“What’s the question about your field that you dread being asked?”

That is the new Edge symposium, with many excellent luminaries, including Jens Ludwig, Richard Thaler, and Raj Chetty from economics, with Sendhil Mullainathan playing host and interlocutor.  Chetty serves up these answers:

Here are three questions that come to mind that I dread answering as an economist working on policy issues:

1. If you were in charge, what policies would you enact today to raise growth rates and incomes for the average family in America?

2. Why do American students perform poorly relative to students in other countries and how can we fix education in the U.S.?

3. When are house prices going to recover to pre-recession levels?


I am a lawyer and the question I dread most is "Can I just ask you a few questions, as a friend?"

It's almost as bad when asked by strangers at a cocktail party.

Just hand them all a business card and tell them to see you at the office where you can properly evaluate the issues in a confidential setting with access to "the law".

Wow, really? Parties are beyond the reach of the law? That's the only way I can imagine you dont have access to it there.

If lawyers are like most knowledge monopoly professions, the real answer is "it's complicated, and we like to keep it that way. Pools don't heat themselves, ya know. "

No, he just can't bill people there

I've found the easiest way to deal with this is not to have friends.

How can you do that and be a lawyer?

I liked these:

"How does one justify having worked and continuing to work in the financial sector?" ( Emanuel Derman )

"What happened before the big bang?" (Lawrence M. Krauss)

"Why is my phone bill so high?" - I own a company that writes software for telephone companies.

When our educational results are normed for ethnic backgrounds, we do just fine.

American kids of Asian extraction do better than Asian kids.
American kids of African extraction do better than African kids.
American kids of Latin American extraction do better than Latin American kids.
American kids of European extraction do as well or better than European kids.

This also is the case at the state level. During the recent ructions over public employee unionization, the Wisconsin teacher's union was sneering at Texas and its bare-bones educational system. However, with proper adjustments for demography, it is clear that Texas is doing a better job teaching its children of all backgrounds than Wisconsin is. Black, Hispanic, and white children in Texas all do better than children of the same background in Wisconsin. The only reason Wisconsin shows modestly better test scores overall is because it is as lily-white as the MS-NBC newsroom.


Right. This was particularly clear on the 2009 PISA reading test:

Asian Americans outscored every Asian country, and lost out only to the city of Shanghai, China's financial capital.

White Americans students outperformed the national average in every one of the 37 historically white countries tested, except Finland (which is, perhaps not coincidentally, an immigration restrictionist nation where whites make up over 95% of the population).

Hispanic Americans beat all eight Latin American countries.

African Americans would likely have outscored any sub-Saharan country, if any had bothered to compete. The closest thing to a black country out of PISA's 65 participants is the fairly prosperous oil-refining Caribbean country of Trinidad and Tobago, which is roughly evenly divided between blacks and South Asians. African Americans outscored Trinidadians by 25 points.

Now that's just one subject for one year on one international test, but the general pattern is that American students do fairly well compared to their ancestral homelands. (Of course, education in America is hugely expensive as well.)


Have you done a similar ethnic analysis using PISA-Science data or PISA-math data (not necessarily 2009 in view of the data limitations)?

Would be interesting to see the trends. The stereotype says Asians are better at math / sci than reading.

I think PISA doesn't test every subject every cycle. I believe they rotate Science, Math and Reading tests every few years.

I too would be interested in the American breakdown by ethnic group compared to various other countries. A whole lot of American 'poor' scores may well just be an artifact of demographics.

As I recall, PISA tests everything, but they don't *report* the detailed breakdown for every subject for every cycle. There may be some way to buy access to that data, but the detailed data for most of their testing is limited-access.

I have my quarrels with Pisa and how they select participants, but that aside, I am not so sure about the overall quality of schooling in the us. I have been to the us during high school and they lagged a year on German schools.

And if I look at the results the us is behind Germany, which is a solid native white country. There is also suisse and France which are better. Seems like the theory doesn't hold.

However on high rating Asians, I think it is the "jew effect", meaning that as a immigrant from a high competition society you are even more motivated to excell, because of parents, culture or whatever.

I really don't think it is valid to compare Asian-Americans with Asians and African-Americans with Africans. US living standards and development are much higher in these countries; of course the American counterparts will do better. It would be notable if any comparisons were even close!

Not sure I understand. Why would one expect Asian-Americans to outperform native Japanese and Koreans? These are some of the most highly developed and technologically advanced countries in the world.

Gymnasium != high school. I went to Gymnasium, and the courses my high school age children take are 1-2 years more advanced than any courses my old Gymnasium is currently offering. Calculus, biochemistry, genetics etc are way more challenging (not compared to my schooling but to the current curriculum).

We are in the Midwest however.


"However on high rating Asians, I think it is the “jew effect”, meaning that as a immigrant from a high competition society you are even more motivated to excell, because of parents, culture or whatever."

Might explain why American Asians excel. But how can that explain why Asian-Asians excel? Hong Kong, Korea and Singapore are all ahead of most white nations.

I don't get the point of normalizing by ethnic groups. Is America somehow the only country where ethnic diversity exists? The only thing normalizing does is expose the inefficiency of our education system, which promotes inequality. What's the proportion of Blacks vs. Whites in France (hint: it's higher there)? Why do French kids on average perform better than ours on those tests? Other countries get results out of their children regardless of race. How do they do it? Ethnic diversiy is very present everywhere else, including China (I know, they all look alike, right?).
And what's the point of comparing ethnic groups with countries of origin? Are you not aware that a disproportinate amount of immigrants that come from overseas are from the educated elites in their countries?
Last, we should from making generic statements about how American kids would perform better than (insert blank), as those claims are proven false time and time again.
What's the logic behind pointing to a small group somewhere to refute an argument about the average population?
Can we stop making up excuses and start owning up to our failures?

From the Wapo link provided:
"Also, first- and second-generation immigrant blacks have high academic standards because of a belief in the relationship between education and the American Dream but from a recognition of sacrifices by family in their home country. Last, African and Caribbean blacks have a strong belief in their ability to succeed because they had firsthand examples of black professionals in their native lands."

If beliefs were snowflakes man would be lost in a whiteout of beliefs, blind in his own front yard and having to feel his way back to his front door.

On the genetic architecture of intelligence and other quantitative traits
Steve Hsu


Not sure what your point is but my main argument is that achievement has nothing to do with ethnic background. That the link attributes the high achievement of Black immigrants in part to belief does not negate said achievement. By focusing on one paragraph which cites belief as one of several factors you're missing the pointentirely. I don't think debating the meaning of intelligence should be the focus here either. I'll just quickly point out there are many forms of intelligence. The BBC did an interesting piece on the subject. It's on youtube. please, watch it if you have the time. Another recommendation: please take the time to investigate the educational achievement of African immigrants compared to White U.S. natives. My guess is you'll learn something that seems counterintuitive.

" What’s the proportion of Blacks vs. Whites in France (hint: it’s higher there)"

I'm not sure if you think France is much, much, much blacker than it is, or if you think there are very, very few blacks in the US. Either way, you are very, very wrong. The ratio in the US is about 5:1. In France it's more like 15:1-30:1

Other countries get results out of their children regardless of race."

Not really.

I opened up a can of worms I didn't intend to. The Black White ratio in France varies from 6:1 to 15:1 depending on whether you include overseas territories, first and second generation immigrants, illegal immigrants and even mixed race people who are not necessarily considered Black, contrarily to the norm here. Adding the fact that french law prohibits the collection of racial statistics and french society does not consider does not consider all Blacks as one ethnic group, you and I can argue all day long and both be right and wrong. So let me remove my foot from my mouth and concede that France was a bad example and distracted from the main point.
Let me clarify: if the original comment was implying that ethnic diversity is the reason for the U.S. poor showing, that argument is strongly refuted by the results from China and India, some of the most ethnically diverse countries in the world. If the argument was that it's all Black people's fault, which I know is not the original commenters point but has been implied by others, that argument is refuted by the educational achievements from African immigrants which trumps that of White natives. As for the comparison between minorities and their home countries, I think I exposed the major flaw in that line of reasoning clearly enough. Even Canada, which is our closest relative in terms of culture and racial mix, is outperforming us by far. As for your last comment, if countries with ethnic diversity similar to ours manage to get better overall scores than we do, where do you come from with your "not really?" I actually had the chance to leave on different continents and I can assure you that having a Black minority is not a situation unique to us.

No, it wasn't implying that "diversity" was the reason. You can have lots of sorts of diversity without low scores.

India, btw, scores really, really, really badly, and works against your point (but your point wasn't even wrong)

Saying that African immigrants have a high rate of school completion and earn a lot of degrees is not really germane to the question of why the US performs poorly on standardized achievement tests relative to other OECD nations. When people talk about the US "falling behind" other countries educationally, they're not referring to school completion rates or graduate degrees awarded, they're looking at standardized test results.

How do African immigrants fare on the SAT, ACT, GRE, etc? Does anyone know?


Non-responsive. Again, you are pointing to school graduation rates, when the idea that the US lags other developed countries educationally is based entirely on standardized test results.

Actually India fairing poorly makes my point that ethnic diversity is not a factor. But I don't know what we're talking about anymore: PISA scores? Educational achievement? In the ranking I saw India was #2 behind China in PISA math score. The first 2 comments on this thread were trying to make a claim that America was competitive if broken down by ethnic group. That's what I was invalidating.

I see your point. The data you're asking for is highly specialized and requires a little more research but implied in my response is the idea that since African immigrants perform better in schools they would do so in standardized tests. This is certainly the trend for the data we do have regarding Asian Americans Whites and native African Americans. I don't think the notion of Americans lack of competitiveness is solely due to PISA scores. Success rates of immigrants in American schools tell the same story. Anyone who's ever done a comparison in graduation requirements by country knows this. In fact, anyone who's ever been interested in learning about other countries' school systems understands this. Certainly anyone who's ever studied abroad knows this. Do some research on Indian students attending American top Universities because school is too competitive in India.
The profileration of IB programs in the U.S. is a consequence of this knowledge. While IB is considered "advanced" in the U.S., it's the minimum standard in a huge part of the rest of the world. PISA scores are just the easiest way to make the point because they appear simple and unbiased. Notice how people get twisted in knots over straightforward standardized tests, how easy would it be to explain graduation requirements, curriculum differences and graduation rates?

India was just about at the bottom of the last PISA rankings. So it's kind of hard to believe that Indians come to top US universities because Caltech is just so much easier than Uttar Pradesh U.


The number of blacks of all origins scoring at high levels on standardized tests is tiny, so the number of immigrant types scoring highly is, at best, very small.

And I mean small like 29 black students total scoring at the level of an average top 14 law school student on the LSAT

The BBC, the economist, and MIT's technology review also have reports on this phenomenon if you want to dig deeper. Same goes with Korean kids attending highschool in the U.S.

I'm a researcher in health economics. When I say that I get "*confused look* what's health economics". My usual response is "economics applied to health". Surprisingly people don't find that helpful.

"When my parents or my baby with Down Syndrome stand in front of Obama’s 'death panel,' what should they say about their 'level of productivity in society' so they are judged worthy of health care?"

Tell me where I can sign up to be on the Death Panel.


Cut a 7-figure check to the DNC and get in line.

Or apply for a Texas State Penitentiary union job.

As an internal government regulatory auditor the question two questions I easily dread are:

1. Can I get a waiver for that?

2. So if I don't do it what will actually happen?

Both are easy to answer, the problem is the answers undermine the intent and marganalize my efforts.

These 50 words tell me more about the workings of government than any civics textbook I've read.

Funny, because I get paid handsomely to deal with your ilk (who are generally reasonable folk, no offence) and I dread the interrogative statement, "soooo I have this audit....."

I preempt it all by introducing myself as, "a craniologist slash glottochronologist that dabbles in industrial refuse art, commenting on blog posts and--- hey are those drinks over there?!"

#2 assumes American students do perform relatively poorly, which as mentioned above is in dispute. #3 assumes prices will reach their previous peak again, which I happen to think is true in the long-enough run, but is not a logical necessity.

3. When are house prices going to recover to pre-recession levels?

Real or nominal?

"Can't a computer do that as well as you can?"

As a member of a company trying to put together software for Dodd-Frank compliance in the commodity space, the questions are:
(1) What was the CFTC trying to do there?
(2) Is a "swap future" a swap or a future?
(3) When do the rule concerning [insert subject here] is going to be finalized?
More here: http://on.wsj.com/16BqD0D

Are we supposed to interpret "dread" as "These questions are hard," or as "I am uncomfortable answering these questions?"

Maybe 1 is hard, but 2 and 3 seem more like "You are not going to like the answer."

Apparently, the worst question I can ask my staff, is "Why aren't you delegating that..?"

I believe that Tino Sanandaji over at Super-Economy has done a good job of refuting the premise of question 2 here: http://super-economy.blogspot.com/2010/12/amazing-truth-about-pisa-scores-usa.html

So what's your fourth most dreaded question?

No. 3: We had a housing bubble. So why is getting prices back to top bubble levels a good thing? Wouldn't something along the lines of "affordable housing" be a better goal?

I love how millions of Americans are engaged in massively leveraged real estate speculation with government subsidies that they lobby ruthlessly to maintain, then act like there's something wrong with Wall Street!

Being in the actuarial profession, the worst questions is: "What are you?"

Inevitably you answer, "an actuary" and get some question like "is that something with birds?" Then you get a few words into the description and eyes glaze over. "Math and statistics having to do with insurance..." **SNORE**

What about "I work for an insurance company. I calculate when you're going to die/ how likely you are to get into an accident/how much your premiums should be." Easily understood, though unlikely to help if you want people to like you.

If you want people to dislike you, you could just say "I work for an insurance company" and leave it at that.

"I look at pension plans to see if they actually have enough money invested to pay out the retirement benefits that they've already promised to their participants."

Sadly, pension plan actuaries - especially those who focus on government plans - often do very newsworthy work.

"Is Wal-Mart good or bad?"

As an engineer the question would be how certain are you the theory and simulation matches reality. Actually I would forward that to climate science and economics too. And perhaps would you trust a driverless car. It's never good to know too much about the people developing these ;-) it might scare you.

What a bunch of pansies. Most of those questions on Edge would fall in the category of "difficult", "impossible to answer definitively", or "mildly annoying". A dreaded question is one that threatens to undermine the core assumptions of one's field, or the core of one's identity as a scholar.


As a leadership coach I am often asked "How do you guarantee results?". "And if you are so good, why aren't you a famous leader?"

So who knows the names of Tiger Wood's coaches?

Just those profession descriptions themselves make me cringe:

"Leadership Coach". "Social Media Consultant".

Anyone who pays attention to golf, I imagine. I don't but I know I've read their names before.

I like how the guy from the University of Chicago used the question about violence in Chicago to go on a rant about gun control.

Why do professors work so hard to avoid teaching and do research?

This sounds a bit like the bank (Allied?) commercial with the economist (forget who but want to say Barro) who is asked where interest rates are going tomorrow.

Why has the US allowed so many Indian workers, most of them somewhat less than mediocre in technical skills, to come and work here for essentially less than prevailing wages and thereby kill the market for US trained computer science majors?

Business likes cheap labor and politicians like to do nice things for business.

"kill the market for US trained computer science majors"


"Does this make me look fat?"

I'm a teacher, and when I get a version of question #2, I point out thatthe premise is wrong, and give them some version of this:


Most annoying from acquaintances:
"Can you fix my computer/phone/television for me?"

This is after I tell them I am an electrical engineer. The answer is no. I can't even fix my own. But, they often seem suspicious.

Most dreaded from my bosses:
"Can you write a schedule for me?"

I work at a small company that often makes proposals for and receives research style work from the DOD, DARPA, etc. After I write the schedule, I am peppered with suggestions and pressure like "can't this be a week shorter", "do we have to do that much testing, etc." This continues until magically the schedule (and therefore the cost) matches the size of the available grant. I feel like I am their cya. The fall guy if we get accused of fraud. It was my schedule after all.

Most feared over all
"I am officer/IRS investigator/Enviornmental Protection agent/etc. so and so can you answer a few questions for me?"

"What relevance does asymptotic complexity have to real-world computing?"

Well, we pose it ourselves even in a (semi-serious) April Fool's post---see remarks about AC^1 here.

Chetty's his inclusion of number 2 shows that he is too honest for his own good. He may be admitting that he must not tell the truth when that is asked.

Comments for this post are closed