You probably need something to cheer you up after the tax follies of this week. So how about a scholarly examination of how many Americans were killed by atomic bomb testing in the 50s? I think that should do the trick.

Here’s the background. After 1949, the US moved most of its atomic bomb testing from the South Pacific to a test facility in Nevada. These were all above-ground tests that generated quite a bit of radioactive fallout, including an especially dangerous isotope of Iodine called Iodine-131. The map below shows the deposits of Iodine-131 from a typical series of tests done in 1953:

As you’d expect, the highest concentrations are immediately downwind of the Nevada Test Site, with a couple of odd hot zones in New Jersey and upstate New York. Luckily, these are mostly areas of sparse population. Unluckily, this doesn’t matter much because it’s not the primary way that Iodine-131 kills people. Most of it ends up being carried by high-altitude winds and then deposited by rainfalls throughout the country. From there it gets into pastureland and then into the milk supply, where it attacks the thyroid. In a new working paper (i.e., still a bit preliminary) Keith Meyers made use of extensive datasets on milk consumption and local death rates and produced a map that shows which areas were most heavily affected:

It turns out that the victims were mostly quite far away from Nevada. A combination of extensive dairy farming and high populations meant that most fatalities were in the upper Midwest extending all the way over to the Eastern seaboard. A rough calculation suggests that the total death toll from testing during the 50s clocked in at about 400,000, far higher than most previous estimates. Children were disproportionately affected because they drink more milk and have smaller thyroids.

But there is, surprisingly, some good news here too. Starting in 1958, and then made permanent by the Partial Nuclear Test Ban Treaty, atmospheric testing was halted. Meyers figures that the testing which was moved underground between 1958-1992 probably saved millions of lives. So it could have been a lot worse.

UPDATE: The original post said that underground testing saved 12-24 million lives, which is indeed the rough estimate in the paper. However, that comes with a number of caveats, and the actual number is most likely smaller. I’ve changed the text to reflect this.

The US Senate has released a list of the workplace harassment claims that it’s settled since 1997. There are a total of 13 settlements totaling $600,000, and since everything is better in chart form, here’s what the settlements were for:

That looks like more than 13, doesn’t it? And more than $600,000, too. That’s because some of the settlements are for multiple issues. Violations of the Family and Medical Leave Act, for example, are almost always paired up with something else.

Anyway, it looks like age discrimination is the biggest issue in the Senate, which is kind of ironic considering that the average age of senators is, like, 103 or something. But I guess they prefer offices full of young whippersnappers who will work 20 hours a day and demonstrate endless fealty because their vaulting ambitions haven’t yet been snuffed out.

This list, by the way, is limited to settlements by the offices of actual senators. There’s a separate list for other Senate offices, and that one includes a $421,000 settlement for “race discrimination and reprisal.” That’s by far the largest settlement on either list, but if you think of it instead as a 28 bitcoin settlement, it doesn’t seem so bad, does it?

Christmas in California #4. Location: Davenport Island, Huntington Harbour.

This is true, but only for the federal exchange. Many of the largest state exchanges are open for several more weeks.

This just came in from CMS:

Exchange open enrollment for 2018 coverage ended w/ approx 8.8M people enrolling in coverage. Great job to the @CMSGov team for the work you did to make this the smoothest experience for consumers to date. We take pride in providing great customer service.

— CMS Administrator (@SeemaCMS) December 21, 2017

Presumably this number is solely for Obamacare signups on the federal exchange, and it’s only slightly lower than last year’s 9.2 million. That’s not bad.

But wait. We already have 2.8 million signups via the state exchanges. That’s 11.6 million total, and many of the state exchanges are still open. California, for example, doesn’t close enrollment until the end of January. So signups on the state exchanges will surely finish up at—what? 3.5 million? If that’s the case, Obamacare would end up with total enrollment (federal + state) of 12.3 million, slightly higher than last year.

But wait again. Even on the federal exchange, enrollment is still open in states affected by hurricanes. It’s not impossible that Obamacare enrollment by the end of January could be as high as 13 million. We’ll see.

Maybe I’ve done some arithmetic wrong here. But it sure looks like Donald Trump’s efforts to sabotage Obamacare only made people more aware that it was out there and they should sign up for it.

UPDATE: Yes, I did some arithmetic wrong. 8.8 + 2.8 = 11.6, not 12. I’ve revised the text.

Greg Lovett/The Palm Beach Post via ZUMA

From a Palm Beach Post story by Christine Stapleton and Lawrence Mower about Donald Trump’s purchase of Mar-a-Lago in the early 90s:

Trump’s promise couldn’t be in writing, Trump attorney Rampell told the council, according to meeting minutes and transcripts. If the council insisted Trump’s commitment be in writing, his donation might be disqualified by the IRS as a charitable contribution.

The backstory here is that the city of Palm Beach wasn’t happy with Trump’s plans for preserving the Mar-a-Lago estate. So Trump’s lawyers proposed a deal: if the city council approved Trump’s plan to create a private club, Trump would turn over much of the interior of the mansion to the National Trust for Historic Preservation. But the deal had to be under the table. Trump wanted a tax deduction for his charitable contribution to the NTHP, and it’s not charitable if it’s done in return for something, is it?

So, wink wink nudge nudge, and the deal went through. But there was never any paper trail, so the IRS approved the charitable deduction. Life is good when you’re a millionaire.

Why are prime-age men dropping out of the labor force? And what can we do about it? A few days ago the Two Jasons tackled this question and came up with very different answers. Here is Jason Richwine criticizing Jason Furman:

How about restricting low-skill immigration to encourage recruitment of Americans? No, Furman says, because — well, actually, he does not mention immigration at all, not even to dismiss its importance. Omitting the i-word in discussions of labor-force dropout is an unfortunate habit on both the left and the right. Amy Wax and I wrote our Inquirer op-ed (based on a much longer essay in American Affairs) to show that employers turned to immigrants as the native work ethic declined. As evidence, we point both to the much higher labor-force participation of low-skill immigrants compared to low-skill natives, as well as to the near-universal preference expressed by employers for immigrant labor. Restricting the flow of foreign workers would generate a major incentive for business owners, politicians, and opinion leaders to reintegrate American men into the labor force. It is, in our opinion, a crucial part of any reform strategy.

I don’t have a huge dog in this fight since, by lefty standards anyway, I’m pretty open to policies that humanely restrict illegal immigration. But I have a couple of questions for Jason R that I’m genuinely interested in. First, there’s this:

Labor force participation of women rose steadily starting with the feminist revolution of the 60s. However, participation peaked in the late 90s and has been declining ever since. In other words, for the past couple of decades this is a problem that’s not restricted just to men. Is illegal immigration affecting the labor force participation of women the same way?

Second, if illegal immigration really is a factor, then labor force participation of native workers should be correlated with the amount of illegal immigration. But it doesn’t seem to be:

I don’t want to make too much of this, since the Great Recession obviously had a big impact on both illegal immigration rates and labor force participation. Still, overall labor participation turned down in 2000 and then turned down even more in 2009. But illegal immigration rates didn’t change in 2000, and then changed in the wrong direction in 2009. Lower immigration rates should have increased labor force participation. What’s more, labor force participation among men has been dropping steadily since the 60s, with the rate hardly changing at all even as illegal immigration has spiked up and down.

There are so many other, bigger factors that affect labor force participation that I don’t really expect much from this comparison. But I’d expect at least a glimmer of something if unauthorized workers from Mexico really are causing native workers to just give up on the job market. Why don’t we see anything?

Alex Edelman/CNP via ZUMA

John Judis doesn’t think the Republican tax bill will hurt them in the 2018 midterms:

I don’t buy the argument — voiced by Democratic pundits, political consultants, and even a few economists — that the bill will doom the Republicans to defeat in 2018 and even 2020. Like many things I read or hear these days from liberals, it’s wish fulfillment disguised as analysis.

Democrats argue that the bill will be unpopular because it increases inequality by giving huge tax breaks to the rich and corporations. But most American voters don’t object to inequality and to the rich per se….The tax bill does give immediate benefits to the middle and lower classes. These include the increase in the child credit and standard deduction and lower rates….It is likely in the short run, that is, during Donald Trump’s presidency, to prolong the recovery and hold off an eventual downturn….Finally, I hear liberals and Democrats pointing to polls showing the tax bill is unpopular. I distrust these polls….Whether a policy is popular or not is usually settled during campaigns when the candidates try to interpret its results.

I halfway agree with Judis. The middle class does benefit in the short run from the tax bill, and George Bush’s tax cuts demonstrated pretty clearly that most people are satisfied with a modest tax cut even if they know it’s mainly a sop to disguise a huge tax cut given to the rich. And to Judis’s arguments I’d add another: Corporations employ an army of tax attorneys, and by November 2018 they’ll know exactly what the tax bill does for them. Ordinary folks don’t, and they won’t. It’s not until they start doing their taxes a few months later that they’ll really figure it out.

However, the sentence I highlighted is key, and it won’t necessarily work in Republicans’ favor. A popular president would be able to make a good case for the tax bill, but Trump isn’t popular and isn’t likely to be a year from now. It’s a lot easier for Democrats to paint the tax bill as a giveaway to corporations and the rich if the electorate already dislikes Trump and is primed to believe this. Popularity casts a golden sheen over everything a president does, but the opposite is true too. The tax bill will suffer merely for its proximity to Trump.

All this depends, of course, on Democrats making a good case against the bill. That’s not a slam dunk given their recent history, but the opportunity is certainly there.

This has nothing to do with politics of any kind at the moment, but Tyler Cowen pointed me to a new paper that’s pretty interesting. The authors have compiled a massive new database that shows the returns on treasury bills, treasury bonds, stocks, and housing for 16 different countries over the past 150 years. The countries covered include the United States, Japan, and 14 major European economies, and that allows the paper to show us average returns for advanced economies worldwide, rather than just a single country with its idiosyncratic features. Here, for example is the average real rate of return for governments bills and bonds:¹

These are both safe assets, but their returns are surprisingly volatile. In wartime, returns plummet, but even in peacetime they range from -3 percent to 8 percent. And another thing: until 1990, returns on bills and bonds were pretty close, usually within about 1 percentage point of each other. But since 1990, they’ve diverged considerably. For the past 30 years, long-term bonds have consistently returned about 4 percentage points more than short-term bills. It’s not clear why this is, although presumably it means that investors have had expectations of declining inflation for quite a long time. Here’s a chart that combines all safe assets to provide a clearer view of the volatility of returns:

Even not counting wartime, the return on safe assets has roller coastered between -2 percent and 7 percent. As you can see, there’s no apparent relationship with economic growth, and the authors point out that the average return on safe assets over the past 150 years is roughly 2 percent, about what it is today. This suggests that current returns on government bonds are nothing out of the ordinary.

But what about risky investments? Here are returns on housing and on stocks:

With the exception of World War I, housing returns have been remarkably stable. With only a couple of exceptions, the return has been between 5 percent and 8 percent year in and year out.² Equities do about as well on average, but are far more volatile, ranging from 0 percent to 15 percent depending on the strength of the economy. If we then take a look at both risky and safe assets together, we can see the evolution of the risk premium over time. Here it is:

With the exception of the huge change during the Depression and World War II, the trend is fairly clear. From 1870 to 1970, the risk premium gradually rose from about 4 percent to 7 percent. Then there was a big drop, and ever since the mid-80s the risk premium has hovered around 2-4 percent. For the better part of a century, investors required a premium of 4-7 percent to entice them to invest in equities instead of treasury bonds. But following the collapse of the Bretton Woods regime in the 70s and the global financial deregulation of the 80s, investors have required a premium of only 2-4 percent. This is, presumably, because the neoliberal revolution prompted investors to have greater confidence in steady economic growth and therefore in steady corporate growth.

I don’t have any big conclusions to offer about all this. Mainly, it’s just interesting raw data, something to provide context when we discuss things like flight to safety or the equity premium puzzle. As this data shows, real returns on both safe and risky assets bounce around a lot, so it’s usually best not to overreact to changes that, in historical context, might not be as dramatic we think.

¹The data throughout these charts is for real rates of returns. That is, everything has been adjusted for changes in the inflation rate.

²All of these charts show 10-year averages, which is why the great housing boom of the early aughts looks less impressive than you’d think. The stratospheric period of the housing boom lasted only a few years, so the point at 2001 (an average of 1996-2006), which represents its peak, includes only about three years of enormous returns averaged with seven years of merely high returns.

Bill Clark/Congressional Quarterly/Newscom via ZUMA

Atrios says:

Gotta Give People Something to Vote For

The issue has never been annoying bernie bros who piss you off on the internet. It’s the vast number of people who don’t bother voting because why should they…. “Trump sucks” might win 2018, but if Dems don’t deliver then “Dems suck” will win in 2020.

This is true, but it’s also a problem. Donald Trump will be president until 2020 no matter how well Democrats do in next year’s midterm elections. Given that, there’s really no way for them to deliver anything. And even if they do, Trump will probably get the credit for it. Presidents always do.

Still, if there is anything that Dems could credibly promise, my best guess is some version of universal health care. The basic pitch would be that they’ve learned their lesson: incremental change doesn’t work because Republicans will sabotage it the first chance they get. If we want real health care for everyone, with no nonsense about out-of-network swindles or narrow networks or skyrocketing costs, then we need something like Medicare for All. Maybe that’s what Dems should propose. Maybe something else. Or maybe an improved version of Medicare for All (probably my choice)¹ that phases in over time. And liberals being liberals, it should be accompanied by a hundred-page white paper explaining precisely how it would work and precisely how it would be paid for.

I dunno. Do you think Democrats could get their entire caucus to buy into this? After the events of the past year, you’d sure think so. But then, we’re talking about Democrats here. You just never know.

¹Or maybe an improved version of Medicaid for All.

Matt Yglesias writes today that, to the extent tax cuts have any effect at all, it’s only if people think they’re permanent. But with political polarization so strong, there’s no way that anyone can believe a tax change passed on a party-line vote is likely to last:

These days, nobody likes preachy blue-ribbon commissions, and bipartisanship is a total nonstarter in Congress. But if you actually want to generate supply-side tax effects, you need to create an expectation of permanence. And bipartisanship seems like one of the only plausible ways to accomplish that.

This gets at my biggest gripe about the Republican tax bill. If they had truly wanted to pass corporate tax reform—of the kind passed in 1986—they could have gotten plenty of Democrats on board. Then there would have been no need to worry about all the weird reconciliation rules and no need to worry that it might get reversed a few minutes after Democrats are next in power. That’s because actual reform really is pretty bipartisan, and there are plenty of Democrats willing to look at ways to simplify the corporate code, close special loopholes and subsidies, and then lower rates.

But that was never what Republicans wanted to do. They like to call their plan a “sweeping” tax reform, but it’s not. It’s just a hodgepodge of changes designed to reduce corporate taxes—which are already among the lowest in the world—and act as a vehicle for a few other Republican hobbyhorses, like reducing the estate tax and killing the Obamacare mandate. It’s a nice payoff for corporations and the rich, but that’s it. There’s barely a paragraph of genuine reform in the whole thing.