Is That Your True Rejection? by Eliezer Yudkowsky @ Cato Unbound

by XiXiDu1 min read7th Sep 201183 comments

44

PoliticsEvolutionEvolutionary PsychologyWorld Modeling
Frontpage

A response essay written by Eliezer Yudkowsky posted at Cato Unbound for the issue Brain, Belief, and Politics:

Is That Your True Rejection? by Eliezer Yudkowsky

Eliezer Yudkowsky suggests that the partial mutability of human traits is an auxiliary reason at best for Michael Shermer’s libertarianism. Take that fact away, and Shermer’s politics probably wouldn’t go with it. Yudkowsky says that his own small-l libertarian tendencies come from the long history of government incompetence, indifference, and outright malevolence. These, and not brain science, are the best reasons for libertarians to believe what they do.

Moreover, we make a logical error when we infer shares of causality from shares of observed variance; the relationship between nature and nurture is cooperative, not zero-sum. One thing, however, is clear: Human genetic variance is tiny, as indeed it must be for human beings all to constitute a single species. Environmental manipulation can only achieve so much in part because of this universal human inheritance.

The lead essay has been written by Michael Shermer:

Liberty and Science by Michael Shermer

Michael Shermer discusses scientific findings about belief formation. Beliefs, including political beliefs, are usually the result of automatic or intuitive moral judgments, not rational calculations. One cluster of those intuitions presumes that human nature is malleable; these usually produce a liberal politics. Another group of intuitions presumes that human nature is static; these tend to produce conservatism. But Shermer argues that humans really fall somewhere in between — malleable, within some important limits. He argues that this set of findings should produce a libertarian politics.

83 comments, sorted by Highlighting new comments since Today at 7:50 PM
New Comment
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

From the beginning of Eliezer's piece

I think my libertarianism rests chiefly on the empirical proposition—a factual belief which is either false or true, depending on how the universe actually works—that 90% of the time you have a bright idea like “offer government mortgage guarantees so that more people can own houses,” someone will somehow manage to screw it up, or there’ll be side effects you didn’t think about, and most of the time you’ll end up doing more harm than good, and the next time won’t be much different from the last time.

I think there's an interesting congitive phenomenon going on here. If you tell the average Joe that you propose to deregulate some industry or activity, they easily call to mind the potential negative consequences - "they'll just pollute as much as they want now" or "only the biggest companies will buy up the fishing quotas" etc. What the average Joe doesn't think about are the potential negative effects of regulation - unintended consequences and regulatory capture for example. Regulatory capture is a sort of funny case because though the concept was invented by a Marxist (if memory serves), only economists and libertarians s... (read more)

At the obvious risk of making this too political (in my defense, everyone else started it), I'm not sure "no one except us ever realizes how bad regulation is" is the real problem here.

In my metacontrarianism post I talked about how when someone discovers a counterintuitive idea that only smart people can understand, they can become very excited and use it as a membership test for the Secret Society Of Smart People, to the point where they enter a Happy Death Spiral and blow it out of all proportion.

(for example, it's easy to think of ways modern Western civilization is better than primitive tribal civilization - more wealth, less prejudice, fewer tapeworms. There are also a few counterintuitive hard-to-understand ways primitive tribal civilizations are better than us - closer family bonds, more attuned to the natural world. And so many intellectuals focus on these to the exclusion of everything else and talk about how Western society is hopelessly corrupt and evil. Where they should be saying "Western civ is great, but not quite as great as the average person might think," we instead get "Western civ is awful." This passes unchallenged because the goo... (read more)

3Matt_Simpson10y(I was carefully trying to not sound like "us vs. them." Did that not come through?) The only reason I think I partially disagree is that the reason many of the people who are against regulation are actually, well, against regulation, isn't because they've thought about the potential negative consequences. They'll cite things like freedom or a disdain communism, etc. And on the opposite political issues (e.g. military interventionism) they are more likely to be in favor of government heavy solutions. With that being said, your point is duly noted. There are probably more people who are against regulation because they see potential negative consequences than my comment acknowledged.
0michael6110yIn my view signalling a strong view for or against regulation suggests the need to properly think through the idea of contexts. The situation in which a regulation is applied is essential to determining its usefulness. For example; in my business we have had serious and expensive problems with substandard copper pipe imported from under-regulated manufacturers. Short term cost advantages turned into long term cost disadvantages. The difference at the initial construction end was less than $500.00. A re-pipe, in a condominium development costs between $50,000 and $100,000 per unit. In this case deregulation didn’t work. There are many counterexamples, the point I’m trying to make is that unless you are prepared to delve into the specifics the general case for/against regulation is not all that useful. "most of the time you’ll end up doing more harm than good", its the most of the time assumption that I have the most difficulty with, what's your base rate? "and the next time won’t be much different from the last time." seems like an arbitrary application of the planning fallacy, why is it any different if you do or you don't?
1Matt_Simpson10yThese statements from Eliezer's piece are empirically based (he says so in the piece). So the short answer is, that's where the base rate comes from and that's why this isn't the planning fallacy. (You could challenge the empirics of course).
0michael6110yI'm not sure, I could do that. I work in a highly regulated business. Urban housing has a range of zoning and technical regulations 90% of which work 100% of the time. The other 10% seen like normal considerations, changes in fashion and legacy issues that are always being resolved. For me a general case for or against regulation wouldn't map on to my experience of the world.
9Raemon10yI haven't actually heard the term "regulatory capture" before, but the concept of big businesses owning regulators is a pretty universal rallying cry of the left.

"Regulatory capture" does not mean "the concept of big businesses owning regulators". It is, rather, a specific proposed explanation for why this takes place, and why it is likely to take place: namely, people working for and invested in a regulated industry have a greater interest in the outcome of regulation than other citizens do, and therefore are likely to expend more effort to influence the regulators.

Notably, regulatory capture does not explain "big business owning regulators" in terms of the members of either big business or regulatory agencies being unusually selfish, evil, or corrupt individuals. It predicts that you can't fix regulation by installing more honest or virtuous regulators, because the problem is structural rather than personal.

Here's an example:

I, as a human citizen who likes breathing air without radioactive soot in it, have an interest in shutting down coal-fired power plants. However, the magnitude of my interest in shutting them down is much less than the magnitude of a coal plant employee (especially a well-paid one who is also a shareholder!) in keeping that plant running. Coal soot slightly increases my chance of lung can... (read more)

3Matt_Simpson10yThat's not how I was using the term. I was using it as a label for a fact - that over time, the industry being regulated by an agency tends to eventually take over that agency. Public choice economics, whatever it's limitations, is trying to explain this fact. Note that my usage is consistent with wikipedia (link in parent and grandparent): (emphasis mine)
2randallsquared10yIt usually goes without saying, but we assume for such examples that the aggregate harm done over all parties is higher than the aggregate good. The problem here is that the expected utility of acting against coal plants is too small to be worth doing at all. That is, once we solve microtransactions (in practice, rather than in theory), this problem may be much more tractable.
0scientism10yThe problem here is that the only institution that does have an interest in the long-term health of the entire population and does have the ability to overrule the interests of the coal plant owners - the state - is crippled under a liberal democracy because its members only have an interest in being re-elected.
3[anonymous]10yActually I think the state's interest in the long-term health of the population isn't really an enduring feature, its more of a plausibly useful strategy. A state can reasonably defined as a local monopolist of violence, generally speaking its long term interest is basically to maintain that monopoly. There are imaginable scenarios where this involves hurting the long-term health of its entire population if say the result makes the population more docile or if they improve rent extraction that the state needs to maintain its monopoly ect. Its even possible to imagine stable states that hurt the long-term health of their citizens so much that they die too fast for fertility to compensate, or rather infertility rates rise until they are sub replacement. Since its easy to import new workers (who may or may not be citizens). The obvious example of this seems to be places where the state relied on the slave owning class and the nature of the work produced by the slaves was so demanding it required constant new imports to maintain populations. A hypothetical (or so I hope) example would be a state using dietary and industrial regulations to ensure that the population as a whole was more docile than average due to hormonal and developmental changes while suffering greater infertility. The slow bleeding of citizens seem easy to replace with hard working first generation immigrants who might not be motivated to cause a fuss over any faults in the system because the system is overall much better than the one they came from.
1fubarobfusco10yI'm pretty sure public choice theory does not adopt the intentional stance with regards to institutions, and so does not assume that they have interests.
3Logos0110yAnd yet I very rarely see progressives or socialists offer up any solution to that other than expanding the power of the regulators.
3Manfred10yWell, there's the obvious (though obvious != effective, of course) solution of making it even harder to get hired by a company that does work or the government after leaving a government job, and imposing stricter hiring rules on regulators. This is the idea behind rhetoric about "getting rid of the revolving door."
0taelor10ySuch a policy would require regulators to enforce, who would themselves be subject to regulatory capture.
0Logos0110yThe problem of regulatory capture has very little to do with regulatory-body agents and more to do with the governing body's rules being set by the groups with the greatest interest in them: the groups governed.
0Manfred10y"Regulatory capture" describes a state of affairs. If it just meant the basic incentives common to all regulation ever, it would be a useless term. And it causes harm not based on the basic incentives just sitting there, but based on a specific state of affairs existing. To prevent the harm, therefore, anything will do as long as it stops a harmful state of affairs from happening.
0Logos0110yThe thing is, there are methods of establishing regulatory bodies that do not create a governing body beholden to the entities regulated. Consider the Sierra Club, for example; they regulate environmental impact but don't answer to the corporate interests they oversee/certify, but instead to their shareholders/members directly.
9Vaniver10yRemember that time someone gave the Sierra Club a bunch of money, explicitly so they would drop opposition to immigration from their platform, and they did?
2Randaly10yNope, they didn't. The Sierra Club never, ever supported immigration restriction; the one time it came up on the ballot, it was voted down by a ratio of 3:2. (By, yes, the Club members in an election, not the Board of Directors.) Election results here. [http://sierraclub.org/bod/electionresults.pdf]
2Vaniver10yFrom 1989 [http://www.susps.org/history/popreport1989.html] to 1996 the position was that immigration should be "stabilized"- i.e. reduced to the level where the entire US's population does not grow. In 1996 the position was changed to neutrality; from 1998 to the present the club has been split on whether or not to oppose immigration (with the majority favoring neutrality).
0lessdazed10yI assume that any result that close was probably susceptible to being determined by who controlled the wording of the ballot question and the use of official resources in campaigning. Apparently [http://rmc.sierraclub.org/pandp/1998-02/imm-c.html] A was put on the ballot by gathering signatures, and then B was put on with wording designed to counter to it. Apparently the organization's leadership largely supported B. Those are great advantages for B, and it didn't even get more than 3/5 of the popular vote. Who could vote for A instead of B, when B is a vote for "the empowerment and equity of women" and "...address[ing] the root causes of migration by encouraging sustainability, economic security, health and nutrition, human rights and environmentally responsible consumption"? Applause, applause. [http://lesswrong.com/lw/jb/applause_lights/]
1lessdazed10yHere [http://articles.latimes.com/2004/oct/27/local/me-donor27/2]
4Manfred10yIf the Sierra Club had the same impact as a national government's environmental agency, there would probably be lobbyists for the Sierra Club too, and cushy jobs for ex-administrators.
0Logos0110yUnlike a centralized agency if the Sierra Club fell to that sort of behavior its membership base would move on to other groups.
3Manfred10yThe same argument applies to political leadership under representative democracy - people could just vote for someone better, or, say, vote for someone who will reform the regulators. Political leaders take advantage of things like group identification, imperfect information, monopoly power, and advertising budgets to keep the status quo intact under pretty much all normal circumstances. These options are also available for business leaders.
0Logos0110ySure. There's a few problems with applying the argument that way, though. 1. Regulatory bodies are not selected democratically. 2. Candidates (as done in the US) are very significantly vetted by non-democratic processes. 3. Once elected, until removed from office a vote is non-rescindable. None of these three items are true of groups in competition with one another. There's no such thing as a perfect system anywhere, politically speaking. All we can do is attempt to mitigate the number of blatant problems and then work to maintain the standards of what we end up with. (Mitigate-then-suppress negative elements.) While business leaders can do that sort of thing, it's difficult for them to get away with it if there's rigorous competition and strong disclosure. There are ways to achieve both of those items, but I haven't mentioned them here because they're only relevant if and only if we actually agree to investigate libertarian solution-spaces, as opposed to comparing ideal states of varying political agendas.
0Manfred10y1) Indirect power over regulators is still power - if the consumers demand it, it will happen. 2) So are businesses. An equivalent in business would be the cost of entering the market that keeps small competitors out - this varies for different sectors, and is most visible in infrastructure-heavy businesses like water companies. Larger companies also get economies of scale that new ones can't. The overall effect in both politics and business is a status quo bias. 3) This is a good point - I think the problem might be more about the time lags between promise, delivery, and evaluation. Some businesses do weakly exploit similar phenomena, but not nearly like politics. Still, a large private regulator would probably have more problems with this than current businesses. A similar, though weaker, effect might also be achieved in businesses by evading accurate evaluation, either through not releasing information, selective release of information, or just plain advertising. I'm willing to discuss theoretical libertarian solutions, but if we just compare current government and business without solutions, I don't see much reason why private regulators would be less corruptible than government ones. The remaining libertarian position is then that solutions to these problems would be much more effective for businesses than for governments.
1Logos0110yThe question isn't of capacity but of magnitude of signal. Are you aware of just how many different regulations there are which create the effect of mitigating the presence of smaller businesses? Just one example; each and every employee costs a small business (under 50 people) something like 25-40% more (in the US) than it does a large corporation (over 1,000) in regulatory-burden-per-employee. There are also a number of diseconomies of scale which our current government scheme protects larger corporations from. Also, it's largely a myth that economies of scale can only be achieved by larger corporations. Smaller businesses that focus on providing a widely-distributed product to a number of companies can create the same effect. Case in point: managed systems administration services firms. Why pay your own full-time employee when you can pay a company to provide a suite of employees who each specialize in their field, and only pay for the hours you really use (and thus get greater expertise at a quarter the cost?) And then there's payroll companies; all they do is provide payroll. Again, economies of scale in the hands of smaller businesses. Which is why I included "strong disclosure" as a necessary element for a libertarian society. Personally I would see this done by a social campaign of vigorously restoring the belief in and appreciation for investigative journalism, and further by widely broadening the scope, scale, and definition of fraud-type crimes. (I would in fact make it a penalizable act to simply make a mistaken statement in a public venue, though I'd make it at most a misdemeanor whose penalty would be somehow tied to the monetary damages resultant to the statement, with a minimum of something like $20.00. This would breed a culture of desire for truthfulness. You can't get around half-lies and the like -- but you can impede them.) Well, I'll admit to having worded that poorly. I mainly meant that I didn't want to go too far down that rabbit hole. :
1Matt_Simpson10yYes, but the left (and the much of the right) doesn't usually have much foresight about this happening. They can think through the negative consequences of deregulation, but typically not regulation or re-regulation. If I wasn't clear, that's what I was talking about. They always seem to propose another regulatory solution to regulatory capture without thinking about how this solution will keep from being captured. (Also, I'm talking about the general public, not the political intellectuals who are typically much more sophisticated.)
3Oligopsony10yThere's nothing really funny here: regulatory capture is a theory of how the social democratic mixed economy etc is prone to certain sorts of institutional failure in accomplishing its goals, so of course the various sorts of people who don't believe in that will find it appealing. Likewise, you'll see both Marxists and libertarians readily endorse the idea that the purpose of most welfare provisions is to control and manage the destitute population, and so on. If only economists and libertarians seem to think about it, that's probably because you know a lot more libertarians.
0Matt_Simpson10yYeah, you're right. I was mainly getting at the fact that marxists are generally associated with the left and the left does seem to be in the group of people that don't think about regulatory capture, but my language didn't really convey that.
2PhilGoetz10yThe impulse to regulate derives from the human tendency to live in hierarchical societies, such as monarchies. It's been a long time since we developed governments other than monarchies, that work by assuming people are corrupt and constructing a system that makes it unlikely for anyone to be especially oppressed. But our instinct is always to forget that is how our government works, and think our problems can be solved by getting the right person as president, CEO, or whatever. (Why has no one developed a democratic corporation?) Exactly the same tendency makes us instinctively think that problems should be solved by having a (problem domain) king who solves them.

(Why has no one developed a democratic corporation?)

They're called cooperatives.

3michael6110yother examples, Mondragon Corporation and many private companies have democratic decision policies
1Matt_Simpson10yYes, but you've merely labeled the tendency. You haven't explained it. I can spin some plausible sounding evolutionary explanations, but what I'm really interested in is the level of cognitive science.
6Scott Alexander10yIf I had to come up with a cognitive science explanation for pro-regulation, regardless of political considerations around whether regulation was good or bad, it would be that failures from too little regulation are obvious, direct, and heartrending (child dies of toxic unproven medicine) and failures from too much regulation are distributed and invisible (child dies of cancer, with no one knowing that a cure sits in a lab somewhere but it's too expensive to license it). This is because regulation, as a specific action taken to stop a problem, gets to optimize for fighting the most obvious, scary problems in the most direct way - whereas nonregulation, as a null action, doesn't get to optimize for that at all.
0Matt_Simpson10yThis is the "seen and unseen" reasoning from my original comment, but something more seems to be going on with regulation. Why are people more quick to point out corruption in a corporation than in a government? Or am I just wrong about this fact? Maybe the seen and unseen reasoning explains the difference in this particular context and in other contexts there is no difference.
-1PhilGoetz10yRight now, Virginia is regulating abortion clinics, making them meet hospital standards, in order to protect womens' safety. Yet I don't think there are any known cases of failures from too little regulation in Virginia abortion clinics. At least, I haven't heard any brought up.
5Unnamed10yAbortion is not a good example to use here since it is often overregulated intentionally by politicians who oppose abortion and want to make abortions harder to get. There is no bias to explain because the policy fits their goal of restricting abortion.
1Scott Alexander10yDo you mean no cases of failure from too little regulation that couldn't have been solved more cleverly by nonregulatory means, no cases of failure where the "solving" regulation didn't have problems of its own, or just that you literally can't think of any any cases ever of failures from too little regulation? If you really mean the last, then avoiding modern-day issues so we don't get into a fight, the Triangle Shirtwaist Factory, thalidomide babies, and the Exxon Valdez oil spill seem like go-to historical examples.
2PhilGoetz10yI mean that I haven't heard anybody in the debate say, "Person X went to an abortion clinic in Virginia, and something bad happened that would have been prevented by these rules." So the impulse to regulate isn't due just to the easy availability of instances of under-regulation over instances of over-regulation.
3[anonymous]10yPresumably "Person X went to an abortion clinic in Virginia, and an abortion happened" fits the bill for a lot of participants in that debate.
0[anonymous]10yThere was a pretty notorious case of a bad abortion clinic in Philadelphia: http://healthland.time.com/2011/01/21/philly-abortion-horrors-what-matters-is-how-and-not-when-an-abortion-is-done-says-expert/ [http://healthland.time.com/2011/01/21/philly-abortion-horrors-what-matters-is-how-and-not-when-an-abortion-is-done-says-expert/] Whether the proposed regulation could help prevent similar cases in VA, I have no idea, but if it means more oversight, you know, it seems plausible.

I am impressed that a political dialogue occurred without the throwing of epithets about the motives of the parties involved. I think I might have even... actually learned something. This feels very weird. :)

7jsalvatier10yCato Unbound is often quite good.
3Vaniver10yReally, any discussion where people agree on basic premises can be quite good, and any discussion where people don't has a very difficult time being good. (Though I agree that Cato puts out quality work in general.)