Discussion of “Credit Supply and the
Housing Boom” by Alejandro Justiniano,
Giorgio E. Primiceri, and Andrea
Tambalotti
Robert E. Hall
Hoover Institution and Department of Economics
Stanford University
NBER with the help of Susan E. Woodward
8th Conference on Monetary Economics
Banco de Portugal
13 June 2015
·
1

The paper argues that an expansion in mortgage funding accounted for the runup in housing prices, not so much reductions in the collateral constraint on borrowers
2
This is not such a novel view—it is close to a consensus among housing experts
The long essay by Levitin and Wachter, cited in the paper, says the following in its abstract:
[Our paper] demonstrates that the bubble was a supply-side phenomenon attributable to an excess of mispriced mortgage finance: mortgage-finance spreads declined and volume increased, even as risk increased—a confluence attributable only to an oversupply of mortgage finance.
·

The paper argues that an expansion in mortgage funding accounted for the runup in housing prices, not so much reductions in the collateral constraint on borrowers
This is not such a novel view—it is close to a consensus among housing experts
2
The long essay by Levitin and Wachter, cited in the paper, says the following in its abstract:
[Our paper] demonstrates that the bubble was a supply-side phenomenon attributable to an excess of mispriced mortgage finance: mortgage-finance spreads declined and volume increased, even as risk increased—a confluence attributable only to an oversupply of mortgage finance.
·

The paper argues that an expansion in mortgage funding accounted for the runup in housing prices, not so much reductions in the collateral constraint on borrowers
This is not such a novel view—it is close to a consensus among housing experts
The long essay by Levitin and Wachter, cited in the paper, says the following in its abstract:
[Our paper] demonstrates that the bubble was a supply-side phenomenon attributable to an excess of mispriced mortgage finance: mortgage-finance spreads declined and volume increased, even as risk increased—a confluence attributable only to an oversupply of mortgage finance.
·
2

The role of the interest rate in allocating risk
The paper also mentions the role of the declining world interest rate
3
The interest rate governs the terms of trade between risk lovers and risk avoiders
The risk lovers borrow from the risk avoiders and hold the riskier investments
Many different institutions have emerged to tranche risk in this way: banks, hedge funds, and securtizations with credit enhancement to create low-risk instruments (AAA-rated tranches) for some clienteles, and high-risk instruments (equity tranches) for others
·

The role of the interest rate in allocating risk
The paper also mentions the role of the declining world interest rate
The interest rate governs the terms of trade between risk lovers and risk avoiders
3
The risk lovers borrow from the risk avoiders and hold the riskier investments
Many different institutions have emerged to tranche risk in this way: banks, hedge funds, and securtizations with credit enhancement to create low-risk instruments (AAA-rated tranches) for some clienteles, and high-risk instruments (equity tranches) for others
·

The role of the interest rate in allocating risk
The paper also mentions the role of the declining world interest rate
The interest rate governs the terms of trade between risk lovers and risk avoiders
The risk lovers borrow from the risk avoiders and hold the riskier investments
3
Many different institutions have emerged to tranche risk in this way: banks, hedge funds, and securtizations with credit enhancement to create low-risk instruments (AAA-rated tranches) for some clienteles, and high-risk instruments (equity tranches) for others
·

The role of the interest rate in allocating risk
The paper also mentions the role of the declining world interest rate
The interest rate governs the terms of trade between risk lovers and risk avoiders
The risk lovers borrow from the risk avoiders and hold the riskier investments
Many different institutions have emerged to tranche risk in this way: banks, hedge funds, and securtizations with credit enhancement to create low-risk instruments (AAA-rated tranches) for some clienteles, and high-risk instruments (equity tranches) for others
·
3

Figure 3
U.S. securities outstanding
Total
Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Equity global-saving-glut paper, 2011
Non-AAA RMBS (private label), CMBS, and ABS
2003 2007
Risk-avoiding portfolio, from Bernanke’s
Share AAA: 37.4% Share AAA: 36.0%
Held by global saving glut countries
2003 Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
Share AAA: 76.2%
Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
Held by Europe
2003
Share AAA: 31.9%
2007
Share AAA: 77.5%
2007
Share AAA: 33.2%
Held by U.S. residents
Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
2003 2007
Share AAA: 35.8% Share AAA: 32.0%
Note: RMBS: residential mortgage-backed securities; CMBS: commercial mortgage-backed securities; ABS: asset-backed securities other than RMBS and CMBS.
Source: Staff estimates based on Flow of Funds and Treasury International Capital system data.
23
4

Figure 3
U.S. securities outstanding
Total
2003
Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
Share AAA: 37.4%
Held by global saving glut countries
2003 Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
Share AAA: 76.2%
2007
Share AAA: 36.0%
2007
Share AAA: 77.5%
Held by Europe
Treasury securities
Agency debt
AAA Corp
2003
Risk-loving portfolio
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
Share AAA: 31.9%
2007
Share AAA: 33.2%
Held by U.S. residents
Treasury securities
Agency debt
AAA Corp
AAA RMBS (private label), CMBS, and ABS
Non-AAA Corp
Non-AAA RMBS (private label), CMBS, and ABS
Equity
2003 2007
Share AAA: 35.8% Share AAA: 32.0%
Note: RMBS: residential mortgage-backed securities; CMBS: commercial mortgage-backed securities; ABS: asset-backed securities other than RMBS and CMBS.
Source: Staff estimates based on Flow of Funds and Treasury International Capital system data.
23
5

30-year fixed-rate mortgage interest rate and 10-year Treasury bond rate
20
18
16
14
Mortgages
12
10
8
6
4
2
10 ‐ year
Treasurys
0
1971 1975 1979 1983 1987 1991 1995 1999 2003 2007 2011 2015
6

Mortgages v. Treasurys during the expansion, crisis, and recovery
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
2000
Spread   of   mortgages   over   10 ‐ year   Treasurys
2002 2004 2006 2008 2010 2012 2014
7

How strong is the effect of the lower interest rate?
The Gordon dividend-growth model applied to housing may help on this: f p = r − g a lower interest rate r or a higher flow-value growth rate g raises the price of housing p
8 g = r − f p
·

How strong is the effect of the lower interest rate?
The Gordon dividend-growth model applied to housing may help on this: f p = r − g a lower interest rate r or a higher flow-value growth rate g raises the price of housing p g = r − f p
·
8

Ratio of flow value of owner-occupied real estate to market value
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
1990 1994 1998 2002 2006 2010
9

Implied nominal expected growth rates of flow value
0.06
0.05
0.04
0.03
0.02
0.01
0.00
‐ 0.01
‐ 0.02
‐ 0.03
‐ 0.04
1990 1994 1998 2002 2006 2010
10

Conclusions about the role of the interest rate
Over the period from 1993 through 2006, the implied expected growth rate was close to constant, meaning that all of the rise in housing stock prices relative to flow prices came from the falling interest rate
11
But the collapse of prices starting in 2007 was not at all the result of rising interest rates, but rather of a large decline in the expected growth of the flow value of housing
·

Conclusions about the role of the interest rate
Over the period from 1993 through 2006, the implied expected growth rate was close to constant, meaning that all of the rise in housing stock prices relative to flow prices came from the falling interest rate
But the collapse of prices starting in 2007 was not at all the result of rising interest rates, but rather of a large decline in the expected growth of the flow value of housing
·
11

Supply of housing
The model assumes a fixed supply of houses—an increase in demand raises house prices, and does not stimulate new construction and the shift of land into housing and out of other uses
12
In reality, many markets have elastic supply of new houses and house price did not rise very much during the bubble
A paper by Albert Saiz has demonstrated this point, and provides strong evidence of the heterogeneity of housing markets along this dimension
·

Supply of housing
The model assumes a fixed supply of houses—an increase in demand raises house prices, and does not stimulate new construction and the shift of land into housing and out of other uses
In reality, many markets have elastic supply of new houses and house price did not rise very much during the bubble
12
A paper by Albert Saiz has demonstrated this point, and provides strong evidence of the heterogeneity of housing markets along this dimension
·

Supply of housing
The model assumes a fixed supply of houses—an increase in demand raises house prices, and does not stimulate new construction and the shift of land into housing and out of other uses
In reality, many markets have elastic supply of new houses and house price did not rise very much during the bubble
A paper by Albert Saiz has demonstrated this point, and provides strong evidence of the heterogeneity of housing markets along this dimension
·
12

Albert Saiz,
QJE
1285
13
(a)
12.5
12
11.5
11
0 .5
1
Inverse of supply elasticity
Log median house value
1.5
Fitted values
2
3 (b)
2.5
13
2
1.5
0 .5
1
Inverse of supply elasticity
Log price 2000 – log price 1970
1.5
Fitted values
F
IGURE
II
Estimated Elasticities and Home Values (2000)
(a) Levels, (b) changes.
2

Modeling the opening of the mortgage market
The paper’s model takes the opening to be a simple upward shift in an exogenous constraint on all mortgage finance
14
This assumption contradicts the evidence that the traditional
30-year fixed mortgage is tightly linked to other debt markets
More subtle modeling would recognize the specific importance of new types of mortgages, which extended mortgage financing to borrowers previously excluded
·

Modeling the opening of the mortgage market
The paper’s model takes the opening to be a simple upward shift in an exogenous constraint on all mortgage finance
This assumption contradicts the evidence that the traditional
30-year fixed mortgage is tightly linked to other debt markets
14
More subtle modeling would recognize the specific importance of new types of mortgages, which extended mortgage financing to borrowers previously excluded
·

Modeling the opening of the mortgage market
The paper’s model takes the opening to be a simple upward shift in an exogenous constraint on all mortgage finance
This assumption contradicts the evidence that the traditional
30-year fixed mortgage is tightly linked to other debt markets
More subtle modeling would recognize the specific importance of new types of mortgages, which extended mortgage financing to borrowers previously excluded
·
14

Figure 8
Expansion of non-traditional mortgages
Value of mortgages outstanding
Billions of dollars
12000
Total
Subprime
Variable rate, other prime, and alt-A
Prime fixed-rate and FHA/VA
10000
8000
2000 2001
Source: Federal Reserve Board staff estimates.
2002
RMBS, CMBS, and ABS outstanding, by rating
2003
All outstanding
Unrated
Junk, private label
Other Investment Grade, private label
AAA-rated, private label
Agency MBS*
2004 2005 2006 2007
0
15
Billions of dollars
9000
8000
7000
6000
5000
4000
3000
2000
1000
2001 2002 2003
28
2004 2005 2006 2007
0
2000
Note: RMBS: residential mortgage-backed securities; CMBS: commercial mortgage-backed securities; ABS: asset-backed securities other than RMBS and CMBS.
Source: Flow of Funds and staff estimates based on Dealogic data.
6000
4000
2000

The fax machine removed a key mortgage bottleneck
16
The fax was the first practical way to distribute the terms of mortgages in the wholesale market to independent brokers, who proved to be adept at helping previously excluded classes of borrowers obtain loans
·

The fax machine removed a key mortgage bottleneck
The fax was the first practical way to distribute the terms of mortgages in the wholesale market to independent brokers, who proved to be adept at helping previously excluded classes of borrowers obtain loans
·
16

Concluding remarks
The paper is on the right track in emphasizing the supply of finance to housing
17
Rough calculations suggest that all of the price increase resulted from declining interest rates; innovation in mortgage forms and extension of eligibility to less credit-worthy borrowers is secondary
So the paper’s characterization of the rise in supply might better be the connection to the world interest rate, rather than a shift of a fixed constraint.
·

Concluding remarks
The paper is on the right track in emphasizing the supply of finance to housing
Rough calculations suggest that all of the price increase resulted from declining interest rates; innovation in mortgage forms and extension of eligibility to less credit-worthy borrowers is secondary
17
So the paper’s characterization of the rise in supply might better be the connection to the world interest rate, rather than a shift of a fixed constraint.
·

Concluding remarks
The paper is on the right track in emphasizing the supply of finance to housing
Rough calculations suggest that all of the price increase resulted from declining interest rates; innovation in mortgage forms and extension of eligibility to less credit-worthy borrowers is secondary
So the paper’s characterization of the rise in supply might better be the connection to the world interest rate, rather than a shift of a fixed constraint.
·
17