The Worsening Distribution of Income: The Top 1% vs. the Other 99%

Posted on by

The Occupy Wall Street and related groups have brought to the fore concerns about the distribution of income in the US.  And there is validity to these concerns, as the distribution of income has deteriorated markedly in recent decades, starting with the Reagan period in the 1980s.  The share of the top 1% has well more than doubled, to a level not seen since the 1920’s just prior to the Great Depression, with this coming out of a declining share of the bottom 90%.

The graph above is taken from material assembled by Professor Emmanuel Saez and his colleagues and co-authors, and is available through his web site (link here). The specific data here is from a July 2010 update of material originally published by Professor Saez with Professor Thomas Piketty in 2003, with data now through 2008.  Professor Saez and colleagues from around the world have assembled an amazing set of data for a large number of countries, using tax return data to produce long time series of family (tax unit) income levels.  Such data can go much farther back in time than available income surveys will allow, to produce historical data otherwise unavailable.  While there are drawbacks (for example, tax units are not always the same as the household units one would prefer), nothing else can span such periods of time.

The US data show that there was a previous boom in the 1920s in the income share of the top 1%, peaking at 23.9% of national income in 1928.  But the share then fell in the Great Depression and during World War II, and with the reforms and structural changes implemented during that period, eventually came to about a 10% share in the 1950s and 1960s, and to about 9% in the 1970s.  But it then started to grow, and reached a 23.5% share in 2007.  It fell back in 2008, with the onset of the financial collapse in the last year of the Bush Administration.  But as noted in this blog posting of November 26 on this site, profits rebounded in 2009 to now, so it is likely the top 1% share has also rebounded.

It is also interesting that this increase in income concentration is essentially only at the very top:  the top 1% is seeing almost all of the increase.  Those in the 90 to 95% percentile, and in the 95 to 99% percentile, have seen some trend rise in their income shares in recent decades, but they have been close to flat.  Rather, the higher share of the top 1% has come out of a falling share of the bottom 90% (the shares of all groups together must sum to 100%).  The share of the bottom 90% has dipped to about 50%, from a level of about 65% (plus of minus a percentage point or two) from 1942 to 1982.  This is a remarkable change after four decades of such stability.

Another interesting calculation, done by Professor Saez, found that 52% of the increase in real national income between 1993 and 2008 accrued to the top 1% of families, even though this top 1% only had a 14% share of national income at the start of this period (and 21% at the end).  That is, more than half of income growth over this 15 year period went to just the top 1% of the population, and where this group accounted for only a 14% share of income at the beginning.  Had the growth been balanced, the top 1% would have obtained 14% of the income growth, not 52% of it.

The data here does not explain why there has been this increasing concentration of income in the US.  One has also seen increasing concentration in other countries around the world in recent decades (the data is available here), which suggests some basic global structural changes are part of the cause.  But the global pattern has not been as extreme as in the US, suggesting that policy changes in the US begun under Reagan and continued since, are also part of the cause.

Non-Defense Federal Government Employment Has Fallen Under Obama, and Grew Under Bush

Posted on by
(change, in thousands of jobs) Jan 2001 to Jan 2005 Jan 2005 to Jan 2009 Jan 2001 to Jan 2009 Jan 2009 to Oct 2011
Federal Govt Employment -35 66 31 49
Defense Civilian Employees -27.5 25 -2.5 63.5
Federal excl Defense -7.5 41 33.5 -14.5

Federal Government employment, other than civilian employees in the Defense Department, has fallen during the Obama Administration.  In contrast, it grew under Bush.

While the numbers are small, in particular relative to national employment (the Federal Government only employs about 2.8 million workers, out of a US labor force of 154 million, or just 1.8%), it is helpful to get the facts straight in the light of the continued Republican attacks that the Federal Government has boomed under Obama, and accounts for the continued weak economic and employment growth of the US.  The spokeswoman for Republican Congressman and Majority Leader Eric Cantor (Megan Whittemore), for example, charged in an email sent to PolitiFact (link here), that the only job growth that can be attributed to the 2009 Stimulus program was in government.  Yet as we saw in a posting made yesterday at this site (link here), total government employment in the US (mostly state and local) has fallen by close to 600,000 since Obama took office.  The purpose of this new post is to focus on what has happened to the Federal Government employment alone.

Federal Government employment is only less than 13% of total government employment in the US, so the changes here will not much matter overall.  But it is interesting that while there has been a very small growth in overall Federal employment since Obama took office (of just 1.8% total, or 0.6% annually), it has all been due to growth in civilian employees at the Defense Department.  The table above, drawn from data issued by the Bureau of Labor Statistics (US Department of Labor), presents the numbers.  The figures by the BLS on Defense Department employees are not seasonally adjusted, so none of the figures in the table above are either, for consistency.  However, seasonal adjustment does not make much of a change in Federal Government employment figures in any case.  The most recent available figures are for October.  It should also be noted that all employment figures of the BLS are for the civilian population, and hence exclude active military personnel in all categories.

As is seen, while the number of all Federal employees rose by 49,000 under Obama, the Defense Department civilian employees grew by 63,500, so that Federal employment excluding Defense fell under Obama by 14,500.  It is also interesting to note that Federal employment grew under Bush, all in his second term, with an increase of 33,500 non-Defense Federal workers over his two terms together (and by 41,000 in his second term alone).

In sum, Federal Government employment grew under Bush.  Under Obama, non-Defense Federal workers have declined, and overall they have grown only because of additional Defense Department civilian workers.  All the numbers are relatively small, in particular relative to the size of the full US labor force.  But the assertion by many Republican politicians that the Federal workforce has exploded under Obama is false.

Contracting Government Has Hurt Job Growth

Posted on by
(change, in thousands of jobs) Jan 2001 to Jan 2005 Jan 2005 to Jan 2009 Jan 2001 to Jan 2009 Jan 2009 to Nov 2011
Total Employment -16 +1,110 +1,094 -1,855
Private Sector -916 +263 -653 -1,262
Government Sector +900 +847 +1,747 -593

Obama has repeatedly and emphatically been charged by Republican politicians as fostering  a huge expansion in government job growth, with this a major cause for the weak recovery in private job growth.  The facts do not support this.  The government sector has in fact been contracting sharply during the Obama period, in distinct contrast to the expansion during the Bush presidency, and it is this contraction which indeed can explain a significant share of the drop in overall jobs in the economy.

The table above, drawn from Bureau of Labor Statistics (US Department of Labor) figures on employment levels by the major sectors, shows the change in the number of those employed, for the periods between January 2001 and January 2005 (the first Bush term), between January 2005 and January 2009 (the second Bush term), between January 2001 and January 2009 (the two Bush terms together), and between January 2009 and November 2011 (the most recent figures, for the Obama term so far).

In the first Bush presidential term, overall job growth was basically zero.  But it is striking that it only comes to zero because a decline of 916,000 private jobs is almost fully offset by a nearly identical rise in government jobs of 900,000.  Note that while the time periods we are examining are the presidential terms, government job growth is largely affected by changes at the state and local level, as these account for about 87% of government jobs.

During the second Bush term, private sector job growth became positive, by a modest 263,000 for the period (with growth early on offset by the downturn in his final year), while government job growth continued at a roughly similar positive rate as during his first term.  With both positive, total job growth was then about 1.1 million.

For the Bush presidency as a whole, it is then interesting to note the overall job growth of about 1.1 million (all in his second term), only came about due to a growth in government jobs of about 1.75 million:  Private jobs in fact fell by about 650,000.

During the Obama presidency so far, government job growth went into reverse, with a decline of almost 600,000 jobs.  It is interesting that the pace of the decline for the first 34 months of Obama’s 48 month term (that is, through November 2011), matches almost exactly the pace of the increase during either of the Bush terms.

Suppose government job growth had increased during the Obama period at the pace it had during the Bush terms.  There would then have been a growth in government employment of over 600,000, rather than a decline of almost that amount, for a swing of 1.2 million jobs.  Assuming the same decline as now of over 1.2 million private jobs, overall jobs would have still fallen, but by only about 600,000 rather than over 1.8 million.

Of course, with such a different policy on government job growth, one would not expect private job growth to be the same.  Conservatives might argue that the government job growth would “crowd out” the private sector, leading to even an even larger fall in private jobs.  But there is no evidence to support this, in an environment where unemployment is high and interest rates on government borrowing are close to zero as the economy suffers from a liquidity trap.

Indeed, the basic insight of John Maynard Keynes is that in such a situation, government job growth (and its accompanying spending) will not only not displace private job growth, but will add to it at a multiple of what is spent directly, as the newly employed by the government will add to demand for privately produced goods and services as they spend their wages.  A reasonable estimate of this government employment multiplier would be at least two, and many would argue higher.  At a multiplier of two (that is, each additional government job leads to one additional private job, for two total), the swing in government employment of 1.2 million (that is, a rise of 600,000 rather than a decline of about 600,000), would have led to 1.2 million additional private jobs, and total employment growth would then have been a positive of 600,000 rather than a negative of over 1.8 million.  This is a swing of 2.4 million jobs.  With current unemployment in the US of 13.3 million, and a civilian labor force of 153.9 million, the unemployment rate would then be 7.1% rather than the current 8.6%.  It was 7.8% when Obama took office.

One can quibble with the specific figures, and what multiplier to assume.  But the basic point is that the contraction in government employment in recent years (primarily at the state and local levels) is a major reason why the overall job picture is still so bad.  If government employment had continued to expand at the pace it had during either of the Bush terms, rather than contract at a similar pace, then under quite plausible estimates, unemployment would be less now than when Obama took office.