Month: November 2022

The American economy gained 261,000 jobs last month. We need 90,000 new jobs just to keep up with population growth, which means that net job growth clocked in at 171,000 jobs. The headline unemployment rate increased to 3.7%.

This is a respectable report, but the bad news is that about 300,000 people left employment and entered the unemployment roles. That's why the unemployment rate went up from 3.5% to 3.7%. Almost all of that came from people losing their jobs or finishing up temporary gigs. None of it came from voluntary leavers.

And now, as a special treat (!), here's a look at the employment diffusion index through October:

This is a measure of the percentage of industries that are expanding. It includes all industries that are hiring more workers + half the industries that are flat. The latest figure is 62%, which is good, but not as good as it was last year. As with so many other metrics, it peaked early in 2022 and has been dropping ever since. This is a sign that employment is not as robust as we think. More on that later.

In earnings news, blue-collar wages were up 4%. Adjusted for inflation this comes out to -0.1%.

In the LA Times today, Anita Chabria reports that Gov. Gavin Newsom is sick and tired of California's lack of progress on homelessness:

On Thursday, Newsom hopes to shake up both our dissatisfaction and the status quo by rejecting every local homeless action plan in the state, demanding more intensity when it comes to getting people off the streets and into homes.

....If need be, he said, he’ll be the “mayor of California,” in places that can’t or won’t do it themselves — a sentiment that is going to throw down “a bit of a gauntlet” and make a lot of people angry (including mayors and other politicians who are “actually friends, like some of our friends that come over for the holidays”).

....What’s at stake is a delay in the next big round of funding that cities, counties and continuums of care are expecting — roughly at least $600 million set to be divided among 75 governments and service providers that applied for the grant money, either alone or in groups.

Newsom gestures here in the right direction—local control is a horror show—but then goes off course by using money as his primary leverage.

Money is always a problem, especially in big cities where building shelters is inherently expensive. But it's not the main problem. The biggest problem is that everyone wants shelters, but only if they're built someplace else. Unfortunately, "someplace else" doesn't want them either. And most communities will tie things up in court endlessly if you try to force shelters on them.

The only real answer is for Newsom to persuade the legislature to pass some kind of safe harbor law. This would override CEQA, our notorious lawsuit machine, and provide developers with a smallish set of essential conditions that would guarantee they could build with only minimal contact with the judiciary. At that point, local communities that want to build shelters would be able to do it, and with additional legislation Newsom could force recalcitrant communities to do it too.

Would the California legislature ever pass a law like this? The magic eight ball says "unlikely." But what else would work? California has 160,000 unhoused people, of which about 100,000 are unsheltered. It's just not possible to build that many beds without a very big stick, and that means money + the legal means to keep development out of constant legal purgatory.

This is from an op-ed piece at Fox News a couple of weeks ago:

Recent crime numbers support the notion that the city is becoming less hospitable. New York crime, at this writing, is up 33% overall. But significantly, there’s been a 36% jump in robberies and a 16% jump in felony assaults.

These are the kinds of numbers I've been hearing, and I finally got curious enough to go take a look for myself. Unfortunately, all this did was confuse me some more. Here's a CompStat chart that compares 2022 to 2021. Each dot represents the number of incidents during the month ending on each date. For example, the last pair of dots represents the month of October in 2022 and 2021:

So the numbers in the op-ed are approximately right, though they've dropped a bit since the day it was written. But the weird thing is that crime hasn't really gone up in 2022. From October of 2021 to October of 2022, crime is up only 6%.

Then why the big numbers? Answer: the NYC crime rate dipped peculiarly from November of 2020 through February of 2021 before returning to its usual rate in September. During that time the number of major crime incidents was much higher in 2022 than 2021.

But if you look at the first date on the chart (October 2021 compared to October 2020), crime starts out 14% higher. On the last date, it's up only 6%. Basically, everything has caught up, and the big yearly increase is solely because of the big dip in the middle of 2021.

This doesn't mean New York has solved its crime problem. Major crimes are up 21% from October of 2020 to October of 2022. On the other hand, crime has been dropping since August and is slowly getting back to its old level.

Do you see why I'm confused? There are a whole lot of numbers you can make a case for, and I'm not really sure which one best describes facts on the ground in New York.

Things are simpler in Los Angeles. CompStat reports that violent crime is up only 3.6% from 2021 to 2022, so there's no big problem to begin with.

Needless to say, it would be better if crime were down instead of up. Still, it doesn't really look to me like there's a huge crime wave in New York right now, and there's certainly no big crime wave in Los Angeles. But the numbers are confusing. I might change my mind if I figure out an even different way of looking at them.

Remember that patch of Dame's Rocket blowing in the wind that I showed you a couple of months ago? Here it is again, but in a normal photograph that shows what it actually looks like. It's a pretty little flower when you get close up.

A couple of days ago I posted this chart showing that corporations were paying their workers less but marking up their products far, far beyond the rate of inflation:

But Matt Yglesias says there's nothing either wrong or unusual about this:

Excess demand and profit-seeking by greedy corporations are not competing explanations for inflation — what happens when demand soars above productive capacity is that prices rise and profits go up. https://t.co/xckpVY14vl

— Matthew Yglesias (@mattyglesias) November 2, 2022

This is a good point, but the problem is that it's very hard to prove. What we'd like to have is a good measure of demand, but we don't have one. None of us can peer into the minds of men and figure out how much they want to buy. Nor can we know at any given time how much future pain they're willing to endure in order to buy stuff now.

This means we have to look at metrics that are—we think—symptoms of demand rising above supply. In the case of food and oil, it's fairly easy to see that shortages—natural ones in the case of food, manufactured ones in the case of oil—have pushed supply below demand and that consumers are mostly willing to pay more until supply and demand are once again in equilibrium.

But for other things it's not so easy. Is there really a shortage of Pepsi, for example? Is it true that PepsiCo can't easily increase production? I have my doubts, but I can't prove anything.

More generally, though, we can look at various metrics and draw some conclusions. For example, the Fed says that supply chain pressures have mostly gone away:

Other metrics tell us that earnings are down, savings are being used up, and the money supply is dropping. Put all this together and it suggests that supply has recovered to normal and demand is probably back to its pre-pandemic normal as well.

The other thing to consider is the huge size of the rise in markups. Markups generally go up after recessions, but markups today are higher than they've ever been since we started keeping records. Here's a different calculation of markups over the long term from the Roosevelt Institute:

The Roosevelt figures suggest that the rise in markups is less than the BEA numbers show, but the increase is still steep and the absolute level is higher than it's ever been. This suggests, once again, that companies are raising prices more than usual and more than they're being forced to by inflation.

Bottom line: Matt's point is a good one, but I'm not sure the evidence suggests that supply is in bad shape anymore or that demand is really a lot higher either. Company greed might be less than I thought it was, but it's still responsible for a fair chunk of our current inflation.

I have a big pile of charts to show you. Don't worry, though, they'll go pretty quickly.

The first is the M2 money supply. It spiked upward after the $3.2 trillion COVID rescue bill was passed in March 2020, and then continued growing. But in January it peaked and has been falling ever since:

Next is both the number of job openings and the actual number of new hires. These metrics were rising until early this year. Then they peaked and they've been declining ever since:

You are probably sensing a trend. Next up is a related measure, the growth of the employment level. It too was growing until January, at which point it started to drop at a fairly dramatic rate:

Next let's look at hourly earnings. They've been dropping ever since early 2021, but in January of this year they started to really drop:

Next is personal savings. Savings spiked after each stimulus bill, the last of which was passed in March of 2021. Savings were back down to normal by May and have been declining ever since:

Next is credit card balances. In January 2021 they suddenly started to go up, a signal that people are making ends meet only by putting purchases on plastic as wages begin to dry up:

Finally we have average rents, which rose considerably in the first half of 2021. However, the growth rate of rents has been dropping ever since August 2021, and last month the growth rate was negative:

All of these things are either inflation precursors (M2 money, employment) or actual elements of inflation (wages, rent). All of them have been falling since the start of the year, and some of have been falling even longer.

None of them affect the inflation rate immediately. They all have lags. The M2 money supply takes about a year to affect inflation, while others have an impact in less time. This is why inflation peaked in January and is already dropping:

By the end of the year pretty much everything related to the inflation rate will finally be having an impact. But all of them started long before the Fed began raising interest rates.

This pisses me off. Inflation will be under control by the start of 2023 and will continue to drop throughout the year. This will happen after four or five big interest rate jumps, and people will therefore hail the Fed for conquering inflation. Like this from NPR today:

The Federal Reserve ordered another big boost in interest rates on Wednesday....The rate, which was near zero in March, has jumped 3.75 percentage points in the last eight months. That's the most aggressive string of rate hikes in decades, but so far it's done little to curb inflation.

"Interest rates have risen at a whiplash-inducing speed, and we're not done yet," said Greg McBride, chief financial analyst at Bankrate. "It's going to take some time for inflation to come down from these lofty levels, even once we do start to see some improvement."

The implication here is that we need more and more rate increases, but it's not true. Like everything else, increases in the fed funds rate have a lag—probably of at least six months and maybe as much as a year. It's true that the Fed's recent increases haven't had an effect, but that's not because the Fed hasn't done enough of them. It's because they simply haven't had time to have any effect.

In other words, whatever happens in January of 2023 will have nothing to do with the Fed and everything to do with other things that started dropping in 2021 and early 2022.

By the middle of 2023 I suspect we'll be firmly into a recession thanks to those interest rate increases. And that recession will come at the worst possible time, just as inflation is reined in and the economy has cooled of its own accord. We won't want that recession by then, but we'll get it anyway thanks to Jerome Powell.

Twelve years ago I wrote a screed about how banks charge high swipe fees on credit cards and then rebate a part of those fees to their richest customers:

This is fundamentally my problem with overdraft and interchange fees: they’re surreptitious ways for the poor to subsidize the rich. There’s no law against that, of course, but the practice is so grotesque that in this case I’m perfectly willing to make one.

....[One solution is to] get rid of reward cards, which are surely one of the most ridiculous and unjustifiable frauds ever invented. Seriously: banks deliberately overcharge their customers and then rebate a fraction of it in the most circuitous and confusing way possible? And to make it worse, they do it in a way deliberately designed to transfer wealth from the poor to the rich? Karl Marx probably wouldn’t have been cynical enough to predict that banks would ever operate like this.

This is pretty obvious stuff that hardly needs an academic study to "prove" it, but recently we got one anyway:

We assess interchange income at the card level to be 1.5 percent of the purchase volume for classic cards and 2.5 percent for reward cards.

....We find that high-FICO consumers benefit from reward programs at the expense of low-FICO consumers and estimate an annual redistribution of of $15.1 billion....We conclude by documenting that the costs and benefits of credit card rewards are unequally distributed across geographies in the United States. Credit card rewards transfer income from less to more educated, from poorer to richer, and from high- to low-minority areas, thereby widening existing spatial disparities

So there you have it. Banks charge higher swipe fees for reward cards—mostly paid by merchants—and then rebate that money in the form of rewards mostly paid to customers who are middle class and above. The net result is that richer, more sophisticated cardholders end up paying less than poorer, less sophisticated cardholders. And that's before we even get to the higher interest rates they're charged on unpaid balances, which is at least notionally justifiable.

But reward cards don't just redistribute billions of dollars from poor to rich. They also redistribute it from high school to college grads and from Black/Brown customers to white ones. Not just unfair, but racist too.

NOTE: This paper is fairly complicated, and not easy to excerpt. If you have questions about methodology and so forth, I'm afraid you'll have to click the link and dive in yourself.

I hate to pick on my old friends at Mother Jones, but Abigail Weinberg recently posted a story about right-turn-on-red (RTOR), a longtime staple of California roads that expanded to most of the country 50 years ago. It's received wisdom on the left these days that roads are for bicycles, not cars, so naturally the conclusion of the piece is right in the headline:

It’s Time to Ban “Right Turn on Red”

It’s an obsolete relic of the 1970s oil crisis. It’s dangerous to pedestrians. And, if you drive a car in the United States, you likely do it every day. It’s time to get rid of right-turn-on-red.

....The data on right-turn-on-red crashes might be scarce, but the existing studies suggest that these types of collisions—while rare—frequently involve a pedestrian or cyclist. Cars, instead of hitting other cars, often hit humans. Now, there’s a growing movement for cities to do away with the traffic law altogether.

Last week, the Washington, DC, city council voted to ban right-turn-on-red (RTOR) at most city intersections....Critics of the DC bill have pointed out the lack of data showing the dangers of RTOR, but many people who don’t use cars know instinctively how dangerous turning vehicles can be. “Our current safety studies fail to capture the reality of the constant near misses and confrontations that result between these motorists and pedestrians which can be observed daily just by observing a typical busy intersection with RTOR,” [Bill] Schultheiss says.

This is a single-source story featuring a bicycle expert named Bill Schultheiss who very obviously has a point of view to press. Because of that, he apparently didn't mention that although data on RTOR is thin, it's not nonexistent. Here's what a ten-minute Google search turned up:

• 1995: National, NHTSA, 1982-92. There were a total of 0-84 RTOR fatalities, probably toward the lower end of that range. About half involved pedestrians (44%) or bicyclists (10%). Roughly, then, there are probably 2-3 RTOR fatalities per year involving pedestrians and bicyclists.
• 2002: San Francisco, Fleck & Yee, 1956, 1994-96. This is a review of two existing studies. RTOR collisions are a tiny fraction (<1%) of all right-turn collisions. The accident rate is lower than right-turn-on-green. In addition, two other papers reviewing different states show no detectable change in right-turn injury rates before and after RTOR was adopted.
• 2008: Connecticut, Office of Legislative Research, 1994-2006. Out of 663 total fatalities, four were at RTOR intersections, all involving pedestrians or bicyclists. That's one fatality every three years. Of those, one involved a drunk bicyclist, one involved a wheelchair user who crossed against the signal; and one involved a bicyclist who failed to yield when he should have.
• 2009: New York City, NYDOT, 2006-08. This is a study of injuries before and after RTOR was allowed at specific intersections. Conclusion: "Accident rates not affected." About a third of injuries involved pedestrians (25%) or bicyclists (4%).

As a born-and-bred Californian, I naturally think of RTOR as God's will. Still, I don't have any special axe to grind here. I just think those of us who write about public policy—even if it's advocacy journalism—have an obligation to perform at least a minimal bit of research before we come out with guns blazing. It may be that there's some concrete evidence out there showing that RTOR really should be banned, but the research I could find shows just the opposite. RTOR doesn't seem to make more than a sliver of difference one way or the other.

This is from the Rodin Museum. It's the guy who inspired John Travolta's dance moves in Pulp Fiction.

Just for the record, I don't care about either Elon Musk or Twitter. Both fall into the category of "obviously flawed but not as bad as the haters claim." Nor do I care who has a blue checkmark or how much it costs. I also don't care what kind of moderation regime Musk sets up, since he'll soon learn that his options are limited. What's more, even in the extreme case that Twitter fails completely or becomes unusable, it will have only a modest effect on both me and the wider world.

Now, can we all stop obsessing over it? Let's just let Musk do his thing and then we'll see.