Monthly Archives: April 2015

Contextualizing GDP growth in the U.S.

Today the Bureau of Labor Statistics published the first look at Q1 Gross Domestic Product and its components. The headline number was revealed to be a surprisingly low 0.2 percent. This is alarming, but not just because it defied the consensus prediction. Any single estimate, especially a preliminary one, is not in itself alarming. What’s alarming is the trend.

Take a look at the following image of the past six quarters’ GDP growth rate and its main components (expressed in terms of how each component adds up to the aggregate GDP number):

image (2)

We see weak Consumption growth, above all.  The net exports number is negative, but that won’t last (more on that below) and besides it was negative last quarter, too. What surprises me is the disappearance of Investment. Realize that a year ago the economy wasn’t just stalling, it was contracting. Rather than get worked up about one quarter, let’s look at the whole recovery to date.

In the next chart, consider how the current recovery’s average growth rate (and average component growth rates) compares to previous expansions.  I calculated each of the expansions based on NBER business cycle dates, trough to peak.

image (1)

What do we see?

1.  The new normal (this expansion) does not look very healthy. Overall, the GDP growth rate is half of the rate in the 80s expansion.  Even more interesting is what looks like a clear linear trend of slower overall GDP growth from 4.4 to 3.8 to 2.9 to the current 2.2.  But why?  Keep in mind that the lengths of the periods being averaged are not equal. The 80s expansion is the 3rd longest in US history at 108 months, while the 90s expansion is the 1st longest at 128 months.  In contrast, the 2000s expansion is “merely” 81 months which, oddly, is also the length of the current expansion.

2. Net exports are a non-story. Currently, the impact of net exports is a mild reduction to topline GDP growth in recent years, maybe 0.1 percentage points, the same as every previous expansion.  Moreover, you don’t find the gains of trade in these topline figures, rather they come from cost efficiency in intermediate goods. Regardless, people spinning blame for weak growth this quarter or this year on Greece or China are not checking their history.  We’re lucky there IS a China (there essentially wasn’t one in the 1980s). Scapegoating foreigners is always lame, but especially here.

3.  The components do not show weakness in Investment during our era.To be honest, I thought “I” would be revealed as weaker today relative to recoveries past. To the contrary, what’s most different in the expansion since 2009 is the weaker Consumption and Government.

I wasn’t satisfied with the second chart because it doesn’t put each recovery in context of the depth of the recession the economy “bounces back” from.  This is important because in 2009, the White House and CEA were claiming that the recovery was likely to be very strong because the recession had been very deep. Greg Mankiw posted an astute comment on his blog at the time which has been on my mind ever since: “The purpose of this picture is to show that deeper-than-average recessions are followed by faster-than-average recoveries. ” (read it here). Naturally, he was attacked by Paul Krugman for daring to question what was unquestionable: the certainty of a strong recovery (stronger still thanks to the Stimulus).

It is striking that the debate raged for a year or two – there’s always tomorrow, right? — but nobody believes today what apparently everybody believed in 2009.  This really begs the question, why is the economy NOT rebounding back to trend?  My Hoover colleague, John “Grumpy” Cochrane, has some interesting things to say about unit roots and trends, but the bottom line is that the very idea of Potential GDP has become slightly metaphysical. I have mixed feelings about it.

The third chart provides some useful context, and in this one you should know the recessions are of roughly equal duration (4-6 quarters).  Now THIS is a picture worth a thousand words:

image (3)

The ’01 recession is an odd duck — average GDP growth was bizarrely positive, albeit tiny. Consumption growth was not derailed at all.  If you believe that boom is proportional to bust, then the mildness of the 00s expansion makes sense. But that means the current expansion does not. And the very sharp drop in Consumption in the 08 recession and current recovery signal that something permanent shifted negatively during the financial crisis.

A final, vital piece of context is that all growth is good growth. What I mean is that our collective focus on “weak growth” is an oxymoron. Even the weakest growth means that the economy is richer than it was the quarter before. By comparing 2015 to 1985, the actual underlying economies are radically different in ways from the composition of the labor force to the nature of money itself. But still, I’m left with a sense that the engine of US growth — our institutions — have changed for the worse. That’s alarming.

Thoughts on Jobs Day

Today’s jobs report is a lot like the last one and the one before: mixed signals.  The unemployment rate is unchanged at 5.5 percent. Payroll growth is positive but timid at 126,000 when economists had apparently expected twice that.  Seven cents were added to the average hourly wage, which is faster than normal. And finally, worst of all, continuing the trend of the past 5 years, the labor force declined by nearly 100,000 people.

Those are mixed signals, but it is exactly what I expected. The most alarming central fact of the economy’s new normal is that it is bleeding people. Here’s a quick and dirty facts from BLS data on the labor force participation rate overall (LFPR) and the rate for people between the ages of 25-54.  This disproves the contention that the change is just about demographics.

-3.5  = Change in LFPR since 2007
-2.2  = Change in LFPR 25-54 since 2007
+1.1  = Change in Unemployment rate since 2007

When the Federal Reserve looks at today’s report, it has to conclude that this may be as good as it gets. Frankly, you do not want to see the unemployment rate get much lower than 5.5 or else it will signal an overheating economy.  The last time the unemployment rate went below 5 percent, it was 2005. But they were being pushed down by ultra low (at the time) interest rates set by the Fed in 2003. Real estate markets were inflated. Starting in mid 2004 – when the unemployment rate was exactly what it is today (5.5%) – the Fed began raising interest rates, which it did multiple times to cool the engine. We all know how that ended. Consider:

-3.4  = Change in LFPR since 2004
-1.8  = Change in LFPR 25-54 since 2004
+0.0  = Change in Unemployment rate since 2004

This is the new normal. So why is LFPR down three and a half percent?  That’s the magic question.