We define "prime" mortgages as those extended to borrowers with a credit score greater than or equal to 660 and a LTV ratio of 80 percent or less. "Near-prime" mortgages are defined as those extended to borrowers with a credit score greater than or equal to 580 but less than 660 and a LTV ratio less than or equal to 90 percent, or to borrowers with an LTV ratio greater than 80 percent and less than or equal to 90 percent and a credit score above 660. "Subprime" mortgages are defined as those extended to borrowers with a credit score less than or equal to 580 or an LTV ratio greater than 90 percent.
Figure 2.1 provides data on the US first-lien conventional mortgage market. In the top row is total mortgages outstanding. In the second row, these mortgages are broken out into prime and non-prime mortgages. In the third row on the left, prime mortgages are broken out into prime conforming and prime jumbo mortgages. In the third row on the right, non-prime mortgages are broken out into near-prime and subprime mortgages. In the fourth row, prime, prime jumbo, near-prime and subprime mortgages are broken out into non-securitized and securitized mortgages. In the fifth row, the number of non-securitized and securitized mortgages that are fixed rate are provided. And in the sixth row, the number of non-securitized and securitized mortgages that are adjustable rate are provided.
Figure 3.1 graphically portrays the economic capital concepts described in the text. In the top panel, the vertical axis, which measures capital in percent of mortgage-backed assets, ranges from 0 to 10 percent. And the horizontal axis, which measures the credit quality index, ranges from 0.10 (for high-risk mortgages) to 1.0 (for low-risk mortgages). FRB economic capital estimates that are based on stress-based losses range from about 5.5 percent to about 20 basis points and these estimates with respect to the credit quality index trace out a pattern that is convex to the origin. Also in the top panel of figure 3.1 are estimates of Basel II capital requirements that assume a 0.15 asset correlation and that are based on (1) FRB economic capital estimates and (2) stress-based expected losses. The estimate of Basel II capital requirements for mortgages of different credit qualities are always equal to or above the corresponding FRB economic capital estimate for the same credit quality. The outer boundary of the Basel II regulatory capital estimates is convex to the origin and is a fairly close approximation to the FRB credit risk model economic capital estimates.
In the bottom panel of figure 3.1, the vertical axis, which measures capital in percent of mortgage-backed assets, ranges from 0 to 10 percent. And the horizontal axis, which measures the credit quality index, ranges from 0.10 (for high-risk mortgages) to 1.0 (for low-risk mortgages). The convex outer boundary of the Basel II regulatory capital estimates is presented and ranges from about 5.8 to about 20 basis points. This boundary is always below our prudent economic capital estimates that are consistent with BBB+ to A- ratings with the distance between the convex outer boundary of Basel II regulatory capital estimates and the prudent economic capital estimates becoming smaller as one moves from high-risk mortgages (with credit quality index values close to zero in value) to low-risk mortgages (with credit quality index values close to one in value). The prudent economic estimates range from about 9.25 percent for mortgages with a credit quality index equal to 0.10 to about 22 basis points for mortgages with a credit quality index equal to 0.94. In contrast, Basel I capital requirements for first-lien whole loans are equal to 4 percent regardless of the value of the credit quality index. With regard to mortgage-backed securities, three regulatory capital requirements are presented: (1) the Basel I capital requirements for AAA-rated securities are equal to 1.6 percent, regardless of the value of the credit quality index; (2) the Basel II capital requirements for AAA-rated securities are equal to 0.56 percent, regardless of the value of the credit quality index; and, (3) GSE minimum capital requirement for credit risk equals 0.45 percent, regardless of the value of the credit quality index.
Rating Category | Rating Designation Examples |
Risk Weight (in percent) |
---|---|---|
Highest or second-highest investment grade | AAA, AA | 20 |
Third highest investment grade | A | 50 |
Lowest investment grade | BBB | 100 |
One category below investment grade | BB | 200 |
Long-term Credit Rating and/or Inferred Rating Derived from a Long-term Assessment (1) |
Risk-weight for Senior Positions (in percent) (2) |
Base Risk Weights (in percent) (3) |
---|---|---|
AAA | 7 | 12 |
AA | 8 | 15 |
A+ | 10 | 18 |
A | 12 | 20 |
A- | 20 | 35 |
BBB+ | 35 | |
BBB | 60 | |
BBB- | 100 | 100 |
BB+ | 250 | 250 |
BB | 425 | 425 |
BB- | 650 | 650 |
Below BB- and unrated | Deduction | Deduction |
Source: Basel Committee on Banking Supervision (2004, pp. 128-129). Deductions of investments will be 50 percent from tier 1 and 50 percent from tier 2 capital.
Figure has top and bottom panels that correspond to the Basel I regime and Basel II regime, respectively. In each panel, the vertical axis measures the mortgage rate and the horizontal axis measures the probability of not defaulting on the loan, which ranges from 0 (high risk) to one (low risk). Mortgage funding cost curves for a given mortgage market segment are presented.
For the Basel I regime (top panel), the marginal cost function assuming a fixed percent of capital per dollar of mortgage assets (e.g., 4 percent) for depositories is represented by a convex to the origin line and the marginal capital cost associated with prudent economic capital is represented by a steeper convex to the origin line. These marginal costs intersect with each other for moderately high-risk loans with a fairly low probability of not defaulting. Because mortgage securitizers may also have capital requirements that are a fixed percent of capital, the mortgage securitization cost function is parallel to, but closer to, the horizontal axis than is the marginal cost function for depositories. A vertical line, determined by securitizer underwriting standards indicates the lowest borrower quality level that the securitizer is willing to accept. Mortgages with probabilities of not defaulting less than this level are either not extended or are held by depository institutions. Another vertical line, determined by limits to mixing high-risk and low-risk mortgages to attain an average regulatory capital level consistent with the regulatory (fixed 4 percent) minimum is to the right of the first vertical line. Mortgages with lower-risk (that is, with higher probabilities of not defaulting) are held by depository institutions. And, mortgages with probabilities of defaulting between the two vertical lines are securitized. The leftmost cross-hatched area is bounded above by the marginal cost function assuming a fixed percent of capital per dollar of mortgage assets, bounded below by the marginal capital cost associated with prudent economic capital, and bounded to the right by the vertical line determined by securitizer underwriting standards. The rightmost cross-hatched area is bounded above by the mortgage securitization marginal cost function, bounded below by the marginal cost associated with prudent economic capital, and bounded to the right by the vertical line determined by limits to mixing high-risk and low-risk mortgages to attain an average regulatory capital level consistent with the minimum.
For the Basel II regime (bottom panel), the adopters cost function is identical to the marginal capital cost associated with prudent economic capital. The nonadopters cost function is equal to the adopters cost function for mortgages to the left of the leftmost cross-hatched area (identical to the leftmost cross-hatched area in the top panel); it is equal to the marginal cost function assuming a fixed percent of capital for the leftmost cross-hatched area; it is equal to the adopter cost function between the cross-hatched areas; it is equal to the mortgage securitization marginal cost function for the rightmost cross-hatched area (identical to the rightmost cross-hatched area in the top panel); and it is equal to the adopters cost function to the right of the rightmost cross-hatched area. The most risky mortgages (those with probabilities of not defaulting that are very low and to the left of the leftmost cross-hatched area) are either not extended or held by depository institutions. And the least risky mortgages (those with probabilities of not defaulting that are very high and to the right of the rightmost cross-hatched area) are held by depository institutions. Mortgages with probabilities of not defaulting that correspond to the two cross-hatched areas are held by adopters. And mortgages with probabilities of not defaulting that lie between the two cross-hatched areas are securitized.
Figure 4.2 presents possible equilibria in a given segment of the mortgage market for the Basel II regime. The vertical axis measures the mortgage rate and the horizontal axis measures the probability of not defaulting on the loan, which ranges from 0 (high risk) to one (low risk). The adopter and nonadopter cost functions for mortgages in this figure are identical to those that were presented in the bottom panel of figure 4.1. Demand curve D1 intersects these curves to the left of the leftmost cross-hatched area and demonstrates that interest rates would not change with the implementation of Basel II since both adopters and nonadopters cost functions are equal to each other for these very high-risk mortgages. Alternatively, demand curve D2 intersects these curves inside the leftmost cross-hatched region. In this region, Basel II implementation could potentially lower mortgage rates, since deadweight losses (represented by the cross-hatched area) would potentially be reduced. Alternatively, demand curve D3 intersects the adopter and nonadopter curves where they are equal to each other because the mortgages are securitized and because market-determined capital would back mortgages of such quality. In this case, mortgage rates would be unaffected by Basel II implementation. Alternatively, demand curve D4 intersects the adopter and nonadopter curves in the rightmost cross-hatched region, where the mortgage securitization marginal cost function is relevant for nonadopters, but the marginal cost associated with prudent economic capital is relevant for adopters. In this case, mortgage rates could fall with the implementation of Basel II capital standards, since deadweight losses (represented by the cross-hatched area) would be reduced.
Note: Standard errors and t-statistics (in parentheses) were corrected for multiple imputation using tools available on the Survey of Consumer Finances website. See Kennickell, McManus, and Woodburn (1997) for a discussion of the techniques that were used. Parameters that are significant at the 10 percent level have one asterisk next to them, and those that are significant at the 5 percent level have two asterisks next to them.
EXPLANATORY VARIABLES | DEPENDENT VARIABLE: Spread of the Mortgage Rate over a Constructed Freddie Mac Benchmark Rate | ||||||||
---|---|---|---|---|---|---|---|---|---|
Model 1 | Model 2 | Model 3 | |||||||
The Effect for Prime Loans (1) |
Additional Effect for Near-Prime Loans (2) |
Additional Effect for Subprime Loans (3) |
The Effect for Prime Loans (4) |
Additional Effect for Near-Prime Loans (5) |
Additional Effect for Subprime Loans (6) |
The Effect for Prime Loans (7) |
Additional Effect for Near-Prime Loans (8) |
Additional Effect for Subprime Loans (9) |
|
INTERCEPT | -0.084 (-0.76) |
-0.007 (-0.03) |
0.539 ** (3.30) |
0.632 (0.24) |
0.621 (0.24) |
3.484 ** (2.45) |
1.593 (0.65) |
1.549 (0.56) |
4.792 ** (2.33) |
CREDIT QUALITY MEASURES | |||||||||
Credit Score | -0.001 (-0.36) |
0.001 (-0.20) |
-0.003 (-0.76) |
-0.002 (-0.78) |
0.001 (0.13) |
-0.001 (-0.36) |
|||
Loan-to-Value Ratio | 0.009 (1.42) |
0.014 (-0.40) |
-0.010 (-1.08) |
0.006 (0.94) |
-0.006 (-0.19) |
-0.011 (-1.23) |
|||
LOCATION | |||||||||
East |
-0.172 (-1.09) |
-0.069 (-.025) |
0.259 (0.57) |
-0.457 (-1.09) |
|||||
Mountain | -0.926 ** (-3.14) |
-0.537 (-1.19) |
0.852 (0.82) |
0.141 (0.22) |
|||||
Pacific | 0.217 (0.91) |
0.257 (0.64) |
-0.302 (-0.35) |
0.035 (0.06) |
|||||
Rural | 0.766 ** (2.60) |
-0.198 (-0.46) |
2.106 ** (1.96) |
1.405 ** (2.34) |
|||||
Herfindahl Index | 0.648 (0.63) |
2.477 (1.23) |
-2.757 (-0.90) |
-1.766 (-0.73) |
|||||
YEAR INDICATOR VARIABLES | |||||||||
1999 | -0.696 ** (3.20) |
-0.366 (-1.24) |
0.073 (0.14) |
-0.501 (-.92) |
|||||
2000 | -0.737 ** (-3.39) |
-0.651 * (-1.78) |
0.375 (0.55) |
-0.252 (-0.37) |
|||||
2001 | -0.140 (-0.50) |
-0.454 (-1.22) |
0.705 (0.82) |
0.942 (1.51) |
|||||
LENDER TYPE | |||||||||
Depository | -0.114 (-0.23) |
-0.284 (-0.35) |
-0.447 (-0.43) |
||||||
Mortgage Company | -0.169 (-0.31) |
-0.406 (-0.48) |
-0.940 (-0.92) |
||||||
Credit Union | -0.207 (-0.278) |
-- | -1.231 (-1.00) |
||||||
Family or Friend | -4.36 ** (-4.41) |
3.353 (1.06) |
2.560 * (1.79) |
||||||
R-Square (Adjusted R-Square) | 0.0483 (0.0429) | 0.2386 (0.1904) | 0.4432 (0.3245) |
Rankings by Asset Size Are Computed as of 2003
* Bank holding company information derived as the sum of subsidiary bank data. Core deposits consist of transactions deposits, savings accounts, and small time deposits. Managed liabilities mainly consist of deposits booked in foreign offices, large time deposits (with values equal or greater than $100,000) and subordinated notes and debentures.
BANK HOLDING COMPANY SIZE Balance Sheet Items |
1999 | 2001 | 2003 | |||
---|---|---|---|---|---|---|
Billions of Dollars | Percent of Total Assets | Billions of Dollars | Percent of Total Assets | Billions of Dollars | Percent of Total Assets | |
10 LARGEST | ||||||
Total Liabilities | ||||||
Core Deposits * | 899 | 30 | 1199 | 30 | 1387 | 30 |
Managed Liabilities * | 1401 | 47 | 1816 | 45 | 2138 | 46 |
Other | 447 | 15 | 693 | 17 | 773 | 16 |
Total Equity Capital | 212 | 7 | 289 | 7 | 328 | 7 |
Total Assets | 3011 | 100 | 4048 | 100 | 4695 | 100 |
NEXT 10 LARGEST | ||||||
Total Liabilities | ||||||
Core Deposits * | 244 | 36 | 300 | 37 | 345 | 36 |
Managed Liabilities * | 267 | 39 | 287 | 35 | 337 | 35 |
Other | 37 | 5 | 58 | 7 | 68 | 7 |
Total Equity Capital | 50 | 7 | 65 | 8 | 82 | 9 |
Total Assets | 675 | 100 | 817 | 100 | 957 | 100 |
NEXT 30 LARGEST | ||||||
Total Liabilities | ||||||
Core Deposits * | 297 | 42 | 360 | 42 | 418 | 41 |
Managed Liabilities * | 239 | 34 | 269 | 31 | 314 | 31 |
Other | 22 | 3 | 61 | 7 | 68 | 7 |
Total Equity Capital | 56 | 8 | 80 | 9 | 97 | 10 |
Total Assets | 706 | 100 | 864 | 100 | 1015 | 100 |
ALL OTHER REPORTING BANK HOLDING COMPANIES | ||||||
Total Liabilities | ||||||
Core Deposits * | 840 | 51 | 818 | 49 | 989 | 47 |
Managed Liabilities * | 510 | 31 | 382 | 23 | 542 | 26 |
Other | 126 | 8 | 277 | 17 | 355 | 17 |
Total Equity Capital | 128 | 8 | 143 | 9 | 185 | 9 |
Total Assets | 1639 | 100 | 1656 | 100 | 2115 | 100 |
Source: Bank Holding Company Consolidated Reports and bank Call Reports as of December 31.
Figure 5.1 presents median estimated prudent economic capital (in basis points) for mortgages in our three broad credit risk market segments.
For mortgages in the prime mortgage market segment for which prudent economic capital falls below 45 basis points, the median prudent economic capital estimate is 20 basis points. For mortgages in the prime market segment for which prudent economic capital lies between 45 and 400 basis points, the median prudent economic capital estimate is 65 basis points. There were no mortgages in the prime mortgage market segment for which prudent economic capital estimate was in excess of 400 basis points.
For mortgages in the near-prime mortgage market segment for which prudent economic capital falls below 45 basis points, the median prudent economic capital estimate is 30 basis points. For mortgages in the near-prime mortgage market segment for which prudent economic capital lies between 45 basis points and 400 basis points, the median prudent economic capital estimate is 150 basis points. For mortgages in the near-prime mortgage market segment for which prudent economic capital is in excess of 400 basis points, the median prudent economic capital estimate is 475 basis points.
There were no mortgages in the subprime mortgage market segment for which prudent economic capital was less than 45 basis points. For mortgages in the subprime mortgage market segment with LTVs less than or equal to 90 percent and for which prudent economic capital lies between 45 basis points and 400 basis points, the median prudent economic capital estimate is 190 basis points. For mortgages in the subprime mortgage market segment with LTVs greater than 90 percent and for which prudent economic capital lies between 45 basis points and 400 basis points, the median prudent economic capital estimate is 295 basis points. And for mortgages in the subprime mortgage market segment for which prudent economic capital was greater than 400 basis points, the median prudent economic capital estimate in 725 basis points.
Figure 5.2 presents the percent of loans in each mortgage market segment where (1) prudent economic capital falls below 45 basis points; (2) prudent economic capital lies between 45 and 400 basis points, and (3) prudent economic capital is in excess of 400 basis points. In the prime market segment, 88 percent of loans have prudent economic capital estimates less than 45 basis points and 12 percent of loans have prudent economic capital estimates that lie between 45 and 400 basis points. In the near-prime mortgage market segment, 7 percent of loans have prudent economic capital estimates less than 45 basis points, 84 percent of loans have prudent economic capital estimates that lie between 45 and 400 basis points, and 9 percent of loans have prudent economic capital estimates in excess of 400 basis points. In the subprime mortgage market segment, about 43 percent of loans have prudent economic capital estimates that lie between 45 and 400 basis points, and 57 percent of loans have prudent economic capital estimates in excess of 400 basis points.
Figure 5.3 provides data on the US first-lien conventional mortgage market. In the top row is total mortgages outstanding. In the second row, these mortgages are broken out into prime and non-prime mortgages. In the third row on the left, prime mortgages are broken out into prime conforming and prime jumbo mortgages. In the third row on the right, non-prime mortgages are broken out into near-prime and subprime mortgages. In the fourth row, prime, prime jumbo, near-prime and subprime mortgages are broken out into non-securitized and securitized mortgages. In the fifth row, the number of non-securitized and securitized that are fixed rate are provided. In addition, for fixed non-securitized mortgage estimates we provide the maximum number of loans contested by adopters and nonadopters over a long-period, and for fixed securitized mortgages we provide estimates for the maximum number of mortgages contested by adopters and securitizers over a long period. Finally, in the sixth row, the number of non-securitized and securitized mortgages that are adjustable rate are provided.