The 3 Ps: Part 2 of Productivity, Population, and ‘Product’

Welcome back!   This is a continuation of my last blog, Part 1, and here it is if you need a refresher:

The 3 Ps: Productivity, Population, and ‘Product’ – Part 1

I’m going to start by quoting myself:

“I contend something is very different today.    Specifically, demographic trends, the efficacy of labor, the impact of both on current and future GDP trends, and, finally, the implications of these long-term trends on investment asset values.”

In my last blog, I explained how – in my opinion – the market’s continuously diminishing ROI expectation has the inverse affect on investment asset value.  Today, I’ll explain why I believe this is not a short term trend; it is the result of structural changes in our country and our economy.

On August 11th, in a paper entitled Credit Markets Review and Outlook produced by Moodys and their capital market research, the following commentary appeared:

“The five-year average annualized growth rate of labor productivity’s moving yearlong average has just endured its worst showing amid a mature upturn since the 1940s at least.

During the five-years-ended June 2016, the moving yearlong average of US labor productivity rose by merely 0.6% annualized, on average.  In stark contrast, productivity’s average annualized growth rates over the two previous contiguous five-year spans were 1.8% as of June 2011 and 3.1% as of June 2006.

In terms of a moving yearlong average, labor productivity’s 5-year average annualized growth rate set a record high at the 3.9% for the span ended June 1966 and then slowed considerably to December 1982’s record low 0.3%. The latter growth rate was skewed lower by the recessions of 1980 and 1981-1982. The same surge in energy costs that gets some of the blame for the double-dip recession of the early 1980s also has been viewed as being responsible for productivity’s weak showing throughout much of the 1970s.

However, productivity has yet to demonstrably benefit from the latest plunge by energy prices. After 1982’s trough, productivity’s 5-year average annualized growth rate climbed all the way up to a June 2004 peak of 3.55%.  Some credit 1983-2004’s pronounced acceleration by productivity to the introduction of the personal computer, powerful microchip technologies, the internet, and wireless telecommunications technologies.  

Since mid-2004, productivity’s five-year average annual growth rate has slowed to the 0.6% of June 2016. Productivity’s average during the year-ended June 2016 barely rose by 0.2% annually, which suggests that productivity’s five-year average annualized growth rate has yet to bottom.“

Hmmm…a lot to take in…so let me summarize:   Labor productivity – a measurement of which, when growing, simply suggests more ‘stuff’ is being made in the same amount of time – by Moody’s measure is sliding toward zero.  Here’s a chart of their 5-year average growth rate:

Again, note this is a 5-year average growth rate.   Which suggests recent readings are much worse.  And, in fact, they are.

To demonstrate this fact, I’ve prepared the graph below – again, using data from my friends at FRED – showing the growth in both real GDP and nonfarm labor ‘hours worked‘ on the same chart, from the time both were tracked beginning in 1947:

Note this is an ‘index value’ chart where both metrics are relative to a point in time – in this case, Q2 of 2009 – when the index value = 100.   As both metrics are shown as an index value, we can easily see growth/movement trends.

Q2 of 2009 is selected as the index point simply because this is where the value of both charts equals 100.   And it’s the only point of intersection.  Take another look.  Which means, moving to the very left hand side of the chart, we see in Q2 of 1947 Real GDP had an index value of 13.5 and, at the identical time, the ‘Nonfarm Business Sector: Hours of all Persons’ had an index value of 48.2.

Which means that between 1947:Q2 and 2009:Q2 (62 years) real GDP grew from 13.5 to 100 and hours worked grew from 48.2 to 100.  Said this way, its easy to see that real GDP has grown at a much greater rate relative to the number of hours needed to create it.

Said another way, in pretty much every year since 1947, Americans were able to produce more GDP in the same amount of time.   Why?   Economists would attribute the increased efficacy to either (1) technological improvement, (2) capital investment, or some combination of both.

But following the Great Recession something seems to have changed.  Take a look at the chart below.   Here I’ve manipulated the index chart a bit differently.  I’ve actually divided the real GDP index values into the ‘hours worked’ values in order to determine the number of hours required to generate one unit of real GDP over time.

This chart is clear:  Something became drastically different following the Great Recession.

In 2000, 1.26 “hours worked” generated 1.0 unit of GDP.   Between 2000 and 2010, this ratio improved, as it has every year since 1947.   The amount of time needed to produce 1.0 unit of GDP continued to decline – reaching about 0.97 hours in mid-2010.   It took less time to make the same amount of stuff – productivity was improving.   And, as a result, real GDP growth was robust.

But in the last 5 or 6 years, something happened.   Something changed.  Labor stopped becoming more efficient.  It actually takes 0.972 hours today to produce 1.0 unit of GDP – just a hair more time than in 2010.  As a result, in the last 6 years, our GDP has only grown at a rate equal to growth in the number of hours worked.

This is why real GDP has been constrained in the past 5 or 6 years.  Our GDP is growing at a rate that is almost 100% correlated with hours worked.

Hmmm…what’s going on?  Is this a trend … or is it a structural, more permanent change?   Has technology stopped advancing?  No…that’s not it.   Have capital investment stopped.  No, capital investment has plateaued recently, but spending does continue at consistent levels.

According the the Centers for Disease Control and Prevention (CDC), the US fertility rate fell to the lowest point since record keeping started more than a century ago.    In the first quarter of 2016, the fertility rate was only 48.6% of ‘peak fertility’ back in 1957.

Here’s a chart from the CDC, tracking births and fertility since 1920.   Notice the trends since 2000:  We saw a spike in both around 2007, but it’s been downhill since then.

Of course, this trend becomes meaningful about 16 years from now.  Because that’s when a baby born today turns 16 and potentially becomes part of the US labor force, or as it’s called by the BLS, part of the “Civilian Non-Institutional Population.”

But while the civilian noninstitutional population measures the potential worker, the actual number of workers in the US economy is much lower.   For some reason, or multiple reasons, fewer potential workers are working.   The ‘participation rate’ of workers has reached an all time low.   After peaking about 15 years ago, it has steadily slipped…sliding lower and lower:

Summing it up

In the final analysis, there are only 2 ways to significantly grow the US GDP in coming years.

First, if current productivity trend remains intact, the US must grow the number of hours worked (the result of more people working) which will translate into a growing GDP.

The second,  if current labor force trends remains intact, the US must increase labor productivity which will result in robust GDP growth.

Unfortunately, neither seems likely to occur.   Suggesting US real GDP growth will remain sluggish for years to come.

At the same time, Dr. Larry Summers, the ex-Harvard professor and ex-Treasury Secretary, believes a thing he (and other economists) call ‘secular stagnation’ is running rampant. 

Secular stagnation, per Dr. Summers, is “an imbalance resulting from an increasing propensity to save and a decreasing propensity to invest.  The result is that excessive saving acts as a drag on demand, reducing growth and inflation, and the imbalance between savings and investment pulls down real interest rates.”

I agree.   This view seems to reflect and explain the recent economic experience of industrialized nations.   Further from Dr. Summers, “Real interest rates are very low, demand has been sluggish, and inflation is low, just as one would expect in the presence of excess saving. Absent many good new investment opportunities, savings have tended to flow into existing assets, causing asset price inflation.”

So, are we experiencing asset price inflation?   Or is this the “new normal?”

At the risk of later being proven wrong, I suggest this is the new normal.   The US is awash with investment capital at a time when real GDP and labor force growth are stunted.   None of these trends will end soon.

Which means we’re likely to experience very low ROIs and elevated asset prices for a long, LONG time to come.

• Terry Liebman

(to read Dr. Summers entire blog, click this link:  http://larrysummers.com/2016/02/17/the-age-of-secular-stagnation/#more-6233)