Initial claims for unemployment benefits fell to 404,000 last week, better than forecast, but still about 25K more than what is needed for the start of a meaningful recovery, and the less volatile 4-week moving average dropped 4,000 to 411,750, with continuing claims dropping 26K to 3.86 million, though the total number of people getting benefits, despite the fact that the “99ers” have run out of benefits, rose by more than 400K to 9.61 million.

In related news, manufacturing employment grew for the first time in more than a decade.

Employment in the US rose by just 103,000, and private payrolls rose by just 113,000, with unemployment falling to 9.4%, though over ⅔ of that is because of people becoming discouraged and leaving the workforce.

Initial unemployment claims rose by 18,000 to 409,000 last week, though the 4-week moving average fell by 3,500 to 410,750, with continuing claims fell by 47K to 4.103 million, and I can’t seem to find the data on extended claims.

In related, and rather surprising news, payroll processor ADP is predicting that private employment increased by 297,000 last month.(!)

If true, this is good news, though I’ll wait for the official data tomorrow.

Initial unemployment claims fell below 400,000 for the first time in over 2 years, hitting 388,000, down 34K, and the less volatile 4 week moving average fell by 12.5K to 414,000.

We are actually at a number that if sustained, might show a meager recovery in the unemployment rate.

The longer term numbers, continuing claims rose by 57K to 4.13 million, and extended benefits fell by 151.5K to 4.53 million.

I would expect the numbers to rise next week, the Christmas season is over, and this sort of move one week tends to have a rebound, but if somewhere around 375K is the new normal, then things are turning up a big.

We had 420,000 initial claims last week, a small drop, but the less volatile 4-week moving average rose for the first time in 6 weeks, from 423,500 to 426,000, though both continuing and extended claims fell.

Initial claims, and unemployment, is too damn high, but it continues to trend a bit better.

Well, it turns out that the ADP estimate was wrong, and the job creation numbers were absolutely pathetic, there were only 39,000 jobs created in November, somewhere around 100K less than is needed to account for growth in the labor market, the unemployment rate went up 0.2% to 9.8%.

And in case you are wondering, that increase in unemployment is not an artifact of the discouraged looking for work, because there was no increase in the labor participation rate, which means that we just had a really bad month.

One month does not a double dip make, but when you realize that this is juxtaposed with the stimulus money tapering down, it is an issue of concern.

Then again, my anecdotal data, job shops calling me, is pretty strong, so YMMV.

We get the official numbers on Friday, but ADP’s private employment survey shows a 93,000 increase in private employment in November, which is the best number in about 3 years, though this alone is still a bit smaller than the natural growth in the labor market.

Also on the plus side, we have the Federal Reserve Board’s  Beige Book showing mild expansion, construction spending rising in October, and car sales for GM, Ford, and Chrysler up significantly in November.

On the so level, we have the Institute for Supply Management’s manufacturing index falling, but still showing slow growth.

On the minus side, we saw mortgage applications falling sharply last week.

Admittedly, it is only one week, but this week’s jobless claims numbers are very good, 407,000 initial claims, the lowest in almost 2½ years, the 4 week moving average fell to 436,000, continuing claims fell by 142K to 4.18 million, and emergency claims fell by 262,000 to 4.66 million.

This is getting close to the level where we can actually start seeing some real recovery in the job market.

I still think that we will see a double dip as what remains of the stimulus runs out, but I am less confident of that than I was, for example, a few months agol.