The Wage-Productivity Gap - Anger on Main Street America

Updated May 1, 2020

There is a simmering anger in America, an anger that is largely responsible for the election of the current administration in Washington.  Millions of American voters feel that they simply cannot get ahead financially no matter how hard they try or what they do, particularly since the massive economic displacement that has taken place during the COVID-19 pandemic.  In this posting, I want to look at an interesting metric that may help to explain this growing feeling of disenchantment with the Great American Dream.

In a recent analysis by the Economic Policy Institute or EPI we find the following graphic:

As you can see on the graphic, when you compare the growth in the average hourly compensation of non-management workers and the growth in net productivity (defined as the growth of output of goods and services minus depreciation per hour worked) you find the following keeping in mind that 1948 is the base year used for comparison:

1.) From 1948 to 1972 - average workers' hourly compensation rose by 92.2 percent and average net productivity rose by 92.2 percent.  Over this timeframe, workers' compensation more-or-less rose in tandem with net productivity, in fact, if we look at data during the 1950s and early 1960s, workers' compensation often rose at a faster annual rate than net productivity (i.e 1957 - compensation rose by 37.1 percent compared to a rise of 30.0 percent in net productivity).  

2.) In 1966, when compared to 1948, the growth in workers' compensation rose by 64.7 percent compared to a rise of 70.0 percent in productivity, the first year that productivity significantly outstripped growth in workers' compensation.

3.) By the mid-1970s, the growth in workers' compensation was significantly outpaced by productivity when compared to the base year.  For example, in 1976, the increase in workers' hourly compensation was 89.4 percent compared to productivity growth of 103.6 percent.

4.) This trend in lower workers' hourly compensation and higher net productivity growth has continued uninterrupted with compensation rising by 115.6 percent and net productivity growing by 252.9 percent when compared to the base year of 1948.

While this situation is bad enough, research by the Pew Research Center shows that, while American wages have risen over the past 40 years, when measured against price increases (i.e. in real terms), the purchasing power of America's working class has barely moved over the past four decades as shown here:

The economic stagnation experienced by Main Street America, particularly those who reside in the middle class, is clearly driving American voters into a search for political change.  It is the driving force behind the political right's longing for the America of the 1950s and 1960s when American workers were actually paid fairly for their productivity and the political left's moves toward implementation of the "nanny state" that uses socialist programs to answer the question of wages that are too low to get ahead.

Let's close with one final thought.  Just in case you thought that American workers were all in the same boat, these two graphics will clearly show you that a certain working Americans have benefitted significantly over the past four decades:

God bless America...well, at least a subset of America.  While we are being told that we are all in  this pandemic situation together, it's pretty clear that some of us have way more skin in the game than others. 

Productivity - The Federal Reserve's Fly in the Ointment

Since September 2007 when the Federal Reserve responded to the economic crisis facing the United States by dropping the federal funds rate from 5.25 percent to between 0 and 0.25 percent, the Fed has done its very best to keep the economy running on all cylinders, unfortunately, even with its massive monetary experiment and its bloated balance sheet, there is one aspect of the economy that has not responded as it normally would.  This key aspect of the economy is critical to government budgetary planning, particularly given that the Trump Administration's plans call for a 3 percent economic growth rate.

Let's open with the Bureau of Labor Statistics definition for output per hour.  According to the BLS, labor productivity or real output per hour is calculated by dividing an index of real output by an index of hours worked of all persons including employees, proprietors and unpaid family workers.  Here is a graph from FRED showing the real output per hour of all persons in the nonfarm business sector (i.e. productivity):

As you can see, it appears as though the curve is flattening after the Great Recession, suggesting that productivity growth has declined in comparison to other economic expansions.  This graph showing the year-over-year percentage change in real output shows just how dramatic the slowing in real output has been since the first quarter of 2011:

Other than the short period between the first quarter of 2009 and the first quarter of 2011, productivity growth has been anaemic, particularly when we compare productivity growth levels in the latest period of economic expansion to that of previous expansions.  To that end, here is a bar graph showing the average rate of productivity growth for all periods of economic expansion since 1970:

Even when we include the burst of productivity growth between early 2009 and 2011, productivity growth in the latest expansion shows the lowest average growth rate going back nearly five decades.  If we look at the growth rate of productivity since early 2011, year-over-year growth drops to an extremely low level of 0.54 percent, the slowest productivity growth rate in the past fifty years and only slightly higher than the period between early 1979 and late 1982 when the U.S. economy experienced two short-lived recessions and a 19 percent Federal Funds Rate.

Why has the growth rate of productivity dropped so significantly since the Great Recession?  This can be attributed to several factors:

1.) a lack of business capital investment in new technology.

2.) a drop in the productivity payoff from digital technology.

3.) a lack of training in new skills for workers.

4.) weak demand.

Here is what Stanley Fisher, Vice Chairman of the Federal Reserve had to say about the weakness of productivity growth in late 2016:

"Understanding the recent weakness of productivity growth is central to addressing the longer-run challenges confronting the economy. Productivity growth over the past decade has been lackluster by post-World War II standards. Output per hour increased only 1-1/4 percent per year, on average, from 2006 to 2015, compared with its long-run average of 2-1/2 percent from 1949 to 2005. This halving of productivity growth, if it were to persist, would have wide-ranging consequences for living standards, wage growth, and economic policy more broadly. A number of explanations have been offered for the decline in productivity growth, including mismeasurement in the official statistics, depressed capital investment, and a falloff in business dynamism, with reality likely reflecting some combination of all of these factors and more.

We should also consider the possibility that weak demand has played a role in holding back productivity growth, although standard economic textbooks generally trace a path from productivity growth to demand rather than vice versa. Chair Yellen recently spoke on the influence of demand on aggregate supply.3 In her speech, she reviewed a body of literature that suggests that demand conditions can have persistent effects on supply.  In most of the literature, these effects are thought to occur through hysteresis in labor markets. But there are likely also some channels through which low aggregate demand could affect productivity, perhaps by lowering research-and-development spending or decreasing the pace of firm formation and innovation. I believe that the relationship between productivity growth and the strength of aggregate demand is an area where further research is required.

I will conclude by reiterating one aspect of the low interest rate and low productivity growth problems that I have mentioned previously--the fact that, for several years, the Fed has been close to being "the only game in town," as Mohamed El-Erian described it in his recent book.5 But macroeconomic policy does not have to be confined to monetary policy. Certain fiscal policies, particularly those that increase productivity, can increase the potential of the economy and help confront some of our longer-term economic challenges. While there is disagreement about what the most effective policies would be, some combination of improved public infrastructure, better education, more encouragement for private investment, and more effective regulation all likely have a role to play in promoting faster growth of productivity and living standards. By raising equilibrium interest rates, such policies may also reduce the probability that the economy, and the Federal Reserve, will have to contend more than is necessary with the effective lower bound on interest rates." (my bold)

  

Basically, the Federal Reserve has no real idea on how to improve productivity growth levels.  One thing that they do seem to acknowledge is that slowing productivity will have a long-term impact on profit growth, living standards, wage growth and future economic policies.  This will obviously pose long-term challenges for the American economy, particularly during the recovery period after the next economic contraction.  If we think that the economy has not grown at levels seen in past recessions since the Great Recession, dropping productivity growth levels will make the next recovery look like it is barely a recovery at all, a reality that will significantly impact Washington's fiscal planning.

What's Behind the Collapse of America's Manufacturing Sector?

As we are all aware, Donald Trump is on a mission to bring jobs back to the United States.  In large part, his plans include renegotiating trade deals that have led to job losses in what was once America's economic foundation; manufacturing.

Here is a graph showing what has happened to manufacturing jobs in the United States since 1939:

At 12.392 million jobs, America's manufacturers have shed 7.161 million jobs from the peak of 19.553 million in June 1979, a decline of 36.6 percent.  As well, since March 2010 when manufacturing jobs hit their post-Great Recession nadir of 11.453 million, the addition of 939,000 jobs is a very poor showing, particularly given the lengths that the Federal Reserve has gone to since 2008.

While there is no doubt that freer trade has led to offshoring of American manufacturing jobs, a study entitled "The Myth and the Reality of Manufacturing in America" by Michael Hicks and Srikant Devaraj at Ball State University strongly suggests that other factors are at play as you will see in this posting.

Let's start with a graphic that shows the U.S. manufacturing production index from 1919 to 2014:

As you can see, the Great Recession put significant downward pressure on national manufacturing production, pushing it well below the trend line.  Nonetheless, in real dollars (inflation adjusted), the value of manufacturing production continues to climb and, by 2014, it had recovered completely from the setback of the Great Recession. 

Here is a graphic showing the value of all manufacturing as well as nondurable goods (those used for less than a year) and durable goods (those used for more than a year) from 1987 to 2014:

While the real value of nondurable goods (in light blue) has been static for more than a decade, the real value of nondurable goods (in dark blue) continues to rise.  As well, you can see that the real value of all manufacturing goods continued to climb throughout the two and a half decades with short interruptions during recessions.  This clearly shows us that manufacturing is still a very important part of the U.S. economy.  In fact, while the total value of manufactured goods fell by 11.5 percent during the period from 2006 to 2009, rose by 32.9 percent over the period from 2009 to 2013, leaving us with overall growth of 17.6 percent from 2006 to 2013.

Now, let's look at how productivity has changed in key manufacturing sectors.  Here is a table showing how the average product of labor changed between 1998 and 2012 and the GDP growth for each sector over the same timeframe:

As you can see, when adjusted for inflation, productivity grew for all sectors over the period from 1998 to 2012, ranging from a low of 6 percent in the nonmetallic mineral products sector to a high of 829 percent in the computer and electronic products sector with an average of 90 percent for all manufacturing sectors.  This compares to overall production value increase (i.e. GDP growth) of 32 percent for all sectors.

What caused this substantial increase in productivity?  The authors note that the increase in productivity is largely related to the adoption of automation and information technology.  The authors focused on the period from 2000 to 2010 when the U.S. economy experienced the largest decline in manufacturing employment in history as shown here:

In January 2000, there were 17.284 million manufacturing jobs in the United States, by the beginning of 2010, this had dropped to 11.46 million, a decline of 5.824 million.  The authors  calculated the total number of employees needed to produce the 2010 levels of production using the 2000-level worker productivity levels.  Had the economy kept the level of productivity from the year 2000 and applied that to the 2010 production levels, manufacturers would have required 20.9 million workers rather than the 12.1 million that were employed in the sector. 

The study then examines the job losses in manufacturing that are attributable to trade, changes in domestic demand for goods and changes to productivity.  The authors found that losses in productivity and trade varied by sector but, overall, the share of job losses in manufacturing were relatively small when it came to international trade (13.4 percent of jobs lost) compared to losses related to increases in productivity (87.8 percent of jobs lost).  This means that only 750,000 of the jobs lost in manufacturing during the period between 2000 and 2010 were related to international trade compared to job losses of over 5 million that were related to increased productivity (i.e. the adoption of automation and information technology). 

From what we can see in this study, very little of the job losses in the manufacturing sector are related to international trade.  It appears that, unless America's domestic manufacturers are willing to forgo the productivity gains made by adopting automation, the U.S. economy will find it increasingly difficult to create significant numbers of manufacturing jobs.  Gone are the glory days of the factory worker no matter what Donald Trump may want.