August payrolls – in line, except for awful AWE

A James Alexander post

The auguries for the August payrolls have not been good judging by industry-side surveys of August activity, and they still disappointed expectations.

Monthly jobs growth was relatively weak. With the participation rate flat there were not enough jobs created to keep pace with entries into the labour force and so the unemployment rate ticked up to 4.9%. No big deal and certainly nothing to move markets or expectations about Fed action.

What should trigger Fed action and more concern generally is the very weak Average Weekly Earnings (AWE) number. It is derived from two more commonly watched numbers, Average Hourly Earnings (AHE) and Average Weekly Hours (AWH). Hourly earnings had made some progress over the past 18 months, rising from a risible 2% YoY growth to a slightly less risible 2.5% or even 2.6%. Hawks had gotten very excited seeing incipient take-off in wage inflation, especially when annualizing a 3-month trend etc. More careful analysis showed that this very modest growth had been accompanied by lower hours worked per week, thus suppressing AWE growth to just 2% or so.

JA Weekly090216

The August data showed both weaker growth in AHE and another drop in AWH (plus a revision down in the July hours) leading to weekly earnings growth dropping from 2.1% YoY to just 1.5%. It has to be remembered that this is nominal growth and so really depressing for wage earners. Other non-wage costs are rising, like medical cover but it remains incredibly dull for employees.

Incredibly, and perhaps tellingly about the market’s view of the FOMC, the chances of a September rate rise stayed put at 24% but the chances of a December hike rose from 54% to 60%. The market is always right, after all. Both equities and the USD initially fell, but then went up, as might have been expected. Even the long bond yield initially rose, before paring back.

A “strong” employment report in a “weak” economy?

The headline numbers for February: 242 thousand jobs and 4.9% unemployment rate.

Let´s give these numbers some “structure”. The unemployment rate is the result of two forces that reflect economic decisions by individuals and firms. The first is the employment population ratio (EPOP). The second the labor force participation ratio (LFPR). The unemployment rate is equal to 1-(EPOP/LFPR).

The charts show the behavior of unemployment together with its two determinants over the three most recent cycles. The first two (1990, 2001) happened during the “Great Moderation”, while the third (2007) takes place during the “Depressed Great Moderation”.

Employ Report 02-16_1

The green bars denote recessions (as determined by the NBER). The yellow bars denote periods of falling unemployment.

A “healthy” fall in unemployment occurs simultaneously alongside a “strong” rise in EPOP and a “moderate” rise or stable LFPR.

In the present cycle, the fall in unemployment reflects a stable and then “moderate” rise in EPOP and a falling LFPR. It´s another “animal” altogether!

Many say that this reflects mostly structural/demographic changes, like baby-boomers’ retirement. The coincidence in time of the structural/demographic factors with the onset of the “Great Recession” that witnessed the largest drop in nominal spending (NGDP) since 1937 is “too much to swallow”. More likely strong cyclical factors were responsible for bringing forward in time decisions that otherwise would have taken place over the next several years. In this sense the problem is mostly “cyclical”.

Although the economy is still adding jobs at a rate that could be called “healthy”, the relatively low quit rate (which tends to rise when employment opportunities are “bountiful”), the relatively high duration measures of unemployment (indication that opportunities are not “bountiful”) and the relatively weak (if you take into consideration the depth of the employment drop) job growth, indicate that the labor market is some distance away from having fully recovered. In other words, like the economy, it´s still weak!

More evidence on the “special” nature of the present (2007-) cycle.

We start with the chart for NGDP (peak to peak over the cycles)

Employ Report 02-16_2

The 1990 and 2001 cycles are, except for length, identical. The present cycle suffers from lack of spending.

That´s mirrored in the behavior of RGDP

Employ Report 02-16_3

Now, when you look at the behavior of employment, the 2001 cycle shows a “surprising” result. From the behavior of NGDP and RGDP one would expect it to also mirror what happened in the 1990 cycle.

Employ Report 02-16_4

But when you look at the behavior of productivity, the employment difference becomes understandable. The 2001 cycle was “productivity rich”.

Employ Report 02-16_5

The present cycle, on the other hand is both “employment poor” and “productivity poor”. In addition, it is also “price & wage poor”

Maybe it´s not farfetched to associate much of the overall “poorness” of the present cycle to the sorry state of nominal spending, i.e. aggregate demand!

The story behind the “scariest jobs chart ever”

Bill McBride (Calculated Risk) posts Update: “Scariest jobs chart ever” where he presents a version of this chart

Scary Chart_1

And writes:

This graph shows the job losses from the start of the employment recession, in percentage terms, compared to previous post WWII recessions.  Since exceeding the pre-recession peak in May 2014, employment is now 3.5% above the previous peak.

Note: I ended the lines for most previous recessions when employment reached a new peak, although I continued the 2001 recession too.  The downturn at the end of the 2001 recession is the beginning of the 2007 recession.  I don’t expect a downturn for employment any time soon (unlike in 2007 when I was forecasting a recession).

By choosing to break with the way he presented the chart originally, with lines ending when the previous employment peak was reached, he ‘pollutes’ the chart, and detracts attention from an interesting characteristic.

Note that the 2001 Employment Recession is one of the shallowest, only less so than the 1990 Employment Recession; but the second most persistent, only less so than the present Employment Recession.

Interesting question that comes up from eyeballing the chart:

What accounts for the depth and persistence of the employment recessions?

The short answer: monetary policy, whose stance is well described by the growth of NGDP.

The chart below is a powerful illustrator.

Scary Chart_2

The 1981 employment recession was deep. The fall in NGDP growth was high. The objective of monetary policy at the time was to bring inflation down significantly. It came from double digit rates to the 3%-4% range. The intensity of the employment pick-up is commensurate with the strength of the rise in NGDP growth.

In the 1990 cycle, monetary policy was slightly less tight than in the 2001 cycle and NGDP growth increased sooner, making the 1990 cycle a little less deep and less persistent than the 2001 cycle.

The 2007 cycle is the clincher, and “living proof” of the responsibility of monetary policy in generating the “Great Recession”. Note that for the initial months/quarters since the cycle peak, the fall in employment during the 2007 cycle was par with the fall in employment in the 1990 and 2001 cycles. As should be expected, the behavior of NGDP growth during this stage was very similar in the three cycles.

But suddenly the Fed makes the second largest error since its inception in 1913. The massive drop in NGDP growth, which remains for an “eternity” in negative range, “destroys” employment. The timid monetary pick up, both in relative and absolute terms, fully explains the persistence of the employment recession in the present cycle.

Decades in the future, when the history of the last 30 years will be written by dispassionate researchers, one interesting footnote will surely describe the “NK Trip”, where NK does not stand for “New Keynesian”, but for Narayana Kocherlakota and will show that the “last step” in the “trip” completely denies the “first step”. Furthermore, the “ingredients” deemed important in the stories told today: the house price boom & bust, “greedy bankers”, “spindrift consumers”, will be seen as bit players, in part victims of the monetary “mayhem” brought on by the Fed!

The Fed must be happy. The employment numbers “confirm their view”!

From the WSJ “Economists react”:

“For the [Federal Open Market Committee], the unexpected 292,000 jobs added in December, and the upward revisions to October and November, will bolster their case for four rate hikes this year. Markets are desperate for good news, and now have it, but it comes with a caveat because better employment growth means the Fed will raise rates faster.” —Beth Ann Bovino, Standard & Poor’s Ratings Services

From Bill Gross:

The Fed does believe that jobs and the unemployment rate is critical to future inflation over the medium term,” Gross said. “So the three or four Fed steps that Stan Fischer and Janet Yellen seem to confirm are probably on track, at least in their verbiage.”

They are right. That´s the way the Fed thinks. But it is dangerous and wrong! Insisting on the mistake will likely lead to another recession. Larry Kudlow (HT Lars Christensen) also rants on the Fed´s mistaken view.

Fed should not make policy based on labor: Kudlow

Some illustrations for the past 22 years will help identify the mistake.

The first chart shows no connection between unemployment and inflation. While unemployment has “travelled widely”, core inflation has remained low and stable (on average below the 2% target)

Fed Happy_1

Note how inflation and unemployment reacted to the productivity shock of the late 1990s. Note also how the negative AD shock of 2008 affected unemployment and inflation.

Many at the FOMC like to appeal to “special causes” for the low inflation. A favorite is oil prices. However, the next chart shows that while oil prices have “jumped & tumbled”, core inflation has remained subdued.

Fed Happy_2

Note that in 2007-08 the economy was buffeted by both a negative AD shock and a negative supply (oil) shock. In that case both the AD curve and AS curve shift to the left. This solves the “puzzle” frequently mentioned of “why didn´t inflation fall even more”. The negative AS shock “supported” inflation. However, the negative AD shock was so strong that it ended up “annulling” the supply shock, with oil prices tumbling, at which point inflation dropped.

And Lacker is worried about an “inflation conundrum” (title has changed)!

Now, let´s look at things from a NGDP perspective.

Fed Happy_3

Much better, I think. The productivity shock after 1997 reduced unemployment and inflation. Keeping monetary policy (NGDP growth) steady was key to the outcome. However, in 1999/00, NGDP growth was a bit excessive and the reaction resulted in too much tightening. In 2003-05, the mistake was corrected, with unemployment falling and inflation remaining close to target.

Bernanke reacts to the oil shock by tightening monetary policy (allowing NGDP growth to fall unadvisedly (“let´s ball drop”). The mistake was compounded by “losing the ball”, after which the “net was permanently lowered” i.e. NGDP growth never rose sufficiently to take spending to the previous trend level.

Additional information is provided by the NGDP growth, RGDP growth and oil prices:

Fed Happy_4

During the first leg of the oil shock in 2002-05, monetary policy was “expansionary”, to offset the previous “tightening”. Note from the second chart from the top, that inflation remained well behaved. Despite the oil shock, real output growth rebounded. Bernanke “dropping the ball”, combined with the second leg of the oil shock, strongly decreased real growth. When the “ball was lost”, real growth tanked. The price of oil tumbled as a result of the strongly negative AD shock.

The oil shock effect on output was something about which Bernanke had written in 1997, concluding:

Substantively, our results support that an important part of the effect of oil price shocks on the economy results not from the change in oil price per se, but from the resulting tightening of monetary policy.

This finding may help explain the apparently large effects of oil price changes found by Hamilton (1983) and many others.

He certainly proved his point in 2007-08!

What the above discussions tells me is that monetary policy should not be informed by the rate of unemployment, what´s happening to oil prices or whatnot. A focus on NGDP growth relative to a “desired trend path” is all that´s needed.

The Fed seems to be making the same error again, this time by focusing on unemployment instead of oil prices.

Ever since the NGDP growth “net” was brought (insufficiently) up in 2010 and until mid-2014, trend NGDP growth was close to 4%. Since then, it has trended down dangerously:

Fed Happy_5

So it´s not at all surprising to see real and financial indicators also going south. Just to give two examples; the Institute of Supply Management diffusion index (ISM) and long term inflation expectations from the 10 year breakeven.

Fed Happy_6

If these trends continue, and the Fed seems set on keeping them going, the likelihood of a recession will be increasing as the year progresses.

Update: Stephen King gets it right:

Conventional wisdom suggests lower oil prices should provide a shot in the arm for the global economy. In current circumstances, however, lower oil prices may simply be providing over-enthusiastic central bankers with an opportunity to shoot themselves in the foot.

Give the Fed a new compass. We´re going in the wrong direction

According to the news:

Friday’s employment report clears the way for the Federal Reserve to raise short-term interest rates by a quarter-percentage point at its Dec. 15-16 policy meeting, ending seven years of near-zero interest rates.

The Fed can reasonably well control nominal spending (NGDP) growth. Stable NGDP growth at the appropriate level well defines what good monetary policy is supposed to look like.

If that´s true, when NGDP growth falters, things like employment growth will register the “punch”, just as it will “blossom” when monetary policy pulls NGDP growth up. Stable NGDP growth goes hand-in-hand with stable employment growth (only thing is if NGDP level falls short, so will the level of employment)

Examples from the mid-1990s and early 2000s show the Greenspan years. For the last ten years, we have been under Bernanke and Yellen. The pictures are illustrative. (The montlhy NGDP numbers come from Macroeconomic Advisers)

Throughout the period, inflation was not a problem. By the mid-1990s, it had reached the “low and stable” target of the time. Ironically, after the numerical 2% target was set in January 2012, inflation has languished, but is still “low and stable”!

Employ Report 11-15_1

Employ Report 11-15_2

But if you zoom in on the past 15 months, things seem “fishy”. For all the Fed´s “communication”, the truth is that they have been tightening policy. NGDP growth is coming down which was shortly followed by decreasing employment growth. Won´t even mention inflation.

Employ Report 11-15_3

To wrap up, where´s the much touted wage growth-inflation nexus so cherished by some at the FOMC?

Employ Report 11-15_4

Great harm might be on the way!

PS If you don´t believe me about the “beauty” of stable nominal spending, believe George Selgin:

a central bank that allows the overall volume of spending to collapse has blown it, no matter how much emergency lending it undertakes.  Indeed, to the extent that a central bank engages in emergency lending while failing to preserve aggregate spending, it may be guilty of compounding the damage attributable to the collapse of spending itself with that attributable to a misallocation of scarce resources in favor of irresponsibly-managed firms.

The “eager beavers” at the FOMC must be disappointed

I gather this is true from comments by Bill McBride (Calculated Risk). For the last several months, his conclusion was always that “…this was a “solid” report”. Today´s report was just “decent”!

In fact, given that this was the 6th year of recovery employment report, it was dismal! One pointer, if the participation rate had not dropped 0.2 points the unemployment rate, instead of falling to 5.3%, would have climbed to 5.7%.

No mystery that despite the “low” unemployment rate, wages stayed put, and remained at their 2% line year over year. Given that the “eager beavers” are watching wage growth with “hawk-eyes”, they must feel “depressed”!

In his preview of the Employment Report yesterday, Tim Duy concluded:

Bottom Line: Incoming data continues to support the case that the underlying pace of activity is holding, alleviating concerns that kept the Fed on the sidelines in the first half of this year. I anticipate the employment report, or, more accurately, the sum of the next three reports, to say the same. Accelerating wage growth could very well be the trigger for a September rate hike, while Greece could push any rate hike beyond 2015. I myself, however, tend to be optimistic the Greece situation will not spiral out of control.

Seems he´s optimistic about everything, forgetting someone took out the “firing pin”. Shortly he´ll be talking December 15…March 16…

It was mind-boggling to hear this conclusion in Ed Lazear´s (a former CEA president) interview about the jobs report:

Interviewer: The numbers we have do not correlate with the zero interest rate policy. If you were at the Fed…

Ed: The numbers don’t give reason to raise interest rates but there´s no reason to keep rates low because that´s not helping the economy very much.