In a letter to clients which quickly went public Thursday afternoon in N.Y., former investment bank Goldman Sachs (GS) warned that September’s NFP (scheduled for release at 08:30 EDT) would show a loss of 250,000 jobs, worse than the 200,000 it previously estimated. The S&P fell 0.92% in the last hour of trading as the dollar gained against the PEAK currencies (pound, euro, aussie and kiwi).

GS chief U.S. economist Jan Hatzius cited the Monster index of on-line hiring, the ISM employment index, consumers’ assessments of job availability, and the total number of individuals receiving continuing claims for unemployment insurance (including those for extended benefits) as reasons for upping their earlier assessment. Economists had been estimating the loss of jobs to be 175,000.

As far as the unemployment rate is concerned however, the bank maintained its 0.1 percentage point increase.

This sets up an especially interesting situation for traders, because it’s possible that the “news” of a higher loss of jobs has already been priced in. That could mean that a surprise to the upside (meaning less jobs lost than thought) is likely to be followed by the rapid ascent of stocks (and commodities) in conjunction with the PEAK currencies against the USD.

That being said, the NFP is a notoriously difficult report to estimate, primarily because the Bureau of Labor Statistics (BLS) uses something called the “birth death ratio” in its calculation of monthly job creation (or loss as the case will probably be). This ratio has nothing to do with the births and deaths of persons; it’s basically a guesstimate of how many companies (and their associated jobs) actually opened up shop or closed their doors over the previous month.

What’s also likely to hurt the September jobs number (to say nothing about October) is that basically all of the car manufacturers said sales fell in September as the “cash for clunkers” program ended. Companies reported on Thursday that September deliveries were reduced by 23% from August to the second slowest rate this year.

In any event, it will take years for the unemployment rate to return to anything that resembles normal (say 5%). For example, the economy would need to create about 15 million jobs over the next 5 years in order to get there, an average of 250,000 per month, but the problem with that becomes evident when looking at average monthly job creation numbers:

The best year since 1989 was 2006, with an average monthly growth of 232,000. If you average the ten years from 1999, you get monthly job growth of 50,000. If you take the average job growth from 1989 until now, you get an average of 91,000 a month. In fact, the best ten year period (1991-2000) yielded an average of only 150,000 a month.

With that, what are the chances of seeing a double dip recession at some point along the way? Back in 1937 (when the economy was in the process of recovery from the great Depression), Congress and President Roosevelt decided the thing to do was to balance the budget, which got terribly out of whack as the government pumped money into the system just as it’s doing now. So, they raised taxes and created another recession which basically lasted until war production ramped up in the early 1940’s.

Now, the Obama administration is intent on making the economically suicidal move to raise the top tax rate by 10% as it lets the Bush era tax cuts expire. Popular myth is that those who pay the highest tax rate are Wall Street fat cats but the reality is that 75% of them are the small business owners, the ones who are responsible for the large majority of new jobs that are actually created in the country. Taxing them will lead to fewer jobs being created (as they will have less to invest) and of course, less consumer spending.