Tuesday, September 2, 2014

August 2014 Employment Preview

Here's a look at some of the employment stats I follow.  First are the comparisons of total unemployment with insured unemployment.  In the first graph, the last data point represents currently reported August insured unemployment and a projected August unemployment rate of 5.84%.  This might seem a bit optimistic.  I do think we are due for a downtick, and a rate under 6% is quite possible.  This is the number shown by the basic regression shown in the next graph.  Mainly, I used that number in the first graph to give a visual picture of the possible range of outcomes.  As you can see, it wouldn't be outside the basic trend range.  On the other hand, it is also possible for this relationship to bounce around for a few months, so that another month at 6.2% wouldn't be out of line.  The fact that insured unemployment has leveled out this month mitigates against the case for a gap down in total unemployment.  But, 6.3-5.7% is probably the 2 standard deviation range this month.

Taking a look at flows, I think we have a mixture of new cyclical trends and some statistical noise.  The readings this month will depend on how much we are seeing of each.  Net flows from NtoU have been low over the past year.  The increased rate of decline in unemployment over the past year has been roughly divided between a decline in net NtoU flows and an increase in net UtoE flows.  This is visible in the slopes of the trendlines.  As the recovery matures, net NtoU will probably settle in at about 0.1% per month and net UtoE will settle in at about 0.15% per month.

While net NtoU has declined, net NtoE has improved.  Essentially, workers who were out of the labor force are now more likely to re-enter the labor force with a job.  This is a very strong sign.  This flow has never had a positive moving average during the 25 years of data that we have, so the recent trend has been very strong.  I wouldn't expect this flow to have a sustained level above -0.05%, and that would be the sign of a very strong labor market.  This flow doesn't really affect unemployment directly, but it does indirectly.  Normally, this late in a recovery, the number of unemployed workers would be at full employment levels, so we would see a decline in the net UtoE flow and a strong economy would be pulling the net NtoE flow up as marginal or opportunistic workers enter the labor force.  In this economy, the countervailing flow may be a reduction in net NtoU instead of a reduction in net UtoE.  The level of unemployment as we end the year will reflect this unknown.  I suspect that we will continue to see strong net UtoE flows because of the continuation of re-employment patterns we have been seeing in the unemployment duration data, the positive unemployment insurance trends we have been seeing, and the continuing improvement in JOLTS data (hiring, job openings, and quits).

thousands per month (slope)
I think there is a decent chance that we'll see a downswing in the unemployment rate this month as we walk back some statistical noise from last month.  The flows to E (both UtoE and NtoE) have been subdued the past couple of months, which seems unlikely, given the strong indications of hiring we have seen from the JOLTS data and the Establishment Survey.  I think it is likely that, given the see-saw pattern of the flows data, both of these flows will see-saw up this month.

Also, the EtoU flow has remained fairly flat for the past couple of years, despite significant declines in insured unemployment rates and initial claims.  Previously, when unemployment claims were at current levels, this flow tended to be about 0.1% lower.

And, we can see in the durations data that last month's unemployment seems to have come in high, as 3 of the categories were up with no apparent cause.

My gap down didn't happen last month, but I'm sticking with my story.  I think there is a good chance of seeing a move down to 5.8%-5.9% in the reported unemployment rate for August.  As the employment market settles into more stable late-recovery patterns and the cohorts of unemployed that are being re-employed more quickly continue to move out further into the long-duration category, we will see the unemployment rate drop 0.2-0.3% over the remaining 4 months of 2014, leaving us at 5.5% to 5.7% in December.  I believe my forecast from earlier in the year remains accurate, and that we could see most of that gap taken out in August.

Monday, September 1, 2014

Minimum Wage Levels, by state

Political Calculations has a great post with some interactive maps that allow you to compare minimum wage levels and cost of living among the 50 states.

This map is interactive at Political Calculations
I have been a little worried, because it seems as though a decent number of states are starting to increase minimum wages above the Federal level.  As Political Calculations has pointed out, they were starting to do that about 15 years ago, before the national level caught up with most of them in 2007-2009.

But, looking at his recent post, it seems that most of the increases are in states with higher costs of living.  A comparison of minimum wage employment as a proportion of the labor force suggests that at levels below about 28% of the average wage, at the national level, the legal minimum wage mostly appears to stop being a binding constraint.  We are probably nearing that level now at the national level.  So, if states with high average wages raise their legal minimum wage levels moderately, it may not create a significant amount of labor dislocation.

As shown in this map, many of the high-wage/high-cost states have relatively low minimum wage levels, when adjusted for cost of living.  Washington and Oregon are the only states with grossly inflated minimum wage levels, adjusted for cost.

This might signal a build up to an eventual national hike in the next few years.  And, if that does happen, the effect of the minimum wage on employment might be similar to past experience, even with a few states already above the new minimums, because it would be having an effect in the states where it is the most binding.

Friday, August 29, 2014

Long Term Non-employment

The Financial Times has an interesting article on long term unemployment.  It is a review of this presentation at Jackson Hole by Jae Song and Till von Wachter.

They utilize longitudinal microdata to track the employment status of displaced workers over time.  Using "non-employment" in lieu of "unemployment", they find little difference between the re-employment behavior of workers in the recent recession and workers of previous recessions.  Here is a graph of very long term non-employment persistence from the paper.  While there has been some change in the trend over time, there is little difference between recessionary times and expansionary times.  Also, there is little difference between the outcomes since 2008 and previous periods.  This is a fairly shocking finding, considering the well-known existence in the BLS's Household Survey of a large quantity of very-long-term unemployed workers.

Here is a graph of unemployment duration by age.  Keep in mind that short term unemployment durations are pretty normal now, so all of this extra average duration is coming from 1/6 of the pool of unemployed workers.  So, estimating from BLS data, it looks like about 5.2% of the labor force is unemployed in a fairly normal labor market, with average unemployment durations of less than 20 weeks.  Then, there is another 1% of the labor force that is unemployed, with an average unemployment duration of more than 120 weeks.

At first glance, these data are telling two different stories.  Here are some more graphs from the article.  First, this graph shows the fraction of the labor force that has been non-employed for one year, two years, etc.  This shows that extended unemployment was slightly worse in the 2008 recession than it had been in 1980, but not excessively worse.

The next graph (Figure 12A from the paper) shows the re-employment behavior of displaced workers in four recessions.  The recovery to employment in the 2008 recession has followed a pattern similar to previous recessions.  It is worth noting that a displacement episode appears to lead to a permanent 10% reduction in employment among the affected workers.

Finally, in the Figure 6A from the paper, we can see the surprising finding that the level of long-term nonemployment has been unusually low in the 2009-2011 period.

This suggests that much of the very long term unemployment in the current count is mostly a categorization issue related to workers whose behavior hasn't been materially different from previous recessions, but who may have been more likely to refer to themselves as unemployed in surveys because of subtle framing effects related to public labor policies, such as emergency unemployment insurance (EUI).  If that is the case, it would suggest several implications:

1) Unemployment has been overstated, relative to previous recessions.  This would apply to the approximately 1% of the labor force that currently is categorized as unemployed with very long unemployment durations.  It would also apply to the U-6 unemployment rate that includes marginally attached and part time workers.  The paper outlines how there tends to be marginal employment activity over time with long term non-employed workers, after a displacement, but that over time permanently non-employed workers become a larger proportion of the remaining non-employed.  Following this pattern, we should continue to see a reduction in U-6 unemployment.  I suppose that we might end up with a permanently self-identified population of unemployed workers, but I think it is more likely that to the extent that this group reflects the displaced workers who permanently leave the labor force, they will slowly begin to self-identify as not-in-the-labor-force.

So, if this is the case, the current unemployment rate, stated comparably to previous periods, might be in the low 5%s.  This would explain how real wages have been higher than we should have expected, given the unemployment rate.  The authors also point out that the permanently non-employed displaced workers tend to be older, which also might explain why unemployment in this recession tended to be excessively high for older age groups.

Here is a graph of employment flows.  Note that the most unusual movement in the recent recession was the unusual increase in flows between unemployment and "Not in Labor Force".  Flows between Employment and "Not in Labor Force" and between Employment and Unemployment reached levels slightly worse than the 2000 recession and about the same as the labor market in 1995 when these data series begin.  And, these sets of flows are currently back at normal recovery levels.  This would lead one to expect a business cycle where unemployment topped out in the high single digits and is back around 5%.  The outlier here is the flows between unemployed and "Not in Labor Force", which moved much higher, relative to the other flow sets, and which remain elevated.  Could this unusual movement reflect a change in self-identification and categorization among marginally attached workers who, in previous downturns, simply identified as "Not in Labor Force"?

This also comports with the recent high level of job openings and the idea that, adjusted for demographics, JOLTS indicators point to a historically comparable unemployment rate around 5.7 (which, given our current demographics would come in at around 5.3%).

2) I have been too hard on EUI.  If this paper is on to something, then EUI didn't change labor behavior significantly, so it shouldn't be blamed for the long-term unemployment problem or for significant hysteresis in the labor market.  These are workers who are mostly just being labeled differently within a fairly typical labor market behavior.  I would still argue that it might not be the most efficient redistribution program, but this paper seems to support the argument that the apparent increase in unemployment durations from EUI comes mostly from movement between "Not in Labor Force" and Unemployment, not from delays in re-employment.

To the extent that this data is informative, it might suggest that in the next downturn, an extremely generous EUI program won't necessarily be that damaging to the labor market - it will just appear to be.

These implications would all generally point to a more optimistic picture of the current economic context.  It would mean that historically comparable Labor Force Participation took a deeper cyclical dive than the reported numbers suggest.  Although, the adjusted statistic would show a dip earlier in the recession, with stronger recovery since then.  But, it also means that we are currently basically recovered and that the labor recovery was stronger and sooner than we thought it was.  Much of the remaining reductions in unemployment would typically be recorded as re-entries into the labor force.

Finally, these findings show how beneficial functional NGDP targeting could be.  There is something to be said for the creative destruction that might come out of a difficult economic period.  But, I think it's incorrect to argue for unnecessary economic disruptions.  The aggregate costs surely outweigh the benefits.  This paper points to significant permanent disemployment coming from economic dislocations.  If some of these labor disruptions are a result of suboptimal monetary policy, and if more stable nominal demand could prevent some of these dislocations, it could lead to higher labor force participation and utilization over time.  I don't think higher labor force participation should be considered a goal, a priori.  But, the permanent disemployment from these dislocations are almost certainly inefficient and are not remotely optimal for the affected workers, so in this case, it would represent improvement.


Wednesday, August 27, 2014

Squaring the Circular Logic of the Interest Rate-Leverage Connection (Updated)

I have outlined theoretically and empirically how the equilibrium level of corporate leverage, counter-intuitively, could rise when interest rates rise.  Here is a graph of corporate debt and equity values over the past half century (adjusted for inflation), with interest rates and equity premiums.

The next graph compares corporate equity and debt levels as a proportion of operating profits to nominal 10 year interest rates.

Now, regardless of any problems with this theory, it is clear that there is very little connection between relative debt growth rates and interest rates, and there is no evidence of firms significantly leveraging up on debt when rates are low (either real or nominal rates, short or long rates).  Except for recent periods of cyclical disequilibrium, debt levels have remained at a relatively stable proportion of operating profits over a long period of time.

In fact, due to the relative stability of debt levels, I have pointed out that the changing leverage comes about mostly through changing equity values.  So, generally when stock market levels are high and PE ratios are high, this is not a product of frothy markets fed by debt-accumulating corporations.

I recently was looking at Robert Shiller's CAPE ratio, and noted that its fluctuations follow a very similar pattern to the inverse level of interest rates and the inverse level of leverage.  In other words, when CAPE is high, leverage (Debt to Enterprise Value) tends to be low and interest rates tend to be low.  But, you might notice some circular logic here.  Enterprise value is a product of price.  Since debt levels are fairly stable, the change in leverage comes mostly from a change in enterprise value, which is mostly going to come from a change in share price.

So, how do we know that this isn't just an accounting identity?  To start, as I mentioned above, debt doesn't have to be stable.  In fact, I think it probably bucks conventional intuition that debt levels don't rise relatively when interest rates decline.

There is a kind of a paradox here.  Because, when corporations manage their capital, which partly entails minimizing their weighted average cost of capital (WACC), they use the market value of their equity to estimate costs.  In other words, they are looking at the cost of debt and the cost of equity with the same Debt to Enterprise Value model that I am using here.  They should be targeting their debt levels based on relative market values and costs of their debt and equity.  And, empirically, the pattern of D/EV matches what we would expect to see from that model of capital management.  If so, how are all of the adjustments to aggregate capital structures being carried by changes in equity, which is mostly a product of changes in equity share price?

I think we have to think about corporate capital as having many competing constraints, and the level of debt as a multiple of NOPAT probably hints at a lot of those constraints.  Here is the standard cost of capital formula:

If we assume away some market frictions, the cost (K) of each form of capital should reflect the same set of risks, so that, theoretically, except for the tax consequence, a firm should be indifferent about the capital mix.  This is the Modigliani-Miller theorem.  But, I think what we see in the graph above is that, in actual capital markets, in the aggregate, the cost of debt starts to accelerate when debt gets above 5 or 6 times net operating profits after tax (NOPAT).  If these constraints are in play, then, even if a firm is managing capital with the WACC model in the current tax context, with dynamic capital costs, then the absolute level of debt compared to output will tend to settle in a tight range regardless of more general risk premiums.  These other constraints, which affect K(d) as leverage increases compared to NOPAT, will make EV/D ~ PE look like an accounting identity.

In fact, it appears that there was a slight shift upward in the mid-1980's, to a slightly higher debt/NOPAT ratio.  And, using the implied corporate interest rate from the Federal Reserve Z.1 nonfinancial corporation financial tables (interest paid / Credit Market Instruments), there appears to have been a corresponding increase in credit spreads (the black line).  This could reflect any number of changes, including what appears to be an increase in average debt maturities.  This also happens to coincide with the development of the high yield bond market.  So, there may have been a one-time shift from financial innovations that allowed firms, in the aggregate, to increase debt/NOPAT levels.  But, it appears to be associated with a steep rise in K(d).

When low rates lead to a tendency for deleveraging, firms may engage in some pure capital management - buying down bonds, issuing new shares, changing dividend or buyback policies, etc.  But, these moves have expenses that may not be justified over the long term.  When equity becomes such a large portion of enterprise value, incremental increases in operations can have a large effect on share value.  So, even though debt will tend to remain at the maximum optimum level relative to NOPAT, firms will have incentives to increase NOPAT through operations management - more efficient use of working capital, etc.  In this way, NOPAT and debt levels can grow, and the high relative equity value will mean that this operational growth will be especially multiplied in the total value of the firm.  The original NOPAT expansion will create value, and then the expansion of debt made possible by the NOPAT expansion will create additional value.

So, we should expect to see NOPAT growth during times of low interest rates and low corporate leverage.  This is what we find.  NOPAT grows when leverage is low.  (Keep in mind that leverage was especially low in the late 1990's because of excessive growth expectations, and the high points after that were times of disequilibrium during demand crises - notice that leverage peaks when NOPAT crashes.  Firms quickly deleveraged as demand recovered.  So, cyclical fluctuations have been especially high during this period, but cyclically adjusted leverage has been close to 30% throughout this time.)

I am sure there are other narratives that can be built around other hypotheses, so this information doesn't create any exclusivity for my hypothesis.  But, the data can support my hypothesis.

Interestingly, GDP hasn't followed the trend of NOPAT.  Before the 1970's, real GDP was growing quickly.  And, real GDP growth leveled off during the high rate period, though not as sharply as NOPAT, understandably.  But, instead of reaccelerating, GDP growth has continued at that lower real growth rate as interest rates have declined.  I suspect there is a combination of innumerable explanations for this, including corporate foreign profits, demographic consumption and saving patterns, and maybe even issues with the difficulty of measuring consumption value in the age of the internet.

It's a shame that this sort of subject tends to be reined in by us vs. them thinking regarding corporate profits versus other household income, because there are probably many interesting things going on here - many of them reasonable and helpful for general prosperity - that would only be visible to the curious and non-judgmental observer.

My Attempt at a Curious & Non-Judgmental Interpretation:

For the investor, the real NOPAT chart shows how there is plenty of room for more growth in corporate valuations, simply coming from recovery in NOPAT.  At this point in the cycle, equities probably still offer a decent expected return without depending on expansion of valuation multiples.  (If the expansion continues, allowing for a recovery in interest rates, equities will also benefit from corporate re-leveraging as rates increase and enterprise value expands.  This would allow more NOPAT expansion and equity growth even as some valuation multiples retract.)

Regarding GDP, I suggest that the housing "bubble" was a reasonable and predictable result of the baby boomer phenomenon.  Boomers are utilizing real estate as a chimera of consumption and savings.  Homes are storing value for future consumption.  From 1997 to the peak in 2006, $8.3 trillion was accumulated in real estate equity.  The rise of home values tends to be blamed for creating unsustainable spending.  I believe this has it backwards.  Mortgages rose less than equity did.  Homes, on net, were replacing spending, not funding it.

While this seems like a reasonable phenomenon to me, it isn't particularly efficient.  The homes are storing value for future consumption, but this is only partly through home production.  Much of the savings simply creates a temporary nominal increase, just as if a rush of savers bids up the price of bonds.  While the homes are a store of value, they don't serve as a basis for economic growth and productivity.  Saving through the foreign operations of US corporations seems like a more efficient means of consumption smoothing.  But, both are inevitable.

I don't think either of these savings vehicles are picked up particularly well as either consumption or savings, so since the mid-1990s, when the baby boomers started entering this phase, GDP has been understated.  And, there has been a real loss in potential GDP growth, due to the allocation of capital to real estate in lieu of more dynamic productive assets.  (On the other hand, there is no unironic way to complain about stagnation and low productivity on a blog.)  In effect, we have all been waiting for the productivity slowdown coming from the aging boomers.  But markets are forward looking, and to the extent that trades through time are available (as they are through real estate), economic trends will be traded and shared through time.  Markets are simply allowing for an exchange with the future, which is manifest as a seemingly premature slowdown in GDP.

(Some people argue that home prices should be factored into inflation measures.  They are dangerously wrong.  But, I will note that if we did count home appreciation as inflationary GDP growth, nominal GDP growth would have been much higher in the 2000's, with most of the increased nominal income going to middle class households.  But, such is the mania behind our current malaise, that broad nominal appreciation of the most widely held durable asset in the economy is vaguely accounted for only as a cost.  So, the colloquial story of the 2000's goes something like:  high home prices led to unsustainable consumption that created a false level of nominal GDP.  And real GDP was much lower than reported, because inflation would have been much higher if home prices were accounted for in inflation measures
.......this is mistakes on top of mistakes on top of mistakes, believed with religious fervor.  I am also frequently told, with the same conviction, that the falling labor force participation of 25-54 year olds is a sign of economic desperation.  I am also frequently told, with the same conviction, that the rising labor force participation of 55+ year olds is a sign of economic desperation.  Religious fervor is helpful to believe such things.  Likewise, as we all know, US corporations are unpatriotically moving out of the US to avoid taxes.  And, we all know, actual US corporate taxes aren't high at all.  They have so many tax breaks that their effective tax rates are much lower than the headline rate.  Both things, believed with fervor.)

Since 2006, the Fed has devastated the real estate savings that the baby boomers had accumulated.  With another 5-10% price accumulation, homes will probably be back near a healthy equity/mortgage balance, at which time, after an 8 year (wow) pause, boomers can accumulate marginal new real estate equity.

Of course, wide swaths of the American public, from all sides of the political spectrum, are explicitly calling for the disruption of both of these vehicles for deferred consumption - calling corporations with operations abroad unpatriotic and warning of the developing new housing "bubble".  Many of these outspoken critics of this phenomenon are boomers themselves, who, no doubt, in their personal financial dealings are holding large positions in real estate and multi-national corporations, specifically to prepare for their future retirements.

I've said it before.  H.L. Mencken was a Pollyanna.

PS.  It also happens that the one-time increase in female labor force participation coincided with the high leverage period.  Female LFP peaked and began to trend down along with male LFP coincidentally when leverage declined in the 1990's.  So, part of the dilemma may arise from this.  This may have caused the 1960s-1980s GDP to be unusually high, while this + the age-demographic decline in LFP caused the 1990s+ GDP to be unusually low.

Added:  I should have done this before I posted.  Here is the GDP graph, adjusted for the size of the labor force.  This has a shape much like the graph of NOPAT, except, instead of rising and falling inversely with leverage, GDP/LF rises and falls with Enterprise Value.  This is affected by the Equity Risk Premium (ERP), in addition to risk free interest rates, so GDP growth stalls and returns a little earlier in the 1960s-1980s period, because ERP declined before interest rates did.  (Note: the trend lines on the NOPAT graph were mathematically derived, but I just eyeballed these.)

Tuesday, August 26, 2014

More Non-Evidence of Part-Time Work Shift

I was thinking about labor supply and demand, and the anecdotal evidence from employers that they are shifting work to part time because of the ACA.  I haven't seen definitive evidence of a significant shift to part time work in the data for number of employees.  Here is the chart of part time employment from yesterday's post.

After an initial shock and a shift from "non-economic" to "economic" reasons, the total number of part time workers has been moving down to about 1% above the pre-recession level.  This is typical for a recession.  The size of the shift up in "economic" part time workers, and the slow subsequent decline are unusual, but these trends pre-date the ACA, so it seems inaccurate to pin much if any of the trend on the ACA.

I have posited that one underappreciated factor here is that the binding constraint here may be the supply of willing part time workers.  But, if that is the case, if employers are experiencing higher costs for full time workers, then we should see an increase in pay for part time workers, as labor supply and demand settle at a new equilibrium that accounts for the new cost structure.  If the quantity of available part time workers is relatively inelastic, maybe this adjustment would mainly play out in wages.  Employers would need to increase the wages on part time jobs to entice workers who preferred full time work to move to a suboptimal work scenario.

But, there doesn't appear to be anything here, either.  Relative weekly wages for full and part time workers have moved in lock step over the past 14 years.

So, it still seems like there is a disconnect between the survey data, anecdotal evidence, and labor data.

Monday, August 25, 2014

ACA Survey from the Philadelphia Fed

Here is a survey from the Philadelphia Fed with some interesting answers regarding the ACA (HT: Patrick Sullivan).  You can click on the links for details.  The gist of it is that the manufacturers reported that, as a result of the ACA, they have decreased the number of full time workers, increased the number of part time workers, raised prices on their products, raised contributions, premiums, and deductibles on their employee health plans, and reduced coverage.

First, a caveat.  It seems to me that the jury is still out regarding part time employment.  There is a lot of anecdotal and survey evidence of this effect, but I don't see strong evidence of it in the data.  Now, it could be that there are a number of supply and demand factors at work in the part time employment realm.  I think that as the economy recovers, the binding constraint on the total level of part time workers will be the supply of part time workers.  In the meantime, I don't see any obviously odd movements in part time employment.  I'm not sure how to square this with employers reporting that they are transitioning to part time work.  There is a lot going on there down in the weeds.

Here is the graph of part time workers, as a proportion of the labor force.  The sharp rise in part time workers for economic reasons clearly happened during the recession, pre-ACA.  Possibly the recovery could have been sharper.  But, I don't see any smoking gun here, regarding an ACA-related jump in part time employment.

The average weekly hours worked from the monthly establishment survey also doesn't seem out of line compared to long term trends.


from the BLS
This all is a reminder of the difficulty of accounting for the costs of these sorts of things.  There have been a few policies that I have harped on, such as minimum wage hikes and Extended Unemployment Insurance, with specific claims about how they harmed workers.  But, in those cases, the policies had a very pointed and direct effect on some specifically identifiable people, so we can identify some sharp empirical evidence about their effects.

But, with so many regulations, like the ACA, the effects are tough to suss out.  Any effects there might be probably just slightly bend the curve of growth of human quality of life, and the damage comes in the long term.

Most of the changes the employers point out probably are accounted for as increases in nominal GDP.  There is a lot of economic activity engaged in compliance activities.  And, for regulations like this, it comes down to how the hedonic changes in price levels are accounted for.  When you lose access to your family doctor, does the BLS account for this as a decrease in quality?  Frankly, in an industry so screwed up, I don't understand how they can begin to account for economic activity.  What is the market price for a medical procedure that is "billed" at $2,500, but settled with your insurance company for $200, with a $20 co-pay?

It seems to me that much of the added costs of the ACA, the results of this survey being good examples, will play out as inflationary.  But, on the other hand, something like secondary education isn't accounted for at all in inflation measures.  That kind of government expenditure is simply recorded as real economic activity.  (Please correct me in the comments if I'm wrong.)

So, while GDP is useful for much analysis, with government expenditures at nearly half of GDP and a growing number of mandates, it ceases to be a good measure of absolute quality of life.  Some measure of the quality of public services and mandates would be required.  I'm not aware of any measure that can remotely address that.  A problem with programs like ACA is that they muddy the usefulness of indicators like GDP.  I'm not sure they even effect growth rates coherently, except in the very long term.  Government programs certainly have a knack for growing, which will usually show up as real growth in the GDP.  The same goes for mandate-based rent-seeking.  The production of ethanol, which probably put millions of acres of US farmland to use with no net positive economic benefit, added $44 billion to GDP in 2013.  (And created nearly 400,000 jobs!)

In 2012, total US government expenditures (including transfers) amounted to $47,159 per household.  The median household US income in 2012 was $51,017.  The median US household is left with $3,000 to spend on government mandates and all private personal consumption (housing, food, transportation, etc.)  To the extent that the median household engages in any private consumption at all, it is because our progressive tax code has essentially transferred those resources from high income households to median income households.  What is the value of that $47,159 of expenditures?  Could it possibly be anywhere near the amount spent?  Does it really take essentially all of the labor of the median US household just to support public services?  How can essential public services (not including off-budget mandates) require more expense, per capita, in real dollars, than the entire pre-1960 US economy?  Some people argue that we should return to 1950's tax rates.  If it is the essential services and social support of the 1950's that they want, shouldn't they argue for a return to 1950's public budgets?

Saturday, August 23, 2014

Minimum Wages, Crime, and Unemployment

Results from longitudinal panel data regarding the effects of minimum wage hikes on people who were working at minimum wage level jobs.  Appears to lead to unemployment and crime, especially among teens.  (HT: CafeHayek):
Does crime respond to changes in the minimum wage? A growing body of empirical evidence indicates that increases in the minimum wage have a displacement effect on low-skilled workers.  Economic reasoning provides the possibility that dis-employment may cause youth to substitute from legal work to crime. However, there is also the countervailing effect of a higher wage raising the opportunity cost of crime for those who remain employed. We use the National Longitudinal Survey of Youth 1997 cohort to measure the effect of increases in the minimum wage on self-reported criminal activity and examine employment-crime substitution. Exploiting changes in state and federal minimum wage laws from 1997 to 2010, we find that workers who are affected by a change in the minimum wage are more likely to commit crime, become idle, and lose employment. Individuals experiencing a binding minimum wage change were more likely to commit crime and work only part time. Analyzing heterogeneity shows those with past criminal connections are especially likely to see decreased employment and increased crime following a policy change, suggesting that reduced employment effects dominate any wage effects. The findings have implications for policy regarding both the low-wage labor market and efforts to deter criminal activity.

Friday, August 22, 2014

Readings on Fed control and risk premiums

Eugene Fama, The Review of Asset Pricing Studies, from December 2013:
In sum, the evidence says that Fed actions with respect to its target rate have little effect on long-term interest rates, and there is substantial uncertainty about the extent of Fed control of short-term rates. I think this conclusion is also implied by earlier work, but the problem typically goes unstated in the relevant studies, which generally interpret the evidence with a strong bias toward a powerful Fed.

An interesting attempt at isolating compensated equity risk, using a measure called "Excess Conditional Value at Risk":
Research that has led to the low-volatility anomaly in cross-sectional stocks from a similar universe indicates that volatility is not compensated with a volatility premium. The authors find evidence of a risk premium, but it depends on the definition or measure of risk. Tail risk measures the probability of having significant losses, and should be what investors care about the most. This article investigates several risk measures, including volatility and tail risk, and finds that volatility is not compensated. Tail risk, however, is compensated with higher expected return in both U.S. and non-U.S. equity funds.

Thursday, August 21, 2014

Extremely Positive News on Economic Mobility

Richard Reeves from the Brookings Institution has released this video about income mobility (HT: EV).



I suspect that the statistics he cites will be a very positive surprise to most people.

For all people born in the bottom income quintile, only 1/3 will remain there.  For a statistically average poor family with three children, one child will grow up to remain in the bottom quintile.  One child will move out of the bottom, but will have income below the median, and one child will have an income in the top half.

For black Americans, in particular, the outcomes are not this good.  But, even for black children born in the lowest quintile, half will move into a higher quintile.

For white Americans born in the lowest quintile, amazingly, the distribution of incomes is roughly flat across quintiles, with the odds of ending up in any quintile falling between 16% and 23% - near to what Reeves calls an "opportunity utopia".  That almost seems unbelievable.  That utopia would be a worthy goal, but I would never have guessed a society could come this close.  With perpetually poor places like Appalachia, or with areas characterized by vastly different standards of living, from West Virginia to Connecticut, how could this be possible?  But, apparently, it is.

Then, he shows the results for children born to parents who are never married versus those born to married parents, with no adjustment for race.  Remarkably, the results are very similar.

For poor children of never-married parents, about half will move to a higher quintile.

For poor children of married parents, again, they are in an "opportunity utopia".  In fact, for a child born into the lowest quintile to married parents who stay married, the lowest quintile is the quintile they are least likely to end up in.  I repeat, for a child born into the lowest quintile to married parents who stay married, the lowest quintile is the quintile they are least likely to end up in.  For children of parents who are married, this is better than a utopia.  Do we even have a word for that?  Is America dreaming?  I wouldn't know how to dream of something better than utopia. But for people who are married when they have children and stay married, we are living in it.

A child born into the bottom quintile to married parents who stay married has an 83% likelihood of moving up to higher income quintiles!

The result is the same regarding education.  If you are born poor and you drop out of high school, you have about a 46% chance of moving up.  If you are born poor and you graduate from college, you are living in the uber-utopia.

Here are a couple of graphs from the report:

This is the distribution of adult incomes of children born into families with married parents.  That is amazing.  And, in the report, the charts are interactive, so you can click on "Never Married" and then click on "Continuously Married", and watch the probabilities change.  And, despite the tone of the accompanying narrative, it is amazingly clear that the marital status of the parents and the education level of the children are much more important than the quintile the children are born into.  College graduates who were born into the first quintile have much better outcomes than high school drop outs who were born into the fifth quintile, for instance.

I haven't seen the raw data, so I don't know how much of the difference in the racial outcomes is due to family structure and education.

Next is a graph showing the amount of time parents spend with their kids, based on the parents' education.  The report points out that a gap has developed between the more educated and the less educated parents.  But, I can't help but noticing that parents with a high school education or less spend nearly 3 times more time with their children than the average parents did 30 years ago, regardless of education!  That is incredible!  As a nation we should be very proud of this development!

On the other hand, some people look at this graph and bemoan that educated parents spend almost 40 minutes more with their children than than less educated people.  This can only be described as a pathological reaction.  It would be interesting if someone set up a study where they presented made up social economic statistics with a helpfully negative tone.  How positive would the reported outcomes have to be before the subjects reached a consensus that inequality wasn't a high priority problem.  The issue has such value because there may be no threshold high enough for it to cease to be a purported issue.  In the report, this graph is shown in order to visualize the "gap".  Nothing positive is said about the trend.  It's as if the reader is expected not to see anything positive going on.  The extreme rise in parental attention among both groups is like the ape on the basketball court.

Is there any way to square the image of poor families ever more desperate to hang on to their disintegrating share of hope with this finding that the most vulnerable parents spend nearly 3 times more time with their kids than educated parents from 30 years ago did?  If spending time with kids is important, then why are outcomes for poor children worsening even while their family time skyrockets?  Are poor kids worse off because children of educated parents have more parent time?  Is there a fixed-pie of parental attention, such that, when educated parents spend time with their children, children of less-educated parents are harmed?  Can someone pencil that out for me?  How did I even end up with a functional life when my own parents barely even noticed me, by today's standards?

The video ends with, "We have a big problem, and we need big solutions."

Wednesday, August 20, 2014

July Inflation Update

Here are updates on a couple of charts I introduced in this post. 

I have been looking at core inflation without shelter.  To the extent that very low long term interest rates are driving home prices, shelter inflation may be a countercyclical indicator.  (To be clear, I think rent equivalent is the appropriate inflation measurement to use.  The notion that a context of moderate rent inflation, very low interest rates, and rising home values is inflationary is dangerously wrong.)  In this context, a crippled credit market is artificially curbing demand for homes, reducing the supply of new homes.  The effect is that an undersupply of homes is leading to rent inflation.  As credit markets loosen, home supply will respond.  Oddly, in this way, demand recovery will be a supply recovery.  Recovery of real estate credit markets may actually suppress nominal GDP relative to real GDP, as it is (correctly) measured.

In the first graph, we can see that core-less-shelter inflation swung down.  All the net core inflation in the month came from shelter.  This is the second month of moderating inflation after several months of increasing inflation, which clouds the picture of where inflation my be leading as QE3 ends.

The continued strength in shelter inflation suggests that we are still early in the credit market recovery.  The real estate recovery should create a bit of a virtuous circle, as the continued re-inflation of property values feeds back into credit creation and reduces mobility frictions.  As long as this is allowed to continue, home prices should continue to rise, home supply should increase, and rent inflation should moderate.

The strong level of rent inflation is a sign of a recovery in household formation, which is a signal of the strengthening underlying economy.  Bank credit continues to show strength as the QE3 taper winds down.  The first half of 2014 has been the first time since the crisis that bank credit has been expanding in quantities similar to pre-crisis levels.  However, most of the increase has been through very strong Commercial and Industrial Loan growth.  As real estate credit continues to recover, real growth should continue to get a goose.  I will look at this some more in follow up posts.