Initial claims for jobless benefits drop sharply (46,000 in 2 weeks)

[denem]

Source: Associated Press

WASHINGTON — New applications for unemployment benefits fell sharply last week as General Motors and other manufacturers skipped their usual summer shutdowns.

The Labor Department said Thursday that new claims dropped by 29,000 to 429,000, the lowest level since August 2008.

It was the second straight week that initial claims dropped sharply and the third drop in the last four weeks. Claims fell by 17,000 in the previous week.

Separately, the Commerce Department said that wholesale prices fell for a third consecutive month, pulled down by another drop in energy costs and the biggest plunge in food costs in eight years. But excluding those two volatile commodities, inflation was relatively flat.

Read more: http://www.google.com/hostednews/ap/article/ALeqM5gNiyJ905Ho0Ur96V2TQhsBX19lGwD9GVG9O00

---

1. These are NEW claims 2."lowest level (of New claims) since August 2008", but not in percentage terms.

[DCBob]

1. So when does the "depression" start?

since we are supposedly in the midst of one.. so says Krugman.

[denem]

2. Er ... (1900s) -Panic, (1920s) -Drepression (1950s) -Recession

(1990-2000 Japan) -Lost Decade. Pick and choose the words. Two economists in a room give three opinions.

[The Northerner]

3. +1

[impik]

4. Terrible news. Sad day here.

[denem]

6. Yeah. Unrecc'd within minutes.

[Lost4words]

25. You do realize DC almost never feels the full crunch or recessions.

ups and downs of the country. I lived there for 27 years so I am speaking in terms of experience. T Inside the Beltway is a unique area compared to the rest of the country. Its the seat of the federal government which isn't small and all those small business that support the federal system.

[Dawgs]

5. So, is this good news or bad?

I just don't know anymore.

[denem]

7. Economists = Good ... DU = Bad/Lies/Meaningless/U6=Depression

[Dawgs]

8. That's what I thought. Thanks!!

:hi:

[Statistical]

9. Lol. Funny and sadly accurate summary.

[caseymoz]

21. It's good in the way a cool summer week somewhere is good news on global warming.

The question is, how good is it so far, really?

We have enough of these weeks and it might finally mean something. And we must have a lot of them substantial considering how many jobs we lost and how many people have given up looking.

[earthside]

10. "Seasonally adjusted."

Frankly, these weekly new unemployment claims numbers are fairly meaningless except when taken as a very broad indication of a trend.

Since they are "seasonally adjusted", meaning that they are manipulated statistically, one never knows just how accurate they are and what unique factors might be skewing them up or down.

I heard an analysis on a radio news report this morning that since auto factories kept working this month, instead of taking their usual summer shut-down, this week's claims numbers were actually not good once the statistical "adjustment" was taken into account.

I think the trend tells us that we've entered a period of economic stagnation.

[dmallind]

12. Quick question - when the seasonally adjusted numbers are higher than actual

Do you take as much effort to point out that it's all statistical smoke and mirrors and things are really better than reported?

[Statistical]

24. Seasonal adjusted is a very similar concept

They take last 20 years. Take average for the 20 years and the average for each month.

Say Average for all 20 years is 400,00.
Say average for January is 350,000
Say average for June is 420,000

50,000 gets subtracted from any Jan #.
20,000 gets added to any Jun #.

Without seasonally adjusting the initial claim numbers would be meaningless.

They would almost always rise in January and fall in Nov. It would tell you nothing at all as to how this year is different (good or bad).

[Po_d Mainiac]

26. Douch Bank had this to say

Temporarily laid-off workers file for unemployment benefits during this period, thereby causing a spike in applications. Seasonal adjustment factors attempt to account for this, but since the schedules change every year, the claims data are prone to large week-to-week volatility. GM recently reported that they were not shutting as many assembly plants as usual due to demand for various models, so as a result the “usual” seasonal factors may be over-compensating and driving initial jobless claims artificially lower.

Read more: http://www.businessinsider.com/deutsche-bank-dont-get-too-excited-about-todays-positive-unemployment-news-2010-7#ixzz0to7xoAaz

[mahatmakanejeeves]

11. From the Employment and Training Administration website

Unemployment Insurance Weekly Claims Report

[GOTV]

13. Has anyone done an updated "bikini" graph?

The last I saw ended in January. It would give us an idea of the trend.

[dmallind]

14. That uses the monthly numbers that we won't get for 3 weeks

But yes it would be good to see the updates. You'd have the census shift to explain, but it would show improvement either way.

[GOTV]

16. I just posted one which at least goes through June but ...

... I don't understand your caveat about the census shift. Could you explain that?

[GOTV]

17. Oh wait, I think I figured it out ...

... you're talking about the census workers whose work is done and are rejoining the ranks of the unemployed. Correct?

[dmallind]

18. Sure - we gained 431000 jobs in May, and lost 125000 in June right?

But the majority of both were Census jobs with known timelines. So it's not reeally indicative of the underlying and painfully slow recovery to use either - one being too optimistic and one being too pessimistic. IIRC private employers added about 125K and about 83K respectively. Now of course the Census jobs really happened so we actually got the first set of numbers, but it would skew the trend heavily.

[GOTV]

19. OK, but by November that will be one blip and the overall trend should appear.

The bikini graph has several outlier data points but still tells a pretty good story for Obama. By November we'll be able to see whether we were teetering on the edge of a second recession, or did the recovery stall or continue to pick up steam.

There were extenuating circumstances for June but if July's numbers are down, and then August, and then September we can expect a bloodfest in November

[dmallind]

20. Agreed - but little chance of losses in July I think. Aug/Sep who knows?

[GOTV]

22. Ok Here's new one taking into account your numbers

The light blue shows the official numbers with the dark blue showing the estimated private sector numbers.



It seems to suggest some possible backsliding on the recovery

[dmallind]

23. Even though it looks "worse" in some ways I like that one better yes

BTW not my numbers - BLS numbers. Don't want to claim credit for work they did.

Yes it is undeniable this is a soft slow recovery, but at least we avoid the Census-driven contraction which is especially problematic for perceiving underlying trends. I think April was Census-driven in some part too BTW but let's not get too pedantic. Essentially for last few months pricvate sector has grown at a pretty anemic and variable but still positive rate.

And while that's no cause for cartwheels, I think that's the resposnible and real message to communicate. jobs are growing, but slowly and at a very fragile pace. That's why we need to keep spending stimulus monies and not worry too much about deficits (and I say that as somewhat of a deficit hawk in normal times - but these are not normal times!). That's why we need to keep extending UE benefits to tide over the long term jobless until the pace improves. That's why we can at the same time point to success - job growth is real and has been for a while - as well as poibnt to the need for more action and help - it's not fast enough and it needs to be nurtured very closely.

[GOTV]

15. found one