Search Results for: elite

Elite Biases Make Policy Biases

A 2014 paper predicted U.S. policy changes over four years for 1,779 issues, using the positions of four groups of influencers: business-based interest groups (55), mass-based interest groups (31), median public opinion (6), and elite public opinion (100), i.e. that of people at the 90th percentile of income. (I’ve listed their relative influence in parenthesis. Criticism says mid-class (not poor) influence is bigger.) While elite and median public opinion had a 0.78 correlation, the other pairs were uncorrelated. (A poll sets median influencer at 92% income percentile.)

What this says is that, even in a democracy, the ~90th percentile rich have the most influence, business interest groups have about half as much, and mass interest groups have about a third as much. We less rich folks only get what we want, to the extent we do, mainly because these elites mostly agree with us, and because we sometimes influence mass interest groups.

This median influencer household has income of $210K/yr and wealth of $1.2M, and households above this cut pay 70% of US Federal income taxes. This income is near the median doctor ($207K) and U.S. District Judge ($218K), more than the median full professor ($141K), lawyer ($139), lobbyist ($115K), judge ($109K), and CEO ($103K), and much more than the median federal civil servant ($64K) and high school teacher ($63K). (The median household made $64K, while the median CEO of the top 500 firms made $12.7M.)

These elites who set policy get most of their status and income from labor, not capital, and they are quite comfortable with, and in fact love, large bureaucratic organizations. Their highest hopes tend to be of gaining positions in, getting promoted in, or creating, such organizations. When they have dreams for the world, they dream of new versions with higher mandates and bigger budgets. (Think socialism.)

They can distinguish each other by their elite accomplishments, school credentials, org affiliations, and styles of talk, dress, etc. And their internal dynamics are dominated by status and gossip. That is, they are very social and join mutually-supporting coalitions which help get them the right jobs, party invites, speaking invites, etc. Via extensive gossip, they quickly form an apparent consensus on the policy issues of the day, on who is higher status among them, and on who should be ostracized and expelled from their ranks. Today these elite communities of gossip and status are integrated across the world.

Simple as it is, this account of who most influences policy seems to me promising as the basis of a theory of policy bias. That is, the natural biases of the group who most influences policy may plausibly explain many of our overall policy biases.

For example, policy set by elites may give elites too much benefit of the doubt, and defer too much to their status-gossip system. As elites tend to see their internal status-gossip processes as sufficient to discourage malfeasance and encourage excellence, they tend to see little need for other forms of track records, incentives, or accountability within elite professions and organizations, including government agencies. They see themselves as mostly good people, trying to do good things, who should be supported not hassled.

As another example, when there are groups that elites see as more outside of themselves, as rivals competing with them for power, then elites may push for policies that control, suppress, and disrespect such rivals.

The most obvious candidate for such a rival group is business. Even though these elites are richer than most of us, like most of us they focus more on those who are above them in status, relative to those who are below. Furthermore, the study above says that business is in fact their main rival for influence over policy. And while most business profits go to elites, elites don’t think of themselves as having their main influence on the world via business; elites instead identify more with their roles as org leaders and elite gossipers.

Furthermore, while elites see themselves as mostly well-meaning good people, they see business as transparently and dangerously selfish. Elites see businesses as tending to do what makes them more money, even when their leaders are ostracized and not invited to the right parties. Meaning that the usual pressures that work on most elites may not work on business and the super-rich. Thus elites support harsh, intrusive, and punitive business taxes, regulations, and legal liability. Yes when the super-rich are taxed, these elites are also taxed, but that may seem worth the price to take them down a peg or two. Most ordinary people miss this conflict by not distinguishing these two different kinds of “rich”.

Even though ordinary people seem to have little influence on policy, and mostly agree with elites on policy, elites are still wary of them as individuals. After all, we outnumber them at least five to one, we might revolt, and they must rely on us to do most of the things that need doing. So as employees, we must be tracked, assigned, and incentivized. As consumers and investors, we must be regulated. As authors and voters, our thoughts must be shaped and channeled via teachers, censors, media, interest groups, and politicians. As potential criminals we need to be tracked and threatened with punishment. And the poorest of us need even more direct management, such as via social workers and parole officers. All of which not only keeps us under control, but asserts elite status via the fact of their managing such controls.

Mass-based interest groups mostly don’t seem to scare elites as a whole, because usually such groups are dominated by elites at their top levels. It is only when a mass-based group seems to oppose elites as a whole that elites close ranks and warn against the dangers of such “populism”. While our society gives a lot of lip service to populism, populism is usually crushed aggressively whenever it actually seems threatening.

So how does this theory do empirically? It seems to me that policy does tend to be overly trusting of elites and their status-gossip system, and overly punitive and disrespectful of rival groups. For example, policy pushes us to pick docs, lawyers, and other prestigious professionals based more on the prestige of their affiliations, and less on track records or incentives. Business does seem greatly overly regulated, and taxes seem overly punitive. And policy seems to rely too much on the consensus of elite gossip, relative to more accurate sources like experts or prediction markets.

While roughly half of all regulation of individuals seems to be justified as protecting people from themselves, warnings seem just as helpful but would be far less controlling. Free speech (really free hearing) would be as effective at informing as is censorship. Pandemics could be more efficiently handled via law. And the poor could be helped more via simple cash transfers instead of expensive intrusive management of their lives.

Our legal system has high costs of suing people (from not using lotteries) but no required liability insurance. This makes law available to elites to sue each other, and to punish business, but not available to ordinary people to sue elites or each other. Elites can protect themselves well from ordinary people via strong prosecutor powers of plea bargaining together with broad surveillance and huge numbers of crime laws on the books, and also judges who are elites and give elites the benefit of the doubt. Oh and living, shopping, and working in separate neighborhoods.

And that’s my simple theory of who runs society, and policy biases that naturally result from their rule.

GD Star Rating
loading...
Tagged as: , ,

What is ‘Elite Overproduction’?

Elite overproduction … describes the condition of a society which is producing too many potential elite-members relative to its ability to absorb them into the power structure. … a cause for social instability, as those left out of power feel aggrieved by their relatively low status. … explained social disturbances during the late Roman empire and the French Wars of Religion, and [Turchin] predicted that this situation would cause social unrest in the US during the 2020s. (More)

Toynbee argues that the ultimate sign a civilization has broken down is when the dominant minority forms a Universal State, which stifles political creativity. He states:

First the Dominant Minority attempts to hold by force – against all right and reason – a position of inherited privilege which it has ceased to merit; and then the Proletariat repays injustice with resentment, fear with hate, and violence with violence when it executes its acts of secession. (More)

There’s a simple and plausible income interdependence scenario where inequality matters little for policy: when [welfare] outcomes depend on [income] rank. … This applies whether the relevant rank is global, comparing each person to the entire world, or local, comparing each person only to a local community. [A] 2010 paper … makes a strong case that in fact the outcome of life satisfaction depends on the incomes of others only via income rank. (More)

I often hear about how many elite areas have become more competitive, and more stressful. People have to do and be more than they once did to succeed. There are ominously more hopefuls who will be disappointed, and becomes disgruntled. But people have always cared a lot about status, and tried hard to rise in status. And if status is mainly about one’s percentile rank in some overall ranking, what could have changed?

Some ideas:

  • The variance in money or popularity has changed; it is now more “winner-take-all”, so status gets you more.
  • There is more, or less, mobility in status over time, perhaps to different degrees at different status levels.
  • What were once many disconnected status hierarchies have merged into fewer more global rankings.
  • The relative weight on prestige versus dominance has changed; the one that is now bigger is more stressful.
  • As we get rich, we more satisfy our basic needs (and get more status drunk), and so care more about status.
  • We have gotten better at measuring status (e.g., via social media), making it more visible, so we care more.

Here is my related hypothesis: we now put more weight on many smaller lower-noise status markers, instead of fewer bigger noisier markers. In particular, we put more weight on markers of connections to statusful people and institutions.

For example, early in ancient empires, many rose in status via winning military battles, or perhaps by building new trading regimes. But later in such empires, status was counted more in terms of your connections to other statusful people. Which led to neglect of military success, and thus empire collapse.

So early on, ambitious soldiers tried to figure out how to win battles, and to get involved in promising battles. But it was hard to guess just how to do this, and outcomes were noisy functions of efforts. So no one could be very sure of their future status, or with whom to associate to gain status. But later on, ambitious soldiers would need to come from the right family, and make good new social connections. So they worked to make sure they wore the right clothes, went to the right events, flattered the right people, joined the right groups, and so on. In this world, they could more easily see who was higher status.

As another example, back in my day physics classes gave lots of hard problems that most students couldn’t do. So there was a lot of noise in particular grades, and students cared as much or more about possibly doing unusually well as doing unusually badly. One stellar performance might make your reputation, and make up for lots of other mediocre work. But today, schools give lots of assignments where most get high percent scores, and even many where most get 100% scores. In this sort of world, students know it is mostly about not making mistakes, and avoiding black marks. There is otherwise little they can do to stand out.

For a third example, it seems to me that in academia people now care more about the status of your journal articles and job institutions, and less about what exactly you said in those articles or did in those jobs. And theory, where a new entry might surprise everyone with its great power, has been displaced by lower variance empirics, where success depends more on access to funding and data, on mastering hard in-fashion stat techniques, and on having the right social connections.

In all of these examples, the new focus is on the low, not the high, end of the distribution of outcomes for each event or activity. The new focus is more on social connection and less on the non-social world. And people can better see their current status, and estimate their future status. All of these changes seem to me to naturally feel more “competitive”, producing more “anxiety”.

Added 22Aug: As status marker weights and groups sizes are always changing, there are always groups rising and falling overall in status. Yes, a high status group that is rising in size and falling in status might see that as a time of “elite overproduction”, but since that sort of thing is happening quite often why would we say it happens overall especially more at certain times?

GD Star Rating
loading...
Tagged as:

How High Our Elite Tax?

In the ancient world, invading armies were mostly men who, if they won, stayed and took over the top status slots in the new society. After a few generations, the locals would almost completely submit, and see invader descendants as the highest status elite members of “us”.

When invaders had distinctive looks, styles, etc. then those became marks of high status. People would seek such marks for they and their family, and try to associate with those who had such marks. To get people to support their ventures, they’d try to give them the appearance of elite support, via attracting elite members, investors, customers, suppliers, regulators, etc. As any substantial elite opposition could doom their ventures.

These elites would of course charge a price for their support, and such prices would add up to an elite tax on the society. The size of this tax would depend on how many such elites had to be “paid off” in this way to make anything happen, and how well suited were such elites for the roles they took on. In the limit of many available elites who were actually just directly the best people for their roles, this elite tax was zero. But the more that ventures had to pay off elites who were not otherwise the best for their roles, and have people do things in other than the best ways, in order to gain that crucial perception of elite support, the larger the tax.

Societies in history have varied in how they define status, and thus have varied in how much they waste in efforts to achieve status. When war was important, for example, societies that defined status in terms of military accomplishment had in essence a lower tax, and thus a competitive advantage over rivals. Today when innovation is important, then organizations where status is rewarded for promoting innovation can be at a similar advantage.

In academia, we often see similar effects on smaller scales. Particular fields are taken over by a mutual admiration insider’s club, and then everyone in that field must pay tribute to these insiders to get anything done. If you do not sufficiently praise, cite, fund, hire, etc. such insiders, you will be excluded from the field.

So how big a tax does our society pay for the perception of elite support? Much of status in our world is set via graduating from elite schools and being hired for elite jobs, and many say that we have a “meritocracy” in picking applicants to such positions only on the basis of directly relevant abilities. But of course we know this is only partially true.

As I’ve discussed before, in many areas today elites from around the world have merged into a single elite community which share common standards on who is elite and what criteria they use to decide status. As this world elite faces no competition from other worlds, our world is now more vulnerable to drifting status criteria; competition between societies won’t suppress wasteful criteria. For example, if elites everywhere measure status via years of pointless school, then the whole world could just keeping doing too much such school.

One interesting way to try to measure our elite tax is to look at events where the tax rose quickly and dramatically. Sometimes a particular field is low status, with people there paid accordingly, and then suddenly the field rises greatly income and status. At which point high status people enter and take over the highest status positions of that field. Such elite entrants tend to be young and lack experience in that field, and so tend to denigrate the old and experienced there. They push for generic elite practices that may not be the best for this area.

For example, tech used to be nerdy and lower status, and then made a lot of money and rose in status. Top school kids decided to take jobs in tech, especially right after investment banking jobs dried up after the 2008 crash. And so top school kids took over tech, putting a new extra premium on youth and top school degrees. Tech priorities and practices changed, including a lot more interest in woke politics.

Studies that compared the productivity, innovativeness, and social value produced by tech before and after this transition might give us valuable data on our elite tax. Similar studies might be done regarding other suddenly-prestigious areas before and after their status transition. Seems to me an important topic to study.

GD Star Rating
loading...
Tagged as:

More on Experts Vs. Elites

When a boss issues a new order, usually the main thing he or she is fighting with is the effects of his (or a prior boss’) previous orders. It can take time to undo their effects. And subordinates who fear that yet newer orders will come down before they can make enough changes might prefer to drag their feet, to see if these current orders will last.

Some responded to my last post on experts versus elites by saying how good it is that elites often overrule experts, as experts get it so wrong so often. As with early in this pandemic. But the experts are less of an autonomous force here, and more just the repository of previous elite instructions. If pandemic experts had it wrong before about masks or travel bans, that is mostly because elites previously pushed them to adopt such policies. For example, our continued ban on challenge trials is due to how med ethics experts have interpreted prior elite instructions. Experts won’t change their mind on this until elites tell them they are allowed to change their minds. In contexts where elites are typically so pushy, it can be hard to tell what experts would decide in their absence.

In economics, it usually feels pretty obvious what the elites want us to say. Not all economists do what they are told, but the major institutions and their elite leaders seem mostly willing to go along, and so what the public mostly hears is economists saying what elites want us to say. When elites change their minds, our major institutions also quickly change their minds.

Now I had been thinking this is all bad news for the new kinds of institutions I want to introduce, as I had been assuming they would be framed as new expert institutions. And yes all this suggests a distrust of formal expert mechanisms that can’t be easily overruled by elite opinion. But maybe I have been too hasty about how new institutions might be framed.

Consider the widespread hostility to “market manipulation”, such as seen in the recent Gamestop stock price episode. Or consider movies like Boiler Room, Glengarry Glen Ross, Wall Street, and Wolf of Wall Street. Typically, financial markets are chock full of “manipulation”, in the sense that most traders are trying to talk and spin to get others to agree with and follow their trades. Sometimes they succeed, and sometimes they fail, but that mostly doesn’t bother people. What bothers people most is when they see clearly low status low prestige people seeming to greatly influence prices, especially in ways that seem unlikely to last. (Elite manipulations tend to last.)

Consider also that elites only rarely complain about errors in speculative market prices, such as stock prices or currency prices. They mainly complain when they think they can find non-elite folks to blame for such prices. Together, these facts suggest to me that most elites may see speculative market prices as something that elites create. They know that there is a lot of money at stake in such markets, and that many big powerful rich elite players play heavily in such markets. So perhaps elites usually accept the verdict of such prices as a verdict of elites!

If this were true, then the prospects for improving our social consensus via improving speculative markets would be far higher than I’d ever hoped! If we could get thick markets trading on many more topics, then elites might well defer to those price estimates in their elite conversations, and push experts to also accept such estimates.

Of course, even if elites would accept a price estimate when it exists, this doesn’t mean most are eager for such any particular price to exist. Rivalrous elites constantly try to undermine each other, including via undermining the organs that rival elites use to express their opinions. If if the prices existed for a while, I predict elites would cave and defer to them, at least until they could kill them.

To signal to all that they are dominated by elites, I do think it important that a lot of money seem be riding on these market prices. Mere prediction tournaments or polls of experts just will not do. Even real money markets with small stakes may not be taken seriously enough.

My proposal for Fire-the-CEO markets seems like it could work here. Though I’ve been waiting for 25 years now for someone to take up this idea.

Added 8Feb: I see now why my usual answer to “what should I read?”, namely “textbooks”, falls on deaf ears. People are looking for elites to read, not experts.

GD Star Rating
loading...
Tagged as: ,

Experts Versus Elites

Consider a typical firm or other small organization, run via a typical management hierarchy. At the bottom are specialists, who do very particular tasks. At the top are generalists, who supposedly consider it all in the context of a bigger picture. In the middle are people who specialize to some degree, but who also are supposed to consider somewhat bigger pictures.

On any particular issue, people at the bottom can usually claim the most expertise; they know their job best. And when someone at the top has to make a difficult decision, they usually prefer to justify it via reference to recommendations from below. They are just following the advice of their experts, they say. But of course they lie; people at the top often overrule subordinates. And while leaders often like to pretend that they select people for promotion on the basis of doing lower jobs well, that is also often a lie.

Our larger society has a similar structure. We have elites who are far more influential than most of us about what happens in our society. As we saw early in the pandemic, the elites are always visibly chattering among themselves about the topics of the day, and when they form a new opinion, the experts usually quickly cave to agree with them, and try to pretend they agreed all along.

As a book I recently reviewed explains in great detail, elites are selected primarily for their prestige and status, which has many contributions, including money, looks, fame, charm, wit, positions of power, etc. Elites like to pretend they were selected for being experts at something, and they like to pretend their opinions are just reflecting what experts have said (“we believe the science!”). But they often lie; elite opinion often overrules expert opinion, especially on topics with strong moral colors. And elites are selected far more for prestige than expertise.

When an academic wins a Nobel prize, they have achieved a pinnacle of expertise. At which point they often start to wax philosophic, and writing op-eds. They seem to be making a bid to become an elite. Because we all respect and want to associate with elites far more than with experts. Elites far less often lust after becoming experts, because we are often willing to treat elites as if they are experts. For example, when a journalist writes a popular book on science, they are often willing to field science questions when they give a talk on their book. And the rest of us are far more interested in hearing them talk on the subject than the scientists they write about.

Consider talks versus panels at conferences. A talk tends to be done in expert mode, wherein the speaker sticks to topics on which they have acquired expert knowledge. But then on panels, the same people talk freely on most any topic that comes up, even topics where they have little expertise. You might think that audiences would be less interested in hearing such inexpert speculation, but in fact they seem to eat it up. My interpretation: on panels, people pose as elites, and talk in elite mode. Like they might do at a cocktail party. And audiences eagerly gather around panelists, just like they would gather around prestigious folks arguing at a cocktail party about topics on which they have little expertise.

Consider news articles versus columnists. The news articles are written by news experts, in full expert mode. They are clearly more accurate on average than are columns. But columns writers take on an elite mode, where they pontificate on all issues of the day, regardless of how much they know. And readers love that.

Consider boards of directors versus boards of advisors. Advisors are nominally experts, while directors are nominally elites. Directors are far more powerful, are lobbied far more strongly, and are paid a lot more too. Boards of advisors are usually not asked for advise, they are mainly there to add prestige to an organization. But prestige via their expertise, rather than their general eliteness.

Even inside academic worlds, we usually pretend to pick leaders like journal editors, funding program managers, department chairs, etc. based mainly on their expert credentials. But they also lie; raw prestige counts for a lot more than they like to admit.

Finally, consider that recently I went into clear expert mode to release a formal preprint on grabby aliens, which induced almost no (< 10) comments on this blog or Twitter, in contrast to far more comments when arguable-elites discuss it in panelist/elite mode: Scott Aaronson (205), Scott Alexander (108), and Hacker News (110). People are far more interested in talking with elites in elite mode on most topics, than in talking with the clear relevant experts in expert mode.

All of which suggests that my efforts to replace choice via elite association with prediction markets and paying for results face even larger uphill battles than I’ve anticipated.

Added noon: This also helps explain why artists are said to “contribute to important conversations” by making documentaries, etc. that express “emotional truths.” They present themselves as qualifying elites by virtue of their superior art abilities.

See also: More on Experts Vs. Elites

GD Star Rating
loading...
Tagged as: , ,

The World Forager Elite

My last post was on Where’s My Flying Car?, which argues that changing US attitudes created a tsunami of reluctance and regulation that killed nuclear power, planes, and ate the future that could have been. This explanation, however, has a problem: if there are many dozens of nations, how can regulation in one nation kill a tech? Why would regulatory choices be so strongly correlated across nations? If nations compete, won’t one nation forgoing a tech advantage make others all the more eager to try it?

Now as nuclear power tech is close to nuclear weapon tech, maybe major powers exerted strong pressures re how others pursued nuclear power. Also, those techs are high and require large scales, limiting how many nations could feasibly do them differently.

But we also see high global correlation for many other kinds of regulation. For example, as Hazlett explains, the US started out with a reasonable property approach to spectrum, but then Hoover broke that on purpose, to create a problem he could solve via nationalization, thereby gaining political power that helped him become U.S. president. Pretty much all other nations then copied this bad US approach, instead of the better prior property approach, and kept doing so for many decades.

The world has mostly copied bad US approaches to over-regulating planes as well. We also see regulatory convergence in topics like human cloning; many had speculated that China would be defy the consensus elsewhere against it, but that turned out not to be true. Public prediction markets on interesting topics seems to be blocked by regulations almost everywhere, and insider trading laws are most everywhere an obstacle to internal corporate markets.

Back in February we saw a dramatic example of world regulatory coordination. Around the world public health authorities were talking about treating this virus like they had treated all the others in the last few decades. But then world elites talked a lot, and suddenly they all agreed that this virus must be treated differently, such as with lockdowns and masks. Most public health authorities quickly caved, and then most of the world adopted the same policies. Contrarian alternatives like variolation, challenge trials, and cheap fast lower-reliability tests have also been rejected everywhere; small experiments have not even been allowed.

One possible explanation for all this convergence is that regulators are just following what is obviously the best policy. But if you dig into the details you will quickly see that the usual policies are not at all obviously right. Often, they seem obviously wrong. And having all the regulatory bodies suddenly change at once, even when no new strong evidence appeared, seems especially telling.

It seems to me that we instead have a strong world culture of regulators, driven by a stronger world culture of elites. Elites all over the world talk, and then form a consensus, and then authorities everywhere are pressured into following that consensus. Regulators most everywhere are quite reluctant to deviate from what most other regulators are doing; they’ll be blamed far more for failures if they deviate. If elites talk some more, and change their consensus, then authorities must then change their polices. On topic X, the usual experts on X are part of that conversation, but often elites overrule them, or choose contrarians from among them, and insist on something other than what most X experts recommend.

This looks a lot like the ancient forager system of conflict resolution within bands. Forager bands would gossip about a problem, come to a consensus about what to do, and then everyone would just do that. Because each one would lose status if they didn’t. In this system, there were no formal rules, and on the surface everyone had an equal say, though in fact some people had a lot more prestige and thus a lot more influence.

This world system also looks new – I doubt this description applied as well to the world centuries or millennia ago, even within smaller regions. So this looks like another way in which our world has become more forager-like over the last few centuries, as we’ve felt more rich and safe. Big world wars probably cut into this feeling, so there was probably a big jump in the few decades after WWII, helping to explain the big change in attitudes ~1970.

Elites like to talk about this system as if it were “democratic”, so that any faction that opposes it “undermines democracy”. And it is true that this system isn’t run by a central command structure. But it is also far from egalitarian. It embodies a huge inequality of influence, even if individuals within it claim that they are mainly driven by trying to help the world, or “the little guy”.

This system seems a big obstacle for my hopes to create better policy institutions driven by expert understanding of institutions, and to get trials to test and develop such things. Because as soon as any policy choice seems important, such by triggering moral feelings, world elite culture feels free to gossip and then pressure authorities to adopt whatever solution their gossip prefers. Experts can only influence policy via their prestige. Very prestigious types of experts, such as in physics, can win, especially on topics about which world elites care little. But otherwise, elite gossip wins, whenever it bothers to generate an opinion.

That is, the global Overton window isn’t much wider than are local Overton windows, and often excludes a lot of valuable options.

Notice that in this kind of world, policy has varied far more across time than across space. Context and fashion change with time, and then elites sometimes change their minds. So perhaps my hopes for policy experiments must wait for the long run. Or for a fall of forager values, such as seems likely in an Age of Em. Alas neither I nor my allies have sufficient prestige to push elites to favor our proposals.

Added 11p: It seems to me that the actual degree of experimentation and variance in policy is far below optimum in this conformist sort of policy world. We are greatly failing to try out as many alternatives as fast as we should to find out what works best. And we are failing to listen enough to our best experts, and instead too often going with the opinions of well-educated but amateur world elites.

Added4p: As John Nye reminds me, in the early years of a new tech, only a few nations in the world may be able to pursue it. They then set the initial standards of regulation. Later, more nations may be able to participate, but risk-averse regulators may feel shy about defying widely adopted initial standards.

GD Star Rating
loading...
Tagged as: , , ,

Enforce Common Norms On Elites

In my experience, elites tend to differ in how they adhere to social norms: their behavior is more context-dependent. Ordinary people use relatively simple strategies of being generally nice, tough, silly, serious, etc., strategies that depend on relatively few context variables. That is, they are mostly nice or tough overall. In contrast, elite behavior is far more sensitive to context. Elites are often very nice to some people, and quite mean to others, in ways that can surprise and seem strange to ordinary people.

The obvious explanation is that context-dependence is gives higher payoffs when one has the intelligence, experience, and social training to execute this strategy well. When you can tell which norms will tend to be enforced how when and by whom, then you can adhere strongly to the norms most likely to be enforced, and neglect the others. And skirt right up to the edge of enforcement boundaries. For weakly enforced norms, your power as an elite gives you more ways to threaten retaliation against those who might try to enforce them on you. And for norms that your elite associates are not particularly eager to enforce, you are more likely to be given the benefit of the doubt, and also second and third chances even when you are clearly caught.

One especially important human norm says that we should each do things to promote a general good when doing so is cheap/easy, relative to the gains to others. Applied to our systems, this norm says that we should all do cheap/easy things to make the systems that we share more effective and beneficial to all. This is a weakly enforced norm that elite associates are not particularly eager to enforce.

And so elites do typically neglect this system-improving norm more. Ordinary people look at a broken system, talk a bit out how it might be improved, and even make a few weak moves in such directions. But ordinary people know that elites are in a far better position to make such moves, and they tend to presume that elites are doing what they can. So if nothing is happening, probably nothing can be done. Which often isn’t remotely close to true, given that elites usually see the system-improving norm as one they can safely neglect.

Oh elites tend to be fine with getting out in front of a popular movement for change, if that will help them personally. They’ll even take credit and pretend to have started such a movement, pushing aside the non-elites who actually did. And they are also fine with taking the initiative to propose system changes that are likely to personally benefit themselves and their allies. But otherwise elites give only lip service to the norm that says to make mild efforts to seek good system changes.

This is one of the reasons that I favor making blackmail legal. That is, while one might have laws like libel against making false claims, and laws against privacy invasions such as posting nude picts or stealing your passwords, if you are going to allow people to tell true negative info that they gain through legitimate means, then you should also let them threaten to not tell this info in trade for compensation.

Legalized blackmail of this sort would have only modest effects on ordinary people, who don’t have much money, and who others aren’t that interested in hearing about. But it would have much stronger effects on elites; elites would be found out much more readily when they broke common social norms. They’d be punished for such violations either by the info going public, or by their having to pay blackmail to keep them quiet. Either way, they’d learn to adhere much more strongly to common norms.

Yes, this would cause harm in some areas where popular norms are dysfunctional. Such as norms to never give in to terrorists, or to never consider costs when deciding whether to save lives. Elites would have to push harder to get the public to accept norm changes in such areas, or they’d have to follow dysfunctional norms. But elites would also be pushed to adhere better to the key norm of working to improve systems when that is cheap and easy. Which could be a big win.

Yes trying to improve systems can hurt when proposed improvements are evaluated via naive public impressions on what behavior works well. But efforts to improve via making new small scale trials that are scaled up only when smaller versions work well, that’s much harder to screw up. We need a lot more of that.

Norms aren’t norms if most people don’t support them, via at least not disputing the claim that society is better off when they are enforced. If so, most people must say they expect society to be better off when we find more cost-effective ways to enforced current norms. Such as legalizing blackmail. This doesn’t necessarily result in our choosing to enforce norms more strictly, though this may often be the result. Yes, better norm enforcement can be bad when norms are bad. But in that case it seems better to persuade people to change norms, rather than throwing monkey-wrenches into the gears of norm enforcement.

So let’s hold our elites more accountable to our norms, listen to them when they suggest that we change norms, and especially enforce the norm of working to improve systems. Legalized blackmail could help with getting elites to adhere more closely to common norms.

GD Star Rating
loading...
Tagged as: , ,

Elite Evaluator Rents

The elite evaluator story discussed in my last post is this: evaluators vary in the perceived average quality of the applicants they endorse. So applicants seek the highest ranked evaluator willing to endorse them. To keep their reputation, evaluators can’t consistently lie about the quality of those they evaluate. But evaluators can charge a price for their evaluations, and higher ranked evaluators can charge more. So evaluators who, for whatever reason, end up with a better pool of applicants can sustain that advantage and extract continued rents from it.

This is a concrete plausible story to explain the continued advantage of top schools, journals, and venture capitalists. On reflection, it is also a nice concrete story to help explain who resists prediction markets and why.

For example, within each organization, some “elites” are more respected and sought after as endorsers of organization projects. The better projects look first to get endorsement of elites, allowing those elites to sustain a consistently higher quality of projects that they endorse. And to extract higher rents from those who apply to them. If such an organization were instead to use prediction markets to rate projects, elite evaluators would lose such rents. So such elites naturally oppose prediction markets.

For a more concrete example, consider that in 2010 the movie industry successfully lobbied the US congress to outlaw the Hollywood Stock Exchange, a real money market just then approved by the CFTC for predicting movie success, and about to go live. Hollywood is dominated by a few big studios. People with movie ideas go to these studios first with proposals, to gain a big studio endorsement, to be seen as higher quality. So top studios can skim the best ideas, and leave the rest to marginal studios. If people were instead to look to prediction markets to estimate movie quality, the value of a big studio endorsement would fall, as would the rents that big studios can extract for their endorsements. So studios have a reason to oppose prediction markets.

While I find this story as stated pretty persuasive, most economists won’t take it seriously until there is a precise formal model to illustrate it. So without further ado, let me present such a model. Math follows. Continue reading "Elite Evaluator Rents" »

GD Star Rating
loading...
Tagged as: , ,

Bowing To Elites

Imagine that that you are a politically savvy forager in a band of size thirty, or a politically savvy farmer near a village of size thousand. You have some big decisions to make, including who to put in various roles, such as son-in-law, co-hunter, employer, renter, cobbler, or healer. Many people may see your choices. How should you decide?

Well first you meet potential candidates in person and see how much you intuitively respect them, get along with them, and can agree on relative status. It isn’t enough for you to have seen their handiwork, you want to make an ally out of these associates, and that won’t work without respect, chemistry, and peace. Second, you see what your closest allies think of candidates. You want to be allies together, so it is best if they also respect and get along with your new allies.

Third, if there is a strong leader in your world, you want to know what that leader thinks. Even if this leader says explicitly that you can do anything you like, they don’t care, if you get any hint whatsoever that they do care, you’ll look closely to infer their preferences. And you’ll avoid doing anything they’d dislike too much, unless your alliance is ready to mount an overt challenge.

Fourth, even if there is no strong leader, there may be a dominant coalition encompassing your band or town. This is a group of people who tend to support each other, get deference from others, and win in conflicts. We call these people “elites.” If your world has elites, you’ll want to treat their shared opinions like those of a strong leader. If elites would gossip disapproval of a choice, maybe you don’t want it.

What if someone sets up objective metrics to rate people in suitability for the roles you are choosing? Say an archery contest for picking hunters, or a cobbler contest to pick cobblers. Or public track records of how often healer patients die, or how long cobbler shoes last. Should you let it be known that such metrics weigh heavily in your choices?

You’ll first want to see what your elites or leader think of these metrics. If they are enthusiastic, then great, use them. And if elites strongly oppose, you’d best only use them when elites can’t see. But what if elites say, “Yeah you could use those metrics, but watch out because they can be misleading and make perverse incentives, and don’t forget that we elites have set up this whole other helpful process for rating people in such roles.”

Well in this case you should worry that elites are jealous of this alternative metric displacing their advice. They like the power and rents that come from advising on who to pick for what. So elites may undermine this metric, and punish those who use it.

When elites advise people on who to pick for what, they will favor candidates who seem loyal to elites, and punish those who seem disloyal, or who aren’t sufficiently deferential. But since most candidates are respectful enough, elites often pick those they think will actually do well in the role. All else equal, that will make them look good, and help their society. While their first priority is loyalty, looking good is often a close second.

Since humans evolved to be unconscious political savants, this is my basic model to explain the many puzzles I listed in my last post. When choosing lawyers, doctors, real estate agents, pundits, teachers, and more, elites put many obstacles in the way of objective metrics like track records, contests, or prediction markets. Elites instead suggest picking via personal impressions, personal recommendations, and school and institution prestige. We ordinary people mostly follow this elite advice. We don’t seek objective metrics, and instead use elite endorsements, such as the prestige of where someone went to school or now works. In general we favor those who elites say have the potential to do X, over those who actually did X.

This all pushes me to more favor two hypotheses:

  1. We choose people for roles mostly via evolved mental modules designed mainly to do well at coalition politics. The resulting system does often pick people roughly well for their roles, but more as a side than a direct effect.
  2. In our society, academia reigns as a high elite, especially on advice for who to put in what roles. When ordinary people see another institution framed as competing directly with academia, that other institution loses. Pretty much all prestigious institutions in our society are seen as allied with academia, not as competing with it. Even religions, often disapproved by academics, rely on academic seminary degrees, and strongly push kids to gain academic prestige.

We like to see ourselves as egalitarian, resisting any overt dominance by our supposed betters. But in fact, unconsciously, we have elites and we bow to them. We give lip service to rebelling against them, and they pretend to be beaten back. But in fact we constantly watch out for any actions of ours that might seem to threaten elites, and we avoid them like the plague. Which explains our instinctive aversion to objective metrics in people choice, when such metrics compete with elite advice.

Added 8am: I’m talking here about how we intuitively react to the possibility of elite disapproval; I’m not talking about how elites actually react. Also, our intuitive reluctance to embrace track records isn’t strong enough to prevent us from telling specific stories about our specific achievements. Stories are way too big in our lives for that. We already norms against bragging, and yet we still manage to make our selves look good in stories.

GD Star Rating
loading...
Tagged as: , , ,

Wanted: Elite Crowds

This weekend I was in a AAAI (Association for the Advancement of Artificial Intelligence) Fall Symposium on Machine Aggregation of Human Judgment. It was my job to give a short summary about our symposium to the eight co-located symposia. Here is what I said.

In most of AI, data is input, and judgements are output. But here humans turn data into judgements, and then machines and institutions combine those judgements. This work is often inspired by a “wisdom of crowds” idea that we often rely too much on arrogant over-rated experts instead of the under-rated insight of everyone else. Boo elites; rah ordinary folks!

Many of the symposium folks are part of the IARPA ACE project, which is structured as a competition between four teams, each of which must collect several hundred participants to answer the same real-time intelligence questions, with roughly a hundred active questions at any one time. Each team uses a different approach. The two most common ways are to ask many people for estimates, and then average them somehow, or to have people trade in speculative betting markets. ACE is now in its second of four years. So, what have we learned?

First, we’ve learned that it helps to transform probability estimates into log-odds before averaging them. Weights can then correct well for predictable over- or under-confidence. We’ve also learned better ways to elicit estimates. For example, instead of asking for a 90% confidence interval on a number, it is better to ask for an interval, and then for a probability. It works even better to ask about an interval someone else picked. Also, instead of asking people directly for their confidence, it is better to ask them how much their opinion would change if they knew what others know.

Our DAGGRE team is trying to improve accuracy by breaking down questions into a set of related correlated questions. ACE has also learned how to make people better at estimating, both by training them in basic probability theory, and by having them work together in teams.

But the biggest thing we’ve learned is that people are unequal – the best way to get good crowd wisdom is to have a good crowd. Contributions that most improve accuracy are more extreme, more recent, by those who contribute more often, and come with more confidence. In our DAGGRE system, most value comes from a few dozen of our thousands of participants. True, these elites might not be the same folks you’d have picked via resumes, and tracking success may give better incentives. But still, what we’ve most learned about the wisdom of crowds is that it is best to have an elite “crowd.”

GD Star Rating
loading...
Tagged as: , ,