Advanced Macroeconomics II. Fiscal Policy

Transcription

1 Advanced Macroeconomics II Fiscal Policy Lorenza Rossi (Spring 2014) University of Pavia Part of these slides are based on Jordi Galì slides for Macroeconomia Avanzada II.

2 Outline Fiscal Policy in the Real business cycle model Debt Dynamics The e ects of scal shocks Fiscal evidence on countries debt and debt consolidation

3 Public Expenditure: provision of public goods and servises. Redistributive issues - expenditure for goods and servises. - transfer: redistributive (unemployment subsidies, pensions) - subsidies: rms subsidies - interests on public debt Taxes: Finance the provision of public goods and servises. Redistributive and incetive issues - consumption taxes - capital income taxes - pro ts taxes - labor income taxes The di erence between public expenditure and taxes is the de cit, nanced through the emission of public debt Fiscal policy decision =) A ects the level of the economic activity.! Fiscal policy is exogenous: independent from uctuations! Fiscal policy is used as an instrument of business cycle stabilization

4 Fiscal Policy in the RBC model GOVERNMENT BUDGET CONSTRAINT: with a balanced budget The Government runs a balanced budget Example I: with only lump-sum taxes G t = T t Example II: with lump-sum taxes and labor income taxed G t = T t + τ n w t N t G t can be represented by an AR(1) process Gt Gt 1 ln = ρ G g ln + ε g,t G If G does not enter into the utility function, i.e. U (C t, G t, N t ) or alternatively it does not enter into rms production function, i.e. Y = AF (N, K, G ) =)it is a pure waste for the economy. In this case government expenditure is nonproductive.

5 IRFS to a government spending shock in the basic RBC 0.15 y 0 c 1 i n 0 w 0 k x 10 3 r 1.5 spesa Result: government spending implies a large crowding-out e ect on private consumption

6 IRFS to a government spending shock in the basic RBC The RBC model predicts a decline in private consumption in response to a rise in government spending. (crowding out e ect on consumption) With in nitely-lived Ricardian households, an increase in (nonproductive) government spending purchases ( nanced by current or future lump-sum taxes), lowers the present value of after-tax income, and thus generates a negative wealth e ect on consumption (Aiyagari et al., 1990; Baxter and King, 1993; Christiano and Eichenbaum, 1992). The quantity of labor supplied at any given wage increases. Equilibrium: lower consumption, lower real wage, higher employment (increase in hours) and higher output. The increase in employment leads, if su ciently persistent, to a rise in the expected return to capital, and may trigger a rise in investment. Otherwise investment decrease.

7 Fiscal Policy in the RBC model GOVERNMENT BUDGET CONSTRAINT: general form Government budget constraint In period t: G t + (1 + r t 1 )B g t 1 = τn t w t N t + T t + Bt g Household budget constraint of the RBC model modi es as follows: C t + B t = (1 τ n t )w t N t + (1 + r t 1 )B t 1 + D t T t De nitions: G t : public expenditure Bt G : stock of Government Bond - Public Debt (one period, risk free) r t : interest rate on public debt. τ n t : labor income tax T t : lump-sum tax (no distorsionary).

8 Fiscal Policy in the RBC model The equation for the labor supply becomes: For U (C t, N t ) = C t 1 U n,t + (1 τ n t )w t U c,t = 0 σ 1 σ v N +ϕ t 1+ϕ bw t = σbc t + ϕbn t, in log-deviation from the steady state τ n 1 τ n bτn t where τ n is the steady state of the labor income tax. The production function in log-deviation Y t = A t N 1 by t = a t + (1 t α α) bn t

9 Equilibrium Goods market in log-deviations Y t = C t + G t by t = (1 s g )bc t + s g bg t. where s g G /Y. Labor market σbc t + ϕbn t τ n 1 τ n bτn t = ba t αbn t Bonds market bb t = bb g t br t = σe t f bc t+1 g Technology by t = ba t + (1 α) bn t

10 Equilibrium output From the labor market, equilibrum after some algebra, σ (1 α) + (1 + ϕ) (1 s g ) (1 α) ϕ by t = ba t (1 s g ) (1 α) (1 α) +σ s g τ n bg t 1 s g 1 τ n bτn t solving for by t by t = (1 s g ) (1 + ϕ) σ (1 α) + (1 + ϕ) (1 s g ) ba t (1 α) σs g + σ (1 α) + (1 + ϕ) (1 s g ) bg t + (1 s g ) (1 α) τ n σ (1 α) + (1 + ϕ) (1 s g ) 1 τ n bτn t

11 Equilibrium of hours, consumption and real wage Substituting the equilibrium output into the production function we nd bn t = n (bg t, bτ n t, ba t ). Substituting by t into the aggregate resource constraint we nd bc t = c (bg t, bτ n t, ba t ) Finally substituting bn t and bc t into the labor supply we nd bw t = w (bg t, bτ n t, ba t ) Find bc t = c (bg t, bτ n t, ba t ), bn t = n (bg t, bτ n t, ba t ) and bw t = w (bg t, bτ n t, ba t ) by your own Finally we also know that br t = σe t f c t+1 g

12 Discussion - E ects of di erent scal shocks (τ n t, g t ) - Ricardian Equivalence Ricardiana - Fiscal rules and Stabilization policies - Structural De cit: a big European Problem

13 Empirical evidence on the e ects of scal shocks Identi cation problem: simultaneity. Discretionary changes in taxes are likely to a e ect GDP contemporaneously, but aggregate uctuations will also contemporaneously a ect commonly used tax measures (such as tax revenues). Consider the following y t = α 0 + b τ t + u t Any measure τ t which is a function of factors also contemporaneously a ecting output, cannot be used to consistently identify the e ects of tax changes. The chosen tax measure would be contemporaneously correlated with the error term u t, violating the standard requirement for consistent estimation of the coe cients. Blanchard and Perotti (2002). Try to identify only the "structural" shocks to revenues: uncorrelated with contemporaneos shocks.

14 Empirical evidence on the e ects of scal shocks The Macroeconomic E ects of Tax Changes: Romer & Romer (AER 2010) - narrative approach - exogenous tax changes: legislated changes (not automatic) tax changes introduced for: - need to reduce the public de cit inherited - aim to achieve a long-term debt target

15 Empirical evidence on the e ects of scal shocks - Romer and Romer 2010 AER Equations estimated: y t = α + K k=0 β k T t k + u t and y t = α + K k=0 β k T a K t k + k=0 γ k T i t k + u t

16 Empirical evidence on the e ects of scal shocks - Romer and Romer 2010 AER

17 Empirical evidence on the e ects of scal shocks - Romer and Romer 2010 AER

18 Empirical evidence on the e ects of scal shocks - Romer and Romer 2010 AER

19 Debt Dynamics Understanding the E ects of Government Spending on Consumption: Galí, López-Salido y Vallés (JEEA 2007) - A model with both standard Ricardian and liquidity constrained households (consume all their disposable income) - Public expenditure (purchases of good and services) g t = K φ 0 k x t k=1 k + ε g t - Macroeconomic E ects: z t = K β k ε g t k=0 k + u t

20 The E ects of Government Spending Shocks - Galì, Lopez-Salido and Vallet JEEA 2007

21 The E ects of Fiscal Consolidation E ects of Fiscal Consolidation: IMF WEO Oct. 2010) - 15 Countries, countries-year scal consolidation measures (aim: de cit reduction) - Equations estimated: z t = α + K k=0 β k f t k + u t z t :GDP, unemployment

22 The E ects of Fiscal Consolidation - IMF WEO October 2010

23 The E ects of Fiscal Consolidation - IMF WEO October 2010

24 The E ects of Fiscal Consolidation - IMF WEO October 2010

25 The E ects of Fiscal Consolidation - IMF WEO October 2010 Factors that usually soften the short-term impact of scal consolidation. Central banks cut of interest rates and the currency falls in value. This helps cushion the impact on consumption and investment, and boosts exports. Fiscal consolidation is less costly when markets are more concerned about scal sustainability. Consolidations based on spending cuts are less painful than those based on tax hikes. This is largely because Central banks cut interest rates more after spending cuts.

26 The E ects of Fiscal Consolidation - IMF WEO October 2010 FISCAL CONSOLIDATION IN THE LONG RUN: scal consolidation has a positive impact on output. In particular, lower debt tends to reduce real interest rates and debt service costs, which allows for future tax cuts. By boosting private investment, this increases output in the long term. The authors simulations suggest that the contraction in output may be more than twice as large as their baseline estimate when central banks cannot cut interest rates, and when the adjustment is synchronized across all countries. Nevertheless, for economies considered at high risk of sovereign default, short-term negative e ects are likely to be smaller.

27 Debt Dynamics B g t = (1 + r t 1 )B g t 1 + G t τ n t W t N t T t De nition of De cit: DEF t B g t B g t 1 = r t 1B g t 1 + G t τ n t W t N t T t De nition of Primary De cit: or DEF p t G t τ n t W t N t T t DEF p t G t T t DEF t r t 1 B g t 1 + G (Y t ) DEF p, t G (Y t ) T (Y t ) T (Y t )

28 Debt Dynamics De nitions b g t B g t /Y t ; def t DEF t /Y t ; def p t deft DEFt /Yt ; def p, t DEF p t /Y t DEF p, t /Yt ; g t Y t Y t 1 Y t 1 De cit and discretionary scal policy def t = r t 1 b g t 1 + def p, t + cyclical comp.

29 Debt Dynamics I Considering the de cit: B g t = B g t 1 + DEF t or, as a share of output b g t = g t 1 b g t 1 + def t thus b g t = g t g t 1 b g t 1 + def t

30 Debt Dynamics I Stationarity condition: bt g = if g 1 + g bg t 1 + def g > 0 In the steady state: show the phase diagram b g = 1 + g g def

31 Debt Dynamics I: Phase Diagram with g > 0

32 Debt Dynamics II Considering the primary de cit: B g t = (1 + r t 1 )B g t 1 + DEF p t as a share of output bt g 1 + rt 1 = 1 + g t 1 b g t 1 + def t p thus bt g rt 1 g t 1 = 1 + g t 1 b g t 1 + def t p

33 Stationarity condition bt g r = g 1 + g b g t 1 + def p if: Steady State: r < g b g = 1 + g g r def p

34 Dynamics of Primary Debt: r < g

35 Dynamics of Primary Debt: r < g

36 Dynamics of Primary Debt: r > g

37 Dynamics of Primary Debt: r b g t 1

38 Alternatives to an adjustment of the primary de cit Higher growth Default or debt restructuring (or in ation) "Financial Repression" Sale of public assets (40% of GDP in advanced economies) Empirical Evidence

39 Government Balance: Fiscal Monitor 2014

40 Fiscal Balance: Fiscal Monitor 2014

41 Fiscal Trend: Fiscal Monitor 2014

42 Public Debt: Fiscal Monitor 2014

43 Trend in Public Debt: Fiscal Monitor 2014

44 Gross Financing Needs: Fiscal Monitor 2014