Abstract
This Selected Issues paper discusses the impact of global financial turmoil on Japan. It describes how close integration in global financial markets has deepened and expanded the channels for spillovers. The paper discusses options for increasing the consumption tax to finance the public pension system. It explains that while parametric reforms have reduced fiscal pressures, the basic pension is becoming a less adequate safety net for the retired. The paper also attempts to explain the declining share of labor income.
II. Issues and Options for Pension Reform1
A. Introduction
1. Japan has a universal pension system consisting of two basic tiers. As in other G7 countries, the Japanese pension system has redistributive and insurance aspects. Thus, the first tier provides a flat-rate redistributive pension (National Pension, NP) and the public second-tier provides a defined-benefit layer that increases as past earnings increase (Employees’ Pension Insurance, EPI). The full NP benefit for a person who has contributed for 40 years is now about ¥66,000 per month. NP premiums are compulsory for every resident of Japan aged between 20–60 and benefits are payable from age 65.2 To maintain a moderate replacement ratio while keeping compulsory contributions at a reasonable level, the government has contributed about ⅓ of pension payouts by the NP program (about ¥7 trillion or about 1½ percent of GDP as of FY2005). By OECD standards, Japan’s pension system has a relatively low replacement ratio (OECD 2007a, Whitehouse 2007). However, due to the wide coverage of the NP, the pension system is more redistributive, and provides relatively more security to low-incomer earners than the OECD average.
2. Recent reforms have improved the equity and sustainability of the pension system. The combination of a pay-as-you-go system and an aging society in Japan has already required pension contributions to be raised above expected benefits for younger generations (Hatta, Horioka, and Suzuki 2007). At the same time, aging has occurred more rapidly than expected and even the higher contributions have been inadequate to protect pension reserves. Against this background, the reform of the public pension in 2004 limited scheduled increases in contribution rates and reduced replacement rates through a system of macroeconomic indexation that linked the growth of benefits to variables such as longevity. An increase in the government’s contribution to the NP from ⅓ to ½ of payouts was also planned from FY2009. With these reforms, the reserves of the pension system will be drawn down only gradually such that the ratio of reserves to annual benefits expenditure still exceeds unity in 100 years.
Projected Pension Benefits and Reserve 1/
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Ministry of Home, Labor, and Welfare (MHLW).1/ Total of the National Pension and Employee Pension Insurance.Projected Pension Benefits and Reserve 1/
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Ministry of Home, Labor, and Welfare (MHLW).1/ Total of the National Pension and Employee Pension Insurance.Projected Pension Benefits and Reserve 1/
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Ministry of Home, Labor, and Welfare (MHLW).1/ Total of the National Pension and Employee Pension Insurance.3. As a part of the 2004 reform, the government committed to the difficult task of raising taxes—implicitly the consumption tax.3 Compared to other G-7 countries, Japan’s tax burden, particularly for goods and services is lower. However, in 2006, proposals were floated to raise the consumption tax rate as part of a general fiscal consolidation strategy that also involved sizable medium-term expenditure reductions, but did not gain traction. And the financing of the increased government contribution to the pension system planned for FY2009, as well as for other expected increases in social security expenditure, has yet to be secured.
Revenue Structures in the G7, 2005
(In percent of GDP)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: OECD (2007 b), Revenue Statistics, 1965-2006.Revenue Structures in the G7, 2005
(In percent of GDP)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: OECD (2007 b), Revenue Statistics, 1965-2006.Revenue Structures in the G7, 2005
(In percent of GDP)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: OECD (2007 b), Revenue Statistics, 1965-2006.4. A number of factors have led to uncertainties about whether the pension system provides a sufficient safety net. In 2005, about 5 percent of the population over 20 were effectively outside the NP.4 While most employed people are paying premiums for the NP, a sizable number of self-employed participants, in particular from younger generations, are in arrears.5 An increasing number of non-regular workers are also not covered by the EPI. In addition, recent well-publicized lapses in pension administration have created anxiety even for those who participate in the pension system (Box II.1) As pension benefits finance a significant portion of living expenses after retirement in Japan, the risk of an increasing number of pension-less people has led to significant public concern.
Pension Record Problem1/
In 2007, the government acknowledged losing track of more than one sixth of pension payment records (50 million). Most of the unidentified records were related to the transition to a new unified pension ID system from the old system under which transactions in different pension systems were recorded separately and individuals had multiple identification numbers. Although the government was hopeful that the missing records could be found and included in the unified system before pension age so that benefits would not affected, current pensioners became concerned that they might not receive pension benefits reflecting past payments correctly.
To address the problem, Diet members proposed a bill in 2007 to suspend the statute of limitation of pension rights retroactively. In addition, while continuing its effort to identify records, the government started validating past transactions with all participants and pensioners to confirm information recorded in the system. Despite these efforts, about 20 million records were still to be identified as of January 2008.
1 Source: Social Insurance Agency, and the Nikkei Shimbun (daily newspaper)5. At the same time, other social-security expenditures will rise and need substantial financing. Japan’s social security related expenditures have been relatively contained so far, but are expected to rise rapidly going forward. Among them, medical and welfare expenditures, which do not benefit from the macroeconomic indexation that applies to pensions, are expected to grow rapidly. As the recent report from the Tax Commission (2007) stressed, the tax system faces challenges to support the safety and security of people, through ensuring the sustainability of the social security regime. In this regard, a consumption tax increase is the principal candidate as a possible revenue source.
Elderly Ratio and Social Security Expenditure
(OECD countries) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Source OECD.1/ Elderly ratio is the ratio of people over 65 to the total population. 2004 data for elderly ratio and 2005 data for social security expenditure.Elderly Ratio and Social Security Expenditure
(OECD countries) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Source OECD.1/ Elderly ratio is the ratio of people over 65 to the total population. 2004 data for elderly ratio and 2005 data for social security expenditure.Elderly Ratio and Social Security Expenditure
(OECD countries) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: Source OECD.1/ Elderly ratio is the ratio of people over 65 to the total population. 2004 data for elderly ratio and 2005 data for social security expenditure.
Projected Social Security Expenditures
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW.Projected Social Security Expenditures
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW.Projected Social Security Expenditures
(In trillions of yen)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW.B. Priority Issues for Further Pension Reform
6. The 2004 reform reduced pressures on the pension scheme, but issues beyond pension sustainability have recently drawn attention.
High inequality between generations: With aging advancing quickly, intergenerational inequality has become intense. The net benefit rate, defined as the discounted present value of net benefits from the pension system in relation to discounted present value of wages, has declined from 10 percent for those born in 1940 to about negative 5 percent for those born in 1980 (Hatta, Horioka, and Suzuki, 2007). With regard to the wider range of net services provided by the public sector, Auerbach and others (1998) conducted a generational accounting exercise for a range of OECD countries. While most countries have common patterns in intergenerational benefit distribution, with the elderly beyond 65 gaining a net benefit from the government, Japan was found to have the largest generational imbalance between current and future generations.
Hollowing out of the pension system: The number of people outside the NP has been declining, but still stands at 3.4 million. The ratio of people with arrears has been rising since FY2002, and by FY2005 was 40 percent for cohorts of Category-1 insured in the 25–35 age bracket. Various factors have been cited to explain the increasing number of people opting out of the pension system, such as heavy contribution rates, myopic behavior, and administrative inefficiencies. The evidence on which factors dominate is not clear-cut. Suzuki and Zhou (2005) do not find a cohort effect where later generations tend to evade more compared to earlier generations. Rather, the authors found a strong tendency for people to lose the incentive to participate at around age 35 when there is no hope of fulfilling the 25 years of premium payments that is required under the NP system. A survey by the MHLW (2008) found that most people cited high contributions and low credibility in the pension system as the reasons for non-compliance, with both factors becoming more important compared to a survey conducted three years earlier. Nevertheless, these results need to be interpreted with caution, as even workers with ¥10 million in annual income cited the high NP premium (fixed at about ¥170 thousand per year) as a reason for non-compliance.
Impact on the labor market: To raise economic growth potential amid a declining population, renewed attention has been paid on the impact of the pension system on the labor market. While some progress has been made in the past (Box II.2), the current beneficial treatment of spouses of EPI participants hinders female participation as spouses have to pay insurance only if they earn above a ¥1.3 million of annual income threshold. There is also a jump in costs faced by employers who try to hire regular-workers instead of non-regular workers without coverage, while the system of reducing benefits for older high-income earners creates an incentive to exit from the labor market.
Arrears of pension insurance
(Category-1 insured, ratio of arrears/participants, in percent)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW (2008).Arrears of pension insurance
(Category-1 insured, ratio of arrears/participants, in percent)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW (2008).Arrears of pension insurance
(Category-1 insured, ratio of arrears/participants, in percent)
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The MHLW (2008).Past Pension Reforms Aimed at Improving Labor Market Participation
The government has taken the following steps to improve economic efficiency through the pension system.
To facilitate labor participation of older workers, people aged 65 and above will be given an option to postpone receiving pension with increased accrual rates applied as compensation in 2007. The system to automatically cut 20 percent of benefits for active workers in their early 60s was abolished in 2005.
To prevent citizens from falling short of achieving the minimum 25 years participation requirement, a moratorium was introduced for the youth in 2005.
A defined contribution pension plan was established in 2002. Rules for early withdrawal have been relaxed since then.
In 2007, the government submitted a bill to expand EPI coverage to non-regular workers who work for more than 20 hours per week in companies with over 300 of employees and who earn more than ¥ 98,000 monthly wage. The bill is now under deliberation.
Administrative efficiency: While the number of countries where the tax collection agencies collect pension premiums is growing (Canada, the United Kingdom., the United States, Sweden and many Eastern and Central European countries), the National Tax Agency and the Social Insurance Agency collect tax and pension premiums separately in Japan (as in France and Germany). With the collection function involving similar tasks in both agencies,6 some argue that a unified collection system would be inherently more efficient (Ross, 1997), but this debate has yet to be settled as the track record of countries with a unified agency system is still quite short.
C. Policy Options
7. The government has begun a discussion on revamping the pension system. The Council on Economic and Fiscal Policy (CEFP) initiated discussions on comprehensive revenue reform in 2007 and set out five key pillars for a broader revenue reform of the tax and social security systems. These are to correct inequalities among generations; maximize incentives to work at various stages of life cycle; improve credibility and transparency of the social security system; achieve both medium-term fiscal consolidation of the general government and the pension system; and enhance economic vitality (CEFP 2007). Following these discussions, the Cabinet established the Committee on Social Security Review (CSSR) in 2008 to discuss broader social security issues including the pension system. The discussions are still at an early stage, but one focus has been on proposals to link an increase in the consumption tax to funding of the pension system.
Pros and cons of a tax-financing option
8. Financing basic pensions entirely through taxes addresses certain key objectives. In terms of which taxes should finance pensions, the consumption tax would likely be the natural choice as it provides a steady revenue flow for pension payments, distorts economic activity less, and would reduce intergenerational inequality through its wider tax base. A tax-financed basic pension could eventually cover all people over pension age. Possible merits include:
Effect on Workers’ Household with Different Age Groups
(Thousand yen per month) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The Committee on Social Security Review (2008).1/ Tax will replace pension premium prospectively.2/ A reduction in corporate matching payment is added to household by the same amount.Effect on Workers’ Household with Different Age Groups
(Thousand yen per month) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The Committee on Social Security Review (2008).1/ Tax will replace pension premium prospectively.2/ A reduction in corporate matching payment is added to household by the same amount.Effect on Workers’ Household with Different Age Groups
(Thousand yen per month) 1/
Citation: IMF Staff Country Reports 2008, 254; 10.5089/9781451820713.002.A002
Source: The Committee on Social Security Review (2008).1/ Tax will replace pension premium prospectively.2/ A reduction in corporate matching payment is added to household by the same amount.Reducing intergenerational inequality: Simulations presented to the CSSR included four different transition schemes regarding the effective date of change in financing method and the extent of conditioning pension rights on past pension premiums. These simulations envisage an increase in the rate of the consumption tax (currently 5 percent) by 6-12 percentage points to finance broader coverage of the NP through FY2050 (while abolishing pension premium and corporate matching payments).7 For example, one of the cases (Case II) assumes that the new scheme starts in FY2009 with consumption tax replacing NP contributions, while benefit payouts would be reduced in proportion to past unpaid premiums before FY2009.8 Financing needs under Case II would be ¥9 trillion in FY2009, but would grow to ¥32 trillion in FY2050 (equivalent, respectively, to a 3½ and 6 percentage point hike in the consumption tax rate). These plans would help improve current intergenerational inequality as they require contributions from the elderly through the consumption tax, while younger generations would benefit from lower contributions (see chart)
Broader coverage of pension system: The reform would alleviate concerns over life after retirement by providing a basic pension to the elderly on more generous terms. Up to 3.4 million people who are currently outside of the pension system would be covered by the NP. This also addresses the concern that a large number of the current younger generation with arrears will become pension-less. Labor participation by homemakers could also be facilitated by eliminating any distinction between the income of spouses of EPI participants and others.
Improving efficiency: There could be some efficiency gains, but many steps are required before yielding tangible results. For example, Sweden consolidated collections to the tax authority in mid-1980, which involved coordinating the tax base, simplifying calculations for contributions, and using a common personal identification number.9 However, in Japan’s case, the room for efficiency gain might not be significant, as the EPI would still need management, and substantial resources would still need to be directed to record keeping and interfacing with participants in the wake of the pension records problem.
9. The proposals may serve as a vehicle to achieve political support for a consumption tax increase, but have several drawbacks that would need to be addressed through careful design. Taxpayers may be more likely to accept a tax hike if framed in terms of the need to finance future social security expenses. The main challenges include:
Transition process: Managing a smooth transition to the new system will be challenging, as Japan’s pension system has been premium-based for decades. Considering the amount of time that is likely to be needed to nurture consensus for the reform, disincentives to make pension contribution in the meantime should be minimized. Options that do not take account of past premiums have clear disadvantages in this regard.
Redistribution among households with different income: The simulations indicated that the tax-financing option in general would put somewhat more burden on low-income households as they are now exempted from premiums but would pay higher consumption taxes.
Other reforms
10. Further parametric reform would reduce fiscal pressure and promote labor participation. OECD (2007) noted that raising the pension age has been the most common feature of recent reforms in OECD countries. Pension eligibility ages are set to rise or have risen in the United States and Germany (both to 67) and the United Kingdom (to 68). If Japan’s pension age was raised to 67 (from 65), this would improve the balance in the pension system by about 0.4 percent of GDP and create an incentive for older workers to keep working. However, lifting the pension age should be pre-announced well in advance and coordinated with raising the retirement age so as not to distort retirement planning.
11. Measures to protect low-income earners would also be desirable for public consensus making. Expanding the coverage of the EPI to certain nonregular workers is now under legislative review. Beyond this, the minimum requirement of 25 years contribution could be shortened if the reform package were not to include a tax-financing option. Coordination with assistance to the poor is required, first to maintain a minimum standard of living, and second to reduce redundant social security provisions. From this viewpoint, means testing of the NP could be introduced in Japan as in other G7 countries. There is a stronger case for means testing for the tax-financing option, as universal coverage calls for more public spending and income redistribution, while a premium-based system is run based on the “self-help” principle. However, monitoring the income of the self-employed, which is essential for means testing, has been very challenging in Japan.
12. At the same time, efforts to improve labor participation should be continued. Wedges creating disincentives to work for women, the youth and the elderly people should be reduced. Benefits given to spouses of EPI participants need reviewing (if the reform package were not to include a tax-financing option). Current efforts to widen the coverage of the pension system are in the right direction and welcomed while defined contribution (DC) pension schemes could be promoted further through user-friendly tax reform such as allowing individual contributions beyond corporate contribution and allowing entry to employees who are working in companies without a DC system. Although Japan’s elderly labor already has a high participation, this could be raised further by reducing the pace of reduction in pension benefits when they have income from work.
D. Conclusion
13. Challenges to the pension system have been mounting, but have prompted a national discussion that potentially provides an opportunity for ambitious reform. The 2004 reform reduced pressures on the pension system significantly. However, given concerns on rapid aging and economic disparities, a fundamental question has arisen as to whether the pension system provides a sufficient safety net. With repeated delays amid political opposition to consumption tax hikes, the government will need to convince the public that the reform will address concerns about sustainability and inequality. The government is also appropriately paying attention to the pension system’s impact on labor market participation.
14. Tax-financing options have some merits, but require careful design. Tax-financing would help improve current inter-generational inequality, and provided more security after retirement through wider coverage. On the downside, a smooth transition to the new system will be challenging, as Japan’s pension system has traditionally been premium-based. The slight increase in the burden to low-income household also needs to be addressed (e.g., through coordination with social assistance programs.) Moreover, a tax-financed system would likely require means testing, which has long been an issue in Japan due to the opaqueness of self-employed businesses.
References
Auerbach, Alan J., Laurence J. Kotlikoff, and Willi Leibfritz, 1998, “Generational Accounting Around the World,” IMES Discussion Paper Series, No. 98-E-2, (Bank of Japan, Tokyo)
Council on Economic and Fiscal Policy, 2007, “On the Promotion of Integrated Reform of Social Security and Tax System,” October 17s, 2007, <http://www.keizai-shimon.go.jp/english/expert/pdf/2007/23th_01.pdf.>.
Committee on Social Security Review, 2008, Syakai-Hosyo-Kokumin-Kaigi Niokeru Kento Ni Shisuru Tameni Okonau Kouteki-Nenkin-Seido Nikansuru Teiryotekina Simulation (Quantitative Simulation for Contributing Discussions in the Committee on Social Security Review),” May 18.
Hatta Tatsuo, Charles Y. Horioka, and Wataru Suzuki, 2007, “Aging, Saving, and Public Pensions in Japan,” Asian Economic Policy Review (2), pp. 303–319.
Iakova, Dora, 2004, “Pension Reform Issues in Japan,” Selected Issues, IMF Country Report, No. 04/247 (Washington: International Monetary Fund).
MHLW, 2008, “Heisei 17 Nen Kokumin Nenkin Hihokensya Zittai Chosa (Survey of participants to the National Pension System).”
Organization for Economic Co-operation and Development, 2007a, Pension at a Glance: Public Policies across OECD Countries, OECD, Paris.
Organization for Economic Co-operation and Development, 2007b, Revenue Statistics, 1965-2006, OECD, Paris.
Ross, Stanford G., 1997, “Common Issues of Social Security and Taxation Systems,” in Interactions of Social Security and Tax systems, in Social Security Documentation (25), International Security Association, Geneva.
Sakamoto, Junichi, 2005, “Japan’s Pension Reform’” Social Protection Discussion Paper (541), World Bank, Washington DC. <http://siteresources.worldbank.org/SOCIALPROTECTION/Resources/SP-Discussion-papers/Pensions-DP/0541.pdf.
Suzuki, Wataru, and Yanfei Zhou, 2005, “Kokumin Nenkin Mikanyu-sya No Keizai-Bunseki (Analysis on non-participants to the National Pension System: with particular focus on cohort effect of non-participants),” Seido No Keizai-Bunseki Discussion Paper (75), September 2005 (University of Tokyo, Tokyo).
Tax Commission, 2007, Basic Idea for Fundamental Reform of Tax System, November, <http://www.mof.go.jp/english/tax/e0711a.pdf>
Whitehouse, Edward, 2007, “Pensions Panorama,” World Bank, Washington D.C.
Williams, David, 1997, “Legal and institutional aspects of social security and taxation reforms,” in interactions of social security and tax systems, in Social Security Documentation (25), International Security Association, Geneva.
Prepared by Masaaki Iizuka.
There are also mutual aid associations with coverage of government employees and private school employees. This paper primarily focuses on the NP and to a lesser extent on EPI, see Sakamoto (2005) or Iakova (2004) for details on the other structures.
Public support for raising taxes has been difficult to achieve. Japan introduced a consumption tax of 3 percent in 1989 and raised the rate to 5 percent in 1997. Changes in the prime minister and his Cabinet followed in both cases, although the tax hike was not the only explanatory factor.
Those people either do not participate in the NP system or have arrears of more than two years.
The public pension system has various categories of participants with differing contribution obligations. Category-1 insured includes the self-employed, farmers, and students. This group pays a fixed premium per month. They can be exempted from payments of contributions based on their status, but benefits would be reduced. Category–2 insured includes private company workers and public service employees. The premium for this group is remuneration based and borne evenly with the employer. Category–3 insured are insured spouses of Category–2 participants. They do not directly pay premium.
The Social Insurance Agency will be required to ask the Tax Agency for forcible collection of deliberate delinquency after privatization in 2010.
These increases will possibly be in addition to a 1 percent hike in the consumption tax rate that might be used to finance the increase in the government’s contribution to the NP envisaged under the 2004 pension reform (from ⅓ to ½ of NP payouts).
Other cases are: I) all pension-age citizens immediately receive benefits irrespective of past contributions and III) all pension-age citizens would receive a pension plus a benefit based on past contributions. Case IV is a more generous version of III. The required consumption tax increases differs under each scenario.
See William (1997) for more details on coordinating issues.
