Tuesday, May 27, 2008

Expert's Expectations Expected

Oil settles below $129

If the market follows its typical seasonal pattern, Schork thinks crude's "highs have been put in."

I have two holes to blow in his theory. Here comes the first hole. I put 32 years (starting in 1976, as far back as I could go) of monthly gasoline data through the Census seasonal adjustment program. Want to guess what the output said?

WARNING: Series should not be a candidate for seasonal adjustment because the spectrum of the prior adjusted series (Table B1) has no visually significant seasonal peaks.

I'm not even going to supply a chart this time. There's nothing to see. Instead, I offer the seasons when gasoline prices peaked for the year so that you can see the lack of things to see for yourself (sarcasm and humor intended). For the following data, I am starting spring in March, summer in June, fall in September, and winter in December. For simplicity, I am using the calendar year to determine it (i.e., January, February and December of 2007 are considered 2007's winter).

Summer: 1977, 1980, 1983, 1985, 1992, 1993, 1994, 1995, 2000, 2004, 2006
Fall: 1976, 1987, 1988, 1990, 1997, 2002, 2005
Winter: 1978, 1979, 1982, 1986, 1991, 1998, 1999
Spring: 1981, 1984, 1989, 1996, 2001, 2003, 2007

Summer: 34%
Fall: 22%
Winter: 22%
Spring: 22%

Here comes the second hole in the expert's logic. It is still spring (March, April, May). Even if you do somehow manage to buy the argument that this data has some seasonally significant trends (which is very much open for debate at best since the program's output says otherwise), then we're not even to the seasonally statistically important part yet (summer). So why would we therefore assume prices have already peaked based on seasonal concerns?

That's not to say that prices aren't peaking. Who knows? I'm simply arguing that the expert's logic has two seasonally adjusted holes blown right through it. Lies, damn lies, and statistics strikes again. Is it any wonder that I find it so hard to trust any financial advice these days? Everyone on Wall Street has their own agenda and very little of it has anything to do with protecting us. That's about the only certainty there is these days.

Am I done yet? Nope. As I have stated in recent posts, I have a tremendous fear of under heckling. I rank it right up there with my fear of borrowing money to sustain the unsustainable.


As consumer confidence wanes and prices at the pump continue to climb, demand for gasoline has softened. Tuesday's report provides further "anecdotal evidence that demand is drying up," Schork said.

What's the anecdotal evidence ("Compare with hasty generalization") and/or hasty generalization ("basing a broad conclusion upon the statistics of a survey of a small group that fails to sufficiently represent the whole population") coming out of China? Are we still sending billions of people our money through a massive trade deficit so that they can learn to drive? Or have they finally figured out that our paper IOUs just don't buy as much as they once did? Here's some scientific evidence, for what that's worth these days. It states that China's increasing its refining capacity by about one-third in the next few years.

China's Energy Mix in 2008

In 2008, China will bring four new refineries into operation: China National Petroleum Corp. (Sinopec) in Qingdao, Shandong, and Quanzhou, Fujian; China National Offshore Oil Corp. (CNOOC) in Huizhou, Guangdong; and PetroChina Co., Ltd. in Dushanzi, Xinjiang. Overall, China's refineries will supply an additional 400,000 bpd of fuel, more than double last year's increase, according to a Reuters survey. China plans to expand its refining capacity by about one- third by 2010.

Source Data:
EIA: Unleaded Regular Motor Gasoline Retail Prices, U.S. City Average
The X-12-ARIMA Seasonal Adjustment Program

1 comment:

Stagflationary Mark said...

If the market follows its typical seasonal pattern, Schork thinks crude's "highs have been put in."

One more thought. Here are the months of the recent years when the price of crude oil peaked for the year.

WTI - Cushing, Oklahoma
http://tonto.eia.doe.gov/dnav/pet/pet_pri_spt_s1_m.htm

2007: November ($94.77)
2006: July ($74.41)
2005: September ($65.59)
2004: October ($53.28)
2003: February ($35.83)
2002: September ($29.66)
2001: February ($29.61)
2000: November ($34.42)

Can you find the typical seasonal pattern Schork seems to see in May? I sure can't.