Pay attention to pundits who make patently obvious predictions; ignore those who make brilliant predictions
This post was triggered by a comment in a recent Tyler Cowen post:
Finally, Rajan is a case for testing Krugman’s oft-stated view that we should listen most seriously to those who have made good predictions in the past. Rajan was probably the best, more accurate, most serious, most detailed, and most non-Chicken Little predictor of the financial crisis. You might think that means he gets listened to today, or given the benefit of the doubt on interpretation, but apparently not.
That was a brilliant set of predictions by Rajan, and thus it can’t be compared to the sort of thing Krugman was talking about. Both market monetarists and Krugmanites have been consistently right about the economy since late 2008. We’ve pointed out that US interest rates are likely to stay low, despite the huge budget deficits. We’ve pointed out that inflation is likely to stay low, despite predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb.
In my view, these predictions were patently obvious, requiring no special insight. The bond market told us interest rates would stay low. The TIPS markets signaled that inflation would remain low. Even a bright high school student could make these sorts of predictions.
In contrast, Rajan made a brilliant prediction, the sort of prediction that earned John Paulson billions of dollars. That’s why everyone should take Krugman and we market monetarists seriously, and no one should pay attention the Rajan.
This should be obvious to my regular readers. But just in case anyone is a bit slow, let me fill in the details:
1. Brilliant predictions of major crises are just luck, according to the EMH.
2. Even if they aren’t luck, they have no policy implications. The government regulators responsible for controlling bubbles won’t have the skill of a Rajan or a Paulson, otherwise they’d be running a hedge fund or teaching at the UC Business School. (And that’s assuming these guys didn’t just get lucky. Given the investment record of Paulson, Roubini, et al, since the crisis, I’m inclined to believe the EMH, and attribute their 2008 success to luck. But again, even if it was skill, it’s an incredibly rare skill with no policy implications.)
3. In contrast we can and should expect policymakers to pay attention to problems that are patently obvious to the financial markets. The markets said that the Rajan’s of the world were wrong about monetary policy in 2010. So were the Austrians and the old-style monetarists. And so were the Keynesians who believed in liquidity traps, who believed QE2 would not and could not boost asset values. Who was right? The market monetarists were the most right, with the Krugmanite Keynesians coming in second. We saw the need for more monetary stimulus in 2009 and 2010, when others were skeptical. Our predictions had important policy implications—they told policymakers that NGDP growth was likely to be suboptimal over the next few years. And we were right.
In fairness to Rajan, I imagine he based his prediction on a sound model of what was wrong with the US financial system. And he was probably right. Let’s suppose he argued there was too much risk-taking. That would be correct—our regulatory system biases banks toward taking excessive risks. But he wouldn’t have been right because he correctly predicted an opportunity for John Paulson to make billions of dollars, that part was luck. Rather he was right because sound economic principles point to our financial system being riddled with moral hazard. He would have been just as right about excessive risk if there’d been no crisis. In contrast, market monetarists made correct predictions based not on luck, but market forecasts. This technique won’t always work, but on average we’ll continue to be right far more often than our opponents.
PS. In 2000 a small increase in interest rates drove Japan right back into deflation. In 2010 Rajan proposed a 200 basis point increase in the Fed’s target at a time when core inflation had fallen to 0.6%. If his policy advice had been followed we would now be in Great Depression II.
PPS. I certainly agree with Tyler that Rajan should be given the benefit of the doubt when being interpreted. As I explained in my previous post, I have a much more nuanced take on his views than many of his critics.
Tags:
10. May 2012 at 06:22
“That was a brilliant set of predictions by Rajan, and thus it can’t be compared to the sort of thing Krugman was talking about. Both market monetarists and Krugmanites have been consistently right about the economy since late 2008. We’ve pointed out that US interest rates are likely to stay low, despite the huge budget deficits. We’ve pointed out that inflation is likely to stay low, despite predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb.”
I disagree with your take on “brilliant predictions”. The problem with “brilliant predictions” is that those brilliant predictors are, by definition, well ahead of the crowd. These predictions are considered by conventional pundits wrong right up until the time those predictions come true. It is only an issue of timing and I would consider it a virtue if Rajan is seeing further down the road than others. And, I’m not aware that Rajan has made any definitive statements on *when* interest rates will rise or *when* inflation might take off. If Rajan’s predictions come true (again) most of his detractors will likely come up with alternative reasons to explain why those things happened. I suppose we will hear something like “a broken clock is right at least two times each day”.
As far as interest rates are concerned, do we currently have a “market” that the EMH might apply to?
10. May 2012 at 06:54
I have to laugh a little that the implication here is that even when your opponents are right, they’re wrong.
10. May 2012 at 06:56
“Even if they aren’t luck, they have no policy implications. The government regulators responsible for controlling bubbles won’t have the skill of a Rajan or a Paulson … ”
I know you agree with this, but it bears repeating that controlling bubbles is not even a proper role for regulators. Bubbles aren’t really a problem for the economy, they are only a problem for the people who lost money on them.
As far as the government is concerned, it should only care about bubbles to the extent they affect the overall economy, or the extent government backed accounts are invested in them.
Obviously, NGDP targeting takes care of the first. The second could be alleviated by tighter regulation of FICA/Fed backed accounts, or completely taken care of by full reserve accounts.
10. May 2012 at 07:01
IMO, it was patently obvious back in ’07 that hard money monetary policy (on top of existing mortgage problems) would produce a financial crisis.
Rajan is taking credit for predicting the outcome, when he was really predicting an unknowable conditionality. All forecasts should be conditional, and we’d have a better sense of who’s smart if we took note of that.
10. May 2012 at 07:13
I am sure Rajan is a smart guy but you know some economists both made the prediction that some banks would get into trouble because of moral hazard (liar loans) and also have been pretty close to the MM and PK camp on the need/effectiveness of stimulus (for example). The FDIC manual does have a chapter on Continental Illinois and Penn Square which reads like Lehman and WAMU.
one brilliant play does not get you into the hall of Fame.
which is a long-winded agreement.
10. May 2012 at 07:18
Suppose that everyone in the world was convinced that the EMH is right, would the EMH still apply given that no one would make the attempt to bid prices up or down from their prior level, thinking that any knowledge of potential mispricings must already be priced in? If the EMH, under a broad interpretation, is right, wouldn’t this point also more or less be reached “efficiently” so we should expect it to already be the case? How is this not a paradox?
Also: can the market forecast anything efficiently except bad monetary policy? Was the stock market not “caught by surprise” by monetary policy failure in 2008? What would account for this exception? Is your view that the market was right from an ex-ante stochastic, probabilistic point of view (i.e. its “bet” more highly weighted the more likely outcome, but the less likely outcome occurred)?
Similar question about using TIPS spreads as an indication of inflation expectations: if the fed commits to targeting a 2% inflation rate, why would anyone bid the spread away from 2%? In the longer run, to bet against the fed is a guaranteed loss no matter how fanatically the fed has to loosen/tighten to get to it’s target. This is especially true if the fed monitors the TIPS spread to see how well it is doing. Knowing this, is there really any information in the TIPS spread other than “the fed targets a 2% CPI inflation rate”?
10. May 2012 at 07:32
This sin’t about investing or EMH.
This is about POLITICAL ECONOMY.
And the kind of people DeKrugman likes are going to get LESS. Period. The End.
“Rather he was right because sound economic principles point to our financial system being riddled with moral hazard.”
This is the lede.
SINCE our financial system being riddled with moral hazard…
SINCE of Fiscal situation is untenable….
The Fed has a moral obligation to reward sound fiscal policy, and the correct amount of regulation.
Scott, this is very similar to the Saudis talking about oil supply when the prices get run up.
They say, “there is plenty of oil” and they mean it.
Monetary policy is not a democratic institution. If you ASSUME it is, you will drive yourself nuts.
—–
WE ARE PAST THIS.
Scott has ADMITTED he is willing to do 4.5% NGDPLT starting today with NO MAKEUP.
That is the headline!!!!!!
That is a brand new assertion one that gets past all this noise, it assumes that the Fed will FORCE the Federal Govt. into AUSTERITY…
BUT!!!
Scott argues, that EVEN without make-up inflation, NGDPLT will keep our economy growing, we will NOT SUFFER from Federal cuts in spending under Scott’s program.
Scott, will confirm my statements as true. He may prefer make up, but his plan works without it, and AUSTERITY in GOVT. SPENDING will have no effect negative effect.
10. May 2012 at 07:33
“In 2000 a small increase in interest rates drove Japan right back into deflation.”
Really? Don’t you think the dotcom bust and its effect on one of the more tech-heavy economies might have had something to do with it?
10. May 2012 at 07:40
When one Frenchman predicts the movie will be bad it’s false to say that all Frenchmen said that the movie would be bad.
The fact that the Fed is both paying interest on reserves & and holding interest rates at zero is well enough understood by “Austrians”.
What happens down the road will depend on many particular things that happen on down the road, including Fed action.
We haven’t seen the end game yet.
Scott writes,
“We’ve pointed out that inflation is likely to stay low, despite predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb.”
10. May 2012 at 07:43
The EMH is false, so you can’t make this inference.
“Brilliant predictions of major crises are just luck, according to the EMH.
10. May 2012 at 07:45
The only true versions of the EMH tell us NOTHING about structural relations of major crises, as even Fama would tell you …
10. May 2012 at 07:47
Sumner comes up with _another_ great argument for ending the Fed:
“Even if they aren’t luck, they have no policy implications. The government regulators responsible for controlling bubbles won’t have the skill of a Rajan or a Paulson, otherwise they’d be running a hedge fund or teaching at the UC Business School.”
Unfortunately, it’s false, because its based on a false understanding of human beings — a typical economist’s fallacy.
10. May 2012 at 07:50
BE HONEST, Scott.
There were Austrians who followed Bagehot’s Rule and advocated monetary policy to counter-act the secondary deflation, see the remarks of Horwitz, Selgin and others, i.e. the “Austrians” with actual academic positions.
All of these dishonesty fouls are fouling up your effort to make a sincere and honest argument.
10. May 2012 at 07:53
If you rig the game with a series of falsehoods, of course you are going to win the game, Scott.
But you will also need to end the game of telling us how “sincere” you are ….
10. May 2012 at 08:03
We’ve pointed out that inflation is likely to stay low
Except inflation isn’t “low”. Inflation is “high.”
Price inflation:
If we use 1990 calculation methodology, inflation is at 6%.
If we use 1980 calculation methodology, inflation is at 10%.
Money supply inflation:
M2 is currently growing at 9% annualized.
despite predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb.
Except all that money printing was indeed an inflationary time bomb. Austrians don’t make time based predictions. You can’t say they were wrong about price inflation, because they didn’t say anything like “By 2012, there will be X% price inflation.”
If you heard any Austrians say “hyperinflation”, you have to understand that there is no time associated with this, because the Austrians are deducing and dependent on human choice, which Austrians hold cannot be predicted a priori.
In other words, the fuse of the inflationary time bomb never stopped burning.
I know you WANT to believe so badly that NGDP framework schools are justified in concluding the Austrians were wrong (despite Austrians never agreeing to the year 2012 as the finish line) so that you can say “Look everyone! The NGDP framework has the best predictive power!” but you can’t win by not even correctly identifying what the Austrian school says.
I defy you to show how the Austrian school economists showed, via deduction, that we will have hyperinflation by 2012.
The bond market told us interest rates would stay low. The TIPS markets signaled that inflation would remain low.
Like the MBS derivatives market told us in 2006 that home default rates would stay low?
Bond market investors are not omniscient. You’re just using the bond markets in an ex post hoc ergo propter hoc way on the basis of claiming that 2012 is the finish line year. If you made the prediction in 2006 that home default rates would stay low, and let’s say they did stay low, then you would have been here saying anyone could have predicted this, by simply looking at MBS market prices.
In short, you’re falsely setting up the finish line to have already been passed, as if Austrians even agreed to that finish line.
Like always, Austrians will be proven right once again as history unfolds, and market monetarists will be proven wrong, just like Austrians proved the regular monetarists wrong, and the Keynesians wrong.
10. May 2012 at 08:24
Really good blogging.
So far, I have witnessed about seven hyperinflations in the USA, as predicted by various money fetishists, gold nuts, currency kooks and old-school monetarists.
Funny thing about the seven hyperinflations—seems like prices budged up only very slowly. Seems like unit labor costs actually declined in the latest hyperinflation. Seems like commercial space is renting for less than it did five years ago. Seems like I can get more and more at various $1 stores.
Hyperinflation—what a bunch of dunces inhabit our real estate investment circles. In a hyperinflation, you leverage up, buy property, watch your equity explode, and then sell. Money in the bank. Instead, property is becoming worth less in nominal dollars, even in a hyperinflation. What gives?
True story. I just bought a dozen large eggs at a 99 cent store. In 1952, Ralph Kramden (Jackie Gleason) complained that eggs in Brooklyn were 79 cents. Honeymooners. What a great show.
10. May 2012 at 08:24
Good post Scott.
BTW major freedom.
“If we use 1990 calculation methodology, inflation is at 6%.
If we use 1980 calculation methodology, inflation is at 10%.”
This again?! So you argue that either we are radically underestimating nominal gdp growth or that real gdp is contracting at between 3 and 7% a year? And you argue that even though unemployment is going sideways not exploding?
You know that the things we consume in 2012 are very different to the things we consumed in 1980 yeah? Do you think that might influence the figures you’re using?
You’re “any accelerating inflation at some point in the future proves my gang right” schtick is quite cute though. Awwww.
10. May 2012 at 08:29
I have a brand-new calculation method for inflation. It’s whatever I say it is. So there.
10. May 2012 at 08:33
Left Outside:
This again?! So you argue that either we are radically underestimating nominal gdp growth or that real gdp is contracting at between 3 and 7% a year?
Saying something about price inflation is different from saying something about nominal GDP, nor real GDP.
Is real GDP contracting? I would say it’s at least stagnating, if we use real wages as a proxy.
You know that the things we consume in 2012 are very different to the things we consumed in 1980 yeah?
You misunderstand. The basket changed yes, but the key is that so has the methodology. That the basket changes over time does not mean there has to be methodology changes over time. If it did mean that, we’d have to change methodology every day, indeed every hour, which would make historical comparisons impossible.
Do you think that might influence the figures you’re using?
No. Basket content and inflation calculation methodology are two different things.
You’re “any accelerating inflation at some point in the future proves my gang right” schtick is quite cute though.
I don’t know what you mean by “schtick.” That is what the Austrians are actually saying, not what Sumner is implying they are saying by declaring 2012 as the finish line year.
10. May 2012 at 08:34
So far, I have witnessed about seven hyperinflations in the USA, as predicted by various money fetishists, gold nuts, currency kooks and old-school monetarists.
yeah, where’s the hyperinflation:
http://research.stlouisfed.org/fred2/graph/?g=78p
maybe hypo-inflation??
find an old sears catalog and see what you would want to buy out of it at yesterdays prices (in many cases the prices are the same or the quality has improved; a car is not much more expensive than in 1988 because while gas has gone up in $$, gas mileage has also gone up).
10. May 2012 at 08:49
don’t be stupid, of course the basket changed. you want to buy those 1980s polyester pants or Commodore 64s? You still have an Atari. my phone has 10000x more memory than my first computer for 1/20 of the $$ cost.
Your statement totally makes sense if you want to revert back to the 14th century, which from other statements you have made i strongly suspect would be preferable to you.
10. May 2012 at 08:50
dwb:
yeah, where’s the hyperinflation
Yeah, where’s the national home price decline?
Oh, right, there it is.
See what I did there?
10. May 2012 at 08:50
I got a haircut last week for $9.99. I remember getting a haircut in 1989 for $8.99 – a whopping 11.1% hyperinflation over 23 years. 11.1 divided by 23 is 0.48% per year. Wait, it’s even less than that because of compounding. After plugging it into a compound interest calculator I see it’s 0.46% inflation per year.
At this rate, we’ll all have to bring wheelbarrows of cash to the barber!
10. May 2012 at 08:52
To be fair, I have less hair now, so perhaps the value of the haircut has gone down. I think they charge the same for everyone though.
10. May 2012 at 08:53
dwb:
don’t be stupid
Don’t be an imbecile?
of course the basket changed.
I didn’t bring up that point. Left Outside did.
you want to buy those 1980s polyester pants or Commodore 64s? You still have an Atari. my phone has 10000x more memory than my first computer for 1/20 of the $$ cost.
You too are conflating basket content with inflation calculation methodology. How surprising.
Your statement totally makes sense if you want to revert back to the 14th century, which from other statements you have made i strongly suspect would be preferable to you.
I am not saying revert to the same basket. I am not even calling for a reverting to 1990 or 1980 inflation calculation methodology. I am just saying that if the BLS used the same methodology for calculating inflation today, as they did in 1990, then reported inflation would be HIGHER today.
10. May 2012 at 08:56
its all about relative prices and the cost of near substitutes, and productivity. You cant just look at one price. you have to look at substitutes and the total cost. Car is a great example: 1988 gas was $1, today its $4, but car gas mileage has gone up substantially so that the total 5-year cost of ownership has not gone up much.
http://www.slate.com/blogs/moneybox/2012/05/08/stop_what_you_re_doing_and_go_buy_a_house.html
10. May 2012 at 08:57
Negation of Ideology:
Inflation measures aggregate prices, not just the prices of haircuts.
And personal anecdotes are not evidence. In my personal experience, a haircut in 1990 where I live cost about $6.00. Now in that same location, where my Dad still goes, it costs $20.00. That’s $14.00 inflation on $6.00, or 233% rate of inflation.
There are also personal computers that used to cost $5000. Now they are $500, so huge deflation there.
The point is that single examples and single personal experiences are not how to calculation price inflation.
10. May 2012 at 09:00
“We’ve pointed out that inflation is likely to stay low, despite predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb.”
You cannot say that. Or at least, you have to say whether or not you are talking about short or long-term. I don’t think Austrians or monetarists have ever precisely defined the time it would take for new money to generate inflation, especially during a lengthy crisis that never ends. They just say it would happen….at some point.
Banks’ reserves at the central banks have exploded in the last few years (+7411% between 2008 and 2011 in the US). Those are excess reserves, maintained by the banks to contain any refinancing or liquidity crisis that could arise given the current highly uncertain economic environment.
The money has just not been “released” yet in the real economy. So the effects are close to nothing at the moment.
This does not mean that the reserves won’t be released (and multiplied for some of them) through credit generation by the banks once they have more confidence in the future.
10. May 2012 at 09:00
dwb:
its all about relative prices and the cost of near substitutes, and productivity. You cant just look at one price. you have to look at substitutes and the total cost. Car is a great example: 1988 gas was $1, today its $4, but car gas mileage has gone up substantially so that the total 5-year cost of ownership has not gone up much.
It’s the hedonic corrections that have led to the BLS reporting no price inflation if in 1990 steak cost $5.00 a pound while hamburger cost $1.00 a pound, if in 2010 steak costs $15.00 a pound while hamburger costs $5.00 a pound, because the BLS will say “You can eat hamburger instead of steak, and it will cost you the same, hence no price inflation.”
They really do this. They say if steak gets too expensive for people to buy that they resort to lower priced hamburger, that people are not paying higher prices.
10. May 2012 at 09:00
great, here is the price index of what people are actually spending money on:
http://research.stlouisfed.org/fred2/graph/?g=78x
nope still no hyperinflation!
10. May 2012 at 09:10
Let’s add to that the billion price index, annual ~ 2%: http://bpp.mit.edu/usa/
10. May 2012 at 09:19
My understanding of EMH is weak, so correct me if I’m wrong, but isn’t it possible to make brilliant predictions of financial crises without “getting lucky” based on information that is not publicly available, and still satisfy the semi-strong EMH? And isn’t “superior understanding of economics” potentially an example of such information?
“But he wouldn’t have been right because he correctly predicted an opportunity for John Paulson to make billions of dollars, that part was luck. Rather he was right because sound economic principles point to our financial system being riddled with moral hazard. He would have been just as right about excessive risk if there’d been no crisis.”
I don’t understand the distinction that Scott is making here. Wouldn’t excessive risks by definition lead to a crisis eventually? Perhaps the difference is that Rajan couldn’t necessarily make money off his prediction, as he couldn’t say exactly when the crisis would happen based on fundamental analysis alone, and so couldn’t safely short the housing bouble. Or could he? What exactly was Rajan saying in the lead up to the crisis?
10. May 2012 at 09:21
yes, i know about the changes.
according to the Economist Big mac index, the US cost of a big mac has increased ~2.3% since 1988.
Still no hyperinflation!
10. May 2012 at 09:22
dwb:
great, here is the price index of what people are actually spending money on:
http://research.stlouisfed.org/fred2/graph/?g=78x
nope still no hyperinflation!
Austrian economists have not predicted hyperinflation by 2012.
I know it makes you mad, because you feel like Austrians “aren’t playing by the rules” of making time based predictions, so it appears as if they are sneakily trying to get away with being able to say anything they want and never be wrong, but that isn’t the motivation for why Austrian economics doesn’t make time based predictions. It’s because the epistemological foundations of Austrian economics prevents such a thing from being possible. It’s strictly a result of the “rules” of Austrianism.
Austrianism isn’t an empirical prediction based school. It is a deduction school akin to mathematics.
10. May 2012 at 09:25
dwb:
yes, i know about the changes.
Then why did you conflate basket content with calculation methodology? Maybe you forgot?
according to the Economist Big mac index, the US cost of a big mac has increased ~2.3% since 1988.
Still no hyperinflation!
It’s obvious there is no hyperinflation now. You’re not saying anything significant against the Austrians. The Austrian economists didn’t say there would be hyperinflation by 2012.
Continually pointing out goods that have increased in price only modestly, is like preaching to the choir.
10. May 2012 at 09:33
you said:
Price inflation:
If we use 1990 calculation methodology, inflation is at 6%.
If we use 1980 calculation methodology, inflation is at 10%
There is no deductive reasoning in Austrianism. Lorenzo had a good post a while back, all it holds dear is an axiom. Science demands testable hypothesis and predictions, thats what differentiates it from theology.
10. May 2012 at 09:37
dwb,
If the Sears catalogue for 1950 has a Rolex Oyster watch, I’ll take it, even though the BLS would probably tell me that watches deflated by a zillion percent since then!
10. May 2012 at 09:43
Major – Two points:
1. “The point is that single examples and single personal experiences are not how to calculation price inflation.”
I agree. But when you claim that the statistics are wrong and inflation is really high, all you have left is anecdotal evidence. The MIT project and the BLS calculation say inflation in low. It’s probably even lower than the official statistics because of quality improvements. And it correlates with my personal observations. Some things, like health care and gas have gone up a lot; many have stayed about the same; some have gone down. Hair cuts are a good example because there’s no quality improvement or substitution effect.
Should I believe some internet Austrian or my own lying eyes?
2. Your Dad is getting ripped off.
10. May 2012 at 09:45
dwb:
you said:
Price inflation:
If we use 1990 calculation methodology, inflation is at 6%.
If we use 1980 calculation methodology, inflation is at 10%
Your point?
There is no deductive reasoning in Austrianism.
Yes, there is.
Lorenzo had a good post a while back, all it holds dear is an axiom.
Lorenzo is also clueless. No, it’s not true that “all” Austrianism holds dear is an axiom.
The axiom is the praxeological aspect of Austrianism. But it’s not exhaustive. The interpretation of history, i.e. past observable data, is the thymological aspect of Austrianism. As R. Long notes: “Praxeology without thymology is empty, and thymology without praxeology is blind.”
Lorenzo is probably the last person in the world anyone would want to know what Austrianism is about. He has never read an original Austrian treatise, and his own economics are full of errors and fallacies.
Science demands testable hypothesis and predictions, thats what differentiates it from theology.
No, that’s what positivism demands. “Science” INCLUDES mathematics, and yet mathematics doesn’t make testable predictions or hypotheses.
Would you call mathematics “theology”? No, but you WANT to call Austrianism theology so that you can believe to have refuted it.
Weak.
10. May 2012 at 09:51
Negation of Ideology:
1. “The point is that single examples and single personal experiences are not how to calculation price inflation.”
I agree. But when you claim that the statistics are wrong and inflation is really high, all you have left is anecdotal evidence.
Actually no. I am not saying “the statistics” are wrong, I am saying the particular statistics you are looking at are lower than they would be if 1990 calculation methods were used. That’s it. This doesn’t require anecdotal evidence.
The MIT project and the BLS calculation say inflation in low.
Based on a particular calculation methodology, yes.
It’s probably even lower than the official statistics because of quality improvements.
The BLS already takes into account quality improvements.
And it correlates with my personal observations.
It doesn’t correlate with my personal observations.
Some things, like health care and gas have gone up a lot; many have stayed about the same; some have gone down. Hair cuts are a good example because there’s no quality improvement or substitution effect.
Hair cuts aren’t the only things with prices. Again you are cherry picking data and insisting that personal anecdotes are evidence.
Should I believe some internet Austrian or my own lying eyes?
You don’t have to “believe” anyone. And the choice isn’t between believing me and your personal anecdotes. It’s between actual prices, and reported prices. They are two different things.
2. Your Dad is getting ripped off.
Prices don’t lie. Inflation calculations do not discriminate among “ripped off prices” and “non ripped off prices.” My Dad isn’t getting ripped off because he values the haircut more than the $20. He’s getting a good deal.
10. May 2012 at 10:02
@rebeleconomist,
fair enough about the Rolex. i did find a “self winding” watch in the 1952 catalog.
there is a toaster oven in i think the 1969 catalog that is virtually the same as the.one we have now, it was the same price when we got it a few years back.
oh and none of this adjusts for wages. a more appropriate index is the number of hours i have to work and on that measure the standard of living has improved enormously.
you can see some catalogs at:
http://www.wishbookweb.com/
10. May 2012 at 10:10
dwb:
1988 gas was $1, today its $4, but car gas mileage has gone up substantially
That “but” is irrelevant.
10. May 2012 at 10:34
Unless you are in the Molotov Cocktail business, the only reason you buy gasoline is to put in an engine, so that “but” is absolutely relevant – its a package deal, so if energy efficiency goes up you use less gas.
10. May 2012 at 10:43
Scott,
The discussion about Rajan’s essay has frightened me severely, in that I took the essay to also mean things which it clearly did not. As one of the marginalized, when I hear discussion about bloated finance sectors, bloated residential construction and more productive work, here’s what a ‘dimwit’ like me thinks:
1) simpler financing and housing options on the way for people who don’t want or need McMansions, just little respectable spaces.
2) opportunities to rethink the work we really want and need.
Instead, the discussion was really intended, by many, to nip in the bud any further ideas of growth. That means in spite of all the bickering and arguing, ‘opposing’ teams are actually moving closer to the table in the closed room, to divide up what shrinking pie there still is. One team wants to think social security and medicare will last a little longer, the other team tells them ‘sweet dreams’ now let’s get this settled.
And the marginalized people still stand outside the door of that closed room, waiting for a chance to be let in and offer suggestions, for after all they wanted to be a part of economic life and opportunity. They wanted to help find ways that it would not get to the point of a shrinking pie in the first place. Why would anyone want to think we can wait a year or two before social security and Medicare disappear before finding solutions for them?? But the other side just wants to tell them, ‘sweet dreams’ let sign and get this over with.
So I have to ask the powers that be, please please please don’t sign those papers yet. Don’t let things get so bad that there is no other choice but to sign, when millions of people are begging to be a part of the solution.
10. May 2012 at 10:46
also, to your hamburger: i looked up the price of live cattle in 1988 vs 2012 and it was ~58 vs 107, about 2.5% pa. feel that hyperinflation!
10. May 2012 at 11:12
also, to your hamburger: i looked up the price of live cattle in 1988 vs 2012 and it was ~58 vs 107, about 2.5% pa. feel that hyperinflation!
1. Nice cherry picking of start and end year.
2. There is no hyperinflation.
3. From 2000 to today, cattle prices went from 70 to 116, which is 4.3% annualized.
4. From 2010 to today, cattle prices went from 80 to 116, which is 20.4% annualized.
4. Money inflation is accelerating.
10. May 2012 at 11:12
also, to your hamburger: i looked up the price of live cattle in 1988 vs 2012 and it was ~58 vs 107, about 2.5% pa. feel that hyperinflation!
1. Nice cherry picking of start and end year.
2. There is no hyperinflation.
3. From 2000 to today, cattle prices went from 70 to 116, which is 4.3% annualized.
4. From 2010 to today, cattle prices went from 80 to 116, which is 20.4% annualized.
5. Money inflation is accelerating.
10. May 2012 at 11:16
dwb:
Unless you are in the Molotov Cocktail business, the only reason you buy gasoline is to put in an engine, so that “but” is absolutely relevant – its a package deal, so if energy efficiency goes up you use less gas.
Wrong. If the context is price inflation, then you cannot excuse price inflation by saying “Oh well, I just consume less of it anyway.”
For a major reason why cars today have better gas mileage is precisely because gasoline is so much more expensive. If it was as cheap as it was 20 years ago, then there would have been a far smaller incentive for car makers to increase the gas mileage of the cars they make. We might still be driving gas guzzlers, who knows.
10. May 2012 at 11:17
Sorry for the double post. I thought the first one didn’t get through after seeing no post after repeated screen refreshes.
10. May 2012 at 11:21
In other words, dwb, if there was no gasoline price inflation, and the capital that went into making cars more fuel efficient was instead used elsewhere in the economy to boost efficiency and productivity, then we would have been better off than we are now.
You are ignoring opportunity costs.
10. May 2012 at 11:27
I think Professor Cowen was just being tongue-in-cheek there.
He was referring to Krugman’s repeated claim that since he has been right the past few years then people should listen to him.
10. May 2012 at 11:29
i dont think i am cherry picking, commodites are volatile but the average is not accelerating.
http://research.stlouisfed.org/fred2/graph/?g=78T
yes, people have figured out how to do more with less. whether its cause or effect is not relevant. oil is a finite resource, so the fact that the total cost of ownership of a car has not changed much (plus, better pollution controls) is a sign of advancing standard of living.
the correct metric for all these numbers is really the average number of hours worked.
10. May 2012 at 11:31
The only hyperinflation I see is in the quantity of posts from Major_Freedom. You could apply hedonic quality adjustments, but that would only make this hyperinflation worse.
10. May 2012 at 11:42
dwb:
i dont think i am cherry picking, commodites are volatile but the average is not accelerating.
I think you are cherry picking because you’re including years of past data that pull down the yearly average price inflation, which is OK in principle, but during those earlier years the people you say are talking about hyperinflation, weren’t talking about hyperinflation then.
yes, people have figured out how to do more with less.
They wouldn’t have to do that if the prices were lower and their real incomes were otherwise higher.
whether its cause or effect is not relevant.
It is incredibly relevant, because if the cause of inflation is a redirection of scarce capital that softens the blow of those higher prices, then that means other things could not have been produced with that same capital.
You are ignoring opportunity costs. You are saying that higher prices of fuel accompanied by higher fuel efficiency means people are no worse off. That ignores what could have been produced with that capital had the prices not risen and not provided any incentive to soften the blow of it through increased fuel efficiency.
Increased fuel efficiency has a cost, you do know that, don’t you? It’s not automatic. It’s not infinite. If gasoline prices didn’t increase, on the basis of no money printing, then capital would have gone into other things instead of softening the blow of gasoline price inflation through investing in increased fuel efficiency.
oil is a finite resource, so the fact that the total cost of ownership of a car has not changed much (plus, better pollution controls) is a sign of advancing standard of living.
All resources are finite. That isn’t a proper justification.
Even if you can show the total costs of car ownership hasn’t changed much (I don’t think that’s right), there is still the increased costs that we pay now compared to what we otherwise would have paid had more capital gone into other things besides ensuring the price of cars is kept down to keep them affordable.
the correct metric for all these numbers is really the average number of hours worked.
Actually, the correct metric for all these numbers is really the productivity of labor, which can be inferred from real wages. And if we go by real wages, we find that real wages have stagnated since the early 1970s.
10. May 2012 at 11:42
Scott, just a thought – Wouldn’t someone’s criteria for “obviousness” probably depend a little bit on their analytical framework or ideology? You cited the “predictions by Austrians and old-style monetarists that all the money printing was an inflationary time bomb,” despite what you interpret to be obvious and straightforward evidence otherwise – “The bond market told us interest rates would stay low. The TIPS markets signaled that inflation would remain low.”
I’m not an economist, I’m a biomedical engineer, so I have no analytical framework to begin with (except for categorically rejecting that praxeology nonsense, in my framework it is “patently obvious” that applying deductive reasoning to the chaotic and complex human mind is a stupid endeavor).
What I’m asking is, if I find nothing particularly obvious, where do I start an analysis? (besides reading and trying to learn from your blog, that is. True story, attempting to understand your blog in the past few years has taken me from economic illiteracy to a passable ability to discuss current events with my peers. So, Thanks for that.)
10. May 2012 at 11:42
Steve:
The only hyperinflation I see is in the quantity of posts from Major_Freedom.
Burn!
You could apply hedonic quality adjustments, but that would only make this hyperinflation worse.
You mad?
10. May 2012 at 11:55
Burn.
10. May 2012 at 11:57
If all of the oil vanished from the world tomorrow, its price would become infinite and therefore MF would say we had infinity% inflation.
10. May 2012 at 12:00
I’m not an economist, I’m a biomedical engineer, so I have no analytical framework to begin with (except for categorically rejecting that praxeology nonsense, in my framework it is “patently obvious” that applying deductive reasoning to the chaotic and complex human mind is a stupid endeavor).
Is it even possible to create an analytical framework that avoids deductive logic?
10. May 2012 at 12:03
Alex:
If all of the oil vanished from the world tomorrow, its price would become infinite and therefore MF would say we had infinity% inflation.
No, those who are arguing against me on this board will be saying that, because I am the one saying don’t cherry pick data for single goods when talking about price inflation.
Plus I define inflation as rise in the money supply, so even if the price of oil skyrockets, without an increase in money supply it would not lead to general price inflation, since whatever additional money people spend on oil, would have to come at the expense of less spending elsewhere.
10. May 2012 at 12:12
Ryan:
Is it even possible to create an analytical framework that avoids deductive logic?
No. Those who reject deduction in economics are blind astrologers who have no sound theory to correctly interpret past data.
RJ:
I’m not an economist, I’m a biomedical engineer, so I have no analytical framework to begin with (except for categorically rejecting that praxeology nonsense, in my framework it is “patently obvious” that applying deductive reasoning to the chaotic and complex human mind is a stupid endeavor).
You just deduced, and you just verified praxeology, by saying that.
Every person on the planet who believed praxeology is nonsense, doesn’t understand praxeology. It is literally irrefutable. It would be like arguing the case that arguments don’t exist, or acting to prove that action does not exist.
The human mind is chaotic and complex? How would you be able to even know anything is “patently obvious” unless you were aware of some “simplistic” deductive thinking process to get to that conclusion?
Maybe you should read about something before you criticize it? Praxeology is just the study of logical constraints in human action. It’s not making any statements about the chemical properties of brains.
10. May 2012 at 12:19
MF:
No, those who are arguing against me on this board will be saying that, because I am the one saying don’t cherry pick data for single goods when talking about price inflation.
Except that your insistence on using the same basket of goods to measure CPI in 1990 as in 2012 would, in fact, result in you measuring infinity% inflation.
My example demonstrates precisely WHY the increased fuel efficiency of automobiles correctly mitigates the impact of rising oil prices on the CPI. If the methodology for calculating the CPI didn’t account for that, it would measure infinity% inflation if all the oil disappeared.
10. May 2012 at 12:33
Ryan, I think so, but I am of course open to counterexamples. As far as I understand it, deductive reasoning is the derivation of specific conclusions from general principles. I am a neuro guy, and I know it’s very difficult to extract any general principles about the brain. Two notes, and I am curious to hear what you think –
1) A spanish pathologist in the early 1900s established the structural properties of the brain, collectively known as the “Neuron Doctrine,” which describes the individual components (neurons, axons, synapses) and how they function, and for many years this was the fundamental knowledge from which all other analyses started. But we’ve been finding out that many of these general assumptions aren’t true – for example, a) neurons can synchronize and work in groups, violating the “the single neuron is the computational unit” part of the doctrine, and b) there has been some revision of how we think about synapses, both in the forms they can take and the types of transmitters that pass through them.
2) So the “top-down” approach described above can sometimes fail, because the initial premises were wrong. A lot of neuroscience actually takes the opposite approach. Right now I’m doing dynamical systems analysis of neural networks, which involves abstracting the properties of one neuron to a whole group of neurons to see what behavior emerges, which is about as inductive as it gets. And this sort of thing is INCREDIBLY useful – it led to cochlear implant technology (and huge advances in neuroprosthetics in general), and more significantly, it allows us to see how the computational properties of neurons emerge from their biological properties.
Basically the short answer is yes, I do think it is possible, but I don’t really think about my discipline this way so my thoughts are rough and cursory.
Sorry for the non-economics related clutter.
10. May 2012 at 13:09
MF, you wrote
“The human mind is chaotic and complex? How would you be able to even know anything is “patently obvious” unless you were aware of some “simplistic” deductive thinking process to get to that conclusion?”
I’m not sure what you mean here. I think you’re talking about an individual being deducing and knowing things. That does take deductive thinking. What I want to say is that it’s wrong to derive a theory of human action using deductive reasoning, because most (if not all) of the a priori assumptions made will be wrong.
“Maybe you should read about something before you criticize it? Praxeology is just the study of logical constraints in human action. It’s not making any statements about the chemical properties of brains.”
The computational properties of the brain are derived from its chemical properties. Stated another way – the chemical properties of the brain are exactly what allow you to think. The general assumptions you make in a study of “logical constraints in human action” are just that, assumptions, not a priori facts or definitions. If you want to study constraints in human action, you have to start with evidence and facts about how our brains work, not “common-sense” assumptions about what people will do in a given situation.
That being said, you’re right, there is a lot I have yet to read and learn. But not from someone who believes there’s such a thing as “irrefutable.” So in that light, can someone else refer me to some literature?
10. May 2012 at 13:20
Alex Godofsky:
Except that your insistence on using the same basket of goods to measure CPI in 1990 as in 2012 would, in fact, result in you measuring infinity% inflation.
You’re the third person now to have conflated my argument over calculation of inflation methodology, with an argument over what should and should not be in the basket of goods.
For the third time, I am not making an argument about what should and should not be in the basket of goods. I am talking about the methodology for how price inflation of any basket of goods is calculated.
If the price inflation for today’s basket of goods is calculated using the methodology used for years up to 1990, before it underwent a drastic change, then reported prices would be higher than they are reported now.
My example demonstrates precisely WHY the increased fuel efficiency of automobiles correctly mitigates the impact of rising oil prices on the CPI.
Reduced consumption of said goods is a mitigation that leads to reduced standard of living, because the capital that went into increasing the fuel efficiency of cars could have went into other projects, increasing the efficiency and productivity elsewhere, which would have otherwise had the net effect of us having a higher standard of living.
If the methodology for calculating the CPI didn’t account for that, it would measure infinity% inflation if all the oil disappeared.
What disappearing oil? Oil isn’t disappearing. Crude oil has been crude oil for hundreds of years.
10. May 2012 at 13:24
I think the inflation discussion misses the point entirely that while the prices of things go up in a nominal sense, the entire price level includes income. So what if you paid $6.00 for a haircut 20 years ago and pay $20.00 now. It matters not, unless you bury your income in the form of dollar bills in a hole in the yard, stuff your mattress, or what have you expecting it to either retain its value or appreciate. Then, you will have a problem. If you want to do that, it is much better to convert income into something that has value, then stuff your mattress with that, or bury it, etc.. I don’t get why the concept of a nominal hedge is so difficult comprehend.
The AS/AD model can help explain why prices of some things might be rising while incomes, employment and output are at depressed levels. But of course that is skirted entirely by the Chicken Littles, making the issue of rising prices concerning necessities to be something it likely isn’t. It really is too bad because all the hyperinflation propaganda floating around obscures supply side issues from public view and they just simply won’t be appropriately addressed because people have been worked into a frenzy by Jacksonian propagandists and are angry about the wrong things; and of course none of the right things are happening either because of it.
10. May 2012 at 13:25
Scott, please come back, this is turning into another MF saga.
10. May 2012 at 13:32
RJ:
“The human mind is chaotic and complex? How would you be able to even know anything is “patently obvious” unless you were aware of some “simplistic” deductive thinking process to get to that conclusion?”
I’m not sure what you mean here. I think you’re talking about an individual being deducing and knowing things. That does take deductive thinking. What I want to say is that it’s wrong to derive a theory of human action using deductive reasoning, because most (if not all) of the a priori assumptions made will be wrong.
You mean the axiom that humans act is wrong? It can’t be wrong, because any attempt to refute it would have to be understood as an action itself, that has the specific goal of refuting a particular argument. You can’t refute the action axiom if you require action in your disproof.
What incorrect deductions are you talking about? If any proposition is deduced from the action axiom, and the logic is correct, the proposition is correct. Not saying everyone who tries this will ipso facto be correct, any more than a physicsts who says positivism works isn’t saying everyone who attempts it won’t make a mistake.
“Maybe you should read about something before you criticize it? Praxeology is just the study of logical constraints in human action. It’s not making any statements about the chemical properties of brains.”
The computational properties of the brain are derived from its chemical properties. Stated another way – the chemical properties of the brain are exactly what allow you to think.
I don’t really need a lesson in something I’m not interested in, that doesn’t serve to justify your criticism against praxeology.
The general assumptions you make in a study of “logical constraints in human action” are just that, assumptions, not a priori facts or definitions.
No, they are a priori true, because it is not even possible to contemplate it being refuted at any point in the future. The fact that the action axiom cannot be refuted without self-contradiction will be true millions of years from now, just like it is true today.
You are fallaciously calling irrefutable axioms “assumptions”, precisely because you haven’t read about it and don’t understand how to grasp the foundation of it.
If you want to study constraints in human action, you have to start with evidence and facts about how our brains work, not “common-sense” assumptions about what people will do in a given situation.
Yes, and the collection of evidence and facts and inducing conclusions from them ALSO presupposes a slew of propositions that must be true before you can even claim that your method of inquiry is doing what you ostensibly claim it is doing.
For example, what is the foundation for the validity of “evidence gathering”? When you say we must observe data before we can say anything, how in the world can you even go about this if you didn’t already know what the significance and implications are for “observation” itself? The validity of observations of course can’t be grounded in observation, or else one would be engaging in circular logic.
That being said, you’re right, there is a lot I have yet to read and learn. But not from someone who believes there’s such a thing as “irrefutable.”
Haha, you just contradicted yourself. By saying “irrefutable propositioning” is impossible, you are effectively saying irrefutable propositioning has been refuted. You are telling me that it is irrefutable that there is no such thing as irrefutable propositions.
You can’t do that.
So in that light, can someone else refer me to some literature?
You won’t find it from others. They are in the same boat of ignorance (no offense) as you.
The literature you are looking for, contains arguments in favor of irrefutable propositioning. Pretty much exactly what you are doing with me, except it takes greater care to not make contradictions like the ones you are making here.
10. May 2012 at 13:34
Saturos:
Scott, please come back, this is turning into another MF saga.
Saturos, I am responding to however many posts are directed at me. If that makes you mad, then your problem is with the fact that I am having something like 5 concurrent conversations going on at once.
10. May 2012 at 13:39
We need to find a way to translate MF’s posts into bitcoin.
Come to think of it, how do we know MF does not get paid by the letter?
10. May 2012 at 13:41
Alex:
If the methodology for calculating the CPI didn’t account for that, it would measure infinity% inflation if all the oil disappeared.
“What disappearing oil? Oil isn’t disappearing. Crude oil has been crude oil for hundreds of years.”
What I meant here is that the methodology used to calculate inflate changed in 1990, not by changing the content of the basket of goods, which is always changing by the way, but by the way the price inflation is calculated for whatever goods are in the basket.
If the methodology for calculating inflation had to change every time the content of the basket changed, then we wouldn’t see occasional methodology changes once every 20 years or so. No, we’d see yearly, even daily changes.
You are constantly conflating content of the basket, with the methodology for price inflation calculation of the basket. They are two different things.
10. May 2012 at 13:41
“You mean the axiom that humans act is wrong?”
The axiom that humans act purposefully to accomplish their goals is wrong. That’s what I’m saying. Many studies have shown that your brain has already decided to do something long before you are aware of the decision.
10. May 2012 at 13:43
“Come to think of it, how do we know MF does not get paid by the letter?”
That conclusion is not praxeologically sound.
10. May 2012 at 13:44
dwb:
We need to find a way to translate MF’s posts into bitcoin.
I’m flattered you think my posts can be converted into purchasing power.
Come to think of it, how do we know MF does not get paid by the letter?
Because those who disagree with you could only be thinking what they think because they’re purposefully lying for money. I smell…conspiratard wafting through the air.
10. May 2012 at 13:45
Saturos:
“Come to think of it, how do we know MF does not get paid by the letter?”
That conclusion is not praxeologically sound.
Actually it is. Thinking, making payments, typing letters, and making conclusions, are all praxeological concepts.
10. May 2012 at 13:47
@Bonnie,
10. May 2012 at 13:57
RJ:
“You mean the axiom that humans act is wrong?”
The axiom that humans act purposefully to accomplish their goals is wrong. That’s what I’m saying. Many studies have shown that your brain has already decided to do something long before you are aware of the decision.
This is just the newest version of the same old materialist argument.
Studies haven’t shown that the brain acts, then we become consciously aware of it. Researchers have only inferred that conclusion from the data they observed by using the theory of materialism. But if that’s the case, then we’d have to understand their researching as an action intended to learn something new that was not known before, because if we don’t, then we can’t even label their arguments as refutation of human action. We’d have to consider their research as nothing but automatic responses and pure instincts, with no argumentative implications, much like the wind could whistle through the trees, uttering the sound “humans…don’t…act.”
I can’t take these conclusion as a refutation of human action, because refutations themselves must be categorized as a purposeful action intended to do something. If the conclusions made really were true, then the conclusions themselves would have to be considered nothing more than just sounds and bleeps and scratches on paper that are not intended to do anything, including learning.
Brain activity that occurs a few milliseconds before conscious awareness of an activity, doesn’t imply the brain is making a choice and then humans are aware of what their brains chose. For how do we even know that the brain activity being considered, is what they are claimed to be? Even the primary researcher in this field isn’t saying that humans are automatons. He isn’t saying his conclusions should be taken to mean brains do something and then humans give themselves an illusion that they had a choice. He isn’t saying what you claim is the result of such research, which can only mean you are advancing an agenda.
10. May 2012 at 14:00
Bonnie:
So what if you paid $6.00 for a haircut 20 years ago and pay $20.00 now.
So it means the price of a haircut has gone up. That’s it. If you want to make a DIFFERENT argument, about real wages, or real incomes, then sure, but that doesn’t mean prices haven’t risen. The context was price inflation.
You’re right that nominal prices don’t tell the whole story IF you want to tell a story about standards of living.
10. May 2012 at 14:13
dwb:
actually that’s what i was trying to get at: what really matters is not how much things cost in $$ terms, but how long you work for a given basket of goods (productivity). If I made $6/ hr 20 years ago and now I make $40/hr, then 20 years ago i had to work 1 hr while today only a 1/2 hour.
in 1850 I would have had zero time to blog because I would be spending it all planting crops. We could have never fed 7 Bn people with the technology we had in 1850. the price level is irrelevant; what matters for standards of living is productivity.
Bingo. Which is why the main problem of economics is not how to increase demand, but how to increase sustainable production.
How is sustainable production promoted?
Individual property rights, saving and investment, economic freedom, contract protection, etc. Would you agree with this general assessment?
OK, now take these principles, and apply them to not only oil, textile, food, housing, medicine, and vehicle production, but money production as well.
The same reason why communist style central planning of oil, textile, food, housing, medicine, and vehicle production is less productive than capitalist style decentralized division of labor laissez-faire production of these goods, is the same reason why decentralized division of labor laissez-faire production of money is more productive as well, and why communist style central planning of money lowers total productivity from what it otherwise would have been.
It’s why a centrally managed “gold standard” is suboptimal, and why a centrally managed “fiat standard” is suboptimal, and why a private property free market division of labor in money production is optimal.
Let money production run according to profit and loss, the same way it works so well in potatoes and computers.
10. May 2012 at 14:22
MF – This is not what I wanted to talk about and so I will end the sidetracking right here. The childish interpretation of my point is “humans are automatons”. The nuanced and correct interpretation is that 90% of your decision-making processes are hidden from you, and thus the common intuition about motivation is usually wrong or incomplete.
Back to my original question to the rest…you all are an accomplished and well-read lot, and I would appreciate some intro material to sink my teeth into.
10. May 2012 at 14:38
Side note:
“If no action were to be taken by the fiscal authorities, the size of the fiscal cliff is” so large that there’s “absolutely no chance that the Federal Reserve would have any ability whatsoever to offset that effect on the economy,” Bernanke said.
http://www.bloomberg.com/news/2012-05-10/bernanke-speaks-about-risks-from-end-of-pro-growth-plans.html
10. May 2012 at 14:58
RJ:
MF – This is not what I wanted to talk about and so I will end the sidetracking right here. The childish interpretation of my point is “humans are automatons”. The nuanced and correct interpretation is that 90% of your decision-making processes are hidden from you, and thus the common intuition about motivation is usually wrong or incomplete.
Except you didn’t say that, you said the absolutist:
“Many studies have shown that your brain has already decided to do something long before you are aware of the decision.”
There is no room for 10% planning here.
You talk about nuance, but I didn’t see any in the initial post.
“Back to my original question to the rest…you all are an accomplished and well-read lot, and I would appreciate some intro material to sink my teeth into.”
1. Economic Science and the Austrian Method
2. Ultimate Foundation of Economic Science
These are good, short introductions.
10. May 2012 at 15:00
A new record was just set by Major_Freedom!
10. May 2012 at 16:04
I just watched Larry Kudlow debate Larry Summers on Kudlow’s show. I feel like I wanna shoot myself.
Kudlow basically denies there’s a demand shortage, and Summers is talking only about fiscal stimulus.
Market Monetarists need to start showing up on the CNBC’S and Bloomberg’s of the world.
10. May 2012 at 16:14
Just hoping one of the professional economists here can briefly explain:
How is real gdp growth possible if real wages have been falling? What am I missing? Is nt gdp total spending on output (and the national income)? So if real wages have fallen, how has rgdp grown? Aren’t falling wages = falling income?
Thanks all.
Big fan of the blog, and all the lively comments.
10. May 2012 at 16:22
marcus nunes:
A new record was just set by Major_Freedom!
Interestingly, this accompanied a new record for posts directed at me in the same thread. Funny how those things work out, huh?
10. May 2012 at 16:23
CA:
There isn’t a demand shortage. Never has been, never will be, as long as humans act.
10. May 2012 at 17:42
there is definitely a demand shortage for MF posts, no supply shortage though.
10. May 2012 at 18:09
What do you get when you recycle MF posts?
10. May 2012 at 18:17
Who, exactly, is Major Freedom?
10. May 2012 at 18:19
I think the inflation discussion misses the point entirely that while the prices of things go up in a nominal sense, the entire price level includes income. So what if you paid $6.00 for a haircut 20 years ago and pay $20.00 now.
Of course. USA 1/1982 to 1/2012…
CPI increase: +136.6%, +2.9% annual
Personal income per capita: +263.7%, +4.4% annual
It matters not, unless you bury your income in the form of dollar bills in a hole in the yard, stuff your mattress, or what have you expecting it to either retain its value or appreciate.
Sure. From 1982 to 2012 inflation by CPI averaged 2.9% annually, while bank saving account interest averaged 4.5%. Plus, you can use bank savings via your ATM card and they are insured against loss if stolen!
So why would anyone flunk an IQ test by keeping currency in a mattress? When the paleo-Austrians on Reddit say “the dollar lost 54% of its value!” due to that inflation, can they point to anyone who actually suffered that loss on the income they saved?
Then, you will have a problem.
Though if one’s mindset is to keep currency in the ground or a mattress for 30 years, then one has other problems too.
10. May 2012 at 18:21
“If no action were to be taken by the fiscal authorities, the size of the fiscal cliff is” so large that there’s “absolutely no chance that the Federal Reserve would have any ability whatsoever to offset that effect on the economy,” Bernanke said.
The Tax Policy Center on this:
~~~~
Taxes would increase by 2.5 percent of Gross Domestic Product in a single year, the Congressional Budget Office estimates. Nominal spending would fall for the first time since 1955…
The deficit would fall, all right. The Congressional Budget Office figures the deficit would decline from 7 percent of GDP this year to 3.7 percent in 2013 and to a very manageable 1.5 percent by 2015.
It would, that is, if the economy didn’t collapse.
~~~~~
Austerity, American style.
10. May 2012 at 18:25
“We need to find a way to translate MF’s posts into bitcoin.”
I’m flattered you think my posts can be converted into purchasing power.
It would an interesting test of the quantity theory of money.
10. May 2012 at 18:34
i dont think i am cherry picking, commodites are volatile but the average is not accelerating …. yes, people have figured out how to do more with less. whether its cause or effect is not relevant. oil is a finite resource, so the fact that the total cost of ownership of a car has not changed much (plus, better pollution controls) is a sign of advancing standard of living.
Not to go too far off topic, but as energy consumption/ “sustainability” is always a hot political-rhetorical issue, some under-appreciated data:
USA from 1979 to 2009…
[] Real GDP per capita: +59%
[] Energy consumption per capita: -14%
[] Cost of energy as % of GDP: -35%
[] Energy consumption per real dollar of GDP: -47% [EIA]
The same is true in other advanced economies. E.g. energy consumption since the 1990s has declined in absolute terms in those with low population growth, such as the UK and Germany.
Most of all the to-do about energy is about the supply side, assuming demand can only up. But the demand-side future is a lot brighter than the rhetoric usually paints it.
Now back to NGDP…
10. May 2012 at 20:18
dwb,
there is NO AD shortage for MF posts that can’t be fixed by unleashing retards into the public square posting spam SO AGGRESSIVELY that dwb becomes thankful to read a MF post.
Don’t you get tired of having metaphors give you deep and emotionally scaring anal probes?
10. May 2012 at 20:44
“Who, exactly, is Major Freedom?”
Someone who needs to start his own blog, lol.
10. May 2012 at 20:45
@Hugh D’Andrade,
You wrote:
“Who, exactly, is Major Freedom?”
He’s a dynamic masked hyperinflation fighting superhero, dedicated to thymology, praxeology and the Austrian way!!!
http://www.dailymotion.com/video/x3mz9r_blankman-theatrical-trailer_shortfilms
Oops, wrong superhero.
P.S. And now, for the bonus question. If everything reminded Sigmund Freud of sex, everything reminded Milton Friedman of the money supply, and everything reminds Scott Sumner of NGDP level targeting, what does everything remind Major Freedom of? (Tick, tock, tick, tock,..)
P.P.S. Tune in next post and see Major Freedom single handedly take on the entire Money Illusion community of regular commenters and kick their hedonic, substitution loving @$$&$!
10. May 2012 at 20:58
On the subject of making predictions here is an article about Ronald Coase who just turned 101. (May we all live so long and healthfully!)
Nobel Laureate: ‘I’ve Been Wrong So Often, I Don’t Find It Extraordinary At All’
by David Kestenbaum
“I recently had a brief conversation with Ronald Coase.
“I’m 101 at the moment,” he told me. “I get older by the minute.”
Coase is a legend in economics. He won the Nobel prize. He has a theorem named after him. But China’s rapid emergence as a global economic power “” one of the most important developments of the past generation “” took him completely by surprise.
“I thought it would take 100 years, if not more,” Coase said.
It seemed striking that an economic legend could be so wrong about such an important subject. I asked Coase what he made of this.
“I’ve been wrong so often I don’t find it extraordinary at all,” he said.
Coase just co-wrote a new book. It’s called “How China Became Capitalist.”
http://www.npr.org/blogs/money/2012/05/09/152197483/nobel-laureate-ive-been-wrong-so-often-i-dont-find-it-extraordinary-at-all
10. May 2012 at 23:43
RJ –
Nothing you said about neuroscience invalidates deductive reasoning. Regardless of *how* we mechanically engage in deductive logic, the process occurs. That process is exactly why we have things like math, logic, and systematic analysis in general… including neuroscience. If you’re going to toss out deductive logic, then you have to also toss out all conclusions derived from it…
…But, oops, that would be engaging in deductive reasoning. If X then Y is a logical chain. We’re stuck.
How on Earth can a person stop using deductive logic? It is practically a default state of human cognition. Are you sure you’re a neuroscientist?
11. May 2012 at 00:21
MF He has never read an original Austrian treatise Up to your usual standards of evidence I see. I guess reading Human Action by von Mises does not count as reading “an original Austrian treatise”. Given the post dwb referred to was my reactions to a paper by Steve Horwitz, saying that in response to dwb’s citing of that post was particularly silly.
11. May 2012 at 00:22
Ryan,
I’m not sure how to answer your questions. Could you give me an example that supports what you’re saying? Again, as far as I understand it, deductive reasoning involves generating specific conclusions from true general principles, and there aren’t many of those regarding the brain. I can’t think of any statement that holds true for all computational units that is actually useful for analysis or finding new information.
Just to be clear, I’m not saying that we don’t engage in deductive reasoning, or that neuroscience invalidates deductive reasoning. I’m saying that all in all, deductive reasoning is a poor tool for understanding the brain, because there are no “top-down” truths that allow you to understand the smaller components. There is a lot of inductive processing in neuroscience, where we take the properties of one neuron and abstract them to a population of neurons, but even that is not so much ‘reasoning’ as it is convenient mathematical simplification. For example, one of the most famous neuron models uses what’s called Cable Theory, which assumes all neurons in a population have the same membrane potential, which is found by observing individual neurons and establishing a baseline.
I am much less well-versed in the application of the philosophical terms than I am in neuro, so I acknowledge I may be misinterpreting these terms. I appreciate constructive criticism and like to learn, so things like this –
“Are you sure you’re a neuroscientist?”
are unnecessary.
11. May 2012 at 00:25
In fact, I might ask in response to your statement
“If you’re going to toss out deductive logic, then you have to also toss out all conclusions derived from it…”
What conclusions has neuroscience derived from deductive reasoning?
But this is an economics blog, and I feel silly writing so many comments about the brain. Back to Mankiw’s textbook…
11. May 2012 at 00:37
“If we use 1980 calculation methodology, inflation is at 10%.”
If we used the 1980 methodology when home prices were used instead of owners’ equivalent rent like today, CPI in recent years would most likely be showing deflation or near deflation because of plunging home prices.
11. May 2012 at 00:53
You know, Major Freedom, from the odd bits of your comments I read, I suspect that I might often agree with you, but you are making so many comments, which are often ill-tempered reflex rebuttals of trivial points, that I tend to just skip over all of them. So you are wasting your time and getting in the way of what could be a useful exchange of informed and thoughtful views.
11. May 2012 at 01:18
@Not an Economist
You raise an interesting point, which I think people are going to have to undestand in the coming years.
Real GDP is essentially domestic output in terms of items – cars, haircuts etc. Real wages, however, are wages deflated by, typically, the average price of a worker’s consumption basket. Since some of that basket – eg oil – comes from overseas, its size depends not only on how productive domestic workers are, but also what the terms of trade are between their products and the items that they consume. Therefore, if for some reason (which could be as simple as a shift in bargaining power as in the case of OPEC in the 1970s), the relative price of imports rise, a country could find that its workers are getting poorer despite producing more (eg because they are shipping more than their increased output of cars to the Middle East for the same volume of oil).
This matters I think because developed economy workers expect that a growing economy will make them richer, but, even if their economy grows they may be disappointed, because the world has become a more competitive place.
11. May 2012 at 01:19
Quoting Scott: “Regarding debt, if there is too much lending the issue should be addressed via regulation”
How exactly is this consistent with today’s post, in which you reiterate your support for the EMH – implying it’s impossible to know what “too much lending” is – and note regulators are not smart enough to spot the very same bubble that you claim they can prevent?
11. May 2012 at 02:24
Major Freedom,
“If we use 1980 calculation methodology, inflation is at 10%.”
If we use the 1980 calculation methodology, then the US economy is the worst in the world and the US could learn a lot from stronger economies like France-
http://www.shadowstats.com/alternate_data/gross-domestic-product-charts
Yes, the US economy has had a few quarters of growth in the last 10 years. That’s not a reductio ad absurdum.
11. May 2012 at 03:12
There’s a difference between a prediction that gets the causal mechanics of what actually happened correct, and some just saying ‘there will be a crash there will be a crash there will be a crash’ until there is. The former cannot be attributed to luck.
11. May 2012 at 04:07
W. Peden:
If we use the 1980 calculation methodology, then the US economy is the worst in the world and the US could learn a lot from stronger economies like France-
Worst in the world? How does that follow?
11. May 2012 at 04:09
RebelEconomist:
You know, Major Freedom, from the odd bits of your comments I read, I suspect that I might often agree with you, but you are making so many comments, which are often ill-tempered reflex rebuttals of trivial points, that I tend to just skip over all of them. So you are wasting your time and getting in the way of what could be a useful exchange of informed and thoughtful views.
???
I don’t see how this applies to anything I wrote. Ill-tempered? Speak to my interlocutors.
11. May 2012 at 04:17
Again, you all don’t like MF long posts, EXACTLY the same way
I don’t like overpaid public employees…
Yet, no one calls for Robot spam to fill Sumner’s blog, to make you like MF posts.
None of you even want to discuss, what Scott’s plan does to public employees.
And Scott agrees with me.
Does that mean you are all comfy with what is coming for public employees under NGDPLT?
Do you think it is clear to DeKrugman?
11. May 2012 at 04:32
RJ –
Maybe the whole source of the confusion is that you think economics is about understanding the human brain. It’s not.
You should actually read what Mises wrote about praxeology. He went to painstaking lengths to outline specifically where psychology ends and economics begins. Everything you’re referring to is what Mises called an “ultimate unknown” (from an economic standpoint). Mises took psychology as exogenous to economics, outside its scope.
It was only after clearly delineating between the questions that economics can and cannot address that Mises engaged in praxeological teleology.
In short, I think Mises’ approach is the one that best addresses your concerns, and yet you’ve completely written it off out of pure misunderstanding. That’s unfortunate.
11. May 2012 at 04:39
And in response to your second question, it’s not a matter of neuroscience, it’s one of epistemology. Neuroscientists engage in theory whether or not they choose to acknowledge that fact. In order to test a hypothesis in a laboratory, one must first know what one is testing. That knowledge is an aprioristic theory.
Science – even (or especially?) lab science – *is* logic.
11. May 2012 at 04:43
Sorry Morgan, this will be super-long. Since I didn’t get around to replying to Major_Freedom on an earlier question I had, I will clutter up this posts’ comment thread a bit more.
Major_Freedom, this is in reply to your answer to my question “can [research reveal and predict the likely economic action of humans]”? You wrote:
The very process of learning of these alleged constancies, presupposes a changing subject matter. You are obligated to hold that research endeavors results in the researcher to be a changed person. After all, research itself is ostensibly a process whereby the researcher learns something that they did not know before. Moreover, and this is the key point, when people learn different things, new things, they act differently. Knowledge influences what the researcher, indeed what everyone, does.
So it would be like studying a particular molecule whose chemical properties and reactive behaviors keep changing every time you study it, because the process of studying it changes the nature of the molecule itself. It would be wrong then to assume that the molecule’s properties and reactive behaviors that existed last year are the same this year.
So humans are like molecules whose properties change in the very process of learning about the history of the molecule.
Thus, there are no constancies in human action that can be inferred from past history. The past is 100% unique. Nothing that can be observed in the past can possibly elucidate any constant relationships in human action. Humans are entities that learn. We are not automaton chemicals and atoms and molecules. Chemicals and atoms do not ACT. They do not behave with a purpose. Humans do. This is why the methodology that the Fed economists use is completely antithetical to the nature of humanity, and why their equations that presuppose constancy, keep blowing up in their faces.
I appreciate the reply about why Austrians would reject the notion of using research to predict economic actions of humans. Yet I find issue with some of the underlying assumptions.
You seem to be almost referencing the Heisenburg uncertainty principle, which states that the act of measuring anything changes that which was measured, preventing accurate projection of the future position of that particle. That may be true, but it doesn’t mean that we cannot make projections accurate enough for use. The effect of measurement of an object, while existing, is so miniscule that the measurement is still accurate enough. I may measure the weight of a piece of metal, and a few atoms could call from the object in the course of moving it about, but that does not prevent my measurement from being close enough for use.
And you say that the past is 100% unique, and there are no constancies in human action. “We are not automaton chemicals and atoms and molecules.” But this seems to deny the existence of the natural world. Human beings are made of chemicals, and while an “chemicals and atoms do not ACT”, all human actions do involve chemicals and atoms. How can one jump to the assumption that a machine made from chemicals would not be subject to any of the same predictability that chemicals themselves display when isolated in a lab? In so much as this is claimed, it is a mark of our ignorance of certain chemical interactions, and not that the whole of the chemicals in a human somehow adopt a magical unpredictability.
And I do not want to depressingly steal away all of the magesty of human beings and free will, as our ignorance of all the chemical states of a human body do permit the appearance of free will. But we do not need to be an all-knowing God in order to determine any useful future projections based on our past observations of humans. Think about the research that exists where there are drugs that can be administered regularly to people to alleviate depression. I’ve personally known someone who does become sullen, withdrawn, and saddened while not on a particular antidepressent. But while on the antidepressent, they become more optimistic, outgoing, and act more friendly and positive with people in their life. Here is an example of a chemical affecting the actions of a human – and affecting them in a reasonably predictable manner! I did not have to predict *who* this person would decide to smile at, but rather the antidepressent seems to moved their brain into a different equilibrium wherein they have a tendency toward optimism and seeking engagement with other humans (to try and apply an economics analogy). And that was predicted – and did come true.
So this would seem to belie the notion that future human action cannot be inferred from past human actions. Humans are not exempt from the natural world, but we are very complex, and that is what makes predictions difficult. But not impossible. Humans are more similar than different. We all use the same chemical building blocks in our DNA and RNA. Our DNA is more alike than different. We are natural beings, not some particle moving in a truly random manner.
So you may be correct to say there are “no constancies” based on the past, it is an extraordinary leap of faith to assume that there is an utter absence of common human tendency based on the past. We do not need to make predictions worthy of God to still make predictions that are born out true enough in the majority of cases, and thus such predictions are useful. And that is why I think that research can help in predictions of human economic action.
All best,
Jason Odegaard
11. May 2012 at 04:51
Desolation Jones:
If we used the 1980 methodology when home prices were used instead of owners’ equivalent rent like today, CPI in recent years would most likely be showing deflation or near deflation because of plunging home prices.
Not sure how the fall in home prices would affect overall CPI, but even if it did, nobody buys the CPI anyway. Every individual buys their own basket of goods, and for most people, they are paying prices that are rising a lot more than 2% per year.
dwb:
there is definitely a demand shortage for MF posts, no supply shortage though.
It’s funny how you can only ever perceive demand and supply shortages and excesses (or else how could you justify state intervention?), even in blog posts. How oh how can the Fed “fix” these shortages and excesses? Problems problems everywhere without mommy and daddy around!
Becky Hargrove:
What do you get when you recycle MF posts?
The same thing you get with Sumner’s followers on this blog: Unoriginal parrots.
Hugh D’Andrade:
Who, exactly, is Major Freedom?
Who? Who is but the form following the function of what and what I am is a man with an anonymous handle.
Mark A. Sadowski:
He’s a dynamic masked hyperinflation fighting superhero, dedicated to thymology, praxeology and the Austrian way!!!
Thanks for putting in thymology. You’re far more informed than Lorenzo. Speaking of which:
Lorenzo from Oz:
Up to your usual standards of evidence I see. I guess reading Human Action by von Mises does not count as reading “an original Austrian treatise”. Given the post dwb referred to was my reactions to a paper by Steve Horwitz, saying that in response to dwb’s citing of that post was particularly silly.
Yeah there’s a problem with that. I say you don’t read that which you criticize because of the content of your criticisms. Criticisms are of course fine to make if you want to make them, it’s just that the content of your criticisms show you don’t even understand the material you are criticizing. You are misrepresenting Austrian arguments. I’ve been directed to your blog on more than one occasion from some of the other posters here for alleged “smackdowns” of Austrian theory, and every single time, it doesn’t take me usually more than two paragraphs in, to see glaring errors and/or omissions, which is particularly frustrating, because I really do believe them when they say you have smacked it down.
11. May 2012 at 05:12
Scott,
Is this a controlled experiment, to see how long it takes your comment thread to spontaneously combust while you are away?
11. May 2012 at 05:13
@Morgan,
i like your sense of humor, but no: you cannot fix an AD shortage by increasing supply. you know what happens at a big wind farm when the wind blows and there is no demand? prices go negative (which means wind producers are paying to put their electricity on the grid).
The problem here is that there is no free market price for MF posts. If there was, the price would be negative.
11. May 2012 at 05:29
Rebel Economist,
Thanks for your thoughts.
From my college days, I kept an intermediate level macro text by Mankiw (4th edition), and in it he speaks of the real wage as being the nominal wage over selling price of the good being made (for example, a bakery worker making bread at $20 per hour that sells for $2 per loaf would have a real wage of 10 loaves of bread, essentially). It seems that a productivity gain (the bakery worker making more bread per hour) would lower the selling price of bread, raising his real wage.
The real wage you are referring to — and the one MF linked to above — is nominal wage deflated by the price of the basket of goods. Are the concepts basically the same, just expressed differently?
And if I understood you correctly, RGDP growth shows we are increasingly more productive at making things (per hour, for example), but falling real wages show that our productivity gains are not keeping pace with those of the rest of the world — NOT that nominal wage growth is lagging behind the rate of inflation.
Thanks for your help in understanding this.
11. May 2012 at 05:31
Here is CLASSIC COMEDY:
“Elliott Stegall, 51, is pursuing a Ph.D. in film studies at Florida State University while he teaches two English courses at Northwest Florida State College in Niceville, Fla.
To help support their two young children, he and his wife rely, in part, on food stamps, Medicaid and aid from the USDA program, Women, and Infants and Children (WIC).
“I tend to look at my experience as a humanist, as someone who is fascinated by human culture,” he told the Chronicle. “Maybe it was a way of hiding from the reality in which I found myself. I never thought I’d be among the poor.”
He and his wife also have worked part-time jobs as house painters and cleaners and food caterers.
“As a man, I felt like I was a failure. I had devoted myself to the world of cerebral activity. I had learned a practical skill that was elitist,” he said. “Perhaps I should have been learning a skill that the economy supports.”
ROFL. At least he is smart enough to think critically.
http://abcnews.go.com/Business/growing-number-americans-phds-receiving-food-stamps-aid/story?id=16310858#.T60T0F2ZgeN
11. May 2012 at 05:37
In the off chance that some may have time for something other than reading Major’s posts;
http://www.aei.org/events/2012/05/11/is-uncertainty-responsible-for-high-unemployment/
Live until 11:30 eastern.
11. May 2012 at 05:38
dwb, please compare that to increasing the marginal utility of the folks who currently don’t have jobs by printing money.
These are the least likely to be valuable workers in our economy.
There’s no “really should be working at a wage they want to receive” workers sitting on the couch.
The economy is not failing them. They have failed. They are in debt from studying for a PHD in Film Studies. Apparently the only thing they can actually do is paint houses.
My Guaranteed Income plan will make sure they go paint houses.
I’m not sure your printing money accomplishes the same thing.
11. May 2012 at 05:57
News headlines: “Stocks of the major U.S. banking companies fell sharply in premarket U.S. trading Friday, led lower by J.P. Morgan Chase & Co. after the bank revealed a $2 billion trading loss in a previously little-known corner of the bank’s vast operations.”
http://online.wsj.com/article/BT-CO-20120511-710858.html
Alternative headline:
“EMH for the win! Non-JP Morgan investors gain $2 billion in trading profits!”
11. May 2012 at 06:11
Jason Odegaard:
I appreciate the reply about why Austrians would reject the notion of using research to predict economic actions of humans. Yet I find issue with some of the underlying assumptions.
You seem to be almost referencing the Heisenburg uncertainty principle, which states that the act of measuring anything changes that which was measured, preventing accurate projection of the future position of that particle. That may be true, but it doesn’t mean that we cannot make projections accurate enough for use. The effect of measurement of an object, while existing, is so miniscule that the measurement is still accurate enough. I may measure the weight of a piece of metal, and a few atoms could call from the object in the course of moving it about, but that does not prevent my measurement from being close enough for use.
I suppose I used the Heisenburg uncertainty principle as an analogy, because most people who think the Austrians wrong on this point tend to understand it better when it is “translated” into conventional, positivist concepts and terminology. Please don’t take the Heisenburg analogy as an exact mirror, because even with atoms and molecules whose study is affected by it, are still treated as behaving according to constant causal operative factors. For example, the double slit electron diffraction pattern ALWAYS appears in the double slit experiment. It is inferred from this that even though we cannot predict single particles, we can get a good enough overall prediction of what will happen to “the electrons.”
Yes, the uncertainty with electrons derived from measuring them is very small, but with humans, it is not a question of being small or large, it is a question of being present or not.
With human action, there is not even a constancy akin to constant diffraction patterns. When I say that, I don’t mean I am saying this based on observation and making inductions, in which case there’s always a chance we might find a constancy at some point. No, I mean that prior to even going about studying this, I am (we all are) logically obligated to treat human action as inherently changing and non-constant, because it is presumed that going forward, we’ll be able to learn by using the positivist methodology. No positivist can possibly deny this. If humans could not learn via positivism, then positivism could not even do what it ostensibly claims it can do, which is provide humans with new knowledge. Even positivists are compelled to admit that their methodology logically requires humans to learn over time, and hence change over time.
The connection to non-constancy in human action is that when humans learn over time, we act differently too. And because we learn in a priori unpredictable ways (or else we would be able to predict what we will learn in the future, thus becoming omniscient in the present), it means that our actions are a priori unpredictable as well. When I say unpredictable, I mean there are no scientific constants that can be discerned from past learning and past actions, which will forever apply to the future regardless of what we will learn and do in the future. We cannot discover any “Planck”-like constants in human action like there are in the physical world. Even if such constancies really do exist, which I cannot say one way or the other, humans will never know them. For in order to know them, we would have to know the universe from a bird’s eye view, which of course is impossible, since we can only understand the world through being ourselves.
Now, does this mean that predicting human action is completely impossible? Well this is where it gets tricky to explain. It is impossible to predict human actions in “scientific” ways that presume constancy in relations, such as every single mathematical equation in economics and finance. It is not however impossible to predict human actions in entrepreneurial ways that presume anticipation and expectations. Entrepreneurs can and do anticipate future human actions, and are able to earn profits by doing so. But when they do so, it’s not because they discovered a constancy in human action that they are exploiting, the way positivist economists believe to be happening.
If anyone really did discover a constancy in human action, then they would have found a God’s constant that will be true no matter what humans tried to do going forward, and no matter what they learned going forward. It would be a constant in human that even they themselves could not overcome even if they tried. In other words, that constant would allow the individual who knows it, to see their own future actions as if they were taking place right now. This is why you typically see Austrians calling positivist economists “astrologers”. It’s because positivist economists are just modern day versions of them. It may appear as more intellectually rigorous and scientific, but it’s just the same ancient desire to see into one’s own future by reading tea leaves.
And you say that the past is 100% unique, and there are no constancies in human action. “We are not automaton chemicals and atoms and molecules.” But this seems to deny the existence of the natural world. Human beings are made of chemicals, and while an “chemicals and atoms do not ACT”, all human actions do involve chemicals and atoms. How can one jump to the assumption that a machine made from chemicals would not be subject to any of the same predictability that chemicals themselves display when isolated in a lab? In so much as this is claimed, it is a mark of our ignorance of certain chemical interactions, and not that the whole of the chemicals in a human somehow adopt a magical unpredictability.
Yes, this is a common response. The thinking goes “You’re asking us to believe the almost theological belief that humans are somehow different than everything else. That everything else around us is based on constancy, that everything else is made of chemicals and atoms and molecules and energy. Why not people as well? After all, aren’t we just chemicals and atoms and molecules and energy too? What makes humans so special? Obviously Occam’s razor suggests that positivism, although it is not as good in economics as it is in physics, is still the way to go, and so we shouldn’t “give up” the fight to studying and trying to find such constancy.”
To answer your point, I will again invoke an analogy at this point. Do you know about emergence? What I am saying can be grasped by the idea of “strong emergence.” Strong emergence takes place when a system has qualities that cannot be directly traced to its constituent components. It a system with self-“supervening.” It can be summed up in the oft repeated expression “the whole is greater than the sum of its parts.” There are properties that arise from the interaction of components, that cannot be deduced or induced from the components themselves. For example, it is impossible to know everything about water by knowing, even 100%, isolated oxygen atoms and isolated hydrogen atoms. There are properties that emerge from the interaction of oxygen and water that do not exist with oxygen and hydrogen alone.
In this way, human action is an emergent property that cannot be reduced or traced back to whatever constituent components it is “made of.” For in order to be able to do so, humans would have to be God. We would have to know the universe and everything about it from a God’s eye view, so that we can know everything about every possible combination of every constituent component, before they even take place.
Human action is an irreducible emergent behavior that MUST be taken as a given and cannot be traced back to constituent causes. It is this aspect of Austrian epistemology that APPEARS to be “theological” and “dogmatic”, when in reality it is accepting a major constraint in human knowledge, namely, knowledge acquisition itself.
Emergent behavior in human action is like a self-learning system that cannot know its own future of learning until it actually goes out and learns it, but as soon as it learns it, it is now a NEW emergent behaving system. Humans are evolving in what we are made up of. Our brains are changing over time. Your brain one year ago is not the same brain as it was yesterday. Albert Einstein pre-1905 was very different compared to Albert Einstein post-1905. He acted differently. Not even Albert Einstein could predict what he will come to know 1905, and thus how he will act on the basis of his knowledge as such.
And I do not want to depressingly steal away all of the magesty of human beings and free will, as our ignorance of all the chemical states of a human body do permit the appearance of free will. But we do not need to be an all-knowing God in order to determine any useful future projections based on our past observations of humans. Think about the research that exists where there are drugs that can be administered regularly to people to alleviate depression. I’ve personally known someone who does become sullen, withdrawn, and saddened while not on a particular antidepressent. But while on the antidepressent, they become more optimistic, outgoing, and act more friendly and positive with people in their life. Here is an example of a chemical affecting the actions of a human – and affecting them in a reasonably predictable manner! I did not have to predict *who* this person would decide to smile at, but rather the antidepressent seems to moved their brain into a different equilibrium wherein they have a tendency toward optimism and seeking engagement with other humans (to try and apply an economics analogy). And that was predicted – and did come true.
This is an excellent example. Ask yourself this: Given that you have discovered a pill that when ingested, would make a depressed person happier, have you discovered a constant in human action? Or did you learn of a particular history for a particular person? Do you know if I will take that pill in the future? Do you know if your friend will take that pill in the future? What you see is the past. You see someone taking a pill only when they take it. You did not actually discover a constancy in action any more than you didn’t discover one by realizing that jumping out of a flying place without a parachute will make the person’s body fall to the Earth and splatter. What you discovered is something akin to “If I eat good and nutritious foods, I will get ill less often.”
This is not what is being referred to when one says there is no constancy in human ACTION. That MAY be a constancy in human biology (I can’t say it’s a constancy because it is possible that such pills can have unintended affects on those who take them), but it’s not a constancy in human action. A constancy in human action would be of the form “If A, B, and C are true, then humans will ALWAYS act in accordance with X.” X can even be “Y with 50% frequency and Z with 50% frequency.”
So this would seem to belie the notion that future human action cannot be inferred from past human actions. Humans are not exempt from the natural world, but we are very complex, and that is what makes predictions difficult. But not impossible. Humans are more similar than different. We all use the same chemical building blocks in our DNA and RNA. Our DNA is more alike than different. We are natural beings, not some particle moving in a truly random manner.
Ah, try not to make the mistake of conflating unpredictability with randomness. I can be unpredictable to you, but completely non-random in my mind. I could have a very detailed pattern of movement I intend to do, and from your perspective, it appears I am acting completely randomly because you are unaware of any constancy.
What do you mean by “humans are not exempt from the natural world”? Are you saying humans have to behave and be like atoms and molecules and things around us because it would be silly to say only humans are different? Or are you saying we’re material and physical and so we must be able to learn about the same sorts of constancy in ourselves?
What is so wrong about recognizing a difference in human action that does not, or cannot be perceived, in anything else that does not act? Why can’t intentional purposeful behavior that transcends automatic reflexes to stimuli be a part of the natural order? Why can’t non-constancy in human action be a naturally occurring emergent property of certain types of atoms and molecules and energy in certain types of orientation?
If the universe is capable, through the eternal laws, of containing unique entities that act in ways that cannot be predicted by those same entities, then why can’t humans be the first instance of this on the planet Earth? Why can’t unpredictability arise in a physical entity on the basis that it cannot know what it itself will learn over time, before it learns it?
It’s not a matter of being complex, it’s a matter of what is possible for any physical entity that exists. How can any physical entity contain information of the entire universe, which would be required for it to even know of a constancy? Humans are contingent. We are the result of billions of years of time, of evolution. We learn over time. How in the world can the actions of a species that learns over time in non-constant ways, be predictable according to constancy? It makes no sense to believe it is possible.
Your examples above are not about human action. They are about human biology.
I will argue humans are more different than similar, but that’s me.
So you may be correct to say there are “no constancies” based on the past, it is an extraordinary leap of faith to assume that there is an utter absence of common human tendency based on the past. We do not need to make predictions worthy of God to still make predictions that are born out true enough in the majority of cases, and thus such predictions are useful. And that is why I think that research can help in predictions of human economic action.
OK, if you think it’s possible, I challenge you to show me just one example of a constancy that has been discovered in human action, that is akin to Planck’s constant. Your example of a person turning from “depressed” to “happy” is a psychological disposition. It is not action based. I could say something similar, in that “most people” become “happier” when they are in a relationship with someone else. I could come up with a general percentage of “something over half of all people”. But did I discover a constancy in action? No, because I am not actually saying anything about what they are DOING.
So tell me a constancy in human action, not psychological dispositions, that is always true the way Planck’s constant is always true. If you can’t, then I ask that you recognize my argument as valid.
11. May 2012 at 06:16
There’s no “really should be working at a wage they want to receive” workers sitting on the couch.
i agree, but when a bunch of sharks are fighting over food, and you want to replace it with more nutritious but more condensed food, then you will get your hand bit off trying to take away the food. You need to wait until they are full and off doing something else.
Which is to say: I agree we need reforms (can we start by shutting down banks that take 2 bn risks with taxpayer subsidized excess deposits please). Many companies are on some sort of distortionary corporate welfare right now which I’d love to axe. Moving monster.com to a type of auction model would be a great idea (there is a lot of $$ surplus in headhunting). I’m on board with the secret plan to reduce govt, but you have to push those reforms at full employment, otherwise, as happened in France and Greece, voters will reject them. Even worse: when voters reject them, it could be a very long time before they consider them again.
11. May 2012 at 06:58
dwb,
my hand won’t get bit off, liberals will cry. WHO CARES if DeKrugman thinks the economy he prefers is being held hostage?
The facts are the facts, the DECIDERS get to decide.
That’s the IMPORTANT LESSON. DeKrugman’s interests have to come last.
This is about status and power. The plan doesn’t have to be secret. We need everyone to take note and learn… if you aren’t productive, you are a beggar, and beggars can’t be choosers, Capitalism trumps Democracy… so don’t fight capitalism, make your Democracy tune itself to Capitalism’s frequency.
11. May 2012 at 07:04
Morgan,
Why do you think that mass devaluation of knowledge in general is indicative of social progress?
11. May 2012 at 07:15
Scott — Do you like corporate credit spreads as a market based indicator of the business cycle?
11. May 2012 at 07:20
RJ –
“What I’m asking is, if I find nothing particularly obvious, where do I start an analysis? (besides reading and trying to learn from your blog, that is. ”
Here’s my suggestions. For Scott Sumner’s views, his National Affairs article is the best link, in my opinion.
http://www.nationalaffairs.com/publications/detail/re-targeting-the-fed
Next, read Irving Fisher’s debt deflation theory of depressions:
http://fraser.stlouisfed.org/docs/meltzer/fisdeb33.pdf
Then, two books by Milton Friedman:
Money Mischief: Episodes in Monetary History
A Program for Monetary Stability (1983 version)
Money Mischief is an easier read, Program gets more into the details of monetary policy. Read all four and I’d estimate you know more about money than 90% of people.
I hope Scott is not offended that I recommended twice as many sources from Friedman than from Sumner!
11. May 2012 at 07:26
@Not an Economist,
I guess Mankiw’s explanation is simplified to a one-item economy. But in the real world, our consumption is more complicated, and we effectively trade some bread for shelter, milk, gasoline etc, and if bread is cheaper, we can buy less of those things, even if we can afford more bread.
The one-item and the multi-item economies are not the same. In a one-product economy, increased productivity would definitely make us better off.
You do understand me correctly, and I agree the most likely explanation is that the rest of the world is improving productivity faster than “us” (ie the old industrial world). In a nutshell, a billion plus Chinese decided to stop worrying whether they should stop at red lights and drive on the right, and started to compete with us in producing many similar goods – eg tyres. Good for them, good for the producers of limited resources that we all buy such as oil, bad for us. It might have been better for us if Nixon had not visited China! This IS the same thing as nominal wages not keeping up with the rate of inflation – money is the numeraire in which terms of trade are expressed.
11. May 2012 at 07:27
Morgan with all your talk of “the deciders get to decide” I’m starting to think you’re George W. Bush’s long lost twin.
11. May 2012 at 07:49
I agree the most likely explanation is that the rest of the world is improving productivity faster than “us” (ie the old industrial world).
not sure i 100% agree with that statement. they are improving “faster”, but from an extremely low level. The US is extremely productive, so productive that you will never get 3000 San Fransicsoans to line up for a Foxconn job because the alternatives are much better for them. The US actually does have a pretty substantial manufacturing and agriculture base and we export lots of capital and ag goods (for example) but the capital goods we export are high-end things which require a lot for skill. GM for example, recently moved a hybrid car plant from Mexico to Maryland.
I also dont agree that this is “bad” for us because they consume more oil. Gasoline efficiency has improved in the US to where we are now a net exporter of gasoline products. The overall energy intensity of the US economy has been declining for a long time as we employ new tech. Some companies are exploring building big plants to make diesel and heating oil from shale gas because the spread is very high (shale is very cheap).
I don’t really see a story where converging standards hurt us.
11. May 2012 at 08:17
@dwb
Troublemaker!
You are confusing levels and changes. The US is still more productive than China, which is why US workers are still paid a lot more than Chinese workers, even if US workers’ pay is slightly declining in real terms.
I chose oil as an example because Americans are painfully aware of rising gasoline prices, but, as you show, oil is perhaps not the best example for the US, not least because the US is a major producer of oil. Insert other internationally traded commodity less produced by the US, say rubber? Actually, the US should do relatively well, because it does have a large endowment of the commodities in limited supply, not least land.
Roughly, converging standards mean that nations tend towards getting a share of the global output pie proportional to their population. As production becomes more efficient, we can expect the pie to grow a little – ie economic growth. If a country’s share falls faster than the pie grows though, it will get poorer. Is it that hard to see?
11. May 2012 at 08:29
@Morgan,
From the article on Elliot Stegall,
“I never thought I’d be among the poor.”
The interesting thing is when I went back to school to study economics I *was* unemployed and poor and it was really just a way of getting by and passing the time. I never gave *any* thought to where it was going. I actually *never* thought it might lead to a PhD, and I never realized that PhDs in economics actually get *jobs*. (As Maynard G. Krebs would say, “work?!?”)
In some ways economics is a better field to be in than my previous field, which was math. (For whatever reason all kinds of institutions employ economists: universities, government, businesses, global institutions, etc. Even more if you’re willing to sell your soul to ride the Koch brothers gravy train.) So I kind of fell bass ackwards into opportunities with absolutely no foresight and all of the deftness of Mr. Magoo. (Ironically, even the mention of my choice of dissertation topic [tax structure and economic growth], which I chose strictly for fun, makes the lizard brains in the private sector twitch with excitement.)
Now, the first clue to Elliot’s weirdly dire predicament is that while he is teaching (english), he’s not even teaching in his own field (film studies). Real jobs are hard to come by at universities these days. As the article points out 70% of faculty members are off the tenure track. But if you can’t even get a lowly subsistence level adjunct position in your own field that suggests that you have probably made an amazingly poor choice.
If he actually thought a doctorate in film studies was going to open doors for him perhaps he should have stopped watching films and pontificating on them long enough to do some market research. (I watch films and pontificate on them for free all of the time. Come to think of it, maybe I should invite people over to my house and charge them tuition to watch TV.)
Which makes me wonder, why does he still consider film studies a “practical” skill?
11. May 2012 at 08:43
@RebelEconomist
yes, i agree. however, i dont think the US predicament is lack of productivity, its lack of demand.
11. May 2012 at 08:57
@dwb,
Yes. I was attempting to explain why US workers can get poorer despite increasing real GDP, but if real GDP contracts, then they are even worse off. How much of that fall reflects demand, and how much reflects supply (structural factors), is another, perennial, question though.
11. May 2012 at 09:26
RebelEconomist / dwb,
Thanks to both of you for engaging with me on this, but if I am hearing you right, don’t declining real wages (nominal wages failing to keep up with inflation) cast doubt on the assumption of long run monetary neutrality? I thought money was not supposed to be able to influence real variables in the long run, such as real wages. If it does, don’t we have some explaining to do as market monetarists if our nominal stabilization contributes to declining real wages?
I chose my name (Not an Economist) in case these are silly questions, and I suppose I am missing something obvious. But just when I thought I was starting to understand this blog, I’ve now got myself all tied in knots over this.
11. May 2012 at 09:39
Morgan,
“Why do you think that mass devaluation of knowledge in general is indicative of social progress?”
Becky, the only way anything improves is if it is constantly subjected to aggressive efforts to make it cheaper (big technical improvements sought for profit by greedheads).
Meaning we have to want the status quo to be under siege by the private market, and we have to keep the greedheads that are winning from using govt. to protect themselves from other greedheads.
This is as perfect a system as we can build.
But, I have no clue what you mean by knowledge in this instance. Where I come from is memetics vs. personal experience.
There is an ever increasing human knowledge base, and we want as many brains on the planet, to receive as much input as possible.
Then there is personal experience, this is where we get real output. It is input + value add (personal experience).
So I have a hard time figuring out what you mean by the “devaluation” of knowledge.
Maybe the above helps you re-ask?
11. May 2012 at 09:46
Morgan,
The reason I get somewhat bent about your arguments: you can come across as though the world would be a better place, without the pursuit of the knowledge-based dream. You’re right in that money does a crummy job at supporting such dreams. I got my first clue as to that problem in the fall of 1973, when my first voice teacher (I was a music major) worried about not being able to keep her job unless she completed her PhD. Plus, the music teachers (back then) weren’t comfortable with the fact that not enough teaching jobs actually existed,so many of them refused to prepare students for popular entertainment and trained them clasically instead. Not many people then really knew what kind of world to prepare us for, including our parents. I never quit wondering, how does society make room for those dreams when money can’t do the job.
11. May 2012 at 09:53
Morgan,
This comment got posted at the same time, will think about your question and reply.
11. May 2012 at 10:04
@Not an Economist,
The above arguments are independent of monetary policy; they would be equally true, for example, in a gold standard regime. NGDP targeting is really about stabilising output and employment rather than the long run, fundamental changes I was discussing.
11. May 2012 at 10:10
Not an economist:
“don’t declining real wages (nominal wages failing to keep up with inflation) cast doubt on the assumption of long run monetary neutrality?”
monetary policy always has some redistributive effects. The aim is to pick the policy with the least redistributive effect (and: doing nothing also has redistributive effects as well).
Declining real wages means means lack of productivity.
Also, I am skeptical about stagnation hypothesis in general – for many reasons.
11. May 2012 at 10:37
“don’t declining real wages (nominal wages failing to keep up with inflation) cast doubt on the assumption of long run monetary neutrality?”
Declining real wages means means lack of productivity.
Well, remember that “wages” are just a subset of employee compensation, and a diminishing subset as ever more employee compensation is taken in the form of benefits, which are not counted in “wages”. E.g.:
http://research.stlouisfed.org/fred2/series/COMPRNFB
Nonfarm Business Sector: Real Compensation Per Hour, 2005 = 100
1982-01-01 76.390
2012-01-01 103.171
That’s up 35%, adjusted for inflation and including benefits.
Be wary of people who invoke “wages” in argument without noting that they are only a portion of what workers get paid.
11. May 2012 at 10:55
The right way to look at future oil supply is EROEI – that is how many barrels of oil equivalent in energy you need to produce a barrel of oil. At some point it doesn’t pay to produce oil for energy. It might still pay to produce it as a petrochemical feedstock, but that won’t run your car or heat your house.
This is complicated by the possibility of substituting one kind of energy for another like natural gas for oil. So we can never run out of oil. We can, however reach a point where we can no longer depend on it as a net source of energy.
The slopes of the price and supply curves will be different, and a 10% increase in price won’t lead to a 10% increase in available oil, and somewhat before EROEI 1.0 It will be probably be to use other forms of energy directly than to use them to pump oil for energy production at a net energy loss. It will be only an expensive feedstock (natural gas or agricultural products will probably be cheaper).
Activities that require oil and accept no substitutes will become expensive, and activities where there are substitutes will be priced according to the price of the substitutes.
For instance, suppose I told you the supply of oil was unlimited at a price of $600 a barrel in real 2012 dollars (which may very well be true). Would you say there was no oil shortage? Would oil production in barrels decrease?
At some price point it will pay to use enough of some other form of energy to trap and convert atmospheric CO2 into hydrocarbons. The supply would then be limited only by the available energy and the earth’s ability to radiate enough heat to keep the temperature livable.
For some reason I don’t get a lot of comfort from that.
11. May 2012 at 11:08
Negation, Thanks a lot! I appreciate it.
11. May 2012 at 11:11
@Jim Glass,
yes i agree.
11. May 2012 at 12:29
Morgan,
Basically I want to make sure that austerity doesn’t mean the loss of ‘knowledge capture’ structures without sufficient replacement…make that ‘better than ever’ replacement. We just need time to make that happen. You have confidence in secure sources (“There is an ever increasing human knowledge base…”) however that base could be really damaged if people only see what they are losing. If people do not have a new direction they can turn to with confidence, that pool of knowledge might not be there when it is needed most.
For people not to lose confidence, knowledge needs to move away from the ‘random mining’ that acknowledged and refined the pool (or vein) in the first place. Business will only continue to need random mining, but then business can no longer be everything to every one. When people did not have to worry about paid work, it was easy to maintain the formal wealth capture structures of the 20th century. But in order to keep knowledge wealth now, we have to securely reinforce the knowledge base so that new wealth continues to build even outside of monetary reward, just as it did for so long. Otherwise the weight of those left out will drag down the structure itself.
You say that capital and main street wealth come first (before the monetary support of education in general) – which is true even if it sucks, but why say capital is better? Capital is nothing without knowledge as you well know. They need equal footing and respect, even if it is not possible to achieve that through the use of money. Some people will not want to listen to arguments about restructuring knowledge utilization, if they also hear that the purveyor of knowledge is second fiddle to the capitalist.
11. May 2012 at 12:34
Jim Glass,
Great point. I wasn’t intending to “argue” the matter — I was simply concerned about what declining real wages might mean and was seeking some clarification (as I have tried to proclaim my relative ignorance on these matters throughout the whole conversation).
You all have been very helpful – thank you!
11. May 2012 at 14:19
Vivian, I think we can now safely say his policy recommendation in 2010 was WRONG.
Ryan, Not wrong, just right and lucky.
Rademaker, You said;
“Similar question about using TIPS spreads as an indication of inflation expectations: if the fed commits to targeting a 2% inflation rate, why would anyone bid the spread away from 2%? In the longer run, to bet against the fed is a guaranteed loss no matter how fanatically the fed has to loosen/tighten to get to it’s target. This is especially true if the fed monitors the TIPS spread to see how well it is doing. Knowing this, is there really any information in the TIPS spread other than “the fed targets a 2% CPI inflation rate”?”
This is Bernanke and Woodford’s “circularity problem.” The trick is not to have the market give an unconditional forecast of inflation, but rather to forecast the instrument setting that produces 2% inflation.
Rebeleconomist, You said;
“Really? Don’t you think the dotcom bust and its effect on one of the more tech-heavy economies might have had something to do with it?”
Yes, and we have the same sort of problems now, which is why higher rates might produce the same effect.
Greg, The majority of Austrians were arguing against monetary stimulus–and that includes academic Austrians.
JN, You said;
“I don’t think Austrians or monetarists have ever precisely defined the time it would take for new money to generate inflation, especially during a lengthy crisis that never ends. They just say it would happen….at some point.”
This is wrong, at least some Austrians predicted inflation soon. Monetarists say it takes about 6 to 18 months after the monetary injection. It doesn’t matter how long you wait–they are wrong, and will remain wrong.
Saturos, I presume he gave dates, otherwise why would Tyler have praised him as offering the most accurate predictions.
Meegs, I don’t think he was being tongue in cheek, but it has no bearing on my argument either way–as I’m sure many agree with Tyler.
RJ, I suppose you could study economics textbooks. I hope to do a book next year.
Not an economist, Wages are very different from income.
MMJ, Regulators cannot spot bubbles, but they can spot moral hazard–it’s all over (FDIC, TBTF, the GSEs, etc.) My first best policy is to get rid of those. Second best is to do regulations like a ban on subprime mortgages with FDIC insured funds.
Unlearningecon, You said;
“The former cannot be attributed to luck.” Of course it can.
Tommy, I don’t know much about them. I suppose they have some information on RGDP.
Everyone, Sorry if I missed a few. 147 comments is too many for me. If it’s important ask again and I’ll come back Sunday and answer.
12. May 2012 at 00:39
Scott, don’t worry, most of them were MF posts 😉
12. May 2012 at 04:48
How can describing the exact mechanics of a crisis that actually comes to pass be attributed to luck ?
12. May 2012 at 04:55
OK I realise we will talk past each other if this continues.
If I say ‘x is going to happen because of y’, then x happens, then on closer investigation it appears that y was a substantial factor, this is surely not luck?
If I say ‘x is going to happen’ and x happens, that is luck.
13. May 2012 at 14:37
UnlearningEcon, If I say the banks are going to get in trouble because they made zillions of sub-prime mortgage loans to anyone that can fog a mirror, does that make me a genius? Obviously if the banks get in trouble it’s going to be due to bad loans, what else could it be?
I don’t doubt he knows more about banking that I do, and hence can describe the specific types of loans in more detail. But it still was luck that he predicted the crisis. If the market didn’t expect the crisis (and it didn’t) then it was most likely luck–unless you think he’s smarter than the market.
14. May 2012 at 01:11
“Monetarists say it takes about 6 to 18 months after the monetary injection.”
My understanding was that the 6 to 18 months rule applied to a more “normal” situation, during which banks don’t accumulate excess reserves without lending them out as they are currently doing.
14. May 2012 at 08:33
Becky Hargrove:
Although I can’t be absolutely sure, I suspect that there is an education bubble, brought about by years of inflation, credit expansion, low interest rates, and federal backed guarantees.
Just like house flippers throughout the 1990s and early to mid 2000s believed housing was a sacred investment, so too do people believe that is the case with education. Education has acquired the same unfortunate sacredness. It bleeds through your posts. Many, many others share your sentiment. It is not to be questioned, or criticized. Thou shalt acquire a student loan, and thou shalt attend a university or college. Period. End of story.
In a free market, only the capital available at the margin will be freed up to expand education, before all other lines are relatively satisfied first. But in our inflationary economy, investors have no idea how much capital is really available and how much consumers are willing to put into education through their saving.
The result, I suspect, is an education bubble that has not only made tuition prices skyrocket, which is the rather uninteresting part of inflation, but it has also brought about an entire sub-economy utterly dependent on student loan money continually increasing.
The larger universities and colleges are like miniature cities, complete with infrastructure, consumer goods retailers, and other amenities. All of this investment depends on students continuing to get more and more credit expansion from the banks with help from the Federal Reserve and Treasury to backstop them. It’s like the Fed along with Fannie and Freddie in housing, but in education instead.
Just imagine what is going to happen to all the investments that depend, directly and indirectly, on student loan money continually increasing.
14. May 2012 at 09:07
hehe, education bubble. lets all be more stupider.
meanwhile:
http://www.slate.com/blogs/moneybox/2012/05/14/thiel_capital_seeks_well_credentialed_young_analyst.html
14. May 2012 at 09:10
by the way, the job descriptions calls for:
ahh, the sweet sweet irony.
14. May 2012 at 17:30
JN, I was talking about the impact of increases in the monetary aggregates.
dwb, Would one out of three be good enough for Thiel to hire me?
14. May 2012 at 18:21
dwb,
I’m pretty stumped by the numerous arguments for people to quit educating themselves! I would rather see locally credentialed entrepreneurial education for low income people, who have every reason to educate themselves for a lifetime. But I guess some folks are feeling threatened by too many smart people…
14. May 2012 at 21:26
@Scott,
Understanding economics is “helpful but not necessary” in this position so you are overqualified. I guess it’s better to be lucky than right. anyway, the JP Morgan gig is a better deal: big bonus if you are right, big retirement package of you lose 2Bn taxpayer dollars.
@Becky,
i’m stumped too, especially since he has a JD from Stanford.
18. May 2012 at 10:44
[…] lot of economists have made a big deal of Raghuram Rajan’s piece “The True Lessons of the Recession: The West […]