Full text of The Southwest Economy : Third Quarter 2015

View original document

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

DALLASFED
THIRD QUARTER 2015

Southwest
Economy

}

Wage Flexibility in Texas
May Ease Impact of
Tighter Monetary Policy
PLUS
} Mexico’s Four Economies Reflect Regional Differences,
Challenges

} Texas Maintains Top Exporter Standing While Its Trade
Remains Concentrated

} On the Record: Greece’s Fiscal Woes Among Issues
Hobbling Euro Zone Rebound

} Spotlight: Diversified Houston Spared Recession … So Far

PRESIDENT’S PERSPECTIVE

great laboratory
}Inof the
the 50 united
states—with its
diversity of institutions,
business climates and
geographies—economic
forces play out in
different ways.

hile I was growing up in Kansas, my parents often
stressed to me that hard work breeds success. My
father was a jewelry salesman serving small and
mid-sized stores in Texas and the Southwest. During school vacations, I frequently joined him on his travels
and learned about the challenges of running a small business. My mother worked as a real estate agent and taught
me the importance of helping people reach their objectives.
Those experiences helped make me who I am and
stayed with me during my careers in financial services and
academia, which regularly brought me back to Texas. They
certainly guide me as I begin as president of the Federal
Reserve Bank of Dallas.
I strongly believe jobs are central to economic well-being, a point reaffirmed in this issue of Southwest Economy.
In the great laboratory of the 50 united states—with its diversity of institutions, business climates and geographies—
economic forces play out in different ways. One important
dynamic is the tradeoff between wage growth and unemployment, as Anil Kumar writes in “Wage Flexibility in Texas
May Ease Impact of Tighter Monetary Policy.”
The work of economist A.W. Phillips and his Phillips
curve suggests that as unemployment falls and slack is
wrung out of the labor market, wages tend to rise, boosting
inflationary pressures. This relationship is one of the indicators Federal Reserve policymakers take into account when
considering whether to change the short-term federal funds
interest rate.
Wages tend to be more flexible in Texas than in other
states, Kumar finds. There are a number of reasons for
that—among them, a lower minimum wage here and more
unionization and greater labor market regulation elsewhere. Thus, during economic weakness, Texas workers
accept less for their efforts and employers resort to fewer
layoffs. In better times, that flexibility helps wages climb
faster here.
Kumar’s article, along with the others in this issue,
underscores the importance of regional differences in
understanding economic change and the implications for
monetary policy.
This research is one example of the terrific work being
done here at the Dallas Fed. I look forward to meeting and
learning from you as I begin this new chapter and help
guide the Eleventh District through the exciting period
ahead.

Robert S. Kaplan
President and Chief Executive Officer
Federal Reserve Bank of Dallas

Wage Flexibility in Texas
May Ease Impact of
Tighter Monetary Policy
By Anil Kumar

}

t times of rising unemployment, wage growth tends to
slow.
This inverse relationship
is one of economics’ most enduring
tenets and is captured in the work of
economist A.W. Phillips and his Phillips curve.1
The Phillips curve helps determine
the amount of expected price or wage
inflation for a given change in the unemployment rate.2
The Phillips curve also remains an
important tool for gauging the responsiveness of real (inflation adjusted)
wages to unemployment. The steeper
the curve, the more flexible or responsive are wages to unemployment
rate shifts. The degree of wage rigidity
helps policymakers assess the ability
of monetary policy to affect output and
unemployment.
The Phillips curve for Texas is
steeper than the one for the U.S., based

ABSTRACT: Because wages
are more flexible in Texas than
in other parts of the U.S., the
state’s unemployment rate will
be less prone to rise when
interest rates increase.

Chart

Steeper Phillips Curve Indicative of Flexible Labor Markets

Real wage growth (percent)

Texas
U.S.

Texas Phillips Curve

2
0
–2
–4

on a review of state-level unemployment rate data from the Bureau of
Labor Statistics (BLS) and hourly wages
from the Census Bureau’s Current Population Survey (CPS) (Chart 1).3 The
steeper Phillips curve and greater wage
flexibility suggest that when interest
rates rise, unemployment will increase
less in Texas than elsewhere.
Monetary policy can affect individual states differently because they
vary widely in the timing, duration and
stage of their business cycles and in the
extent of labor availability, or slack.4
Moreover, states’ economies differ significantly with regard to industry composition, the presence of small versus
large banks, and firm size—factors that
can cause states to respond differently
to monetary policy shocks.5
Because monetary policy is formulated at the national level, the sensitivity of wage growth to unemployment
rate change generally focuses on activity across the country. But this national
viewpoint often masks significant local
differences. Conversely, state-level information yields more precise measurement of the Phillips curve relationship
nationally. It also helps us understand
the local effects of monetary policy
changes in places such as Texas.

Unemployment rate (percent)

NOTE: Each dot represents annual average real wage growth and unemployment rate for a particular year from 1982 to
2013 for the U.S. and Texas.
SOURCES: Bureau of Labor Statistics’ Current Population Survey; Census Bureau; author’s calculations.

Real wage growth tends to accelerate more rapidly in Texas than
the nation when unemployment is low
and decelerate more sharply when
unemployment is high, as depicted
in Chart 1. The graphic is drawn from
aggregated CPS data for Texas and the
U.S. from 1982 to 2013. The unemployment rate is calculated as the number
of unemployed as a percent of all workers in the labor force. The real wage

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

measure excludes overtime pay and
fringe benefits.
The linear fit on the chart shows
that the relationship between real wage
growth and the unemployment rate
has a steeper slope in Texas than in
the nation, indicating that wages are
more flexible in Texas. A percentagepoint decline in the unemployment
rate leads to real wage growth of 0.65
percentage points in Texas, compared
with 0.42 percentage points for the
U.S. The response of inflation-adjusted
wage growth to a given change in the
unemployment rate is therefore about
0.23 percentage points stronger than in
the nation.
The heightened flexibility of Texas
wages means they are more responsive
to changes in the unemployment rate
and adjust more freely. Texas ranks
high among states on this measure of
wage flexibility and is in the top quintile of responsiveness of wage change
to movements in the unemployment
rate (Chart 2).

Greater Wage Flexibility
The presence of wage rigidity is
fundamental to the existence and persistence of unemployment. In standard
economic models that assume flex-

Chart

ible wages, unemployment arises only
because workers are in the process of
a job search or transitioning between
jobs. Wages adjust instantaneously
to clear the labor market. When such
models are extended to incorporate
real wage rigidity, structural or involuntary unemployment arises because
the number of job seekers exceeds the
number of workers firms are willing
to hire at the prevailing real wage. An
oversupply of labor is created.
Why can’t wages adjust freely so
that supply and demand of workers is
in balance? There are several potential
explanations.
First, a job can be viewed as an
implicit contract between workers and
firms in which risk-averse employees
trade greater job security for more
stable, though less lucrative, pay.6 Second, many firms voluntarily pay above
market-clearing wages to encourage
worker effort rather than engage in
costly labor monitoring to prevent
shirking.7 Such efficiency wages also
limit worker turnover, helping firms
save on new-employee training. Third,
labor market imperfections such as
internal labor markets—typically, the
filling of positions from within companies rather than through open compe-

Texas Ranks High Among States in Real Wage Flexibility

tition—also prevent wages from fully
adjusting.8
Additionally, some government
policies prevent wages from falling
enough to clear the surplus of workers
over jobs. For example, more generous
unemployment benefits raise the wage
at which workers are willing to accept a
new job. Indeed, higher jobless benefits
raise the wage a firm must offer to attract
available workers. Minimum-wage laws
similarly hinder free adjustment of pay.
The degree of wage rigidity is
correlated with other characteristics
of labor markets. The prevalence of
unions in certain industries is an
important impediment to full adjustment of wages. Wage rigidity is further
correlated with manufacturing’s share
of the economy and the concentration
of public sector employment.
The presence of immigrant labor
with less bargaining power than native
workers often mitigates wage rigidity.
Such workers are also less likely to be
covered by union agreements. Moreover, undocumented immigrants may
be more willing than others to work for
less than the minimum wage.
Finally, wages tend to be more
rigid in large companies than in small
firms that can monitor worker effort
more easily without having to pay efficiency wages to induce effort.
Given these explanations for wage
rigidity, it is not surprising that wages in
Texas are more flexible. The state has a
lower minimum wage than other large
states, provides less-generous unemployment benefits than the national average
and has less union participation than the
rest of the country. Immigrant workers
are more common in Texas, where rightto-work rules and lighter government
regulation help the state rank high on
business-climate indicators.

Assessing Policy Implications

Lighter shades indicate more flexible wages:*
–1 to –.6

–.6 to –.48

–.48 to –.38

–.38 to –.3

–.3 to –.2

*Shades represent quintiles of states ranked by wage flexibility on the basis of the slope of the Phillips curve; for
example, the lightest shade (–1 to –.6) indicates that the 20 percent of states with the most flexible wages have slopes
between .6 and 1.
SOURCES: Bureau of Labor Statistics’ Current Populations Survey; Census Bureau; author’s calculations.

The consequence of wage rigidity
can become particularly apparent during an economic downturn, when firms
often choose between two options to
reduce labor costs: cut wages and hours
or lay off workers. If lowering wages is
difficult, layoffs become the preferred
choice. Because the supply of workers

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

then exceeds demand at the prevailing
pay, such wage rigidity is correlated
with unemployment and other measures of labor market slack.9 Inflexible
wages can also contribute to unemployment persistence—when joblessness in
one period fails to disappear in the next,
a phenomenon called “hysteresis.”10
Wage rigidity not only has a direct
effect on the unemployment rate, it
plays a key role in monetary policy’s
impact on employment and output.
Economists have long suggested that
monetary policy shocks can affect the
real economy only if wages and prices
are inflexible. The greater the wage
rigidity, the more pronounced the
impact of monetary policy on real personal income, gross domestic product
and unemployment.
A contractionary monetary policy
shock—for example, higher interest
rates—could produce larger and more
persistent increases in unemployment
in states with significant wage rigidity.
States with more flexible wages, such
as Texas, will more easily adjust to an
interest rate change. Previous research
has also suggested that because of relatively stronger economic conditions in
Texas than in the rest of the U.S., shortterm interest rates could have been
higher here than the near-zero rate that
policymakers installed after the Great
Recession began.11

recent years in both Texas and the U.S.,
with the slope edging closer to zero.
For the nation, the predicted
decline in real wage growth for a
1-percentage-point increase in the
unemployment rate—in absolute-value
terms—peaked at 0.44 percentage
points in 2006 and declined to 0.36 in
2013. The decline in Texas was even
sharper—from 0.88 to 0.67. Increased
wage rigidity is thought to be a key
explanation for a surprising lack of wage
stagnation during the Great Recession
and for weak real wage growth during
the recovery.13
If employers cannot sufficiently
lower wages when the economy
slumps, they will be slow to increase
wages when conditions improve.
Several factors may have contributed
to generally heightened wage rigidity
nationally and in Texas since 2008.
First, wage rigidity tends to be
countercyclical, and increased rigidity during downturns typically lingers
before subsiding.14 Another possible
explanation is the phased increase in the
federal minimum wage, from $5.15 to
$7.25 per hour, between 2007 and 2009.
Apart from the national impact, the
higher minimum wage may also have
contributed—with some lag—to the
post-2009 spike in wage rigidity in Texas.
The minimum wage increase mattered more in Texas than in the U.S.,

lower
}While
unemployment
rates lead to greater
wage growth, higher
unemployment
rates do not lead to
proportionately lower
wage growth due to the
relative inability of firms
to reduce wages.

Comparing Texas, U.S.
Measuring the response of wages
to the unemployment rate over time
helps draw the distinction between the
U.S. and Texas. The depiction of the
Phillips curve relationship in Chart 3
suggests that wages in the state were
more sensitive to changes in unemployment than they were nationally
during the period studied, 1999 to 2013.
The Phillips curve’s slope—the
change in wage growth for a given
change in the unemployment rate—is
estimated in decimal form for each
year, using data from 1982 through the
year shown. For example, the slope
for 1999 is based on 19 years of data
from 1982 to 1999; the slope for 2013
was based on data from 1982 to 2013.12
Wages have become less flexible in

Chart

Real Wages More Flexible in Texas Even as Flexibility Declines

Phillips curve slope

–.3

U.S.
–.36

–.5

–.44

–.43

–.44

Texas
–.7

–.9

–.67

–.91

–.88
–1.01

–1.1
’99

’00

’01

’02

’03

’04

’05

’06

’07

’08

’09

’10

’11

’12

’13

NOTE: The slope of the Phillips curve for the U.S. and Texas is estimated using a regression of real wage growth on
the unemployment rate since 1982. The estimation accounts for other factors that differ across states and over time.
SOURCES: Bureau of Labor Statistics’ Current Population Survey; Census Bureau; author’s calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

on average, because many other states
already had a higher minimum wage
than the federal level. Additionally,
Texas has a larger share of hourly paid
workers who were likely affected by the
increase. That said, the sharper spike
in the state’s wage rigidity vis-à-vis the
nation may simply reflect more volatile
labor market data at the state level.

Wage Growth Feeding Inflation
The Phillips curve slope also may
vary with the unemployment rate.
When economic conditions deteriorate
and unemployment is high, firms have
an incentive to lower pay to cut labor
costs. While raising wages when the
economy is hot and unemployment is
low presents no particular challenge
for firms, lowering wages when unemployment is greater is more difficult
and results in a relatively flatter Phillips
curve. Though this characteristic is
difficult to detect at the state level, its
presence can be easily established
nationally and has important monetary
policy implications.
The national Phillips curve slope
is significantly steeper when the
unemployment rate is below its longterm average than when it is above
the average (Chart 4). An important
implication is that continued declines
in unemployment when the rate is
already low may lead to significantly
stronger real wage growth that can feed
into overall inflation.
The Phillips curve slope at belowaverage unemployment has been
stable at about -0.5, except for a period
between 2003 and 2006 when wage
flexibility at lower levels of unemployment hit a high. A potential explanation is a decline in public sector employment during those years that likely
enabled wages to adjust more easily.
The slope of the Phillips curve
at above-average unemployment
remained largely stable until the onset
of the Great Recession, although it has
drifted toward zero since then, becoming less negative. This is not surprising
because the data since 2008 correspond with a period when the unemployment rate was high and real wage
growth was rather subdued.

Additionally, the downward
movement in the Phillips curve slope
following 2008 may partly reflect the
effect of the minimum-wage increase
that was fully phased in during 2009.
The extended availability of unemployment benefits coming out of the Great
Recession also may have impeded
adjustment of wages because the
payments effectively raised the wage
firms needed to pay to attract potential
workers.
Another reason real pretax wages
may be more rigid post-2009 is that the
“payroll tax holiday”—a temporary reduction in the payroll tax from 6.2 to 4.2
percent—was in effect between 2011
and 2013. This may have induced firms
to limit increases in the pretax wage as
worker take-home pay rose because of
the tax-rate cut.

Differences Among States
The varied responses of wages in
high- and low-unemployment rate situations have important implications for
wage growth, particularly if there are
significant differences in joblessness
among states. Indications of a widening
gap between high- and low-unemployment scenarios heightens the probable
effect on wage growth.
Using data through 2000, previous
research reveals that cross-state differences in labor market slack amplify

Chart

the wage-growth response of a given
change in the unemployment rate.15
If unemployment rates are uniform
across states and equal the national
long-term average of about 6 percent,
the model used for Chart 4 implies
modest real wage growth of about 0.1
percent in 2013.16
If the unemployment rate is 5 percent in half the states and 7 percent in
the rest, the national average remains
at 6 percent, the model predicts real
wage growth of 0.66 percent for lowunemployment states and real wage
deflation of 0.19 percent for highunemployment states, making average
real wage growth 0.24 percent.
Clearly, predicted wage growth
when the unemployment rate differs
across states is higher than when the
unemployment rate is uniform. Thus,
for a given national unemployment rate,
greater divergence in labor market slack
is associated with higher wage pressure.
The economic explanation for why
cross-state diversity in unemployment
rates yields higher wage growth stems
from downward wage rigidity. While
lower unemployment rates lead to
greater wage growth, higher unemployment rates do not lead to proportionately lower wage growth due to the relative
inability of firms to reduce wages.
A measure of unemployment rate
variability across states shows that it is

Phillips Curve Steeper When Unemployment Is Low

U.S. Phillips curve slope

–.30
–.35

–.31

Above-average unemployment
–.40

–.38

–.40
–.45
–.50

Below-average unemployment

–.51

–.55

–.53

–.60
–.65
’99

–.63
’00

’01

’02

’03

’04

’05

’06

’07

’08

’09

’10

’11

’12

’13

NOTE: The slope of the Phillips curve for the U.S. is estimated using a regression of real wage growth on the
unemployment rate on data since 1982. The estimation accounts for other factors that differ across states and
over time.
SOURCES: Bureau of Labor Statistics’ Current Population Survey; Census Bureau; author’s calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

significantly below the levels of the late
1980s and has remained largely stable
since 1990 (Chart 5).17
The jobless recovery that followed
the 2001 recession appears to have affected most states similarly, mitigating
cross-state variability in unemployment
rates. As a result, state-level differences
account for wage pressures to a much
smaller extent than in the 1980s. But
insofar as modest cross-state differences in labor market slack persist, they
remain a source of wage pressure.

Prospect of Higher Wages
Despite consistent tightening of
labor market slack, wage growth has
been remarkably restrained during
the long recovery. One explanation
is that unemployment rates haven’t
fallen far enough. But as the economy
gains more steam and the unemployment rate drops further, the traditional
responsiveness of wages—illustrated by
the Phillips curve relationship—should
reappear and begin to spur wage
growth.
Tighter monetary policy may be
warranted if and when wage growth
picks up and starts feeding into
consumer prices. A steeper Phillips
curve and more flexible wages in Texas
relative to the nation suggest that, all
else equal, the state will experience a
smaller increase in labor market slack
when interest rates rise.

Chart

Kumar is a senior research economist in
the Research Department at the Federal
Reserve Bank of Dallas.
Notes
The inverse relationship between unemployment and
wages was originally found in “The Relation Between
Unemployment and the Rate of Change of Money Wage
Rates in the United Kingdom, 1861–1957,” by A.W.
Phillips, Economica, vol. 25, no. 100, 1958, pp. 283–99.
2
Closely linked to the Phillips curve is the concept of the
natural rate of unemployment—a jobless rate consistent
with stable inflation. 3 Hourly wages were measured
following the procedure in “Creating a Consistent Hourly
Wage Series from the Current Population Survey’s
Outgoing Rotation Group, 1979-2002,” by John Schmitt,
Center for Economic and Policy Research, 2003, p. 64.
4
See “Business Cycle Phases in U.S. States,” by Michael
T. Owyang, Jeremy Piger and Howard J. Wall, Review
of Economics and Statistics, vol. 87, no. 4, 2005, pp.
604–16.
5
See “The Differential Regional Effects of Monetary
Policy,” by Gerald Carlino and Robert DeFina, Review
of Economics and Statistics, vol. 80, no. 4, 1998, pp.
572–87.
6
For details, see “Implicit Contracts and Underemployment
Equilibria,” by Costas Azariadis, Journal of Political
Economy, vol. 83, no. 6, 1975, pp. 1,183–202.
7
For details, see “Efficiency Wage Models of
Unemployment,” by Janet L. Yellen, American Economic
Review, vol. 74, no. 2, 1984, pp. 200–05.
8
For more on wage rigidity explanations, see
Fundamentals of Labor Economics, by Thomas Hyclak,
Geraint Johnes and Robert Thornton, Mason, Ohio: SouthWestern/Cengage Learning, 2012.
9
See “The Determinants of Real Wage Flexibility,” by
Geraint Johnes and Thomas J. Hyclak, Labour Economics,
vol. 2, no. 2, 1995, pp. 175–85.
1

See “Hysteresis and the European Unemployment
Problem,” by Oliver J. Blanchard and Lawrence H.
Summers, in NBER Macroeconomics Annual 1986,
Volume 1, ed. Stanley Fischer, Cambridge, Mass.: MIT
Press, 1986, pp. 15–90.
11
See “Would a Texas Central Bank Set Rate Higher?” by
Janet Koech and Mark A. Wynne, Federal Reserve Bank of
Dallas Southwest Economy, no. 2, 2014, p. 15.
12
See “A Closer Look at the Phillips Curve Using State
Level Data,” by Anil Kumar and Pia Orrenius, Federal
Reserve Bank of Dallas Working Paper no. 1409, May
2014.
13
See “Why is Wage Growth So Slow?” by Mary C. Daly
and Bart Hobijn, Federal Reserve Bank of San Francisco
Economic Letter, no. 1, 2015.
14
See “The Path of Wage Growth and Unemployment,” by
Mary C. Daly, Bart Hobijn and Timothy Ni, Federal Reserve
Bank of San Francisco Economic Letter, no. 20, 2013.
15
See “U.S. Regional Business Cycles and the Natural Rate
of Unemployment,” by Howard J. Wall and Gylfi Zoega,
Federal Reserve Bank of St. Louis Review, vol. 86, no. 1,
2004, pp. 23–31.
16
Although real wage growth predicted by the model
appears low, data from the Bureau of Labor Statistics show
that average hourly earnings grew just 0.3 percent in 2013.
17
Variability across states is measured using the coefficient
of variation, which equals the standard deviation of the
unemployment rate across states divided by its mean.
10

Dispersion in Unemployment Rates Falls, Stabilizes

Unemployment rate variability across states*

.35

.34

.30

.25
.23
.20
.19
.15
1980

1990

2000

2010

2020

*Standard deviation of the unemployment rate across states divided by its mean.
SOURCES: Bureau of Labor Statistics’ Current Population Survey; Census Bureau; author’s calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

ON THE RECORD
A Conversation with Mark A. Wynne

Greece’s Fiscal Woes
Among Issues Hobbling
Euro Zone Rebound
While the U.S. has emerged from the global economic downturn, the
path for the euro zone has proven bumpier. Senior economist Mark A.
Wynne, vice president and director of the Globalization and Monetary
Policy Institute in the Research Department at the Federal Reserve
Bank of Dallas, explores the reasons and outlook.
Q. Why has the euro zone’s economic recovery from the global financial
crisis lagged behind the U.S. recovery? Is the situation improving?
The euro area suffered two big
shocks in recent years: first, the shock
associated with the global financial
crisis that was centered in the United
States, and second, a euro-area-specific
shock due to problems in a number of
geographically peripheral countries
(Cyprus, Portugal, Ireland, Greece and
Spain).
Economic activity in the euro zone
significantly contracted between first
quarter 2008 and second quarter 2009.
After the economy resumed growing, it
stalled in early 2011 before it could attain
its precrisis level of economic output.
The second contraction lasted through
early 2013. Although the euro zone
economy has since been in recovery, the
latest estimates show real gross domestic product (GDP) remains below first
quarter 2008 levels.
Some of the hardest-hit countries
are doing better—Ireland, Spain and
Portugal, in particular. Italy has taken
longer to turn around but seems to have
done so this year. Of all the peripheral
countries, Greece has experienced the
biggest collapse. There were signs that it
was beginning to come back, but recent
developments seem to have snuffed out
the fragile recovery.

into difficulty for different reasons. In
Ireland and Spain, public finances were
in very good shape in the run-up to
the financial crisis, but both countries
experienced enormous housing booms
fueled by low interest rates that dwarfed
the boom we experienced in the U.S. In
the cases of the U.S., Ireland and Spain,
loans linked to real estate development
went bad, creating problems in the
banking sector.
In Ireland, the government guaranteed the liabilities of the banking system
and nationalized two of the largest banks
in the country. This in turn put public
finances on a dangerous trajectory and
eventually necessitated a bailout from
the European Union and the International Monetary Fund (IMF). A similar
situation arose in Spain, although in that
instance it was the Spanish banking system rather than the Spanish government
that was bailed out.
In Greece, the problems stemmed
from a pattern of public spending and
taxation that was simply unsustainable.
In 2009, Greece ran a government budget deficit equal to more than 15 percent
of its GDP, which is more than five times
the supposed maximum of 3 percent
for euro zone members. The absence
of a formal fiscal or banking union as
concomitants to the monetary union
launched in 1999 complicated dealing
with these problems.

Q. What contributed to the European sovereign debt crisis?

Q. Has Europe’s malaise harmed the
U.S. economy?

In 2011, different countries got

It probably contributed to the

sluggish pace of recovery in the United
States by reducing demand for U.S.
exports. For all its problems, Europe
remains one of the more important and
wealthier economic regions in the world
and, as such, is an important trading and
investment partner of the United States.
In addition to slow growth impacting
demand, financial volatility in the euro
area can lead to capital flows out of the
area to “currency safe haven” countries
such as Switzerland and the United
States. This tends to increase the value
of our currency, making it harder for our
exporters to compete globally.

Q. Is there anything the U.S. can do
to aid the euro zone recovery?
Not really. The Europeans need
to figure out for themselves what form
they want their monetary union to take.
In its original conception, there was to
be no banking or fiscal union and no
bailouts. Potential members had to meet
specific criteria to join, and once in, had
to adhere to certain rules. For a variety
of essentially political reasons, the rules
were bent to admit some countries and
then subsequently broken by others.

Q. Is recent improvement in Europe
the result of quantitative easing by
the European Central Bank (ECB)
earlier this year or have there been
structural changes?
I think quantitative easing—the
purchase of bonds and addition of euros
to the monetary supply—has helped.
But perhaps more important was the
promise in mid-2012 by ECB President
Mario Draghi to “do whatever it takes”
to preserve the single currency, and
the subsequent announcement of the
so-called Outright Monetary Transactions—a plan to buy sovereign debt of
euro zone countries under specific circumstances—to back up that promise.
There have also been structural
reforms. For example, in Spain it is now
easier to register new companies. Similar
steps have been taken in Portugal and
Greece. But the payoff from structural
reforms takes time. In the short run, such
reforms may even temporarily depress
economic activity as capital and labor are
reallocated to more productive activities.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

all the problems that the euro has
}For
experienced in recent years, it has nevertheless
brought real benefits.

Q. The euro area includes the richest nations in the world, yet the
challenges seem unending. Could it
be that adopting a common currency—the euro—was a bad idea?
I think it is fair to say that most
North American economists (and a good
number of European economists as well)
felt that the idea of such a diverse group
of countries sharing a common currency was doomed to fail at some point
because the countries in question did
not constitute what economists refer to
as an “optimum currency area.” This is
an idea that is more than a half-century
old and originated with Robert Mundell
(who won a Nobel Prize for his work)
asking the question: When is it a good
idea (from an economic perspective) to
stipulate the use of a currency within a
geographic boundary that coincides with
a political boundary?
In North America, an east-to-west
border determines where U.S. and
Canadian dollars are used. But one could
just as easily imagine drawing a northto-south line that would demarcate currency zones independent of the political
boundary. Under what conditions might
it make more sense for the eastern U.S.
and eastern Canada to share a common
currency, and for the western U.S. and
western Canada to share another currency?
As economists began thinking about
these issues, they highlighted a number
of considerations key to a successful
monetary union between a group of
sovereign nations—things such as the
degree of integration between the nations, mobility of labor and capital, the
similarities and differences in the structure of their economies and the flexibility
of wages and prices.

On the economic side, advocates of
the single currency pointed to the fact
that a single internal market within the
U.S. functions a lot better because all 50
states use the dollar. One of the longterm economic goals of the European
project was to create a common single
market in Western Europe that would be
as integrated and seamless as in the U.S.
But there was always an important political dimension, an idea that by sharing a
common currency, a shared European
identity would emerge independent of
national identities, thereby advancing
the goal of “an ever-closer union” among
the peoples of Europe.
The architects of the treaty that provides the legal and institutional basis for
the euro were well aware of the concerns
expressed by many economists, and
to that end they specified a set of rules
governing which countries could join the
single currency and how those countries
were to behave once they were in. Unfortunately, these rules were not rigorously
enforced, and this contributed to the
recent crisis. Skeptics also pointed to the
absence of a fiscal union to accompany
the monetary union as a key design flaw.
The argument was that the U.S. monetary
union works so well in part because of
the insurance provided to individual
states by the federal government.
For example, when Texas experienced the oil bust in the 1980s, the
adjustment here was eased by the fact
that we paid in less in taxes to the federal
government as economic activity contracted, and we received more in the way
of benefits. In addition, the burden of
bailing out depositors in the many financial institutions that failed was shared
among all 50 states rather than falling
on just Texas. There is no comparable
arrangement in Europe. Another factor
that makes the U.S. monetary union
work well is the high degree of labor
mobility between individual U.S. states,
facilitated in no small part by the fact that
we all speak the same language. Legally,
there are no barriers to labor mobility in

Europe, but informal barriers due to differences in language and culture remain.
But what the crisis really revealed
was that the absence of a banking union
to accompany the monetary union was
an even bigger design flaw and, surprisingly enough, not one that many of the
skeptics seemed to have anticipated. For
all the problems that the euro has experienced in recent years, it has nevertheless brought real benefits, and even in
some of the hardest-hit crisis countries,
support for the shared currency remains
relatively high.

Q. What is the outlook for Greece?
The Great Depression was the most
traumatic event in our nation’s history. At
the Depression’s depth, the unemployment rate approached one-quarter of the
U.S. labor force. Greece is experiencing a
comparable economic trauma.
Earlier, I mentioned that the
architects of the monetary union had
established a set of rules for euro
membership. One of these is a limit on
government deficits of no more than 3
percent of GDP. Greece did not get to
join the euro in 1999 when the project
was launched because it failed to meet
this condition and various other criteria
for membership. But it was admitted in
2001. Just three years later, Greece’s public accounts were revised to show deficits
exceeding the 3 percent limit every year
from 2000 to 2003. But the proximate
cause of the crisis was the revelation in
late 2009 following a general election that
the deficit for that year would not be the
3.7 percent of GDP originally reported
but instead would be closer to 12.5 percent of GDP—more than four times the
euro-area treaty limit. Greece has been in
a state of crisis since then.
Is there a scenario in which Greece
leaves the euro? Yes. But it would do
little to fix the deeper problems Greece
is wrestling with and could prove to be
destabilizing for the rest of the euro area
and for the global economy.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

Mexico’s Four Economies Reflect
Regional Differences, Challenges
By Jesus Cañas and Emily Gutierrez

}

ABSTRACT: The economic
potential of Mexico’s four
regions is defined by their
industrial makeup, income per
capita and how much of the
labor force operates outside
the formal economy. Recent
government reforms could
promote growth and reduce
regional inequality.

exico is a country of contrasts,
its geography varying from
deserts to jungles, mountains
to beaches. Such differences
extend to the economic characteristics
of Mexico’s four regions: the manufacturing north, the agrarian north-central, the service-based central and the
energy-producing south (Chart 1).
Such economic specialization has
contributed to significantly different
levels of development—evident in persistent and often worsening disparities
in standards of living.1

Regional Diversity, Growth
Mexico’s affluent north is characterized by a large manufacturing base,
which sharply diverges from the poverty-stricken south, a hub of energy activity. The central region benefits from the
sprawling reach of Mexico City, one of
the world’s largest metropolitan areas

Chart

and the heart of the Mexican economy,
while the agriculturally driven northcentral zone makes a much smaller
economic contribution.2
Each region’s industrial base helps
explain these regional income and
growth disparities. Researchers use
location quotients (LQs) as a means of
identifying dominant or prominent industries in an area.3 An LQ is a region’s
share of output in a specific industry divided by the national share of
output in that same industry. When an
industry’s LQ exceeds 1, the industry
accounts for a larger portion of output
in the region than in the nation as a
whole; the larger the LQ, the greater
the industry’s importance.
For instance, agriculture in the
central region has an LQ of 0.5, indicating the industry’s share of gross
domestic product (GDP) is half the
national average. The north-central

Mexico’s Four Economic Regions Are Diverse

North
GDP share: 22.1%
Pop. share: 17.9%

North-central
GDP share: 18.2%
Pop. share: 21.1%

Central
GDP share: 38.8%
Pop. share: 37.9%

South
GDP share: 20.9%
Pop. share: 23.0%

NOTE: 2013 values were used to create shares.
SOURCES: Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography); Consejo
Nacional de Población (National Council of Population).

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

region has an LQ of 2.2 for the same
sector, indicating a GDP share more
than twice the national average. Taken
together, LQs highlight what makes a
region unique.
LQ analysis of the northern economy shows a high concentration in manufacturing, which isn’t surprising given
that it’s home to almost 3,000 manufacturing plants (Table 1). The north
has capitalized on the manufacturing
symbiosis between Mexico and the U.S.,
posting the highest regional economic
growth between 2003 and 2013.
The northern region—particularly
the states of Chihuahua and Coahuila—
boasts a world-class automotive industry that includes General Motors and
Ford operations. It is also home to a
cluster of auto parts manufacturers that
have made Mexico the No. 1 supplier
of parts to the U.S. market since 2001.
Additionally, the north has a highly
competitive electronics manufacturing industry in Baja California and has
solidified its aerospace manufacturing
sector in Sonora and Chihuahua.4
The north-central region specializes in agriculture—Sinaloa is a major
tomato producer, and about 80 percent
of the avocados consumed globally are
produced in Michoacán. This region is

Table

also the transportation hub of Mexico.
Tourism, as reflected in a high LQ for
leisure and hospitality, is an important
economic engine in the region, driven
by attractions in Jalisco (Puerto Vallarta) and Baja California Sur (Cabo
San Lucas). The north-central region
grew at about the same rate as the central region over the 10-year period.
Central Mexico, which includes
Mexico City, also performed well, with
its GDP growing on average 2.7 percent
annually in inflation-adjusted terms
over the period. As the high LQs across
most of the service industries suggest, the central region is the country’s
financial center and provides business
services to the domestic market and to
international companies and investors.
This densely populated region is
home to more than 45 million people
within a 200-mile radius of the nation’s capital. It has first-class road and
rail networks and is only a few hours’
drive from major ports on the Pacific
and Gulf of Mexico. In addition, major
transnationals such as Nestlé and Telmex and strategic government-owned
enterprises like Pemex have major
offices in this region.
The south is the slowest-growing
region, expanding 1 percent annually

densely populated
}This
central region is home
to more than 45 million
people within a 200-mile
radius of the nation’s
capital. It has first-class
road and rail networks
and is only a few hours’
drive from major ports
on the Pacific and Gulf
of Mexico.

Industry Location Quotients by Region
North

Annual average growth rate (2003–13)

North-central

Central

South

3.0

2.7

1.0

Agriculture

0.9

2.2

0.5

1.0

Mining

0.6

0.4

0.1

3.7

Construction

1.1

1.2

0.8

1.2

Manufacturing

1.4

1.0

0.6

Trade, transportation & utilities

1.0

1.1

0.8

Information

0.9

0.8

1.5

0.5

Financial activities

0.9

1.0

1.2

0.8

Professional & business services

0.9

0.5

1.6

0.5

Education & health services

0.9

1.1

0.9

Leisure & hospitality

0.7

1.2

0.9

1.3

Other services

0.8

0.9

1.3

0.8

Government

0.8

1.0

1.2

0.8

Goods-producing industries

Service-providing industries

NOTE: Location quotients greater (less) than 1 represents a gross domestic product concentration higher (lower) than the
national average in a given region.
SOURCES: Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography); authors’ calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

south heavily relies
}The
on energy-related
activity, with most of
it concentrated in two
states: Campeche and
Tabasco. A big factor
behind the south’s
anemic growth is the
steady decline in oil
production since 2004.

in real (inflation-adjusted) terms over
the 10-year period. The south heavily
relies on energy-related activity, with
most of it concentrated in two states:
Campeche and Tabasco. A big factor
behind the south’s anemic growth is
the steady decline in oil production
since 2004. Moreover, the Mexican
energy industry is wholly controlled
by Pemex—the national monopoly—
whose energy revenues flow to the
federal government and largely bypass
the local area. That is in contrast to
energy-dependent regions in the U.S.,
which benefit directly from oil and gas
production.
Although Mexico has implemented
initiatives to overhaul its oil industry,
Pemex continues to control operations,
beginning with exploration and extending to transport, refining and retail sales.5
Additionally, the south is the region with the lowest levels of education
and highest concentration of poverty,
labor informality and social unrest. 6

Regional Income Gaps
The north and central regions are
diverging from the north-central and
south, recent data show. The contrast
with the southern region is even more
pronounced when discounting the oilrich states (Chart 2).
The uneven regional growth rates
go back many years and have allowed
the north and central regions to grow

Chart

richer relative to the rest of the country
(Table 2).
GDP per capita was $12,627 in
the north and $10,415 in the central
region in 2013. Output per capita in
the north-central ($8,777) and south
($8,573), meanwhile, trailed the nation
as a whole. When the oil-producing
states of Campeche and Tabasco are
excluded from the south, output is significantly lower, $6,583 per capita.
Table 2 also shows labor informality and poverty rates by region. Generally, where labor informality is found,
poverty abounds. The southern region,
where close to 70 percent of the labor
force works in the informal sector, is
also the poorest area of Mexico.

Role of Reforms
Recent labor, energy, financial and
fiscal reforms could contribute to a
reduction in regional inequality.
Federal labor law includes increased flexibility in hiring and payment of wages that could help workers
move from informal to formal employment. Energy reform aims to introduce
competition in refined products and
electricity markets, allowing private investment to flow into the sector, particularly into oil and gas exploration. The
reform also will allow private participation in the sale, transport and distribution of energy products. Changes that
allow more competition and foreign

Income Divergence in Mexico Remains the Norm

Real GDP per capita (thousands of pesos)

160
140

North

120

Central

100

South
North-central

80
60
40
2003

South without Campeche and Tabasco

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

SOURCES: Instituto Nacional de Estadistica y Geografia (National Institute of Statistics and Geography); authors’
calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

Table

Labor Informality Tied to Poverty in Mexico
Per capita GDP
(dollars)

Informal labor
(% of labor force)

Poverty rate
(% of total population)

Total Mexico

10,193

North

12,627

Central

10,415

South

8,573

South
without Campeche and Tabasco

6,583

North-central

8,777

NOTE: Data are from 2013.
SOURCES: Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography); Consejo Nacional
de Población (National Council of Population); Consejo Nacional de Evaluación de la Política de Desarrollo Social (National
Council for the Evaluation of Social Development Policies).

investment could spark regional growth
in energy-dependent areas similar to
that seen in recent years in Texas regions such as the Eagle Ford Shale.7
Comprehensive reform of the
financial sector includes improving
small-business access to the financial
system and increasing credit availability. Finally, fiscal reform designed to
increase the tax base could accelerate
government revenue diversification,
allowing public investment to flow into
needed areas. However, some evidence
suggests that existing regional public
investment has gone to “pork barrel”
projects—those satisfying a political
debt—rather than to redistribution or
efforts to boost regional growth.8

Jesus Cañas, Federal Reserve Bank of Dallas Southwest
Economy, second quarter, 2014.
6
More specifically, informal labor is defined as private
sector workers who are not reported to the government and
thus do not pay employment taxes or receive governmentmandated benefits and pensions.
7
For more information, see “Oil Boom in Eagle Ford Shale
Brings New Wealth to South Texas,” by Robert W. Gilmer,
Raúl Hernandez and Keith R. Phillips, Federal Reserve
Bank of Dallas Southwest Economy, second quarter, 2012.
8
For more information, see “Political Competition and
Pork Barrel Politics in the Allocation of Public Investment
in Mexico,” by Joan Costa-i-Font, Eduardo RodriguezOreggia and Darío Luna Plá, Public Choice, vol. 116, nos.
1-2, 2003, pp. 185–204.

The central region benefits from
greater diversification because of its
access to bigger and wealthier areas of
the country.
Economic expansion in the south,
with its high poverty levels and labor
informality, will continue to lag behind
the nation. However, recent labor,
energy, financial and fiscal reforms
could help close the gap in the medium
to long term by increasing investment
and labor mobility.
Cañas is a business economist and
Gutierrez is a research analyst in the
Research Department of the Federal
Reserve Bank of Dallas.
Notes

Looking Forward

For purposes of this analysis, Mexico’s 32 states are
divided into: north (Baja California, Chihuahua, Coahuila,
Nuevo León, Sonora and Tamaulipas); north-central
(Aguascalientes, Baja California Sur, Colima, Durango,
Jalisco, Michoacán, Nayarit, San Luis Potosí, Sinaloa and
Zacatecas); central (Distrito Federal, Estado de México,
Guanajuato, Hidalgo, Morelos, Puebla, Querétaro and
Tlaxcala); south (Campeche, Chiapas, Guerrero, Oaxaca,
Quintana Roo, Tabasco, Veracruz and Yucatán).
2
There is no official regional classification system in
Mexico’s national statistics. The grouping of states used
here is based on Banco de México’s regional economic
report series.
3
Banco de México’s criteria for the grouping of the states
and the location quotient (LQ) technique were followed to
determine the economic base of each region. Output was
aggregated by industry and region to obtain a regional
numerator for the LQ calculation.
4
See “The Maquiladora’s Changing Geography,” by Jesus
Cañas and Robert W. Gilmer, Federal Reserve Bank of
Dallas Southwest Economy, second quarter, 2009.
5
See “‘Reforma Energética’: Mexico Takes First Steps
to Overhaul Oil Industry,” by Michael D. Plante and
1

Regional inequality continues to
haunt Mexico. The dynamic north and
central regions contrast with the lackluster north-central region and dismally
performing south. Economic growth
over the past decade, mainly due to
external factors such as high oil prices
and strong global demand, has proven
insufficient to mitigate inequality.
As long as the U.S. economy
continues expanding, it’s likely the
north will grow faster than the rest of
the country. Mexico manufacturing
is highly dependent on U.S. demand,
with 80 percent of exports going to the
U.S. market. Growth in the north-central and central regions will continue
to be more closely tied to the national
average because both predominantly
serve the domestic market.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

NOTEWORTHY
EDUCATION: Texas Near Bottom in Spending for Public Schools

exas ranked 45th nationally in kindergarten through 12th grade public education spending per
pupil, according to the recently released 2013 U.S. Census Bureau Public Education Finances report. The state spent $8,299 per student compared with a national average of $10,700. While greater
expenditures do not guarantee better outcomes, schools with more resources tend to have students with
higher educational attainment.
Following the recession, Texas expenditures per student declined in 2011 and 2012 from a high of
$8,746 in 2010. State spending picked up in 2013, though Texas’ rank has slipped steadily since 2010. Relative to per-pupil spending in other states and the District of Columbia, California ranked 36th ($9,220),
while New York was first ($19,818). There are about 5 million students in Texas public schools—the
second-largest enrollment after California. Of the Texas total, 13 percent are black, 30 percent are white
and 51 percent are Hispanic.
Texas public education received 11 percent of its funding from the federal government, 39 percent
from the state and 50 percent from local resources in 2013. Nationally, including Texas, about 60 percent
of per-pupil expenditures was spent on instruction and roughly 35 percent on support services. The remaining 5 percent in Texas funded items such as textbooks, transportation and employee retirement.
—Emily Gutierrez

TAXATION: Dallas County Property Values Rise 7.5 Percent in 2015

he Dallas Central Appraisal District—Texas’ second-largest appraisal district by market value (behind Harris County)—reported a 7.5 percent increase in the taxable value of property, totaling $188
billion this year. This follows a 6.7 percent increase in 2014.
While residential makes up the largest of the three categories of Dallas County property values (45
percent), commercial property rose the most in 2015, accounting for almost half of the overall increase.
Property taxes, typically accounting for more than 60 percent of Dallas County government revenues, are
expected to rise 5.3 percent in the current fiscal year. A steeper increase is likely next year.
Of the school districts located entirely within Dallas County, Sunnyvale Independent School District
recorded the largest percentage increase in property values—10.8 percent—while the Dallas Independent School District had the highest total property value.
The appraisal district determines the value of properties—preliminary values are released in May
and the final valuations in July—located within Dallas County; taxes are collected by the Dallas County
tax assessor and then distributed to cities, school districts and other local jurisdictions. Proposals for how
to spend the additional dollars abound and include more funds for schools, hospitals, jails and animal
control.
—Sarah Greer

INCOME: Obama Plan to Give More Managers Overtime Pay

orkers sometimes find that becoming a manager means a little extra pay and many more hours.
An Obama administration plan may change that. It would almost double the Fair Labor Standards
Act minimum weekly salary of $455, allowing management employees to become salaried and excluded from overtime. Retail and hospitality industries generally pay managers less and have them work
longer hours than many other businesses and, thus, the change may affect them most.
The act, approved in 1935, exempts “executive, administrative and professional” employees from
overtime—generally 1.5 times the hourly wage—provided their pay exceeds the threshold.
The president’s plan, which requires U.S. Labor Department rulemaking after a public comment period, comes at a time of relatively low unemployment. The retail and accommodation and food services
sectors account for more than a quarter of employment in San Antonio, compared with just over a fifth
in Dallas and Houston, according to data compiled by the Federal Reserve Bank of Dallas. Overall, wage
rates in Texas tend to trail nationwide averages.
The U.S. Chamber of Commerce predicts that employers may respond by promoting fewer managers
and reducing hours worked. Some firms could use more part-time workers. However, in a still-tight labor
market, those options may be limited.
—Michael Weiss

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

SPOTLIGHT

Diversified Houston Spared Recession … So Far
By Jesse Thompson

il and gas exploration, production and services firms
nationwide drastically cut
spending and employment
after oil prices plunged 40 percent in
the second half of 2014. Thousands
subsequently lost their jobs as the
U.S. rig count leveled off at 861 in June
2015—1,064 fewer than in October
2014. The industry’s capital expenditures, typically for equipment and
facilities, have been cut.
In Houston, headquarters of the
energy industry, manufacturing was
the first sector to respond, losing 15,000
jobs by June (after peaking at 261,300 in
December)—the largest decline since
the Great Recession. Fabricated metals
was particularly hard hit, with employment falling at an annual rate of 16.4
percent in the first half of 2015; support
activities for mining slid at an annualized 17 percent during the period.
Still, negative spillover to the rest
of the Houston-area economy has appeared only slowly. Despite significant
losses in manufacturing and oilfield
services, total jobs in Houston only
declined by an annualized rate of 0.6
percent in the first half of 2015. While
not a large reduction, this represents
a reversal from the 4.1 percent pace of
job growth last year.
Three factors may be limiting the
impact of the exploration and production downturn.
First, Houston is the center of the
nation’s refining and petrochemical
industries, which benefit from low oil
and gas prices. Petrochemical production is booming. Construction of new
facilities by firms such as Chevron
Phillips Chemical and Dow Chemical,
and the thousands of workers needed
for the build-out, will help prop up employment until at least 2017. Second,
conservative bank lending practices,
increased hedging against oil price declines and a low “opportunity cost” for
investing in the energy industry have
arguably kept the rate of bankruptcies

Diversification Makes Houston Industry Mix More Like U.S.
Index (0 to 1)*

.95
.94
.93
.92
.91
.9
.89
.88

’76 ’78 ’80 ’82 ’84 ’86 ’88 ’90 ’92 ’94 ’96 ’98 ’00 ’02 ’04 ’06 ’08 ’10 ’12 ’14

*Higher values mean regional industry mix resembles national industry mix.
NOTE: Industry definitions changed in 1990.
SOURCES: Bureau of Labor Statistics; seasonal and other adjustments by the Federal Reserve Bank of Dallas.

and mergers and acquisitions relatively
low among exploration and production
companies. Third, the region’s industry
mix has become more diversified.
An index measuring how Houston’s
industry composition is similar to the
nation’s shows that from 1982 to 2004,
Houston became more like the U.S.
(see chart). Of particular importance,
professional and business services and
health services, as a share of Houston
employment, grew 6.5 percentage
points to 26 percent from 1990 to 2014.
Above-average wages in these
industries helped real (inflationadjusted) per capita income grow 63
percent locally between 1990 and 2013,
compared with a 43 percent increase
nationally. The housing boom of the
mid-2000s boosted construction’s share
of the region’s economy. The proportion
of manufacturing and wholesale trade
employment also grew during the period, and the shale revolution allowed
the energy industry to expand after
the Great Recession. Mining’s share of
Houston employment—which tumbled
from a peak of 7 percent in 1982 to a
low of 2.6 percent in 2000—stood at 3.8
percent in 2014. Energy sector growth
spurred a flurry of commercial and
residential real estate development as

the energy sector consolidated into the
region.
With so much recent economic development tied to oil and gas, some have
questioned how diversified Houston has
become, especially since exploration and
production firms outsource many legal,
professional and financial services.
In an econometric model that incorporates real U.S. gross domestic product
(GDP), exploration and production firms’
real capital expenditures and Houston
employment from 1991 through 2014,
a 30 percent decline in exploration and
production capital expenditures—such
as occurred in first quarter 2015—yields
a 1 percent drop in Houston employment
(about 30,000 jobs) by year end, holding
all else constant.
That is one-quarter of the 3.9 percent
employment loss that would have occurred in the pre-1990 era. The model
also suggests that Houston’s reaction to
U.S. GDP growth was 68 percent larger
post-1990—meaning that even serious
oil industry declines can now be mostly
offset by economic growth elsewhere.
On balance, these changes indicate that
Houston’s oilfield connection, while
strong, has weakened. By becoming more
like the U.S. economy, the region can better weather oil market volatility.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

Texas Maintains Top Exporter Standing
While Its Trade Remains Concentrated
By Janet Koech and Mark A. Wynne

}

ABSTRACT: While Texas has
become the nation’s topexporting state, benefiting
from trade of intermediate
goods to Mexico and a global
presence as an energy hub,
its export activity remains
concentrated relative to the
U.S. and other states.

exas exports more goods than
any other state and is a big recipient of foreign investment.
The state is the third-most
globalized in the U.S. based
on foreign-owned companies’ employment and export-based manufacturing,
according to globalization scorecards.1
Texas has become increasingly
integrated with the rest of the world
and dependent on foreign markets for
its economic growth. Exports equal 17
percent of Texas’ total output, almost
twice the nation’s average of 9 percent.
Manufactured goods exports supported
more than 1 million jobs in Texas in
2014—about 17 percent of all exportrelated jobs in the nation.2
With greater international linkages
comes greater international exposure.
Increased interconnectedness of world
economies means that the effects of
economic booms as well as slowdowns
spread across geographical boundaries.
Dependence on a few export partners
and products can make exports and exporting states sensitive to developments
in the recipient countries. A state with
a diverse range of export products and
export destinations is typically more
likely to withstand shocks to particular
industries or countries.
Diversification of Texas’ trade with
the rest of the world may be viewed
along two dimensions. The diversity
of Texas export destinations provides
one guide. Do most of our exports go
mainly to our neighbor to the south or
do we trade with a range of countries?
The composition of the state’s export
basket is another measure. Are Texas’
exports comprised primarily of energy
and related products, or are a range of
products involved? 3

State’s Largest Trade Partners
Texas, after surpassing California
as the top exporting state in 2002, sold
$288 billion worth of goods overseas in
2014.4 From 2000 to 2014, the state’s real
(inflation-adjusted) exports increased

at an average annual rate of about 7
percent, faster than the nation’s annual
average of 4 percent.
Perhaps not surprisingly given Texas’ geographic proximity to Mexico and
preferences under the North American
Free Trade Agreement, the state heavily
exports south of the border. Mexico
accounts for about 36 percent of Texas’
foreign sales (Chart 1). Much of this
trade involves intermediate goods; U.S.
companies have plants in Mexico that
manufacture and assemble products
from the intermediate inputs for reexport to the U.S. By one estimate, the U.S.
content in imports from Mexico is 40
percent.5
Texas’ top three foreign markets
accounted for more than half its total
exports in 2014 compared with 42
percent for the U.S. and 35 percent for
California. Over the past decade, however, Texas expanded its sales abroad to
new destinations including to rapidly
growing emerging market economies.
Real exports to China increased 17 percent on average in 2000–14, while those
to Brazil expanded at an average annual
rate of about 14 percent during that
period. Emerging markets’ demand for
petroleum and coal products has been a
boon for Texas exports.

Trade Activity Index
For the purpose of comparing
Texas’ trade patterns with other states
and the U.S. as a whole, it is useful to
summarize them in a single number.
The Herfindahl index is a widely used
measure of industry concentration
and is calculated as the sum of the
squares of export shares of each country
constituting a state’s total exports. High
values indicate that a state’s exports are
highly concentrated; low values suggest
that a state exports to a wide variety of
countries.
This measure of the degree of concentration of Texas exports has evolved
over time (Chart 2). Perhaps not
surprisingly, Texas’ exports are highly

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

concentrated, much more so than the
national average or No. 2 exporter, California. The increase in diversification in
the 2000–07 period is notable. Texas’ diversification by export destination index
decreased to 0.13 in 2007 from 0.23 in
2000 (a decline in the index implies an
increase in diversification). This coincides with the state’s increased exports
to many emerging market economies.
Since 2007, the diversification index has
mostly held steady because of the Texas
boom in shale oil, which led to addi-

Chart

tional exports to Latin American countries. Mexico is the single biggest market
accounting for more than one-fifth of
the state’s petroleum product exports.
Texas ranked 37th among the states
in terms of diversification of trading partners in 2014, compared with
California at No. 6. Florida is the most
diversified state, while North Dakota is
the least diverse. Texas and California
both inched up one spot in the diversification ranking between 1997 and 2014
(Table 1).

ranked 37th
}Texas
among the states
in terms of diversification
of trading partners
in 2014, compared
with California at No.
6. Florida is the most
diversified state, while
North Dakota is the
least diverse.

Mexico Is Texas’ Main Export Destination

Percent

100
90
80
70
60
50
40
30
20
10
0
1998

2000

2002

2004

2006

2008

2010

2012

2014

Rest of the world

Brazil

Netherlands

Canada

Other emerging market economies

South Korea

China

Mexico

SOURCES: World Institute for Strategic Economic Research (WISER) state export data; authors’ calculations.

Chart

Export Destinations More Concentrated in Texas than U.S.

Herfindahl index (higher score = less diversified)

.25
.20
Texas
.15

U.S. average

.10
.05
0

California

1998

2000

2002

2004

2006

2008

2010

2012

2014

NOTE: The average U.S. index is computed as the export-weighted average of all states’ Herfindahl indexes.
SOURCES: World Institute for Strategic Economic Research (WISER) state export data; authors’ calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

Table

growth in
}Strong
emerging economies,
especially in Asia,
generated demand
for energy products,
which now account for
a much larger share of
the state’s exports
than a decade ago.

Most diversified states

Least diversified states

Diversification Rankings by Export Destination

State

2014
Herfindahl
index

2014 total
state exports
(billions of
U.S. dollars)

Florida

0.027

57.30

0.036

23.23

Maryland

0.043

11.97

0.058

5.21

Louisiana

0.043

65.05

0.036

18.73

Georgia

0.046

38.75

0.067

12.95

Massachusetts

0.053

27.22

0.081

16.53

California

0.055

170.72

0.066

99.16

Texas

0.143

287.60

0.161

76.18

New Mexico

0.218

3.71

0.190

1.78

Michigan

0.248

55.64

0.359

32.25

South Dakota

0.249

1.59

0.165

0.52

Maine

0.303

2.73

0.157

1.72

North Dakota

0.631

5.26

0.310

0.78

2014
state
ranking

1997 total
1997
1997
state exports
state Herfindahl
(billions of
ranking
index
U.S. dollars)

NOTE: An increase in the index shows a decrease in export diversification.
SOURCES: World Institute for Strategic Economic Research (WISER) state export data; Haver Analytics; authors’ calculations.

Products Sold
The types of goods Texas exports
provide another measure of diversification. The state’s largest exports are petroleum and coal products (19 percent),
computers and electronics (17 percent)
and chemicals (16 percent).6 Petroleum
products and chemicals use oil and gas
as inputs and are highly sensitive to oil
price changes.
After the 1980s oil price collapse
and ensuing recession, Texas diversified its economy away from oil and gas,
marked by the rise of the high-tech and
telecommunications industries in the
1990s. By 2000, computers and electronics constituted 29 percent of Texas exports and petroleum and coal products
had fallen to 4 percent. Since then, as a
result of oil prices that were rising and
technological innovations in drilling,
the energy sector reappeared as a major
driver of Texas growth. Moreover, strong
growth in emerging economies, especially in Asia, generated demand for
energy products, which now account for
a much larger share of the state’s exports
than a decade ago.
The Herfindahl index can also
quantify the degree of concentration of

Texas’ exports in particular products.
Texas’ exports are less concentrated in
particular product categories than the
nation’s average (Chart 3). The degree
of product concentration has remained
remarkably constant over time. While
California’s exports are more diversified
in terms of destinations, they are more
concentrated in terms of products, at
least until recently. In 1997, computer
and electronic products accounted for
48 percent of California’s total exports.
In 2014, that share had fallen to 25 percent, and the combined export shares of
the state’s top three products in that year
amounted to 45 percent of total exports.
Virginia and Pennsylvania are
the most diversified states in terms of
product composition of their export
baskets, while Wyoming and Vermont,
which admittedly have very few exports,
are the most specialized (Table 2).
While Washington state exported more
than $90 billion worth of merchandise
in 2014, exports of transportation and
equipment (primarily aircraft, engines
and parts assembled by Boeing) accounted for 57 percent of total exports,
making its export mix one of the nation’s
most concentrated. Texas ranked 18th

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

Chart

data suggest a small but negative correlation between the importance of exports
to a state’s economy and the level of
product diversification.7 Exports tend to
be less diversified for those states where
foreign sales account for a larger share of
state output.
Research shows that Texas’ comparative advantage in energy-related
industries has increased in recent years,
which is consistent with the shale oil and
gas boom that dominated state economic growth. The state has also increased
competitiveness in heavy machinery and
transportation equipment industries.8
The state’s comparative advantage in
select industries might explain why the
products Texas exports have remained
fairly concentrated.

Export Products Less Concentrated in Texas than U.S.

Herfindahl index (higher score = less diversified)

.25
California

.20
.15

U.S. average
.10

Texas

.05
0
1998

2000

2002

2004

2006

2008

2010

2012

2014

NOTE: The average U.S. index is computed as the export-weighted average of all states’ Herfindahl indexes.

Texas Trade Arrangements

SOURCES: World Institute for Strategic Economic Research (WISER) state export data; authors’ calculations.

Table

Most diversified states

Least diversified states

Diversification Rankings by Export Products
2014
state
ranking

State

2014
Herfindahl
index

2014 total
state exports
(billions of
U.S. dollars)

1997 total
1997
1997
state exports
state Herfindahl
(billions of
ranking
index
U.S. dollars)

Virginia

0.043

19.04

0.081

12.76

Pennsylvania

0.048

40.09

0.064

16.07

North Carolina

0.050

31.14

0.041

16.40

Maine

0.057

2.73

0.096

1.72

Illinois

0.058

67.85

0.102

26.45

California

0.071

170.72

0.207

99.16

Texas

0.087

287.60

0.094

76.18

New Mexico

0.277

3.71

0.672

1.78

Washington

0.305

90.48

0.333

32.75

Alaska

0.348

5.15

0.158

2.72

Wyoming

0.363

1.75

0.753

0.56

Vermont

0.410

3.64

0.658

3.81

in the country, performing better than
most U.S. states.

Diversification or Specialization
Lack of diversification of export
products and destinations can expose a
state to increased export earnings volatility. Plunging oil prices, for example, have

contributed to a 10 percent decline in
Texas exports over the last year.
Export volatility can be mitigated by
expanding the variety of products and
countries. However, there are also gains
from specializing in products in which
a state has comparative advantage and
can produce most efficiently. Indeed, the

Rapid internationalization of the
U.S. economy has spread unevenly
across regions and states. Similarly,
diversification of states’ exports varies
across the nation, evolving over time.
While Texas is one of the most globalized
states, it is also one of the least diversified
in terms of with whom it trades.
Mexico is the destination for more
than one-third of the state’s total exports.
Increased mobility for goods, labor and
capital generally entails greater exposure
to global economic pressures and risks.
However, U.S. exports to Mexico are
largely intermediate goods, assembled or
processed into final goods and reimported back to the U.S. for consumption.
Studies show that trade flows associated with such production sharing tend
to be closely related to the economic
activity in the source country of intermediate goods.9 U.S. exports to Mexico
are, therefore, more tied to U.S. demand
than to changes in demand in Mexico,
and Texas exports to Mexico may be
sheltered from economic fluctuations in
Mexico.
Texas export products are more
diversified than the national average but
more concentrated than California and
16 other states. The state largely relies
on exports of chemicals, computers and
electronic products and petroleum and
coal products, which collectively account
for over half of total Texas exports.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2015

(Continued on back page)

Federal Reserve Bank of Dallas
P.O. Box 655906
Dallas, TX 75265-5906

PRSRT STD
U.S. POSTAGE

PAID

DALLAS, TEXAS
PERMIT NO. 151

Texas Maintains Top Exporter Standing
While Its Trade Remains Concentrated
(Continued from page 19)

Koech is an assistant economist and
Wynne is a vice president and associate director of research in the Research
Department at the Federal Reserve Bank
of Dallas.
Notes
See “The 2014 State New Economy Index:
Benchmarking Economic Transformation in the States,”
by Robert D. Atkinson and Adams B. Nager, Information
Technology and Innovation Foundation, June 2014.
2
Data on jobs supported by state exports are produced
by the U.S. Department of Commerce’s International
Trade Administration.
3
For a similar analysis of Texas exports see “Globalizing
Texas: Exports and High-Tech Jobs,” by Anil Kumar,
Federal Reserve Bank of Dallas Southwest Economy,
September/October 2007.
1

DALLASFED

State exports data are from the Census Bureau’s Origin
of Movement exports series. These data measure exports
based on where the goods began their journey of exit
from the United States. The transportation origin of
exports is not always the same as the location where the
goods were produced, so data should be interpreted with
caution.
5
“Give Credit Where Credit is Due: Tracing Value Added
in Global Production Chains,” by Robert Koopman,
William Powers, Zhi Wang, Shang-Jin Wei, National
Bureau of Economic Research, Working Paper no. 16426,
September 2010.
6
Values for second quarter 2015.
7
The correlation between the diversification of state
exports as measured by the Herfindahl index and the
shares of exports in total state output is -0.26.
8
For more details on Texas comparative advantage,
see “Texas Comparative Advantage and Manufacturing
4

Southwest Economy

is published quarterly by the Federal Reserve Bank of
Dallas. The views expressed are those of the authors and
should not be attributed to the Federal Reserve Bank of
Dallas or the Federal Reserve System.
Articles may be reprinted on the condition that the
source is credited and a copy is provided to the Research
Department of the Federal Reserve Bank of Dallas.
Southwest Economy is available on the Dallas Fed
website, www.dallasfed.org.

Federal Reserve Bank of Dallas
2200 N. Pearl St., Dallas, TX 75201

Exports,” by Jesus Cañas, Luis Bernando Torres Ruiz,
and Christina English in Ten-Gallon Economy, Sizing Up
Economic Growth in Texas, Palgrave Macmillan, 2015.
9
“Trade, Production Sharing, and the International
Transmission of Business Cycles,” by Ariel Burstein,
Christopher Kurz, and Linda Tesar, National Bureau of
Economic Research, Working Paper no. 13731, January
2008.

Mine Yücel, Senior Vice President and Director of Research
Pia Orrenius, Keith R. Phillips, Executive Editors
Michael Weiss, Editor
Kathy Thacker, Associate Editor
Dianne Tunnell, Associate Editor
Ellah Piña, Graphic Designer

Full text of The Southwest Economy : Third Quarter 2015

FRASER