A dynamic new Keynesian life-cycle model: Societal aging, demographics, and monetary policy

https://doi.org/10.1016/j.jedc.2007.09.002Get rights and content

Abstract

In this paper, we first construct a dynamic new Keynesian model that incorporates life-cycle behavior a la Gertler [1999. Government debt and social security in a life-cycle economy. Carnegie–Rochester Conference Series on Public Policy 50, 61–110], in order to study whether structural shocks to the economy have asymmetric effects on heterogeneous agents, namely workers and retirees. We also examine whether considerations of life-cycle and demographic structure alter the dynamic properties of the monetary business cycle model, specifically the degree of amplification in impulse responses. According to our simulation results, shocks indeed have asymmetric impacts on different households and the demographic structure does alter the size of responses against shocks by changing the trade-off between substitution and income effects.

Introduction

Societal aging is one of the biggest economic issues facing many industrial countries. In Japan, in particular, society is aging so rapidly that not only is the working population (those older than 15 but younger than 65) already shrinking, but the total population is also expected to start decreasing by 2007. This movement suggests that the central bank should have an even greater interest in how monetary policy affects heterogenous agents, namely workers and retirees, differently and how the consideration of this demographic structure may alter the reaction of variables to structural shocks.1 Seminal research by Woodford (2003) depicts the various forms of the dynamic New Keynesian model corresponding to different economic conditions and has had a significant influence on central banks’ views of monetary business cycle. However, to date very little research has paid attention to monetary business cycle model with heterogenous agents, particularly within a life-cycle setting.2

In this paper, we first set up a dynamic stochastic general equilibrium model with nominal rigidities and investment adjustment costs that incorporates life-cycle behavior a la Gertler (1999). Then, we show whether the structural shocks to the economy have asymmetric effects on heterogeneous agents and whether the considerations of the life-cycle and demographic structure alter the dynamic properties of the solution, under different settings of life-cycle behavior.3 Of course, as mentioned in Bean (2004), it is true that ‘the glacial nature of demographic change appears to suggest that the implications for monetary policy should be modest’. We, however, believe that it becomes more important for central banks to acknowledge the asymmetric effects on heterogeneous agents within a life-cycle economy with a stationary population, since societal aging in many industrial countries necessitates the consideration of the distributional consequences of monetary policy.4 Furthermore, central banks must always understand the monetary transmission mechanism as well as macroeconomic responses to structural shocks in detail. Therefore, we focus on the impulse responses of this life-cycle economy. We know that the impact of societal aging on general equilibrium has two separate aspects, and it is important to distinguish between the two. First, the ‘transition’ toward the aging society can be most naturally considered in terms of a macro shock, which will affect monetary policy decisions. Second, the impulse responses derived from a dynamic new Keynesian model in a ‘stationary population’ may be quite different for an elderly society than for a young society. For the purposes of the current paper, we focus on the second of these two aspects.

Bean (2004) summarizes the previous findings in this field, pointing out their implications for a central bank: (1) demographic developments represent a macroeconomic shock, which may lead to abrupt movements in asset prices and sharp movements in saving behavior; (2)the natural rate of interest falls both along the transition path and in the steady state; (3) the natural rate of unemployment may also be affected through the matching mechanism5; (4) the wealth channel is likely to become a more important transmission channel of monetary policy than intertemporal substitution; (5) the Phillips curve is flatter due to immigration and the increased participation of retired workers whose supply of labor is considered to be relatively elastic; (6) the constituency for keeping inflation low will be larger thanks to higher average wealth accumulation; and (7) societal aging may induce diversification and risk-shifting with a securitized market rather than bank-intermediated finance, which has implications for financial stability. Although not all the topics raised by Bean (2004) can be covered in this paper, we formally verify (4) using the dynamic general equilibrium model with sticky price and life-cycle behavior. Yet, as societal aging deepens, although income effect becomes stronger for retirees, that of workers becomes weaker. Hence, in aggregate, a tightening monetary policy shock still has negative impacts on the aggregate demand. Even though retiree's consumption increases, worker's consumption or the investment needs to be lowered since there is no expansion in the production frontier. At the same time, we will show that (2) is not a general result due to the endogenized labor participation by retirees. Furthermore, we find that since retirees do not work as much as workers, they benefit less from improved technology, a point not by raised by Bean (2004). Therefore, the monetary response to a positive technology shock can be smaller in a greyer society. Anecdotal evidence on these points abounds, but there have been very few studies that have tackled this problem in a theoretically consistent dynamic general equilibrium framework, 6 which is the workhorse model for modern monetary policy analysis. Our main conclusions are as follows. Since retired people rely more on interest income than on wages from their labor supply, shocks indeed have different impacts on different households. Furthermore, the demographic structure does alter the size of the response to shocks by changing the degree of the trade-off between substitution and income effects. Therefore, societal aging and life-cycle considerations, have important implications for monetary policy.

Another interesting point is to compare optimal monetary policy in this life-cycle economy to the one obtained in a standard model with homogeneous agents. Stochastic welfare analysis, however, raises difficult issues such as what discount rate should be used by firms and which welfare measures are appropriate with the setting of the model in this paper. As for the latter, it is not trivial to compute aggregate welfare with heterogenous agents. In other words, it is difficult to weigh the welfare of each heterogenous agent in order to obtain aggregate welfare measure. The former is an especially large concern with incomplete markets and heterogenous agents, because marginal utility growth rates are no longer equalized across agents and therefore the choice of the appropriate discount factor is no longer obvious. 7 Thus, we restrict our analysis to the case of perfect foresight with one-time unanticipated shocks that may be auto-correlated as examined in Ghironi (2006), Kilponen and Ripatti (2006), and Kilponen et al. (2006). In a perfect-foresight setting, all assets including the one-period bond must yield the same return. As a result some very difficult (and as yet unsolved) problems regarding asset valuation can be side stepped.

This paper is put together as follows. In section two, we describe the model employed in this analysis. Then, section three discusses the nature of our stationary population under different demographic structures. We show that the natural rate of interest is quite different among the different demographic setups. In section four, we show the impulse responses and show that shocks indeed have different impacts on different households and the demographic structure does alter the size of responses to shocks by changing the trade-off between substitution and income effects. Finally, section five concludes.

Section snippets

Model

The model examined here is based on Gertler (1999). We add a sticky price mechanism with endogenous capital whose accumulation is subject to an investment adjustment cost. This type of the model has been referred to as the ‘canonical model’ by Edge (2003). We have six agents in this model, namely firms, capital producers, financial intermediaries, households, a government, and the central bank. As explained above, we limit our analysis to the cases with perfect foresight to avoid aggregation

Properties under stationary population

Parameters in the baseline life-cycle economy are depicted in Table 1. Since the model is solved with quarterly frequency, parameters are on quarterly terms.

In our baseline life-cycle economy, we assume that people work from age 21 to 65, which is defined by ω, while people remain in retirement from age 66 to 75, defined by γ, based on Auerbach and Kotlikoff (1987). Other parameters which define life-cycle behavior are exactly the same as those in Gertler (1999) except for newly added

Dynamic properties

In this section, we answer the two major questions raised in this paper, namely whether the structural shocks to the economy have asymmetric effects on workers and retirees and whether the considerations of life-cycle and demographic structure alter the dynamic properties of the solution. For this aim, we first analyze the impulse responses to a level technology shock. Then, we investigate the transmission mechanism of a monetary policy shock. The reason why we choose these two shocks is that a

Conclusion

In this paper, we construct a dynamic new Keynesian life-cycle model based on Gertler (1999). We show that it is of great importance for a central bank to consider the demographics for sound monetary policy. First, the natural rate of interest differs as demographics change. Second, the structural shocks to the economy have asymmetric effects on heterogeneous agents, namely workers and retirees. Differences in the labor productivity between workers and retirees make a positive technology shock

Acknowledgments

Previously circulated under the title: ‘Monetary Policy in a Life-Cycle Economy: Distributional Consequences of Monetary Policy Rule.’ We thank Martin Bodenstein, Dale Henderson, Keisuke Otsu, Laura Piscitelli, Antti Ripatti, Masashi Saito and three anonymous referees for invaluable inputs and have benefited from discussions with Stefania Albanesi, Kosuke Aoki, David Aikman, David Backus, Tamin Bayoumi, Toni Braun, Larry Christiano, Shigeru Fujita, Mark Gertler, Marc Giannoni, Fumio Hayashi,

References (35)

  • D. Andolfatto

    Business cycles and labor-market search

    American Economic Review

    (1996)
  • A. Auerbach et al.

    Dynamic Fiscal Policy

    (1987)
  • Bean, C., 2004. Overview. In: Global Demographic Change: Economic Impacts and Policy Challenges. A Symposium Sponsored...
  • O. Blanchard

    Debt deficits and finite lives

    Journal of Political Economy

    (1985)
  • Carceles-Poveda, E., Coen-Pirani, D., 2005. Shareholders unanimity with incomplete markets. GSIA Working Papers...
  • L. Christiano et al.

    Nominal rigidities and the dynamic effects of a shock to monetary policy

    Journal of Political Economy

    (2005)
  • Doepke, M., Schneider, M., 2005. Redistribution of nominal wealth and the welfare cost of inflation. Mimeo. UCLAand...
  • Cited by (45)

    • Aging and the real interest rate in Japan: A labor market channel

      2023, Journal of the Japanese and International Economies
    • Population aging, government policy and the postwar Japanese economy

      2022, Journal of the Japanese and International Economies
      Citation Excerpt :

      Kitao (2015) and Kitao (2018) quantify the fiscal cost of population aging in Japan within an overlapping generations model with a social security system. Fujiwara and Teranishi (2008) incorporates the life-cycle framework of Gertler (1999) into a New Keynesian model to analyze the effect of population aging on the amplification of monetary policy shocks. There are also several related studies that analyze the impact of productivity growth on economic growth in a representative agent neoclassical growth model.

    View all citing articles on Scopus
    View full text