Kevin Drum

I don't really understand how you can conduct a poll in a place like Gaza, but the Palestinian Center for Policy and Survey Research says they can do it. Here's one result from their latest survey:

After nearly a year of the most brutal pummeling imaginable, only 28% of Gazans still think Hamas can win the war. But still: 28%!

In other results, Hamas has lost a bit of support but is still the most popular single faction in Gaza. And Gazans really, really don't like the United States. I wouldn't be surprised if we poll lower than Israel.

On a side note, there's also this:

This is something I've been following for a while, and this survey seems to confirm that food shipments are improving—at least in southern Gaza. PSR doesn't do fieldwork in northern Gaza, so we have nothing on that one way or the other.

Here's the growth of mobile cellular subscriptions in a few selected countries:

Note that this is per 100. Most rich countries have more cell subscriptions than they do people.

The top region is Hong Kong at 292 (China as a whole clocks in at 125). The United States is a hair above the world average at 110. The lowest number is North Korea at 23, edging out South Sudan's 30.

Regions of the world look like this:

• East Asia: 130
• Middle East and North Africa: 124
• Europe: 123
• Latin America: 109
• North America: 108
• World: 108
• Sub-Saharan Africa: 89
• Pacific islands: 76

Here is the number of murders each year in Springfield, Ohio, both before and after the large-scale arrival of Haitian immigrants:

Daniel Driscoll, the Republican top prosecutor in Clark County, adds this:

During the time that I’ve been with the prosecutor’s office, which is 21 years now, we have not had any murders involving the Haitian community — as either the victims or as the perpetrators of those murders.

Springfield has seen a recent increase in the rate of aggravated assault, but it's a statistical artifact caused by the Springfield police department changing its definition of the term.

Here's a funny thing. The Wall Street Journal, searching for a trend based on Kamala Harris's debate revelation that she owns a gun, says that Democrats are suddenly buying lots of guns. Their data for this comes from NORC at the University of Chicago, but so does mine and it's not the same:

According to the NORC GSS data I downloaded ten minutes ago, gun ownership among Democrats went up a grand total of 1.3 percentage points last year and 2.7 percentage points over the past decade. It's currently at 25.3%, but the Journal puts it at 29.2%.

This is a head scratcher. I've tried all sorts of ways of manipulating the NORC data to reproduce the Journal's figures, but I can't do it. And I'm not sure what the point is anyway, other than trying to find something that supports a thesis. The basic data is what it is, and it suggests only the tiniest uptick in Democratic gun ownership over the past decade.

Here's an odd thing. With my old camera and old flash unit I could never take a picture of Hilbert. Charlie was fine, but Hilbert's eyes always came out pitch black. This happened even though I always use bounce flash, not direct flash.

But with my new camera and new flash everything is fine. As you can see, Hilbert's eyes are lovely and bright.

I can't explain this. It's the same brand of flash. It sits on the hot shoe, just like the other one. It's pointed directly upward, as always. I wonder what the difference is?

Here's the blaring headline at the top of the Washington Post this morning:

In a hyper-technical sense this is true. The new Georgia rule doesn't require a hand count of how people voted, but it does require a hand count of the number of ballots to make sure it matches the machine total. The Post doesn't tell us that until the ninth paragraph.

It's still probably a dumb idea, and maybe even illegal too—according to Georgia's Republican attorney general, anyway. There's no reason to doubt machine tallies, and the whole thing is most likely designed to generate discrepancies in case Trump loses and needs something to yell about.

Still, the headline is wildly misleading. Readers deserve better.

Alex Tabarrok points today to an interesting RAND study of prescription drug prices. It turns out that we pay much higher prices than other countries for new, branded drugs, but other countries pay much higher prices for older generics:

But I found this even more interesting:

U.S. prices for brand-name originator drugs were 422 percent of prices in comparison countries, while U.S. unbranded generics, which we found account for 90 percent of U.S. prescription volume, were on average cheaper at 67 percent of prices in comparison countries, where on average only 41 percent of prescription volume is for unbranded generics.

Why do other countries use so few generics? Obviously they have less price pressure to do so since brand-name drugs are relatively cheap compared to the US. But is there more to it? National health care systems could nevertheless reduce drug spending if they pushed generics harder, and it would be low hanging fruit for health systems that are always under budgetary pressure. So why not? Are Europeans just more suspicious of generics than Americans?

I don't remember why I got curious about this, but earlier this evening I looked up the word colorway to find out where it came from. Why did the folks in the sneaker industry invent this odd word?

They didn't. According to the OED, the earliest use is 1941 in the Guardian. Another source places its origin in the textile industry in the late 19th century. And the Google Ngram viewer has instances starting in 1844. Here's the first century of colorway (as always, click to embiggen):

There a sudden outburst in 1844 in the US, followed by very rare and sporadic usage through 1940—also mostly in the US. Here's the second century:

Common usage doesn't start until 1952, followed by a steady rise and then a huge spike in the US in 2003—presumably the result of its adoption by the sneaker industry.

Anyway, colorway just means a particular combination of two or more colors, but I can't find an explanation of where it came from. The best guess seems to be that it's a shortening of "a way of using color," but I'm not sure I find that very convincing. But I don't have any better ideas.

The New York Times celebrates the stock market's record high today with a standard chart showing the S&P 500 since the year 2000. Nothing wrong with that.

But if you really want to know how the market is doing, you should do two things. First, adjust for inflation. Who cares if the S&P is up 5% if inflation is also running at 5%?

Second, you should look at total returns, which includes reinvested dividends. Here it is:

Since 2000, the S&P 500 is up 120% after adjusting for inflation. Not bad. But if you put actual money into a fund that reinvested dividends, it would have returned 250% by now. Even better!

Note, however, that even with reinvested dividends the total return from 2000 to 2013 was 0%. Equities are a good bet in the long run, but sometimes the long run is pretty long.

According to the American Historical Association, states have been busy in recent years adding new requirements to the social studies curriculum:

Since 2017 there's been an uptick in the number of states requiring the study of specific groups (Black, LGBT, etc). Since 2012 there's been a steady increase in civics and US government requirements. In the entire period since 2000, diversity wins have outscored civics wins 155-110.