2. Building Blocks of Modern Housing Finance¹

For many people, homeownership is the “American Dream,” as well as an indicator of status, position, and individual identity. But for most people, the dream can’t come true without taking advantage of lending practices that have evolved in the U.S. since pre-Civil War days, with the birth of the savings and loan industry.

But first, how does America stack up against other nations in terms of home ownership? The answer to that question has varied over time.

In 1890, the U.S. homeownership rate was at 17.9%, compared to 6.7% for Europeans. By the middle of the twentieth century, that rate had risen above 61% in the U.S., but European countries were gaining as well. Their rates were 50% for Belgium, 33% for France, 13% for Germany, 26% for Sweden, and 43% for the United Kingdom.²

Figure 2.1 shows more recent data, with homeownership rates varying from a low of 38% in Switzerland to a high of 97% in Lithuania. Of the 46 countries in the figure, only Switzerland and Germany (43%) fell below 50%. This has been attributed to cultural factors, very low rents, and conservative mortgage lending.³ Italy, Greece, and Spain have much higher rates of homeownership, reflecting cultural values, discriminatory policies toward private rental housing, and weaker support of “social” rental housing (low-cost public housing owned and managed by government or nonprofit organizations).⁴ Fisher and Jaffe have found that, even though several partial factors are associated with high or low rates of homeownership, no single explanation can account for all global patterns. In their words, “any explanation of worldwide homeownership rates must be limited from a generalizable proposition to an anecdotal explanation with limited empirical content.”⁵

Note: Based on the latest available data from sources listed below. In countries such as Brazil, where there are favelas, it is not clear exactly how these are treated in terms of the homeownership rate that is provided. Also, it is not clear in another country such as South Africa what is included or excluded in the homeownership rate. The sources listed do not always provide sufficient detail to elaborate on these issues.

Sources: EMF Hypostat (2009) for E.U. countries and Iceland, Russia, Norway, and Turkey; Whitehead (2010) for Australia and Canada; Pollock (2010) for Japan, Israel, New Zealand, and Singapore; U.S. Census Bureau for United States; Euroconstruct (2008) for Switzerland; Gao (2010) for China; United Nations (2001) for Argentina, Brazil, South Korea, and South Africa; and Soula Proxenos (2002) for India.

Figure 2.1 Homeownership rates (%) around the world.

Figure 2.2 provides information on the ratio of home mortgage debt to gross domestic product (GDP), to accompany the homeownership rates just discussed. As you can see, Switzerland has the highest ratio, even though it has the lowest homeownership rate among the countries in the figure. This reflects a high cost of housing due to substantial increases in housing prices over the past decade and a sizable group of wealthy domestic and foreign-born (often transient) individuals who can afford more expensive homes. Germany has a mortgage debt–to–GDP ratio that is relatively low, reflecting its low rate of homeownership. Overall, the ratio for the 27 European Union countries was 52% in 2009, compared to a U.S. ratio of 67%.

Note: Based on the latest available data. EMF Hypostat (2009) provides the latest data as of 2009, and Warnock and Warnock (2008) for the average data from 2001 to 2005.

Sources: EMF Hypostat (2009) for E.U. countries and Iceland, Russia, Norway, Turkey, and United States; Warnock and Warnock (2008) for the other countries.

Figure 2.2 Home mortgage debt to GDP (%) in countries around the world.

According to Bardhan and Edelstein, “The large differences in the national mortgage markets reflect the fact that mortgage markets retain strong national characteristics...as a result of the differences in the historical, demographic, political and regulatory environments in which mortgage lenders operate.”⁶

In some countries, such as France, mortgage interest is not subsidized, yet the rental market is subsidized and heavily regulated. This policy contributes to lower homeownership rates, which, in turn, leads to lower mortgage debt–to–GDP ratios.⁷

More generally, as urban property values increase and more developed mortgage capital markets emerge in cities, a higher proportion of homes can be financed by mortgages in areas of rapid urbanization and industrialization. In this way, homes not only become secure shelter, but also provide potential income (through rentals and boarding) and serve as collateral for borrowing.

Modern Housing Finance Takes Shape in the United States⁸

The financial system in a modern economy facilitates the transfer of resources from savers to borrowers. This allows the productive sectors to invest in capital necessary for growth. The financial system also allows consumers to adjust to variations in income over time, to smooth consumption. Homeownership, of course, requires financing, given the price of housing relative to the typical homeowners’ income. The modern financing of housing largely began with the savings and loan industry in the United States.

Origin and Development of Savings and Loan Associations: The First Hundred Years (1830–1930)

The first American savings and loan (S&L) institution, the Oxford Provident Building Association, was founded in Frankford, Pennsylvania, on January 3, 1831. The association was organized to enable its member shareholders, most of whom were textile workers, to pool their savings so that a subset of them could obtain financing to build or buy homes. Every member was to be afforded the opportunity, over time, to borrow funds for this purpose, with the association terminating after the last member was accommodated. Association membership was geographically restricted: No loans were made for homes located more than 5 miles from Frankford.

This first savings and loan was organized as a mutual institution and therefore was owned by its shareholder members. New members were elected by ballot at shareholder meetings. Shareholders were expected to remain with the institution throughout its life, but those who wanted to withdraw their shares were allowed to do so if they gave a month’s notice and paid a penalty of 5% of their withdrawal. The association’s balance sheet consisted of mortgage loans as assets and ownership shares as liabilities, with relatively little net worth. These shares were the precursor of the savings deposits held today.⁹

The founders of the Frankford S&L envisioned an institution with a limited life in which the shareholders and the borrowers eventually became one and the same. The institution served the important role of consolidating the savings of a group of local individuals and rechanneling the funds to these same individuals in the form of mortgage loans.¹⁰ Thus, the Frankford association filled a niche that was woefully underserviced by financial institutions at the time. Only much later did borrowers and savers become separate and distinct customers of these institutions.

Originally, institutions of this type were called building societies because they bought land and built homes. When they began lending to members to build their own homes, they were referred to as building and loan associations. After the 1930s, they tended to be called savings and loan associations, following the name given to the newly created federally chartered associations. Until the 1980s, all new federal associations had to begin as mutuals. With the passage of the Garn–St. Germain Depository Institutions Act of 1982, however, the federal regulator of these institutions, the Federal Home Loan Bank Board (FHLBB), could provide federal charters to new stock associations.

Perhaps it is not surprising that this financial need went largely unfulfilled by commercial and mutual savings banks.¹¹ These financial entities preceded the development of S&Ls, but neither catered to the home real estate market. Commercial banks issued liabilities consisting primarily of currency and demand deposits that were acceptable to their customers because they were meant to be backed by self-liquidating commercial loans. These banks catered to the short-term business loan market.

Mutual savings banks did not issue currency or accept demand deposits but were involved with the savings of the general public. Unlike savings and loans, however, they were originally philanthropic. Their intent was to provide financial services for the small saver, which required that their deposits be more flexible in terms of amounts and maturities and correspondingly required a much more flexible asset portfolio than just mortgages. Each type of institution specialized in a particular market, and the specialization was reflected in the balance sheets of these financial firms.

After the organization of the Frankford S&L, similar institutions spread throughout the United States—for example, entering New York in 1836, South Carolina in 1843, and what is now Oklahoma in 1890. Accurate data on the number and assets of S&Ls before 1900 are not readily available. However, data for that year indicate that the total number of such institutions was 5,356, and total assets were $571 million. By 1930, before growth was interrupted by the Great Depression, the number of S&Ls had grown to nearly 12,000, and total assets were almost $9 billion.¹² S&L assets increased substantially as the number and income of shareholders grew.

As these associations spread throughout the country, innovations began to occur. For example, the self-terminating type of institution was replaced by a more permanent type, and the borrowers were separated from the savers. Thus, these firms began to operate with a long-term horizon in mind, and they began to accept shareholders who were not obliged to take out mortgage loans. This not only enlarged the pool of potential shareholders, but also emphasized the savings aspect of membership in an association. So the link between borrower and saver began to dissipate despite the mutual form of organization under which these firms usually operated. These institutions still generally did not take deposits, per se; in many states, in fact, they were precluded by law from doing so. Not until the advent of federal deposit insurance for savings and loans in the 1930s did the taking of deposits as such become widespread.¹³

Accompanying the growth of the savings and loan industry during its first century were state and eventually federal regulations. As the roles of saver and borrower became more distinct, and as the shareholders or owners became less directly involved in managing the associations, the public desire for monitoring and supervision grew. The state governments, which were the only chartering agents for the associations until the Great Depression, therefore became more involved with monitoring these early institutions.¹⁴

State supervision evolved from reports to officials, to permissive monitoring by regulators, and then to required periodic examinations.¹⁵ In this way, states were able to show “their disapproval of loans for purposes not strictly within the building and loan field, which [was determined to be] the financing of single-family residences.”¹⁶

In addition, “a rather common power given to the state official in charge of the supervision of building and loan associations [was] that of refusing to grant charters where there [did] not seem to be a necessity for another building and loan.”¹⁷ Thus, monitoring and supervision had three objectives: preventing and detecting fraud, limiting S&L lending to home loans, and curbing competition within the industry. This lenient regulation was fostered by the scarcity of reported failures of savings and loans in their formative years. However, failures increased during the economic downturn of the 1890s.

A major result of this downturn was the virtual elimination of the so-called national institutions. This type of institution developed in the 1880s by gathering deposits and making mortgage loans on a national scale through the use of branch offices and the mail. Although the downturn of the 1890s hit both nationally based and locally based institutions, local associations attributed their problems to the improper loan strategy and subsequent failures of the national institutions. They claimed that customers were unable to distinguish between the two types of firms during the panic of 1893. The increased competition engendered by the nationals also led to the establishment of the U.S. League of Local Building and Loan Associations. This group became very influential during the subsequent years and successfully lobbied the state legislatures to curb the activities of the national associations, a move that eventually drove the nationals out of business.¹⁸

Although the local savings and loans had effectively removed competition from the national associations, competition from commercial banks was beginning to develop. In the early 1900s, national banks were informed that they were not prohibited from accepting savings deposits. Moreover, Federal Reserve member banks were given an incentive to use this source of funds when a lower reserve requirement was placed on savings accounts than on demand deposits.

On the asset side, competition for residential mortgages was also beginning to develop between savings and loans and banks, albeit to a much lower degree. Without active secondary markets and with still somewhat restrictive regulations, the two types of depository institutions found that comparative advantages in information collection and processing, as well as the favorable tax treatment afforded savings and loans, still led to fairly identifiable balance sheet differences.

Thus, as the economic boom of the 1920s began, the banks and S&Ls maintained different balance sheets, competed only indirectly, and were regulated to a different degree and by different levels of government. The federal regulators were most interested in commercial banks and the payments mechanism, and the state governments were most directly involved with savings and loans and their role in facilitating homeownership.

It is interesting to note that, before World War II, noninstitutional sources were also major providers of home finance. Frederiksen (1984) reported that, in the late 1800s, about 55% of mortgages in the country were held by local investors who made the loans or sold the property themselves, and about 18% were held by nonresident investors. “[I]n America,” he wrote, “the making of a mortgage loan is essentially a local transaction.”

Frederiksen further noted that “it is entirely satisfactory only when the investor is personally familiar with the property mortgaged, and the insurance is kept up, and when, furthermore, he is able at any time to take steps to protect himself in case of default.”

Frederiksen’s study indicates that the mortgages averaged less than half of the value of the security and that less than half of the property in America was under mortgage.¹⁹ Interestingly, when local investors were replaced increasingly by more formal and more regulated sources after World War II, two major real estate crises occurred, with the most recent one more widespread and costly than the earlier crisis.

As you can see in Table 2.1, S&Ls were a major provider of funding until 1980. Afterward, commercial banks became more important than S&Ls. But in recent decades, government-sponsored enterprises (GSEs) have dominated the field.

Table 2.1 Nonfarm Residential Mortgage Holdings, by Type of Institution^a (1900–2010)

Sources: U.S. Federal Reserve Flow of Funds and Bureau of the Census, Statistical Abstract Supplement, Historical Statistics of the United States, 1961.

As shown in Figure 2.3, financing of homeownership differs substantially across countries. In the U.S., securitization has clearly become important, whereas in Denmark, covered bonds dominate. In other countries, such as Australia, Japan, Austria, Finland, France, Germany, and Greece, homeownership has been financed largely through the use of deposits at financial institutions.

Note: Based on the latest available data from the following sources.

Sources: Author’s calculation. U.S. Federal Reserve Flow of Funds (2009) for the United States; Hypostat (2009) for E.U. countries; Odaira and Takado (2008) and Japan’s Government Housing Loan Corporation; and Australian Bureau of Statistics.

Figure 2.3 Sources of funding for home mortgages in selected countries, 2009.

Savings and Loan Associations and the Great Depression

There have been only two periods in the first 150 years of savings and loan history in which the institutions suffered large-scale failures. The first was the severe economic downturn of the 1890s, and the second was the Great Depression of the 1930s.

During the Great Depression, savings and loans did not accept demand deposits and, therefore, did not suffer the runs that reportedly plagued commercial banks. Nevertheless, their members had to draw upon their savings to maintain consumption. S&Ls were hard pressed to cope with these withdrawals because their assets were almost entirely mortgages, and they prided themselves on maintaining low liquidity levels. Moreover, reserves for losses were relatively low because “many state laws...discouraged the accumulation of reserves and some supervisory authorities practically forced the distribution of all earnings.”²⁰ As withdrawals mounted and assets declined in value due to delinquencies and defaults, S&Ls failed. These failures severely limited the flow of funds to housing.²¹

This disruption in the housing market finally changed the role of the federal government in the regulation of the savings and loan industry. First, on July 22, 1932, President Hoover signed the Federal Home Loan Bank Act. This act set up the Federal Home Loan Bank System, consisting of 12 regional Federal Home Loan (FHL) banks under the supervision of the Federal Home Loan Bank Board (FHLBB) in Washington. The main purpose of the system was to financially strengthen member savings and loan associations by providing them with an alternative and steady source of funds for promoting homeownership.²² Member savings and loan associations included all federal associations and any state-chartered institutions that voluntarily chose to become and qualified to be members. The system was designed so that the FHL Banks could issue bonds in the capital markets and thus be able to provide advances to healthy and reasonably safe institutions. It was not intended to bail out failing thrifts.²³

Second, the Home Owners’ Loan Act was signed June 13, 1933. Although the main purpose of the act was to facilitate the refinancing of existing mortgages in distress cases, many borrowers seeking the more favorable interest rate and other terms the government offered were also able to obtain loans. This led many borrowers to deliberately default on their existing loans, thus exacerbating the problems of the S&Ls.²⁴ (Amid the mortgage market meltdown from 2007 to 2010, some Americans may also have had an incentive to default on loans, due to favorable government support or an inability of private lenders to seek adequate legal recourse against defaulters.)²⁵

Another purpose of the 1933 act was to allow the Federal Home Loan Bank Board to charter federal savings and loans. This was not an attempt to drive the state-chartered institutions out of business. Instead, the aim was to establish S&Ls in places where the state institutions were providing insufficient service. Furthermore, the federal S&Ls were to be mutual-type associations and to operate only in the local areas where they were chartered.²⁶

Finally, Title IV of the National Housing Act, enacted June 27, 1934, created the Federal Savings and Loan Insurance Corporation (FSLIC) to provide deposit insurance for savings deposits at savings and loans. Membership in the FSLIC was made compulsory for federal associations and optional for state-chartered associations. With the establishment of the FSLIC, the savings and loans were placed on equal footing with commercial banks, which were insured by the Federal Deposit Insurance Corporation (FDIC). Eventually, the FDIC became the administrator of the federal deposit insurance fund for savings and loans as well.

Postwar Growth and Diversification in the Savings and Loan Industry

Following the Great Depression and World War II, savings and loans experienced tremendous growth for close to four decades. They surpassed mutual savings banks in terms of total assets for the first time in 1954 and grew to half the size of the commercial banking industry by the end of 1980. This expansion was spread throughout the entire industry, with large and small institutions participating.

The magnitude of the redistribution is remarkable. Private financial assets in 1945 totaled $247 billion. Of this amount, savings and loan associations held a meager 3%, compared to 65% for commercial banks. By 1975, however, savings and loans had increased their share of the total to 16%, while the share for commercial banks had dropped to 37%. Mutual savings banks and life insurance companies also lost considerable ground during this period.

Moreover, although the share of total financial assets accounted for by all the depository financial services firms declined to 51% from 76%, the share of S&Ls quintupled. The relative growth in S&Ls is even more impressive in light of the increased competition among financial institutions.

In the 1950s and 1960s, more than half of the deposits in commercial banks were noninterest-bearing checking accounts or demand deposits (which savings and loans did not offer). However, by the late 1970s, demand deposits made up only about one-fourth of commercial bank deposits. At savings and loans over this same period, time and passbook savings accounts remained about the same, 75% to 80% of total deposits, and S&Ls offered virtually no transactions accounts. The commercial banks were therefore beginning to compete more directly for the saver’s dollar, but savings and loans had not yet received unrestricted permission to compete for the commercial banks’ bread-and-butter accounts.

In addition to facing competition from other depository institutions, some nondepository financial institutions were being developed to help small savers combat the ravages inflicted on their portfolios by inflation and high interest rates. Laws and regulations generally prohibited savings and loans and other depository institutions from paying interest on checking deposits and limited the rates of interest and maturities on savings and time deposits. As a result, customers of these institutions increasingly began to go to the relatively unregulated nondepository financial firms in search of higher interest rates.

This interaction between market interest rates and regulatory constraints led to the development of money market mutual funds. These funds attracted customers by offering a new type of account. Not being subject to depository institution regulations, this account offered higher rates of interest than other financial depository service firms could pay, and it included limited check-writing privileges.

By the late 1970s, these money market funds, which had been introduced in 1972, had grown large enough to be a serious competitive threat to depository financial institutions. This threat spurred the development and regulatory approval of new market-based accounts offered by depository institutions, such as the money market certificate in 1978, whose rate was tied to the six-month Treasury bill rate.

Although the 1970s marked the beginning of what became extensive changes on the liability side of the depository institutions’ balance sheet, not until the 1980s did the asset side catch up with the changes.

Congress rejected variable-rate mortgages, which existed in the early 1970s in some states, such as Wisconsin and California, on a national basis in 1974. Although federally chartered savings and loans were allowed to issue variable-rate mortgages in states where state-chartered institutions were permitted to do so, not until January 1, 1979, were all federally chartered S&Ls allowed to offer variable-rate, graduated-payment, and reverse-annuity mortgages on a national basis.

Tax law in the savings and loan industry came into play in 1951. Before the Revenue Act of 1951, savings and loans were exempt from federal income taxes. Although this act terminated their tax-exempt status, savings and loans nonetheless were able to avoid paying taxes because they were permitted to essentially deduct up to 100% of taxable income through a bad-debt reserve.

In 1962, however, another Revenue Act was passed that reduced the bad-debt deduction to 60% of taxable income, subject to a qualifying asset restriction. This restriction stated that, for a savings and loan to be eligible for the maximum deduction, 82% or more of its assets had to consist of cash, U.S. government securities, and passbook loans, plus one-to-four family residential property loans. The deduction was zero if these assets fell below 60%. The Tax Reform Act of 1969 modified this restriction by permitting a savings and loan association to base its bad-debt deduction on taxable income, loss experience, or percentage of eligible loans.

Because the vast majority of associations used the taxable-income method, the deduction was reduced in scheduled steps from 60% of taxable income in 1969 to 40% in 1979.²⁷ The Tax Equity and Fiscal Responsibility Act of 1982 further reduced the bad-debt deduction to 34% in 1982 and then to 32% in 1984. The Tax Reform Act of 1986 reduced the bad-debt deduction as a percent of taxable income to 8% in 1987. Thus, over time, the tax laws have provided a large but diminishing incentive to invest in eligible mortgage-related assets.

In addition to the tax laws, regulations pertained to FHLBB member associations, FSLIC-insured associations, federal associations, and state associations. The Interest Rate Control Act of 1966, for example, gave the FHLBB the authority to set rate ceilings, which until then had been nonexistent, on the savings deposits of member associations. This ceiling was set initially at one-half of 1% but later was reduced to one-fourth of 1%, above the ceiling rate that commercial banks were permitted to pay on savings deposits. The ceiling represented an attempt to provide a competitive edge to savings and loans to garner funds for the residential housing sector. This differential was abolished in January 1984, and all rate ceilings for depository institutions were eliminated in March 1986.

The regulations for federal associations were initially quite direct in their intention to limit lending to local home mortgage loans, which meant loans secured by houses within 50 miles of the association’s home office.

In 1964, federal associations were permitted to make unsecured personal loans for college or educational expenses, the first time they had been allowed to make loans for any purpose other than acquiring real estate. In the same year, the geographical limit for mortgage loans was extended to 100 miles. Congress later extended this limit to encompass the association’s home state—and beyond that for the largest savings and loans. Then in 1983, the FHLBB permitted federal associations to make loans nationwide. Unless prohibited by state law, state associations with FSLIC insurance were permitted to do the same. In 100 years, the savings and loan industry had come full circle—nationals were once again alive and well.²⁸

Federal associations were also permitted in 1964 to issue mortgages and buy property in urban renewal areas and to buy securities issued by federal, state, and municipal governments. In 1968, these associations were allowed to make loans for mobile homes, second or vacation homes, and housing fixtures. Thus began the entry by savings and loans into business areas long viewed as the exclusive domain of commercial banks.

Turbulent 1980s for the Savings and Loan Industry

As interest rates rose unexpectedly and fluctuated widely in the late 1970s and early 1980s, it became clear that many savings and loans were ill equipped to handle the new financial environment. Their newly authorized market-rate deposits were rapidly escalating the institutions’ cost of funds, while the largely fixed-rate mortgage portfolios were painfully slow to turn over.

The result was rapidly deteriorating profits and a significant increase in failures. The problems persisted—even as interest rates declined in 1982 and the maturity-mismatch problem lessened—due to a growing deterioration in the quality of assets held by many associations.

The savings and loan industry’s ratio of net worth to total assets fell from more than 5% at the end of 1979 to 3.4% at the end of 1985. During this same period, more than 500 savings and loans failed and an additional 400 or so were left with negative net worth. By the end of the decade, approximately 500 more associations had failed and the government had bailed out the industry. A few years later, the insurance fund for savings and loans was merged into the insurance fund for commercial banks.

The turbulence of the early 1980s, however, did more than reduce the number of institutions. It permanently affected the way savings and loans were to do business. Instead of offering just savings and time deposits, these institutions began to offer transaction accounts, large certificates of deposit, and consumer repurchase agreements—virtually as wide a selection as that of any commercial bank. On the asset side, these institutions went beyond mortgages to hold consumer loans, commercial loans, mortgage-backed securities, and a wide variety of direct investments. From then on, savings and loans differed from commercial banks more as a matter of degree than of kind.²⁹ The distinctions among the depository financial services firms became forever blurred.

Federal Government Involvement in Mortgage Markets³⁰

The initial spread of home mortgages came after the National Banking Act in 1863, which enabled an expansion of bank charters and greater lending. Mortgage insurance helped smooth the way for credit that fueled homebuilding. The idea of a guarantee was later extended to municipal bonds, which furthered infrastructure financing that created the framework for further residential expansion. Mortgage insurance later expanded throughout the twentieth century (see Table 2.3 later in the chapter).

Since the 1930s, the federal government has played an increasingly important role in the allocation of mortgage credit. Instruments of federal policy used for this purpose include or have included loans insured and guaranteed by the Federal Housing Administration and Veterans Administration; secondary mortgage transactions by the Federal National Mortgage Association, Federal Home Loan Mortgage Corporation, and Government National Mortgage Association; interest rate subsidies; tax expenditures; and direct loans. Federal regulations have been enacted to affect the behavior of mortgage lenders in the pursuit of social objectives. These regulations include the Fair Housing Act (Title VIII), the Equal Credit Opportunity Act, the Home Mortgage Disclosure Act, and the Community Reinvestment Act.

Housing policies are clearly a part of the stabilization, allocation, and distribution activities of the federal government. The first major federal housing initiatives, enacted in the National Housing Act of 1934, as mentioned earlier, were part of an economic recovery program implemented during the Great Depression. Though stabilization of economic activity has remained an important objective, allocation and distribution objectives have become increasingly important. Indeed, the 1949 Housing Act explicitly acknowledged such aims, proposing a “goal of a decent home and a suitable living environment for every American family.”

An implicit but important goal of federal housing policies has been to encourage the acceptance of greater risk in mortgage markets. Encouraging greater risk taking may be socially desirable for reasons of economic efficiency and distributional equity. However, attitudes toward risk by private lenders and federal, state, and local financial regulatory agencies may prevent mortgage transactions that would be profitable for borrowers and lenders. In such cases, appropriately designed federal mortgage insurance programs may enhance the efficiency of mortgage markets. For numerous reasons, low-income applicants for mortgages are likely to be more risky. Improving the access of such individuals or groups to mortgage credit through government actions can be a means of achieving greater distributional equity.

Table 2.2 shows that the United States is one of relatively few countries among those listed in which the government provides support to residential mortgage markets.³¹ As a result of the recent mortgage market meltdown in the United States, the role of the government in mortgage markets has been reconsidered. However, as of the writing of this book, no legislative action had been taken to make any major changes.

Table 2.2 Government Mortgage Market Support

Source: Michael Lea, International Comparison of Mortgage Product Offerings, Special Report, Research Institute for Housing America, Mortgage Banker Association, September 2010.

It should also be noted that the Canada Mortgage and Housing Corporation, which is a 100% government-owned and -controlled corporation, insures (guarantees) mortgage loans and securitizes some of the insured loans. Pollock points out that CMHC is “in one sense...a combination of FHA and Ginnie Mae.” He adds that it insures roughly half of Canadian mortgages, which is the same proportion as the combined Fannie Mae and Freddie Mac in the United States.³² Homeownership rates are nearly identical in the two countries, as shown in earlier Figure 2.1.

Furthermore, despite all the U.S. government support for homeownership, Australia, Ireland, Spain, and the United Kingdom all have higher homeownership rates with far less government support.

Development and Impact of Government and Private Insurance for Home Mortgages³³

The first private mortgage insurance company was established in Rochester, New York, in 1887.³⁴ During the 1920s, the U.S. mortgage market relied heavily on mutual savings banks, savings and loan associations, insurance companies, and commercial banks. These four types of institutions accounted for 74.4% of the total new mortgage loans made on one- to four-family houses from 1925 to 1930.³⁵

The typical mortgage terms on loans these institutions made during this period were quite different from those prevailing in subsequent periods, including the present. During the 1920s, mortgages were written with the term to maturity not exceeding 12 years and with loan-to-value ratios close to 50%. In the 1930s and 1940s, however, these mortgage terms were significantly liberalized. By 1947, the term to maturity approached 20 years and the loan-to-value ratio was roughly 70%. In more recent years, both of these factors were further liberalized.

During the 1930s, the housing and banking industries virtually collapsed. In 1930–1933, more than 8,800 banks failed. In 1933 alone, 3,891 banks suspended operations.³⁶ Total housing starts fell 70%, from 2,383,000 in 1926–1930 to 728,000 in 1931–1935. It is estimated that only 150,000 persons were employed in on-site construction in 1933.³⁷

At the same time, approximately half of all home mortgages were in default, and foreclosures were occurring at the phenomenal rate of more than 1,000 per day.³⁸ Nonfarm real estate foreclosures reached a maximum of 252,000 in 1933. Foreclosures declined only slightly at first, to 229,000 in 1935, but then declined more rapidly to 185,000 in 1936, 151,000 in 1937, and 59,000 in 1941.³⁹

Among the responses of the federal government to these events were the establishment of the Home Owner’s Loan Corporation (HOLC) in 1933 and the passage of the National Housing Act of 1934, which created the FHA mortgage insurance programs.

HOLC was established to buy mortgages in default and threatened with foreclosure. It was therefore directly concerned with mortgage debt and only indirectly, if at all, with the availability of new mortgage credit. At its peak in 1935, HOLC held more than 15% of all U.S. residential mortgage debt. HOLC was expected to incur large losses as a result of its activities—primarily the extension of emergency loans on a long-term, self-amortizing basis. However, when liquidated in the 1940s, HOLC fully repaid all its Treasury borrowings and actually showed a small profit.

By contrast, the National Housing Act of 1934 was designed to increase the availability of new mortgage credit and thereby encourage the revival of the housing industry. The principal instrument was Section 203(b) of the National Housing Act. (Since the passage of this Act, new FHA mortgage programs have been implemented as amendments to this act and are commonly known by their section number and letter.)

Mortgages insured under Section 203(b) were secured by the Mutual Mortgage Insurance Fund (MMIF). The creation of the Federal National Mortgage Association (FNMA, also known as Fannie Mae) in 1938 provided additional impetus to 203(b) mortgage activity because FNMA was authorized to buy such mortgages. FNMA therefore made FHA mortgages extremely liquid by providing a ready secondary market for the longer-term type of mortgages offered under Section 203(b).

The main feature of Section 203(b) was the provision of mortgage insurance to all borrowers at a uniform premium. Each 203(b) loan was to be evaluated on the basis of economic soundness to ensure the solvency of the MMIF. Though no formal definition of economic soundness was provided in the legislation, limits were placed on the maximum mortgage amount and the maximum loan-to-value ratio. However, general consensus in the literature states that FHA implemented Section 203(b) mortgage insurance by imposing minimum values on neighborhood quality, property quality, and borrowers’ credit worthiness. These criteria were implemented by conducting a property inspection, and maximum permissible values were established for monthly payment-to-income ratios. Furthermore, the actual operation of the Section 203(b) program indicated that some urban areas were excluded from FHA insurance. This practice of exclusion was characterized as “redlining” because excluded areas were said to be marked by a red line at FHA offices.

High levels of mortgage insurance activity experienced under Section 203(b) during the 1930s, 1940s, and 1950s, along with sizable surpluses in the MMIF, indicate that, during this period, the bulk of FHA-insured mortgage loans were profitable. Indeed, as early as 1938, the maximum loan amount and the maximum loan-to-value ratio were increased based on favorable loss experience. Subsequently, these maximums were further increased. Government insurance transactions generally met or exceeded the criterion of actuarial or economic soundness. Note, however, that it is possible for the average transaction to earn a profit even though the marginal loan transaction, at the highest loan-to-value ratio, may suffer a loss.

The U.S. Congress has traditionally set the maximum mortgage amount that can be insured under FHA programs. In 2008, under the basic Section 203(b) single-family mortgage insurance program, the limit was $362,790. Thirty years earlier, in 1978, the limit was $60,000, which was in force since 1977. Before then, the cap was $45,000, a limit introduced by the Housing and Community Development Act of 1974. Until the passage of this act, the mortgage limit was $33,000.

If housing prices rise more rapidly than these congressionally determined mortgage limits, the maximum permissible loan-to-value ratios must necessarily fall. This reduction in the real value of mortgage limits in inflationary periods induces borrowers to shift to conventional mortgage loans.⁴⁰

The 1950s also saw the revival of the private mortgage insurance (PMI) industry, which began to offer insurance for conventional mortgage loans for the first time since the 1930s.⁴¹ The industry had collapsed during the Great Depression.⁴² As the result of legislation passed in 1956 in Wisconsin, the Mortgage Guarantee Insurance Corporation began operating in 1957. Subsequently, more PMIs were permitted to operate as additional states passed enabling legislation permitting this kind of insurance.

The PMIs became increasingly important in the mortgage insurance market thereafter. The standard mortgage insurance policy indemnifies the beneficiary against losses in the event of default, with the amount indemnified depending on the special coverage chosen.⁴³

An increase in the demand for conventional loans, of course, also results in an increase in the demand for PMI.⁴⁴ When the U.S. housing market collapsed in 2007 and 2008, mortgage insurers paid $15 billion in claims.⁴⁵

PMI competes with FHA insurance by offering lower premiums for safer mortgages. FHA generally sets an insurance premium of 0.5% of the outstanding mortgage amount, collected over the life of the loan on a current basis, on all its loans. In contrast, PMI premiums vary according to the loan-to-value ratio of the mortgage, the percentage of the mortgage amount insured, and the choice of prepayment option with fixed length of coverage.

Of course, these premiums are set lower than the FHA premium because the PMI companies insure only relatively low-risk mortgage loans, a practice known as “cream skimming.” FHA is left with the rest. However, the FHA demonstrated positive profits associated with higher loan-to-value mortgages that were not being exploited by conventional mortgage lenders. Increased PMI activity decreases the volume of Section 203(b) insurance activity and raises the loss rate, thereby reducing the surplus in the MMIF. Both of these outcomes have been observed.

This does not mean, however, that the FHA mortgage insurance program has necessarily been a failure. The FHA has made a number of important contributions, including assisting in popularizing and standardizing the fully amortized, fixed-interest, level-payment mortgage; lengthening the term of the mortgage; increasing the loan-to-value ratios on residential mortgages; developing minimum property standards, standardized appraisals, and the standardization of the mortgage contract; and providing information on risks of default that was then available to private mortgage lenders and insurers. All these factors, of course, have contributed to the development of a truly national mortgage market. Furthermore, the FHA thus far has never requested an appropriation for MMIF.⁴⁶

The accomplishments should not, however, be overstated. The surplus in the MMIF in earlier years may have demonstrated to private mortgage lenders that relatively high loan-to-value ratios were potentially profitable. It is uncertain, however, whether private lenders could have increased their loan-to-value ratios before the introduction of FHA mortgage insurance. In this regard, Leo Grebler states:

The restrictions on loan-to-value ratios and maturity of mortgage loans imposed by the National Banking Act do not apply to loans insured by the Federal Housing Administration or guaranteed by the Veterans Administration. This is also true for the limitations in many of the state banking laws. While similar exceptions apply to other mortgage lending institutions, they are more potent in the case of commercial banks because their conventional lending activity is more severely limited by existing laws.⁴⁷

This quotation suggests that private lenders making conventional mortgage loans may have been unable in earlier years to liberalize their loan terms because of federal, state, and/or local regulations and laws. Over time, as these regulations and laws were liberalized, lenders’ offer curves shifted upward to take advantage of opportunities flowing from the ability to lend on less restrictive terms. The recent housing problems in the United States indicate what happens when loan terms become too lenient.

The introduction of PMI encouraged lenders to make larger loans because, with insurance, conventional loans could be sold to such institutions as FNMA and FHLMC (discussed in the next section). Until these institutions were created or were permitted to buy conventional insured mortgages, however, lenders were undoubtedly reluctant to liberalize their loan terms, particularly when faced with usury laws or “soundness” requirements imposed by federal or state regulatory agencies.

Table 2.3 shows that mortgage insurance is available in many other countries. The insurance is offered by the private sector, as well as by government entities.

Table 2.3 Mortgage Insurance Timeline

Sources: Loîc Chiquier and Michael Lea, Housing Finance Policy in Emerging Market, 2009; Dwight Jaffee, Monoline Restrictions, with Applications to Mortgage Insurance and Title Insurance, 2006; and www.soa.org/library/monographs/finance/housing-wealth/2009/september/mono-2009-mfi09-herzog-history.pdf.

The Golden Age of Government-Sponsored Enterprises (GSEs)⁴⁸

The U.S. established several government-sponsored institutions to support the housing sector (see Table 2.4).

Table 2.4 Housing Government-Sponsored Enterprises

Source: Barth, et al., The Rise and Fall of the U.S. Mortgage and Credit Markets: A Comprehensive Analysis of the Meltdown (Hoboken: John Wiley & Sons, 2009).

First, as noted earlier, the Federal Home Loan Bank Act set up the Federal Home Loan Bank System in 1932, consisting of 12 regional Federal Home Loan Banks under the supervision of the FHLBB in Washington. Its main purpose was to financially strengthen member savings and loans by providing them with an alternative and steady source of funds to promote homeownership. It now provides such funding to all depository institutions.

Second, the FNMA was established in 1938 to buy home mortgages and thereby created a secondary market for such mortgages.

Third, the Federal Home Loan Mortgage Corporation (FHLMC, also known as Freddie Mac) was established in 1970 to increase the availability of residential mortgage credit by contributing to the development and maintenance of the secondary market for residential mortgages.

Because the FHLMC primarily buys conventional mortgages and then issues securities backed by those mortgages, a process known as securitization, its creation increased the liquidity of this type of mortgage loan. More important, however, is the FHLMC policy (required by its enabling legislation) of limiting its purchases of conventional mortgages to those in which the borrower has at least 20% equity in the property or in which a lower borrower equity is accompanied by private mortgage insurance so that the effective exposure risk is reduced to 80% of the loan amount. Clearly, this policy increased the demand for PMI. About the same time that the FHLMC was created, the Emergency Home Loan Financing Act of 1970 authorized the FNMA to buy conventional mortgages and also securitize home mortgages.

The Federal Housing Enterprises Financial Safety and Soundness Act of 1992 established goals for Fannie Mae and Freddie Mac for financing affordable housing and housing in inner cities and rural and other undeserved markets. In 1996, the affordable housing goal was increased from 40% to 42% of their financing to go to borrowers with low and moderate incomes for each year from 1997 through 2000. This goal was boosted to 50% for the years 2001–2004 and raised still higher, as shown in Table 2.5.

Table 2.5 Housing Goals Set for Fannie Mae and Freddie Mac

Sources: Federal Register, Milken Institute.

There are also goals for underserved areas (low-income and/or high-minority census tracts and rural counties) and special affordable housing (very low-income families and low-income families living in low-income areas.)

Figure 2.4a shows the ratio of total mortgages outstanding to GDP over the past century. Figure 2.4b shows the growing importance of financial institutions such as the FHLMC and FNMA in financing home ownership over the past three decades. (As a result of the mortgage market meltdown, however, their future remains uncertain as of the writing of this book.) Similar to the FHLMC, the FNMA can buy high loan-to-value ratio conventional loans only if these loans have private mortgage insurance. FNMA activities and policies therefore also increased the demand for PMI.⁴⁹ In the early 1970s, regulations were also promulgated permitting thrift institutions to originate mortgages at 95% of value when the individual loans are insured.

Sources: U.S. Federal Reserve Flow of Funds, Bureau of the Census, Statistical Abstract Supplement, Historical Statistics of the United States, 1961; and Bureau of Economic Analysis.

Figure 2.4a Total mortgages outstanding as percentage of GDP.

Sources: U.S. Federal Reserve Flow of Funds, Bureau of the Census, Statistical Abstract Supplement, Historical Statistics of the United States, 1961; and Bureau of Economic Analysis.

Figure 2.4b Total home mortgages outstanding and share of home financing provided by government-sponsored enterprises.

The securitization of residential mortgages has clearly spread beyond the United States during the past 30 years, as shown in Table 2.6. Other developments have also facilitated the financing of homeownership, such as covered bonds in Denmark and Pfandbrief in Germany. Clearly, however, the use of securitization and covered bonds to fund home purchases is found in more mature economies due to their more complex legal and financial issues.

Table 2.6 Mortgage-Backed Securitization Timeline

Source: Milken Institute, Capital Access Index, 2005.

Federal Policies to Address Special Housing Problems in Inner Cities⁵⁰

In the 1950s and 1960s, federal housing policy focused increasingly on inner cities. This represented a basic shift from the primary emphasis placed in the 1930s and 1940s on increasing the supply of adequate housing. This change was initially reflected in the Omnibus Housing Act of 1954, which, under Section 220, attempted to expand housing credit and production in urban renewal areas and, under Section 221(d)(2), provided mortgage insurance to families displaced by urban renewal. This legislation referred to the unmet goals of the National Housing Act of 1949 as the primary mandate for policy rather than revival of the housing industry or the financial system.

In addition, the Federal National Mortgage Charter Act of 1954 established the first special assistance functions (which require governmental financial support) to be carried out by the FNMA. These new programs were established with their own statutory provisions and insurance funds to permit them to function almost independently of the FHA. This was done to insulate the original FHA mortgage insurance fund, supporting programs such as Section 203 single-family home mortgage insurance from the effects of the relatively liberal underwriting terms of each new program.

Liberalization of mortgage terms relative to those in effect for the regular government mortgage insurance programs was a major feature of the 1954 Housing Act and the 1954 Mortgage Charter Act. This liberalization was basically achieved in three different ways. First, the “economic soundness” test for the proposed construction was replaced with an “acceptable risk” test, which required that the property to be insured meet “such standards and conditions as the Secretary shall prescribe to establish the acceptability of such property for insurance under this section.” Second, the maximum insurable mortgage loan was based on “replacement cost” rather than the more conservative estimate of long-range “value.” Third, the maximum allowable ratio of loan to “replacement cost” was increased. In some cases, the maximum term of the mortgage was also lengthened, thereby permitting lower monthly payments.

The trend toward liberalization continued with the enactment of Sections 231 and 202 in the Housing Act of 1959 and with the Housing Act of 1961. Section 231 in the 1959 act provided for generous insurance terms on housing for the elderly. In addition, Section 202 provided loans at subsidized interest rates to developers of private housing for the elderly.

The 1961 Housing Act further relaxed the terms of government mortgage insurance and broadened the coverage of Section 221(d)(2) insurance to include low- and moderate- income families. This section enabled such families to acquire housing at low down payments.

The 1961 Housing Act provided for subsidized, below-market interest rate (BMIR) mortgage insurance. The relevant section is 221(d)(3), which allowed loans at low interest rates to nonprofit or limited dividend corporations or cooperatives for the construction of modest housing for moderate-income households. Section 221(d)(3) is nominally structured as an interest subsidy program. However, lenders made such loans only because FNMA (later GNMA) immediately bought the BMIR mortgage at par, meaning FMNA (GNMA) was the actual lender. Hence, Section 221(d)(3) incorporated elements of a public interest subsidy and a direct loan program. Through the provisions of Section 221 (d)(3), the 1961 Housing Act reduced the reliance on mortgage insurance in government housing policy.

The one exception to this tendency was the previously described expansion of mortgage insurance under Section 221(d)(2). However, the mid-1960s saw a renewed emphasis on the role of federal mortgage guarantee programs in encouraging private lending in declining inner-city areas. This was undoubtedly due in part to the tendency of the FHA to follow conventional lenders in “treating loans in older urban areas cautiously—resulting in charges of redlining.”⁵¹ A congressional study in 1965, for example, argued that only a small fraction of FHA-insured mortgages were for existing homes in blighted central city areas.⁵²

In late 1965, FHA Commissioner Phillip Brownstein responded to these criticisms by issuing directives to FHA regional offices.⁵³ These directives urged that FHA activities in older inner cities not be confined to urban renewal areas and that Section 221(d)(2) mortgage insurance be used in nondeclared urban renewal areas and even in “neighborhoods in which blighting influences have started decay.” These directives helped define the notion of acceptable risk underlying Section 221(d)(2) insurance. However, the economic soundness criterion was still used to evaluate mortgage insurance provided under Section 203(b).

In 1966, Section 203(b) insurance was modified by Section 302 of the Demonstration Cities and Metropolitan Development Act, which added Section 203(1) to the National Housing Act of 1934. This new section applied the acceptable risk criterion to Section 203(b) insurance and specifically noted the need for providing mortgage insurance in inner-city areas, including those experiencing or threatened by riots and disorders. Moreover, in summer 1967, Commissioner Brownstein urged FHA regional offices not to designate entire areas as ineligible for insurance and not to limit FHA activity in the inner city to the Section 221(d)(2) program. FHA approvals in high-risk areas rose from 200 to 1,000 per week during 1967.⁵⁴

Section 103 of the Housing and Urban Development Act of 1968 repealed Section 203(1), replacing it with Section 223(e). This section allowed mortgage insurance to be extended under any FHA program in areas where economic soundness or related considerations would normally preclude eligibility.

The 1968 act designated troubled urban areas as worthy of special consideration, including waiver of statutory limitations concerning loan-to-value ratio, size of unit, or maximum mortgage amount. The aim of Section 223(e) was to provide insurance in “older, declining urban areas” where one or more eligibility requirement for mortgage insurance could not be met, provided that (1) the area was “reasonably viable” and (2) the property was “an acceptable risk.” The terms of the mortgages insured under this provision were to be designed with consideration for the needs of “families of low and moderate income in such areas.”

It was explicitly recognized that Section 203(b) and Section 221(d)(2) mortgage insurance issued pursuant to Section 223(e) would not be economically sound. Consequently, the 1968 act established a Special Risk Insurance Fund (SRIF) for which appropriations were authorized. This fund was to be used to fulfill insurance obligations under the subsidized programs and certain other mortgage insurance programs, including Section 223(e). As expected, these programs immediately incurred losses.

The effects of Section 223(e) quickly manifested. FHA activity in central cities rose quickly during 1969, with Section 223(e) activity leading the way in declining areas. A dramatic increase in foreclosures and losses followed. In response, the FHA tried to improve the administration of Section 223(e). The resulting administrative changes reduced default losses and the level of FHA activity. In 1976, fewer than 7,500 mortgages, or less than 14% of the program’s peak volume in 1969, were insured pursuant to Section 223(e). These efforts notwithstanding, the insurance position of the SRIF—that is, the excess of insurance reserves over estimated reserve requirements—was still –$394 million as of June 30, 1975.

FHA insurance activity became tantamount to a large-scale experiment in inner-city mortgage lending. Units that might never have qualified for private or government-insured financing in the past were approved for FHA mortgage insurance.

Default rates for mortgages insured under Sections 203 and 221 began rising after the passage of Section 223(e) in 1968. These rates peaked in 1972, reaching levels of 2.15% for Section 203(b) mortgages and 6.11% for Section 221 mortgages. Compared to pre-1968 default rates, these figures represent an increase in 203(b) and 221 defaults of roughly 50% for Section 203(b) mortgages and more than 100% for Section 221 loans.

Caution should be exercised in attributing these shifts to Section 223(e). However, the orders of magnitude observed for the 203(b) program are quite consistent with cross-section estimates of relative default probabilities obtained from FHA data. It is also noteworthy that default rates did decline after the administrative reforms implemented in response to the loss experience under 223(e).

These outcomes promoted considerable debate concerning the purpose of legislation that encouraged, if not forced, the FHA to expand its activities in the inner city. Was the mandate simply to stop large-scale redlining (refusal to lend in certain sections of cities)? Alternatively, was the mandate to assume the burden of providing high-risk credit in the hope of saving inner-city neighborhoods?

Whatever the intent, expansion of FHA insurance in the inner city can be viewed as an experiment in determining the influence of neighborhood characteristics on default. The pattern of making loans through federal insurance programs that private lenders viewed as “too risky” bears a superficial resemblance to the pattern observed in the initial Section 203(b) program during the 1930s. However, the two historical episodes have at least two major differences.

First, expanded FHA activity in risky neighborhoods resulted in the approval of many economically unsound mortgages. As discussed in the next section, some community groups said FHA-insured mortgages accelerated neighborhood decline by undermining incentives for property maintenance.⁵⁵ Private lenders did not likely view this as an experiment worthy of emulation.

Second, even if lenders wanted to follow the FHA example, neighborhood characteristics are more difficult to quantify than loan terms, which are easily reduced to a loan-to-value ratio or mortgage term. As a result, it is more difficult to translate neighborhood characteristics into measures of risk.

Government Regulation of Conventional Mortgage Lenders

The expansion of FHA mortgage insurance and housing subsidies during the 1960s and 1970s was perceived as less than completely successful in achieving the social objectives intended. Several explanations were offered. Racial discrimination was cited as one barrier to the efficient and equitable functioning of urban housing markets.

Paradoxically, the increase in FHA activity during the 1960s and 1970s was criticized as contributing to urban housing problems. This opinion is reflected in the following statement in a report by the Congressional Budget Office:⁵⁶

Prospective buyers or repairers of homes in redlined areas either are unable to finance their desired actions or must use FHA, VA, and Farmers’ Home insured or guaranteed financing. Unfortunately, federally underwritten mortgages in central city areas have been subject to abuse, often resulting in overpayment for poor quality housing and later abandonment. Lack of conventional financing thus produces substantial losses for potential buyers and sellers in affected areas and for their neighbors as neighborhood decline is hastened.

Concern about discrimination in housing markets was manifested in the Fair Housing Act (Title VIII) in 1968 and the Equal Credit Opportunity Act in 1974. Both regulations defined criteria that lenders may and may not use in their lending decisions.⁵⁷ In general, both acts prohibited lenders from denying or limiting credit solely on the basis of race, sex, creed, or national origin. In addition, the Fair Housing Act permitted lenders to take some neighborhood characteristics into account, but not others. Characteristics that are permissible include

• The condition or design of the property, or of nearby properties that clearly affect the value of that property

• The availability of neighborhood amenities or city services

• The need of the bank to hold a balanced real estate portfolio, with a reasonable distribution of loans in various neighborhoods, types of property, and loan amounts

However, lenders are enjoined from

• Denying or restricting mortgage credit in certain neighborhoods in the lender’s service area because of race, color, religion, or national origin of the residents

• Relying on appraisals that assign a lower value to a neighborhood because of a mix of races and national origins

• Equating a racially mixed neighborhood with a deteriorating neighborhood

• Incorporating the idea that deterioration of a neighborhood is inevitable

• Equating age of the property with the value of the property

• Prescreening loan applicants

The Home Mortgage Disclosure Act of 1975 and the Community Reinvestment Act of 1977 are aimed at increasing the volume of conventional loans in redlined areas. The HMDA requires lenders to disclose the location of their loans, though, interestingly enough, not deposits. The CRA represents an increased effort to prod lenders to expand mortgage lending in older and moderate-income areas in which they have offices.

Lenders deemed in violation of the Fair Housing Act are assumed, a priori, to violate performance standards of the Community Reinvestment Act. Consequently, the forms of lender behavior described previously are also proscribed under the CRA. However, the range of lender behavior subject to scrutiny is wider under the CRA than under the Fair Housing Act.

In particular, the CRA places emphasis on possible “errors of omission” that discourage potential borrowers from applying for loans. The Fair Housing Act singles out errors of commission in the form of prescreening. Prescreening is also viewed with suspicion under the Community Reinvestment Act. However, lenders are also judged on whether they make affirmative efforts to encourage applications for credit.

Specific assessment factors are the institution’s steps to evaluate the credit needs of its community, including efforts to communicate with the community regarding credit services provided by the institution; and the institution’s record of opening and closing offices and providing services at offices.

By implication, lenders that devote more resources to identifying community needs in some neighborhoods than in others, or that open (or close) offices in some neighborhoods but not in others, could violate the standards of the Community Reinvestment Act.

Summary

For millions of households around the globe, housing is one of the most important assets. Homeownership rates differ not only over time in individual countries, but also across countries at the same point in time. The tendency has been for homeownership to increase with urbanization and economic development. Important public- and private-sector innovations in home finance include the growth of specialized institutions, mortgage insurance, securitization of mortgages, and use of covered bonds. These types of innovation play different roles in different countries, with the result that mortgage debt–to–GDP ratios vary substantially. No single strategy will address all the world’s housing needs.

Endnotes

1 This chapter draws heavily on earlier articles co-authored by James R. Barth.

2 Michael R. Haines and Allen C. Goodman, “A Home of One’s Own: Aging and Homeownership in the United States in the Late Nineteenth and Early Twentieth Centuries,” NBER Working Paper, no. 21 (January 1991).

3 See The Economist, “Bricks and Slaughter: A Special Report on Property,” March 5, 2011.

4 See Michael Lea, “International Comparison of Mortgage Product Offerings.” Special Report, Research Institute for Housing America, Mortgage Banker Association, September 2010.

5 Lynn M. Fisher and Austin J. Jaffe, “Determinants of International Homeownership Rates,” Housing Finance International, September 2003.

6 Ashok Bardhan and Robert H. Edelstein, “Housing Finance in Emerging Economies: Applying a Benchmark from Developed Countries,” in ed. Danny Shahar, Seow Eng Ong, and Charles Leung, Mortgage Markets Worldwide (Oxford, U.K.: Blackwell Publishing Ltd., 2009). Also see Michael Lea, “Alternative Forms of Mortgage Finance: What Can We Learn From Other Countries?” paper prepared for Harvard Joint Center for Housing Studies National Symposium, Moving Forward: The Future of Consumer Credit and Mortgage Finance, Harvard Business School, 18 February 2010.

7 See Richard K. Green and Susan M. Wachter, “The American Mortgage in Historical and International Context,” Journal of Economic Perspectives 19, no. 4 (2005): 93–114.

8 Section 3 is drawn heavily from James R. Barth and Martin Regalia, “The Evolving Role of Regulation in the Savings and Loan Industry,” in ed. Catherine England and Thomas F. Huertas, The Financial Services Revolution: Policy Directions for the Future (Boston: Kluwer Academic Press, 1988).

9 The Housing and Urban Development Act of 1968 authorized federal savings and loan associations to offer savers deposit accounts rather than share accounts. The Federal Home Loan Bank Board adopted a regulation implementing this legislation, and all federal associations began offering deposit accounts the following year.

10 The first savings and loan associations were essentially finite-lived mutual funds, investing shareholders’ savings solely in residential mortgage loans. There were no maturity mismatch problems, and informational problems were minimized by direct saver involvement in every aspect of the business.

11 Credit unions did not come into existence until the early 1900s.

12 James R. Barth and M.A. Regalia (1988).

13 See Bodfish, Morton. History of Building and Loans in the United States, Chicago: U.S. Building and Loan League, 1931.

14 See Bodfish (1931) and Bodfish (1935). Note: Apparently, the state governments initially became involved as a cost-effective means for most associations to impose some discipline on all associations, to maintain the reputation of every association.

15 In 1875, New York was the first state to pass legislation requiring the filing of annual reports.

16 See Bodfish (1931).

17 Ibid.

18 See Bodfish and Theobald (1940).

19 D.M. Frederiksen, “Mortgage Banking in America,” Journal of Political Economy 2, no. 2 (March 1984): 203–234.

20 See Bodfish (1931).

21 According to Bodfish (1935), “One-half of the counties in the United States as a result of the Great Depression now had no mortgage loan institutions or facilities.”

22 The institutions initially eligible for membership were savings and loan associations, insurance companies, and mutual savings banks. The latter two types of institutions generally did not join the system.

23 See Bodfish and Theobald (1940).

24 See Bodfish (1935).

25 The federal government, for example, introduced the Home Affordable Modification Program in March 2009 to encourage lenders and services to modify the terms of the mortgage contract in a way that increases affordability for homeowners. The government introduced the Emergency Homeowner’s Loan Program in August 2010 to provide interest-free loans to borrowers to pay arrearages plus a portion of their monthly mortgage when the borrowers experienced a significant loss of income. (See James Orr, John Sporn, Joseph Tracy, and Junfeng Huang, “Help for Unemployed Borrowers: Lessons from the Pennsylvania Homeowners’ Emergency Mortgage Assistance Program,” Current Issues in Economics and Finance 17, no. 2 (2011).

26 The act specified that most loans had to be mortgage loans and had to be secured by houses within 50 miles of the association’s home office.

27 In 1975, 83% of all savings and loan associations used the taxable income method. In 1982, the corresponding figure was 65%.

28 Before March 1981, however, FHLBB policy prohibited interstate branching by federally chartered savings and loan associations. Since then, the policy has been modified to permit federal associations to branch on equal terms with state associations and to branch more freely as a result of the acquisition of supervisory or failing institutions. In 1994, the Riegle-Neal Interstate Banking and Branching Efficiency Act allowed for nationwide banking of depository institutions through acquisitions and mergers. Then in 2010, the Dodd-Frank Act allowed all depository institutions to engage in nationwide branching.

29 Indeed, savings and loan associations are now permitted to accept demand deposits insured by the FDIC and make commercial loans, which is a function that, according to the Bank Holding Company Act, legally defines a bank.

30 This section is drawn heavily from James R. Barth, Joseph Cordes, and Anthony Yezer, “Federal Government Attempts to Influence the Allocation of Mortgage Credit: FHA Mortgage Insurance and Government Regulations,” The Economics of Federal Credit Activity, U.S. Congressional Budget Office, Washington, DC, October 1980.

31 According to O. Emre Ergungor, “Homeowner Subsidies,” Economic Commentary, Federal Reserve Bank of Cleveland, February 23, 2011, the fiscal year 2010 budget indicates that “the U.S. government will spend $780 billion in tax expenditures over the next five years to subsidize housing through mortgage interest and property tax deduction.”

32 Alex Pollock, Testimony to Subcommittee on Security and International Trade and Finance: Committee on Banking, Housing, and Urban Affairs, United States Senate, September 29, 2010.

33 This section is drawn heavily from James R. Barth, Joseph Cordes and Anthony Yezer (1980).

34 Thomas N. Hertzog, History of Mortgage Finance with an Emphasis on Mortgage Insurance, American Society of Actuaries, 2009.

35 See Leo Grebler, David M. Blank, and Louis Winnick, Capital Formation in Residential Real Estate (Princeton, N.J.: University Press, 1956), Table 55.

36 Board of Governors of the Federal Reserve System, Banking Studies (Baltimore, Md.: Waverly Press, 1977).

37 U.S. Department of Housing and Urban Development, Future Role of FHA, 1977.

38 Ibid.

39 Leo Grebler, “The Role of Federal Credit Aids in Residential Construction,” Occasional Paper 39, National Bureau of Economic Research, 1953.

40 These mortgage limits, which apply nationwide, will also have regional effects due to different rates of growth in housing prices. One might therefore expect more conventional mortgage loans to be made in the East than in the West. This is consistent with the evidence.

41 The modern PMI industry was born in the 1950s but traces its origins to the late 1800s and the founding of title insurance companies in New York. That state passed the first legislation authorizing the insuring of mortgages in 1904 and then, in 1911, allowed title insurance companies to buy and resell mortgages—comparable to today’s secondary mortgage market. To make mortgages more marketable, companies offered guarantees of payment as well as title, thus establishing the business of mortgage insurance (see 2009–10 Fact Book, Mortgage Insurance Companies of America).

42 According to Dwight Jaffee, “Monoline Restrictions, with Applications to Mortgage Insurance and Title Insurance,” Review of Industrial Organization 28 (2006): 83–108, there were conflicts of interest within the mortgage insurance industry. He points out that the “largest conflict was that the mortgage insurers were also acquiring mortgages, then reselling them within insured pools (an early form of mortgage securitization). As mortgage default rates rose, the insurers fraudulently placed bad loans in insured pools.”

43 Ibid. Jaffee points out that the coverage amount is typically the first 20%–30% of the lost mortgage principal. He also notes that, in Australia, the standard policy covers 100% of the loan amount. Also see Thomas N. Herzog, “History of Mortgage Finance With an Emphasis on Mortgage Insurance,” Society of Actuaries, 2009.

44 It should be noted that, under temporary authority granted in 1974, the Government National Mortgage Association, which was chartered by federal law in its present form in 1968, also purchased conventional insured mortgages. However, GNMA’s general purchase authority is restricted to the purchase of mortgages insured or guaranteed by the federal government.

45 Ibid.

46 See Congressional Budget Office, “An Overview of Federal Support for Housing,” Economic and Budget Issue Brief, November 3, 2009.

47 Leo Grebler, The Role of Federal Credit Aids in Residential Construction (NBER, 1953).

48 This section is drawn heavily from James R. Barth, Joseph Cordes, and Anthony Yezer (1980); and James R. Barth with Tong Li, Wenling Lu, Tripon Phumiwasana, and Glenn Yago, The Rise and Fall of the U.S. Mortgage and Credit Markets: A Comprehensive Analysis of the Meltdown (Hoboken, New Jersey: John Wiley & Sons, 2009).

49 It should be noted that, under temporary authority granted in 1974, the Government National Mortgage Association, which was chartered by federal law in its present form in 1968, also bought conventional insured mortgages. However, GNMA’s general authority is restricted to the purchase of mortgages insured or guaranteed by the federal government.

50 This section is drawn heavily from James R. Barth, Joseph Cordes, and Anthony Yezer (1980).

51 Department of Housing and Urban Development, “Future Role of FHA,” 1977.

52 Defaults on FHA-Insured Home Mortgages, Detroit, Michigan, H. Rept. 1152, 92:2 (1965): 51.

53 “Real Estate Settlement Costs, FHA Mortgage, Foreclosures, Housing Abandonment, and Site Selection Policies,” House Committee on Banking and Currency, 1965.

54 Peter M. Greenstone, C. Duncan MacRae, and Carla I. Petrone, The Effects of FHA Activity in Older, Urban, Declining Areas; A Review of Existing, Related Analysis, Research Report,” (Washington, DC: The Urban Institute, February 1975).

55 For a discussion of this issue, see Kenneth F. Phillips and Michael B. Teitz, “Housing Conservation in Older Urban Areas; A Mortgage Insurance Approach” Berkeley Institute of Governmental Studies, University of California, 1978.

56 Congressional Budget Office, Housing Finance; Federal Programs and Issues, September 23, 1976.

57 For detailed discussion of the problem of identifying and detecting redlining, see James R. Barth, Joseph J. Cordes, and Anthony M.J. Yezer (1980).