Full text of The Southwest Economy : Third Quarter 2013

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 2013

Southwest
Economy

}

Did Home Equity Restrictions Help Keep
Texas Mortgages from Going Underwater?
PLUS
ffGetting Prices Right: Addressing
Mexico’s History of Fuel Subsidies

ffFor Hispanics, Border Wage Gap
Reflects Education, English Divide

ffOn the Record: Outgoing D/FW Airport Chief Cites
International Growth as Key to Future

ffSpotlight: Firms Expect Health Act to Raise Labor Costs

President’s Perspective

escaped the worst
}Texas
of the housing bubble’s
excesses, and it may
provide policymakers
some guidance on how
the nation can avoid
another such calamity.

period of easy mortgage financing and subprime lending was among the catalysts of the
financial crisis of 2007–09, our nation’s worst
downturn since the Great Depression. Relaxed
credit standards contributed to a housing-frenzy-fed
bubble that ultimately burst, leaving our financial system
in disarray and homeowners “underwater” on their
mortgages, owing lenders more than their houses were
worth. The resulting wave of foreclosures—from which
we have only recently begun to emerge—contributed to
a tepid economic recovery.
Texas escaped the worst of the housing bubble’s
excesses, and it may provide policymakers some guidance on how the nation can avoid another such calamity,
economists Anil Kumar and Ed Skelton write in this issue
of Southwest Economy.
Since becoming part of the United States in 1845,
Texas has ascended from frontier state to economic
powerhouse. But the legacy and laurels of our early days
are imbedded in the very fabric of the state. We have
remained fiercely protective of personal property and
opposed to its forced seizure by creditors. Under the
homestead exemption, the Texas Constitution offers hallowed protections from the forced sale of house and land.
Similarly, the Texas Property Code exempts from confiscation some personal property, including “two firearms”
and “two horses, mules, or donkeys and a saddle, blanket,
and bridle for each.”
A quarter century ago, Texans’ strong ethic of personal responsibility was challenged, when loose lending
and risky bets attacked our financial underpinning. From
1987 through 1991, Texas experienced 729 bank failures—including most of the state’s biggest institutions—
representing 38 percent of the nation’s bank closures.
From pain, grew wisdom: Texans took on relatively less
household leverage and laid the foundation for a more
resilient economy. And with lessons learned about the
need for prudent oversight following our state’s oil and
real estate bust of the 1980s, the Texas Legislature wisely
decided to constrain home equity lending to no more
than 80 percent of a home’s value. The limit effectively
restrained the borrowing that figured in much of the
collapse elsewhere. For Texas, the difficult lessons of a
previous era were aptly applied.

Richard W. Fisher
President and CEO
Federal Reserve Bank of Dallas

Did Home Equity Restrictions
Help Keep Texas Mortgages
from Going Underwater?
By Anil Kumar and Edward C. Skelton

relatively stable
}With
house prices in Texas,
the incidence of
underwater mortgages
was a fraction of what
occurred in hard-hit
states.

Chart

he bursting housing bubble
heralded the onset of the Great
Recession in December 2007,
bringing record post-war
joblessness and long-term unemployment. As house prices sharply declined,
so did home equity, an important source
for funding consumer spending. Many
homeowners discovered their mortgages
were “underwater”—they owed more
than their houses were worth. A rising
incidence of such negative equity helped
produce soaring mortgage default rates
and foreclosures.
In Texas, the story was different.
The Federal Housing Finance Agency
house price index fell less than 1
percent in Texas from its peak in 2007
to its trough in 2011, while it plunged
20 percent nationally (Chart 1). With
relatively stable house prices in Texas,
the incidence of underwater mortgages
was a fraction of what occurred in
hard-hit states such as California, Florida, Arizona and Nevada. Even among
subprime borrowers—those consti-

Relatively Stable House Prices Contributed to Smaller Spike
in Texas’ Subprime Delinquencies

Index, seasonally adjusted

Percent of mortgages, seasonally adjusted

240
220
200

U.S. subprime
serious delinquencies

U.S. FHFA
house price index

35
30
25

Texas FHFA
house price index

20
180
160

Texas subprime
serious delinquencies

140

15
10
5
0

120
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
SOURCES: Mortgage Bankers Association; Federal Housing Finance Agency; Haver Analytics.

tuting the greatest credit risk—Texas
default rates remained well below the
national average during the recession
and subsequent tepid recovery.
The Texas experience presents an
important case study, in part because
of a unique state law. Texas is the only
state with a regulation limiting home
equity borrowing. After purchase,
mortgage debt along with any new
borrowing—including home equity
loans—cannot exceed 80 percent of a
home’s market value unless the new
debt funds home improvements.
Consequently, the state has lower
levels of “cash-outs”—owners taking
money out of their houses during a refinancing. Through the boom, the rates
of mortgage debt growth and consumer spending in response to house price
appreciation were more restrained in
Texas. And during the downturn, the
state’s cap on home equity borrowing
may have also helped homeowners
avoid incurring negative equity and,
with it, the excessive mortgage default
rates that occurred elsewhere.
There has been little research that
statistically analyzes the role of the
home equity regulation in keeping
underwater mortgages, default rates
and foreclosures in Texas below the national average. Understanding the impact of the state’s restrictions on home
equity borrowing may aid policymaker
efforts to protect consumers and rein
in risky lending practices that led to the
mortgage and financial crises.1
Two additional factors have been
widely cited to explain the Texas soft
landing amid the Great Recession.
First, house price expectations were
less exuberant in the early to mid2000s, as Texans recalled the significant correction in the late 1980s that
followed the mid-1980s oil price bust.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

homeowners
}Texas
were less aggressive
in taking money out of
their homes during the
2002–06 housing boom
than others nationally.

Second, during the recent housing
boom, Texas price pressures were more
contained than in many other states
because of an abundance of land and
less-stringent zoning requirements that
made it easier to meet demand.
Overall, the state’s relatively stronger economy also played a role, with a
smaller rise in the unemployment rate
and a shorter downturn.

Mortgage Debt Growth
Texas homeowners were less aggressive in taking money out of their
homes during the 2002–06 housing
boom than others nationally, patterns
of mortgage debt and house price

Chart

growth indicate.2 Chart 2 plots the
growth of mortgage debt and house
prices using county-level data constructed from a large consumer credit
database assembled by the Federal
Reserve Bank of New York.
The upward-sloping line overlying
the data points in Chart 2A suggests
that as house prices rose nationally during the boom, homeowners
increased their mortgage debt. On the
other hand, the line in Chart 2B for
Texas counties has a small negative
slope, indicating that Texas homeowners did not increase their mortgage obligations in response to modest house
price gains.

House Price, Mortgage Debt Relationship Holds
for U.S., Misses for Texas

A. Mortgage Debt Growth Responds to House Price Increases in U.S.
Mortgage debt (percent change)

236
186
136
86
36
–14
–64
–14

46
66
86
House price index (percent change)

106

126

B. Little Correlation Between Mortgage Debt, House Prices in Texas
Mortgage debt (percent change)

113
103
93
83
73
63
53
43
33
23
13

House price index (percent change)
SOURCES: Authors’ calculations based on Federal Reserve Bank of New York Consumer Credit Panel/Equifax data.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Texas’ Home Equity Regulations
Under the 1876 Texas Constitution,
residents could only borrow against
the equity in their home for improvements (see “Texas’ Homestead Exemption Evolves Through the Years”). Aside
from very limited exceptions, cash-out
refinancings were prohibited. The only
reliable way to get money out of a house
was to sell it.
A 1997 constitutional amendment
allowed closed-end home equity loans—
an obligation to repay by a specified
date—but the loan plus the primary
mortgage could not exceed 80 percent
of the value of the home.3 Although
such lending increased after 1997, the
state’s regulations capping home equity
extraction are believed to have played
an important role in helping the state
navigate the post-2007 mortgage crisis.
Homeowners in many states, meanwhile,
extracted a significant portion of their
house price gains during the housing
boom.

Texas’ Homestead Exemption Evolves Through the Years
Texas’ ban on home equity lending was based on the state’s broad homestead
exemption, which excluded a portion of a property’s value from property taxes and
protected a primary residence from forced sale or seizure as long as mortgage and
tax payments were current. The exemption, however, also prevented homeowners
from withdrawing equity from their homes.
1839: The Republic of Texas passed a prohibition on the forced sale of homesteads for all but a very limited number of reasons. This prohibition was a reaction to
the Panic of 1837, when a number of people lost their farms and homes to foreclosure.
1845: After joining the United States, Texas enshrined the homestead exemption
in the first Texas Constitution. The 1845 Texas Constitution forbade the forced sale of
a homestead of up to 200 acres or a value of up to $2,000. This prohibition continued
in the 1861 and 1865 Texas Constitutions. The 1869 Constitution maintained the
prohibition and raised the protected value to $5,000.
1876: Initially, the Texas Constitution allowed homeowners to borrow against
equity for home improvements. More broadly, a lien could be granted on the homestead to fund:
•
•
•
•

The initial purchase or to secure the mortgage.
Unpaid taxes.
Home improvements.
Owelty of partition (to convert to full property ownership from part ownership,
most commonly after divorce).
• Refinancing of existing liens plus refinance costs.

Measuring Negative Equity
Before the 1997 amendment, Texans
borrowed aggressively when acquiring
a house. One way to measure the law’s
impact is to see if the mortgage-loanto-home-value ratio declined. Analyzing a sample of first liens confirms the
hypothesis that it did.4 First mortgages
originating in Texas before 1997 had an
average initial loan-to-value (LTV) ratio
of 86 percent, 9 percentage points higher
than in the rest of the nation. Texans, on
average, are relatively more credit constrained and need a larger first mortgage
to buy a house. But the average initial
LTV ratio among mortgages originating
after 1997 declined to 80 percent in Texas,
and the gap with the nation shrank to 6
percentage points.
The decline in the proportion of
mortgages with initial LTV exceeding 80
percent is even more striking since 1997,
further suggesting that the reform likely
induced Texans to limit their initial loan
amount on first mortgages as home
equity loans became available (Chart 3).
All else equal, mortgages with a
smaller initial LTV ratio are significantly
less likely to default. Consequently,
the 1997 amendment set the stage for

The exemption applies to 10 acres for urban homesteads, 100 acres for rural
households held by a single adult and 200 acres for a rural homestead occupied by a
family. The homestead exemption has no dollar limit.
1986: The federal Tax Reform Act of 1986 made home equity loans more attractive by phasing out the tax deductibility of interest paid on other, nonmortgage
consumer loans. However, Texas’ homestead exemption precluded such home equity
lending.
1997: Texas voters passed a constitutional amendment allowing closed-end
home equity loans effective Jan. 1, 1998. It stipulated that a home equity loan plus
the primary mortgage be less than 80 percent of the value of the home.
1999: Another amendment made the guidelines for reverse mortgages consistent
with those outlined in federal law. Because of inconsistencies between the U.S. and
Texas, reverse mortgages weren’t made in Texas before 2000.
2003: Texas voters passed an amendment allowing open-end home equity loans,
so-called home equity lines of credit.1 Total debt secured by the home still cannot
exceed 80 percent of a home’s value. Funds from a home equity credit line cannot
exceed 50 percent of the value of the home at the time the home equity line of credit
is made. Advances must be in increments of $4,000 and cannot be made via debit
card, credit card or preprinted check.
2007: Minor revisions in the home equity lending amendment were passed. The
changes modified the procedures for obtaining and granting a home equity loan and
changed the deadline for designating property as agricultural.

Note
A home equity line of credit works like a credit card. Borrowers can borrow up to a set limit determined by
the lender. The loans have a variable interest rate, so payments vary according to the interest rate and amount
of credit used.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

Looking just at first mortgages
understates the extent of negative equity
because there could be two or more
mortgages secured by a house. It is possible for a home to be worth more than
the first mortgage but less than all the
mortgages combined. Information on all
mortgages securing a house is available
for nonprime borrowers—a group believed to be at the center of the mortgage
crisis.
The gap in the incidence of underwater mortgages between Texas and the
rest of the nation among nonprime borrowers is particularly striking (Chart 4).6
The proportion of nonprime borrowers underwater in the other 49 states
reached a high of 54 percent in 2011,
while in Texas it peaked at 10 percent.
The depth of negative equity among underwater nonprime mortgages was also
significantly lower in Texas. Mortgage
debt among Texas’ underwater homeowners exceeded the home value by an
average 14 percent in 2008 compared
with 32 percent for the rest of the country.
The state’s restrictions on home
equity borrowing cannot be given all the
credit for the lower incidence of negative
equity in Texas. The absence of a housing
bubble in Texas clearly contributed but
cannot possibly account for the entire
difference. Nationally, borrowers extracting equity from their homes played a
major role in pushing homeowners
underwater; the 80 percent cap on
home-equity-based borrowing in Texas
mitigated that urge to pull money out.7

Share of Mortgages with Loan-to-Value Ratio
Above 80 Percent Dropped in Texas

Percent
70

Texas

U.S. minus Texas

50
42

40
32
30
20
10
0
Before 1997

After 1997

NOTE: Home purchase mortgages before and after 1997 Texas Constitution change.
SOURCES: Authors’ calculations based on data provided by LPS Applied Analytics.

Chart

As the Housing Bubble Burst, Underwater Nonprime
Mortgages Skyrocketed Outside of Texas

Nonprime mortgages underwater (percent)
60

50
U.S. minus Texas

Texas

0
2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

SOURCES: Authors’ calculations based on data provided by CoreLogic.

long-term benefits to Texans in terms of
avoiding a severe housing slump.

Texas vis-à-vis the Nation
Underwater mortgages are a key
factor contributing to default, recent
research suggests. Households owing
more than a house is worth may engage
in “strategic default”—choosing to walk
away even though they can still make
their payments. Although mortgage
balance information was current in the
database used for Chart 3, home value
data pertained only to the date of pur-

chase. To overcome this limitation,
we updated initial home values using
monthly house price data by ZIP code
compiled by CoreLogic, a financial
analytics firm.5
When U.S. house prices peaked in
2006, the incidence of underwater first
mortgages was less than 1 percent of all
first mortgages in Texas and the U.S. The
national share of underwater first mortgages reached 12 percent in 2008 and
continued climbing before peaking at 27
percent in 2011. That compared with 7
percent in Texas in 2011.

Extracting Home Equity
At some point after purchase,
home equity consists of three components: initial equity at the time of
purchase plus any change in house
price since mortgage origination minus
any change in mortgage balance since
origination. Initial equity is often positive since LTV at origination is typically
less than 100 percent. House prices fell
sharply after 2006 in many states, eroding much of the 2000–06 boom’s home
equity gains.
Detailed homeowner-level data on
the amount of home equity extracted is
not readily available. But a close look at
the purpose for which a lender approved

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

Texas Subprime Borrowers Lagged the Nation in Cashing Out
During Housing Boom

Subprime mortgages involving cashout refinancing (percent)
60

Texas
U.S. minus Texas
49

48
39

35
32

2000

2001

2002

2003

2004

2005

2006

2007

SOURCES: Authors’ calculations based on data provided by CoreLogic.

a mortgage can shed some light on how
much of the house price gains Texans
extracted vis-à-vis the nation.
Most mortgages are either for
outright purchase of a home or for
refinancing into a new mortgage to take
advantage of lower interest rates. Many
other borrowers refinance to increase
their mortgage amount and withdraw
accumulated equity in their homes.
About half of the subprime mortgages
that originated across the nation, excluding Texas, between 2000 and 2006
involved cash-out refinancing. In Texas,
this share was much smaller (Chart 5).
Texas has a younger population and
homeowners with lower credit quality,
who typically borrow more aggressively
against their home equity. Alone, these
factors would imply a higher incidence
of cash-out refinancing. The belowaverage share of such activity among
Texas’ subprime mortgages lends
credibility to the explanation that the
80 percent home equity borrowing cap
likely contributed to the lower incidence
of negative equity.

Likely Mortgage Default Impacts
It is difficult to precisely assess the
impact of home equity restrictions on
Texas’ relatively lower mortgage default
rates. The state restriction was in place
well before the onset of the housing and
mortgage crisis. Additionally, since the

regulation covered all Texans, there is no
unaffected group to contrast with those
impacted. Comparisons with the rest
of the nation cannot fully disentangle
the role of the home equity borrowing
restrictions because states differ in many
other ways.
Nevertheless, at the peak of the
housing crisis, the share of subprime
mortgages underwater in Texas was
40 percentage points below the rest of
the nation, with serious delinquencies
among subprime borrowers about 10
percentage points lower. Much of this difference remains even after accounting for
such factors as the state’s relatively lower
unemployment rate, differences in credit
scores, smaller house price declines and
differences in other demographic and
economic factors such as age, sex, race,
education and household income.
Rules governing home equity
borrowing are not uniform across the
U.S., and Texas’ rules are significantly
more stringent. The data suggest that
the tighter regulations in Texas helped
keep underwater mortgages and default
rates from rising by as much as they did
elsewhere. By extension, lower default
rates and fewer underwater homeowners
might also have helped Texas avoid the
subsequent sharp drop in home prices
other states experienced.
To be sure, these benefits did not
come without attendant costs. Just as the

restrictions helped Texas navigate the
housing downturn, the same restrictions could have constrained consumer
spending growth during the boom by
preventing homeowners from fully tapping their housing wealth. At the same
time, this may have helped limit swings
in consumer spending. Moreover, the
inability to access housing wealth may
have driven some credit-constrained
Texans to more-expensive credit card
debt, unsecured consumer debt or even
payday loans. Any estimate of net benefit
of Texas’ home equity regulations must
also account for such costs.

Kumar is a senior research economist
and advisor in the Research Department and Skelton is a business economist in the Financial Industry Studies
Department at the Federal Reserve
Bank of Dallas.

NOTES
Internationally, loan-to-value limits seem to lead to more
stable housing markets. See “Cycle-Resistant Credit
Systems: Learning from Hong Kong’s Experience,” by Ying
Guan, Jeffery W. Gunther and Sophia Tsai, Federal Reserve
Bank of Dallas Economic Letter, vol. 5, no. 6, 2010.
2
For the nation, a recent paper found that every 10 percent
increase in home equity caused household borrowing to
increase by 5 percent. See “House Prices, Home EquityBased Borrowing and the U.S. Household Leverage Crisis,”
by Atif Mian and Amir Sufi, American Economic Review, vol.
101, no. 5, 2011, pp. 2132–56.
3
Closed-end home equity loans usually have a term of
between five and 15 years, with a fixed interest rate and the
same payments each month.
4
To confirm the hypothesis, we analyzed a 5 percent
random sample of first-lien mortgages from a large database
provided by LPS Applied Analytics. The database covers
about two-thirds of all installment-type loans in the U.S.
5
For methodology used, see “Below the Line: Estimates of
Negative Equity Among Nonprime Mortgage Borrowers,” by
Andrew Haughwout and Ebiere Okah, Federal Reserve Bank
of New York Economic Policy Review, vol. 15, no. 1, 2009,
pp. 31–43.
6
The analysis is performed using a 5 percent sample of a
private-label securities database available from CoreLogic.
The data mainly consist of approved loans given to
nonprime borrowers based on insufficient documentation
(subprime) or alternative documentation (alt-A). Unlike the
data used for Chart 2, this database does have information
on combined LTV on the first mortgage as well as all
subordinate liens on the home securing the first mortgage.
7
See “Recovering from the Housing and Financial Crisis,”
by John V. Duca and David Luttrell, Federal Reserve Bank of
Dallas Economic Letter, vol. 5, no. 7, 2010.
1

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

On the record
A Conversation with Jeff Fegan

Outgoing D/FW Airport
Chief Cites International
Growth as Key to Future
Jeff Fegan, chief executive officer of Dallas/Fort Worth International
Airport, is retiring after 19 years of leading what has become the
world’s fourth-busiest airport. He reflects on the facility’s evolution
and its impact as a driver of regional economic activity.

Q. How much have D/FW Airport’s
operations grown during your tenure
and how has the airport’s stature
changed in the past two decades?
At the time of my appointment as
CEO for D/FW, it had flights to 24 international destinations and 113 domestic
markets and was listed among the
busiest airports in the world. Today,
D/FW has service to 201 total destinations; 53 of those are international
cities and the other 148 are domestic.
More importantly, D/FW continues
to serve the economic interests of the
Dallas–Fort Worth region by attracting
new air service, giving our community
direct access to five continents and
connectivity to the entire world.
D/FW has now become one of
the most highly regarded airports in
the world for customer service, having
been ranked among the top 10 large
airports worldwide for seven straight
years in surveys by Airports Council
International. That organization constantly asks hundreds of thousands of
international passengers about their
airport experiences.

Q. What makes D/FW different from
other large U.S. airports? What
role does the vision of the airport’s
early backers have in its continuing
operations?
D/FW Airport differs through its
governance; it is owned by the cities of
Dallas and Fort Worth and operated by
a board of directors that includes the
mayors of the two owner cities. That

governance and the board’s support of
our management team make it possible
for D/FW to seek continuous improvement and a culture of excellence and
to look for ways to set itself apart from
other airports. That kind of culture of
excellence shows itself in many ways,
such as when you see D/FW leading the
airport industry on initiatives like sustainability, airfield safety, fire training,
research and customer satisfaction.
The airport’s founders had a vision
that it needed a lot of land. They were
very wise to gather 18,000 acres of farmland and put it to use as an airport—
that land continues to pay dividends to
this day. D/FW uses it not only as a buffer zone for noise, but also for appropriate commercial development—leases
and warehouses, for example—that we
are doing in cooperation with our city
neighbors. As D/FW develops its land,
those projects add revenue, making
the airport more cost competitive for
airlines to operate here.

Q. In the early 1980s, D/FW was a
hub to three airlines; now it’s just
American Airlines. How did that happen, and what are the trade-offs of
single-carrier dominance?
When Braniff Airways collapsed
in 1982, the hub-and-spoke-system we
know today [in which an airline feeds
its passengers into a hub for connecting
service] was really in its infancy. It was an
outgrowth of airline deregulation from
the late 1970s, as was Braniff’s collapse.
American Airlines really developed the

hub system to its full potential using
D/FW as its largest hub and home base,
to the point that Delta eventually shrank
its hub at D/FW in 2005.
American holds about 85 percent
of the market share at D/FW. However,
that does not mean that airlines don’t
have competition at D/FW because
our airport has low-cost carriers serving about 40 markets and competitive
flights to most major cities. We’ve added
a record amount of new airline service
in recent years, including international
carriers such as Qantas and Emirates,
and we’ve seen a lot more low-cost
service, with carriers such as Virgin
America, JetBlue and Spirit Airlines
starting service. With single-carrier
dominance, as you put it, the D/FW
market also enjoys a lot more air service
than other population centers of its size
because it is American’s largest hub.

Q. How will D/FW’s competitive position change with the end of Wright
Amendment limitations on flights at
nearby Dallas Love Field in 2014?
We don’t anticipate we will see a
lot of changes at D/FW based on the
end of these restrictions. The airlines
already serving D/FW today will
continue to do so. You might see some
additions of new cities from Love Field,
but those cities are already served by
airlines out of D/FW, so the competition will simply increase.
We are working very hard on
improving our facilities, primarily
through the Terminal Renewal and
Improvement Program. We will renovate D/FW’s four original terminals
with new infrastructure, new parking,
new concessions, new technology and
new amenities for passengers, to be
completed in 2017. The terminals date
back to the opening of the airport, in
1974, and a lot has changed since then,
so we are adapting to the new needs of
the modern traveler.

Q. How has the revenue stream
from more than 100 natural gas
wells on airport property changed
how D/FW does business? Do they
provide a buffer against the cyclical nature of air travel?

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

continues to serve the economic interests
}“D/FW
of the region by attracting new air service, giving
our community direct access to five continents
and connectivity to the entire world.”

Q. How well has D/FW competed
against other airports for international routes? Why was Chicago
O’Hare able to get Beijing and
India flights that D/FW sought?
The revenue stream from the natural gas wells has not changed anything
about D/FW business operations. From
the start, the natural gas revenues were
earmarked for capital accounts, and so
the money [$15 million in royalties in
2011, about $8 million in 2012 and an
anticipated $4 million this year] is saved
and invested in new projects or new
construction, and not day-to-day operations. The gas revenue doesn’t provide
any buffer against ups and downs in the
air travel market because the revenue is
set aside.

Q. What flights are D/FW’s biggest
revenue generators? How have the
sources of airport income changed
during your tenure?
Our most popular destinations
continue to be Cancun, Las Vegas, London, Orlando and some of the major
population and business centers such
as New York, Chicago, Los Angeles and
Washington, D.C. Regarding airport
income, one of the major changes that
we have engineered at D/FW over the
years is the model of how our revenue
is generated. At one point, we earned
65 percent of revenue from airline
sources, such as landing fees and
rents. Over the past 15 years, we’ve
managed to reverse that so that D/FW
earns 65 percent of its revenue from
our own sources such as concessions,
parking, rental cars or commercial
development. That was a significant
achievement because by diversifying
and increasing nonairline revenue
sources, D/FW has remained a very
cost-competitive airport for airlines to
do business.

D/FW has competed very well for
new international routes. There’s not
another airport in America that has
added 15 new international routes and
seven new passenger carriers in the
past 30 months. We’ve done very well in
adding service to regions of the world
that we were most interested in, such as
Dubai and Australia. We’re still keenly
interested in new service to China or
other Asian destinations, and we hope
to see something materialize there
soon. All this additional service means
we’ve seen our international passenger
traffic grow by 11 percent this year, and
we are seeing new flights to emerging
economies, such as the new, expanded
service from American Airlines into
South America, specifically Brazil and
Peru, with new service to Bogota, Colombia, later this year.
D/FW is really helped in the pursuit
of new air service primarily by the
strength of our home travel market,
connectivity through our hub and the
airport’s cost-competitive structure. As
to why D/FW was not able to get service
to Beijing or India as Chicago O’Hare
did, one of those decisions was driven
by a labor [union] issue and the other by
the strength of the market in relation to
how it could serve the destination. We
do have a growing market in the Dallas–
Fort Worth area—the fastest-growing
metropolitan area in the U.S.—and that
fact will also help D/FW gain more air
access to more air markets in the future.

The international air freight business is really a key driver for us in terms
of revenue, and we think it can improve
with more efforts to integrate our services with the needs of cargo carriers and
freight forwarders. The international air
freight industry is currently in a bit of a
downturn, but we’ve managed to keep
pace with the world trends, so we think
cargo will continue to be a strong slice of
our market. The biggest change we are
seeing is the effort to adopt new technologies to make shipping more efficient.
D/FW and Alliance Airport are both
great, providing outstanding locations
and services to their customers. I think
what sets D/FW apart is the ability to
interact with a more diverse portfolio of
freight forwarders and shippers here, a
lot more capacity for flights and faster
access to more major highways.

Q. How do you assess the future of
D/FW? What changes are planned
in the near term and the long
term?
I think the future is very bright for
D/FW. Our team is really working hard
to update our original terminals. For
the long term, major projects such as
Dallas Area Rapid Transit rail service
from Dallas are underway, and we hope
to have the Fort Worth Tex Rail project
on the planning cycle sometime soon.
We’ve already begun adding new gates to
Terminal B and have plans in the works
if other new gate needs arise. We’ve
never wanted to be an impediment to
economic growth for the D/FW area, and
by having capacity for airlines to grow,
D/FW can continue to drive the local
economy forward.

Q. How has the air freight business changed and what does D/FW
offer freight customers that the
newer Fort Worth Alliance Airport
doesn’t?

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Getting Prices Right: Addressing
Mexico’s History of Fuel Subsidies
By Michael D. Plante and Amy Jordan

}

In recent years, Mexico’s
retail gasoline and diesel
prices lagged behind
skyrocketing crude
oil prices, creating a
de facto government
subsidy.

he government of Mexico,
the world’s eighth-largest oil
producer in 2011, and its oil
and gas sector are practically
inseparable. Petróleos Mexicanos, or
Pemex, is the Mexican state-owned
monopoly responsible for producing
and refining crude oil in the country. It
also imports and distributes petroleum
products, such as gasoline and diesel.
Mexican President Enrique Peña
Nieto has sought a range of reforms
through the Pact for Mexico initiative,
the product of a broad-based coalition
of the nation’s main political parties.
Substantial energy sector changes are a
significant part of the ongoing overhaul.
Although the way fuel prices are determined is not explicitly discussed in the
pact, it is one aspect of Mexican energy
policy that could face reform. In Mexico,
as in many other developing countries,
the government sets fuel prices, which
can then behave very differently from
those in a country such as the U.S.,

Chart

where prices are determined in a free
market (Chart 1).
In recent years, Mexico’s retail
gasoline and diesel prices lagged
behind skyrocketing crude oil prices,
creating a de facto government subsidy.1
This support seeks to protect vulnerable
groups in society, such as the poor, from
high fuel prices or to provide them with
extra income. However, in Mexico, as in
many other countries, the subsidies have
mainly benefited higher-income groups.
These subsidies, among the largest in
the world, have cost billions of dollars in
recent years.
While ongoing government increases have brought domestic prices
closer to true economic cost, experience
shows that subsidies often reappear if
the government controls prices. Mexico’s
spate of major reforms presents a unique
opportunity to reexamine how domestic
fuel prices are set in the country.
Several options could produce
a more economically efficient out-

Mexico’s Gasoline Prices Reflect Price Controls

Pesos per liter

Dollars per liter

1.2
United States
1

.8
9
Mexico

6
.4
3

.2
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

SOURCES: Bureau of Labor Statistics; Bloomberg; authors’ calculations.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

Mexico Tax on Gasoline, Diesel Indicates Emerging Subsidy

Pesos per liter

4
3
Gasoline

2
1
0
–1
–2
–3
–4
–5

Diesel

–6
–7

1995

1997

1999

2001

2003

2005

2007

2009

2011

NOTE: Zero tax rate means retail cost equals cost of producing the fuel; negative tax rate implies consumer subsidy.
SOURCES: Organization for Economic Cooperation and Development; authors’ calculations.

come—for example, automatic adjustment of fuel prices based on strict
application of a formula that keeps
retail prices in line with the true cost, or
a more radical reform that liberalizes
the production, distribution and sale of
petroleum products (a measure that has
been recently proposed).

Fuel Price Systems
Retail fuel prices are set in a free
market in the U.S. and in many other
countries. Prices change frequently in
response to market conditions, and in
extreme situations they can rise or fall
dramatically in a short period. Under
this system, retail prices generally reflect the cost of the fuel being sold.
Not every country follows this approach. Some use a formula, calculated
periodically and taking into account
factors such as cost of production, to
determine retail fuel prices.
In still other countries, the government sets retail fuel prices in an ad hoc
manner. Prices change infrequently
and unpredictably and can remain
fixed for long periods. For example, in
Indonesia diesel prices have been fixed
at 4,500 rupiah per liter since 2008—a
period over which the currency depreciated 9 percent against the dollar.
When fuel prices are not set in
a free market, retail prices may not
reflect the underlying costs of the

fuel for an extended period. When
production costs exceed retail costs,
consumers benefit while some other
entity sustains a loss—typically the
government or a government-run oil
company.

Pricing in Mexico
The Mexican government sets retail gasoline and diesel prices monthly.
A formula provides an estimate of
Pemex’s production, distribution and
retailing costs, but the government can
choose a different price if it so desires.
Retail prices can be set above cost

Table

(consumers essentially pay a tax on the
fuel) or below cost (consumers receive
a subsidy).
Surging crude oil prices in recent
years have increased production costs
for gasoline and diesel. Mexican retail
prices, however, haven’t kept pace.
The result: Gasoline and diesel, which
were consistently taxed before 2006,
have been largely subsidized since
then (Chart 2). The gasoline subsidies
have averaged 1.2 pesos per liter per
year since 2006 (about 40 U.S. cents per
gallon); 1.8 pesos for diesel (59 cents
per gallon).

Misplaced Help
Officials often implement fuel
subsidies to protect vulnerable groups
of society from high fuel prices or to
provide extra income to those groups
at a low cost. However, the research
on fuel subsidies shows that they are
usually ill-targeted—that is, the poor do
not typically receive a large portion of
the benefits. Higher income groups can
afford greater fuel consumption while
poor households often lack the income
to even purchase the goods that use the
fuel—such as a car.
Mexico is confronted with this problem, data show. The bottom 20 percent
of the income distribution purchased
only 3 percent of the gasoline and diesel
in 2010, the last year for which such
figures are available (Table 1). Because

Mexico’s Distribution of Spending on Gasoline
and Diesel by Income Decile in 2010
Decile

Total (percent)

1.2

1.8

2.4

3.6

4.7

6.6

8.6

12.5

19.1

39.4

NOTE: Deciles are increasing in income, where decile 1 represents the lowest 10 percent of the income
distribution and decile 10 represents the highest 10 percent of the income distribution.
SOURCE: Mexican Secretariat of Finance and Public Credit.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

Mexico’s Spending on Oil Subsidies Rises Again

Billions of U.S. dollars

Percent of GDP

2.5
Subsidy cost

the goal of the
}Ifsubsidies
is to protect
the lowest income
brackets, the cost to do
so is heavy.

Percent of GDP

1.5

2007

2008

2009

2010

2011

NOTE: The IEA includes subsidies on gasoline, diesel and other oil products, such as LPG (propane).
SOURCES: International Energy Agency; International Monetary Fund; authors’ calculations.

the subsidy is applied per unit of fuel,
the data show that 97 percent of the
assistance went to the top 80 percent of
income earners in Mexico.
This is in line with other countries.
The International Energy Agency (IEA)
estimated that out of the $409 billion
spent globally in 2010 on all subsidies
covering consumption of fossil fuels
(oil, natural gas and coal as well as the
electricity they produced), only $35 billion, or 8 percent, reached the bottom
20 percent of the income distribution.
If the goal of the subsidies is to
protect the lowest income brackets, the
cost to do so is heavy. Of the $15.9 billion Mexico spent on subsidies in 2011,
roughly $15.4 billion went to higher
income groups—in other words, it cost
$15.4 billion to provide about $500 million in aid to Mexico’s poorest.

Costly Subsidies
Mexico’s subsidies have been
expensive on an absolute dollar basis
(Chart 3, left axis).2 While Mexico’s fuel
subsidies in recent years were below
the peak levels in 2008 (when subsidies
exceeded $20 billion), they still exceeded
$15 billion in 2011. Preliminary Mexican
government data suggest that the 2012
total will be close to the cost in 2011.
Sometimes it’s useful to consider
the size of a subsidy relative to gross
domestic product (GDP). This method

helps illustrate how big a burden the
subsidy might impose on the economy,
taking into account the country’s ability
to pay. As a share of GDP, Mexican fuel
subsidies were at least 1 percent of GDP
in four of the last five years for which
data are available (Chart 3, right axis). By
comparison, expenditures on education
amounted to 3.5 percent of GDP in 2010;
health spending, 2.8 percent; and pensions, 1.2 percent.
Relative to other countries, Mexico
typically ranks high in terms of the
subsidies’ dollar value. Mexico ranked
seventh in such spending in 2011, according to IEA data. Only Saudi Arabia
($46.12 billion), Iran ($41.39 billion),
India ($30.86 billion), Venezuela ($21.97
billion), Iraq ($20.35 billion) and China
($18.45 billion) spent more. However, on
the basis of subsidies as a percentage of
GDP, Mexico ranked relatively low—19th
out of 33 countries in 2011 (Chart 4).

Other Negative Impacts
Subsidies work by artificially reducing prices for fuel, making it relatively
cheaper than other goods. Households
and firms respond, changing their
economic decisions. This introduces
distortions in the economy that can
hinder performance. For example,
households may choose to consume an
outsized amount of fuel and to consume
less of other goods because of pricing,

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

Subsidies Spending as a Percentage of GDP in 2011:
Top 10 Countries and Mexico

Percent

12
10

Mexico, the best
}For
way to prevent subsidies

8
6

from reoccurring would
be liberalization of fuel
production and product
distribution.

(6
)
Eg
yp
t(
7)
Al
ge
ria
Tu
(8
rk
)
m
en
ist
an
(9
)
Ku
wa
it (
10
)
M
ex
ico
(1
9)

Lib
ya

bia
(

(4
)
Sa

3)
Ve
n

ez
ue
la

ua
do
r(

Ira

SOURCES: International Energy Agency; International Monetary Fund; authors’ calculations.

negatively impacting non-energy sectors
of the economy.3
Fuel subsidies can also negatively
affect the environment. Burning gasoline
and diesel produces air pollution. To the
extent that fuel subsidies lead to overconsumption, air pollution increases.
This can impose a cost on the population
in general because many forms of air
pollution negatively affect public health.

households are free to use the extra
income to purchase what they most
desire. They may buy more fuel, but they
might also spend the extra income on
something they value more, such as food
or clothing. By not affecting the price of
fuel, transfers may avoid other negative
implications of fuel subsidies, including
environmental problems.

Protecting the Poor

The Mexican government has had
an explicit policy of increasing retail
prices by small monthly increments
since 2010, aiming to eventually remove
subsidies. Contrary to what has occurred
in many other fuel-subsidizing countries,
no major political upheaval has accompanied the price increases, which as of
2013 brought prices to near global parity.
Although the subsidies are fading,
price rises don’t guarantee the subsidies’
permanent elimination. In other countries where governments control fuel
prices, energy subsidies have often reappeared over time, often after an event
caused sharply higher world fuel prices.
Governments generally hesitate to pass
on such rapidly rising costs.
For Mexico, the best way to prevent
subsidies from reoccurring would be liberalization of fuel production and product distribution. As long as the market is

The International Monetary Fund
has worked with a number of countries
to reform fuel pricing and currently
recommends removing subsidies and
replacing them with measures that
specifically aid the poor. Transfers of
cash targeted at lower-income households—rather than applying the subsidies to the entire population—is one
preferred method. Mexico, in fact, could
implement such a reform through its
already-established antipoverty program
Oportunidades. It has operated since
1997 and is specifically geared to provide
aid to Mexico’s poorest.4
There is also an important theoretical justification for using targeted
transfers instead of fuel subsidies.
Unlike fuel subsidies, transfers do not
artificially reduce fuel prices. Instead of
being induced to purchase more fuel,

Recent Events

reasonably competitive, retail prices in
Mexico would reflect the true economic
costs of producing and distributing
fuel. A second-best option would be
implementation of a more mechanical
fuel-pricing formula that automatically
adjusts monthly. This would smooth
short-term, retail-price fluctuations
while ensuring that they reflect the fuel’s
underlying cost over the medium term.

Plante is a research economist and
Jordan is a research analyst in the
Research Department of the Federal
Reserve Bank of Dallas.

Notes
Although not the focus of this article, the Mexican
government also subsidizes the purchase of LPG (propane).
2
Comprehensive estimates on the costs of fuel subsidies,
in dollars, are available from the IEA from 2007 to 2011 for
more than 30 countries. More information on how these
estimates are calculated can be found on the IEA’s website,
www.worldenergyoutlook.org/resources/energysubsidies/
methodology/.
3
For a more detailed analysis on how fuel subsidies can
distort an economy, see “The Long-Run Macroeconomic
Impacts of Fuel Subsidies,” by Michael D. Plante, Federal
Reserve Bank of Dallas Working Paper no. 1303, March
2013.
4
For more information on Oportunidades, see “As Mexico’s
Social Safety Net Grows, Issues Arise,” by Melissa LoPalo
and Pia Orrenius, Federal Reserve Bank of Dallas Southwest
Economy, Second Quarter, 2013.
1

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

NoteWorthy
WATER: Lawmakers Seek Funds to Expand Supplies, Ease Growing Demand

exans’ increasing thirst for water can’t continue. Demand is projected to rise 5.4 percent by 2020
while supply contracts 3.3 percent, according to the Texas Water Development Board, the agency
responsible for water planning and data collection.
State lawmakers will ask voters in November to fund $2 billion in water conservation and infrastructure improvements that backers say would increase supplies 50 percent by 2060.
Farmers use the most water in the state, about 56 percent, while urban areas account for 27 percent
of consumption.
Texas water comes from two sources. Surface water in lakes and rivers supplies 62 percent of municipal water, and groundwater drawn from underground aquifers supports 80 percent of farm use.
The state owns surface water, holding it in trust for the public. Residents and water authorities apply
for the right to use the water. Rights holders then sell the surface water to the public, pricing it to reflect
purification and transportation costs, but not its scarcity.
The state does not assign rights to groundwater, which largely depends on rain for replenishment.
Aquifer water becomes private property once landowners draw it from the ground and hold it.
—Edward Rodrigue

IMMIGRATION: Border Apprehensions in Rio Grande Valley Surge

or the first time in almost 50 years, the Rio Grande Valley sector is the most active section of the
2,000-mile U.S.–Mexico border for migrant apprehensions. While arrests along the entire border
were down 78 percent in 2012 from a peak in 2000, apprehensions in the border’s easternmost section rose 55 percent in the October 2012–June 1 period, compared with a year earlier. The Border Patrol
arrested 282,414 crossers over the length of the Mexican border in the recent period, 94,305 of them in the
Rio Grande Valley sector, which extends 120 miles and includes McAllen and Brownsville, Texas.
Border Patrol statistics suggest the South Texas crossers are not the same migrants who came in
through the Arizona desert, previously the most active section of the border. Migrants from countries
other than Mexico have traditionally favored the lower Rio Grande Valley; Reynosa, across the border
from McAllen, is the endpoint of a railroad that Central American migrants ride from southern Mexico.
Apprehensions of migrants from countries other than Mexico accounted for half of all Rio Grande Valley sector apprehensions in fiscal 2012, approaching levels last seen in 2006.
A lack of economic opportunities in Central America, along with increasing drug violence and instability, has prompted young workers to migrate, even as some of those factors have ebbed in Mexico.
—Melissa LoPalo

TRADE: Along Texas Gulf, Exports Pick Up as Oil Imports Decline

il imports, including crude oil and refined products such as gasoline and diesel, totaled 52 million
barrels at the Houston–Galveston Customs District in June 2013, the latest U.S. International Trade
Commission data show. The amount has declined by 20.4 million barrels, or 28 percent, since the
beginning of 2012, and was off 8.4 million barrels from January 2013, a 14 percent decrease. Oil imports
accounted for more than half the total value of all district imports in June 2013.The customs district is
made up of the ports of Corpus Christi, Freeport, Galveston, Houston, Port Lavaca and Texas City.
By themselves, crude oil imports to the district totaled 44.3 million barrels in June 2013, down 10
percent from the beginning of the year and 29 percent from January 2012. The decrease in oil imports
comes amid rising domestic crude production, particularly in Texas and North Dakota.
Increased domestic crude production has led to sharply higher exports of petroleum products.
Product exports from the district reached 32.2 million barrels in June 2013—up 10 percent since June
2012. U.S. exports of petroleum products have consistently exceeded imports since June 2011. Total
U.S. exports reached 82.6 million barrels in May 2013—a 14 percent increase from January 2012, according to U.S. Energy Information Administration data.
—Amy Jordan

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

spotlight

Firms Expect Health Act to Raise Labor Costs
By Amy Jordan and Pia Orrenius

ederal and state regulations
govern many aspects of
health insurance, which 159
million U.S. residents obtain
through an employer. The Affordable
Care Act (ACA), signed into law in
2010, imposed conditions that include
a requirement that companies employing 50 or more full-time-equivalent
workers provide affordable, broadbased insurance.1
The act stipulates that coverage
be of “minimum value”—paying at
least 60 percent of medical costs—and
affordable—with a premium representing no more than 9.5 percent of
a worker’s annual earnings. It also
imposes fines on companies of 50 or
more workers that fail to offer coverage for full-time employees starting in
January 2015 and on individuals who
go without health insurance.2
For small companies—those with
fewer than 50 full-time-equivalent employees—the law provides incentives
to offer coverage, including a tax credit
and access to a small-business health
insurance exchange.3
Some employers and legislators
are concerned that the ACA’s provisions will drive up labor costs, which
could damp hiring and economic

growth. Others worry that the law may
distort the labor market, providing
incentives to convert full-time workers
to part-time or trim payrolls to below
the 50-employee threshold. However,
the impact on hiring could be muted
if employers shift compensation from
wages to health benefits, leaving total
compensation unchanged. To discern the sentiments of area firms, the
Federal Reserve Bank of Dallas asked
respondents to its Texas Business Outlook Surveys (TBOS) a series of special
questions about the act.
More than 400 Texas businesses in
manufacturing (Texas Manufacturing
Outlook Survey), services (Texas Service Sector Outlook Survey) and retail
(Texas Retail Outlook Survey) participate in the monthly TBOS surveys, and
262 responded to questions regarding
how health care reform will affect their
labor costs, wages and benefits, and
staffing decisions.4 The great majority of respondents—77 percent—said
they expect labor costs to increase. By
sector, the results were 81 percent in
manufacturing, 75 percent in services
and 77 percent in retail.
To counteract new cost pressures,
more than half (54 percent) indicated
they would adjust compensation struc-

Employers Likely to Alter Compensation Structures Under Act
Will likely:
Manufacturing

13.5

24.3

8.1 5.4

Pass some/all increased
costs on to employees

48.7

Reduce wages or other
benefits to help offset
Services

30.4

11.4 6.3 10.1 3.8

Minimize raises and
bonuses to help offset

Alter health care coverage
to reduce cost increases
Retail

35.7
0

14.3
40

7.1

Percent

Discontinue coverage
and pay penalty

42.9
60

SOURCE: Texas Business Outlook Surveys, Federal Reserve Bank of Dallas.

Other
100

tures. Of those anticipating adjusting
compensation, 28 percent said they will
likely pass on some or all additional
health care costs to employees, while 12
percent said they will likely reduce employee wages or other benefits to offset
the costs (see chart for results by sector).
The Dallas Fed also asked companies to report on planned staffing
changes. A majority (64 percent) said
they expect no change or are uncertain.
However, 28 percent anticipate reducing
the number of full-time employees, with
18 percent planning to increase parttime, contract or temporary workers.5
Implementation of the ACA employer mandate may not significantly
affect large companies that already
offer workers health insurance.
Initially, some employers may experience increased labor costs as more
workers sign up for coverage. This surge
could temporarily hurt profits and hiring. However, according to the TBOS
survey, the increased cost may cause
firms to shift compensation from wages
to health care, which is what happened
in Massachusetts following implementation of a similar law there.

Notes
The ACA defines full-time workers as those working at least
30 hours per week. Under the law, combinations of part-time
employees count toward the threshold; for example, two
half-time employees (15 hours per week) are equivalent to
one full-time employee.
2
The government announced July 2 that it would implement
the employer mandate in January 2015, one year later than
originally planned. Plans are still on track for uninsured
individuals to sign up for coverage through individual
insurance exchanges starting Oct. 1, 2013, for coverage
beginning Jan. 1, 2014. Fines for individuals who decline
coverage will begin in 2014 and become payable in 2015.
3
Under the ACA individual mandate, everyone must
have medical insurance, obtained through their own or
their spouse’s employer, the government (Medicaid and
Medicare) or health insurance exchanges.
4
See current and historical releases of TBOS as well as the
results of the April 2013 special questions at www.dallasfed.
org/research/surveys/index.cfm.
5
Respondents were asked to select all options that apply,
so percentages sum to more than 100 percent.
1

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

For Hispanics, Border Wage Gap
Reflects Education, English Divide
By Christina English

}

While the earnings
differences based on
educational attainment
may not be surprising,
border wage gaps
by ethnicity are
remarkably large.

he Texas side of the U.S–
Mexico border is one of the
nation’s poorest regions, with
30 percent of the people living
in poverty, nearly double the national
average. In this area, where 86 percent
of the population is Hispanic, a wide
gap separates the earnings of Hispanics
and non-Hispanics.
Low educational attainment is the
primary reason Hispanics on the border
earn less, but low English fluency also
matters. Adjusting for differences
in English ability and education all
but eliminates the border wage gap
between Hispanics and non-Hispanics.
This implies that lower overall skills are
driving the Hispanic wage deficit along
the border and that earnings can be
increased by improving both English
proficiency and education, particularly
the high school completion rate.

Low Education, Lower Earnings
Hispanic residents along the
border are far less educated than their
non-Hispanic counterparts and Texans
overall. In the border region, about
one-third of Hispanics did not complete
high school, compared with 7 percent of
non-Hispanics, according to the Census
Bureau’s American Community Survey
(Chart 1).1
The low number of Hispanics along
the border pursuing higher education
produces further differences. While
more Hispanics than non-Hispanics
have only a high school credential (25.8
percent versus 22.3 percent), far fewer
have a college degree (15 percent versus
37.9 percent).
Education is highly correlated with
wages. College graduates along the
border make the most, with average
annual earnings of $71,000 among
non-Hispanics and $50,300 among
Hispanics (Table 1). Among those with

less than a high school education,
annual wages total $23,300 for nonHispanics and $16,700 for Hispanics. In
each group, those not completing high
school earn only about one-third of
college graduates’ wages.

Drivers of Hispanic Earnings
While the earnings differences
based on educational attainment may
not be surprising, border wage gaps
by ethnicity are remarkably large. The
higher wages of non-Hispanics suggest
that well-paying jobs exist in border
cities, so what is behind the relatively
low earnings for Hispanics?
One possible explanation is that 30
percent of Hispanics along the border
were born outside the United States,
usually in Mexico. Some may be in the
country illegally, which likely depresses
their wages. Further, college-educated
immigrants may have trouble getting
their foreign degrees recognized and
valued by U.S. employers.
Previous research has suggested
Mexican migrants in Texas border cities accept lower wages to be closer to
family in Mexico.2 So while 70 percent
of Hispanic residents along the border
are U.S. born, they may also have family or relatives in Mexico.
English fluency is another factor
potentially limiting Hispanic earnings
that hasn’t been fully explored. The
American Community Survey queries
respondents on English proficiency on
a scale of 1 to 5, where 1 is “speaks no
English” and 5 is “speaks only English.”
Hispanic border residents rated
themselves a 3.2 on average, far lower
than the 4.7 reported by non-Hispanic
border residents. Even college-educated
Hispanic border residents rate their
English comparatively low, at 3.8, nearly
a full point lower than similarly educated
non-Hispanics (4.7). Hispanics along

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Chart

On Border, Hispanics Trail Non-Hispanics in Education

Percent

40
Hispanics

Non-Hispanics

30
25

has shown
}Research
limited English

20
15
10
5
0

Less than high school

High school

Some college

College

NOTE: Data include both native and foreign-born Texas border residents, age 25–64.
SOURCES: American Community Survey 2009–11; author’s calculations.

Table

Hispanics Along Border Earn Less at Each Education Level
Hispanics

Non-Hispanics

Less than high school

$16,700

$23,300

High school

$24,600

$34,900

Some college

$31,500

$44,800

College

$50,300

$71,000

proficiency puts certain
jobs out of reach, with
the least fluent working
in relatively low-paying
jobs requiring fewer
skills.

NOTES: Average yearly wages of people age 25–64 on Texas border who worked full time. Wages are inflation adjusted
(expressed in 2011 dollars) and rounded to the nearest 100.
SOURCES: American Community Survey 2009–11; author’s calculations.

the border also rate their English skills
slightly lower than Hispanics in the rest
of Texas rank theirs (3.5). The difference
in English skills is consistent across all
education levels and persists even for
native-born Hispanics along the border,
who rate their skills at 3.7 versus 4.0 in
the rest of Texas.
Research has shown limited English
proficiency puts certain jobs out of reach,
with the least fluent working in relatively
low-paying jobs requiring fewer skills.3
Poor language skills also may be
a reason some Hispanics stay in the
Spanish-friendly border area rather
than go farther into Texas, where lesser
English skills would be a greater day-today problem.

Wage Gap Causes
The wage gap between Hispanics
and non-Hispanics illustrates just how
much lower Hispanic earnings are across
Texas.4 Table 2, row 1 shows Hispanic

wages along the border and statewide
without accounting for worker characteristics that influence earnings power
and may narrow differences.5 At first
glance, Hispanic border residents earn
about half the wages of non-Hispanics
along the border; in the rest of Texas,
Hispanics take in 40 percent less than
non-Hispanics.
Age and gender do not explain
much of the difference in Hispanic
earnings. Adjusting for these demographic characteristics leaves the
wage gap with non-Hispanics largely
unchanged (Table 2, row 2).
Controlling for years of education
(Table 2, row 3) drastically reduces
the wage gap along the border from
49 percent to 19 percent. This means
over half of the observed difference in
earnings is due to the lower educational attainment of Hispanic border
workers. In the rest of Texas, educational attainment has an even larger

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Table

Lower Education Accounts for Over Half of Hispanic Wage Gap
Remaining wage gap with local
non-Hispanics (percent)

raising their
}Byeducational
attainment
to non-Hispanic levels,
full-time Hispanic
workers would increase
average earnings by
$9,600 a year, a 34
percent jump to $37,800.

Border residents

Non-border residents

–51

–40

Demographic differences

–49

–42

Add: educational attainment

–19

–6

Add: English skills

–5

–4

Unadjusted
Adjusted for:

NOTE: Average yearly wages of people age 25–64 on Texas border who worked full time.
SOURCES: American Community Survey 2009–11; author’s calculations.

impact, shrinking the wage gap to 6
percent from 42 percent.
Adding controls for English ability
(Table 2, row 4) further shrinks the
wage gap for Hispanic border workers
from 19 percent to 5 percent, suggesting a lack of English proficiency plays
a large part in limiting earnings. But in
the rest of Texas, differences in English
ability reduce the wage gap only 2
percentage points because average
English skills are higher than along
the border.
While educational attainment is
the primary driver of the wage gap in
both locations, English ability is also
an important factor along the border.
Since controlling for English fluency
and education level nearly eliminates
the wage difference, Hispanics who
work full-time have lower average
earnings because the skill gap is so
large.

Potential for Higher Earnings
The impact of more education
on the earnings of Hispanic workers
along the border can be extrapolated, based on the statistical analysis
underlying the results in Table 2. By
raising their educational attainment
to non-Hispanic levels—holding all
other characteristics the same (sex,
age, experience, etc.)—full-time Hispanic workers would increase average
earnings by $9,600 a year, a 34 percent
jump to $37,800 from the current
$28,300.6
Much of the education difference
reflects Hispanics’ lower college graduation rates. The typical Hispanic worker
with a full-time job and some college,

but no degree, would increase annual
earnings by $12,700 by completing
college—a 38 percent rise to $46,000
from $33,300 (see “Hispanic Women
on Border Gain the Most from College
Degree,” page 19).
Finishing high school also pays
dividends. A full-time Hispanic
worker on the border with a high
school credential (diploma or General
Educational Development certificate)
could achieve a 72 percent earnings
bump, to $28,700 from $16,700. This
large percentage increase reflects the
low earnings power of individuals with
less than a high school education (who
average just seven years of schooling).
Improved English fluency would also
raise income, although gains would not
be quite as large.
This is a statistical exercise, and
the scenarios are hypothetical. The
full extent of projected gains won’t
likely be attained; even if all Hispanics
reached the education levels of nonHispanics, some jobs would be out of
reach without adequate English skills
or legal status—something not captured by the data. Also, the statistical
review assumes that, except for education, everyone is similar. However,
education levels typically reflect other,
unobserved factors that influence
income, such as ability, family wealth
and access to credit.
The hypothetical education increases are therefore likely to overestimate the true impact on earnings. Still,
the broader impact on the wage gap
is instructive. It is largely the product
of education and English fluency,
through which a solution also lies.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

Benefits of More Education
Hispanics along the Texas–Mexico
border experience poverty rates that
are more than double those of nonHispanics. The income inequality partly results from a cycle of low English
skills limiting educational attainment.
When passed on to children, this can
lead to reduced high school graduation
rates among even U.S.-born Hispanics.7
Improving English fluency and
educational attainment could help narrow the wage gap between Hispanics
and non-Hispanics. Boosting Hispanic
workers’ self-reported English proficiency roughly 20 percent (to 4) and
increasing high school attainment (to
12 years from 11.6) would increase
annual average earnings 20 percent, or
$5,600.
This suggests policymakers should
focus limited resources on improving
English-language proficiency and raising the high school completion rate.
Children can be targeted though the
school system and adults via continuing-education classes.

English is a research analyst in the
Research Department of the Federal
Reserve Bank of Dallas.

Notes
This article is based on research for “Immigration and
Education: Setbacks and Opportunities for Earnings
Along the Texas–Mexico Border,” by Christina Daly,
Journal of Borderlands Studies, vol. 27, no. 3, pp.
287–98.
1
The data are from the U.S. Census Bureau’s 2009–11
American Community Survey. The survey yields data on
areas with populations exceeding 20,000. The data are
organized by Public Use Microdata Area and are deemed
to be border if they contain a county that borders Mexico.
Mentions of the border refer to the Texas–Mexico border.
2
See “Mexican Migrants Stay in Border Comfort Zone,”
by Pia Orrenius and Madeline Zavodny, Federal Reserve
Bank of Dallas Southwest Economy, May/June 2008,
pp. 10–13.
3
See “English Skills, Earnings and the Occupational
Sorting of Mexican Americans Along the U.S.–Mexico
Border,” by Alberto Dávila and Marie T. Mora,
International Migration Review, vol. 34, no. 1, 2000,
pp. 133–57.
4
See “Texas’ Latino Pay Gaps: Taking a Closer Look,” by
Emily Kerr, Pia Orrenius and Madeline Zavodny, Federal
Reserve Bank of Dallas Southwest Economy, First
Quarter, 2010.
5
The log of real yearly earnings among full-time border

workers age 25–64 is used as the dependent variable
in least squares regressions on the Hispanic dummy
variable (row 1), adding demographic information (row
2), education attainment (row 3) and English skills
(row 4). The wage gap is the coefficient of the Hispanic
dummy variable.
6
To calculate the hypothetical wage, the mean of each
variable is plugged into the regression equation used to
generate Table 2, column 1, row 4. The education means
of Hispanics are then replaced with average education
for non-Hispanics. The benchmark earnings are simply
the unconditional average earnings in the American
Community Survey data for the relevant Hispanic
population (all full-time Hispanic workers, age 25–64;
those with some college; and those with less than a high
school education).
7
See “A Conversation with Richard Fry: Why Hispanic
Education Deficits Persist,” Federal Reserve Bank of
Dallas Southwest Economy, First Quarter, 2010, pp. 8–9.

suggests
}This
policymakers should
focus limited resources
on improving Englishlanguage proficiency
and raising the high
school completion
rate.

Hispanic Women on Border Gain the Most from College Degree
Individuals completing college in the U.S. enjoy a “college premium”—
a large increase in earnings over what high school graduates earn.
The premium is particularly significant for Hispanics along the Texas
border. Estimates suggest that those who work full-time earn 90 percent
more than if they had not received a college degree. The result matches
non-Hispanics’ experience in the rest of the state.1 Meanwhile, Hispanic
college graduates not on the border earn only 77 percent more than high
school graduates, suggesting Hispanics along the border can gain relatively more by obtaining a college education.2
Hispanic women along the border have the highest college premium
in Texas, earning 105 percent more than if they had not obtained a degree.
One possible explanation is the overall scarcity of highly educated workers in the region, which is likely even more pronounced among bilingual
Hispanic women.
For example, bilingual teachers along the border are paid more for
their ability to communicate with Spanish-speaking students and Englishspeaking administrators.3 Farther inland, the shortage of bilingual teachers
may not be as large and, therefore, the premium is lower.

Notes
1
The college premium is the coefficient on the dummy variable for completing college, compared
to the omitted category of high school graduate in least squares regressions with the log of
real yearly earnings among full-time border workers age 25–64 as the dependent variable with
controls for experience and gender.
2
Other studies have also found higher returns to education along the border where highly
educated workers are more scarce. See “The Border: Is It Really a Low-Wage Area?” by Lori L.
Taylor, Federal Reserve Bank of Dallas Border Economy, June 2001, pp. 6–8.
3
See “Districts Seek Bilingual Education Teachers in Mexico,” by Cynthia Garza, Houston
Chronicle, Feb. 21, 2007, www.chron.com/news/houston-texas/article/Districts-seek-bilingualed-teachers-in-Mexico-1828254.php.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2013

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

PRSRT STD
U.S. POSTAGE

PAID

DALLAS, TEXAS
PERMIT NO. 151

Snapshot Republic of Texas Dollar Slumped
when Public Doubted Its Worth

he saying “cash is king” is based on the assumption
that as the most liquid form of exchange, money in
hand is superior to, say, the debt obligations of private
companies. The independent Republic of Texas tested that
proposition during its existence, from 1836 to 1845.
The period coincided with a severe depression in the
U.S. that lasted from 1837 to 1845 and influenced events in
the republic, whose constitution gave lawmakers “the power
to coin money,” provided it was gold or silver. Metal coins of
any type were scarce in the U.S. and more so in Texas, which
never issued them.
In January 1839, during the administration of Mirabeau
B. Lamar, a new type of promissory note called the “red
back” was authorized. As James Michener observes in his
1,000-plus-page historical narrative Texas, a fictional work
grounded in fact, the currency didn’t garner wide acceptance during its three years of circulation:
“But always the limiting factor was this strangling lack
of currency, and when the distressed nation, swamped with
debt, tried to salvage itself by printing two million dollars’

DALLASFED

worth of ‘red-back bills’ supported by no collateral except
the government’s word and faith,
citizens evaluated the situation
realistically,” Michener writes.
“On the first day it was issued, a
dollar bill was worth fifty cents, a few days later thirty cents,
then ten cents and four cents, until it bottomed out at an appalling two cents.”
By February 1842, not even the republic’s government
had any faith in its currency. The Texas dollar was no longer
acceptable for payment of taxes. In its place, obligations
issued by state banks were legal tender, along with “shinplasters,” company-issued notes in denominations of less than
a dollar that first appeared in fall 1837. The Texas Congress
had previously moved to restrict such currency to ensure the
primacy of its own money.
—Adapted from the Journal of Political Economy
(April 2013) and Handbook of Texas

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 free of charge by
writing the Public Affairs Department, Federal Reserve
Bank of Dallas, P.O. Box 655906, Dallas, TX 75265-5906;
by fax at 214-922-5268; or by telephone at 214-9225254. This publication is available on the Dallas Fed
website, www.dallasfed.org.

Harvey Rosenblum, Executive Vice President and
Director of Research
Mine Yücel, Director of Research Publications
Pia Orrenius, Executive Editor
Michael Weiss, Editor
Jennifer Afflerbach, Associate Editor
Kathy Thacker, Associate Editor
Samantha Coplen, Graphic Designer
Ellah Piña, Graphic Designer

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

Full text of The Southwest Economy : Third Quarter 2013

FRASER