[SfN] Census Bureau Stats on Econ. Inequality

[steven paul varel]

Hey everybody,

I came across a rather stubborn Gore supporter today on the quad, and we
got into quite a debate.  During the conversation we disagreed on topics
such as the effects of welfare reform, the role of trickle-down economics,
the economic condition of the social democracies in Western Europe, and
the trends in income inequality in the US.  It was his position on this
last issue that motivated me to check the situation out myself.  He
claimed that economic inequality had decreased since 1995 as a result of
lower unemployment caused by policy supported by the Clinton/Gore
administration.  I thought you may be interested in what I found.  

I went to the Census Bureau's website at www.census.gov.  The Census
Bureau uses two measures to judge income inequality:  the Gini ratio and
the % of income by quintiles.  Both of these indicate increasing income
inequality since 1995.  Just in case anyone does not know, the Gini ratio
is a method of measuring inequality which rates the amount of inequality
from 0 to 1, 0 being perfect equality (a situation in which all economic
resources are distributed evenly) and 1 being perfect inequality (a
situation where all resources are concentrated in the hands of one person
and no one else has anything).  The Gini ratio has increased form .399
in 1967 to .457 in 1999.  It has increased steadily since 1980.  During
the Reagan administration it increased from .403 to .427, an increase of 
24 points. When the Clinton/Gore administartion took over in '92, the Gini 
ratio was .434; and the latest recorded data on the Gini ratio show that
in 1999, after seven years of a Clinton/Gore white house, the Gini ratio 
has increased 23 points.  This shows the Clinton/Gore administration to be
as bad the Reagan era.  In addition, the Gini ratio has increased seven
points since 1995.  

The other method of measuring inequality used by the Census Bureau
involves dividing the population into fifths or quintiles and measuring
the % of aggregate income recieved by each fifth.  This measure shows
similar increases in income inequality.  In 1967 the bottom fifth received 
4% of the overall income, the second fifth got 10.8%, the third fifth got
17.3%, the fourth fifth got 24.2%, and the top fifth got 43.8%.  During
the Clinton/Gore administration, the bottom fifth fell from 3.8% to 3.6%,
the second fifth from 9.4% to 8.9%, the third from 15.8% to 14.9%, the
fourth from 24.2 to 23.2, and the highest fifth rose from 46.9% to 49.4%
of the total income with the top 5% rising from 18.6% to 21.5%.  There
were also decreases in every quintile except the top one in the period
from 1995 to 1999.  

Is the increase in inequality recorded in these two measures statistically
significant?  Well, I think the jump in the Gini ratio from 1992 to 1999
must be.  As for numbers that show a decrease in the income share each of
the bottom 4 quintiles coupled with an increase in the income share of the
top quintile, I think that the 3% increase in the top five percent must be
seen as statistically significant.  Since this group must have taken this
3% from somewhere, it is reasonably to conclude that the bottom four
quintiles must have decreased in income share at some significant level.  

As for the data showing the increasing inequality from 1995-1999, it is
difficult to tell if this represents much real change.  It is difficult to
measure the extent of change in such a small time period with these
measures.  It really does not matter whether there was a significant
increase, however, in order to show that there definitely was not a
decrease.  It would be very difficult to make a case for decreasing
income inequality during the Clinton/Gore administration or during the 
period from 1995-1999 using this data.  

Is this data biased somehow?  It seems doubtful that the Census Bureau
data would be biased in such a way that it would show greater income
inequality.  The Census Bureau is an established, highly reputed
organization that uses the best statistical methods available.  

So what explains the fact that income inequality has risen (or at least,
not decreased) when unemployment has supposedly decreased over the period
from 1995-1999 (I say supposedly because I have not checked any specific
sources for this information, but I have heard it enough from seemingly
reliable sources so I assume it is true, but this can sometimes be a
dangerous assumption)?  I think the answer to this question lies in the
structural shifts in the US economy.  Over the last few year the US has
undergone a process of deindustrialization due largley to globalization
and the rise of the multinational corporation which has caused a lost of
many middle class jobs to countries where corporations can find cheaper
labor and less taxes.  With the decreasing in industrial jobs, there have
been a growing number of jobs in the service and technology sectors.
Technology sector jobs are usually upper middle class positions with good
pay, whereas service sector jobs typically pay $5-10 an hour.  Since the
service sector is increasing more rapidly than the technology sector,
wages for most workers are generally being driven down.  This leaves the
upper class taking advantage of free trade while the middle and lower
classes are slowly losing ground.   

But isn't it true that real income by household has been rising at all
levels even though it has been rising faster at the upper end?  Well, this
trend was occurring for years, every while real wages were decreasing.
One reason for this is the fact that women have increasingly joined the
workforce over the last few decades.  Thus, in any given household, it is
more likely that there are two people working in the family.  If this is
considered, the income trends seem quite alarming since real income by
household has only increased slightly at the lower levels.  Two people are
now making little more than one person used to.  In addition, it seems
that real income by household may be on the verge of falling at the lower
levels for the first time in decades because of the fact that the increase
in service sector jobs has been more rapid than the increase in technology
jobs.  

This is why I am sick of hearing about all of the economic prosperity that
the democratic platform and its supporters continue to go on and on about.
The increasing income inequality will only continue to increase the wealth
inequality.  Several times I have heard Nader quote the statistic that 1%
of the population owns as much wealth as the bottom 95%.  Why are there
even people talking about how great the economy is?  I think part of this
can be explained by the fact that Americans have always accepted a high
degree of inequality due to historical, cultural, and religious processes
of legitimation of inequality.  This may be why a large potion of the
economists today only consider certain economic indicators such a Gross
Domestic Product in determining if the economy is doing good.  I guess it
depends how you define a good economy:  by the GDP or by the amount of
equality it allows.  It seems that some sort of balance between the two
might be desirable, but what kind of balance.  I think one way to gain
more information on this topic is to look at Western European social
democracies.  This is where I must stop because I do not know enough about
the economic situations in these countries in terms of GDP or amount of
inequality.  I also do not know what different variables might come into
play that may have an effect on their economic and social success that are
separate from public policy (such as historical trends, culture, etc.).

Anyway, a lot of you probably already know this stuff.  I think I did this
more for my self than for the purpose of getting out the information.  I
just needed to confirm some stuff I already knew (at least as well as
you can know anything, which is to say not very well as the philosophical
body of thought called skepticism has shown us), but questioned for a
moment when I was presented with data that now seems largely false.  I 
hope you found some of it interesting.  Have a good weekend.