The General Equilibrium Effects of Fiscal Policy with Government Debt Maturity

Abstract

This paper highlights the importance of accounting for both the maturity structure of government debt and the composition of fiscal instruments when studying the macroeconomic effects of fiscal policy. Using a dynamic stochastic general equilibrium (DSGE) model featuring a debt maturity structure and six exogenous fiscal shocks spanning both the expenditure and revenue sides, we show that long-maturity debt systematically weakens the expansionary effects of fiscal policy under dovish monetary policy, particularly in response to increases in government purchases, government investment, and capital income tax cuts, where long-term financing leads to the significant crowding-out of private activity. In contrast, short-term debt financing yields output multipliers that often exceed unity. The maturity structure also alters the relative efficacy of fiscal instruments: while labor income tax cuts produce the largest multipliers under short-term debt, government purchases become more potent under long-term debt financing. We also show that the stark difference between short- and long-term debt becomes muted under a hawkish monetary regime. Our results have important policy implications, suggesting that the maturity composition of public debt should be carefully considered in the design of fiscal policy, particularly in high-debt economies.

1. Introduction

Government debt management is now a pressing issue for fiscal policy, not only due to elevated debt levels but also with regard to the evolving maturity structure of government debt, an aspect that has received relatively little attention in the literature. A recent prominent work by De Graeve and Mazzolini (2023) shows a long-run positive co-movement between government debt levels and debt maturity for almost all OECD countries. Figure 1 depicts the average maturity of outstanding US government debt since 1960. The data shows the active reshuffling of debt maturity over time, with maturities shortening prior to the mid-1970s and lengthening markedly thereafter. Since the 2008 global financial crisis (GFC), we have seen a rising average debt maturity alongside historically large debt issuance.

This paper studies how the maturity structure of government debt influences the macroeconomic effects of fiscal policy and the magnitude of fiscal multipliers. To exactly address this question, we develop a dynamic stochastic general equilibrium (DSGE) model augmented with a maturity structure of government debt and a rich set of fiscal policy instruments both on the revenue and expenditure sides of the US government budget. The use of DSGE models in fiscal policy analysis has gained momentum recently. As noted by Sbordone et al. (2010), the general equilibrium feature and clear specification of stochastic shocks in DSGE models imply an important interaction between policy actions and households or firms’ behavior, thus making DSGE models a reliable tool for evaluating structural changes in alternative policy instruments. To reflect the prolonged period of low interest rates in the US since the GFC, our benchmark specification adopts a dovish monetary policy stance.1 On the expenditure side, we distinguish among government consumption (i.e., government purchases of goods and services), investment and employment. On the revenue side, our model includes distortionary taxes on labor income, capital income, and consumption.

We compare the responses of the main macro variables to six fiscal shocks in an economy with one-period public debt, the typical assumption in the literature, and in an alternative setting with long-term debt. Under accommodative monetary policy, we find that fiscal expansions, whether via higher spending or tax cuts, have uniformly positive effects on private consumption, investment (though small and short lived following government purchases and investment shocks), total employment, and real wages when debt is single-period. By contrast, the presence of long-maturity debt systematically weakens the expansionary effects across all fiscal instruments, with particularly pronounced crowding-out following increases in government purchases, investment, and reductions to capital income tax rates. This is mainly due to an additional channel for debt revaluation: a dramatic drop in bond prices at the time of the shock absorbs the fiscal expansion and postpones higher inflation into the future. This tradeoff is not present if there is only short-term debt.2

Differences in impulse responses translate directly into differences in fiscal multipliers. When fiscal expansions are financed with short-term debt, output multipliers are substantially larger, often exceeding unity, as documented in many theoretical models (see Woodford, 2011; L. Christiano et al., 2011; Zubairy, 2014; Jo & Zubairy, 2025). In contrast, long-term debt financing produces multipliers that are much lower, and frequently negative especially for private consumption and investment, consistent with the findings of Ghomi et al. (2025). Debt maturity also alters the relative efficacy of different fiscal policy instruments. With short-term debt, labor income tax cuts yield the highest multipliers on output, consumption, investment and employment. With long-term debt, however, increased government purchases generate the largest output and employment multipliers, as well as the highest impact multiplier for consumption. On the other hand, under a hawkish monetary regime, where the central bank responds aggressively to inflation, the stark difference between short-term and long-term debt becomes negligible, highlighting the importance of fiscal-monetary interactions in shaping fiscal outcomes.

Our paper contributes to the vast literature that investigates the effects of fiscal policy on aggregate economic activity and the transmission mechanism of those effects in macroeconomic models. While most theoretical studies predict that output rises following expansionary fiscal shocks, they often differ regarding the implied effects on private activities like consumption, investment and employment (see Edelberg et al., 1999; Burnside et al., 2004; Galí et al., 2007; Bilbiie, 2011; Rannenberg, 2021). We contribute to this literature by explicitly modeling the maturity structure of government debt to examine how the size and sign of fiscal multipliers vary with debt maturity. We demonstrate that the government debt maturity is a key determinant of the transmission of fiscal policy. On the one hand, short-term vs. long-term debt financing can account for differences in fiscal multipliers on private activity like consumption and investment. On the other hand, our finding that fiscal expansions financed with long-maturity debt produce output multipliers less than unity even in a low-interest-rate environment suggests that the typical specification of short-term debt may overstate the expansionary effects of fiscal policy.

Two papers most closely related to ours are Leeper et al. (2017) and Ghomi et al. (2025), both of which cover how debt maturity affects fiscal multipliers in New Keynesian DSGE models. However, their analyses focus only on aggregate government spending shocks without distinguishing among expenditure components. We extend their work in two key ways. First, we introduce a richer set of fiscal instruments that includes six distinct fiscal shocks both on the spending and revenue sides. Second, we assess how debt maturity shapes the transmission of each fiscal instrument, providing a more granular and policy-relevant comparison across fiscal tools. This distinction is critical, as prior research has documented that different types of government spending shocks have dramatically different effects on the economy (see Wynne, 1992; Finn, 1998; Fatás & Mihov, 2001; Pappa, 2009; Ramey, 2012; Boehm, 2020). To the best of our knowledge, this is the first paper that comprehensively studies the role of government debt maturity in determining fiscal transmission and multipliers within a DSGE framework that features both distortionary taxation and a detailed decomposition of government expenditures. This extension not only enhances the empirical realism of the model but also allows us to provide novel policy implications regarding debt management and fiscal policy design.

2. Model Specification

This section presents our model which incorporates a government debt maturity structure, building on the framework of Leeper et al. (2017). Moreover, we extend their work by separately modeling government purchases, investment, and employment, following Finn (1998) and Pappa (2009). In total, our model includes six exogenous fiscal shocks affecting both government expenditures and revenues defined on distortionary tax rates. Nominal rigidities are introduced as adjustment costs for wages and prices, following Rotemberg (1982) and Ireland (1997).

2.1. Households

The economy is populated by a large number of identical households. The lifetime utility of each household i is a separable function of its composite consumption, $c_t^{*}\left(i\right)$, and labor, $n_t\left(i\right)$, given by

$$\begin{array}{c}\hfill E_t\sum _{t=0}^{\infty }{\beta }^t\left(ln\left(c_t^{*}\left(i\right)-h{c}_{t-1}^{*}\right)-{\displaystyle \frac{n_t{\left(i\right)}^{1+\eta }}{1+\eta }}\right)\end{array}$$

(1)

where $\beta \in \left(0,1\right)$ is the discount factor. $h\in \left(0,1\right)$ governs the degree of external habits in consumption. $c_t^{*}\left(i\right)$ consists of private and government consumption goods, $c_t^{*}\left(i\right)\equiv c_t^p\left(i\right)+{\alpha }^g{c}_t^g$, where parameter ${\alpha }^g$ governs the degree of substitutability of the consumption goods. When ${\alpha }^g<0$, private and public goods are complements (Karras, 1994; Bouakez & Rebei, 2007), whereas ${\alpha }^g>0$ implies that these are substitutes with each other (L. J. Christiano & Eichenbaum, 1992; Ambler & Paquet, 1996). The case of complementarity will be the relevant one below. Government consumption is regarded as exogenous.

Households receive after-tax wage and rental income, lump-sum transfers from the government, $P_t{z}_t\left(i\right)$, and profits from firms, $D_t$. They spend income on private consumption, $c_t^p\left(i\right)$, investment in future capital, $i_t^p\left(i\right)$, and on government bonds. The nominal flow budget constraint of households is given by

$$\begin{array}{c}\hfill \begin{array}{cc}\hfill & P_t(1+{\tau }_t^c)c_t^p\left(i\right)+P_t{i}_t^p\left(i\right)+R_t^{-1}{B}_t^s\left(i\right)+P_t^m{B}_t^m\left(i\right)+{\displaystyle \frac{{\varphi }^w}{2}}{\left({\displaystyle \frac{W_t\left(i\right)}{W_{t-1}\left(i\right)}}-{\overline{\pi }}^w\right)}^2{W}_t\le \hfill \\ \hfill & B_{t-1}^s\left(i\right)+\left(1+\rho P_t^m\right)B_{t-1}^m\left(i\right)+(1-{\tau }_t^n)W_t\left(i\right)n_t\left(i\right)+\left[\left(1-{\tau }_t^k\right)R_t^k{u}_t\left(i\right)-a\left(u_t\right)\right]k_{t-1}^p\left(i\right)+D_t\left(i\right)+P_t{z}_t\left(i\right)\hfill \end{array}\end{array}$$

(2)

Nominal private consumption, $P_t{c}_t^p\left(i\right)$, is subject to a sales tax ${\tau }^c$. $W_t\left(i\right)$ is the nominal wage, and the labor income $W_t\left(i\right)n_t\left(i\right)$ is taxed at the rate ${\tau }^n$. Following J. Kim (2000), wage rigidities are introduced through the cost of adjusting nominal wages, assumed to be quadratic and zero at the steady state. ${\overline{\pi }}^w$ is the steady-state wage inflation, which is assumed to be equal to price inflation $\overline{\pi }$. Rental income on capital, $R_t^k{u}_t\left(i\right)k_{t-1}^p\left(i\right)$3, is taxed at the rate ${\tau }^k$.

There are two types of bonds: one-period nominal private bonds, $B_t^s\left(i\right)$, in zero net supply with price $R_t^{-1}$, where $R_t$ is the gross nominal interest rate set by the monetary authority, and a more general portfolio of nominal government bonds, $B_t^m\left(i\right)$, in non-zero net supply with price $P_t^m$. Following Woodford (2001), all government bonds consist of perpetuities with coupon payments that decay exponentially at a constant rate $\rho$ so that bond issued in period t pays ${\rho }^T$ dollars $T+1$ periods later for $T\ge 0$ and decay factor $0\le \rho <{\beta }^{-1}$. Varying the parameter $\rho$ varies the average maturity of government debt. The classic “consol” is a security with $\rho =1$.

$u_t\left(i\right)$ denotes the utilization rate of capital. Choosing capital utilization incurs a cost equal to $a\left(u_t\right)$ per unit of capital, which is parameterized by a quadratic function $a\left(u_t\right)={\zeta }_1\left(u_t-1\right)+{\displaystyle \frac{{\zeta }_2}{2}}{\left(u_t-1\right)}^2$4. Note that $u=1$ and $a\left(1\right)=0$ in the steady state. We also define ${\displaystyle \frac{a^{″}\left(1\right)}{a^{\prime }\left(1\right)}}\equiv {\displaystyle \frac{{\zeta }_2}{1-{\zeta }_2}}$ following Smets and Wouters (2007).5 Private capital accumulates according to

$$\begin{array}{cc}\hfill k_t^p\left(i\right)& =\left(1-\delta \right)k_{t-1}^p\left(i\right)+\left[1-S\left({\displaystyle \frac{i_t^p\left(i\right)}{i_{t-1}^p\left(i\right)}}\right)\right]i_t^p\left(i\right)\hfill \end{array}$$

(3)

where ${\delta }^p$ is the depreciation rate and $S\left(·\right)={\displaystyle \frac{\kappa }{2}}{\left({\displaystyle \frac{i_t}{i_{t-1}}}-1\right)}^2$ denotes an adjustment cost function, proposed by L. J. Christiano et al. (2005) such that $S\left(1\right)=S^{\prime }\left(1\right)=0$, and $S^{″}\left(1\right)\equiv \kappa >0$.

Government Debt Maturity Structure

Let ${\widehat{x}}_t=ln{x}_t-lnx$, where $lnx$ is the steady-state value for the variable $x_t$. Then combining household’s Euler equations for two types of government debt gives the following log-linearized debt maturity structure equation:

$$\begin{array}{cc}\hfill {\widehat{R}}_t& =-\left({\widehat{P}}_t^m-{\displaystyle \frac{\rho \beta }{\overline{\pi }}}{E}_t{\widehat{P}}_{t+1}^m\right)\hfill \end{array}$$

which represents an equilibrium restriction on the expected movements of asset prices. Solving this equation forward and using transversality determines the following relation as

$$\begin{array}{cc}\hfill {\widehat{P}}_t^m& =-\sum _{j=0}^{\infty }{\left({\displaystyle \frac{\rho \beta }{\overline{\pi }}}\right)}^j{E}_t{\widehat{R}}_{t+j}\hfill \end{array}$$

The long-term debt is priced as the expected present discounted value of all future one-period interest rates. The average maturity of the debt, defined in Macauley (1938), is the weighted average of time until each of the debt’s future payment, with the weights determined by the proportion of the debt’s present value on each payment date. Hence, in the model, the average maturity for long-term debt equals

$$\begin{array}{cc}\hfill AD& =\sum _{j=0}^{N}j·{\displaystyle \frac{{\left(\frac{\rho \beta }{\overline{\pi }}\right)}^j{E}_t{\widehat{R}}_{t+j}}{{\sum }_{j=0}^N{\left(\frac{\rho \beta }{\overline{\pi }}\right)}^j{E}_t{\widehat{R}}_{t+j}}}=\sum _{j=0}^{N}j·{\displaystyle \frac{{\left(\frac{\rho \beta }{\overline{\pi }}\right)}^j}{{\sum }_{j=0}^N{\left(\frac{\rho \beta }{\overline{\pi }}\right)}^j}}={\displaystyle \frac{1}{1-\frac{\rho \beta }{\overline{\pi }}}}\hfill \end{array}$$

Then the parameter $\rho$ can be written as

$$\begin{array}{cc}\hfill \rho & =\left(1-{\displaystyle \frac{1}{AD}}\right){\displaystyle \frac{\overline{\pi }}{\beta }}\hfill \end{array}$$

Given the average maturity of government debt, $AD$, $\rho$ will be determined by the steady-state inflation rate and the discount factor rate. A central focus of our analysis will be the consequences of variations in the maturity structure of government debt for the effects of fiscal instruments.

Both empirical and theoretical studies have shown that the fiscal multiplier can exceed unity when the government finance its deficit by issuing short-term debt (see Blanchard & Perotti, 2002; Zubairy, 2014; Broner et al., 2022 and Jo & Zubairy, 2025). We demonstrate that these sizable multipliers can decrease significantly when the average maturity of government debt is longer. The rationale is that for government debt to be stabilized, large increases in inflation and output are typically needed. With longer debt maturity, however, inflation adjustments can be spread over a longer horizon. Thus, current inflation rises by less than in the case of short-term debt financing. This leads to a smaller decline in the real interest rate, weakening the intertemporal substitution effect that ordinarily stimulates aggregate demand. In subsequent sections, we present our model simulations to formally analyze how debt maturity influences the effectiveness of different fiscal instruments.

2.2. The Labor Market

As in Finn (1998), households supply labor both to private-sector firms and to the government. Total employment can be written as

$$n_t=n_t^p+n_t^g$$

(4)

In the private labor market, a competitive labor agency combines the differentiated labor services into a homogeneous composite labor input that, in turn, is sold to pirate intermediate firms according to

$$n_t^p={\left[{\int }_0^1{n}_t^p{\left(i\right)}^{{\displaystyle \frac{{\theta }^w-1}{{\theta }^w}}}di\right]}^{{\displaystyle \frac{{\theta }^w}{{\theta }^w-1}}}$$

where $0\le {\theta }^w<\infty$ is the elasticity of substitution among different types of labor. The competitive labor agency maximizes its profit subject to the above production function, taking all differentiated labor wages $w_t\left(i\right)$ and the aggregate wage $w_t$ as given, yielding

$$n_t^p\left(i\right)={\left({\displaystyle \frac{W_t\left(i\right)}{W_t}}\right)}^{-{\theta }^w}{n}_t^p$$

(5)

We assume that the wage rate in the government sector is equal to the one in the private sector.6

2.3. Production

2.3.1. Final Good Firm

A perfectly competitive final goods firm produces the final good $y_t$ from a continuum of differentiated intermediate goods $y_t\left(j\right)$. The final goods aggregator is $y_t={\left[{\int }_0^1{y}_t{\left(j\right)}^{\left({\theta }^p-1\right)/{\theta }^p}dj\right]}^{{\theta }^p/\left({\theta }^p-1\right)}$, where $\theta >1$ governs the degree of substitution between the inputs (see Dixit & Stiglitz, 1977). Taking the price of intermediate goods $P_t\left(j\right)$ and final goods $P_t$ as given, the profit maximization of the final goods firm yields the demand schedule $y_t\left(j\right)={\left(P_t\left(j\right)/P_t\right)}^{-{\theta }^p}{y}_t$, where $P_t={\left[{\int }_0^1{P}_t^{1-{\theta }^p}\left(j\right)dj\right]}^{1/1-{\theta }^p}$.

2.3.2. Intermediate Firms

Intermediate goods firms are monopolistically competitive. Following Pappa (2009), each firm j produces output according to

$$\begin{array}{c}\hfill y_t\left(j\right)={\left[n_t^p\left(j\right)\right]}^{1-\alpha }{\left[u_t\left(j\right)k_{t-1}^p\left(j\right)\right]}^{\alpha }{\left(n_t^g\right)}^{\nu }{\left(k_{t-1}^g\right)}^{\mu }\end{array}$$

(6)

where $n_t^p\left(j\right)$ and $u_t\left(j\right)k_{t-1}^p\left(j\right)$ are private labor and effective capital inputs employed by firm j, and $n_t^g$ and $k_{t-1}^g$ are the government employment and capital. The parameters $\nu$ and $\mu$ regulate how government inputs affect private production: When $\nu$ or $\mu$ is zero, the government employment (capital) is unproductive. Each intermediate firm minimizes its operating costs by choosing private labor and capital, taking the input prices $w_t$ and $r_t^k$ and government inputs as given, subject to its production function. All firms are identical and thus rent inputs at the same price. Cost minimization yields

$$\begin{array}{cc}\hfill {\displaystyle \frac{u_t\left(j\right)k_{t-1}^p\left(j\right)}{n_t^p}}& ={\displaystyle \frac{\alpha }{1-\alpha }}{\displaystyle \frac{W_t}{R_t^k}}\hfill \end{array}$$

(7)

$$\begin{array}{cc}\hfill M{C}_t& =\mathsf{\Theta }{W}_t^{1-\alpha }{\left(R_t^k\right)}^{\alpha }{\left(n_t^g\right)}^{-\nu }{\left(k_{t-1}^g\right)}^{-\mu }\hfill \end{array}$$

(8)

where $\mathsf{\Theta }={\alpha }^{\alpha }{\left(1-\alpha \right)}^{1-\alpha }$. $M{C}_t$ is the nominal marginal cost.

As with wage rigidities, each intermediate firm faces a cost to adjusting its price given by ${\displaystyle \frac{{\varphi }^p}{2}}{\left({\displaystyle \frac{P_t\left(j\right)}{P_{t-1}\left(j\right)}}-\overline{\pi }\right)}^2{P}_t{y}_t$ (see Ireland, 1997)7. Each firm chooses its price $P_t\left(j\right)$ to maximize the expected discounted present value of profits

$$\underset{P_t^{*}\left(j\right)}{max}{E}_t\sum _{s=0}^{\infty }{Q}_{t,t+s}\left[P_{t+s}\left(j\right)y_{t+s}\left(j\right)-M{C}_{t+s}{y}_{t+s}\left(j\right)-{\displaystyle \frac{{\varphi }^p}{2}}{\left({\displaystyle \frac{P_{t+s}\left(j\right)}{P_{t+s-1}\left(j\right)}}-\overline{\pi }\right)}^2{P}_{t+s}{y}_{t+s}\right]$$

(9)

subject to the demand for $y_{t+s}\left(j\right)={\left(P_{t+s}\left(j\right)/P_{t+s}\right)}^{-{\theta }^p}{y}_{t+s}$. $Q_{t,t}=1$, $Q_{t,t+s}={\beta }^s{\displaystyle \frac{{\lambda }_{t+s}}{{\lambda }_t}}$ is the stochastic discount factor.

2.4. Fiscal Policy

The government’s flow budget constraint is given by

$$\begin{array}{cc}\hfill P_t^m{B}_t^m+P_t{\tau }_t^c{c}_t^p+{\tau }_t^n{W}_t{n}_t+{\tau }_t^k{R}_t^k{u}_t{k}_{t-1}^p& =\left(1+\rho P_t^m\right)B_{t-1}^m+P_t{c}_t^g+P_t{i}_t^g+W_t{n}_t^g+P_t{z}_t\hfill \end{array}$$

(10)

where the government collects tax revenues and sells debt portfolio to finance its interest payments, consumption and investment purchases, and employees’ compensation. $i_t^g$ is government investment and the government capital stock evolves according to

$$\begin{array}{cc}\hfill k_t^g& =\left(1-\delta \right)k_{t-1}^g+\left[1-S\left({\displaystyle \frac{i_t^g}{i_{t-1}^g}}\right)\right]i_t^g\hfill \end{array}$$

(11)

$S\left(·\right)$ denotes adjustments to government capital, which is the same as in the private sector.

Fiscal rules include a response of fiscal instruments to the market value of the debt-to-GDP ratio to ensure the determinacy of equilibrium, a nonexplosive solution for debt, and an autoregressive term to allow for serial correlation. Fiscal instruments follow the rules

$$\begin{array}{cc}\hfill {\widehat{x}}_t& ={\psi }^x{\widehat{x}}_{t-1}-(1-{\psi }^x){\gamma }^x{\widehat{s}}_{t-1}^b+{\nu }_t^x\hfill \end{array}$$

(12)

$$\begin{array}{cc}\hfill {\widehat{\tau }}_t^{\mathcal{J}}& ={\psi }^{{\tau }^{\mathcal{J}}}{\widehat{\tau }}_{t-1}^{\mathcal{J}}-(1-{\psi }^{{\tau }^{\mathcal{J}}}){\gamma }^{{\tau }^{\mathcal{J}}}{\widehat{s}}_{t-1}^b+{\nu }_t^{{\tau }^{\mathcal{J}}}\hfill \end{array}$$

(13)

$$\begin{array}{cc}\hfill {\widehat{\tau }}_t^c& ={\psi }^{{\tau }^c}{\widehat{\tau }}_{t-1}^c+{\nu }_t^{{\tau }^c}\hfill \end{array}$$

(14)

where $x=c^g,i^g,n^g$, $s_{t-1}^b\equiv {\displaystyle \frac{P_{t-1}^m{B}_{t-1}^m}{P_{t-1}{Y}_{t-1}}}$ and $\mathcal{J}=k,n$. $\psi \in (-1,1)$ governs the degree of the persistence of the process, and $\gamma >0$ triggers a correction of fiscal expansion when the government debt deviates from its stead state value. Consumption tax rates are assumed to follow an exogenous log linear AR(1) process.8 ${\nu }_t^s={\rho }^s{\nu }_{t-1}^s+{\sigma }^s{\epsilon }_t^s$ for $s=\left\{x,{\tau }^{\mathcal{J}},{\tau }^c\right\}$ and ${\epsilon }_t^s\sim \mathbb{N}\left(0,1\right)$.

2.5. Monetary Policy

The monetary authority follows a Taylor rule. It adjusts the gross nominal interest rate ($R_t$) in response to the deviations of inflation (${\pi }_t$) and output ($y_t$) from their respective steady-state levels:

$$\begin{array}{c}\hfill R_t=R_{t-1}^{{\psi }^r}{\left[\overline{R}{\left({\displaystyle \frac{{\pi }_t}{\overline{\pi }}}\right)}^{{\varphi }^{\pi }}{\left({\displaystyle \frac{y_t}{\overline{y}}}\right)}^{{\varphi }^y}\right]}^{1-{\psi }^r}\end{array}$$

(15)

where $0\le {\psi }_r<1$ is the interest rate smoothing parameter, $\overline{R}$ is the equilibrium real interest rate, and ${\varphi }^{\pi }$ and ${\varphi }^y$ are the policy response parameters to the inflation gap and the output gap, respectively.

Market Clearing and Aggregation

The aggregate resource constraint is given by

$$\begin{array}{cc}\hfill y_t& =c_t^p+i_t^p+c_t^g+i_t^g+ad{j}_t\hfill \end{array}$$

(16)

where $ad{j}_t$ stands for real adjustment costs

$$\begin{array}{cc}\hfill ad{j}_t& ={\displaystyle \frac{{\varphi }^w}{2}}{\left({\displaystyle \frac{W_t}{W_{t-1}}}-{\overline{\pi }}^w\right)}^2{\displaystyle \frac{W_t}{P_t}}+{\displaystyle \frac{{\varphi }^p}{2}}{\left({\pi }_t-\overline{\pi }\right)}^2{y}_t+a\left(u_t\right)k_{t-1}^p\hfill \end{array}$$

3. Calibration

We solve the model by approximating the equilibrium conditions around a nonstochastic steady state using the Sims (2002) gensys algorithm. The full list of our parameter choices is given in

Table 1. Parameter calibration.

Parameter Description	Values
Preference and HHs
$\beta$, discount factor	0.99
h, habit formation	0.9
$\eta$, inverse Frisch labor elasticity	2
$\delta$, capital depreciation rate	0.025
${\alpha }^g$, subs. of private/public cons.	−0.25
$k^g/k^p$, steady-state capital ratio	0.34
$n^p/n^g$, steady-state employment ratio	5.02
Frictions and Production
$\alpha$, productivity of private capital	0.33
$\nu$, productivity of govt employment	0.25
$\mu$, productivity of govt capital	0.1
${\theta }^j,j=p,w$, price/wage elasticity of demand	6
$\omega$, Calvo pricing	0.85
${\varphi }^j,j=p,w$, price/wage adjustment cost.	${\displaystyle \frac{\omega ({\theta }^j-1)}{(1-\omega )(1-\omega \beta )}}$
${\zeta }_2$, capital utilization	0.15
$\kappa$, investment adjustment cost	5
Monetary/Fiscal Calibrations
$c^g/y$, steady-state govt purchases-to-GDP ratio	0.052
$i^g/y$, steady-state govt investment-to-GDP ratio	0.074
${\psi }^s,s=c^g,i^g,n^g,{\tau }^c,{\tau }^n,{\tau }^k$, lagged resp. for fiscal instruments	0.98
$s^b$, steady-state market value debt-to-GDP ratio	1.501
${\tau }^c$, steady-state consumption tax rate	0.023
${\tau }^n$, steady-state labor tax rate	0.217
${\tau }^k$, steady-state capital tax rate	0.250
Regime D (benchmark)
Monetary Policy
${\varphi }^{\pi }$, interest rate resp. to inflation	0.83
${\varphi }^y$, interest rate resp. to output	0.27
${\psi }^r$, resp. to lagged interest rate	0.68
Fiscal Policy
${\gamma }^s,s=c^g,i^g,n^g,{\tau }^n,{\tau }^k$, fiscal instruments resp. to debt	0
Regime H
Monetary Policy
${\varphi }^{\pi }$, interest rate resp. to inflation	2.15
${\varphi }^y$, interest rate resp. to output	0.93
${\psi }^r$, resp. to lagged interest rate	0.79
Fiscal Policy
${\gamma }^s,s=c^g,i^g,n^g,{\tau }^n,{\tau }^k$, fiscal instruments resp. to debt	0.05

Note: Parameters are calibrated at a quarterly frequency.

. Our model is calibrated at a quarterly frequency based on the US postwar data from 1960Q1 to 2019Q4.9 The setting chosen for many of the parameters is standard. Thus, the discount factor ($\beta$) is set at $0.99$. Parameter $\eta$, the inverse Frisch labor supply elasticity, is set at 2. The quarterly depreciation rate for capital ($\delta$) is set at $0.025$, which implies an annual depreciation rate of $10\%$. The habit formation coefficient (h)h is set at $0.9$ as in Boldrin et al. (2001). ${\alpha }^g$ is set at $-0.25$, implying complementarity between government and private consumption. The price elasticity of demand for intermediate goods (${\theta }^p$) and labor wage elasticity (${\theta }^w$) are set to be 6. The capital utilization rate (${\zeta }_2$) and the investment adjustment cost ($\kappa$) are set to $0.15$ and 5, respectively. We set the price and wage adjustment cost parameters, ${\varphi }^p$ and ${\varphi }^w$, to match the probability of keeping prices unchanged using Calvo pricing to be 0.85, implying more than one-year average duration of prices/wages being kept unchanged.

Allowing for productive government employment and capital poses some discipline on the calibration of productivity of government employment and capital to prevent unrealistic steady-state values due to increasing returns to scale. Since more recent studies suggest lower values for the productivity of public capital (see Bermperoglou et al., 2017), we use $\mu =0.1$. The productivity of government employment ($\nu$) is restricted to $0.25$ following Pappa (2009). We set the capital ratio ($k^g/k^p$) to $0.34$ using investment in fixed assets from the national income and product accounts (NIPAs) and the private-to-public employment ratio ($n^p/n^g$) to $5.02$ using data from the Current Employment Statistics (CES), both based on the postwar average values for the US. The private capital share in the production function ($\alpha$) is set to $0.33$ as is commonly used in the literature.

Steady-state fiscal and monetary variables are calibrated to match the postwar mean values in the US. The government purchase-to-output ratio ($c^g/y$) is $0.052$. The government investment-to-output ratio ($i^g/y$) is $0.074$. The market value of government debt-to-GDP ratio ($s_b$) is $1.501$. The average labor tax rate (${\tau }^n$) is $0.217$, the capital tax rate (${\tau }^k$) is $0.25$ and the consumption tax rate (${\tau }^c$) is $0.023$. In the benchmark specification (Regime D), we assume a dovish central bank that responds only weakly to inflation to keep the balance between output and inflation, with the long-run coefficients on inflation (${\varphi }^{\pi }$) and output (${\varphi }^y$) set at 0.83 and 0.27, respectively. The interest rate smoothing parameter (${\psi }^r$) is set at 0.68. The fiscal policy in this case does not respond to government debt as indicated by $\gamma =0$.

To understand how monetary policy affects the transmission of fiscal shocks, we also consider an alternative specification (Regime H), where a hawkish central bank reacts aggressively to inflationary pressures. Here, ${\varphi }^{\pi }$, ${\varphi }^y$, and ${\psi }^r$ are set at 2.15, 0.93, and 0.79, respectively. Additionally, the fiscal policy includes a mild response to debt, with $\gamma =0.05$, to ensure equilibrium determinacy. The monetary policy parameters for both regimes are based on Taylor rule estimates from Clarida et al. (2000). Finally, the persistence parameter for all fiscal rules ($\psi$) is set at $0.98$, following the estimates of Leeper et al. (2017).

4. Simulation Results

Figure 2, Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7 display the impulse responses to a 10% increase in different components of government spending and tax rates with 1-, 20-, and 40-period average maturities of government debt. All responses are expressed in percentage deviations from the respective steady-state values. The bottom right panel of each figure shows the path of the shock. We first report the results for the case of one-period debt, the typical assumption in the literature, and with extended debt maturities. In our benchmark model, the monetary policy is assumed to be accommodative. We then discuss how the maturity structure of government debt interacts with fiscal shocks under alternative monetary policy regimes.

4.1. Government Spending Shocks

When there is only short-term debt (solid lines), a sharp decline in real interest rates by the dovish monetary policy in response to all government spending shocks we consider is the dominant force in the transmission mechanism. Lower real rates trigger the typical reactions: households raise consumption and investment demand, which increase output significantly. Although all fiscal shocks are positively autocorrelated, the persistence of their effects (relative to the steady state) varies. Specifically, the positive response of consumption to government purchases shocks die out quickly, while it persists in a more pronounced way after shocks to government employment and investment. It is worth noting that the positive demand effect in the goods market may or may not be prorogated to the labor market depending on the type of government spending shocks. Innovations to government purchases and investment shocks prompt firms to employ more labor to meet production needs, increasing equilibrium total employment through an increase in private employment. Although the labor supply may also increase due to a negative wealth effect, the positive demand effect dominates any other factor at play, pushing real wages up.

In contrast, an exogenous increase in government employment crowds out private employment since it expands the government’s usage of private resources. Although total employment increases, this growth occurs entirely through government employment, reflecting a reallocation of labor from the private to the public sector as noted by Finn (1998). Consequently, real wages rise significantly in response to the sharp decline in labor supply available in the private sector. This crowding-out effect is the main channel through which government employment shocks exert effects on the economy. The negative correlation between private employment and real wages implies an inward shift in the labor supply curve along a relatively stable labor demand curve. In comparison, the positive association between private employment and real wages following shocks to government purchases and investment reflects an outward shift in labor demand along an (almost) fixed labor supply curve.

Long-maturity government debt. The presence of long-maturity debt systematically reduces the probability of large responses, compared to when there is only one-period debt. This is due to an additional channel for debt revaluation: bond prices that reflect expected inflation over the entire duration of debt. Once debt maturity extends beyond a single period, government spending shocks trigger a dramatic drop in bond prices at the time of the shock, which absorbs the fiscal expansion and postpone higher inflation into the future. As a result, the contained current inflation is transformed into less decline in real interest rates by the dovish monetary policy, weakening the interest rate channel through which monetary accommodation stimulates demand. In particular, with an average maturity of 10 years (dotted-dashed lines) debt, all government spending shocks lead to an immediate increase in both short- and long-term real interest rates. This upward pressure on interest rates reduces demand, wages and employment. This finding resonates with prior work by Cochrane (2001), Sims (2013), and Leeper and Leith (2016), all of whom emphasize the role of long-term debt in shaping inflation dynamics.

The inclusion of long-term debt also alters the transmission mechanisms of government spending shocks qualitatively with differential economic effects across different components of government spending. Notably, while consumption rises in response to government purchases shocks when debt is short term, this effect is overturned once debt maturity is extended. Higher government purchases financed by longer-maturity debt lead to an immediate and persistent decline in consumption. This finding is particularly striking, given that our model allows for government purchases complementing private consumption in the household’s utility function, a key channel for government spending raising consumption emphasized in the literature (see Linnemann, 2009; Bermperoglou et al., 2017). Thus, our results demonstrate that even under accommodative monetary policy and complementarity assumption, longer-maturity debt can significantly weaken the stimulative effect of fiscal expansions. As illustrated in Figure 2, the decline in consumption following purchases shocks is a consequence of negative wealth effect: the market value of government debt falls below the steady state substantially due to the precipitous decline in bond prices. Lower consumption demand, in turn, induces firms to employ less labor and capital, which explains the weaker responses of employment and capital utilization compared to when all debt is one-period. Output increases only through higher government purchases. Note that when the debt maturity is extended to 10 years, the negative wealth effect outweighs the positive labor demand effect, resulting in a decline in real wages.

The broader real effects of government investment shocks are qualitatively similar to those of government purchases shocks. As with purchases shocks, the expansionary effect of investment shocks diminishes with longer debt maturities. Quantitatively, however, extending debt maturity reduces the positive impacts of government investment shocks on private economic activity more than it does for government purchases shocks. As shown in Figure 3, private investment falls by nearly twice as much under investment shocks compared to purchases shocks. Moreover, total and private employment responses turn negative when the debt maturity extends, a reversal that does not happen with purchases shocks. This stronger crowding-out effect arises from an immediate and sizable increase in real interest rates following investment shocks under long-term debt. Also, unlike government purchases, government investment is modeled as productivity-enhancing. As a result, increases in government investment generate larger fiscal deficits, requiring the government to issue more debt. Despite falling bond prices, the total market value of outstanding liabilities increases, opposite to the negative debt revaluation after purchases shocks. However, the contractionary effect of higher real interest rates dominates the wealth effect from rising debt valuations, leading to a broad-based decline in private sector activity.

Turning to government employment shocks, a distinct pattern emerges as shown in Figure 4. The crowding-out of private employment by public employment is significantly reinforced by longer debt maturities. With short-term debt, total employment rises, suggesting only partial crowding-out of private jobs. But under longer maturities, total employment falls persistently, suggesting that public employment more than fully crowds out private employment. The tighter labor supply coupled with rising real interest rates even leads to a short-lived decline in output under long-term debt. Interestingly, in contrast to government purchases and investment, consumption rises steadily across all maturities following government employment shocks. On the one hand, real wages increase even when debt is long term, boosting household income. On the other hand, the market value of government debt rises, generating a positive wealth effect. Together, these two forces sustain higher consumption despite less favorable financing conditions.

Overall, our findings reveal a common constraint imposed by long-term debt on the effectiveness of fiscal stimulus for all government spending shocks. At the same time, the transmission of fiscal shocks depends critically on the type of government spending. These results highlight the importance of accounting for both the maturity structure of government debt and the composition of government expenditures when evaluating the macroeconomic consequences of fiscal policy.

4.2. Shocks to Tax Rates

Next we look at the effects of tax rate innovations. Figure 5, Figure 6 and Figure 7 plot the impulse responses to a negative 10% shock to the tax rate on, respectively, labor income, capital income and consumption.

An exogenous reduction in tax rates lowers expected tax revenues, raises households’ wealth, and increases aggregate demand and output. As with government spending shocks, the stimulating effects are strong when there is only short-term debt, where the dovish monetary policy leads to a sharp decline in real interest rates. In particular, a reduction in the labor income tax rate boosts disposable income, encouraging consumption, while also lowering the distortionary wedge in the labor market. This leads to an outward shift in labor supply. Firms increase the output to meet the additional aggregate demand and they do so by increasing the labor and capital demand, driving up real wages and capital price. The resulting positive comovement between private employment and real wages implies that labor demand dominates the increase in labor supply, at least under short-term debt.

The general pattern of responses to lower labor income tax rates carries over to responses to capital and consumption tax reductions but with some key differences. A reduction in the capital tax rate primarily affects the economy by raising the after-tax return on capital, which stimulates investment. It also leads to an outward shift in capital supply, putting downward pressure on the rental rate of capital and encouraging firms to expand their capital stock over time. Differently, a reduction in the consumption tax rate directly increases the households’ real purchasing power by lowering consumer prices, which boosts consumption. However, consistent with the model-based literature, we find that the aggregate impact of consumption tax cuts is modest, partly due to the relatively small share of consumption taxes in the US GDP.

Long-maturity government debt. As with government spending shocks, including the long-term debt not only weakens the stimulative effects of exogenous decreases in tax rates but alters the shock transmission by dampening the decline in real interest rates and shifting more of the fiscal burden into the future as well. Whether the overall impact of tax reductions under long-term debt remains mildly expansionary or turns contractionary depends on the relative strength of opposing forces at play.

In the case of a reduction in the labor income tax rate, longer debt maturity reduces but does not eliminate their expansionary effects on output, consumption, investment, and employment. As shown in Figure 5, longer debt maturities soften but not reverse the decline in real interest rates following the labor tax cut. Thus, the positive income effect from lower labor taxes continues to support consumption and labor supply though to a lesser extent than under short-term debt. This stands in contrast to increases in government spending, which often trigger crowding-out effects under long-term debt. As aggregate demand weakens markedly with rising debt maturity, firms reduce their hiring and investment. As a result, even though employment continues to rise, the increase in labor supply outpaces labor demand, leading to a reversal in the real wage response, from positive to negative, as maturity lengthens. This decoupling between employment and wages reflects the diminished effectiveness of tax cuts in stimulating private demand when financing constraints are spread over longer horizons.

In contrast, the expansionary potential of capital income tax cuts is largely eliminated as debt maturity rises. Investment incentives induced by capital tax cuts are highly sensitive to expectations about future interest rates. As shown in Figure 6, unlike labor tax cuts, longer debt maturities lead to an immediate and sharp increase in real interest rates following a capital tax reduction. This rise in real rates undermines the intended investment stimulus, outweighing the benefits from a lower tax rate. Both consumption and investment fall significantly under long-term debt, mainly driven by tighter financing conditions. As a result, total and private employment decrease drastically from the outset. The joint decline in consumption and employment renders falling real wages. Our findings show that capital income tax cuts are particularly sensitive to the structure of debt financing: when financed through long-term debt, capital income tax cuts may become ineffective or even contractionary in the short run.

The effects of consumption tax cuts under long-maturity debt also exhibit a notable decline in effectiveness, with transmission dynamics that initially resemble those of labor tax cuts. As can be seen in Figure 7, real interest rates decline less but remain negative with longer debt maturities as with the case of labor income tax cuts. As debt maturity increases, the boost to consumption persists, but it comes increasingly at the expense of household saving, thereby crowding out private investment. This compositional shift in aggregate demand toward consumption and away from investment in turn weakens the overall impact on output and employment.

Overall, consistent with government spending shocks, the presence of long-term debt systematically weakens the effectiveness of tax cuts. Among the three tax instruments, capital income tax cuts are most sensitive to the maturity structure of government debt. This greater sensitivity arises because investment, which is the key transmission channel for capital tax cuts, is particularly responsive to expected real interest rates. As shown in Figure 6, longer debt maturities lead to a sharper increase in real interest rates following a capital tax cut, undermining the intertemporal substitution effect and significantly raising the cost of capital. This sharply reduces investment, and in turn, leads to contractionary effects on consumption, employment, and real wages. By contrast, labor and consumption tax cuts continue to generate modest expansionary effects, as their transmission channels are less reliant on forward-looking investment dynamics and more tied to household income and consumption smoothing.

Taken together, our analysis on tax cuts reinforces the broader conclusion that long-term debt systematically weakens the stimulative effects of fiscal policy. While labor and consumption tax reductions retain some expansionary effects, capital income tax cuts exhibit growing contractionary effects as debt maturity extends.

4.3. Fiscal Multipliers

To summarize the quantitative effects of our six fiscal shocks, we report in

Table 2. Fiscal multipliers.

			Avg. Maturity	1	20	40	1	20	40	1	20	40	1	20	40
		Time Horizons		Output Multiplier			Consumption Multiplier			Investment Multiplier			Employment Multiplier
Increase in	${\mathbf{c}}^{\mathbf{g}}$	Impact		2.06	1.35	1.24	0.48	0.18	0.13	0.14	−0.09	−0.12	1.02	0.71	0.66
		10 qtrs		2.03	0.83	0.65	0.47	−0.06	−0.14	0.06	−0.30	−0.35	0.91	0.44	0.37
		20 qtrs		1.91	0.69	0.51	0.41	−0.17	−0.25	0.01	−0.35	−0.41	0.83	0.37	0.30
		40 qtrs		2.02	0.67	0.47	0.43	−0.25	−0.35	0.05	−0.37	−0.43	0.84	0.37	0.30
	${\mathbf{n}}^{\mathbf{g}}$	Impact		0.39	−0.44	−0.65	0.40	0.05	−0.04	0.28	0.02	−0.05	−1.23	−1.59	−1.69
		10 qtrs		1.78	0.06	−0.34	1.09	0.30	0.12	0.59	0.10	−0.02	−0.78	−1.44	−1.60
		20 qtrs		1.95	0.31	−0.11	1.26	0.46	0.26	0.59	0.11	−0.01	−0.76	−1.36	−1.51
		40 qtrs		2.30	0.57	0.13	1.54	0.64	0.41	0.66	0.14	0.01	−0.67	−1.28	−1.43
	${\mathbf{i}}^{\mathbf{g}}$	Impact		1.73	0.94	0.76	0.24	−0.09	−0.17	0.11	−0.14	−0.20	0.84	0.50	0.42
		10 qtrs		1.78	0.48	0.21	0.36	−0.24	−0.37	−0.01	−0.37	−0.45	0.68	0.20	0.10
		20 qtrs		1.73	0.40	0.11	0.38	−0.26	−0.40	−0.08	−0.45	−0.53	0.57	0.11	0.01
		40 qtrs		1.93	0.46	0.13	0.51	−0.25	−0.42	−0.06	−0.49	−0.59	0.54	0.04	−0.06
Reduction of	${\mathbf{\tau }}^{\mathbf{n}}$	Impact		0.94	0.13	0.00	0.42	0.07	0.02	0.30	0.04	0.00	0.44	0.08	0.02
		10 qtrs		2.29	0.30	0.06	1.15	0.20	0.08	0.69	0.09	0.02	1.03	0.20	0.09
		20 qtrs		2.38	0.36	0.12	1.27	0.26	0.13	0.71	0.10	0.03	1.03	0.23	0.13
		40 qtrs		2.87	0.48	0.20	1.64	0.35	0.20	0.86	0.13	0.04	1.18	0.28	0.17
	${\mathbf{\tau }}^{\mathbf{k}}$	Impact		1.42	0.49	0.29	0.27	−0.09	−0.17	0.22	−0.05	−0.11	−0.05	−0.40	−0.48
		10 qtrs		2.17	0.38	0.06	0.56	−0.17	−0.30	0.46	−0.06	−0.16	0.16	−0.44	−0.55
		20 qtrs		2.32	0.45	0.11	0.63	−0.16	−0.30	0.53	−0.02	−0.12	0.17	−0.41	−0.52
		40 qtrs		2.77	0.62	0.23	0.87	−0.11	−0.28	0.69	0.05	−0.07	0.27	−0.36	−0.48
	${\mathbf{\tau }}^{\mathbf{c}}$	Impact		0.88	0.19	0.08	0.41	0.12	0.07	0.25	0.03	−0.01	0.40	0.09	0.05
		10 qtrs		1.63	0.38	0.19	0.90	0.33	0.24	0.34	−0.03	−0.09	0.69	0.19	0.11
		20 qtrs		1.68	0.42	0.22	1.02	0.42	0.32	0.27	−0.10	−0.16	0.69	0.21	0.14
		40 qtrs		1.87	0.49	0.27	1.20	0.49	0.37	0.26	−0.16	−0.23	0.74	0.25	0.17

Notes: Table reports present-value fiscal multipliers across the time horizons and average maturity of government debt when the monetary policy is accommodative. For instance, the output multiplier in response to an increase in $c^g$ is computed as the present value of additional output over a k-period horizon produced by an exogenous change in the present value of $c^g$, which also varies depending on the maturity structure of government debt. A value of 1 corresponds to one-period bond, 20 represents a debt maturity of 20 periods, and 40 indicates a debt maturity of 40 periods.

the fiscal multipliers on output, private consumption, private investment and private employment implied by our simulations. We focus on present-value multipliers to assess the full dynamics associated with exogenous fiscal actions. For instance, the present-value output multiplier to an exogenous change in government purchases is defined as

$$\begin{array}{c}\hfill \mathrm{Government} \mathrm{Spending} \mathrm{Multiplier}\left(k\right)={\displaystyle \frac{E_t{\mathsf{\Sigma }}_{j=0}^k\left({\mathsf{\Pi }}_{j=0}^k{\left(1+r_{t+j}\right)}^{-1}\right)\Delta y_{t+j}^m}{E_t{\mathsf{\Sigma }}_{j=0}^k\left({\mathsf{\Pi }}_{j=0}^k{\left(1+r_{t+j}\right)}^{-1}\right)\Delta {\left(c^g\right)}_{t+j}^m}}\end{array}$$

where $r_{t+j}$ is the model-implied real interest rate, which is included to properly discount future macroeconomic effects. m denotes the average maturity of government debt. At $t+k$ for any $k>0$, this measures the present value of the cumulative change in output over k periods produced by an exogenous change in government purchases, also in the present value. At $k=0$, the present-value multiplier equals the impact multiplier. Private consumption, investment, and employment multipliers to government purchases and other spending shocks are defined analogously.

The present-value tax multiplier is defined as

$$\begin{array}{c}\hfill \mathrm{Tax} \mathrm{Multiplier}\left(k\right)={\displaystyle \frac{E_t{\mathsf{\Sigma }}_{j=0}^k\left({\mathsf{\Pi }}_{j=0}^k{\left(1+r_{t+j}\right)}^{-1}\right)\Delta y_{t+j}^m}{-E_t{\mathsf{\Sigma }}_{j=0}^k\left({\mathsf{\Pi }}_{j=0}^k{\left(1+r_{t+j}\right)}^{-1}\right)\Delta {\left(T^n\right)}_{t+j}^m}}\end{array}$$

where $-\Delta T^n$ denotes decreases in real labor income tax revenue following a negative labor income tax rate shock. This corresponds to the standard tax multiplier for output (i.e., the response of output to a change in tax revenues). Private consumption, investment, and employment multipliers to other tax shocks are defined analogously.

The difference in the impulse responses translates into variations in fiscal multipliers. Extending the average maturity of government debt not only systematically reduces the magnitude of output, consumption, investment, and employment multipliers but also, in many cases, reverses their signs. Output multipliers are sizable, generally greater than one under short-term debt, but fall below one at all horizons (except for the impact multiplier after government purchases shocks) as maturity rises, consistent with findings in Ghomi et al. (2025). The deterioration in multiplier values with debt maturities is especially evident for private demand. Consumption multipliers, which are uniformly positive under one-period debt, decline sharply and often turn negative under extended maturities, particularly in response to increases in government investment or purchases and reductions in capital income tax rates. Similarly, investment multipliers shift from mild positive to negative values as maturity increases. Notably, investment multipliers remain negative following government investment shocks across all maturities and horizons, suggesting the persistent crowding-out of private capital formation. Employment multipliers become negative under longer maturities after capital tax cuts and government employment shocks. In the latter case, the multiplier exceeds unity in absolute value, confirming that each additional public-sector job displaces more than one job in the private sector.

Beyond the magnitude of individual multipliers, the maturity structure of public debt also governs the relative importance of different fiscal policy instruments. Under short-term debt, the largest fiscal multipliers across output, consumption, and employment are found for cuts in the labor income tax rate as reported in Table 2. These effects also build over time, with multipliers exceeding unity at longer horizons, consistent with the findings of Zubairy (2014). In contrast, under long-maturity debt, government purchases shocks produce the largest multipliers for output and employment at all horizons, and the highest impact multiplier for consumption. Accordingly, our model implies that fiscal environments with predominantly short-term debt are most conducive to tax-based stimulus, particularly through labor income tax cuts. By comparison, fiscal expansions via government spending become relatively more effective when debt maturities are extended. In addition, while prior research has identified labor income tax cuts as generating the strongest fiscal multipliers among revenue-side measures (see Coenen et al., 2012), our results suggest that such findings hold primarily under the standard assumption of short-term debt. Once longer maturities are introduced, the strongest fiscal multipliers are no longer exclusive to labor tax cuts. Instead, each type of tax cut may exert its largest impact on different economic variables.

4.4. Transmission Mechanism in an Alternative Policy Regime

Leeper et al. (2017) demonstrates that the effects of fiscal policy are highly conditional on the interaction between monetary and fiscal authorities. To examine this monetary–fiscal interplay more closely, Figure A1 and Figure A2 in the appendix plot the impulse responses to each of our six fiscal shocks under an alternative regime, regime H, where the monetary authority raises the interest rate aggressively in response to inflation while the fiscal authority slightly adjusts expenditures and tax rates to stabilize debt. All other parameters are held constant from regime D, ensuring that any differences across two regimes solely stem from distinct policy behavior.

In regime H, hawkish monetary policy raises real interest rates sharply in response to fiscal expansions, curbing the increase in inflation and reducing private demand substantially. This neutralizes the expansionary effects of fiscal stimulus. Moreover, the debt revaluation through bond price and inflation channels is largely muted under this policy regime. As a result, the full burden of debt stabilization is placed to future fiscal contractions as emphasized by Corsetti et al. (2012). Consequently, while the maturity structure of outstanding debt plays a central role in regime D, it has minor effects in regime H.

Our regime-based analysis offers a theoretical lens for interpreting the wide-ranging empirical estimates of fiscal multipliers. The two most widely discussed and visible shifts in US monetary policy—the hawkish stance of the Volcker era in the early 1980s when the Federal Reserve aggressively raised interest rates to combat inflation (see Bernanke & Mihov, 1998; Judd et al., 1998), and the dovish stance following the 2008 GFC when the Fed rapidly lowered policy rates to the zero lower bound (see Nikolsko-Rzhevskyy et al., 2014; Dean & Schuh, 2021)—serve as historical analogs to our model’s regimes H and D, respectively. This mapping helps rationalize the muted fiscal multipliers around 1980 and the stronger effects observed after 2008. By incorporating a government debt maturity structure, our model also contributes to bridging the theoretical and empirical gaps, especially where the effectiveness of fiscal policy appears sensitive to both the monetary stance and the composition of public debt.

5. Concluding Remarks

In this paper, we investigate how the maturity structure of government debt influences the macroeconomic effects of fiscal policy and the magnitude of fiscal multipliers under a dovish monetary policy regime. To this end, we have developed a DSGE model augmented with long-maturity debt and six exogenous fiscal shocks spanning both the expenditure and revenue sides of the government budget.

Our key finding is that longer debt maturities systematically dampen the expansionary effects of fiscal policy, particularly for increases in government purchases, investment, and capital income tax cuts. This occurs because long-term debt spreads inflation adjustment over time, leading to a more muted short-run decline in real interest rates and thus weakening the intertemporal substitution channel that drives private demand. By contrast, under a hawkish monetary regime where fiscal policy is less active, the influence of debt maturity is significantly muted.

Conventional studies based on short-term debt specifications may therefore overestimate the size of fiscal multipliers. Our findings suggest that, just like level of government debt, the maturity structure of public liabilities should be carefully considered when designing and implementing fiscal policy. For countries with high debt levels and a more heavily reliance on long-term debt, the fiscal design may need to shift toward instruments less dependent on interest rate channels, such as direct government consumption or public employment programs. One possible extension of this work would be to endogenize debt maturity decisions within a portfolio optimization framework to better capture the evolving fiscal landscape faced by modern governments.