The Long Decline in Per Capita GDP Growth

While economic growth in the United States is looking fairly robust as shown on this graph:

...in fact, if we look at GDP corrected for inflation and on a per capita basis, it is quite clear that the economy has not really done all that well since the end of the Great Recession.

Here is a graph from FRED that shows real domestic growth per capita since 1948:

You will notice that there was a significant drop in real per capita GDP during the Great Recession, in fact, per capita real GDP dropped from a peak of $49,506 in the third quarter of 2007 to a low of $46,781 in the second quarter of 2009, a drop of 5.5 percent.  Since then, it has recovered to $50,805 in the third quarter of 2014, the latest quarter for which data is available.  This works out to an increase of 8.6 percent from the 2009 low point.

While this looks relatively healthy, there is another way to look at the data.  Since the standard measure of GDP in the United States is expressed as the compounded annual rate of change from one quarter to the next, let's look at the per capita real GDP in those terms as shown on this graph from FRED:

When we look at economic growth in these terms, the latest recovery certainly doesn't look quite as healthy as it did after previous recessions.    

Here is a table that provides the average compounded annual rate of change of per capita real GDP for each of the periods between recessions since 1961:

Here is the same data in graphic form:

It is interesting to note that after each recession since the recession of 1974, the per capita annual growth rate of real GDP has dropped, hitting a multi-decade low since the Great Recession.  This quite clearly shows us that United States economic growth rates have been slowing for decades.

While we are all aware that GDP data is one of the most heavily revised lagging indicators, a brief look at the history of GDP per capita shows us that this recovery has been far more modest than any recovery in the past five decades, no matter how often the data is revised. 

Why Has GDP Growth Been So Modest?

Updated January 2015

We all know that there is a close relationship between personal consumption and GDP growth as shown on this graph from FRED:

As the decades have passed, personal consumption has made up an ever-increasing component of GDP, rising from a low of 58.5 percent in 1967 to its current level of 68.2 percent.  That said, if we focus on the tail end of the curve showing the data since the end of the Great Recession in June 2009, this is what we find:

It's obvious that the personal consumption component of GDP has not risen since the end of the Great Recession, unlike its pattern in since the late 1960s.  Although consumer spending has risen, its contribution to GDP has not grown in the past five and a half years.  This has put downward pressure on GDP growth rates.

Let's look at a brief essay by Daniel Aronson and Andrew Jordan, both at the Federal Reserve Bank of Chicago that may explain why economic growth has generally been so modest since the end of the Great Recession.  In their essay, the authors look at the relationship between real wage growth and the labor market.  They note that there is a close relationship between the share of the labor force that is medium-term unemployed (five to twenty-six weeks of unemployment), the share that is involuntarily working part-time (less than 35 hours weekly) and real growth in wages (after inflation). 

Let's start with this graph showing real hourly wage growth between 1979 and 2914:

With recessions shaded grey, we can quickly see that in the period between 1991 and 2001 during the jobs boom, real wage growth was substantial and remained above zero until around 2005.  After the 2001 and 2008 - 2009 recessions, real wage growth has stalled or fallen.

Here is a graph showing short-term unemployment (less than 27 weeks), long-term unemployment (more than 27 weeks) and involuntary part-time workers as a percentage of the workforce from 1979 to June 2014:

In the past, the national unemployment rate has been a useful predictor of real wage growth.  As unemployment rose, real wage growth fell.  This relationship has broken down over the past five years.  Given that the unemployment rate fell from a peak of 10 percent in 2009 to 6.1 percent in June 2014, had the historical relationship between real wage growth and unemployment held, real wage growth would have been 3.6 percentage points higher by mid-2014 than it was.

Calculations by the authors show several key things for average wage earners since the end of the Great Recession:

1.) a one percentage point increase in the short-term unemployment rate results in a change in real wage growth of -0.4 percentage points.

2.) a one percentage point increase in the long-term unemployment rate results in a change in real wage growth of -0.38 percentage points.

3.) a one percentage point increase in the medium-term unemployment rate results in a change in real wage growth of -0.71 percentage points

4.) a one percentage point increase in the part-time for economic reasons underemployment rate results in a change in real wage growth of -0.40 percentage points.

As well, the authors' calculations show that the negative impact on real wage growth is higher for those that earn less meaning that the impact of the slack labor market on real wage growth is highest on those who earn the least.

The authors calculate that if labor market conditions were the same as what they were in the period between 2005 and 2007, average real wage growth would have been one-half to one percentage point higher in June 2014 than it was.

While the current job market has improved since its low point after the Great Recession, its current weakness, particularly for those that are medium- to long-term unemployed and those that are underemployed for economic reasons, is having an impact on wage growth.  Without growth in wages, consumers are forced to either take on debt or reduce spending.  In our consumer-oriented society, so much of economic growth hinges on never-ending increases in consumer spending and until consumers feel that they are wealthier or at least see some real growth in their wages, it will be more difficult to achieve the economic growth rates that we were accustomed to in the past.

  

Household Debt and Delinquency Levels and Their Impact on GDP Growth

The recent Quarterly Report on Household Debt and Credit from the New York Fed shows an interesting change in the trend of household debt in the United States.

Here is a graphic showing the total household debt balance, including both housing and non-housing debt for the third quarter of 2013:

On September 30, 2013, total consumer indebtedness was $11.28 trillion.  This is up 1.1 percent on a quarter-over-quarter basis, however, it is still 11 percent below the peak consumer debt level of $12.67 trillion in the third quarter of 2008.  That said, this is the largest increase in consumer debt seen since the first quarter of 2008 when the Great Recession was just an infant.  It is also important to note that the housing market was severely over-valued at the beginning of 2008, putting significant upward pressure on the dollar value of housing loans, a situation that does not exist in most real estate markets today.  This renders a comparison between the current total consumer indebtedness level and that seen in 2008 rather difficult. 

Here is a breakdown of the increases by type of debt:

Mortgage debt: increased by 0.7 percent to $8.43 trillion

Non-housing debt increased by 2.7 percent to $2.85 trillion

Of non-housing debt, auto loan balances increased by $31 billion, student loans increased by $33 billion and credit card balances increased by $4 billion.  Auto loans increased to $97.4 billion, the highest level since the third quarter of 2007.

Here is a bar graph showing the total debt balance and a breakdown of its composition:

Now, let's look at the delinquency status of household debt in the United States.  Here is a bar graph showing the percentage of loans broken down the degree of delinquency:

It is interesting to note that while the percentage of loans that are more than 30 days delinquent has dropped from its peak of 11.9 percent in the first quarter of 2010, at 7.4 percent, it is still nearly double the 3.5 to 5 percent range experienced prior to the Great Recession.   Of the total outstanding household debt, $831 billion is considered delinquent with $600 billion considered seriously delinquent (at least 90 days late).  Here is a graph showing the new delinquent balances by the type of loan:

For the first time since the end of 2012, the total balance of new delinquent loans grew, hitting nearly $200 billion.  You will also note that the total delinquent loan balance is still well above the pre-Great Recession level of between $135 billion and $150 billion.

Despite the improvement in the economy, about 355,000 consumers had a bankruptcy notation added to their credit reports, roughly the same number as the year before.  Here is a graph showing the number of consumers with new foreclosures (in blue) and new bankruptcies (in red):

I find it interesting that the number of new bankruptcies has not dropped significantly since the middle of 2011 and that it is still quite elevated compared to levels seen from early 2006 to mid-2007.

The Federal Reserve has been using its "printing presses" to get the economy running on all cylinders.  Consumer spending is key to economic growth in America.  As shown on this graph from FRED, personal consumption expenditures make up nearly 69 percent of GDP, a multi-decade high:

With interest rates sitting at all-time lows and household deleveraging well underway, consumers are now showing signs that they are willing to take on increasing levels of debt to increase their expenditures which have risen from a low of just over $9.8 trillion in late 2008 to their current level of $10.52 trillion as shown on this graph:

What I find interesting is that, as shown on this graph, personal consumption expenditures are outgrowing inflation by over 2 percentage points:

What concerns me is the still elevated level of household debt delinquencies.  With household debt levels now on the rise and the threat of interest rate increases looming, only time will tell whether tapering will put upward pressure on already high delinquency rates, forcing consumers to reduce their debt levels.  With nearly 70 percent of GDP stemming from consumer expenditures, any reduction in consumer spending for any reason will put further downward pressure on what is already anemic economic growth.

America's New and Improved GDP - You Really Can Rewrite History

Who says that you can't rewrite history?  The recent announcement by the Bureau of Economic Analysis (BEA) announcing the results of the 14th comprehensive analysis of the National Income and Product Accounts (NIPA), also known as the nation's GDP by the sweaty masses, is the BEA's way of rewriting the past, all the way back to 1929 when most of our great grandparents or grandparents were suffering through the worst economic contraction in recent economic history.

Every five years or thereabouts, the BEA examines the source data behind the U.S. GDP data and revises what goes into calculating the size of the American economy and the level of economic growth.  Once these calculation changes are adopted, to maintain some semblance of consistency, the changes are applied all the way back to 1929.

Let's open by looking at how GDP is calculated.

GDP = gross investment + government spending + private consumption

Note that government spending is net of imports.  By increasing the influence of any one of these factors, GDP will rise, showing the importance of the BEA's revision process.

This time, there are four main conceptual changes:

1.) Research and Development is now capitalized.

2.) Entertainment, literary and artistic originals are now capitalized.

3.) Residential housing ownership transfer costs are capitalized on an expanded basis.

4.) The accrual treatment of defined benefit pension plans has changed.

In addition, major "table" changes were made to reflect:

1.) Intellectual property products.

2.) New pension tables.

While a discussion of most of these changes would be enough to induce involuntary sleeping among the non-accounting people among us, there is one key change that is rather surprising; the inclusion of the capitalization of entertainment, literary and artistic originals and intellectual property products.  By affecting these changes, the BEA is increasing the size of the gross investment number with the new inclusion of research and development spending, art, music, film royalties, books and theatre.  The United States is way ahead of the curve on this one, being the first in the world to adopt such a wide range of "investment" items.  Unfortunately, it means that we can no longer compare apples to apples when comparing U.S. economic growth rates to those of the rest of the world.   That's a topic for another posting.

First, let's look at how much these changes to the calculation of GDP has actually affected GDP growth rates back to 1929:

Naturally, the changes have less impact the further back that we go, however, over the past five years (between 2007 and 2012), the impact is noticeable.  Average annual economic growth, although very, very tepid compared to the past 80 years, moves up from 0.6 percent under the old schema to 0.8 percent, an increase of 33.3 percent.  It also pushes up the annual economic contraction rate between the fourth quarter of 2007 and the second quarter of 2009 from -3.2 percent to -2.9 percent.  See, that bad old Great Recession was nowhere nearly as bad as we all thought!

The biggest positive impact of the new GDP calculation was in 2012 as shown on this bar graph noting that the dark blue bar shows the total impact of the changes and the other bars show the impact of each factor on the total:

The new and improved GDP for 2012 rose by 2.8 percent rather than the even more pitiful 2.2 percent previously touted by the BEA.  The biggest improvement of all was seen in the first quarter of 2012 as shown on this bar graph:

For that quarter alone, the newly minted GDP growth rate of 3.7 percent was nearly twice the old rate of 2.0 percent.  By any measure, that's a robust period of economic growth, similar to what was experienced, on average, between 1929 and the Great Recession as you noted on the first graphic.

What does all of this mean?  GDP for 2012 is now $16.2445 trillion, up from $14.4179 trillion in 2009 and $14.4803 in 2007.  For 2012 alone, the upward revision was $559.8 billion when compared to the bad, old way of doing things.  This also has an additional unintended consequence; the U.S. debt-to-GDP level now looks somewhat less frightening than it did before.  With the debt including both debt held by the public and intragovernmental debt sitting at $16.433 trillion on December 31, 2012, the new and improved GDP numbers give us a debt-to-GDP ratio of "only" 101 percent.  Under the old scheme, the debt-to-GDP was a completely unacceptable 104.8 percent.  Sadly, all of that is academic now since the debt has long passed the level seen back at the end of 2012.

While I understand that the methodology behind the GDP calculation has to change to reflect changes in technology and other key parts of the economy, the apparent newfound, relatively robust, growth in the economy since  the end of the Great Recession is little more than smoke and mirrors.  In addition, looking back at the historical record, we can see that the level of economic growth in this cycle is still pathetic no matter what accounting magic is used by the BEA to make it look better.