Fiscal Multipliers and Heterogeneous Agents

Transcription

1 Fiscal Multipliers and Heterogeneous Agents Yongquan CAO April, 6 Abstract The paper analyzes the impacts on different agents from the fiscal shocks under a heterogeneous agent New Keynesian model, where the Ricardian Equivalence does not hold anymore. Different individual gets both quantitatively and qualitatively different impacts from the same fiscal shocks. The financially constrained agents tend to have larger consumption responses than others. Also, the heterogeneous impact of the fiscal policy hinges also on how the government spending is financed and monetary and fiscal regimes. Keywords: Fiscal Policy, Incomplete Markets. JEL Classification: E6, D Committee Members: Eric Leeper(Chair), Todd Walker and Grey Gordon. I am grateful to my third year paper committee for their continuous encouragement and support. This draft is still preliminary and is prepared for Jordan River Conference on April 9th, 6. Department of Economics, Indiana University, Bloomington, IN 7. yongcao@indiana.edu

2 Introduction During the Great Recession, monetary policy is constrained by zero lower bound on the nominal interest rate and the government relies more on expansionary fiscal policy to stimulate the economy. The effectiveness of the fiscal policy is getting more attention from researchers. At the same time, the growing income and wealth inequality suggests that evaluating the effectiveness of fiscal policy through aggregate variables is not enough, since the fiscal policy may not influence the individual equally For example, Jappelli and Pistaferri () estimates the MPC for the household with different wealth and finds that wealth poor households have larger MPC than wealth rich households. I want to address the following questions: How does the fiscal policy affect the agents heterogeneously? More specifically, do the fiscal instruments or the monetary regime matter? To answer these questions, I build a heterogeneous agent New Keynesian Model with monetary and fiscal authorities. I study the two kinds of fiscal shocks, i.e. government spending shock and lump-sum transfer shock, and three ways of fiscal financing, i.e. lumpsum tax, labor income tax, and inflation, seperately. Our first contribution is that the financially constrained agents consume relatively more than others in most of the cases when an expansionary fiscal shock hits the economy. The major reason is that their MPC is much larger than others. The expansionary fiscal shock boosts the aggregate demand and drives up the wages income for the individuals. Then financially constrained individuals consume the most of the extra wage income. The second contribution is that the lump-sum tranfer shock financied only by the lumpsum tax has an impact on output and consumption, which is not the case in the Ricardian Equivalence environment. The borrowing limit and incomplete market prevents the some indiviuals to insure fully against the future shocks, so the Ricardian Equivalence breaks down. The financially constrained agents will consume almost all the increase in wage income when the lump-sum transfer shock hits, and will have to consume less when the lump-sum tax is imposed in the long run. The third contribution is that how the fiscal shocks are financied matters. In general, the lump-sum tax financing can generate larger multipliers in both output and consumption than the labor income tax financing. The shocks that are financed by inflation generate largest consumption multiplier among all the experiments. The paper is related to a vast amount of literature on the effectiveness of fiscal policy. Leeper, Traum, and Walker () performs both prior and posterior predictive analysis to show the transmission mechanisms that underlies the fiscal multipliers and empirical estimates. Ilzetzki, Mendoza, and Végh () claims that the magnititude of fiscal multipliers The income gini increases from.6(989) to.67() and wealth gini increases from.79(989) to.8() according to Rios-Rull and Kuhn (6)

3 depend on the country-specific characteristics. Galí, López-Salido, and Vallés (7) shows that introducing the rule-of-thumb agent, who has large MPC, can reconcile the empirical findings and the macroeconomic model. Some empirical findings (Brinca, Holter, Krusell, and Malafry ()) suggest the inequality affects the fiscal multipliers. Heathcote () finds the short-run effects of tax change in a heterogeneous agent model is larger than the one in representative agent model. Brinca et al. () also builds a heterogeneous model to show that the amount of people who face borrowing limit and total wealth in the economy matters regarding fiscal multipliers. Ferriere and Navarro () and Oh and Reis () claim by transfering resources from rich people to poor people, government can effectively boost the economy. The paper is also related to the heterogeneous agent New Keynesian literature (McKay and Reis (6),Gornemann, Kuester, and Nakajima (),Kaplan, Moll, and Violante (6), Ravn, Sterk, et al. () and Kaplan and Violante ()). The rest of paper is organized as follows. In section, I will introduce a heterogeneous agent New Keynesian model with patient and impatient households. Section contains the calibration of the parameters. Section discusses the results. Section concludes. The model The economy is populated by two groups of households. The first group is more patient and have access to the complete asset market. This group of agents can be treated as a representative agent since they can insure against all the idiosyncratic risk. The other is less patient and can only trade one-period risk-free bond. Besides, they have idiosyncratic labor income risk.. Patient household s problem The representative patient household choose{c t, n t } to maximize the subject to E t= [ β t n +ψ ] t log c t ψ + ψ [ ] p t (( + τt c b t )c t + k t+ ) + b t+ b t = p t i t + d t + ( τ l p t)w t sn t + ( + r t )k t + T t t () () where k is capital, b is nominal bond holdings, d is a dividend from owning the intermediate goods firms, w is the aggregate wage, and s is the skill of the patient household. τ l t is a proportional labor income tax, τ c t is the consumption tax, and T t is the government transfer.

4 Euler equations for patient household are ψ n ψ t = ( + τt c)c ( τt)w l t s () t [ ] + i t ( + τt c)c = βe t t π t+ ( + τt+ c )c () t+ [ ] ( ) ( + τt c)c = βe t + rt+ t ( + τt+ c )c () t+ Finally, notice that the patient household s stochastic discount factor is λ t,t+s = β s ( + τ c t+s)c t ( + τ c t )c t+s (6). Impatient households problem There is a measure ν of impatient households indexed by h [, ν], so that an individual variable, say consumption, will be denoted by c t (h). They have the same period utility function as patient households, but they are more impatient: β < β. I assume that the impatient households do not own shares in the firms or own the capital stock. However, their savings can be used to finance capital accumulation by lending to the patient households through bonds market. Individual impatient households choose consumption, hours of work, and bond holdings {c t (h), n t (h), b t+ (h)} to maximize: subject to E t= [ β t n t (h) +ψ ] log c t (h) ψ + ψ p t ( + τ c t )c t (h) + b t+ (h) = ( + i t )b t (h) + p t [( τ l t)w t s t (h)n t (h) + T t ] (8) b t+ (h) (9) Unlike patient households, impatient households face uninsurable idiosyncratic risk to their skill, s t (h). I assume that s follows a Markov chain where the transition matrix Ω = ω ω ω ω ω ω ω ω ω (7)

5 . Final goods producers A competitive sector for final goods combines intermediate goods according to the production function: ( µ y t = y t (j) dj) /µ () where y t (j) is the input of the j th intermediate input. µ > is the desired markup, the elasticity of substitution. µ µ is The representative firm in this sector takes as given the final-goods price p t, and pays p t (j) for each of its inputs. Cost minimization together with zero profits imply that:. Intermediate firm ( ) pt (j) µ/( µ) y t (j) = y t () p t ( ) µ p t = p t (j) /( µ) () There is a unit measure of intermediate-goods monopolistic firms, indexed by j [, ]. Their production function is: y t (j) = a t k t (j) α l t (j) α () where a t is productivity, which follows AR(), i.e. a t = ρ a a t + ε A t, ε A t is capital used, and l t (j) is effective labor. N(, σ A ), k t (j) Intermediate firms set prices subject to nominal rigidities a la Calvo (98) with a probability of price revision θ. max E t [θ p t subject to: ] ( θ) s λ t,t+s d t+s (j) s= () d t (j) p t(j) y t (j) w t l t (j) (r t + δ)k t (j) () p t ( ) pt (j) µ/( µ) y t (j) = y t () p t y t (j) = a t k t (j) α l t (j) α () Since the intermediate firms are owned by the patient households, they discounted future profits using the (6).

6 . Monetary and fiscal policy The government levies a proportional labor-income tax (τ l ), lump-sum tax (T ) and consumption tax (τ C ) to finance government spending and interest payments on this debt. The government budget constraint is G t + + i [ ν ] [ ν ] t B t = B t+ + τ l π tw t s t (h)n t (h)dh + sn t + τt c c(h)dh + c t T t ( + ν) t (6) Monetary policy follows a simple Taylor rule: i t = ī + φ M π t z M t (7) where φ M determines the degree of nominal interest rate response from the change in price level. z M t follows AR(), i.e. z M t = ρ z z M t + εm t, ε M t N(, σ M ). where The fiscal rules are ( Bt ) φg G t = Ḡ zt B G (8) ( ) φτ τt l = τ l Bt l z B τ l (9) ( ) φτ c τt c = τ c Bt z τ c t () T t = T B ( Bt B z G t follows AR(), i.e. z G t = ρ z z G t + εg t, ε G t N(, σ G ). z τ l t z τ c t follows AR(), i.e. z τ l t = ρ z z τ l t + ετ l t, ε τ l ) φt z T t () t N(, σ τ l ). follows AR(), i.e. z τ c t = ρ z z τ c t + ετ c t, ε τ c t N(, σ τ c ). zt T follows AR(), i.e. zt T = ρ z zt T + εt t, ε T t N(, σ T ). According to Leeper (99), I define two monetary and fiscal regimes here. In regime M, the monetary authority raises the interest more than one for one increase in inflation and fiscal authority passively adjusts surplus to fulfill the government budget constraint. In regime F, the monetary authority raises the interest less than one for one increase in inflation and fiscal authority actively adjusts surplus to pin down the price level in the economy. In this paper, I don t provide the specific range of parameter values. But, I will provide the some particular set of parameter values which satisfy the description of each regime. 6

7 .6 Market clearing The labor market clearing condition is l t (j)dj = v s t (h)n t (h)dh + sn t () The demand for labor from intermediate firms is equal to the sum of effective labor supply by households. Each period the government issues debt, B t. The market for debt clears if B t = Goods market clearing condition is.7 Equilibrium v b t (h)dh + b t () ν c t + c t (h)dh + G t + k t+ ( δ)k t = y t () An equilibrium in this economy is a collection of aggregate quantities (y t, k t, d t, c t, n t, b t+ ); aggregate price (p t, r t, w t ); impatient household policy functions (c t (h)), n t (h); a distribution of household over assets, skill; individual firm variables (y t (j), p t (j), k t (j), l t (j), d t (j)); and government instruments (B t, i t, G t, τ C t, τ l t, T t ) such that: patient households maximize () subject to the budget constraint (), impatient households maximize (7) subject to the budget constraint (8), final good producers behaves according to () and (), intermediate goods firm solve () subject to () and (), fiscal policy follows the (8)-() and maintains government budget constraint (6), and monetary policy follows (7), market clears for labor (), capital, bonds () and goods (). the distribution of household over asset and skills evolves in a mannaer consistent with shocks and decision rules. Calibration Table shows the calibration of parameters. All the parameters related to the policy rules (φ s) will be specified in each case. The transition matrix of skill shock is calibrated such 7

8 Symbol Parameter Value Source (Target) β Discount factor of patient HH.989 McKay and Reis (6) α capital share.96 McKay and Reis (6) µ desired price markup. Basu and Fernald (997) β Discount factor of impatient HH.979 McKay and Reis (6) s Skill level of patient HH ν population of impatient HH θ Calvo price stickiness /. average duration =. ψ Labor supply.7 Average hours worked. ψ Labor supply Frisch elasticity = / (s s s ) Skill units of impatient HH ( ) Table : Calibration of parameter that there are the equal proportion of impatient households with each skill at the steady state. Results. Steady State The steady state is calculated when the aggregate shocks are all shut down, which is close to the steady state of Aiyagari (99). Figure shows the individual saving policy functions of the impatient household. For the worst skill (s in figure ) case, Since no borrowing is allowed for impatient household, households will choose to consume all wealth and save nothing when their wealth is low. The group of people, who behave like hand-to-mouth agents, have marginal propensity to consume (MPC). Figure shows the ergodic distribution of wealth for impatient households, I find that the group of households mentioned above account for around of the impatient household population at the steady state.. Lump-sum tax financing In this section, I restrict the fiscal authority to only use lump-sum tax to finance its deficits(φ T =., φ τ C = φ τ l = ), and monetary authority behave actively by raising nonimal interest rate by more than one for one inflation (φ M =.). Figure shows The impulse response of aggregate variables for one standard deviation of government spending shock. As government spending increases, output, aggregate labor and inflation increases. The nominal interest rate increases to fight against inflation. But the consumption is crowded out by the increases in government spending. 6 of total population 8

9 However, not all households in the economy consume less. Figure shows the the impulse response of the consumption for different wealth cohort for one standard deviation of government spending shock. The first quintile of impatient households consumes most of the increase in wage income due to high MPC, on the other hand, the rest of groups choose to save more against future increases in taxes. In a Ricardian Equivalence economy, the increase in a lump-sum transfer which is financied by the future lump-sum tax will have no impact on the choice of comsumption. I treat the lump-sum tax as a non-distorted tax. However, it is not the case for this model economy. Figure shows the impulse response of aggregate variables for one standard deviation lump-sum transfer shock. The increase in transfer will boost the consumption and inflation in the short run. However, total investment decreases, and output and aggregate labor barely increase. In the disaggregated level, according to figure 6, the influences of the lump-sum transfer shocks are different. The first quintile impatient households consume more, since their MPC is relatively high. On the other hand, the other groups save more for the following two reasons. First, their MPC is relatively low. Second, the saving appears to be more attractive comparing to Ricardian Equivalence economy since the first quintile of households group barely saves the extra lump-sum transfer. Specifically, the increase in consumption of the first quitile is stronger than the second quintile. The rest of groups show the decrease in consumption, and the more wealthy the households are, the more decrease in consumption they have.. Labor income tax financing In this section, I restrict the fiscal authority to only use labor income tax to finance its deficits (φ τ l =., φ τ C = φ T = ), and monetary authority still behaves actively (φ M =.). Figure 7 shows The impulse response of aggregate variables for one standard deviation of government spending shock. The effects are similar with the lump-sum tax financing case qualitatively. Output and aggregate labor supply increase and aggregate consumption is crowded out. Future labor income tax discourages the labor supply more in the long run since it is more distorted comparing to the lump-sum tax. The figure 8 shows the impulse response of the consumption for different wealth cohort for one standard deviation of government spending shock. I observe the similar pattern as figure. The increase in government spending boosts the consumption only in the first quintile among the impatient household population, but for all the other groups, consumption drops. Furthermore, the more wealthy the households are, the more severe the drop in consumption there are among them. Comparing figure, figure 9 shows some differences in the dynamics of both the labor 9

10 supply and the aggregate output. Since the labor income tax distorts the labor choice more than the lump-sum tax, both output and labor supply drops. All the other aggregate variables remain the same signs. Also, figure 6 and figure are similar. However, the drops in comsumption are more persistent for wealth rich groups, including third quintile, fourth quintile, fifth quintile and patient households, in the long run.. inflation financing (regime F) In this section, I restrict the fiscal authority to only use inflation to finance its deficits(φ τ l = φ τ C = φ T = ), and monetary authority pegs nominal interest rate (φ M = ). This is corresponding to the regime F mentioned in the literature (Leeper (99)). Figure shows The impulse response of aggregate variables for one standard deviation of government spending shock. As the government spending increases, the output, consumption inflation and labor increases. The nominal interest rate is unchanged. Since the nominal interest is pegged, all the increase in inflation happens in the short run. In a disaggregated level, according to figure, I find that the consumption of patient household drops, since their savings suffer more loss from an increase in inflation than rest of groups. Figure shows The impulse response of aggregate variables for one standard deviation of lump-sum transfer shock. Transfer shock seems to be more effective to boost the economy, comparing to figure. In the disaggregated level, according to figure in the short run consumption increases for all the groups. However, the consumption of patient household drops below zero sharply. Consumption of all the groups remains to be above zero for the rest of periods afterward. Figure -8 shows the case when φ M =.. In regime F, the monetary policy controls the timing of inflation. As φ increases, more inflation will happen in the future instead of now. Patient households do not have the initial increase in the consumption. Also, their consumption takes more time to converge back to the steady state.. Fiscal multipliers In this section, I want to quantify the short run and long run effect of fiscal shocks by calculating the fiscal multiplier. For the short run, I calculate the impact multiplier, which is the change in the variables at t= divided by the change in the fiscal shock at t=. For the long run, I calculate the present value multiplier, which is the present values of the change in the variable divided by the present values of the change in the fiscal shocks. In table A, I calculate the multipliers for output, aggregate consumption and consumption for each wealth group in four different cases mentioned above. I can observe many inter-

11 esting patterns in this table. For now, I will just pick three sailent features in those tables. First, the first quintile of impatient households consumption impact multipliers is larger than any other group. Second, fiscal multipliers depend on the fiscal financing. Comparing to labor income tax financing, the lump-sum tax financing generates larger both output and consumption multipliers. Inflation financing delivers the largest aggregate consumption multipliers among all cases. Third, long run effect and short run effect of fiscal shocks are different qualitatively. Conclusion In this paper, I show that the impact of fiscal shocks is different for different wealth groups. The heterogeneous impact depends on the ways of fiscal financing. This provides the policymakers different views of looking at the effectiveness of fiscal shocks. The future research may focus on the specific transmission mechanisms of heterogeneous impact of fiscal shocks.

12 References S Rao Aiyagari. Uninsured idiosyncratic risk and aggregate saving. The Quarterly Journal of Economics, pages 69 68, 99. Susanto Basu and John G Fernald. Returns to scale in us production: Estimates and implications. Journal of political economy, ():9 8, 997. Pedro Brinca, Hans A Holter, Per Krusell, and Laurence Malafry. Fiscal multipliers in the st century. Journal of Monetary Economics,. Guillermo A Calvo. Staggered prices in a utility-maximizing framework. Journal of monetary Economics, ():8 98, 98. Axelle Ferriere and Gaston Navarro. The heterogeneous effects of government spending: It s all about taxes.. Jordi Galí, J David López-Salido, and Javier Vallés. Understanding the effects of government spending on consumption. Journal of the European Economic Association, ():7 7, 7. Nils Gornemann, Keith Kuester, and Makoto Nakajima. Monetary policy with heterogeneous agents.. Jonathan Heathcote. Fiscal policy with heterogeneous agents and incomplete markets. The Review of Economic Studies, 7():6 88,. Ethan Ilzetzki, Enrique G Mendoza, and Carlos A Végh. How big (small?) are fiscal multipliers? Journal of Monetary Economics, 6():9,. Tullio Jappelli and Luigi Pistaferri. Fiscal policy and mpc heterogeneity. American Economic Journal: Macroeconomics, 6():7 6,. Greg Kaplan and Giovanni L Violante. A model of the consumption response to fiscal stimulus payments. Econometrica, 8():99 9,. Greg Kaplan, Benjamin Moll, and Giovanni L Violante. Monetary policy according to hank. Technical report, National Bureau of Economic Research, 6. Eric M Leeper. Equilibria under activeand passivemonetary and fiscal policies. Journal of monetary Economics, 7():9 7, 99. Eric M Leeper, Nora Traum, and Todd B Walker. Clearing up the fiscal multiplier morass: Prior and posterior analysis. Technical report, National Bureau of Economic Research,.

13 Alisdair McKay and Ricardo Reis. The role of automatic stabilizers in the u.s. business cycle. Eonometrica, 8(): 9, 6. Hyunseung Oh and Ricardo Reis. Targeted transfers and the fiscal response to the great recession. Journal of Monetary Economics, 9:S S6,. Morten Ravn, Vincent Sterk, et al. Job uncertainty and deep recessions. Centre for Macroeconomics (CFM) Discussion Papers,,. Michael Reiter. Solving heterogeneous-agent models by projection and perturbation. Journal of Economic Dynamics and Control, ():69 66, 9. Jose-Victor Rios-Rull and Moritz Kuhn. update on the us earnings, income, and wealth distributional facts: A view from macroeconomics. Quarterly Review, (April): 7, 6. A Solution method I follow Reiter (9) to solve the equilibrium. There are steps. Discretize the saving policy function x(h) and cross-sectional distribution of wealth F x for impatient household. Compute the steady state of the economy when the aggregate shock is zero. But the idiosyncratic shocks s i,t remain the same as the full model. Compute a first-order perturbation of the steady state solutiono in all variables of the model. The solution will be linear in aggregate state variables but nonlinear in the individual shocks. Table : Government Spending Impact Multiplier Y C C C C C C Ce 6 lump-sum tax labor tax inflation (φ M = ) inflation (φ M =.) first quintile of impatient household consumption; second quintile of impatient household consumption; third quintile of impatient household consumption; fourth quintile of impatient household consumption; fifth quintile of impatient household consumption; 6 patient household consumption.

14 Table : Transfer Impact Multiplier Y C C C C C C Ce 6 lump-sum tax labor tax inflation (φ M = ) inflation (φ M =.) first quintile of impatient household consumption; second quintile of impatient household consumption; third quintile of impatient household consumption; fourth quintile of impatient household consumption; fifth quintile of impatient household consumption; 6 patient household consumption. Table : Government Spending Present Value Multiplier Y C C C C C C Ce 6 lump-sum tax labor tax inflation (φ M = ) inflation (φ M =.) first quintile of impatient household consumption; second quintile of impatient household consumption; third quintile of impatient household consumption; fourth quintile of impatient household consumption; fifth quintile of impatient household consumption; 6 patient household consumption. Table : Transfer Present Value Multiplier Y C C C C C C Ce 6 lump-sum tax labor tax inflation (φ M = ) inflation (φ M =.) first quintile of impatient household consumption; second quintile of impatient household consumption; third quintile of impatient household consumption; fourth quintile of impatient household consumption; fifth quintile of impatient household consumption; 6 patient household consumption.

15 .6 Individual policy function s. s s. asset t asset t Figure : Individual saving policy functions of impatient household distribution distribution.... Ergodic Distribution All types asset s asset distribution.. s asset distribution.. s asset Figure : The ergodic distribution of bond holding for impatient agents at the steady state

16 .. Output -. Consumption. Inflation labor input (skill-weighted)..... Nominal interest rate spending Figure : The impulse response of aggregate variables for a government spending shock financed by lump-sum tax only Patient HH Consumption Impatient HH st quintile 6 Impatient HH nd quintile Impatient HH rd quintile Impatient HH th quintile Impatient HH th quintile Figure : The impulse response of the consumption for different wealth cohort for a government spending shock financed by lump-sum tax only 6

17 Output Consumption Inflation labor input (skill-weighted) 6 Nominal interest rate Lump-sum transfer Figure : The impulse response of aggregate variables for a lump-sum transfer shock financed by lump-sum tax only Patient HH Consumption Impatient HH st quintile Impatient HH nd quintile Impatient HH rd quintile Impatient HH th quintile Impatient HH th quintile Figure 6: The impulse response of the consumption for different wealth cohort for a lumpsum transfer shock financed by lump-sum tax only 7

18 .... Output Consumption.. Inflation labor input (skill-weighted)... Nominal interest rate...8 spending Figure 7: The impulse response of aggregate variables for a government spending shock financed by labor income tax only Patient HH Consumption Impatient HH st quintile 6 Impatient HH nd quintile Impatient HH rd quintile - Impatient HH th quintile - Impatient HH th quintile Figure 8: The impulse response of the consumption for different wealth cohort for a government spending shock financed by labor income tax only 8

19 Output Consumption. Inflation labor input (skill-weighted) -. Nominal interest rate. Lump-sum transfer Figure 9: The impulse response of aggregate variables for a lump-sum transfer shock financed by labor income tax only Patient HH Consumption - - Impatient HH st quintile Impatient HH nd quintile.. -6 Impatient HH rd quintile Impatient HH th quintile Impatient HH th quintile Figure : The impulse response of the consumption for different wealth cohort for a lumpsum transfer shock financed by labor income tax only 9

20 . Output. Consumption 8 Inflation labor input (skill-weighted) Nominal interest rate spending -. - Figure : The impulse response of aggregate variables for a government spending shock financed by inflation only Patient HH Consumption - Impatient HH st quintile. Impatient HH nd quintile Impatient HH rd quintile.. Impatient HH th quintile Impatient HH th quintile Figure : The impulse response of the consumption for different wealth cohort for a government spending shock financed by inflation only

21 . Output Consumption. Inflation labor input (skill-weighted) Nominal interest rate. Lump-sum transfer Figure : The impulse response of aggregate variables for a lump-sum transfer shock financed by inflation only Patient HH Consumption Impatient HH st quintile. Impatient HH nd quintile Impatient HH rd quintile. Impatient HH th quintile Impatient HH th quintile.... Figure : The impulse response of the consumption for different wealth cohort for a lumpsum transfer shock financed by inflation only

22 ..8.6 Output 6 Consumption 8 6 Inflation labor input (skill-weighted).8.6 Nominal interest rate..8 spending Figure : The impulse response of aggregate variables for a government spending shock financed by inflation only and φ M =. Patient HH Consumption - Impatient HH st quintile. Impatient HH nd quintile Impatient HH rd quintile Impatient HH th quintile Impatient HH th quintile.. - Figure 6: The impulse response of the consumption for different wealth cohort for a government spending shock financed by inflation only and φ M =.

23 . Output Consumption Inflation labor input (skill-weighted) Nominal interest rate. Lump-sum transfer Figure 7: The impulse response of aggregate variables for a lump-sum transfer shock financed by inflation only and φ M =. Patient HH Consumption - Impatient HH st quintile. Impatient HH nd quintile Impatient HH rd quintile Impatient HH th quintile. Impatient HH th quintile..... Figure 8: The impulse response of the consumption for different wealth cohort for a lumpsum transfer shock financed by inflation only and φ M =.