Full text of Economic Trends (Federal Reserve Bank of Cleveland) : Economic Trends - September 2003

View original document

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

1
•

•

The Economy in Perspective

FRB Cleveland • September 2003

Man was born to be rich, or, inevitably grows rich
by the use of his faculties; by the union of thought
with nature. Property is an intellectual production.
…An infinite number of shrewd men, in infinite
years, have arrived at certain best and shortest
ways of doing, and this accumulated skill in arts,
cultures, harvestings, curings, manufactures, navigations, exchanges, constitutes the worth of our
world to-day.
—Ralph Waldo Emerson,
“Wealth,” in The Conduct of Life (1860)
Work in progress…Work in America is under
scrutiny once again. Despite nearly two years of
steady gains in U.S. production and income, total
employment is still below the peak of the previous
business cycle in March 2001. Manufacturing jobs
seem particularly vulnerable, having declined by
roughly 15 percent since the recession began. Workers have reportedly become less loyal to their
employers because of downsizing and well-publicized
corporate accounting scandals. Even retirees, who
depend on former employers for pension income—
and often for medical benefits—increasingly worry
about their retirement security.
Unquestionably, work life in America is changing;
but then again, it has been changing for quite some
time. Over the past several decades, we saw men’s
labor force participation rate decrease and women’s
increase. We witnessed a rise in the number of selfemployed and part-time workers, as well as employers’ greater reliance on employment service companies to supply them with temporary help. We also saw
declines in the percentages of people belonging to
unions, working in the manufacturing sector, and
staying with only one or two employers for most of
their working lives. The value of learning increased,
with the most educated people enjoying the highest
incomes and the lowest unemployment rates.
Despite all of these changes, most Americans’
real incomes rose during the past several decades,
and each generation attained a higher standard of
living than the one before. Employment rose dramatically during the last business cycle expansion—
after a “jobless recovery”—and the unemployment
rate fell so low that analysts feared a new round of
wage hikes and inflation. Was the exceptional labor
market performance of the past decade just a
stroke of luck, or was it the result of the trends
mentioned earlier? Put another way, have the many
changes in U.S. labor markets helped American
workers or harmed them? What further developments lie ahead, and how are they likely to affect
our economic well-being?

These are “big picture” questions to be sure, and
deserve more serious commentary and debate than
this short essay can provide. Nevertheless, as we
gear up for an election year that is likely to focus
attention on manufacturing employment and global
trade’s impact on U.S. workers, some big picture
thinking might provide a useful frame of reference.
The fact is that since its inception, this nation has
been a work in progress, an experiment built on the
premise that free people can govern themselves. As
this experiment has lumbered forward through time,
Americans have seen the benefits of thrift, education,
and trade among nations. They have prospered
because they have been free to challenge the status
quo, to innovate, to acquire property, and to keep the
benefits of their labor. Whenever prospects seemed
brighter in another calling or another place, people
pulled up stakes and moved on. Adaptation has been
the soul of the pioneer spirit that still lives on.
American society has changed profoundly during
the nation’s relatively short lifetime, but except for
the most wrenching events, such as the Civil War
and the Great Depression, the transformation has
been gradual and driven forward by commerce.
Much is made today of the notion that the United
States is a nation of consumers, but historically
speaking, we are more accurately described as a
nation of business and a nation with confidence in
the future. This confidence—persisting despite the
recognition that the future demands certain breaks
with the past—is what has encouraged Americans
to embrace change.
Ralph Waldo Emerson witnessed this country’s
transformation from 13 colonies to an immense
landmass stretching from the Atlantic to the Pacific.
He saw a nation rise on the bounty of nature, but
soar on its willingness to allow the established
order to retreat and make room for the future. His
essay on wealth asks,
…how did our factories get built? how did North
America get netted with iron rails, except by the
importunity of …orators, who dragged all the prudent men in? Is [theirs] the madness of many for
the gain of a few? This speculative genius is the
madness of few for the gain of the world.…Wealth
brings with it its own checks and balances. The
basis of political economy is non-interference. The
only safe rule is found in the self-adjusting meter
of demand and supply. Do not legislate. Meddle,
and you snap the sinews with your sumptuary
laws. Give no bounties: make equal laws: secure
life and property, and you need not give alms.
Open the doors of opportunity to talent and
virtue, and they will do themselves justice…

2
•

•

Inflation and Prices
12-month percent change
4.00 CPI AND CPI EXCLUDING FOOD AND ENERGY

August Price Statistics

3.75

Percent change, last:
2002
a
a
a
1 mo. 3 mo. 12 mo. 5 yr. avg.

3.50
CPI

3.25

Consumer prices

3.00

All items

2.0

1.3

2.1

2.4

Less food
and energy

2.5

1.9

1.5

2.2

2.0

2.50

Medianb

2.7

2.0

2.1

2.9

3.0

2.25

2.75

Producer prices

2.00

Finished goods

1.7

1.4

3.0

1.8

1.2

Less food
and energy

2.4

0.8

0.2

0.9 –0.5

1.75
1.50
CPI excluding food and energy

1.25
1.00
1995

1996

1997

1998

1999

2000

2001

2002

2003

Millions of barrels
240 U.S. GASOLINE INVENTORIES d

Billions of cubic feet
4,000 U.S. NATURAL GAS INVENTORIES c,d
3,500

230
3,000
220
2,500
210

2,000

1,500
200
1,000
190
500
180

0
2001

2002

2003

2001

2002

2003

FRB Cleveland • September 2003

a. Annualized.
b. Calculated by the Federal Reserve Bank of Cleveland.
c. Working gas in underground storage.
d. The shaded band indicates the range between minimum and maximum values for weekly data from 1998 through 2002.
SOURCES: U.S. Department of Labor, Bureau of Labor Statistics; U.S. Department of Energy, Energy Information Administration; and Federal Reserve Bank
of Cleveland.

Consumer and wholesale prices rose
at an annualized rate of about 2% in
July, partly restrained by falling energy
costs. Excluding food and energy
goods, the Consumer Price Index
(CPI) and the Producer Price Index
1
(PPI) each rose about 2 /2%, both well
above their 12-month growth trends.
Despite their restraining influence
on the major price statistics in July,
energy costs have given a strong
upward push to the aggregate price
measures over the past year. And several sources indicate that energy costs
may continue to exert pressure on

consumers’ pocketbooks and businesses’ income statements in the immediate future. Certainly, spot prices
for a wide range of energy goods rose
sharply this summer, but over the past
month or so, upward pressures have
shown signs of relenting.
In June, Federal Reserve Chairman
Greenspan testified before the House
Committee on Energy and Commerce
that “[t]oday’s tight natural gas markets have been a long time in coming,
and futures prices suggest that we are
not apt to return to earlier periods of
relative abundance and low prices

anytime soon.” He noted that unlike
the market for crude oil, where “American refiners have unlimited access to
world supplies” and can therefore adjust readily to any imbalance between
domestic consumption and domestic
supply, it is more difficult to meet domestic demand for natural gas by
means of imports. Indeed, at the
time of Chairman Greenspan’s testimony, natural gas prices had risen
from a low of $2.55 (per million Btu) in
July 2000, to $3.65 in July 2002, and to
$6.31 in July 2003.
(continued on next page)

3
•

•

Inflation and Prices (cont.)
12-month percent change
5 CORE CPI GOOD AND SERVICES

12-month percent change
4.25 CPI AND TRIMMED-MEAN MEASURES
4.00

3.75

CPI core services

CPI
3.50
3

3.25

Median CPI a

3.00
2
2.75
CPI core goods
2.50

2.25
2.00

1.75
1.50

–1

CPI, 16% trimmed mean a

1.25

–2

1.00
1995

1996

1997

1998

1999

2000

2001

2002

1995

2003

1996

1997

1998

1999

2000

2001

2002

2003

12-month percent change
5.0 YEAR-AHEAD HOUSEHOLD INFLATION EXPECTATIONS c

Four-quarter percent change
7
PRODUCTIVITY AND COSTS b
6

4.5

5
4.0

4
Output per hour

3.5

3
2

3.0

2.5

0
2.0

–1
Unit labor costs

1.5

–2
–3
1995

1996

1997

1998

1999

2000

2001

2002

2003

1.0
1995

1996

1997

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

a. Calculated by the Federal Reserve Bank of Cleveland.
b. Nonfarm business sector.
c. Mean expected change in consumer prices as measured by the University of Michigan’s Survey of Consumers.
SOURCES: U.S. Department of Labor, Bureau of Labor Statistics; Federal Reserve Bank of Cleveland; and University of Michigan.

Since his testimony, pressures on
the natural gas market seem to have
eased somewhat; energy companies
were reported to be replenishing
reserves at a record pace. Natural gas
inventories, though still somewhat
low, have now come back within
their five-year ranges and are more
able to meet cold-weather demands.
The market for gasoline has also
been feeling some strain recently; U.S.
fuel reserves are reportedly near a
three-year low. Reduced supplies and
the strong summer demand created by
increased travel have combined to

push gasoline prices up more than
50% between May and August,
although futures markets indicate that
as much as two-thirds of that rise could
reverse itself by the end of the year.
Inflationary patterns remain rather
subdued overall but are still quite
mixed by category. Goods prices continue to post outright declines,
whereas services prices are rising at an
annualized pace of about 3%. This
large discrepancy between goods and
services makes it difficult to discern
the economy’s underlying inflation
trend. A recent surge in U.S. labor
productivity, against a backdrop of

relatively modest growth in labor
compensation, has put substantial
downward pressure on the per-unit
labor costs of U.S. output. (Unit labor
costs have been on the decline for
much of the past three years.) Inflationary expectations may also be
trending lower, at least according to
the University of Michigan’s survey
data, which show that U.S. households anticipate price increases of
3
about 2 /4% over the next 12 months,
about one percentage point lower
than the rate they expected three
years ago.

4
•

•

Monetary Policy
Percent
1.375 IMPLIED YIELDS ON FEDERAL FUNDS FUTURES

Percent
8 RESERVE MARKET RATES

Effective federal funds rate a
1.250

August 25, 2003

1.125

Intended federal funds rate b

May 7, 2003

4
August 13, 2003
1.000

3
Primary credit rate b

July 24, 2003

2
0.875
Discount rate b

June 24, 2003
0.750
May

0
2000

2001

2003

2002

Percent, quarterly
9 FEDERAL FUNDS RATE AND INFLATION TARGETS

July

Sept.
2003

Nov.

Jan.

Mar.
2004

Percent
2.0 YIELD SPREAD: EFFECTIVE FEDERAL FUNDS RATE
MINUS THE TWO-YEAR TREASURY BILL
1.5

1.0

0.5

–0.5

–1.0

–1.5

Inflation targets: 4% 3% 2% 1% 0%
(federal funds rates implied by the Taylor rule) c

Federal funds rate

–2.0
–2.5

1998

1999

2000

2001

2002

2003

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

a. Weekly average of daily figures.
b. Daily observations.
c. The formula for the implied funds rate is taken from Federal Reserve Bank of St. Louis, Monetary Trends, January 2002, which is adapted from John B. Taylor,
“Discretion versus Policy Rules in Practice,” Carnegie-Rochester Conference Series on Public Policy, vol. 39 (1993), pp. 195–214.
SOURCES: Board of Governors of the Federal Reserve System, “Selected Interest Rates,” Federal Reserve Statistical Releases, H.15; and Bloomberg Financial
Information Services.

The Federal Open Market Committee made no change in its target for
the federal funds rate at its August 12
meeting. The Board of Governors of
the Federal Reserve System also left
the primary credit rate unchanged.
Before that meeting, market participants had believed there was some
chance of the FOMC lowering the target rate, a possibility reflected in
yields in the federal funds futures
market. Since then, however, the
thought of rates lower than 1% has
diminished, and the market seems to

be entertaining the thought of rates
increasing in early 2004.
One popular benchmark for the
federal funds rate is the Taylor rule,
which posits that the FOMC chooses
the target rate as a balanced response
to weakness and inflation. The form
of this rule depends on the weights
given to inflation, output, and the
assumed inflation target. Since fall
2002, rates have been well below those
suggested by the Taylor rule, even
assuming a rather high target inflation
rate of 4%.

Another benchmark—whether the
fed funds rate is “neutral” or “in line
with the market”—is based on the
ideas of Swedish economist Knut
Wicksell. It compares the fed funds
rate with other market rates. One
logical comparison is with the yield
on the two-year Treasury note, whose
maturity is long enough to ensure
that it is not just a reflection of immediate policy expectations. Although
the spread between these two rates is
now negative, it shows somewhat
less easing than the Taylor rule
benchmark would suggest.

5
•

•

Money and Financial Markets
THREE-DIMENSIONAL YIELD CURVE, 2003

As of 5/9
As of 6/20
As of 4/18

Percent
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5

As of 8/15

As of 6/27
0

0.25

0.50

1.00

As of 7/25

2.00

3.00
5.00
Years to maturity

Percent
6
YIELD CURVE a,b

7.00

10.00

20.00

Percent, weekly average
1.5 TREASURY-TO-EURODOLLAR (TED) SPREAD c

5
July 19, 2002

1.2

July 25, 2003

4
0.9
August 22, 2003
3
0.6
2

0.3

0
0

10
15
Years to maturity

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

a. Average for the week ending on the date shown.
b. All yields are from constant-maturity series.
c. Yield spread: three-month euro minus three-month constant-maturity Treasury bill.
SOURCES: Board of Governors of the Federal Reserve System, “Selected Interest Rates,” Federal Reserve Statistical Releases, H.15; and Bloomberg Financial
Information Services.

The yield curve has steepened since
last month, but this change represents
more than a bounce-back from the
summer’s exceptionally low longterm rates; current yields show an
increase over April and May as well.
The 10-year, three-month spread,
often cited as a predictor of future
economic growth, stands at a robust
350 basis points (bp), up from 325 last
month and 277 last year. Other
spreads also look promising. The TED
spread—the difference between

eurodollar deposits and Treasury
bonds, widely thought to reflect concern over international tensions—
remains quite low by recent historical
standards.
Although headlines usually focus
on nominal interest rates, the economy is affected by real rates, that is,
rates adjusted for inflation. Treasury
inflation-indexed securities (TIIS),
which adjust their principal and
interest for inflation, provide a direct
measure of real rates. Inflation expectations may also be used to estimate

real rates; the Pennacchi approach
estimates 30-day real rates. Short rates
have remained steadfastly negative
throughout 2003. Although long rates
have shown more variability, their
current value is near its level at the
beginning of the year.
Real interest rates matter because
they influence investment. One must
be careful to consider the appropriate
real rate, however, since most projects
implicitly embed a subtle option—
the option to wait. If real rates rise, the

(continued on next page)

6
•

•

Money and Financial Markets (cont.)
Percent
8 PENNACCHI MODEL a

Percent, weekly
7 TREASURY INFLATION-INDEXED SECURITY AND BERK RATE

7
6

Berk rate

30-day Treasury bill

5
5
4
3

2
3

10-year TIIS yield

0
2
–1

Estimated real interest rate

–2
1998

1999

2000

2001

2002

2003

1
1998

1999

2000

2001

2002

2003

Percent
3.0 IMPLIED INFLATION

Percent, weekly
2.5 BERK RATE MINUS 10-YEAR TREASURY
INFLATION-INDEXED SECURITY

Estimated expected inflation rate
2.5
2.0

2.0
1.5
1.5
10-year Treasury minus 10-year TIIS yield
1.0
1.0

0.5

0.5
1998

1999

2000

2001

2002

2003

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

a. The estimated real interest rate is calculated using the Pennacchi model of inflation estimation and the median forecast for the GDP implicit price deflator
from the Survey of Professional Forecasters. Monthly data.
SOURCES: Board of Governors of the Federal Reserve System, “Selected Interest Rates,” Federal Reserve Statistical Releases, H.15; Bloomberg Financial
Information Services; and Jonathan B. Berk, “A Simple Approach for Deciding When to Invest,” American Economic Review, vol. 89 (1999), pp. 1319–26.

waiting option has two contrary effects: The present value of
future profits is lower with higher interest rates, but delaying investment
also looks worse. Thus, the increase in
real rates has an ambiguous effect
on investment.
One way to adjust for this problem
is to use bonds that embed the option to wait. Fortunately, such
“callable” bonds, which the issuer
can buy back at a pre-specified price,
do exist. The chart at the lower left
takes a common callable bond, the

30-year Government National Mortgage Association bond, and subtracts,
as an estimate of inflation, the yield
difference between a 10-year Treasury bond and a 10-year TIIS. Both
the 10-year TIIS rate and the optionadjusted rate (dubbed the “Berk
rate,” after the economist who developed these ideas) have been increasing lately, but the Berk rate has risen
nearly 80 bp since early June, compared with the TIIS’s 60 bp rise.
The counterpart to real rates is
expected inflation. Although the

shorter-term measure from the Pennacchi model shows little movement,
longer-term expectations have gone
up more than 0.5% since early June.
Late July and early August saw increased volatility and a large spike in
yields on interest rate swaps. The market has since settled down, but the
event taught some useful lessons.
The price of a simple bond varies
inversely with market interest rates.
This makes sense because tomorrow’s
dollar is worth less today if the interest
rate is higher. How much a bond’s
(continued on next page)

7
•

•

Money and Financial Markets (cont.)
Price
1.1 PRICE VERSUS YIELD FOR A SIMPLE BOND
UNDER DIFFERENT MATURITIES a
1.0

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

1(1+r)

0.2

(1/(1+r)) 2

0.1

Price
1.1 DURATION AS SLOPE a

1/(1+r)
0.1

(1/(1+r)) 10

0
0

5
6
4
7
Interest rate, percent (r)

Price

5
6
4
7
Interest rate, percent (r)

Percent, daily
0.8 YIELD SPREAD

NEGATIVE CONVEXITY a,b

0.7

0.6

0.5

0.4
10-year interest-rate swap minus 10-year Treasury bond
0.3

0.2
7/1

Interest rate, percent (r)

7/11

7/21

7/31
2003

8/10

8/20

8/30

FRB Cleveland • September 2003

a. Author’s calculations.
b. As interest rates rise, the duration of a bond increases.
SOURCE: Bloomberg Financial Information Services.

price changes when the interest rate
changes depends on several things. A
longer-maturity bond is more sensitive
to interest rate changes because rates
are compounded over time.
Most bonds are more complicated,
with multiple payments and added
features, but a version of the same relationship between price and interest
rate still holds. Financial professionals
call the relation between interest rates
and a bond’s price the duration. This is
a weighted average of the maturity of

all payments, coupon and otherwise,
that tells how much the price of a bond
changes when interest rates change. It
can be thought of as the slope of the
price/interest rate curve, as shown in
the chart at the upper right.
Duration itself depends on interest
rates: The slope of the line is flat at
high interest rates and steep at low interest rates. How duration changes
depends on how “curvy” the line is,
which is termed convexity. Most
bonds have positive convexity—their
price line bends inward. But some

(many mortgage-backed securities,
for example) have negative convexity—their curve bows outward. When
interest rates started rising in late July
and people stopped refinancing their
mortgages, the duration of mortgagebacked securities began to increase.
Financial managers realized that this
increased their portfolios’ sensitivity
to interest rates and attempted to
reduce duration (and sensitivity). This
effort led them into the swaps market,
provoking the spike.

8
•

•

The Wealth of Nations
Index, 1971 = 100
195 PER CAPITA REAL GDP: NORMALIZED VERSUS SELF IN 1971
185

Index, 1971 = 100
235 PER CAPITA REAL GDP: NORMALIZED VERSUS U.S. IN 1971
215

Italy

175

195
165

Switzerland a

U.S.

175

155
Canada

145

155
Canada

135

Italy

U.S.

125
115
115

Switzerland a
95

105
95

75
1971

1975

1979

1983

1987

1991

1995

1999

Index, 1971 = 100
210 PER CAPITA REAL GDP: NORMALIZED VERSUS SELF IN 1971

1971

1975

1979

1983

1987

1991

1995

1999

Index, 1971 = 100
45 PER CAPITA REAL GDP: NORMALIZED VERSUS U.S. IN 1971
40

190

Argentina
35
170
30

Brazil
150

Uruguay
130

Brazil

Uruguay
110

Argentina

15
Peru

10
Peru

5
1971

1975

1979

1983

1987

1991

1995

1999

1971

1975

1979

1983

1987

1991

1995

1999

FRB Cleveland • September 2003

a. For Switzerland, 1997 was the last year for which data were available.
SOURCES: U.S. Department of Commerce, Bureau of the Census and Bureau of Economic Analysis; and United Nations, Department of Economic and
Social Affairs.

How does one measure the wealth of
nations? One possibility is using per
capital gross domestic product
(GDP) to see how a particular nation’s wealth has grown over time or
to compare changes in the wealth of
various nations.
Among Western developed countries, the U.S. and Italy had comparable growth rates between 1971 and
1999 (an annual average of about
2.2%). Over the same period,
Canada’s GDP growth rate (about
1.8%) lagged that of the U.S. and Italy,

but Switzerland’s rate was only about
0.6%. On the basis of growth rates
alone, one might conclude that the
U.S. is more prosperous than
Switzerland, but this would be incorrect. Calculating Switzerland’s per
capita GDP as a fraction of the U.S.
figure shows that Switzerland is the
wealthier of the two, but its advantage
was eroded between 1971 and 1999.
If both countries continue to grow at
rates comparable to that period, it
would take about 10 years—from
1999—for the U.S. to become as
wealthy as Switzerland.

Comparing some South American
countries’ average annual GDP growth
rates shows that Brazil is similar to the
U.S., and Uruguay and Argentina are
similar to Switzerland. Peru’s annual
average was negative over this period.
Compared to the U.S., however,
these countries are poor. At the end
of the sample period, Brazil’s wealth
was less than 25%—and Peru’s was
slightly more than 10%—of U.S.
wealth. Argentina is the richest in the
South American group, but its wealth is
only about 40% of the U.S.
(continued on next page)

9
•

•

The Wealth of Nations (cont.)
Index, 1971 = 100
200 PER CAPITA REAL GDP: NORMALIZED VERSUS SELF IN 1971

Index, 1971 = 100
25 PER CAPITA REAL GDP: NORMALIZED VERSUS U.S. IN 1971
South Africa

180
20

160
15
Congo
140
Congo
10

Benin
120

Zimbabwe

South Africa
5

100

Benin
Zimbabwe

0
1971

1975

1979

1983

1987

1991

1995

1999

1971

1975

1979

1983

1987

1991

1995

1999

Index, 1971 = 100
700 PER CAPITA REAL GDP: NORMALIZED VERSUS SELF IN 1971

Index, 1971 = 100
30 PER CAPITA REAL GDP: NORMALIZED VERSUS U.S. IN 1971

600

500

20
China
Thailand

400

15
Malaysia

300

Thailand

Indonesia

Indonesia
200

Malaysia

China
0

100
1971

1975

1979

1983

1987

1991

1995

1999

1971

1975

1979

1983

1987

1991

1995

1999

FRB Cleveland • September 2003

SOURCES: U.S. Department of Commerce, Bureau of the Census and Bureau of Economic Analysis; and United Nations, Department of Economic and
Social Affairs.

In Africa, Congo showed tremendous GDP growth in the late 1970s
and early 1980s but lost nearly all of it
by the end of the 1977–99 period.
Over the same period, the growth
rates for Zimbabwe and South Africa
were essentially zero. Comparison
with the U.S. shows how very poor
some African countries are. Benin’s
per capita GDP is only about 3% of
the U.S. figure, and Zimbabwe’s is
about 6%.
A group of Asian countries experienced exceptional growth over the

sample period. For example, China’s
wealth increased almost 600%, in
contrast to the U.S.’s increase of
about 85%. The Asian group’s slowest growth occurred in Indonesia,
which increased its wealth more
than 200%. Compared to the U.S.,
however, these countries remain
very poor. Per capita GDP in
Malaysia, the richest of this group,
was about a quarter of the U.S. figure. China is now the poorest in the
group, but if its per capita GDP continues to grow at the sample period’s

average annual rate, a phenomenal
7%, China would take about 55 years
to match the U.S.’s 1999 wealth. If the
U.S. continues to grow at its sample
period rate, China would need about
80 years to catch up. Of course,
China’s ability to continue growing at
a 7% rate is questionable.
One caveat: The annual average
market exchange rates were employed
to convert all series into U.S. dollars,
and comparisons may be sensitive to
the exchange rate used.

10
•

•

Economic Activity
Percentage points b
3.0 CONTRIBUTION TO PERCENT CHANGE IN REAL GDP

a,b

Real GDP and Components, 2003:IIQ
(Preliminary estimate)

Annualized
Change, percent change, last:
billions
Four
of 1996 $
Quarter
quarters

Real GDP
73.5
Personal consumption 62.4
Durables
55.8
Nondurables
5.6
Services
13.7
Business fixed
investment
22.7
Equipment
19.5
Structures
3.7
Residential investment
4.5
Government spending 34.5
National defense
40.6
Net exports
–33.3
Exports
–3.2
Imports
30.1
Change in business
inventories
–25.7

3.1
3.8
24.1
1.1
1.5

2.5
2.9
8.2
3.3
1.7

8.0
8.2
7.2
4.5
8.2
45.9
__
–1.2
7.9

1.2
4.0
–7.3
6.2
4.0
13.8
__
–1.0
2.9

2.5

Last four quarters
2003:IIQ

Personal
consumption
2.0

Government
spending

1.5
1.0
Residential
investment

0.5

Exports
0
Business fixed
investment
–0.5
–1.0

Change in inventories

Imports

–1.5

Percent change from previous quarter b
4.5 REAL GDP AND BLUE CHIP FORECAST

Index, 1985 = 100
150 CONSUMER EXPECTATIONS

4.0

140

110

130

105

120

100

Final percent change from advance estimate
Preliminary estimate

3.5

Blue Chip forecast c

3.0

Index, 1996:IQ = 100
115

110

2.5

95
Survey of Consumer Sentiment

30-year average
100

2.0
1.5
1.0

0.5

Survey of Consumer Confidence

IIQ

IIIQ
2002

IVQ

IIQ

IIIQ
2003

IVQ

IQ
2004

65
1997

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

NOTE: All data are seasonally adjusted.
a. Chain-weighted data in billions of 1996 dollars. Components of real GDP need not sum to the total because the total and all components are deflated using
independent chain-weight indexes.
b. Annualized
c. Blue Chip panel of economists.
SOURCES: U.S. Department of Commerce, Bureau of Economic Analysis; Conference Board; University of Michigan; and Blue Chip Economic Indicators,
August 10, 2003.

The preliminary estimate of real gross
domestic product (GDP) for 2003:IIQ—
the first of two planned revisions to
the advance estimate that the Commerce Department issued in July—
shows a greater increase in output
growth than initially reported. The
quarterly increase (3.1% at an annualized rate, 0.7 percentage point higher
than the advance estimate), results
primarily from revisions to personal
consumption expenditures (up $7.3
billion), net exports (up $10.0 billion), and state and local government
spending (up $2.1 billion). These
revisions were partly offset by a

$3.0 billion downward revision to private inventory investment.
The revisions did not affect which
categories contributed most to the
2003:IIQ increase in real GDP, which
continued to be personal consumption expenditures, federal defense
spending, and nonresidential fixed
investment. Inventory investment
and imports continued to exert a significant drag on real output growth.
Survey measures of consumer expectations rebounded in 2003:IIQ,
consistent with the improving economic environment suggested by the
rising rate of output growth. However,

although the university of Michigan’s
Survey of Consumer Sentiment finds
that consumer optimism has returned
to its pre-recession level (consistent
with the Blue Chip forecast of stronger
growth in the second half of 2003),
the Conference Board’s Survey of
Consumer Confidence suggests that
Americans remain less sanguine about
the economy than they were in early
2001. One explanation for the difference is that the Michigan survey asks
respondents to look a year ahead,
whereas the Conference Board asks
them to look ahead only six months.

11
•

•

Labor Productivity
Percent change
15 PRODUCTIVITY GROWTH

Percent change
15 OUTPUT GROWTH

Average, 1948–2000
10

Average, 1948–2000

2001 onward

5
1990–2001

1990–2001
2001 onward

–5

–5
0

3
5
6
8
9
4
7
Quarters from previous business cycle peak

Percent change
15 HOURS GROWTH

Percent change
15 UNIT LABOR COSTS GROWTH

Average, 1948–2000
1990–2001

Average, 1948–2000

2001 onward

1990–2001
2001 onward
–5

–5
0

3
9
8
5
6
4
7
Quarters from previous business cycle peak

3
5
6
8
9
4
7
Quarters from previous business cycle peak

FRB Cleveland • September 2003

NOTES: All data are for nonfarm businesses and are seasonally adjusted. Shaded bands indicate a 95% confidence interval for the 1948–2000 average.
SOURCE: U.S. Department of Labor, Bureau of Labor Statistics.

Growth in labor productivity (nonfarm business output per hour) has
been phenomenal in this business
cycle so far, much faster than in the
1990–2001 cycle. Cumulative growth
in productivity is up 10% from the last
peak, about three percentage points
higher than the postwar average for
this point in the cycle. The postwar
average includes the “golden age of
productivity growth,” which boosted
incomes in the 1950s and 1960s, as
well as the period of slower growth

that followed the 1970s oil crisis.
To judge just how well labor productivity has performed, consider that
the current cycle’s productivity
growth would be at the high end of
the range that typified the golden age
from 1948 to 1973. What accounts for
this strong performance?
This time around, nonfarm business output initially held up better
than in the average postwar cycle,
growing almost 5% from the last
peak, but its performance over the

last three quarters has been less
impressive. Output growth is now at
the low end of the typical postwar
range for this series, so it cannot be
the driving force behind the strong
productivity numbers.
In the calculation, this leaves
hours, whose dramatic drop seems
to be the main cause of the vigorous
productivity numbers. Although the
current business cycle started out in
a fairly typical way, hours have continued to drift down. By comparison,

(continued on next page)

12
•

•

Labor Productivity (cont.)
Percent change
20 MANUFACTURING PRODUCTIVITY GROWTH

Percent change
20 MANUFACTURING OUTPUT GROWTH

15
1990–2001

Average, 1948–2000
10

10
2001 onward

1990–2001
5

5
Average, 1948–2000
0

–5

–10

2001 onward

–15

–15
0

3
5
6
8
9
4
7
Quarters from previous business cycle peak

Percent change
20 MANUFACTURING HOURS GROWTH

Percent change
20 MANUFACTURING UNIT-LABOR-COSTS GROWTH

Average, 1948–2000
1990–2001

2001 onward

1990–2001
–5

–5
Average, 1948–2000

–10

2001 onward
–15

–15
0

3
5
6
8
9
4
7
Quarters from previous business cycle peak

FRB Cleveland • September 2003

NOTES: All data are seasonally adjusted. Shaded bands indicate a 95% confidence interval for the 1948–2000 average.
SOURCE: U.S. Department of Labor, Bureau of Labor Statistics.

total hours growth in the so-called
jobless recovery of the early 1990s
looks positively robust.
Productivity growth has an impact
on firms’ cost structures. Unit labor
costs combine compensation figures
with productivity data to measure
how much a typical firm spends to
produce its output. Unit labor costs
have also performed atypically, falling
about 3.25% instead of rising the
usual 5% at this point in the cycle.
All else equal, this should boost
firms’ profits.

Labor productivity growth, up
12.4%, has been even stronger in manufacturing, where it is well above the
range for a representative recovery. On
the other hand, manufacturing output
growth has been disappointing. The
current cycle began fairly typically, but
output growth has stalled over the
past three quarters.
With output flat, the only way to
achieve strong productivity growth is
a drop in hours, and this is what has
occurred. Manufacturing hours have
declined about 15% since the last

business cycle peak, far more than
the average postwar decline of about
4% for this point in the cycle.
Unlike overall nonfarm business,
manufacturing’s unit labor costs have
hardly changed over this cycle, which
implies that compensation growth
has largely kept up with productivity
growth. Nonetheless, manufacturing’s unit labor costs are running
below typical levels for this point in
the business cycle.

13
•

•

Labor Markets
Change, thousands of workers
300 AVERAGE MONTHLY NONFARM EMPLOYMENT CHANGE

Labor Market Conditions

250

Average monthly change
(thousands of employees)a

Preliminary

200

2001
–149

2002
–39

–1
7
–9
2
–11

–124
–1
–123
–88
–35

–64
–4
–57
–41
–16

–46
10
–55
–40
–15

–26
19
–44
–19
–25

Service providing
162
Information
15
Financial activitiesb
6
PBSc
40
Education and health
32
Leisure and hospitalityd 22
Government
22

–25
–15
7
–63
51
–2
46

25
–14
5
–10
37
7
16

–3
–11
12
11
18
4
–10

–67
–16
–1
–28
24
5
–26

Payroll employment

150

Goods producing
Construction
Manufacturing
Durable goods
Nondurable goods

100
50
0
–50
–100

Jan.–July Aug.
2003
2003
–49
–93

2000
161

Revised

Average for period (percent)

–150

Civilian unemployment
rate

4.0

4.8

5.8

6.0

6.1

–200
1999 2000 2001 2002

IIIQ IVQ
2002

June

IQ IIQ
2003

July Aug.
2003

Percent
65.0 LABOR MARKET INDICATORS

Percent
6.5

Employment Change in U.S. and Fourth
District States, March 2001 Peak to July 2003

Employment-to-population ratio
64.5

6.0
Percent change in
nonfarm employment

Percentage point change in
unemployment rate

5.5

64.0

63.5

5.0

63.0

4.5

62.5

4.0

U.S.

–2.0

2.0

Ohio

–3.2

2.6

Pennsylvania

–1.6

1.3

West Virginia

–1.5

2.0

Kentucky

–3.0

1.1

Civilian unemployment rate
3.5

62.0
1995

1996

1997

1998

1999

2000

2001

2002

2003

FRB Cleveland • September 2003

NOTE: All data are seasonally adjusted.
a. Data are according to the North American Industrial Classification System.
b. Financial activities include the finance, insurance, and real estate sector and the rental and leasing sector.
c. Professional and business services, including professional, scientific, and technical services, management of companies and enterprises, administrative and
support, and waste management and remediation services.
d. Leisure and hospitality includes arts, entertainment, and recreation, as well as accommodation and food service.
SOURCE: U.S. Department of Labor, Bureau of Labor Statistics.

Nonfarm payroll employment fell
93,000 jobs in August. Net losses were
revised from 44,000 to 49,000 jobs for
July and from 72,000 to 83,000 jobs for
June. Employment has declined about
900,000 jobs since the recovery began
in December 2001; the drop from the
previous peak is about 2.8 million.
In August, declines were widespread
in both sectors, goods-producing
(down 26,000) and service-providing
(down 67,000). In goods production,
manufacturing employment dropped
another 44,000 jobs. In the past three
years, manufacturing has declined
about 2.7 million jobs, including

431,000 jobs this year alone. Construction remained strong, adding 19,000
jobs. Information services continued to
shed jobs, 16,000 of them in August.
Since its peak of 1.3 million jobs in
March 2001, this industry has lost
about 212,000 jobs. Professional and
business services lost 28,000 jobs in August and 232,000 jobs since its March
2001 peak. Since February of this year,
state and federal government has continued to shed jobs amounting to
131,000. Education and health services
added 24,000 jobs; leisure and hospitality added 5,000, the third consecutive monthly gain this year.

The unemployment rate inched
down by 0.1 percentage point to 6.1%
in August, and the employment-topopulation ratio remained unchanged
at 62.1%.
Ohio’s nonfarm employment fell
3.2% between March 2001 and July
2003, farther than either the average
U.S. drop of 2.2% or the drop in
other Fourth District states. Similarly,
Ohio’s unemployment rate has
jumped 2.6 percentage points since
the 2001 peak, more than either the
U.S. or the other Fourth District states.

14
•

•

Elderly Americans
PERCENT OF PEOPLE OLDER THAN 65

MEDIAN AGE

U.S. median: 35.3 years

U.S. average: 12.4%

Lower than nation
Within + 1 percentage point of nation
Higher than nation

Lower than nation
Within + 1 year of nation
Higher than nation

INCREASE IN PEOPLE OLDER THAN 65, 1990–2000

Places with the Highest Concentrations
of Elderly Peoplea

U.S. average: 12%

Percent of
population over 65
Clearwater, FL

Less than 10%
10–13.9% (near U.S. average)
14–25%
Greater than 25%

21.5

Cape Coral, FL

19.6

Honolulu, HI

17.8

St. Petersburg, FL

17.4

Hollywood, FL

17.3

Warren, MI

17.3

Miami, FL

17.0

Livonia, MI

16.9

Scottsdale, AZ

16.7

Hialeah, FL

16.6

FRB Cleveland • September 2003

a. Places with populations of at least 100,000.
SOURCE: U.S. Department of Commerce, Bureau of the Census.

As life expectancies have risen, meeting the needs of increasing numbers
of older Americans has become a
pressing concern. Medicare, prescription drug coverage, nursing
homes, and other quality-of-life issues affecting the elderly have moved
to the forefront of public policy debate. These issues affect states and
cities alike throughout the country,
but are more urgent in places with a
higher level or a faster growth rate of
elderly people.
In 2000, the median age in the U.S.
was 35.3 years—half the people were

older and half were younger. In most
states, the median age was close to
the nation’s; the median ranged from
a low of 27.1 years in Utah to a high of
38.9 in West Virginia. In the New England states, along with West Virginia,
Iowa, Montana, and Florida, median
ages were above average; in many
states in the South and West they
were below average.
In the U.S. as a whole, 12.4% of the
people were older than 65, the age
the Census Bureau designates as
“elderly.” Higher- than-average shares
of elderly people were concentrated
in the Central region, in parts of

New England, and in Florida, the
state with the highest share (17.6%).
States with below-average shares
were in the Mid-Atlantic region, most
of the West, and Georgia, Texas, and
Alaska, which was the state with the
smallest share (5.7%).
Between 1990 and 2000, the number of elderly people increased 12%.
In the New England, Midwest, and
Central regions, the elderly population changed little during this time,
but it grew rapidly in the Southwest
and Southeast; in Nevada alone, the
number increased more than 70%.
(continued on next page)

15
•

•

Elderly Americans (cont.)
Percent
20 PEOPLE OLDER THAN 65 IN THE FOURTH DISTRICT

MEDIAN AGE
U.S. median: 35.3 years

Older than 65
Change from 1990 to 2000

Lower than nation
Within + 1 year of nation

Higher than nation

0
U.S.

PEOPLE OLDER THAN 65 IN FOURTH DISTRICT COUNTIES
U.S. average: 12.4%

Kentucky

Ohio

Pennsylvania

West Virginia

Concentrations of Elderly People
in Fourth District Citiesa
Percent of
population over 65
Pittsburgh, PA

16.4

Erie, PA

15.4

Akron, OH

13.5

Toledo, OH

13.1

Cleveland, OH

12.5

Cincinnati, OH

12.3

Lower than nation
Within + 1 percentage point of nation

Dayton, OH

12.0

Lexington, KY

10.0

Higher than nation

Columbus, OH

8.9

FRB Cleveland • September 2003

a. Places with populations of at least 100,000.
SOURCE: U.S. Department of Commerce, Bureau of the Census.

Most of the cities with the largest
concentrations of elderly are located
in the South, with Florida topping the
list. Only two northern cities, Warren
and Livonia, Michigan, made the list.
Most counties in the Fourth Federal
Reserve District have median ages that
exceed the U.S. average. They range
from Athens County, Ohio (25.7 years),
to Forest County, Pennsylvania (44.2
years). The few counties where the
median age is lower than average are
those with colleges or universities
(such as Wood County, Ohio, or
Rowan County, Kentucky), counties
along the growing I-75 corridor in

Kentucky, and Holmes County, Ohio,
with its large population of Amish.
Counties whose median ages approximate the national average are located
in other growing areas of southern
Ohio.
In every state of the Fourth District, the share of elderly residents
exceeds the national average; both
Pennsylvania and West Virginia have
more than 15%; Kentucky and Ohio
are closer to average. However, this
age group grew at a much slower rate
in the District than in the nation
between 1990 and 2000.
The region as a whole has much
higher concentrations of elderly than

the national average; all Fourth
District counties in West Virginia and
Pennsylvania are above average (with
Forest County, Pennsylvania, again
highest at 20%). The below-average
percentages of elderly are located
in the Columbus, Cincinnati, and
Lexington areas.
Of the nine largest cities in the
District, three have concentrations of
13% or more; only two have 10% or
less. Like the counties, the northern
cities of Ohio and Pennsylvania have
higher proportions of elderly than do
the central and southern cities.

16
•

•

Federal Home Loan Banks
Billions of dollars
600 ASSETS

Billions of dollars
800 LIABILITIES
700

500

Deposits and borrowings

Advances

Consolidated obligations

Investments

600

Other liabilities
Capital

Other assets

400

500
300

400

300
200
200
100
100
0

0
1994

1995

1996

1997

1998

1999

2000

2001

2002

2003 a

1995

1996

1997

1998

1999

2000

2001

Percent
8 CAPITAL

COMPOSITION OF OTHER ASSETS a
Interest receivable
3%
Derivative assets
3%

Net mortgage loans
94%

1994

2002 2003 a

Billions of dollars
40
35

7
Percent of assets

0
1994

1995

1996

1997

1998

1999

2000

2001

2002

2003 a

FRB Cleveland • September 2003

a. Data are through 2003:IIQ.
SOURCES: Federal Home Loan Bank System, Quarterly Financial Report, August 13, 2003, and annual reports.

The 12 Federal Home Loan Banks
are stock-chartered, governmentsponsored enterprises whose original
mission was to provide short-term
advances to member institutions,
funded with those institutions’
deposits. Membership was open to
specialized housing-finance lenders,
mostly savings and loan associations
and mutual savings banks. With continued shrinkage of their traditional
clientele and ongoing consolidation
of the financial system, the FHLBs
have been reinventing their role
in financial markets. Their advances,
which now represent an important
source of funding for member

institutions’ mortgage portfolios, rose
to $506 billion at the end of 2003:IIQ,
far outstripping all their other investments and assets.
By far the largest share of FHLBs’
assets came from the $710 billion
of consolidated obligations of the
Federal Home Loan Bank System—
bonds issued on behalf of the 12
FHLBs collectively. The market considers these bonds to be implicitly
backed by the U.S. government; consequently, the FHLBs can raise funds
at lower rates of return than AAArated corporations. Member institutions’ deposits and short-term borrowings, along with other liabilities,

provided only a miniscule share of
funds. The FHLBs have added to
their capital as they have grown, but
asset growth has outstripped capital
growth; the capital-to-asset ratio fell
from 5.8% in 1996 to 4.6% at the end
of 2003:IIQ.
In 1997, the Chicago FHLB initiated
the Mortgage Partnership Finance Program, through which it began investing directly in mortgages in addition to
supporting members’ own mortgage
portfolios through advances. All
FHLBs currently purchase mortgages
directly from member institutions.
The FHLBs now hold $90 billion in
mortgages, more than double what
(continued on next page)

17
•

•

Federal Home Loan Banks (cont.)
Millions of dollars
4,000 COMPOSITION OF INCOME

Millions of dollars
2,500 NET INCOME

3,500

Net interest income
Net noninterest income

3,000

2,000

2,500
2,000

1,500

1,500
1,000

1,000

500
0

500

–500
–1,000

1994 1995 1996 1997 1998 1999 2000 6/01 2001 6/02 2002 6/03 a

Percent
0.7 PROFITABILITY

Percent
10 RETURN ON EQUITY

Ratio
27

9
0.6

Net interest margin

24
Equity multiplier c

0.5

6
15
0.4

5
12

Return on assets

0.3

0.2

1
0.1
1994

0
1995

1996

1997

1998

1999

2000

2001

2002 2003 b

0
1994

1995

1996

1997

1998

1999

2000

2001

2002

2003 b

FRB Cleveland • September 2003

a. Data are through 2003:IIQ.
b. Data for 2003:IIQ are annualized.
c. The equity multiplier is the ratio of total assets to equity.
SOURCES: Federal Home Loan Bank System, Quarterly Financial Report, August 13, 2003, and annual reports.

they held a year ago, and mortgage
portfolios are projected to be a major
source of their asset growth in the
future.
FHLBs’ earnings grew steadily
from 1994 through 2000 before declining in 2001 and 2002. However,
their net income of $907 million for
the first six months of 2003 exceeds
the $862 for the same period in 2002.
FHLBs’ net interest income (interest income less interest expense) rose
from $735 million in 1992 to $3,311
million at the end of 2000. For the first
six months of 2003, their net interest
income of $1,392 million was down

from $1,433 million for the same period in 2002. The most important reason for the increasingly negative
spread between non-interest income
and non-interest expense since 1993 is
a steady increase in FHLBs’ operating
expenses, especially for employee
compensation and benefits.
Improvements in earnings and net
interest income have resulted from
strong asset growth rather than
greater underlying profitability. Return
on assets fell from 75 basis points (bp)
in 1991 to 31 bp at the end of 2001.
The annualized return on average
assets through 2003:IIQ is 23 bp.

Profitability has been hurt by the net
interest margin’s decline from 45 bp at
the end of 2001 to an annualized 36 bp
for the first six months of 2003. Moreover, FHLBs’ net interest margins are
far lower than the 300 bp to 400 bp
typical of depository institutions.
Finally, despite continued increases in leverage since 1996, return
on average equity fell from 6.3% at
the end of 2001 to 4.9% in the first six
months of 2003. These persistently
weak returns on assets and equity
put further pressure on the FHLBs to
undertake nontraditional lines of
business in search of higher returns.

18
•

•

Foreign Central Banks
Percent, daily
7 MONETARY POLICY TARGETS a

Trillions of yen
–35

CHANGES IN POLICY RATES SINCE LAST FOMC CUT

–30
–25

Malta

–20

Taiwan

–15

5
Bank of England

European Central Bank

South Korea

–10

2
Federal Reserve

Philippines

–5

–1

New Zealand

Bank of Japan
–2

Sweden

–3

Czech Republic

–4

–5

–6

–7

35
10/28

4/1
2001

12/22

5/26
2002

Canada
Bulgaria

7/20
2003

Israel
Thailand
Indonesia

Members of the Bank for International
Settlements
Argentina
Australia
Bosnia and Herzegovina
Brazil
Bulgaria
Canada
China
Croatia
Czech Republic
Denmark
Estonia
European Central Bankb
Hong Kong

Hungary
Iceland
India
Japan
Latvia
Lithuania
Macedonia
Malaysia
Mexico
New Zealand
Poland
Romania
Russia

Sri Lanka
Peru

Saudi Arabia
Singapore
Slovakia
Slovenia
South Africa
South Korea
Sweden
Switzerland
Thailand
Turkey
United Kingdom
United States
Yugoslaviac

South Africa
Norway
Lithuania
Brazil
Mexico
Turkey
Venezuela

–2

–4

–6

–8

–10

–12

Percent

FRB Cleveland • September 2003

a. Federal Reserve: overnight interbank rate. Bank of Japan: a quantity of current account balances (since December 19, 2001, a range of quantity of current
account balances). Bank of England and European Central Bank: two-week repo rate.
b. European Central Bank plus each of its 12 constituent nations’ central banks.
c. Constitutional changes in February 2003 transformed the Federal Republic of Yugoslavia into the State Union of Serbia and Montenegro, with two separate
central banks. The legal status of the Yugoslav issue of the capital of the Bank for International Settlements is currently under review.
SOURCES: Board of Governors of the Federal Reserve System; Bank of Japan; European Central Bank; Bank of England; Wholesale Markets Brokers
Association; and Bloomberg Financial Information Services.

None of the four major central banks
has changed its operating target since
July 10, when the Monetary Policy
Committee of the Bank of England
reduced its policy rate from 3.75% to
3.5%. On balance, since the Federal
Open Market Committee’s last rate reduction, most of the other rate-setting
central banks tracked by Bloomberg
also have lowered their operating
targets at least 25 basis points.
A central bank typically acts as the
bank for a nation’s banks, which use
central bank deposit transfers to settle
their debts to one another. The Bank

for International Settlements (BIS) is a
central bank for central banks, established in 1930 to facilitate payment of
international reparations from World
War I. In addition to its banking functions, the BIS is an important center
for economic and financial research
that facilitates international discussion
and helps coordinate decision making
among central banks.
The Reserve Bank of New Zealand
announced in August that it has
accepted an invitation to join the BIS,
becoming the sixteenth new member since 1996. The BIS has been

enlarging its membership to “underpin its increasingly global activities
and the regional interests of its shareholding members.” New Zealand’s
central bank says that it expects benefits to include “increased access to
foreign currency funding, which in a
crisis could enhance the Bank’s
capabilities to intervene in markets to
maintain stability. Being a BIS shareholder also more generally strengthens the Reserve Bank’s connections
with the international central banking community, which in a period of
financial stress could be very useful.”

Full text of Economic Trends (Federal Reserve Bank of Cleveland) : Economic Trends - September 2003

FRASER