Monday, December 30, 2013

Abundance Requires Real-Time Knowledge of Scarcity

Most politics consists of factions talking past one another in a process of loyalty signaling.  Obamacare is no exception.

Supporters have defended the law by describing individuals who have qualified for health insurance through the exchanges at lower prices than they had previously paid.  This defense is effective within the progressive paradigm, which is built around the goal of transferring resources to affiliates by taking it from non-affiliates, via the state, with the goal of imposing justice.

Please allow me to talk past those defenders in support of a different paradigm.

Obamacare as an expression of progressivism seems especially vulgar to me.  The law is such a Rube Goldberg device for the delivery of health care, there is no way of knowing what groups of citizens end up paying and who will benefit.  In that way, I think it fails, even as a progressive policy.  As a first order effect, it does appear that fines and penalties are imposed on businesses, especially on larger businesses, and subsidies are paid to families, pro-rated based on income, and this does reflect a typical line-up of progressive affiliations.  But even before we leave the first order effects, there appear to be arbitrary transfers from young people to older people, and many instances where people across the socio-economic spectrum are losing coverage based on some arbitrary combination of the law's requirements.

Beyond the first order outcomes, there appear to be penalties imposed on employers for hiring the poor and incentives for employers to drop or change coverage in any number of arbitrary ways.  The only things we can say for certain about the law at this point are that (1) it creates a non-transparent method for state-direction in the disposition of health care and (2) it obfuscates the cost of that disposition.

Abundance is not a naturally occurring outcome.  The natural human condition is one of poverty or of temporary abundance quickly nullified by Malthusian reality.  We only escaped that inevitability when some miraculous combination of cultural and political developments created an economic context roughly described as:
From each according to his need, to each according to his ability.
This means that the measures of need and ability are distilled through the individual; disposition and costs are personal and transparent.  This is the only sustainable source of abundance.  Obviously charity complements this ethic, by fulfilling needs left unmet.  But, the engine of abundance requires charity work to be parallel to this ethic - not in opposition to it.  (The corollary - From each according to his ability, to each according to his need - describes both (1) as-yet-unattained utopia and (2) slavery.)

So, if we want to rid ourselves of abundance, a very effective first step would be to obfuscate sources of scarcity and hide costs.  Public policy has been engaged in this for the better part of a century, most notably in education and health care, possibly never more so than in Obamacare (excepting the NIRA).  And, consumption of these services is understandably enveloping more and more of our productive capacity.

When I hear the above defense of Obamacare, I hear someone saying that what the law has accomplished so far is the destruction of information about scarcity.  It is a fetishization of egalitarianism and of healthcare.  The vast amount of that destruction could be avoided by simply handing out cash.  The problem is that, in many cases, if households received $10,000 in cash instead of a health insurance subsidy, they would choose to spend it on something other than health insurance.  That would be informational, but we have a hear no evil, see no evil policy.  We don't want information.  We want our political fetishes to be stroked.

A national income would seem to be such an improvement over this mess.  If every household started with a credit of some $10,000's - whatever we deemed appropriate as a minimum income - and then paid a flat income tax rate from the first dollar earned, then we could stop beating abundance down with all of these obfuscations of scarcity information.

This policy wouldn't end politics.  The optimal policy would leave us with some deserving and unlucky citizens who need more and some undeserving parasitical citizens who will be taking advantage of the system - in fact, we'd all probably find a little bit of ourselves in both columns.  We would be able to fight over the weight we should assign to those individual outcomes in managing transfers and determining the appropriate level of generosity.  And we would all want to express our political power over recipients by judging the ways they spend or don't spend their proceeds, so we will always be tempted to re-impose policies that undermine the individual liberty that is necessary for information about scarcity to be disseminated.  That temptation is so strong a part of human nature, I don't see how we will ever escape it.

Therefore, I'm not completely confident in the success of a national income policy.  Reforms in the 1990's replaced some former welfare policies with policies that lowered the net de facto tax rate on poor workers and removed some of the debilitating disincentives to work.  These were unalloyed improvements in the system of social support.  This led to a situation where more adult workers are taking low-paying jobs, because they can take those jobs without giving up all of their income supplements.  That was the point of the reform.  But, now, some people point to these workers and interpret the situation as corporate welfare, as if the employers are pocketing the public subsidies.

The progressive foundational tendency to view the population as collections of people with different levels of power which we need to help equalize leads to a view of prices resulting from a power imbalance between negotiating parties.  A worldview that treats corporations in general as monopolists naturally discounts the informational value of prices.  This is the foundation for support of minimum wage increases.  A model that denies any employment loss from the minimum wage is a model that attributes wage levels to powerful monopsonist employers.  I am afraid that there will always be a political force undermining the important informational function of prices.

The problem is that lack of bargaining power is a product of lack of productivity, so the empirical distinction that leads people off-track is a subtle one - too subtle for a counterargument to be persuasive.  If Malcolm Gladwell and I both walked into a publisher's office, he would be able to negotiate a much more lucrative contract than I would.  He would be a lot more powerful than I would be in that negotiation.  He would have power because he knows how to write books that a lot of people will buy.  His negotiating power is information.  It tells us something about the value publishers expect his books to create.  If we collect anecdotes about evidence of undeserved or abused power, and we use those experiences as a reason to disarm all negotiating power, we are destroying information, and thus, abundance.

I would note, in addition, that the sources of power imbalances beyond informational signals of value are much, much smaller than they are made out to be.  Take Wal-Mart as a typical example of the large, powerful corporation that takes advantage of a power imbalance to hire workers at low wages (so the story goes).  We have a similar imbalance of power as Wal-Mart customers.  Do you feel overwhelmed by the negotiating power of Wal-Mart when you shop there?  Are you beaten down and forced to pay above market rates for diapers and paper towels because of the power imbalance?  On the contrary.  In fact, if you pick up a $12 package of paper towels at Wal-Mart and tell the check-out clerk that you saw them at KMart for $9, the clerk will most likely deduct $3 without hesitation.  Power imbalances are much less important in the marketplace than what they are given credit for.

PS.  Paul Krugman gives a good example of how this paradigm, which begins with a virtuous intuition to take the side of the least powerful, can so quickly become bigotry:
Now, you may believe that employment is a market relationship like any other — there’s a buyer and a seller, and it’s just a matter of mutual consent. You may also believe in Santa Claus. The truth is that employment is, in many though not all cases, a power relationship. In good economic times, or where workers’ position is protected by legal restraints and/or strong unions, that relationship may be relatively symmetric. In times like these, it’s hugely asymmetric: employers and employees alike know that workers are easy to replace, lost jobs very hard to replace.
And may I suggest that employers, although they’ll never say so in public, like this situation? That is, there’s a significant upside to them from the still-weak economy.
I've mentioned before how bigoted thinking can very innocently arise from the innocent human quest for ideals (especially when our worldview feeds on a selective denial of scarcity).  Advocacy lends itself to us-vs.-them thinking, and it is very hard to maintain advocacy in large groups without accruing some bigoted tendencies.

Take Dr. Krugman's comment above.  Now, subtly switch the identities of his narrative.  Imagine it's 2009, you need your house painted, and house painters are dying for work.  You can really feel around for a low ball price from a good painter.  So, in this version of the story, you are the "employer" and small businesspeople are the "workers" in a poor economy.  Everything in that narrative is still basically true.

So, as a general proposition, in the market for home maintenance, is it not a matter of mutual consent?  Is it basically a power relationship?  Do legal restraints on your ability to freely contract with home painters help the home painter?  Do they help you?  Should home painters belong to a cartel that limits your choice of provider and price?  Home painters are always very easy to replace.  They clearly need you more than you need them.  And, were you delighted that the economy was in the stinker, so you could get such a great deal?

The truth is, in that specific case, you would have been happy to get such a service at such basement level prices.  But, there's no doubt that you would have traded that cheap service for a more prosperous community in a heartbeat.

When we set up paradigms that lock certain parties into the role of "other", we withdraw an important part of the process of engaging and judging, and we become opponents of truth-seeking, in, frankly, a rather ugly way.  We make a mistake if we believe that this problem is limited to contexts where the other is a race, or an ethnicity, or a sexual identity.

Wednesday, December 25, 2013

What Quits Are Telling Us

Demographic Adjustments to Quits

It occurred to me that another area where demographics (the baby boomers) would be skewing a statistical trend is the JOLTS data.  Older workers have longer durations of unemployment and lower unemployment rates, which suggests that older workers tend to have much lower employment churn than younger workers.  This should mean that, all else equal, current quit rates should be lower than we might have seen in earlier time periods.

JOLTS data doesn't include age information, but by using age-specific unemployment and unemployment duration data, I have been able to infer the quit rates of each age group.  In order to make this inference, I need to assume that the distribution of unemployment durations and the proportion of separations that are quits are fairly uniform among age groups.  These assumptions appear to be reasonable.  Here is a comparison of actual quits to the estimated quits that my age-specific model produces:

Now that I have an age-specific model, I can adjust the quits data for changes in age distribution.  This graph compares actual quits to the quits we might be seeing if age demographics had remained as they were in December 2000, when JOLTS data started:

This is not quite as stark a difference as I had expected.  But, demographics is creating a sizable downward shift in the quit rate of a little more than 200,000 quits per month, or about 0.15%, compared to what we would likely see with a younger labor force.

Whereas the previous graph showed Quits with labor force age distribution held equal to December 2000, this graph shows Quits with everything else held equal and only labor force age distribution changing.  This gives us a kind of full-employment demographic trend line for Quits.  Again, there is a small drop in expected quits, though not as much as we see from demographic effects on the labor force participation rate.

Quits Compared to the Unemployment Rate

So, demographics are probably skewing labor churn down a little bit, but the basic narrative from quits isn't overwhelmingly changed by it.  In either case, we are roughly back to the nadir of the quits rate in the previous recession.  We do need to keep in mind, though, that the previous recession was pretty shallow in terms of labor disruptions.  Unemployment topped out at 6.3%.  (I've reviewed in many previous posts why a declining labor force does not negate this fact.)  So, while quits look weak in absolute terms, they are quite strong in relation to the unemployment rate.  Quits have recovered to the same level they were in 2003 when the unemployment rate was only 6%.

As I have discussed before, recent minimum wage hikes and Emergency Unemployment Insurance (EUI) may have inflated the current unemployment rate, currently by around 1 1/2%.  That would leave an unemployment rate of 5.5% related to non-MW and non-EUI factors.  If we think about the effect of these legislative policies on quits, this could explain an unexpectedly strong quits rate.  Low-wage workers who are legislatively priced out of labor markets and workers with extra incentives for extending their duration of unemployment would provide less competition to quitting workers.  So, if my estimates are somewhat accurate, we would expect that last 1 1/2% worth of unemployed workers to exert less downward pressure on the quits rate, although they would still exert some.

All in all, I would say that the quits data makes me more confident about my estimates regarding the causes and level of current unemployment, and their effect on the labor market.

Here is a scatter graph of the quits rate over time compared to the unemployment rate:

The quits rate compared to the unemployment rate (in blue) looks to me like it has been following a pretty stable relationship.  If I adjust quits for demographic effects (the red line), quits look especially strong coming out of the deepest point of the recent recession.  If I adjust quits for demographics and also adjust unemployment for the distortions caused by MW and EUI (the purple line), then the relationship is similar to the relationship that we see without any adjustments.  This could be a coincidence, or it could mean that my estimates of structural distortions of Quits and Unemployment are both fairly accurate, or it could mean that my estimates are both similarly inaccurate.

In all three cases, I don't see any evidence of an especially weak quits rate.  In the time series above, the trend in growth rate of quits does appear to be lower in the current recession than it had been in the previous recession, but I suspect that this has to do with the inverse relationship between quits rates and the unemployment rate.  This will only be answered with new data as the unemployment rate drops further.  Quits might actually accelerate in 2014, which would be another tailwind for markets next year.  Regardless of whether we use the raw labor numbers or my adjusted ones, the Quits rate looks pretty normal, even a little strong, compared to the trend that would smoothly lead Quits back to normal recovery levels.

This may be a case where the demographic adjustment is important.  The unadjusted quits data may not look as strong as quits in the previous recovery, due to the older labor force.  This may be one of many indicators where a bullish cyclical signal will be retarded by demographic effects.

Forgetting Half of Supply and Demand

I think this is another context where analysts ignore the role and the power of labor supply as an influence on the labor market.  There are two factors pressing downward on the Quits rate.

One is a dearth of labor demand, signaled through an elevated unemployment level.  This is the influence typically cited.  But, this is only half the story.

The other influence is the aging of the labor force.  Here, it is the relatively powerful position of labor suppliers that is pushing quits down.  Older workers have less churn because they have lower unemployment rates and longer unemployment durations.  This is the case because they have more control over their work, more seniority, more income and wealth, less to gain from a career shift, and more frictions in job searching because they tend to have more specialized and senior positions.

The demographic influence isn't huge, in absolute numbers, but if we compare where quits are now, at about 2350 per month, to where we would expect them to be as we neared full employment, around 2800 per month, the supply factor and the demand factor each seem to be contributing about 50% of the difference.

This is one more case where the tendency to view labor markets as if laborers are powerless causes commentators to come to easy conclusions about these issues too quickly.

Monday, December 23, 2013

A couple more thoughts on Emergency Unemployment Insurance

One wrinkle in the changes in North Carolina labor statistics in the period since they cut Emergency Unemployment Insurance (EUI) is that more North Carolinians have left the unemployment rolls than were even enrolled in EUI.  Since June, more than 75,000 North Carolinians have exited unemployment, but there were probably only about 30,000 to 35,000 receiving EUI.

This could be because there are many other factors affecting unemployment, so EUI may account for less than half of the net changes.

The End of EUI as Fiscal Stimulus

But, this gets me thinking about the general tenor of discussions about this issue, and one overlooked factor that, to me, seems too obvious to leave aside.  There are many pundits who describe EUI as a fiscally stimulative policy.  But, especially to the extent that we continue to see a large number of long term unemployed workers, and also especially considering we are 5 years past the shock in real national production, any stimulative effect is likely to be nominal (that is to say, inflationary).

So, if we start from a context where a worker is unemployed and is not receiving benefits, changing to a context where he remains unemployed, but receives benefits, nominal GDP might be increased (assuming a stable monetary policy), but real production at this point is arguably not significantly changed.

But, if at this juncture, a cancellation of benefits leads a portion of EUI recipients to exit unemployment and become employed more quickly than they would have otherwise, then this clearly is a real fiscal stimulus.  This will lead to an increase in real GDP.

Now, the overlooked issue is that an increase in employment of this type will have complementary effects.  There will be jobs that are complementary to the jobs which are now being filled more quickly, so that there will be some "multiplier" effect from these former EUI workers.  And, the new real increase in production should also lead to other complementary increases in economic activity.  It is possible that the increase in employment due to the end of EUI will be larger than the actual number of EUI recipients.

Evan Soltas linked to this paper that outlined a logical argument for why moderate unemployment insurance might increase production in the long run because it helps facilitate better job matching, so that productivity improves.  But, what that logic is saying is that UI makes up for higher current unemployment with future productivity.  Isn't that the opposite of the logic of all the other Keynesian fiscal stimulus policies?  Aren't we supposed to be willing to take small decreases in future production in order to increase current production immediately?  If these workers are becoming employed as a result of the end of EUI, isn't the end of EUI as effective a fiscal stimulus as we could ask for?

EUI as Bizarro Social Safety Net

Duration of unemployment increases with age and with education levels.  This is because older, more educated workers have a better personal safety net, and have more discretion in their work decisions.  Even in the oft-referenced situation where older unemployed workers are ironically denied jobs because they are overqualified, the operating issue there is that, because workers in their position tend to have so much more discretion than other workers, they cannot credibly signal a commitment to the job in question.  Duration is negatively correlated with desperation.

Here's what median duration of unemployment looks like, by age:
There are several items to note here:

1) I haven't included 16-24 year olds here.  The median for all adults (16+ year olds) is so low because the median duration for 16-24 year olds is very low and 16-34 year olds tend to have a much higher unemployment rate than 35+ year olds, with the unemployment rate decreasing with age.

2) So, 16-34 year olds represent a large proportion of the unemployed, and generally the most vulnerable, since their incomes tend to be lower and they have less savings.  Median unemployment for these groups has never been above 10 weeks in any previous recession.  So, only a very small percentage of the most vulnerable population has ever benefited from extended UI.  Even in the current recession, these age groups topped out at a little over 20 weeks, so that most of them would not have used it.

3) Even among the older groups, median durations topped out at around 15 weeks or less in previous recessions, so that most workers were not affected by EUI.  And, the workers who would have used EUI would have tended to be the workers with the highest incomes, the most savings, and the most discretion as a result of their own personal safety nets.  Notice how the 55-64 and 65+ age groups are the age groups that show the most inflation in unemployment duration during the recent recession.  To the extent that former EUI recipients leave the labor force, we can infer that the vast majority are the workers who were the oldest and wealthiest and who are able or nearly able to qualify for Medicare & Social Security benefits.  The GAO survey results corroborate this inference.

In sum, desperate workers generally must find work as quickly as possible, and are not in a position to use EUI.  Twenty-six weeks covers the vast majority of them.  For each week that is added to UI benefits, the policy will be increasingly targeted to workers who do not need the benefit.  I think it would be very illuminating to see the age, average wealth, and lifetime income of workers who were taking EUI benefits beyond 50 weeks.

But there is one additional point I want to make about this graph.  This graph only goes back to 1977 because that's how far the age-specific data goes.  But, the trend shown here goes all the way back to 1948.  Median unemployment durations have never been more than 10 weeks at any time since WWII.  The anomaly in the latest recession is stark and unique.  This tremendous and unprecedented jump in unemployment duration just happens to coincide with a tremendous and unprecedented jump in EUI benefits.  How much of our current perception of a broken labor market is the result of this one policy?  Have we found a policy that is self-sustaining?  It creates unemployment, and then the existence of that unemployment is cited as the reason why we need the policy?

It so happens that unemployment under 26 weeks is very nearly back to normal.  Most of the remaining excess unemployment is long-duration unemployment.  That's quite a coincidence.  At this point, most of the long-term unemployed are not on EUI, but this raises the question about whether former EUI recipients were incentivized to extend their unemployment duration against their better interests, and now find that they have hobbled themselves with a giant red flag when they try to re-enter the labor market.  (No problem.  Just blame the employers.)

This also poses another problem for the idea that EUI creates a net benefit in the long term by increasing subsequent worker productivity.  If the typical EUI recipient will be earning a full-time income for a decade or less, the payoff of marginally higher productivity will not be substantial, certainly not enough to justify adding a year or more to their unemployment duration.

The Social Cost of Moral Posturing

I can understand the feeling that some people have as guardians of the public safety net.  They feel that there are political jackals about that are always trying to undermine the system and leave vulnerable families in dire straits.  But, if any discussion about the effectiveness of our current programs is met with a knee-jerk reaction of Dickensian doomsaying, then this social posture is the enemy of a well-functioning safety net.  If we can't have reasonable public discussions about the effectiveness of these programs, then they will inevitably be ineffective.  Expensive and ineffective social programs are a much larger danger to an effective public social policy than are programs that are too small.

In these discussions, it is important to keep in mind the most vulnerable citizens who might be affected, and to make sure they are protected.  But, this goal is not served by public flagellation and concern-peacocking.  There are valid reasons why this policy is not in the country's best interest.

And, please note, I am not casting aspersions at recipients of these benefits, or making any moral judgments about anyone's decision to accept or not accept them.

Sunday, December 22, 2013

A Cool Census Map

This is a really cool map, from the Census Department:

You can see statistics down to the level of the census tract.

A couple things it really helps to visualize are, (1) how segregated we are in so many ways, (2) how much labor force participation correlates with incomes, (3) and, hence, how segregated we are by labor force participation, and (4) how much high incomes are building around Washington, D.C.

I am sure there are many other interesting things to capture from this map.  I think Choropleth mode is the best way to look at it.

The income inequality among neighborhoods is really easy to see here.  The incomes around the boroughs of NYC are interesting.

Friday, December 20, 2013

More on North Carolina Unemployment Insurance

Evan Soltas, who was the one that got me looking into this, has an update on the issue here:

He adds this graph:

This graph makes it look like the reduction in unemployment is completely related to exits from the labor force.  I'm not sure about the graph.  It's comparing a current rate, to the year-over-year change in another rate, to the year-over-year change in a raw quantity.  There is enough noise in these series already.

Here, I have changed the three measures to month-over-month, seasonally adjusted changes, in thousands of persons, just showing the last two years for more clarity:
FRED Graph

In the last 4 months, the North Carolina unemployment rate has dropped by an average of more than 0.4%, per month.  At the national level, only about 0.75% of the labor force is even on EUI at this point, which adds another wrinkle to the interpretation here.

In the end, while messy data series are the means we use to argue, my reading of Evan's argument is that he considers any drop in the unemployment rate that results from the drop in EUI benefits to be problematic.  If they drop out of the labor force, he suspects that EUI would have kept them engaged in job hunting.  If they become employed, he suspects that they would have benefited by waiting longer to find a better job.

So, data or no data, this really comes down to a philosophical or narrative choice.  We have to be careful here to avoid "mood affiliation", or narrative thinking where we imagine a typical worker.  There were probably around 35,000 people collecting EUI in North Carolina, with 35,000 narratives.  So, when Evan prints things like,
"I think that it is likely that, on the margin, the "active search" requirement is more powerful in keeping people in the labor force than is desperation."
I think he's making several errors of binary, narrative thinking.  Isn't there a middle ground between bureaucratic requirements and desperation that is more populated than these two poles? (Long term unemployed do skew older and more educated.  The GAO found that of the people who exhausted EUI, more of them ended up moving into social security programs than into SNAP programs.  And less than 1/4 had left the labor force.)  Is it the goal of every person with long-duration unemployment to be in the labor force as often and as soon as possible?  I don't actually quite know what to make of this sentence.  If we are thinking on the margin, are we thinking of someone ruled by desperation?  I wouldn't expect the desperate person to make their labor force participation dependent on their benefit status.  I would expect the marginal person to be, say, a guy in his 50's whose wife is working, who almost has the house paid off, who has saved some money for retirement, but not quite as much as he would like.  He sincerely would like to take a job when the right one comes along.  Etc.  He probably doesn't need 99 weeks of UI, not that he can't use the money.  He isn't in any position to claim another type of public support.  It seems reasonable to collect EUI.  Nobody's shaming him for it.  But, he's not going to go apply for SNAP.

For the single 30 year old mother who has been out of work for 18 months, there are other programs.  She will apply for SNAP, and she should.

On the issue of former EUI recipients who are now employed, earlier than they might have been, Evan cites this paper from Daron Acemoglu and Robert Shimer about the productivity benefits of unemployment insurance.  But, I believe that paper concerns itself with moderate levels of UI around 26 weeks.  This roughly describes the level North Carolina has reverted to, and that the US will probably be returning to.  I don't believe that paper offers a defense of 99 week, or even 75 week UI, and Evan is ignoring the problem of hysteresis for longer spells of unemployment.

This isn't a decision between a carefully calibrated safety net and no safety net at all.  This is a decision between a broad, temporary policy, which has uncertain ramifications and a return to the normal, highly developed system of social support that the country generally supports as a general proposition.  The food banks Evan references are part of that support system, as well as many public programs that dwarf the reach of EUI.

(A couple closing comments, here.)

Edit: It was bothering me that the graphs in this post were based on data from both the establishment and the household survey.  Here is the graph with all of the numbers from the household survey, so the summed changes in employment and unemployment should equal the change in the labor force, for each month.  The basic trends are the same, but this is a little more coherent to me (green is employment, blue is unemployment, red is labor force, all are expressed as monthly changes in thousands of people):
FRED Graph

November Update on North Carolina Labor and Unemployment Policy

Here are the updated charts:

Considering how strong the national employment report was in November, this is a very strong continuation of the trends in North Carolina.

The unemployment rate in North Carolina dropped 0.6% in November, and has dropped by 1.5% since July.

Relative to national trends, the North Carolina unemployment rate has dropped by 1.0% since July.  At this point, the movement from unemployment looks roughly split.  During this time, North Carolina's Employment-to-Population ratio is up 0.4%, relative to the national number, and its Labor Force Participation rate is down 0.3%.

These are noisy data series, and I still expect the following few months to help break out the noise from the trends.  The October data might have overstated the success of the North Carolina policy.  The November data might be understating it.

I know this is just one state, and that there are a lot of moving parts that make it difficult to draw distinctions with certainty.  But, this is extremely good news.  If this is indicative at all of what we can expect at the national level after EUI lapses, then we could be seeing an unemployment rate under 6% next year.

PS.  One reason I suspect that the activity between January and July may be generally unrelated is that we don't see much change in regular unemployment insurance claims until June:
FRED Graph

And this graph of North Carolina employment shows how difficult it is to differentiate anything from the noise.  The problem is that we consider changes in unemployment of a couple of percent to be catastrophic, so that any issue can be either written off as a rounding error or be cast as the end of the working class, using the same data.  The dip in early 2013 is the mysterious period of employment loss, and the little pick up there at the end is the part that would suggest that EUI was causing disemployment issues.  If we weren't looking for any effect, this line would just look like a normal data series with nothing unusual going on after 2009 - with a few blips and wiggles along the way.
FRED Graph

Here's another update.  And, a couple of closing comments.

Wednesday, December 18, 2013

More on Minimum Wage & Labor Force

The Atlantic recently ran a decent article on the minimum wage, which, although it tilted to a positive spin, did a better than usual job of covering the topic.  That article referenced an Economic Policy Institute report that claimed 21.3 million workers would see wage increases if we hiked the MW to $10.10.

They assume no job losses, which makes the 21.3 million figure useful to me as an estimate of the number of workers currently making $10.10 or less.

I previously had graphed the level of MW employment at different MW rates, going back to 1979, as published by the BLS.  I found a surprisingly tight relationship:

It occurred to me that we can test the EPI assumptions by taking the 21.3 million workers they estimate would be affected by a MW hike to $10.10, and compare that to the historical number of MW workers when the MW has been at that level.

This would put MW at about 50% of the average hourly wage, and MW workers at about 13.5% of the total labor force.  The EPI report assumes no loss of jobs.  If their assumptions are correct, then this should plot near the long term trend line.  If there are job losses, then this would plot above the line, and we might infer that the distance between the trendline and the plot of all workers affected would give us an indication of the amount of job loss caused by the wage floor.  Here is the comparison:

The EPI assumption of no job loss would predict a quantity of MW workers that is much higher than historical experience.  If this long-term trend holds, we would expect 8 million workers to lose their jobs.  My regressions of past MW correlations with employment would have predicted lost employment of 5 to 10 million jobs.  I would expect the effect to strengthen as the MW level rises to a higher portion of the average wage, since there will be a denser number of workers at those levels.

The EPI report appears to depend on the following logical series:

1) The demand for low wage labor has an elasticity of zero, therefore no jobs will be lost.

2) Only 20-50% of the cost will be reflected in price increases, so 50-70% of the increased wages will be in the form of a transfer from employers to low wage employees.  (Note, then, that the inelasticity of demand for labor does not come from an ability to pass on costs, but comes from an assumption that low wage employers, as a group, have an enormous level of monopolistic profits, a level so high and so universal that a doubling of labor costs in less than a decade would not lead to a net loss of a single job, even as they face declining profit margins as a result.)

3) Low wage workers will be more likely to spend the transferred cash than employers.

4) Spending is better for the economy than saving.  The boost to the economy from the extra spending will lead to the creation of 140,000 extra jobs!

Each of these steps invites some level of skepticism, which makes the expectations of the report unlikely.  And, if we accept these assumptions at face value, the results would require us to expect a level of job retention wholly unrelated to historical experience, as shown above.

Sadly, I don't doubt that we will have a chance, eventually, to test the theory again, with low wage workers as the test subjects.  As the EPI report notes, while many current MW workers are young and single, as MW gets higher in relation to average wages, the costs and benefits of the policy fall more heavily on poor working adults.

This analysis and my estimate of job losses in the recent recession stemming from MW increases would both suggest that as a broad rule of thumb, for every two workers whose wages are below the new MW level, one will lose their job.  Keep in mind that since these tend to be workers on the edges of the labor force, many of those former employees leave the labor force instead of showing up as unemployed.

Tuesday, December 17, 2013

The Yield Curve is a Call Option

At the zero bound, the forward yield curve should be understood as a Call Option on Interest Rates.

Here are the current forward rates in the Eurodollar market:

The main tweak that we need to make, in terms of understanding the yield curve this way, is to reconsider forward rates in terms of time instead of a rate.  When we do that, we have the three basic elements of the Black-Scholes formula:

For a call option, the determinants are:

S = the current price of the equity
r = the risk free interest rate
sigma = implied volatility of the equity price
K = the strike price of the option contract
T = time to expiration

In the context of forward interest rates, first, we need to assume that if short term rates rise from the current level near zero, they will rise at a relatively constant rate for some period of time.  This assumption is not substantially different from the experience of recent decades.

Then, we need to redefine the determinants:

S = the date of the first rate increase
r = the rate at which interest rates will rise after the first increase
sigma = the implied volatility of the potential movement of the date of the first rate increase
K = the expiration date of a position in Eurodollar futures

Here, the time factor, T, does not effect r, but it would be an implied part of sigma, with the typical decay pattern as the date of the first rate hike approaches.  (Note that, where time in a typical call option would normally be a known quantity, here it would fluctuate with changes in S.)

So, looking at the chart above, we can look at the current set of Eurodollar futures contracts (blue line), and break it apart into the elements of an option contract.  The red line would be the expected payouts at expiration.  The kink in the curve is the date of the first rate hike, the slope of the curve is the rate of the rate increases, and sigma is the level of uncertainty about changes in the expected date of the first increase.  Sigma determines the curvature of the yield curve.

In a normal option contract, premiums would reflect a combination of a time premium and a volatility premium.  As I model this, since our new r variable is not time dependent, there is no separate time premium.  There is only a volatility premium.  So, sigma is what bridges the difference between the expected values at expiration and the current contract prices.

We could imagine a different way to consider the model.  For instance, if the Fed announced credibly that they were going to start raising rates in September 2015 - no sooner and no later - then sigma would go to zero, S would be stationary, and the payout would be determined solely by r.  In that context, changes in market sentiment regarding future rates would be expressed through changing expectations in the rate of rate changes coming out of the zero bound, and the slope of the yield curve would be the only tradable factor.

Now that we have formulated yields in this way, we can more clearly see the appropriate methods for taking a range of positions.

For instance, going forward in time is like going into the money.  The farther into the money you go, the more an option behaves like the underlying security and the less it acts as an insurance policy.  So, if you want to hedge a portfolio that will benefit if interest rates increase but will suffer with low rates, a hedge against low rates might be most effective if part of the hedge was positioned as a long Eurodollar contract (a short position, in terms of the interest rate) in the March to September 2015 range.  This would have a payout similar to a short option, with a limited premium payoff potential and a large potential downside if it moves in-the-money (if the Fed starts raising rates earlier).  You would receive a premium for this hedge as sigma deteriorates over time.

On the other hand, a speculative position for rising interest rates should be positioned farther out on the curve, into late 2016 or later.  A speculative position taken on contracts expiring earlier than that will pay a premium, as it were, as sigma deteriorates.  Right now, the market expects the first rate hike around September 2015.  If you believe the first rate hike will instead happen around June 2015, and you sell September 2015 Eurodollars (which is a long interest rate position), you will have a loss even if your forecast is correct and the rate hike happens in June.  This is because you will have been paying for implied insurance due to the sigma related premiums for those contracts embedded in the current yield curve.

It should be theoretically possible to take several positions along the curve which would net out to give you a pure position with exposure to only one of these factors.  To speculate on a rate hike happening before the current market expectation, A position of 2 short June 2016 contracts and 1 long June 2017 contract should provide no net exposure to changes in slope (r), but this combination would be equivalent to holding a single short position at June 2015.  The advantage of this position is that it would not be exposed to the sigma premium that the raw June 2015 contract would be.  In fact, again in theory, this should be an arbitrage opportunity, since that synthetic position could be hedged against a long June 2015 contract.  The long June 2015 contract would "earn" a premium, as the yield curve slowly meets the expected payoff curve over time.  These positions should roughly net to zero over time, and the speculator would earn a premium of about 35bp over that time.  At current contract margin requirements, that would provide a return of about 25% in less than 2 years.

Monday, December 16, 2013

Fed Funds Rate and S&P500 returns

Here is the flip side of the return to equities after a rise in the Fed Funds Rate.  This graph shows the monthly performance of the S&P 500 after the initial decrease in the Fed Funds Rate (implemented during an inverted yield curve that signals a coming recession).
During the period starting in 1989, investors avoided a bear market.  But, the next two episodes saw large declines in equities.  It looks, in any case, like we might be able to expect a period of several months in which to measure whether the apparent monetary easing due to falling rates is effective before committing to a hedging strategy or a speculative short position.
In fact, while this includes only three events, these events suggest that serial correlation in these events is significant, and that the market signal in the few months after the initial rate cut might indicate the continued movement of the index over the cycle.
There is an awful lot of pressure out there for a hawkish Fed stance, so when the next cycle comes, I'm afraid that rate cuts will be tepid, and returns will look more like 2000 & 2007 than like 1989. 

Friday, December 13, 2013

Fear of the Fed

I keep reading that equity funds are seeing outflows because investors fear the taper and the inevitable rise in interest rates.

Here is what everyone is afraid of:

The horror!!

Labor force participation trends and assumptions

John Taylor references research by Christopher Erceg and Andrew Levin (this link is to an April 2013 version of the paper Taylor references with a September 2013 date) (HT: Marcus Nunes).  I am disposed to agree with Taylor that the recovery has been slow, partially, due to economic policies.  But, here he is saying that labor force participation (LFP) is low because of these cyclical pressures, and that the economy is much worse than the unemployment rate (UER) would make it seem.  My past reviews of the topic have convinced me that the LFP decline reflects some very typical cyclical movements, and that all of the unusual decline has a demographic source.  So, I took his opening comment: "Research by Christopher Erceg and Andrew Levin is providing solid evidence that the decline in the labor force participation rate since 2007 has been due to cyclical factors" as a challenge.  How could reasonable research coming from the Federal Reserve reach such a different conclusion from what I have found?  Maybe I'm getting it wrong.

I think there are two main factors that are leading to the different conclusions.  My version of the difference is:

1) The study relies on November 2007 projections of LFP from the BLS.  In hindsight, the BLS got it wrong.

2) The minimum wage increases of 2007-2009 pulled a lot of younger workers out of the labor force.  I would agree that this is bad policy, and it led to a kind of cyclical drop in LFP, but the marginal effect of that policy on the rate of change in today's LFP should be small.  The continued downward drift of LFP today is due to demographics.

LFP Trends

Two items from the paper that make this clear are:

First, we can see that the unexpected drop in LFP does not come from workers 55 years and older.  They roughly followed the trend that the BLS forecasted in 2007.

16 to 24 year olds took a huge hit from trend.  We could expect this age group to have more cyclical behavior, but I believe much of this is due to minimum wage increases.

The vast majority of the hit to LFP, in absolute numbers, comes from the 25-54 age group.  The paper acknowledges that there had been a long term downward trend in these age groups.  So, I would expect the actual drop in LFP of -1.5% to reflect the long term trend of -.5%, plus a drop in LFP of about 1%, which would represent about .5% of cyclical movements above and below trend as the economy switched from a very hot labor market with 4.5% UE to a recessionary market with up to 10% Unemployment.  But, strangely, the BLS had projected an increasing LFP for this age group.

Stacking Error on top of Error

Treating the top of the labor market boom of the 2000's as the new normal caused a dual error - forecasts from 2007 started too high, and the trend slopes were adjusted too high.  Here, I will simply compare the least squares linear trend of the long term LFP rates of each age group to the trends set by the BLS in 2007.

Here, we see that, compared to long term trends, the 2007-2012 period looks perfectly normal for 35+ year olds.  The 25-34 year old group has an unusually sharp cyclical deviation, which is now reverting to the mean.  But, the 2007 BLS forecasts all either start above the trend mean or have an inflated slope, or both.  In hindsight, in a data series with such a long pattern of linear behavior with mean reversion tendencies, it was clearly an error to forecast LFP's that would accelerate from a frothy labor market.

The trends in the female LFP's aren't as long, but in 2007, there was a good 10 years of data suggesting a very similar pattern of falling within each age group at a rate of 1% to 1.5% per decade (the coefficients in the trendline equations are based on quarterly units).

Perhaps, due to the shorter history of linear behavior, the BLS can be forgiven for expecting more positive movement from the female series.

In either case, I'm afraid that any analysis that uses the 2007 BLS forecasts as a baseline is simply measuring the understandable failure of the BLS to predict the exact timing of the business cycle.

In hindsight, the naïve linear trends appear to be a much more reasonable baseline than the BLS forecasts.  In every case, the BLS forecasts predict higher LFP's than the naïve trendlines, generally overstating expected levels by 1 to 3 percentage points by 2012, compared to the naïve trends.

Cyclical Decreases in LFP

The Fed paper includes this graph:

The regression produces the following result:

LFP = -0.40 - 0.30*UER

Of which the authors comment:

These regression results provide stark evidence that cyclical factors have been crucial in explaining the recent decline in prime-age LFPR. The coefficient on the lagged change in prime-age unemployment is highly significant (t-statistic of -3.9); that is, the state-level data exhibit a strong negative correlation between changes in LFPR and lagged changes in unemployment for prime-age adults. In contrast, the regression intercept is not statistically significant from zero (t-statistic of -0.97), indicating that the data provides no support whatsoever for structural interpretations of the drop in prime-age LFPR. In effect, the state-level data indicates that the aggregate decline in prime-age LFPR since 2007 can be fully explained by the persistent shortfall in labor demand.

That seems like a really strange conclusion to me.

1) The bulk of the long-term trend in declining LFP comes from workers aging out of the prime working age.  Right now, in 2013, there is a downward trend from that effect, and it will be a headwind for the aggregate LFP, regardless of what cyclical issues there are among the prime age group.  These inter-age-group changes cut about 0.65% off of aggregate LFP from 2007-2012.  I don't see how finding cyclical LFP changes within an age group addresses this.

2) Using 25-54 year olds as the prime age group is a problem, because 45-54 year olds have a markedly lower LFP than 25-44 year olds.  Right now, baby boomers are bulking up the 45-54 year old category, so there is a temporary downward trend among this group that amounted to about 0.55% during this period.

3)  They conclude that the intercept, -0.40, is not statistically significant from zero, so they proceed to attribute the entire decline in LFP to cyclical factors.  I would say that -0.40 is not statistically significant from -0.55, which is the secular decline in LFP we would have expected from this age group over this time period.  So, roughly 30% of the LFP decline among 25-54 year olds was the result of long term aging trends.

4) If statistical analysis can come that close to the expected intercept (-0.40 compared to -0.55) and still be interpreted as having an intercept that is not significantly different from zero, then I have doubts about the ability of that analysis to say anything at all.

Taylor attributes this graph to the authors:

This graph looks to me like the Unemployment Rate line assumes that the "normal" LFPR would be practically flat.

On the contrary, the graph below shows a LFP trend line based on long term age-group trends.  This downtrend is accelerating.  Currently, LFP is declining by about 0.1% a year due to long-term trends across the prime age groups, and by about 0.2% a year due to aging baby boomers.

LFP went from about 0.75% above trend in 2007 to about 0.75% below trend at the end of 2012.  This is not outside the range one might expect from a deep recession.  It is about the same as the drop during the 1980-1982 recession, but that recession has a growing LFP trend, due to the entrance of more women into the labor force, so if you don't correct for trend, it looks like there was no LFP retrenchment during that recession.

At the rate the LFP is naturally declining, we will be back to trend by early 2016, even if the Employment to Population Ratio doesn't increase at all.  We are the mirror image of the period of the 1980-82 recession.  At that time, a cyclically rising LFP rose very quickly, and a cyclically dropping LFP looked flat.  Today, a cyclically rising LFP is flat, and a cyclically dropping LFP drops like a stone.  So, comparing the 2007 inflated BLS trends to 1980 is like stacking errors on top of errors on top of errors.

I would agree that policy issues have made this worse, but not in a cyclical way.  JOLTS measures (hires, quits, openings, etc.) are still at low levels, but they have been growing at rates similar to rates in the previous recovery:

I have attributed the entire drop below LFP trend to the minimum wage.  And, the remaining unusual level of unemployment is attributable to the minimum wage and emergency unemployment insurance.  I would call these pro-cyclical structural problems.

Both demand-side Fed solutions and supply-side economic structural solutions can help a little bit, but I think only a little bit.  On the other hand, the pro-cyclical nature of these policies should eventually create some catch-up growth in the labor force and in employment.

Thursday, December 12, 2013

A Natural Experiment on Emergency Unemployment Insurance in North Carolina

Update:  Here is an update with November numbers.

In February 2013, North Carolina passed a law that at the end of June, would end Emergency Unemployment Insurance (EUI).  This gives us a kind of natural experiment to get an idea of the effect of EUI on employment.

The apparent effects were large and immediate.

Normally, 0.6% of the labor force has been unemployed for more than 26 weeks.  Currently, that amount is about 2.6%.  From that population, about 0.7% of the labor force is on EUI.

My estimates, from the model I have been reviewing here in recent posts, estimates that the unemployment rate (UER) is inflated by about 1% due to EUI's effect on extending the average duration of unemployment.

This should mean that the Employment to Population Ratio (EPR) is deflated by about 0.6%.

My estimate for the positive effect of EUI on Labor Force Participation (LFP), which, unlike the other factors, is more of a broad guess, is currently about 0.15% of the labor force.

What Has Happened in North Carolina?

It appears that fairly immediately after the bill passed, even before it became enforced, all three indicators fell precipitously in North Carolina.  Real world data always throws us some curve balls.  The re-entry of 1% of the North Carolina labor force back into employment should have increased the EPR.

Maybe there is some other factor that has caused all these measures to decrease in North Carolina.  On net, since January, 1% of employed North Carolinians have left the work force, and in addition, 1% of unemployed North Carolinians have left the work force.

You could argue that the Keynesians were right, that UEI is stimulative, and that when North Carolinians stopped receiving benefits, they left the labor force out of discouragement.  Furthermore, they reduced their consumption, which caused more North Carolinians to lose their jobs.  But, this would be unlikely to lead to reduced UER.

I measured the level of each statistic, relative to the national level, and normalized them to the average levels for the 12 months of 2012.  Then I compared them:

From the point when the law was passed until it was implemented, compared to national averages, approx. 1% of North Carolina's adults ceased to be employed workers, on net, and left the labor force.  An additional 0.1% moved from being unemployed to being out of the labor force.  Compared to the 2012 average levels, both LFP and EPR decreased by about 0.7%.

Since the law took effect, the Labor Force has held steady, and 0.5% of adults, who were unemployed, have become employed, boosting EPR by 0.5% and lowering the UER by 0.8%.

Maybe the movements before July 2013 were unrelated to UEI.  I suppose I could create a narrative where North Carolinians, foreseeing a cut in benefits, cut back on consumption immediately upon passage.  This led to a Keynesian contraction.  After benefits were cut in July, workers were more incentivized to become re-employed, so the supply-side story kicked in, causing employment to rebound.

When I first saw this situation, I thought the drop in LFP might suggest that I have underestimated the effect of EUI on LFP.  But, it now looks to me like practically all of the LFP shrinkage was related to a decline in employment, which was either a Keynesian result of the cut in benefits, or was not related to the EUI law.

But, it looks plausible that the North Carolina experience will support both of my estimates that (1) unemployment is about 1% higher than it would be without EUI and that (2) LFP is slightly higher (less than 0.2%).

Tuesday, December 10, 2013

Minimum Wage As a Proportion of the Labor Force

Following up on the recent series of posts, I used annual data from the BLS, going back to 1979, to compare the relative level of minimum wages to the proportion of workers working at minimum wages.  (Here is a follow-up to this post.)


Some inferences we might be able to draw from this relationship:

1) The relationship is surprisingly linear, which suggests that a linear relationship between minimum wage levels and employment may be a reasonably accurate assumption.

2) For each 1 percentage point increase in the MW, relative to the Average Wage (AHETPI), the wages of an additional approx. .44% of workers fall below the MW.

3) Historical correlations associate about .4%-.44% in lost employment for each 1 percentage point increase in MW/AW.  This suggests that a total of .9% of workers are affected, with about 1/2 losing employment.

4) In the 2007-2011 period, 1.6% of additional workers' wages fell below the new minimum.  An estimated 3.4% of workers were associated with MW job losses.  The relative amount of estimated lost employment was especially high in this episode.  As can be seen in the graph, this could be because the minimum wage had reached a level below the natural minimum.  After the first two MW hikes in 2007 & 2008, the proportion of the labor force at or below MW was still at 1.7% in 2008.  After the third rise in 2009, the proportion of workers earning MW finally topped out at 2.8% in 2010.

This probably means that I need to revisit my employment estimates for the recent period, because a linear relationship probably overstated the effect of the first hike in 2007.  That hike probably didn't have as great an effect as the nominal dollar amount would indicate.

5) At least for the period covered by this data, the reduction in the reach of the MW seems to correspond to the leftward shift of the Beveridge Curve.  The recent significant shift to a broader MW also corresponds to the recent shift of the curve back to the right.

6) In a context of increasing income inequality, or variance in wages, we might expect a shift up in the relationship (more MW workers at a given MW level).  We see the opposite, though.  There appear to be periods roughly parallel to the trend, but with a shift down in the 1990's, so that there are relatively fewer MW workers now than there were in the 1980's.  This also shows up if we graph the relationship in terms of constant dollars instead of as a proportion of average wages.
This could be the result of a change in the middle class life cycle.  Young people have much lower labor force participation than they used to.  So, this could be a product of middle class workers choosing not to work during the low-wage portion of their potential work life.

7) In 1980, about 60% of combined MW workers were at minimum wage and about 40% were below minimum wage.  By 2008, nearly 90% of the combined MW workers were below the minimum.  I take this as another sign that the MW was at an ineffective level by that time.  This suggests that most jobs reported as MW involved tips or some similar characteristic.  By 2012, the portion of combined MW workers working below the minimum was back down to 56%.  This figure topped out in 2008, even though the first MW hike was in 2007, which again suggests that I need to lighten up my estimates of job losses during the early part of the recent episode.

8) Women are much more likely to work at MW than men:

Yet, my preliminary tests of the data show that, with some differences in character, male and female employment are roughly similar in their sensitivity to MW hikes.  This poses a challenge to my model.  Women are concentrated at MW jobs at nearly 3 times the rate of men.  That should lead to greater sensitivity to job loss.

There are caveats.  Female employment has different cyclical behavior than male employment.  And, whatever the differences in labor force behavior among the genders, their proportions are pretty stable, so whatever differences there are, they shouldn't undermine the stable relationships of the aggregate population.

It is also likely that some of the employment losses from MW shocks comes through supervisory positions or other jobs with complementary relationships to MW jobs.  This is one reason why studies that track individuals working at MW would not capture the complete effect of MW hikes.

But, still, the inability to explain this would be a strike against my model.  The higher proportion of young workers at MW meant higher sensitivity to shocks, and I would have expected to see the same thing among genders.

Minimum Wage, Demographics, Emergency Unemployment Insurance, and Labor Force Participation from 2007-2011

In the previous post, I estimated the effects of these issues on unemployment.  Now, I'm going to look at Labor Force Participation (LFP).

Edit:  After looking some more at the relationship between the MW and the proportion of workers at or below MW, I noticed a non-linearity at the low end of the range, where MW levels were in 2007.  The first hike in 2007 hardly budged the proportion of workers at MW.  This was likely because the legislated MW had fallen below the typical voluntary MW.  I have updated the graphs and numbers below to reflect that deviation from the trend.

Minimum Wage

I discussed this in the previous post, and went into more details in the posts before that.  Even though there have been relatively few separate minimum wage episodes over the past 60 years, I was able to establish a strong relationship between the 6 month change in employment over the course of the typical episode and the scale of the MW increase in relation to the average wage.  The forecast specified by that relationship is what I use here to estimate the effect on labor force participation (LFP).  Much of the lost employment coming from MW hikes appears to lead to lower labor force participation instead of unemployment.


I have discussed this quite a bit.  Most recently here.  This shouldn't be complicated.  There are stable long-term trends among different age groups.  As the population bulge enters the older age groups, this causes the reported labor force participation rate to shrink.

Emergency Unemployment Insurance

EUI plays a small role here.  I referenced a GAO report here that found about 1/5 of EUI recipients who had aged out of EUI exited the labor force.  An analysis of these factors should include the effect that EUI should have on inflating LFP.  I presume that, on the margin, this is not controversial.  On this particular effect, I don't have any clever methods to get at a number, so I have simply estimated this effect to be equal to 10% of the cyclical long term unemployed.  This estimate leads to a fairly small number, in any event.  I don't believe I am double counting with the EUI unemployment effect, because that measure specifically did not attribute any additional unemployed workers to the effect.  It only accounted for longer durations of unemployment among the given number of unemployed.

Here is the estimated effect of these factors on LFP.
graph before update

The minimum wage effect is estimated to outweigh the demographic effect since 2009.  Adding these effects up and comparing them to the actual change in LFP over this period, we see this comparison:

Considering that these three forecasts come from three separate contexts, it's kind of surprising that they would add up to explain the entire recent this much of the character and the decline in the LFP.  But, maybe they do.

Here is the chart, extended to 2013.  As with the unemployment chart, I am less confident in the Minimum Wage effect once we get past 2011.  But, we can see here that as the EUI and MW factors dissipate, the actual LFP is now sloping below the estimate derived from these factors.

graph before update

Unemployment, Labor Force Participation, Market Frictions, and the Definition of Cyclical

Here is a long term chart of the cyclical deviation of LFP, compared to the inverted Unemployment Rate:

Here is that chart, with the adjustments from 2007-2013 that I have just estimated, removed.  If my estimates are correct, this would represent the truly cyclical behavior of the current labor market.

graph before update
With the adjustments, unemployment looks similar to previous downturns.  But, adjusted LFP seems just a little too high.

It could be that I have overestimated something.  It could be that my model doesn't capture enough unemployment among the MW job losers, so that adjusted unemployment and adjusted LFP are both too high.  The adjustment to reduce the estimated effect of the 2007 mimimum wage hike, as discussed above, does create a more realistic looking cyclical LFP path in the early part of the crisis.  In the 2nd version of the new graph, I changed the assumption about EUI and the labor force.  If we assume that, instead of 10% of the excess long term unemployed, that a full 25% of excess long term unemployment will transfer out of the labor force, then the cyclical trend looks very normal.

Obviously, tweaking other estimates here and there would move the demographically detrended LFP up or down, slightly.  The point here isn't to claim exact numbers, but simply to suggest that reasonable estimates of the effects of these policies and long term trends can explain all of the supposedly unusual labor market behavior.  At least, I think I've given some plausibility to the idea that there is no big mystery in recent LFP behavior.

One other comparison I will make with the adjusted data is to compare the standard Beveridge Curve (the relationship between unemployment and the number of job openings) and an adjusted Beveridge Curve that utilizes my policy-and-demographic-adjusted unemployment rate.  My adjusted unemployment rate pulls the Beveridge curve back into the recent conventional range.
graph before update

So, if my adjustments are accurate, does this shift in the Beveridge curve point to labor market frictions and rigidities.  Historically, the Beveridge curve has shifted slowly through decades:

Is this just a reflection of shifting levels of labor market frictions?  How much have policies such as the Minimum Wage been reflected in these changes?

The other question that comes to mind is, "What is the difference between structural and cyclical or demand-driven shocks?"

Disemployment from the minimum wage is clearly a structural shock, but it plays out like a cycle.  EUI also looks pro-cyclical to me.  It's countercyclical for the unemployed person who is provided with a safety net.  But, it's pro-cyclical to the analyst looking at the unemployment rate.

So, the discussion seems a little confusing to me if someone says, "Bad news.  It looks like the LFP drop is cyclical.  Unemployment should be even higher than it looks."  Once you correct for the demographic issue, do you answer, "No, it's not.  Much of this might be a minimum wage or EUI effect."?  Or, do you answer, "Yes.  You're right.  Much of it might be a minimum wage or EUI effect."?

I believe I have two more posts coming on the topic, if anyone is still reading at this point...

Minimum Wage, Demographics, Emergency Unemployment Insurance, and Unemployment from 2007-2011

So, after reviewing the historical evidence on increases in the national minimum wage, I have some method for estimating each of these factors on employment in the latest downturn.

Edit:  After looking some more at the relationship between the MW and the proportion of workers at or below MW, I noticed a non-linearity at the low end of the range, where MW levels were in 2007.  The first hike in 2007 hardly budged the proportion of workers at MW.  This was likely because the legislated MW had fallen below the typical voluntary MW.  I have updated the graphs and numbers below to reflect that deviation from the trend.

Minimum Wage:
This is based on the forecast described in the previous few posts, based on the relationship between the change in the minimum wage as a proportion of average wages, and the change in employment, during episodes of minimum wage increases from 1956 to 2004.  For total employment, during two-year periods following an initial MW hike, the null is rejected at the 10% level.  For follow-up MW hikes in episodes with more than one increase, the null is rejected at the .001% level if we include all periods of time, and is rejected at the 10% level for the 1956-2004 period.
The null hypothesis could not be rejected when tested in terms of unemployment instead of employment for the period ending in 2004.  But, the employment forecast gives us the more important number, in terms of expected jobs lost.  The unemployment forecast simply helps parse that out between workers who leave the labor force and workers who are categorized as unemployed.  With that caveat, I have used the forecast from those tests for the estimated unemployment due to MW hikes.

This estimate is based on a pattern I touched on here.  Unemployment duration tends to grow with age, and education.  I think this tendency can basically be thought of as labor market friction.  Older workers may have greater skill specialization, which creates searching costs.  They may also have more back-up resources, and a greater ability for consumption smoothing.  All of these factors may lead to greater cyclicality in the unemployment rate during times when there are more older workers.  This estimate is based on simple relative average unemployment durations for each age group, with unemployment increasing in proportion to average duration.
This is one of the earliest effects of the baby boomer aging process.  This effect is already diminishing because as older workers enter the oldest age categories, their labor force participation and general level of unemployment decline enough to counteract this effect on unemployment.

Emergency Unemployment Insurance:
I touched on this issue in this post.  Here is a graph of the relationship between unemployment durations of less than 26 weeks and durations of 26 weeks or more, going back to 1960.  Short term unemployment durations and long term unemployment durations moved with some uniformity through many different business cycles over that time, until the implementation of 99 week EUI.  David Grubb from the OECD explains how the level of EUI support was vastly longer than in any previous period.

My estimate for how much this policy affected the unemployment level treats this, much as in the demographic effect, as a labor market friction.  I compared the proportional level of durations over 26 weeks to the typical level we have seen in business cycles over the past 3 decades, and attributed that difference to the EUI policy.  One might be able to argue that there are some other factors involved here, but decades-long pattern of this relationship suggests that those other factors have been dwarfed.  Further, my measure ignores any influence EUI might have had in increasing the average unemployment duration of workers unemployed for less than 26 weeks.  This is clearly a conservative assumption.  On net, until I figure out how to sensibly estimate EUI influence on durations under 26 weeks, this estimate will probably be somewhat conservative.

Here are the unemployment estimates:

graph before update

This chart covers the time period until July 2011.  This is the period where I have a monthly model for minimum wage related employment changes.  These estimates suggest that if we did not have these demographic and policy issues, this would have been a typical unemployment cycle.  (The pale blue line is where I estimate we would be without these policies.)

Here is the graph extended to June 2013.  We are getting far enough away from the MW hikes that the recovery in low-wage employment is mostly dependent on broader economic recovery, Fed policy, etc..  As the earlier posts on MW showed, MW episodes that were small enough would eventually see catch-up employment growth as the economy naturally outgrew the wage-price floor.

graph before update
We may be near a normal labor market, if not for these issues.  These issues might continue to inflate the unemployment rate for a few years, but they shouldn't be permanent.  The Fed hasn't been too loose up to this point, but if they start to err on the side of being stimulative in the face of an unemployment rate that stays stubbornly above 6%, the hidden core of the labor market which may already be back to recovery levels, might create unexpected inflation.  The scale of the recent MW hikes was large enough that there may be low wage workers who continue to be priced out of the market if we don't experience some inflationary adjustments.  There may also be some frictions for long-term unemployed who want to go back to work.  So, unemployment declines might slow down at a higher range than has been the recent experience.  But, it is possible that if federal policies allow these issues to decrease in importance, that could help to reaccelerate the decline in unemployment.

Next up, Labor Force Participation.