Category: Statistics

Missed Jobless Thursday, My Bad

No Comments

Initial claims increased to 744,000, up by 16,000, and the second straight increase.

I am thinking that initial claims will remain in the mid 700K range for a while yet: 

Workers are slowly pulling away from unemployment assistance as a U.S. economic revival picks up speed, with initial filings for benefits holding near pandemic lows and the number of people receiving help dropping.

Initial jobless claims, a proxy for layoffs, increased by a seasonally adjusted 16,000 last week to 744,000, the Labor Department said Thursday. The four-week average, which smooths out volatility in the figures, rose slightly to 723,750 from 721,250. Claims are still well above the weekly average of around 220,000 in the year before Covid-19’s arrival.

The continued high rate of filings comes amid other signs of recent labor-market improvement. U.S. employers added 916,000 jobs in March, and the unemployment rate slipped to 6.0%, from 6.2% in the prior month.

………

Moreover, the number of people receiving unemployment assistance is slowly declining. Continuing claims, which provide an approximation of the number of people receiving benefits, at the end of March reached their lowest level of the pandemic, declining slightly to 3.73 million. A broader reading that includes state and pandemic-related federal programs also eased slightly to 18.16 million.

As always, I am more of a pessimist, particularly with Coronavirus cases spiking and talk about the re-imposition of some restrictions becoming more common.

Good Monthly Jobs Report

No Comments

The Scariest Jobs Chart Ever


Workforce Participation Rate

There were 916,000 added to non-farm payrolls in March, which is a very good performance, though not anywhere near close to what it was during the before time. (See graph pr0n)

Note that NFP is still down over 8 million from its peak, and the employment-population ratio is lower than it has been since the depths of the 1981 recession, and that the percentage of women in the workforce has risen steadily since that time, so an apples to apples comparison (correcting for women in the workforce and an aging population) might still have us back to somewhere in the great depression.

We’re still in a hole, but it is not as deep:

U.S. hiring surged in March as the economic recovery accelerated, the start of what economists say could be a sustained run of job growth to industries, regions and workers hardest hit during the pandemic.

U.S. employers added a seasonally adjusted 916,000 jobs in March, the best gain since August, the Labor Department said Friday, and the unemployment rate, determined by a separate survey, fell to 6.0%, a pandemic low. Still, as of March, there are 8.4 million fewer jobs than in February 2020 before the pandemic hit.

The jobs rebound is gaining renewed momentum as more people are vaccinated against Covid-19, states lift restrictions on business activity, and consumers grow more comfortable dining, shopping and traveling outside their homes.

Note also that Covid numbers are spiking again, so we may see a reversal. 

………

Friday’s report showed hiring rose in most industries, led by a gain of 280,000 in the category that includes restaurants and hotels. Employment also rose sharply in construction, most manufacturing sectors and public and private schools. Temporary help and auto manufacturing, where a semiconductor shortage has idled assembly plants, were weak spots.

So a lot of the growth was in lo wage jobs.

………

Some economists project job growth will top one million in April. Further out, economists surveyed by The Wall Street Journal project employers will add an average of 514,000 jobs each month over the next year, for a total of more than six million. That would mark the best 12-month stretch of job creation in decades but leave overall employment totals below where they stood before the pandemic.

I’ll always be a pessimist, which comes from the fact that the “experts” have consistently given an excessively rosy assessment of economic outcomes of recessions throughout my entire adult life. 

The levels of inequality in our society will weigh down any recovery.

And Back Up Again

No Comments

U.S. initial unemployment claims rose by 61,000 to 719,000 last week, which means ……… Hell, I don’t know.

The 4 week average fell though.

The weekly average for the before times was a little bit over 200K:

Filings for unemployment benefits rose last week but remained near their lowest levels since the pandemic’s onset, amid signs of a broader U.S. economic recovery.

Workers filed 719,000 initial jobless claims, on a seasonally adjusted basis, in the week ended March 27, the Labor Department said Thursday. The increase followed a downward revision to 658,000 initial claims the prior week, the lowest point since the pandemic hit in March 2020.

The four-week moving average, which smooths out volatility in the numbers, fell to 719,000, also a low during the pandemic.

I’m not sure what it all means in terms of trends, but until the weekly number drops below 300K, I would not take my foot off of the gas pedal.

This Can’t Be Good

No Comments

With all the signs pointing to explosive economic growth in 2020, economists are generally predicting a growth rate greater than 5%, the news that personal income fell by $1,516.6 billion (7.1%) from January to February is a pretty dire data point. 

Obviously, there were some freak conditions in February, the massive Texas blackouts come to mind, but this is pretty grim news:

Hence, when the opening line of the press release for this report tell us “Personal income decreased $1,516.6 billion (7.1 percent) in February“, that means that the annualized figure for US personal income in February, $19,945.6 billion, was $1,516.6 billion, or roughly 7.1% less than the annualized personal income figure of $21,462.2 billion for January; the actual change in personal income from January to February is not provided…similarly, annualized disposable personal income, which is income after taxes, fell by nearly 8.0%, from an annual rate of an annual rate of $19,210.5 billion in January to an annual rate of $17,678.2 billion in February…the components of the monthly decrease in personal income, which can be seen in the Full Release & Tables (PDF) for this release, are also annualized figures…in February, the reason for the $1,516.6 billion annualized decrease in personal income was a $1,584.1 billion annualized decrease in government social benefits to individuals, which was only slightly offset by a $37.7 billion annualized increase in business & farm proprietors’ income and a $15.6 billion annualized increase in interest and dividend income…wages and salaries, which fell by an annualized $0.2 billion, were barely a factor in February’s personal income change . . .

Not great economic news.

Jobless Claims Finally Fall Below Pre-Pandemic Record

No Comments

There were 684,000 initial claims, less than the pre-pandemic peak of 695,000.

Yes, this is unmitigated good news:

Jobless claims fell to their lowest level of the pandemic last week as stronger hiring and consumer spending drive a U.S. economic revival.

Worker filings for unemployment benefits, a proxy for layoffs, fell to 684,000 last week from 781,000 a week earlier. Claims are now at the lowest point since mid-March of last year, before lockdowns triggered millions of layoffs. They are also below the pre-pandemic high of 695,000, a threshold not crossed for 52 weeks.

“The recovery is really hitting full steam again, and all of the conditions will be in place for a real, explosive liftoff in the summer when hopefully we’ve reached a higher vaccination threshold,” said Julia Pollak, labor economist at jobs site ZipRecruiter.

………

Economists surveyed by The Wall Street Journal this month raised their average forecast for 2021 economic growth to 5.95%, measured from the fourth quarter of last year to the same period this year, from a 4.87% projection in February’s survey. The higher figure would mark the fastest such pace in nearly four decades, following a steep downturn last year.

If these predictions are accurate, this is a blistering pace of economic growth.

Remarkably Toxic Individuals

No Comments

It turns out that only 12 people are responsible for the overwhelming majority of anti-vaxx content online.

It really is amazing what a few horrible people can do:

They’ve been dubbed the “Disinformation Dozen”: 12 individuals or organizations are tied to up to 65 percent of anti-vaccine content circulating on major social media networking sites, according to an analysis of popular anti-vaccine content on Facebook and Twitter.

………

The report accuses Robert F. Kennedy Jr. — who was banned from Instagram last month — Joseph Mercola, Ty and Charlene Bollinger — whose Twitter accounts were briefly suspended at the beginning of the pandemic — Sherri Tenpenny, Rizza Islam, Rashid Buttar, Erin Elizabeth, Sayer Ji, Kelly Brogan, Christiane Northrup, Ben Tapper and Kevin Jenkins of spreading disinformation and claims that their social media accounts “have repeatedly violated Facebook and Twitter’s terms of service agreements.” And the CCDH has receipts — the report is full of screenshots of “example violations” that range from misleading to antisemitic.

It’s not a surprise.  The past few years have shown us the potential effects of a few toxic individuals in the right (wrong?) places.

Speaking of Leverage

No Comments


Indeed, WTF


Increased leverage goeth before a fall

It appears that the use of margin trading, one of the things that triggered the stock market crash of 1929, is spiking, which is a pretty good indicator to me that we are headed to another market panic:

In the current craze that encompasses everything from sneakers and NFTs to stocks, where valuations don’t matter because of widespread certainty that valuations will be even greater in a few days, and where folks are chasing lottery-type returns, supported by the Fed’s interest rate repression and $3 trillion in asset purchases, and by the government’s trillions of dollars of handouts and bailouts – well, in this perfect world, there is a fly in the ointment: Vast amounts of leverage, including stock market leverage.

Margin debt – the amount that individuals and institutions borrow against their stock holdings as tracked by FINRA at its member brokerage firms – is just one indication of stock market leverage. But FINRA reports it monthly. Other types of stock market leverage are not reported at all, or are disclosed only piecemeal in SEC filings by brokers and banks that lend to their clients against their portfolios, such as Securities-Based Loans (SBLs). No one knows how much total stock market leverage there is. But margin debt shows the trend.

In February, margin debt jumped by another $15 billion to $813 billion, according to FINRA. Over the past four months, margin debt has soared by $154 billion, a historic surge to historic highs. Compared to February last year, margin debt has skyrocketed by $269 billion, or by nearly 50%, for another WTF sign that the zoo has gone nuts:

………

And it’s risky leverage for the borrower. It seems like risk-free leverage when stocks go up, but when your stocks do the unheard-of and tank below a certain level, your broker will ask you to put more cash into your account or sell stocks into the tanking market, whereby you then join the legions of forced sellers.

In the past, a big surge in margin balances tended to precede history-making stock market declines:

………

Leverage is the great accelerator of stock prices, on the way up, and on the way down. Purchasing stocks with borrowed money creates buying pressure, and prices rise, and rising prices increase the margin balances a portfolio can support, and this encourages more stock-buying on margin.

On the other hand, selling stocks to deal with margin calls adds more selling pressure to an already declining market. The more prices fall, the more selling pressure there is from frazzled forced sellers trying to deal with margin requirements.

When market correct, this is going to be very ugly.

H/t Naked Capitalism.

And Unemployment Goes Up

No Comments

Initial jobless claims rose by 45,000 to 770,000 last week.

Not good news:

Worker filings for jobless benefits are hovering near the pandemic’s lowest levels, adding to evidence of recent economic improvement.

Jobless claims rose last week to 770,000—still elevated above the pre-pandemic peak of 695,000—but have declined since January. The four-week moving average, which smooths out weekly volatility, fell last week to about 746,000, near November’s pandemic low.

An increase in Texas accounted for about half of last week’s overall rise in jobless claims, which could reflect delayed effects from last month’s winter storm, some economists said.

More broadly, declining jobless claims are one sign of economic improvement. U.S. employers added 379,000 jobs in February, and the unemployment rate ticked down to 6.2%. The U.S. manufacturing industry has exhibited steady signs of expansion. JPMorgan Chase & Co.’s tracker of credit- and debit-card transactions showed consumer spending climbed in early March.

Economists expect widespread distribution of vaccines and a fresh round of government stimulus to fuel growth in the first half of this year.

They are predicting that economic growth this year will be about 6.5% in the US, largely because of government stimulus.

Me, I’ll take the under, but the fact that people are lauding fiscal stimulus, as opposed to waiting for monetary stimulus to eventually do its thing, is a nice change in the conventional wisdom.

 

Better, But Still Not Good, Initial Claims Data

No Comments

Initial claims fell from 754,000 to 712,000 last week, indicating an improving, though still dismal, job market:

New filings for unemployment benefits last week neared their lowest level since the pandemic fueled a surge in layoffs last March, adding to evidence of renewed labor-market growth.

Jobless claims, a proxy for layoffs, fell to a seasonally adjusted 712,000 in the week ended March 6, down about 200,000 from an early January peak and close to a pandemic low point reached last November.

The four-week moving average, which smooths out volatility in week-to-week numbers, was 759,000 for the week ended March 6, slightly higher than the previous pandemic low recorded last November. The weekly average in 2019, the year before the pandemic started, was 218,000.

The recently passed $1.9 Trillion stimulus bill should lead to further improvements. 

It should be noted though that the employment population ratio is still crap, and improving VERY slowly.

I Missed This on Friday

No Comments


The Return of the Scariest Job Chart Ever


Workforce participation is still at a 45 year low

Largely because we changed cell phone providers, and my attempts to BYOD have been ineffective. (Not having a cell phone right now is a major drag).

The February employment report came out on Friday, and it was generally positive from a month-to-month perspective, but the job numbers are still worse than they were at the depths of the 2007-2012 recession.

As Calculated Risk observes: (They are also responsible for the graph pr0n)

The current employment recession was by far the worst recession since WWII in percentage terms.

At the worst of the Great Recession, employment was down Down 6.29% from the previous peak.

Currently employment is down 6.21% – the current unemployment situation is about the same as the worst of the Great Recession (and there was no pandemic to contend with in 2009).

I think that saying that we, “Are not out of the woods yet,” is too week a metaphor.

I don’t think that we have even reached the halfway point in Mirkwood.

About F%$#ing Time

No Comments

It appears that at least some economists are willing to learn, and they have that the headline unemployment rate is artificially low because it does not take into account discouraged workers. 

This has been true basically forever, but economists, who favor low wages for everyone but economists, and people who sit on their tenure committees have only now begun to realize this:

When Brianna Kipnis was laid off from a fitness start-up last June, she thought it would be nice to take a month off before returning to the jobs market. She cancelled the lease on her New York City apartment and moved in with her parents in neighbouring New Jersey.

………

The hopelessness felt by Kipnis and many others is one of the reasons that US policymakers, from the Federal Reserve to the Biden administration, have lost faith in the unemployment rate as an indicator of the strength of the jobs market.

The rapid decline in the US jobless rate has so far exceeded the forecasts of private sector economists and Fed officials alike. The latest reading, for February, will be published on Friday.

But the headline figure has obscured far less encouraging trends in America’s labour market, and is now considered an incomplete and unreliable guide to the trajectory of the US recovery.

“Published unemployment rates during Covid have dramatically understated the deterioration in the labour market,” Jay Powell, the Federal Reserve chair, said during a speech last month, noting that a more realistic unemployment rate was closer to 10 per cent.

Powell is not an economist, which is the second time that this has happened since (I think) William McChesney Martin left the post in 1970, (G. William Miller was in for about a year in the late 1970s, and his time in office was ……… problematic) and I would argue that he has been the best Fed Chair since then. 

You know what you call 1000 economists at the bottom of the ocean?  “A good start”

It’s Unemployment Claim Thursday

No Comments

And in a REMARKABLY circuitous headline, the Wall Street Journal announces that, “U.S. Jobless Claims Hold Nearly Steady,” because they rose only 9,000 from the (revised upward from 730,000 to 736,000) jobless claims of the week before.

It is a remarkably awful headline, and you know that if claims had fallen by 9,000 it would have been called a drop:

Filings for unemployment benefits in the latter half of February reached their lowest level in nearly three months amid signs of slow labor-market improvement.

The Labor Department said jobless claims, a proxy for layoffs, rose slightly to 745,000 for the week ended Feb. 27, from a revised 736,000 the prior week. The four week moving-average, which smooths out week-to-week volatility in claims numbers, was just under 800,000, its lowest level since early December.

So the numbers went up, and are still higher than they were in early December.

Screwing with headlines to minimize this is not a good look.

Tomorrow’s job numbers should be interesting.

FWIW, I don’t think that the Texas energy f%$#-up had much to do with this number, while there were certainly many people in Texas unable to work because of their delusional free-market energy dystopia, it is abundantly clear that none of them could file, because there was no power to run the unemployment offices.

Another Thursday, Another Unemployment Report

No Comments

And this time, initial claims went up:

Worker applications for unemployment benefits rose during the first half of February, pausing a downward trend that pointed to an improving labor market amid other signs that the economic recovery is picking up.

The Labor Department on Thursday said the increase to 861,000 last week was accompanied by a 55,000 upward revision of claims in the prior week, on a seasonally adjusted basis. That put the four-week moving average, which smooths out week-to-week fluctuations, at 833,000, slightly lower than the prior week and near the top of a roughly 750,000 to 850,000 range since last October.

Jobless claims—a proxy for layoffs—have remained above the pre-coronavirus pandemic peak of 695,000 since the start of the pandemic last March.

Joe, get those f%$#ing checks out the f%$#ind door, and $2000, not the f%$#ing ridiculously means tested $1600 you are trying to sell right now.

Microflaccid Office Fail

2 Comments

It turns out that one of the major data exchange formats for genetics is Microsoft Excel, and we have now discovered that the Redmond company’s flagship spreadsheet program has been autocorrecting the data into oblivion:

For many people, working with error-ridden spreadsheets is a way of life. This takes on added meaning for genomics researchers, who study the building blocks of life. It turns out that their work, too, is rife with dodgy spreadsheets.

A new paper has revealed the vast extent of errors in published genomics research, which is down to an unfortunate quirk of Microsoft Excel. A trio of scientists in Australia scanned 7,500 Excel files with gene lists accompanying 3,600 papers in 18 journals over a 10-year period. One-fifth of the files had easily identified errors, which is “quite striking and a little bit embarrassing,” says Mark Ziemann of the Baker IDI medical research institute in Melbourne, one of the paper’s co-authors.

What happened? By default, Excel and other popular spreadsheet applications convert some gene symbols to dates and numbers. For example, instead of writing out “Membrane-Associated Ring Finger (C3HC4) 1, E3 Ubiquitin Protein Ligase,” researchers have dubbed the gene MARCH1. Excel converts this into a date—03/01/2016, say—because that’s probably what the majority of spreadsheet users mean when they type it into a cell. Similarly, gene identifiers like “2310009E13” are converted to exponential numbers (2.31E+19). In both cases, the conversions strip out valuable information about the genes in question.

What on earth inspired all these researchers to use what can only be described as the greasy kid stuff of analysis and data storage for this purpose?

It’s nucking futz.

Charter School Fail

No Comments
In a bit of news that should surprise no one, it turns out that charter school students do slightly worse later in life than public school students:

Charter school boosters have many arguments in favor of fostering a publicly-financed, privately run parallel education system. But at the end of the day, their model should help kids learn more, perform better, get good jobs and earn a higher salary than they might have otherwise, right?

By that metric, it appears that Texas’ charter schools have failed, according to a large-scale study of kids from the K-12 system through early adulthood.

The analysis was conducted by Will Dobbie, an assistant professor at Princeton, and Roland Fryer, the Harvard economist who in recent years helped Houston ISD adopt charter school methods (you might also remember his name from research on Houston’s police-involved shootings). It uses data from Texas state agencies that tracks student achievement and demographics from primary school, through college, and on to the labor market.

Texas is the ideal laboratory for this kind of study. It introduced charter schools way back in 1995, and they now enroll 3.5 percent of the public school population. The schools have thus had time to refine their methods and work out some kinks, while their students have had time to test their mettle in the labor market.

< The findings: On average, charter schools have no meaningful effect on test scores or employment, and actually have a slight negative impact on earnings. The results are slightly better for so-called “no excuses” charters, which feature stricter discipline and extended instructional hours — they increased test scores and four-year college enrollment and had no effect on earnings. Regular charter schools boosted two-year college enrollment, but depressed test scores, four-year college enrollment, and earnings.

The idea behind charter schools has always been that unleashing the market on education will create amazing result.

It has been about as effective as the idea of “Unleashing Chiang” (Kai Shek) on the communists in mainland China was.

You can read the full study here.

How Many Times Does This Lie Have to Be Disproved?

No Comments

Time and time again, when arguing for outsourcing and skill based immigration programs like H1B and L1A programs, business argue that there are simply not enough skilled workers in the US.

Time and time again, these claims have proved to be complete fabrications:

For years, employers, pundits and policymakers alike have bemoaned the lack of qualified workers available to fill vacant manufacturing jobs in the U.S.

Despite the prominence of the skills-gap debate, a new paper co-written by a University of Illinois expert in labor economics and workforce policy finds that the demand for higher-level skills in U.S. manufacturing jobs is generally modest.

Three-quarters of U.S. manufacturing plants show no sign of hiring difficulties for open positions, says new research from Andrew Weaver, a professor of labor and employment relations at Illinois.

“Not a week goes by without someone declaring that a huge skills gap exists in the U.S. workforce,” he said. “A lot of ink has been spilled on this topic, but it’s frequently without evidence. The popular sentiment encourages people to think that employers have high skill demands, but U.S. workers just aren’t up to snuff, and that’s why manufacturing work is being outsourced overseas.”

However, the results show that U.S. manufacturers are generally able to hire the skilled workers they seek.

“We estimate an upper bound of job vacancies due to a potential skills gap of 16 to 25 percent of manufacturing establishments – a finding that sharply contrasts with other surveys that have reported figures of more than 60-70 percent,” Weaver said.

It’s not that business cannot find appropriately skilled employees, it’s that they don’t want to pay them a fair market wage, and so they try importing workers and exporting jobs.

Factoid of the Day

No Comments

I always knew the people who you know are the ones most likely to kill you.

I did not know this: (BTW, kickass discussion of statistics of incomplete data here)

Americans are afraid of many threats to their lives – serial killers, crazed gunmen, gang bangers, and above all terrorists – but these threats are surprisingly unlikely. Approximately three-quarters of all homicide victims in America are killed by someone they know. And the real threat from strangers is quite different from what most fear: one-third of all Americans killed by strangers are killed by police.

Our law enforcement system is broken.

OK, that bit I knew.

Economy Grew at Less Than ½ of Forecasts in Q2

No Comments

Forecasts were generally at a 2.5% annual rate, and the actual rate was 1.2%:

The U.S. economy expanded less than forecast in the second quarter after a weaker start to the year than previously estimated as companies slimmed down inventories and remained wary of investing amid shaky global demand.

 Gross domestic product rose at a 1.2 percent annualized rate after a 0.8 percent advance the prior quarter, Commerce Department figures showed Friday in Washington. The median forecast of economists surveyed by Bloomberg called for a 2.5 percent second-quarter increase.

 The report raises the risk to the outlook at a time Federal Reserve policy makers are looking for sustained improvement. While consumers were resilient last quarter, businesses were cautious — cutting back on investment and aggressively reducing stockpiles amid weak global markets, heightened uncertainty and the lingering drag from a stronger dollar.

So it’s 3 months until the general election, and the economy is slowing down.

Lovely.

Good Jobs Stats this MOnth

No Comments

287,000 new jobs in June:

Quashing worries that job growth is flagging, the government on Friday reported that employers increased payrolls by 287,000 in June, an arresting surge that could reframe the economic debate just weeks before Republicans and Democrats gather for their conventions.

The official unemployment rate did rise to 4.9 percent, from 4.7 percent, but that was largely because more Americans rejoined the work force. And average hourly earnings ticked up again, continuing a pattern of rising wages that brought the yearly gain to 2.6 percent.

“Wow, this one takes my breath away,” said Diane Swonk, an independent economist in Chicago.

An unexpectedly grim employment report in May combined with Britain’s vote to leave the European Union had fanned wider concerns that the American economy was in danger of stalling. During its meeting last month, the Federal Reserve unanimously decided to postpone increasing the benchmark interest rate.

It’s just one month, and part of that number is Verizon strikers returning to work, and it follows a horrible May.

About all we can say is that the Fed almost certainly won’t do anything with rates before the election.