Saturday, November 7, 2020

Weekly Indicators for November 2 - 6 at Seeking Alpha


 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

With the almost certain election of Joe Biden to the Presidency, we can expect a much more organized and effective Federal response to the pandemic beginning on January 20. If the pandemic is brought under control by spring sometime (and maybe a vaccine becomes available), the ordinary function of the leading economic indicators will be resumed - which means they start to have extra importance right about now.

As usual, clicking over and reading will bring you up to the virtual moment in terms of what those indicators forecast, and reward me a little bit for giving you that information.

P.S.: Because I am in much better frame of mind for some reason over the past few days, I will probably put up an edition of the Coronavirus Dashboard later today or tomorrow, as well as a post-mortem of the election nowcasts. 

Friday, November 6, 2020

October jobs report: extremely strong monthly gains overall, but at this rate still another 18 months from full jobs recovery


 - by New Deal democrat

  • 638,000 million jobs gained. The gains since May total about 55% of the 22.1 million job losses in March and April. The alternate, and more volatile measure in the household report was 2,243,000 jobs gained, which factors into the unemployment and underemployment rates below.
  • U3 unemployment rate declined -1.0% from 7.9% to 6.9%, compared with the January low of 3.5%.
  • U6 underemployment rate declined -0.7% from 12.8% to 12.1%, compared with the January low of 6.9%.
  • Those on temporary layoff decreased -1,432,000 to 3,205,000.
  • Permanent job losers decreased by 72,000 to 3,684,000.
  • August was revised upward by 4,000. September was also revised upward by 11,000 respectively, for a net gain of 15,000 jobs compared with previous reports.
Leading employment indicators of a slowdown or recession

I am still highlighting these because of their leading nature for the economy overall.  These were positive: 
  • the average manufacturing workweek rose  0.3 hours from 40.2 hours to 40.5 hours. This is one of the 10 components of the LEI and will be a strong positive.
  • Manufacturing jobs rose by 38,000. Manufacturing has still lost -621,000  jobs in the past 8 months, or 4.8% of the total. 55% of the total loss of 10.6% has been regained.
  • Construction jobs rose by 84,000. Even so, in the past 7 months -294,000 construction jobs have been lost, 3.8% of the total. About 75% of the worst loss of 15.2% loss has been regained.
  • Residential construction jobs, which are even more leading, rose by 18,000. In the past 8 months there have still been -6,400 lost jobs, or about 0.8% of the total.
  • temporary jobs rose by 108,700. Since February, there have still been -342,700 jobs lost, or 11.7% of all temporary help jobs.
  • the number of people unemployed for 5 weeks or less declined by -52,000 to 2.5 million, compared with April’s total of 14.283 million.
  • Professional and business employment rose by 208,000, which is still -1,149,000, or about 5.3% below its February peak.

Wages of non-managerial workers
  • Average Hourly Earnings for Production and Nonsupervisory Personnel: rose $0.05 from $24.77 to $24.82, which is a gain of 3.6% in the 8 months since the pandemic began. Gains had previously reflected that job losses were primarily among lower wage earners, who have been disproportionately recalled to work. That we have increased employment and increased wages as well is a very positive development.

Aggregate hours and wages:
  • the index of aggregate hours worked for non-managerial workers rose by 1.2%. In the past 8 months combined this has nevertheless fallen by about -6.3%.
  •  the index of aggregate payrolls for non-managerial workers rose by 1.4%. In the past 8 months combined this has nevertheless fallen by about -3.0%. Close to 80% of the loss from February to April has been made back up.

Other significant data:
  • Full time jobs were responsible for 1.0 million of the gain in the household report.
  • Part time jobs were responsible for 1.2 million of the gain in the household report.
  • The number of job holders who were part time for economic reasons increased by 383,000 to 6.683 million. This is still an increase since February of 2,365,000.

Important note: There was a decline of 268,000 in government jobs. This included -147,000 census workers and -159,000 teachers!


This was an extremely strong report. About the only negative was the big decline in education jobs, which may have been a quirk of seasonality.  Everything else was positive.

The household report’s gain of over 2 million jobs was responsible for most of the good news, including the decline in the un- and under-employment rates, and gains in both full and part-time jobs, as well as gains in aggregate hours and payrolls.

Further, all of the leading jobs sectors showed gains. This bodes well for the months ahead.

Nevertheless, only a little over half of all of the jobs lost due to the pandemic have come back, and the rate of increase since June has slowed to a *relative* crawl. It would take another 18 months, at the rate of this month’s job gains, to get back to the number of jobs that existed in February.

Thursday, November 5, 2020

Slow pace of improvement in jobless claims continues


 - by New Deal democrat

This week’s new jobless claims were essentially unchanged (but at their pandemic low), while continued claims continued their decline, also to a new pandemic low.

On a non-seasonally adjusted basis, new jobless claims declined by only 543 to 738,166, just above October 3’s revised pandemic low 731,249. Seasonally adjusted claims declined by 7,000 to 751,000, a new pandemic low (which was also last week’s number before revision this week). The 4 week moving average also decreased by 4,000 to 787,000, also a new pandemic low. Here is the close up since the end of July - for comparison, remember that these numbers were in the range of 5 to 7 million at their worst in early April: 

Continuing claims (which lag initial claims typically by a few weeks to several months) on a non-adjusted basis declined by 537,898 to 6,951,731. With seasonal adjustment they declined by 538,000 to 7,285,000. Both of these are new pandemic lows:

Continuing claims are now about 70% below their worst level from the beginning of May, but are still about 900,000 - 1,300,000 higher than their worst levels of the Great Recession.

The very slow improvement in layoffs has generally continued, similar to the same slow continued improvement in most of the “weekly indicators” I update each Saturday.  I continue to harbor serious doubts whether that will continue to be the case as cold weather forces some venues like outdoor dining to close again, and the pandemic continues to surge yet again, albeit with lower levels of deaths than last spring.

Wednesday, November 4, 2020

The Morning After


 - by New Deal democrat

Last night was truly a dark night of the soul. The rule of law in an American Republic simply will not withstand the re-election of Donald Trump, together with a continuing GOP majority in the Senate. Not to mention the prospect of continuing to shelter in  place for perhaps years as the COVID pandemic is allowed to run free.

As of this morning, absentee and mail-in ballots are still being counted in many swing States. As a result, as of when I type this, Biden has carried Arizona, assumed a slight lead in Wisconsin, maintains a slight lead in Nevada, is only ~13,500 away from taking the lead in Michigan with about 10% of ballots still to be counted, and the NYT says that Georgia is back in play and slightly leading towards Biden. And for what it’s worth, once California counts all its ballots, Biden is likely to have won something like 52.5% of the popular vote.

In other words, as of right now, the most likely outcome is somewhere between the two below possibilities:

If Biden were to take all of those, plus PA, which may not finish counting ballots until Friday, and NC (which still has not counted some ballots), here’s what the Electoral Map will look like:

If he simply maintains his leads, and takes the lead in Michigan, here is what the Electoral Map will look like:

Note that even in more pessimistic second map, Biden will wins a majority of 274 Electoral votes.

Put another way, basically the Presidential map looks very much like my final forecast map with the 3% error band, where we give Trump all of the toss-ups, but Biden still wins:

Still, the fact remains that under the above scenarios there appears to have been roughly a 3%-4% “hidden” Trump vote. That also was apparent in the Senate races, where the two Democrats with under 4% leads - Gideon in Maine and Cunningham in NC - lost. As of now, the Democrats have 48 seats, and the GOP 50. If Perdue continues to lead in Georgia, the GOP will retain its majority, although it is possible the Democrats could pick up the other Georgia seat in a runoff election (but don’t bet on it). Also, it appears that the Democrats also lost some Congressional seats, although they will most likely maintain their majority. 

A President Biden with a GOP Senate is going to be foreclosed from filling any judicial vacancies, and may not even be allowed to fill cabinet positions. Needless to say, budget hardball would continue. In short, he will be completely hamstrung in his first two years.

The bottom line: nearly 50% of your fellow Americans are perfectly happy with the idea of turning this country into a full-on herrenvolk banana republic.

Sent from my iPad

Monday, November 2, 2020

September housing construction and October manufacturing both on a tear


 - by New Deal democrat

It’s the first of the month, so we get the last laggard for September (construction spending) and the first read on October (ISM manufacturing). Both were very positive in their important components.

While total construction spending was only up 0.3% from a downwardly revised August, private residential construction spending (i.e., non-public housing construction) increased 2.8%:

This is yet more evidence of a very strong housing rebound brought about by record low mortgage rates.

The ISM manufacturing index, including its new orders component, was even more positive. The total index increased 3.9 to 59.3, the highest reading since 2018. The new orders component was even stronger, up 7.7 to 67.9, also the highest since 2018:

Manufacturing is on a tear.

Housing and manufacturing together are the most positive components of the entire economy right now. The former is a long leading indicator, the latter a short leading indicator.  Bring the pandemic under control and the US economy is going to do extremely well in 2021.

Sunday, November 1, 2020

The final 2020 Presidential Electoral College forecast: Biden 350, Trump 181, 7 toss-ups


 - by New Deal democrat

Here is my final nowcast for the Presidential 2020 election, 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. For this final installment, I am also including forecasts from 3 “fundamentals” models, including the one based on the Index of Leading Indicators that already forecast a result over half a year ago.

At only 2 days from Election Day, the polls are probably less than 1.5% off the final result. All of the fundamentals are known, and Trump’s attempt at an “October surprise” fizzled. There is some evidence of previously undecided partisan voters “coming home” in the past week or two. Aside from that, all that is left are brazen judicial shenanigans. In this last regard, here is a really helpful map of what days we can expect the ballot counting to be complete in various States, courtesy of Mike Sherlock a/k/a Mish:

By November 6, unless the Supreme Court wants to end the American Republic in the next 5 days, we should have results from Pennsylvania and enough other States to know definitively who the victor is, if we don’t know already.

Once again, unlike 2016, Biden’s lead been at very least steady for nearly 5 full months:

Not only that, but as of yesterday about 67% (about 94 million!) of the total number of ballots compared with 2016 have already been cast:

Trump has not made any extraordinarily incompetent and cruel mistakes since early  October, and as it has always done in the past, his approval ratings have recovered to their normal range. This week his approval declined by -0.4% to 42.2%, while his disapproval rate also declined -0.9% to 52.7% — again, well within the normal range of approval going back over 3.5 years: 

So, here are the final two maps through October 31 for the Presidential election. To refresh, in the first map:

- 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.

This week New Hampshire improved to “solid Biden,” and Nevada to “likely Biden.” Meanwhile Florida decline back to “toss-up,” Alaska to “likely Trump,” and Utah to “solid Trump.”
Even so, Biden still has 279 “solid” or “likely” Electoral votes, enough to win without any “leaning” or toss-up States. 
This final week I am also including a map that includes any State with a 1%+ differential to “lean Democrat/GOP” category:
In perhaps the biggest shock of the 2020 race, Georgia looks likely to join North Carolina as a Blue State, following Virginia as the 3rd east coast State of the old Confederacy to do so. Ohio and Texas are slightly likely to remain Trump States. The only true remaining toss- ups are  Iowa and Maine’s 2nd Congressional District.
Biden’s support remained at 50%+ in Michigan, Minnesota, Nevada, and Wisconsin. But P en. nsylvania and Nevada declined below 50%, although above 49%. North Carolina also remained at 49%+. Florida declined below 49%, joining Arizona. Florida is the only State I am really worried about “flipping” from Biden to Trump, as it appears enough Cuban  Americans and elderly retirees have been “coming home” to the GOP:
But perhaps more importantly, here is Pennsylvania, hanging tough for Biden. Although I won’t post the graphs, Minnesota, Wisconsin, and Michigan are similar:
 Biden’s lead in these States has been durable, and he has an actual majority in all but Pennsylvania. In short, this year the “blue wall” is extremely likely to hold. And if it does, Trump’s chances of victory are foreclosed.

Before I close out this exercise, I want to reiterate forecasts from several other good “fundamental” models.  
The first focuses on whether there has been an increase in the unemployment rate during  the election year by more than +0.3%.  Needless to say, the increase from 3.5% in February to 7.9% in September bodes ill from Trump:

Secondly, Douglas Hibbs’ “Bread and Peace” model, which relies on the Q2 and Q3 ch ange in real disposable income and then subtracts for military casualties, gives contradictory signals depending on whether or not you include pandemic fatalities as being similar to military fatalities. Here’s the model’s performance through 2016:

  Hibbs, who is still alive, has not updated his model for 2020.
The Congressional stimulus earlier this year strongly boosted personal income. Since there have been negligible military casualties, another researcher duplicating Hibbs’   methodology recently estimated that the model would predict a roughly 52% vote share victory for Trump. But in 1952 and 1968, military casualties of roughly 35,000 out of a roughly 175 million US population subtracted 7% to 8% from the incumbent share. If one  plugged the pandemic casualties adjusted for current US population, the model would subtract almost 4x that number, or roughly 30%, from Trump’s share!

Finally, my favorite “fundamentals” model takes the candidates’ Q1 polling results and a. djusts for the reading of the index of leading indicators through that point. In Q1, Biden already had a slim lead over Trump in national polls, and then the pandemic did this to the Index of Leading Indicators:

I’ve included the graph through June just for comparison purposes.

In short, the Index of Leading Indicators forecast a very dramatic drop in the economy,    which did happen and from which the economy has only partially recovered. Here is Prof. Menzie Chinn’s graph of the 4 big coincident indicators through October:

In short, with the unusual exception of one that focuses on personal disposable income,    the “fundamentals” models strongly indicate a big Trump loss.  Further, the model that is based no the Index of Leading Indicators foretold this over half a year ago, as I wrote here LINK.
In sum, I am increasingly confident of a Biden victory, perhaps a decisive one, with the final Electoral College tally of Biden 350, Trump 181, with 7 toss-ups.

The final 2020 Senate nowcast: 51 Democrats, 48 GOPers, 1 true toss-up


 - by New Deal democrat

Here are the final maps for the Senate, including all polls since October 1. I’ll post the Presidential nowcast separately later, since it is going to be significantly longer this week. 

To refresh, In the first map below:

- 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.

This results in a map showing 50 Democratic seats, 43 GOP, and 7 toss-ups.

In the 2nd map below, any race with a gap of 1%+ but less than 5 is designated a “lean:”

There are 5 Senate races where the GOP is ahead by more than 1% in the polls, but less than 3%. One of those is the Georgia Senate special election, which is likely to head to a run-off in January. If all of the D and R candidate votes go over to the two survivors, the GOP candidate is a slight favorite. The only Democratic pickup is Iowa. The one true toss-up is the Ossoff vs. Perdue race in Georgia.

So, the final 2020 forecast for the Senate is that the Democrats will win a *very* slim majority, and lose 4 or 5 more heartbreakingly close contests.