Category: Uncategorized

“In College Endowment Returns, Davids Beat the Goliaths” starts off by pointing out that small college endowments and large ones had roughly the same return:

the smallest endowments— those under $25 million — edged out the biggest endowments, averaging a five-year annualized return of 10.6 percent to the $1 billion-plus category’s 10.4 percent.

[Don’t expect these returns going forward for your own portfolio! We’re still getting a dead cat bounce from the Collapse of 2008.]

Curiously, the Times considers 10.6 versus 10.4 to be a significant difference in favor of the smaller endowments, but the real story is in the next paragraphs:

Even more surprising, the top-performing endowments over 10 years among all schools reporting data weren’t giants like Harvard and Stanford or even Yale … the top-performing colleges are two Virginia universities whose financial resources amount to a negligible fraction of the typical Ivy League endowment. … Radford University, which ranked first, has an endowment of $55.5 million, and Southern Virginia University, which was second, has an endowment of just $1.1 million. Radford’s annualized 10-year return is 12.4 percent, and Southern Virginia’s is 11.2 percent. For the most recent fiscal year, 2015, Radford earned over 13 percent and Virginia Southern’s return was 10.5 percent. The average return last year for all endowments was just 2.4 percent.

Should we be surprised that a particular portfolio of $1.1 million outperformed a portfolio of $37.6 billion? Perhaps, but not if there is a bell curve of performance and there are a lot more $1.1 million portfolios than there are $37.6 billion ones.

The Gates Foundation went down this road after finding that a lot of the best high schools in the U.S. were small. The conclusion was that smallness leads to better academic performance. It later transpired that small high school performance fell on a bell curve, just the same as large high schools, but with a somewhat higher standard deviation (and therefore it was easier to find good-performing small schools). See this analysis in Marginal Revolution. (And, of course, don’t forget that a high school falling on the center of the U.S. bell curve would be considered an emergency situation in Finland, Shanghai, Singapore, or South Korea! See “Smartest Kids in the World Review”.)

Readers: Did the New York Times fall into the same statistical trap that led the Gates Foundation to squander hundreds of millions of dollars? If so, why are we so prone to this one? Can we blame the lack of emphasis on null hypothesis testing in AP Statistics?

 

Full post, including comments

Guantánamo Diary by Mohamedou Ould Slahi contains a Jihadi’s view of light aircraft. Our tax dollars paid for him to fly from Senegal to Mauritania to Jordan to Afghanistan to Cuba. Here are Mr. Slahi’s notes on the trip from Senegal to Mauritania:

But it didn’t take very long to realize I had my own plane to myself. As soon as the guy returned with my stamped passport, all five of us stepped toward the runway, where a very small white plane was already running its engines.

The plane was as small as it could be. We were four, and barely managed to squeeze ourselves inside the butterfly with heads down and backs bent. The pilot had the most comfortable place. She was a French lady, you could tell from her accent. She was very talkative, and rather on the older side, skinny and blond.

The bigger guard and I squeezed ourselves, knees-on-faces, in the back seat, facing the inspector, who had a little better seat in front of us. The plane was obviously overloaded.

I heard her at one point telling him that the trip was only 300 miles, and would take between 45 minutes and an hour, depending on the wind direction. That sounded so medieval.

I hate traveling in small planes because they’re shaky and I always think the wind is going to blow the plane away.

My company seemed to have a good time checking the weather and enjoying the beach we had been flying along the whole time. I don’t think that the plane had any type of navigation technologies because the pilot kept a ridiculously low altitude and oriented us with the beach.

The account of the trip from Mauritania to Jordan is hard to reconcile with geography. This is a 2600-mile (nautical) trip so a fuel stop makes sense, but not in Cyprus, which is about the same distance from Mauritania as Jordan. And it certainly wouldn’t make sense to make a fuel stop between Cyprus and Amman, which is what Slahi describes. Maybe this is why the U.S. government has been interrogating him for 15 years. There is no way to get a straight story out of him! (The one area where Slahi is consistent and clear is in assigning blame for terrorism: “The whole problem of terrorism was caused by the aggression of Israel against Palestinian civilians, and the fact that the U.S. is backing the Israeli government in its mischiefs.”)

 

Full post, including comments

We like to think about the U.S. as a meritocracy compared to Third World countries and Old World countries. Achievement is how you get to the top. But what would young people see when they look at recent Presidents and Presidential candidates?

• George W. Bush: son of former president
• Barack Obama: supported by many voters on the basis of his skin color
• Hillary Clinton: wife of former president; supported by many voters on the basis of her current gender identification (but will Hillary still identify as a woman in January 2017?)
• Donald Trump: child of rich parents

We would have to go back 24 years, to the 1992 election of Bill Clinton, to find an example of meritocracy in action at the Presidential level. How will parents be able to tell their children “You could grow up to be President” if there are hereditary, marital, skin color, or gender requirements?

Full post, including comments

One thing that I’ve noticed in talking to high school students from high-income families in which both parents have a lot of fancy degrees is that a lot of these young people want to major in “international relations” once they get to college. And in fact quite a few colleges offer this major (partial list; list of schools outside the U.S.).

This leads to the stupid question of the day: If so many people around the world are studying international relations, why are countries constantly fighting? (if not always in a shooting war then at least some sort of verbal conflict)

Slightly less stupid question: What do young people actually learn when they study “international relations”? Presumably they can’t get hands-on experience negotiating agreements!

Full post, including comments

“How Society Pays When Women’s Work Is Unpaid” is an interesting New York Times article.

Cultural change is also important, [Melinda] Gates said.

She recalled being unhappy about the long commute to her oldest daughter’s preschool. Mr. Gates, then chief executive of Microsoft, said he would drive their daughter two days a week.

“Moms started going home and saying to their husbands, ‘If Bill Gates can drive his daughter, you better darn well drive our daughter or son,’ ” Ms. Gates said. “If you’re going to get behavior change, you have to role-model it publicly.”

In other words, she married a guy who brought $80 billion into the household, but ended up being unhappy with the division of labor in the household.

A subtext in the article and in reader comments is that men are able to earn more money because women do “unpaid work” around the house. This is kind of a cornerstone of the American family law system, i.e., that the lower-earning spouse somehow contributed to the higher-earning spouse’s ability to earn (successful women who pay child support and alimony to ex-husbands so that they can have sex with younger/hotter women end up not being too happy about this theory), though it is no longer an assumption in Germany.

Certainly there do seem to be a lot of high-earning people with spouses who don’t work for wages. But maybe that is simply due to a combination of high tax rates for those marginal earnings and also the fact that increasing after-tax household earnings by 5 percent won’t significantly improve living standards. (And typically sufficient cash can be extracted through litigation following a divorce. See how Jamie Cooper-Hohn collected about $500 million from the English courts without working for wages.) Consider Sheryl Sandberg. She earned over $1 billion, mostly from Facebook shareholders, without a stay-at-home spouse. Judith Faulkner‘s husband has continued to work as a physician (source) and that hasn’t stopped Ms. Faulkner from earning nearly $3 billion by building Epic Systems. Hillary Clinton would perhaps argue that Judith Faulkner could have made a lot more if she had been a guy, but does anyone argue that she would have earned more if her husband had quit his doctor job?

New York Times readers don’t seem to have much doubt as to the potential earnings boost from an adult at home 9-5. Here’s a reader comment:

if the man is able to gather assets into the marriage because he has his wife at home doing his laundry, cooking his meals, and caring for his children, the assets reasonably and ethically belong to both parties. If he wants to negotiate a rate of pay with her, wherein he pays her for all the things she does for the family, then it would make more sense to talk about who earned what if it becomes time to divide the property.

I agree with this comment’s second point, which is not too different from what happens in some European jurisdictions where lifetime alimony is not available. If a couple makes the decision that one will stay at home, the working partner puts money into the stay-at-home partner’s retirement account. This way they don’t spend 100% of their assets (and their children’s assets) on legal fees to have a judge figure out what is the fair division of assets and income post-divorce.

The first point raises a question, however. I responded with

Is there a basis for the assumption that a stay-at-home spouse increases a person’s earnings? Do companies find out that an employee has a stay-at-home spouse and say “Wow, here’s your fat pay raise”? In nearly every part of the U.S. a child can be parked with the government until 3 pm and then be seamlessly handed off to an “after-school program” until dinner time. If a child is in school or in an after-school program from 8 am to 6 pm, how does the presence or absence of a stay-at-home spouse affect the earnings of a worker?

You mention laundry. There are services that will pick up and drop off laundry. You mention cooking meals. Americans have been known to survive on take-out or pre-prepared meals from supermarkets.

If you’re right on the impossibility of making money without having a stay-at-home spouse, how is it that single people are able to earn significant money? Who does their laundry and cooks their meals?

Of course it is nice to have a wonderful home environment and there is a lot of value delivered to a family by an adult who enhances that environment, but I am not sure how we get from that to the assumption that this affects the earnings of a full-time worker who is part of that household.

Readers: What do you think? Does consuming a home-cooked meal enable you to earn more money? Does sleeping in an elaborate suburban home enable you to earn more money than if you lived in a smaller full-service apartment? Can you earn more money if you have a stay-at-home spouse doing child care compared to if kids are parked in commercial care?

[Personally I was at my most productive when I lived in a modest rental apartment and consumed most meals from restaurants and/or corporate cafeterias. For one thing, I didn’t spend half of my life on the phone with Whirlpool and GE trying to get appliances repaired!]

Full post, including comments

Due to the large number of American voters who want a planned economy and the tendency of American politicians to bury the nation in debt and pension obligations, a large proportion of our economy is already “socialist” in the senses that (a) government ministries decide what kind of housing people will occupy, how much they will spend on food every month, and what kind of health care they will get, and (b) the government needs to collect roughly 50 percent of GDP in taxes in order to stay solvent (currently less is being collected, but insolvencies, such as Detroit and Puerto Rico, are becoming more commonplace).

In other words, we may already have become a “socialist” country without any public debate on whether or not that is what we want.

This Wall Street Journal article by Larry Lindsey, a former Federal Reserve governor, has some interesting statistics on the extent to which our economy is now centrally planned:

In 1968, government transfer payments totaled $53 billion or roughly 7% of personal income. By 2014, these had climbed to $2.5 trillion—about 17% of personal income. Despite the redistribution of a sixth of all income, inequality measured by all three of the Census Bureau’s indexes is far higher today than in 1968.

Transfer payments under Mr. Obama increased by $560 billion. By contrast private-sector wages and salaries grew by $1.1 trillion. So for every $2 in extra wages, about $1 was paid out in extra transfer payments—lowering the relative reward to work. Forty-five million people received food stamps in mid-2015, an increase of 46% since the end of 2008. Similarly, 71.6 million individuals were enrolled in Medicaid and the Children’s Health Insurance Program, an increase of 13.3 million since October 2013.

In 2008, during the deepest recession in 75 years, 13.2% of Americans lived below the government’s official poverty line. The Great Recession officially ended in June 2009, but in 2014, after five years of economic expansion, 14.8% of Americans were still in poverty. The economy was better, and there were a lot more handouts, but still poverty rose.

The structure of American households shows how this happened. From 2008 through 2014, the most recent year for which we have data, the number of two-earner households declined. These two-earner households have become the backbone of the American middle class.

Research by the Hamilton Project and the Urban Institute show that when families with children making between $20,000 and $50,000 attempt to have a second earner go back to work, the effective tax rate on the extra earnings—including lost government benefits such as food stamps, the earned-income tax credit, and medical support payments—is between 50% and 80%.

While the number of two-earner households declined during the first six years of the Obama presidency, the number of single-earner households rose by 2.6 million and the number of households with no earners rose by almost five million. In other words, two thirds of the increase in the number of families under Mr. Obama was accounted for by households with no one working.

Some of this is territory covered in The Redistribution Recession, but the relevant point is that these dramatic changes in the structure of society have occurred without any real political debate. No politician stood up and said “Who agrees with me that sitting at home playing Xbox should result in roughly the same spending power as working 40 hours per week at a lower-wage job?” (As noted in an earlier posting, paying people to sit home idly is neither a necessary nor even a conventional feature of “socialist” economies; the Soviet Union didn’t have able-bodied people without jobs.)

With Bernie in the White House Americans would be forced to answer questions such as “What does American socialism mean” and “Under what circumstances should someone be provided with a lifetime of free housing, food, and health care?” Perhaps the answers would result in a similar system to what we have now, but at least it would then be an affirmative answer rather than a place to which we were gradually nudged.

Related:

• “A Brief History of Divorce in America” — how we got to a situation, without any voter input, where having sex with three cardiologists (e.g., in Massachusetts) would yield the same spending power as going to medical school and becoming a cardiologist.

Full post, including comments

One of Hillary Clinton’s main campaign points is that women don’t get paid, on average, as much as men. She’s about to make the “gender gap” stat worse, though, by quitting a $20 million/year job giving talks to Goldman Sachs et al. and taking a $400,000/year government job (President of the United States). So Hillary will be taking dramatic action, once in office, to address a problem that she herself exacerbated!

Full post, including comments

“Save the children… from having to use Windows 10.” What if you were a schoolkid in rural Oaxaca (about an hour’s drive from the part the tourists enjoy) without Internet and wanted to learn from Wikipedia, Khan Academy, etc.? Kids on Computers has figured out how to set up labs in schools with a Raspberry Pi driving each child’s monitor. The Raspberry Pi runs a version of Unix, a web browser, a video player, etc.

Do you want to be a hero to some young Mexicans? Instead of spending time expressing righteous anger on Facebook regarding Donald Trump’s existence, you could work long enough to earn $68 pre-tax dollars and then make a tax-deductible purchase of a Raspberry Pi and have it shipped to my house or Avni Khatri’s. Email me (philg@mit.edu) if you need the address. Avni can send you a letter for the IRS. Look on the bright side of spendthrift government: With Bernie’s new tax rates, maybe you’ll get 90 percent of the $68 back!

[You might ask why I’m not buying a huge stack of these myself. It seems that there is a limit to how many a single individual can purchase.]

Full post, including comments

A friend who constantly advocates for Bernie Sanders, higher taxes, solar power, government-set wages for women, bigger government, etc. posted the following on Facebook:

… it is a sense of “female social superiority” as in, females generally speaking are more community oriented, more nurturing and more considerate of others.

I responded with

If you changed gender and became a female, Q, would you then become more nurturing and considerate of others? If Bernie Sanders became Bernice Sanders would she be more community-oriented?

That lead to the following exchange:

• Him: I have no idea. If I were a bird I would be able to fly, but currently we have no way to effect such a transformation. We do know that there are gender differences that affect behavior; there was a recent article about gender differences in stress response that I thought was an interesting example.
• Me: I don’t know of an available procedure for you to become a bird, Q. But isn’t it possible for you to identify as a woman starting tomorrow?
• Him: Possibly, but we have no reason to believe that my self-identification as a woman would sufficiently alter my behavior
• Me: So you’re saying that a transgender woman would not share these positive characteristics with a cisgender woman?
• Him: It is an unknowable question since two natural women may not share these properties or possess them in equal measure smile emoticon

Full post, including comments

I’ve spend part of this school year tutoring a student in AP Statistics. The course seems to be designed to help reinforce the classic paper “Why Most Published Research Findings Are False”. If the goal of high school is to prepare young people for citizenship and making decisions at work, AP Statistics seems like a spectacular failure. If the goal were to help people make decisions, wouldn’t the core of the class be hypothesis testing? People who take high school stats aren’t necessarily going to become professional statisticians but at least if hypothesis testing were the core they could be appropriately skeptical with regard to claims regarding, e.g., the efficacy of a new drug that taxpayers were being asked to pay for.

Hypothesis testing, however, is relegated to the end of the class/textbook/prep books. Thus for most of the class students have little context for why they are learning most of the material. Is it critically important to know what percentage of people think X based on a survey? Maybe for political pollsters who can then tell candidates how to pander to voters, but the most interesting uses of statistics seem generally to involve answering a question about whether an intervention was effective or not.

What do readers who are stats experts say? Would it make sense to weave hypothesis testing into the entire course?

Related:

• “Supplementary materials for teaching AP statistics?”
• “The Wrong Way to Teach Math” (New York Times article from February 27 using AP stats as the poster child for U.S. incompetence at designing math syllabi)
• “Everybody’s getting paid, but Raheem still can’t read.” (from “Schooled” in New Yorker)

Full post, including comments