| | Statement of William E. Spriggs, Ph.D., Professor and Chair, Department of Economics, Howard University Testimony Before the Full Committee
of the House Committee on Ways and Means January 23, 2007 I wish to thank Committee
Chairman, Congressman Charles B. Rangel for the invitation to offer this
testimony. It is an honor and a privilege to offer this evidence.
Since 2001, the U.S. economy has
been in a recovery as defined by the National Bureau of Economic Research’s
Business Cycle Dating Committee. Gross Domestic Product, the broad measure of
the nation’s economic activity has grown, aggregate consumption has increased
and the unemployment rate has fallen. Those are broad measures to confirm a
sense that the economy is in recovery. However, data from the U.S. Census
Bureau shows this recovery has also seen a fall in the inflation adjusted median
income for working-age families, poverty rates rise, the share of private-sector
workers covered by employer-provided health insurance fall and the share of
private-sector workers covered by employer-provided pension plans. These latter
trends, if not corrected soon, will have important implications for fiscal
policy.
Wage and income inequality have
been on the rise in the United States since the late 1970s. It appears that the
bulk of increase in wage inequality between workers of different education
levels took place in the 1980s. That period was marked with an increase in the
premium paid to college-educated workers, relative to the wages of high
school-educated workers, and prompted debate among economists that the economy
was now experiencing growth that increased rewards to skills; or “skill biased
technological change.” During the 1970’s, men with college education actually
suffered a drop in their premium, from about 25 percent higher than their high
school-educated counterparts to a low of about 20 percent by the decades end in
1979. But, in the 1980s, the premium for men grew to reach 35 percent by
decade’s end, a significant increase.
But, economists are not in
agreement that the skill biased technological change can explain the growth in
wage inequality that has taken place since the 1970s.
A major concern, is that the build-up in wage inequality between workers of
different education levels took place in the 1980s before the broad introduction
of computing, and did not expand greatly in the 1990s when computer use became
ubiquitous and productivity increases returned to their long-run trend after a
slowdown in the 1970s and 1980s. What has continued to rise, and what is the
larger component of over-all wage inequality, are differences in the earnings of
workers who have equal education and experience. Some economists attempted to
reconcile this apparent anomaly by arguing for increases in the returns to
unobserved productivity characteristics (like school quality) that were linked
to schooling differences.
But, a closer look suggests that using more accurate data, and controlling for
shifts in the structure of employment, would show the rise in the returns to
unobserved characteristics took place in the 1980s as well. And, the greatest
growth in inequality among similarly educated workers is among college and
graduate-educated workers, not among workers with high school or less education.
There are other problems with the
skill-biased hypothesis. One of them is that there is great variation between
states in the amount of wage inequality within states, even though the same
technology that drives skill demands affects all states. Observing inequality
within states over time, economists have noted that the decline of large
manufacturing accounts for increases in overall wage inequality within states.
This is not consistent with the skill-biased hypothesis.
Another is that returns to skills by race diverged during that period, which
would be inconsistent if firms truly faced skills shortages that bid up the
wages of skilled workers.
It would also be difficult to explain the much higher use of skilled workers by
foreign-owned companies operating in the U.S. than for domestic producers, and
for the racial disparity in the skilled work forces between foreign-owned and
domestic firms suggested by the patterns of employment in those sectors with
high foreign direct investment in the U.S.
Economists have found the decline of unionization in the 1980s, and the effects
of trade to be important in explaining the growth in overall wage inequality.
The importance of unions and labor market institutions are not consistent with
the skill-biased hypothesis.
Perhaps more of an issue is the
break between productivity gains and wages. During this recovery, productivity
has continued to grow at its post-1995 rate, suggesting a return to its long-run
trend. Yet, median wage levels have not kept paced with inflation. Fast
productivity growth is a way to keep inflation in check, but also a way to
improve the lifestyles of America’s workers. Yet, noted economist Robert Gordon
has found that only the wages of those in the very top ten percent of earnings
have kept above productivity growth over the 1966-2001 period. The
redistribution of gains to the top explains the stagnation of those in the
middle.
During this recovery, wage
inequality has continued to grow. It has grown not because of an increase in
the returns to education, because in the initial phases of the recovery, the
wage premium of college educated workers fell, as they became the larger share
of the long-term unemployed.
Instead, it has been the continued expansion of inequality of earnings for
workers who are similarly educated. Apparently, an important source of the
growth of that inequality is traced to declines in the inflation-adjusted value
of the minimum wage.
But, the other source is the
redistribution of corporate income, from wages to capital income. The latest
data from the Bureau of Economic Analysis shows that the share of
corporate-sector income going to wages is down to its lowest share in over 25
years, according to an analysis done by the Lawrence Mishel and Jared Bernstein
at the Economic Policy Institute.
They also point to new figures from the Congressional Budget Office showing an
increased concentration of corporate capital income among America’s richest one
percent. The latest CBO figures show that almost 60 percent of capital income
goes to the top one percent in the U.S. income distribution.
During this recovery, U.S. Census
data show that income for those in the bottom twenty percent, those in the
middle twenty percent and those in the top twenty percent have all fallen. Yet,
aggregate consumption has increased. This anomaly has occurred, because the
aggregate savings level of Americans has become negative, and household debt has
risen dramatically. But, a closer look at the data shows that those in the
bottom twenty percent have in fact suffered from a drop in consumption. Real
wages for them have fallen, and because they are credit constrained, they have
not borrowed to maintain consumption. Those with middle incomes have apparently
maintained consumption, with some modest borrowing, and some modest benefit from
lower taxes. The big gains in consumption have come from those at the top of
the income distribution, where incomes in the highest ranges have gone up, and
by borrowing, and from larger benefits from tax cuts. The relative gains in
consumption by those in the top twenty percent were more rapid than during the
1980s or 1990s recovery. By 2005, the top twenty percent of the income
distribution accounted for almost 40 percent of all consumption. The bottom
twenty percent consumed only 8.2 percent.
Of course, this personal borrowing
spree is not sustainable. Household debt is growing at annual rate of almost 11
percent during this recovery, compared with a more modest growth of 3.7 percent
in household net worth, leading to a very high household debt to asset ratio.
This means that unless incomes rise to sustain consumption growth, instead of
borrowing, the permanent incomes of Americans are falling. That is, at some
point, consumption must fall so households can balance their incomes.
Further, with the federal budget
deficit, it means that the nation has been borrowing from the rest of the world
at an astounding rate to fuel our consumption. The current account deficit has
mushroomed from about 4 percent of GDP in 2001 to 6.8 percent, as of the third
quarter of 2006. That is a significant claim on future U.S. income by foreign
interests.
There are several important fiscal
policy implications from these current trends. The lower permanent incomes of
Americans, particularly those in the bottom eighty percent of the income
distribution, means they will face real constraints that will ill prepare them
to take on added responsibilities, such as the current shifts away from
employer-provided health care, and the changes in their household balance sheets
toward increased risks resulting from current shifts away from employer
defined-benefit retirement plans.
There are already implications
from the shift of shared prosperity that ended in the 1970s. The shift to
rising incomes only at the highest ends of the income distribution has led to a
significant drop in revenues for the Social Security system, despite continued
growth in the economy, and an apparent return to long run productivity growth
that was not anticipated in the early 1980s. If the Social Security system were
to return to receiving revenue on ninety percent of payroll, almost 40 percent
of the projected shortfall in benefits could be accounted for.
The shift in the nation’s income
shares, toward a lower share of national income in the form of wage and salary
means that tax revenues from earned income, as opposed to capital income, will
need to be re-calibrated. Continued heavy reliance on earned income as a source
of revenue will mean that a rising burdens will be placed on earned income to
pay off current federal obligations. Yet, if the current trends to do not
change, it will already be the case that wage earners will face lower permanent
incomes than the earners anticipated.
| |
