We have an economy with a positive first derivative and negative second derivative—everything is continuing to improve but it improving at a slower pace than before.
Normally 661,000 jobs would be something to celebrate. But when you’re 11 million jobs short of where you were in February the slowing pace of recovery is a worry.
Three reasons for it:
1. Easy recovery already happened. Has been people being called back from temporary layoff, permanent unemployment rising.
2. CARES Act expired.
3. Virus resurgence.
Notably in September there were 661,000 jobs added (payroll survey) while 1.5m reduction in temporary layoff (household survey). That is worrying because the fuel of labor market recovery is going away.
Also notable, the labor force participation rate has not moved since July. Normally we would expect a strengthening economy to have an increase in participation rates. Moreover, if the $600 was having a large disincentive effect that should have raised participation.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Shelter inflation was moderate and the three month moving average continues to basically trend down, albeit slowly.
But you can't just assume elevated items like shelter will get better but that everything opposite won't get worse.
And that's what we've (predictably and predicted) seen: goods inflation was negative for a while but turned positive for 4 straight months. Offset shelter cooling.
Tariffs & exchange rates. A short explainer of the simple case of 10% across-the-board tariffs. Let's start with no retaliation.
Brief version: Tariffs will strengthen the U.S. dollar which will reduce impact on consumers but exacerbate it for exporters.
Three cases:
1. No exchange rate effect. In this case imports are 10% more expensive for consumers. Exports are the same (because the exchange rate did not change) and the trade deficit shrinks. The entire tariff is paid by Americans.
2. USD appreciates by X% where 0 < X < 10%. Imports are 10% - X more expensive & consumers cut back on imports. But the xr appreciation also makes it more expensive for foreigners to buy exports so exports fall. Trade deficit effect is ambiguous & foreigners pay part of tariffs.
Labor market tightness has stabilized over the last several months after a loosening steadily through the summer. Job openings were up and quits down. My favorite metric, job openings per unemployed, was stable.
Here are openings and quits. They've been telling a somewhat contradictory story in recent months as openings are up and quits are down.
The economy remains on the Beveridge curve--admittedly the tight part of it.
Broadly speaking what has happened is core services inflation as only slowed a little (less than people were hoping on lagged shelter) while goods prices have started rising--with unusually large auto price increases in November that could still be hurricane-related.
I believe it is useful to make small contributions to big things (many engaged in doing that now) & also bigger contributions to small things.
On the later, in @BostonGlobe I argue for zoning reform to enable Cambridge to help build more than 1,000 additional housing units.
A🧵
States and localities can resist the likely regressive thrust of federal policymaking while doing what they can to build a more progressive, inclusive and upwardly mobile society.
To do that we need cheaper housing.
And to do that we need more housing.
VP Harris was right to set a goal of building 3 million housing units. On a proportional basis that would require 1,050 from Cambridge. Unfortunately on current course we'll get 100. But with reforms proposed by the City Council that could be raised to more than 1,000.
I know many skeptics of prediction markets. I don't have an ideological faith in them (OK, maybe quasi ideological). But the empirical evidence is they have worked really, really, really well. And did again on Tuesday night.
A short 🧵 about this remarkable picture.
Markets gave Trump a 60% chance. How does that prove they know what they're doing? If Harris won could say, "but she had a 40% chance" so wasn't wrong.
That's correct. Can only judge when you've seen them many, many times. Do 60% chance things happen 60% of the time?
In Ec10 we should them 15 million data points from sports betting from @andrewlilley_au comparing the prediction market probability to the outcomes.
And guess what: if you collect 100 markets with a 6% chance of a team winning and look at the results you'll see them win 6 times.