Household Leveraging and Deleveraging

Karen Dynan

Brookings Institution

May 20, 2010

2

Household Leveraging and Deleveraging

Q4

.6

.8

1

1.2

1.4

19601960 19651960 1965 19701960 1965 1970 19751960 1965 1970 1975 19801960 1965 1970 1975 1980 19851960 1965 1970 1975 1980 1985 19901960 1965 1970 1975 1980 1985 1990 19951960 1965 1970 1975 1980 1985 1990 1995 20001960 1965 1970 1975 1980 1985 1990 1995 2000 20051960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010Source: U.S. Flow of Funds Accounts

Ratio of Debt to Disposable Personal Income

3

Big Questions

• About the past: What explains the pre-crisis uptrend in debt? Were these factors a good or bad thing for households and the economy?

• About the present: How is the deleveraging occurring? Is the deleveraging a good or bad thing for households and the economy?

• About the future: How much more deleveraging should we expect?

4

Why Did Household Debt Rise?

“The most important factors behind the rise in debt and the associated decline in saving out of current income have probably been the combination of increasing house prices and financial innovation.”

Dynan and Kohn, The Rise in U.S. Household Indebtedness: Causes and Consequences (2007)

5

Home Prices and Rising Debt

Home Mortgage Debt

Other Debt

Q4

0

.2

.4

.6

.8

1

1.2

1.4

0

.2

.4

.6

.8

1

1.2

1.4

Ra

tio

19801980 19851980 1985 19901980 1985 1990 19951980 1985 1990 1995 20001980 1985 1990 1995 2000 20051980 1985 1990 1995 2000 2005 2010Source: U.S. Flow of Funds Accounts

Composition of the Aggregate D/Y

Home Mortgage Debt

House Prices

-20

-10

0

10

20

-20

-10

0

10

20

4-Q

tr P

erce

nt C

hang

e19801980 19851980 1985 19901980 1985 1990 19951980 1985 1990 1995 20001980 1985 1990 1995 2000 20051980 1985 1990 1995 2000 2005 2010

Source: U.S. Flow of Funds Accounts and First American CoreLogicDebt data through 2009q4. House price data through Feb. 2010

Home Mortgage Debt and House Prices

6

Financial Innovation and Rising Debt• “Democratization” of credit explains only a small

part of the rise.

• Rather, the key factors were easier access and lower cost of credit for those who already had access to debt markets.

• Note that quantifying the role of FI difficult:

» FI broad and has occurred gradually.

» FI has interacted with other factors.

7

Evolution of Median Household Debt by Demographic Group

Consistent with the incremental and thorough-going nature of financial innovation, increases in borrowing have been gradual and widespread.

8

Consequences of Greater Credit Availability

“Developments in lending practices and loans markets that have enhanced the ability of households and firms to borrow … should be added to the list of likely contributors to the mid-1980s stabilization.”

Dynan, Elmendorf, and Sichel, Can Financial Innovation Help to Explain the Reduced Volatility of Economic Activity? (2006)

9

Greater Access to Credit Probably Explains Reduced Sensitivity of Consumption to Income Declines after Mid-1980s

10

BUT, downsides to greater access to credit have become painfully clear …

11

Share of Income Committed to Debt Service Has Risen• From Dynan (Journal of Economic Perspectives,

2009):

» Median DSR rose from .05 in 1983 to .13 in 2007.

» Percent of households with DSR > 0.40 rose from 4% in 1983 to 11% in 2007.

• Households with higher DSRs more likely to have trouble making payments.

12

Greater Leverage Has Increased Exposure to House Price Shocks

Change in Wealth Implied by a 20% Decline in Change in Wealth Implied by a 20% Decline in House Prices (as a Fraction of Income)House Prices (as a Fraction of Income)

19831983 19951995 20072007

All HouseholdsAll Households -.30-.30 -.33-.33 -.49-.49

Households in Households in Lowest 1/3 of Lowest 1/3 of Income DistnIncome Distn

-.46-.46 -.57-.57 -.77-.77

Source: Dynan (2009) based on the Survey of Consumer Finances.Source: Dynan (2009) based on the Survey of Consumer Finances.

13

Lessons about Household Leverage• Useful for getting around income constraints …

» To smooth consumption.

» To pursue higher returns via investment.

• But exposes households to more risk if things do not turn out as expected.

• FURTHER, financial crisis has taught us that it can hurt a much broader group than those who borrow if risk-taking is correlated across households.

14

Household Deleveraging

15

Both Mortgage and Consumer Debt Are Shrinking

-50

00

500

1,00

01,

500

Bill

ion

s o

f Do

llars

2005 2006 2007 2008 2009 2010

q1 q2 q3 q4 q1 q2 q3 q4 q1 q2 q3 q4 q1 q2 q3 q4 q1 q2 q3 q4 q1

Source: U.S. Flow of Funds Accounts and Fed G.19 data release.2010q1 consumer credit is estimated

Household Debt Growth

Home Mortgage Debt Consumer Credit Debt

16

How is deleveraging occurring?

17

Importance of DefaultsIf charge-offs had not been unusually elevated …

Consumer credit would have declined 2¾% in 2009 instead of declining 4½% (charge-offs explain 60% of decline).

Mortgage debt would have been flat in 2009 instead of declining 2% (charge-offs explain all of decline).

Credit Card Loans

Other Consumer Loans Q4

Q4

0

2

4

6

8

10

0

2

4

6

8

10

Per

cent

age

of lo

ans

cha

rge

d of

f (an

nual

ize

d)

20002000 20022000 2002 20042000 2002 2004 20062000 2002 2004 2006 20082000 2002 2004 2006 2008 2010Source: FFIEC Consolidated Reports of Condition and Income

Charge-offs of Consumer Loans

Q4

At Commercial Banks

AllMortgages

0

.5

1

1.5

2

2.5

0

.5

1

1.5

2

2.5

Per

cent

age

of d

ebt c

harg

ed

off (

annu

aliz

ed)

2000q1 2002q3 2005q1 2007q3 2010q1Sources: Federal Reserve Board and Fed Economist Jim Kennedy

Charge-offs of Residential Mortgage Debt

18

Evidence of Supply Constraints

Other Loans

Credit Cards Q1

-20

0

20

40

60

-20

0

20

40

60

Ne

t Per

cent

of B

anks

Tig

hten

ing

Sta

ndar

ds

20002000 20022000 2002 20042000 2002 2004 20062000 2002 2004 2006 20082000 2002 2004 2006 2008 2010Source: Fed Senior Loan Officer Survey

Banks Tightening Consumer Lending Standards

All

Prime

Nontraditional

Subprime

Q1 0

25

50

75

100

0

25

50

75

100

Ne

t Per

cent

of B

anks

Tig

hten

ing

Sta

ndar

ds

20002000 20022000 2002 20042000 2002 2004 20062000 2002 2004 2006 20082000 2002 2004 2006 2008 2010Source: Fed Senior Loan Officer Survey

Banks Tightening Mortgage Lending Standards

19

More Evidence of Supply Constraints• Loan officer survey: 2/3 reported loan standards

for nonprime households unlikely to return to long-term averages “for foreseeable future.”

• Private nonprime mortgage market still closed.

• Lower credit card limits, higher LTVs on new car loans.

• More austere regulatory environment (CARD act, new HOEPA regs, push for CFPA).

20

Demand for Loans Also Down

• Not surprising in a downturn when demand for consumer goods and services falling.

• 1/5 of senior loan officers (on net) reported that demand for consumer loans had fallen relative to 3 months earlier in May 2010 (well after consumption trough).

• But, signal from other possible indicators of greater prudence among households fairly weak.

21

Disentangling Supply vs. Demand

Looking at price is a traditional way to see if demand or supply effects are more important.

But these measures tell us only part of story because there is non-price rationing too.

Used Cars

30 Yr ConventionalMortgages

New Cars

Apr

Feb

Feb

0

5

10

15

0

5

10

15

Ra

te

20002000 20022000 2002 20042000 2002 2004 20062000 2002 2004 2006 20082000 2002 2004 2006 2008 2010Sources: Freddie Mac Primary Mortgage Market Survey. Fed. Reserve Release G.20.Car loan rates are from subsidiaries of the three major U.S. auto manufacturers.

Selected Household Interest Rates

22

Is the Deleveraging a Good Thing or Bad Thing for the Economy?• Short run: Inability / unwillingness to borrow is

dampening the pace of recovery.

• Longer run: Deleveraging puts households and the broader economy in a more solid and sustainable position.

» Regardless of how it occurs.

» Lower debt service payments have already made much more cash available for spending.

23

Aggregate DSR Has Fallen by 1.25 Percentage Points

Q4

10

11

12

13

14

10

11

12

13

14

De

bt S

erv

ice

Pa

ymen

ts a

s a

Per

cent

age

of D

PI

19801980 19851980 1985 19901980 1985 1990 19951980 1985 1990 1995 20001980 1985 1990 1995 2000 20051980 1985 1990 1995 2000 2005 2010Source: U.S. Federal Reserve.

Household Debt Service Ratio

24

Looking Ahead

25

High D/A Suggests Potential for Considerably More Deleveraging

Q4

.12

.14

.16

.18

.2

.22

.12

.14

.16

.18

.2

.22

Ra

tio

19801980 19851980 1985 19901980 1985 1990 19951980 1985 1990 1995 20001980 1985 1990 1995 2000 20051980 1985 1990 1995 2000 2005 2010Source: U.S. Flow of Funds Accounts.

Aggregate Household Debt/Asset Ratio

26

Future Deleveraging Most Likely to Occur in the Mortgage Area

• Mortgage charge-offs likely to remain high.

» Foreclosures are on the rise again after having been forestalled by HAMP and other factors.

• Even as lenders’ standards ease, new borrowing should be dampened by lack of home equity.

» CoreLogic: Close to 30% of mortgage borrowers have little or no equity (50% or more in the hardest-hit states).

27

Scope for Further Deleveraging in the Consumer Credit Area Less Clear• Delinquency rates are very high, particularly for

credit cards, and soft labor markets should keep them high => more charge-offs to come.

• But borrowing is likely to pick up with consumer demand.

• Q1 saw first increase in outstanding consumer credit since 2008:Q3 (up 0.4% but if charge-offs not elevated the increase would have been around 3%).

28

Conclusions

• More deleveraging seems likely, especially in the mortgage area.

• The deleveraging reflects a combination of households defaulting and borrowing less (different in some ways but gets them to the same place).

• Big unanswered question: Who is deleveraging?