In the face of lies, damn lies and statistics: What will the BOJ do?

February 16, 2007
By Ken Worsley


In last week’s edition of its venerable newspaper, The Economist made the argument that the Bank of Japan could be on the verge of raising interest rates, given strong expectation for GDP growth (which turned out to be true) and pressure from the G7 (which was weaker than expected, especially from the influential American side).

The Economist barely made mention of Japan’s domestic fundamentals in its piece, aside from GDP growth. We know that prices have barely risen, household spending is flat, and wages have barely grown (0.1% or 0.2% last year, depending on the source). If interest rates are indeed a tool intended to keep inflation in check, why raise them now, when there is no apparent inflationary pressure in Japan?

The Economist argues that the widespread practice of carry trading (estimated at a value of $1 trillion) is one of the gravest threats to the world economy, as it nearly did massive damage in 1998, and is thought to be much larger now. One of the troubles, of course, is that much of the carry trade (which involves assuming debt in a currency with a low interest rate - such as Japan’s, and then buying assets in another currency with a higher rate of returns) is done by hedge funds, uses complex derivatives, and is thus off the books.

Well enough. I was cynical (to say the least) about Japan’s 4.8% reported quarterly GDP growth when it was announced. However, the New York Times has now come out and said that the Japanese Cabinet Office’s statistics on both GDP and household spending might be misleading, and that no hike in interest rates is justified. Takahide Kiuchi, a senior economist at Nomura Securities in Tokyo, weighed in with this quote:

Consumer spending remains stagnant. These growth figures don’t change the economy’s overall outlook. There’s nothing here to justify a rate hike.

True enough. Other economists have said that the government’s claimed 1.1% raise in household spending is due to faulty data and misleading survey techniques. You don’t say…

Now things get interesting. Is the data being set up to allow the BOJ to raise rates? Or will political pressure remain on the BOJ to hold them steady?

And, since when is there an empirically verifiable connection between interest rates and foreign exchange rates?

Comments

2 Responses to “In the face of lies, damn lies and statistics: What will the BOJ do?”

  1. Adamu on February 16th, 2007 10:57 am

    It is shocking (but not suprising) how deceptively the GDP stats were presented. I mean, look at this chart from — it seems like they don’t need to be legally ordered to trumpet what the government wants them to. I mean, they didn’t even calculate the “annualized” rate correctly — it would be 4.9% if they assumed the same growth rate each quarter. Not only that, it’s quarter on quarter growth, not growth as compared with the same quarter last year.

    Richard Jerram, who was quoted in the NYT article, is apparently a perennial doubter of Japan’s economic statistics. Here he is in an interview:

    I think poor quality data is a reflection of the nature of the bureaucratic system. Each ministry collects data and produces related releases. Even when a release is erratic, there rarely seems to be pressure for reform. Setting up a statistical office to manage the quality and process would seem sensible, but would offend too many interests.

    The greatest problem is in measurement of consumer spending, where all of the official releases seem seriously flawed. I have constructed my own measures of consumer activity from other sources that seem more reliable, but it is always tough to convince people when the headline data releases are showing random behavior.

    In general I don’t think the data is manipulated for political gain, which is an accusation that you sometimes see. And to be fair, the Cabinet Office has been trying to improve the quality of the releases, especially GDP.

    If the data itself aren’t manipulated for explicit political purposes, the presentation of that data certainly is.

  2. Ken on February 16th, 2007 2:21 pm

    If the data itself aren’t manipulated for explicit political purposes, the presentation of that data certainly is.

    I agree 100%, but that’s been everywhere and since the dawn of time. We keep talking about that with, for example, data on foreign crime in other blogs. I’d be willing to bet that those cave paintings in France were somehow meant to be politically manipulative.

    Jerram is very correct about different ministries each posting their own statistics and them often being wildly different:

    Setting up a statistical office to manage the quality and process would seem sensible, but would offend too many interests.

    Absolutely. As would appointing ’special advisers’ to help with administrative reform. Wonder how that ‘National Security Council’ is working out…

Got something to say?