Monday, October 19, 2020

Two noteworthy tweets

 

 - by New Deal democrat

It’s a slow economic news week. Housing starts and permits are reported tomorrow, and jobless claims and existing home sales on Thursday. I’ll update the Coronavirus Dashboard Wednesday.  So for today, two nuggets.

1. Nate Silver discovers behavioral psychology:



This has been my paradigm for months. Panic breeds compliance with mask-wearing and social distancing. Complacency breeds risk-taking. Over time both trends wane, breeding the conditions necessary for the opposite outcome. Not only has this been true in almost all US States, but we have now seen the same dynamic play out in Europe.


Nice to see that Nate Silver is learning about learning.

2. What is the solution to the Supreme Court?

Hoarse is right, although I suspect the Supreme Court will immediately issue injunctions against laws passed by the Democrats, or uphold injunctions issued by lower courts, and won’t wait for 2023 to strike them down. 

In the short term, the Democrats can certainly expand the Court. Of course, the GOP will retaliate the next time they are in power. And on we go, destroying the institution (to be fair, the movement conservatives on the Court, after Bush v. Gore and Shelby County, not to mention fast-tracking Trump appeals while slow-walking those of Congressional Democrats, have brought this on themselves.)

The better long term solution, which I have previously endorsed, as have others like Matt Yglesias, is for a single 18 year term on the Court, with appointments made during the 1st and 3rd year of each Presidential term, after which the Justices can become “Senior Justices” who sit on the Appellate Courts for life or until they choose to retire.

Lower courts already have “senior judge” status - but it is *voluntary.* I doubt Clarence Thomas or Samuel Alito will voluntarily retire during a Democratic Presidency. Making it mandatory would require a Constitutional Amendment. Good luck with that.

So the bigger quandary with the solution, as I see it, is how to club the GOP with the short term retaliation for their own misbehavior (see Garland, Merritt) while offering a neutral longer term solution that they could endorse, and hopefully without necessity of a Constitutional Amendment.

I have a couple of ideas, but they require a longer post. Once I flesh them out, I’ll have more to say.

Sunday, October 18, 2020

The 2020 election nowcast: Biden widens national lead; Senate races likely to follow Presidential result in each State

 

  - by New Deal democrat

Here is my weekly update on the 2020 elections, based on State rather than national polling in the past 30 days, since that directly reflects what is likely to happen in the Electoral College. 

At only 16 days from Election Day, the polls, while actually nowcasts rather than forecasts, are probably less than 2% off the final result. With the exception of the last Presidential debate, and any *significant* “October surprise,” all of the fundamentals of the election are already “baked into the cake.” Because some GOP voters will likely still “come home” in the next two weeks, I expect the race to tighten a little bit.

There are two big takeaways from the present situation:

1. In the Presidential election, Biden’s lead has not just been steady, but on a national level has been pulling decisively away from Trump, to the biggest lead of the entire year:


2. The Senate elections show very little variation from Presidential polling in the affected States. The only 4 States in which contrary results at the two levels look reasonably possible are Georgia, Iowa, North Carolina, and South Carolina. 

With those statements out of the way, here is this week’s update.

Trump has continued to have a bad October. In the past week his approval rating declined by another -0.8%, while his disapproval rate rose a full 1.0% to 53.2% - which is still within the normal range of approval going back over 3.5 years: 


Here is this week’s updated map through October 17 for the Presidential election. To refresh, here is how  it works:

- States where the race is closer than 3% are shown as toss-ups.
- States where the range is between 3% to 5% are light colors.
- States where the range is between 5% and 10% are medium colors.
- States where the candidate is leading by 10% plus are dark colors.

A reminder: When we get to the last week before the election I will probably move any State with a 1%+ differential to “lean Democrat/GOP,” rather than stay with 3%.

 
There was something remarkable this week: not a single State changed ratings. That’s how static this race is. Biden’s “solid” plus “likely” Electoral College votes remained at 279 this week, below their peak of 302 over two months ago - but Biden still doesn’t need Florida or Arizona in order to win.

Biden’s support remained at 50%+ in Michigan, New Hampshire, Pennsylvania, Wisconsin, and Minnesota. While Arizona declined to 49%, Florida and North Carolina were 49%+. This makes it extremely hard for Trump to mount a successful comeback in those States.

Turing to the Senate, there were 5 small changes this week, as Alaska moved from toss-up to lean GOP, Maine moved from likely to lean Democrat, Montana moved from lean GOP to toss-up, and New Mexico moved from likely to solid Democrat:


At current polling, if Democrats win all those seats in which they are favored, they will have 51 Senate seats. There are also 5 races now rated “toss-ups,” as to which the most important question is how many undecided voters “come home” to the GOP candidate. Still, it is quite likely that the Democrats wind up with a majority in the next Senate.

Saturday, October 17, 2020

Weekly Indicators for October 12 - 16 at Seeking Alpha

 

 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

What was most noteworthy about the past week is the confirmation that consumer spending, so far, has continued to hold up even as emergency Congressional assistance has been terminated for a month and a half.

As usual, clicking over and reading will bring you up to the moment on all of the important economic data, and will reward me with a penny or two in my pocket for putting the information together for you.

Friday, October 16, 2020

Real retail sales continue (inexplicably) strong, still bode well for employment

 

 - by New Deal democrat


This morning we got two important monthly September reports: industrial production and retail sales.

I have more to say about industrial production, and some general economic analysis about retail sales, which are pending at Seeking Alpha. I will post a link once that article goes up. UPDATE: Here’s the link: Link

For this blog, let’s focus on how real retail sales are likely to affect employment. 

Just to start, here is are CPI adjusted retail sales. You can see that they have actually jumped once the Congressional stimulus kicked in, and have remained well ahead of pre-pandemic levels:


As I have said many times in the past, consumption slightly leads employment. It has almost universally done so for the entire 70+ year history that both measures have been kept. Basically, demand for goods and services drives hiring to fulfill that demand (or at least to an increase in hours employed) typically within a few months later.

I am frankly very surprised that sales have held up in the past several months, despite the ending of emergency Congressional assistance. But done so they have - although it may largely be a function of seasonal distortions caused by the pandemic.

But, turning to employment, historically consumption has led employment (/2) by several months (albeit with lots of noise), and has an even closer relationship with aggregate hours (all shown YoY below):



Here is a close-up on the past year:


Because sales have made a full recovery, and indeed on a YoY basis are accelerating, I expect employment and hours worked to continue to show gains for the next several  months, although at a slower pace similar to what we have seen in the August and September reports. 

Thursday, October 15, 2020

Jobless claims: only one week’s data, but cause for significant concern

 

 - by New Deal democrat

Today marked the biggest increase in new jobless claims in two months, and one of the two biggest increases since May, while the slightly lagging continuing claims continued to decline.

On a non-seasonally adjusted basis, new jobless claims rose by 76,670 to 885,885. After seasonal adjustment (which is far less important than usual at this time), claims rose by 53,000 to 898,000. The 4 week moving average also increased by 8,000 to 866,250: 


Here is a close-up of the last four months highlighting the overall glacial progress in initial claims since the beginning of August:


Continuing claims declined on a non-adjusted basis declined by -1,188,202 to 9,631,790. With seasonal adjustment they declined by 1,165,000 to 10,018,000. On the bright side, both of these numbers are new pandemic lows:


 Continuing claims are now about 60% below their worst level from the beginning of May, but remain about 3 to 3.5 million higher than their worst levels during the Great Recession.

Only one week’s data, but it is a significant concern that claims have risen, after largely stalling for two months. The situation is at best only improving at a snail’s pace, and at worst is deteriorating again as we head into winter and a likely renewed increase in COVID cases.

Wednesday, October 14, 2020

Coronavirus dashboard for October 14: winter is coming

 

 - by New Deal democrat

Total US confirmed cases: 7,806,805*
Average cases last 7 days: 51,038
Total US deaths: 215,887
Average deaths last 7 days: 714

*Actual cases probably more like 14 million, or over 4% of the US population

Source: COVID Tracking Project

Today let’s take a look at the most recent upsurge in COVID not just in the US, but in the entire West.

Here is the 7 day average of new cases per capita in the US, Canada, and the 5 most populous countries in Europe:


Every single country, even Germany, is experiencing an upsurge. France, Spain, and the UK are having an even worse outbreak than the US.

Here are deaths per capita. Note that this graph only shows the past 12 weeks. Early on the US, UK, Spain and Italy all had daily deaths on the order of 14 to 18 per capita, dwarfing everything since:


We know that these will follow new cases with a 2 to 4 week lag. Spain already exceeds the US. France and the UK are following behind.

All things considered, Canada and Germany are still doing quite well. But the increase in cases there certainly points to how hard it is to keep this virus contained.

Within the US, here is the spaghetti chart of cases in all 50 States plus DC and the territories:


Note that the Dakotas and Montana now have worse outbreaks than even NY did in April.

Here are deaths:


In April, NY had deaths of about 50 per capita every day, so even North Dakota, at 14 per day, while the worst since then - even surpassing Arizona’s summer spike - is nowhere near that bad - yet.

Here are the bottom 10 jurisdictions for infections:


Only Maine and Vermont among the 50 States still have the pandemic controlled.

Finally, here are the bottom 10 for deaths:


Again, only Maine and Vermont have still fully contained the virus.

And winter is coming.

Tuesday, October 13, 2020

Consumer prices rise a “normal” 0.2% in September

 

 - by New Deal democrat

In September Consumer prices rose a “normal” 0.2%, the first such typical increase since the pandemic began (blue in the graph below):



For the past 40 years, recessions had typically happened when CPI less energy costs (red) had risen to close to or over 3%/year. We are nowhere near that now (last 15 years shown in graph):


Because wages are “stickier” than prices, typically as recessions beat down prices (or at least price increases), in real terms wages rise. That has been the case for the coronavirus recession as well:


As a result, as I’ve noted for the past several months, real hourly wages for non-supervisory workers have finally exceeded their previous 1973 peak:


Of course, this is also because lower wage workers disproportionately lost their jobs during the pandemic. Until several months ago, they were buoyed by emergency Congressional assistance. One of the big mysteries in the past few months is why the expiration of those benefits has not caused a steep drop in consumer spending. Retail sales will be reported on Friday, and that is when we will know the answer for September.

Monday, October 12, 2020

College educational attainment by age demographic

 

 - by New Deal democrat


There is no economic data today due to the Columbus Day observance.


So let me drop this graph of a metric I have been trying to find, of college educational attainment by age demographic, that I finally came across a couple of days ago:


It is a commonplace that among Whites at least, support for Democrats is highly correlated by college education. What is unclear is whether that is actually a function of education itself, or is simply confounded by age group.

The above graph shows are steady rise in the percentage of both men and women with college degrees over the past 80 years. But we know that the strongest support for the GOP, and for Trump in particular, is in the age cohort of younger Boomers and the first half of Gen X, roughly birth years 1960-1970. That equates with college degrees being granted about 20 years later, from 1980 to 1990 or so.

Yet there is no notable exception in the growth of the percent of college educated persons during the 1980s. That strongly suggests that it was the condition of the economy in particular during the 1970s and 1980s that led to such strong GOP identification among that age group, rather than their educational attainment.

I still haven’t found any good information that provides cross-tabs on both educational attainment and age group, to more specifically sort out the strongest correlation. So for now, this will have to do.

Sunday, October 11, 2020

The 2020 Presidential and Senate nowcast: birds coming home to roost


 - by New Deal democrat


Here is my weekly update on the 2020 elections, based on State rather than national polling in the past 30 days, since that directly reflects what is likely to happen in the Electoral College. 

Now that we are only 23 days from Election Day, let me update and revise a few comments. First of all, while polls are really only nowcasts - snapshots of the present, rather than forecasts - at this point they are probably only about 2% off the final result: 


At the moment, according to 538, Biden is up 52.2% to 41.9%. Even if each % moved 2% favorable to Trump, that would still be a 50.2% to 43.9%, or 6.3%, advantage for Biden. That’s a pretty solid lead.

A dditionally, all of the important COVID and economic information is already “baked in the cake.” There is nothing that is going to greatly change perceptions for either outlook (unless one of Trump’s inner circle were to die of the disease). The one issue that is likely to be very front and center in the next several weeks is the Supreme Court. So far I have not seen any polling indicating which if either of the candidates might benefit.

Further, I expect that we are already seeing marginal voters “come home” to their respective parties, as will be apparent in the maps below. 

Because of the vast increase in the numbers of early voters in States where voting is already underway, it’s not clear at all that polls in late October or the first several days of November should receive increased weight. Beginning next week, my plan is to slightly increase the weighting of any polls conducted during October vs. September, and leave it at that. 

Finally, because the polling has become extremely persistent, when we get to the last week before the election I will probably move any State with a 1%+ differential to “lean Democrat/GOP, rather than stay with 3%.

With those statements out of the way, here is this week’s update.

Trump had a particularly bad past 12 days - and as a presumed result of which, his approval rate has declined all of -0.6%, while his disapproval rate has increased by +0.9%. Needless to say, this is well within the normal range of approval going back 3.5 years: 


Here is this week’s updated map through October 10 for the Presidential election. To refresh, here is how  it works:

- States where the race is closer than 3% are shown as toss-ups.
- States where the range is between 3% to 5% are light colors.
- States where the range is between 5% and 10% are medium colors.
- States where the candidate is leading by 10% plus are dark colors.


Two weeks ago there were a number of changes adverse to Biden. This week they have all disappeared. Additionally, both Indiana and Mississippi went from “solid” to “likely” GOP, on the basis of relatively weak new polls. But most importantly, Florida has moved back from toss-up to “lean Biden.” I have seen lots of information that seniors have decisively soured on Trump, and it appears to be showing up there.

Biden’s “solid” plus “likely” Electoral College votes remained at 279 this week, below their peak of 302 two months ago. In other words, Biden doesn’t even need Florida or Arizona in order to win.

Biden’s support also improved to 50%+ in Arizona, in addition to Michigan, New Hampshire, Pennsylvania, Wisconsin, and Minnesota. This makes it extremely hard for Trump to mount a successful comeback in those States.

Turing to the Senate, there were 3 changes: Montana went from toss-up to “lean GOP,” while Iowa, importantly, went from toss-up to “lean Democrat.” Additionally, we have more clarity in the 2nd Georgia “jungle primary” election. It appears that Democrat Warnock will make it through to the finals. But once there, the combined GOP vote looks to be nearly 10% more than the combined Democrat vote. I have thus moved the race to “likely GOP”:


At current polling, if Democrats win all those races in which they lead, they will have 51 seats. The number of “toss-ups” has declined to 5.

A Biden victory and a majority Democrat Senate looks increasingly likely, which is probably why we are beginning to see a few Senate Republicans publicly put distance between themselves and Trump.


Saturday, October 10, 2020

Weekly Indicators for October 5 - 9 at Seeking Alpha

 

 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

Because the pandemic so quickly and completely put the economy in a coma back in April, and the economy remains at severely depressed levels by most metrics compared with a year ago, in June I started to compared current levels with the worst pandemic readings.  This week I went further, and calculated the best readings since the pandemic began for most indicators, and am tracking those as well.

The results as of now confirm that the economy has *slowly* continued to improve into the autumn compared with summer.

As usual, clicking over and reading will give you a comprehensive look at where the economy is right up until this week, and it will also reward me with a penny or so for my efforts in keeping you up to date.

Friday, October 9, 2020

Coronavirus dashboard for October 9: everybody has to touch the hot stove at least once

 

 - by New Deal democrat

Total confirmed US infections: 7,605,218*
Average infections last 7 days: 46,869

Total US deaths: 212,762
Average deaths last 7 days: 717

*Actual number is probably 5 to 7 million higher, or about 4% of the total US population
Source: COVID Tracking Project

In the last 4 weeks, the average number of new infections has risen again. Here is the breakdown by regions: