Economic White Paper: Cutting Butter with a Saw?

Takamitsu SAWA (Professor, Kyoto University)

The 2005 government white paper on the Japanese economy and public finances, which the Cabinet cleared earlier this month, has a chapter titled "From Public to Private: Restructuring the Government Sector and Its Challenges." It makes the following points:

* Big government could have negative effects on economic activity as it blocks the economy from achieving an efficient distribution of resources.

* A qualitative review of spending programs is needed to limit the scale of spending and the growth of the "national burden rate" (the ratio of taxes and social security contributions to national income).

* Privatization has various merits, such as improving efficiency and diversifying services through ingenuity and resourcefulness. Privatizing the postal services will help invigorate the economy not only by improving managerial efficiency but also by creating mechanisms that allow for efficient private-sector use of huge amounts of postal savings and life insurance funds.

* Smaller government should raise the administrative and fiscal efficiency of local governments. This will require expanding their discretionary authority, or streamlining the centralized across-the-board administrative system. Local tax and fiscal reforms and municipal mergers (involving cities, towns and villages) are steps in this direction.

To prove the first point, the report cites data from panel discussions conducted by 30 OECD countries from 1988 to 2003. The data shows a negative correlation between real economic growth and the government-spending share of gross domestic product.

In other words, the real economic growth rate of a country with a high spending-to-GDP ratio is generally low. Conversely, the growth rate is high for a country where the ratio is low. This, however, does not necessarily suggest that Japan will be able to boost its growth rate over time if it reduces government spending relative to GDP.

The second point is presented in a more elaborate way -- first by a poll in which people are asked to choose "the most desirable" policy for (a) social security payments, (b) spending on public services (education, defense, police), (c) public-works expenditure, and (d) the national burden rate. Pollees are then asked to make a similar choice concerning a set of eight policy patterns on the assumption that the policies selected reflect the "sense of utility" among pollees.

To my knowledge, such an elaborate statistical method has not been used in any similar foreign government publication. Authors of Japanese white papers seem to be using unnecessarily complicated methods -- as if to cut butter with a saw. By the same token, statistical results of a poll can be distorted, not only qualitatively but also quantitatively, if an overly sophisticated method is used.

The third point is illustrated by graphs showing how per capita profits and productivity of three privatized entities (NTT, JR and Japan Tobacco) changed over time.

Staff reduction was the biggest factor contributing to productivity gains. The reason is simple: Cutting jobs, which was difficult under state management, was made easier by privatization, thus raising managerial efficiency. Privatization increased efficiency not necessarily because it exposed the privatized companies to market competition, but rather because it freed them from the yoke of excess employment that had restricted their freedom under public management.

Generally, national or public management is the norm for network services (telecommunications, electric power, mail, railways, expressways) in their initial stages, partly because of the enormous investments required to build basic infrastructure and partly because of their universal nature. Experience shows, however, that generally these services are eventually privatized or liberalized by shifting ownership and operation of the infrastructure to the private sector.

As for the fourth point, an attempt is made to prove that expanding the size of a local government will reduce the per-resident administrative cost. Specifically, populations of local districts (cities, wards, towns and villages throughout the country) and per-resident administrative costs are logarithmically plotted. The graphs show a negative correlation, up to a point, between the two -- namely, that enlargement of local districts through mergers help improve administrative efficiency.

John Maynard Keynes once said a line should be drawn between what government should and should not do. The government of Japan must figure out, from the viewpoint of costs and benefits, how it should carry out what it needs to do.

Let me give two examples. First, it should be remembered that not all public-works projects are bad. Some are needed to make this country a better place to live. Given tight budgetary restraints, the government should set its priorities carefully while paying due attention to costs and benefits.

Second, social security benefits should be provided not as "safety-net" payments but as "trampoline" funding for investments in human capital. In other words, the principal role of welfare should be to reward those who take the risk of changing jobs.

(This article appeared in the July 26, 2005 issue of The Japan Times)