Chapter 6


Charter Repeal Would Magnify Jumbo Market Weaknesses

Some advocates for charter repeal argue that because the jumbo market operates without access to Freddie Mac and Fannie Mae, the entire conventional market could continue to function as well as today without the government charters. Compared to the conforming market, however, the jumbo market exhibits many undesirable characteristics.

Private-label securities, often used to finance jumbo mortgages, lack the standardization, the volume and the corporate guarantees of Freddie Mac and Fannie Mae securities. Thus, the private-label market is much less liquid, and the investor base is much more narrow. As a result, jumbo mortgage rates are higher and more variable across regions, especially for areas experiencing economic weakness. In addition, more jumbo borrowers are compelled to choose ARMs. These systemic weaknesses likely would characterize the entire conventional market if Freddie Mac’s and Fannie Mae’s charters were repealed.

In addition, the use of today’s jumbo market as a model for understanding how the conventional market would function without the charters ignores the extent to which the jumbo market piggybacks on the conforming market. That is, jumbo borrowers benefit from the existence of a well-functioning conforming secondary market and the significant innovations that have been developed by Freddie Mac and Fannie Mae.

A. What Is the Jumbo Market?

A jumbo mortgage is a conventional mortgage whose loan balance at origination exceeds the Freddie Mac and Fannie Mae conforming loan limit. Jumbo originations generally comprise between 15 to 20 percent of the dollar volume of single-family originations. During 1995, jumbo originations totaled approximately $110 billion, or 17 percent of all single-family originations.1

In the primary market, there is little difference between the type of lenders originating jumbo loans and those originating conforming mortgages. Many lenders originate both types of loans. Commercial banks, including their mortgage banking subsidiaries, are the dominant lenders in both markets, accounting for 47 percent of conventional, conforming originations and 45 percent of the large jumbo loans made in 1994. Independently owned mortgage companies and thrifts also account for similar shares of the conforming and jumbo markets. 2

In the secondary market, radical differences between the jumbo and conforming markets emerge. While Freddie Mac and Fannie Mae play a critical role in assuring the availability of low-cost mortgages in the conforming market, the jumbo secondary market has no entities of that scope. As a result, the jumbo market is characterized by a far greater share of mortgages retained in the portfolios of depositories, and a far smaller share sold into the secondary market. Similarly, a much smaller volume of jumbo loans are financed with mortgage-backed securities, and the investor base is much more concentrated. In both 1994 and 1995, less than one in five jumbo loans was securitized; mortgage companies, including those specializing in wholesale activities, issued about two-thirds of the securities, followed by investment banks (19 percent) and depositories (10 percent).3

B. Jumbo Mortgage Rates Are Higher than Conforming Rates

Mortgage rates on jumbo loans are higher than rates on conforming loans, particularly for fixed-rate mortgages. From July 1986 to March 1996, the jumbo-conforming interest-rate spread on 30-year fixed-rate loans ranged from 0.11 to 0.70 percentage points, as shown in (Exhibit 16). The average spread over that period was 0.43 percentage points. The shaded area represents a core spread range of 0.26 to 0.60 percentage points.4 While this range reflects the differences borrowers may detect between rates on fixed-rate jumbo and conforming loans, it underestimates the benefits that Freddie Mac and Fannie Mae provide. To estimate the impact of repealing the charters, two other factors must be considered.

About 20 to 25 percent of jumbo mortgages have loan balances near the conforming loan limit. Mortgage rates on these loans are about 0.15 percentage points lower than rates on larger jumbos because near-conforming jumbos compete directly with conforming loans and may be purchased by Freddie Mac and Fannie Mae if the loan limit increases. 5 Because this price reduction is already embedded in the average jumbo rate, adjusting the actual average spread for this spillover effect of the two companies on near-conforming loans increases the range of the jumbo-conforming spread by about 0.03 to 0.04 percentage points. In addition, because average fixed costs of loan originations decline as loan size increases, jumbo rates should be, all else equal, about 0.05 to 0.10 percentage points below conforming rates. 6 A conservative estimate of the combined value of these two adjustments is 0.07 percentage points. Adding this amount to the observed jumbo-conforming spread for fixed-rate mortgages increases the range to 0.33 to 0.67 percentage points. Today, for example, consumers might pay 7.50 percent for a fixed-rate conforming mortgage, for which they otherwise would have to pay between 7.83 and 8.17 percent. If the charters were repealed, however, everyone in the conventional market would pay these higher rates or more because of the difficulties that the current jumbo financing vehicles would have handling the volume of the entire market.

Freddie Mac and Fannie Mae also lower rates on ARMs, but the effect is much smaller than for fixed-rate loans. Depositories have advantages in originating ARMs for their portfolios because they can match their assets and liabilities by funding these mortgages with deposits. First-year rates on one-year, Treasury-indexed conforming ARMs are about 0.1 percentage points below rates on similar jumbo ARMs, that is, conforming ARMs tend to have lower initial rates. This differential generally disappears once the ARMs adjust.7

Homebuyers are very sensitive to the significant interest-rate differences between jumbo and conforming fixed-rate mortgages. Year after year, Freddie Mac purchases a large number of mortgages at the conforming loan limit. This clustering suggests that many homebuyers in areas with high housing costs increase their down payments to avoid using jumbo loans with their higher interest costs.

C. Jumbo Rates Vary More by Region

Prior to the creation of Freddie Mac, mortgage rates varied significantly across the country. As a result of the national scope of the activities of Freddie Mac and Fannie Mae, conforming mortgage rates are today highly uniform. In contrast, jumbo mortgage rates vary significantly within and among different regions of the country. Exhibit 17 shows that for the U.S. as a whole, jumbo fixed-rate mortgage rates are 54 percent more variable than conforming rates, based on a nationwide sample of lenders offering both products. In the Midwest, jumbo rates are more than twice as variable as conforming rates.8

Sensitivity to regional economic downturns explains much of the high variability of jumbo loan rates. During a regional recession, property values soften. In response, depositories attempt to shore up their capital bases, often reducing their portfolio lending. 9 This restriction in credit availability manifests itself as a combination of higher loan rates and stricter underwriting standards by portfolio lenders.

Exhibit 18 highlights one instance during the 1990s when a regional economic downturn caused local jumbo rates to rise above the national average jumbo rate. Housing values peaked in 1989 in Massachusetts and then fell sharply in 1990 and 1991 as the New England economy went into recession.10 During this period of local economic recession and weakness in the housing market, mortgage rates on jumbo fixed-rate loans in Boston spiked 0.2 percentage points relative to the national average jumbo rate. In contrast, conforming mortgage interest rates averaged slightly less than the national average conforming rate, providing conforming borrowers continued access to low-cost mortgages.11

In the event of charter repeal, these regional differences likely would prevail in the overall conventional market. In fact, regional differences probably would exceed current jumbo market imbalances because today’s conforming borrowers, as a group, require significantly more credit than jumbo borrowers.

The return to a regional constellation of mortgage markets could hit rural areas particularly hard. Studies of credit and banking markets generally have found that areas characterized by low levels of competition among financial institutions also exhibit higher loan rates and fees. 12 Following charter repeal, this pattern likely would reemerge in home mortgage lending. Lacking direct access to national markets, rural and small urban areas likely would become less competitive and higher cost markets.

In contrast, mortgages flow through today’s vast conforming market to all parts of the nation with little variation in price. Loan terms are largely equivalent in both rural and urban markets, and the number of originations per property is generally constant across the country.13 Whether they reside on mountain tops or in large metropolitan areas, Americans have access to low-cost mortgages.

Finally, regional downturns would cause some lending and insuring institutions to exit the housing finance system. The focused charters of Freddie Mac and Fannie Mae ensure that they will continue to provide a stable supply of low-cost mortgages for qualified borrowers in all markets.

D. ARM Lending Is Much Greater in the Jumbo Market

Because of the activities of Freddie Mac and Fannie Mae, mortgage interest rates on conforming fixed-rate mortgages are at least one-half of a percentage point lower than they otherwise would be. ARM rates, however, are similar in the jumbo and conforming markets. As a result, the difference between rates on fixed-rate loans and ARMs is much greater in the jumbo market, giving jumbo borrowers a stronger incentive to choose ARMs. For example, in January 1996, average first-year interest rates on jumbo ARMs were 2.0 percentage points below rates on jumbo fixed-rate mortgages. In contrast, mortgage rates on conforming ARMs were 1.5 percentage points below rates on conforming fixed-rate mortgages. 14

As a result of the wider spread, the ARM share of lending in the jumbo market is generally more than double the share in the conforming market, as shown in Exhibit 19. In choosing a jumbo ARM over the more costly fixed-rate jumbo loan, families are accepting the risk of potentially sharp upward increases in monthly payments. Because of the preponderance of ARMs in the jumbo market, more interest-rate risk is being borne by individual borrowers--not savvy institutional investors who have the ability to manage it.

E. The Jumbo Market Is Less Liquid than the Conforming Market

The major weaknesses of the jumbo market, such as higher and more variable mortgage rates, stem from the lower level of liquidity in that market compared to the conforming market. Less liquidity affects homeowners; it translates into higher mortgage rates.

One way to measure liquidity is in terms of trading volume. A large amount of trading is generally a sign of a liquid market. In 1994, more than 2.7 million newly originated conventional conforming loans were sold by lenders, compared with 1.4 million government-insured or guaranteed loans and 0.2 million jumbo loans.15

High trading volume in the conforming mortgage market is supported by a very liquid market for mortgage-backed securities. During 1995, nearly $8 trillion in Freddie Mac, Fannie Mae and Ginnie Mae mortgage securities were traded, equivalent to an average of five trades per security over the course of the year. 16 This high degree of liquidity enables depositories and other holders to trade these securities readily in response to changes in the market or in their portfolio needs.

Liquidity also can be measured by using bid-ask spreads. The bid price represents what a buyer is willing to pay for a security, while the ask price represents what a seller is willing to receive for a security. A large difference, or spread, between the two prices reflects trader expectations that finding a buyer or seller is relatively difficult. The higher the bid-ask spread, the higher the risk of the transaction and the greater the mark-up the trader expects to earn. In contrast, a low bid-ask spread implies a wide agreement on prices. Transactions involving Freddie Mac and Fannie Mae mortgage-backed securities have low bid-ask spreads because these securities can be bought and sold almost instantaneously. Low bid-ask spreads are one hallmark of a highly liquid market.

Bid-ask spreads for Freddie Mac and Fannie Mae securities can be seen on computerized broker screens. On these screens, mortgage traders post actual prices at which they are willing to buy and sell securities at any given time. These screens allow trades to be made through a third party; at the time of the trade the buyer and the seller are not identified to each other. Active trading on broker screens is another sign of a liquid market. In contrast, because of the need for detailed information about the issuer and the underlying mortgages, private-label securities are not generally traded in this way.

Exhibit 20 compares bid-ask spreads for Freddie Mac and Fannie Mae securities with bid-ask spreads for Treasury securities and for highly rated private-label mortgage-backed securities. Reflecting the high degree of liquidity for Treasury, Freddie Mac and Fannie Mae securities, their bid-ask spreads are low. Bid-ask spreads for private-label securities, in contrast, are several times those of Freddie Mac and Fannie Mae mortgage-backed securities or Treasury securities. While bid-ask spreads may be the last thing on a potential homebuyer’s mind, low spreads indicate investors are attracted to housing finance--and that means mortgages are low-cost and plentiful.

The liquidity advantage of Freddie Mac and Fannie Mae securities arises from greater issuance volume, standardization and the brand recognition and guarantees of the corporations’ mortgage-backed securities compared to private-label issues. Freddie Mac and Fannie Mae are well known in financial markets. In the private-label market, in contrast, there are a large number of issuers that are not nearly as well known to investors (which imposes research costs on each issue). In addition, the number of securities per issuer is small. For a given security issue, the pool of underlying mortgages tends to be much more homogeneous for Freddie Mac and Fannie Mae securities compared to private-label securities. These features make trading in private-market securities highly cumbersome. Investors must analyze the structure of each security, the composition of the pool of mortgages, and the extent and type of the credit enhancement. These difficulties are magnified over time. Initially, this information is disclosed by the issuer. In later trading, it is more difficult, for example, to assess whether the remaining credit enhancement is sufficient for the remaining mortgages. In contrast, updated information about the mortgages underlying Freddie Mac and Fannie Mae securities is readily available, and investors can rely on their corporate guarantees for the life of the securities.

The weaknesses in the jumbo market are reflected in the lower securitization rates for jumbo mortgages shown in Exhibit 21. Over the 1987-95 period, the securitization rate for jumbo loans averaged 19 percent, peaking at 25 percent in 1993; in contrast, the securitization rate for conventional conforming loans averaged 55 percent. Government-insured and guaranteed loans are securitized at a rate of 92 percent.

The weaknesses that impair trading in private-label securities contribute to the reduced liquidity of the overall jumbo mortgage market. Liquidity is not an abstract concept: if the entire housing finance system had to rely on illiquid markets, mortgage flows would be less certain and borrowers would pay higher rates.

F. The Jumbo Mortgage Market Has a Narrower Investor Base

Freddie Mac and Fannie Mae have succeeded in attracting a wide array of investors to housing finance. The growth in the investor base for mortgages has taken place primarily in the conforming market, but the jumbo market also has benefited. However, private-label securities remain much less liquid than the mortgage-backed securities guaranteed by Freddie Mac and Fannie Mae. Despite the temptation of higher yields, many investors stay away from the private-label market because of their preferences for liquid, lower-risk mortgage securities. As a result, the investor base for jumbo mortgages is much more narrow than the conforming investor base.

Exhibit 22 shows that in 1991 thrifts and commercial banks held 50 percent of conforming mortgages compared to 65 percent of jumbo mortgages. 17 Dominated by depositories, the jumbo mortgage market resembles the pre-1970 conventional market. Its highly concentrated investor base makes it less resilient to reductions by any one investor type, causing mortgage rates to rise during periods of stress.

One symptom of a narrow investor base for jumbo mortgages is increased fluctuation or volatility in the price of the mortgage-related instruments over time. Highly volatile securities carry more risk because the return on investment is less certain. As compensation for this additional risk, jumbo investors demand higher yields, which translate into higher mortgage interest rates for jumbo borrowers. In contrast, Freddie Mac’s and Fannie Mae’s superior access to investment funds keeps rates low and reduces volatility, resulting in more moderate movements in conforming rates.

One way to assess volatility is to compare the degree to which conforming and jumbo rates vary over the course of a year relative to a common benchmark: the yield on the ten-year Treasury bond. 18 Exhibit 23 shows that jumbo loan rates are consistently more volatile than conforming rates; in 1995 they were nearly twice as volatile.

The landscape of investors in a post-repeal housing finance system likely would resemble a hybrid of the investors in today’s jumbo and conforming markets. Because of the advantages that flow from federal deposit insurance and low-cost FHLB advances, depositories likely would play a much larger role. Deposits--which finance mortgages as well as business and other consumer credit needs--have grown by a factor of six, from $0.6 trillion in 1970 to $3.4 trillion today. In contrast, single-family conventional mortgage debt grew more than 15-fold over the same period, from $0.2 to $3.1 trillion. The significant disparity between the growth of deposits and the growth of mortgage debt suggests that depositories would need to increase deposit rates to attract additional deposits. This increased cost would be passed on to borrowers in the form of higher mortgage rates.

Alternatively, issues of private-label securities could increase to finance more home mortgages. The private-label securities market has severe limitations, however. By virtue of their charters and their financial strength, Freddie Mac and Fannie Mae provide credible corporate guarantees of the timely payment of the principal and interest on mortgage-backed securities. These guarantees provide investors with the assurance that securities issued by the two firms are high quality and will remain high quality throughout the 30 years of a mortgage’s life.

In contrast, private-label issuers rarely provide a corporate guarantee for their issues because they are not adequately capitalized to do so and the cost of raising additional capital is prohibitive. To achieve the same high rating of their securities, these firms must provide additional credit enhancement. Alternatives include pool insurance, which is expensive and rarely used; similar to issuers, mortgage insurers cannot provide a credible guarantee for 30 years without excessive capital costs. More common is the senior-subordinated structure in which payments from the mortgage pool are split into a senior (usually AA-rated or AAA-rated) piece and a subordinated (rated below investment grade, often B-rated or lower) security. The senior piece can be sold to a broad range of investors, but the low-rated security, which contains most of the credit risk of the pool of mortgages, is much more difficult to sell. The market for “B-pieces” is very limited because, commensurate with the risk of these securities, many institutional investors are required to hold significant capital for these assets. 19 In addition, expected income from the typically small B-security may not sufficiently reward investors for committing the often substantial resources necessary to evaluate the entire transaction. If the Freddie Mac and Fannie Mae charters were repealed, subordinated issuance would have to increase 500 percent to ensure sufficient capital to meet borrower needs. Higher capital requirements and restrictions on these securities would have a chilling effect on investor interest in mortgage-related securities.

In addition to the difficulty in finding investors for subordinated classes, senior classes of private-label issues would be less liquid and more expensive to issue than Freddie Mac and Fannie Mae securities. For example, each transaction would require rating-agency review of the composition of the mortgage pool to determine the size of the subordinated class needed. Still, senior class investors would not know if the subordinated class would offer protection for the life of the securities.

The intricacies of structuring mortgage investments affect homeowners. If, in the event of charter repeal, senior-subordinated structures became much more widely used, then security yields would have to increase to cover higher capital costs and continue to attract sufficient investors. These higher costs would be passed onto homeowners in the form of higher mortgage rates.

G. The Jumbo Market Piggybacks on Freddie Mac and Fannie Mae

In addition to the inherent weaknesses of the jumbo market, the use of that market as a model for post-repeal housing finance system overlooks the vast difference in the size of the two markets. For the jumbo market to do the work of today’s conforming market supported by Freddie Mac and Fannie Mae, a combination of private-label issuance and depository holdings would have to increase substantially. Such a massive increase could not be accomplished without substantial increases in mortgage costs and periodic credit shortages.

The use of the jumbo market as a post-repeal model also ignores the degree to which it now piggybacks on the benefits of the conforming mortgage market. For example, jumbo borrowers benefit from the ongoing standardization of the mortgage process and the growth of the investor base--all of which take place under the leadership of Freddie Mac and Fannie Mae. Consumers choosing jumbo loans that are near the conforming loan limit also benefit by having slightly lower rates than other jumbo borrowers.

In addition, the ability of jumbo borrowers to lock-in a given mortgage rate one or two months before the mortgage actually closes is a direct result of the highly liquid market for Freddie Mac and Fannie Mae mortgage-backed securities. This feature of the mortgage process is highly valued by borrowers, saving them interest costs and providing great peace of mind.

Because of the extent the jumbo market piggybacks on the conforming market, if Freddie Mac’s and Fannie Mae’s charters were repealed, all conventional borrowers would have to depend on a market less able to meet their needs than even today’s jumbo market.

H. Charter Repeal Would Bring Undesirable Jumbo Characteristics to the Conventional Market

Jumbo mortgage rates are higher and more variable; and jumbo borrowers, because so many are compelled to take out ARMs, bear greater risk of losing their homes through default. These problems would be magnified across the much larger conventional mortgage market if the charters of Freddie Mac and Fannie Mae were repealed.

These weaknesses arise from the lack of government-chartered corporations operating in the jumbo market. Without access to Freddie Mac and Fannie Mae, the secondary market for jumbo loans is hampered by lack of geographic diversification causing higher capital costs, a shallow investor base, a thin securities market, cumbersome trading, high research costs and greater risk.

The jumbo market does benefit from the conforming market’s standardization, innovations and securities for hedging. However, without the continuing operation of Freddie Mac and Fannie Mae, these benefits would soon disappear. For all these reasons, the jumbo market is an unacceptable substitute for the conforming mortgage market supported by Freddie


Footnotes:
1. Total single-family originations during 1995, including government-insured or guaranteed loans, were estimated at $636 billion by HUD, while conventional originations were estimated at $562 billion. Based on the Federal Housing Finance Board’s Mortgage Interest Rate Survey, 20 percent of the volume of conventional home purchase originations were jumbo loans during 1995. Thus, jumbo originations totaled about $110 billion (17 percent of $636 billion).
2. Tabulations from data reported under the 1994 Home Mortgage Disclosure Act. The data include loan amount and property location but not the number of units in the property. Thus, originations at or below the one-unit loan limit were used to tabulate the lender distribution for the conventional conforming market, yielding 47, 28 and 25 percent shares for commercial banks (including mortgage banking subsidiaries owned by banks), independently owned mortgage companies and thrifts (including mortgage banking subsidiaries owned by thrifts), respectively. Originations above the four-unit loan limit must be jumbo loans and were used to tabulate the lender distribution for the jumbo primary market, obtaining 45, 30 and 25 percent shares for commercial banks, mortgage bankers and thrifts, respectively.
3. The Resolution Trust Corporation accounted for 3 percent of issues. Inside Mortgage Securities, February 16, 1996.
4. The core spread represents a one standard deviation band about the mean. The difference in jumbo and conforming mortgage rates is particularly sensitive to changes in the cost of credit enhancement on jumbo loans and prepayment expectations.
5. Patric H. Hendershott and James D. Shilling, “The Impact of the Agencies on Conventional Fixed-Rate Mortgage Yields,” Journal of Real Estate Finance and Economics (1989:2), p. 112.
6. The average cost of originating a loan for a lender is approximately 2 percent of the loan balance; Mortgage Bankers Association, 1994 Cost Study, Operations Report No. 17, Table 2. If one-quarter to one-half of these costs are fixed, and the average jumbo loan is three times the amount of the average conforming loan, then jumbo loan rates should be 0.05 to 0.10 percentage points below conforming rates. A similar effect can be derived from the loan amount coefficient reported by Hendershott and Shilling, op. cit., p. 113.
7. “ARMs Lose Popularity as Interest Rates Drop in 1995,” Secondary Mortgage Markets, February 1996, p. 5. If the interest-rate cap is binding at the first adjustment, the conforming ARM will still have an interest rate that is slightly below the jumbo ARM, on average.
8. Findings taken from a survey of 88 lenders offering both jumbo and conforming fixed-rate, 80-percent LTV ratio mortgages, conducted the week ending February 16, 1996. Variability is measured by the coefficient of variation, which is equal to the standard deviation divided by the mean interest rate (adjusted for points).
9. See Joe Peek and Eric S. Rosengren, “Bank Real Estate Lending and the New England Capital Crunch,” Journal of the American Real Estate and Urban Economics Association, Spring 1994, pp. 33-58; and Diana Hancock and James A. Wilcox, “Bank Capital, Loan Delinquencies, and Real Estate Lending,” Journal of Housing Economics, June 1994, pp. 121-46.
10. This analysis uses the Conventional Mortgage Home Price Index rather than actual prices to show the relative change.
11. Mortgage rate data compiled by HSH Associates show that mortgage rates on jumbo fixed-rate loans in the Boston metropolitan area averaged 0.2 percentage points above the national average jumbo rate during the fourth quarter of 1990. In contrast, conforming rates in Boston averaged up to 0.1 percentage points below the national average each quarter of the recession.
12. R. Alton Gilbert, “Bank Market Structure and Competition: A Survey,” Journal of Money, Credit, and Banking, 16 (1984), pp. 617-45; and Timothy H. Hannan, “Bank Commercial Loan Markets and the Role of Market Structure: Evidence from Surveys of Commercial Lending,” Journal of Banking Finance, 15 (1991), pp. 133-49.
13. Effect of Federal Home Loan Bank System District Banks on the Housing Finance System in Rural Areas, report prepared for the Federal Housing Finance Board by ICF Incorporated, April 23, 1993, pp. 9-12; and William B. Shear, James Berkovec, Ann Dougherty and Frank E. Nothaft, “Unmet Housing Needs: The Role of Mortgage Markets,” Journal of Housing Economics, September 1995, pp. 299-300.
14. HSH Associates, week ending January 5, 1996.
15. Unpublished tabulation of 1994 Home Mortgage Disclosure Act data for single-family home-purchase or refinance loans. Conforming loans were defined to be loans at or below the one-unit loan limit; thus some conforming loans secured by two- to four-unit properties are included in the jumbo trading volume. Also, only loans originated or purchased within 1994 are covered in the data, excluding the bulk of seasoned loan trades.
16. “U.S. Government Securities Dealers Transactions,” Federal Reserve Bulletin, Table 1.42, various issues. At mid-year, $1.5 trillion in Freddie Mac, Fannie Mae and Ginnie Mae mortgage-backed securities were outstanding, implying that each security dollar traded an average of five times over the year.
17. Conforming loan percentages include holdings of whole loans and Freddie Mac and Fannie Mae mortgage-backed securities. Whole loans placed in private-label securities could not be reallocated to the holder of the security. Data refer to holders as of 1991 of loans originated 1989-91. Most recent unpublished data from the Residential Finance Survey (1991).
18. Exhibit 23 contrasts the difference between the highest and the lowest spread between the prevailing conforming 30-year, fixed-rate mortgage rate and the ten-year Treasury rate with the maximum jumbo-ten-year Treasury spread. The difference between the lowest rate for the year and the highest rate for the year was used to measure the volatility of the series.
19. For depositories, the capital requirement for the B-piece is generally the lesser of the dollar amount of the B-piece or 4 percent of the dollar value of the whole pool (including the senior piece). For example, a bank holding a Freddie Mac or Fannie Mae security would have a capital requirement of 1.6 percent. In contrast, a bank holding a B-piece equal to 4 percent of the total pool amount would have to hold capital equal to 100 percent of the B-piece--or 62 times more capital. For life insurance companies that issue senior securities and retain the subordinate piece, regulators require the companies to hold the lesser of the statement value of the subordinated piece or the required capital for the whole pool. For insurance companies that purchase B-pieces the capital charge ranges up to 30 percent of book value, compared with 0.3 percent for high-rated securities. The capital requirement for depositories holding highly rated senior pieces is also higher than it is for holding Freddie Mac and Fannie Mae securities.



© 2009 Freddie Mac close window