The NBER is about to publish the proceedings of a conference held in March 2002 and entitled "Structural Impediments To Growth" (available online HERE ). Among the many intresting papers is one by Robert Dekle of the University of Southern Carolina entitled "The Deteriorating Japanese Fiscal Situation:Future Prospects in an Era of Population Ageing (available online HERE ):
The deteriorating Japanese fiscal situation has attracted worldwide attention. If the Japanese fiscal situation does not improve, the huge public debt is expected to sharply increase Japanese interest rates; lower Japan�s international credit worthiness; and adversely impact the welfare of future generations. In this paper, we assess what current Japanese government fiscal policies mean for the future evolution of the Japanese public debt, and the economy in general, given the inevitable aging of the population. Owing to a very weak domestic economy, that lowered tax revenues and raised government spending, the Japanese fiscal balance worsened dramatically over the last decade. The government budget, in surplus until 1992, turned negative in 1993, and subsequently, deficits have continued to worsen, reaching over 10 percent of GDP in 1998.
Correspondingly, Japanese government debt ballooned. The government debt-GDP ratio increased by over 10 percent per year in the late 1990s. By 2000, Japan had the largest debt-GDP ratio among the OECD countries. The Japanese fiscal situation may continue to look grim, especially given the rapid aging of the population. The population aging is expected to slow economic growth, and raise future government healthcare and social security expenditures. The percentage of the population over the age of 65 has grown rapidly, and now stands at about 15 percent. By 2020, that percentage 2 should approach 25 percent, and by 2050, 33 percent.
In this paper, we first review how Japan got into this current fiscal mess in the mid- to late-1990s, from a healthy fiscal situation in the early 1990s. We next perform some calculations of debt dynamics, and show that with unchanged fiscal policies, Japan�s public debt-GDP ratio rises to between 260 and 380 percent in 2020 and to between 700 to 1300 percent in 2040-- clearly unsustainable levels. We perform calculations showing that for the Japanese public debt to be sustainable, significant increases in taxes, or cuts in government spending are necessary.
Finally, we report the results of a previous simulation exercise that explicitly incorporated, the effects of an aging Japanese population. In the simulation exercise, we explicitly modeled the interplay between government fiscal policies and household and corporate behaviors. This is important, since in reality, fiscal policies can clearly affect private behavior, and these changes in private behavior may, in turn, influence the dynamics of government debt.
The simulation exercises showed that absent cuts in government spending, taxes would need to increase from the current 28 percent of GDP to over 40 percent of GDP by 2020, for the government debt to be sustainable. The tax increases and the inevitable aging of the population are projected to sharply reduce household saving rates. As the labor force declines, and the need to equip workers with capital decreases, corporate investment rates are also projected to fall. The surge in Japanese government debt in the 1990s has raised questions about the sustainability of this debt, especially given the rapid aging of the Japanese population.
The dynamics of the debt to GDP ratio are highly sensitive to real interest rate and real GDP growth rate assumptions. We assume that between 2000 and 2040, real GDP growth averages about 1.2 percent, annually. We make two real interest rate assumptions, of 3 percent and 6 percent. Since the early 1980s, Japanese real interest rates havaveraged about 3 percent; the 3 percent real interest rate assumption is also used in other studies to project the path of future Japanese government deficits (IMF, 2000; Jinno and Kaneko, 2000).
In the future, however, the Japanese government may no longer be able to borrow at these low domestic rates, and may have to borrow at higher international real rates. Recently, the nominal interest rate on five to six year maturity Japanese Government Bonds (JGBs) have yielded close to 4 percent. Since inflation rates are essentially zero, the real rate on Japanese government borrowing is now close to 4 percent. The 6 percent real interest rate assumption reflects the average real cost of borrowing in international financial markets in the 1990s.
In this paper, we argued that the future prospects for improvements in the Japanese fiscal situation are grim, unless the government carries out significant fiscal reform--by raising taxes or cutting spending. For example, we showed that under unchanged spending policies, taxes would need to increase from the current 28 percent of GDP to over 40 percent of GDP by 2020, for the government to be solvent. Japanese citizens should brace themselves for painful adjustments in the near future, in the form of lower public services or higher taxes. A way out of these painful adjustments, however, is for strong growth in real GDP to resume. For example, if real interest rates are 3 percent, a real GDP growth rate of slightly in excess of 3 percent can imply falling debt-GDP ratios.
In this paper, we assumed that real GDP growth averages 1.2 percent per year from 2000 to 2040. This assumed labor augmenting technical progress of about 1.2 percent per year, or total factor productivity growth (TFP) growth of 2.0 percent per year. TFP growth of 2.0 percent is an assumption of the high side, since it is about equal to Japan�s average TFP growth between 1970 and 1990, and Japan in the future is not expected to be as innovative as in the past (Branstetter and Nakamura, this volume).
What lowers GDP growth from 2.0 percent to 1.2 percent is the dramatic 0.8 percent annual decline in the labor force, caused by the aging of the population. Thus, one way to increase GDP growth is to raise the labor supply. Ono and Rebick (this volume) argue for removals of structural impediments that restrict the movement of labor between firms and discourage women from participating to a greater extent. Another possibility that has received scant attention until now is to promote immigration into Japan. Of high priority are further studies on the impact of increased foreign immigration on Japanese growth, saving, andthe government debt.
Whilst I consider this paper to be useful in focusing attention on a problem which is going to make itself pretty well known to us soon enough, I have my doubts about the validity of these kinds of projections, simply because we do not really know yet what is going to happen in various key areas. In the first place we simply lack experience of the economic consequences of systematic population reduction (since the time of the black death at any rate, and the little we know about this particular moment in European history would tend to suggest it was none too poitive as a precedent). So when we say that possibly economic growth will reduce by 1% per year (or simlilar), we really are guessing.
One of the first problems is to make forward projections based on previous experience. One god example of this is illustrated by the Federal Reserve paper below which demonstrates how projections for Japanese economic growth were systematically revised downwards in the 1990's. Another problem is one of principal: what is the validity of making a partial equilibrium analysis of a process which is thought to be passing through a structural adjustment, and not steadily approximating to an equilibrium state.
For example Dekle experiments with various possible tax and spending scenarios. But both of these variables are likely to have important secondary consequences on employment creation and consumption. How sensitive, for example, is Japanese employment to every additional 1% hike in taxes? Or how sensitive are the values of accumulated private savings (housing values, equities etc) to demographic processes and declining GDP, ie what is the likely result of every 1% decline in GDP on equity values, and every 1% decline in population on property values? And how would this evident wealth effect translate itself into secondary consequences for the growth of GPD. That is far from having a system approach a stable equilibrium path, we could see one which is cascading away from one.
Another major incognito is the real level of outstanding Japan government debt. A wide variety of numbers are tossed in the air, largely depending on what factors are included and which excluded. The official net debt of the Japanese government is only of the order of 45% of GDP. A small enough figure it would seem. But this does NOT include unfunded social security liabilities - a not unimportant isue in a rapidly ageing society. Adding these in you reach a figure of between 70% and 80%, depending. But then ther are, as this paper spells out three more catgeories of unfunded liabilities.
Firstly there are potential losses on government assetts - eg soft loans that may never be repaid. Doi and Hoshi (see above) estimate that these liabilities could reach 9% of current GDP.
Secondly there are explicit government guarantees for private sector lending. This could amount to some 10% of current GDP.
Finally there are implicitly guaranteed private sector loans, in particular recognised in terms of the governments liability to recapitalise the banking system in the eventuality that this is neded - this could be a further 2% of current GDP.
It is by accumulating all of these contingent liabilities that the figures for Japanese government debt in the order of 130% of GDP are arrived at.
This paper is useful if only for the fact that it makes plain the serious scale of the Japanese problem, and for the fact that it at least concludes with a recognition that the problem of Japanese society confronting the ned for increased immigration is receiving 'scant attention'. This situation I fear, even if it were to change, would be a case of too little, too late for reasons whiach I will explain elsewhere and often.