Allocating Roles to Politicians and Public Servants

Date December 7, 2010
Speaker HATTA Tatsuo (President, National Graduate Institute for Policy Studies)


HATTA TatsuoThe "power elite" who shape public opinion and government policy are selected in various ways. Politicians are chosen through elections, and that is the basis for their authority. Public servants, meanwhile, have been successful in examinations.

Politicians and public servants are therefore selected according to completely different criteria. So what role should each of them play in determining policy? Today, I would like to share with you my views on how people chosen through elections and people chosen through examinations should be assigned roles in the policy-making process.

Types of government policy

Government policies can be broadly divided into two types: income redistribution policies and efficiency improving policies. The former can be likened to slicing up a pie, while the latter is about making the pie bigger.

Income redistribution is an extremely big issue, and one that economists have wrestled with for years. Views on income redistribution are strongly influenced by personal values, so elected politicians should be the ones to make the final decision on how far to redistribute income.

On the other hand, policies aimed at expanding the pie through more efficient allocation of resources are based not on values but on analysis. Most policy reforms will benefit some but disadvantage others. Yet if the gainers could retain some residual after they had hypothetically compensated the losers, we call that policy an "efficiency-improving policy" or "efficiency-improvement policy." The economy of a country will grow over the long term if efficiency-improving policies are carried out successively in that country. However, in principle, the effects of reform can be predicted only after economic analysis, so public servants, academics, and think tanks clearly have a role to play. Policies aimed at increasing efficiency have traditionally been the purview of public servants, but in the case of new, distinctive issues, academics should play the central role or work jointly with public servants.

However, as the recent "screening of public-sector projects" has shown, some projects can be identified as inefficient even without resorting to economic analysis. As an example, schools built by the JICA in Cambodia cost three times more than schools constructed by other countries there. This is because they contracted all the work to Japanese companies. The stated reason is that they want to build high-quality schools. However, given the choice, Cambodians would probably rather have three times as many schools constructed, even if they are of the type built by other countries. Even without economic analysis, it is possible to infer that what is going on is little more than waste and corruption. As a result, the screening of public-sector projects is probably suited to remedying such problems.

Another example of a situation where economic analysis is unnecessary was the case of the light bulbs used in traffic lights. Compared with light bulbs, LEDs have three advantages: They (1) use less electricity, (2) are brighter, and (3) last longer, and thus have extremely low maintenance costs. Nevertheless, the switch to LEDs took a long time because of resistance from manufacturers of light bulbs.

It should be noted that even straightforward reforms such as weeding out waste from the Cambodian school construction project and switching to LEDs will inevitably result in some people being disadvantaged. Enhancing efficiency means that the construction firms that built the expensive schools and the companies that manufactured light bulbs lose out. Nevertheless, in the case of these reforms, the benefit to society as a whole is clearly greater than the losses suffered by a certain section of it.

Another, slightly more complex, example is the case of the modern postal service. When Hisoka Maejima decided to launch a postal service in Japan in 1871 after seeing those in operation in the West, he came up against adamant opposition from the traditional couriers which had established their service during the Tokugawa period. And because the couriers wielded a lot of political power, Maejima very much struggled. In the end, he came up with the solution of hiring the couriers for the new postal service. Even so, the couriers suffered a lot in the transition to their new jobs, though the welfare benefits of launching the postal service, a new "technology," were clear.

A more complex example is the Ministry of International Trade and Industry's policy of switching from coal to oil in the early 1960s. After the war, as part of the government's "priority-production" policy, the coal industry was heavily protected by the government and developed into an industry so strong that Japan was self-sufficient in all types of coal except coke. It also provided a lot of jobs, with Mitsui's Miike mine alone employing 300,000 people. However, as soon as it became possible to import oil from the Middle East, oil could be imported for unimaginably low prices.

When I was in high school, most homes didn't have televisions, and we often had school trips to watch recorded films. One of them was a documentary film titled "Karakorum". It followed a team of explorers comprising anthropologists and biologists from Kyoto University as they journeyed from Iran to the Karakorum mountain range. The first scene showed a fire burning in the Iranian desert, and it was explained that what was burning out there in the desert was Middle Eastern oil, and that it might account for a significant portion of the world's oil supply in the future. As this was back in the 1950s, this prediction was taken merely as a possible scenario.

In the 1960s, that prediction became reality. In 1961, when I entered university, a local public bathhouse owner told me that he wanted to cut costs by heating the baths with oil instead of coal, but that some law or ordinance required that he use coal, in order to protect the coal industry. Such anomalies were already becoming clear in 1961, and before long the government took the decision to start importing oil on a large scale. As a result, however, huge numbers of jobs were lost in the coal industry.

Policies aimed at making the pie bigger obviously result in both winners and losers. It is necessary to analyze whether the policies will actually increase the size of the pie. That is the role of public servants, academics, and think tanks.

However, it should be noted that a numerical cost-benefit analysis is not necessary for justifying many policy reforms including trade liberalization; simple reasoning of economics may be enough to decide whether liberalization is advisable. When external diseconomies are involved, however, a cost-benefit analysis is necessary in determining whether the policy is efficiency-improving.

Efficient resource allocation

I'd now like to turn to the question of what kind of policies can expand the size of the pie, and also to take a closer look at the roles that markets and the government need to play in this expansion.

If things are left to the market, resources will flow in the direction of higher productivity, where higher rewards are paid. In this sense, markets allocate resources efficiently. When a market fails, however, the government must intervene in the market to attain efficiency. There are four types of market failure: externalities (i.e. external economies and diseconomies), public goods, asymmetry of information, and economies of scale.

It is actually possible to prove the proposition that if there were no market failures, leaving things to the market would lead to an efficient resource allocation. This proposition is called the "Basic Theorem of Welfare Economics."

If it were possible to increase a person's utility without reducing that of anyone else's, the situation would have been a wasteful, inefficient one. "Efficient resource allocation," therefore, is defined as the allocation where increasing someone's utility (i.e., standard of living) would require someone else's utility to be reduced.

Let us use a diagram to look at efficient resource allocation. The solid curve in Figure 1 is the "utility frontier" (also known as the "utility-possibility frontier"). The frontier indicates the maximum utility level of person B for each utility level of person A when a given level of technology is applied to fixed levels of resources in an economy. The utility combinations along the frontier therefore are attained under efficient allocations of resources. In other words, at any point of the frontier, the only way A can increase his standard of living is by reducing that of B's. At point J, however, there is still waste, so both can increase their standard of living. Put simply, the aforementioned Basic Theorem of Welfare Economics can be rephrased as follows: In the absence of market failure, leaving things to the market will put the economy on the frontier.

Maximizing social welfare

Let us now assume that the utility frontier has been reached thanks to the market mechanisms and to the government's corrective measures of market failures. The utility frontier, however, contains an infinite number of points. Which of them is best from a social-welfare perspective?

The criterion that yields the answer to this question is social welfare, which reflects value judgments concerning income distribution. The social welfare function assesses the level of social welfare for various combinations of utility among individuals. The indifference curves shown in Figure 2 put combinations of utility levels for person A and person B in order. Obviously, the higher the curve, the better the situation is from a social perspective. Point L represents equality, but if we move to point J, making A slightly poorer but B a lot wealthier, the value to society will be the same as at L. In addition, making B extremely wealthy and A extremely poor will constitute a similar level of social welfare to a situation where A is wealthy and B is poor. These curves represent a specific social welfare function, i.e. value judgment. In Figure 2, social welfare is maximized at point M.

In order to move from a point such as Q to point M, the government needs to use income redistribution policy measures. Meanwhile, different political parties have different value systems, as well as the resulting social welfare functions. It is the role of politicians to combine the utility frontier and the social welfare function of their parties and indicate the point to which their party endeavors to lead the country, i.e. the position at which it believes social welfare will be maximized.

Efficiency improvement policy

In reality, however, our economy is not on the utility frontier, because it is filled with market failures. In addition, there are barriers to entry or other regulations or laws that hinder the movement of resources, which are called government failures. As a result, the actual economy lies within the frontier. The dotted line represents "utility-possibility frontiers when distortions exist." The combination of the utilities of the two people will move along each line when income is redistributed between the two.

Let us assume that we have moved from point J to point E in Figure 1 following the trade liberalization of agriculture. The living standards of farmers (person A) have dropped a little, but other people (person B) have benefited. However, a subsequent redistribution of income would take us to point K. Here, although B would compensate A, B would still gain a little for himself. As a result, both sides could put themselves into a better situation than before. Thus, if income redistribution after the initial reform could lead to a move into the oblique line area with J as its apex, the policy is said to be one that improves efficiency.

Hence a policy is "efficiency-improving" if the new utility possibility curve is uniformly outside of the utility possibility curve compared with before the reform.

The impossibility of compensation

Efficiency improvement policies have been criticized because most of these policies make someone worse off. Such a criticism could be avoidable if a proper compensation is provided with each individual policy to ensure that both person A and person B end up better off, realizing a Pareto improvement. In terms of Figure 1, this would mean not moving to point E and stopping there, but rather redistributing income to move to point K. By carefully awarding compensation to accompany each policy, a Pareto improvement could be realized. Naturally, social welfare would improve.

However, providing compensation is very difficult. Take the introduction of an antitrust law, for example. The loss suffered by a monopoly may be estimated by the difference between its profits before and after the introduction of the law. However, if the government compensates the company by giving it the amounts equivalent to the fall in its profits following this reform, the company would lose any incentive to improve its business. In other words, basing compensation on the loss caused by the reform is not an option.

On the other hand, in order to finance the compensation, the government has to tax individual consumers according to the amount they gain from the introduction of the antitrust law. However, calculating the value of these gains is difficult. Therefore, the government could instead tax them based on the amount they personally consume. Under such taxes, however, the consumers would reduce their consumption in order to avoid the tax payments. A moral hazard therefore also arises here.

Accordingly, compensation cannot be awarded on a case-by-case basis. All that can be done is to take some measures of compensation to alleviate the pain when a reform induces a drastic change to a major industry.

Successive social welfare improvement and decentralization

The policy principle of adopting only those policies that improve social welfare may be called the "principle of successive welfare improvement." Although it is difficult to attain a Pareto improvement after each reform, it may still be possible to adopt the principle of successive welfare improvement. There are few problems with this principle, however.

First, politicians have to make all the decisions themselves in choosing a welfare improvement policy. Value judgments need to be made, which involves a lot of effort and effectively makes such decision-making impossible. There are so many policies that it is impossible to make decisions on each and every one of them based on values. Second, decisions are based on the value judgments of the politicians, and therefore the criteria lack transparency. The basic shortcoming of the principle of successive welfare improvement is that policy making decisions cannot be decentralized, which makes this principle unrealistic.

Long-term welfare improvement delivered by efficiency-improvement policies

The shortcoming of an efficiency-improving policy was that it would generally reduce the utility level of some. John Hicks and Harold Hotelling, who are among the creators of the concept of consumer surplus, however, were optimistic about the long-run potential of efficiency-improvement policies.

Hotelling, for example, wrote this about the Tennessee Valley Authority (TVA) to the following effect: "The TVA will benefit those living near the Tennessee Valley by providing flood control. On the other hand, taxpayers who do not live near the valley will face higher tax bills. However, projects like the TVA are implemented nationwide, and looked at as a whole, they will probably provide all people with benefits that exceed their losses from the tax increases."

The policy principle of implementing all of the efficiency-improvement policies may be called the "efficiency-improvement principle." Hotelling's view is: "applying the efficiency improvement principle would make everyone better off in the long run."

The concept of social welfare didn't exist at that time, but, if it had existed, he might have said: "applying the efficiency improvement principle will improve social welfare in the long run." Such a statement would allow the possibility that even in the long run some people might find themselves a little worse off than before the initial situation.

Premises for employing the efficiency-improvement principle

It is true that in many situations, the efficiency-improvement principle can improve social welfare in the long run. On occasion, however, welfare-improvement policies may be preferable to plunging forward under the efficiency-improvement principle. Let us imagine, for example, that a World Bank-financed dam is going to be built in a developing country in which efficiency-improvement policies are rarely pursued. If the dam is built, people's lives will improve significantly, though the people living in the villages that will be submerged by the dam will obviously suffer. Moreover, because no other policies to improve efficiency are pursued in this country, the dam will mark a once-in-a-lifetime policy shift. So those that benefit will benefit enormously, yet those that suffer will suffer enormously. As a result, it may be possible to argue that one-off efficiency-improvement projects should not be implemented.

However, the situation in countries like this is quite different from that of Japan. In countries that meet the following three criteria, it will probably be advisable to employ the efficiency-improvement principle.

First, the country is pursuing numerous other policies aimed at improving efficiency. If this is the case, the income redistribution effects of each may cancel each other out.

Second, citizens are free to choose their occupation and where they live. Even in the case of the dam I discussed earlier, if the residents are free to choose where to live, they will move to locations that will benefit from the project over the long term. If they can't, the dam should not be constructed.

Third, a strong safety net is in place. Even if the other two criteria are met, bad luck might still mean that a person who suffered from the policy is unable to benefit from any other policy. However, a safety net might give that person the chance to rebuild their life.

The impossibility of mixing both principles

If the efficiency-improvement principle is employed, the bureaucracy can select efficiency-improvement policies according to objective criteria. This will help the economy develops dynamically. Nevertheless, politicians will sometimes want to assess improvements in welfare, and in doing so will incorporate some value judgments relating to the distribution of wealth. In fact, the attempt to combine the efficiency-improvement policies with policies to improve social welfare may be a mark of the current political leadership. However, if both successive welfare-improvement policies and efficiency-improvement policies are employed, there is a risk of a Pareto deterioration. Let us now examine this.

With point J in Figure 4 as the starting point, we assume that two government ministries implement policies. Let us assume, for example, that the Ministry of Economy, Trade and Industry pursues an efficiency-improving policy and the economy moves from point J to point E. Because E is on the utility-possibility curve (b), which lies above (a), the policy increases efficiency. This policy reduces social welfare, however, increasing inequality. Next, let us assume that the Ministry of Health, Labour and Welfare implements a separate welfare-improving policy, which takes the economy from point E to point Q. Clearly, point Q is less efficient than point E, but social welfare is improved.

However, point Q, which is where the economy is taken by the combination of the policies of both ministries, reduces the standard of living of all citizens compared with point J. Specifically, point Q reflects Pareto inferior to point J. This is the result of the two ministries basing their policies on different criteria. The negative aspects of both policies remain, so both efficiency and welfare decline.

If either the principle of steadily improving welfare (i.e. the policy of always employing welfare-improvement policies alone) or the efficiency-improvement principle are maintained from start to finish, this kind of Pareto deterioration cannot occur. We are therefore forced to make a straight choice of either one or the other.

In developed countries like Japan, it can be said that employing the efficiency-improvement principle and using systems for redistributing income at the same time is conducive to long-term development.

HATTA Tatsuo

The significance of political leadership

If the efficiency-improvement policy is pursued, the utility frontier will gradually be approached. If income redistribution policy from the rich to the poor is carried out concurrently, point M in Figure2 will eventually be reached. Alternatively, we might start the redistribution policy after reaching point Q initially through efficiency-improving policies. If efficiency-improving policies are combined with income redistribution policies in this way, point M can also be reached in the long term.

In this way, the fundamental role of politicians is to redistribute income under the efficiency-improvement principle. However, they also have other roles that fall under this principle:

(1) Measures to alleviate the effect of abrupt change

Because reforms aimed at improving efficiency produce both winners and losers, measures will often be required to mitigate abrupt income declines caused by the reforms.

When oil import was liberalized in early 1960s, for example, the government introduced measures to mitigate the burdens of ex-coalminers. To this end, the government established a body called the Employment Promotion Corporation, which constructed large numbers of apartments in places like Tokyo and Osaka and made them available to former coalminers. The corporation also provided subsidies to companies employing ex-miners. This policy of switching from coal to oil was a structural reform that Japan can be proud of. What was great about this policy was that money was invested not in former mining towns such as Chikuho and Yubari, but in places to which the ex-miners were moving, like Tokyo and Osaka. It therefore encouraged them to relocate.

Adopting measures that help mitigate the abrupt implementation of policy is one of the roles of politicians. In other words, the role of politicians is to regulate the speed of implementation of the efficiency-improving policies that government officials have selected.

(2) Prioritizing

In order to formulate an efficiency-improving policy, alternatives need to be compared, comparisons with other countries need to be made, concerned parties need to be consulted, and interests need to be coordinated. A huge amount of effort and energy is therefore required. As a consequence, it is necessary to choose which of the countless reform agendas should take priority. This is a role for politicians.

(3) Creating an environment in which government officials can perform their functions effectively--civil-service reform

Japan's public servants have not properly fulfilled their role of pursuing pie-expanding policies. Instead, they have devoted their energies to protecting their own vested interests. They have also formed ties with those Diet members who support specific ministries, in order to protect interest groups in the private sector. The Japanese civil-service system has in some ways forced public servants into doing such things.

The current system pressures government officials to retire before the mandatory retirement age, thus officials have to establish cozy ties with private-sector companies to find positions after retirement.

Another role of politicians, therefore, is to implement civil-service reforms that allow officials to keep working until the mandatory retirement age. This would create an environment in which government officials can concentrate on developing policies so as to expand the pie without being influenced by interest groups.

(4) Supervising the bureaucracy

The bureaucracy needs to be monitored whether policies being formulated are really going to be effective for improving efficiency. To this end, making them disclose information is fundamental, but it is also important for politicians to establish an official body for performing such checks.

The Cabinet Office's Regulatory Reform Council (this body seems to have its official name changed every three years, so I will use the name to which it is generally referred) is charged with checking whether the various government ministries and agencies are allocating resources efficiently from the perspective of the nation's citizens as a whole. As soon as a ministry seems to be veering towards protecting the interests of industry groups, it demands the disclosure of various types of information. If it finds that the ministry is pursuing a policy that will promote inefficiency, it points it out and let them redress the policy. If the ministry is unconvinced, the Cabinet Office minister consults the minister in charge of the ministry. The council is therefore a means for politicians to take the lead in monitoring the performance of public servants.

In addition, when a ministry tries to protect its interests by manipulating or distorting reforms, it will be necessary for steps to be taken, under the leadership of the prime minister, to restrain the ministry's power. This was the role of the Council on Economic and Fiscal Policy. Through this Council, political leadership functioned to eliminate monopolies held by single ministries over policy decisions, always ensuring that another, alternative analysis is performed, and to thereby establish an environment in which public servants can perform the roles they were originally employed for to the full.

The two councils functioned reasonably well for a time under the government of the Liberal Democratic Party. Towards the end of the LDP's time in power, however, the government caved in to pressure from interest groups and certain government ministries and watered down the power of the bodies. The degree to which these bodies function constitutes a barometer of the strength of the government's political leadership.

(5) Political leadership during the process of transition

The new administration has made "politicians' initiative" (i.e. having politicians, not public servants, take the initiative in policy setting) one of its major slogans.

In the initial phase of this shift to politicians' initiatives, politicians from the ruling party determined each and every policy directly. They tried to exercise political leadership far in excess of its role as mandated by the efficiency-improvement principle.

This can be seen as an attempt to implement the principle of welfare-improvement. However, as a result of this initiative, (1) politicians could only deal directly with so much and very few policies were actually implemented, and (2) the opinions of the minister, vice minister, and parliamentary secretary representing the interests of each ministry were seriously divided among different ministries, so the government did not present a social welfare function. As a consequence of these problems, it became clear that giving politicians complete control over the implementation of policies aimed at steadily improving welfare is fraught with difficulties.

Alternatively, politicians' initiatives may be seen as the new administration's attempt to demonstrate to government officials how the pie-expansion policies should be formulated. In its attempt to engineer a regime change, the new administration may have had no choice but to flex its muscles in this way.

Nevertheless, a key task when promoting political leadership is to create an environment in which public servants take back their original roles. Once such an environment is created, public servants should feel free to use their analytical capabilities and access to information to focus on policies aimed at increasing the size of the pie without the interference by politicians.

Perhaps, the greatest failure of the "politician's initiatives" in the current administration is its failure to promote the civil service reform.

Policy formation and social science

The formulation of policies aimed at improving efficiency (i.e. policies designed to increase the size of the pie) must be positioned as a role of the bureaucracy. Once society is changed into one where the bureaucracy focuses on implementing efficiency-improving policies, the society will be filled with vitality. I believe that by defining the role of public servants in this way, social sciences come to play an extremely important role.

In many other countries it is becoming normal for master's or doctoral graduates to become public servants, and two thirds of students at the National Graduate Institute for Policy Studies already come from overseas. These students return to their countries and advance their careers, but in Japan this doesn't happen. Most Japanese students do not attempt to acquire a master's or doctoral degree to obtain a job in a government ministry.

Society as a whole should be made to recognize that the role of public servants is to increase the size of the pie. Also, an education system and promotion system for government ministries should be reformed to that end. In particular, the government should be made to employ more officials who possess analytical abilities necessary to implement efficiency-improvement policies.

On the other hand, taking measures to redistribute income and ease dramatic impacts of reforms are major roles for politicians. Another extremely important role for them under the current plight of Japan is to allow public servants to devote themselves to pie-expansion policies and establish an environment in which they can do so free of distractions.

Q&A session


You said that in Figure 4, a policy of moving from point J to point E would, for example, be the role of public servants, and that it would be implemented by the Ministry of Economy Trade and Industry. But suppose politicians implement income-redistribution policies. Would they involve moves from point E to point Q, for example?


Different politicians may take different approaches of redistributing income, starting at point J. For example, a politician may choose to redistribute income by cash transfer. Through this redistribution, the economy will move along a distorted utility-possibility curve from point E to a point like K.

However, politicians may attempt to redistribute income at point E by strengthening employment regulations, in the name of reducing differences in the treatment between regular employees and those dispatched by staffing agencies. This policy may bring us to a point like Q, which is less efficient than E.


Some policies, like pensions and health insurance, are redistributive, but also offer a tradeoff between efficiency and fairness. If public servants are always trying to boost efficiency, they might, for example, keep scaling back pension schemes. Is that acceptable?


That isn't the case. Obviously, income should be redistributed through progressive taxation, inheritance taxes, and welfare payments. Public pensions, however, should not be redistributive; they should be administered by the government for the sake of efficiency.

For example, the National Pension Plan (Kokumin Nenkin) is necessary to prevent a moral hazard in the system of welfare benefits. If there were only a system of welfare benefits but no national pension system, it would be pointless for a low income person to put aside savings for retirement. For them, it would be better to spend all their money so that they can receive welfare benefits. This means that if a society provides welfare benefits it also needs a national pension plan that forces people to save and provide for in their old age. Thus, the goal of a national pension plan is therefore to prevent a moral hazard, which means that it is essentially aimed at efficiency.

The Employee's Pension Plan (Kousei Nenkin), meanwhile, exists to correct market failure, caused by adverse selection with private-sector pensions. Since the private sector insurance companies are unable to know detailed health information of each participant, they are not able to set contribution rates according to his or her expected lifespan. Instead, they keep contribution rates at a uniform level, regardless of each individual's expected remaining lifespan.

Accordingly, people whose parents died young because of illness and therefore could be at a higher risk of falling ill themselves are unlikely to enroll, believing that they will be able to get by on just their savings. As a result, only those who believe they will live a long time take out pensions. This means that only people with many more years left to live are enrolled. As a consequence, the rates of return offered by private-sector pension are extremely low, making them unattractive to ordinary people with average remaining lifespans.

To overcome this drawback, Employee Pension Plans, in which everyone is obliged to enroll, have been established. This is also to ensure that efficient resource allocation is attained in the pension service.

Preventing moral hazard in the welfare payment system and avoiding adverse selection in an annuity service are irreplaceable functions of public pension plans. On the other hand, there are many policy tools to redistribute income from the rich to the poor, e.g. making taxes more progressive, raising inheritance taxes, increasing welfare benefits, and providing scholarships for college education. Redistributive functions should be given those policy tools. I believe we can better evaluate the performance of the public pension plans if only the irreplaceable functions of the public pension plans are assigned to them.

One of the reasons the Japanese pension system has become such a monster is because it has been loaded with redistributive functions, making it difficult to straightforwardly evaluate the system from the viewpoint of efficiency.


The politicians who make the decisions and the politicians who take into account the opinions of the electorate really ought to be the same people, but very often they are not. Since the days of Aristotle, democracy has never been the best system. Dictatorships and aristocracies also have their strong points. To overcome the drawbacks of democracy, presidential or prime ministerial systems have been established. "Aristocratic" public servants and scholars have been also given roles, and I feel that a good balance has been struck. I would like to hear your thoughts about such decision-making and the taking into account of public opinion.


There are two ways to take into account the views of citizens: markets and elections. So I think the question is how roles should be allocated to each. In Aristotle's era, the role of markets may not have been sufficiently recognized. It has become gradually clear, however, that markets play an extremely important role in reflecting the preferences of the people in the delivery of goods and services. It is true that markets fail sometimes. Nevertheless, most of the market failures can be overcome without elections.

One of the tools to reduce market failures is cost-benefit analysis. In 1974, for example, Kakuei Tanaka began redistributing income to the provinces and put a stop to the rapid economic growth of the 1960s. In those days, though, few cost-benefit analyses were performed. Government ministries and scholars should really have analyzed whether dishing out money to the regions was going to benefit the country as a whole in the future. However, at that time, when Marxian economics was still dominant, nobody even thought about doing so. However, I think that decisions on regional policy in the future should take cost-benefit analysis into account.

By the way, in today's lecture, I have only mentioned the role of politicians in the context of resource allocation. I have ignored issues like foreign policy, which carry a lot of uncertainty and cannot be solved through economic analysis alone. In other words, in areas where market information is useful, public servants should make as much use of it as possible. And in areas where it isn't so useful, they should leave the decisions to the politicians. For example, it would be difficult to determine policy on North Korea using market information alone. Such policies cannot be decided without reference to public opinion.


The problem with efficiency-improvement policies is how to award compensation. Such policies significantly increase tax revenues, and something should be done with those revenues. Unfortunately, however, this isn't always the case. Policies like that don't actually work in practice. A prime example of a compensation policy is the one trillion yen per year (six trillion yen in total) to cover costs accompanying compliance with the Uruguay Round treaties. In a sense, though, this was the worst policy failure we have ever seen. Compensation policies have never been successful in Japan. In the end, another huge sector emerges, and success is achieved by that sector getting absorbed. The current Democratic Party of Japan (DPJ) government has come out of nowhere and started trying to do things, and I think it's in an extremely difficult situation.


The major advantage of the efficiency-improvement principle is that even if compensation isn't awarded, it is possible that most people benefit over the long run. For example, when Russia first shifted to a market economy, the situation was dire, and many thought that it might prove impossible. However, when viewed over the long term, such shock treatment proved to be a great success.

I believe that Japan might also be in need of a number of similar reforms that are not accompanied by compensation. A good example is policy that allows fixed-term employment contracts to be renewed without limitation by reforming the employment regulations. Such reform may initially appear to expand inequality by increasing the number of fixed-term employees. In the long term, however, it would probably reduce inequality; the regular workers with vested interests would experience a decline in living standards, while the living standards of those with fixed-term contracts would rise by a large margin, as a result of an improvement in efficiency.

Though I am afraid many people will find this hard to believe, I think the root of the problem is that there is no fundamental understanding of the role of the market among Japanese intellectuals. I would like to call on academia, the bureaucracy, and particularly the media, to educate the public on this.

Let me demonstrate a case where compensation is worthwhile. With regard to fisheries, the Olympic system is currently employed, which encourages fishermen to rush to catch fish until the total quota is reached. As a result, fish are caught no matter how small they are, so the population of young fish is wiped out. If this system was abandoned, and each boat was assigned an annual quota such that they could fish carefully during the year, only large fish, which can be sold at a much higher price per kilogram, would be caught and the young fish would be protected. However, fishermen would suffer considerably until the young fish matured. They would therefore need to be compensated over the short term.


In the past, when I researched industrial adjustment policy, I found that in addition to policies on textiles and coal, the policy of reducing the number of fishing boats adopted after Japan's defeat in international negotiations concerning the establishment of 200-mile economic exclusion zones functioned fairly well as an industrial adjustment policy. The problem is that politicians take the lead with some policies even though they shouldn't. The current DPJ government's policy of providing income compensation to farmers on a household to household basis, for example, means that even welfare farmers with other jobs or businesses receive extra income.

If you have any ideas about who should fulfill the role of monitoring politician-led policies like this, which provide income compensation even to those who don't need it, I would like to hear them.


The current household-base income compensation is designed as a reward for people who have reduced acreage, clearly reducing efficiency. I agree with you that politicians are responsible for this inefficiency. Efficiency will be improved if household-based income compensation were introduced to replace the acreage reduction.

What is really desirable is to establish a mechanism that enables the government officials of independent ministries to point out when politicians and officials of the ministries in charge of policy implementation are going against efforts to improve efficiency. At a minimum, the Regulatory Reform Council and the Council for Economic and Fiscal Policy should be restored. This notion of dividing policies into those that can be politician-led and those that shouldn't needs to spread to journalists, academics, public servants, and, in fact, everybody. Unless this is done, I believe that things will just keep ending in chaotic debate.

Moving toward RIETI's future

I believe that RIETI has played an extremely important role ever since its establishment in analyzing various areas of policy in Japan. I do not think it a good idea to give specific government ministries monopolies on academic analysis in the respective areas of the administration. Independent checks need to be performed continually in various fields of research. I believe that RIETI has successfully performed this function, and I hope that RIETI will continue to fulfill this role in the future.

*This summary was compiled by RIETI Editorial staff.