The stabilizing effect of fiscal policies on the dynamics of effective demand and income distribution in Japan

Abstract

In this study, we build a Kaleckian model incorporating fiscal policies and investigate the possibility that fiscal policies have had a stabilizing influence on the dynamics of effective demand and income distribution in Japan. In a Kaleckian model with a fluctuating income distribution rate, the combination of the demand and distributive regimes may destabilize the system, and it is important to clarify the policy approach to stabilize it. For this purpose, we specify dynamic system consisting of three variables, the rate of capacity utilization, wage share, and the level of government expenditure. Then we estimate the parameters in this model using data relating to the Japanese economy from 1977 to 2007. As a result, we obtained the following findings: first, even if the combination of demand and distributive regimes has the potential to destabilize the system, the system may stabilize if counter-cyclical fiscal policy has a sufficient effect on the effective demand and if the government does not take an excessive budgetary austerity stance. Second, in Japan, the dynamics of effective demand and income distribution may have been unstable, and fiscal policy has not contributed to systemic stability. Third, the main reason Japan's fiscal policy did not have a stabilizing effect was not the stance of the government's fiscal policy, but rather the insufficient effect of the policy on effective demand. These results suggest the existence of different factors that stabilized the Japanese economy.

Appendices

Appendix 1. Proof for the equation of demand regime

We can express the rate of capacity utilization as follows:

$$ u_{t} = \frac{{Y_{t} }}{{Q_{t} }}, $$

(A1)

where ${Y}_{t}$ denotes the real output at time $t$, and ${Q}_{t}$ the capacity. Therefore, we obtain the following equation:

$$ \widehat{{u_{t} }} = \widehat{{Y_{t} }} - \widehat{{Q_{t} }}. $$

(A2)

Next, we specify the equations for $\widehat{{Y}_{t}}$ and $\widehat{{Q}_{t}}$ as follows¹⁹:

$$ \widehat{{Y_{t} }} = \varepsilon_{1} \left( {u_{t} - \overline{u}} \right) + \varepsilon_{2} \psi_{t} + \varepsilon_{3} G + \varepsilon_{C} , $$

(A3)

$$ \widehat{{Q_{t} }} = \zeta_{1} \left( {u_{t} - \overline{u}} \right) + \zeta_{2} \psi_{t} + \zeta_{3} G + \zeta_{C} . $$

(A4)

In Eq. (A3), if the Keynesian stability condition is satisfied, ${\varepsilon }_{1}<0$ holds. The sign of ${\varepsilon }_{2}$ is ambiguous because the effect of an increase in wage share on the effective demand can be positive or negative depending on the demand structure. We can also assume that ${\varepsilon }_{3}$>0 because government expenditure is one part of effective demand. ${\varepsilon }_{C}$ is the constant term and represents the sum of investment demand and net export when the rate of capacity utilization is standard level and both wage share and government expenditure are zero.

In Eq. (A4), we assume that ${\zeta }_{1}$>0 because an increase in output raises capacity through the acceleration of capital accumulation. In general, a decrease in profit share slows down capital accumulation, and we assume that ${\zeta }_{2}<0$. We can also assume that ${\zeta }_{3}$>0 because an increase in government expenditure raises capacity through the acceleration of public capital formation. However, as this positive effect is thought to be smaller than the direct effect of an increase in government expenditure on the real output, we assume that $0<{\zeta }_{3}$<${\varepsilon }_{3}$. ${\zeta }_{C}$ is the constant term and represents the effect of capitalists’ investment on capacity when the rate of capacity utilization is standard level and both wage share and government expenditure are zero.

By substituting Eqs. (A3) and (A4) into (A2), we obtain the following equation:

$$ \widehat{{u_{t} }} = \left( {\varepsilon_{1} - \zeta_{1} } \right)\left( {u_{t} - \overline{u}} \right) + \left( {\varepsilon_{2} - \zeta_{2} } \right)\psi_{t} + \left( {\varepsilon_{3} - \zeta_{3} } \right)G + \left( {\varepsilon_{C} - \zeta_{C} } \right). $$

(A5)

From Eqs. (1) and (A5), the following equations hold:

$$ \phi_{1} = \varepsilon_{1} - \zeta_{1} , $$

(A6)

$$ \phi_{2} = \varepsilon_{2} - \zeta_{2} , $$

(A7)

$$ \phi_{3} = \varepsilon_{3} - \zeta_{3} , $$

(A8)

$$ C_{u} = \varepsilon_{C} - \zeta_{C} . $$

(A9)

From Eq. (A6), we can confirm that ${\phi }_{1}$<0 holds because ${\varepsilon }_{1}<0$ and ${\zeta }_{1}$>0. In Eq. (A7), ${\phi }_{2}>0$ holds when ${\varepsilon }_{2}$>0 or ${\zeta }_{2}<{\varepsilon }_{2}<0$. In contrast, ${\phi }_{2}<0$ holds when ${\varepsilon }_{2}<{\zeta }_{2}<0$. Therefore, we can confirm that the sign of ${\phi }_{2}$ is ambiguous depending on the demand regime. From Eq. (A8), we obtain ${\phi }_{3}$>0, because we assume that $0<{\zeta }_{3}$<${\varepsilon }_{3}$. From Eq. (A9), we can confirm that ${C}_{u}$ is a constant term because ${\varepsilon }_{C}$ and ${\zeta }_{C}$ are constant.

Appendix 2. Proof for the fiscal policy regime

Government expenditure can be divided into two parts as follows:

$$ G_{t} = G_{1t} + G_{2t} , $$

(A10)

where ${G}_{1t}$ denotes spending for discretionary fiscal policy and ${G}_{2t}$ constant spending (ex. social security expenditure). From Eq. (A10), we obtain the following equation:

$$ \dot{G}_{t} = \dot{G}_{1t} + \dot{G}_{2t} , $$

(A11)

where the superscript dots represent time derivatives. By dividing both sides of Eq. (A11) by ${G}_{t}$, we obtain the following equation:

$$ \frac{{\dot{G}_{t} }}{{G_{t} }} = \frac{{\dot{G}_{1t} }}{{G_{t} }} + \frac{{\dot{G}_{2t} }}{{G_{t} }}. $$

(A12)

Now, we assume that the government determines the level of ${G}_{1t}$ by considering the economic condition and the voters’ response to the budget deficit. Therefore, we can formulate the first term on the right-hand side of equation (A12) as follows:

$$ \frac{{\dot{G}_{1t} }}{{G_{t} }} = \lambda_{1} \left( {u_{t} - \overline{u}} \right) + \lambda_{2} \cdot \left( {G_{t} - \overline{G}} \right). $$

(A13)

The first term on the right-hand side of Eq. (A12) indicates that the government increases expenditure to achieve economic stabilization when the rate of capacity utilization falls below the standard level; therefore, ${\lambda }_{1}<0$. The second term on the right-hand side of Eq. (A12) indicates that the government controls the level of expenditure according to voters’ response to changes in deficit-financed bonds. If the tax revenue is constant, an increase in ${G}_{t}$ leads to an increase in deficit-financed bonds. If voters tend to call for the government to fix finances, ${\lambda }_{2}<0$ holds. In contrast, when voters do not pay much attention to deficit-financed bonds and tend to support more expansion of government expenditure, ${\lambda }_{2}>0$ holds. Therefore, the sign of ${\lambda }_{2}$ is ambiguous depending on the fiscal policy regime.

We also assume that the level of ${G}_{2t}$ growth at a constant rate over time because the scale of constant spending by the government will increase due to the expansion of social security systems and other factors. Therefore, we can formulate the second term on the right-hand side of Eq. (A12) as follows:

$$ \frac{{\dot{G}_{2t} }}{{G_{t} }} = C_{G} , $$

(A14)

where ${C}_{G}$ is a positive constant term.

By substituting (A13) and (A14) into (A12), we can obtain Eq. (10) and confirm the signs of ${\lambda }_{1}$ and ${\lambda }_{2}$.