امکان سیاسی اصلاح امنیت اجتماعی با یک ساختار دو لایه

This paper investigates the welfare implications and political feasibility of social security reforms with a two-tier structure in Japan. We evaluate social security reforms from two points of view: (i) the ex-ante expected value of future generations, and (ii) whether current generations prefer reform to the status-quo system, which we call political feasibility. To evaluate the reforms, we use a large-scale overlapping generations model with idiosyncratic income risk and a two-tier structure. The first tier guarantees a basic pension and the second tier consists of the earnings-related part. Calibrating the parameters of the model to the Japanese economy, we compute the transition path and the two welfare criteria. We find that, given the two-tier structure in Japan, an increase in the basic pension and the abolition of the earnings-related part of the social security system improve the welfare of future generations, and ensures political feasibility when a consumption tax is the source of revenue.

Many developed countries including Japan have become aging societies. Faced with aging, governments in such countries have taken social security reforms into account seriously to sustain the system. Many studies propose social security reform options for enhancing the social welfare of future generations. However, it is known that there is a status-quo bias problem in implementing reforms, given the pay-as-you-go system ( Conesa and Krueger, 1999). It may not be so difficult to find social security reform options that improve the welfare of future generations if current generations willingly pay the reform cost. However, such a plan does not usually receive the political support of current generations, and is thus politically infeasible. Therefore, one difficulty is how to implement social security reforms with the agreement of current generations, who have already made, at least in part, consumption-saving decisions over the lifecycle. In what follows, we address the following question: Are there social security reform plans that improve the welfare of future generations with the political support of current generations? Given a two-tier structure of the social security system in Japan, we show that there exists a social security reform that both satisfies political feasibility, and enhances the welfare of future generations. OECD (2007) separates the role of the social security system into two parts: (a) the insurance part and (b) the redistribution part. Concerning the insurance part, it is widely believed that the social security system should be actuarially fair. On the other hand, in the redistribution part, a minimum floor is required for the consumption or redistribution of resources through the social security system. For this reason, in many countries, the social security system has a two-tier structure. The first tier comprises three types of redistribution schemes: basic pension schemes, resource-tested plans, and the minimum pension. 1 How the first tier is constructed differs significantly across the developed countries. For example, in the US, the government imposes a resource test for the receipt of a public pension. On the other hand, in Japan all eligible retirees receive the same amount of basic pension. Some countries employ a mixture of the three roles. The second tier comprises two typical forms of social security systems, defined benefit and defined contribution. Although a limit is set on the second tier, it is basically earnings-related. Therefore, although we consider Japanese economy in this paper, considering social security reforms with a two-tier structure will be applicable to other countries. To consider social security reforms, we employ an overlapping generations (OLG) model with idiosyncratic income risk. Our model is based on Conesa and Krueger, 1999 and Nishiyama and Smetters, 2007, who extend the steady state equilibrium model constructed by Huggett (1996) to the transition dynamics. We extend the model to include a two-tier structure of Japanese economy. There are infinitely many households who face idiosyncratic income shocks, and the government manages the social security system as a pay-as-you-go system. We calibrate the parameters of the model for the Japanese economy and calculate the steady state and the transitional path. We choose the Japanese economy as a target for the following two reasons: First, the Japanese economy employs a two-tier structure: the basic pension and the earnings-related part.2 Second, Japan is one of the most rapidly aging countries in the world. A population projection indicates that the percentage of retired households will exceed 40% by 2055. Therefore, the Japanese economy provides a good example by which to consider social security reforms. To consider social security reform, we basically propose three reform options, and a mixture of these options: (1) the abolition of social security reform, (2) introducing a consumption tax as a source of revenue for the basic pension, (3) introducing a capital income tax as a source of revenue for the basic pension. Neither the abolition nor the privatization of the social security system is a new idea. For example, Storesletten et al. (1999) consider alternative social security arrangements, including abolition using an OLG model calibrated to the current US system by steady state comparison. Conesa and Krueger, 1999 and Nishiyama and Smetters, 2007 extend their research to investigate the transition path of privatization. Conesa and Krueger (1999) conclude that privatization improves the welfare of future generations, although it is not supported by hypothetical voting. Because our model includes a two-tier structure, we need to reconsider whether these results apply in our setup. In addition, we consider the choice of sources of revenue to sustain the social security system. In many countries, pensioners (retirees) do not pay social security contributions. One reason why aging reduces the sustainability of the social security system is that, in many countries, social security contributions are based on a payroll tax. Recently, many researchers focus on tax reform by using the consumption tax for social security reforms (e.g., Kotlikoff et al., 1999). A consumption tax may be a good option for improving the welfare of transition generations because it collects revenue from rich retirees. We add a third plan, capital income tax, as an option that has the potential to improve the welfare, because it too collects revenue from rich retirees. Many theoretical studies have been conducted on social security reforms using OLG models. Moreover, because research on social security reform requires numerical values such as tax rates, quantitative studies of the social security system have attracted attention since the pioneering research by Auerbach and Kotlikoff (1987). Recent research focuses on social security reform in an economy with heterogeneous agents, due to its redistribution effect. For example, İmrohoroğlu et al., 1995, Storesletten et al., 1999 and Huggett and Ventura, 1999 investigate an efficient social security system using steady state comparison. Huang et al., 1997, De Nardi et al., 1999, Conesa and Krueger, 1999 and Nishiyama and Smetters, 2007 consider the transitional dynamics of social security reforms. Our research is an extension of the result by Conesa and Krueger (1999). They show that, given pay-as-you-go systems, there arises the status-quo bias problem for social security reform. Conesa and Garriga (2003) discuss this point, and show that the elimination of compulsory retirement can offset the status-quo bias problem. Bütler (2000) also investigates the political feasibility of several reform options by calibrating the model to the Swiss economy. Using an extension of the Auerbach and Kotlikoff (1987) model, Okamoto et al. (2009) consider the effect of demographic change on welfare and the growth rate in Japan. Their research is more specific to the Japanese economy, and focuses on the actual public pension system in Japan. Compared to their research, we include idiosyncratic income risks, which imply substantial intragenerational income and wealth inequality. Okamoto et al. (2009) conclude that a progressive consumption tax improves the welfare of the economy. Our conclusion is as follows. Although abolition, or financing choice of the basic pension alone, improves the welfare of future generations, it is not supported by current generations. We find that (i) abolition of the earnings-related part (second tier) and (ii) introducing a consumption tax to finance the basic pension improves the welfare of future generations. The abolition of the earnings-related part improves welfare because the real return from second tier is dominated by the equilibrium interest rate. Moreover, financing the basic pension by consumption tax improves welfare measured in ex-ante value, due to the insurance effect. In addition to these, (iii) expanding the basic pension to guarantee the current middle and old poor households receives political support. Therefore, a combination of these options provides a politically feasible social security reform that improves future generations’ welfare. This paper is structured as follows. Section 2 provides the details of our model. Section 3 calibrates the parameters of the model for the Japanese economy. Section 4 discusses policy experiment plans, and compares the steady states of the model. Section 5 considers the transitional dynamics, and the political feasibility of several reforms. Finally, Section 6 concludes the paper.

In this paper, we investigate the welfare implications and political feasibility of social security reforms, using an overlapping generations model with idiosyncratic income risks. We examined such alternative reform options as (1) the abolition of the social security system, (2) the introduction of a consumption tax to finance the basic pension, and (3) the introduction of a capital income tax, and a mixture of these options based on a two-tier structure. We show that introducing a consumption tax and a capital income tax as the tax base of the basic pension improves welfare from a steady state comparison. Moreover, the abolition of the social security system creates large welfare gains for future generations. To establish the political feasibility of implementing these plans, we evaluate alternative reform options based on two criteria: (i) ex-ante welfare and (ii) the political support of current generations. We find that, given a two-tier structure, abolishing the earnings-related part and expanding the basic pension improve the welfare of future generations, and is also politically feasible when the basic pension is financed by a consumption tax. Therefore, this plan offers a solution to the “status-quo bias” problem. We believe that this result actually offers a feasible plan for aging countries that have a social security system with a two-tier structure. However, the social security reform we proposed does not imply an efficient reallocation. Note that in our model the optimal basic pension is zero. Thus the politically feasible plan tends to move away from the optimum. Recently, Conesa and Garriga (2008) consider the optimal fiscal policy design for social security reforms. Conesa et al. (2009) find that the optimal capital income tax is positive, and that the progressivity of the labor income tax should be low. Although we do not consider the fiscal policy mix, the progressivity of alternative tax options may enhance our results. In addition, we omit a government debt for computational reasons. It is not clear whether inclusion of the government debt into our model changes our results. We try to consider this point in the next step. In this paper, the political feasibility mechanism is very simple. Thus, more sophisticated modeling of the politico-economic decisions regarding alternative reforms, such as Corbae et al. (2009), may prove to be a promising approach. In addition, considerations of a general altruistic linkage in political mechanisms are necessary because political decisions regarding social security reforms affect future generations welfare significantly. These points will also be considered in our future research.