Hello Stagflation!

[Surlethe]

Total bullshit. The unemployment rate is calculated using the Current Population Survey, which uses a census method.

A mere 60,000 households out of a population of 300,000,000 people is truly representative? WTF? How can that possibly work?

If it is sufficiently random, the sample is representative of the population. In general, the probability that the sample is not representative of the population is inversely proportional to the sample size; in this case, 60,000 households should be more than enough to make prohibitively low the probability that the unemployment estimate is wrong.

[Broomstick]

What assurance do we have that these households are, indeed, selected randomly?

[Master of Ossus]

[snip]

I'm not entirely certain this is related, but fears of inflation and the falling dollar seem to be driving a surge in commodities prices (as commodities are generally seen as a hedge against inflation), which makes the cost of doing business increase as well. This is apparently driving a reduction in consumer spending. If (as I've often heard it stated) the American economy is largely driven by consumer spending, and demand for consumer goods and services is decreasing, why would the economy add jobs? Shouldn't you see layoffs because of increased costs and decreased sales?

All economies are driven primarily by consumer spending. The point is that reducing the interest rate at which banks can borrow money spurs investment spending, too, by reducing the required rate of return for businesses that are contemplating investing. These marginal businesses will then invest in new capital.

[Master of Ossus]

A mere 60,000 households out of a population of 300,000,000 people is truly representative? WTF? How can that possibly work?

The accuracy of a random sample from a large population is related to the absolute number of its participants; not the proportion sampled. That's how political pollsters can make reasonably accurate predictions about the population's voting trends from samples of about 3,000. And, btw, I love how you compare households to members of the population.

BTW, you ought to know this. I thought you had worked in the healthcare industry. When was the last time you saw a new drug or procedure that had been tested on millions of people? For that matter, when was the last time that you saw 60,000 people as a sample in clinical trials? This never concerned you, then, but as soon as it's in a different field you leap on it as if it's a ginormous methodological flaw?

And mere assurances from the agency conducting the census that this is representative aren't enough - after all, the CPS people have an interest in maintaining their system is accurate whether or not it actually is, not to mention the government has an interest in keeping the unemployment rate low which, if you can't do it in reality, could be done by tweaking the reporting. Rather like the folks who say the economy is doing well because the stock market is up and ignore everything else.'

I love the way that you shifted your goalposts there. When it turned out that you were utterly bullshitting about your first complaint about the CPS, you come up with a new problem with it: "mere assurances from the agency conducting the census that this is representative aren't enough." The CPS methodology is open to peer review. They actually send out all of their data to researchers and academics each year, as wel as a voluminous manuscript on how they generated it (in text files, it constitutes several hundred megabytes of information). Is peer review insufficient to guarantee sufficient accuracy to base government decisions on? Moreover, this entire discussion is in the context of whether or not the definition of unemployment is reasonable for formulating the definition of stagflation. That has much more to do with the CONSISTENCY of the unemployment definition than with its accuracy.

Also note the following from your own article:

Yes, there is only one official definition of unemployment and that was discussed above. However, a number of analysts believe this measure to be too restricted, that it does not adequately capture the breadth of labor market problems.

I pointed that out earlier, you fucking moron. I disagree with the way that they tabulate unemployment. However, that is the official US government measure of unemployment, and it's the one that is regularly used by government agencies and the Federal Reserve. Indeed, I think that the BLS probably would have changed the definition of unemployment, by now, except that other government agencies and researchers count on them to maintain consistency on such important measures (it was a big deal, for instance, when Greenspan got the Fed to switch inflation calculation to a chain-weighted system even though a huge number of economists believed it to be a more accurate method of estimating the CPI, and was really only a comparatively small change from the existing system).

[Master of Ossus]

What assurance do we have that these households are, indeed, selected randomly?

It's not random. The population of the US is divided into primary sampling units and households are then selected randomly from within those sampling units. This is not the same thing as a random sample of the US, since their system also needs to have data that is accurate down to smaller granules (for instance, a researcher working on the State of Idaho could easily isolate observations from that state from the general population and calculate an unemployment rate only for the state of Idaho that would be accurate).

As for "assurances" that the methodology is acceptable, the BLS methods are all subject to peer, and (in fact) public review. And, btw, how the fuck is the BLS even theoretically able to assure people that the sample is random. "Well, guys, we decided that this month, we're going to survey every household across the nation that ends in the number 00011. Oh, wait, that's not a random sample. Oh, shit. Well, okay. To make sure that morons like Broomstick don't get on our cases, let's just use the RAND corporation's booklet on random numbers to pick out households. Oh, shit. That's not a random sample, either." Do you see the issue, here? The defining characteristic of a random sample is its randomness. The randomness of a sample can only be measured ex ante, and it is impossible to prove that a sample is random after-the-fact, except by recourse to the methodology used to select the sample... which, of course, is random.

[Broomstick]

A mere 60,000 households out of a population of 300,000,000 people is truly representative? WTF? How can that possibly work?

The accuracy of a random sample from a large population is related to the absolute number of its participants; not the proportion sampled. That's how political pollsters can make reasonably accurate predictions about the population's voting trends from samples of about 3,000. And, btw, I love how you compare households to members of the population.

Well, since nowhere in that article do I see a definition of how many people are in what they consider a typical household how in the hell am I supposed to make and apples:apples comparison? IS there a definition of "household" that includes numbers of people?

BTW, you ought to know this. I thought you had worked in the healthcare industry. When was the last time you saw a new drug or procedure that had been tested on millions of people? For that matter, when was the last time that you saw 60,000 people as a sample in clinical trials? This never concerned you, then, but as soon as it's in a different field you leap on it as if it's a ginormous methodological flaw?

First of all, one does not normally test drugs on a completely random segment of the population. There are a few studies, such as the Nurses's health study, that follows vary large numbers of people for years but they are the exception. When testing a new drug you test in clinical trials of people suffering from the condition in question, preferably with the elimination of inconsistencies. So, for example, a drug for high blood pressure would be tested on people with high blood pressure, not the general population, and you would have to either make sure all study participants are suffering from the same co-morbid conditions - say, all diabetics with high blood pressure - or not suffering from co-morbid conditions at all. The randomization comes into play when assigning people to experimental vs. control groups. In other words, no, it's not the same as randomly sampling the US population.

And, as I said in an earlier post, what assurance do we have that the method of sampling used it truly representative?

And mere assurances from the agency conducting the census that this is representative aren't enough - after all, the CPS people have an interest in maintaining their system is accurate whether or not it actually is, not to mention the government has an interest in keeping the unemployment rate low which, if you can't do it in reality, could be done by tweaking the reporting. Rather like the folks who say the economy is doing well because the stock market is up and ignore everything else.'

I love the way that you shifted your goalposts there.

I'm not shifting the goalposts, I'm asking for independent verification that this method is on the level because, quite frankly, I am NOT an expert in statistical analysis and I know it.

I'm an American, which means I inherently distrust my own government. I want someone without a vested interest in the outcome to verify that what they describe here is legit. Which, by the way, Surlethe did at least in part.

The CPS methodology is open to peer review. They actually send out all of their data to researchers and academics each year, as wel as a voluminous manuscript on how they generated it (in text files, it constitutes several hundred megabytes of information). Is peer review insufficient to guarantee sufficient accuracy to base government decisions on?

Point me to this information. Do you have links to it? Something whereby I can go to my local library and look up this verification?

Back to medical testing - it's not enough to say "it's peer reviewed" - you have to list such reviews. And the reputation and/or possible conflict of interest of the party doing the review also can be a factor in assessing whether or not that review is worthy of consideration.

That has much more to do with the CONSISTENCY of the unemployment definition than with its accuracy.

So... as long as it's consistent you're saying it's totally OK if it's wildly inaccurate? As long as it's consistently inaccurate? Is that really what you're saying here?

Back to medical science - consistent results that are also inaccurate are considered bullshit at best.

However, that is the official US government measure of unemployment, and it's the one that is regularly used by government agencies and the Federal Reserve.

But, according to you, it's OK for it to be inaccurate as long as it's consistent... even though policy is set by these numbers? Is that how we want to run a country?

You know, I don't care how much a fucking bother it is to change things - if the current system is giving inaccurate numbers change it. If it's accurate then the methods used should easily withstand questioning.

[Beowulf]

BLS data There you go Broomstick. Go nuts.

[Chardok]

BLS data There you go Broomstick. Go nuts.

...You're an evil, evil man...

[Broomstick]

BLS data There you go Broomstick. Go nuts.

Thank you.

It will, of course, take me a little time to digest all that.

[Alan Bolte]

I notice you didn't address the first half of my post, MoO, where I said I understand that under normal conditions the Fed would be doing the right thing, but suggested that we might be experiencing extraordinary conditions, where reducing that interest rate is less effective, at best. Really, it reads like you just took two of my sentences and restated them in a way you thought sounded better. If there were some inaccuracies in the way I stated the issue, you could at least point them out.

[J]

With regards to the current economic situation and stagflation, we have to look at the big picture, in other words, what got us into this mess. Then we'll look at the claims by economists and Fed officials and see whether we'll pull out of it just fine by the end of the year as they say or if we are truly in deep doo-doo.

So what are the root causes of the current mess?

1. Implosion of the housing bubble, starting with the subprime sector and spreading into the rest of the market. This has wiped out a large source of consumer spending as many people rely on the equity in their homes to fuel their spending. People are also defaulting and walking away from their homes after ending up with negative equity, which is bad news for the financial institutions which use their mortgages as the base of their highly leveraged and questionable investment schemes. This is also costing towns & cities a lot in tax revenues as property values continue to fall.
2. Collapse in value of highly leveraged and unregulated investment vehicles, starting with subprime-backed investments and spreading into the rest of the sector, caused by the above. This has now wiped out hundreds of billions of dollars from banks & other financial institutions, and destroyed almost all the liquidity in the banking system.
3. The bonds auction market has locked up in recent weeks putting yet more strain on financial institutions. As a side effect, this has added a large cost to municipalities, school boards, hospitals, port & transportation authorities and quite a few others. There's a town in California that's about to go insolvent, and this is after it's laid off a bunch of workers to cut costs which has done wonders for unemployment.
4. The Federal Reserve bank's continued rate cutting in an attempt to keep the economy going is tanking the US dollar and setting the stage for inflation to take off. Helicopter Ben has lived up to his name and then some.
5. US consumers have run up record levels of debt, and now have no way to spend money to stimulate the economy

Condition #1 is only going to get worse, as I've mentioned before it generally takes around 6 months from the time home owners stop sending their cheques in until the home is foreclosed, the homes which are being foreclosed right now were the ones where the owners stopped paying last August, not too long after the subprime problem began. Since then mortgage defaults have only gone up and it's only going to get worse since the peak month for subprime & ARM rate resets is March, that's when the rates go up and start off a whole new wave of defaults which will then take half a year or so to work their way to foreclosure. It then takes a little while longer for the investment vehicles based on those homes & mortgages to go into a funk.

Condition #2 isn't going to improve until the housing situation is resolved, and furthermore, liquidity will be absent and lending locked up until most financial institutions open up their books for audits and disclose how much exposure they have to various toxic CDO's, whether the bonds they're packaging are really AAA, and so on and so forth. The system will not function properly until all the writedowns are taken and the appropriate disclosures & audits are made to restore trust. Only then can the recovery begin. This is going to take some time to work out, it certainly won't be done by Q3 or Q4 this year.

Condition #5 isn't going to fix itself until American learn to save their money and pay down their debts. From this Bloomberg article, it seems like Americans are beginning to grow a brain. Excerpts:

Feb. 28 (Bloomberg) -- The stimulus plan Congress approved this month may provide less of a jolt to the U.S. economy than intended, as most Americans plan to save rather than spend their tax rebates, a Bloomberg/Los Angeles Times survey shows.

Only 18 percent of respondents said they will spend their rebate on purchases, while slightly more than three in 10 said they prefer to use the money to pay off debt, and a third said they'll pocket it.

``People in Washington assume that about 40 percent of the money will be spent,'' said Douglas Elmendorf, a senior fellow at the Brookings Institution, a Washington-based research organization. ``Much less would be disappointing.''

Plans to spend the rebate were consistent across income groups: About two in 10 respondents in households with less than $40,000 annual income as well as those making more than $100,000 said they'll spend the money. A plurality of respondents in the lower-income group said they'll pay off debt, while saving the money was the most popular plan among the wealthier group.

Problem is if consumers aren't spending, the economy isn't going to get much stimulation which makes pulling out of a downturn rather difficult.

On top of that we have a couple more trends which point towards bad news. There is a strong correlation between the Consumer Confidence Index and the state of the economy, take a look at this chart and you can see where it lines up with the early 90's slump, 9/11, and the Dot Com bust. You can also see that it's been trending downwards in recent months and is currently sitting at 75, generally accepted as recession territory.

On the manufacturing side, durable goods orders are down but more importantly, inventories have been moving up for months which implies that product isn't moving. Rising inventories of durable goods also has a good correlation to economic downturns.

And just to make thing more fun, we also have to deal with rising energy prices which are beginning to put a nice crimp in consumer spending.


Put everything together and it adds up to this:

• The US is in a recession, whether people want to admit it or not
• Recovery will not happen this year, there's a lot of crap in the system which hasn't even begun working itself out. A lot of things need to be sorted out before the recovery can begin, and it's likely to get a lot worse before things start improving
• Helicopter Ben will continue dropping interest rates and throwing money into the system, further weakening the dollar and running up inflation risks.

It's my opinion that the talking head economists and Fed officials are completely full of shit when they claim that this is just a minor hiccup, and that recovery will be well on its way by Q3 this year. This is an opinion shared by Karl Denninger who runs the MarketTicker forums as well as many on The Oil Drum. Not to mention Jim Kunstler.

I believe we're in for a world of trouble, if we're lucky and everything goes right we'll get by with a moderate recession with a mild dose of stagflation. More likely, we're in for a fairly deep recession with a fair helping of stagflation and other nasties thrown in.

[Master of Ossus]

Well, since nowhere in that article do I see a definition of how many people are in what they consider a typical household how in the hell am I supposed to make and apples:apples comparison? IS there a definition of "household" that includes numbers of people?

You could have compared the number of households in the survey to... the number of households in the US (about a hundred million).

First of all, one does not normally test drugs on a completely random segment of the population.

But presumably they select from a random sample of the population with a condition that they drug/treatment is meant to treat, no? Or, at least, from a subsection of that group that constitute the entire population of the study.

There are a few studies, such as the Nurses's health study, that follows vary large numbers of people for years but they are the exception. When testing a new drug you test in clinical trials of people suffering from the condition in question, preferably with the elimination of inconsistencies. So, for example, a drug for high blood pressure would be tested on people with high blood pressure, not the general population, and you would have to either make sure all study participants are suffering from the same co-morbid conditions - say, all diabetics with high blood pressure - or not suffering from co-morbid conditions at all. The randomization comes into play when assigning people to experimental vs. control groups. In other words, no, it's not the same as randomly sampling the US population.

I agree. The CPS does not randomly select from the entire US population, either. But the issue is whether the NUMBER of people in a particular random sample is more critical or the PERCENTAGE of the overall population with a particular condition that you're looking to treat. Clearly the absolute number is the more important statistic for any large sample size.

And, as I said in an earlier post, what assurance do we have that the method of sampling used it truly representative?

It's based on the US census. I linked you to the page in which the BLS describes its methodology.

I'm not shifting the goalposts, I'm asking for independent verification that this method is on the level because, quite frankly, I am NOT an expert in statistical analysis and I know it.
I'm an American, which means I inherently distrust my own government. I want someone without a vested interest in the outcome to verify that what they describe here is legit. Which, by the way, Surlethe did at least in part.

How in heck is saying "If it is sufficiently random, the sample is representative of the population. In general, the probability that the sample is not representative of the population is inversely proportional to the sample size; in this case, 60,000 households should be more than enough to make prohibitively low the probability that the unemployment estimate is wrong" more responsive to your question than saying that the number of households is clearly the more valid measure, as opposed to the fraction of households? I even gave you an example of political polling.

Point me to this information. Do you have links to it? Something whereby I can go to my local library and look up this verification?

I gave you a fucking link to it. Please read my posts.

Back to medical testing - it's not enough to say "it's peer reviewed" - you have to list such reviews. And the reputation and/or possible conflict of interest of the party doing the review also can be a factor in assessing whether or not that review is worthy of consideration.

1. WHAT possible conflict of interest does the BLS have, today, that they have not had across time?
2. Academia uses the CPS extensively because it's such a well-respected source of information. For a summary of criticisms of the CPS, see (e.g.,) Bollinger 1998.

Back to medical science - consistent results that are also inaccurate are considered bullshit at best.

But in medicine, they're not used in the definitions of other things. Consistency is obviously a more important measure for evaluating their utility in DEFINING something else. In this case, saying that an unemployment rate above 5% means something can ONLY be reasonable if the definition of unemployment remains constant. Were you to change the definition of unemployment, then clearly the 5% marker could also change without altering the fundamental point.

But, according to you, it's OK for it to be inaccurate as long as it's consistent... even though policy is set by these numbers? Is that how we want to run a country?

You know, I don't care how much a fucking bother it is to change things - if the current system is giving inaccurate numbers change it. If it's accurate then the methods used should easily withstand questioning.

What, precisely, is the problem with the unemployment rate that requires it to be changed? It measures a particular thing. Once you recognize that it measures a particular thing, then you can utilize that to produce consistent results. The main objection that I have for the definition is that it defines "unemployment" instantaneously, instead of across time. I don't think their definition "wrong," because they've defined "unemployment" a particular way.

Do you seriously not understand how something like "unemployment" in a statistical survey can be defined in various ways that are all defensible? Given this ambiguity in the term, what is so unreasonable about the definition that the CPS has chosen to use that it should be changed? It's not even inaccurate--it's VERY accurate, but it measures a particular thing and it does not necessarily measure the definition that you or I or someone else might want it to measure. That doesn't mean that it's invalid.

[Master of Ossus]

I notice you didn't address the first half of my post, MoO, where I said I understand that under normal conditions the Fed would be doing the right thing, but suggested that we might be experiencing extraordinary conditions, where reducing that interest rate is less effective, at best. Really, it reads like you just took two of my sentences and restated them in a way you thought sounded better. If there were some inaccuracies in the way I stated the issue, you could at least point them out.

Can you provide evidence to the effect that banks are so terrified of the credit crunch that they have stopped lending AT ALL to people, as opposed to merely tightening the purse strings? I see no theoretical or empirical reason to accept the idea that banks will no longer respond at all to interest rates highly dubious, since we can observe that banks are still in the business of credit. Moreover, you ARE the one who bears the burden of proof if you're suggesting that an observed historical trend will not continue.

[Master of Ossus]

So what are the root causes of the current mess?

1. Implosion of the housing bubble, starting with the subprime sector and spreading into the rest of the market. This has wiped out a large source of consumer spending as many people rely on the equity in their homes to fuel their spending. People are also defaulting and walking away from their homes after ending up with negative equity, which is bad news for the financial institutions which use their mortgages as the base of their highly leveraged and questionable investment schemes. This is also costing towns & cities a lot in tax revenues as property values continue to fall.

But it has a direct effect on prices, which drop as a result of all of this. This will tend to counter inflation.

[*]Collapse in value of highly leveraged and unregulated investment vehicles, starting with subprime-backed investments and spreading into the rest of the sector, caused by the above. This has now wiped out hundreds of billions of dollars from banks & other financial institutions, and destroyed almost all the liquidity in the banking system.

I don't even understand how destroying liquidity in the banking system is possible in the way that you describe. The whole point of the Federal Reserve is that it can create or destroy liquidity in the financial markets at will.

[*]The bonds auction market has locked up in recent weeks putting yet more strain on financial institutions. As a side effect, this has added a large cost to municipalities, school boards, hospitals, port & transportation authorities and quite a few others. There's a town in California that's about to go insolvent, and this is after it's laid off a bunch of workers to cut costs which has done wonders for unemployment.

I don't understand what you mean by an auction market "locking up." Moreover, small businesses and governments go insolvent all the time. Why is this creating a particularly large jump in unemployment, and why in the world would it lead to inflation?

[*]The Federal Reserve bank's continued rate cutting in an attempt to keep the economy going is tanking the US dollar and setting the stage for inflation to take off. Helicopter Ben has lived up to his name and then some.

First of all, Bernanke also mentioned repeatedly that he's more concerned now with inflation than with unemployment, which is something that Greenspan essentially never said. Second of all, this is precisely one of the unemployment-fighting measures that you poo-poohed in the opening post.

[*]US consumers have run up record levels of debt, and now have no way to spend money to stimulate the economy[/list]

So with a large amount of debt, consumers can spend to stimulate the economy, but with even a slightly higher level of debt they can't do that, anymore?

I fundamentally do not even understand two of your concerns. Moreover, virtually all of these concerns except for the specific housing market problems (which would tend to reduce inflation in the economy) could have been argued during the last recession (or the one before that...) but we did not observe stagflation during that time.

[Broomstick]

Moreover, small businesses and governments go insolvent all the time. Why is this creating a particularly large jump in unemployment, and why in the world would it lead to inflation?

While businesses go under all the time, it's only a small percentage of businesses. Beyond a certain percentage you start getting problems you didn't see before. Maybe you haven't been unemployed before, MoO, but when you lose your main income source you go on this thing called a "budget" where you stop buying anything but true necessities - housing, food, fuel, maybe clothing. The number of foreclosures on housing seems to indicate that many households can't even manage that much. People aren't buying luxuries like brand new vehicles, new electronic toys, eating at expensive restaurants, taking trips, upgrading services... the result of that being a sudden drop in demand for all goods and services outside of necessities, which means THOSE companies cut back on employees or even go out of business.... The economic multipler can work both ways - increasing jobs when things are good, and leading to a decrease when too many companies go under.

Why is that beyond your comprehension?

Admittedly, I haven't considered the effect on inflation, but to me it's obvious why a lot of companies going under or cutting back on employees leads to a sudden spike in unemployment

[*]US consumers have run up record levels of debt, and now have no way to spend money to stimulate the economy

So with a large amount of debt, consumers can spend to stimulate the economy, but with even a slightly higher level of debt they can't do that, anymore?

Yeah, because 1) it's harder to get loans/credit now and 2) at some point you run out of money/credit. I'm sorry it's so hard for you to understand that money is not a limitless resource. At some point you really CAN spend your last dollar.

What, precisely, is the problem with the unemployment rate that requires it to be changed? It measures a particular thing. Once you recognize that it measures a particular thing, then you can utilize that to produce consistent results.

What's wrong with it? I'm still appalled at your statement:

That has much more to do with the CONSISTENCY of the unemployment definition than with its accuracy.

What OTHER area of human endeavor is consistency more important than accuracy? I can't imagine an engineer stating "well, I don't know how far out of spec this steel may or may not be, but because we're really consistent with the production techniques however much it is out of spec it is consistently out of spec". Would you want to live or work in a building using such steel?

Now, in an area like, say, weather prediction there are issues of inaccuracy that have to do with the system being studied and computer capacity and the like, but meterologists don't spew bullshit like "well, there might be better ways of recording/predicting weather than what we're using, but we won't use them because it's more important we get consistent results than accurate ones." I'm sure it's inconvenient and expensive to continually upgrade computers and software and launch new weather sattelites and all that but they do it anyway.

If there is a better way to measure an economic indicator then fucking use it and don't bitch to me about "consistency" - consistent inaccuarcy doesn't impress me. I don't fucking care if it's inconvenient. Or just admit that the so-called "science" of economics is no more a hard science than psychology - if consistency is more important than accuracy then it's not, and never will be, a science on the level of physics or engineering.

Or just admit you worded your post badly.

[Darth Wong]

So with a large amount of debt, consumers can spend to stimulate the economy, but with even a slightly higher level of debt they can't do that, anymore?

I'm not particularly interested in a drawn-out economics discussion, but the logic here seems bizarre. You seem to be saying that if a system can handle X amount of consumer debt without problems, then it must be able to handle X + Y amount of consumer debt without problems. Why? Is there some logic principle operating here that I've never heard of?

[Master of Ossus]

So with a large amount of debt, consumers can spend to stimulate the economy, but with even a slightly higher level of debt they can't do that, anymore?

I'm not particularly interested in a drawn-out economics discussion, but the logic here seems bizarre. You seem to be saying that if a system can handle X amount of consumer debt without problems, then it must be able to handle X + Y amount of consumer debt without problems. Why? Is there some logic principle operating here that I've never heard of?

It seems extraordinarily unlikely that this is some sort of on-off switch, as J's statement implies. Presumably, there is some continuous function whereby at X+Y level of debt they spend marginally less, but not ZERO unless Y is incredibly large--that is not the sort of situation that we're observing, here, and particularly because the US dollar has depreciated and therefore the nominal debts incurred are less costly in terms of real value.

[Master of Ossus]

While businesses go under all the time, it's only a small percentage of businesses. Beyond a certain percentage you start getting problems you didn't see before. Maybe you haven't been unemployed before, MoO, but when you lose your main income source you go on this thing called a "budget" where you stop buying anything but true necessities - housing, food, fuel, maybe clothing. The number of foreclosures on housing seems to indicate that many households can't even manage that much. People aren't buying luxuries like brand new vehicles, new electronic toys, eating at expensive restaurants, taking trips, upgrading services... the result of that being a sudden drop in demand for all goods and services outside of necessities, which means THOSE companies cut back on employees or even go out of business.... The economic multipler can work both ways - increasing jobs when things are good, and leading to a decrease when too many companies go under.

Why is that beyond your comprehension?

It's not "beyond [my] comprehension." It's just not what's happening, right now. As I have repeatedly stated, the unemployment rate is about 5%. I know that you're unemployed, but for every one of you there are 19 others who still have jobs. We're not seeing a snowball effect leading to a large recession, Broomstick.

Admittedly, I haven't considered the effect on inflation, but to me it's obvious why a lot of companies going under or cutting back on employees leads to a sudden spike in unemployment

Where is this "sudden spike in unemployment?"

Yeah, because 1) it's harder to get loans/credit now and 2) at some point you run out of money/credit. I'm sorry it's so hard for you to understand that money is not a limitless resource. At some point you really CAN spend your last dollar.

Right. And this would happen simultaneously to all consumers, rather than as some sort of a gradual process?

What, precisely, is the problem with the unemployment rate that requires it to be changed? It measures a particular thing. Once you recognize that it measures a particular thing, then you can utilize that to produce consistent results.

What's wrong with it? I'm still appalled at your statement:

That has much more to do with the CONSISTENCY of the unemployment definition than with its accuracy.

What OTHER area of human endeavor is consistency more important than accuracy? I can't imagine an engineer stating "well, I don't know how far out of spec this steel may or may not be, but because we're really consistent with the production techniques however much it is out of spec it is consistently out of spec". Would you want to live or work in a building using such steel?

Broomstick, you moron, the point is that in order to define Y by X, X has to be measured consistently. To change the definition of X also changes the definition of Y, in such a circumstance. Why is this hard for you to follow? Moreover, the US unemployment rate isn't "wrong," it just measures something very specific. I don't like the way that it's defined, but that just means that I would define it differently. It doesn't mean that we should suddenly change the system. And, frankly, that's clearly what I'm talking about above (a statement you totally ignored in favor of going back to a previous statement rather than one meant to clarify it).

Now, in an area like, say, weather prediction there are issues of inaccuracy that have to do with the system being studied and computer capacity and the like, but meterologists don't spew bullshit like "well, there might be better ways of recording/predicting weather than what we're using, but we won't use them because it's more important we get consistent results than accurate ones." I'm sure it's inconvenient and expensive to continually upgrade computers and software and launch new weather sattelites and all that but they do it anyway.

You don't see the weather guys running around changing the definition of "cloudy" to mean "having one cloud over a period of 24 hours." It's not a matter of improving accuracy. It's a matter of DEFINITIONS. The US unemployment rate is DEFINED to be a particular thing, and people who follow the US unemployment rate accept that this is what the measurement is telling them.

If there is a better way to measure an economic indicator then fucking use it and don't bitch to me about "consistency" - consistent inaccuarcy doesn't impress me. I don't fucking care if it's inconvenient. Or just admit that the so-called "science" of economics is no more a hard science than psychology - if consistency is more important than accuracy then it's not, and never will be, a science on the level of physics or engineering.

You really don't get it, do you? This is a problem of a DEFINITION of a measurement.

Or just admit you worded your post badly.

It's possible that I worded my post badly, but the substantive meaning of the post was abundantly clear from the context. It's not a matter of accuracy, because the unemployment rate is an extremely accurate measure of what it is defined to measure--precisely the fact that is involved, here.

[Edi]

So what are the root causes of the current mess?

1. Implosion of the housing bubble, starting with the subprime sector and spreading into the rest of the market. This has wiped out a large source of consumer spending as many people rely on the equity in their homes to fuel their spending. People are also defaulting and walking away from their homes after ending up with negative equity, which is bad news for the financial institutions which use their mortgages as the base of their highly leveraged and questionable investment schemes. This is also costing towns & cities a lot in tax revenues as property values continue to fall.

But it has a direct effect on prices, which drop as a result of all of this. This will tend to counter inflation.

Price drops will counter inflation because then people will more readily spend money. In this case, it's going to do absolutely squat, because people have NO money to spend anywhere and what they do spend it on (food, energy, other essentials) is not affected one whit just by the housing market crashing and burning down around your ears. They won't buy anything else and you get the problems Broomstick outlined.

[*]Collapse in value of highly leveraged and unregulated investment vehicles, starting with subprime-backed investments and spreading into the rest of the sector, caused by the above. This has now wiped out hundreds of billions of dollars from banks & other financial institutions, and destroyed almost all the liquidity in the banking system.

I don't even understand how destroying liquidity in the banking system is possible in the way that you describe. The whole point of the Federal Reserve is that it can create or destroy liquidity in the financial markets at will.

For a guy who claims to understand economics, you sure seem clueless. The banks loaned a shitload of money out to people who could not realistically pay those loans back. They then claimed that loaned out money and its prospective interest as assets and sold them to other institutions with clever little gimmicks and supposedly their value went up. This was repeated several times. And when the underlying assets behind it all turned out to be illusion as the loans default and nobody gets anything back, it all gets wiped out in a cascade. It's a pyramid scheme.

So the central bank can only print more money to increase supply and that increases inflation.

[*]The Federal Reserve bank's continued rate cutting in an attempt to keep the economy going is tanking the US dollar and setting the stage for inflation to take off. Helicopter Ben has lived up to his name and then some.

First of all, Bernanke also mentioned repeatedly that he's more concerned now with inflation than with unemployment, which is something that Greenspan essentially never said. Second of all, this is precisely one of the unemployment-fighting measures that you poo-poohed in the opening post.

Doesn't matter whether people have jobs or not if they can't afford basic necessities because the prices go up due to inflation. Too many people in the US already struggle to make ends meet.

[*]US consumers have run up record levels of debt, and now have no way to spend money to stimulate the economy

So with a large amount of debt, consumers can spend to stimulate the economy, but with even a slightly higher level of debt they can't do that, anymore?

If they are already using debt to fuel spending, the whole economy is running on fumes already and those fumes are now running out. When the gas tank is empty, the car just won't fucking move. This is not difficult to understand.

I fundamentally do not even understand two of your concerns. Moreover, virtually all of these concerns except for the specific housing market problems (which would tend to reduce inflation in the economy) could have been argued during the last recession (or the one before that...) but we did not observe stagflation during that time.

Are you familiar with the concept of synergy, that the whole can be greater than the sum of its parts? This exactly what you have there, and in this case the synergy effect comes from a combination of factors that are bad enouh on their own.

If you are so fundamentally incapable of understanding this discussion and its basics, so incapable of seeing the big picture instead of just single pieces of the puzzle, you should just stay out of the discussion until you get well enough acquainted with the real world to be able to competently assess it.

[Admiral Valdemar]

There are concerns that people are paying for their heating oil and petrol via credit card. I don't think I need to draw a picture on how fucked up that is and how it means bad things are on the way if the situation carries on this same route.

[Broomstick]

It's not "beyond [my] comprehension." It's just not what's happening, right now. As I have repeatedly stated, the unemployment rate is about 5%.

Yet you yourself have stated that the accuracy of the method used to determine that figure is not known...

We're not seeing a snowball effect leading to a large recession, Broomstick.

Not yet.

Admittedly, I haven't considered the effect on inflation, but to me it's obvious why a lot of companies going under or cutting back on employees leads to a sudden spike in unemployment

Where is this "sudden spike in unemployment?"

I was speaking in a general and not a specific sense, dickhead. Please try to keep up. Even if we don't have that condition now it has occured in the past and it is conceivable to normal human beings that this situation could occur again in the future.

Yeah, because 1) it's harder to get loans/credit now and 2) at some point you run out of money/credit. I'm sorry it's so hard for you to understand that money is not a limitless resource. At some point you really CAN spend your last dollar.

Right. And this would happen simultaneously to all consumers, rather than as some sort of a gradual process?

It could. As I said, there is historical precedence for sudden upsets and economic collapses. It's sort of the flip side of the economic boom/gold rush effects.

What, precisely, is the problem with the unemployment rate that requires it to be changed? It measures a particular thing. Once you recognize that it measures a particular thing, then you can utilize that to produce consistent results.

What's wrong with it? I'm still appalled at your statement:

That has much more to do with the CONSISTENCY of the unemployment definition than with its accuracy.

What OTHER area of human endeavor is consistency more important than accuracy? I can't imagine an engineer stating "well, I don't know how far out of spec this steel may or may not be, but because we're really consistent with the production techniques however much it is out of spec it is consistently out of spec". Would you want to live or work in a building using such steel?

Broomstick, you moron, the point is that in order to define Y by X, X has to be measured consistently. To change the definition of X also changes the definition of Y, in such a circumstance. Why is this hard for you to follow?

I get what you're saying - you're saying that the accuracy of the measure doesn't matter as long as the measuring tool is consistent. Why don't you understand that for many, if not most, areas of human endeavor that is NOT an acceptable standard?

Moreover, the US unemployment rate isn't "wrong," it just measures something very specific.

You also stated that the method used to measure it was consistent but of unkown accuracy. Again, how can that be justified?

I don't like the way that it's defined, but that just means that I would define it differently. It doesn't mean that we should suddenly change the system.

If changing the system would give us more accurate information then yes, we should change it.

Now, in an area like, say, weather prediction there are issues of inaccuracy that have to do with the system being studied and computer capacity and the like, but meterologists don't spew bullshit like "well, there might be better ways of recording/predicting weather than what we're using, but we won't use them because it's more important we get consistent results than accurate ones." I'm sure it's inconvenient and expensive to continually upgrade computers and software and launch new weather sattelites and all that but they do it anyway.

You don't see the weather guys running around changing the definition of "cloudy" to mean "having one cloud over a period of 24 hours." It's not a matter of improving accuracy. It's a matter of DEFINITIONS.

If you can't define the accuracy of what you're measuring perhaps you should change the definition, then.

The US unemployment rate is DEFINED to be a particular thing, and people who follow the US unemployment rate accept that this is what the measurement is telling them.

What good is measuring X - however it is defined - if we don't know how accurate that measurement is? With something as multi-factorial as the economy how can you know whether or not you're overlooking something if you don't even know how accurate your measurements are?

You really don't get it, do you? This is a problem of a DEFINITION of a measurement.

You really don't get it, do you? What you state as obvious is voodoo to the average reader, where it doesn't just flat-out defy common sense and real-world observation.

Or just admit you worded your post badly.

It's possible that I worded my post badly, but the substantive meaning of the post was abundantly clear from the context.

It wasn't clear - that's what I'm asking you questions. I fully admit I am not an authority on either statistics or economics, that's WHY I ask questions in these sorts of threads. You, on the other hand, seem to expect everyone to be an expert in these matters, expert apparently being defined as "agrees with everything MoO says without question".

It's not a matter of accuracy, because the unemployment rate is an extremely accurate measure of what it is defined to measure--precisely the fact that is involved, here.

But you stated that the accuracy of the method wasn't certain. The article you linked to also stated that some authorities questioned the accuracy of the method - how can it be "extremely accurate" AND the accuracy be questioned at the same time? That doesn't make sense.

[aerius]

There are concerns that people are paying for their heating oil and petrol via credit card. I don't think I need to draw a picture on how fucked up that is and how it means bad things are on the way if the situation carries on this same route.

It's all fun and games until people start paying their income taxes and mortgages with credit cards.

[Master of Ossus]

But it has a direct effect on prices, which drop as a result of all of this. This will tend to counter inflation.

Price drops will counter inflation because then people will more readily spend money. In this case, it's going to do absolutely squat, because people have NO money to spend anywhere and what they do spend it on (food, energy, other essentials) is not affected one whit just by the housing market crashing and burning down around your ears. They won't buy anything else and you get the problems Broomstick outlined.

No. Price drops will counter inflation because inflation, by definition, is a rise in the price level of goods and services. Price DROPS, therefore, have a direct negative effect on inflation. Moreover, I like how you remove housing from the list of "essentials" and say that the other goods are unaffected by housing. That may be true, but it's irrelevant. A price drop in any one of those areas will decrease inflation, and in fact promote DEflation.

For a guy who claims to understand economics, you sure seem clueless. The banks loaned a shitload of money out to people who could not realistically pay those loans back. They then claimed that loaned out money and its prospective interest as assets and sold them to other institutions with clever little gimmicks and supposedly their value went up. This was repeated several times. And when the underlying assets behind it all turned out to be illusion as the loans default and nobody gets anything back, it all gets wiped out in a cascade. It's a pyramid scheme.

So the central bank can only print more money to increase supply and that increases inflation.

The central bank doesn't print money; that's the Treasury Department. The Central Bank's mechanism is much more powerful--it creates and destroys liquidity at will. Do you not understand how the very existence of a central bank makes it nonsensical for some other force to destroy liquidity in the financial markets?

Doesn't matter whether people have jobs or not if they can't afford basic necessities because the prices go up due to inflation. Too many people in the US already struggle to make ends meet.

What are you talking about? I suppose that is hypothetically true, but:
1. It's not happening right now.
2. It's not what we're talking about. Unemployment is required for stagflation to exist--the point of the entire thread. J has argued that stagflation will continue and worsen significantly, which requires that unemployment rise significantly. So it does matter whether people have jobs or not. It similarly matters if there's inflation or not, since stagflation requires both.

If they are already using debt to fuel spending, the whole economy is running on fumes already and those fumes are now running out. When the gas tank is empty, the car just won't fucking move. This is not difficult to understand.

But that state has existed for a long time. Why is it that, suddenly, we have reached such a debt level that consumers are no longer capable of buying things at all? Why would we not expect to see some gradual process as SOME consumers reach such a high debt level, and others only do so a few months or years down the road?

Are you familiar with the concept of synergy, that the whole can be greater than the sum of its parts? This exactly what you have there, and in this case the synergy effect comes from a combination of factors that are bad enouh on their own.

Prove it. Show that the magnitude of all of these things that J has listed will have a greater impact on STAGFLATION in combination than the sum of their parts. Incidentally, note that many of the effects of these factors will individually run COUNTER to stagflation.

If you are so fundamentally incapable of understanding this discussion and its basics, so incapable of seeing the big picture instead of just single pieces of the puzzle, you should just stay out of the discussion until you get well enough acquainted with the real world to be able to competently assess it.

Blow me, Edi. You are obviously unfamiliar with the whole concept of economic evaluation, but more fundamentally you have failed to apply basic logic to this discussion: you don't even recognize who has the burden of proof!

[Master of Ossus]

Yet you yourself have stated that the accuracy of the method used to determine that figure is not known...

No, I haven't. It is accurate to within +/-.2 percent at the 95% certainty level for any given month. Over a longer timeframe its accuracy improves proportionally.

I was speaking in a general and not a specific sense, dickhead. Please try to keep up. Even if we don't have that condition now it has occured in the past and it is conceivable to normal human beings that this situation could occur again in the future.

I'm not interested in what is conceivable or theoretically possible or not. We're talking about what's going on right now, and in the immediate and foreseeable future--which is the entire point of the thread. Thanks for trolling.

It could. As I said, there is historical precedence for sudden upsets and economic collapses. It's sort of the flip side of the economic boom/gold rush effects.

Indeed? Where is the historical precedent for the mechanism that you describe? Moreover, if it's not happening right now, then why are we even discussing it? The whole point of this thread is whether stagflation is occurring RIGHT NOW, and whether it will worsen in the immediate future. We're not particularly interested in things that are theoretically possible in some arbitrary future point.

I get what you're saying - you're saying that the accuracy of the measure doesn't matter as long as the measuring tool is consistent. Why don't you understand that for many, if not most, areas of human endeavor that is NOT an acceptable standard?

Bullshit. I'm saying that the measure is consistent and accurate for what it purports to measure. That is a perfectly acceptable standard when consistency is important, too. I'm sick of your strawmen.

You also stated that the method used to measure it was consistent but of unkown accuracy. Again, how can that be justified?

Nowhere have I said that it is of unknown accuracy. It is accurate to within +/-.2 percentage points at the 95% confidence level, as was clearly stated in the link that I gave you earlier. It is clear that dishonesty does not phase you, provided that it allows you to continue on your soapbox tirade about unemployment in this country. Studies designed to look for nonsampling error within the CPS have consistently found that its impact on month-to-month is negligible (see previous link).

If changing the system would give us more accurate information then yes, we should change it.

But it wouldn't be more accurate. It would just be different.

Here's the problem, Broomstick: separate the following people into "employed" and "unemployed" categories:
1. A graduate student who is paid a salary of $10,000/year for working in a professor's laboratory. The work actually takes her over 80 hours/week.
2. A 16-year-old high school student who has been trained to work at McDonald's, has never been fired, and who is on good terms with the manager. At any time, he could simply call the manager to request hours, and he knows that the manager would happily give him anywhere from 5-30 hours per week, but hasn't felt like working and so he has not gone back in the last six weeks.
3. A 30-year-old farmer who lives on his family farm. Last week, he worked 45 hours there, but was not paid any wages for his work.
4. A 28-year-old artist who owns and runs her own studio. Last week, she finished a sculpture that she had been working on for some time, and also worked on some other pieces, but she did not sell anything. Next month, she has an important showing scheduled with major art critics, for which she is busily preparing.
5. A 40-year-old father of three who was recently laid off from General Motors. He got a job last week for a company that repaired refrigerator cars, but quit almost immediately because he decided that conditions in the garage were unsafe. He still receives unemployment benefits from GM which should help him get by, but is concerned that the automobile industry where he lives is contracting, and is considering moving to another state where a rival company has set up an auto factory. In the meantime, he's odd-jobs for neighbors, who pay him around $10/hour. He expects to be able to get 5-10 hours of work, next week, in this way.

Even if you can make judgments on each of these, can you see how reasonable people might make different calls as to whether or not each person is employed? Personally, I would say that the HS student who could work at McDonald's is employed. The BLS says that he's unemployed. I'd say that the 40-year-old who just quit from the auto garage is unemployed. The BLS says he's employed. Reasonable people can go either way on some of the other ones, too. I'm not going to say that they're wrong, nor am I going to say that going with my definitions instead of the ones that the BLS uses would improve accuracy--it wouldn't--it would just measure something different than what is being measured, now.

If you can't define the accuracy of what you're measuring perhaps you should change the definition, then.

Accuracy IS defined by what you're measuring--there's just an issue as to what people are measuring.

What good is measuring X - however it is defined - if we don't know how accurate that measurement is? With something as multi-factorial as the economy how can you know whether or not you're overlooking something if you don't even know how accurate your measurements are?

Because there have been multiple studies done that have shown that the BLS' measure is accurate to the point where only the sampling error matters, and the non-sampling error is trivial for all of the key metrics on a national scale.

You really don't get it, do you? What you state as obvious is voodoo to the average reader, where it doesn't just flat-out defy common sense and real-world observation.

[snip]

It wasn't clear - that's what I'm asking you questions. I fully admit I am not an authority on either statistics or economics, that's WHY I ask questions in these sorts of threads. You, on the other hand, seem to expect everyone to be an expert in these matters, expert apparently being defined as "agrees with everything MoO says without question".

You came into the thread claiming that the CPS measured unemployment in a way that it does not, thus indicating that you felt you had some knowledge which evidently you did not have (since you were wrong). I don't see why I can't expect some level of knowledge from someone who claimed very forcefully something that was wrong.

But you stated that the accuracy of the method wasn't certain. The article you linked to also stated that some authorities questioned the accuracy of the method - how can it be "extremely accurate" AND the accuracy be questioned at the same time? That doesn't make sense.

It goes back to the definition problem. Statistically, the CPS is VERY precise, but of accuracy that is arbitrary depending on whether you agree with its definitions or not.

[Alan Bolte]

Can you provide evidence to the effect that banks are so terrified of the credit crunch that they have stopped lending AT ALL to people, as opposed to merely tightening the purse strings? I see no theoretical or empirical reason to accept the idea that banks will no longer respond at all to interest rates highly dubious, since we can observe that banks are still in the business of credit. Moreover, you ARE the one who bears the burden of proof if you're suggesting that an observed historical trend will not continue.

I didn't mean to imply that banks are not lending at all, that's silly. The Fed has stated that it's disappointed by the response to the last two interest rate drops, and I'm left wondering why dropping the rate further at this point would be any more effective. I was proposing a hypothetical mechanism by which the extent that banks are "tightening the purse strings" is exceeding the extent that the Fed is loosening theirs, which is the only reason I can see for the Fed to be disappointed by the effects of their rate cut. I can't seem to locate the article on businesses that are suddenly having trouble getting loans, whether in an absolute sense or because the cost of borrowing has increased. I'll have to get back to you on that.

Unlike the rest of the people here, I'm not entirely convinced about inflation, at least not in the near term. It's true commodities prices are on the rise, but fear of inflation tends to drive money into commodities, and there seems to be plenty of fear about inflation, probably caused in part by the rate cuts. Thus, the present commodities prices could be driven primarily by speculation. Commodities aren't the only price that matters, though I'm sure it seems that way to the some right about now.

If the money supply is shrinking, and you're pumping money into the supply, then either you're blunting deflation or you're successfully countering it. With housing prices falling, we've clearly got some sort of deflationary pressure. Whether we're experiencing overall deflation of the money supply means you have to ask what is money? People who were treating their home equity lines of credit like cash in an account don't have nearly as much to spend anymore. Investors that were trading certain kinds of paper like cash are finding the market suddenly illiquid. Was that money? Is that money gone? When is credit not money? This is where my already tenuous grasp on the situation breaks down completely.

What I'm hearing from Ossus is that there are no limits (legal or practical) on the rate at which the Fed can add or remove liquidity from the economy, and that if you can just keep the money supply on track the economy will always be just fine, regardless of other factors. Both of which I find hard to believe. I guess it depends on your definition of "just fine." Am I understanding you correctly, sir?