Monday, August 31, 2015

Comparing labor market recoveries


 - New Deal democrat

With few exceptions, people don't get a job for social reasons.  They go to work each day in order to earn money to purchase necessities, discretionary goods, and to save for future needs.  In short, they work because of cold, hard cash.

So why is it that most economic writers appear to think the defining element of a labor market recovery after a recession is the number of jobs created? Isn't the better measure the amount of cold, hard cash it is delivering to workers?  That, dear reader, is measured by real aggregate wages, and that, I believe, is the best measure of labor market recoveries.

Let me give you a few examples.

First, compare an economy that creates 1 million 40 hour a week jobs at $10/hour, with an economy that creates 2 million jobs at 10 hours a week at $10/hour.  If we were to count by job creation, the second economy would be better.  But that's clearly  not the case.  The second economy is paying out only half of the cold hard cash to workers as the first.

Next, let's compare two economies that both create 1 million 40 hour a week jobs, but one pays $10/hour and the other pays $12/hour.  Clearly the second economy is better.  It is paying workers 20% more than the first.

Finally, let's compare two economies that create 1 million 40 hour a week jobs at $10/hour.  In the first economy, there are 3% annual raises, but inflation is rising 4%.  In the second, there are 2% annual raises, but inflation is rising 1%.  Again, even though the second economy is giving less raises, it is the better one -- those workers are seeing their lot improve in real, inflation-adjusted terms, whereas the workers in the first economy are actually losing ground.

In each case, the economy creating more jobs, or more hourly employment, is inferior to the economy  that pays more in real wages to its workers,  In other words, the best measure of a labor market recovery is that economy which doles out the biggest increase in real aggregate wages.

So let's compare the increase in real aggregate wages -- the total wages paid to all nonsupervisory workers, adjusted for inflation, from their bottom in each recession.  Since that was 5 years and 9 months ago for our current recovery, that will be our measuring stick.  This is calculated as follows: average hourly earnings for nonsupervisory workers, times average hours worked, times the number of jobs, and then divided by the consumer price index, with the result indexed to 100 at the bottom.  Here are the results:

Trough Peak #months Real wage
growth %
 Wage growth
per month %
Wage growth
at 69 months
1/64*  8/6967* 30.2 .45  30.0%
11/705/7430   18.5 .62 11.8%***
4/753/79 47 20.8 .4411.5%*** 
7/801/81 62.4 .40  9.3%***
11/8210/8983 21.6  .26  18.4% 
2/9211/00  10533.8 .32  21.7%
4/03 9/07 53 10.5 .208.0% 
10/097/15**   69** 16.3 .2416.3%

*start of series

**to date

***Measured through the next recession, as recovery was short-lived

The interactive FRED graph of real aggregate wages can be found here.

An interesting aside is that the "Reagan recovery" of the 1980s is either great or mediocre depending on where you measure.  In the first 14 months, real aggregate wages grew 7.9%, or a blistering .56%  a month! In the last 69  months, however,  they grew only 9.3%, or a miserable .13% a month.  Had I measured back from the end rather than forward from the beginning, the current labor market recovery would be stronger.

 We can immediately see the effect of labor bargaining power, as all of the economic expansions before the 1980s showed far faster real aggregate wage growth per month than any expansion since.   On a per-month basis, the current labor market recovery is only better than the George W. Bush recovery.  Over the total recovery, the current recovery is only better than the George W. Bush recovery, and the 6 month recovery at the end of Jimmy Carter's term.

But because it is longer lasting, however,  after 69 months the current labor market recovery  has a better record than the short-lived recoveries of the 1970s and the brief 1981 recovery as well.   By this  measure it only  lags the 1990s recovery as well as the 1960s, and, depending on whether you measure forward from the beginning  or back from the end of the 1980s recovery, the current recovery is slightly worse, or significantly better than that labor market recovery.  Should the current recovery last another 24 months, it will probably surpass the Reagan recovery by either measure.

The bottom line is that the record  of this labor market recovery is mixed: poor monthly growth due to nonexistent labor bargaining power; but good total growth based on its persistence.  

Friday, August 28, 2015

Corporate profits in Q2 set another record


 - by New Deal democrat

Since corporate profits are a long leading indicator for the economy, and stock prices a short leading indicator, that means ... I have an important post up at XE.com .

BTW, sorry about the light posting here this week.  I have been working on a bunch of things, but partly due to real life, and partly due to this week's exciting action at the dog track, it made sense to post more topical financial markets stuff over at XE.

Wednesday, August 26, 2015

A simple point about the Chinese stock market crash


 - by New Deal democrat

As I wrote yesterday, the only thing that has been added to the economic news this summer, that we didn't already have last fall, is the Chinese stock market crash.  And wherever that crash is going to bottom out, it is most of the way there.

Here is a graph of the Shanghai index for the last 25 years:



Notice that the recent market bubble and burst aren't as severe as the 2008 crash, in which the Shanghai composite lost about 70% of its value from peak to trough.  And then bounced back smartly.

Here's a close-up of 2015:



The Shanghai composite is already down 45%. Most of the damage has been done.

No inside information, but the big selloffs in the late afternoons the last two days have to odor of forced liquidation.  We saw a similar pattern in October 2008 after the September crash.  The forced liquidation reached a climax, and that was the bottom.

By the way, here is what American corporate insiders have been doing:



They've been buying, and that was before the big downturn in the last week.

My focus in this blog is on the economy, not on investing. And my focus on the economy is mainly about jobs and wages.  While I have a sneaking suspicion that one or more hedge funds will be carried out on their shield in the very near future, I just don't see this as having a big impact on the US economy at large, although job and wage *growth* will probably stall or decelerate.

Bottom line: whatever collateral damage a Chinese stock market crash might do to the US economy is almost certainly already priced in at this point.

Tuesday, August 25, 2015

China sneezes, and the US catches ... the sniffles


 - by Neew Deal democrat

I have a new post up at XE.com, entitled "The China Syndrome."

Saturday, August 22, 2015

Weekly Indicators for August 17 - 21 at XE.com


 - by New Deal democrat

My Weekly Indicator post is up at XE.com.

The US$ is "King of the World" at the moment. The resulting weakness in markets has brought mortgage rates back into positive territory for US consumers.

Friday, August 21, 2015

WE'RE DOOOOMED!!!! S&P goes negative YoY


 - by New Deal democrat

I have a special comment up at XE.com.

Given what has been happening with corporate profits, and the likely impact of China's currency devaluation, this really shouldn't be a surprise.

The good news is, if the Fed is wrong, it can reverse course. The bad news is...


 - by New Deal democrat

To find out, click on over to my new post up at XE.com.

See, I can write a clickbait headline just like Business Insider!

Thursday, August 20, 2015

The implications of the child care cost crush for median household income and "shadow unemployment"


 - by New Deal democrat

The other day I showed that there is compelling evidence that the primary reason for the long term decline in the Labor Force Participation Rate in the 25 - 54 age range is the increasing real cost of child care, coupled with stagnant to declining real wages in the lower paying jobs typically taken by the second earner in a two earner household.

Today I have a few more precise graphs, and discuss the implications for median household income and the issue of "shadow unemployment" or "missing workers."

First of all is a graph of the increase in the number of those aged 25-54 who are neither employed nor unemployed, but out of the labor force entirely:



Unfortunately this is not avaiable on FRED, but via the BLS, here are the number of people in that above group who tell the Census Bureau each month that they want a job now:

thie number of people aged 25-54 who told the Census Bureau that they were out of the labor force, but wanted a job now, peaked in 2011.

Subtracting the second number (second column below) from the first (first column below)  gives us the number of people aged 25-54 who tell the Census Bureau they *don't* want a job now (third column below):

1999: 17.5m   2.0m   15.5m
2011:  20.9m   3.0m   17.9m
2015:  22.4m   2.5m   19.5m

What I showed the other day is that the big reason for the increase in this last number is the cost of child care vs. wages, which over the long term has become increasingly onerous.  Another piece of supporting evidence comes from the below chart in the 2014 Pew Research Center's report on women in the job market:




The Pew study found that as of 2012 the vast majority --  85% -- of stay at home married mothers say the reason for not working is to take care of their children.  Note also that the number of women staying at home is much larger for age groups 25-54 than for those either young or older.  The only thing not making this the ultimate "smoking gun" supporting the child care cost crunch thesis is that the study makes no comparison with 1999.

Implications for real median household income

Like the LFPR, real median household income has declined across most age cohorts since 2000.  Here is a graph I have run a number of times, showing real median household income for various age groups (h/t Doug Short):



Aside from the unemployment rate, working age people dropping out of the labor force due to increasing child care costs is the most important reason for the failure of median household income to return to its level in 2000.  After all, giving up a second income necessarily means that the household's income is less than otherwise.

Implications for "shadow unemployment"

Finally, here is a graph of what the Economic Policy Institute calls "missing workers:"
  


The EPI defines "missing workers" as
potential workers who, because of weak job opportunities, are neither employed nor actively seeking a job. In other words, these are people who would be either working or looking  for work if job opportunities were significantly stronger. Because jobless workers are only counted as unemployed if they are actively seeking work, these “missing workers” are not reflected in the unemployment rate.
Note that this has been rising since 2007, unlike virtually all other metrics.

I have been critical of this measure, which is based on projections from a 2006 Regional Fed study. The Census Bureau's series "Not in Labor Force, Want a Job Now" measures exactly what the EPI claims to be calculating, and that peaked out about 4 years ago. To the point, parents who stay at home to raise their children and tell the Census Bureau month after month that they  quite simply don't want a job now are not "missing" from the unemployment rate.

But since increased wages, especially for lower quintile jobs, change the equation for families considering whether to have the second spouse work, i.e., increased real wages should result in more spouses opting to work, the Center's "missing workers" statistic, including its continued slow rise, makes perfect sense if we describe it not as those who have dropped out of the work force, but would return if the job market was strong, but rather as those who would return to work if wages were better.

The effects of the child care cost crush not only unravel the mystery of the decline in the prime working age Labor Force Participation Rate, but it is also the missing piece in the puzzle of the similar secular decline in real median household income, and the rise in "missing workers." Once we include these effects, almost all of the mysteries in current labor market statistics disappear.

Wednesday, August 19, 2015

July consumer prices: all intact trends continue


 - by New Deal democrat

This morning's CPI report for July was a big yawn, coming in at +0.1% just as expected, and YoY CPI also unchanged at +0.2%.  OMG, the Fed simply *must* raise rates as a precaution!!! < / snark >

So let's look at a couple of useful bits of information.

First, inflation continues to be confined almost exclusively to housing.  The below graph comparing CPI for shelter (blue) vs. everything else (red) continues to show housing inflation a little on the "hot" side, while everything else remains in the most severe deflation in the last 50 years outside of the Great Recession:



Next, we can now show real retail sales for July, which continue to show an improving trend, although they did not make a new high:



Finally, over the longer term YoY real retail sales tend to lead YoY employment.  The recent deceleration in growth of real retail sales is another reason to suspect that YoY job growth will fade at least a little bit, with reports under 225,000 or maybe even under 200,000 a month:



Basically, all of the existing trends are still intact.

Tuesday, August 18, 2015

July housing permits and starts: no cause for celebration or concern


 - by New Deal democrat

I have a new post up at XE.com analyzing this morning's housing permits and starts report.  We are neither Delivered nor Doomed.

Do rising child care costs & stagnant wages explain the declin e in the Labor Force Participation Rate ?


 - by New Deal democrat

What is driving the long term decline, shown in the graph below, in the labor force participation rate (LFPR) among prime working age adults, ages 25-54?  




A couple of weeks ago I wrote that it wasn't poor job prospects.  In that post I ran the following comparison.  First, I calculated the number of persons in age group both unemployed and also not in the labor force for age group 25-54.  That is the (red line in the graph below).  Then I calculated what the  number would be if we added together the unemployed in that age group, plus a proportionate share of  those Not in the  Labor Force, but Who Want a Job Now (blue):


If the primary explanation for the decrease in the employment-population ratio for ages 25-54 were the inability to find a job, the result of the two calculations should be the same.  In fact, the second calculation (in blue)  gives me a much lower value.   That tells us that the primary driver of the increase in persons not being employed in age group 25-54 is not poor job prospects. 

There is compelling evidence that  the main culprit is increased child care costs, paired with stagnant or declining real wages.
    
Let's start with a framework for the discussion.  Many families face a familiar choice between hiring a contractor vs. doing it yourself for a variety of tasks.  For example, do I mow my lawn myself, or do I let a landscaper do it? All else being equal, higher costs for hiring a contractor will cause some increase in the number of homeowners who decide to do the task themselves.  The same process would apply to other tasks, like hiring a painter or someone to install a new automatic garage door.

But since all of the above tasks can be done on nights or weekends, a choice to perform them doesn't mean the homeowner doesn't have to give up time from work to go the DIY route.

Child care stands alone.   A choice not to send children to day care inevitably means limiting at least one spouse to part-time work, or dropping out of the labor force entirely.  In other words, the cost of day care must be weighed against the income earned from a job.  This is a balance: the higher the cost of daycare, or the lower income earned from a job, the more we should expect a parent to choose part-time work or dropping out of the labor force entirely.

So, what has been happening to the costs of child care?  And what has been happening with real wages, especially in the lowest two quintiles of paying jobs, which a second income earner in a household is likely to be in (the primary earner by definition holding a higher paying job)?  There is strong circumstantial evidence that both the real cost of day care has risen, and real pay in the lowest two quintiles has been particularly hard hit.  In other words, the data supports a hypothesis that the secular decline in the LFPR among working age families has largely being driven by the issue of child care. 

To begin with, the real, inflation adjusted cost of child care has been soaring for 30 years with nary a break:


The Bloomberg article last week which featured that graph explained:
 There’s been a growing push for childcare workers to be better trained and educated, and the costs associated with those efforts make it tougher for managers to scrape up pay increases for existing personnel, said Anna Carter, president of the Chapel Hill, North Carolina-based Child Care Services Association.The mismatch in supply and demand has made childcare a “broken market,” said Marcy Whitebook, director and founder of the Center for the Study of Child Care Employment at the University of California at Berkeley.
"Mothers who do work are paying more than ever for child care. In  inflation-adjusted dollars, average weekly child care expenses for families with working mothers who paid for child care (24% of all such families) rose more than 70% from 1985 ($87) to 2011 ($148), according to research  by the Census Bureau. For those families, child-care expenses represent 7.2% of family income, compared to 6.3% in 1986 (the earliest year available).
 Meanwhile, in particular since the Great Recession, median pay for the lowest quintile of jobs declined more than any other category, into 2014, the latest data I could find, from the National Employment Law Project:


Last year, the average cost of child day care in the US was $11,666, according to  Babycenter.com.    Meanwhile, an earner at the top of the 5th quintile, working a full time job, is making about $19,000 a year; a worker at the top of the 4th quintile, about $25,000. 

That means a household with a second earner whose job pays among the lowest 20%, and well into the lowest 40%, might actually have a net financial gain by foregoing the second income to dispense with day care costs, and get the benefit of having a parent at home full time to run the household and be with the kids when they aren't in school.

While I have been unable to locate any data specifically quantifying the average or median incomes of primary vs. secondary earners in a two income hosusehold, a 2009  study by the BLS found that the median part time secondary earner makes 62.8% of part time primary earner  But if anything, that percentage is too high, since most two income households are probably either both full time workers, or a full time worker and a part time worker.

More to the point is a December 2013 paper by the Hamilton Project, which found that the median 1 earner family income was ~$40,000, while the median 2 earner family income was ~$60,000.  


By implication, the median part time earner only makes about 1/2 of primary earner, $20,000 vs. $40,000.  Even that disparity is probably underestimating the true number, since the primary earner by definition is earning more, and is probably earning more than the median of all workers. 

Remember that we are dealing with the *change* in the cost-benefit analysis that families make.  As a simple matter of supply and demand, that day care costs have far outrun inflation, and that real wages among the lower two quintiles of jobs have declined, and declilned by more than the upper quintiles, implies that more parents would make the choice to stay at home.

This hypothesis gains some support by correlating well with the fact that LFPR for those aged 55-54 did not decline during 2000 - 2007, but rose:


By this time, the kids are in college if not already out of the house.
An even more compelling correlation is  the fact that the decline in the LFPR for men (-4%) has outpaced the decline in the LFPR for women(-3%): 
   

Thirty years ago, men who might otherwise have wanted to stay at home to raise their kids faced intense societal pressure to work anyway.  Now, not so much.  According to the Washington Post last year:
 The number of stay-at-home Dads has doubled in the last 25 years, reaching a peak of 2.2 million in 2010, according to a new report by the Pew  Research Center. ....

The new Pew Research Center report found that in 1989, only 5 percent of the 1.1 million at-home fathers said they were home to be primary caregivers. That share has increased four-fold now to 21 percent, a sign not only of the power of economics in reshaping traditional family structures, but of shifting gender norms.
  
"The assumption that a lot of people make is that the number of stay-at-home dads went up because of the recession. And while that’s absolutely true, even if you take out that trend altogether, the fact is the number has been going up over time, regardless. And the biggest increase is in the share of fathers who want to stay home to take care of kids," said Gretchen Livingston, author of the new report. "That’s very striking." 

The Pew Research study findings are nearly a "smoking gun" in support of the thesis  that that  the increase in stay at home parents are the force driving the decrease in the prime working age LFPR.

Without quoting the Pew study at length, the numbers are very instructive.  Over the same time period, the total number of stay at home parents for all reasons increased by 1.5 million, with 700,000 of the increase being women.  Between 1989 and 2012, the total number of stay at home dads, for all reasons, grew by 808,000.  Of that 808,000,
  • 364,000 stayed at home to raise the kids
  • 295,000 due to inability to find work
  • 165,000 to go to school, being retired, or other reasons
  • 84,000 due to disability
In other words, even shortly after the end of the recession, 45% of the entire increase in stay at home dads since 1989 was in order to raise the children.  While by no means a perfect fit, that is still pretty close to a "smoking gun."

Update:  The BLS tables show that between 2001 - 2014, the number of people age 25 - 54 who are not in the labor force and did not want a job increased 3.3 million from 17.8 million to 21.1 million. No graphs, but the raw data can be found here.  While not directly comparable, Pew showed an increase of this number by 2.6 million from 1989 through 2012, the lion's share of which occurred after 2000.  That means the only missing piece of the puzzle is how much of the increase in women staying at home, about 600,000 from 1989-2012 as derived from the Pew data, is for child care reasons.

Update 2: I have now found an equivalent study by the Pew Research Center about stay at home moms.  The study states:
 
The share of mothers who do not work outside the home rose to 29% in 2012, up from a modern-era low of 23% in 1999, ... A growing share of stay-at-home mothers (6% in 2012, compared with 1% in 2000) say they are home with their children because they cannot find a job.

A little math shows that the number of stay at home moms for reasons *other* than inability to find a job has risen by 4.5% from 22.8% in 1999 to 27.3% in 2012.  The Pew study on women does not break down this remaining number by those staying home due to disability, retirement, education, or other reasons besides raising children.  
Of course this is not conclusive, and correlation is not necessarily causation.  In particular, I have no information as to how the real  median wage of second earners has changed over time.

To conclude, in short:
  1. the decision to be a stay at home jparent in a two earner family should rise when child care costs rise and/or lower quintile wages fall.
  2. child care costs have been rising inexorably for several decades
  3. wages for lower quintile jobs have been particularly hard hit since the Great Recession
  4. data from a study by the Pew Center show that 45% of the increase in the number of stay at home dads is in order to raise children
  5. although we do not have the equuivalent data for women, the cirucmstantial evidence is compelling that the i ncreased real costs of child care vs. wages are the primary explanation for the secular decline in the LFPR among working age families over the last 15 years.

Monday, August 17, 2015

Weekly Review Columns

I'm back from vacation.  Here are last week's reviews and this week's previews, all over at XE.com

http://community.xe.com/blog/xe-market-analysis/international-preview-aug-16-21

http://community.xe.com/blog/xe-market-analysis/us-equity-and-economic-review-august-10-14

http://community.xe.com/blog/xe-market-analysis/international-economic-week-review-aug-10-14

The labor market isn't tight, just employers' wallets


 - by New Deal democrat


Via Mark ThomaProf. Stephen Williamson writes a nice, thorough post on the state of the labor market, which is well worth reading in full.

One point on which I disagree with his analysis has to do with the Beveridge curve, i.e., the tradeoff between the unemployment rate and job openings, which he graphs here:


He writes, 
Early in the post-recession period, people were speculating as to whether the rightward shift in the Beveridge curve was due to cyclical factors (the Beveridge curve always shifts rightward in a recession) or some phenomenon related to mismatch in the labor market (the unemployed don't have skills that match well with the posted vacancies). Perhaps surprisingly, the Beveridge curve has not shifted back, with the end of the Great Recession now more than 6 years in the rearview mirror. That would seem to put the kabosh on cyclical explanations for the phenomenon. But it's not clear that mismatch fares any better in explaining the Beveridge curve shift. If that's the explanation, why doesn't the mismatch between the searchers and the searched-for go away?"
And he concludes his review in part thusly:
Employment growth is currently strong, and by most measures the labor market is currently somewhat tight to very tight. 
In other words, the labor market is tight because even with a U3 unemployment rate of 5.3%, job openings are soaring. Although he doesn't mention it, there are more job openings now than even during the height of the tech bubble at the beginning of 2000!

I disagree.  And the reason can be found with wages.  Below is a graph of job openings, minus new hires (i.e., the number of unfilled job openings) (blue) vs. YoY nominal wage growth (red):



By and large, the two series moved in tandem since the JOLTS series began in 2000, through 2010.  Since then, there have been three distinctive episodes:

1. During 2011-12, wage growth declined. Unfilled vacancies rose.
2. In 2013-early 2014, wage growth improved. Unfilled job opensings plateuaed.
3. Since mid-2014, wage growth has flagged.  Unfilled job openings have soraed   

 Note that at 2%, YoY wage increases are well below even the 2003-07 average of about 3.5%.

This is a market which is not clearing.  Prospective employers and employees are at a standoff. To put it less charitably, employers can't find workers to fill their job openings ... for the paltry wages they want to pay them.  With near record corporate profits, that isn't a tight labor market. It's tight-wads.

Saturday, August 15, 2015

Weekly Indicators for August 10 -14 at XE.com


 - by New Deal democrat

My weekly indicator post is up at XE.com.  Needless to say, the biggest issue has to do with the revaluation of the Yuan.

Friday, August 14, 2015

July industrial production: a solid start to Q3


 - by New Deal democrat

Like retail sales earlier this week, industrial production rose strongly, also mainly on the back of motor vehicles.

The overall index is still just shy of its all-time high set last autumn:



Below are the three main component groups - manufacturing, mining, and utilities:



Manufacturing is the largest sector, so the big increase there overwhelmed the decrease in utilities.  Mining, of course, has suffered as part of the Oil patch, but had a little rebound in July.

Industrial production is basically the King of Coincident Indicators.  The NBER historically has almost always dated the beginning and end of recessions as of the date industrial production turns. To put this in a longer term context, as shown in the below graph, the recent downturn wasn't enough to set off recession alarm bells:



Not a perfect report, but a solid one. The pessimists will point out how much the increase relies on autos, and seasonal issues with retooling, but since motor vehicles are the second most important sector after housing, the increase remains real and important.  This with retail sales earlier this week and the employment report last week have gotten the 3rd Quarter off to a good start.


I hereby volunteer


 - by New Deal democrat

So, no sooner do I tell you that there is no point to my reporting on daily economic statistics, because I am going to agree with Bill McBride, a/k/a Calculated Risk, 90% of the time, than the bastard goes and takes a 10-day vacation.

On top of that, Hale Stewart, the alleged proprietor of this here blog, also went on vacation and didn't even bother to tell me!  I think I am going to change the name of the blog and see how long it takes him to notice.

Well, since *somebody* has to stay behind and keep the internet turned on, I suppose I have no choice but to report for duty. Hopefully industrial production, housing permits, and CPI won't blow up the place while I am watching them. < / sigh >

Thursday, August 13, 2015

More evidence we are past the midpoint for jobs growth


 - by New Deal democrat

Yesterday's JOLTS report adds to the evidence that this expansion is past its midpoint and that the expansion of jobs is heading towards maturity.

In the last expansion, the only one for which JOLTS provides complete data, hiring stalled out for over a year while job openings continued to grow:



A close-up of the last year shows that for the last 9 months, the same thing has happened.



Similarly, layoffs and discharges bottomed out shortly after hiring stalled:



A close-up of the last year shows a bottom in layoffs and discharges last November, one months before the peak in hiring to date.




Further, as I noted on Tuesday, a weakening of the Labor Market Conditions Index tends to lead a decline in YoY jobs growth by about 6 - 12 months:



That is further evidence that we should expect less YoY growth in jobs for the remainder of this year as compared with 2014.

Finally, unsurprisingly housing permits lead jobs growth as well:



While a steep decline to a stall in housing, as happened in 2014, has not always led to a stall in jobs, usually it has led to at least some weakening, sometimes slight, sometimes very marked.  Since the lead time varies between 6 to 18 months, we are about due for last year's weakness in housing to lead to some weakness in payrolls.

What would lead to a positive reboot in jobs growth? A decline towards new lows in mortgage rates.  Failing that, we are in the latter stage of the jobs expansion.

Wednesday, August 12, 2015

Why there is no "second housing bubble"


 - by New Deal democrat

I have a new post up at XE.com, explaining why I reject the idea that housing has entered a second "bubble."

Tuesday, August 11, 2015

The Labor Market Conditions Index as a leading indicator


 - by New Deal democrat

Via Naked Capitalism, Bill Mitchell writes that the Fed's "Labor Market Conditions Index is Weakening."  As usual, I hate the present progressive tense, since there is nothing about the index that forecasts what it itself will do over the coming months. Further, since it is still positive, labor market conditions are still above trend - just much less so than last year.  But the Index itself turns out to be a useful addition to the forecasting toolbox.

The LMCI is based on 19 components.  Click on the link above if you want further information on its makeup.  What is important to me is that it has a 40 year history of signaling a turn in the economy.

Here is the entire history of the index:



Here's what the data behind the Index shows: (1) the index has with one exception (1981's double-dip) always failed to make a new high for at least 12 months before the next recession, sometimes much longer than that; and (2) the index has always dropped below 0 and stayed negative for 6 months or somewhat more before the onset of the next recession.

Here is the Index over the last 5 years:



The Index made its cycle high at the beginning of 2012.  It made a secondary high in early 2014. But it has not turned consistently negative.

The Index become more valuable when it is teamed with two other measures: the YoY% growth in nonfarm payrolls, and interest rates.

As shown in the graph below, the LMCI consistently leads the YoY% growth in jobs by 6 - 12 months, but YoY job growth (red) is a much smoother measure:



Thus job growth serves as an important confirmation for the long leading part of the Index.

But a downturn below 0 in the LCMI, even partnered with a downturn in the growth of YoY jobs, hasn't always meant recession, as in the 1980s and 1990s.  That's where interest rates come into play.  In the 1980s and 1990s, as shown in the graph below (green), a marked increase in interest rates led to the declines in both the LMCI and YoY job growth.  



When that upturn in interest rates abated, the LMCI, and jobs, came back.  It was only when both weakened to the point where the LMCI was consistently negative, before interest rates declined, that a recession ensured.

Here is the same information for the period 2000 - present:



While the "taper tantrum" in interest rates briefly put a dent in the LMCI and YoY job growth, interest rates have calmed down since. In other words, more confirmation that we are probably past mid-cycle, but no imminent threat of an actual downturn.

On the negative side, since the LMCI does lead the much smoother YoY growth in jobs, it strongly suggests that YoY payroll growth is going to decline over the next 6 months or so.  And that can only happen if those payroll numbers generally come in under 225,000, and probably even below 200,000 through next winter.

All in all, the LMCI is a useful addition to the forecasting toolbox.

Monday, August 10, 2015

Forecasting the 2016 election economy: what are the best metrics?


- by New Deal democrat

How well do economic conditions predict the outcome of Presidential elections?  Can we forecast the most important of those conditions in advance? If so, what can we say even now about 2016?

That's what I am examining in this series, Forecasting the 2016 Election Economy.

Fortunately, I don't have to start from scratch.  As I noted in my inaugural post, most significantly, in 2011, Nate Silver undertook an examination of 43 economic variables, and their potency in predicting the election outcomes, going back all the way to the 1950s.  Additionally, James Surowiecki has suggested that growth in real personal income in Q2 of the election year is the most determinative economic fact.  Also, Professor of Economics Douglas Hibbs has used what he calls:
... the Bread and Peace model to explain presidential voting outcomes. The model claims that just two fundamental variables systematically affected post-war aggregate votes for president: (1) weighted-average growth of per capita real disposable personal income over the term, and (2) cumulative US military fatalities due to unprovoked, hostile deployments of American armed forces in foreign wars.
I'll look at that model, which has a generally good record, but missed badly in 2000 and 2012, in a later post.

In this post, I still start by looking at 5 of the variables identified by Nate Silver as being the most dispositive.  To cut to the chase, in order of their predictive importance, here's the top 20 variables he found most predictive of election results, in order of their value:



Below, I have taken those 5 indicators, and charted them all the way back to the 1952 election, or the start of the series when applicable, as to their YoY change (q/q change for ISM) during Q2 and Q3 of each election year.  All of these are adjusted for inflation when relevant:

Y earISM
Mfg
 Change
Nonfarm
payrolls
Change
 unemplymt
rate
Real
personal
income
Change
Emp/Pop
ratio
19561.0/0.43.9/2.5 -0.2/0.05.6/4.11.2/0.5
1960-5.9/-4.41.9/1.4 0.1/0.22.9/3.20.1/0.1
19649.3/13.22.6/3.0-0.5/-0.55.8/6.46.1/5.5
19685.6/2.93.2/3.4-0.2/-0.35.1/5.50.8/0.3
19729.8/12.33.2/4.3-0.2/-0.45.1/6.30.4/0.6
19769.2/4.73.6/3.2-1.3/-0.84.5/4.31.6/1.2
1980-17.6/-6.50.8/-0.31.6/1.8-0.7/-0.8-0.1/-0.8
19849.2/3.05.0/4.9-2.7/-2.07.7/8.32.7/1.9
19886.9/6.23.2/3.2-0.8/-0.54.5/0.50.8/0.7
19924.0/2.30.3/0.60.6/0.73.2/3.3-0.4/-0.2
19960.7/0.82.0/2.2-0.2/-0.44.3/4.40.3/0.7
20003.1/0.72.5/2.1-0.4/-0.26.1/6.50.2/-0.1
200410.8/8.61.1/1.5-0.5/-0.70.2/3.50.0/0/4
2008-0.9/-13.8-0.1/-0.70.8/1.3-0.3/-1.2-0.5/-0.9
20123.0/0.91.7/1.6-0.9/-1.03.7/3.80.6/0.7

For reasons of space limitation, for now I omitted the series for real GDP, real final sales, real aggregate payroll growth, and real disposable personal income per capita.

 The  top 5 variables identified by Silver mostly  give me the same result: if they were consistent, Humphrey  would have won in 1968, and Ford in 1976. Both Humphrey and Ford had objectively great economic numbers to run on.   Further, while  poor economic performace as in 1980 and 2008 are a death sentence for the incumbent party, mediocre numbers are less dispositive. If they were, Clinton in 1992, Bush in 2004, and Obama in 2012, all would have lost.

That being said, the single variable with the best record is the change in the unemployment rate.  It has correctly forceast 13 of the 15 elections since 1956. If the unemployment rate is rising in Q2 and Q3 of the election year, the incumbent party is going to lose.  If it is declining, with two exceptions, (Humphrey in 1968, Gore in 2000) the incumbent party is going to win.

Meanwhile, Surowiecki's measure of real personal income calls for Clinton and Obama to lose their re-election bids, and also misses 1968 and 1976.

Note that when a  trend has markedly accelerating or decelerating during the election year, the voters have reacted to that.  In 2000, it was clear by the end of Q3 that the economy was decelerating markedly.  The reverse was true in 2004.
  
Looking at Silver's most valuable indicators, in general good growth will lead to re-election of the incomumbent, and usually re-election of the incombent party. A downturn, unsurprisingly, results in the incoumbent or their party being ousted. Mediocre growth leads to more incomsistent results, although the change in the unemployment rate stands out as having the most promise.

That being said,  none of even the best economic variables have an infallible record. The economiy is not always the dominant issue in an election.  For lack of a better word, the economy can be trumped by "moral" issues -- winning or losing a war, a big increase or decrease in crime, Watergate and the Nixon pardon, The Most Expensive Blow Job in History, and even in 2004 the spectre of gay marriage.

In my next post, I will look at one of the long leading economic indicators, and what it can already tell us about employment and unemployment next year.

Sunday, August 9, 2015

Value added


 - by New Deal democrat

Why should you read me?

After all, there are about a bazillion economic commentators out there, a significant number of whom feel obligated to give their take on the latest statistics.  As far as I am concerned, for reasoned commentary on daily statistics, 90% of the time you could read Bill McBride a/k/a Calculated Risk, and Doug Short (maybe the most informative graphs in the entire econoblogosphere), and just stop there.  There's no point in my feeling the need to add my $.02.  

So what is it that I bring to the table?  Fitting the data into the business cycle.

I know of nobody else in the econoblogosphere who so relentlessly parses the economic data into long and short leading indicators, coincident indicators, short and long lagging indicators, and midcycle indicators -- and backs up the categorizations with as much historical data as possible, sometimes going back an entire century.

So when I read somewhere that, e.g., employment will drive consumer spending, or housing, I already know that the historical data shows exactly the reverse: housing drives employment with a 12 month or more lag; consumer spending drives employment with a much shorter lag.  I know that, over the longer term, corporate profits lead stock prices, not the other way around. I know that bank lending only turns up once an expansion has started.  I know that the trend in hiring, whether up or down, slows and then turns first in comparison with the trend in firing. And so on.

In short, I see myself as the ECRI (Economic Cycle Research Institute) for those who don't have the $64,000 (last I heard) annual subscription fee to spare.

And I am particularly focused on how that data will play out in the lives of ordinary workers. A modern industrial democratic society ought to be improving the lot of the vast majority of its people.  If it's not. something is seriously amiss.

A side effect of my focus is that unlike most observers, I do not fall prey to the fallacy of projecting the present trend into the future.  In November 2006 I forecast the last recession being 1 year ahead.  In January 2009 I forecast that the bottom of the Great Recession was likely to be that summer.  For the last 6 years, with various amounts of waxing and waning, I have been bullish on the economy, and relentlessly eviscerating Doomers.  I still am bullish.

Now more and more of the midcycle indicators have appeared to hit their inflection points.  That just means the numbers underlying the expansion will remain positive but in general be gradually less so as time goes on. 

Just remember, when the time comes - and it hasn't yet and may not for quite a while - that I yellow-flag the next recession, all of the coincident numbers like industrial production and employment will still be quite positive.  You will think I have gone over to the Dark Side.  I won't.  I will simply be "skating to where the puck will be," in the words of hockey great Wayne Gretsky.  That's why you read me.

Saturday, August 8, 2015

Weekly Indicators for August 3 -7 at XE.com


 - by New Deal democrat

My Weekly Indicator post is up at XE.com.  Last week's poor tax withholding report was as anticipated an aberration, but most coincident indicators remain weak.

On a broader note, I am increasingly convinced that we are past the midpoint of this expansion - probably in the 6th inning, to use a baseball analogy.  But that doesn't mean that I have turned negative, or that a recession is near.  Basically it just means that we are getting less "bang for the buck" from the fundamental trends that have underpinned the expansion.

Friday, August 7, 2015

Demographic pressure is driving demand in the US housing market


 - by New Deal democrat

My second article on the US housing market, explaining how real demand is driving prices, is up at XE.com.

The Pied Piper of Doom and The Automatic Earth: commodity collapse edition

- by New Deal democrat

Doom sells.  That's the simple fact.  Whatever you may think of their substantive record, you have to ruefully appreciate Zero Hedge's business model: if Doom sells, then the way to really monetize economic commentary is to aggregate Doom!  And boy howdy has it ever worked. For them.

I mention this because there is yet another Doomgasm going on in the place where I started bloggging, this time about the collapse in commodity prices.  So why be bothered?

Because here's a few of the comments by the financially unsophisticated readers, particularly after the Pied Piper of Doom touted the blog "The Automatic Earth:"

"Articles like this make me feel like deer in the headlights.
"Really, if true, how the hell is one expected "to prepare" for that scenario?"
---
"Yes, please.  I am reading the Ilargi piece."
--- 
"Reading the comments to this article, I gather one is to liquidate all assets in banks and brokerages...then, what? Keep them under the mattress?"

Causing this kind of unfounded panic makes my blood boil (although, to his credit, the author of the article is not calling for an economic crash, despite the Apocalyptic title he chose for his piece).

So let's cut to the chase.  What is the meaning of the commodity collapse"

1. sucks to be China
2. really sucks to be a supplier of China
3. Teh Awesome to be a consumer of products.

Why?  Prices, supply and demand for industrial commodities are set globally, not locally. A comodity price decline usual ly happens because manufacturers are using less commodities, i.e., there is a  manufacturing recession. Commodiity prices go down more than produceer prices, and producer prices go down more, or increase less than consumer prices.

Here's the best example I have read.  In the late 1800s, one of the most important things happening in the US was the building of railroads.  Each year new track would be laid down.  So suppose in a given year railroads lay down 4,000 miles of track.  Then the next year they lay down 2,000 miles of new track.  Even though there is now a total of 6,000 miles of new track, the suppliers of track saw their sales fall by 50%. Ouch! But rail passengers benefitted from the rail lines.

Ironically, the best evidence indicating that consumer countries have little to fear from a commodity collapse comes from ECRI, which sevcral years ago published a presentation on "Yo Yo economies", in which they wrote:
The rising export dependence of these [suppliers of suppliers] economies, with growing involvement in global supply networks, makes it increasingly difficult for economies to decouple, especially for suppliers of early-stage goods that have embedded themselves further up the supply chain and farther away from the final consumer. This makes them highly vulnerable to the Bullwhip Effect and at the mercy of cyclical fluctuations in end-user demand growth.
[my emphasis]

It is necessarily true that if supplier economies are especially vulnerable to cyclical fluctuations, then economies which are primarily consumers of end stage goods, like the US, are the least vulnerable. If supplier economies are especially unable to decouple, then it followers that consumer economies are the  most likely to be able to approach decoupling.

Elsewhere in their presentation they wrote:
[D]eveloping economies are very much subject to the Bullwhip Effect, where small fluctuations in consumer demand growth get amplified up the supply chain into big swings in demand as we move away from the consumer. So, smaller shifts in end consumer demand growth translate into larger fluctuations in intermediate goods demand, and even bigger ones in input material demand, and especially, raw material prices.

Even a modest decline in consumer spending growth in developed economies like the U.S. and Europe can help trigger a significant downdraft in the level of demand from suppliers and, in turn, a serious downturn in the level of demand for “suppliers to  suppliers.” 
In other words, an absolute contraction in supplier countries can be caused by simply continued growth, but at a slower rate, in a consumer country. Which means that converse is also true: an observed contraction in supplier economies (like China) does not necessarily mean that there is a contraction in consumer economies (like the US). Rather, consumer economies may simply continue to grow, just at a slower pace.

As if that weren't clear enough, ECRI supplied this very helpful graph:



Notice that the arrow at the top for consumer countries in ECRI's diagram is still pointing UP.

Think of the middle line as China, the bottom line Australia and Canada, and the top line as the US. In fact, the situation is probably better than that for the US, because the premise of the graph is that there is a consumer slowdown. ECRI then calculates the effect on suppliers.  But here, the consumer slowdown is taking place in China. So there is no reason to think that other consumming countries, like the US, will slow down.

But to turn back to the article in question, as to which the Pied Piper of Doom  touts a recent piece on the blog "The Autormatic Earth."  What were the y saying back in the Great Recession?  Well, here's what they though was goiong to happen as of December 12, 2008:
Last month I started warning of millions of lay-offs to come in the US economy. Since then, we've seen a November job loss of 533.000, as well as predictions of pink slip totals of a million and more every single month in 2009. ...  
[S]upermarket shelves will be as empty as electronics stores .... 
Banking and investing as we've known it won't be back for at least decades. The upcoming round of bank failures, which will leave very few, if any, standing, cannot be prevented. 
Umm, no.

The stock market bottomed in March 2009.  What were they saying then?

On March 2 they wrote
The  US Government throws 30 billion more pearls before the swine at AIG.... [note: the US actually made a profit on its AIG investment, which was paid back in full]
The downturn is nowhere near over, or at a bottom, or anything like that.  It will not be a straight line down, but the trendline is certain.  Many pundits now claims that a Dow at 5000 is some sort of bottom, but don't count on it.  As for home prices, they will keep tumbling and lose at least 80% of their peak values.
[note: house  prices bottomed at the beginning of 2012, about 30% under peak values].  
On March 4 they wrote:
Almost 700,000 Americans lost their jobs in February, which is 100,000 more than in January.  The revised versions of those numbers will be between 10% and 20% higher.  .... So I assumed that for the rest of 2009 pink slips will rise by 100,000 every month compared with the month before.  That is, for the sake of scribblings I pretend that the economy will not deteriorate exponentially. I know that takes a lot of pretending, but since the results are bad enough even when I pretend, I'll stick with it.
The total number of jobs lost would be about 13.7 million on December 31. 

Peak job losses were recorded at -824,000 in April.  While January was revised higher, February wasn't. Job losses didn't increase 100,000 per month, let alone exponentially, on average they *decreased* 100,000 per month after that. So they were only off by about 10,000,000 jobs in their "conservative" estimate.

And on March 9, 2009, the exact day of the stock market bottom, they wrote
Warren Buffet says that stocks are the best investment ... It won't be for all of the millions of investors who take his advice. Why would anyone listen to [Buffet], ... it's beyond me.
And the two author of the piece today and the Pied  Piper of Doom, who is touting The Autormatic Earth ?  Here's their financial insight,  only  two months after the stock market  bottom, on  May 12, 2009,:

Pied Piper of Doom: "Shifted to 75% Cash/Bonds Friday!"
The author of the piece about the commodity crash today: "The rally is done. .... If you are buying now you are simply giving your money away. Now is the time to go to cash. Maybe in a few months things might be different."

Of course, instead the stock indexes have more than doubled since then.  

So anybody who takes advice from them about what to do with their money is playing with fire, to say the least. Caveat reader.