Daily Speculations

The Web Site of Victor Niederhoffer & Laurel Kenner

Dedicated to the scientific method, free markets, deflating ballyhoo, creating value, and laughter;  a forum for us to use our meager abilities to make the world of specinvestments a better place.

 

Home

Write to us at: (address is not clickable)

8/7/04
Victor Niederhoffer: Flagition and Friday

One of most regrettable things about looking over the shoulder of someone is that you never know what they're thinking after the initial reason for a trade is given. It is particularly regrettable in this regard that during the biggest weekly decline in the stock market over past 10 months, we suspended our predictions. We will continue that policy for the indefinite future. But for those who are interested, here is Vic's current thinking on what the numbers mean. And one can hypothesize that after such a week, and run of bad days for the stock market and good days for the bond stock differentia, Vic's is hobbling down to Wall Street with computer output in hand. How long will it be one wonders until the big allocaters between bonds and stocks note that there has been a divergence in relative returns, and % of their assets employed ? And cane in hand seeing the spin put on the latest employment release brings to mind that Gladstone and Disraeli, during the 1870's, liked to refer to each other as the most flagitious person in England. 80 years ahead of the now commonly accepted insights of public choice economics, that politicians act in their own selfish interests rather than ours, Albert Nock generalized the concept. He suggested that the best model for all politicians was to treat them as flagitious criminals, intent on manipulating all events, all announcements for their own personal benefit. The immediate political reaction to the employment number is classic in this regard. Kerry and Altman immediately said that the lower job growth meant the economy was "making a U-turn (to the lower)", or "obviously not creating jobs at a rate than even keeps up with the growth of population". Treasurer Snow however was "encouraged by the fact that unemployment rate came down". And Chief of Staff Card was encouraged by the much larger increase in employment of the household survey. I would go with Nock on this one and say that the key is which party has more operatives in the Labor Department that have the power to make assumptions on things like the number of jobs created by new businesses relative to seasonal adjustments. Further I would add the basic point that the labor theory of value has been discredited by all except market watchers. Productivity of jobs, how much output they create in combination with physical and human capital is the key to our prosperity not the number. Amid the flagition I thought it might be well to provide some order to the inordinate tendency of Fridays to mark major new highs and new lows. One notes for example that this Friday bonds and bunds closed at a 4 month high, stocks at a 7 month low (thus making the highly predictive Fed Model even more salient than usual), gold closed at 14 day high (above the round). And such markets as all the foreign stock markets, corn, copper, silver, natural gas, almost all the foreign stock markets closed at major new lows or highs as well. Is this a random phenomenon? To test this I wrote a few statements in Liberty Basic, a pleasant occupation for an old speculator, and Mr. Downing finished it up. Because of drift in markets, a closed form solution doesn't seem completely illuminating. The probability of new 5 day highs and lows in the stock market ( i.e. Friday higher than Thursday, Wednesday, Tuesday, or Monday ) classified by day of week is as follows:

Weekday % NEW HIGHS % NEW LOWS
MON 31 22
TUE 30 24
WED 26 23
THR 29 25
FRI 32 26

One notes that Friday extremes for the period 1996 to present,( about 425 days and 110 extremes for each day of the week) are most likely but that the divergences from change, at a first blush are not great.

 

8/9/04
Rudolf Hauser Comments

Vic writes,

"I would go with Nock on this one and say that the key is which party has more operatives in the labor department that have the power to make assumptions on thing like the number of jobs created by new businesses relative to seasonal adjustments"

What public choice theory tells us is that people act in their own interests rather than in the interests of the country, the company they work for or a political party. At times they may view the interests of the institution as in their own interest and at other times not. Some people have passionate political convictions so they may view the success of their favored party as in their interests. Most favor a party only to the extent they see it favoring them. Elected officials and political appointees have obvious reasons to see a political party's interests as their own. A lifetime career government bureaucrat does not. He wants more money and power for his part of government and perhaps a better job or position elsewhere outside government. Given that Democrats tend to favor bigger government than Republicans, bureaucrats are more likely to favor the former, particularly if they are in a department mainly concerned with Democratic favored programs. That connection is far less obvious in the BLS. Fudging statistics would not be good for their budget if the party not favored thereby comes to power. It would not be good for their reputations or career prospects if caught. That is more likely if there are those sympathetic to more than one party working with you, increasing the chance of getting caught. It seems far more likely that someone may have a biased view on a statistical approach to a problem and that might cause them to manipulate the numbers in one way or the other. Or based on other data, they feel they have a view what the data should look like and then try to find ways that might justify moving the data in that direction. The main pressure might come if someone up the chain of command had political interests and tried pressuring those below to move the statistics in a particular way. But our system it might be difficult to just say I want lower employment gain numbers. He or she would more likely have to pressure them on a statistical issue. But the top people are political appointees - GOP appointees in this case, and the report did not overall favor the Republicans. My point is that the public choice view of this issue is not as simple as asking how the bureaucrats voted.