Data are from the Residential Finance Survey of the 1970 Census of Housing.

Yield spreads are the annual averages of the weekly difference between the 30-year, fixed-rate mortgage rate from Freddie Mac’s Primary Mortgage Market Survey and the ten-year, constant maturity Treasury yield from the Board of Governors of the Federal Reserve System.

Exhibit 4: Regional Differences in Mortgage Rates Have Declined

Interest-rate differences are annual averages of end-of-quarter rates from Freddie Mac’s Primary Mortgage Market Survey. The regions are: Northeast (CT, DE, DC, ME, MD, MA, NH, NJ, NY, PA, PR, RI, VT, VI, VA, WV), Southeast (AL, FL, GA, KY, MS, NC, SC, TN), North Central (IL, IN, IA, MI, MN, ND, OH, SD, WI), Southwest (AR, CO, KS, LA, MO, NE, NM, OK, TX, WY) and West (AK, AZ, CA, GU, HI, ID, MT, NV, OR, UT, WA).

Exhibit 5: Freddie Mac Cushions Regional Housing Recessions

House-price growth is calculated from the Conventional Mortgage Home-Price Index. The purchase ratio is the percent of Freddie Mac’s purchases of conventional, newly originated single-family mortgages from a state divided by the percent of single-family originations in that state, as estimated in HUD’s Survey of Mortgage Lending Activity, multiplied by 100.

Exhibit 6: The Benefits of Freddie Mac’s Purchases Are Widely Distributed

Incomes are reported in 1994 dollars, with 1995 incomes adjusted by the annual change in the Consumer Price Index. Freddie Mac’s distributions are based on the number of single-family, owner-occupied, conventional, first mortgage loans purchased each year. The primary market consists of single-family, owner-occupied, conventional loans originated for home purchase or refinance with loan amounts no greater than the one-unit conforming limit, as reported in the Home Mortgage Disclosure Act (HMDA) public-use file for 1994. An adjustment is made to estimate the income distribution without mobile home loans. The mobile-home adjustment assumes 5 percent market share for mobile-home lending, with the income distribution for mobile-home borrowers calculated from the American Housing Survey for the United States in 1993. Census tract income is from the 1990 Census of Population and Housing.

Exhibit 7: Freddie Mac Purchases Loans with a Wide Range of Down Payments

Freddie Mac’s down-payment distribution is based on the number of single-family, owner-occupied, conventional, first mortgage loans purchased.

Exhibit 8: Highlights of Freddie Mac’s Innovations

Exhibit 9: Mortgage-Backed Securities Have Become a Major Investment Vehicle

Mortgage-backed securities include only residential mortgage securities. Treasury securities reflect total marketable interest-bearing public debt. Corporate debt reflects industrial revenue bonds, corporate bonds and commercial paper of nonfarm, nonfinancial companies. Data are from the Board of Governors of the Federal Reserve System.

Exhibit 10: Sources of Mortgage Funds Have Broadened

Secondary market entities consist of Freddie Mac, Fannie Mae, Ginnie Mae and dealers’ securities holdings. Thrifts include savings and loans, saving banks and credit unions. Other includes mortgage companies, finance companies, state and local credit agencies, individuals and other investors. The year-end 1980 and 1994 distributions are calculated from estimates of mortgage debt holdings prepared by the Board of Governors of the Federal Reserve System and Freddie Mac estimates of the investor profile in mortgage passthrough securities for 1980 and Inside Mortgage Securities estimates for 1994 (August 11, 1995).

Exhibit 11: Freddie Mac and Fannie Mae Foster Securitization

Single-family mortgage originations for 1992-94 for the United States are from HUD’s Survey of Mortgage Lending Activity, and for the other nations are from Michael J. Lea and Steven A. Bernstein, International Housing Finance Sourcebook 1995, The International Union of Housing Finance Institutions, September 1995. Securitization volume for the United States is from Secondary Mortgage Markets, Summer 1995, Table A-8 for Freddie Mac, Fannie Mae and Ginnie Mae (excluding mortgage-backed securities backed solely by multifamily mortgages), and from Mortgage Market Statistical Annual for 1996, Inside Mortgage Finance Publications, p. 176, for private-label issues. Securitization volume for the other nations are derived from data in "The MBS Goes Global," Mortgage Banking, May 1996, pp. 78-87, Figures 1 and 2.

Stockholders’ equity reflects common shares outstanding, as reported in Freddie Mac and Fannie Mae annual reports, various years. Mortgage portfolios include mortgages pooled and sold as mortgage-backed securities as well as mortgages and mortgage-backed securities held in portfolio.

Data reflect year-end 1995. Data for FHA, VA and FmHA insured or guaranteed single-family mortgage debt outstanding are from the Board of Governors of the Federal Reserve System. Data for state and local governments are estimated using data for state and local credit agencies from the Board of Governors of the Federal Reserve System and the conventional share of state and local government and housing finance agency holdings from the Residential Finance Survey of the 1990 Census of Housing. Data for federally insured depositories are estimated using data from the Board of Governors of the Federal Reserve System, Consolidated Report of Condition and Income, Thrift Financial Report and the Residential Finance Survey. Data for Federal Home Loan Banks were obtained from the Federal Home Loan Bank System 1995 Financial Report, May 10, 1996, and includes held-to-maturity and available-for-sale securities. Freddie Mac and Fannie Mae data are from the respective companies. Data for mortgage companies, individuals and others are from the Board of Governors of the Federal Reserve System.

Exhibit 14: Government Support Is Pervasive in the Primary Market

Data are from the HMDA public-use files for 1990-94.

Exhibit 15: The Mortgage Market is Dynamic and Competitive

Freddie Mac’s and Fannie Mae’s purchases only include mortgages originated less than one year before purchase; purchases by Fannie Mae of newly originated mortgages are estimated based on data from Fannie Mae’s quarterly investor/analyst report and from Bond Buyer, Monthly Factor Report. Private-label securities data are based upon data from the Mortgage Market Statistical Annual for 1996, p. 176. Conventional single-family originations are from HUD, Survey of Mortgage Lending Activity.

Yield spreads are quarterly averages of effective interest rates based on the first week of each month for 30-year, fixed-rate mortgages, with points converted to yield assuming each point equals 0.20 percentage points. Conforming rates are from Freddie Mac’s Primary Mortgage Market Survey and jumbo rates are from HSH Associates.

Exhibit 17: Jumbo Rates Are Less Uniform

Interest rates on conventional, 30-year, fixed-rate mortgages are from Freddie Mac’s Primary Mortgage Market Survey (February 16, 1996), with points converted to yield assuming each point equals 0.20 percentage points. Regions are identical to the four regions defined by the U.S. Census Bureau. Percentages reflect the coefficient of variation (standard deviation divided by mean) of jumbo loans divided by the coefficient of variation of conforming loans, less one and then multiplied by 100.

Exhibit 18: Regional Economic Declines Cause Jumbo Mortgage Rates to Rise in Local Markets

House-price growth in Massachusetts is calculated using the Conventional Mortgage Home-Price Index. The difference between the Boston and national average jumbo interest rate is calculated using a quarterly average of the jumbo 30-year, fixed-rate mortgage rate from the first week of each month for Boston and the United States from HSH Associates.

Exhibit 19: ARM Lending Is Much Greater in the Jumbo Market

The ARM percentage of conventional, purchase-money originations secured by one-unit homes is from the Federal Housing Finance Board, Mortgage Interest Rate Survey.

Exhibit 20: High Bid-Ask Spreads for Private-Label Issues Mean Low Liquidity

Bid-ask spreads were obtained from interviews with security traders. Security prices are typically quoted in 32nds of a percentage point.

Exhibit 21: Securitization of Jumbo Loans Lags Rest of Market

Single-family origination data for FHA, VA and conventional mortgages are from HUD, Survey of Mortgage Lending Activity; jumbo originations are assumed to equal 20 percent of the dollar volume of conventional originations. Private-label issuance is from the Mortgage Market Statistical Annual for 1996, p. 176. Unpublished tabulations from the 1991 Residential Finance Survey for recent originations show that 54 percent of private-label pools are loans below the conforming loan limit; thus, 46 percent of private-label issuance is taken to represent jumbo securitization. Conventional conforming securitization includes 54 percent of private-label issuance and all Freddie Mac and Fannie Mae securities containing single-family conventional loans, as reported in Secondary Mortgage Markets, Summer 1995, Table A-8, with updates to 1995. FHA, VA and FmHA securitization includes securities issued by Freddie Mac and Fannie Mae and loans placed in Ginnie Mae mortgage-backed securities.

Exhibit 22: Most Jumbo Loans Are Held by Depositories

Data are from the Residential Finance Survey for loans originated between 1989 and 1991. Holdings of Freddie Mac and Fannie Mae mortgage passthrough securities are allocated back to their respective investors using data from the Consolidated Report of Condition and Income, the Office of Thrift Supervision’s Thrift Financial Report and the American Council of Life Insurance’s Statistical Bulletin.

Exhibit 23: Jumbo Rates Are More Volatile than Conforming Rates

Interest rates on conventional, 30-year, fixed-rate home mortgages are from HSH Associates. The ten-year, constant-maturity Treasury yield is from the Board of Governors of the Federal Reserve System.

Originations of conventional, conforming, 30-year, fixed-rate mortgages secured by one-unit homes by ZIP code are calculated from the Federal Housing Finance Board’s Mortgage Interest Rate Survey for 1994. Median family income and minority share data by ZIP code are from the 1990 Census of Population and Housing. For metropolitan areas, relative income is the median income of families residing in the ZIP-code area divided by the median family income of the overall metropolitan area.

Exhibit 25: Estimated Decline in Homeownership Rates if Freddie Mac’s and Fannie Mae’s Charters Were Repealed

Estimated declines in homeownership are from Susan M. Wachter, James R. Follain, Peter Linneman, Roberto G. Quercia and George McCarthy, Implications of Privatization: The Attainment of Social Goals, report submitted to HUD, Tables 4 and 6.

Exhibit 26: Freddie Mac and Fannie Mae Securities Are Widely Held

At the end of 1995, Freddie Mac and Fannie Mae had $1.0 trillion in mortgage-backed securities and $0.2 trillion in long-term debt (debentures, notes and bonds due after one year) outstanding, as reported in Freddie Mac’s and Fannie Mae’s annual reports for 1995. The investor distribution for long-term debt reflects dealers’ estimates for Freddie Mac debt for 1995. The amount of Freddie Mac and Fannie Mae mortgage-backed securities held by depositories is from the Consolidated Report of Condition and Income and the Thrift Financial Report. The percentages of mortgage-backed securities held by life insurance companies and pension and mutual funds are from the Mortgage Market Statistical Annual for 1996, p. 359, and assumes that Freddie Mac’s and Fannie Mae’s percentages are the same as for overall securities outstanding.