The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
December 14, 1979
More or less Equal?
A mysterious trend in income distribution has
emerged over the past decade. Certainly the
economy has been growing, although
somewhat erratically in the last few years,
so that living standards on average have
improved somewhat. But relative income
matters too, particularly to those at the
bottom, and recently they have been losi ng
ground.
Most Americans would probably guess,
if asked, that the distribution of income has
been growing more equal. Up until about
1968, they would have been right. Butthis
trend was abruptly reversed at the end of the
last decade, and since then the pre-tax
income distribution (inclusive of transfer
payments) has become decidedly less equal.
In focusing on pre-tax income, of course,
we potentially exaggerate the increase
in inequality. A progressive tax system can
help to even out these disparities. Also, many
low-income families receive subsidized
services (health and housing services, for
example) which are not counted in the
transfer payments included here.
Nevertheless, the initial distribution
of rewards is still important, especially if tax
revolts and limitations on government
spending lead to reductions in redistributive
effort.
Measuring the shift
Before trying to decide how this change came
about, it would be worthwhile to define the
problem-in other words, to decide what
we mean by equality of income distribution,
and how it can get better or worse. After all,
we know the distribution of income in the
United States is not exactly equal. Not all
of us live like the Kennedys or the
Rockefellers. On the other hand, the dire
poverty that exists in many countries doesn't
seem to be prevalent here. One popular
measure of equality is the Gini coefficient
of family-income concentration. This
is a summary measure of the share of income
each fraction of the population, by family
unit, receives. The more equal the
distribution, the lower the Gini coefficient.
Another useful measure is the quintile
distribution, which shows the percentage
of total income accruing to each fifth of the
population, from poorest to richest.
As it turns out, both measures tell a similar
story. The Gini bounced around quite a bit,
but moved generally down as distribution
became more equal until 1968. Since then
it has gone steadily up, wiping out about half
the gain of the 1950-68 period. The quintile
distribution, shown in the chart, indicates that
the shifts have been widespread, with the
poorest two-fifths los i ng out to the upper twofifths since 1968, while the middle group has
stayed about even. The biggest losers are
in the second quintile, which had an even
smaller share in 1977 than in 1950. The big
winners over the past quarter-century were
not the very richest, though. They were the
relatively well-off people in the fourth·
quintile. But since 1968, the rich have gotten
relatively richer too.
Worl{er's share
Have the capitalists, in true Marxist fashion,
been enriching themselves at the expense
of the workers? Because property income
is highly concentrated among upper-income
groups, any gain in the overall share of capital
at the expense of labor wou Id tend to increase
the Gini coefficient. But far from this being
the case, the trend has been in the opposite
direction-the labor share of income has
been steadily increasing. In 1950, workers
received 66 percent of the total, but in 1968
they received 73 percent-and in 1977,
76 percent. This should have made the
distribution more equal, not less.
If workers as a whole have been receiving
more but poor people are getting less,
perhaps better-paid workers are gaining
at the expense of the lower-paid. Here the
However, this trend does not explain the
reversal since 1968, because the increase
in the number of workers perfamily was even
stronger in the period between 1950 and
1968. The increase in labor-force
participation rates was quite pronounced
in the earlier period in all but the lowest
quintile.
evidence becomes rather murky. Incomes
of managers and administrators increased
faster than those in any other occupation
between 1968 and 1977, putti ng them ahead
of professional and technical workers as the
most highly paid group. But at the same
time, sales and clerical workers gained
on professional and technical employees,
who are higher on the white-collar ladder.
Differentials also narrowed within the
blue-collar group, as both operatives and
laborers gained a littl€ in relation
to more-highly paid craftsmen.
Since changing incomes within occupations
do not explain the growing disparity
of income distribution, what about shifts
in employment between occupations with
different distributional patterns? Income from
some types of work, notably farm and
nonfarm self-employment, tends to be much
more unequally distributed among its
recipients than income from either bluecollar orwhite-collar jobs. And there are also
substantial differences within these
subgroups. Thus increased employment
in some occupations with less equally
distributed income would raise the Gini
coefficient, while increased employment
in the jobs where income is more evenly
distributed would lower it toward a more
balanced distribution.
In contrast, a tendency toward growing
differentials was quite evident during
the earlier (1950-68) period. Sales and
clerical workers lost in relation to
professional and technical people,
operatives' and laborers' wages grew more
slowly than those of craftsmen, and all
categories of blue-collar workers received
relatively less than their white-collar
counterparts in 1968 than in 1950. Again, the
big success story of the 1950-68 period was
managers and administrators, as they almost
eliminated the salary gap between
themselves and top professionals.
Unfortunately for our thesis, there have been
no major shifts recently in the occupational
distribution of employment. There was some
movement out of farming and out of self-paid
managerial jobs-which should have
increased income equality-but the
movement was small. In contrast, there were
sizable shifts between 1950 and 1968, which
worked in the direction of greater equality.
The big moveme'!ts were out of self-paid farm
and nonfarm jobs and into salaried
professional, technical and managerial
employment, all of which have relatively low
indices of income concentration. Moreover,
considering the fact that farm incomes are
generally much lower than nonfarm
incomes, the movement away from farming
tended to increase the share of lower-income
groups during that period.
More prosperous workers
Another possibility is that prosperous families
have been sending more members into the
work force than the poor, and thus increasing
their share of total income. And yes, in fact,
they have. Between 1968 and 1977 the
average number of earners per family fell in
the lowest two quintiles, while it rose among
the upperthree. The women's movement and
a decade-long rise in unemployment rates
may together account for this conflicting
picture. Cyclical unemployment tends to
raise income concentration because it affects
lower-income groups more heavily than
higher-income groups. Since 1977 was at a
lower point in the cycle than 1968, this may
have had some effect. And since husbands
with relatively high salaries also tend to have
well-paid wives, any increase in female
participation in upper-income groups would
have a magnified effect on family inequality.
Causes summarized
Here, then, is at least part of the answer.
2
probably will be limited at some point,
simply because labor's share is already threefourths of the total. And as long as other
recent trends continue-such as relatively
large income gains by managers and
administrators, and rising labor-force
participation rates among upper-income
families-pretax family income may
continue to be distributed in an increasingly
unequal fashion. To the extentthat
subsidized services to the poor are not
eliminating the differentials, this poses
challenges for progressive income-tax and
transfer policy.
Income equality increased in the earlier
period because people in self-employed
work where income is more concentrated
moved into wage and salary jobs. But this
trend had run its course by 1968. Then the
other unequalizing forces took over, in the
form of increasing wage differentials and
relatively higher participation rates among
upper-income families. Even the continued
shift away from property and toward labor
income has not been able to offset this trend.
What are the prospects for the future?
Probably toward greater inequality, at least
for a while. The shift toward labor income
Jane Haltmaier
Percentage Points
3
CHANGES IN THE SHARE OF FAMILY INCOME, BY QUINTILE
2
1950-1968
1
o
-1
Lowest
Fifth
Second
Fifth
Third
Fifth
-2
3
Fourth
Fifth
SS'd1~
l.Sl:II.:1
UOlSU!4SI?M" 41?ln .. uoSaJO .. I?pl?AaN .. o41?PI
HI?MI?H .. I?!UJoJ!lI?:)
l?uozpV" 1?>lsl?lV
(G)~~~~\ill~Jr~ \ill~CS
~ (G) ~\ill~<£I
.J!I~::> IO:>SpU~J:I u~S
(;SL ·ON llWHld
OI\7'd
1~\7'lSOd ·s·n
11\fW SSV1::> lSlJl:I
~.mJr~~~CQI TI~Jr~1P~~
JJ \ill~mJJJJ\2?dr~crn
BANKING DATA-TWELFTH FEDERAL RESERVE DISTRICT
(Dollar amounts in millions)
Selected Assets and Liabilities
large Commercial Banks
Amount
Outstanding
Change
from
11/28/79
11/21/79
Change from
year ago @
Dollar
Percent
+ 15,884
+ 13.34
+ 15.87
+ 15,330
+ 2,630
+ 9.24
+ 8,709
+ 25.71
NA
NA
NA
NA
Loans (gross, adjusted) and investments*
134,929
+ 48
71
Loans (gross, adjusted) - total#
111,911
50
31,094
Commercial and industrial
Real estate
42,578
+ 175
23,997
Loans to individuals
+ 49
- 149
Securities loans
1,409
771
9.45
U.s. Treasury securities*
7,384
+ 91
15,634
Other securities*
+ 1,325
+ 9.26
+ 28
-1,166
43,388
Demand deposits - total#
+ 2,477
+ 6.05
- 153
630
30,920
Demand deposits - adjusted
+ 2.08
+
1,847
6.07
244
Savings deposits - total
28,592
Time deposits - total#
58,330
+ -147
+ 17.75
+ 8,791
Individuals, part. & corp.
49,784
+ 73
+ 9,432
+ 23.37
(Large negotiable CD's)
+ 11.91
21,753
171
+ 2,315
Comparable
Weekly Averages
Weekended
Weekended
year-ago period
of Daily Figures
11/21/79
11/28/79
Member Bank Reserve Position
10
Excess Reserves (+ )/Deficiency (-)
19
+ 31
+ 187
Borrowings
107
+ 36
4
197
Net free reserves (+ )/Net borrowed( - )
88
Federal Funds - Seven large Banks
Net interbank transactions
+ 712
+ 269
+ 989
[Purchases (+ )/Sales (-)]
6
Net, U.s. Securities dealer transactions
+ 357
+ 81
[Loans (+ )/Borrowings (-)]
* Excludes trading account securities.
# Includes items not shown separately.
@ Historical data are not strictly comparable due to changes in the reporting panel; however, adjustments
have been applied to 1978 data to remove as much as Possible the effects of the changes in coverage. In
addition, for some items, historical data are not available due to definitional changes.
Editorial comments may be addressed to the editor (William Burke) or to the author .... Free copies of
this and other Federal Reserve publications can be obtained by calling or writing the Public Information
Section, Federal Reserve Bank of San Francisco, P.O. Box 7702, San Francisco 94120. Phone (415)
544-2184.
~~JJ~~~~CQ[
@FedFRASER