Why asset bubbles are a part of the human condition that regulation can’t cure | Hacker News Reader

Why asset bubbles are a part of the human condition that regulation can’t cure

theatlantic.com

47 pointsiamelgringo28 comments

The title is a unbelievable deceptive because, unless I missed something, all of these experiments were done on completely unregulated toy markets.

I would be curious about what would happen if, instead of rewarding people directly with the returns on their trades and their dividends, you gave them a baseline salary that they can never go under, then gave them a bonus based on a percentage the gained market value of their holdings during any particular round, but ended their game if they ranked below 50% of the other players for two rounds in a row then gave them severance equal to five times their baseline salary plus their average bonus per round over the entire game.

In short, I'd be interested in how far fiduciary investment bubbles would deviate from the bubbles produced by direct investors in these same safe markets, and whether changing any of the variables above could bring them closer together.

Wow, my first sentence was made incoherent by backediting: "The title is unbelievably deceptive because..."

If asset bubbles are part of the human condition, then why have they been so rare? There was tulip mania in Holland in 1637. The South Sea Bubble of 1719-1721. The canal overbuilding bubble in England in the 1820s. In the USA, there was the Florida land boom in the 1920s, ending in 1926, and then from 1926 to 1929 there was the stock market bubble. There was the Japanese real estate bubble of the 1980s, and a similar, smaller bit of overbuilding in the USA. Then there was the dotcom bubble in the USA stock market during the 1990s, and then the real estate boom of the 00s.

Those are the big ones.

Why were there no bubbles before 1637? In a word: government. Under a feudal regime, people are not free to do what they want with their money, therefore there were no bubbles. People were not free to bid up the prices of assets - most assets are illiquid under feudal regimes. The fact that a heavy-handed monarch can suppress markets and therefore make bubbles impossible is a fairly strong argument toward government effectiveness in stopping bubbles. Likewise, the lack of asset bubbles in the USA from 1932 to about 1982 also suggests the effectiveness of government in suppressing bubbles.

Of course, one might argue that we want bubbles, that such freedom of investment has positive consequences, or that heavy-handed government, while creating some benefits, also imposes large costs that outweigh whatever benefits are being offered. But that would be a separate issue.

Actually, you've missed quite a few big ones, have you read Kindlebarger?

http://www.amazon.com/Manias-Panics-Crashes-Financial-Invest...

I wouldn't say that bubbles are rare at all, and it wouldn't surprise me if there were lots of them before 1637 that just aren't well recorded. The limited scope of trade in those days would mean that such events weren't as disruptive as they are today.

I thought about also listing the panics, and I started to, but then I realized they were off-topic. This article is about asset bubbles, not about panics. Therefore, the crash 1825, in England, the crash of 1873, in the USA, the crash of 1907 in the USA - all of those are off-topic.

I suppose one could make the argument that before a panic, bank accounts are overvalued assets.

That's just it, you could define it however you want, but as soon as you do, the whole conversation falls apart. If you define all panics as asset bubbles, then pretty soon you need a new term to mean what "asset bubble" used to mean. I think a lot of these conversations fall apart over semantics. So I tried to use the phrase asset bubble in something like the sense that the article above seems to.

I'm not sure what you mean when you say If you define all panics as asset bubbles, then pretty soon you need a new term to mean what "asset bubble" used to mean. "Asset bubble" must always be followed by "panic" by definition, otherwise you're actually seeing an increase that could be called "fairly priced" instead of "inflated." Look at the five year chart on AAPL, is it experiencing a bubble at a P/E of around 21, or is the stock somewhere close to a fair valuation? What about AMZN trading at a P/E of around 63? As far as I'm concerned, there's no way to be certain of the answer to this question until the future happens: if the stock crashes, then it was a bubble, otherwise it was fairly priced. If you have a better answer to this question than that, then you are either a wealthy (or soon to be wealthy) portfolio manager, or you should seriously consider that as an occupation if making a lot of money is your goal.

Yes, but not every panic is caused by an asset bubble. The panics that are caused by asset bubbles are a subset of the of panics. There are other panics that have other causes - both supply side shocks and demand side shocks can lead to a panic. Thus, I left out a lot of panics, which were not caused by asset bubbles.

What if AMZN had a P/E of 630? What if it had a P/E of 6300? What if it had a P/E of 63,000? What if had an infinite P/E, by which I mean large losses and a sky high price?

You need to be careful about what examples you choose. If you only choose reasonable examples, then you are really not making a strong case.

Actually, it was the banks lending money to whoever was investing in the overvalued asset du jour and then not getting that money back that actually caused the panic. In the days before the FDIC, any significant question that was raised as to a bank's solvency was a death sentence because depositors had to be among the first at door to get their money out or it was all lost.

A lot of the things you cite became bubbles BECAUSE of government involvement. In the US and Japan, government is subsidizing the cost of borrowing.

It is not freedom that is the problem. It is the socialization of risk that is the problem. In the US, the government insures private pension plans, guarantees home loans, bails out automakers and financial "innovators". There's no incentive to be prudent.

I don't think socialization of risk was a factor in the tulip mania that gripped Holland in the 1630s. You can argue that socialization of risk has been factor that has made bubbles much more common over the last 30 or 40 years. That would go to my what I said before - government can stop bubbles, if it wants to. Or it can start them.

Without government intervention, bubbles remain rare, which undercuts the thesis of the above article.

I wonder why this was down voted?

Does someone think the socialization of risk was a factor in the tulip mania that gripped Holland in the 1630s?

Or that asset bubbles are rare unless the government gets involved?

What, exactly did someone disagree with?

The causes of each bubble are different. Some theorize that monetary expansion and a conversion of futures contracts for tulips into options contracts played a significant role in the tulip mania.

Government can stop bubbles, disease, cheating on one's spouse, failure to wash dishes and a thousand other sins by simply ending freedom. Big deal. Any kidnapper can do that.

You are right to say that it is a big deal. Your point I think largely disproves the article referenced above. Clearly, government regulation can stop asset bubbles. It has before and I'm sure it will again. The government can stop anything by ending freedom.

The problem with government involvement is that it bails out or assumes the costs of disastrous behavior as it regulates to reduce risk.

Then over time the lobbyist weaken the regs and another disaster develops, even bigger this time, with the taxpayer holding the bag. Only this time everybody piles in because they know the government will clean it up.

And the government, taking the path of least resistance, plays along, amplifying the problems, even passing more laws to subsidize the risky behaviors. They'll get you a tax write-off, help the first time buyers, lower the interest rates, get rid of the capital gains tax. You name it. But the last thing they will do in the midst of a bubble is take action to counter it.

Then once the bubble collapses, everybody else pushes for regulation again to restart the whole cycle, most never questioning whether government should be involved in the first place.

And government, in control of the debate, finds the exploiters of their policies and offers them up as the chief perpetrators. And live to trash the economy another day.

I think we are in agreement that government has the power to stop bubbles with regulation, if it wants to.

As I wrote above: "Of course, one might argue that we want bubbles, that such freedom of investment has positive consequences, or that heavy-handed government, while creating some benefits, also imposes large costs that outweigh whatever benefits are being offered. But that would be a separate issue."

Are you responding to the article above?

The people in their study (a group of 12, "bunch of volunteers, usually undergraduates but sometimes businesspeople or graduate students") don't seem like a group that would actually understand the value of their assets despite the value being explicitly clear for purposes of this study. They should do the study with a group of 100 financial experts and see if the results are the same.

That would be interesting. Get a bunch of sufficiently sensible people, and give them enough time to think things through, and I doubt they'll have a bubble.

Of course this doesn't solve the real-world bubble problem, where we have absolutely no idea what the future payout of an asset will be.

The people in their study don't seem like a group that would actually understand the value of their assets

The same could be said of the people who were trying to get rich quick by flipping houses. Every one I heard seemed to think housing prices could never go down.

It sounds like the biggest problem with this study (and with many markets) was that there was no way to efficiently sell short.

If it's only possible to make money when asset prices go up, some rational individuals will buy in even when they think the price is already too high, because until they own the asset, they can't profit from it.

There's a quote that goes "Your capacity to short a market is dwarfed by the market's ability to defy gravity".

The housing bubble began in 2002. If you had shorted it as soon as it was 10% "too high", you would have had to wait 6 years before cashing in. The yield/year is miserable, and meanwhile, your total net worth would have been negative around 2006. (More likely, the bank would have forced you to "cash out", at great loss)

Even if we grant the basic premise of the article (that regulation can't stop bubbles), that still doesn't rule regulation out. After all, it still doesn't prove that the government can't do anything to lessen the impact of a bubble.

It sounds like that experiment is flawed, but the article doesn't go into detail about what the participants are told. At any rate, you can't draw the conclusion it does from that experiment as given and then extrapolate it to the outside world.

IF you put any group of people together for an experiment, and give them nothing else to do, they are going to trade. especially fake money because there's no real risk. Unnecessary trading is going to result in abberant activity.

The bubbles we've had in our economy are due specifically to government forcing interest rates below their market level. You do that and you give institutions the power to make free money-- buy from the government at %x, invest it in low risk situations like mortgages at %Y and then pay back the government in devalued currency (due to inflation) at %Z. In each of those steps you make money...so, its a no-brainer.

If you want to stop having bubbles, remove the power to set interest rates from the federal reserve / government. Then the market has a chance to self regulate.

Right now the primary price signal that would keep bubbles from inflating is being held down, and the natural bubble prevention is being short circuited.

So, it is no wonder that bubbles are happening. The sad thing is that this is not a very controversial or difficult subject, but it is rare to find an article in a publication like the Atlantic that seems even aware of the effect of interest rates on interest bearing securities!

The reason that the mainstream media and government pretends like they aren't manipulating us into bubbles is very simple: Letting interest rates operate normally would also put a restraint on reckless government spending.

> If you want to stop having bubbles, remove the power to set interest rates from the federal reserve / government. Then the market has a chance to self regulate.

The article's assertion is that the problem is people themselves, so removing the Federal Reserve would make no difference. Anyway, we've already done that experiment. The Federal Reserve was not created until 1913 (as a result of the Panic of 1907). Prior to that, the unregulated U.S. economy had regular boom-bust cycles about every 10-20 years: The Panic of 1819, 1837, 1857, 1873, 1884, 1893, 1907.

Of those, speculation was part of the Panic of 1819, and directly responsible for the Panic of 1837. The Panic of 1857 was caused by banks lending too much (arguably a bubble in something). The Panic of 1893 was caused by speculation in railroads (it sounds similar to the recent speculation in housing). The Panic of 1907 was precipated by a failed short squeeze on United Copper, causing a quick mini-bubble where its price doubled in a day and then halved a few days later when it the squeeze failed. This left the Otto Heinze unable to pay his loans that he used to buy United Copper shares, causing a chain reaction in the banking system. The negotiations of J.P. Morgan in NY described in the wikipedia article (http://en.wikipedia.org/wiki/Panic_of_1907) sounds very reminiscent of what happened with the banks in the Panic of 2009.

(See http://en.wikipedia.org/wiki/Panic_of_1819, http://en.wikipedia.org/wiki/Panic_of_1837, http://en.wikipedia.org/wiki/Panic_of_1857, http://en.wikipedia.org/wiki/Panic_of_1893)

I think it is very clear that, left to themselves, financial markets cannot regulate themselves.

I don't know if the Federal Reserve has reduced bubbles, since we had the Panic of 1929 (too much borrowing), the 1987 Black Monday stock market crash, the Savings and Loans bankruptcies that may have resulted from that, the dot-com bubble of 2001, and the Panic of 2009 (housing). I'm not sure that the Federal Reserve has reduced bubbles (that being a function of human greed), but it has reduced the aftermath. Read how long the recession/depressions lasted in the 1800s. I'll take the Federal Reserve.

> IF you put any group of people together for an experiment, and give them nothing else to do, they are going to trade. especially fake money because there's no real risk. Unnecessary trading is going to result in abberant activity.

The implication is that it's not fake money:

    The difference between a good trading performance and a bad one is about $80
    for a three-hour session, enough to motivate cash-strapped students to 
    do their best.

Now, there's an argument to be explored there about whether the behaviour of a cash-strapped student accurately represents that of an institutional investor, but I think it's probably as close as it's reasonable to expect.

Reply on news.ycombinator.com