Wednesday assorted links


Talk of inequality is partly about rallying one part of the Democratic base -- the rich, educated, urban class that is resentful of those who are much richer. It is also an attempt to blend their concerns about inequality with a broader dissatisfaction on the part of the general population with bankers and the financial elite. For example, Piketty's book fires up that resentment even though rich "banksters" represent high end "labor" and not gains from r, as even Solow acknowledged but tried to dodge.

This is satire, right?

"Piketty’s book fires up that resentment"

The general population is never going to read "Capital in the Twenty-first Century". They probably won't watch it if HBO makes it into a movie. They're always going to feel resentment toward anyone perceived as having made a lot of money without getting blisters and sweaty. But they're not writing about it and commenting on it on national television. That's being done by academics and pundits who've never missed a meal and whom the unwashed masses regard also regard as parasites. Sadly these non-entities just don't know how bad they've got it without being informed by guys in ivory towers.

Yes, all of that is (not) true. It has (not) been well-studied and proven to repeatedly by sociologists (armchair intellectuals) and political scientists (partisans).

The Democrats do well among the richest ethnic groups and among the poorest. Talking of inequality works well when talking to the poorest. It is a promise to give more money to Blacks and single mothers - the two largest supporters of Barak Obama for instance.

I doubt it will do well among their richer supporters.

So they need some sort of code so that everyone will understand that what they need by inequality is taking from people who don't vote for them and giving to people who do. "Diversity" as applied to university campuses is a great example of this. What the academic de facto affirmative action programs do is keep out the sort of people who vote Republican - poorer, White, rural, southern - and allow in anyone who votes Democrat. You can see this by suggesting that Jewish Americans are over-represented at college and so in the interests of fairness and diversity, their numbers should be reduced.

Not an idea with a great deal of support.

"Do well among the richest ethnic groups and among the poorest"? Do you mean they do well with whites, Asians and poor people? A lot of whites vote Republican. Poorer people do vote Dem, while richer people vote Rep. But there are also significant trends by sex, age and college educated.

Real affirmative action, which liberals support, actually often gives extra points to rural and southern students (assuming it is not a southern school). What do you mean de facto academic affirmative action lets in Dems? Are you saying that de facto acdemic affirmative action (read: no affirmative action) lets in poor black kids?

You are so wrapped up in your unfounded assumptions and narrow worldview you can't being to understand how ridiculous you sound.

Jan May 14, 2014 at 9:31 pm

No, I do not mean they do well with Whites, Asians and poor people. I mean they do well with some White ethnic groups and with poor people. A lot of Whites do vote Republican. Especially if they are from the South. Poorer people do vote Democrat too. So what? While richer people vote Republican on the whole but it is not all that strong. The North-East is strongly Democrat regardless of how wealthy people are.

Which is pretty much what your link shows. Wealth is actually not that much of a strong predictor of how someone will vote. Especially if they are from Connecticut. It is fairly good as a predictor if you are poor - but that is probably also a racial difference. It is measuring how good race is as a predictor. As we have seen with Obama, the strongest predictor of his voters is race - Blacks were overwhelmingly pro-Obama. Wealth has something to do with it, because the next strongest pro-Obama group was single women. Who are presumably often poor on paper.

I would look at ethnicity. From what I can see of your link, being rich is not much of a predictor of how people vote, presumably because, as I said, the Democrats are strongly supported by some of the richest ethnic groups in America. The Jewish American community is a good example. They should, logically, vote Republican. But significant numbers don't.

As for your claims about "Real affirmative action", you must be kidding. If it allows in rural and southern Whites, who do you think it excludes? It cannot preferentially admit everyone. What we know is that it allows more Blacks is, fewer Asians, and fewer poor Whites. Universities are not allowed to call their Admissions programs Affirmative Action openly any more. So they have claim it is about "diversity". Which works to allow more and weaker Black candidates in. More and weaker Hispanics too by and large. Fewer Asians. Fewer poor Whites. Rich Whites are more or less unaffected. They can afford the other things they need like a fake charity building houses in the Third World funded by Daddy and so on. That is, the people who are admitted are groups that tend to vote for the Democrats - Blacks, Hispanics, rich people from the North-East. People who are excluded are mainly Asians, who lean Democrats but may still be up for grabs, and poor Whites.

Now all this is a re-statement of what I said before and you should have understood the first time.

You are so wrapped up in your unfounded assumptions and narrow worldview you can’t being to understand how ridiculous you sound.

Given you have not understood a word I said, this criticism is pathetic. Give me a valid criticism, show me the slightest sign you have understood what I said, and then we will talk about how ridiculous it is.

In the meantime, would you agree with the statement that Jewish Americans are over-represented on campuses across America and that in the interests of a diverse student body that reflects America, there ought to be a limit on how many are admitted?

Your first comment didn't make sense, because it overgeneralized voting preferences and on some accounts it was just wrong. And your understanding of how affirmative action is usually implemented is weak. You probably don't know that almost every affirmative action program has both socioeconomic (your poor whites) components, especially for underrepresented rural areas, and racial components. And you think rich whites (who are often Republican, remember) can get around diversity goals by having a lot of volunteer work, but Asians can't do the same thing? Also, as recently as the late 1990s, Asians pretty reliably voted Republican.

You desperately want affirmative to be some conspiracy to undermine access to college by Republicans. Guess what--most 18 year olds don't even strongly identify with a major political party, but by the time they finish college they tend to lean Democrat. You should actually be working to keep people out of college. They are more likely to vote R that way.

You have not shown it was wrong in any way whatsoever. It did generalize. I am not writing a book.

I am sure every program has a socioeconomic component. The fact is it does not work to admit poor rural Whites. I am sure Asians can work around. I know at least one who did so. It is just harder if you start out with less social capital. That is probably why universities discriminate against Asians. They tend to vote Republican. That does seem to be changing though. We will see what big colleges do.

I don't want it to be anything. I don't even think it is a conspiracy. It is just people like them choosing other people like them and keeping people not like them out. Bigotry, not conspiracy. I am sure most 18 year olds do not identify strongly with any political party. And I am sure that after a careful selection process that happens to exclude the sort of people who vote Republican, and four years of leftist indoctrination, they skew Democrat. Big deal.

You did not answer the Jewish-American question - would you support holding the number of Jewish students admitted to college down closer to their number in the population? In the interests of diversity of course.

I would agree that Jews are a higher share of college students than their proportional population. That's just a fact. I would not support limiting the number of Jewish students, nor any other group in particular, to a specific proportion of admitted students. One can implement affirmative action without making the increased enrollment of underrepresented come at the expense of any one specific group. You do understand that, right? Allowing in more black kids doesn't mean not letting in more Jews specifically (rather than whites, suburban kids Asians, whatever). It can be spread around.

@6 -- Perhaps the NY Times should sue Google in the European Court of Human Rights to demand such embarrassing old links be removed from its search engine.

The NYT was against the minimum wage and leading Democrats were sponsoring Reagan's watershed tax cut bill. And today, in the NYT you can read about how politics has since lurched to the far right......

Well, he did reduce taxes for the top income bracket, which fell from 70% in 82 to 28% in 86.

But yes, of course the Republicans remain the party of moderates and the rational center. Not allowing emotion or prejudice to outweigh consensus expert opinions, and what with their stalwart defense of voting rights and civil liberties for all Americans. Their refusal to allow big government enter the personal lives of Americans. Reasoned, intelligent, and righteous - as always.

Those flakes at the NYT, I don't even think they realize what fools they appear to be to the rest of us.

In your shoes, I wouldn't have tackled Erdmann's comment head-on either. Well played.

Give them the seriousness they deserve.

I should add that I'm not in your shoes and glad of it.

One can, without being a hypocrite, support cutting taxes from 70% to 28%, but call further tax cuts too much.

Maybe I could build a parallel case with the NYT editorial calling for an increase in the EITC instead of a higher minimum wage, by saying that they got the EITC but further raises in the EITC would be too much. It's hard to really understand "too much EITC" though.

Reducing marginal rates and reducing taxes are not the same thing. At high income levels, the legal definition of taxable income has far more impact on how much tax you pay than the marginal rate does, and the Reagan "cuts" were accompanied by big changes in that definition.

Of course two alternatives they propose are expanding EITC and direct wage subsidies. I wonder which of these choices conservatives today feel most comfortable with?

EITC. That question wasn't nearly as "tricky" as you probably imagined it to be.

I didn't say it would be tricky, but have you seen any conservatives in Congress pushing it? My point is that it is imaginary support.

You can read of Republicans' quickly evolving position on this issue once Obama basically said, "You're right. Let's do EITC expansion!"

What would you say is the biggest difference between the EITC and a wage subsidy?

I think wage subsidies would be invisible to the average worker, who would just see higher wages. The responsibility for taking advantage of the subsidies would be shifted to employers, rather than requiring low-income workers educate themselves on the details of the tax code.

Only after the EU rules that corporations are 'persons.' Don't hold your breath, though.

@6: Correlation or causation?

Rather # 3

I wondered that myself. According to the article, the high inequality Democratic districts are urban and the high inequality Republican districts are heavy-immigrant southern districts. In other words, the Democratic districts have high inequality because the middle class fled them and the Republican districts have high inequality because the poor are moving into them.

All I see is correlation.

One issue is that congressional districts are usually highly unnatural geographic units because of gerrymandering. At the state level I see no clear correlation between inequality and political leanings. Red southern states are unequal, blue MA, CA and the NY tri-states are unequal, red plane states are equal and blue New England, Minnesota, Hawaii and PacNW are equal.

My guess is that the district level inequality is driven by the fact that Republican districts are more gerrymandered. This is demonstrated by the fact that Republicans congressmen are overrepresented in the House relative to the number of votes they received. Gerrymandered districts are less likely to intra-district inequality, because the right-leaning upper-middle class neighborhoods are separated from the left-learning lower-middle class neighborhoods.

We don't need a minimum wage of zero--we need more unpaid internships everywhere. No regulation on working conditions to let the job creators create more jobs (that pay nothing).

#2: This would fit nicely with the age old axiom of adversity building character. But, there's all the problems with how you define the terms. Yesterday's narcissism is today's confidence. Of course, we no longer have recessions so the whole idea falls apart over time.

#3: Some interesting insights about the culture gap, but I'll note that the pod people running this joint live in those areas worrying about inequality. Flyover country does not get a say in the matter.

#6: Another example of how everything has moved Left. Today, no respectable public conservative would make that argument. I've lived long enough to see moderate democrats turn into extreme right-wing extremists, despite not changing their position on a single issue.

#6: I've seen it time and time again. Those who were once moderate Republicans are now considered liberal lefties by their own party. Their politics haven't changed at all. Example issues on which they did not change their mind, but have miraculously transformed into liberals -- abortion, health care reform, belief in the existence of manmade climate change and evolution, etc etc.

Z: Great point on 6. My position has long been that abortion should be legal with sensible caveats (not after fetus viability, no major surgery on minors without telling the parents); gays should be able to enter in civil unions with all the rights of married couples; and we should allow copious immigration while enforcing the actual laws.

These were all moderate positions not long ago. They are now all hate crimes.

Jim, you do this - you go run that platform in a republican primary anywhere but California or the Northeast. Then please tell us more stories of how it was the lefties who sent you packing.

Not much of a reader, are you?

I said nothing of Republican primaries. I said that these positions were considered moderate -- or even center-left, really -- just a few election cycles ago.

Now the left considers them hate crimes.

That sounds like a fairly centrist Democrat position from the 1990's. Do you think you could run on that platform as a Democrat in California or the NorthEast?

I worked for a congressman in the 1980's who was considered a "moderate Democrat." The mere fact no one ever uses the term "moderate Democrat" today is a clue. Most of those guys from the 80's would not only be conservative Republicans today, nitwits like Jan would be on calling them extreme right-wing extremists of the most extreme kind.

So narrowminded. Look at what I just posted below about Landrieu. Look at all the moderate Republicans who have been run out of Congress for being moderates. I'd list them for you but you probably know exactly who they are and refuse to acknowledge it because it doesn't fit into your worldview.

What you're describing is pretty close to Mary Landrieu's (D) position on all those things.

So, she believes on sensible restrictions on abortions and civil unions for gays? She's a firm supporter of Don't ask Don't tell? Those are all moderate Democratic positions from the 1990's.

Yeah. She believes in sensible restrictions on abortions and does not support gay marriage (though she is ok with other protections for them).

Nobody said anything about DADT. She doesn't support it, but you have to understand that on two specific social issues the whole country has evolved: gay rights and marijuana. There are many other issues, such as abortion, on which the country, and especially conservatives, have headed for the reddest legal line they can find. Look at the number of abortions over time. Look at all the state restrictions on it. Does that in any way signal a shift leftward to you?

"Yeah. She believes in sensible restrictions on abortions and does not support gay marriage (though she is ok with other protections for them)."

"Sen. Mary Landrieu, D-La., is part of a diminishing number of Senate Democrats who have not endorsed same-sex marriage. "

I think you are correct about Landrieu, and also simultaneously reinforce the original posters point. Democrats have moved to the Left as a party when somebody like Senator Landrieu would have been a centrist 20 years ago, is now to the parties right. It's also noteworthy that a Democrat would win in a solidly Red state like Louisiana. Apparently, Red staters will vote for Democrats that are centrist.

I don't think Landrieu would have been a moderate Dem 20 years ago, probably a centrist. I recognize that the country will move left on some issues and right on others over time -- it's not consistent. But on balance I think politicians have gotten more conservative in recent years. The problem is that you're using the two major social issues that the country has admittedly veered left on to illustrate your point. If you look at other topics, such as taxation, school vouchers, and support for unions you will find that everyone seems to have shifted right over the last 10 years.

In any case, I think both parties have become more homogenous in the last 20 years. There are basically no Republicans who supports new taxes and almost no Dems who suggest taking up arms in foreign conflicts these days.

#5 - Actually, the robot cars are just copying humans.

Cars drive closer to helmet-wearing cyclists than to non-helmet-wearing cyclists. The driver assumes that the non-helmet-wearer is less experienced and therefore requires more space. Counter-intuitively, wearing a helmet puts you at greater risk of accident; although it also provides more protection if an accident does happen.

"Drivers passed an average of 8.5 cm (3 1/3 inches) closer with the helmet than without."

#6 - The oddest thing about that editorial is not seeing the NY Times question the minimum wage, but seeing the NY Times (or for that matter almost any publication) make an argument on the grounds of efficacy and not status signaling. It feels so quaint.

You could have stopped at "make an argument".

# 3
Ms. McArdle is wrong here on several levels. First Democrats generally have not talked about inequality but about fairness. It was Romney who heard the words “fair share” and thought “class war.” Second, it not obvious that fairness has not been a good issue for Democrats. Third which states are Democrat and which Republican has little to do with national fairness issues like EITC v minimum wage, top marginal tax rates, deductions v partial tax credits.

You are correct. Here's a quote from Charlie Brown's little sister Sally from Christmas 1965: "All I want is what I have coming to me, all I want is my fair share."

"First Democrats generally have not talked about inequality but about fairness."

That seems like a far fetched comment.

"Remarks by the President on Economic Mobility - Barack Obama December 4th 2013"

The word "inequality" was used 26 times in that speech.

It was Romney who heard the words “fair share” and thought “class war.”

When people are saying that those paying literally orders of magnitude more in taxes than they themselves are are not paying their "fair share" in taxes, how can you not infer class warfare?

In wars, pretty bad things happen to the losers. In the closest we have come to a straight redistribution issue, the repeal of the GWB tax cuts the losers, lost a couple of percentage points of their highest bracket, non-investment income and in return the payroll tax rate went up. War is Hell, right? .

"In wars, pretty bad things happen to the losers."

So are you conceding the point that this argument is all about class warfare?

>almost all the areas with the worst inequality are already controlled by Democrats

Gee. I wonder if we can learn anything from that.

That inequality/envy is in fact a winner for Democrats at the state and local levels?

Most of those areas have been Democratic for decades.

@5: The author is allegedly an "affiliate scholar" at the Stanford Law School, and he can write an article on robots deciding which people to kill and maim without even mentioning the legal implications? I can think of no better way to convince a jury to deliver bankruptcy via punitive damages than to introduce into evidence the logs and algorithms showing how someone's carefully programmed robot decided to maim or kill a particular victim.

And when I produce evidence that the companies that do NOT optimize expected consequences kill at twice (or whatever) the rate of the first? Do you think the jury will be too stupid to understand that, or they will understand it but not care?

Or that (to use the example in the paper) it hit and killed the mini driver, which was let us say fully predictable, when there was ample evidence to show that it if had hit the Volvo instead there's only a 10% [*] chance someone would have died? What if I take _that_ to the jury? The robot is not "deciding to kill or maim" someone, else it's a somewhat different (maybe easier) problem; the robot is trying to reduce the chance of that happening.

(Or if not 10%, what about 99% or 1%? Is there any value that you would be willing to take to the jury on behave of the Mini-driver's estate? What about 0.0000001%?)

>Do you think the jury will be too stupid to understand that, or they will understand it but not care?

Maybe I'm too cynical but I'd expect most jurors to fit into one of those two categories.

"the robot is trying to reduce the chance of that happening"

Reduce the chance of what, exactly, happening? In your hypothetical, the robot is not trying to reduce the chance of the Volvo driver being killed or injured. And even if the Volvo driver suffered only a minor injury, it is that minor injury that you are hypothetically but specifically being sued for. The injury that your robot caused when it decided to use the Volvo as part of its crumple zone. Your robot which is a totally unsympathetic inhuman machine built by an unsympathetic inhuman greedy corporation with huge money bins.

That the only physically possible alternative would be the certain death of the Mini driver, is almost irrelevant legally speaking. You aren't being sued for anything you did or did not do to the Mini driver, you aren't being sued for the "optimized expected consequences" of your actions generally; tort law doesn't work that way. Not in the United States, nor anywhere else that I know of. Nor is there any provision for an un-suit for anti-damages on account of your having benefited another by steering your car away from him. You are liable for the harm you or your agents cause. You are not rewarded for the harm you or your agents avoid causing, or even actively prevent. There is a "necessity" defense in tort law that might be relevant, but AIUI it generally doesn't work in personal-injury cases.

An algorithm like "steer so as to minimize the possibility of the most-imminent collision; when in doubt brake straight ahead / while staying in the current lane", are likely to be more defensible. If your robot winds up killing the guy in the Mini, at least there was never a point where it actively steered towards the Mini. That counts for a lot, and brings the question back to just why the Mini was so close ahead that even robotic reflexes couldn't stop in time. You might win that argument, even with a wholly unsympathetic inhuman robot as your chief witness.

IMO, only in the US would this been seen as a big ethical dilemma.

Your point I think is that if I make a product that ends up being applied in a no-win (or rather, either way _might_ lose) situation, no matter what I chose I may be literally ruined (bankrupt if a company). So the problem. If you had a sane legal system where the risk was you faced was to "justly compensate the victim, with punitive damages if applied somewhat proportionate to your actual wrongdoing"
the law of large numbers and/or insurance (same idea) would get us where we need to be. But "no easy option; nothing is certain; but if your choice turns out to be wrong (even if best in probability or expectation) it's the end of the line for you" you'll find all sorts of hard moral dilemmas. I really wonder whether this line of thinking is unique to the US.

I don't know of any legal system that does a really good job of balancing Type I and Type II risks. It almost certainly can't be done with tort or liability law, because half of the outcome is based in harms avoided, and how do you turn those into legal actions?

Even imperfect systems can work (somewhat) - even if the hypothetical death of the Mini-driver is deemed irrelevant; even if the law ignores the concept of "expected" deaths.
You create a device that puts itself into these dilemmas - so you provision enough directly (or by insurance) to cover "no win" situations. Remember, if you
hit the Mini and the jury believes that aiming the other way would "probably" save lives, you are likely toast. So either decision might end up with a dead person with
a grievance against you.

It's the "one difficulty decision; your choice turned out wrong (whether wise in expectation or not) - you are RUINED" (the original post: "deliver bankruptcy")
that breaks this. You can no longer provision or plan for the algorithm that saves most lives in expectation (or any other sensible metric you care to name)
since the first time you do the right thing (by any metric you care to name) you are toast. If the liability is large but bounded we (society) gets to tell an automaker (e.g.)
what to optimize. If the liability is certain ruin the message is "don't make a product that will ever get itself in hard-decision situations, no matter how beneficial
overall". I contend that we don't want the latter. I contend that few countries outside the U.S. have the
"it turned out wrong, too bad, RUIN FOR YOU; MEGA LOTTERY WIN FOR THE
ESTATE" mentality in its tort system.

I really wish this forum had an edit-after-post function (or even review before posting) :-(
There are numerous minor bugs, but for the major one please ...
s/first time you do the right thing/first time your arguably-rational decision turned out to have a bad outcome/.

I think I'm really unclear. Back to Jon Schillings post, what if the tort system and the Jury pretty much always in such cases
delivered a price of say $10M per death (20, 30, the point will remain the same?) - because after all, the driverless car killed (or could have avoided) this particular person's death
so the estate should be paid), and _furthermore_ plus additional punitive damages based on and proportionate to provable negligence and/or noncompliance with current standards.
But though high, this is ho longer necessarily the death sentence (or for a corporation, bankruptcy as Schilling's post).

This would change things COMPLETELY. Upends the financial question facing car manufacturers. Upends he legal question. Upends The moral question.
Yetmost of the world outside the US is like this.

#3) Democrats tend to hold seats in high inequality areas. So while talk of inequality may not help them gain seats, it fortifies the ones they already have.

Inequality is one of those words, like vegetables, that, although important, just doesn't resonate. The Democrats have to rephrase the issue. Just like when people think they can eat an apple for breakfast, a banana for lunch, and some strawberries with dinner, they get all the "fruits and vegetables" they need, the word doesn' t resonate. The concept of inequality does, however.

Comments for this post are closed