[kenrexford]

OK, maybe someone can explain to me why I'm being tricked by anti-Obama folks on this one.

I just reviewed an article on line where it is claimed that the economy is improving as far as jobs. Here was the logic.

We thought we would lose 580,000 more jobs.

However, we only had 550,000 new jobless claims.

But, we had 70,000 more in the already-claimed category, apparently because people expected to find jobs and get off of unemployment failed to do so.

Now, to my strange way of thinking, if we thought we would have 580,000 more people unemployed than earlier, but we actually have 620,000 more people unemployed, then things are worse than we even expected. Am I missing something here?

But, then I get even more confused, as the whole "we think" number seems rather arbitrary. I could say that I thought we would lose 1,000,000 jobs and thereby claim that we just picked up 380,000 jobs, a huge success. Or, I could say that I thought we'd stay even and claim a 620,000 loss as a bad thing.

Beyond all of that, though, how is 620,000 more people unemployed a good thing? It seems to me like a bad economic problem.

I also wonder why we don't have job numbers like the DOW runs. If at 10,000 it drops 1,000 points, that's a 10% drop. If the DOW is at 5000 and drops 1000, that's a 20% drop. So, if we have 5% unemployment and go to 6% unemployment, then we have lost 1 out of every 95 jobs. If we are at 8% unemployment and go to 9% unemployment, we have lost 1 out of every 92 jobs.

That might not seem like that much of a difference, but this gets even more complicated if you think along the terms of assured jobs. If government accounts for, say, 20% of the job force (low, I know), then the "flexible" market seems to be only 75 or 72.

Thinking along this further... Suppose that any economy, no matter how bad, will have at least 80% employed. Maybe the other 20% reflects real possible movement. Well, then a move from 5 to 6 unemployment is losing 1 out of every existing 15 of these possible-movement jobs. A move from 9 to 10 reflects losing 1 out of every existing 11 of these flex jobs.

In other words, if the government tinkering can only be expected to really affect 20% of the market, then a change from 9.4% to 9.6% unemployment might be worse than a change from 4% to 6% unemployment, in theory. This is because both the percentage drop of playable rooks is larger on the high end and because perhaps that dip in the higher end has more net impact on the entire economy, by reducing incomes, reducing taxes, etc.

When I think all of this through, I keep ending up with the conclusion that this actual headline is utter hogwash spin.

[G_R__E_G]

You're not stupid.

[TimG]

I did not read through your whole post, but suspect this is a case of being able to support just about anything with a barrage of percentages.

I caught a snippet of a discussion with Barbara Boxer on NPR recently. She said that insurance companies have enjoyed a 400+% increase in profits in the last 6-7 years and thought that this was atrocious.

I'm on pace to win 300% more masterpoints in 2009 than I won in 2008. Sounds impressive until I admit to having only won 15 masterpoints through the first 7 months of 2008.

[matmat]

Enjoy.

http://www.youtube.c...h?v=2yhN1IDLQjo

[awm]

Unemployment numbers are notoriously difficult to measure. What does it even mean to be unemployed? Obviously, that you don't have a job, but it is much more complicated when you consider that there are people who aren't really looking for work, people who are working part time and would like to work full time, people who are retiring, etc.

It's pretty easy to spin these numbers. Also, employment numbers are very much a lagging indicator since it normally takes some time for businesses to lay people off when the economy is lousy and also takes some time for them to increase hiring when the economy is recovering. We're trying to figure out to what degree (if at all) the government stimulus is helping the economy, but this is basically impossible because we don't know how the economy "would've been" if no stimulus had occurred. So you see a lot of the numbers comparing how things are to how they were "predicted to be" which don't really measure a whole lot. Of course, all of this applies both to the Obama administration's numbers and the Republican opposition's numbers (basically they are both spin and have little relevance).

[helene_t]

awm, on Aug 6 2009, 07:17 PM, said:

Unemployment numbers are notoriously difficult to measure. What does it even mean to be unemployed?

I used to work for Statistics Netherlands (a government office producing official statistics).

What surprised me was that almost all their statistics were based on interviews rather than registrations from other government offices. For example, unemployment statistics were made by calling random people and asking them if they considered themselves unemployed. This is different from Denmark, where the numbers were based on the number of people registered as job seeking with the municipal employment agencies. (This was back in 99, the methods may have changed since then).

Obviously there are advantages to both approaches.

[jjbrr]

[ ] stupid
[x] not stupid

[matmat]

It bothers me how certain entities try to exploit others' phobia and poor knowledge of math.

Recently a Fresh and Easy (i.e. tesco US) opened up near me. Now, maybe it's just my imagination, but it seems to me that they package their foods in unusual weights (7, 9, 11oz, etc), making an on the fly comparison with other stores hard, unless you can do math in your head or carry a calculator (and know how to use it!).

[Al_U_Card]

"Lies, damn lies and statistics." Mark Twain

[PassedOut]

awm, on Aug 6 2009, 01:17 PM, said:

Also, employment numbers are very much a lagging indicator since it normally takes some time for businesses to lay people off when the economy is lousy and also takes some time for them to increase hiring when the economy is recovering.

Yes. During the Reagan recession of 1982-83, unemployment remained high long after the other indicators turned positive. The Bush recession is even worse, so the best we can expect now is a slowing rate of job loss, sad to say.

[TimG]

matmat, on Aug 6 2009, 02:41 PM, said:

It bothers me how certain entities try to exploit others' phobia and poor knowledge of math.

Recently a Fresh and Easy (i.e. tesco US) opened up near me. Now, maybe it's just my imagination, but it seems to me that they package their foods in unusual weights (7, 9, 11oz, etc), making an on the fly comparison with other stores hard, unless you can do math in your head or carry a calculator (and know how to use it!).

In Maine, supermarkets are required to display a per unit price along with the item price. So, all cereals have a price per ounce listed along side the total price.

It's not perfect, for instance the same units are not always used, but it does generally make things easier.

[y66]

Quote

how is 620,000 more people unemployed a good thing?

It's not good. The only thing that's good is that unemployment is not getting worse as fast as it was earlier in the year, as passedout says.

[kenrexford]

y66, on Aug 6 2009, 09:04 PM, said:

Quote

how is 620,000 more people unemployed a good thing?

It's not good. The only thing that's good is that unemployment is not getting worse as fast as it was earlier in the year, as passedout says.

Yeah, but who would expect that even if things were really bad?

The average company might have, say, 50 employees as a core group, but flexes between 50 and 100 depending on how the market is. Losing 50 from 100 is easy to expect, but losing one more, down to 49, means the company is about to shut it's doors. So, if we lose 30 in one month, then 20, then 10, it looks like we are doing somehow better by this logic. However, we have lost 60% of the flexible jobs, then 66%, then 100%.

Or, how about fishing with a net? You cast the next the first time and get a lot of fish. Each additional cast brings in fewer because there are fewer fish to catch and they are more scattered. Do the fish think that the carnage is getting better?

[barmar]

Unemployment numbers can go up without anyone losing their jobs. When someone graduates from college and doesn't have a job lined up, they increase the unemployment numbers. Conversely, when someone retires, the number of employed people goes down, but the number of unemployed doesn't go up (or vice versa if a replacement is hired immediately).

When the US government reports unemployment rates, they also report "new jobless claims". This is the number of people who were laid off that month. This is more accurate than trying to estimate it from the change in the unemployment rate.

[Mbodell]

There are also at least 6 different official unemployment rates released from the BLS each month (and two different surveys: Household survery versus payroll survey). U-3 is the commonly reported rate, although a bunch of folks think U-6 more closely matches reality/how we did things in 30s, WWII, 70s, etc.

The idea that we lost many hundreds of thousands of jobs last month and it is a good thing is bizarre that is really more the expectation game than reality. Similarly, U-3 went down very slightly last month from like 9.5 to 9.4 but it wasn't really good news as it was largely a result of more people becoming discouraged and no longer counting in the U-3 measure.

There was a good piece on the PBS newshour by their economics correspondent 2 years back about unemployment talking about how discouraged workers, those in jail or on probation or with criminal records, plus those disabled, plus those working part time jobs could make an official U-3 of 6% hide numbers that were really more like 20% of the population when you count people more generously.