Below are the provisional minutes of Forum #10, prepared by the Secretariat of the ESRI Economic Policy Forum (for details, see the entire discussion at our Japanese website).

Date	September 30, 2002, 14:00-17:00
Panelists	Takashi Oshio	Associate Professor, Tokyo Gakugei University	(Keynote speech)
	Toshiaki Tachibanaki	Professor, Kyoto University	(Keynote speech)
	Tatsuo Hatta	Professor, Tokyo University
	Yasuhiko Yamazaki	Professor, Sophia University
Moderator	Etsushi Kitamura	Executive Research Fellow, Economic and Social Research Institute, Cabinet Office,Government of Japan
Language	Japanese only

   At the beginning of the discussion, Associate Prof. Takashi Oshio and Prof. Toshiaki Tachibanaki gave the keynote speeches. A panel discussion was then held, including questions from the general public and an exchange of opinion.
 

1.  Outline of the keynote speech by Associate Professor Takashi Oshio (Tokyo Gakugei University)
 

  (Perception of the present situation and basic policy on reform)
 

• My primary topic is the pension system of employees. The present public pension system creates a significant intergenerational gap; it also generates a massive and unfunded pension net liability. The biggest problem of the present system is that the government makes public commitments to the pension benefits that exceed the tax burden that the working generation is capable of bearing. At the end of fiscal 2000, pension net liability totaled 644 trillion yen (employee pension amounted to 552 trillion yen). If no measures would be taken, the pension net liability would definitely increase further. The present system would further widen the gap between generations.
   
• The target of pension system reform can be described by just one word-"downscaling" of the public pension. The government has to stop making promises to the people that are unlikely to be kept. It is necessary to admit honestly that one cannot make something out of nothing. Ideally, it is necessary to restrict the public pension to basic pensions and to finance them along with medical and long-term care insurance for elderly people by the consumption tax.
   
• As an actual proposal toward 2004 reform, I think the Swedish system deserves to be considered as an alternative option.
   

  (Fiscal Principles of the Public Pension as It Ought to Be)
 

• The transition to a funded system is ideal in the sense that it redresses the balance of pension revenues and expenditures and, thus, does not incur new pension net liability. Moreover, it can correct the intergenerational gap. Nevertheless, the important problem of how to write off the existing past pension net liability remains a problem.
   
• On the other hand, the Swedish system, which has been attracting attention recently, does not incur new net liability and contributes to the correction of the intergenerational gap provided the insurance rate is fixed. However, the problem of the past pension net liability disposal remains in this case as well.
   
• Regardless of the chosen system, it is indispensable to downscale the public pension. Besides, both systems will generate additional burdens during the transition period. Both systems are also associated with risk of uncertain income in the post-retirement years. The comparative merits and demerits are not definite, but the Swedish system is easier to implement. An immediate transition to the funded system will pose a serious financial problem of the pension benefits to pensioners.
   

  (Other Issues)
 

• It is difficult to justify complete privatization. From the standpoint of operational costs and risk management, it is often pointed out that an aggregate operation by government is more desirable. If the present system is preserved, however, enrollment will bring certain substantial negative rates of return (those born in 2000 will lose approximately 10 percent in respect to the life-time earnings; state contributions are not included)
   
• It is desirable to create a pension system that will be neutral with respect to the decision to become an individual proprietor, public official, or private corporate employee, or to switch jobs several times. To this end, it is necessary to make the first floor portion in the form of a basic pension common to all the citizens in analogy with the present system and the second floor portion in the form of the individual accounts based on the principle of accumulated contributions. The ultimate idealized vision of the system is stimulating concentration on individual accounts by providing incentives through the tax system.
   

2.  Outline of the Keynote Speech by Prof. Toshiaki Tachibanaki (Kyoto University)
 

  (Problems of the Present System)
 

• The public pension is a system that provides insurance for the uncertainty of time of death and shortage of post-retirement income. As its aim is risk sharing, the system cannot be established without confidence with respect to social solidarity. Consequently, under normal circumstances intergenerational and intra-generational losses and gains are not supposed to intrude. Nonetheless, in reality people are sensitive to losses and gains; a system with as few obvious losses and gains as possible is better. To this end, it is good to ensure financial resources through taxes.
   
• A social insurance system that provides services according to individual characteristics, like that found in Japan, is called "communalism" in philosophy and ethics. It presents universalism as an antagonistic concept. I think a system based on universalism, in which all citizens enter the unique system, is ideal. If we adopt it, we can avoid the following problems; the biggest drawback of the Japanese social insurance system is that there are too many insurance systems.
   

  (Sustainable Pension System and Transition Policy)
 

• My proposal for the reform of the pension system is consistent with the idea of universalism, especially the system restricting the public pension to the basic pension and paying fixed benefits. For this reason, a pay-as-you-go, tax-based system is desirable. The amount of benefits is fixed at 150,000 - 170,000 yen, covered by a progressive consumption tax.
   
• The merits of tax-based system are 1) it provides security with a minimum social safety net and leaves little room for the discussion of losses and gains; 2) it ensures a broad, thin, and sure collection of fiscal resources; 3) it allows economization on collection costs due to unification of collection; 4) it contributes to the revitalization of private enterprise, as enterprises will be free from the burden of insurance premiums; 5) because the impact of indirect taxes on labor supply and savings is smaller than that of insurance premiums, it will not lower the economic growth rate.
   
• The demerits of the tax-based system: 1) it is criticized for being the same as "Daily Life Protection," but I think this criticism is wrong; 2) it loses incentives as the burden is not linked to benefits; 3) it will tend to create a stiffer, inflexible financial system.
   
• The Swedish system is based on a deemed pay-as-you-go system with defined contribution proportional to income in principle; it has possibility of large fluctuations of the benefits, and the risk for low-income earners is high. This point is in conflict with my proposal. Besides, in Sweden the proportion of the minimum security is much lower than my plan. In this connection, according to my proposal, the consumption tax rate will rise to approximately 15 percent, while there will be no need to pay social insurance premiums.
   
• It is necessary to consolidate the insurance system for pension and medical care respectively. In the medium-term perspective, it is good to consolidate medical care and long-term care.
   
• During the transition process to the new system, reducing the benefits of recipients already receiving them would be a violation of the Constitution. Furthermore, it is necessary to take care not to cut drastically the benefits of people who have already paid their insurance premiums.
   
• The reform of social security system, including the pension system, implies that the decision of such problems as tax-based systems or premium-based systems, fixed-benefit or defined contribution, pay-as-you-go system, or funded system is a national problem requiring a national referendum.
   

3.  Panel discussion

(1)  Comments

(Prof.Tatsuo Hatta, Tokyo University)
 

• It is necessary to consider the raison d'etre of the basic pension in connection with Daily Life Protection. It has become natural that people who cannot save enough for their post-retirement period become the recipients of Daily Life Protection, and this causes a moral hazard. To prevent this, it is necessary to force them to accumulate an amount comparable to Daily Life Protection. At the very least, the basic pension should be set at the level compatible with Daily Life Protection.
   
• Basic pension can be interpreted as a kind of negative income tax. All people receive a fixed amount of benefits after retirement. The low-income earners receive money from the country on balance, while the high-income earners pay money to the country on balance. The tax-based system is natural for the basic pension.
   
• Another reason for recommending a tax-based system is the difficulties involved in collecting insurance premiums by the Social Insurance Agency.
   
• A public pension proportionate to earnings has meaning because of the adverse selection problem. If we entrust this to the private sector, only those who think they will live a long life will join. In fact, the cases of pure life annuity are rare and their profitability is low. This is primarily due to the strong impact of adverse selection. Consequently, it is necessary to force those who think they will live a short life to join as well. Thus, it will be possible to provide an average pension. The problem of whether to entrust the practical operation of the funds to the private sector or to adopt operation by government is another story. We should make enrollment compulsory, and at the same time provide operational freedom.
   
• The major bottleneck in operation of accumulated pension funds is the impossibility to cope with the risk of interest fluctuations. The retired person will receive accumulated funds in the form of a life annuity, but the amount of benefits will depend crucially on the interest rate prevailing at the moment of retirement. The government, being different from the private sector, can only perform simple risk-management strategies. Hence, in addition to compulsory enrollment it is necessary to use private management.
   
• The Swedish system is unlikely to be feasible politically as Japan's rates of aging are faster than Sweden's, implying that the curtailment of benefits will be sharp. In case of transition to the funded system, it is essential to clarify the proportion of the contributed insurance premiums that will be repaid to an individual and the proportion used for disposal of past liabilities. The central task is to divide them in accounting. It is out of the element of the Swedish system.
   

(Prof.Yasuhiko Yamazaki,Sophia University)
 

• A serious problem of the present system is intergenerational fairness. The present level of pension among the elderly is somewhat excessive as compared to that of still-working generation. Nevertheless, it is necessary to look at the social security system in general, including medical and long-term care, in addition to pensions. Elderly people are now charged somewhat for their medical and long-term care, so even the present level of pensions means a real cut in benefits.
   
• The employee pension system was established during the war. As the postwar period was characterized by high inflation, it is necessary to consider the pension of generation born before the middle of 1930s, and, consequently, the disposal of liabilities separately.
   
• Under the social security system of the pay-as-you-go system, child-raising has the same value as premiums. In Germany, it was judged unconstitutional that premium for long-term care insurance was the same among families despite whether it had a child or not, and now it is pointed out that consideration for the value of child-raising is insufficient in connection with pension and health insurance. If we change into a funded system, the dependence on the future generations will diminish, thus, allowing us not to think about child-raising.
   
• When the proponents of a funded system make mention of pension, it usually means "second floor portion." However, under the present system the share of the first floor portion is larger. Almost all specialists assume that basic pension, medical and long-term care for the elderly are organized as a pay-as-you-go system. Consequently, we have to bear the responsibility in respect to future generations. This is an issue of social justice. A decision to have children or not is a matter of individual freedom that should not be interfered with. If we fail to assess properly the value of child-raising, however, the country is doomed.
   
• In connection with intra-generational fairness, the imparity between corporate employee groups and individual proprietors groups raises a serious problem.
   
• To design real intra-generational fairness, first, it is possible to fix the amount of burden and benefits in the first floor portion of the corporate employee groups as well. The other way is to adopt the Swedish system, eliminating the wall between employees and self-employed workers and setting the contributions and benefits in accordance with income. The problem is how to assess the income exactly.
   
• Concerning the response to the changes in work arrangement and lifestyle, the wall between employees and self-employed workers, and full-time and part-time workers has become thinner. This aspect makes the Swedish system more appealing.
   
• As employers have to pay a half amount of the applied employees' premiums, they tend to escape from this burden. Even if government relaxes the requirement of application to increase the number of applied workers, especially part-time workers, employers in fact often change working hours and/or wages not to meet the requirement. Hence, it is necessary to levy premiums on paid salaries despite of whatever kind of workers is employed. I propose pro forma standard taxation based on the total sum of wages. This is particularly necessary to facilitate the application of the elderly.
   
• Since fiscal 1994, pension system for old-age employees placed stress on the incentives of the workers; the sum of wage and pension benefit is now secured to increase even if benefit has to be cut when wage rises; but what is more important is incentives to employ the elderly. Employment of workers not less than 60 years old contributes pension finance because their benefits will be lowered. However, the present pension system makes it more convenient for employers to "pension off" elderly people. As a result, the retirement costs are shared by everybody. In other words, it causes an external diseconomy.
   
• As for women's pension, at least in the field of social security, we should go for split income system in the future.
   
• Both the funded system and the pay-as-you-go system have merits and demerits. We should appropriately combine the both system and explicitly incorporate the funded part into the employee pension.
   
• Using national tax resources to dispose the past liability on the employee pension system is beyond all reason. Those resources, if used at all, should be used to dispose past liabilities on basic pensions of the national pension plan.
   
• Concerning the choice between tax-based system and premium-based system, almost no countries among the developed ones finance pensions, medical and long-term care with taxes.
   
• Concerning the Swedish system, some pay attention to the fact that this system is proportionate to income, and that minimum security is funded by taxes, while others point out that the insurance rate is fixed and benefits are adjusted automatically. The former aspect is certainly appealing, but I do not think it is feasible in the near future. In connection with the latter, if we fix the insurance rate at the present level, it would lead to the sharp curtailment of benefits. First, we should step up our efforts toward raising insurance premiums. Only after should we examine the possibility of adopting the Swedish system.
   

(2)  Discussion among Panelists

(Oshio)
 

• It is wrong to think that a tax-based system is better because it has less-obvious losses and gains. Even with this system, it is necessary to look at what part of the money paid out will be received back. Now, tax and social insurance premiums are mixed and, thus, there is a lack of clarity about the role distribution. Shifting the financing of the basic pension part to financing by taxes will clarify this point.
   
• I think that the basic pension of 15,000 - 17,000 yen would be too much and require means test.
   
• The idea to make the first floor portion compulsory funded is lucid theoretically, but I think that the effect of compulsory saving is unclear. There may be a lot of loopholes.
   
• According to the examples the of foreign countries, privatization proved to have a merit of flexible response due to derivative management, but it has a problem with regard to handling costs.
   
• Adoption of the Swedish system will leave no choice but to bequeath a huge burden of 550 trillion yen in past liabilities to future generations. In this case, the merits of the transition to the funded system disappear. It may even be a violation of the Constitution, but we have to write off property rights of the present recipients. According to my calculations based on the fixed insurance rate of 15 percent, eventually it is necessary to cut benefits in half.
   

(Tachibanaki)
 

• The proposed amount of the basic pension, of 15,000 - 17,000 yen, is not groundless. It was calculated based on the actual social condition of elderly people, taking into account the amount necessary to manage their everyday life. In this case it is necessary to raise the consumption tax to 15 percent. Instead it will make the payment of insurance premiums unnecessary. It is preferable that the nation make that decision.
   
• The government share of contributions to the basic pension is one-third; it was decided to increase that to one-half. However, this plan is unlikely to be implemented. In my opinion it is necessary to make 100 percent of the sum financed by taxes. There are relatively many countries financing basic pensions by tax revenues. This plan, however, is seriously criticized because there are no examples in major G5 countries. As for me, I think the tax- based system is a good method to prevent the hollowing-out of the employee pension system.
   

(Yamazaki)
 

• If a tax-based system is the right choice for a basic pension, medical care insurance and long-term care insurance must be also based on a tax-based system according to the same logic. However, there are almost no countries with tax-based systems for all three areas.
   
• It is a mistake to think that a tax-based system will prevent hollowing-out. Even for income tax, only three-fourths can be collected from individual proprietors. This portion is equal to that of the national pension. Tax and insurance premiums are different in crucial ways. A person not even paying taxes can receive services, but someone not paying an insurance premium will not receive benefits.
   
• Raising insurance rates is also difficult. We want to put an end to passing the bills on to future generations, but in reality the previous reform froze the increase in insurance rates, contributing to the increase in bills.
   
• We have to pay attention to the impact of indexed system if we raise consumption tax rate to finance pension; the rise in consumption tax rate must be deducted from inflation rate when the benefit is raised by indexation. The working generation cannot be guaranteed an increase in salary in accordance with the increase in consumption tax, so it is necessary to share the burden with elderly people.
   

(Hatta)
 

• With respect to the difference between tax and insurance premiums, it is said that in case of insurance-based basic pension those who do not pay cannot receive the benefits, but the primary purpose of the basic pension then becomes difficult to understand. Those who do not pay a premium after all will be saved by livelihood protection. Such conditions that those who do not pay contributions cannot receive benefits contradicts the essential purpose of the basic pension in terms of economics. Consequently, the basic pension cannot be anything but a tax-based system.
   
• Some argue that a tax-funded system is strange, but my proposal is that it is funded within the generation in general, covered by consumption and income taxes that are not equal per capita. As for an individual, there are no ties between the fixed benefits and taxes. This system allows us to escape causing fiscal troubles to future generations.
   
• Basic pension of 170,000 yen is not bad; it is nearly equal to the amount received by a taxi driver in a city near by Tokyo. It is necessary to lower the level of livelihood protection and to adopt elements of a negative income tax. It is essential not to reduce the total sum of livelihood protection benefits of working elderly people as it is now; it should be reduced only in part.
   

4.  Open Discussion with Audience

(Audience)

• First, I think we should divide the problem of funded and pay-as-you-go systems, and the problem of benefit-based system or contributions-based system. We have almost no options if these problems are argued politically. It will be an economically significant discussion if we think to what extent it is possible to raise the burden and what institutional arrangements should be designed to cope with the risk based on this assumption.
• Second, concerning the difference between tax and insurance premiums, it is beyond economics; after all it is a problem of labeling and what is important is the extent to which the political bind will become looser.

(Yamazaki)

• In connection with the first aspect, it is important to combine these two systems appropriately and to divide definitely the past and future liabilities.
• With respect to the second point, I do not think it is only a labeling problem. It is important whether it will be possible to secure contribution incentives in case of taxes. The problem is that among individual proprietors there are those who can pay but do not fulfill their payment obligations. This is the main reason of distrust.

(Hatta)

• The U.S. payroll tax is said to increase payment incentives, as it is allotted to social insurance premiums. It is good to discuss bringing into focus the percentage of income or consumption taxes allotted to the basic pension. Besides, it is necessary to divide them clearly in accounting. We should make it clear that the part of the paid sum that does not return is the appropriation for past liabilities and that, at the same time, disparities among generations in this part are not large; this condition is important for us to be able to discuss this problem openly.

(Oshio)

• The part that can be managed based on actuarial principles should be financed by premiums, but the other part should be financed by taxes. Concerning the question about the choice of systems, it is necessary to bring into the question the fact that the present system is not a pay-as-you-go system in the full sense. The funded system can not be operated on the basis of benefits; the risk is associated with the fluctuations of interest rates and faced by the future elderly. On the other hand, the risk of the pay-as-you-go system is associated with the increase rate of nominal wages; it emerges at work on the basis of benefits and after retirement on the basis of premiums. There is not much difference between them when calculated in present discounted value. Eventually, taking portfolio theory of risk management into account, it becomes clear that a balance between a funded system and a pay-as-you-go financing system is necessary. However, I do not think that the pay-as-you-go part of the system will become significant.
   

(Audience)

• It was suggested in the previous question that economists tend to interpret tax and insurance premiums as the same things. I think this is a rather crude way of thinking, as shown in the answers from the panelists.
• At the same time, I think you, as economists, should demonstrate what kind of incentives will be created if your proposals are realized and how the society will change as a result.
• One of the problems of the present Japanese economy is that wealthy elderly people do not consume. I would like to ask what measures should be taken to make the elderly to consume.

(Tachibanaki)

• In this connection I studied on the procurement method that would be the most suitable from the viewpoint of effectiveness and fairness. According to the results of the simulation based on the general equilibrium fiscal model, a progressive consumption tax is the best way.

(Hatta)

• It is said that high premiums distort the economy and reduce labor incentives, but it is a wrong argument. The biggest merit of the funded system is that, even in case of high premiums, they are perceived as a part of income or a form of savings and, thus, they do not distort labor incentives. Only the part exceeding the insurance premiums in the case of a funded system-in other words the part that will not return, can be considered to have a labor disincentive effect. However, the problem of the present system is that this cannot be grasped, and it leads to distrust. If we make clear accounting of a funded financing system, it will be possible to eliminate this distrust.
• To stimulate consumption of the elderly, inheritance taxes must be raised. It is natural to assume that if the elderly use money in life they will use it on long-term care. However, due to the adverse selection problem, they rely on the family. In exchange for the care they bequeath a legacy. Raising the inheritance tax would take this part, forcing the elderly to use long-term care services provided by the market. Furthermore, to make an environment where the elderly can move to smaller houses and increase consumption, it is necessary to amend capital gains taxes. To avoid the efforts of tax saving, postponement of tax payment should be allowed until death and the tax should be levied on the remained property after death.

(Oshio)

• For ordinary people, the supply of labor is not so flexible as that considered by economists. Incentives are connected with benefits if anything. It is important to advocate consumption among the young people rather than the elderly. In this case it will be important to lessen the concerns stemming from the uncertainty of the social security system.

(Hatta)

• For ordinary people, the pension system is not considered to be connected with incentives to supply labor, but this is not true if examined rigorously. It greatly affects the decision of women to work, and it influences the choice of time for retirement. It is important to make the system neutral in respect to these points. Regardless of this, it is necessary to clarify accounting to secure intergenerational fairness and stability of the system.
   

(Audience)

• I think every person should think how individuals and families should take their own risk-responsibility. I would like to ask what you think about the roles of public and private sectors in the social security system and about an ideal society.

(Oshio)

• Concerned with financial resources of pension system, "public" resources today mean in fact "private" resources in the future. Therefore, conflict between "public" and "private", which I think has become weaker than in the past, is conflict between oneself in the future and oneself at present. We can not but make much of "private" standpoint at present in consideration of the impact of population dynamics on the performance of government.

(Tachibanaki)

• Speaking about the connection between the private and public sectors, it is high time for Japanese to choose whether communalism or universalism, as I mentioned in the beginning. From the standpoint of universalism, I assert that individuals expect the public sector to provide a minimum level of security.

(Hatta)

• From the standpoint of economics, social insurance system can be justified only in cases of income redistribution and market failure. Daily life protection and basic pension is associated with the former. If there can be adverse selection, social insurance system can not but be adopted with respect to the part of proportionate compensation. The adverse selection is not just a theoretical problem; it exists in reality. In the case of pensions, the fact that the number of lifelong insurances is small and that their profitability is extremely low implies the existence of adverse selection.

(Yamazaki)

• We had better introduce the advantages of private sector such as market mechanism and principle of self-responsibility into public sector. It is possible that the second floor portion is operated in that way with transparent accounting.
   

(Audience)

• Private enterprises are financially strapped as pension contributions are increasing step by step. Do we have to keep the huge reserve funds in the present situation of the employee pension deficit?

(Yamazaki)

• The present situation in reserve funds is serious; they do not have the necessary amounts. It is necessary to expand reserves now; it is not a time to cut them.

(Hatta)

• The debt of employee pension is far bigger than the reserves, so we can not take out the reserves. If we reduce the reserves now, we will have to pay more tax in the future.

(Oshio)

• We should not cut the reserves under the current system. However, if we adopt the Swedish system or limit the role of pension only to the first floor portion, such huge amount of reserves is not necessary.