Lately, international investors have become enthusiastic about Japan's economy and stock market.

Several American investment banks have trumpeted that the turnaround of Japan's economy has arrived. Foreign investors have been pouring money into Japanese stocks, boosting the yen in the process. Officially, Japan's real GDP grew in the second quarter by a sensational 3.9%, beating America's growth rate of 3.3%.

Our knowledge about Japan's economy comes exclusively from the monthly reports of its central bank, the Bank of Japan. Its September report stated:

"Economic activity still continues to be virtually flat as a whole, although signs of improvement have been observed in such areas as the environment for exports. With regard to final demand, business fixed investment is recovering gradually."

"Meanwhile, private consumption continues to be weak, housing investment remains sluggish, and public investment is declining. Net exports are virtually flat. Industrial production continues to be basically level in response to these developments in final demand, and corporate profits are on a moderate uptrend."

This hardly reads like the description of an economic recovery. But how to reconcile this dismal description of the economic situation with the officially reported stellar real GDP growth rate of 3.9% for the second quarter? Putting it bluntly: It is exactly the same statistical hoax as the 3.3% simultaneously reported for the U.S. economy. Japan's statisticians have learned from their American colleagues how to conjure up the perception of an economic recovery that does not exist.

Yet in Japan's case, there were some critical comments in the press. A report in the Financial Times quoted an economist as saying that "the government deliberately manipulated statistics" using in particular an "incorrect measure of deflation."

This, in turn, provoked a reply from the head of the Japan's Department of National Accounts, a Cabinet Office. In his letter to the FT, the director explained that the "basic price statistics used in Japan's GDP deflators, such as the consumer price index and corporate goods price index, adopt the hedonic method to track price changes accompanying quality improvement in it goods" - stressing that this is common international practice.

To be precise: Just like the practice of annualizing quarterly numbers, this is an American, not an international, practice. Without the annualization, reported real GDP growth would have been 1% in the quarter, and 3% year-over-year. In today's world of sluggish growth, that would also be impressive, if its main source was not hedonic pricing of computers. Under its impact, nonresidential investment grew 4.7%, or a stunning 20.2% at annual rate.

Measured in current prices, a radically different picture emerges: Japan's nominal GDP grew during the quarter by a dismal 0.3%, or 1.2% at annual rate. Year-over-year, it was up 0.5%. Repeating the first sentence of the above citation from the Bank of Japan: "Economic activity still continues to be virtually flat as a whole."

Assessing the Japanese economy's performance and prospects, there is a broad choice. If you want to see an economy and a stock market powering ahead, focus on the 3.9%, as measured in real terms and annualized; if you want to see an economy that remains stuck in the post-bubble aftermath, focus on the 0.3%, as measured in nominal terms and without annualization.

Such a vast difference in measured economic activity is, of course, laughable. One of them must be completely out of whack with economic reality. For us, there is no question which one - the 3.9%. It has two main statistical sources that we regard as outright phony: first, annualized quarterly figures; and second, a very low GDP deflator.

Take these two statistical gimmicks away, and you end up with the earlier mentioned mini-growth rate of 0.3% in nominal terms for the second quarter. From the above citations of the Bank of Japan, we conclude that it, too, favors this measure.

But which of the two is the better measurement of economic activity? "Real" figures attempt to measure the physical volume of goods produced and sold in the economy by adjusting nominal figures with calculated inflation rates. Falling prices essentially increase real GDP growth, and rising prices decrease it. By contrast, the nominal figures simply add up the amount of money that has been spent on various demand components.

We have to say that we are principally opposed to translating falling prices into rising real GDP growth. It corresponds, of course, with the opposite practice to deduct inflation rates from nominal GDP growth. Mechanically, both seem equally logical, but economically, this equal treatment makes no sense. Consider that the steeper the fall of prices, the higher the economy's real GDP growth rate.

For us, the decisive consideration is that the real figures tell us nothing about the movements of incomes and profits. These only show in nominal figures, based on money transactions. This may have been less important in times of high inflation, but in our time of protracted economic sluggishness, the money measure is far more important than the volume measure.

Regards,