Japan’s Prime Minister Junichiro Koizumi produced a budget last week that cut $24 billion, or about 0.5 percent of Gross Domestic Product, from Japan’s public spending, which he believes, against much “expert” and media opposition, is the solution to Japan’s recession.
The Organization for Economic Co-operation and Development produces statistics, dating back to 1960 in many cases, for total public spending in 28 of their 30 member countries as a percentage of GDP. With this evidence, one can look both at the absolute level of public spending and at public spending growth, and reach a verdict as to whether public spending, or a sharp rise therein, causes and prolongs recessions.
The verdict? To a great extent, Guilty as Charged!
Public spending of course fluctuates from year to year, and in general rises in recessions as a percentage of GDP, because tax revenues are lower than planned, while social spending is higher. However the causality here is the opposite of what we are looking for: recessions are here causing rises in public spending as a percentage of GDP, not the other way around.
Over forty years, however, the business cycle evens out, and so by regression analysis the trend of public spending as a percentage of GDP (let’s refer to this hereafter as the “government take”) can be determined. Three figures are important here; the “slope” (the amount the government take rises or falls each year), the mean level the government takes over a period, and the correlation coefficient, the extent to which the government take rises uniformly over time. (In some cases, this coefficient is very low, showing that random or cyclical fluctuations are much more important than the steady increase in government take).
In the case of the U.S., for example, the mean level of government take over the 43 years 1960-2002 (taking the OECD’s 2002 projection as the end-point) is 30.59 percent of GDP; it is increasing by 0.14 percent per annum, and about 20 percent of the variation in government take is represented by the increase over time. In other words, since 1960, the U.S. government take has remained relatively low, and has only modestly increased — it is now, for example, lower than in 1980. Federal spending has increased somewhat as a percentage of GDP over the period, but this has been balanced by a decrease in state and local spending.
The highest average government take over the period is that of Belgium, at 50.40 percent, just above Sweden at 50.38 percent. The lowest is that of South Korea, at 19.40 percent, with Turkey next at 24.16 percent. Sweden’s government take increased rapidly until a peak of 67 percent in 1993, but has since been dropping, while Belgium’s has been increasing slowly over the whole period, albeit with a significant drop, from 53 percent to 45 percent, since 1993 — the “Maastricht Criteria” for Euro membership did indeed have a beneficial effect in some countries. The highest current government take is Sweden, at 52 percent of GDP, followed by Denmark, France and Greece at 50 percent.
The highest “slope” over 25 plus years (data for some countries is available only since 1990 or so) is by far Greece, at 0.95 percent per annum, followed by Spain (0.74 percent), Sweden (0.72 percent) Portugal (0.69 percent), Switzerland (0.68 percent) Italy (0.60 percent) and Japan (0.57 percent.) Only a few countries have a negative “slope” over more than 25 years — the most notable example is Ireland, with a negative “slope” of 0.76 percent per annum over 1977-2002.
It’s also quite interesting to look at “slope” over the two twenty year periods where data are available. In general, the government take was increasing gently during the 1960’s and rapidly during the 1970’s, because of the economic difficulties of the latter decade, and because in many quarters the view of government spending was generally positive. The highest slopes in 1960-81 were much higher than over the period as a whole — Belgium at 1.71 percent (1970-81), Portugal at 1.57 percent (1969-81) and Sweden at 1.49 percent (1960-81) and the Netherlands at 1.41 percent (1969-81). Given that Portugal began from a very low base under Antonio Salazar, one can see clearly how Belgium and Sweden got public spending as high as they did, and the Netherlands, by 1981, had become the inventor of the “Dutch disease.”
Since 1981, growth rates in government take have been much lower, as the countries concerned have become more aware of the economic costs of such growth (to which we shall return later). All four countries with the highest “slope” in 1960-81 had negative “slope” in 1981-2001, in other words, the government take has declined. In the latter period, since 1981, the highest “slopes” are from Greece (0.78 percent), Switzerland (0.77 percent, 1988-2001), Finland (0.55 percent) and Japan (0.37 percent).
The correlation coefficients, over the period as a whole, are also of interest. The highest, at 88 percent, is from Japan, followed by Greece (78 percent) and Spain and France (both 69 percent). These four countries, therefore, have been the most determined at increasing the government take; neither political change nor economic recovery has affected their enthusiasm for extracting money from taxpayers or the bond markets.
The correlation between government take and economic growth is of course not a perfect one. Statistically, it is possible with ease only to look at growth since 1990, a period when OECD growth rates have ranged from Ireland’s 7.1 percent to Switzerland’s 0.9 percent (ignoring the Eastern bloc, where special considerations apply).
Of the four countries with the highest post 1981 “slope” Switzerland has the slowest growth rate in the OECD, Japan (which has the highest determination to increase the size of government) the second slowest, and Greece and Finland are both significantly below average. Third through sixth positions are occupied by Italy (fifth highest average “slope” over 40 years) Germany (part of which is of course the former East Germany), France (third highest current government take and fourth highest “determination”), Sweden (highest government take and third highest “slope”) and Belgium (highest average government take over the 40 years).
At the other extreme, the three highest growth rates over 1990-2000 are by Ireland (the only substantial negative “slope” over 1981-2001), Korea (lowest government take in 2001) and Luxembourg (below average government take and negative “slope” in 1990-2001).
Alternatively, by running a regression, one can determine that slope in 1988-2001 is 37 percent correlated with economic growth over that period, while mean government take is 60 percent correlated. Since slope and mean government take are, somewhat surprisingly, slightly negatively correlated with each other, a total of about 50 percent of the differences between different countries’ growth rates can be explained, not by education systems, natural resources, monetary policy or economic policy in general, but by government take and its rate of growth. If government take rises substantially, the economy stalls; it’s as simple as that.
A great deal of economic reality is explained by this paradigm. Euro-sclerosis is clearly mainly due to high levels of government take and, in some countries, determination to increase government take year after year. Only Ireland, which found a “holy grail” of low corporate tax rates almost by accident in the 1980’s, through the “Shannon licenses,” has been able to break free of the European disease. Switzerland, so successful an economy before 1980, has become itself Euro-sclerotic since then, raising the government takes at a rapid rate. The Dutch disease, and its Swedish equivalent, were cured when their governments finally got serious about keeping public spending under control.
As for Japan, it is very clear from these statistics that rising, and now excessive government spending are indeed responsible for Japan’s economic woes. Japan ranked No. 1 of OECD countries in 1960, with the lowest government take, and was still No. 5 out of 25 as late as 1990-1993 — the first years of the Japanese recession were, as has been many times written here and elsewhere, without doubt due to the collapse of the 1980’s asset price bubble. However, since 1993, Japan’s absolute and relative position has deteriorated sharply, to the extent that she now ranks 19th out of 25, with among the higher government takes, a very substantial relative move in only eight years. In absolute terms, Japan’s deterioration has been severe; from government take of 38.47 percent in 1992 to 46.04 percent in 2001, an increase of 0.95 percent per annum. A reining in of this misery machine by Koizumi, even if it is only by ½ percent of GDP, is long overdue, but may well be insufficient over the short term in restoring Japan’s economy. Over a 2-3 year period, however, if Koizumi maintains tight fiscal discipline, Japan’s absolute position will improve slowly, and her relative position is likely to improve substantially, as other countries battle with recession.
As for the United States; the source of its success in the 90’s is very clear. The government take declined substantially, from 34.80 percent of GDP in 1992 to 29.56 percent in 2001, and this has no doubt been a substantial contributor to the booming economy. However, the credit for this is due, not to President Clinton or Congress, since federal “take” actually increased over the period, but to fiscal discipline in the states and localities, primarily by the Republican Governors and state assemblies elected in 1994 and thereafter.
In the future, there are two possible paths. A recession is already upon us, and as in Japan in 1990-92 or the U.S. in 1929-31, will initially get substantially deeper, as the negative wealth effect from the asset bubble bursting hits. During that period, the government take will inevitably increase, and indeed is already doing so as state budgets swing into deficit.
If this trend is allowed to continue, with a Democrat congress in 2002 and President Hillary in 2004, then the future trend of the U.S. economy is very clear, from the Japanese example, as well as that of Sweden in the middle and late 80’s, the Netherlands in the early 80’s, Switzerland since 1990 and of course the U.S. in 1931-41. In all these cases the government take was allowed to increase substantially, thus deepening the trough and delaying and weakening the recovery from the recession.
If on the other hand, fiscal discipline is maintained (which in turn will require a modicum of electoral discipline from the U.S. people in not choosing false Socialist nostrums) then recovery from the recession can be relatively quick, as it was in Neville Chamberlain’s Britain, and Kim dae jung’s South Korea.
The choice is there, ladies and gentlemen.
-0-
(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)
This article originally appeared on United Press International.