Great article. 12 step programs based on the 12 steps/traditions of AA are an ideal incubator for contrarian ideas. This article reminded me of the story of evolution pertaining to black/white moths on white oaks during the industrial error. Prior to industry, white moths dominated bc black moths highly contrasted to the white oak trees & were easily spotted by predators. Yet as industry kicked in, the oaks became covered by soot. Suddenly the black moth became dominant. So after centuries of one trait dominance, survival demanded a change. Amazing! In a similar fashion, I've found Alcoholics Anonymous to be a community where my ideas of emotional sobriety can be nurtured. Ultimately, if I can help the newcomer find sobriety, then my ideas take dominance, regardless of that my sobriety is less than one year old. In this way, 12 step programs are unions through and through, allowing people to harmonize & essentially thwarting group think. Amazing, amazing stuff. If anyone is curious, feel free to tweet me @vovrv.
I would distinguish between rational groupthink and emotional groupthink.
Groupthink can be rational. A group opinion is more likely to be right than my opinion, so it is rational to adopt the group opinion. But if everyone does this, it leads to information cascades, where an idea that gets a head start becomes entrenched even though nobody would believe it on their own.
Avoiding rational groupthink requires people to distinguish between their informed opinion, what they believe after hearing from the group (which should typically match the group opinion); and their naive opinion, what they would believe if they didn't know the group opinion. The group consensus should be based on combining naive opinions rather than informed opinions. People need to reveal both and make clear which is which. They would still use their informed opinions to make decisions, but they need to also expose their naive opinions. (Unfortunately I can't see how to apply this to markets, where only the informed opinion will guide people's trades, leading to the familiar problem of bubbles.)
Emotional groupthink is where people try to punish those who depart from the consensus. I confess I don't understand this part of the problem. But perhaps some fraction of groupthink is rational and could be addressed as I have described.
Aren't prediction markets your neglected contrarian idea? What's the 'prediction market' that will overcome the existing groupthink opposed to prediction markets that will impel people to adopt prediction markets?
A prediction market itself could possibly provide prestige prizes.
I haven't read Robin's tracts. He's probably addressed the following. But I think you could have an option to pre-register publicly whatever bets you like. Whoever makes $10 million in one field publicly will start to be listened to, you can bet. Not to mention a billion. People with worse-than-average predictions are already incented to depart the prediction market because they lose money - eventually, anyway, though not immediately - and the best are already incented to dominate the market by putting more money in. Yet people will probably look to the opinion of such magnates as well as looking at the position of the prediction market itself, since the latter's signal is slightly worsened by entries from people who haven't decided yet whether they are consistently beating the average or not.
It's like Warren Buffet. When Buffet says something about finance, business, or the economy, one might suspect what he says of possibly being self-interested. But suspect him as being full of hot air? Not likely to be true - at least not compared to just about anyone else.
Obviously prediction markets are good, but another incentive for accurate contrarianism might be high prestige-prizes for people who have gone out on a limb and were right about something important when most of their colleagues thought they were wrong.
if you want not to have another bubble like the housing bubble, you have to make sure that there are well-paying high prestige jobs for the contrarians. not going to happen.
Fighting 'group think' often involves a group effort. If one is recruiting for this effort, who is a better candidate - someone who is an established authority whose opinion will be respected, or someone who is younger (professionally) with less invested in the current status quo? Of course 'better' can be evaluated in terms of the costs of recruitment, as well as the potential reward for the 'cause'...
And what about those in the middle of their careers? What is their likelihood of jumping ship and fighting the status quo, and being productive at doing so? How does one evaluate who is likely to be 'good' in the cause of fighting group think in a particular area?
Last night I posted on the NYT blog a structural reform of the NIH budget disbursement. It's yet to be mod-approved. This is more a brainstorm than something I'm confident in.
Basically (with some revision from last night's version) this system hands NIH grant approval over to 200 dictators. It partakes, somewhat, from Mencius Moldbug.
- The premise is that committees suck. This is just an intuition I have. I have no data.
- Determine the top 20% of NIH-funded biomed profs. This is by vote of all the profs. Each has one vote. Those you vote for may not fund you at any time in the future.
- From the thousands of electi, the 200 dictators are selected at random for 15 year terms. The dictators will get half the NIH budget, with the other half controlled the old way. Thus each dictator controls, absolutely, 1/400 of the ~30 billion budget, about 100 million per year. He funds grant proposals at his sole discretion, from any biomed field, consulting anyone or no one, requiring whatever application procedures he wants.
- So that people will want to be dictators, there should be a way for them to continue their own research programs. Their historic mean level of funding will be doubled, and awarded automatically for the 15 years. Since they will have little time for their research programs, their labs will be mostly run by an appointee, selected at their sole discretion, who can be fired at will.
- Possibilities for the impeachment of dictators are extremely conservative.
- Note that even if a number of dictators do poorly, it doesn't matter that much. The eggs are certainly not in one basket.
- Viola. A lot like what we have now, only all the power is several/individual rather than joint.
- Something comparable could be used by academic departments for hiring and tenure decisions.
What about the argument that government altered the market - via implicit government financial backing of Fanny Mae, by Bush the Lesser's deregulation of leveraging, via interest rates, by letting banks merge only if they implied with gov's egalitarianist demands re loan approvals. Even if you stipulate that all these things were good, that doesn't change the fact that they altered the market.
In contrast, the prediction market on future housing prices is a pure and free market. It incorporates expert opinion on how the actual housing market is being impacted by not being free and pure.
But if instead of proposing some structural, cultural, or institutional reform to achieve this, you just argue “there can be group think, so my contrarian view deserves more attention” you are really just using group think as an excuse to try to jump the queue.
This seems like something of a strawman - I don't see it being implied that mere susceptibility to groupthink should lead to an idea's being dismissed, but rather to a marginal skepticism of arguments that draw on expert consensus for support. In fact, I think Wade's passage is meant to highlight an example of the asymmetrical application of this weighting to consensuses we don't like, which is exactly what you're complaining about in this thread.
It may be possible to distinguish two problems and solve only one without necessarily solving the other.
1) How can we cause the groups in question to overcome groupthink?
2) How can individuals or subgroups overcome groupthink?
It's not always clear to me whether in discussions about groupthink (of which there are many here) the intent is the more ambitious (1) or the less ambitious (2). A prediction market seems to be aimed at (1), but the many past discussions about overcoming bias (which are now mostly moved over to "less wrong", making the current blog somewhat misnamed) seem to be aimed at individual self-improvement and thus at (2).
It's hilarious that, after identifying the housing bubble as a prime example of groupthink, you then propose prediction markets as the solution. The housing market is a prediction market. That's how the groupthink occurred. That's how we got into this mess.
The article: Ben "Systemic Risk" Bernanke proves that Bernanke knowingly maintained a strict monetary policy long after he knew of the sub prime problem as he knew it would cause of the "Depression".
It shows that he probably engineered it on purpose!
If you want to sleep tonight, Don't Read It!
"In contradiction to the prevalent view of the time, that money and monetary policy played at most a purely passive role in the Depression, Friedman and Schwartz argued that "the [economic] contraction is in fact a tragic testimonial to the importance of monetary forces" (Friedman and Schwartz, 1963, p. 300)......
The slowdown in economic activity, together with high interest rates, was in all likelihood the most important source of the stock market crash that followed in October.
In other words, the market crash, rather than being the cause of the Depression, as popular legend has it, was in fact largely the result of an economic slowdown and the inappropriate monetary policies that preceded it.
Of course, the stock market crash only worsened the economic situation, hurting consumer and business confidence and contributing to a still deeper downturn in 1930."
Governor Ben S. BernankeMoney, Gold, and the Great Depression.At the H. Parker Willis Lecture in Economic Policy, Washington and Lee University,Lexington, Virginia.March 2nd, 2004
You can read also: Preparing for the Crash, The Age of Turbulence Update: 22/07/09., which tries to accomplish Greenspan Mission Impossible:
That is mission impossible. Indeed, the international financial community has made numerous efforts in recent years to establish such oversight, but none prevented or ameliorated the crisis that began last summer. Much as we might wish otherwise, policy makers cannot reliably anticipate financial or economic shocks or the consequences of economic imbalances. Financial crises are characterised by discontinuous breaks in market pricing the timing of which by definition must be unanticipated - if people see them coming, then the markets arbitrage them away."
Alan GreenspanThe Age of Turbulence: Adventures in a New World [Economic Order?].
Plea for a New World Economic Order. explains the nature and causes of economic depressions and proposes a plausible alternative solution.
Great article. 12 step programs based on the 12 steps/traditions of AA are an ideal incubator for contrarian ideas. This article reminded me of the story of evolution pertaining to black/white moths on white oaks during the industrial error. Prior to industry, white moths dominated bc black moths highly contrasted to the white oak trees & were easily spotted by predators. Yet as industry kicked in, the oaks became covered by soot. Suddenly the black moth became dominant. So after centuries of one trait dominance, survival demanded a change. Amazing! In a similar fashion, I've found Alcoholics Anonymous to be a community where my ideas of emotional sobriety can be nurtured. Ultimately, if I can help the newcomer find sobriety, then my ideas take dominance, regardless of that my sobriety is less than one year old. In this way, 12 step programs are unions through and through, allowing people to harmonize & essentially thwarting group think. Amazing, amazing stuff. If anyone is curious, feel free to tweet me @vovrv.
"Emotional groupthink is where people try to punish those who depart from the consensus. I confess I don’t understand this part of the problem."
read my next post (should be up in a day or so and I think it has a good metaphor for this).
I would distinguish between rational groupthink and emotional groupthink.
Groupthink can be rational. A group opinion is more likely to be right than my opinion, so it is rational to adopt the group opinion. But if everyone does this, it leads to information cascades, where an idea that gets a head start becomes entrenched even though nobody would believe it on their own.
Avoiding rational groupthink requires people to distinguish between their informed opinion, what they believe after hearing from the group (which should typically match the group opinion); and their naive opinion, what they would believe if they didn't know the group opinion. The group consensus should be based on combining naive opinions rather than informed opinions. People need to reveal both and make clear which is which. They would still use their informed opinions to make decisions, but they need to also expose their naive opinions. (Unfortunately I can't see how to apply this to markets, where only the informed opinion will guide people's trades, leading to the familiar problem of bubbles.)
Emotional groupthink is where people try to punish those who depart from the consensus. I confess I don't understand this part of the problem. But perhaps some fraction of groupthink is rational and could be addressed as I have described.
Aren't prediction markets your neglected contrarian idea? What's the 'prediction market' that will overcome the existing groupthink opposed to prediction markets that will impel people to adopt prediction markets?
Ha!
I agree with all you guys, groupthink is bad.
A prediction market itself could possibly provide prestige prizes.
I haven't read Robin's tracts. He's probably addressed the following. But I think you could have an option to pre-register publicly whatever bets you like. Whoever makes $10 million in one field publicly will start to be listened to, you can bet. Not to mention a billion. People with worse-than-average predictions are already incented to depart the prediction market because they lose money - eventually, anyway, though not immediately - and the best are already incented to dominate the market by putting more money in. Yet people will probably look to the opinion of such magnates as well as looking at the position of the prediction market itself, since the latter's signal is slightly worsened by entries from people who haven't decided yet whether they are consistently beating the average or not.
It's like Warren Buffet. When Buffet says something about finance, business, or the economy, one might suspect what he says of possibly being self-interested. But suspect him as being full of hot air? Not likely to be true - at least not compared to just about anyone else.
Obviously prediction markets are good, but another incentive for accurate contrarianism might be high prestige-prizes for people who have gone out on a limb and were right about something important when most of their colleagues thought they were wrong.
if you want not to have another bubble like the housing bubble, you have to make sure that there are well-paying high prestige jobs for the contrarians. not going to happen.
Fighting 'group think' often involves a group effort. If one is recruiting for this effort, who is a better candidate - someone who is an established authority whose opinion will be respected, or someone who is younger (professionally) with less invested in the current status quo? Of course 'better' can be evaluated in terms of the costs of recruitment, as well as the potential reward for the 'cause'...
And what about those in the middle of their careers? What is their likelihood of jumping ship and fighting the status quo, and being productive at doing so? How does one evaluate who is likely to be 'good' in the cause of fighting group think in a particular area?
Last night I posted on the NYT blog a structural reform of the NIH budget disbursement. It's yet to be mod-approved. This is more a brainstorm than something I'm confident in.
Basically (with some revision from last night's version) this system hands NIH grant approval over to 200 dictators. It partakes, somewhat, from Mencius Moldbug.
- The premise is that committees suck. This is just an intuition I have. I have no data.
- Determine the top 20% of NIH-funded biomed profs. This is by vote of all the profs. Each has one vote. Those you vote for may not fund you at any time in the future.
- From the thousands of electi, the 200 dictators are selected at random for 15 year terms. The dictators will get half the NIH budget, with the other half controlled the old way. Thus each dictator controls, absolutely, 1/400 of the ~30 billion budget, about 100 million per year. He funds grant proposals at his sole discretion, from any biomed field, consulting anyone or no one, requiring whatever application procedures he wants.
- So that people will want to be dictators, there should be a way for them to continue their own research programs. Their historic mean level of funding will be doubled, and awarded automatically for the 15 years. Since they will have little time for their research programs, their labs will be mostly run by an appointee, selected at their sole discretion, who can be fired at will.
- Possibilities for the impeachment of dictators are extremely conservative.
- Note that even if a number of dictators do poorly, it doesn't matter that much. The eggs are certainly not in one basket.
- Viola. A lot like what we have now, only all the power is several/individual rather than joint.
- Something comparable could be used by academic departments for hiring and tenure decisions.
I was gonna say, "yeah - what he said." But....
What about the argument that government altered the market - via implicit government financial backing of Fanny Mae, by Bush the Lesser's deregulation of leveraging, via interest rates, by letting banks merge only if they implied with gov's egalitarianist demands re loan approvals. Even if you stipulate that all these things were good, that doesn't change the fact that they altered the market.
In contrast, the prediction market on future housing prices is a pure and free market. It incorporates expert opinion on how the actual housing market is being impacted by not being free and pure.
But if instead of proposing some structural, cultural, or institutional reform to achieve this, you just argue “there can be group think, so my contrarian view deserves more attention” you are really just using group think as an excuse to try to jump the queue.
This seems like something of a strawman - I don't see it being implied that mere susceptibility to groupthink should lead to an idea's being dismissed, but rather to a marginal skepticism of arguments that draw on expert consensus for support. In fact, I think Wade's passage is meant to highlight an example of the asymmetrical application of this weighting to consensuses we don't like, which is exactly what you're complaining about in this thread.
It may be possible to distinguish two problems and solve only one without necessarily solving the other.
1) How can we cause the groups in question to overcome groupthink?
2) How can individuals or subgroups overcome groupthink?
It's not always clear to me whether in discussions about groupthink (of which there are many here) the intent is the more ambitious (1) or the less ambitious (2). A prediction market seems to be aimed at (1), but the many past discussions about overcoming bias (which are now mostly moved over to "less wrong", making the current blog somewhat misnamed) seem to be aimed at individual self-improvement and thus at (2).
It's hilarious that, after identifying the housing bubble as a prime example of groupthink, you then propose prediction markets as the solution. The housing market is a prediction market. That's how the groupthink occurred. That's how we got into this mess.
The article: Ben "Systemic Risk" Bernanke proves that Bernanke knowingly maintained a strict monetary policy long after he knew of the sub prime problem as he knew it would cause of the "Depression".
It shows that he probably engineered it on purpose!
If you want to sleep tonight, Don't Read It!
"In contradiction to the prevalent view of the time, that money and monetary policy played at most a purely passive role in the Depression, Friedman and Schwartz argued that "the [economic] contraction is in fact a tragic testimonial to the importance of monetary forces" (Friedman and Schwartz, 1963, p. 300)......
The slowdown in economic activity, together with high interest rates, was in all likelihood the most important source of the stock market crash that followed in October.
In other words, the market crash, rather than being the cause of the Depression, as popular legend has it, was in fact largely the result of an economic slowdown and the inappropriate monetary policies that preceded it.
Of course, the stock market crash only worsened the economic situation, hurting consumer and business confidence and contributing to a still deeper downturn in 1930."
Governor Ben S. BernankeMoney, Gold, and the Great Depression.At the H. Parker Willis Lecture in Economic Policy, Washington and Lee University,Lexington, Virginia.March 2nd, 2004
You can read also: Preparing for the Crash, The Age of Turbulence Update: 22/07/09., which tries to accomplish Greenspan Mission Impossible:
That is mission impossible. Indeed, the international financial community has made numerous efforts in recent years to establish such oversight, but none prevented or ameliorated the crisis that began last summer. Much as we might wish otherwise, policy makers cannot reliably anticipate financial or economic shocks or the consequences of economic imbalances. Financial crises are characterised by discontinuous breaks in market pricing the timing of which by definition must be unanticipated - if people see them coming, then the markets arbitrage them away."
Alan GreenspanThe Age of Turbulence: Adventures in a New World [Economic Order?].
Plea for a New World Economic Order. explains the nature and causes of economic depressions and proposes a plausible alternative solution.