Initial claims for unemployment insurance fell by 6,000 last week to 439,000. Since last week’s numbers were revised slightly higher from 442,000 to 445,000, one could also argue that the decline was really 3,000.

In either case it is a moderate decline, and one that is largely in line with expectations. As such, it should not have a major impact on the market.

The four-week moving average fell by 6,750 to 447,250. After peaking about a year ago, initial claims fell sharply for the remainder of 2009, but then had a bit of a retracement earlier this year, as is shown in the graph below. That retracement raised the very real possibility that initial claims were forming a high plateau, the way they did following the last two economic downturns, which reflected very long periods of jobless recovery.

As the horizontal dashed line shows, we are right back down to where we were at the end of the year in terms of the four-week moving average. We cannot quite yet reject the high plateau as the shape of things going forward, but it is looking less likely.

In any case, we are far below where we were at this time last year, when the four-week moving average was at 642,750. That sort of decline looks more like we saw coming out of recessions in the 1970’s and early 1980’s.

Continuing Claims

The news on continuing claims was much more mixed. Regular state-paid claims fell by 6,000 to 4.662 million, and for those claims we are well below the year-ago level of 5.776 million. However, those numbers do not tell the whole story — not by a long shot.

Regular claims run out after 26 weeks, and extremely long periods of unemployment have been one of the hallmarks of this recession relative to previous downturns. It is the thing that has made the Great Recession so “great” (though not in a good way). As of the February jobs report, half of all the unemployed had been out of work for 19.4 weeks, and the average duration of unemployment was 29.7 weeks, and over 40% of the unemployed had been out of work for more than 26 weeks.

After people reach the 26-week mark, they move over to extended unemployment claims, which are paid for by the federal government, as part of the stimulus package. Those extended claims rose by 264,000 to 6.031 million, and are almost triple the year-ago level of 2.188 million.

In other words, in total there are now 10.693 million people getting unemployment benefits, up 259,000 from last week and up 34.3% from 7.964 million a year ago. The duration of unemployment numbers will be one of the more important details of the jobs main course tomorrow.

Still, overall the report is encouraging, and if the appetizer for the jobs data meal was yesterday’s ADP (ADP) report, think of this as the salad course. Personally, I would have liked to have sent the appetizer back to the chef, as it was not very tasty, coming in at a decline of 23,000 rather than the increase of 40,000 that the consensus expectations had ordered.

The salad is just fine — nothing spectacular, but at least it is not wilted. The poor first course and the just OK second course should temper expectations that the main course is going to be spectacularly good.

Sure it’s going to have some good ingredients in it: fresh snapback from the February snow storms, plus lots of Census jobs. However, while those were not in the ADP report, there are a lot of ingredients that are shared between the two recipes.

The jobs data really is the key to the recovery. More people working will do wonders for budget deficits at the federal, state and local level. Not only will people go out and start to do more shopping at J.C. Penney’s (JCP) and Wal-Mart (WMT), but when they do they also pay sales tax, which will help alleviate the budget crunches at the state and local levels.

People who are working end up paying income taxes because they have some income, rather than receive federal benefits such as extended unemployment claims, or increasingly food stamps. Those extended benefits are a very useful part of the Stimulus effort; in fact, the CBO [Congressional Budget Office] scores them as among the highest impact on a dollar spent per job saved or created basis, far higher than any tax cut (although reductions in the employer side of the payroll tax for incremental jobs is also highly effective on a dollar-per-job basis).

That is in addition to the obvious humanitarian benefit of not leaving people with no income at all, and all of their savings depleted and their credit maxed out. However, extended benefits are not a great substitute for a real job. Even a Census job that only lasts six months or so is better than having millions of people just sitting around doing nothing. That sort of extended idleness is not good for them and it is not good for the country.

Dirk van Dijk, CFA is the Chief Equity Strategist for Zacks.com. With more than 25 years investment experience he has become a popular commentator appearing in the Wall Street Journal and on CNBC. Dirk is also the Editor in charge of the market beating Zacks Strategic Investor service.

More about Zacks Strategic Investor >>

Read the full analyst report on “JCP”
Read the full analyst report on “WMT”
Read the full analyst report on “ADP”
Zacks Investment Research