The key question for the upcoming election is, will voters vote based on their own experience - most folks have a job, and most folks they know have a job - or will they vote based on the media-spread misperception based on preliminary and inaccurate data.

If unemployment were running at, say, 35%--most folks would have a job and most folks they know would also have a job.

The point is Mr. Hobbs is setting the bar extremely low.

The real question ought to be an examination of Bush's policies, the stated objectives of those policies, and the measured results.

The measured results have not been good.

The stated objectives of Bush's policies have consistently been revised downward by this appointed administration--and they've failed to meet those lowered objectives WRT jobs.

It's gotten to the point where the WH is now refusing to endorse their own projections WRT job growth. In early February of this year, the WH forecast there would be 2.6M new jobs created in 2004. This means, of course, employment would have to grow at an impressive rate of some 320,000 jobs each month.

Well, February's numbers weren't too hot: a meager 21,000 new jobs (almost all public sector).

What's more, the unemployment rate is always underreported; it only counts those who are actively seeking work and doesn't account for the tens (or hundreds) of thousands who've abandoned actively seeking work.

How many people have "settled" for jobs with lower pay, benefits, or career potential?

Another interesting metric is long term unemplyment; that is, people who spend six months or more unemployed. In 2002, the long term unemployed was 18.3%. In 2003, it was 21.8%. Compare that to 2000's long term unemployed rate of 11.0%.

Jadegold,

The BLS actually measures people who want a job and have given up looking (and those who are working part-time but want to work full-time). See here, U-6. I believe that marginally attached includes the long term unemployed, though I'm not sure. Do you have a link to that data?

I think it is misleading to compare current conditions to those at the peak in 2000. To gauge policy efficacy, why not compare the current economy to the economy at this point in Clinton's first term? In the data I linked to above, the total unemployment (official+marginally attached+unwilling part time) was 10% in Feb 1996 and in Feb 2004 it is 9.6% (seasonally adjusted, both trending down). So it is better now. (Granted it may not continue the trend and we will have to see). Poverty is down (2002 vs 1996, they haven't released 2003 data yet that I can find) and median wage is up (2002 vs 1996, same thing). What statistics do you want to use? I will grant you that the rate of job growth is weak, but it appears to be improving somewhat similarly to the end of the last recession (though it is taking longer). I think that basing things on their projections is silly, by the way.

Regardless of the measures you select, an important point is that Clinton came in with a recovering economy, an increasing stock market, and an improving job market. Bush came in with a peaking economy about to fall and a falling stock market; what Noam Scheiber of TNR calls a "stacked deck" in reference to job numbers (slight paraphrase). The fact that the economy is in about the same shape as in 1996 (at least in a static sense) seems to me to show that Bush policies may at least deserve some credit for dealing with significant problems or at least not so much blame (I think credit and blame for presidents is always exagerrated).

Now, a valid criticism is the deficit. The deficit in 1996 was only about 1.4% of GDP and this year I believe it will be 4.5% or more. This affects the long term health of the economy, but does not affect whether or not the current economy is "good" by other measures. Whether or not it is "worth it" seems to only be answerable in retrospect.

But besides the deficit (which I do think is important) what measures are worse now than in 1996? I don't know of any static measures that are worse off, although, comparing rates, jobs growth is lower than in 1996 (it is more comparable to the late 1991 early 1992 period).

"If unemployment were running at, say, 35%--most folks would have a job and most folks they know would also have a job."

Ummm...

(Checks local reality. Finds no congruence with JadeGold's reality.)

Okay....

J.

Everybody knows OPEC controlls the oil prices. OPEC consists out of America's biggest ennemies Arab tyrants and Muslim fundamentalists. Bush told them to stop oppressing their populations, because this was contributing to terrorism or else he would take care of them the way he took care of Saddam. OPEC wants Bush out. That's why they limit the oil output, increasing the oil price and making sure the economy here doesn't grow and the unemployment stays high. They hope this will lose Bush the election. If Bush loses, America's ennemies will win. It's as simple as that.

Mr. Messina:

The Labor Department tracks employment or job growth along two tracks: payroll data and household surveys. Your link is to a household survey. Most economists, Alan Greenspan included, don't place much stock in the accuracy or reliability of the household survey. However, if you insist, the Household Survey for February show a loss of nearly 270,000 jobs.

A good overview of the differences between payroll data and household surveys.

I think it is misleading to compare current conditions to those at the peak in 2000. To gauge policy efficacy, why not compare the current economy to the economy at this point in Clinton's first term?

I agree that it is misleading to compare snapshots in time. What you have to compare are trends. After all, you wouldn't compare the effectiveness of a diet by merely saying you weighed 180 on date 1 and 185 on date 2. Instead, you would compare the weight lost--over time--between the two diet plans. To do this, you need context, a baseline.

When President Clinton took office, unemployment was 7.3%; by the time of the elections in 1995--that rate was 5.2%. When Bush was appointed, unemployment was 4.2%; it's now 5.6%.

Now, I'd be happy to debate about who inherited a better economy but that's a side issue. What's important is what each man said his plans would accomplish and what they actually accomplished.

President Clinton ran on a plan of deficit reduction. The GOP said this would doom us all; Gingrich said we'd have double digit unemployment, Dole swore this lead to certain recession, maybe even depression. The results are history.

Bush has told us his tax cuts would lead to greater job growth. But as I mentioned earlier, the WH is now refusing to endorse its own projections. Even those which have been scaled back dramatically. The results simply aren't there.

I've read Scheiber and have a much different take. Scheiber seems to be taking the "snapshot" view and ignoring the very real structural shift--or longer view-- that's happening. Where is investment going to come from to produce jobs at sustainable wages? And if that employment doesn't come along, what happens when consumer demand drops? Even Greenspan notes, in his recent testimony before Congress, that we cannot reasonably expect to grow our way out of this mess. There comes a time (and we're not far from it) where the Fed can't jimmy interest rates any further and we won't be running redline in terms of productivity. Then what?

The bottomline is that Bush's plan (tax cuts for the wealthy) have been exceedingly poorly designed in that it's created very little demand.

Well, we'll see once the income tax returns are filed.

And how many millions of people have we added since 2000?

Who are "the wealthy?" Some are finally beginning to say $200K or more a year. But a teacher and a cop on both coasts can make $150K, and $50K puts them into the wealthy class? I'd think they would argue w/some.

One way gas prices could come down is change summer blends. Do we really need over 90 blends for summer?

Put it to America, most don't even know that there are 90 blends across the country. Then the enviros would sue, blah, blah, blah. There are short-term things we can do to ease the crunch. Prices are also high at this point in time because the refineries are mixing the summer blends because 4/1 is right around the corner.

Everybody knows OPEC controlls the oil prices.

Nope. OPEC nations consist of 11 countries which control about 35-40% of the world's oil production. Further, a number of the nations in OPEC (Iran, Libya, Nigeria in particular, although other nations often dissent) often disagree with OPEC production quotas and quite often distribute as much as they can in order to increase their own profits. Further, OPEC nations aren't really keen on cutting production as it means cutting revenues.

So, OPEC has very little control of world oil prices.

We can put that little theory to bed.

Jadegold,

I am aware of the payroll and household data. The data for unemployment is household data, and that is why I linked to it. You spoke of unemployment, that is household data. You said it was underreported, and I wanted to show you that they do report a statistic that encompasses more people. When I spoke of job growth I was using payroll numbers, same as you.

Unemployment:

I compared unemployment just as you did, but you are choosing odd dates. The elections were in Nov. 1996. I compared Feb 2004 to Feb 1996, in other words, the same points in their respective first terms. The unemployment rate (expanded definition) is lower now than it was then and is trending downward, just like then. You once again are comparing unemployment at low point (4.2%) to now, which is misleading. One thing that you didn't mention is that peak unemployment during Bush's first term was 10.3% (again expanded version). In Jan 1994 (the furthest this particular measure goes back) it was 11.8%. If you use the official unemployment rate, the last peak in the unemployment rate was 7.8% in Jun 1992. The peak in Bush II's first term was 6.3%. So Bush II's policies at least seem to have capped the unemployment rate peak at a lower level, although again I don't know how much credit is warranted.

"Side" issue:

The economy they inherited is NOT a side issue. If the previous recession had started two month's after Clinton took office in 1993, his jobs record would have been bad as well. This is precisely why comparing to peaks is misleading. During recessions, jobs are lost, and they continue to be lost even after the official recession is over. Recovery is taking longer after the 2001 recession, that is true. But there are reasons this may be. For instance, the large stock market correction, which seems to have started in middle 2000 and ended at the beginning of 2003. The stock market drop in the previous recession was not as deep (quick look it appears that S&P fell about 15% then versus about 45% current), or as long, it ended in Nov 1990 (again a simple glance).

On deficit politics in the '90s:

Come on. Republicans wanted a balanced budget amendment to the Constitution, that implies zero deficits. I also don't remember the government shutdowns being about the GOP spending too much. The deficit was eliminated for two reasons. First that spending grew slower than the GDP after 1991. After 1994, the GOP affected this. Second, receipts grew faster than GDP. This is a somewhat natural occurence during recovery/growing economy even if taxes aren't increased, by which I mean that you don't have to keep increasing taxes each year for this to occur (In 1993 the "largest tax increase in history" was estimated to be a paltry $250 Billion over 5 years! I don't know what later analysis says actually came in or the effects it had).

Projections:

I said before that I think this is silly. Projections are not guarantees; they are estimates, and are subject to revision or "we were wrong"s. So they estimated that their policies would have an effect by now. So what? The job growth numbers appear to be trending in the right direction. If in one or two months the job numbers are what they predicted for this month, will you change your tune?

The Key Question:
I guess what I am asking is: what is falsifiable about your statements? My statements are falsifiable on the job rate if the rates return to a flat/downward trend like occured in 2003. My statements on the other numbers are easily falsifiable.

Thread Summary(and previous ones to an extent):
You said unemployment was underreported, I showed you an expanded definition that said the same thing as the official rate (lower now than in 1996 and trending downward). Your response was to tell me there is a difference between household and payroll, which I clearly knew, and then compare current unemployment rates to the trough before the recession, which is misleading.

You said 2 million jobs were lost, I said this is from the peak and is misleading. Your response was to say that Bush said he would create jobs by now.

The jobs rate is trending upward, and you claim to care about trends instead of "snapshots" and then point to how low it is in February. To make this somehow consistent with your feeling that snapshots are misleading, you fall back to saying "Their projections were wrong".

Results are results, projections are useless except in gauging how good they are at making projections. Ok, their projections are terrible. But let's talk results. Just give me the stats you want to use and we can discuss.

Mr. Messina: It appears we are talking past one another.

Unemployment: As you note (and hopefully I communicated), there are two sources of unemployment figures: payroll data and household surveys. I related why I believe--as do most economists, not to mention Alan Greenspan--believe household surveys are unreliable and inaccurate.

When I referred to unemployment as being underreported, I am most assuredly referring to payroll data. The unemployment figures are underreported. This is because the Govt. only tracks those workers receiving unemployment benefits. Workers who lose their job and don’t apply for benefits, workers whose unemployment benefits expire, college students who can’t find a job upon graduation are not counted in the monthly unemployment report. In fact, many financial experts believe the actual unemployment rate is about 10%.

WRT unemployment rates at different times--it's a nonissue. Again, I'll plead my case for overall trends, not a month-by-month matchup between 1996 and 2004. As we both know, these rates will fluctuate up or down on a monthly basis.

Side Issue: I'll be pleased to take on this issue. You're assuming Bush was appointed knowing there'd be a recession. If we actually look back to 2001, Bush and his cabinet spent an awful lot of time denying we were in a recession. When recession could no longer could be denied, the Bush WH tacked, claiming it was Clinton's fault.

An important thing to realize is that a recession is defined by when the economy reaches a peak of activity and ends as the economy reaches its trough. A recession is a significant decline in activity spread across the economy, lasting more than a few months, visible in industrial production, employment, real income, and wholesale-retail trade.*

Thus, by NBER's* definition, Bush was appointed with an economy that was in expansion, low unemployment, low inflation, and was running a surplus.

Jadegold

1. I'm pretty sure there is no unemployment data from the payroll (CES) survey. All unemployment data is from the household survey (CPS). Obviously if you ask businesses how many people they are employing, they can't tell you how many they are not employing. There is employment data (CES) and unemployment data (CPS). The CPS also has an employment component that many see as too volatile to use. I will reiterate. The unemployment rates are about the same but lower than they were and are TRENDING downwards. If you think trends are important, what is the problem? Either way, this statistic is improving.

2. The recession began in March 2001, two months after Bush assumed office. He had not signed any economics bills (if any) by that time if I recall correctly. Are you trying to blame the recession on Bush for merely entering the White house? The fact is that the economy was overheated and at the end of its expansion. Again, I don't care what was said, I care about what is and was.

3. On deficit politics, I have already stated that I think that the current deficit (and by extension the policies that led to it) is a problem. That doesn't change what happened in the 90's.

4. Again, if the projections for February are satisfied in April, will you change your tune? Presidents cannot make the economy do anything (another point made by Schreiber). The question "Is the economy bad and/or not improving?" is answerable whether or not the projections are satisfied. The fact that the economy is either about the same point as it was in 1996 and/or is improving in the places it is not is pretty good considering what has happened since early 2001. Now, how much credit to give Bush is debatable, but the fact that it is pretty good and improving is, well, pretty good.

5. Asking you how your argument is falsifiable is not asking you to say what is false about it. Maybe I shouldn't have used that term. I am asking what data would have to be different for you to change your mind about what shape you claim the economy is in. For instance, if the jobs rate in February were +200 thousand, would you change your mind at all? What about if this happens in March or April?

Summary

1. It is misleading because jobs are always lost in a recession. If you compare employment statically from peaks, it will always be lower in the near future. That is the definition of a peak. It is not a fair measure of whether or not the economy is good (historically) and/or improving. The unemployment data tells us that not too many people (historically) are out of work currently. The jobs rate trend (see 2 below) shows that the rate of job creation is improving, even if it is not what Bush projected.

2. The jobs rate is trending upwards. Last February it was (to pick a few data points) -159 thousand, July -45 thousand, January 2004 +97, February +21 (last two subject to revision). It is true that it fell from January to February and at other points in between, but it is trending upwards. Please don't tell me by "trend" you mean just two data points instead of one. Look at the graphs at the BLS site (see here for 1 month, you can also do 3 and 12 to smooth it out a little). It used to be negative, now it is about even/slightly positive (1 month), and if it continues it will be significantly positive (i.e. >+200000). That's trending upwards. This is not to say that it is currently spectacular, I agree it is not. It is also not to say that the trend will definitely continue, it may not. But right now, unemployment is not historically bad for a recovery and the job creation numbers are improving.

Jade,

You're not winning this argument (standard). Michael has the facts to back up his points.

What I notice most, however, is the snide and somewhat pathetic shots about GW being "appointed." Do we need to review the electoral college for you? The popular vote is simply not indicative of the mood of the ENTIRE country. If we went by the popular vote, NYC and LA would choose the president, much to the detriment of the rest of the country. Our forefathers were very wise in this regard.

Let's put that little theory to bed. It's 2004, and the fact dems still bitch and moan about Florida is comical. It won't be that close this time around in Florida.

Noam Scheiber (TNR)...says it's NOT Bush policies. Kerry couldn't do better than Bush did on jobs.
http://www.tnr.com/doc.mhtml?pt=xwkUArrAZheENa9zyjk9Pw%3D%3D

Dan Drezner points out 2.2 million jobs have been created according to the Household Survey;
http://www.danieldrezner.com/archives/001138.html

Speaking of the Household Survey, JadeGold can forget what economists have previously said about it. The major problem has been taken care of. Something about the population data not being properly applied per month but adjusted only in Jan of the following year. (making month by month comparisons iffy) That has been fixed.

That's explained in a bit more detail here. I spent a few hours digesting this juicy analysis of the differences between the Establishment and Household surveys and the HUGE gap between these two at this point in time which is really truly unusual.

http://www.heritage.org/Research/Labor/CDA04-03.cfm

One thing I got from my reading is that far fewer jobs were lost than have been reported due to the way the Establishment data is taken. If you change jobs within the payroll period you are counted twice (which means the number of employed is higher). If not as many people change jobs, fewer are counted twice. And fewer are changing jobs now...there's data to back that up. Exactly how many jobs are falsely 'lost' due to this is not exactly known. This analysis posits as many as 1.3 million. Yes, as many as 1.3 million FEWER jobs lost than we've been hearing.

Also questions in the household survey should be tinkered with a bit. I know when I changed from employee to contractor I myself still considered myself employed by the company I still worked for...not as being self-employed. That misunderstanding probably gives us another 600,000 jobs or so. (These would be picked up by neither survey)

hat tip for all of this is Pejman: http://www.pejmanesque.com/archives/006055.html

Oh, another thing, econopundit had a theory that the tax-cuts actually ADDED 1 million to the number of people in the work force. He ran his data through Fair model I think. No link..it was a bit back and I'm just going from memory.

I love this stuff! I think I've learned more in the last six months of fact-checking stuff, than in all my years in college! LOL

"Bush was appointed with an economy that was in expansion, low unemployment, low inflation, and was running a surplus"

Funny. The day the Martha convictions came down I saw a repeat of an interview with her from October 2000. They were discussing her plans for her company now that the economy was in a downturn.

Saw it with my own eyes (date on the screen). Just thought I'd throw this out...it had me laughing.

But, you see, the economy was in a downturn because it KNEW in advance that Bush was going to get appointed. The market knows all, but tells little.

If only the market had clued in the electorate, the outcome might have been quite different.

And yes, this is sarcasm, mostly in anticipation of a JadeGuy post saying pretty much the same thing.

Oops, forgot to mention that the Heritage Foundation is an evil rightwing think-tank whose every opinion is bought and paid for by big oil, and therefore wrong. Consider it fixed.

One thing you have to remember when debating Jaded and others like her is that they support Democrats, so they KNOW that their leaders are lying. They therefore project and assume that Republicans are also lying.

Syl - here is a link to get you to the Econopundit post you mentioned (involving the Fair Model analysis of the impact of tax cuts on the labor pool.)

Ivan: Rest assured, I'm not concerned about 'losing' this argument. I've challenged those who believe that this economy is so much better than our last democratically-elected President's to make that their campaign theme.

Thus far, the Bush-? 2004 Campaign theme seems to be 'it's not our fault.'

Mr. Messina:

1. You are very much mistaken. Unemployment figures come from the payroll data. In fact, the payroll data is known as the "establishment survey" and is the one used by the BLS and CBO. The payroll (establishment) survey is the one that Wall Street reacts to--and they have an incentive to get it right. The payroll survey has a much bigger sample--600 times the sample size, in fact--there's much less statistical noise in it.

2. The recession happened on Bush's watch. Some might attribute it to the business cycle--but it's immaterial. The real question is how Bush responded. Did his policies prolong the recession or help? I believe there's ample evidence that his policies have not only prolonged the recession but also make a sustained and meaningful recovery very unlikely.

3. I know you find it painful to address this point but I insist; what happened to the GOP demanding a Balanced Budget Amendment to today's 'deficits don't matter?'

I know you know what the answer is.

Further, I'd ask how the GOP could be so wrong as to the effects of Clinton's 93 budget and why this should give us any confidence in what they're proposing today.

4. WRT Scheiber, I think he's mistaken. Presidents can't make the economy doexactly as they wish but they do have some influence. And this gets back to my point that Bush often 'sells' his policies--namely, tax cuts--on the premise it will lead to some great level of employment. The disconnect between his promises and reality is striking.

If Bush is able to create 2.6M new jobs in 2004 as he promised--I'll be amazed. I won't be the only one.

5. See the last para.

The problem is, as each month goes by--Bush falls deeper in the hole. When he promised 2.6M new jobs this year--that means around 320K new jobs each month. So far, he's created ~ 180K in a period where he'd need 640K to maintain pace. He's not keeping pace.

Summary

Bush's employment figures are only slightly trending upward because he's been losing jobs.

One Million Payroll Jobs Short Since October

Jadegold,

1. Please see the link here. Or I can just help you out:

What is the establishment payroll survey?

The establishment payroll survey, known as the Current Employment Statistics (CES) survey, is based on a sample of 400,000 business establishments nationwide. The primary statistics derived from the survey are monthly estimates of employment, hours, and earnings for the Nation, States, and major metropolitan areas. Preliminary national estimates for a given reference month are typically published on the first Friday of the following month, in conjunction with data derived from a separate survey of households, the Current Population Survey (CPS). The CPS is the source of statistics on the activities of the labor force, including unemployment and the Nation's unemployment rate (emphasis mine).

What types of data can one get from the CES survey?

The establishment survey produces nonfarm payroll estimates for: all employees, women workers, production workers, average weekly hours, average hourly earnings (constant dollar and current dollar), average weekly earnings, average overtime, index of aggregate hours and payrolls, and diffusion indexes. All data are available not seasonally adjusted, and some data are available seasonally adjusted.

2. Something is not "immaterial" just because it doesn't support your argument. Bush's "job creation record" is influenced heavily by the fact that two months after he came in, a recession occurred. If he had inherited a growing economy and stock market, and "created" jobs like he has, that would be really bad. Economic plans are not carried out in a vacuum.

3. It doesn't pain me to talk about this (though it does pain me that it happened), it just doesn't affect the particular thing we are talking about in this thread. I have been trying to stay on specific points. What happened? "Compassionate" conservatism + War + Homeland Security bureaucracy + tax cuts. Take your pick. But let's stick to the issues at hand in this thread.

4. Projections are not "promises". Neither are goals. "Will lead" implies that we will get there eventually. Have we yet? No, but trends are improving. Taking a projection/goal, morphing it into a "promise", and then claiming Bush is a lying failure is ridiculous. By that method, he could "create" 1.5 or 2.0 or 2.3 million instead of 2.6 million by November and still be a "liar". You say that you think Scheiber is mistaken. Your argument? Presidents "have some influence". How does that show that Scheiber is mistaken exactly? Apparently, his argument is "immaterial" is well. Because it is really about Bush's economic "lies", right? Projections are projections.

5. So are you are saying "There is no possible evidence (e.g. different numbers) that could theoretically make me think that the economy is not horrible right now. Bush lied!"? If yes, this is no reason to continuing wasting our time. If no, is it only about projections? Then see the last sentence of number 2.

Thanks for the link, Bill!

JadeGold..I think more would 'listen' to you if you could separate the rhetoric from facts. Just saying 'I disagree' is not an argument.

There is a 3 million job discrepancy between the Establishment and Household surveys which is the largest in known history. Dismissing it as some statistical noise is ridiculous.

Syl: The fact is that the establishment survey uses a far bigger sample--as we both know, the larger the sample size, the less chance of significant error.

The household survey has always been considered rather iffy; while it is useful in capturing the self-employed, it is subjective in nature. Among economists, there is no debate that the establishment survey is the more reliable.

Mr. Messina:

1. Again, I'm sticking to my story. The payroll survey is the gold standard. It's what's used by Wall Street and the various economic forecasters to gauge unemployment.

Are these figures revised after the fact? You bet. That's why we always hear about how a given month's unemployment has been revised up or down.

ArgMax

2. That's your interpretation. In general, I'd say its accurate to expect a downturn in employment in a recession. However, once the recession ends, its generally accurate to expect an upturn in employment. By all accounts (really by NBER's), this particular recession was relatively mild and short-lived. Where are the jobs?

I guess the question I'd ask is how long we can continue to blame a mild recession which lasted 7-8 months and ended over 2 1/2 years ago for continuing job woes?

And I'd ask why the economic policies haven't fulfilled Bush's promises WRT employment?

3. OK, you don't want to address it.

I will. From Bush's first press conference on 02/22/01:

"Our budget is fiscally responsible. If enacted, it will reduced the deficit by an unprecedented amount over the next four years."

4. Projections are not "promises".

What are they then? Why put them out? You seem to be making an argument that projections are meaningless.

The fact is that projections are used to 'sell' policies. Bush repeatedly has said his economic policies--namely, tax cuts--will lead to spurring the job market.

BTW, notice that Paul Krugman is using the chart I posted yesterday in his column today?

Yet it hasn't happened.

How does that show that Scheiber is mistaken exactly?

The question should be turned on Scheiber; he is the one making the assertion that a President doesn't have any influence. It's an assertion that's remarkably without supporting evidence.

But let's answer your question of me; I'll defer to the CBO. The CBO seems to believe presidential policy has a fairly significant impact on the economy.

5. Please review the topic of this thread. Bush has made, and is making, promises he simply hasn't kept. Should we continue to reward someone who is constantly telling us things that either aren't accurate or don't reflect reality?

JadeGold, you are factually incorrect in asserting that unemployment numbers come from the Payroll Survey.

Here's a link to the BLS's own web site:

http://www.bls.gov/news.release/empsit.nr0.htm

One heading reads, "Unemployment (Household Survey Data)".

Wall Street may use the Payroll Survey for its estimates of unemployment numbers, but the official unemployment numbers come from the Household Survey.

To the assertion that projections are not promises, you ask, "What are they then? Why put them out?" This is yet another example of your intellectual dishonesty. Projections are NOT promises, and never will be. They are simply the best guess at the future that someone takes with the data at hand. They are put out to show what the data are indicating. They are not guarantees of any kind. And the fact that you conflate the two speaks volumes of your dishonesty.

Anonymous:

I refer you to the ArgMax link in my comment above.

WRT 'projections aren't promises' meme; yes, they are. They are an assertion of your belief that if X happens then Y will happen.

It speaks volumes as to your dishonesty when you claim otherwise. Imagine, if you will, this scenario: you tell your boss that you want to change all the Slurpee machines to one flavor: blue raspberry. Your boss asks why you'd employ this strategy. You respond that you've noticed, as head clerk, that you sell many more blue raspberry Slurpees than any other. Your reasoning is that if all three Slurpee machines are pumping out blue raspberry Slurpees--you could double or triple your sales!

"The fact is that the establishment survey uses a far bigger sample"

The Household survey uses a sample size of 60,000. Are you willing, then, to discount political surveys with a sample size of only 1000?

The more important point is that the two surveys measure different things. The Establishment survey, by design, does not measure non-payroll jobs such as contracting and other self-employment that the Household survey does. It cannot, therefore, be dismissed out of hand. The two surveys historically have a discrepancy for this very reason...but never one as large as now.

You said that the Household survey is more subjective. That very fact is why it should be studied more closely to find a reason for this discrepancy. It seems quite obvious that a structural change is happening in our economy with a major increase in contract workers noted by several economists. This change would not be picked up in the Establishment survey because that tracks only payrolls. Contract workers are not on the payroll. Whether contractors think of themselves as payrolled or self-employed effects how they answer that question on the Household survey. It could, and probably has, made a huge difference in the results.

As for economic projections, do you think the administration pulls these things out of the air? Data is entered into economic models and the models give the results.

Something I haven't seen anyone address: If the number of employed is actually higher than reported (due to errors in Household survey analysis) then productivity is actually LOWER than economists have calculated. Seems to me economists have been surprised by the high productivity numbers...maybe they're not really that high afterall.

Jadegold,

We're getting nowhere, but oh well:

1. That's interesting that they get unemployment data from a data set that doesn't have any unemployment component. Do you think they just subtract employment from the labor force number? Wherever would they get that number... What exactly does your link say about unemployment?

2. Let's look at that. The 1991 recession ended in March 1991. The peak rate of unemployment occurred in June 1992, 15 months later. We can also look at the time it took to get to above +200K/month from the trough of 1 month job creation. For the 1991 recession, it was about 22 months. There's a lag, not an upturn just following the end of the recession.

The 2001 recession ended in November 2001. The peak rate of unemployment occurred in June 2003, 19 months later. A slightly longer lag (though the peak was not as high). Right now we are 28 months after the trough of job creation for the 2001 recession. So yes, there has been much more of a lag here. But taking the easy way out and just saying "Must be Bush's fault" is not good analysis. Or any analysis at all, for that matter.

You claim the recession was mild. Why? Because it only lasted 9 months? What was different about the 2001 recession that might affect things? For one thing (among other possibilities), the stock market dropped significantly until about the beginning of 2003. Maybe you could argue why it didn't affect things and we should treat the 2001 recession as if it were the same as the 1991 recession?

3. Yes, that's correct. I try to stay on topic. I don't need to throw up smokescreens, or respond to them.

4. What are projections, you ask? See number 5. "Will lead" is a future tense. Yes, "it hasn't happened", yet. That's all you can say. Scheiber made his case. You say he is mistaken and then don't say why. Do you think debates are won by the most close-minded person?

5. Perhaps you should use a dictionary to understand the difference between goals, promises, and projections.