Though It Will Likely Be Short, Enjoy the Boom

The dot-com recession that struck in 2001 was the mildest cyclical trough on record. That didn’t prevent, however, a lot of official angst about the affair. First of all, there was that whole immoderate stock bubble sticking out for what would shortly after be called the Great “Moderation.” The small scale of that recession and the one that preceded it in 1990-91, more so the record expansion in between, was what led Economists toward patting themselves on the back.

Following the recession, Federal Reserve Chairman Alan Greenspan told Congress in early 2002 it was likely to get worse before it got better. “Even if the economy is on the road to recovery, the unemployment rate, in typical cyclical fashion, may resume its increase for a time.” The labor force tends to shrink during downturns once marginal workers get the sense there won’t be much work for them.

When the economy begins to heal, sometime after recovery becomes entrenched, these laborers step back into the labor force pushing the unemployment rate uncomfortably higher if only temporarily. But as Greenspan warned in his testimony, this optimistic if counterintuitive rise in the unemployment rate could last for a significant length of time.

What was even more interesting than the general dynamics of labor recovery was some of the specifics. In one contemporary New York Times article describing Chairman Greenspan’s cautious approach, it was noted that these subtleties affect different groups differently.

“Mr. Greenspan, the Federal Reserve chairman, who will appear before the Senate Banking Committee today, did not mention that the lingering effects of high unemployment early in a recovery tend to be concentrated among the unskilled and minorities. This is true even though recessions are becoming more egalitarian.”

This is a phenomenon that economists have tried to explain for decades. In 1955, Melvin Reder proposed one intuitive account. As the economy contracts and companies find it difficult to maintain both their top and bottom lines, they increase skill requirements for any given job rather than cut pay. It was one reason why we have recessions instead of depressions after the 1930’s, as businesses don’t slash wage rates in response to economic events like they once did.

But they have to offset these negative pressures somewhere. For a very long time that has meant the unskilled are left at the end of the line. Normal cyclical forces, which eventually involve this shrinking and then expanding labor force, can account for why untrained labor fares substantially worse under macro scenarios.

As that same New York Times piece suggested more than sixteen years ago, “For this reason, economists often prefer to focus on the employment-to-population rate -- the fraction of the population that is working.”

Economists would prefer you not use that metric any longer. If you did, you would have a hard time with this current economy, starting with how the Great “Recession” doesn’t appear to have been a recession at all. This particular ratio fell and never really came back. Unlike the unemployment rate, a low employment-to-population ratio means businesses in general are using much less labor.

Not once has the unemployment rate behaved as it had during prior business cycles. The labor force contracted, and stayed that way.  Recovery, as should have been the case, was never announced by that expected rise in the unemployment rate when marginal, unskilled workers flooded back in. With the scale of the 2008-09 contraction wiping out so many jobs, that recovery flood would have been epic.

Instead, the unemployment rate just kept falling and falling and falling.

There have been all sorts of absurd explanations put forward for why this time is different. None of them, of course, involve really explaining what it is that is different.

In terms of the labor market, Economists have taken to drug use, meaning that they see the opioid epidemic as souring the American labor force to some large enough extent. This is in addition to Baby Boomers who are reaching retirement age without enough young workers to replace them.

More than those, however, the conventional view is that there is some giant skills mismatch out there. You have to appreciate the irony here, especially in how Economists in 2002 including their chief Alan Greenspan were expecting unskilled labor to bear the brunt of both recession and early recovery. It wasn’t until the economy actually recovered that unskilled labor would be able to rejoin the labor force.

This time around, quite the contrary, we are led to believe that for some reason these particular workers can’t nor won’t be brought back into the economy at all. They certainly haven’t been to this point, and it’s been eleven years already.

They call this a skills mismatch, but there has always been one to some degree. That was largely Greenspan’s point and one further brought forward in the pages of the New York Times.  If this time is different, then it isn’t the general skill level of discrete pieces of the labor force.

The unemployment rate is used way beyond its traditional format. At some level, Economics still believes in the Phillips Curve; that is, there is a trade off between unemployment and inflation. The relationship is far more complex than what was done, improperly, in Mr. Phillips’ name.

Briefly, there is today thought to be some dynamic equilibrium point at which slack in the labor market is fully absorbed by economic growth. Employers have to more fiercely compete for workers, which then pressures wage rates, leading eventually to broad-based gains in consumer prices as companies attempt to pass along those sharply higher labor costs to consumers.

The amount of slack is determined by those potential workers sitting idle on the sidelines. If there are millions, wages won’t rise very fast or far because labor supply would still be more than sufficient to meet labor demand. If there aren’t millions, for whatever reasons including fentanyl, retirees, and this unique skills mismatch, then wages should rise quickly.

And if the economy happens to be booming while the labor market is in this latter position, wages and pay should explode upward in an out and out bidding war for marginal labor.

It is not happening. It’s not. Categorically, all the data we have says wage and pay rates are still historically depressed. Wages are rising, of course, but since the 1930’s they always rise even during the worst recessions. They are not accelerating as they would if the economy was operating at true full employment.

Instead, what we find in labor market data is the condition that was described well enough by 2002 commentary. The unskilled sit there idle still waiting their turn while Economists declare a recovery that doesn’t show up for them.

Let’s put some quick numbers on it. This week the Bureau of Labor Statistics (BLS) reported that Unit Labor Costs fell quarter-over-quarter in Q2 2018. Year-over-year, they increased by a mere 2%, not much different than at any time during the last decade. In 2007, by contrast, the last time the unemployment rate was close to as low as it is now, Unit Labor costs rose by 3% and that wasn’t exactly the most robust jobs market. In the year 2000, they increased by 3.6% once slack became exhausted (meaning the employment-population ratio was at its high).

The BLS also reports that nominal hourly compensation rose just 2% quarter-over-quarter, and was up only 3.2% year-over-year in Q2. The 4-quarter average is now 3.3%. Nominal pay is exactly where we would expect to find any bidding war for workers on an economy-wide scale (meaning beyond specific stories of worker shortages).

In the middle of 2007, quite differently, nominal compensation per hour rose by 4.6% year-over-year with a 4-quarter average of 4.4%. In 2005, it was rising by an average of almost 5%, nearly two percentage points faster than at any time during this so-called recovery dating back to 2009. In the year 2000, the average was at one point 7%.

For a supposedly booming economy supposedly short of workers, skilled or otherwise, it sure seems weird that no one seems willing to pay for them.

There is simply no evidence whatsoever that the labor market supply is tight, let alone extremely so. What we do know is that these other anecdotal explanations could instead be symptoms of bad economy rather than explanations for serious oddities contained within what is being called a boom period. Drug use may have exploded over the last ten years because largely unskilled workers are still being shut out by an economy that has performed in tragically substandard fashion.

According to the Congressional Budget Office (CBO), the so-called output gap has relatedly closed in Q2 2018. In estimates also released this week, the output gap is similar to the idea presented by the Phillips Curve. Indeed, they are interrelated since this economic measure is computed in large part via the unemployment rate.

The output gap starts with economic trend, or potential. This is the constant level the economy “should” be growing at when utilizing all inputs to maximum efficiency. In terms of employment and consumer prices, this is the level of output where businesses effectively use all labor before it becomes inflationary.

During a recession, like the one in 2001, potential doesn’t change. Actual output does, meaning that it falls significantly below trend. But a recession is a temporary deviation from that trend, meaning that through recovery processes like those that bring unskilled workers back into the labor force, actual output accelerates until it resumes that prior potential.

Using the latest CBO estimates, the output gap grew as large as 3% of potential GDP in the first quarter of 2003 – more than a year after the recession had been officially declared as ended. It took some time fifteen years ago for unskilled labor to be so cordially invited back in.

Unlike prior recessions, the contraction in 2008 and 2009 has led to enormous, categorical changes. Because the unemployment rate only falls, and because actual output never accelerates like it would in recovery, the CBO’s models interpret these contradictory signals as if economic potential itself has declined. And not by a little, it’s been a truly massive economic write-off.

If this was all merely a computer game, the CBO just used a cheat code so Economists can declare victory.

If we use the same estimates for economic potential as this one outfit had supplied in January 2007 before anyone had really heard of the word subprime, the output gap in Q4 2017 (the farthest their estimates went, ten years out into the future) would have been an unthinkable 12.9%, not nearly closed as is now suggested. That’s actually larger than it was at the official end of the Great “Recession” in 2009 (10.5%). By those prior standards, the economy has gotten worse, not better.

Even using the drastically reduced trend estimates from just four years ago, the output gap today would still be around 5%. This would be a substantially larger gap than it was at any point during the dot-com recession and its aftermath. It ends up being the only thing the rapidly falling unemployment rate has achieved, to close a huge output gap despite a serious worldwide downturn in between 2014 and today!

What has happened over the intervening decade which has caused the CBO to so radically alter its estimates for potential? One thing is for certain, it was never actual economic growth.

Instead, each time the trend numbers are downgraded it is because of the unemployment rate. As it goes lower, it forces the CBO to believe in it, meaning that they have no other choice but to see it as an accurate reflection of the economy (bureaucratic rigidity and inertia). The farther it falls, the closer to a zero-output gap the models have to output – economic growth or not.

With the rate now less than 4%, and the anticipated rush back into the labor market still absent, by these internal definitions there can’t be any economic gap left – even though there hasn’t once been sustained economic growth the entire decade. This right here is the drastic, cosmic change everyone should instead be focused on.

Everything gets turned backward by the unemployment rate. The skills mismatch and heroin crisis, Baby Boomers and the ever-present anecdotes of this labor shortage, they all are descriptions of an economy that doesn’t exist. These things exist, obviously, but by letting go of the unemployment rate it reorients these same tragic factors back in the same direction so as to be consistent with actual data. The lack of recovery causes them; they don’t explain how zero recovery can somehow at the same time be a full one (no output gap).

During the 2016 Presidential campaign, candidate Donald Trump told prospective voters in New Mexico:

“You hear a 5 percent unemployment rate. It’s such a phony number. That number was put in for presidents and for politicians so that they look good to the people.”

Not all people. Many are still left on the outside looking in. The CBO in 2018 says there aren’t that many at least those who should count, but the CBO in 2007 would have said these millions perhaps tens of millions of Americans actually do matter a whole lot. That was Trump’s message, too. At least it used to be.

Why do all these things change starting in 2008?

We don’t have to look very far for answers. For Economists, however, it is a whole different world, one with which they are no longer at all familiar. It begins with the one factor economic models didn’t model and still largely don’t.

You may have heard something about the dollar lately, especially this week. Like the CBO’s tendency to “reinterpret” economic potential and the terrible, unchanging circumstances for unskilled workers forever on the outside hearing about how great things are somewhere in the world, the dollar changed in 2008, too. And like the labor force, it hasn’t been able to find its way back, either.

Enjoy the boom. Even if it has been more than a numerical fiction, something specific tells me it may not last very long. Again.