View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

A Quarterly Review
of Business and
Economic Conditions
Vol. 25, No. 4

Productivity

Is the Slowdown Due
to Retiring of Boomers?

President Bullard

A Review of 2017’s
Key Policy Presentations

Fourth Quarter 2017

THE FEDERAL RESERVE BANK OF ST. LOUIS
CENTRAL TO AMERICA’S ECONOMY®

Shifting Times
The Evolution of the
American Workplace

C O N T E N T S

4
THE REGIONAL

ECONOMIST
FOURTH QUARTER 2017 | VOL. 25, NO. 4

Shifting Times:
The Evolution of the American Workplace

PRESIDENT’S MESSAGE

10

The Role of Baby Boomers
in Productivity Changes

14

Productivity

Is the Slowdown Due
to Retiring of Boomers?

President Bullard

A Review of 2017’s
Key Policy Presentations

Fourth Quarter 2017

Shifting Times
The Evolution of the
American Workplace

Workers and work have changed dramatically since 1950.
Workers are older, more educated and more diverse. Employment
opportunities have shifted to higher-skilled occupations. Even
jobs that have traditionally required low levels of schooling are
employing people with more formal education.

Looking for the Positives
in Negative Interest Rates
By Brian Reinbold and Yi Wen

18

INDUSTRY PROFILE
Advanced Manufacturing
Vital to Eighth District

By Guillaume Vandenbroucke

By Charles Gascon
and Andrew Spewak

Although the Federal Reserve
has never used negative interest
rates, central banks elsewhere
have used them—and continue to
use them—to encourage people to
shift their investments away from
government bonds to something
that will do more to stimulate the
economy.

Director of Research
Christopher J. Waller
Chief of Staff to the President
Cletus C. Coughlin

Growth in productivity in the U.S.
is noticeably slow these days, as
it last was in the 1970s. The baby
boomers might be a reason why:
They had yet to reach their stride at
work in the 1970s, and now they are
aging out of the workforce.

Deputy Director of Research
David C. Wheelock
Director of Public Affairs
Karen Branding
Editor
Subhayu Bandyopadhyay
Managing Editor
Al Stamborski
Art Director
Joni Williams

Please direct your comments
to Subhayu Bandyopadhyay
at 314-444-7425 or by email at
subhayu.bandyopadhyay@stls.frb.org.
You can also write to him at the
address below. Submission of a
letter to the editor gives us the right
to post it to our website and/or
publish it in The Regional Economist
unless the writer states otherwise.
We reserve the right to edit letters
for clarity and length.

Vol. 25, No. 4

By Alexander Monge-Naranjo and Juan Ignacio Vizcaino

3

The Regional Economist is published
quarterly by the Research and Public Affairs
divisions of the Federal Reserve Bank of
St. Louis. It addresses the national, international and regional economic issues of
the day, particularly as they apply to states
in the Eighth Federal Reserve District. Views
expressed are not necessarily those of the
St. Louis Fed or of the Federal Reserve System.

A Quarterly Review
of Business and
Economic Conditions

THE FEDERAL RESERVE BANK OF ST. LOUIS
CENTRAL TO AMERICA’S ECONOMY®

16

P.O. Box 442, St. Louis, MO 63166-0442.

The Eighth Federal Reserve District includes

all of Arkansas, eastern Missouri, southern
Illinois and Indiana, western Kentucky and
Tennessee, and northern Mississippi. The
Eighth District offices are in Little Rock,
Louisville, Memphis and St. Louis.

By Kevin L. Kliesen

By Subhayu Bandyopadhyay
and Javed Younas

In both developed and developing nations, terrorism destroys
life and property. But developing
countries suffer more in terms of
economic growth, foreign direct
investment and trade.

ONLINE EXTRA

N AT I O N A L O V E R V I E W
Economy Absorbs
Blows of Hurricanes

12 Impact of Terrorism
on Developing Countries

Single-copy subscriptions are free
but available only to those with
U.S. addresses. To subscribe, go to
www.stlouisfed.org/publications.
You can also write to The Regional
Economist, Public Affairs Office,
Federal Reserve Bank of St. Louis,

Advanced manufacturing requires
substantial R&D spending and
workers with a high degree of technical knowledge, for which they
are paid a wage premium. Such
manufacturers have a significant
impact on U.S. production and
exports.

Despite initial forecasts of a sharp
slowdown in third-quarter GDP
growth because of the hurricanes
this summer and fall, the pace of
economic activity turned out to be
stronger than expected. The fourth
quarter is also on track for abovetrend growth.
17

21

First-Time Homebuyers
Are Younger,
Less Creditworthy
By Brian Reinbold and
Paulina Restrepo-Echavarria
The number of first-time homebuyers has declined over the
past 16 years, both in the Eighth
District and the rest of the U.S.
A closer look at the District finds
that these buyers are younger
and less creditworthy than those
homebuyers nationally.

E C O N O M Y AT A G L A N C E

www.stlouisfed.org/re

DISTRICT OVERVIEW

23

RE ADER E XCHANGE

How Fast Will Banks Adopt New Technology This Time?
By Drew Dahl, Andrew Meyer and Neil Wiggins
To get an idea of how fast the banking industry might embrace new
financial technologies—“fintech”—it might be worth looking at how quickly
banks entered the internet age with a website almost a generation ago.
2 The Regional Economist | Fourth Quarter 2017

COVER IMAGE © THINKSTOCK / ISTOCK /ZAPP2PHOTO

P R E S I D E N T ’ S

M E S S A G E

A Year in Review

S

t. Louis Fed President James Bullard, a
noted economist and scholar, has been a
participant in Federal Open Market Committee (FOMC) deliberations since April
2008. Bullard actively engages with many
audiences—including academics, policymakers, business and community organizations,
and the media—to discuss monetary policy
and the U.S. economy and to help further
the regional Reserve bank’s role of being the
voice of Main Street.
Some of his key policy presentations during 2017 are summarized below, in chronological order. To see all of Bullard’s public
presentations, please visit www.stlouisfed.
org/from-the-president.
Five Macroeconomic Questions for 2017

Jan. 12, 2017: In New York, Bullard discussed key questions related to the overall
economy and to the Fed in particular. Bullard
said the St. Louis Fed’s recommended policy
rate (the federal funds target rate) depends
mostly on the safe real rate of return, and
such rates are exceptionally low and are not
expected to rise soon. “This, in turn, means
that the policy rate should be expected to
remain exceptionally low over the forecast
horizon,” he said. “The new administration’s policies may have some impact on the
low-safe-real-rate regime if they are directed
toward improving medium-term U.S. productivity growth.”
The Role of the Fed’s Balance Sheet
for the U.S. Monetary Policy Outlook
in 2017

Feb. 28, 2017: Now may be a good time
for the FOMC to begin allowing the balance
sheet to normalize by ending reinvestment,
Bullard said at George Washington University in Washington, D.C. “Adjustments to
balance sheet policy might be viewed as a
way to normalize Fed policy without relying
exclusively on a higher policy rate path,” he
said. He also noted that current FOMC policy
is distorting the yield curve. “Ending balance

President Bullard (left) often travels throughout the St. Louis Fed’s District to share his views on the economy and to listen to
the perspectives of others. In September, he visited Dot Foods, the nation’s largest food redistributor, in Mount Sterling, Ill.

sheet reinvestment may allow for a more
natural adjustment of rates across the yield
curve as normalization proceeds and for
‘policy space’ in case balance sheet policy is
required in a future downturn,” he said. (The
Fed began gradually reducing the size of its
balance sheet in October 2017.)

rate is likely to be appropriate for this regime
over the forecast horizon. “Many future
developments could impact this policy path,
but the Fed does not need to pre-empt any of
them,” Bullard said.

Current Growth, Inflation and Price
Level Developments in the U.S.

Nov. 14, 2017: Inflation has been mostly
below the Fed’s 2 percent target since 2012
and is unlikely to return to target anytime
soon, Bullard said in Louisville, Ky. “Inflation data during 2017 have surprised to the
downside and call into question the idea that
U.S. inflation is reliably returning toward target,” he said. If the FOMC is going to hit the
inflation target, “it will likely have to occur in
2018 or 2019,” he added.

May 26, 2017: In Tokyo, Bullard said that
U.S. macroeconomic data have been relatively weak, on balance, since the FOMC
met in March and raised the policy rate. For
instance, he noted that U.S. inflation and
inflation expectations have surprised to the
downside in recent months. He also said that
even if U.S. unemployment declines substantially further, the effects on U.S. inflation are
likely to be small. Regarding the U.S. price
level, he said that it “has begun to deviate
noticeably from the 2 percent path established in the mid-1990s.” The price level is 4.6
percent below the previously established path.
The Path Forward for
U.S. Monetary Policy

June 23, 2017: In Nashville, Tenn., Bullard said the Fed can wait and see how the
economy develops before making any further
adjustments to the policy rate. He noted that
the U.S. policy rate has been rising while key
policy rates abroad have remained fixed. He
said the U.S. economy remains in a “regime”
of low growth, low inflation and low interest
rates, and that the current level of the policy

When Will U.S. Inflation Return
to Target?

Assessing the Risk of Yield
Curve Inversion

Dec. 1, 2017: In Little Rock, Ark., Bullard
said that there is “a material risk of yield
curve inversion” over the forecast horizon if
the FOMC continues on its present course
for raising the policy rate, as suggested in
September’s Summary of Economic Projections. Such an inversion—where short-term
interest rates exceed long-term interest
rates—has helped predict recessions in the
past. He noted that yield curve inversion is
best avoided in the near term by caution in
raising the policy rate. “Given below-target
U.S. inflation, it is unnecessary to push
normalization to such an extent that the
yield curve inverts,” he said.
The Regional Economist | www.stlouisfed.org 3

4 The Regional Economist | Fourth Quarter 2017

THINKSTOCK / ISTOCK / KINWUN

L A B O R

Shifting Times
The Evolution of the American Workplace
By Alexander Monge-Naranjo and Juan Ignacio Vizcaino
hat are the main characteristics of American workers?
What types of jobs do they do? Who does what? It
turns out that the answers to these questions have been
changing, in some cases dramatically.
For starters, the basic demographic makeup—age, gender
and race—is very different now than it was nearly 70 years
ago. Second, the educational levels of workers have been
increasing dramatically.1 Third, the occupations or types of
jobs employing American workers are very different now
relative to what American workers were doing just a few
decades ago.
In this article, we explore these shifts in the American
labor force and workplace. We show that the identity,
education and occupations of the average American worker
have all been changing. We also show that there are big
changes in who does what, especially in the higher-skilled
and higher-paying occupations.
Overall, the picture emerging from the data is very clear:
American workers are older, more educated and more
diverse. Because skilled workers are more abundant, the
employment opportunities have been shifting to higherskilled occupations, and this movement has taken place
for workers of all genders and races. Workers with loweror even middle-level skills are likely to face relatively
tougher times because their remaining labor market
opportunities are in the lower-skilled occupations.
Demographics and Education

To characterize American workers over the years, we
collected individual level data from IPUMS-USA on the
age, gender, race, educational level and current occupation

of workers.2 For ease of use, we categorized the nine racial
groups in the database into four broader groups: white,
black, Asian and other.3 Similarly, for educational levels,
we grouped the 11 categories in the data into five broader
groups representing the maximum possible level of education attained by these individuals: primary or less (nursery
through grade 8), secondary incomplete (grades 9-11),
secondary complete (grade 12), college incomplete (one to
three years of college), and college complete or more (four
or more years of college).
The table contains the basic demographic information.
A number of salient features are evident. First, female
workers almost doubled their share in the labor force;
nowadays, they are close to being half of the working population. Similarly, nonwhites as a whole more than doubled
their share, accounting for nearly one in four workers.
An even more dramatic increment is in terms of schooling levels: In 1950, close to 40 percent of workers had only
primary schooling (completed or less); today, the U.S. has
only a negligible fraction of workers with such little formal
education. On the opposite extreme, from having less than
18 percent of workers with at least some college, the U.S.
now has about 60 percent of the labor force with either
some college education or a completed college education.
A closer inspection of the data reveals that much of the
changes took place in the 1970s and 1980s, when the baby
boomers entered the labor market. Figure 1 shows the close
relationship between the average age of American workers
and the fertility rate of previous decades.4
The relatively high fertility rates of the 1950s and 1960s
led to an interesting pattern in the age of active workers

The Regional Economist | www.stlouisfed.org 5

Characteristics of American Workers: 1950-2015
Gender

Race

Education
Secondary
Incomplete

Secondary
Complete

College
Incomplete

College
Complete
or More
8.4%

Year

Average Age

Male

Female

White

Black

Asian

Other

Primary
or Less

1950

37.7

72.6%

27.4%

90.0%

9.6%

0.3%

0.2%

38.8%

19.3%

24.3%

9.3%

1960

40.1

68.0%

32.0%

89.8%

9.3%

0.6%

0.3%

29.4%

22.3%

28.4%

10.4%

9.6%

1970

39.3

62.9%

37.1%

89.2%

9.5%

0.8%

0.4%

17.3%

21.0%

35.4%

13.4%

12.8%

1980

37.4

57.8%

42.2%

87.7%

9.7%

1.8%

0.8%

8.3%

15.4%

38.4%

19.3%

18.6%

1990

38.3

54.7%

45.3%

83.1%

10.0%

2.9%

4.0%

3.8%

9.2%

33.2%

45.5%

8.2%

2000

40.0

53.6%

46.4%

78.8%

10.1%

3.8%

7.3%

2.9%

7.7%

38.1%

41.6%

9.7%

2010

43.1

52.3%

47.7%

76.8%

10.8%

5.3%

7.1%

2.8%

5.4%

33.8%

46.5%

11.5%

2015

43.5

52.7%

47.3%

74.9%

11.5%

5.9%

7.7%

2.5%

4.8%

32.8%

47.5%

12.3%

Farmers
and Farm
Laborers

Laborers

SOURCE: IPUMS.

FIGURE 2

Workers’ Average Age and Fertility
in the U.S.

Shifts in the Shares of U.S. Workers across Occupations

44

4

43

3.5

Age

2.5

41

2

40

1.5

39

1

38

0.5

37
1950 1960 1970 1980 1990 2000 2010

0

Average Age of Workers (left axis)
Fertility Rate (right axis)
SOURCES: For the average age, IPUMS; for the fertility rate,
World Bank via FRED.
NOTE: Total fertility rate represents the number of children
who would be born to a woman if she were to live to the end
of her childbearing years and bear children in accordance
with current age-specific fertility rates.

6 The Regional Economist | Fourth Quarter 2017

Children

3

42

Occupational Share in Total Employment

FIGURE 1

30%
1950

25%

1980

2015

Craftsmen

Service
Workers

20%
15%
10%
5%
0%

Professional Managers,
and Technical Officials,
and
Proprietors

Sales
Workers

Clerical
and
Kindred

Operatives

SOURCE: IPUMS.
NOTE: “Clerical and Kindred” includes those occupations whose clerical duties, such as those related to general office work or duties
pertaining to the operation of various office machines, take up a majority of the worker’s time or for which the major requirement is
the ability to perform the clerical duties. “Operatives” includes those occupations in which duties related to operating and handling
machines take up a majority of the worker’s time.

over the years. First, average ages tended to
increase between 1950 and 1960 as young
female workers in the 1950s left the labor
force to rear children. Later, however, when
the baby boomers’ children entered the
labor force in the 1960s, the average age
started to decline. Yet, with the lower fertility rates observed since the late 1970s and
early 1980s, the average American worker
started aging, a trend that has remained up
until at least 2015, the last year for which
we have data.
To be sure, the baby boomers had more
formal education than their parents, but
the boomers’ education has since been
eclipsed by that of their children. It is
easy to see why the 1970s and 1980s were
years of rapid expansion in the average
educational level of American workers.
After that, a steady increase in education

has been sustained up until 2015, and it is
expected to continue.
These changes in the educational level of
American workers are significant enough
that one would expect to see important
changes in the structure of the economy, i.e.,
in the types of occupations in the economy
and the types of workers filling those jobs.
The data show this vividly.
Changes in Work
and in Who’s Doing What

We now explore the changes in what the
American workers do in the marketplace.
To this end, we grouped workers into the
following nine broad groups,5 ordered
by their skill intensity6: professional and
technical workers; managers, officials and
proprietors; sales workers; clerical and kindred; craftsmen; service workers; operatives

(e.g., machine operators); farmers and farm
laborers; and laborers.7
Figure 2 shows the shares of workers
across the nine broad occupation categories
in the data. For ease of presentation, we
reported on the data only for the beginning, the middle and the end of the sample
period. For each occupation, the first bar in
each case corresponds to American workers in 1950, the middle bar corresponds
to workers in 1980 and the last bar corresponds to 2015, the most recent year for
the data.
Figure 2 shows important changes in
what American workers do. First, there is
a big shift toward professional and technical occupations and toward management.
The first group almost tripled its share over
all workers between 1950 and 2015, from
8.7 percent to 25.4 percent of all workers.
The second group, i.e., the management
positions, almost doubled its share, from 8.8
percent to 14.7 percent. Another occupation
that expanded is service workers, a finding
that is not surprising, given the well-known
movement of the U.S. economy toward services and away from agriculture and manufacturing. This movement also explains the
significant decline in craftsmen, operatives
and farm workers.
Beyond these profound changes in the
occupations or job types, we observed
substantial shifts in the types of workers
that are allocated across the different types
of jobs. Each of the nine panels of Figure
3 shows the share of workers with different schooling levels in each of the nine
broad occupation categories. Obviously, the
educational level of the workforce was very
different in 2015 relative to that of 1950 and
even 1980.
Specifically, consider the notable difference in the schooling attainment of workers
in professional and technical occupations
between 1950 and 2015. In 1950, only half
of these workers had completed a college degree. By 1980, those with college
degrees already made up 60 percent of these
workforces and by 2015 they accounted for
70 percent. In 1950, it was not uncommon
to find workers with only a high school
diploma in professional positions; in fact,
one in 10 of these professional workers
had not finished high school, and up to 6
percent of them did not have any secondary

FIGURE 3A

FIGURE 3B

Schooling of Professional and Technical
Workers in the U.S.

Schooling of Managers

100%

100%

80%

80%

60%

60%

40%

40%

20%

20%

0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

2015

College incomplete
College complete or more

0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

FIGURE 3C

FIGURE 3D

Schooling of Sales Workers

Schooling of Clerical Workers

100%

100%

80%

80%

60%

60%

40%

40%

20%

20%

0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

2015

College incomplete
College complete or more

0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

FIGURE 3F

Schooling of Craftsmen

Schooling of Service Workers
100%

80%

80%

60%

60%

40%

40%

20%

20%

0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

2015

College incomplete
College complete or more

0%

1950

Primary or less
Secondary incomplete
Secondary complete

2015

College incomplete
College complete or more

FIGURE 3E

100%

2015

College incomplete
College complete or more

1980

2015

College incomplete
College complete or more

SOURCE FOR ALL FIGURES ABOVE: IPUMS.

FIGURES IN THIS SERIES ARE CONTINUED ON NEXT PAGE.

The Regional Economist | www.stlouisfed.org 7

FIGURE 3G
Schooling of Operatives
(e.g., Machine Operators)
100%
80%
60%
40%
20%
0%

1950

1980

Primary or less
Secondary incomplete
Secondary complete

2015

College incomplete
College complete or more

FIGURE 3H
Schooling of Farmers
100%
80%
60%
40%
20%
0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

2015

College incomplete
College complete or more

FIGURE 3I
Schooling of Laborers
100%
80%
60%
40%
20%
0%

1950

Primary or less
Secondary incomplete
Secondary complete

1980

2015

College incomplete
College complete or more

SOURCE FOR ALL FIGURES ABOVE: IPUMS.

8 The Regional Economist | Fourth Quarter 2017

education at all. Formally or informally,
this subset of professional workers must
have accumulated technical knowledge on
the job. As Figure 3A shows, this group of
empiricist professionals had all but disappeared by 1980 and was completely gone
in 2015.
Even more striking changes can be seen
in workers occupying managerial jobs. In
1950, managers were predominantly workers with no formal college education: Individuals who had no more than a high school
diploma accounted for more than three
in four of American managers. (In 1950,
27.4 percent of managers had only primary
education and only 11 percent of them had
completed college.)
Figure 3B shows the drastic change that
has taken place: In 2015, virtually all managers had completed at least secondary education, almost three-fourths of them had some
form of college education and 46.4 percent of
them had completed at least a college degree.
The movement toward higher levels of
education can be seen also in all other
occupations, albeit to a different extent. In
all of them, there is an increasing share of
college-educated workers and a decline in
workers with primary education only. The
main difference across occupations is in the
incidence of secondary education (complete
and incomplete) and in workers with some
college education. For example, while in
1950 virtually no operative worker had any
college education, in 2015 more than 30 percent of these operators had some college.
It is noteworthy that the agricultural sectors have attracted—or required—workers
with higher levels of education. Nowadays,
almost 31 percent of these workers have
some college education. Notice that similar
numbers apply to the group of laborers.
Despite some ambiguity in the share of
workers who have completed secondary
school over the years, all occupations in the
country have undergone a process of skill
upgrade, namely the movement in which
the same form of task, job or occupation is
now performed by workers with higher skill
levels. 8 This is most evident when looking
at the share of college-educated workers
performing more and more of all these
broadly defined categories of jobs and also
when looking at the sharp decline in the
share of workers with only primary school

completed. This sharp decline appears even
among farmers and laborers, a solid majority of whom have traditionally had only a
primary school education.
Top-Earning Occupations

We now look more closely at the managerial and professional occupations, the two
occupations that have been expanding at the
fastest pace and that are the ones paying the
highest salaries. Figure 4 breaks down the
composition across gender and race groups
for these two broad categories.
As the two panels of Figure 4 clearly
show, both occupations have traditionally
been performed predominantly by white
workers and, up until recently, by predominantly white male workers. But that has
changed profoundly. In 1950, white males
accounted for more than 81 percent of all
managers and for 51 percent of all professional and technical workers. Interestingly, the predominance of white males in
both groups was even higher in 1960 and
1970, likely reflecting large numbers of
younger, highly educated females leaving
the marketplace to raise children. But by
2015, white males accounted for about half
of the managers and for about 34 percent of
professional workers.
The entry of highly educated white
women is one of the main forces behind this
change. From essentially being a rarity in
the 1950s and 1960s—and even the 1970s—
women in management positions accounted
in 2015 for one of every three managers in
the U.S. White women accounted for even
more of the professional occupations, outnumbering white men in 2015.
A second major force of change is the
entry of nonwhite workers. Indeed, from
virtually being negligible in these two broad
groups of higher-paying occupations, nonwhite workers now account for 20 percent of
professionals and 15 percent of managers.
The rise of women and nonwhite workers in the marketplace can be tied to higher
college enrollment rates over time and to
reductions in educational and labor market
distortions and barriers. In the case of
women, some have argued that technological changes favor female skills and that the
combination of women’s higher social skills
with increased cognitive skills has also
played an important role.9

ENDNOTES

FIGURE 4A

Percent

Race and Gender of Managers in U.S.

1

1950
White Male

1960

1970

White Female

1980
Black Male

1990
Black Female

2000
Asian Male

2010

2015

Asian Female

FIGURE 4B

Percent

Race and Gender of Professional and Technical Workers in U.S.
100
90
80
70
60
50
40
30
20
10
0

See Monge-Naranjo.
IPUMS-USA, University of Minnesota, www.
ipums.org. We discarded individuals whose
employment status is unknown or who are unemployed or are not in the labor force, as classified by
the variable EMPSTAT codes 0, 2 and 3. Also, see
Ruggles et al.
3 In the database, racial categories consist of national
origin groups. Beginning in 2000, the race question changed substantially to allow respondents
to report as many races as they felt necessary to
describe themselves. In earlier years, only one
race response was coded. We grouped nine racial
categories reported in IPUMS-USA into four
broader groups: white (IPUMS-USA: White), black
(IPUMS-USA: Black/African American/Negro),
Asian (IPUMS-USA: Chinese, Japanese, Other
Asian or Pacific Islander) and other (IPUMS-USA:
American Indian or Alaska Native, two major
races, three or more major races). IPUMS-USA
contains separate information on ethnicity, in particular, whether a worker has Hispanic ethnicity.
In a future article, we will focus exclusively on the
participation of Hispanic workers in the U.S. labor
force and in the different occupations.
4 Fertility data come from the World Bank and were
obtained via FRED at https://fred.stlouisfed.org.
5 In order for occupations to be comparable across
time, we used the 1950 Census Bureau occupational classification. Each of the nine categories
groups occupations that are similar in nature
according to their three-digit occupational code,
the smallest level of desegregation the Census
Bureau provides.
6 The occupations with the highest percentages of
workers with the top level of education (college or
more) are deemed those that are most skill-intense.
The top four occupations were the same in 2015 as
in 1950.
7 Observations of individuals with unclassified,
missing or unknown occupations are discarded.
8 See Costinot and Vogel.
9 See Rendall, Cortes et al. and Hsieh et al.
10 This point is forcefully made by Hsieh et al.
2

100
90
80
70
60
50
40
30
20
10
0

1950
White Male

1960
White Female

1970

1980
Black Male

1990
Black Female

2000
Asian Male

2010

2015

Asian Female

SOURCE FOR BOTH FIGURES: IPUMS.

Conclusions

We explored the substantial shifts in the
American labor force and workplace over
almost 70 years, showing that the identity,
education, race and occupations of the
average American worker have all been
changing. We documented big changes in
the types of jobs being done by American
workers and on the assignment of jobs
across workers with different educational
levels and other characteristics.
The data discussed here provide a number
of clear lessons. First, American workers
are older, better-schooled and much more
diverse in terms of race and gender. Second,
employment opportunities have shifted to
higher-skilled occupations. Third, there has
been a generalized process of skill upgrading, as all occupations are employing workers with more formal education.
Needless to say, these changes have led
to additional challenges for some groups
of workers: Those with lower levels of

education may be unable to find jobs in
occupations that their parents held with
much less formal schooling. For those
with higher levels of education, they now
have heightened competition from more
individuals with higher education, including groups that were rarely represented in
these ranks in the past, e.g., females and
nonwhites.
Regardless of how much more challenging labor markets become for everyone,
the aggregate productivity is higher when
the country takes advantage of the talent of
all the demographic groups and not just a
subset of them.10
Alexander Monge-Naranjo is an economist at
the Federal Reserve Bank of St. Louis. For more
on his work, see https://research.stlouisfed.org/
econ/monge-naranjo. Juan Ignacio Vizcaino is
a technical research associate at the Bank.

REFERENCES
Cortes, Guido M.; Jaimovich, Nir; and Siu, Henry.
The End of Men and Rise of Women in the HighSkilled Labor Market. Manuscript, 2016. See http://
faculty.arts.ubc.ca/hsiu/work/endofmen_post.pdf.
Costinot, Arnaud; and Vogel, Jonathan. Matching
and Inequality in the World Economy. Journal of
Political Economy, 2010, Vol. 118, No. 4, pp. 747-86.
Hsieh, Chang-Tai; Hurst, Erik; Jones, Charles I.; and
Klenow, Peter J. The Allocation of Talent and U.S.
Economic Growth. Manuscript, 2016. See http://
klenow.com/HHJK.pdf.
Monge-Naranjo, Alexander. Workers Abroad Are
Catching Up to U.S. Skill Levels. Federal Reserve
Bank of St. Louis’ The Regional Economist, Third
Quarter 2017, Vol. 25, No. 3, pp. 6-7.
Rendall, Michelle. Brain versus Brawn: The Realization of Women’s Comparative Advantage.
Manuscript, 2017. See https://sites.google.com/site/
mtrendall/research.
Ruggles, Steven; Genadek, Katie; Goeken, Ronald;
Grover, Josiah; and Sobek, Matthew. Integrated
Public Use Microdata Series: Version 6.0 [dataset].
Minneapolis: University of Minnesota, 2015. See
http://doi.org/10.18128/D010.V6.0.

The Regional Economist | www.stlouisfed.org 9

D E M O G R A P H I C S

Boomers Have Played
a Role in Changes
in Productivity
By Guillaume Vandenbroucke
© THINKSTOCK / ISTOCK

Productivity 101

A typical measure of productivity is
labor productivity, which is gross domestic
product (GDP) per worker.2 Figure 1 shows
that, in the 1970s, the growth rate of labor
productivity was noticeably low.3 This slowdown started in the 1960s, when the growth
rate of labor productivity started to decline.
The growth rate of labor productivity accelerated between 1980 and 2000. Since 2000,
another decline is noticeable. It is interesting to note that the current state of low labor
productivity growth is comparable to that of
the 1970s and that it results from a decline
that started before the 2007 recession.
10 The Regional Economist | Fourth Quarter 2017

FIGURE 1
The Growth Rate of GDP per Worker in the U.S., 1955-2014
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
2012

2009

2006

2003

2000

1997

1994

1991

1988

1985

1982

1979

1976

1973

1970

1967

1964

1961

1958

0
1955

n the 1970s, the U.S. economy experienced
a prolonged period of low productivity
growth. Nowadays, growth in productivity
is once again slower than expected.
The causes of these slowdowns have been
much debated. The 1970s’ slowdown has
often been associated with, among other
causes, high energy prices following the 1973
oil price shock, increased antipollution regulations and a decline in the quality of education.1
The current productivity slowdown is often
associated with the 2007-08 financial crisis.
In this article, I hypothesize that the two
slowdowns are related to a single, common
factor: the baby boom, that period from
1946 to 1957 when the birth rate increased
by 20 percent. This hypothesis is not to say
that the baby boom was entirely responsible
for these two episodes of low productivity
growth. Rather, it is to point out the mechanism through which the baby boom contributed to both. Exactly how much did the
baby boom contribute to these slowdowns?
The answer to that question is beyond the
scope of this article.

Percent

I

SOURCES: Bureau of Economic Analysis and Bureau of Labor Statistics, via FRED.

To understand how the baby boom may
have contributed to both the 1970s slowdown and the current slowdown, it is worth
taking a detour to think about what makes a
worker productive.
Current thinking is that workers supply
“human capital services” to their employer.
Sometimes one can refer to “skills” or simply
“productivity.” The exact terminology is not
critical. What is critical is the theory that
young workers have relatively low human
capital and that, as they become older, they
accumulate human capital.
The accumulation of human capital can be
achieved in multiple ways. One is simply via
experience: Older workers have more human
capital, i.e., they know more just because they
have done more and have experienced “learning by doing.” Another possibility is that
workers go through periods of formal on-thejob training throughout their careers; so,
they learn more as they grow older. Human
capital is what makes a worker productive:
The more human capital, the more output a
worker produces in a day’s work.

Picture, then, a typical worker’s human
capital profile throughout life. A stylized
representation of this profile is in Figure 2.
In theory, such a human capital profile
implies that a worker’s earnings profile
should look very similar. This is because, in
theory, workers are paid according to their
productivity. Interestingly, this is exactly
the case in the United States: The data show
that the typical earnings profile throughout
a worker’s life increases until it reaches a
peak, usually a few years before retirement.
What, then, does human capital theory tell
us about U.S. productivity?
Who Is More Productive?

Start with a simple example. Suppose that
there are only young and old workers. Each
young worker produces one unit of a good,
while each old worker produces two units
since the old worker has more human capital
(Figure 2). Suppose now that there are 50
young and 50 old workers. The total number
of goods produced is 150 and, therefore, labor
productivity is 150/100=1.5.

1 See Cullison.
2 Another measure of productivity is total factor

productivity, also called multifactor productivity,
which gauges the joint contribution of labor and
capital to output, instead of the contribution of
labor only, as does labor productivity.
3 The growth rate of productivity in Figure 1 was
smoothed to remove frequent variations and to
focus on secular changes.
4 The total number of workers is kept constant in this
example, but that does not matter. Suppose there
were 10 times more workers: 750 young and 250 old.
Labor productivity would still be 1.25.
5 The share of people between ages 23 and 33 is a
proxy for the share of young people. This does not
imply that the old are all the people older than 33.

REFERENCE
Cullison, W.E. The U.S. Productivity Slowdown: What
the Experts Say. Economic Review, Federal Reserve
Bank of Richmond, July/August 1989, pp. 10-21.

FIGURE 2

Human Capital

A Stylized Profile for a Worker’s Human Capital

20

25

30

35

40

45

50

55

60

65

Age
SOURCE: Author.
NOTE: In theory, a worker’s earnings reflect his or her human capital and should be increasing until the earnings reach a peak shortly
before retirement. In the U.S. data, the typical earnings profile of a worker displays this exact pattern.

FIGURE 3
The Growth Rate of GDP per Worker and the Share of 23-33-Year-Olds in the U.S.,

4.0

39

3.5

37

3.0

35

2.5

33

2.0

31

1.5

Percent

1955-2014

29

1.0

27
25
2012

2009

2006

2003

2000

1997

1994

1991

Share of population ages 23-33 (right axis)
1985

1982

1979

1976

1973

1970

1967

1964

0

1961

Growth rate of real GDP per worker (left axis)

1988

0.5
1958

What do these observations mean for
productivity measurement?
It is important to realize that, should the
theory proposed here be correct, there exists
a sense in which the productivity slowdowns
(especially in the 1970s) are statistical artifacts, that is, it may be that the productivity
of individual workers did not change at all
during the 1970s, but that the change in the
composition of the labor force caused the
slowdown in labor productivity. In a way,
therefore, there is nothing to be fixed via government programs. Productivity slows down
because of the changing composition of the
labor force, and that results from births that
took place at least 20 years before.

ENDNOTES

Guillaume Vandenbroucke is an economist at
the Federal Reserve Bank of St. Louis. For more
on his work, see https://research.stlouisfed.org/
econ/vandenbroucke. Heting Zhu, a research
associate at the Bank, provided research
assistance.

1955

Is There a Problem to Be Fixed?

If we knew exactly how much human
capital each worker has, better measures
of productivity could be constructed. This,
however, is a difficult endeavor since human
capital is not directly observable.
The literature devoted to the measurement
of human capital is large. Significant progress
has been made, but much remains to be
learned.

Percent

But what if there were a larger proportion
of young workers? Suppose that there are
75 young and 25 old. The total production
would be 125 and, therefore, labor productivity would be 1.25. Thus, the increased
proportion of young workers reduces labor
productivity as we measure it via output
per worker.4
The mechanism just described is exactly
how the baby boom may have affected the
growth rate of U.S. labor productivity. Look
at Figure 3. The blue line represents the
growth rate of labor productivity, as in
Figure 1. The red line represents the share of
people between the ages of 23 and 33 (relative
to the population between the ages of 23
and 63).5 An increase in the red line means
that the 23-33 population represents a larger
share of the U.S. population. The peak circa
1980 is the direct consequence of the baby
boom: The U.S. birth rate peaked circa 1960,
implying a large share of people in their 20s
circa 1980. Note in Figure 3 that the two lines
move mostly in opposite directions except
during the 2000s. The correlation between
the two lines is, indeed, –37 percent.
Note also that the share of 23-33 year-olds
is increasing since the late 2000s. This can
also be viewed as a result of the baby boom:
The baby boomers are slowly leaving the
23-63 population, tilting the scale toward the
younger population once again. This trend is
noticeably less pronounced, however, during
the 2000s than it was during the 1970s. Thus,
the mechanism discussed here is likely to be
a stronger contributor to the 1970s slowdown
than to the current one.

SOURCES: Bureau of Economic Analysis and Bureau of Labor Statistics, via FRED; and the Human Mortality Database of the University
of California, Berkeley, and the Max Planck Institute for Demographic Research (Germany), available at www.mortality.org
or www.humanmortality.de.
The Regional Economist | www.stlouisfed.org 11

E C O N O M I C

D E V E L O P M E N T

Trade and Terror:
The Impact of Terrorism
on Developing Countries
By Subhayu Bandyopadhyay and Javed Younas
© THINKSTOCK / ISTOCK / MIMADEO

E

conomists Walter Enders and Todd
Sandler defined terrorism as the premeditated use of or threat to use violence by
individuals or subnational groups to obtain a
political or social objective through the intimidation of a large audience beyond that of the
immediate victims. Central to this definition
is the widespread sense of vulnerability that
individuals or businesses in a venue nation—a
country where the violence occurs—must feel.
This sense of vulnerability is particularly
damaging to trade or foreign direct investment (FDI) because foreign nations always
have a choice of conducting business with
less-terror-prone nations. The decline in trade
and foreign investments compounds the difficulties of developing nations, which suffer a
myriad of economic and noneconomic costs
associated with the loss of life and property
from terror attacks. This article focuses on the
economic costs that are imposed by terrorism
on developing nations through diminished
economic growth, trade and FDI.
Terrorism incidents are classified into two
broad categories, “domestic” and “transnational.” Domestic incidents are ones in
which the perpetrators, victims and damaged properties belong to the venue nation.
In contrast, transnational terrorism involves
different nationalities.
The table reports data for total terrorism, domestic terrorism and transnational
terrorism incidents and associated fatalities
and injuries for the 12 most-terrorism-prone
countries in the world and for the world as a
whole.1 These 12 nations account for almost
79 percent of global terrorist incidents. It is
also notable that most of these nations are
developing countries.
It is understandable that developing
nations are more vulnerable to terrorism
12 The Regional Economist | Fourth Quarter 2017

because they are unlikely to have the
resources to adequately fight terrorism. This
problem is often compounded by corruption, poor governance, and lack of proper
judicial systems or rule of law in these
nations. Such institutional shortcomings
breed discontent in the population, which
in turn can spur terrorism.
Notice that almost 87 percent of global
terrorist incidents are domestic (12,899 out of
a total of 14,820). Therefore, the vast majority
of damages due to terrorism are borne exclusively by the citizens of the venue country.
The associated rise in security costs and loss
in productivity of the workforce—through
damages to labor and capital—are likely to
reduce national income.
Transnational incidents, although less
numerous, have significant economic implications, especially through loss in trade and
FDI. Transnational incidents involve foreign
citizens and therefore garner international
press attention. Such publicity makes foreign nations less willing to do business with
a terrorism-prone nation, leading to less
trade and FDI.
Growth Effects

A 2004 study by economists Brock
Blomberg, Gregory Hess and Athanasios
Orphanides used a sample of 177 nations
(developed and developing) over the period
of 1968 to 2000 to estimate the effect of
terrorism on growth rates of gross domestic
product (GDP). They found that transnational terrorism has rather modest effects
on the economy, reducing per capita GDP
growth by 0.048 percent in a given year.
A 2009 paper by Todd Sandler and his
co-author Khusrav Gaibulloev highlighted the
differences between developed and developing

nations by dividing a sample of 42 Asian
nations into seven developed and 35 developing nations. They did not find any significant adverse effect on growth for developed
nations. However, an additional transnational
terrorist incident (per million people) reduced
an affected developing nation’s growth rate
by around 1.4 percentage points.
Foreign Direct Investment

Greater terrorism in a developing nation
raises the risk for foreign investors of not
being able to get the returns to their investments in the future. Such investors will look
for safer alternate nations to invest in.
Economists Alberto Abadie and Javier
Gardeazabal investigated this issue in a paper
published in 2008 and found that there is
substantial diversion of FDI from a venue
nation of terrorism to alternate terror-free
nations. One standard deviation increase in
the risk of terrorism in a particular nation
can reduce the country’s net FDI position by
approximately 5 percent of its GDP.
This is a huge potential loss in capital formation for any nation, but it is especially hard on
a developing nation that seeks to use foreign
investments to fuel its growth. A 2014 paper
by economists Subhayu Bandyopadhyay, Todd
Sandler and Javed Younas focused on a sample
of 78 developing countries from 1984 to 2008.
They found that a one standard deviation
increase in domestic terrorist incidents per
100,000 people reduces net FDI by between
$323.6 million and $512.9 million for the
average sample country, while the comparable
reduction in the case of transnational terrorist
incidents is between $296.5 million and $735.7
million. They also found that foreign aid can
substantially mitigate terrorism-related FDI
damages due to greater aid flows.

ENDNOTE

Terrorism Incidents and Casualties Summed over the Period 2001-2012
Terrorism
incidents

Domestic
Terrorism
fatalities

Terrorism
injuries

15,066

2,737

6,693

14,075

191

407

843

4,371

9,855

2,229

3,614

7,909

78

621

1,716

820

2,995

985

788

2,786

21

20

186

842

2,164

1,680

712

1,829

1,498

92

305

181

810

1,707

2,450

708

1,537

2,307

91

146

126

Russia

722

1,884

3,901

670

1,655

3,654

21

191

214

Philippines

702

862

2,280

621

779

1,960

51

66

239

Colombia

620

1,000

2,171

540

896

1,939

37

47

181

Israel

546

738

3,585

482

551

2,772

42

170

798

Nepal

323

439

713

282

411

607

27

8

69

Turkey

321

292

1,149

264

192

809

32

50

143

Yemen

313

648

685

261

573

627

42

59

52

14,820

33,910

62,651

12,899

26,135

52,179

1,296

6,894

9,273

Country

Terrorism
incidents

Total
Terrorism
fatalities

Terrorism
injuries

Pakistan

3,043

7,282

India

2,438

Thailand

1,027

Nigeria
Somalia

World
(167 countries)

Transnational
Terrorism Terrorism Terrorism
incidents
fatalities
injuries

SOURCE: Global Terrorism Database.
NOTES: Afghanistan, Iraq, Syria, and West Bank and Gaza are not included due to warlike/civil conflict situations there. Total terrorism
incidents and casualties include incidents and casualties from domestic and transnational terrorism and from those terrorism incidents
that cannot be unambiguously categorized into either of the two categories.

International Trade

Terrorism raises the costs of doing business
across national borders. For example, shipping
costs will rise if shippers have to buy insurance to cover possible damages in the ports of
terrorism-prone nations. In turn, such costs
are passed on to the consumers in the form of
higher prices, which will tend to reduce both
exports and imports of terror-affected nations.
Consider a pair of developed nations.
Based on the table, which clearly shows that
the most terror-prone nations are developing nations, we would not expect terrorism
to be a significant deterrent to trade between
this developed country pair. On the other
extreme, consider a pair of developing
nations—and to make the case clear, consider
a pair from the top 12 nations in the table.
For this pair, a good exported by one nation
and imported by the other suffers potential risks in transportation in both nations.
This will contribute to higher trade costs
and prices and be a significant deterrent to
trade. A 2004 paper by economists Volker
Nitsch and Dieter Schumacher found that a
doubling in the number of terrorist incidents
over the period 1960 to 1993 is associated
with a decrease in bilateral trade among 200
countries by about 4 percent.
There is evolving literature on this issue,
with some papers finding more modest
effects of terror on trade. Among other

reasons, this may be due to changes in a
nation’s production patterns in response to
terror-related disruptions. For example, if
terror disproportionately disrupts an importcompeting domestic industry in a developing
nation, that nation may be forced to turn to
imports for the good in question, thus raising
rather than reducing trade.

1

The data are drawn from the Global Terrorism
Database (GTD) online, which records domestic,
transnational and other terrorist incidents that
cannot unambiguously be placed into either of the
two categories (National Consortium for the Study
of Terrorism and Responses to Terrorism, 2014).
For this table, we have summed data over the period
2001-2012.

REFERENCES
Abadie, Alberto; and Gardeazabal, Javier. Terrorism
and the World Economy. European Economic
Review, January 2008, Vol. 52, No. 1, pp. 1-27.
Bandyopadhyay, Subhayu; Sandler, Todd; and Younas,
Javed. Foreign Direct Investment, Aid, and Terrorism. Oxford Economic Papers, January 2014,
Vol. 66, No. 1, pp. 25-50.
Blomberg, S. Brock; Hess, Gregory D.; and
Orphanides, Athanasios. The Macroeconomic
Consequences of Terrorism. Journal of Monetary
Economics, July 2004, Vol. 51, No. 5, pp. 1007-32.
Enders, Walter; and Sandler, Todd. The Political
Economy of Terrorism, Second Edition. New York:
Cambridge University Press, 2012.
Gaibulloev, Khusrav; and Sandler, Todd. The Impact
of Terrorism and Conflicts on Growth in Asia.
Economics and Politics, November 2009, Vol. 21,
No. 3, pp. 359-83.
National Consortium for the Study of Terrorism and
Responses to Terrorism (START). Global Terrorism
Database. 2014. See www.start.umd.edu/gtd.
Nitsch, Volker; and Schumacher, Dieter. Terrorism
and International Trade: An Empirical Investigation. European Journal of Political Economy, June
2004, Vol. 20, No. 2, pp. 423-433.
Sandler, Todd. The Analytical Study of Terrorism:
Taking Stock. Journal of Peace Research, March
2014, Vol. 51, No. 2, pp. 257-71.

Conclusion

We have discussed some of the economic
costs of terrorism. There are myriads of other
costs like destruction of infrastructure, flight
of skilled workers (brain drain) and diversion
of funds to counterterrorism (compared to
funding of health, education, etc.). A comprehensive discussion of these costs is beyond
the scope of this article. However, a greater
understanding of terrorism-related damages
can help governments and multilateral organizations (e.g., United Nations, World Bank)
to better direct scarce resources to mitigate
terrorism-related costs.
Subhayu Bandyopadhyay is an economist at
the Federal Reserve Bank of St. Louis, and
Javed Younas is an associate professor of
economics at American University of Sharjah,
United Arab Emirates. Research assistance
was provided by Rodrigo Guerrero, a senior
research associate at the Bank. For more on
Bandyopadhyay’s work, see https://research.
stlouisfed.org/econ/bandyopadhyay.
The Regional Economist | www.stlouisfed.org 13

M O N E T A R Y

P O L I C Y

Looking for the Positives
In Negative Interest Rates
By Brian Reinbold and Yi Wen

T

he 2007 global financial collapse resulted
in central banks around the world taking unprecedented action to combat weak
aggregate demand in both consumption and
investment. In the United States, the Federal
Reserve implemented a zero-interest-rate
policy, slashing the federal funds rate to the
range of 0-0.25 percent beginning in late
2008. It was seven years later before the Fed
raised rates—and then it was just by 25 basis
points. Today, the target for the fed funds rate
stands at a range of 1.25-1.50 percent.1
Although the U.S. has never used negative
interest rates (NIR), many other industrial
nations have implemented them to spur
their economies and continue to use them.
For example, Denmark, Japan, Hungary,
Sweden, Switzerland and the entire euro
area have implemented negative nominal
interest rates. The nominal interest rate
in the entire euro area has been negative
since 2014. Among this group of countries,
Switzerland has the lowest level, at 75 basis
points below zero. (See the figure.)
The use of negative interest rates raises
three important questions for monetary
theory. First, given the widely held doctrine
of the zero lower bound on nominal interest
rates, how is a negative interest rate policy
possible? Second, if an NIR is possible, will
it effectively stimulate aggregate demand?
Finally, is it desirable to keep the nominal
interest rate very low for so long? This article
addresses these questions.
Different Countries, Different Rates

In general, the overnight lending rate on
loans and deposits from a central bank to
commercial banks is called a policy rate.
In the U.S., this rate is the federal funds
rate. This overnight lending rate is a key
14 The Regional Economist | Fourth Quarter 2017

economic tool for central bankers as it can
be used to adjust the cost of borrowing,
which influences real economic activity.
For example, since the Fed’s lending (or
deposit) rate directly translates into shortterm government bond yields (e.g., through
open market operations), low interest rates
incentivize others to shift investment from
low-yielding government bonds to moreproductive investments.2
The interest rate for the euro area, set by
the European Central Bank (ECB), is the
overnight deposit rate that banks receive.
In Sweden, the official policy rate is the
repo rate, which is the rate of interest at
which banks can borrow or deposit funds
at the Riksbank for a period of seven days.
Normally, the overnight deposit rate is
0.75 percentage points lower than the repo
rate, and the overnight lending rate is 0.75
percentage points higher than the repo rate
at the Riksbank. The monthly average is
reported here.
Japan’s policy rate is the overnight deposit
rate on excess reserve balances.
Switzerland’s central bank does not set a
target interest rate but instead sets a target
range based on the three-month Libor
(London Interbank Offered Rate) for threemonth interbank loans in Swiss francs.
The reported policy rate in the figure is the
midpoint of this range.
The policy rate set by Denmark’s central
bank is the rate charged on certificates of
deposit. The certificates of deposit are sold
on the last banking day of the week and
typically mature one week later.
The rate reported for Hungary is the overnight lending rate on deposits, analogous
to the Federal Reserve’s policy rate, the fed
funds rate.

Conventional Wisdom

Conventional monetary theory always
assumed that the policy rate cannot go
below zero because an individual would not
pay, in theory, to lend out his or her own
money. Instead, people would hoard cash
to prevent nominal rates from falling below
zero. Since the policy rate is closely linked
to the rate of return on short-term government bonds, the bond yield is also assumed
bounded below by zero—the nominal rate
of return on cash.
When the zero lower bound is reached,
this situation is referred to by economists as
a liquidity trap, the point at which further
monetary injections do not stimulate the
economy because people opt to hoard all
cash available instead of investing or spending it. So, further monetary injections by
the government would only end up hoarded
by people or the banking system instead of
being lent out and circulated in the economy. In monetary theory, this situation of
low circulation is also called zero velocity
of money because money is not circulated
in the economy.
However, if there are costs for people or
institutions to hoard cash, then it is possible
for banks to charge depositors by offering
a negative interest rate. This means that
depositors need to pay to have banks hold
cash for them, or commercial banks must
pay to have the central bank keep their
deposits. In this case, the nominal deposit
rate can go negative without getting into the
liquidity trap. Of course, how negative the
nominal interest rate can go depends on the
costs of holding cash in hand.
In other words, negative nominal interest rates are possible because there are
costs to holding cash, especially for large

What the Model Shows

Researchers Feng Dong and Yi Wen
recently created a theoretical model with
costs of holding cash to capture the negative interest rate phenomenon as seen in the
figure. They showed that when aggregate
demand for investment and consumption
is extremely weak, it is optimal for central
banks to implement negative interest rates.4
This policy would potentially reduce the
cost of borrowing and stimulate investment
spending.
In addition, these authors showed that the
competitive interest rate on bank loans may
move more than one-for-one with changes
in the expected inflation rate, in contrast to
the conventional wisdom. The conventional
wisdom holds that given total bank deposits,
a 1 percent increase in the expected inflation
rate would induce a one-for-one increase in
the nominal interest rate on bank loans to
keep the lender indifferent between lending
and not lending. However, this conventional wisdom fails to take into account
the adverse general-equilibrium effect of
inflation on total deposits. If total deposits
decline as a result of the inflation increase,
the competitive nominal interest rate would
increase more than the increase in the
expected inflation rate to keep the lender
just as well off.
Indeed, we know that people opt to hold
less cash when the inflation rate is expected

to be high. This implies that there is less
money to be deposited into the banking
system. So the nominal interest rate on
bank loans has to increase more than the
anticipated increase in inflation for profitmaximizing banks to break even. In this
case, the correct definition of the real interest rate is no longer the difference between
the nominal interest rate and the expected
inflation rate, but something else. This
means that under negative-interest-rate
policy, the conventionally defined real
interest rate (by the Fisherian relationship,
Nominal Interest Rate ≈ Real Interest Rate
+ Inflation) tends to overestimate the level
of the real interest rate (namely, the real
interest rate may be more negative than the
conventional Fisherian principle suggests).
Not So Far-Fetched, After All

Negative interest rates may seem ludicrous since why would an individual buy
a government bond with a negative yield,
but this is what a central bank would like
you to think. The central bank’s goal is to
incentivize agents to shift investments away
from government bonds to something more
productive economically, thus stimulating
the economy.

ENDNOTES
1
2

3

4

As of Federal Open Market Committee meeting in
December 2017.
The overnight rate is the interest rate at which
a depository institution lends funds to another
depository institution (short term), or the interest
rate the central bank charges a financial institution
to borrow money overnight. The rate increases
when liquidity decreases (when loans are more
difficult to come by) and decreases when liquidity
increases (when loans are more readily available).
The Federal Reserve influences the overnight
rate in the United States through its open-market
operations. For example, selling government bonds
can increase the bond yield and the overnight rate
because these sales reduce the money supply to the
economy. Hence, the overnight rate and bond yield
move together.
For example, it is costly to build and secure a large
private vault by private individuals or corporations,
and such facilities yield no gains in normal times.
See Dong and Wen.

REFERENCE
Dong, Feng; and Wen, Yi. Optimal Monetary Policy
under Negative Interest Rate. Working Paper
No. 019A, Federal Reserve Bank of St. Louis,
May 2017.

Yi Wen is an economist at the Federal Reserve
Bank of St. Louis, and Brian Reinbold is a
research associate there. For more on Wen’s
work, see https://research.stlouisfed.org/econ/wen.
Central Banks’ Policy Interest Rates
2
Euro area
Switzerland

Sweden
Denmark

Japan
Hungary

1
Percent

corporations.3 The central bank can also
require (by law) large corporations to keep
their cash, savings and loans in the banking
system when a negative interest rate policy
is implemented. The same argument applies
to commercial banks that deposit their cash
in the central bank. If the effective returns
to cash go negative, then short-term yields
of government bonds can also go negative,
suggesting that there is still demand for
government-issued debt even if it pays a
negative interest rate.
This means that the lower bound of the
nominal interest rate is not zero, but lower
than zero, if there are costs of holding
(hoarding) cash. So long as the negative
interest rate falls short of reaching its lower
bound determined by the cost of holding cash, conventional monetary policies
remain as effective as in the case of positive
interest rates.

0

–1
Jan.’14

May ’14

Sept.’14

Jan.’15

May ’15

Sept.’15

Jan.’16

May ’16

Sept.’16

Jan.’17

May ’17

Sept.’17

SOURCES: European Central Bank, Riksbank, Denmark Nationalbank, Swiss National Bank, Bank of Japan, Central Bank of Hungary,
Haver Analytics, Bloomberg, World Bank, Trading Economics.

The Regional Economist | www.stlouisfed.org 15

N A T I O N A L

O V E R V I E W

Probabilities of Different Levels of Inflation
0.8
0.7

By Kevin L. Kliesen

T

wo major hurricanes hit the U.S. mainland in August (Harvey) and September
(Irma).1 Given the population and economic
significance of the impacted regions, most
forecasters immediately downgraded prospects
for the U.S. economy’s growth of real gross
domestic product (GDP) in the third quarter.
Although the hurricanes reduced U.S.
employment in September, employment
subsequently recovered in October. Despite
initial forecasts of a sharp slowdown in
third-quarter real GDP growth, the pace of
economic activity turned out to be stronger
than expected.
Forecasters continue to see above-trend
real GDP growth in the fourth quarter, bolstered by the burst in economic activity that
normally occurs during the recovery and
rebuilding phase after natural disasters.
Economic Effects of Natural Disasters

Typically, natural disasters disrupt
activity in three key ways. First, disasters
destroy lives, property and other factors of
production. These are termed direct losses.
These losses reduce the region’s and, if large
enough, the nation’s wealth and tend to
adversely affect productivity, income and
profits in the short term.
Second, indirect losses occur as a result
of the disaster’s direct losses. These indirect
losses include disruptions to the supply
chain, upending the efficient distribution
of goods and services, as well as lost sales
and increased costs for businesses. Some of
these losses (e.g., restaurant meals or airline
services) can never be made up.
Finally, natural disasters eventually trigger a rebound in economic activity, as structures, furniture, appliances and vehicles
are repaired or replaced. For example, U.S.
auto sales rose sharply in September and
remained at a high level in October.
16 The Regional Economist | Fourth Quarter 2017

0.6
Probability

Economy
Bounces Back
from Hurricanes

0.57

0.5

0.37

0.4
0.3
0.2

0.05

0.1
0.0

July ’12

Jan. ’13

July ’13

Jan. ’14

July ’14

Jan. ’15

Price Pressures Measure (inflation above 2.5 percent)
Deflation probability (inflation below 0 percent)

July ’15

Jan. ’16

July ’16

Jan. ’17

July ’17 0.00

Inflation between 1.5 and 2.5 percent
Inflation between 0 and 1.5 percent

SOURCE: Federal Reserve Bank of St. Louis.

This chart plots the four St. Louis Fed Price Pressures Measures (PPM). Each series measures the probability that the personal consumption
expenditures price index (PCEPI) inflation rate over the next 12 months will fall within a certain bucket. The four buckets are as follows: below
0 percent, between 0 and 1.5 percent, between 1.5 and 2.5 percent, and above 2.5 percent. For example, the probability for the above 2.5
percent bucket (“Price Pressures Measure”) is 0.05, which indicates there is a 5 percent probability inflation will exceed 2.5 percent over the
next 12 months.
All data for this article are as of Dec. 1.

Developing Momentum

Despite the hurricane-spawned disruptions, U.S. real GDP accelerated at a 3.3
percent annual rate in the third quarter. The
second estimate was modestly stronger than
the advance estimate. The advance estimate
of 3 percent was very close to the St. Louis
Fed’s Economic News Index (ENI) estimate,
which had predicted third-quarter growth
of 2.9 percent.
With the hurricanes in the rearview mirror, the near-term outlook for the economy
is brightening. Business surveys, such as
the purchasing managers reports and the
national homebuilders survey, indicated
high levels of activity in September and
October. Importantly, business-capital
expenditures continue on an upward
trajectory.
Likewise, consumer confidence continues
to trend higher, reflecting record-high stock
prices and healthy labor market conditions.
Indeed, the unemployment rate fell to 4.1
percent in October, its lowest level since
December 2000.
Wage gains have also picked up, albeit at
a sluggish pace. Importantly, labor productivity growth is finally beginning to accelerate, which would be a catalyst for stronger
wage and real GDP growth.
Another factor helping to bolster the U.S.
economy is the improving global economic

outlook, which has triggered an upswing in
U.S. exports.
At the same time, the construction industry has slowed, mostly because of slowing in
the multifamily and commercial segments.
Housing sales have slowed, but homebuilders generally report that this reflects supply
shortages (e.g., labor and lots) rather than a
softening in demand.
The St. Louis Fed’s ENI predicted on
Dec. 1 that real GDP will increase at a
3.1 percent rate in the fourth quarter.
Inflation Developments

The effects of Hurricane Harvey were
notable because it affected the heart of the
nation’s petrochemical industry on the Gulf
Coast. As refineries, pipelines and chemical
production facilities shut down, prices of
gasoline, diesel fuel and petroleum-based
products like resins and plastics rose appreciably; price increases were passed along
to consumers and producers to varying
degrees. However, as production returned to
normal, these supply shortages abated and
prices retreated accordingly.
Likewise, Hurricane Irma roared through
Florida, disrupting its important tourism and
agricultural industries. Food price increases
were already on the upswing since fall 2016,
and Irma may put additional upward pressure on them. The recent fires in northern

E C O N O M Y

A T

A

G L A N C E

All data as of Dec. 1.
REAL GDP GROWTH

PERCENT

4

2

0

–2

Q3
’12

’16

All Items, Less Food and Energy

2

0

–2

’17

2.75

October

’12

’13

’14

’15

’16

’17

RATES ON FEDERAL FUNDS FUTURES ON SELECTED DATES

1.70

5-Year

2.50

1.60

10-Year

2.25

1.50

05/03/17

10/20/17

06/14/17

11/01/17

07/26/17

1.40

20-Year

2.00

1.30

1.75

1.20

1.50

1.10
1.00

1.25

0.90

Nov. 24

’13

’14

’15

’16

0.80

’17

NOTE: Weekly data.

C I V I L I A N U N E M P L O Y M E N T R AT E

1st-Expiring
Contract

3-Month

6-Month

12-Month

CONTRACT SETTLEMENT MONTH

I N T E R E S T R AT E S
4

10

10-Year Treasury

9
3

8
PERCENT

7
6
5

2
Fed Funds Target

1

1-Year Treasury

4
3
’12

October

’13

’14

’15

’16

0

’17

October

’13

’14

’15

’16

’17

NOTE: On Dec. 16, 2015, the FOMC set a target range for the
federal funds rate of 0.25 to 0.5 percent. The observations
plotted since then are the midpoint of the range.

U.S. AGRICULTURAL TRADE

AVERAGE LAND VALUES ACROSS THE EIGHTH DISTRICT

90
Exports

75
60
Imports

45
30

Trade Balance

15
0
’12

September

’13

’14

’15

’16

NOTE: Data are aggregated over the past 12 months.

’17

YEAR-OVER-YEAR PERCENT CHANGE

A third major hurricane, Maria, ravaged
Puerto Rico. Because U.S. GDP and
employment data do not include economic
activity from Puerto Rico, this article does
not discuss the potential economic effects
stemming from Maria on the U.S. economy.

’15

I N F L AT I O N - I N D E X E D T R E A S U RY Y I E L D S P R E A D S

PERCENT

1

’14

CPI–All Items

NOTE: Each bar is a one-quarter growth rate (annualized);
the red line is the 10-year growth rate.

BILLIONS OF DOLLARS

ENDNOTE

’13

PERCENT CHANGE FROM A YEAR EARLIER

4

1.00

Kevin L. Kliesen is an economist at the
Federal Reserve Bank of St. Louis. Brian
Levine, a research associate at the Bank,
provided research assistance. See http://
research.stlouisfed.org/econ/kliesen for
more on Kliesen’s work.

CONSUMER PRICE INDEX (CPI)

6

PERCENT

California may be another source of
additional pressures on food price
inflation.
Despite the uptick in food and
energy prices, the personal consumption expenditures price index was up in
October by only 1.6 percent from a year
earlier. Still, the rise in crude oil prices
in October and November suggests
that inflation could drift higher in the
fourth quarter.
Nonetheless, inflation expectations remain stable, perhaps reflecting
the expectation of further tightening
actions by the Federal Open Market
Committee in 2018, which would be
expected to help stanch rising price
pressures. As of late November, the
St. Louis Fed’s inflation forecasting
model continues to see a low probability of headline inflation exceeding
2.5 percent over the next 12 months.

12
10
8
6
4
2
0
–2
–4
–6
–8
–10

Quality Farmland
Ranchland or Pastureland

2016:Q3 2016:Q4 2017:Q1 2017:Q2 2017:Q3
SOURCE: Agricultural Finance Monitor.

On the web version of this issue, 11 more charts are available, with much of those charts’ data specific to the Eighth District.
Among the areas they cover are agriculture, commercial banking, housing permits, income and jobs. To see those charts, go to
www.stlouisfed.org/economyataglance.
The Regional Economist | www.stlouisfed.org 17

I N D U S T R Y

P R O F I L E

Advanced Manufacturing
Is Vital across Nation,
Including Eighth District
By Charles S. Gascon and Andrew Spewak

Manufacturing has been one of the nation’s largest and most productive sectors
dating back to the Industrial Revolution, and that remains true today despite a
long-term decline in employment.1
s technological progress continues to
alter the landscape of the economy, a
subset of manufacturing industries known as
“advanced manufacturing” serves as a critical
source of growth as these products drive productivity gains throughout the economy.
In some sense, all manufacturing is
“advanced” because it requires specific knowledge and use of modern technology. However,
we refer specifically to the advanced manufacturing sector as industries in which research
and development spending exceeds $450 per
worker and at least 21 percent of jobs require
a high degree of technical knowledge.2 These
two metrics quantify the high level of development, design and technical work that is
needed to initially develop advanced products.
Thirty-five manufacturing industries
outlined in the North American Industry
Classification System (NAICS) qualify as
advanced. Among the largest U.S. advanced
manufacturers are companies that produce
electronics, motor vehicles and fuel. The table
displays the largest advanced manufacturing
firms, based on revenue, in the nation and the
Eighth Federal Reserve District.3
A company is classified as a manufacturing
firm if its main business purpose is to produce
goods, regardless of how much it engages in
the actual production of those goods. Consider
Apple: Its purpose is to produce electronics, so
it is a manufacturing firm even though it contracts production to other suppliers and has
many employees developing software. Similarly, there are two types of manufacturing
18 The Regional Economist | Fourth Quarter 2017

employees: production workers, who physically make the goods, and nonproduction
workers, who work in other occupations that
include administrative, professional, technical
and management positions.
We restrict our analysis to advanced
manufacturing for two reasons. First, these
industries are more productive than the rest
of manufacturing. Although they historically have employed only about 45 percent of
manufacturing employees, their output makes
up to 53 percent of manufacturing output.
Second, there exists a wage premium for
advanced manufacturing employees. The average employee in these industries earns about
40 to 50 percent more than the average private
sector worker, depending on the data source.
As of 2016, the wage premium for nonproduction workers compared with private sector
workers is 72 percent, and the premium for
production workers is 7 percent.4 In contrast,
workers in non-advanced manufacturing
sectors earn essentially the same wage as other
private sector workers.
In this article, we will examine advanced
manufacturing’s long-term shifts, its current
state and its impact on the Eighth District
economy.

© THINKSTOCK / ISTOCK

Largest Advanced Manufacturing
Firms by Revenue
National

Eighth District

1

Apple
(3342)

Emerson Electric (335)
(St. Louis, Mo.)

2

Johnson & Johnson
(3254)

MilliporeSigma (3254)
(St. Louis, Mo.)

3

Gilead Sciences
(3254)

Energizer Holdings (3359)
(St. Louis, Mo.)

4

Intel
(3344)

Hillenbrand (3339)
(Batesville, Ind.)

5

Cisco Systems
(3342)

American Railcar Industries (3365)
(St. Charles, Mo.)

6

General Motors
(3361)

Esco Technologies (3345)
(St. Louis, Mo.)

7

General Electric
(335)

FutureFuel (3251)
(Clayton, Mo.)

8

Amgen
(3254)

Kimball Electronics (3344)
(Jasper, Ind.)

9

Pfizer
(3254)

Escalade (3399)
(Evansville, Ind.)

Exxon Mobil
(3241)

Sypris Solutions (3363)
(Louisville, Ky.)

10

SOURCE: Compustat.
NOTE: Firm location is based on the location of the headquarters,
which is self-reported by the corporation. Company NAICS code in
parentheses. All data are from December 2016 unless otherwise
noted. MilliporeSigma data are from Sigma-Aldrich in December
2014; since then, Sigma-Aldrich has been bought out and merged
into MilliporeSigma.

National Advanced Manufacturing

From January 1997 to the end of the Great
Recession in June 2009, advanced manufacturing lost over 2 million employees. The
biggest losses were in computer electronics
manufacturing, which lost 720,000 jobs, and

primary metals manufacturing, which lost
450,000. As a share of private employment,
advanced manufacturing employment fell
from 7.5 percent to 4.9 percent during this
period. During the recovery from June 2009

Regional Employment

Advanced manufacturing is especially
vital to the Eighth District economy: The
sector employs 7 percent of private sector workers and generates 11 percent of
private output.5 As Figure 1 shows, both
the employment share and growth since
the recession exceed the national averages.
Among District states, the employment

FIGURE 1
Advanced Manufacturing Employment

Employment Growth (2009-2017)

40%
35%
30%
25%
20%
15%
10%
5%
0%
–5%
–10%
0%

Kentucky

Tennessee

Louisville

Indiana

Eighth District
Missouri

Memphis
St. Louis

U.S.
Mississippi

Arkansas
2%

Little Rock

4%
6%
Share of Private Employment (2017)

8%

10%

12%

SOURCES: Bureau of Labor Statistics and authors’ calculations.
This figure shows the advanced manufacturing employment share in March 2017 versus the growth of advanced manufacturing employment from the end of the recession in June 2009 until March 2017. Areas to the right of the vertical line have a higher employment share
than the nation. Areas above the horizontal line have experienced faster employment growth than the nation. Areas in the top-right
quadrant are the best-performing, as both the share and growth exceed the national averages.
NOTE 1: Due to nondisclosure at the county level for some industries over time, estimates for the Eighth District advanced manufacturing sector are calculated as the sum of data for the entirety of all District states except Illinois. We excluded Illinois from our calculations
since most of Illinois’ economic activity stems from the Chicago area, outside the District. The other District states are Arkansas, Indiana,
Kentucky, Mississippi, Missouri and Tennessee.
NOTE 2: In calculating employment for each metropolitan statistical area (MSA), we estimated nondisclosed four-digit North American
Industry Classification System (NAICS) industries by projecting the MSA employment data using the employment growth rate of the MSA’s
largest county. If the data were also nondisclosed in the largest county, we used the state growth rate. If the state data were also missing,
we used the growth rate of the corresponding three-digit NAICS industry.

FIGURE 2
Advanced Manufacturing Wages
25%
Real Wage Growth (2009-2017)

to March 2017, advanced manufacturing
employment increased 6 percent, but the
share fell to 4.5 percent. (See Figure 1.)
Despite gains in recent years, employment in advanced manufacturing has fallen
over 30 percent since 1997. Yet, that is not
necessarily an indication of weakness in
the sector. From 1997 to 2015, real output
increased by over 50 percent due to gains in
labor productivity. In 2015, advanced manufacturing was 40 percent more productive
than the private sector as a whole.
Similarly, advanced manufacturing
remains the largest U.S. exporter. In 2016,
advanced manufacturing accounted for
60 percent of the dollar value of exports,
down slightly from 68 percent in 1997,
but up from 2014.
Moreover, wages in advanced manufacturing are high, with the average worker
making over $1,600 per week. Wages are
highest in computers and electronics manufacturing, at $2,300, and chemical manufacturing, at $1,900. Real (inflation-adjusted)
wages have grown 11 percent since the
recession, with the largest gains in computers and electronics manufacturing. Today,
the average advanced manufacturer makes
$1.53 for every $1 that the average private
sector worker makes. (See Figure 2.)
Most advanced manufacturing jobs are in
large metropolitan statistical areas (MSAs).
Employment is highest in Los Angeles,
which has 232,000 employees, followed
by Chicago, with 143,000, and New York,
with 132,000. These three MSAs account
for 9 percent of advanced manufacturing
employment nationwide. While the total
number of employees is smaller, as a share
of private employment, advanced manufacturing is most heavily concentrated in Midwestern MSAs. The share is highest in Battle
Creek, Mich. (the main product being autos),
followed by Wichita, Kan. (airplanes), and
Columbus, Ind. (machinery).

20%

Louisville

Arkansas

15%
10%

Little Rock

Tennessee
Memphis

5%
0%
$1.10

U.S.

Kentucky
Eighth District
Indiana
Missouri

$1.15

$1.20

$1.25

$1.30

$1.35

$1.40

$1.45

Mississippi
St. Louis

$1.50

$1.55

$1.60

Wage Premium

SOURCES: Bureau of Labor Statistics and authors’ calculations.
Analogous to Figure 1, this figure shows the wage premium for advanced manufacturing workers in March 2017 versus real wage growth
from the end of the recession in June 2009 until March 2017. Areas to the right of the vertical line have a higher wage premium than the
nation. Areas above the horizontal line have experienced faster real wage growth than the nation. Areas in the top-right quadrant are the
best-performing, as both the wage premium and growth exceed the national averages. The apparent negative relationship in the figure
is due to the limited number of observations presented. A sample of all 50 states indicates a modest positive correlation between wage
growth and wage premiums.
NOTE 1: The wage premium is calculated as the amount of money the average advanced manufacturing employee earns for every $1 earned
by the average private sector employee.
NOTE 2: Due to nondisclosure at the county level for some industries over time, estimates for the Eighth District advanced manufacturing sector are calculated as the sum of data for the entirety of all District states except Illinois. We excluded Illinois from our calculations
since most of Illinois’ economic activity stems from the Chicago area, outside the District. The other District states are Arkansas, Indiana,
Kentucky, Mississippi, Missouri and Tennessee.
NOTE 3: Due to nondisclosure at the county level for some industries over time, wage estimates are based off the 3-digit NAICS industries
325, 327, 331, 333, 334, 335, 336 and 339.
The Regional Economist | www.stlouisfed.org 19

In the Eighth District, advanced manufacturing has a relatively large presence,
mostly due to a high concentration of
automotive manufacturing employment.
However, the wage premium for advanced
manufacturing employees, while significant, is smaller regionally than nationally.
Likewise, though real wages are growing
positively in the Eighth District, the pace of
growth lags behind the national average.
Auto production accounts for 39 percent of the advanced manufacturing jobs in Eighth District
states. The auto industry’s share of these jobs is highest in Indiana, Kentucky and Tennessee.

share is largest in Indiana, Kentucky and
Tennessee. Among the District’s four largest MSAs (St. Louis, Mo.; Memphis, Tenn.;
Louisville, Ky.; and Little Rock, Ark.), the
employment share is highest in Little Rock.
Since the end of the recession, advanced
manufacturing employment in the District
states has grown 23 percent, outpacing the
national rate considerably. That translates
to 139,000 new jobs in the District states.
Employment growth has been fastest in the
eastern portion of the District: Kentucky,
Tennessee and Indiana are growing substantially more rapidly than the rest of the
District states. Among the MSAs, Louisville
has experienced the fastest employment
growth since 2009, at 29 percent, followed
by Memphis, at 10 percent.
Auto Manufacturing in the District

Auto manufacturing has a significant presence regionally, employing 39 percent of
advanced manufacturing workers, and has
driven the bulk of advanced manufacturing’s
growth. On net, 90 percent of new advanced
manufacturing jobs since 2009 are automotive.
Among District states, auto manufacturing
employment as a share of advanced manufacturing employment is largest in Indiana,
Kentucky and Tennessee. Among the MSAs,
the auto employment share is largest in
Louisville, at 37 percent, and Memphis, at
15 percent. Recall from Figure 1 that these
areas also experienced the fastest growth in
advanced manufacturing employment.
The Regional Impact

District productivity in the sector mirrors
the nation. Advanced manufacturing in
2015 was 36 percent more productive than
the overall private sector, with the most
20 The Regional Economist | Fourth Quarter 2017

© THINKSTOCK / ISTOCK /ZAPP2PHOTO

productive subsector being transportation equipment manufacturing. Advanced
manufactures are a larger component of
trade for the District than nationally. They
make up 70 percent of the dollar value of
District state exports to the world, above the
1997 share of 64 percent.
Average weekly advanced manufacturing wages in the District are generally below
the U.S. average. However, nominal wages
are lower throughout the private sector in
the Eighth District, mostly because of the
District’s lower cost of living.6 Figure 2 shows
that the District’s wage premium, which
accounts for differences in cost of living,
also tends to fall below the U.S. average. This
result is largely due to the fact that nonproduction workers, who garner higher wages
than production workers, constitute a smaller
proportion of the sector’s workforce in the
District compared to the nation. Of the District MSAs and states, only Mississippi’s wage
premium of 54 percent exceeds the national
average. Among the four MSAs, the premium
is highest in St. Louis, at 51 percent.
Likewise, real wage growth in the District, while positive, is slow. Of the states,
only Arkansas real wages are growing more
quickly than the national average. Among
the MSAs, real wages are growing fastest in
Louisville and Little Rock, at 20 percent and
9 percent, respectively.
Sector Still Significant

Advanced manufacturing employment as
a share of private employment has steadily
declined over the years, but the sector
remains a significant cog in the U.S. economy. Advanced manufacturing accounts
for 7 percent of private output and 60 percent of the dollar value of U.S. exports.

Charles Gascon is a regional economist, and
Andrew Spewak is a senior research associate,
both at the Federal Reserve Bank of St. Louis.
For more on Gascon’s work, see https://research.
stlouisfed.org/econ/gascon.

ENDNOTES
1
2
3

4

5

6

See Kliesen and Tatom.
See Muro et al.
Headquartered in St. Louis, the Eighth Federal
Reserve District includes all of Arkansas and parts
of Illinois, Indiana, Kentucky, Mississippi, Missouri and Tennessee. In our analysis we exclude
Illinois; see endnote 5 for more information.
The Quarterly Census of Employment and Wages
(QCEW) and Occupational Employment Statistics
(OES), both from the Bureau of Labor Statistics,
report industry-level wages. The advanced
manufacturing wage premium is estimated to be
53 percent (QCEW) and 42 percent (OES). The
OES provides estimates for both nonproduction
and production occupations. Throughout the rest
of the article, we will use QCEW data, as they are
better suited for time series and regional analysis.
When available, we have tested the robustness of
our results using the OES data.
Due to nondisclosure at the county level for some
industries over time, estimates for the District’s
advanced manufacturing sector are calculated as
the sum of data for the entirety of all District states
except Illinois. We excluded Illinois from our
calculations since most of that state’s economic
activity stems from the Chicago area, which is
outside the District.
See Coughlin, Gascon and Kliesen for more information on the relationship between cost of living
and income.

REFERENCES
Coughlin, Cletus; Gascon, Charles; and Kliesen,
Kevin. Living Standards in St. Louis and the
Eighth Federal Reserve District: Let’s Get Real.
Federal Reserve Bank of St. Louis Review, Fourth
Quarter 2017, Vol. 99, No. 4, pp. 377-94.
Kliesen, Kevin; and Tatom, John. U.S. Manufacturing and the Importance of International Trade:
It’s Not What You Think. Federal Reserve Bank
of St. Louis Review, January/February 2013,
Vol. 95, No. 1, pp. 27-49.
Muro, Mark; Rothwell, Jonathan; Andes, Scott;
Fikri, Kenan; and Kulkarni, Siddharth. America’s
Advanced Industries: What They Are, Where They
Are, and Why They Matter. The Brookings Institution, February 2015.

O V E R V I E W

First-Time Homebuyers Appear
to Be Younger, Less Creditworthy
in Eighth District

The Eighth Federal Reserve District
is composed of four zones, each of
which is centered around one of
the four main cities: Little Rock,
Louisville, Memphis and St. Louis.

By Brian Reinbold and Paulina Restrepo-Echavarria

FIGURE 1
Total Number of First-Time Homebuyers

States in Eighth District

F

irst-time homebuyers are essential to
the dynamics of the housing market
by allowing current homeowners to trade
up. The number of first-time homebuyers
decreased between 2000 and 2011, and
then started slowly increasing again. (See
Figure 1.) There are many possible reasons
why this happened, such as rising rent
and home prices, rising student debt and
tightening credit standards.
Have there been fewer first-time
homebuyers in the Eighth Federal Reserve
District? In this article, we study the
number and some characteristics of firsttime homebuyers in the Eighth District1
and see how they compare to those at the
national level.
We used the Federal Reserve Bank of
New York Consumer Credit Panel/Equifax to estimate the number of first-time
homebuyers. The FRBNY Consumer
Credit Panel (CCP) consists of detailed
credit-report data, updated quarterly, for
a unique longitudinal panel of individuals and households beginning in 1999. It
provides information on various forms of
debt, including student loans, auto loans
and mortgages.
The CCP is a nationally representative
5 percent random sample of individuals
in the United States with a Social Security
number and credit report.2
We took a 10 percent random sample of
the CCP dataset, so we have a 0.5 percent
nationally representative sample. In other
words, we have 1,347,520 unique records for
this sample in 2016, out of approximately
269,504,000 individuals in the U.S. with a
Social Security number and a credit report.
A drawback of the CCP dataset is that
it only includes homebuyers who finance

400,000

9,000,000

350,000

8,000,000

300,000

7,000,000
6,000,000

250,000

5,000,000

200,000

4,000,000

150,000

3,000,000

100,000

2,000,000

50,000

1,000,000

0

USA

D I S T R I C T

2000
AR

2002
IL

2004

2006
IN

2008
KY

2010
MO

2012

2014

MS

TN

2016

0

USA

SOURCES: Federal Reserve Bank of New York Consumer Credit Panel/Equifax and authors’ calculation.
NOTE: Some parts of these states lie outside of the Eighth District.

their purchase with a mortgage; it excludes
all cash purchases. However, it is likely
that most first-time homebuyers finance
their home.
Following work by Jessica Dill and Elora
Raymond, we used the CCP to estimate
the number of first-time homebuyers.3 We
took the year of the oldest mortgage on
file for individuals within the dataset to
determine the first time they obtained a
mortgage. This analysis does not consider
individuals who transitioned back to renting and then purchased a home later on.4
Figure 1 is a plot of the total number of
first-time homebuyers from 2000 to 2016
by each state in the Eighth District5 and the
whole U.S. The number of first-time homebuyers decreased significantly since 2000. The
decline bottomed out around 2011 and 2012
for the U.S. and most states in the District.
From 2000 to 2011, the rate of decline
for these District states is similar to the 76
percent decline nationwide. Indiana and

Illinois experienced the sharpest decline
during this period, each falling about 80
percent, while Arkansas had the smallest
decline, dropping about 70 percent.
The number of first-time homebuyers bottomed out in 2011 for the nation and most
states in the Eighth District. Since 2011, the
number of first-time buyers nationally has
increased 34 percent. The growth rates since
2011 for Missouri and Tennessee exceeded
the nation’s at 46 and 54 percent, respectively. The rates in Arkansas, Illinois and
Indiana are in line with the nation’s. However, the rate remains flat in Kentucky, while
the rate in Mississippi has actually declined
22 percent since 2011.
Figure 2 plots the median credit score
of first-time homebuyers at the date of
purchase. As we can see, credit worthiness
appears to be of lesser importance in the
states of our District throughout the whole
period of 2000 to 2016; the combined credit
score is lower than the national level.
The Regional Economist | www.stlouisfed.org 21

FIGURE 2
Median Risk Score of First-Time Homebuyers at Date of Purchase

ENDNOTES

730

1

States in Eighth District

720
710

2
3

700

4

States in Eighth District

690
USA
680
670

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

SOURCES: Federal Reserve Bank of New York Consumer Credit Panel/Equifax and authors’ calculation.
NOTE: Some parts of these states lie outside of the Eighth District.
5

FIGURE 3

REFERENCES

Total Number of First-Time Homebuyer by Birth Year
35,000

250,000
200,000

25,000
20,000

150,000

15,000

100,000

10,000

USA

States in Eighth District

30,000

50,000

5,000
0

2000

2002
1985 District

2004

2006
1975 District

2008

2010

1985 USA

2012

2014

2016

0

1975 USA

SOURCES: Federal Reserve Bank of New York Consumer Credit Panel/Equifax and authors’ calculation.
NOTE: Some parts of these states lie outside of the Eighth District.

Qualitatively, however, the District and
national trends behave the same. Credit
requirements eased from 2003 to 2006,
corresponding to the time of the housing
bubble. When the housing bubble burst,
credit significantly tightened as lenders
increased credit worthiness requirements.
As we mentioned, credit scores in the District follow a similar trajectory but began
to increase a year earlier than the national
trend. After the sharp increase, the District
and national trends flattened out in 2009.
Although increasing over the last several
years, the number of first-time homebuyers
is still much lower than the pre-2007 level,
suggesting that tightened lending standards have been a headwind for first-time
homebuyers.
Did this decline affect age groups differently? Figure 3 shows the total number of
22 The Regional Economist | Fourth Quarter 2017

Headquartered in St. Louis, the Eighth District
includes all of Arkansas and parts of Illinois,
Indiana, Kentucky, Mississippi, Missouri and
Tennessee.
See van der Klaauw and Lee.
See Dill.
According to the U.S. Department of Housing
and Urban Development, a first-time home
buyer is “an individual who has no ownership in a principal residence during the 3-year
period ending on the date of purchase of the
property.” Therefore, an individual who buys
their first home, then becomes a renter and
finally purchases a home three or more years
later would be considered a first-time home
buyer again. See HUD.
The sample includes individuals from the entire
state, not just those from the parts of the state
that belong to the District.

first-time homebuyers who were born in
1975 and 1985 for the U.S. and the District,
from 2000 to 2016. The number of first-time
homebuyers for those born in 1975 peaked
in the early 2000s, when they were in their
late 20s, while the number of first-time
homebuyers for those born in 1985 has
remained more constant since 2010.
For those born in 1975, the total number of
first-time homebuyers fell precipitously after
age 30, while the number for those born in
1985 remained fairly constant after 2010. These
results suggest that demand by first-time buyers
is more spread out for later generations.
From these data, we can conclude that
the number of first-time homebuyers in
the District states has a trend which is very
similar to the national level and that credit
requirements are somewhat looser in the
District.

Dill, Jessica; and Raymond, Elora. Are Millennials Responsible for the Decline in First-Time
Home Purchases? Federal Reserve Bank
of Atlanta: Real Estate Blog, May 20, 2015.
See http://realestateresearch.frbatlanta.org/
rer/2015/05/are-millennials-responsible-forthe-decline-in-first-time-home-purchases.
html.
Dill, Jessica; and Raymond, Elora. Are Millennials Responsible for the Decline in First-Time
Home Purchases? Part 2. Federal Reserve
Bank of Atlanta: Real Estate Blog, July 1, 2015.
See http://realestateresearch.frbatlanta.org/
rer/2015/07/are-millennials-responsible-forthe-decline-in-first-time-home-purchasespart-2.html.
U.S. Department of Housing and Urban Development. HUD HOC Reference Guide, Nov. 7,
2012. See https://archives.hud.gov/offices/hsg/
sfh/ref/sfhp3-02.cfm.
van der Klaauw, Wilbert; and Lee, Donghoon.
An Introduction to the FRBNY Consumer
Credit Panel. Federal Reserve Bank of New
York, Staff Report No. 479, November 2010.

Paulina Restrepo-Echavarria is an economist,
and Brian Reinbold is a research associate,
both at the Federal Reserve Bank of St. Louis.
For more on Restrepo-Echavarria’s work, see
https://research.stlouisfed.org/econ/restrepoechavarria.

R E A D E R
A S K

A N

E X C H A N G E
E C O N O M I S T

ASK AN ECONOMIST

LETTERS TO THE EDITOR
Don Schlagenhauf has been an economist at
the Federal Reserve Bank of St. Louis since
2015. His research interest is in macroeconomics
and policy, with emphasis on housing.
He enjoys baseball. Don was born in St. Louis
and has been a lifelong Cardinals fan. In fact,
he is a season ticket holder for the Cards spring
training. For more on his research, see https://
research.stlouisfed.org/econ/schlagenhauf.

These letters pertain to articles in our Third Quarter issue (stlouisfed.org/
publications/regional-economist/third-quarter-2017). The first letter is about
the article Quantitative Easing: How Well Does This Tool Work?
Dear Editor:
I agree with you on the point that QE should not be repeatedly used in
the future as a monetary policy because (1) purchasing private bonds is too
influential to the firms’ financial health, which may result in economic biases;
and (2) public sentiment can no longer be more optimistic than it was from

Q: How did consumer borrowing change after
the Great Recession?

2008. On the other hand, I believe that people’s faith in QE positively worked
at least in the past.
In the analyses with Japan and Canada, you did not mention exchange
rates. However, both Japanese yen and Canadian dollars significantly

A: Following the run-up in household debt during the early 2000s,

changed during the past decade. I also studied international economics

consumers have been steadily reducing their overall debt level

and learned that Canadian transports with the U.S. remarkably increased

(i.e., deleveraging) since the Great Recession ended in June 2009.

after US-Can FTA (1989), and its economy became more reliant on the U.S.

The ratio of household debt to personal income peaked in the mid-

economy. Likewise, Japanese trade volumes and its stock prices are reacting

2000s at nearly 1.2, and it has declined to about 0.9 in the second

in accordance to JPY-USD exchange rates.

quarter of 2017.
However, looking at aggregate data tells us only part of the story.
To better understand the run-up in debt and subsequent deleveraging, Carlos Garriga, Bryan Noeth and I studied patterns in mortgage

Therefore, the fact that Canadian real GDP boosted without QE is
explained by 1) its reliance on US economy, and 2) large fluctuations in
exchange rates.
By the way, nominal GDP in U.S. dollars shows completely different trends.

debt, credit card debt, auto loans and student loans held by differ-

The growths from 2008 to 2015 are: Canada 0.24 percent, Japan –13.06 percent

ent age groups between 1999 and 2013.1

and U.S. 23.11 percent.

Obviously, the biggest change in borrowing over that period has
been mortgage debt. In the early 2000s, average mortgage debt

Emi Igawa, Nagoya, Japan

increased among all age groups, but especially for younger households. In 1999, homeowners with the largest mortgage debt (about
$60,000 in 2013 dollars) were around 45 years old. In 2008, peak
mortgage debt (about $117,000) occurred for those around age 42.
Despite large deleveraging after the recession (particularly among
those younger than 60), average mortgage debt remained higher in
2013 than in 1999.
For the other types of debt, the general patterns we found were:
• Credit card debt also increased, primarily for those older than
30, and then began to decline after 2008. Unlike other types of
debt, average credit card debt in 2013 was below its 1999 level
for most age groups.
• Auto debt also rose between 1999 and 2008, but dropped across all
age groups after the recession. Auto debt then rebounded in 2013.
• Student debt, on the other hand, consistently grew from 2005 to
2013 for all age groups. For those over 50, the rise is likely due to
parents or grandparents taking on loans or co-signing for relatives.
Having debt is not necessarily bad, as it allows individuals to
make up for the mismatch between income and consumption

The second letter comments on the article titled Household Participation
in Stock Market Varies Widely by State.
Dear Editor:
I think the methodology in this analysis is very flawed, and a wide variety
of conclusions could, therefore, be drawn.
Our household falls within the key household income group discussed.
We do all of our savings within tax-deferred retirement vehicles and have
substantial savings, with about 75 percent in equities. We never report
dividends because we own no equities outside the tax-deferred accounts;
so, we are a reason that they report low participation in the stock market.
So an alternative explanation of the data shown in this paper is that the
people in the states with high stock market participation rates are investing
in tax-inefficient vehicles and could benefit from financial advice to put more
or all of their savings into tax-deferred plans. Between Roth and Regular IRAs,
401(k)s, and 403(b)s, there is no reason for anybody making less than $200k
per year to have ANY taxable stock dividends.
We may have a retirement crisis, but it is not because people are not

expenditures earlier in life; consumers just need to be prudent with

buying stocks outside of tax-deferred accounts.

the amount of debt they take on. By studying debt patterns, how-

Raymond D’Hollander, Fayetteville, N.Y., an engineer

ever, we hope to gain a better understanding of the tipping point
between manageable debt and debt levels that expose consumers
to excessive risk.

1

Garriga, Carlos; Noeth, Bryan; and Schlagenhauf, Don E. Household Debt and the
Great Recession. Federal Reserve Bank of St. Louis Review, Second Quarter 2017,
Vol. 99, No. 2, pp. 183-205.

We welcome letters to the editor, as well as questions for “Ask an
Economist.” You can submit them online at www.stlouisfed.org/re/
letter or mail them to Subhayu Bandyopadhyay, editor, The Regional
Economist, Federal Reserve Bank of St. Louis, P.O. Box 442, St. Louis,
MO 63166-0442.

The Regional Economist | www.stlouisfed.org 23

P.O. Box 442
St. Louis, MO 63166-0442
CHANGE SERVICE REQUESTED

The Regional Economist: 100 Issues and Counting

I

n January 1993, the first issue of the Regional Economist
debuted. The three articles focused on health insurance,
the business cycle and exports from our District. The editor
then was James Bullard, now president and CEO of the
St. Louis Fed.
The issue you are reading is the 100th of this quarterly
publication. Much has changed over the past 25 years—the
magazine is bigger, readership has mushroomed (thanks
largely to our web presence), the topics span from the local
to the international, and all articles are now written by our
economists (but still in layman’s language).
To prepare RE for its next 100 issues, we’re introducing
some changes in the coming year:
• Articles will be published online (www.stlouisfed.org/re)
as they are finished—one every two weeks or so.
This will ensure that they don’t become outdated while
waiting for the next quarter’s release. We think online

readers will also appreciate the one-article-at-a-time
approach. (Print subscribers will continue to receive this
magazine—with all of the articles—in their mailbox four
times a year.)
• The online version of RE will be redesigned to reflect
our new approach of continuous publishing. Check it
once in a while to see what’s new. (Readers who already
subscribe to receive an email when a new issue is published will receive in the future an email when each new
article is posted. Sign up for this email newsletter at
www.stlouisfed.org/subscribe/re.)
• The print version of RE will also be redesigned—the first
new look since 2008.
We hope you like the changes.
Subhayu Bandyopadhyay,
Editor
The Regional

Economist

F E D E R
A L

A Quarterly Review
of Business and
Economic Conditions

Income Inequality
It’s Not So Bad
in the United States

Exports to China

District Tops Nation
in Growth of Shipments

A Quarterly Review
of Business and
Economic Conditions

Vol. 16, No. 4
October 2008

Vol. 25, No. 2

THE FEDERAL
RESERVE BANK
OF ST. LOUIS

CENTRAL TO AMERICA’S

Second Quarter

2017

President Bullard

Let’s Start Trimming
Fed’s Balance Sheet

ECONOMY ®

Industry Profile

Growth in Tech
Sector
Returns to Glory
Days

APRIL 2003
R E S E R V
E

B A N K

O F

S A I N T

L O U I S

The Regional

Economist

A Quarterly Review
of Business and
Economic Conditions

A Winning
Combination?

Vol. 17, No. 1

Economics and Sports

El Dorado Promise
Free College Education
Rejuvenates Arkansas Town

January 2009

The Federal reserve Bank oF sT. louis

A Quarterly
Review of
Business and
Economic
Conditions

Split Decisions

How Marriage and
Motherhood
Affect Women’s Wages

CONSUMER CONFIDENCE

What Do They Tell

SURVEYS

Us?

Nation

COMMUNITY PROFILE

Starkville, Miss., a
Stark
Contrast to State’s
Image
WWW.STLOUISFED.ORG

China’s Econo
mic Data
An Accu

Deficits, Debt

INNOVATION

The District vs. the

Community
Colleges

and Looming Disaster
Reform of Entitlement Programs
May Be the Only Hope

rate Reflectio
n,
or Just Smoke
and Mirrors?

ECONOMY

AT

A

THE REGIONAL

GLANCE

ECONOMIST
FOURTH QUARTER

All data as of Dec. 1, 2017.
REAL GDP GROWTH

4

2

0
Q3
’12

’13

’14

’15

’16

PERCENT CHANGE FROM A YEAR EARLIER

4
PERCENT

VOL. 25, NO. 4

CONSUMER PRICE INDEX

6

–2

|

’17

CPI–All Items
All Items, Less Food and Energy

2

0

–2

October

’12

’13

’14

’15

’16

’17

NOTE: Each bar is a one-quarter growth rate (annualized);
the red line is the 10-year growth rate.

I N F L AT I O N - I N D E X E D T R E A S U RY Y I E L D S P R E A D S
2.75

1.70

5-Year

2.50

1.60

10-Year

2.25
PERCENT

RATES ON FEDERAL FUNDS FUTURES ON SELECTED DATES

1.50
1.30

1.75

1.20

1.50

1.10

10/20/17
11/01/17

07/26/17

1.40

20-Year

2.00

05/03/17
06/14/17

1.00

1.25

0.90

Nov. 24

1.00

’13

’14

’15

’16

0.80

’17

NOTE: Weekly data.

C I V I L I A N U N E M P L O Y M E N T R AT E

1st-Expiring
Contract

3-Month

12-Month

I N T E R E S T R AT E S

10

4

9

10-Year Treasury

3

8
7

PERCENT

PERCENT

6-Month

CONTRACT SETTLEMENT MONTH

6
5

2
Fed Funds Target

1

1-Year Treasury

4
3
’12

October

’13

’14

’15

’16

0

’17

October

’13

’14

’15

’16

’17

NOTE: On Dec. 16, 2015, the FOMC set a target range for the
federal funds rate of 0.25 to 0.5 percent. The observations
plotted since then are the midpoint of the range.

U . S . A G R I C U LT U R A L T R A D E

AVERAGE LAND VALUES ACROSS THE EIGHTH DISTRICT

90

BILLIONS OF DOLLARS

75
60
Imports

45
30

Trade Balance

15
0
’12

September

’13

’14

’15

’16

NOTE: Data are aggregated over the past 12 months.

’17

YEAR-OVER-YEAR PERCENT CHANGE

Exports

12
10
8
6
4
2
0
–2
–4
–6
–8
–10

Quality Farmland
Ranchland or Pastureland

2016:Q3 2016:Q4 2017:Q1 2017:Q2 2017:Q3
SOURCE: Agricultural Finance Monitor.

U.S. CROP AND LIVESTOCK PRICES
140

INDEX 1990-92=100

120

Crops
Livestock

100
80
60
40
’02

September

’03

’04

’05

’06

’07

’08

’09

’10

’11

’12

’13

’14

’15

’16

’17

COMMERCIAL BANK PERFORMANCE RATIOS
U.S. BANKS BY ASSET SIZE / THIRD QUARTER 2017
All

$100 million­$300 million

Less than
$300 million

$300 million$1 billion

Less than
$1 billion

$1 billion$15 billion

Less than
$15 billion

More than
$15 billion

Return on Average Assets*

1.08

1.10

1.07

1.13

1.10

1.17

1.15

1.07

Net Interest Margin*

3.15

3.87

3.86

3.85

3.85

3.77

3.80

3.01

Nonperforming Loan Ratio

1.17

1.02

1.06

0.89

0.96

0.82

0.87

1.25

Loan Loss Reserve Ratio

1.27

1.36

1.38

1.30

1.33

1.10

1.18

1.29

R E T U R N O N AV E R A G E A S S E T S *

NET INTEREST MARGIN*
1.17
1.12
1.33
1.33

1.10

.40

.60

Third Quarter 2017

.80

3.60
3.68

Indiana
Kentucky

3.90
3.77

1.10

Mississippi

3.83
3.85

1.12
1.08

Missouri

1.15
1.07

Tennessee

1.00

.20

3.56
3.47

Illinois

1.15
1.14

.00

4.11
4.14

Arkansas

0.98
1.00
0.97

3.73
3.70

Eighth District

1.00

1.20

1.40

PERCENT

3.48
3.46
3.40
3.33

0.0 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50

Third Quarter 2016

Third Quarter 2017

N O N P E R F O R M I N G L O A N R AT I O
0.79

L O A N L O S S R E S E RV E R AT I O

1.09

.20

.40

Third Quarter 2017

.60

.80

1.18
1.17

Kentucky
0.95

Mississippi

1.06
1.25
1.31

Missouri

0.81
0.84

.00

0.71
0.74

Indiana

1.22

0.82

0.69

1.11
1.16

Illinois

0.81

0.60

1.08
1.11

Arkansas

1.02
1.03
1.09

0.70

1.09
1.14

Eighth District

0.94

0.83

Third Quarter 2016

1.00

1.02
1.09

Tennessee

0.99

1.20

Third Quarter 2016

NOTE: Data include only that portion of the state within Eighth
District boundaries.
SOURCE: Federal Financial Institutions Examination Council Reports
of Condition and Income for all Insured U.S. Commercial Banks.
* Annualized data.

1.40

PERCENT

.00

.20

.40

Third Quarter 2017

.60

.80

1.00

1.20

Third Quarter 2016

For additional banking and regional data, visit our website at:
https://fred.stlouisfed.org.

1.40

REGIONAL ECONOMIC INDICATORS
N O N FA R M E M P L O Y M E N T G R O W T H / T H I R D Q U A RT E R 2 0 1 7
YEAR-OVER-YEAR PERCENT CHANGE
United
States

Eighth
District †

Arkansas

2.1%

Total Nonagricultural

1.4%

1.2%

Natural Resources/Mining

8.4

0.5

Construction

2.6

Manufacturing
Trade/Transportation/Utilities

Illinois

Indiana

Kentucky

Mississippi

Tennessee

0.8%

1.8%

1.6%

0.4%

1.5%

–1.6

3.7

1.6

–5.3

2.5

6.5

8.3

1.1

3.2

–1.0

4.8

6.6

–1.7

–2.4

NA

0.8

0.7

2.6

–0.2

1.5

–0.3

0.4

1.6

0.3

0.4

0.5

1.2

–0.6

–0.3

1.5

1.4

0.8

1.6

–2.1

–1.4

–2.5

0.7

–6.6

5.5

–6.0

–4.9

–0.2

Financial Activities

1.9

2.2

0.1

2.6

3.0

2.1

0.4

2.7

1.5

Professional & Business Services

2.9

2.3

3.9

1.8

1.8

5.2

–1.5

3.5

2.0

Educational & Health Services

2.2

1.5

3.8

0.8

2.6

0.8

2.6

1.7

0.9

Leisure & Hospitality

1.8

1.8

3.8

0.9

0.2

0.4

1.3

3.8

3.3

Other Services

1.3

1.3

3.8

0.3

2.7

3.6

–0.3

0.5

1.4

Government

0.3

0.7

–0.6

–0.7

2.6

–0.3

1.3

2.1

1.4

Information

1.5%

Missouri

† Eighth District growth rates are calculated from the sums of the seven states. For the Construction category, data on Tennessee are no longer available.
Each state’s data are for the entire state even though parts of six of the states are not within the District’s borders.

U N E M P L O Y M E N T R AT E S

EIGHTH DISTRICT ADJUSTED GROSS CASINO REVENUE*
II/2017

III/2016

United States

4.3%

4.4%

4.9%

Arkansas

3.5

3.4

4.0

Illinois

4.9

4.7

5.8

Indiana

3.5

3.3

4.4

Kentucky

5.4

5.1

5.0

Mississippi

5.2

5.0

5.8

Missouri

3.8

3.9

4.8

Tennessee

3.2

4.1

4.8

MILLIONS OF DOLLARS

III/2017

800
750
700
650
600
550
500
450

Mississippi

Indiana

Illinois

Missouri

400
350
300

2009:Q1 2010:Q1 2011:Q1 2012:Q1 2013:Q1 2014:Q1 2015:Q1 2016:Q1 2017:Q1

* NOTES: Adjusted gross revenue = Total wagers minus player winnings.
Native American casino revenue is not included. In 2003 dollars.
SOURCE: State gaming commissions.

HOUSING PERMITS / THIRD QUARTER

R E A L P E R S O N A L I N C O M E / S E C O N D Q U A RT E R

YEAR-OVER-YEAR PERCENT CHANGE IN YEAR-TO-DATE LEVELS

YEAR-OVER-YEAR PERCENT CHANGE

6.0

1.3

7.7

14.8

7.2

33.1
20.2

2017

10

0.8
1.1
0.2

Missouri

0.9
1.9

Tennessee

13.2

15

2.4
0.5
0.6

Mississippi

19.1

5

1.8

Kentucky

–9.8

0

1.1

Indiana

8.6

6.2

0.2

Illinois

12.2

0.3

1.5
1.5

Arkansas

15.2
15.5

–15 –10 –5

1.3
1.4

United States

20

25

30

2016

All data are seasonally adjusted unless otherwise noted.

35

40

PERCENT

2.6

0

.50
2017

1.0

1.5

2.0

2.5

2016

NOTE: Real personal income is personal income divided by the PCE
chained price index.

3.0