Full text of The Southwest Economy : Third Quarter 2018

View original document

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

Federal Reserve
Bank of Dallas
THIRD QUARTER 2018

Southwest
Economy

}

New Technology Boosts Texas
Firms’ Output, Alters Worker Mix
PLUS
}}
Texas Property Taxes Soar as Homeowners Confront
Rising Values

}}
On the Record: Shale Renews Excitement in Energy Industry
}}
Parental Borrowing for College Comes with Repayment Issues
}}
Spotlight: Shale Oil Propels U.S. Crude Export Increase
}}
Go Figure: If Texas Were a Country ...

PRESIDENT’S PERSPECTIVE

} Companies provided
insight into new

technologies they
were adopting and
their impact on
operations and costs.

echnology-enabled disruption means technology is increasingly replacing workers. It also
means old business models are being replaced
by new models for delivering goods and services at
lower prices and potentially better convenience. This
powerful trend was the subject of a Federal Reserve
System conference co-hosted by the Dallas and Atlanta Federal Reserve Banks earlier this year.
As a follow-up to the conference, the Dallas Fed
surveyed more than 300 Texas firms to understand
how technology has impacted their businesses. Emily
Kerr, Pia Orrenius and Christopher Slijk detail the
results in their article headlining this quarter’s
Southwest Economy, “New Technology Boosts Texas
Firms’ Output, Alters Worker Mix.”
Companies provided insight into new technologies
they were adopting and their impact on operations
and costs. Most firms said they adopted technology
to boost productivity and keep up with competitors.
Most also said technology had not reduced the numbers of workers they employed but did affect the types
of workers they needed.
Our work in this area suggests that education and
skills training are critical to helping workers adapt to
this trend. For example, community leaders can form
partnerships to more effectively improve early childhood literacy, college readiness, and skills training at
our high schools and community colleges.
Also in this quarter’s issue, Jason Saving discusses
reasons for rising property taxes in his article, “Texas
Property Taxes Soar as Homeowners Confront Rising
Values.” Although property tax rates have generally
not increased in recent years, property tax revenue
has grown at a near 7 percent annual rate, mostly due
to a 40 percent appreciation in the median value of a
single-family home in Texas over the past six years.
As this issue reflects, we at the Dallas Fed are dedicated to producing economic research and analysis
that provide insights on key issues important to our
region and our nation. Through this work, we hope to
make sound monetary policy decisions and inform
other policymakers in a manner that improves the
economic prospects for our region and nation.

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

New Technology Boosts
Texas Firms' Output,
Alters Worker Mix
By Emily Kerr, Pia Orrenius and Christopher Slijk

}
ABSTRACT: A Federal
Reserve Bank of Dallas
survey of manufacturing
and services firms in
Texas found that as
companies adopt new
technologies, the number
of workers is little changed
though the employees’
skill levels shift. While
some manufacturers
see tighter margins as a
result of technology and
globalization, service
sector firms may realize
increased pricing power.

echnological change is the
economy’s greatest opportunity
and its greatest challenge. It affects
almost every aspect of economic activity, impacting outcomes for firms and
workers. Technological change is also
what economists believe drives productivity growth and, thus, higher standards
of living.
Nevertheless, such evolution doesn’t
come easy. During the Industrial Revolution, Luddites famously opposed the
introduction of new machines they felt
threatened their jobs. When it comes
to labor, technology can be a complement, as well as a substitute. Robots
and other automated factory tools substitute for labor on the assembly line.
However, these technologies complement workers who build, program and
repair this type of equipment.
Technological change can also
reach beyond the walls of the firm and
transform how companies interact with
workers and customers. Resulting efficiency gains can lower prices of goods
and services to the point that higher demand increases industry employment.
Ride-sharing platforms such as Uber
and Lyft have significantly lowered the
cost of travel, increasing ridership and,
hence, vehicle-for-hire employment.
The pace of technological change and
adoption varies over time. Research
suggests the aging of the labor force is
leading to an acceleration in automation technology investment and implementation as a substitute for the slower
growth of the prime-aged workforce.1
To gain insight into the role of
technology in business operations in
Texas, the Federal Reserve Bank of
Dallas queried more than 300 firms in
the manufacturing and service sectors
in June. Specifically, companies were
asked about the technologies they plan

to adopt or have already implemented,
why they undertook technological
change and the impact they expect on
firm employment and pricing power.2

Emerging Technologies
The Dallas Fed technology survey
looked at the emerging technologies
in Texas businesses—the ones only
narrowly in use now but on the brink
of wider adoption. When asked which
technologies firms plan to adopt within
the next three years, artificial intelligence was most often cited, followed by
3-D scanning, biometric authentication,
blockchain and 3-D printing (Table 1).
Further analysis shows that significantly more manufacturers than
services firms are planning technology
adoption in the near future (Chart 1).
More than one-fifth of manufacturers
plan to adopt 3-D scanning, a technology that captures a physical object’s
exact shape and specifications into a
digital 3-D representation.
3-D scanning has tremendous utility
in the manufacturing sector for reverse
engineering, product development
and quality control. Nearly one-fifth of
manufacturing firms plan to adopt 3-D
printing, a complementary technology
for prototyping and design iterations,
with additional uses for customization
and low-volume production.
A similar 20 percent of firms plan to
incorporate robotics into manufacturing processes in the near future, adding
to the 20 percent that have already
implemented it.
Among service sector firms, artificial
intelligence tops the list of emerging
technologies, with several companies
mentioning the use of machine-learning platforms for analytics and decision
insights. Biometric authentication—a
technology that can transform access

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

TABLE

Top 10 Emerging Technologies Among Texas Firms
Plan to
adopt (%)

In process
of adopting (%)

Already
adopted (%)

Artificial intelligence (voice recognition,
decision trees, autonomous vehicles, etc.)

3-D scanning

Biometric authentication

Blockchain

3-D printing

Big data

Robotics

Cloud computing/edge computing

Virtual reality/augmented reality

Digital currencies (cryptocurrency, bitcoin, etc.)

NOTES: Shown are the top 10 responses to the question, “Which of the following technologies has your firm already
adopted? Is your firm in the process of adopting? Does your firm plan to adopt within the next three years?” Data were
collected June 12–20, 2018, and 314 Texas business executives responded.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

CHART

Manufacturing Leads Services Firms in Technology Adoption

communication platforms such as
email and Skype, social media, highspeed internet, intranet networks and
mobile apps. Roughly half or more of
firms use these.

Why Firms Adopt New Technology

Overall Jobs Outlook

Adopting new technology is often
expensive and disruptive. Firms may
require financing, or they may draw on

Notably, technological adoption has
not appreciably changed the overall
employment outlook. The majority of

3-D scanning
3-D printing
Robotics
Artificial intelligence
Blockchain

Artificial intelligence
Biometric authentication
Blockchain
Big data
Cloud/edge computing

Manufacturing firms

Services firms

NOTES: Shown are the top five responses by firm type to the question, "Which of the following technologies does
your firm plan to adopt within the next three years?" Data were collected June 12–20, 2018, and 224 Texas service
sector executives and 90 Texas manufacturing executives responded.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

management for physical and digital
resources—is planned for adoption
at roughly 10 percent of firms. This is
followed closely by blockchain, the
decentralized digital ledger technology
underpinning cryptocurrencies such as
bitcoin, and big data.
In taking stock of the technologies
that Texas firms have already adopted,
the top responses are not surprising:

Employment Effects
The Dallas Fed survey next asked
how adoption of new technology will
affect firm employment over the next
five years. Interestingly, technology is
not expected to replace workers on net.
Only 14 percent of firms said technology adoption will decrease their need
for workers, and a similar share said
it will actually increase their need for
workers (Chart 2).
Half of firms expect no impact on
employment, and about a quarter of
firms said the adoption of technology
will change the type of workers needed
but not the number.
On the manufacturing side, Texas
business executives note that production is increasingly automated and
technology-dependent, shifting some
labor demand from blue-collar workers
to programmers, engineers, and robotics and/or computer design specialists.3
In the service sector, executives noted
a shift to workers who are more technologically adept—conversant in analytics, artificial intelligence platforms and
computer programming—and able to
handle more sophisticated demands.

Percent
25

savings. Installation of new equipment
may disrupt operations and likely requires retraining workers and spelling
out new processes. There is always the
risk that the new equipment will not
work as intended. Given the high cost
and uncertainty, the survey asked firms
why they change.
Raising productivity was the No. 1
response, cited as a main reason for
technology adoption by two-thirds of
firms (Table 2). Productivity means
doing more with less—producing more
output with the same or less input.
Services firms secondarily mentioned
remaining competitive and/or fending
off new market entrants as an impetus,
while manufacturing firms disproportionately mentioned lowering costs.
More than half of all respondents cited
increasing output.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

Texas firms surveyed—65 percent—
plan to add jobs over the next five
years. An additional 26 percent said
they would keep employment about
the same, and only 9 percent of firms
indicated they would decrease employment over the period.
Texas manufacturers are particularly bullish, with 78 percent expecting higher headcounts five years out.
Optimism among manufacturers is
likely helped by robust conditions in
the state’s energy industry, bolstered by
sustained, relatively high oil prices over
the past year or so.
Employment projections vary slightly
between large versus small firms.
Interestingly, nearly 20 percent of large
firms surveyed—ones with at least 500
employees—expect to pare headcounts
over the next five years compared with
just 7 percent of smaller firms (Chart 3).
Even still, nearly three-quarters of large
firms plan to increase employment.

Broader Trends, Pricing Power
Globalization and technological
change are two pervasive forces that
define our economic times. Since the
fall of the Berlin Wall in 1989 and the
end of the Cold War, international
trade and exchange have surged. The
expansion of global economic activity
in developing countries has led to falling poverty and other improvements
for some of the world’s poorest populations. But there have also been costs.
Manufacturers in advanced economies, including the U.S., have sustained steep declines in employment.4
Thanks to technological improvement,
however, manufacturing output has
continued to grow.
Against the backdrop of these broader trends in the 21st century, the Dallas
Fed survey asked how these long-term
industry trends—technological change
and globalization—have affected firms’
ability to pass on cost increases to customers over the past five years. About
half of firms noted there was no net
effect on pricing power.
Among the remainder, the breakdown of positive and negative impacts
varied between the service sector and
manufacturing firms (Chart 4). Service

TABLE

Firms Adopt Technology Mainly to Raise Productivity
All Firms
(%) (295)

Services firms
(%) (209)

Mfg. firms
(%) (86)

Raise productivity

Remain competitive/fend off new competitors

Increase output (revenue/sales/production)

Lower costs

Expand into new business lines/markets

Strengthen security and/or protect information

Meet industry standards/government regulations

NOTES: Shown are responses to the question, “What are the main reasons why your firm is adopting these
technologies? You may select up to three.” Data were collected June 12–20, 2018. Figures in parentheses represent the
number of Texas business executives responding.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

CHART

Technology Affects Type, Not Number of Workers Needed
On net, how will the adoption of these technologies affect
employment at your firm over the next five years?

Increases
need for
workers
12%
Changes the
type of workers
needed, not
the number
25%

Decreases
need for
workers
14%

Does not impact
need for workers
49%

NOTE: Data were collected June 12–20, 2018, and 296 Texas business executives responded.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

sector companies were more likely
to report increased pricing power (24
percent) than decreased (19 percent).
Respondents pointed to technology as
key to their ability to raise prices.
A commercial heating, ventilation
and air conditioning company noted:
“We have bought some industry-specific customized tools that allow us to
complete repairs on equipment much
faster than our competitors; we charge
for this since there is a benefit of decreased downtime to our customer.”

Several services firms also touched
upon the significant value in data
analytics—an office moving company
reported that “technology now allows
us to have immediately available metrics to price to a standard and price to
demand. … When costs go up, we can
model what cost sharing we can push
through to our customers.” A law firm
mentioned that because of cost modeling, it shifted from a billable-hours
pricing model to a value-added model,
allowing the firm to capitalize on the

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

CHART

Employment Expectations More Polarized Among Large Firms

Overall, do you expect your firm to increase employment, leave employment
unchanged, or decrease employment over the next five years?
Percent
80

Increase
Leave unchanged
Decrease

70
60
50
40
30
20
10
0

1–499 employees

500+ employees

NOTE: Data were collected June 12–20, 2018, and Texas business executives from 248 small firms and 49 large
firms responded.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

CHART

Technology, Globalization Lift Pricing Power
for Services, Not Manufacturing

On net, how have long-term trends in your industry, such as technological change
and globalization, affected your firm's ability to pass on cost increases to customers
over the past five years?
Increased
Leave unchanged
Decreased

Percent
60
50

The June 2018 Dallas Fed technology survey yielded unique insights into
what technology Texas firms are adopting and why, as well as how they view
their long-term prospects. Firms adopt
new technology to increase productivity and, as a result, their long-term
employment prospects remain bullish.
Texas firms are not adopting technology to shed workers, although onequarter of respondents said adopting
new technology changes the types of
workers needed.
Policymakers and education and
workforce experts should take note:
Employment will continue growing
in Texas firms but the type of skills in
demand is evolving. Just as firms must
be agile and ready to adopt new technology, workers have to be flexible and
attentive to changing job market needs.
Kerr is a senior business economist,
Orrenius is a vice president and senior
economist, and Slijk is an assistant
economist in the Research Department
at the Federal Reserve Bank of Dallas.

“Demographics and Automation,” by Daron Acemoglu
and Pascual Restrepo, National Bureau of Economic
Research, Working Paper no. 24421, March 2018.
2
For complete results, see www.dallasfed.org/research/
surveys/tssos/2018/1806/specquest.aspx. The survey
was a follow-up to the conference, "Technology-Enabled
Disruption: Implications for Business, Labor Markets
and Monetary Policy," held in May at the Federal Reserve
Bank of Dallas and cohosted by the Federal Reserve Bank
of Atlanta. For conference program and agenda, see www.
dallasfed.org/research/events/2018/18ted.
3
A follow-up anecdotal survey was conducted via email
to gather further insights from firms on how technology
affects the type of worker needed and how technological
change and globalization impact pricing power. Twentytwo business executives submitted responses June
29–July 9, 2018. Comments from this survey are referred
to here and as anecdotes in this article.
4
Since 1990, manufacturing employment in Texas
has declined just 10 percent, compared with nearly 30
percent for the U.S. as a whole.
1

30
20
10

All firms

Service sector firms

Manufacturing firms

NOTE: Data were collected June 12–20, 2018, and 203 Texas service sector executives and 83 Texas manufacturing
executives responded.
SOURCE: Dallas Fed Texas Business Outlook Surveys.

time savings of automating repetitivetype work and other business efficiencies technology has prompted.
Conversely, among manufacturers, an outsized share experienced
a declining ability to raise prices, a
response consistent with greater exposure to international competition and
surging imports from China. A textile
manufacturer commented, “Our clients

Increased Firm Productivity

Notes

foreign companies to provide products
similar to ours. They use that information as leverage to keep my prices low.”

are getting very aggressive in sourcing
from all corners of the globe.” A fabricated metals producer mentioned that
“Our domestic customers have many
more options to find lower-priced
products in the international marketplace than ever before and, with the
internet, can find those options easily.”
A high-tech producer said, “My
customers are being approached by

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

Texas Property Taxes Soar as
Homeowners Confront Rising Values
By Jason Saving

}

ABSTRACT: A precipitous
rise in the amount of
property taxes Texans
pay has accompanied
an uncharacteristically
large increase in
property tax valuations.
Because a variety of
local jurisdictions provide
services that elsewhere
are state responsibilities—
particularly public
education—there are
limited ways to rein in
rising property taxes
across Texas.

roperty taxes in Texas have risen
markedly in recent years. Complaints about soaring appraisals
have become ubiquitous as have calls
from the public to “do something”
about ever-rising property tax bills.
The situation has begged the question
of whether these developments have
jeopardized Texas’ status as a relatively
low-tax state, potentially harming longterm economic growth.
Yet Texas is distinctive as one of only
seven states without an income tax at
either the state or the local level, suggesting that sales and property taxes
might be somewhat higher in Texas
than elsewhere.
The amount of property taxes has
jumped in recent years because of not
only tax rate changes, but also rapidly
rising home prices—a product of people
having more housing wealth. The increase has created a sense that the total
state and local tax burden in Texas is no
longer competitive with taxes in other
states, even as the total burden today remains well below the national average.
Still, property tax rates are comparatively high in Texas and pose a greater
burden as personal income rises more
slowly than property values, raising
questions about both the economic arguments for property taxation and the
implications of reducing that burden.

Relatively High Burden
The first step in assessing the
property tax situation is determining
how much higher property tax rates
are in Texas than elsewhere. In 2016,
Texas’ average property tax rate of
1.86 percent was the sixth-highest in
the nation, over 50 percent more than
the national median of 1.19 percent
(Chart 1).1 For a $250,000 house, this
translates into a tax payment of $4,650,
compared with the national average

of $2,975—a sizable burden in a state
whose average income remains slightly
below the national average.
An examination of where all of this
money goes and who imposes property
taxes in Texas sheds additional light.
Numerous local taxing entities provide a wide variety of services. School
districts are perhaps the best known.
Fifty-four percent of Texas property
taxes were paid to school districts in
fiscal 2015, the last year for which full
data are available (Chart 2A).
Property taxes are also levied by counties (17 percent of the burden), cities
(16 percent), special-purpose districts
such as hospital and utility districts,
community college districts, water districts, development/improvement districts and emergency-services districts.

Multiple Local Entities
All told, more than 4,000 local government entities collect property taxes
in Texas. By law, payments must be
based on the current assessed value of
property, though there are exceptions
for property owners’ primary residence
(homestead exemptions), land used for
designated purposes (such as agriculture, which is sometimes eligible to
be taxed at a lower rate) and property
owned by certain people (such as
seniors, who are sometimes eligible to
freeze their payment levels).2
Single-family homes represent 51 percent of the state’s total property tax base,
with commercial and industrial businesses composing another 35 percent
and multifamily residences 6 percent.
In fiscal 2015, property taxes made
up about 42 percent of Texans’ total
state and local tax burden.3 While less
than the 50 percent for sales and use
taxes, it is significant (Chart 2B).
Property taxes are not only high, but
also rapidly rising in the state. Following

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

CHART

Median Property Tax Rate in Texas Exceeds National Average

Percent
3.0

2.5

2.0

Texas

1.5
U.S.
1.0

0.5

NJ
NH
IL
CT
WI
TX
NE
VT
MI
NY
RI
OH
PA
IA
KS
SD
ME
MA
U.S.
AK
MN
MD
OR
WA
ND
FL
MO
GE
OK
IN
NC
KY
MT
MS
VA
CA
AZ
NV
NM
ID
TN
UT
AR
WY
WV
CO
SC
DC
DE
LA
AL
HI

0.0

NOTE: Property tax is calculated by dividing the median real estate taxes paid by the median home price.
SOURCE: Census Bureau, 2012–16 American Community Survey 5-year estimates.

a three-year lull during and immediately after the Great Recession, property
tax revenue has grown at a 6.9 percent
annual rate since 2012, with the proceeds distributed fairly evenly among
schools, counties, cities and other taxing districts (Table 1).
While the available data stop at 2015,
anecdotal evidence suggests the trend
continued in subsequent years. By
comparison, household income grew at
a 2.7 percent annualized rate in 2012–15
and a 2.9 percent rate in 2016–17, adding to the perception that Texans are increasingly burdened by property taxes.

Rising Property Values
Property tax burdens have increased
rapidly in recent years. While it’s possible that tax authorities have raised
rates so quickly that revenue growth
has outstripped home prices, in reality
overall property tax revenue growth for
Texas jurisdictions has actually trailed
real estate price appreciation in recent
years. Annual tax revenue growth trailed
appreciation by 0.9 percentage points in
2012–15 (and the trend has likely continued). Some jurisdictions no doubt
markedly raised their rates, but a better

explanation is needed for why property
tax revenue increased so quickly.
If higher rates can't fully explain rising property taxes, perhaps increased
home valuations can. Chart 3 suggests home values have, indeed, risen
rapidly. Over the past six years, the
Texas median home price has jumped
nearly 40 percent, in line with trends at
the national level. But because Texas
has historically relied more heavily on
property taxes than the national average, Texans are more directly affected
than their counterparts who have low
or no property taxes.
One other factor of note is that recent home-price movements in Texas
have been unusual. Typically, large
swings in national home prices yield
only modest changes in Texas because
the state’s lax zoning, plentiful land, flat
geography and robust economy have
tended to ensure enough supply will be
built to keep pace with demand.
The boom–bust cycle of 2002–08
illustrates this phenomenon. Texas
home prices barely budged as national
home prices appreciated 30 percent
in the first few years of the period and
then fell about 40 percent. Texas’ high

property tax rates didn’t attract a lot of
attention over the period because assessments weren’t growing rapidly.
For reasons not fully understood, this
pattern has been broken in the aftermath of the Great Recession, with both
state and national home prices rising
30 to 40 percent. Does this unusual Texas home-price appreciation signify a
change in fundamentals, such that the
state will experience boom–bust cycles
from which it had been excluded? Or
do the housing data simply reflect other, secular forces that are temporarily
prompting a rise in Texas home prices
that just happens to coincide with the
current national housing boom?
Factors such as the Dodd–Frank
Wall Street Reform and Consumer
Protection Act that affect lending for
housing development, the gradual
de-localization of housing finance
and the appearance of lot-availability
constraints in major Texas metros suggest housing markets may more closely
follow national home price trends than
they once did. Still, it remains far from
certain to what degree the next national housing bust might impact Texas
real estate prices.4

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

Ranking Texas Taxes

CHART

Whatever the long-term outlook, the
current cycle of home-price appreciation has dramatically boosted property
tax revenue. And with property taxes
growing at a rapid pace that is faster
than income, it may be tempting to
conclude that Texas’ low-tax reputation
is now more myth than fact. Such a conclusion would be premature, though.
In the last year for which full data are
available, Texas’ state and local tax burden was 15 percent below the national
average, 30 percent below California’s
and 55 percent below New York’s on a
per capita basis (Chart 4). The differences are even starker when only the state
portion of the burden is considered.
However, the local portion of that
tax burden tells a different story. Texas’
per capita local tax burden ($2,116 per
year) is actually slightly higher than the
national average and, perhaps surprisingly, higher than that in high-tax
California. Measuring the total local tax
burden rather than median property
taxes alone, as seen in Chart 1, reveals
that Texas property taxes are indeed
high by comparison with other states
that may levy other types of local taxes.
This comparative Texas burden may be
why property taxes have become a focal
point of attention in recent years.
Why, then, are there so many types of
local jurisdictions in Texas and why do
they need to raise so much revenue?
There are a couple of reasons. First,
Texas has historically delegated significantly more responsibilities to localities than other states, allowing cities,
counties and school districts to provide
services that are elsewhere handled at
the state level. Such decentralization
stems from a historic distrust of any
single center of power. This is visible in
requirements that are unnecessary in
other states, such as the election (versus appointment) of executive branch
officials and public referendums to
ratify some legislative measures.
As a corollary, Texas transfers a relatively small amount of state revenue to
localities, requiring local jurisdictions
to raise revenue themselves. In Texas,
local governments receive only 23
percent of their revenue from the state;

School Districts Receive Half of Texas Property Tax Revenue;
Property Tax Share Large for Individual Taxpayers
A. Recipients by Taxing Authority

Special
13%

City
16%
School district
54%
County
17%

B. Taxpayer Outlays by Tax Type

Other
7.5%

Property
42%

Sales*
50.5%

NOTES: Property tax data are from fiscal year 2015. Shares are based on the taxable value assigned.
SOURCE: Texas Comptroller of Public Accounts.

in only six other states do localities
receive a smaller percentage (Chart 5).
The U.S. average is 30 percent. Buffeted
by the combination of more responsibility and less state support, the local
tax burden in Texas is relatively high.

Local Property Taxation
This doesn’t mean the property
tax specifically should be the vehicle
through which local revenue is raised,
though some economic arguments
favor locally administered property

taxes. A central argument from the
economic literature is that houses can’t
readily be moved from one jurisdiction
to another, which makes tax avoidance
less of an issue than it would be for, say,
a locally imposed income tax.
Another argument is that property
taxes don’t directly discourage productive activity as income taxes sometimes
do. This by no means implies it is impossible to have a dynamic local area
with such taxes in place, as California’s
Silicon Valley demonstrates. Neverthe-

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

CHART

Texas Median Home Values Have Risen Dramatically
Since 2012, in Sync with Nation

Thousands of real dollars*
350
300

U.S.
$259,081

250
200

Texas
$228,137

150
100
50
0

’00

’02

’04

’06

’08

’10

’12

’14

’16

’18

*Four-month moving average; seasonally adjusted.
NOTE: Data are through April 2018.
SOURCES: Multiple Listing Service; Texas A&M Real Estate Center.

CHART

Total Texas Tax Burden Below U.S. Average;
Local Taxes Alone Tell a Different Story

2015 taxes per capita, dollars
10,000
9,000
8,000

State
Local

7,000

3,946

6,000
5,000
4,000
3,000

2,007
1,836

2,000
1,000

3,875

2,839

2,043

1,612

National average

Florida

4,765
2,116

1,985

Texas

California

New York

SOURCES: Census Bureau; Bureau of Economic Analysis.

less, efficiency arguments suggest income taxes are more appropriate at the
federal level, while property taxes may
be more effectively assessed locally.
Nonetheless, property taxes hit only
one type of asset—housing wealth—
and therefore discriminate against
people who choose to spend their
money on a larger house rather than,

for example, better cars or travel. Is it
fair to tax the one person more heavily
than the other just because their housing preferences are different? It’s also
possible that property taxes discourage
housing consumption to some degree,
though this effect is likely much smaller
when it comes to housing than it would
be for, say, a tax on stamp/coin collec-

tions. There are tax benefits associated
with homeownership, after all, and people cannot easily do without housing.
Local property taxes also enable
some jurisdictions to spend more than
others on public services, if they so
choose. To be sure, it is efficient for
individuals to be able to compare jurisdictions and live in the one whose mix
of taxes and services best matches their
own preferences.
However, when applied to K–12
education in particular—the largest
local program—large differences in
spending per student could potentially
perpetuate patterns of inequality,
leaving the children of poor parents
with less human capital (and lower
future salaries) than their wealthier
peers. Texas mitigates this issue to
some degree with its so-called Robin
Hood system, under which a portion of
school property tax revenue is redistributed to poorer jurisdictions, though
that system is itself controversial.5
A final issue is that an individual’s
property tax burden can rise dramatically when neighborhood property values spike, leading to a sudden unwelcome increase in tax liability for which
the homeowner may be unprepared.
People who cannot pay may be forced
to sell their homes.
These surges can be dealt with by
“circuit breakers” that phase in appreciation over several years, though phase
ins inevitably reduce revenue available
to local jurisdictions and may, depending on the circumstance, make it more
difficult for those jurisdictions to provide the services residents expect.
Texas also limits annual tax rate
increases by allowing voters to hold
a rollback election. While these rules
are complicated and vary depending
on the taxing authority, if the tax rate
increase exceeds a certain percentage,
property owners may elect to reduce
the rate increase in a given year.

Imperfect Funding Method
The bottom line: Property taxes are
an imperfect way to raise revenue. For
this reason, some states emphasize
sales and income taxes over property
taxes. However, Texas’ sales tax rate

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

CHART

State Transfers Provide Relatively Low Share of Local Revenue in Texas

Percent of revenue from home state
60

40
U.S. average: 30 percent

Texas
44th

VT
AR
NM
ND
MI
WI
DE
MN
AK
ID
WY
MS
NV
MT
CA
IN
WV
KS
IA
PA
OR
OH
KY
VA
OK
NC
AZ
UT
AL
LA
NY
WA
ME
CT
MA
MD
IL
NJ
RI
GA
SC
MO
NH
TX
CO
TN
SD
FL
NE
HI

NOTE: Texas localities received 22.9 percent of total revenue from the state in fiscal year 2015.
SOURCE: Census Bureau.

is already among the highest in the
country and has been shown to disproportionately burden the poor, while an
income tax is constitutionally prohibited in Texas and would risk discouraging
work and investment were it somehow
implemented.6 Any taxing system
comes with its own set of advantages
and disadvantages.
Alternatively, property taxes could
be cut without raising taxes elsewhere.
However, significant property-tax-funded functions such as K–12 education,
already well below the national average
in terms of per capita funding, would
fall further, potentially reducing the
quality and quantity of those services.
One solution might be to pair local
property tax cuts with increased state
transfers, though those transfers would
themselves have to be funded through
service reductions or higher taxes at
the state level. This doesn’t automatically make efforts to rein in property
tax growth a “bad” thing, but it does reinforce the need to carefully weigh the
economic arguments, fully cognizant
of both residential tax burdens and
desired levels of government services.

It is eminently possible to address
Texas’ relatively high property tax burden, but doing so inevitably imposes
sacrifices on some, while potentially affecting state and local tax progressivity
and perhaps even future growth rates.
It is also possible that the marketplace will address the issue through a
housing contraction, but that would
dramatically lower home valuations
across the state. Were that to happen,
today’s higher property taxes caused
by soaring home valuations might not
seem like such a bad thing.

Bank of Dallas Southwest Economy, First Quarter, 2017.
5
For example, the Robin Hood system partially but not
fully equalizes per-student funding across jurisdictions,
leaving both donor and recipient districts unsatisfied
with the outcome. Many also argue the system amounts
to a de facto statewide property tax, though the state
Supreme Court ended a lengthy legal battle last year
by affirming its constitutionality. For a more thorough
discussion of these and other economic issues, see
“Improving School Finance in Texas,” by Jason Saving,
Fiona Sigalla and Lori L. Taylor, Federal Reserve Bank of
Dallas Southwest Economy, no. 6, 2001.
6
See “Texas Taxes: Who Bears the Burden,” by Jason
Saving, Federal Reserve Bank of Dallas Southwest
Economy, Third Quarter, 2017.

Saving is a senior research economist
and advisor in the Research
Department at the Federal Reserve
Bank of Dallas.

Notes
Based on the Census Bureau’s American Community
Survey 5-year estimate for 2016.
2
Government-owned facilities are also generally exempt
from property taxation.
3
Texas fiscal years begin Sept. 1. Thus, fiscal 2015
began Sept. 1, 2014.
4
See “Texas Housing Market Soars to New Heights,
Pricing Out Many,” by Laila Assanie, Federal Reserve
1

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

ON THE RECORD
A Conversation with Mine Yücel

Shale Renews Excitement
in Energy Industry
Mine Yücel, a senior vice president and research advisor at the
Federal Reserve Bank of Dallas, leads the Research Department’s
energy group. She joined the bank in 1989 and is a past president
of the International Association for Energy Economics. She
discusses the evolution of the energy sector, its role in Texas and
her perspective as a female energy economist.

Q. What is energy economics and how
does it differ from the study of other
commodities?

natural gas markets and could feel the
economic effects of changing natural
gas prices.

Energy—whether it be oil, gas, coal
or renewables—is an essential input for
the economy. Energy prices affect all
sectors of the economy, from businesses
to households. And because the use of
energy is so pervasive, energy supply
and energy policy have been frequently
deemed important for national security
reasons. This was especially true in the
1970s when U.S. oil production started
declining and oil prices soared due to
geopolitical factors.
When looking at energy markets,
oil really stands out. Oil is priced in
international markets, but the market
is not necessarily competitive. OPEC
controls 40 percent of the global crude
oil market and can influence oil prices.
In the U.S., gasoline and diesel make
up 67 percent of oil consumption, so a
change in oil prices affects consumer
spending directly through fuel and heating oil prices. Oil price changes also
have historically affected gross domestic
product (GDP) growth and inflation,
and, therefore, have been an important
consideration for monetary policy.
We have long had a domestic market
for natural gas. But as U.S. natural gas
production continues to rise, and U.S.
liquefied natural gas exports increase,
we will be more integrated with global

Q. When you started working in
energy economics in the 1980s, what
were the big questions? What are
they now?
I graduated in the early 1980s. Research focused on OPEC, its impact on
oil markets and prices, and the effects of
high oil prices on the macroeconomy.
As OPEC grew and started flexing its
muscles, the price of oil shot up from
$3.50 per barrel in 1973 to around $37
by 1980. There was much research on
OPEC market structure: Was it a cartel,
was it targeting market share or targeting revenue?
Oil prices started declining in 1981
and collapsed to near $14 in 1986. Then,
there were questions about whether the
impact of oil price changes was symmetric—that is, would falling oil prices benefit the economy, just as rising prices
had hurt the economy? There were also
questions about the health of oil-producing states such as Texas, Alaska and
Louisiana whose economies were very
dependent on the oil industry.
Interestingly, we are still researching
these same questions. As the economy
has evolved, questions about the impact of oil price shocks on the economy
remain an ongoing concern, but the

answers have changed somewhat. The
source of the oil price shocks matters,
and shocks have a smaller impact. Also,
as the share of renewables such as solar
and wind has increased, issues about
how to integrate renewables into the
electricity grid have come into focus.

Q. Oil price increases have seemed
to always precede recessions. Is that
still true?
Oil price shocks have preceded 10
of the 11 post-World War II recessions
in the U.S. However, this is not a causal
relationship. The recessions were not
necessarily caused by oil price shocks.
Rather, high oil prices most likely contributed to the weakness of an already
fragile economy before the recessions.
Oil prices affect the economy through
two channels: a price effect and an
allocative effect. For example, when
the price of oil goes up, we all feel the
impact of higher gasoline prices and
higher fuel prices and lower our spending accordingly. This is the negative
price effect.
Because oil and goods and services
related to oil have now become relatively more expensive, there is also a change
in the use of resources that produce
these goods and services—a reallocation of resources because of the relative
price change. This reallocation effect is
also negative.
These two effects are why the economy is affected negatively when oil prices
rise. Of course, producers of oil are now
better off because of the increase. This
positive impact offsets some of the negative effects from the oil price rise.

Q. How has energy’s impact on the
Texas economy changed in your time?
I came to Texas in 1977. We were in
the midst of an oil boom, and the state’s
economy was growing at double-digit
rates. Texas went into a deep recession
when oil prices started declining in 1982
and again when they collapsed in 1986.
The oil and gas industry lost roughly
150,000 jobs, about 2.2 percent of employment, from the peak of the boom in
1982 to early 1987. The second recession

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

drilling and hydraulic fracturing have
} “Horizontal
been amazing technological developments that
completely changed the oil and gas picture in
the U.S.”

coincided with the 1986 Tax Reform Act’s
more stringent treatment of investment
income and the savings and loan crisis,
which also buffeted the Texas economy.
After the bust, the Texas economy diversified, and low oil prices have much
less of an impact on the Texas economy
now. When oil and gas prices collapsed
at the end of 2014, Texas didn’t fall into
recession unlike all other states with
large oil sectors, such as Alaska, North
Dakota, Oklahoma and Louisiana. During the most recent shale oil boom, increased oil production boosted the share
of oil and gas in overall Texas GDP, but
the share of employment has remained
low, reaching a high of only 2.5 percent.

Q. What’s the most exciting oil and
gas industry change you have seen in
your 30-plus-year career?
The most exciting development has
been the shale revolution. U.S. crude
oil imports topped 10 million barrels
per day in 2004, almost twice what we
produced. U.S. crude production had
been declining since the peak in 1970,
and there was much discussion about
whether we were running out of oil.
Horizontal drilling and hydraulic fracturing have been amazing technological
developments that completely changed
the oil and gas picture in the U.S. and the
dynamics of the global oil market.
Now, the U.S. is producing nearly 11
million barrels per day, higher than our
record in 1970. Since 2009, we have increased production by more than 5 million barrels per day. This is basically adding another Iraq into the global oil market.
Technology has helped with energy research as well. The availability of data and
the ease of procuring energy data have

been an immense benefit to research in
energy. This is true for all research fields
of course, not just energy. Improved
econometric techniques are also very
useful in parsing out the impacts of
changes in energy markets and prices.

Q. How do you see energy markets in
the future?
I think we will see increased use of
renewables in the future. How close is
that future, though? Renewables, such
as wind and solar, are used in electricity
generation but are still a small part of
our energy mix. Currently, renewables
account for 13 percent of U.S. energy
production. There are a couple of factors
hindering renewables in the short run.
One is the problem of intermittency. For
renewables to gain wider market share,
we need better storage technology. Battery technology has been improving, but
we’re not there yet.
Another factor: Seventy-one percent
of oil is used in transportation, according to the U.S. Energy Information Administration. Electric vehicles have been
making inroads, but again, battery technology and infrastructure are limiting
factors in the short run. However, France
and the U.K. have said that they will ban
gasoline and diesel cars by 2040, and
India has also declared that all new cars
after 2030 will be nonfossil fuel. Such
policies may hasten the inroads that
electric cars will make.

Q. Which one of your many
accomplishments stands out?

ics (IAEE) in 2011. The association was
founded after the 1970s energy crisis. It
is a worldwide organization that has affiliations and members in more than 100
countries. I’ve been going to IAEE meetings since 1986 and have probably met
all the prominent energy economists in
the world at the conferences. I learned
a lot being a part of the IAEE and made
lifelong friends from all over the world.

Q. You are a prominent female
economist in a male-dominated field.
What advice do you give?
Being a woman in the energy field
may have initially been somewhat of a
hindrance. The profession has slowly
changed though. When I first started
going to energy conferences in the
mid-’80s, there were only a handful of
women in the profession. We all knew
each other quite well.
There are many more women now,
and many young women are entering
the profession because it is such an
interesting area of study. But it is still a
relatively smaller number than the men.
My advice to young economists would
be to do your research and get published
in peer-reviewed journals. That is what
gets you the respect in the profession.
Go to conferences to present your work;
get to know people in your area. That is
how your work gets noticed. Ask the interesting questions. And work on issues
that have broad relevance for industry
and academia.

My proudest accomplishment was
becoming the president of the International Association for Energy Econom-

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

Parental Borrowing for College
Comes with Repayment Issues
By Wenhua Di, Carla Fletcher and Jeff Webster

}

ABSTRACT: As the cost of
college continues to rise,
parents are increasingly
taking out federally backed
loans to help make ends
meet for their children.
Parents, while often
more adept at managing
debt, assume some of
the financial risks of their
offspring who are seeking
higher education.

ollege is an investment that generally yields benefits over a student’s
lifetime in the form of higher
wages, more stable employment and
better benefits. Typically, parents want
to help their children with college costs,
but they don’t always have enough savings to do so.
Since the 1950s, federal government
student loan programs have encouraged postsecondary education. More
recently, federal assistance for parents
wanting to help their children with
college costs came in the form of the
federal Parent Loan for Undergraduate Students (PLUS) program. It was
created in 1980 and assists parents who
are borrowing for their offspring’s college expenses.
Parents have increasingly taken out
PLUS loans, with the average amounts
borrowed growing.1 The parent loan
default rate remains low, though signs
of it increasing have appeared.2
Parents’ repayment behavior differs
from that of students, with parent borrowers presenting their own benefits
and risks. Parent borrowers tend to
have more experience dealing with
debt and more realistic expectations
for repayment than students. At the
same time, parents say taking on loans
for their children may affect their ability to save for retirement and undertake
major purchases.

PLUS Program Growth
Most federal education loans are
loans to students. Stafford loans make
up the largest portion of the borrowing.
As of second quarter 2018, there were
29.5 million subsidized Stafford loan
recipients (receiving relatively favorable terms) and 28.3 million unsubsidized recipients, together representing
$753 billion of the $1.4 trillion out-

standing federal education loans.3 The
approximately 3.5 million parent PLUS
borrowers (8.2 percent of all federal education loan borrowers) were responsible for $87.7 billion, or 6.2 percent of
the outstanding loan debt.
Stafford loans (named after former
Vermont Sen. Robert Stafford) are
based on the level of financial need
calculated using data supplied by
students through the Free Application
for Federal Student Aid.4 Stafford loans,
available to borrowers regardless of
credit score, usually carry lower interest rates than private loans. They also
offer various borrower protections such
as hardship deferments, forbearance,
income-driven repayment options and
public service loan forgiveness.
Stafford loans have annual and aggregate borrowing limits. With rising
college prices and high financial need
among students from middle- and
lower-income families, there are often
substantial gaps students must fill
through savings, paid work and contributions from family and friends.
PLUS loans carry higher interest
rates than Stafford loans and are intended for families who have exhausted student borrowing options.5 PLUS
borrowing limits were modified in 1992
to offer greater flexibility. Parents were
subsequently allowed to borrow up to
the difference between the total cost
of attendance and the amount of other
financial aid, regardless of expected
family contribution, as long as the
parental borrowers did not have an
adverse credit history.6 This modification typically provided parents with the
ability to borrow much larger amounts.
Although PLUS borrowers are fully
responsible for loan repayment, many
proceed because they have altruistic
motives.7 College education typically

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

CHART

Parents' Share of Undergraduate Federal Loan Programs Increases

Billions of dollars
120

Subsidized
Unsubsidized
PLUS

100

’16–’17

’15–’16

’14–’15

’13–’14

’12–’13

’11–’12

’10–’11

’09–’10

’08–’09

’07–’08

’06–’07

’05–’06

’04–’05

’03–’04

’02–’03

’01–’02

’00–’01

’99–’00

’98–’99

’97–’98

’96–’97

NOTE: Each time period refers to a school year.
SOURCE: College Board, Trends in Student Aid 2017.

leads to a host of financial and other
lifetime benefits.8 There could be some
net gains for parents as well. Parents’
net lifetime income may increase as
a result of incurring PLUS debt—if a
child completes a college degree, the
subsequent higher income may offset
the need for other future support from
parents and allow contributions from
children to parents in old age.

Greater Parental Borrowing
PLUS loans comprise an increasing
proportion of federal aid to students
and their families. About 8.6 percent
of the $42.1 billion (in 2016 dollars)
in undergraduate loans originated in
the 1996–97 academic year were PLUS
loans (Chart 1). The share rose to 15
percent of $84.2 billion in the 2016–17
academic year. The $15,878 average
parent loan was $6,251 more than two
decades earlier—much greater borrowing than the average amount of Stafford
subsidized or unsubsidized loans.9
Like other federal education loans,
PLUS loans are usually nondischargeable in bankruptcy. Borrowers may
also have their wages, tax refund and
Social Security benefits garnished if
they default on the loans.

Deteriorating Loan Repayment
Parents can potentially access either
a federal PLUS loan or a private loan.
However, parents with lower credit
scores can’t easily obtain a private
loan, which involves more rigorous
underwriting. So, while PLUS loans are
not need-based and were designed to
support education for families of any
income level, they tend to attract lowerincome borrowers and those who can’t
qualify for private-lender funding.
This “adverse selection” of borrowers
into the PLUS program became more
apparent when conventional underwriting tightened following the Great
Recession.10

TABLE

The U.S. Department of Education
has published default rates for PLUS
loans for fiscal 2006 to 2010 (Table 1).11
The overall default rate increased from
1.8 percent in fiscal 2006 to 5.1 percent in fiscal 2010. The rate more than
doubled for loans involving students
enrolled in proprietary, private nonprofit and public institutions during
the period, with the rate at proprietary
institutions being the highest.12
In response, the Department of
Education tightened the parent PLUS
credit check rules in October 2011.
Loan denials increased 10 percentage
points the following year.13 The denial
rate is also linked to a steep enrollment

Parent PLUS Loan Defaults Increase Throughout Recession

Three-Year Cohort Default Rates

Fiscal year Fiscal year Fiscal year Fiscal year Fiscal year
2006
2007
2008
2009
2010

Parent PLUS overall (%)

1.8

2.2

2.6

3.4

5.1

Parent PLUS proprietary (%)

4.7

5.5

6.3

8.3

13.3

Parent PLUS private nonprofit (%)

1.2

1.6

2.0

2.5

3.4

Parent PLUS public (%)

1.2

1.6

1.9

2.2

3.1

NOTE: Rates are calculated based on borrowers entering repayment after in-school deferment. Proprietary institutions
are generally for-profit private schools.
SOURCE: U.S. Department of Education, 2012.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

TABLE

Loan Characteristics and Education Experience
by Parent PLUS Loan Default Status*
Variables

Borrowers not in
default

Borrowers in
default

Avg. number of PLUS loans per borrower

2.0

1.5

Highest interest among PLUS borrowings (%)

7.4

7.2

19,509

12,403

PLUS Loan Characteristics

PLUS loan beginning balance ($)

PLUS Loan Performance
PLUS loan amount paid down ($)

8,080

109

Delinquency (%)

26.4

96.3

Deferment (%)

13.2

16.0

Forbearance (%)

31.8

56.5

Student Borrowing, Enrollment and
Education Attainment
Children’s Stafford loan amount ($)

18,831

17,015

Second year funded by PLUS loan (%)

19.3

20.2

Third year funded by PLUS loan (%)

17.8

13.5

Fourth year funded by PLUS loan (%)

24.6

12.6

1.5

0.7

Fifth year funded by PLUS loan (%)
Two-year public (%)

3.7

5.0

Four-year public (%)

64.4

57.5

Four-year private (%)

22.4

20.1

Proprietary (%)

5.0

14.8

Minority-serving institution (%)

30.2

44.0

Graduated (%)

48.8

36.8

Withdrawn (%)

25.1

37.1

*Based on borrowers entering repayment in fiscal 2005–10.
NOTE: Loans from Trellis Co. portfolio, shown in nominal dollars.
SOURCES: Trellis Co.; authors’ calculations.

decline in 2011 among historically
black colleges and universities, which
tend to be low-resourced schools with
limited institutional grant funds to
support their disproportionately large
economically disadvantaged student
populations.14
With the policy change, many
parents who relied on the PLUS loans
were shut out. Officials later loosened
the standard to allow greater participation. Still, families with large unmet
need using PLUS are the most likely to
struggle with repayment.15

Loan Outcomes Examined
Studies of PLUS loans have been
limited, largely because of the relatively small share of PLUS loans in the
student loan market and the gener-

ally lower (albeit rising) default rate.
Trends, patterns and the experiences of
parental borrowing emerge in administrative data of PLUS borrowers assembled by the Trellis Co., a nonprofit
student loan guarantor that has helped
administer the Federal Family Education Loan Program in Texas since 1979.
The dataset covers 62,449 parent
PLUS recipients who entered repayment between October 2004 and September 2010, with children attending
Texas institutions.
Trellis’ data track borrowers’ repayment behavior from the beginning of
repayment and continuing for the next
seven years or until the loans were
paid in full, consolidated and changed
guarantor, or the borrower defaulted.
About 8.6 percent of these PLUS bor-

rowers defaulted during the seven-year
period. Those who defaulted obtained
fewer loans with smaller beginning
balances, paid down less of the balance
and had higher levels of delinquency,
deferment and forbearance than those
not in default (Table 2).
Parents who defaulted mostly supported students who took more time
attending school.16 Relative to borrowers not in default, the children of those
in arrears were more likely to enroll in a
two-year public college, a for-profit proprietary (private) school or a minorityserving institution and less likely to
attend a four-year public or nonprofit
private college and to have graduated.17
Multivariate statistical models—a
means of examining the interplay
between several variables and an
outcome—were developed to examine
how some of these factors explain the
likelihood of a PLUS default.18 Holding other factors constant, PLUS loans
are more likely to default if the students also borrow large amounts, have
dropped out of college without a degree
or enroll in a four-year private, proprietary or minority-serving institution.
On the other hand, PLUS borrowers
are less likely to default if they enter
repayment with a higher beginning balance or fund children who have already
completed relatively more schooling,
are enrolled part time or have graduated from college. Parents’ default
probability is much more related to
their children’s college experience than
to the PLUS loan’s characteristics.
A students' college experience may
be tied to family finances, academic
aspirations and borrower risk preferences, all of which can influence repayment behavior.

Outperforming Stafford Loans
The Trellis data also include information on Stafford loans, allowing
review of overlapping parent PLUS
and student Stafford loan data from
September 2006 to August 2009.
Compared with Stafford borrowers,
PLUS borrowers on average took out
fewer loans, had higher initial balances
and paid a higher interest rate. PLUS
borrowers’ children were more likely

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

to attend four-year public or private institutions instead of a two-year public
institution or proprietary college, had
a higher graduation rate, were more
likely to enroll full time and much less
likely to drop out.
Since parents are generally older,
financially more stable and more experienced with debt, it is not surprising
that PLUS loans have better repayment
outcomes than Stafford loans (Chart 2).
For borrowers seven years into repayment in the Trellis portfolio, the default
rate on Stafford loans was 28.3 percent,
18.1 percentage points higher than that
of PLUS loans.
Despite a default rate increase
around the recession, PLUS loans have
been the only federal student loan
program that generates profits for the
government and, thus, helps offset
other federal educational loan program
costs. PLUS loans are forecast to generate a $20.6 billion profit for the federal
government from 2018 to 2028.19
PLUS loans also outperform Stafford loans after controlling for other
factors.20 Student borrowers are more
likely to default if they attend a two-year
public institution, enroll part time or
withdraw without a degree. Attending
a nonprofit, private four-year institution tends to to increase the chance of
default for parent borrowers but not
for students.

Parent, Student Interviews
To learn more about PLUS expectations and experiences, 49 parent borrowers and 36 students whose parents
had borrowed on their behalf were
interviewed. Parent borrowers tended
to have more experience dealing with
debt and had more realistic expectations for repayment than did students.
Overall, the majority of the parents
and students expected the parents to
repay the PLUS loans. The decision to
pay for college through PLUS loans
didn’t always follow thoughtful discussions with students about explicit academic expectations and implications of
ongoing financial obligations.
Parents also reported that PLUS loans
affected their ability to save for retirement and make major purchases. The

CHART

PLUS Loans Outperform Stafford Loans Seven Years
After Repayment Begins
In repayment or paid off

In default

Consolidated

Stafford

PLUS

40
Percent

100

SOURCE: Authors’ calculation based on Trellis data for fiscal 2007, 2008 and 2009.

collegiate pathway to adulthood, when
parental borrowing is involved, seems
to come with parental sacrifice as well
as a transfer of financial responsibility.
Di is a senior economist in the Research
Department at the Federal Reserve Bank
of Dallas. Fletcher is a senior research
analyst and Webster is the director of
research at Trellis Co., a Round Rock,
Texas, nonprofit corporation that seeks
to help students retire education loans
and improve access and outcomes
involving education.

Notes
Parents may also borrow from private lenders with
terms, conditions and interest rates set by the lender
based on the borrower’s creditworthiness. Graduate
students can obtain loans for themselves under a
separate program, also called PLUS. That program is not
the focus of this article.
2
The PLUS loan default rate increased around the
recession. Recent official data are unavailable.
3
Federal Student Loan Portfolio, U.S. Department of
Education, accessed July 20, 2018, https://studentaid.
ed.gov/sa/about/data-center/student/portfolio.
4
Subsidized and unsubsidized loans have the same
interest rate fees. Students who demonstrate financial
need and qualify for subsidized loans do not have the
loan interest accrued while in school or during the
grace period.
5
As of July 1, 2017, the PLUS loan interest rate was
7.0 percent, and the loan fee at disbursement was 4.26
percent, while Stafford loan interest was 4.45 percent and
the loan fee 1.07 percent.
1

Parents need to pass the PLUS loan credit check.
See, “Direct PLUS Loans and Adverse Credit,” U.S.
Department of Education, March 2015, https://
studentaid.ed.gov/sa/sites/default/files/plus-adversecredit.pdf.
7
See “Borrowing Constraints, Parental Altruism and
Welfare,” by Jorge Soares, Journal of Macroeconomics,
vol. 45, 2015, pp. 1–20.
8
“America's Divided Recovery: College Haves and HaveNots,” by Anthony P. Carnevale, Tamara Jayasundera
and Artem Gulish, Georgetown University Center on
Education and the Workforce, June 2016.
9
All calculations are in 2016 dollars. PLUS loans totaled
$12.6 billion in 2016–17. Data are from “Trends in
Student Aid 2017,” College Board, accessed July 20,
2018, https://trends.collegeboard.org/student-aid.
10
Some families prefer PLUS loans because of the
repayment flexibility federal loans offer. PLUS borrowers
can consolidate their loans and join the IncomeContingent Repayment Plan, which is less generous than
most other income-driven repayment plans but caps
payments at a share of earnings.
11
A cohort default rate, the standard measure of
federal education loan performance, is the percentage
of borrowers who enter repayment during a particular
federal fiscal year, Oct. 1 to Sept. 30, and default or fail
to meet other specified conditions prior to the end of the
second following fiscal year.
12
Proprietary postsecondary institutions refer to those
private, profit-seeking colleges that operate as businesses.
13
“Cracking Down on PLUS Loans,” by Libby A. Nelson,
Inside Higher Ed, Oct. 12, 2012, www.insidehighered.
com/news/2012/10/12/standards-tightening-federalplus-loans?.
6

(Continued on back page)

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

SPOTLIGHT

Shale Oil Propels U.S. Crude Export Increase
By Kunal Patel and Grant Strickler

rude oil exports from the U.S.
are rising, reaching 2.2 million
barrels per day (mb/d) in June
2018, triple the 2016 average and the
highest ever for the nation. More than
90 percent of crude exports this year
have originated on the Gulf Coast,
generating jobs, capital and income for
ports in Houston and Corpus Christi.
Such exports were at a trickle before
Congress lifted a federal crude oil
export ban that had been in place since
1975. The change, which took effect in
December 2015, allows U.S. producers
to sell oil directly to the global market
at a time when shale oil production is
high and rising.

CHART

Shale Boom Impacts

0.0

U.S. crude oil production has grown
steadily since 2008, reaching a record
of more than 10 mb/d this year, with
12 mb/d expected by the end of 2019,
according to the Energy Information
Administration. Shale oil accounts for
99 percent of the production growth.
Shale yields a light-sweet crude oil, requiring a simple refining configuration
to produce gasoline and diesel.
As domestic crude production
declined in the 1990s and 2000s, U.S.
refiners made significant investments
in their refining capabilities to process
imported heavy-sour crude, primarily
from Venezuela and nearby Mexico and
Canada. Heavy-sour crude, which is
generally cheaper than light sweet, provided greater profitability for refiners.

Building Infrastructure
With the shale boom, there was a mismatch between the crude oil produced
and domestic refining capabilities,
creating a pricing distortion for domestic production. This mismatch is one
reason Congress removed the export
ban; rising domestic production likely
also made energy security less relevant.
When Congress ended the ban, the
infrastructure needed to export signifi-

U.S. Crude Oil Exports Rising at Accelerating Rate

Million barrels per day*
2.0

Total
Asia & Oceania
Europe, Middle East & Africa
Latin America & Caribbean
Canada

1.5

1.0

0.5

’14

’15

’16

’17

’18

*Three-month moving average.
SOURCE: Energy Information Administration.

cant volumes of crude oil was lacking.
Midstream providers started investing
in export-related infrastructure in the
Houston and Corpus Christi regions in
2016. Exports from both ports increased, with the Port of Corpus Christi
the first in the U.S. to partially load a
very large crude carrier (VLCC), a type
of vessel capable of transporting more
than 2 million barrels of oil.
The Louisiana Offshore Oil Port,
about 20 miles south of Louisiana’s Port
Fourchon, successfully fully loaded a
VLCC last February and became the
first U.S. port to do so. With improvements to export infrastructure and an
increasing supply of light-sweet shale
oil, U.S. exports are poised to continue
expanding.
Before allowing exports, Congress
permitted some small-scale exemptions—almost all (92 percent) destined
for Canada. Now, 42 percent of U.S.
oil exports go to Asia and Oceania; 34
percent to Europe, the Middle East and
Africa; and 19 percent to Canada.

Constraining Factors
While the long-term outlook is bright
for U.S. exports, infrastructure limits
the near term. The Rapidan consulting
group estimates current Gulf Coast export capacity at up to 3.0 mb/d, which
could start constraining exports in as
little as a year, assuming a reduction
of transport bottlenecks in the oilrich Permian Basin in West Texas and
southeastern New Mexico.
Potential Chinese tariffs on U.S.
crude exports could also be a limiting
factor. However, assuming production
growth continues and the construction
of new export terminals is completed,
other trading partners would likely
emerge. The Intercontinental Exchange
is looking to add a futures contract for
crude delivered in Houston, making it
easier for transport companies to purchase crude close to export infrastructure rather than having to source it from
production areas in West Texas.

Southwest Economy • Federal Reserve Bank of Dallas • Third Quarter 2018

GO FIGURE

If Texas Were a Country ...
Design: Emily Rogers; Content: Christopher Slijk & Benjamin Meier

State conducts business
on an international scale
18th largest
world economy*

4th largest
oil producer

19th largest
global exporter

Texas is comparable to entire countries—and so are some of its local areas.

Exports

GDP

$1.7

$1.8

$264

$295

Texas

Canada

Texas

India

Trillion

Billion

Dallas–Fort Worth $511 Billion

Houston

$84 Billion

Belgium $510 Billion

Norway

$89 Billion

Population

Oil

4.6

Million b/d

Texas

4.5

28.3

24.5

Iraq

Texas

Australia

Million b/d

Permian Basin 2.3 Million b/d
Mexico 1.9 Million b/d

Million

Houston 6.9 Million
Libya 6.4 Million

*Ranking based on purchasing-power-parity adjusted gross domestic product, a way of comparing economically differing nations.
NOTES: State data are as of 2017 except oil production, which is as of May 2018. Metro and regional data are as of 2017 for population, as of fourth quarter 2017 for oil production and as of 2016 for gross domestic
product (GDP) and exports. The abbreviation b/d refers to barrels per day. Permian Basin calculation covers 55 counties in West Texas and encompasses the cities of Midland, Odessa and Lubbock.
SOURCES: GDP—Bureau of Economic Analysis and International Monetary Fund World Economic Outlook Database; exports—U.S. Department of Commerce International Trade Administration and International
Monetary Fund; oil production—Texas Railroad Commission
and International
Energy •
Agency;
population—Census
and United
Nations
Department
of Economic and Social Affairs.
Southwest
Economy
Federal
Reserve BankBureau
of Dallas
• Third
Quarter
2018

Parental Borrowing for College
Comes with Repayment Issues
(Continued from page 17)
Before 2010, federal student loans were either made
directly by the federal government or by private lenders
but federally guaranteed. The guaranteed loan program
implemented stricter credit checks on borrowers than the
direct loan program. Officials subsequently made all new
student loans direct loans, though underwriting became
stricter. PLUS default rates are not factored into an
institution’s student loan default rate, which determines
the institution’s eligibility for federal aid. See, “The
Wealth Gap PLUS Debt: How Federal Loans Exacerbate
Inequality for Black Families,” by Rachel Fishman, New
America Foundation, May 2018, https://s3.amazonaws.
com/newamericadotorg/documents/Wealth_Gap_Plus_
Debt_FINAL.pdf.
14

The U.S. Department of Education sets the minimum
total debts with adverse conditions (i.e., accounts in
collection or charge-offs) as exceeding $2,085 (inflation
adjusted, 2015 dollars), instead of any amount. Thus,
fewer borrowers are disqualified.
16
Some students may take more than four years to
complete a standard four-year program.
17
The minority-serving institutions were defined
according to the integrated postsecondary education data
system data, which include historically black colleges
and universities, predominantly black institutions and
Hispanic-serving institutions. Some of minority-serving
institutions are eligible for federal Title III funding under
the Higher Education Act. In Trellis data, the largest
15

historically black colleges and universities in Texas
include Texas Southern University and Prairie View A&M
University; the largest Hispanic-serving institutions
include the University of Texas at San Antonio and the
University of North Texas at Dallas.
18
A logit model and a proportional hazard model are
developed. The results are consistent across econometric
specifications.
19
Authors’ calculation based on “Student Loan
Programs—CBO’s April 2018 Baseline,” Congressional
Budget Office, April 2018, www.cbo.gov/sites/default/
files/recurringdata/51310-2018-04-studentloan.pdf.
20
As shown in a logit regression of the likelihood to
default on a loan and borrower characteristics.

Federal Reserve
Bank of Dallas

Southwest Economy
is published 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 to the Federal
Reserve Bank of Dallas.
Southwest Economy is available on the Dallas Fed website, dallasfed.org.

Marc P. Giannoni, Senior Vice President and Director of Research
Pia Orrenius, Keith R. Phillips, Executive Editors
Michael Weiss, Editor
Kathy Thacker, Dianne Tunnell, Associate Editors
Justin Chavira, Emily Rogers, Darcy Taj, Graphic Designers

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

Full text of The Southwest Economy : Third Quarter 2018

FRASER