Altruists should mostly follow standard investment advice

Recently, I’ve been seeing some really bad investment advice floating around the effective altruist community. This has two parts:

1. Bad arguments for taking extreme levels of risk with money you plan to donate.
2. Needlessly re-inventing the wheel in the area of personal finance, making mistakes along the way.

Instead, I think most EAs should follow standard investment advice, by which I mean the kind of advice you will get in books like Burt Malkiel’s A Random Walk Down Wall Street and William Bernstein’s The Four Pillars of Investing, or from the typical econblogger (for example), or from the good people at Vanguard.

(As an aside: I would advise caution about taking financial advice from anyone in the business of providing financial services, given the conflicts of interest that entails. However, although Vanguard is a mutual fund company, their unique customer-owned structure gives them unusually good incentives to do what’s best for their customers, and their fees are consistently the lowest in the business.)

I do not claim this approach is right for everyone. In particular, it may not be the right approach for people with unusually high net worth and/or levels of financial expertise. But I do think it is right for, say, the typical non-profit employee or Silicon Valley software engineer.

The standard advice, in a nutshell, is that you should invest in some mix of stock and bond index funds. You should lean more towards stocks if you have more tolerance for risk, and more towards bonds if you have less tolerance for risk. And yes, diversifying internationally is a good idea.

Combine the above points with the fact that you can generally take more risks the further you are from retirement, and you get Vanguard’s target retirement funds. Wealthfront, which I currently use, provides a slightly more complicated implementation of the same basic concept.

Now let’s talk about the first issue I brought up in the introduction: bad arguments for excessive risk-taking. The claim is that you should be risk-neutral with money you plan to donate, and therefore be willing to do things even riskier than putting 100% of your money in stocks.

When deciding whether to pursue a particular risky investment, there are two questions to ask:

1. How much risk do I really want to take?
2. Will I be adequately compensated for taking on these risks?

Question (2) can be very hard to answer. I’ve previously argued that startups may no longer offer founders and investors good risk-adjusted returns. But in this post, I’ll try to steer clear of those questions.

Instead, I’ll focus on the relatively simple example of leveraged investing, i.e. investing with borrowed money (or using derivatives to the same effect). Leveraged investing has recently been advocated by Brian Tomasik, and some people have gotten the impression that the EA consensus is you should make leveraged investments with money you plan on donating.

In theory, the problem of leveraged investing is simple, because leverage is supposed to multiply gains and losses equally. In reality, this is only true if you can borrow money cheaply, but the simplified “cheap margin” case is worth thinking about. If you are truly risk-neutral, in the cheap margin world, you should use as much leverage as other people will let you get away with when investing money you plan to eventually donate.

If you know anything at all about finance, alarm bells should be going off. “Use as much leverage as other people will let you get away with” is generally not considered sound financial advice. But it can sort-of be defended in an altruistic context, via arguments that it’s better to have a 20% chance of saving 100 lives than a 90% chance of saving 10 lives, and too deny that is sheer irrational loss aversion.

But if we continue to think about the implications of true risk-neutrality, they get even weirder. Suppose you were truly risk-neutral, and also a competent professional gambler and/or investor, meaning you can reliably identify positive-expected value bets even after accounting for adverse selection. Being completely risk-neutral, your only goal is to maximize the long-run expected value of your bets. What do you do?

The answer to this question is counter-intuitive, and it took me several months of reading and thinking about the issue to understand the answer. But–if you make all the assumptions in the last paragraph–the answer is that you should repeatedly bet every every penny you have on the highest expected-value bet you can possibly find, even when the bet carries a substantial risk of losing everything.

If this sounds like a good way to go broke, it is. Given a sufficient supply of positive-expectation but risky bets, you are virtually certain to go broke in short order.

This is counter-intuitive, because it seems like a terrible way to maximize the long-run expected value of your bets. However, as you look out over your possible futures, you will see ever tinier odds of ever more ridiculous payouts–payouts whose ridiculous size outweighs the small chance of achieving them in expected-value calculations.

Virtually guaranteeing your own bankruptcy seems bad. In fact, I’d argue it is bad–especially if everyone else in your social movement is doing the same, so that if you go bankrupt they probably will too. I mean, forget about the most extreme thought experiments here–it would be bad, for example, if Dustin Moskovitz decided gamble all the money he was planning on donating to Good Ventures on a single business idea.

So instead of “maximize the long-run expected value of your investments”, let’s try tweaking our goal a bit: “maximize the long-run expected value of your investments while minimizing the chance you go broke.” It turns out we can mathematically specify exactly how to do this: it’s called the Kelly criterion.

What does the Kelly criterion say about leveraged investing? Ed Thorpe, a mathematics professor who gained fame and fortune as a a hedge fund manager and blackjack player, once wrote a paper addressing this question. He concluded that if you’re using the Kelly criterion, you should borrow 17% of your portfolio’s value to invest in more stocks.

This is much less leverage than you will get from buying a leveraged exchange-traded fund (which is what Brian Tomasik has advocated). Furthermore, Thorpe assumed the ability to borrow at the same interest rate paid by T-bills, which nobody except the federal government can actually do and most people can’t even approximate.

So for people without access to cheap margin, 100% stocks is probably about as much risk as you should take, even with money you plan on donating. For money you do plan on donating, putting it in Vanguard’s total world stock fund looks like a good choice.

It might seem surprising that sophisticated mathematical analysis would lead to as simple of a strategy as investing all your money in the stock market but not using leverage. But maybe it’s not so surprising. Just by investing in the stock market, you’re already using leverage, in a sense.

This is because corporations borrow to finance new investment all the time. If it were easy to boost returns with leverage, it would suggest corporations aren’t borrowing enough. Furthermore, some corporations specialize in highly leveraged investments: we call these corporations “banks”.

If we step into the world of cheap margin for a moment, it might be possible to improve on Thorpe’s 117% stock portfolio. How? By buying a buying a diversified portfolio of stocks and bonds, and applying leverage to get stock-like returns with less risk. This appears to be basically what hedge fund giant Bridgewater’s All-Weather Fund does.

My guess is that this possibility is probably not relevant to small retail investors, but might be relevant if you have a seven-figure sum (or more) to invest. How much money you have matters because the more money you have to invest, the easier it is to get cheap margin (see Interactive Brokers for an example).

The difficulty of getting sufficiently cheap margin as a retail investor is probably why very few financial advisors recommend leveraged investing to ordinary people. So while there are exceptions, I maintain that most people should just follow the standard advice.

Steven Pinker is wrong about the Cuban Missile Crisis

For the most, I really like Steven Pinker’s book The Better Angels of Our Nature. People who talk about how awful the modern world is are generally just ignoring how much worse the past was. Pinker does a great job of showing this with numbers on how violence has, on the whole, declined over the past several thousand years. It’s not a straight-line decline. Things sometimes get worse in the short run. But the long term trend is very positive.

But–and this is a big but–as I’ve looked more into Pinker’s discussion of the Cuban Missile Crisis, the more I’m convinced that Pinker is just wrong about that. The issue with the Cuban missile crisis is that a sophisticated critic of Pinker can say, “maybe the median amount of violence at any one time has declined, but the risk of rare but huge conflagrations of violence–like nuclear war–has gone up. The Cuban Missile Crisis shows how much danger we’re in of something really catastrophic happening.”

Here’s what Pinker has to say about that. He describes the de-escalation of the crisis, and cites political scientist John Mueller to argue that the de-escalation was not just “an uncanny stroke of good luck.” Pinker writes:

Mueller reviewed the history of superpower confrontations during the Cold War and concluded that the sequence was more like climbing a ladder than stepping on to an escalator. Though several times the leaders began a perilous ascent, with each rung they climbed they became increasingly acrophobic, and always sought a way to gingerly step back down.

I bought a copy of Mueller’s book Retreat from Doomsday to see what it had to say. I was actually disappointed by how short the discussion of the Cuban Missile Crisis was. The book is mostly about the broader question of the decline of “Great Power war” (think the Napoleonic wars or the two World Wars). “How big was the risk of the Cuban Missile Crisis escalating into nuclear war?” only gets a few pages discussion.

Mueller’s basic argument is simple: war was unlikely because neither side wanted war and both were working hard to prevent it. That much is consistent with other things I’ve read, at least when it comes to Khrushchev and Kennedy. But I still think Mueller’s claim that the risk of nuclear war was “next to zero” ridiculous.

Focusing on the top leadership ignores the possibility that the top leadership could have lost control of the situation. More broadly, it ignores the possibility that other people could have been in power when the crisis, or a similar crisis, unfolded. Would Eisenhower, or Johnson, or Nixon, or for that matter Barry Goldwater have been as cool headed as Kennedy was?

I’ve read Robert F. Kennedy’s memoir Thirteen Days, and watched documentary The Fog of War, which consists of interviews with Robert McNamara, who was secretary of defense during the crisis. Both emphasize Kennedy’s exceptional leadership during the crisis, and how he reigned in US military leaders who initially wanted to attack Cuba. Furthermore, McNamara revealed that:

It wasn’t until January, 1992, in a meeting chaired by Castro in Havana, Cuba, that I learned 162 nuclear warheads, including 90 tactical warheads, were on the island at the time of this critical moment of the crisis. I couldn’t believe what I was hearing, and Castro got very angry with me because I said, “Mr. President, let’s stop this meeting. This is totally new to me, I’m not sure I got the translation right.”

“Mr. President, I have three questions to you. Number one: did you know the nuclear warheads were there? Number two: if you did, would you have recommended to Khrushchev in the face of an U.S. attack that he use them? Number three: if he had used them, what would have happened to Cuba?”

He said, “Number one, I knew they were there. Number two, I would not have recommended to Khrushchev, I did recommend to Khrushchev that they be used. Number three, ‘What would have happened to Cuba?’ It would have been totally destroyed.” That’s how close we were.

If Castro and the US joint chiefs of staff had been calling the shots, we easily could have had a nuclear war. Mueller claims that the Soviet leadership had issued a secret directive not to go to war even if the US invaded Cuba. However, according to the website of the JFK Presidential Library, a local commander had been authorized to use nuclear weapons if the US attacked. (It’s possible that both of these things were true–what if the first order was given, but it never reached the local commander in Cuba?)

And about the possibility that Kennedy and/or Khrushchev could have lost control of the situation–during the crisis, Robert Kennedy warned the Soviets of a risk of a coup against the President, and Robert Pastor, McNamara’s son-in-law, has claimed that the younger Kennedy brother wasn’t bluffing–that some of the commanders involved in the crisis had been insubordinate. On the Russian side, naval officer Vasili Arkhipov had to convince the captain of the submarine he was on not to launch a nuclear warhead after they lost radio contact with their superiors.

Arkipov’s heroic action happened fifty-three years ago today. So take a moment to think about how close we close we came to destroying the world that day. Tensions have been rising lately in Eastern Europe. I only wish I knew what I could do to reduce the risk of existential catastrophe from nuclear weapons.

Why I’m pessimistic about cancer cures and radical life extension

Epistemic status: idea I got in college biology nearly ten years ago. My memory of the relevant biology may not be perfect.

Recently, Sarah Constantine has been writing some posts on cancer treatments and how we might create better ones. This has reminded me of some thoughts I’ve had for awhile on cancer cures and life-extension, and why I’m quite pessimistic about the prospects for them.

(This is not a direct reply to Sarah–she hasn’t finished her post series yet, so it’s hard to evaluate her arguments.)

Basically, cancer and aging seem to be two sides of one coin. Cancer is typically described as cells growing out of control, but in a sense cancer is more fundamentally about cells not dying when they should. Conversely, a lot of aging involves slow cell death throughout the body. This means trying to solve both problems is a potentially delicate balancing act.

For example, telomeres seem to play an important role in aging. I won’t get in to the technical details, but basically a lot of people think telomere elongation is promising as an aging treatment. Unfortunately, it turns out that willy-nilly telomere elongation causes cancer.

Conversely, well, I probably don’t need to tell you that radiation therapy and chemotherapy often suck. Cancer cells aren’t alien invaders from outside your body, they’re your own cells gone bad, and how exactly they go bad can vary a lot. So it’s hard to kill cancer without killing some cells you don’t want to kill.

In other words, medicine is very broadly about making sure your cells don’t die when you don’t want them to. Cancer, however, is about cells not dying when you want them to (until you and all your cells are killed by the cancer). This is an example of the general problem with false positives vs. false negatives. The hard thing isn’t reducing one or the other–it’s reducing one without increasing the other.

Similarly, there’s nothing actually difficult about killing cancer cells–cyanide should do the trick. And maybe preventing aging-related cell death is just a matter of throwing telomerase at the problem. Doing both at the same time, however…

If this model is right, solving cancer and aging will likely make solving the problem of AIDS look trivial by comparison. For a long time, AIDS was the big scary disease we couldn’t do anything about, but nowadays antiretrovirals are pretty effective and more recently we’ve gotten good AIDS prevention drugs. But AIDS was easy–we never had to worry about making a person’s viral load too low.

Similarities between anti-gay and anti-sex work arguments

Ozy has complained to me that my post on research on sex work from a couple week ago didn’t go into enough detail. To this I must plead guilty. Partly, once I found the Vox.com article, I got lazy. But there was another factor: I realized I’d seen this movie before.

Nowadays, when you hear evangelical Christians talk about gay marriage, it’s about how if they can’t discriminate against gays and lesbians because of their “sincerely held religious beliefs”, they’re being oppressed. But not long ago–before this started sounding ridiculously homophobic–it used to be common for them to claim that it wasn’t just about religion: homosexuality was a disease and they could prove it.

How? By citing studies based on ridiculously unrepresentative sample populations, for one. For example, they might claim that the median gay man has hundreds of sexual partners during his lifetime–a claim based on studies done in the 1970s, based on self-selected samples (i.e. whoever was most willing to talk about their sex lives to some sociologists), which in some cases relied heavily on advertising the study in gay pickup joints and bath houses. (Source and discussion.)

Similarly, they’ll claim shockingly high rates of sexually transmitted infections among gay men, citing studies which turn out to have recruited their participants primarily from STI clinics. Of course, they’ll never mention this, leaving their readers to assume the studies are representative of all gay men.

Parallel problems exist in studies on sex work. As sociologist Ronald Weitzer explains, claims about all sex workers will be made based on studies of street-based sex workers, or sex workers in jail, or sex workers who reached out to service organizations for help.

In both cases (anti-gay and anti-sex work), writers pushing an ideological agenda will ignore explicit warnings in the studies themselves. For example, in their book Half the Sky, Nicholas D. Kristof and Sheryl WuDunn cite a Norwegian government report as showing Sweden’s sex work laws have been very successful at reducing the number of sex workers in Sweden.

However, if you actually read the report, it’s explicit that the decline was only observed in the number of outdoor sex workers. Not only do we have no idea what happened to the number of indoor sex workers, the report also notes the possibility that sex work shifted indoors due to mobile phones and the internet, not changes in the law.

Anti-gay and anti-sex work writers have also used similar rationalizations for generalizing based on non-representative samples. For example Thomas Schmidt, author of the anti-gay tract Straight and Narrow, argues (p. 128):

Some may claim that the studies cited are not representative of homosexuality as they observe or experience it, to which I can only appeal to the evidence: nearly two hundred sources, multiple sources at key points covering different geographic areas, different times, and samples both clinical and nonclinical, random and recruited.

Similarly, clinical psychologist turned anti-sex work activist Melissa Farley wrote the following in her a reply to Weitzer:

Weitzer bemoans our lack of a random sample. As other researchers of prostitution have noted, it is not possible to obtain a random sample of people currently prostituting (McKeganey & Barnard, 1996). Investigators, therefore, use a variety of techniques to learn about the experience of prostitution for those in it. Generally, smaller numbers of interviewees limit the generalizability of results. We have reported data from a large number of respondents in different countries and in different types of prostitution.

In both cases, the problem, as anyone who understands social science research will realize, is that things like geographic diversity are no substitute for true random sampling. Flawed sampling techniques can easily produce similar biases in multiple locations.

Of course, just because many of the relevant studies are flawed does not mean that gay people or sex workers are not at higher risk for certain problems than straight people or non-sex workers. But that brings me to another similarity between anti-gay and anti-sex work arguments: the assumption that if X is associated with something bad, X itself must be inherently bad.

For example, there is unambiguous evidence that gay teens are at increased risk for suicide. But the cause of that increased risk matters enormously. If homosexuality somehow caused suicide directly, and ex-gay therapy really could cure it, sending teens off to ex-gay camps might be defensible. But because we know the problem is stigma and ex-gay therapy doesn’t work, parents need to realize that sending their son off to an ex-gay camp could kill him.

Similarly, it matters how much of the problems that sex workers face is the result of stigma and criminalization. Furthermore, high rates of mental illness among sex workers may be due to some mentally ill people turning to sex work because they have trouble finding other jobs. That’s an argument for a stronger social safety net, and stronger support for the mentally ill, but simply taking away those sex workers’ one means of supporting themselves seems unlikely to help them.

Some of my readers may assume that the difference between the battle for gay rights and the battle over sex work is that gay rights were only opposed by people who hate gays and lesbians, while opponents of sex work are motivated by a genuine desire to help sex workers. But it used to be common for anti-gay propaganda to strike a similar compassionate pose.

This may be a little hard to imagine now, now that attempts to “cure” homosexuality have been so thoroughly discredited, with even former proponents admitting the therapies do not work. But not long ago, it was easy to find people claiming with a straight face that because homosexuality is so harmful, true compassion towards gays and lesbians meant trying to cure them.

Similarly, rhetoric about “rescuing” sex workers is disingenuous if we’re talking about police raids on sex workers who don’t want to be “rescued.” For example, Kristoff and WuDunn write, without a trace of irony:

It’s enormously dispiriting for well-meaning aid workers who oversee a brothel raid to take the girls back to a shelter and given them food and medical care, only to see the girls climb back over the wall.

Shockingly, Kristoff and WuDunn admit that the police who conduct these raids are often deeply corrupt and “crackdowns” on sex work are sometimes a matter of police demanding larger bribes from brothels. Pardon me if I’m unable to take their “compassion” any more seriously than I take the “compassion” of the anti-gay right.

Fantasy Republican presidential campaign platform

Note: this is the kind of half-baked post I would normally save for Tumblr, but it felt a bit long to leave there.

Is it just me, or do Republican presidential candidates tend to be a lot more exciting than Democratic ones? They throw out crazy policy proposals left and right, and every candidate has a different crazy plan! This might be my left-leaning bias, but somehow I can’t imagine a Democrat proposing something as downright weird as Herman Cain’s “9-9-9” plan from 2012.

Because of this, I think it would be a lot of fun to be a Republican presidential candidate. If I, by some miracle, ended up as the Republican governor of, say, California, I’d definitely be running for president right now, and here would be my platform:

Jeb! says he can do 4% economic growth? Okay, I’ll do 5%! And I’ll solve all our problems with the financial sustainability of Social Security and Medicare at the same time without the need to cut a dime from those programs (unless it’s to get seniors a better deal on healthcare). How?

Create a new class of guest workers visa that will be auctioned off in large numbers. All guest workers who come on the visa will need to pass rigorous background checks. Plus, a portion of the money raised in the auction will be set aside for better border enforcement, and block grants to states to spend on more police. The plan will also help ensure we have enough young workers to take care of our seniors.

Talk about how Muslims should be able to come here if they agree to “accept our values.” Promise to pass an anti-sharia law, secretly base details on a proposal by British secularists.

Of course, no good Republican platform would be complete without tax cuts. So let’s make the top tax rate on all income, regardless of source 28%–a small increase for some types of capital gains, but a big cut for short-term capital gains and labor income.

Also, we’ll massively raise contribution limits on IRAs, possibly to [old limit] or a fixed % of income, whichever is higher. Make it easier for people to take money out of their IRAs early without penalty. Promise to cut corporate tax rates while you’re at it.

How will we pay for such tax cuts? Well, if we’re lucky there will be some money left over from the visa auction, so start there. Partly, we won’t pay for the cuts entirely (I’d quietly ask some economists how much we can increase the deficit without affecting inflation). There will also be noise about closing loopholes and slashing corporate welfare. And finally, we’ll propose a value-added tax.

Of course, value-added taxes (and consumption taxes more generally) are regressive. So you’d want to give people a tax rebate roughly equal to the tax they’d likely pay in in value-added taxes on basic necessities. Speaking of tax rebates, this brings us to the third major leg of my domestic platform: refundable tax credits, and a new tax-refund scheme.

The welfare state, you see, is terribly inefficient. It’s a bureaucracy, after all. So whenever possible, ditch it, and give people refundable tax credits instead. Food stamps? Definitely replace with a refundable tax credits. Welfare? Probably. Maybe you could even replace unemployment and disability with tax credits.

And of course, it’s incredibly unfair to make people wait a whole year to get a refund if the government has overcharged them on their taxes. So start a program to let people get any likely-to-be-refunded tax credits–or refunds that they’re clearly going to be owed because of a change in their circumstances–added to their paycheck. If they’re not working, they can get their choice of a check or direct deposit twice a month.

Attack “Obamacare” as a giveaway to insurance companies. Forcing you to buy their product? C’mon! And talk about how Medicaid is evil for forcing middle-class Americans to become poor. Promise to fix both of these problems.

Promise to restrict abortion up to the maximum of what’s allowed under Roe v. Wade. Accuse competing candidates who promise more than that of lying. When talking about gay marriage, quote C.S. Lewis on divorce a lot. Tell evangelicals that they have a right not to recognize gay marriage, just as Catholics have a right not to recognize divorce.

On foreign policy, mostly talk about the threat from Russia a lot. Say Iran and ISIS are a distraction from Russia. Make lots of noise about sticking up for America’s allies, and making clear to Russia that an attack on any NATO member is an attack on America.

(Figuring out the left-wing spin on many of these positions is left as an exercise for the reader.)

Effective Altruism and mistakes of globalization

Most people in the effective altruism movement are not economic radicals. You’re unlikely to hear them talk about the evils of capitalism, but you’re also unlikely to hear them say that taxation is theft and anarcho-capitalism is the only way to go. Or, as Dylan Matthews once put it:

That lends itself to a particular political bent, which is left of center but technocratic, friendly to markets (when they can be shown to work), and, above all, cosmopolitan… They’re sympathetic to the welfare state but far more jazzed about open borders.

Open borders is probably the most radical political view you’ll regularly hear advocated in the EA movement, but even that can be seen as a logical extension of views about the benefits of free trade that became standard for establishment liberals under Clinton, just applied to labor rather than goods. It’s a view that would have a shot at widespread elite support, if only it weren’t so toxic to the average voter.

I basically share the standard EA view here. My views are left of center, technocratic, friendly to markets, and above all cosmopolitan. Yet lately, I’ve come to see why many people are nervous about this combination of views, particularly the “technocratic” part. Why did my perspective shift? It all started with Greece.

I assume most of my readers know about Greece’s current economic mess–namely, that as I write this, Greece’s unemployment rate is around 25%. I don’t know how many people have followed the economic policy arguments, though. Let me give a short summary.

Normally, a country like Greece would deal with its problems by devaluing the currency. This would be painful, but would boost exports and tourism by making both cheaper, which would in turn bring down unemployment. Greece can’t do this, though, because of the Euro. Greece could solve this problem by leaving the Eurozone, but it hasn’t because lots of people are ideologically committed to the Euro.

The other way to solve this problem is for the rest of Europe to just give Greece a bunch of Euros. This is what happens regularly in the United States to some extent–richer states support poorer states. If unemployment was low in California but at 25% in Alabama, Congress and the Federal Reserve wouldn’t write off Alabama as somebody else’s problem, even if the solution required giving Alabama some money.

But the rest of Europe doesn’t want to do this either. So instead we get demands that Greece basically hand over the running of much of its government to European central institutions. The nicest thing you can say about the policies these institutions have imposed is that they might be sensible under other circumstances, but they have absolutely no chance of getting Greece out of its current mess.

It’s relevant that it’s widely taken for granted that Eurozone economic policies are really set by the richer Eurozone members, particularly Germany. So seeing the Greece crisis unfold got me wondering about other accusations of rich countries having imposed terrible economic policies on poorer ones. Was there any truth in them?

It turns out that among informed people, this isn’t terribly controversial. The answer is “yes”. When I was research this issue, one of the things that sealed it for me was former IMF economist Kenneth Rogoff’s reply to fellow economist Joseph Stiglitz. The reply is bizarrely petulant; a major line of counter-attack is to admit mistakes but then try to shift blame to Stigliz for daring to criticize those mistakes. (It damaged confidence, you see.)

What exactly were the mistakes? Paul Krugman, in his book The Return of Depression Economics gives two main “things the IMF clearly did do wrong”:

First, when the IMF was called into Thailand, Indonesia, and Korea, it quickly demanded that they practice fiscal austerity–that they raise taxes and cut spending in order to avoid large budget deficits. It was hard to understand why this was part of the program, since in Asia (unlike in Brazil a year later) nobody but the IMF seemed to regard budget deficits as an important problem. And the attempt to meet these budget guidelines had a doubly negative effect on the countries: where the guidelines were met, the effect was to worsen the recession by reducing demand; where they were not met, the effect was to add, gratuitously, to the sense that things were out of control, and hence feed the market panic.

Second, the IMF demanded “structural” reform–that is, changes that went well beyond monetary and fiscal policy–as a conditions for loans to afflicted economies. Some of these reforms, like closing bad banks, were arguably relevant to the financial crisis. Others, like demanding that Indonesia eliminate the practice of giving presidential cronies lucrative monopolies in some business, had little if anything to do with the IMF’s mandate. True, the monopoly on cloves (which Indonesians like to put in their cigarettes) was a bad thing, a glaring example of crony capitalism at work. But what did it have to do with the run on the rupiah?

Krugman also writes that “never in history had so many first-rate economists been in positions of so much authority.” So what went wrong? Is modern mainstream economic theory fatally flawed, as some have argued? Krugman has a different explanation:

International economic policy ended up having very little to do with economics. It became an exercise in amateur psychology, in which the IMF and the Treasury Department tried to persuade countries to do things they hoped would be perceived by the markets as favorable.

This analysis foreshadows Krugman’s more recent, more acerbic complaints about belief in the “confidence fairy.”

Krugman’s book makes another interesting point: while most economists at the time (including at the IMF) opposed capital controls, there are still tradeoffs involved in the decision to abandon them. Capital controls have definite costs, but they may be a necessary evil if a country wants both insulation from volatile currency markets and the ability to have a monetary policy (i.e. not end up like Greece).

Joseph Stiglitz, in his book Globalization and its Discontents, goes a little further in his critique than Krugman. Stiglitz is not anti-globalization. He’s clear that trade and markets have lifted millions out of poverty in countries like China, South Korea, and Taiwan. He also criticizes anti-global activists whose agenda involves rich countries strengthening their protectionist policies, something that would almost certainly hurt the global poor (whether those activists admit it or not).

However, Stiglitz argues, it matters a lot how free-market policies are implemented. Pacing and ordering matter. For example, Stiglitz contrasts the bungled transition to capitalism of former members of the Soviet Union–where outside experts pushed for rapid changes, and the result was plundering of national resources by corrupt oligarchs–with China’s slower and more successful transition to more market-oriented policies.

Again, it’s striking how problems were caused not by blind adherence to economic theory, but by neglect of basic economic theory. Here’s Stiglitz on the 1998 Russian financial crisis:

In the weeks preceding the crisis, the IMF pushed policies that made the crisis, when it occurred, even worse. The Fund pushed Russia into borrowing more in foreign currency and less in rubles. The argument was simple: The ruble interest rate was much higher than the dollar interest rate. By borrowing in dollars, the government could save money. But there was a fundamental flaw in this reasoning. Basic economic theory argues that the difference in the interest rate between dollar bonds and ruble bond~ should reflect the expectation of a devaluation. Markets equilibrate so that the risk-adjusted cost of borrowing (or the return to lending) is the same. I have much less confidence in markets than does the IMF, so I have much less faith that in fact the risk-adjusted cost of borrowing is the same, regardless of currency. But I also have much less confidence than the Fund that the Fund’s bureaucrats can predict exchange rate movements better than the market.

The more I read about these issues, the more sympathetic I am to the claim that global development needs to happen in accordance with democratic principles. Sure, voters may not understand the issues as well as professional economists. But voters can, at least, be trusted to care about their own well-being, and the politicians they elect have strong incentives to listen to them.

Foreign technocrats, on the other hand, may be tempted to ignore what they know of economics if it makes them look like responsible adults. Or if it seems likely to protect them against charges of irresponsibility later. Or if it serves the interests of whoever’s paying their salary. At any rate, some of the IMF’s blunders seem unlikely to be ever chosen by a democratically elected government for itself.

Or maybe this impression is misleading. Maybe I’m biased because I live in the United States, a country whose economic policy, if not perfect, has usually been non-disastrous. Maybe if I knew more about the policies of democratically-elected politicians in, say, India, I’d have second thoughts. But I at least understand the apprehension many people have about letting foreign technocrats run things.

I don’t think this is a strong objection to the object-level ideas most popular among effective altruists. GiveWell’s top charities promote narrowly-targeted interventions, not radical and comprehensive economic reform. GiveDirectly, in particular, is about as respecting of poor people’s autonomy as you can get.

Furthermore, I don’t see many EAs arguing that the thing to do to fight global poverty is to push the kind of policy reforms the IMF once pushed. When EAs do advocate for policy change, it tends to be changes to rich country policies, things like open borders. Still, it’s worth, worth being sensitive to these worries, especially as EA explores further into the policy realm.

This also suggests a possible direction for EA policy advocacy: advocacy for making the governance of certain global institutions more democratic. This may well be a pipe dream, but I think it’s worth at least chewing on a bit.

Almost all our research on sex work is terrible

Lately, I’ve been reading up on the empirical research on sex work, and the effects of various policy regimes re: sex work. And the thing is… it’s almost all terrible?

I mean, I haven’t done an exhaustive search. But if there’s a huge amount of good research, well-informed people on all sides of the debate over sex work are inexplicably failing to cite the research that would support their views. Not that the researchers are idiots. They know about the methodological flaws in our studies on sex work, but it turns out that research on extremely marginalized populations is hard. Who knew.

When I first realized this, I thought, “wait a minute, shouldn’t something like the General Social Survey or The Social Organization of Sexuality have good data about sex work?” It turns out those studies provide a decent estimate of the percentage of women in the US who have ever sold sex (it’s probably about 1-2%), but they don’t provide much beyond that.

I guess the problem is that while those studies sets have good sampling techniques and are relatively large (a few thousand people), that’s will translate into at most a few dozen sex workers in the sample. So if you try to analyze that subpopulation, your result is probably going to be non-representative in some ways just by random chance.

But the absence of better research is still a little puzzling. My impression (and this is Ozy’s impression to) is that the research on gays and lesbians is better than the research on sex workers, and “people who have been paid for sex” apparently isn’t a much smaller proportion of the population than gays and lesbians. Of course, part of the difference is greater marginalization of sex workers, but part of it may also be gays becoming a major focus of research due to AIDS.

If you want a good overview of the available research on sex work, this Vox.com article is a good place to start, and matches my independent impressions. The article does claim to find a couple good studies, both of which broadly support decriminalization. (There’s also some data on sex trafficking convictions in Germany, but as the article notes, what that data means is debatable.)

But if you’ve followed the story of replication problems in science, you know that two studies don’t prove a ton. So I’m left with the main reason I support decriminalization being priors. Abuse of domestic workers is a problem (and in fact caused an international incident between the US and India last year), but making it illegal to hire a maid seems unlikely to be the best solution to that problem. I see no reason–aside from moral panic around sex–not to apply the same reasoning to sex work.

Why Eric Drexler’s critics are right

Alyssa Vance wrote a blog post urging people to ignore one of Eliezer Yudkowsky’s critics (who goes by the handle su3su2u1), entirely based on said critic dismissing Eric Drexler as a crackpot. I think Vance has already gotten a lot of spot-on pushback, both in her blog comments and elsewhere (see e.g. here, here, and here), but coincidentally I’d been thinking about writing a post on Drexler recently, and “ignore people who think Drexler is a crackpot” is a hilariously terrible heuristic.

At best, Drexler is someone whose ideas are highly speculative, based on problematic assumptions, and have yet to yield a fruitful research program; and whose understanding of the relevant science appears superficial. At worst, he’s been called a crackpot by many scientists who know what they’re talking about. I’ll let my readers decide whether the label is appropriate.

Some background: Drexler was originally an engineer who had worked for NASA and gotten a master’s degree in aerospace engineering. He got the idea that you might be able to build little robots and factories at atomic scales, and wrote a book on the subject called Engines of Creation, published in 1986. In 1992, he got a Ph.D. under computer scientist Marvin Minsky. You can read his dissertation online.

Eliezer Yudkowsky has claimed that, “if you say that nanomachines cannot work, you must be inventing new physics.” This is a good example of Yudkowsky spouting off on scientific topics without knowing what he’s talking about.

In reality, you can’t derive chemistry from physics. I mean, we can conjecture that in principle you could, given unlimited computing power, but we don’t have unlimited computing power. Wikipedia’s article on computational chemistry is a good introduction here. (Edit: see note below.)

Wikipedia mentions calculating the properties of molecules with 40 electrons, and I’ve heard of doing quantum models of a few hundred atoms with good approximation techniques. But you’re not going to deduce molecular biology from first principles, as a protein can easily have tens of thousands of atoms.

A lot of tech nerds don’t pay much attention to chemistry because it’s not as pure as math or physics, but I think chemistry is the relevant domain of expertise to apply here–Drexler’s claims about molecular assemblers are basically claims about a radical new way to do chemistry.

(I was a biochem major in college for two years before I realized I didn’t want to be a doctor, and it improved my general understanding of the world in ways just studying physics never would have.)

So physicist Richard A.L. Jones is correct when he says:

First, those building blocks–the cogs and gears made famous in countless simulations supporting the case for the singularity–have some questionable chemical properties. They are essentially molecular clusters with odd and special shapes, but it’s far from clear that they represent stable arrangements of atoms that won’t rearrange themselves spontaneously. These crystal lattices were designed using molecular modeling software, which works on the principle that if valences are satisfied and bonds aren’t too distorted from their normal values, then the structures formed will be chemically stable. But this is a problematic assumption.

A regular crystal lattice is a 3-D arrangement of atoms or molecules with well-defined angles between the bonds that hold them together. To build a crystal lattice in a nonnatural shape–say, with a curved surface rather than with the flat faces characteristic of crystals–the natural distances and angles between atoms need to be distorted, severely straining those bonds. Modeling software might tell you that the bonds will hold. However, life has a way of confounding computer models. For example, if you try to make very small, spherical diamond crystals, a layer or two of carbon atoms at the surface will spontaneously rearrange themselves into a new form–not of diamond, but of graphite.

Similarly, Nobel-prize winning chemist Richard Smalley’s criticisms of Drexler are right on target. Vance dismisses Smalley’s criticisms as “centered around a silly analogy comparing molecular chemistry to romance,” but what’s actually going on is Smalley explaining basic chemical principles in a way laypeople can understand. As Smalley says in one of his replies to Drexler:

You cannot make precise chemistry occur as desired between two molecular objects with simple mechanical motion along a few degrees of freedom in the assembler-fixed frame of reference. Chemistry, like love, is more subtle than that. You need to guide the reactants down a particular reaction coordinate, and this coordinate treads through a many-dimensional hyperspace.

Describing this motion through hyperspace as a waltz, as Smalley does, is as good an analogy as any.

To give another not-literally-true, but still illuminating analogy, imagine chemical bonds as springs rather than the sticks you often get in molecular modeling kits. The hyperspace Smalley mentions is the many mathematical degrees of freedom a contraption built from these springs is going to have. Each spring can not only be stretched and bent, they can bounce about chaotically as you’re trying to make your chemical reaction happen.

That’s why Smalley says that in order to have the kind of control of chemical reactions that Drexler envisions, you can’t just have one nanobot arm for each molecule. You’re going to need to control the movement of many atoms all at once. Hence the “fat fingers” problem. Drexler’s reply to Smalley on this point suggests he doesn’t understand chemistry well enough to understand what Smalley is saying.

Smalley, by the way, got his Nobel for discovering so-called “buckyballs”, tiny spheres made out of exactly 60 carbon atoms arranged in a soccer-ball pattern. But these molecules are synthesized by vaporizing graphite, not with nanobots. Today, serious scientific work in “nanotechnology” generally has a lot more in common with Smalley’s work than Drexler’s. That’s why I say Drexler’s ideas haven’t yielded a fruitful research program.

As for whether Drexler is a crackpot, well, different people have different views. From a Wired article on Drexler:

“It’s very impressive, there is no question,” said MIT chemist Rick Danheiser, who served as Drexler’s thesis adviser, in 1992. “I couldn’t have done a better job.”

“It showed utter contempt for chemistry,” countered Danheiser’s colleague Julius Rebek. “And the mechanosynthesis stuff I saw in that thesis might as well have been written by somebody on controlled substances.”

The first paragraph contains a mistake, Danheiser was on Drexler’s committee but Minsky was Drexler’s adviser. Danheiser was the only chemist on the committee, so my guess is that Drexler did not get an especially well-rounded chemistry education when getting his doctorate. I’m speculating, but it’s possible Drexler did impressive work with molecular modeling software, yet his ideas about the significance of his models look, well, to a chemist at least, like something you’d come up with while on controlled substances.

If this sounds implausible, imagine a mathematician who was firmly convinced that P=NP. He did his dissertation on all the astonishing implications this would have! And let’s assume the thesis is technically correct, impressive even. But when he starts saying that the only thing holding back all kinds of technological miracles is the misrepresentations of his ideas being spread by malicious computer scientists… a Ph.D. would be no proof of sanity. Nor would a lack of peer-reviewed journal articles refuting his ideas.

If Vance wants technical publications, chapter 5 of this report essentially says, in very formal language, that we can’t know that Drexler’s theoretical calculations have anything to do with reality. If she’s not satisfied by that, unfortunately the truth is there’s just not much more to say about the issue.

Note: bartlebyshop tells me the Wikipedia article is bad. Sorry, I was looking for a source that vaguely backed up things I’d learned in undergrad. But if anything, the above discussion may actually understate how hard computational chemistry is.

Why animal advocates should read Temple Grandin

Temple Grandin is a professor of animal science at Colorado State, who is also known for her autism rights activism and work as a consultant for the meat industry. Many animal rights advocates do not have a high opinion of her. The Daily Pitchfork is fairly representative:

Grandin is a paid industry consultant. She profits financially by designing industrial slaughterhouses. She supplements her income by writing books and delivering speeches about those designs. Whatever animal welfare advice she offers should always be framed in the context of her monetary connection to industrial agriculture.

It should also be noted that big agriculture—big beef in particular—adores Grandin. She approaches agricultural “reform” from a compellingly safe perspective, one as much informed by her Ph.D. in animal science as her autism…

Obviously, one would think, Grandin’s empathy for these animals runs deep, deep enough at least for us to trust her as a viable source of information on their welfare.

But her real job is to help agribusiness kill them. Grandin argues that industrial slaughter should be as peaceful for animals as possible. But it turns out that cattle rendered calm by Grandin’s architectural designs turn the grimmest work of agribusiness—slaughter—into a more efficient and emotionally palatable process.

Calm cows are more likely to go gently. And cows who go gently—say, by not thrashing around inconveniently before being shot in the head with a steel bolt—enhance productivity. When Grandin’s relationship to industrial agriculture is placed in this context, the welfare benefits her slaughterhouses offer are significantly diminished.

On one side of the scale, cows raised under horrible feedlot conditions are spared a few moments of anxiety before their throats are slit. On the other, industrial agriculture per se is not only reified as a legitimate (and more productive) practice, but it becomes in the eyes of consumers a welfare-oriented endeavor.

However, after seeing Robin Hanson praise Grandin on Facebook, I decided to buy a copy of her book Improving Animal Welfare: A Practical Approach. Much of the book consists of extremely detailed advice on practices for improving animal welfare, not of interest to the general reader, so I ended up only skimming much of it. The book is roughly half Grandin’s writing, and half chapters contributed by a variety of other researchers.

Based on what I did read, I think that if Grandin is a shill for the meat industry, that’s all the more reason for animal advocates to read and quote her work. In spite of the academic tone of the book, and any biases she may have, the picture of animal agriculture she paints is often not pretty.

An important source of information for the book is Grandin’s experiences conducting audits of farms and slaughter houses. Among the things she reports seeing are disturbingly high rates of lameness among farmed animals:

The author observed a big increase in lame slaughter weight pigs between 1995 and 2008. A major breeder of lean rapidly growing pigs did nothing about it in the USA until in some herds 50% of the slaughter weight pigs were clinically lame… This breeder was selecting for leanness, loin-eye size and rapid growth, and over a 10-year period did not notice that there were more and more lame pigs.

She also describes the kind of horrific overcrowding I normally expect to hear about from animal rights advocates:

To have a bare-minimum acceptable level of welfare animals must NEVER be jammed into a crate or pen so tightly that they have to sleep on top of each other. The author has observed some caged layer farms where hens had to walk on top of each other to reach the feeder.

She also discusses the problem of biological overloading, in which breeding animals for rapid growth and extremely high milk and egg production leads to welfare problems such as lameness, increased aggression, and dairy cows that last only two lactations. The problems that have resulted from selective breeding have recently become fairly well known, see for example this Vox.com article titled, “Chickens have gotten ridiculously large since the 1950s”.

One thing I did not realize before reading Grandin’s book, however, is the way that parellel problems are caused by the drugs animals are fed, or, in her words, the “welfare problems caused by over-use of beta-agonists, rBST growth hormone and other performance-enhancing substances.” She writes:

The author has also observed healthy pigs that had been fead too much Paylean(R) (ractopamine) that were too weak to walk from one end of a lairage to the other. This is another example of bad conditions that some people perceived as normal. The author observed pigs 30 years ago that were strong enough to walk up long, steep ramps.

Grandin does claim that some farms are better than others when it comes to animal welfare. I think this is probably true, and important to understand for organizations like the Humane Society of the United States, which work with farmers to improve welfare on their farms.

In general, Grandin’s book has a lot of material that I think would be useful for orgs like HSUS–though I wouldn’t be surprised if HSUS already knew this, and keeps a copy or two of the first edition of Grandin’s book around their offices.

Among the chapters written by other people, one of the best was by philosopher Bernard Rollin, who devotes his chapter to arguing that “animal welfare” cannot be solely defined in terms of what makes animals most productive. One of the major ideas of his chapter is “the end of husbandry.” By “husbandry,” Rollin means the way how farmed animals were raised almost everywhere before the 20th century, practices that were increasingly abandoned as farming became industrialized:

Whereas husbandry animal agriculture stressed putting square pegs into square holes, round pegs into round holes, while producing as little friction as possible, industrialized animal agriculture forced square pegs into round holes by utilizing what I have called ‘technological sanders’, such as antibiotics, hormones, extreme genetic selection, air handling systems, artificial cooling systems and artificial insemination to force animals into unnatural conditions while they none the less remain productive…

In husbandry conditions, animal suffering would usually lead to loss of productivity. Under industry, however, the technological sanders mentioned sever the close connection between productivity and welfare, in so far as the animals may suffer harm in ways that do not impinge on economic productivity.

Rollin also mentions something whose significance he doesn’t seem fully aware of, the concept of ‘sound science’:

When one discusses farm animal welfare with industry groups or with the American Veterinary Medical Association, one finds the same response – animal welfare is solely a matter of ‘sound science’.

Those of us serving on the Pew Commission, better known as the National Commission on Industrial Farm Animal Production, encountered this response repeatedly during our dealings with industry representatives. This commission studied intensive animal agriculture in the USA (Pew Commission, 2008). For example, one representative of the Pork Producers, testifying before the Commission, answered that while people in her industry were quite ‘nervous’ about the Commission, their anxiety would be allayed were we to base all of our conclusions and recommendations on ‘sound science’.

Rollin goes on to argue that what counts as “sound science” depends on whether your goals are mere productivity or you actually care about animal welfare. But what he seems to miss is that the the phrase “sound science” is a corporate buzzword which, in context, generally means “scientific findings amenable to corporations.” It appears to have origins in tobacco-industry PR efforts, and has also been used to spread confusion over other scientific issues like global warming.

Seeing Rollin mention the phrase makes me wonder, “what scientific issues is the factory farm lobby lying about?” I’m not talking about ethical issues, I mean, what scientific issues as straightforward as global warming or the effects of second hand smoke are they lying about? Trying to find this out seems like potentially a good project for an animal advocate to work on. (The website of the Animal Agriculture Alliance looks like a good place to start.)

Grandin herself is quite clear on the difference between productivity and welfare. For example:

Research shows that sows can be highly productive in stalls, but confining a pig in a box where she cannot turn around for most of her life is not acceptable to two-thirds of the public… Confining an animal for most of its life in a box in which it is not able to turn around does not provide a decent life.

Similarly:

When pigs or chickens are jammed too tightly into a house, the welfare and productivity of each individual animal usually declines. Unfortunately, there is an economic incentive to do this because the overall output of eggs or meat per house may be greater.

How, according to Grandin economic incentives aren’t the sole source of problems with animal welfare. Ignorance is also a major source of problems. For example, she discusses how “bad becomes normal”, when animal welfare problems grow slowly and farmers don’t notice, and think what they see is normal and unavoidable. This happens because they don’t know what conditions are like on other farms and therefore don’t realize that not all farms are as bad as theirs.

Managers may also be out of touch with the day-to-day operations of their farms. Grandin talks about her experience giving seminars on handling techniques:

Employees were taught to use behavioral principles of animal movement, stop yelling and greatly reduce the use of electric goads. A year later when the author returned to re-evaluate the handling practices, it was discouraging to observe that many employees had reverted back to their old rough ways. When the manager was informed that the animal handling methods were bad, he was surprised and upset. Because the regression back into the old rough ways had happened slowly over the course of a year, the manager did not notice the slow deterioration of handling practices. Bad practices had become normal because the manager had not measured handling practices in an objective manner.

The importance of measuring animal welfare is a major theme of Grandin’s book. Examples she gives of measures of animal welfare that can be objectively quantified include bruises, sores, rates of lameness and broken bones, and cortisol levels.

Grandin also mentions measuring vocalizations, while being careful to note that many species of animals will avoid showing pain when humans are around. This means that with those animals vocalizations must be measured when the animals do not know they’re being observed (for example, by using video cameras).

Grandin emphasizes the importance of clear quantitative standards for what is acceptable, as opposed to vague requirements. She writes:

Unless a numerical limit is placed on an animal-based measure, it is impossible to enforce it in an objective manner. An example would be 5% as the maximum acceptable percentage of lame animals. Data presented from the studies reviewed in Chapter 1 show that well-managed dairies can easily achieve this. Vague terms such as minimizing lameness should not be used because one auditor may think 50% is acceptable and another may consider 5% lame animals as a failed audit.

Setting the standard at one level may result in even lower rates of problems in practice:

Data from audits done in many beef and pork plants indicated that for a plant to reliably pass an audit of 100 animas at the 1% level their actual falling percentages dropped to less than one in 1000.

Unfortunately, she notes, “there are some politicians and policy makers who make standards vague on purpose.” For example:

Many people are reluctant to assign hard numbers or they make the allowable numbers of bruised, injured, or lame animals or birds so high that the worst operations can pass. For example, the National Chicken Council in the USA set the limit for broken wings during catching and transport at 5% of the chickens. When the more progressive managers improved their catching practices, broken wings dropped to 1% or less. The standard should be set at 1% not 5%.

Similarly:

There have been serious problems with conducting audits when an input standard is vague. One example was a standard for pasture; it stated that animals must have access to pasture. There was no stipulation on how much time the cows had to be on pasture.

Grandin doesn’t say these kinds of quantitative standards should be the only standards used. She agrees some practices should be banned outright, and that subjective standard can be valuable if auditors have seen a wide variety of farms, so they know what better and worse farms look like. However, the idea of measuring welfare does seem to be an important idea. As the saying goes in the business world, “what you measure is what you get.”

Prior to reading Grandin’s book, I tended to think the answer to the question, “what tactics should animal advocates use to advance the cause?” is “a range of tactics.” The work of the Humane Society of the United States, for example, includes undercover investigations, efforts to tighten animal welfare laws, as well working with animal product producers to improve the treatment of animals used in food production. My impressions reading Grandin’s book tended to confirm this view.

She doesn’t go into great depth about activist organizations, but does mention them a couple of times:

Non-governmental (NGO) animal advocacy groups are also a major factor in developing animal welfare standards and legislation. When videos of animal abuses are seen around the world on the Internet, it makes people aware of the issue and they demand improvements.

She also says that, “Pressure from activist groups has also forced the upper management of many big companies to examine the substandard practices of their suppliers.” Here, “big companies” refers to companies such as McDonalds, which have such an important role in the market that suppliers need them more than they need any one individual supplier. This gives them enormous power to enforce changes if they deem it necessary.

One last tidbit that will interest Brian Tomasik, if he’s reading this: Grandin has a section discussing research on whether various types of animals can suffer. She seems fairly confident that fish can, given that they engage in quite complex behavior in response to aversive stimuli. She reports being able to find only one study on whether insects can suffer, but advocates for more research on that topic.

The issue that should make Bernie Sanders unacceptable to progressives

Imagine if a major presidential candidate got up and said that Wall Street supports racial equality because they want to use it as a weapon to suppress the wages of white workers. Imagine if he said that we shouldn’t be helping blacks compete with unemployed white kids. His career would be over, right?

Would it change anything if he claimed to care deeply about blacks? If he claimed to “just have a few concerns about how black workers are being used,” and wanted to protect them from exploitation? If he listed, as an example of his totally reasonable concerns about “how black workers are being used,” a complaint that some of the corporations that hire the largest numbers of African-Americans are also heavily involved in outsourcing and offshoring?

Would it help if he framed his solutions to this “problem” as actually helping blacks? If his platform included statements like “employers should be required to reimburse black workers for housing and transportation expenses” and “substantially increase prevailing wages that employers are required to pay blacks,” without saying that these provisions should apply to workers in general, and arguing that if employers really need to hire African-Americans they should be happy to follow these rules?

Guess what? Take the above hypothetical candidate, replace “blacks” with “immigrants,” and you have Bernie Sanders. ThinkProgress quotes Sanders as saying:

“There is a reason why Wall Street and all of corporate America likes immigration reform, and it is not, in my view, that they’re staying up nights worrying about undocumented workers in this country. What I think they are interested in is seeing a process by which we can bring low-wage labor of all levels into this country to depress wages for Americans, and I strongly disagree with that.”…

“I frankly do not believe that we should be bringing in significant numbers of unskilled to workers to compete with [unemployed] kids,” Sanders said. “I want to see these kids get jobs.”

And as much as I wish I were making up the offshoring thing…

“Last year, the top 10 employers of H-1B guest workers were all offshore outsourcing companies,” Sanders said in a Senate speech in 2013. “These firms are responsible for shipping large numbers of American information technology jobs to India and other countries.”

And here are some of the bullet points from Sanders’ official platform on his website:

• “Employers should be required to reimburse guest workers for housing, transportation expenses and workers’ compensation.”
• “Substantially increase prevailing wages that employers are required to pay temporary guest workers. If there is a true labor shortage, employers should be offering higher, not lower wages.”

To be clear, I’m not calling Sanders a racist. I’m calling him a xenophobe. In America today, bigotry based on skin color is taboo, but bigotry based on where someone was born is 100% mainstream.

Some people have asked me why we should make this issue, of all issues, a litmus test. In my view, it’s precisely because immigration seems marginal to many progressives, precisely because xenophobia is mainstream, that we should be fighting to create a world where the kinds of things Sanders has said have a political cost.

By the way, Sanders’ idea that immigration hurts American workers is as wrong as Donald Trump’s claim that Mexican immigrants are mostly criminals. Economists agree that immigration actually raises wages of native-born workers, and the debate is about how big this effect is.

Who am I rooting for this election? Martin O’Malley. His platform promises aggressive executive action to help immigrants, with bullet points like:

• “Provide Deferred Action to the Greatest Possible Number of New Americans.”
• “Limit Detention to Only Those Who Pose a Clear Threat to Public Safety.”

O’Malley promises to end the practice of prosecuting almost all undocumented immigrants for illegal entry and “direct federal prosecutors to focus on priority cases that advance national security, address violent crime or financial fraud, and protect the most vulnerable members of society.”

He has also attacked visa caps as “putting unrealistic and rigid quotas on who can contribute to our country”, and supports creating an independent agency to enable more flexible immigration policy, as well as moves to “address employment barriers for foreign professionals.” In the current political climate, these statements really stand out.

What about Hillary Clinton? Clinton appears to be positioning herself as Generic Democrat on immigration issues. I expect that if she were elected president, we’d continue to see deportation of child refugees, as is happening under Obama. On the other hand, she hasn’t promoted ideas about immigrants hurting American workers, nor has she proposed measures that would make it harder for immigrants to get jobs.

So I definitely want to see pressure on Clinton to take a clearer pro-immigration stance (something that will happen if O’Malley’s campaign manages to gather any momentum). But the things Sanders has said should just be totally unacceptable.