect is relatively small. tion and signalling sounder fiscal behaviour are Giordano et al. (2007) show that, in Italy, govern- related with increases in stock market prices. Using ment expenditure has positive and persistent effects a panel of Organization for Economic Co-operation on output and on private consumption. and Development (OECD) countries, the author also As for the empirical importance of housing over shows that fiscal consolidation that lead to a perma- the business cycle, there are only a small number of nent and substantial fall in government debt are papers that discuss the empirical link between linked to a stronger increase in stock market prices. economic policy and housing prices, the focus has In terms of interest rates, according to Gale and mainly been on the effects of monetary policy. Some Orszag (2003) there are two important reasons as to examples are McCarthy and Peach (2002) and why budget deficits may raise nominal interest rates: Chirinko et al. (2004). Iacoviello and Minetti (2003) (i) they reduce aggregate savings when private savings emphasize the housing market as creating a credit do not increase by the same amount (no Ricardian channel for monetary policy. Iacoviello (2005) looks equivalence) and if there are no compensating foreign at the monetary policy-house price to consumption capital inflows, which leads to a decrease in the channel and finds a significant effect on house prices. supply of capital; and (ii) they increase the stock of Julliard et al. (2008) suggest that monetary policy government debt and, consequently, the outstanding contractions have a large and significantly negative amount of government bonds (relative to other impact on real housing prices, but the reaction is financial assets). In this case, there is a ‘portfolio extremely slow. On the other hand, monetary policy effect’, as a higher interest rate on government bonds shocks do not seem to cause a significant impact on would be required in order to incentive investors to stock markets. hold the additional bonds. In what concerns fiscal policy, it can impinge on While some studies find that interest rates tend to housing market developments notably via subsidies increase after a rise in the deficit, others do not (Engen and tax measures: taxation of the imputed rental and Hubbard, 2004). The empirical findings seem to value of the house, tax deductibility of interest depend on whether expected or current budget deficits payments, capital taxes on housing gains and Value are used as explanatory variables (Brook, 2003; Added Tax (VAT) on new houses. Given short-run Upper and Worms, 2003; Laubach, 2009), and also inelastic housing supply, fiscal subsidies for buying on whether yield differentials in Europe with houses may end up pushing up its demand and prices. respect to Germany (Codogno et al., 2003) or interest Giavazzi and Pagano (1990) and Alesina and Ardagna (1998) have uncovered the presence of ‘non-Keynesian effects’ (i.e. negative spending multipliers) during large fiscal consolidations, with output rising significantly despite large cuts in government spending. Perotti (1999) also obtains such findings, but only in circumstances of ‘fiscal stress’ (unusually high debt-to-GDP ratios). In addition, Afonso (2010) finds some evidence of expansionary fiscal consolidations, for a few budgetary items (general government final consumption, social transfers and taxes). 4442 A. Afonso and R. M. Sousa rate swap spreads are used as the dependent revenue and government spending need to adjust variable (Goodhart and Lemmen, 1999; Afonso and accordingly. Second, the debt dynamics may influ- Strauch, 2007). ence interest rates as they depend on future expected For Europe, the existing evidence points either to a monetary policy and the risk premium. Third, debt significant (although small) effect (Bernoth et al., may not be neutral and have an impact on inflation 2003; Codogno et al., 2003; Faini, 2006; Afonso and and output (Barro, 1974; Kormendi, 1983; Canzoneri Strauch, 2007), or to the absence of impact (Heppke- et al., 2001; Afonso, 2008). Therefore, it is important Falk and Hufner, 2004). For the US, the effect seems to allow for the fact that government revenues, to be substantially larger (Gale and Orszag, 2002). government spending, real GDP growth, inflation For OECD countries, Ardagna (2009) shows that and the interest rate are linked by the government long-term government bond rates fall in periods of intertemporal budget constraint. budget consolidation and rise when the fiscal position Fiscal policy is characterized as follows: deteriorates. G ¼ f ð Þþ " ð4Þ t t T ¼ gð Þþ " ð5Þ t t III. Modelling Strategy where G is the government spending, T the govern- t t ment revenue, f and g the linear functions,  the G T The modelling strategy adopted consists in the information set and " and " , respectively, the gov- t t estimation of the following Structural VAR (SVAR) ernment spending shock and the government revenue G T shock. The shocks " and " are orthogonal to the t t ðLÞ X þ  d ¼  X þ  X þ þ  d t i t1 0 t 1 t1 i t1 |ffl{zffl} |{z} elements in  . nn n1 We follow a recursive identification scheme and ¼ c þ " ð1Þ assume that the variables in X can be separated into three groups: (i) a subset of n variables, X , whose 1 1t ð1 þ i Þ contemporaneous values appear in the policy func- d ¼ d þð g  tÞð2Þ t t1 t t ð1 þ  Þð1 þ  Þ t t tion and do not respond contemporaneously to the fiscal policy shocks; (ii) a subset of n variables, X , 2 2t v ¼  " ð3Þ that respond contemporaneously to the fiscal policy t t shocks and whose values appear in the policy where " j X , s5t  N(0,), (L) is a matrix valued t s function only with a lag; and (iii) the policy variables polynomial in positive powers of the lag operator L, in the form of government expenditure, G , and/or n the number of variables in the system, " the government revenue, T . fundamental economic shocks that span the space of The recursive assumptions can be summarized by innovations to X and v the VAR innovation, t t X ¼ [X , G , T , X ] and t lt t t 2t d :¼ B /(P Y ), d :¼ B /(P Y ), g :¼ G /(P Y ) t t t t t1 t1 t1 t1 t t t t 2 3 0 0 and t :¼ T /(P Y ). t t t t 11 |{z} |{z} |{z} n 2 n n Equation 2 refers to the government’s intertem- 6 1 1 2 7 n n 1 1 6 7 poral budget constraint, and i , G ,T ,  , Y , P ,  6 7 t t t t t t t 21 22 6 |{z} 7 |{z} |{z} ¼ ð6Þ 6 7 and d represent, respectively, the interest rate (or the 2n 2n 22 6 1 7 4 5 average cost of debt refinancing), government pri- 31 32 33 |{z} |{z} |{z} mary expenditures and government revenues, infla- n n n n 2 1 n 2 2 2 tion, GDP, price level, real growth rate of GDP and The two upper blocks of zeros correspond, the debt-to-GDP ratio at the beginning of the period t. respectively, to the assumptions that the variables Following Favero and Giavazzi (2008), this speci- fication includes the feedback from government debt, in X do not respond to the fiscal policy shock 1t an assumption that is potentially important in the either directly or indirectly. This approach delivers a determination of the effects of fiscal policy shocks for correct identification of the fiscal policy shock but a number of reasons. First, when fiscal authorities not of the other shocks in the system. In practice, care about the stabilization of debt, government we include in our system the same variables as in Since the fiscal policy shock is identified regardless of the ordering restrictions among the nonpolicy variables, we can put arbitrary O in the nonpolicy blocks of  in order to obtain (n  1)n/2 linearly independent restrictions and to get consistent s 0 impulse-responses (Christiano et al., 1999). As a result, the Choleski decomposition emerges as a particular case, where the restrictions are imposed such that  becomes a lower triangular matrix. 0 The macroeconomic effects of fiscal policy 4443 Christiano et al. (2005), but also add housing price government revenue due to the sale of Universal among the X variables, that is, we allow the policy Mobile Telecommunications System (UMTS) 1t authority to react contemporaneously to changes in licenses. the housing market. We also include the stock Due to limitations of the data, housing starts are market index and the exchange rate in X . included only in the US, and among the set of 2t The identification procedure adopted in this article variables included in the SVAR, the average govern- is closer in spirit to Fata´ s and Mihov (2001), who use ment debt cost servicing deserves a special attention. a Cholesky ordering to capture the unexpected Therefore, we obtain the average implicit interest rate variation in fiscal policy. In contrast, Afonso and by dividing the net interest payments by the govern- Sousa (2011) identify the potential impact of fiscal ment debt at time t  1. policy on asset prices, by using a Fully Simultaneous The quarterly fiscal data refers to the Federal System approach in a Bayesian framework. Government spending and revenue in the case of the Nevertheless, the empirical findings deliver qualita- US, and the Public Sector spending and revenue in tively similar results, which allows one to be confident the case of the UK In both cases, quarterly fiscal data on their robustness. Finally, we assess the posterior is available directly from national accounts. As for uncertainty about the impulse-response functions Germany and Italy, we compute the quarterly series using a Monte Carlo Markov-Chain (MCMC) of government spending and revenue using budgetary algorithm. cash data, which is published monthly by the fiscal authorities of both countries. In this case, data for government spending and revenue refer to the Central Government and are available in a cash basis. IV. Empirical Analysis The data cover the following samples: 1970:Q3– 2007:Q4, in the case of the USA; 1971:Q2–2007:Q4, in Building the data set the case of the UK; 1979:Q2–2006:Q4, in the case of Germany; and 1986:Q2–2004:Q4, in the case of Italy. We use quarterly data for four countries: US, UK, Figures 1–4 provide a comparison of the annual Germany and Italy. All the variables are in natural values of such quarterly fiscal data with the annual logarithms unless stated otherwise. national accounts data provided by the European For the identification of the fiscal policy shocks, Commission (Ameco database). It is interesting to the variables in X – the ones predetermined with 1t observe that the patterns of both series are rather respect to fiscal policy innovations – are GDP, similar. Moreover, in most of the cases, the levels private consumption, GDP deflator and private themselves are also close. investment. To these variables, we add: the housing Finally, Fig. 5 plots the observed government debt- price index (or the median sales price of new houses to-GDP ratio and the implicit debt-to-GDP ratio, sold, in the case of the US), the housing starts (only that is, the one that would emerge by including the for the US), and the average cost of government debt feedback from government debt. It shows that, financing (or the yield to maturity of long-term despite some small discrepancies in the case of Italy, government bonds). The variables in X – the ones 2t the implicit series for the debt-to-GDP ratio tracks allowed to react contemporaneously to fiscal policy pretty well the actual series. shocks – are the S&P500 Index (for the US), the FTSE-A11 Shares Index (for the UK), and the MSCI index (for Germany and Italy). As measure of the Results fiscal policy instruments we use either the government The starting point is the estimation of a B-SVAR that expenditures (in which case, the government revenues does not include the feedback from government debt, are included in X ) or the government revenues (in 1t that is, where Equation 2 is not considered. Then, we which case, the government expenditures are included compare the results with the ones that emerge from in X ). We include a constant (or quarterly seasonal 1t estimating specifications (1–3). dummies), and the government debt-to-GDP ratio in Figures 6–9 show the impulse-response functions to the set of exogenous variables. For Germany, we also consider two dummies: (i) one dummy for 1991:Q1, a fiscal policy shock. The solid line refers to the corresponding to the German reunification; and (ii) median response when the VAR is estimated without another dummy for 2000:Q3, to capture the spike in including the debt feedback, and the dashed lines are, A detailed description is provided in Afonso and Sousa (2011). In this context, Perez (2007) argues that, for some EU countries’, intra-annual cash fiscal information can be used to improve the forecasting and monitoring of the annual General Government deficit in terms of ESA95. 4444 A. Afonso and R. M. Sousa Primary spending Total spending Total revenue 35 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 6 16 40 3 8 30 6 20 4 0 0 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1971 1976 1981 1986 1991 1996 2001 2006 From quarterly data AMECO Average Cost debt Government bond yield AMECO From quarterly data AMECO Fig. 1. Quarterly versus annual fiscal data, US Primary spending Total spending Total revenue 60 50 30 30 30 25 20 20 10 10 0 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 140 16 80 3 2 6 20 2 1963 1968 1973 1978 1983 1988 1993 1998 2003 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data EP3H BKTl AMECO Average Cost Debt Government Bond Yield AMECO From quarterly data AMECO Fig. 2. Quarterly versus annual fiscal data, UK Notes: EP3H – ‘Financial Intermediation Services Indirectly Measured (FISIM) adjusted GDP; BKTL – GDP at current prices. respectively, the median response and the 68% US. Figure 6(a) displays the impulse-response func- posterior probability intervals from the VAR esti- tions of all variables in X to a shock in government mated by imposing the government budget con- spending in the US straint. The confidence bands are constructed using a In the case we do not include the debt feedback, it MCMC algorithm based on 50 000 draws. can be seen that the effects on GDP are positive, % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP The macroeconomic effects of fiscal policy 4445 Primary spending Total spending Total revenue 50 52 42 44 40 43 38 42 42 41 1979 1984 1989 1994 1999 2004 1979 1984 1989 1994 1999 2004 1979 1984 1989 1994 1999 2004 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Government debt Interest payments on debt Long-term interest rate 80 4 70 3,5 60 3 50 2,5 2 6 30 1,5 0,5 0 0 1970 1975 1980 1985 1990 1995 2000 2005 1974 1979 1984 1989 1994 1999 2004 1974 1979 1984 1989 1994 1999 2004 From quarterly data AMECO From quarterly data AMECO Average Cost Debt Government Bond Yield AMECO Fig. 3. Quarterly versus annual fiscal data, Germany Primary spending Total spending Total revenue 50 50 40 40 30 30 20 20 10 10 1976 1981 1986 1991 1996 2001 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 From quarterly data AMECO From quarterly data AMECO From quarterly data AMECO Long-term interest rate Government debt Interest payments on debt 140 18 100 10 80 8 60 6 40 4 20 2 0 0 0 1985 1990 1995 2000 1984 1989 1994 1999 2004 1982 1987 1992 1997 2002 From quarterly data AMECO From quarterly data AMECO Average Cost Debt Government Bond Yield AMECO Fig. 4. Quarterly versus annual fiscal data, Italy sizeable and reach a peak at after eight quarters. The ‘Keynesian’ effect in the economy and that there is impact on private consumption and private invest- no ‘crowding-out’. In addition, there is a positive ment is also positive, therefore, supporting the idea effect on the average cost of debt and the price level. that government spending has an expansionary In what concerns the reaction of asset markets, the Fata´ s and Mihov (2001), when looking at the response to changes in different components of government expenditures, find that increases in government consumption are always expansionary, while increases in public investment do not have a significant impact on output. Here, we cannot analyse some of these policy experiments because we have not made explicit the role of different components of government spending. % GDP % GDP % GDP % GDP % GDP % GDP % GDP % GDP % % % GDP % GDP 4446 A. Afonso and R. M. Sousa US UK Debt ratio Debt ratio 50 100 40 80 20 40 10 20 Implicit debt ratio Observed debt ratio Observed debt ratio Implicit debt ratio Germany Italy Debt ratio Debt ratio 0 0 Implicit debt ratio Observed debt ratio Implicit debt ratio Observed debt ratio Fig. 5. Implicit debt ratio and observed debt ratio, percentage of GDP Note: Implicit debt ratio computed via Equation 2. empirical evidence suggests that while there is a GDP are negative, very persistent and the trough is positive effect on housing prices that persists for reached at after 12 quarters. They also reveal a almost 20 quarters, the reaction of stock prices is change in the composition of GDP’s major compo- rather small and negative. nents: while private consumption is negatively When we include the debt dynamics in the model, impacted by a positive shock in government revenues, the effects of a government spending shock on GDP the effect on private investment is positive but quickly become somewhat smaller. The effects on private erodes after eight quarters. That is, in this case, the consumption and private investment also remain ‘crowding-out’ effect works mainly through the positive although smaller. In addition, the effect on consumption channel. In what concerns the reaction the average cost of refinancing the debt is negligible. of asset markets, the empirical evidence suggests that The reaction of asset markets is similar to the model the effects of revenue shocks tend to be negative for without the government constraint: there is a positive housing prices and slightly positive for stock prices. (although smaller) and persistent effect on housing UK. The impulse-response functions to a shock in prices, while the reaction of stock prices is still small due to the debt dynamics and the portfolio government spending in the UK are shown in reallocation. Fig. 7(a). Figure 6(b) shows the impulse-response functions The results show that an increase in government to a shock in government revenue. The results suggest spending has a positive effect on GDP in accordance that government revenue declines steadily following with the predictions of the standard IS-LM model. the shock which erodes after eight quarters. Contrary As for the US, there is evidence of a change in its to a shock in government spending, the effects on composition: private consumption increases but the 1966Q4 1970Q2 1968Q3 1972Q1 1970Q2 1973Q4 1972Q1 1975Q3 1973Q4 1977Q2 1975Q3 1979Q1 1977Q2 1980Q4 1979Q1 1982Q3 1980Q4 1984Q2 1982Q3 1986Q1 1984Q2 1987Q4 1986Q1 1989Q3 1987Q4 1989Q3 1991Q2 1991Q2 1993Q1 1993Q1 1994Q4 1994Q4 1996Q3 1996Q3 1998Q2 1998Q2 2000Q1 2000Q1 2001Q4 2001Q4 2003Q3 2003Q3 2005Q2 2005Q2 2007Q1 2007Q1 1985Q4 1963Q4 1965Q4 1986Q4 1967Q4 1987Q4 1969Q4 1988Q4 1971Q4 1989Q4 1973Q4 1990Q4 1975Q4 1991Q4 1977Q4 1979Q4 1992Q4 1981Q4 1993Q4 1983Q4 1994Q4 1985Q4 1995Q4 1987Q4 1996Q4 1989Q4 1997Q4 1991Q4 1998Q4 1993Q4 1995Q4 1999Q4 1997Q4 2000Q4 1999Q4 2001Q4 2001Q4 2002Q4 2003Q4 2003Q4 2005Q4 2004Q4 2007Q4 The macroeconomic effects of fiscal policy 4447 Fig. 6. Impulse-response functions, US: (a) Spending shock and (b) Revenue shock 4448 A. Afonso and R. M. Sousa Fig. 7. Impulse-response functions, UK: (a) Spending shock and (b) Revenue shock The macroeconomic effects of fiscal policy 4449 Fig. 8. Impulse-response functions, Germany: (a) Spending shock and (b) Revenue shock 4450 A. Afonso and R. M. Sousa Fig. 9. Impulse-response functions, Italy: (a) Spending shock and (b) Revenue shock The macroeconomic effects of fiscal policy 4451 effects on private investment are negative, suggesting These findings do not change significantly when the idea of a ‘crowding-out’ effect via investment (see, the debt dynamics is included in the model. e.g. Blanchard and Perotti (2002)). In what concerns Figure 8(b) plots the impulse-response functions to the reaction of asset markets, the empirical evidence a shock in government revenue. Similar to the US, the suggests that both housing and stock prices fall after results show that government revenue declines the spending shock but the effect on housing markets quickly after the shock, eroding after two quarters. is much more persistent. That is, markets seem to The effects on GDP are positive and support the idea interpret the expansion of government spending as of a ‘crowding-in’ effect as both private consumption signalling a deterioration of public finances and private investment react positively to the shock. (Ardagna, 2009). In contrast, the effects on the Revenue shocks tend to be significant and positive price level and on the average cost of financing the only for housing prices. debt are positive. When the debt dynamics is included in the model, the impulse-response functions do not Italy. We look at the effects of a fiscal policy shock change significantly. in Italy. Figure 9(a) displays the impulse-response Figure 7(b) shows the impulse-response functions functions to a shock in government spending. Despite to a shock in government revenue. They support the a very small positive effect in the first quarters, GDP, idea of a ‘crowding-in’ effect: private investment private consumption and private investment fall, reacts positively to the shock; private consumption suggesting a ‘crowding-out’ effect and a ‘Non- falls in the first eight quarters after the shock but then Keynesian’ pattern. Government spending shocks recovers and the effect becomes positive. The effects have a positive and persistent effect on both the price on housing prices are also positive, but the reaction level and the average cost of financing the debt, in takes place with a lag of around eight quarters. By accordance with the findings of Giordano et al. their turn, stock prices react positively to the shock (2007). Housing prices seem to be positively affected but the effect reaches a peak at around after eight by the shock, while stock prices fall immediately after quarters. Finally, the effects on the price level and the the shock but later start recovering. interest rate are negative and the impact is maximum When we take into account the feedback from at after 8 to 12 quarters. government debt, the results suggest that the effects When we include the feedback from the on GDP, private consumption, private investment, government debt in the model, the effects on the the price level and the interest rate become smaller. majority of the variables included in the model This gives rise to the importance of considering the become smaller. debt dynamics in the model. Figure 9(b) shows the impulse-response functions Germany. We now look at the impulse-response to a shock in government revenue. The effects on functions a shock in government spending for GDP, private consumption and private investment Germany (Fig. 8(a)). The effects on GDP are are negative in the first quarters, but the impacts are negative, reflecting the fall in both private consump- not persistent as they vanish after four to six tion and private investment, and providing evidence quarters and even become positive after that. of a ‘Non-Keynesian’ effect (see, for instance, Looking at the reaction of asset markets, one can Giavazzi and Pagano (1990, 1996), Giudice et al. see that the effects of government revenue shocks (2004) and Afonso (2010)). There is, therefore, tend to be positive for stock prices and negative for evidence of a ‘crowding-out’ effect. Nevertheless, housing prices. This suggests that while the credit- the magnitude of the impact is small and, in this channel (i.e. the fall in interest rates) impacts respect, it corroborates the findings of Heppke-Falk positively in stock markets, for housing markets et al. (2006). Government spending shocks also have that channel is annihilated by the ‘crowding-out’ a negative and persistent effect on the price level and effects. Moreover, it is in accordance with the the average cost of financing debt. Finally, housing argument that the effects of fiscal consolidation on prices seem to increase after the shock while stock stock prices are particularly strong in countries prices fall immediately after the shock, possibly, which typically run higher levels of government reflecting the deterioration of fiscal policy. deficit (Ardagna, 2009). Afonso and Claeys (2008) mention that large revenue reductions unmatched by expenditure cuts have pushed the deficit beyond the 3% threshold in Germany in 2002, putting the country in an excessive deficit situation within the EU fiscal framework. 4452 A. Afonso and R. M. Sousa Table 1. The effect of (d  d*) in a VAR t1 US T G Germany T G t t t t (d  d*) 1970:Q3 to 2007:Q4 0.187*** 0.180*** (d  d*) 1980:Q3 to 2006:Q4 0.211*** 0.021 t1 t1 (N ¼ 150) (0.068) (0.050) (N ¼ 106) (0.112) (0.064) 1970:Q3 to 1987:Q4 0.352 0.176 1980:Q3 to 1990:4 0.122 1.087*** (N ¼ 70) (0.261) (0.156) (N ¼ 42) (0.410) (0.271) 1988:1 to 2007:4 0.203 0.228 1991:Q1 –2006:Q4 0.017 0.075 (N ¼ 80) (0.158) (0.154) (N ¼ 64) (0.192) (0.060) UK T G Italy T G t t t t (d  d*) 1964:2 to 2007:4 0.120*** 0.065** (d  d*) 1986:Q2 to 2004:Q4 0.081 4.125** t1 t1 (N ¼ 175) (0.037) (0.030) (N ¼ 75) (0.476) (1.741) 1964:2 to 1984:4 0.013 0.102 1986:Q2 to 1994:Q4 1.344 3.963 (N ¼ 83) (0.109) (0.089) (N ¼ 35) (1.529) (8.288) 1985:1 to 2007:4 0.143** 0.043 1995:Q1 to 2004:Q4 4.353** 6.642 (N ¼ 92) (0.127) (0.104) (N ¼ 40) (1.838) (6.518) Notes: SEs in brackets. *, ** and *** denote statistical significanceat the 10, 5 and 1% levels, respectively. Fiscal shocks and government debt feedback debt-to-GDP ratio from its sample average only for the full sample. Indeed, one sees that when the debt- We now consider the potential debt feedback and to-GDP ratio is above its historical mean, govern- estimate the following structural VAR: ment primary spending decreases (the coefficient associated to (d  d*) in the government spending X þ  X þ  þ  ðd  d Þ¼ c þ " ð7Þ 0 t 1 t1 i t1 t t1 equation is negative and significant (0.180), while in the government revenue equation is positive and ð1 þ i Þ d ¼ d þð g  tÞð8Þ t t1 t t significant (0.187)). ð1 þ  Þð1 þ  Þ t t For the UK, the results are similar. They show a The specification of Equation 7 is suggested by the stabilizing effect of the debt level on the primary empirical findings in Bohn (1998), who estimates a budget balance that works through the response of fiscal reaction function in which d* is the uncondi- both government revenue and government spending tional mean of the debt ratio and allow us to take into to deviations of the debt from the target level. This is, account the debt feedback. Following Bohn (1998) particularly, the case of the full sample. It is also the and Favero and Giavazzi (2008), we model the target case of the second sub-sample (1985:Q1 to 2007:Q4) level of the debt as a constant on the basis of the but only for government revenue. Therefore, when evidence of stationarity of d. In addition, we impose the debt ratio is above its sample mean, it is possible the government’s intertemporal budget constraint as to observe an increase in government revenue (0.120, described by Equation 8. in the full sample; 0.143, for the second sub-sample). Table 1 reports the estimated coefficients on In the case of Germany, there is evidence of a small (d  d*) in the structural equations of the SVAR t1 stabilizing effect of the debt level on the primary (government spending and government revenue). We surplus that works through the government revenue report the coefficients (and the SEs in brackets) taken (0.211). In addition, the results of the estimation of from the estimation for the full sample and the sub- the model in the first sub-sample (1980:Q3 to samples. In the case of the US, we consider two sub- 1990:Q4) suggest that government spending tends to samples: 1970:Q3 to 1987:Q4, corresponding to the decrease (1.087) when the debt-to-GDP ratio is pre-Greenspan era; and 1988:Q1 to 2007:Q4, after above its sample mean. Greenspan. In the case of the Germany, we also split Finally, for Italy, one can see that when the debt the sample in two periods: 1980:Q3 to 1990:Q4, that ratio is above average, government spending strongly is, before the reunification; and 1991:Q1 to 2006:Q4, falls (4.125). Additionally, in the period 1995:Q1 to that is, after the reunification. For the UK and Italy, 2004:Q4, there is evidence supporting a stabilizing each sub-sample is built by splitting the entire sample effect of government debt on the primary surplus that into roughly two sub-samples of similar size. works through the government revenue: the coeffi- For the US, the results show a significant response cient associated to (d  d*) is positive and large in t1 of revenue and primary spending to deviations of the magnitude (4.353). This is in line with the increase of The macroeconomic effects of fiscal policy 4453 fiscal policy discipline imposed by the Maastricht comments and suggestions, to Boris Hofmann, Treaty. Bernhard Manzke and Sandro Momigliano, for data clarifications and to Silvia Albrizio and Matthijs Lof for research assistance. The opinions expressed herein are those of the authors and do not necessarily reflect those of the ECB or the V. Conclusion Eurosystem. UECE is supported by FCT (Fundac¸ a˜ o para a This article provides a detailed evaluation of the Cieˆ ncia e a Tecnologia, Portugal), financed by ERDF macroeconomic effects of fiscal policy. It, therefore, and Portuguese funds. also deals with a relevant policy question, in partic- Ricardo Sousa would like to thank the Fiscal ular, given the expansionary fiscal policies adopted by Policies Division of the ECB for its hospitality. governments in industrialized countries in the attempt to stabilize the economies after the burst of the financial crisis in 2008. We identify fiscal policy shocks using a recursive partial identification, and estimate a B-SVAR, there- References fore, accounting for the posterior uncertainty of the Afonso, A. (2008) Ricardian fiscal regimes in the European impulse-response functions. In addition, we explicitly Union, Empirica, 35, 313–34. include the feedback from government debt in our Afonso, A. (2010) Expansionary fiscal consolidations in framework. Europe: new evidence, Applied Economics Letters, 17, 105–9. The empirical evidence suggests that government Afonso, A. and Claeys, P. (2008) The dynamic behaviour of spending shocks: (i) have, in general, a small but budget components and output, Economic Modelling, positive effect on GDP; (ii) have a varied effect on 25, 93–117. private consumption and on private investment; (iii) Afonso, A. and Sousa, R. M. (2011) What are the effects of have a positive effect on housing prices; (iv) lead to a fiscal policy on asset markets?, Economic Modelling, quick fall in stock prices; and (v) in general, impact 28, 1871–90. Afonso, A. and Strauch, R. (2007) Fiscal policy events and positively on the price level and the average cost of interest rate swap spreads: some Evidence from the refinancing the debt. In addition, government reve- EU, Journal of International Financial Markets, nue shocks: (i) have a positive effect on GDP and Institutions and Money, 17, 261–76. private investment; (ii) have a varied effect on private Alesina, A. and Ardagna, S. (1998) Tales of fiscal adjust- consumption; (iii) a varied impact on housing prices ment, Economic Policy, 27, 489–545. Ardagna, S. (2009) Financial markets’ behavior around and a positive effect on stock prices; and (iv) in episodes of large changes in the fiscal stance, European general, do not impact on the price level, but have a Economic Review, 53, 37–55. mixed effect on the interest rate. Barro, R. J. (1974) Are government bonds net wealth?, When the debt dynamics is explicitly taken into Journal of Political Economy, 82, 1095–117. account, (long-term) interest rates and GDP become Bernanke, B. and Kuttner, K. (2005) What explains the stock market’s reaction to Federal Reserve policy?, more responsive and the effects of fiscal policy on Journal of Finance, 60, 1221–57. these variables also become more persistent. Bernoth, K. J., Von Hagen, J. and Schuknecht, L. (2003), Finally, the results support the existence of a The determinants of the yield differential in EU stabilizing effect of the debt level on the primary government bond market, Working Paper, ZEI – budget balance that works through both government Center for European Integration Studies. Biau, O. and Girard, E. (2005) Politique budgetaire et spending and revenue in the case of the US and the dynamique economique en France: l’approche VAR UK, through government revenue in the case of structurel, Economie et Pre´vision, 169–171, 1–24. Germany, and through government spending in the Blanchard, O. and Perotti, R. (2002) An empirical case of Italy. This signals that governments tend to characterization of the dynamic effects of changes in adjust their fiscal imbalances in response to govern- government spending and taxes on output, Quarterly ment debt developments. Journal of Economics, 117, 1329–68. Bohn, H. (1998) The Behaviour of US public debt and deficits, Quarterly Journal of Economics, 113, 949–63. Brook, A.-M. (2003) Recent and prospective trends in real long-term interest: fiscal policy and other drivers, OECD Working Paper No. 21, Organization for Acknowledgements Economic Co-operation and Development. We are grateful to Marten Blix, Carlo Favero, Jiirgen Canzoneri, M., Cumby, R. and Diba, B. (2001) Is the price Von Hagen, Ad van Riet and to participants to a level determined by the needs of fiscal solvency, European Central Bank (ECB) seminar for helpful American Economic Review, 91, 1221–38. 4454 A. Afonso and R. M. Sousa Castro, F. and Cos, P. H. (2008) The economic effects of Goodhart, C. and Lemmen, J. J. G. (1999) Credit risks and fiscal policy: the case of Spain, Journal of European government bond markets: a panel data Macroeconomics, 30, 1005–28. econometric analysis, Eastern Economic Journal, 25, Chirinko, R., de Haan, L. and Sterken, E. (2004) Asset 77–107. price shocks, real expenditures, and financial structure: Heppke-Falk, K.H. and Hu¨ fner, F. (2004) Expected budget a multi-country analysis, Emory University, Working deficits and interest rate swap spreads – evidence for Paper. France, Germany and Italy, Discussion Paper No. 40, Christiano, L. J., Eichenbaum, M. and Evans, C. L. (1999) Deutsche Bundesbank. Monetary policy shocks: what have we learned and to Heppke-Falk, K.H., Tenhofen, J. and Wolff, G. B. (2006) what end? in Handbook of Macroeconomics, 1A (Eds) The macroeconomic effects of exogenous fiscal J. Taylor and M. Woodford, Elsevier, Amsterdam, policy shocks in Germany: a disaggregated SVAR pp. 65–148. analysis, Discussion Paper No. 41, Deutsche Christiano, L. J., Eichenbaum, M. and Evans, C. L. (2005) Bundesbank. Nominal rigidities and the dynamic effects of a shock Iacoviello, M. (2005) House prices, borrowing constraints, to monetary policy, Journal of Political Economy, 113, and monetary policy in the business cycle, American 1–45. Economic Review, 95, 739–64. Codogno, L., Favero, C. and Missale, A. (2003) Yield Iacoviello, M. and Minetti, R. (2003) The credit channel of spreads on EMU government bonds, Economic Policy, monetary policy: evidence from the housing market, 37, 503–32. Working Paper, Boston College. Craine, R. and Martin, V. (2003) Monetary policy shocks Jappelli, T. and Pistaferri, L. (2007) Do people respond to and security market responses, Working Paper, tax incentives? An analysis of the Italian reform of the University of California at Berkeley. deductibility of home mortgage interests, European Edelberg, W., Eichenbaum, M. and Fisher, J. (1999) Economic Review, 51, 247–71. Understanding the effects of a shock to government Julliard, C., Michaelides, A. and Sousa, R. M. (2008) purchases, Review of Economics Dynamics, 2, 166–206. Housing prices and monetary policy, Working Engen, E. M. and Hubbard, R. G. (2004) Federal govern- Paper, London School of Economics and Political ment debt and interest rates, NBER Working Paper Science. No. 10681, National Bureau of Economic Research. Kormendi, R. C. (1983) Government debt, government Faini, R. (2006) Fiscal policy and interest rates in Europe, spending, and private sector behavior, American Economic Policy, 21, 443–89. Economic Review, 73, 994–1010. Fata´ s, A. and Mihov, I. (2001) The effects of fiscal policy Laubach, T. (2009) New evidence on the interest rate effects on consumption and employment: theory and evi- of budget deficits and debt, Journal oj the European dence, CEPR Discussion Paper No. 2760, Centre for Economic Association, 7, 1–28. Economic Policy Research. Maclennan, D., Muellbauer, J. and Stephens, M. (1999) Favero, C. (2002) How do European monetary and fiscal Asymmetries in Housing and Financial Market authorities behave?, CEPR Discussion Paper No. 3426, Institutions and EMU, CEPR Discussion Paper No. Centre for Economic Policy Research. 2062, Centre for Economic Policy Research. Favero, C. and Giavazzi, F. (2008) Debt and the effects of McCarthy, J. and Peach, R. (2002) Monetary policy fiscal policy, University of Bocconi, Working Paper. transmission to residential investment, Federal Gale, W. G. and Orszag, P. R. (2002) The economic effects Reserve Bank of ‘New York Economic Policy Review, of long-term fiscal discipline, Discussion Paper, 8, 139–58. Urban-Brookings Tax Policy Center. Mountford, A. and Uhlig, H. (2005) What are the effects of Gale, W. G. and Orszag, P. R. (2003) Economic effects of fiscal policy shocks?, Humboldt-Universita¨ t zu Berlin sustained budget deficits, National Tax Journal, 56, Working Paper SFB No. 649. 463–85. Pe´ rez, J. (2007) Leading indicators for euro area govern- Galı´, J., Lo´ pez-Salido, J. D. and Valle´ s, J. (2007) ment deficits, International Journal of Forecasting, 23, Understanding the effects of government spending on 259–75. consumption, Journal of the European Economic Perotti, R. (1999) Fiscal policy in good times and bad, Association, 5, 227–70. Quarterly Journal of Economics, 114, 1399–436. Giavazzi, F. and Pagano, M. (1990) Can severe fiscal Perotti, R. (2004) Estimating the effects of fiscal policy in contractions be expansionary? Tales of two small OECD countries, Working Paper, University of european countries, in NBER Macroeconomics Annual Bocconi. (Eds) O. J. Blanchard and S. Fischer, MIT Press, Ramey, V. and Shapiro, M. (1998) Costly capital realloca- Cambridge, MA, pp. 75–110. tion and the effects of government spending, Carnegie Giavazzi, F. and Pagano, M. (1996) Non-Keynesian effects Rochester Conference on Public Policy, 48, 145–94. of fiscal policy changes: international evidence and the Rigobon, R. and Sack, B. (2002) The impact of monetary Swedish experience, Swedish Economic Policy Review, policy on asset prices, Finance and Economics 3, 67–103. Discussion Series 2002–4, Board of Governors of the Giordano, R., Momigliano, S., Neri, S. and Perotti, R. Federal Reserve System. (2007) The effects of fiscal policy in Italy: evidence Rigobon, R. and Sack, B. (2003) Measuring the reaction of from a VAR model, European Journal of Political Economy, 23, 707–33. monetary policy to the stock market, Quarterly Journal Giudice, G., Turrini, A. and in’t Veld, J. (2004) Non- of Economics, 118, 639–670. Keynesian fiscal consolidation in the EU? Ex post Upper, C. and Worms, A. (2003) Real long-term interest evidence and ex ante analysis, CEPR Discussion Paper rates and monetary policy: a cross-country perspective, No. 4388, Centre for Economic Policy Research. BIS Papers, 19, 234–257.

Journal

Applied Economics – Taylor & Francis

Published: Dec 1, 2012

Keywords: fiscal policy; Bayesian structural VAR; debt dynamics; C11; C32; E62; H62

There are no references for this article.