View original document

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

t
b
e
D

VO L . 1 , I S S U E 4 , COV E R I N G 2 0 1 6 : Q 4

Auto Debt Expansion Continues to Slow
While Subprime Delinquencies Rise
By Lowell R. Ricketts

W

hile real per capita consumer debt growth
rose across the country in the fourth quarter of 2016, it declined or remained unchanged in
most of the largest metropolitan statistical areas
(MSAs) in the Eighth Federal Reserve District.1
Declines in mortgage debt continued to temper
overall debt growth. Lending in both the auto
and student debt sectors rose, but was uniformly
lower on a year-over-year basis when compared
with our previous report for the nation and for
the District MSAs of Little Rock, Ark., Louisville,
Ky., Memphis, Tenn., and St. Louis.2 Over the
past few years, strong lending of auto and student
debt has buoyed total debt, accounting for the
majority of credit expansion. This report offers a
closer look at auto lending, including the factors
that contributed to the expansion in this sector.
This report uses the latest release of the Federal
Reserve Bank of New York and Equifax Consumer Credit Panel with data as of the fourth quarter
of 2016. The figures in this report help to provide
a focused narrative of the latest developments in
consumer debt across the District and nation. For
a more extensive collection of figures updated
each quarter, see the QDM appendix.3

Latest Developments in
Consumer Debt
1. In the fourth quarter of 2016, real per capita
mortgage debt declined across the United
States and the four District MSAs. The steepest declines were in Louisville and Memphis,
where per capita mortgage debt declined by
3.5 percent and 4.3 percent, respectively.
2. The serious delinquency rate for mortgage

Changes in Per Capita Debt Levels and Serious Delinquency Rates
Year-over-Year Percent Change from 2015:Q4 to 2016:Q4
St. Louis
MSA

2016:Q4

United States

Total

0.4

Mortgage

–0.9

–1.8

–5

–6.7
4.2

Little Rock
MSA

Louisville
MSA

Memphis
MSA

–1.5

–0.8

–0.5

–3.5

–4.3

–11.5

–2.8

–6.8

4

5.4

5.5

% Change in Per Capita Consumer Debt

HELOC
Auto

5.1

0

1.5

Credit Card

2.9

1.4

3.8

0.9

3.4

Student

5.3

5.6

6.7

4.8

6.6

Mortgage

–0.5

–0.2

–0.6

–0.1

–0.7

HELOC

0.1

–0.2

0.1

0.1

0.2

Change in Serious Delinquency Rate

Auto

0.4

0.1

Credit Card

–0.2

–0.1

Student

–0.2

0.4

0.5

1.2

0

0.6

–0.4

–0.1

0.1

0.1

–0.1

SOURCE: Federal Reserve Bank of New York/Equifax Consumer Credit panel.
NOTES: Serious delinquency rates are the shares of outstanding debt that is over 90 days past due. Changes to those rates are given
as the differences in percentage points. Figures are rounded.

debt fell both nationally and across the District
MSAs, with the MSAs reaching their lowest
levels since before the Great Recession. Meanwhile, although the national rate has fallen, it
still remains above its pre-recession low.
3. Per capita HELOC, or home equity line of
credit, debt declined across the nation and all
of the District MSAs. HELOC debt, an important source of consumer debt growth prior to
the recession, continues to show a pattern of
consistent deleveraging.4

4. Growth in per capita student debt slowed for the nation and all
District MSAs. The average growth rate across the four MSAs
fell to 5.9 percent in the fourth quarter, down from 7.2 percent
in the third quarter.
5. Per capita credit card debt continued to grow for the nation as
well as for all District MSAs. At the same time, rates of serious
delinquency fell or remained unchanged. Credit card debt
remains well below pre-recession levels and rates of serious
delinquency.
6. For the third quarter in a row, growth in per capita auto debt
slowed across the District MSAs. While auto debt remains a
significant contributor to overall growth, the robust growth
observed over the past few years continues to abate. With the
exception of Louisville, the slowdown has been more pronounced across the District MSAs than the nation.
7. Repayment difficulty among auto loan borrowers continued to
worsen and at a faster pace since our previous report. Serious delinquency rates in Louisville and Little Rock were 3.4
and 4.3 percent, respectively. Those rates match or exceed the
previous peak seen after the recession.
8. Subprime auto debt comprised 38 percent of outstanding
auto debt in Memphis and 28 percent in Little Rock during
the fourth quarter of 2016. Subprime debt has been growing
fastest in both MSAs and exceeds the amount of prime debt in
those markets.5
9. Serious auto delinquency rates were higher and increasing
faster for subprime borrowers in all four MSAs. Given the
growing share of subprime debt within these MSAs, this may
contribute to further increases in overall serious delinquency
rates. Much of this lending was originated by auto financing
companies rather than banks or credit unions.

FIGURE 1

Contribution
to Percent
Change
Capita Debt
Contribution
to Percent
Change
in Total in
PerTotal
CapitaPer
Debt
Year-over-Year Change from 2015:Q4 to 2016:Q4
1.5

0.4

0.4

0.5

0.6
0.4

−0.3

0

0
Percent

1

1

1
0.5

0.4
−0.2

−0.1

−0.2

−0.2

0.2 0.2

0.1

0.2

0.7

0.6

0.1

0.1

0.2

0.1

0.2

−0.1

−0.2

−0.6
−0.8

–1

−1.2
−1.5

–2
−2.3
−2.6

Total

Mortgage

Auto

HELOC Credit Card
Debt Type

Little Rock MSA
Louisville MSA

United States
St. Louis MSA

Student

Other

Memphis MSA

SOURCE:Federal
FederalReserve
Reserve
Bank
of New York/Equifax
Consumer
SOURCE:
Bank
of NY/Equifax
Consumer Credit
Panel. Credit panel.
NOTES:
each
to total debt. All data labels have been rounded.
F E D E RGrowth
A L R E S EinR V
E B Atype
N K Oof
F Sdebt
T. LOisU relative
IS
FIGURE 2

Total
PerPer
Capita
Auto Debt
TotalReal
Real
Capita
Auto Debt

Index, 2003:Q1=100

120

110

100

90

2005
United States

2010
Year

2015

Louisville MSA

Little Rock MSA

Memphis MSA

2016:Q4

St. Louis MSA

SOURCE: Federal
Bank
of NY/Equifax
Consumer Credit
Panel. Credit Panel.
SOURCE:
FederalReserve
Reserve
Bank
of New York/Equifax
Consumer
NOTES: “Real” data are inflation-adjusted. The gray bar indicates recession.

Waning Momentum for Consumer Debt

NOTES: “Real” data are inflation-adjusted. The gray bar indicates recession.
F E D E R A L R E S E R V E B A N K O F S T. LO U I S

Almost all consumer debt growth continues to be driven by
strong growth in a few categories of debt. Without the contributions of new student and auto debt over the past few years, the
deleveraging seen after the Great Recession would likely have
continued. Since mortgage debt comprises a majority share of
consumer debt, changes within this category have a considerable impact on total debt. Figure 1 shows the growth of various
types of debt as a contribution to the change in total per capita
debt.6 This figure provides a clear picture of the drivers of overall
growth. Continued deleveraging of mortgage debt continues
to curb overall growth.7 Deleveraging has been particularly
strong in both Louisville and Memphis. In Memphis, despite the
strongest growth in student and auto debt among the District
MSAs, total debt has fallen because of the sizeable reduction in
mortgage debt. Very low serious delinquency rates for mortgage
debt suggest that defaults are less of a contributing factor to these
reductions.

FIGURE 3

U.S. Outstanding Auto Balances
by Equifax Risk Score at Origination
U.S. Outstanding Auto Balances by Equifax Risk Score at Origination
400

Billions 2009 $

300

200

100

0

2002

2004

2006

2008

2010

2012

2014

2016

Year
Under 620

620-659

720-759

Over 760

660-719

SOURCE: Federal Reserve Bank of New York/Equifax Consumer Credit Panel.
SOURCE:
Federal Reserve Bank of New York/Equifax Consumer Credit Panel.
NOTES: For inflation adjustment, 2009 dollars are used, in line with the Bureau of Economic Analysis’ personal
NOTES:
For inflation
adjustment,
2009
used,developed
in line with
Bureau
of Economic
Analysis’
consumption
expenditures
chain-type
pricedollars
index. are
Equifax
the the
score
to measure
lending risk.
Typically,
credit scores
below 620 are
considered chain-type
subprime. Aprice
scoreindex.
below Equifax
500 would
qualify asthe
deep
subprime.
These lending
lower
personal
consumption
expenditures
developed
score
to measure
scores
indicate
that
the
borrower
is
more
likely
to
have
problems
repaying
the
loan.
risk. Typically, credit scores below 620 are considered subprime. A score below 500 would qualify as deep
F E D E R A These
L R E S Elower
R V E Bscores
A N K Oindicate
F S T. LO that
U I S the borrower is more likely to have problems repaying the loan.
subprime.

2

FIGURE 4

Outstanding Auto Loan Balances by Equifax Risk Score at Origination
MEMPHIS MSA
MSA
Memphis

1.5

3

1.0

2

Billions 2009 $

Billions 2009 $

LITTLERock
ROCKMSA
MSA
Little

0.5

0.0

1

0
2002

2004

2006

2008

2010

2012

2014

2002

2016

2004

2006

2008

Year
Under 620

620-659

720-759

Over 760

2010

2012

2014

2016

2014

2016

Year
660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

Under 620

620-659

720-759

Over 760

660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

LOUISVILLEMSA
MSA
Louisville

ST. Louis
LOUISMSA
MSA
St.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

5

1.5

1.0

Billions 2009 $

Billions 2009 $

4

0.5

3
2
1

0.0
2002

2004

2006

2008

2010

2012

2014

0

2016

2002

2004

2006

2008

Under 620

620-659

720-759

Over 760

2010

2012

Year

Year
660-719

Under 620

620-659

720-759

Over 760

660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

SOURCE (ALL 4 FIGURES): Federal Reserve Bank of New York/Equifax Consumer Credit Panel.

The Rise of Auto Debt in the Nation and
District MSAs

The robust growth in auto lending has been fueled in part by a
combination of low interest rates, low fuel prices and a strengthening economy following the recession. The interest rate on a
48-month loan from a commercial bank for a new automobile
purchase dropped from close to 8 percent prior to the Great
Recession to an average of 4.3 percent since the second quarter of
2014.8 Meanwhile, auto finance company rates for new car loans
averaged around 5 percent during this same period.9 Between
May 2011 and June 2014, gas prices averaged around $3.50 per
gallon. Since then, the average price per gallon has been about
$2.40.10 In addition, the unemployment rate dropped to 4.7 percent in December 2016 from 9.1 percent in January 2011.11 Taken
together, these developments have provided greater incentives for
consumers to take out loans to make car purchases.
Figure 3 shows how this credit expansion has included consumers across the credit spectrum, including those with records
considered subprime. Softened underwriting standards have
raised concerns regarding the risk associated with the robust
growth in auto debt. On that note, the Office of the Comptroller

The steep upward trend seen in Figure 2 demonstrates that auto
lending has been a substantial source of growth in consumer debt.
While still positive, the year-over-year growth rate for the nation
slowed for the second quarter in a row, representing a drop of 3.2
percentage points from the second quarter of 2016. The slowdown
has been more persistent among the District MSAs, with three successive quarters of diminished growth. Between the first and fourth
quarters of 2016, growth rates for per capita auto debt slowed by
4.9 percentage points in Little Rock; 4.1 percentage points in St.
Louis; 3.4 percentage points in Memphis and 2.5 percentage points
in Louisville. Figure 2 shows that the level of per capita auto debt in
the fourth quarter plateaued for each of the District MSAs except
Little Rock, where it sharply declined. Auto debt remains a source
of relatively strong growth in consumer debt. However, if the
recent deviation from trend is persistent, it may be a more modest
source of growth moving forward.
3

FIGURE 5

Serious Delinquency Rates by Debt Type
MEMPHISDelinquency
MSA
Serious
Rates by Debt Type – Memphis MSA

LITTLE ROCK
MSA Rates by Debt Type – Little Rock MSA
Serious
Delinquency

15
Percent of Outstanding Balance

Percent of Outstanding Balance

15

10

5

0

2005
Auto

2010
Year
Mortgage

Student

10

5

0

2015
HELOC

Credit Card

Auto

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

2010
Year
Mortgage

Student

2015
HELOC

Credit Card

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

LOUISVILLE
MSA
Serious Delinquency
Rates by Debt Type – Louisville MSA

ST. LOUIS
MSA
Serious
Delinquency
Rates by Debt Type – St. Louis MSA
15
Percent of Outstanding Balance

15
Percent of Outstanding Balance

2005

10

5

0

2005
Auto

2010
Year
Mortgage

Student

5

0

2015
Credit Card

10

HELOC

2005
Auto

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

2010
Year
Mortgage

Student

2015
Credit Card

HELOC

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

SOURCE (ALL 4 FIGURES): Federal Reserve Bank of New York/Equifax Consumer Credit Panel.

of the Currency has issued repeated warnings in its semiannual
risk perspective reports. The most recent report noted that delinquencies have increased, as have non-seasonal losses.12 It also
noted that lenders have stretched repayment terms and offered
higher advance rates, resulting in greater loan-to-value ratios.13
These developments have prompted some lenders to tighten
underwriting standards and/or reduce lending activity.14, 15 This
may be reflected to some degree in the growth slowdown seen in
the nation and the District MSAs.
Figure 4 shows that the auto credit expansion has differed substantially across District MSAs in terms of lending to prime and
subprime borrowers.16 Debt originations to subprime borrowers
encompassed a much greater share of overall debt in Little Rock
and Memphis. In the fourth quarter of 2016, subprime debt accounted for 38 percent of outstanding auto debt in Memphis and
28 percent of outstanding auto debt in Little Rock. In comparison, subprime debt accounted for 23 percent of outstanding auto
debt in Louisville and 21 percent in St. Louis. This isn’t a historical deviation for Memphis. Since 2001, the share of subprimeto-total auto debt has averaged around 39 percent. Memphis

also has a greater share of this debt generated by auto financing
companies, which include financial services provided by car
manufacturers or dealerships, as opposed to other lenders, such
as banks or credit unions. In the fourth quarter of 2016, close to
65 percent of outstanding auto debt in Memphis was financed
by auto finance companies. In comparison, financing companies
were lenders for 56 percent of auto debt in Little Rock, 51 percent
of auto debt in Louisville and 47 percent of auto debt in
St. Louis. Auto financing companies often originate a greater
share of subprime loans than traditional financial institutions.
This is consistent in the data: Auto financing companies originated close to 87 percent of subprime auto debt in Memphis and
83 percent of subprime debt in Little Rock. In Louisville, they
generated 77 percent of the subprime auto debt and in St. Louis,
71 percent. The greater shares of subprime debt in Memphis and
Little Rock represent greater risks of rising repayment delinquencies.

4

FIGURE 6

Serious Delinquency Rate for Auto Debt by Equifax Risk Score at Origination

Serious Delinquency Rate for Auto Debt,
MEMPHIS
by
EquifaxMSA
Risk Score at Origination – Memphis MSA

4

4

3

3

Percentage Share

Percentage Share

Serious Delinquency Rate for Auto Debt,
LITTLE
ROCK
MSA
by
Equifax
Risk
Score at Origination – Little Rock MSA

2

1

0

2

1

0
2002

2004

2006

2008

2010

2012

2014

2016

2002

2004

2006

2008

Year
Under 620

620-659

720-759

Over 760

660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

3

3

Percentage Share

Percentage Share

4

2

1

0
2006

2008

2010

2012

2014

620-659

720-759

Over 760

2016

Under 620

620-659

720-759

Over 760

2014

2016

660-719

2

1

0

2016

2002

Year
Under 620

2014

F E D E R A L Delinquency
R E S E R V E B A N K O F Rate
S T. LO Ufor
IS
Serious
Auto Debt,
ST. LOUIS MSA
by Equifax Risk Score at Origination – St. Louis MSA

4

2004

2012

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

F E D E R A L Delinquency
R E S E R V E B A N K O F Rate
S T. LO Ufor
IS
Serious
Auto Debt,

LOUISVILLE
MSAScore at Origination – Louisville MSA
by Equifax Risk

2002

2010
Year

2004

2006

2008

2010

2012

Year
660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

Under 620

620-659

720-759

Over 760

660-719

SOURCE: Federal Reserve Bank of NY/Equifax Consumer Credit Panel.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S
SOURCE
(ALL 4 FIGURES): Federal Reserve Bank of New York/Equifax Consumer Credit Panel.

F E D E R A L R E S E R V E B A N K O F S T. LO U I S

NOTE: Four-quarter moving average.

Continued Signs of Stress in Auto
Serious Delinquency Rates

debt into groups based on the Equifax Risk Score at origination.
Figure 6 shows the risk score is an effective predictor of repayment
difficulty, and that the majority of repayment difficulty is concentrated among the two groups with the lowest scores. Recall that
in Figure 4, Memphis and Little Rock had the greatest growth in
subprime lending. This, coupled with the fact that these borrowers
are more likely to experience repayment difficulties, helps explain
the increase in the overall serious delinquency rates in these MSAs.
In addition, serious delinquency rates among subprime borrowers
are more cyclical, which means these borrowers are more likely to
fall behind on payments during an economic downturn. This can
be seen within the recession shading within Figure 6.
Serious delinquency rates among subprime borrowers in Little
Rock and Memphis have now markedly increased during two
years of an economic expansion. Nationally, some lenders have
reduced their originations and tightened underwriting standards. It remains to be seen if other lenders will expand lending
in response. The response will have important implications for
potential growth and delinquency in auto debt, and the growth
prospects for overall consumer debt.

Our previous report for the third quarter of 2016 highlighted
rising serious delinquency rates for auto debt in the nation and all
District MSAs except for St. Louis. That trend accelerated in the
fourth quarter of 2016, with serious delinquency rates increasing
for the nation and all District MSAs. They also increased by a
greater margin than in the previous quarter.
Figure 5 shows the serious delinquency rates across all types of
debt for the District MSAs. Little Rock and Memphis show the
most pronounced upward trend in serious delinquency of auto
debt. This is consistent with the greater proportion of subprime
auto debt within those markets. In general, Memphis has had
greater rates of serious delinquency over much of the period
shown in the figure, which is the first quarter of 2003 to the fourth
quarter of 2016. The rates correspond with its higher level of subprime auto debt.
The link between overall serious delinquency rates for auto debt
and subprime auto debt is best illustrated by again partitioning
5

Endnotes
1 We adjust debt for both inflation and population growth; these are two factors which contribute to a general upward trend in debt levels.
2 The Eighth Federal Reserve District comprises all of Arkansas and portions
of Illinois, Indiana, Kentucky, Mississippi, Missouri and Tennessee. An MSA
consists of multiple counties and includes the core urban area, as well as any
adjacent counties that have a strong social and economic attachment to the
urban core.
3 https://www.stlouisfed.org/~/media/publications/quarterly%20debt%20
monitor/issue_4/2016Q4%20QDM%20appendix.pdf
4 Deleveraging is the process where consumers pay down debt or shed it in
bankruptcy or foreclosure.
5 We define debt with an Equifax Risk Score less than 620 as subprime and debt
with a Risk Score over 760 as prime. The Equifax Risk Score, like a credit score,
is a measure of lending risk with higher scores indicative of a lower likelihood
of repayment difficulties.
6 Total debt is averaged across the population to account for population growth.
7 It’s important to note that the rate of change is a net flow; it reflects the sum of
debt creation and destruction. The change in mortgage debt is the difference
between new debt creation from borrowing activity (e.g. for home purchases
financed via new mortgage); and debt destruction in the form of repayments
and debt default.
8 Board of Governors of the Federal Reserve System, Finance Rate on Consumer
Installment Loans at Commercial Banks, New Autos 48 Month Loan
[TERMCBAUTO48NS], retrieved from FRED, Federal Reserve Bank of St. Louis,
https://fred.stlouisfed.org/series/TERMCBAUTO48NS, February 14, 2017.
9 Board of Governors of the Federal Reserve System, Average Finance Rate of
New Car Loans at Finance Companies, Amount of Finance Weighted
[RIELPCFANNQ], retrieved from FRED, Federal Reserve Bank of St. Louis,
https://fred.stlouisfed.org/series/RIELPCFANNQ, February 14, 2017.

10 U.S. Energy Information Administration, U.S. Regular Conventional Gas Price
[GASREGCOVM], retrieved from FRED, Federal Reserve Bank of St. Louis,
https://fred.stlouisfed.org/series/GASREGCOVM, February 14, 2017.
11 U.S. Bureau of Labor Statistics, Civilian Unemployment Rate [UNRATE],
retrieved from FRED, Federal Reserve Bank of St. Louis, https://fred.stlouisfed.org/series/UNRATE, February 14, 2017.
12 Office of the Comptroller of the Currency. “Semiannual Risk Perspective from
the National Risk Committee.” Fall 2016, Washington D.C., Pg. 26. www.
occ.treas.gov/publications/publications-by-type/other-publications-reports/
semiannual-risk-perspective/semiannual-risk-perspective-fall-2016.pdf
13 An advance rate is the maximum percentage of the value of collateral that a
lender is willing to extend for a loan. A higher advance rate will translate into
a larger loan offered to the borrower and less insurance for the lender.
14 Wack, Kevin. “Why these big banks are retreating in auto lending.” American
Banker, February 13, 2017. www.americanbanker.com/news/why-these-bigbanks-are-retreating-in-auto-lending
15 McLannahan, Ben. “Santander Consumer continues to put the brakes
on originations.” The Financial Times, January 25, 2017. www.ft.com/
content/347f77fc-e320-11e6-8405-9e5580d6e5fb
16 Auto debt is partitioned by the Equifax Risk Score at time of origination.
Debt that didn’t have a valid Risk Score was omitted from these totals.

6