Question for Bailout Opponents

Suppose the bailout doesn't pass, in any form. What information would you need for you to conclude that you were more likely wrong than right in opposing it, at least in the sense of the economy being worse than it otherwise needed to be? (Setting aside purely moral libertarian concerns.) I.e., if GDP declines 25%, are you still sure that it would have been worse with a bailout? 99%? Is your rejection of the bailout falsifiable?

(I'm not supporting the bailout, by the way. But as someone who is continually, and increasingly, awestruck by the degree of certainty people express in their political beliefs, not to mention their predictions about the future state of the economy, I'm genuinely curious about what results would cause them to re-examine their opinions.)

Share this

How do you prove anything?

I happen to think the world is complex enough that you can't tell case by case whether X caused Y. There are simply so many other causal factors. Really the best you can do is look at a whole bunch of cases and try to exclude alternative explanations, to see if there is anything left.

This applies here as well. If the economy tanks, well, economies do tank from time to time. Who can really say what the cause is in this one case?

Furthermore, remember what Reagan did. It's long enough ago that I'm hazy on details, but the account that I remember is that around the beginning of his presidency, he implemented a fix which had the predictable short term effect of causing a recession, but a predictable longer term effect of fixing what had been ailing the economy in the seventies (e.g. "stagflation"). Think about and ask yourself, does the recession that occurred prove that Reagan's fix was mistaken? Well, no, not once you take into account the larger effect of Reagan's fix. Similarly here: if there's an economic downturn does that prove that whatever the government does, or fails to do, is a mistake? Not necessarily, because it may be a necessary part of a larger net positive effect.

Actually, Volcker, as head

Actually, Volcker, as head of the Federal Reserve, was the one who tightened up the money supply and ended stagflation. I'm not sure how much input Reagan had, as he wasn't even the one who appointed Volcker (that was Carter, though Reagan did reappoint him in 1983).

So it happened

I take that as confirmation. It doesn't matter who did it.

... and proved Keynes wrong

It was Volcker, and things did not behave as Keynes predicted. They did work out the way Austrian theory predicts however.

In the name of balance...

Shouldn't the question also be open to those in favour of the bailout - if it passes, by what measure would it be declared a success and over what period of time?

Absolutely, I'd like to hear

Absolutely, I'd like to hear from them too; I just haven't seen any around these parts.

Players

How is giving 700B to our worst gambling addicts going to do more good than bad? This is what I don't see, but I could be wrong.

Why the Bailout Must Be Tried

The most certain effect of the financial crisis is a shortage of short term funds which are the life blood of the international economy. This has already happened. It is unknown what will correct this. When Lehman tanked it removed $630 billion from the world economy. All investment bankers were getting rich by leveraging their short term money into now worthless sub prime long term derivatives.

When Lehman went under it soon took AIG down because it insured these short term deposits. The resulting panic has huge international implications.

“Officials were particularly concerned with two areas: the credit-default-swap market, where players buy and sell insurance against defaults on corporate and other bonds; and the so-called repo market, where Wall Street banks fund their investments by putting up securities as collateral for short-term loans.

The Fed had been pushing Wall Street firms for months to set up a new clearinghouse for credit-default swaps. The idea was to provide a more orderly settlement of trades in this opaque, diffuse market with a staggering $55 trillion in notional value, and, among other things, make the market less vulnerable if a major dealer failed. But that hadn't gotten off the ground.”

Due to low savings rates the US finances much of its debt by borrowing from foreigners. The US had the reputation as the most stable financial system in the world, but no more. This will inevitably make borrowing harder and more expensive, thus slowing the economy.

I don’t know if this is all available but try : http://online.wsj.com/article/SB122266132599384845.html

Dave

Fiat justicia ruat

Fiat justicia ruat caelum.

Bail the banks out with your own money.

Right now short term funds

Right now short term funds are not the life and blood but the crack cocaine of the international economy. What you're saying merely boils down to it will make the system "stable". Maybe it will but at what cost? Why should we value stability so much?

In capitalism the losers fail and are replaced by better competitors. This mean firms which are better at understanding collateral risk, exposure to systemic events, which are more stable, smarter. Now we can whine about stability, bail out every firm, prevent anyone from firing anyone, heck ! Let's stabilize the price of goods with price control... And what about stocks ? Who needs a volatile stock market. Banking socialism, like any form of socialism is a mistake.

Besides what you advocate is basically a heist. It's berely a stretch to consider that your advocacy of this plan is criminal in itself.

Simple calculations made

Simple calculations made using the Bitur-Camember Law and the de Poncins theorem show the bailout would have produced at the very least a 16% unemployment rise, or 24 560 000 job losses. Actual effect could be up to the double.

And using my very own empirically-calibrated Friedman Law on the GDP associated with would-be bought-out businesses, I'd anticipate a loss of 1.4 trillion $ in GDP, or 11%. Once again, this is a low boundary.

Why unemployment ?

Why unemployment ? Unemployment can only be created by directly interfering with employer / employee relationship.

Why unemployment:

By removing money from the pockets of employers and money from the pockets of employees who'd otherwise consume and incite entrepreneurs to employ, public spending affects employment.

Remove an amount of money equal to one mean net income, and you remove one job, statistically. That's the de Poncins Theorem.

Must be short term

Surely in the longer term there would be an adjustment as wages go down across the board.

You're making the lump of

You're making the lump of labor fallacy.

Not really: the public

Not really: the public spending itself induces a waste roughly equal to its amount (Bitur-Camember Law), and that's where the efforts and wealth (and job opportunities) are destroyed.

Regardless, wages adjust

Regardless, wages adjust downwards.

You're right, and I realized

You're right, and I realized that in my response to Curunir's wager proposal. The decrease in jobs is really a decrease in total job income, so the number of jobs or the mean wages or both can adjust down.

If I understand you

If I understand you correctly, you're saying that there would be at least 22% unemployment (it's 6% now, plus adding on another 16%) if there's a bailout passed. It seems as of today that the bailout might end up being passed after all. Care to place a friendly wager?

(You really should give me odds too, since you're claiming that as a lower bound.)

Why, sure, a friendly wager

Why, sure, a friendly wager sounds fine.

However let me detail what I'm saying will happen from the bailout: the rise in unemployment will happen gradually, and because I think we already have seen about 2% of the whole 16% it should cause, we're talking about an estimate of 20% total unemployment, or a decrease of 16% in net mean income (constant dollar), or a mix of the two. That figure will be attained gradually while the new economic conditions of post-bailout settle in, I'd say maximum 5 years, more likely 3. Are you interested in betting over such a long period of time ? Also, artificial employment such as large "public works" additional spending that is enacted afterwards should be discounted.

Could you define "net mean

Could you define "net mean income (constant dollar)" a little more precisely for me?

Bryan Caplan takes his stand

Bryan Caplan takes his stand on the reverse question, what if the bailout passes:

If the bail-out happens, and unemployment stays below 8% for the next two years, I'm going to become less confident that the bail-out prevented disaster. After all, even a near-miss with disaster should look pretty ugly. Alternately, if the bail-out happens, and unemployment hits 8% or higher during the next two years, I'm going to become more confident that the bail-out prevented disaster. I still won't be convinced, but I'll be less skeptical than I am now.

I really should get around to answering my own question sometime.

Caplan also said, in the

Caplan also said, in the post immediately following that one:

Financial markets are always complicated. Now they're more complicated than ever. But the more complicated the world gets, the more you need your common-sense skepticism.

In other words, when someone asks you for $700B for reasons he can't clearly explain, you hang up the phone.

This sort of question makes

This sort of question makes me appreciate Austrian anti-empiricism.

We are constantly updating our understanding of how the world works based on what has already happened. But it doesn't seem clear to me that we are very good predictors of the future, especially in the realm of the social sciences.

Suppose there is no bailout, GDP declines by 25% over the next few years (how many years?), and this leads you to conclude that a bailout would have been preferable. But you could have just as easily concluded that GDP would have declined anyway, bailout or not, and the bailout would have made things even worse.

Or, you could just expand your time frame: Suppose there is a bailout, GDP initially declines by 25% over the next few years (how many years?), GDP subsequently increases by 50% over a larger number of years (how many years?), and this leads you to conclude... what, exactly?

Is the idea that "falsifiable prediction is an appropriate criterion to use in economics" itself falsifiable?

I agree with you (and

I agree with you (and Constant, above) that you'll never know to certainty what caused a particular economic outcome, but that's not what I asked. My question was what would make you think that you were "more likely wrong than right". And Bryan's totally right in his post: There's just no way that getting more information shouldn't cause some updating of beliefs.

I also agree that empiricism should be tempered by theoretical knowledge. There are very robust predictions of economics, say, a downward sloping demand curve, and you're going to need awfully strong evidence to convince me you've found a counterexample. I just do not think that any predictions about what is going to happen as a result of the recent troubles are even close to that certain, and frankly I'm disinclined to give much credibly to people who talk with surety about it.

And I answered you

My question was what would make you think that you were "more likely wrong than right". And Bryan's totally right in his post: There's just no way that getting more information shouldn't cause some updating of beliefs.

I answered you. I told you that it takes a lot of evidence - a lot more, I think, than you can get from a single event, to rationally do more than slightly modify your degree of belief. I wrote:

I happen to think the world is complex enough that you can't tell case by case whether X caused Y. There are simply so many other causal factors. Really the best you can do is look at a whole bunch of cases and try to exclude alternative explanations, to see if there is anything left.

This is absolutely standard practice. I am not describing anything unusual. Look, for example, at Lott and his critics. Lott says More Guns Less Crime. His critics say otherwise. Did any of these people look at a single gun purchase by a single person to see whether more guns cause less crime? No, of course not. They looked, in effect, at an enormous number of gun purchases all around the country.

Or look at clinical trials. Would you seriously ask a scientist working for pharmaceutical company:

Suppose a sick patient receives your experimental drug. What information would you need for you to conclude that the drug was more likely ineffective than effective, at least in the sense of the patient being no better than he otherwise needed to be? I.e., if the patient gets sicker, are you still sure that the illness would have been worse without the drug? Much sicker? Dies? Is your belief in the efficacy of the drug falsifiable?

The answer here is of course that:

1) Yes, the belief (hypothesis) is falsifiable, but

2) It takes a lot more than experience with just one patient to decide whether the drug is effective or ineffective. That's why clinical studies frequently involve tens, or even hundreds or thousands of subjects.

My question was what would

My question was what would make you think that you were "more likely wrong than right".

This question is incoherent to me. Wrong or right about what? A prediction? Which prediction? Over what time frame? My belief is that government bailouts are a bad idea in general, not least because doing so changes the rules of the game in a way that is detrimental to long term economic outcomes. But since I don't know which direction the economy is headed in the short term to medium term, I don't see how I can predict which changes would be a result of a bailout (or no bailout) and which changes would have happened anyway.

My belief that bailouts are a bad idea is not based on my certainty of what the economy will look like in the short to medium term with or without a bailout. It's based on what I understood the rules of the game to be in the past, under state socialist conditions and under market conditions. Market conditions appeared to lead to more desirable results.

If I was exposed to enough instances of market conditions leading to undesirable results and enough instances of state socialist conditions leading to desirable results, I would update my relative beliefs about the desirability of spontaneous order and planning accordingly. Of course, in a mixed economy it is more difficult to make this analysis, for it is more difficult to tease out which effects were a result of market causes and which effects were a result of government causes.

A praxeological

A praxeological demonstration would do.

Falsifiability

First of all, amongst professional philosophers of science, Popper's "falsifiability" criterion has been recognized as an inadequate demarcation principle for four decades or so... so, by invoking it, you are not appearing savvy, but way behind the times.

Secondly, even if we accept Popper as is, there is, of course, no set of facts which could falsify "the bailout was bad" (or "the bailout was good"), because we aren't dealing with isolated causal factors, and there is no way we could re-run the situation taking the other choice. So the question is nonsense.

Falsifiability

Gene,

I imagine Micha writing exactly your words. So if you were trying to point something out to him then you are preaching to the choir.

"First of all, amongst professional philosophers of science, Popper's "falsifiability" criterion has been recognized as an inadequate demarcation principle for four decades or so... so, by invoking it, you are not appearing savvy, but way behind the times."

First of all, amongst professional economists, Mises' anti-empiricism has been recognized as an inadequate for six decades or so... so, by invoking it, you are not appearing savvy, but way behind the times.

"Secondly, even if we accept Popper as is, there is, of course, no set of facts which could falsify "the bailout was bad" (or "the bailout was good"), because we aren't dealing with isolated causal factors, and there is no way we could re-run the situation taking the other choice. So the question is nonsense."

I would say that the question "is the bailout good or bad" is not a scientific question in the first place, so of course it's not going to be falsifiable.

In the long run we ARE running the experiment again and again with different sets of variables. We know these variables to some degree. They aren't complete unknowns. It's still an empircal process the only problem is that we are not in control of the experiments we can do. We must work with the experiments that history gives us.

The Austrians didn't just deduce all their theory a priori based on a knowledge of human behavior. That's just nonsense. Market behavior is just too complex a beast for someone to just deduce aprior from human action. Did some spear throwing genius living in a barter economy four thousand years ago deduce monetary theory from mere human action? Of course not.

First there was observation of what was happening at a systemic level. Historically the experiment had been run hundreds of times. Rome debased their currency, France did, the Neatherlands had an influx of silver, and so forth. The Austrians took these observations and tried to explain them by producing an economic model.

That is already an empirical connection. Had no economist ever observe fractional reserve banking, or the resulting business cycles then it is most likely that they never would have deduced that one would lead to another. In fact, economists still don't get it even when given the model. Some (most?) really don't comprehend dynamic systems very well.

Especially dynamic systems with emergent behavior like economies. Based merely on "human action" who would even think, firstly that fractional reserve banks would arise? That's not obvious in the first place. If it were then people would have immediately invented them (or immediately shunned them) when any form of money was created.

Why didn't the ancient Greeks figure this out long ago? Did they have an inferior grasp of deductive logic? No. It's because having deduction doesn't mean that all the answers in the world just fall out of the bag immediately.

Furthermore, even if you were one of the clever guys who first figured out fractional reserves and cashed in on it doesn't mean that you figured out all the consequences.

So no economics of any sort was ever deduced aprior from some base axioms. It was empiricism all along. Falsifiability being part of the answer, because if your theory is not falsifiable, well then it really has nothing to say about the world, does it. You can never tell if it's true or if it's false, ever.

Part of the falsifiable model that is made is the axioms themselves. The assumptions about how humans act. Deduction can fail in that regard also, not just simple mistakes, or understandable mistakes in complex systems.

You need the empiricism not just for checking your logic, but also for checking your assumptions. Falsification is a means to check. Not a foundational means but a means none-the-less.

(I'm posting now without further clarification that I think needs to be done, but I've got to go pick up my son, and this is just a comment, and not a dissertation)

and there is no way we could

and there is no way we could re-run the situation taking the other choice

Actually there might be in the far future, as it is physically possible to run counter factual experiments using quantum mechanics :D