Issue Bulletin #199
July 20, 1994
(Archived document, may contain errors)
July 20, 1994
HOUMNG MIS TACKLE NON-FMSTENT CME (HR 3M S. 2M9)
The House and Senate soon will consider legislation to reauthorize the programs of the U.S. Department of Housing and Urban Development (HUD). The House and Senate bills (H.R. 3838, S. 2049) have been marked up by committee and now will be debated in each chamber. This legislation, however, unfortunately is based on a persistent fiction: that there is a shortage of affordable housing and thus a need for more subsidies to home- buyers and renters. In reality, America's housing quality and quantity have been improving for decades, something not apparent from the continuing stream of pronouncements from congres- sional liberals, state and local officials, labor unions, and builders that the nation faces a housing crisis. This notion of a crisis has triggered demands for ever-larger federal out- lays and more stringent regulation. The appropriations bill passed last fall for fiscal year 1994, for instance, reserved $25 billion for HUD, an increase of 27 percent over the fis- cal 1989 outlay of $19.7 billion when President Bush entered office. A large portion of this money is subsidizing publicly or privately owned multifamily rental housing. Yet many public projects are in physical disrepair, plagued with corrup- tion and mismanaged by the public housing authorities who would get the money. More- over, many of the privately owned multifamily rental projects are prime candidates for foreclosure and repossession by HUD. Some are as decayed as any publicly built project.Some of the information in this Issue Bulletin draws upon material in Carl Horowitz, "Washington's Continuing Fiction: A National Housing Shortage:' Heritage Foundation Backgrounder No. 783, August 22, 1990.
1 The $25 billion in the appropriations bill is HUD's portionMe bill also set aside funds for the Department of Veterans Affairs and various independent federal agencies.
The House bill, H.R. 3838, introduced in February by Representative Henry Gonzalez (D-TX), Chairman of the House Banking Committee, reauthorizes HUD spending at $31.2 billion in fiscal 1995 and $33.6 billion in fiscal 1996. The Senate version would open the federal purse even more, providing $28. 1 billion for fiscal 1995, but jumping to $39.1 billion for fiscal 1996. The impetus behind the House and Senate bills, which contain many features sup- ported by Secretary Cisneros and other top Clinton Administration officials, is the notion that housing was neglected during the Reagan and Bush years and that this neglect was manifested in massive cutbacks in the HUD budget. Cisneros said of his proposed legisla- tion that it would put HUD "back in the business as a force for positive change in Amer- ica."2 Yet this will not be accomplished by increasing HUD spending. HUD outlays dur- ing the Reagan years rose from $14.9 billion in FY 1981 to $19.7 billion in FY 1989 and reached $25.2 billion in FY 1993. According to this year's Office of Management and Budget (OMB) projections, they will have risen to $25.5 billion for fiscal 1994 and to more than $28.4 billion in fiscal 1996. Those who would raise federal spending on housing-related agencies such as HUD, the Department of Veterans Affairs, and the Farmers Home Administration know-even if they do not admit-that spending has risen. That is why they regularly cite statistics pur- portedly showing that the private sector no longer produces sufficiently affordable hous- ing. As evidence, they refer to the slight decline during the first half of the 1980s in the overall homeownership ratio (which since has been reversed) and to misleadingly high housing cost-to-income ratios of low-income renters. To be sure, there are housing cost problems. Yet neither the 1990 Cranston-Gonzalez National Affordable Housing Act (P.L. 101-625) nor the current proposals address those problems. Each of these measures fads to recognize that high taxes and overregulation un- dermine efforts to improve the housing stock and its affordability. Each also fails to rec- ognize that politicizing the distribution of housing raises the likelihood that there is less of it to go around. To combat the idea that massive subsidies are needed for housing production and con- sumption, lawmakers need to bear in mind several realities about housing in America: I ) The supply of America's housing stock has been improving steadily over time in both quality and quantity. 2) The demand for additional housing during the 1990s will not be as great as it was during the 1970s and 1980s. 3) The cost of housing, in comparison to incomes, has been getting less burden- some over the last several years, especially with interest rates last year dip- ping to their lowest levels in some 25 years, and now holding steady at about 8.5 percent.
If HUD officials and Members of Congress really wish to address the nation's housing supply and affordability problems, they should promote greater market efficiency. The reauthorization legislation gives lawmakers the opportunity to make needed changes. To improve the reauthorization legislation, Congress should take several steps. Among them, lawmakers should: V Eliminate provisions that would raise loan limits on Federal Housing Admini- stration (FHA) single-family mortgages. V Eliminate the provision that would end the requirement on public housing ten- ants who obtain a job to continue paying 30 percent of their income in rent. V Add a provision eliminating the requirement on HUD to subsidize the rents, for at least 15 years, of low-income tenants in foreclosed FHA-insured muwrfa- mily rental projects. V Fund the HOPE program. V Give HUD the authority to pressure other federal agencies, and state and local agencies, to reduce excessive land and construction regulations that make hous- ing less affordable. America is not reeling from an affordable housing crisis and was not in the grip of such a crisis during the Reagan and Bush years. Housing conditions and affordability rarely have been better. What HUD and other federal agencies should be doing is enabling hous- ing markets, especially in inner-city neighborhoods, to function more as markets. At pre- sent, they function too much as bureaucratic fiefdoms, a problem exacerbated by high lev- els of crime. While the House and Senate packages contain potentially beneficial ele- ments and even mention the need for crime control, too much in them represents business as usual under the guise of a "reinvention" of HUD. They would boost HUD spending by about a third without necessarily addressing the real problems.
THE SUPPLY OF HOUSING IN AMERICA Plenty of Homes. During the past two decades, America has expanded an already plentiful housing stock. According to the most recent American Housing Survey (199 1), there were 104.6 million conventional (non-mobile) dwellings in the United States, more than 50 percent greater than the 68.7 million units existing in 1970. 3Thus, over one-third of all dwellings available in 1991 were built in or after 1970. 4 This ratio, of course, would be higher if one took into account net housing production since 199 1.
Fewer Deficiencies, More Amenities.
Just as the quantity of housing units is impor- tant, so is the quality. That, too, has risen with affluence and with advances in homebuild- ing technology. Early in this century, a three-room cold-water flat was regarded as ade- quate housing for young urban immigrant families. Today it is considered unacceptable for anyone. Using American Housing Survey in- Table I dicators of housing quality, Table I in- Selected Interior Housing Deficiencies: 1991 dicates how far the U.S. has come in Characteristic % af Owner- % Of eliminating substandard housing. Occupied Renter Only a small portion of dwellings How In Roors 0.8% 2.0% now experience interior problems Open cracks or Hobo 3.4 8.1 such as holes in floors, broken plaster Broken Plaster or Peeling Point 3.0 6.1 or peeling paint, and exposed wiring. No Ellectriol Wiring 0.0 0.0 Moreover, those problems occur less Eqxmd Wiring 1.1 2.5 often in owner-occupied units, which Room without Ellectriclool Outlets 1.4 2.4 account for almost two-thirds of all oc- Lacking All of Sam Plumbing 2.2 2.9 cupied dwellings. SOUM. U.S. Eknau d ft Comm, OA=ftn HawkV StwMm Meanwhile, more new housing than ever before is being built with things once consid- ered luxuries, as Table 2 indicates. To some degree, therefore, rising standards of quality are responsible for rising home prices. The number of newly completed private, single- family homes containing central air conditioning, one or more fireplaces, and 2.5 or more bathrooms each rose during 1970-1992 by anywhere from roughly 80 percent to 200 per- cent. Median interior area rose from 1,385 to 1,920 square feet, an increase of 38.6 per- cent.
Homebuilders provide such amenities largely because buyers insist on them. A 1989 survey by the National Association of Home Builders asked: "If you cannot afford to buy the type of home you want, what would you give up to make it more affordable?" Some 36 percent of respondents said they would choose a house with unfinished rooms, and 35 percent would live farther from work or shopping. Only 18 percent would have accepted .. 5 a smaller house, and only 11 percent would have accepted fewer amemues.
THE DEMAND FOR HOUSING IN AMERICA
Population Growth. During America's baby boom, which spanned the years 1946- 1964, there were 20 to 25 live births per 1,000 population. Thereafter the birth rate de- . 6 - clined, so much so that in 1975 the rate was 14.6 births per 1,000 population, rising again to the 16-to- 17-per- 1,000 range in the late 1980s and early 1990s. The aging of the baby boom population has major implications for the demand for housing.
According to Census Bureau Table 2 mid-range projections, the Characteristics of Now Pfivately Owned population will increase from a One-Family Homes Completed: 1970-1992 little under 250 million to al- most 275 million during 1990- Medlan Percent Percent w/ Percent w/ yew Square Feet WCAC One or More 2.5 or More 20M, or by some 2.5 million Fireplaces Bathrooms annually. Households in the 35- 197o 1,385 34% 35% 16% 54 age bracket-mainly the 1980 1,595 63 56 25 baby boomers-will make up 1984 1,605 71 59 28 more than 75 percent of this 19" 1,660 69 63 33 19" 11810 75 65 42 population increase, rising by Im 1190S 76 66 45 more than 19 million. By con- 1992 1,920 77 64 47 trast, the number of Americans age 18-34 (the people who am Total Change +38A% +126.5% +82.9% +193.8% starting households for the first sown: u.s. Bwm of ve cman, ca@ Row seem cn mw m,,*. time, usually as renters) will decline from about 70 million to 63.5 million during this period. As a result, the age bracket most heavily in financial need win be smaller, numerically as well as proportion- ately. Household Growth. This pattern of population growth and age distribution has sig- nificant implications for housing demand, in particular the demand for new housing con- struction in the 1990s. According to the Census Bureau, the total number of households rose by about 1. 18 million annually during 1980-19927 and will rise by roughly the same rate by the year 2000. However, as with population growth, the age distribution of this shift reveals much. Households headed by persons age 35-54 will account for virtually all 8 of this increase. Yet households headed by persons 34 and under actually will decrease by I million. Growth is occurring, therefore, among households headed by adults in their prime years.
CHANGING HOUSEHOLD COMPOSITION
The trend toward smaller proportions of married households had been established by 1980 and since then, even with a falling divorce rate, 9has continued. Table 3 shows that in 1980, 60.8 percent of all households consisted of married couples, fairly evenly di- vided between those with at least one child present in the household and those without. Yet, by 1992 the cumulative proportion of married households, with and without chil- dren, declined to 54.8 percent. Male or female householder families (no spouse present) increased from 12.9 percent to 15.4 percent of the total. 10 Especially marked has been the rise of one-person households Table 3 from 22.7 percent in 1990 to 25.1 Composition of Amedcan Households: percent in 1992, caused in part by 1980 and 1992 the fact that the median age of first Household Type 1980 1 M marriage for men and women has INN rded with Children 30.9% 25.5% increased b TI some three years M:rded without Children 29.9 29.3 since 1970. 31noe-Porent Farnilly (FeMale Headed) 10.8 12.2 Single-Porent Family Nole-Headed) 2.1 3.2 Many lawmakers and orgaru a- Uvkv Ala" 22.7 25.1 tions calling for more housing sub- Other Non-Forrily Households 3.6 4.7 sidies misinterpret the growth in I Towl 100.0% 100.0% non-family households as an up- I bu= U.S. &nNGfftC==,QNWPVUWM RIP&*86fts M Nm WCmwof PW*ft. I= surge in the need for assistance. To be sure, non-married households have incomes well below those of married ones. The median household income of one-person households in 1991 was $15,44 1, while the median figure for female-headed families with children was $17,961. By contrast, the median household income for married households was nearly $41,075. 12 But that does not mean that non-married households are poorly housed or that they must forego food and other basic necessities to acquire good housing without a sub- sidy. The growth of non-family households has been due in large part to the affordability and availability of housing. That is, many of these households would not have formed in the first place were America's housing industry not so productive and adaptive to local market circumstances. In Western Europe, where housing construction levels are far lower than in the United States, marriage is one of the few ways in which a young adult can obtain housing separate from parents. And as a 1992 Heritage Foundation paper made clear, a -pooe' American actually has more housing space per person than the aver- age Western European. 13 It is not the growth of female-headed families that necessitates more welfare. Rather, it is the widened eligibility for welfare that is the major driving force behind the increase in such households in the first place. This includes housing assistance as well as cash pay- ments, Medicaid, and food stamps as its key components. The number of all households (mainly female-headed and/or elderly) receiving HUD housing subsidies alone grew from 3.3 million in fiscal 1981 to roughly 4.7 million in 1993. Subsidization of living expenses makes it more attractive for women with children not to marry and to maintain households separate from their fathers and from other members of their families. One study showed that a roughly $200 increase in monthly welfare benefits causes illegitimate teenage births to rise by 150 percent.
Status, and Living Arrangements: March 1992" (Washington, D.C.: U.S. Govemment Printing Office, 1993). 11 The median ages of first marriage were 22.5 and 20.6 for men and women, respectively, in 1970. They had risen to 25.5 and 23.7 in 1988. See Vital Statistics. 12 U.S. Bureau of the Census, Current Population Reports, P-60, No. 180. 13 See Robert Rector, "How the Poor Really Live: Lessons forWelfiwe Reform," Heritage Foundation Backgrounder No. 875, January 31, 1992. 14 Shelley Lundberg and Robert D. Plotnick, "Adolescent Premarital Childbearing: Do Opportunity Costs Matter?" June 1990, revised version of a paper presented at the Population Association of America Conference, Toronto,
At the opposite end of the age spectrum are the elderly, defined as households headed by persons age 65 or over. For over two decades a fast-growing segment of the popula- tion, 15 these households have relatively few housing problems. For example, 77.3 per- cent of senior-citizen households in 1991 owned their own homes, and 82.4 percent did so "free and clear," without any outstanding mortgage debt. The 1991 median monthly housing cost for all elderly households, owners and renters combined, was $257. More- over, only 2.6 percent of all elderly households lived in dwellings with less than com- plete plumbing facilities and with a median of 5.3 rooms per unit. 16 These are not house- holds requiring subsidies; yet advocates of more subsidies frequently call for more fed- eral aid to the elderly, especially under the Section 202 program, enacted by Congress in 1959 to finance the construction of privately owned rental housing projects for senior citi- zens.
THE MARKET FOR HOUSING
Combining these supply and demand characteristics into an overall market reveals how weak the case is for declaring a national housing emergency or even for enacting new housing assistance programs. By examining several basic indicators-vacancy rates, owner-occupancy ratios, costs, and cost-income ratios-it is evident that the supply of af- fordable housing is not in jeopardy.
INCREASING VACANCY RATES
One of the effects of the high levels of housing construction during the period from the 1960s through the 1980s has been relatively high vacancy rates. This is especially true in the case of housing for rent. In 1983, when many housing advocates declared a crisis, the rental vacancy rate was 5.7 percent. Yet each year since 1986 it has been at least 7 per- 17 cent, despite the decline in new apartment construction since the latter half of the 1980s. Those calling for more federally subsidized housing production as the principal way to eliminate homelessness ignore such realities. According to the American Housing Sur- vey, in 1991 there were over 8.7 million vacant year-round housing units. This represents about 15 vacant dwellings for every homeless person, based on the Urban Institute's original top estimate of about 600,000 homeless persons nationwide. 19 These units, more-over, are not cramped "tenements" or "slums." The 1991 American Housing Survey indi- cated they contained a median of 4.3 rooms per unit, and only some 5.5 percent lacked complete plumbing facilities. Almost 48 percent were single-family homes. 19 THE "DECLINING"' OWNER-OCCUPANcy RATio
A common myth about housing affordability is that young homeseekers no longer can afford to buy a home. Ever since 1977, when studies by the Congressional Budget Office and the Joint Center forUrban Studies of NUT and Harvard University asserted that dwin- Wing proportions of renter households could buy a home, advocates of more federal intru- sion into the housigg market have been sounding alarms about the disappearance of the American Dream.
2(r Typically, they use two statistics to bolster this claim. First they note that in 1980, 64.4 percent of all occupied units were owner-occupied, while in 1985 the figure fell to 63.5 percent. But even if this decline were cause for alarm, the ratio later reversed itself, rising to 64.2 percent in 1991. 21 Second, they often cite a finding from a 1988 report (the first in a continuing annual series on housing affordability by Harvard University's Joint Center for Housing Studies) showing that if the national homeownership rate had held steady since 1973, some two million additional households under age 35 would be homeowners. But what the re- searchers fail to emphasize is that more households today than in 1980 are singles and one-parent families (see Table 3). These households have a much higher propensity to rent than do married couples @2 so a slight decline in the overall ownership rate is not sur- prising. In fact, as HUD senior policy analyst Irving Welfeld has noted, the rates of own- ership among married and nonmarried households each rose after 1980, even though the overall rate declined slightly.
Thus, young adults are not being "locked out of' homebuying opportunities. On the contrary, Chicago Title and Trust's widely respected annual survey of recent homebuyers reveals first-time buyers to be an increasingly significant force. In 1989, for example, 40.2 percent of all homebuyers, were buying for the first time; in 1993 this ratio, thanks to small price increases and falling interest rates, reached 46 percent. 24
EXAGGERATED COST BURDENS FOR HOMEOWNERS
If changes in the homeownership ratio do not point to lessening affordability of homes for sale, neither do cost figures themselves, whether for the homes as such or for the mort- gage money used to finance their purchase. Those sounding the alarm bell often point to how much more owner-occupied homes with mortgages cost than do those without mort- gages. Yet the cost of borrowed money is distinct from the cost of housing itself. The American Housing Survey revealed that the median monthly housing cost for homes with a mortgage in 1991 was $761; for homes without a mortgage, it was $222. 25 To read into this a problem of affordability, however, would be erroneous. A great many of today's owners bought their homes during the late 1970s and 1980s when interest rates on 30-year fixed-rate conventional mortgages were in double digits. By contrast, the rates on 30-year fixed-rate mortgages currently are about 8.5 percent and, for a time late last summer and early fall, fell to the 6.7 percent to 7.0 percent range, according to the Federal Home Loan Mortgage Corporation ("Freddie Mac"), which, among other things, tracks mortgage rates on a weekly basis. Remarked Freddie Mac chief economist Robert Van Order last summer, "The last time interest rates fell below 7 percent was back in 1968."26 Thanks to failing home mortgage interest rates, there has been a surge in the popularity of 15-year and 20-year fixed-rate mortgages. 27 The Federal National Mortgage Associa- tion ("Fannie Mae') and Freddie Mac each reported that in 1993, close to half of all mort- gage refinancings consisted of 15-year or 20-year loans-the highest ratio ever. This trend has helped homes be more affordable for long-term ownership. Not only is the cu- mulative interest on a 15-year loan far lower than on a 30-year loan, but lenders charge interest rates about 0.5 percent lower as well. The homeowner stampede during 1991-1993 to refinance existing mortgages, though slower by now, is one of this decade's more remarkable housing stories. Fannie Mae data released last year indicated that 31 percent of current homeowners have refinanced their mortgages, with almost 20 percent having done so in 1992 or 1993 alone. Another 14 per- cent were considering refinancing. 28 Some 56 percent of the $1.1 trillion in mortgage loan originations in 1993 took the form of loan refinancings, according to Robert Rosenblatt, director of surveys for the Mortgage Bankers Association of America. 29 America's homeowners know a good bargain when they see one. The Harvard Joint Center for Housing Studies data actually provide strong evidence that homeownership is becoming more in reach, not less. Working with the American Housing Survey and other data, the Center's researchers estimated that the annual cost of owning a home (in constant 1989 dollars) fell from $8,895 to $7,077 during 1992-1992. This decline would have been even steeper had expected price appreciation levels not been flat in the last few years. 30 The past decade has been more a buyer's than a seller's market.
Data from Chicago Title and Trust's annual homebuyer survey provide further evi- 31 dence that rising costs, in and of themselves, mean little. During 1980-1993, repeat buy- ers saw the median home purchase price increase from $75,750 to $159,600, or by 110.7 percent. During the same period, however, their median incomes increased from $31,820 to $67,500, or by 112.1 percent. For first-time buyers, median prices increased from $61,450 to $121,100 (97.1 percent) while incomes increased from $27,430 to $53,400 (94.7 percent). Thus, incomes kept pace with prices. That incomes were not higher is due largely to the higher ratio of nonmarried households in the first-time and repeat homebuy- ing populations. Indeed, from 1991 to 1993, the proportion of first-time buyers who were single rose from 21.2 percent to 32.3 percent, according to the survey. These households, as Census figures earlier revealed, have lower incomes than married households. The survey also revealed that although the average mortgage payment as a percentage of income during 1980-1993 rose slightly from 30.3 percent to 30.7 percent for repeat buyers, it fell substantially from 35.6 percent to 32.2 percent for first-time buyers. The argument that a homebuying affordability crisis exists is refuted most thoroughly by examining figures from the National Association of Realtors' monthly publication, Home Sales, which tracks the affordability of existing homes throughout the nation. In 198 1, when President Reagan took office, an existing home cost $68,900; in February 1993 it cost $103,600, an increase of 50.4 percent. Yet price rises are not, in fact, cause for alarm in the context of income growth. NAR has a "Housing Affordability Index," which relates home price to buyer income, given a certain down payment ratio. The Index indicates what proportion of the income needed to qualify for a conventional 30-year mortgage on a median-priced home with a 20 percent downpayment would be possessed by a median-income family. Any Index figure above 100 means that a median-income household has more than enough to qualify for a mortgage on a median-priced home.
In 1981 the Index for all buyers was 68.9. This meant that a median-income family that year had only 68.9 percent of the income necessary to qualify for the purchase of a me- dian-priced home. During the fourth quarter of 1993, the Index had reached a record-high 141.9. In 'other words, taking into account the cost of mortgage credit and growth in real income, a typical home in the U.S. has become more than twice as affordable in a little more than a dozen years.33 For first-time buyers (renters seeking to buy), the Index im- proved markedly from 49.9 to 92.3 from 1981 through the fourth quarter of 1993.
EXAGGERATED COST BURDENS FOR RENTERS
Renters are less well off than owners. That is why most are not owners in the first place. Yet if the cost of ownership is not preventing today's renters from becoming tomor- row's owners, neither is the cost of renting. Between 1980 and 199 1, the median monthly rent (including utilities), according to the American Housing Survey, increased from $243 to $462, or by 90.1 percent. Yet during thisperiod median renter income rose from $10,500 to $20,300, or by 93.3 percent. Housing activists typically employ a rule of thumb for rental affordability. If rent takes up 30 percent or more of a household's gross (pre-tax) income, the dwelling is "unaf- fordable." Superficially, Ta- Tabie 4 ble 4 shows why they have Monthly Rental Costs as a Ratio of Income: 1991 an effective case. According Cost/income All Renter Poor Renter to American Housing Sur- Raft Households Households vey data, some 45.9 percent of all renter households and LM t1lon 6% 22LOW 14com 5% to 9% LOX= 4M000 69.0 percent of below-pov- 10%1o14% Z7XOOO 114000 erty-line households pay at 15% to 19% 4,2MOM 219.0W least 30 percent of their in- 20% to 24% 42MOM 37&WO conics in housing costs. In- 25% to M% Oftwo 547,0M 30% to 34% ZW3,0W 467.0W deed, 2 1.0 percent of all 35% to 39% zwolow W5,0W renter households and 48.0 40% to 49% 2&9.000 71&OM percent of poor renter house- 50% to 59% 1,707,000 637,000 60% to 69% 1,071,000 W3,000 holds pay at least 50 percent 70% or 3.567.000 Z384000 of their incomes on housing Zero or Negalive Income 547.000 490,000 costs. No Cash lient z5MOM 9&000 Median Rcft 28% 55% Using such Census data, (excwhg w two iines) liberal research organiza- Scum 'Arnek= Houft WW tions like the Center on Note 7he %pros In this table are used by advoooles of a huge boost In federal rert subsidlim 7hey clolm that these Ures tow Mat 459 percal of oil rerftm ord 69.0 Budget and Policy Priorities percent of all Poor refto (ockxft Izero or negaft WcorrW and Im cash wr howeholds)M at least 30 percent of their hmm for "r dwWWV. kft the concki- conclude that rental af- don that mod of the poor cormot afford ftr hou*v arid exlst an the pmctke of homd- fordability is slipping from emm Is wroM. AmorV other reasom reported Wcomes of the poor do not account for enorrnow arrmft from bereffiVn4dnd nich as Medloold food d=pL rert subsidies. and even cash. InckidlM these payrnerils as heome would lower the wit burdem of the grasp of the poor.-4 below-poverty Wo households uWantidy
From this, they conclude Tabie,9 that only large increases in Average Annual Income and Expenditures of government support can pre- Households Among Persons 14 and Over 1991 vent the crisis from worsen- ing. QuIntiles Income Total Total Housing of Income Before Taxes Expenditures Expenditures According to the U.S. De- wd 2o p..t $5,981 $13,4M $ 4,9W partment of Labor's Con- Sscmd 20 penent 14,821 1&986 6.178 sumer Expenditure Survey, 1hkd 20 permt 26,073 26,144 7,975 which measures how much Fwrffi 20 pawl 4OM8 36,151 1Q6M the households of different N9W 20 permt 81,594 57,597 16,898 income levels spend on key consumer items (food, hous- Soumr U.S. DWaftWdLabm, anm&mdAminixi, @LSReW835,Dmff6er1M.) ing, etc.), 35 per-household I I expenditures in 1991 averaged $13,464 among the lowest-quintile-income households, with pre-tax incomes averaging $5,98 1. In other words, the typical "poor" household, ac- cording to the Labor Department's statistics, spends well over $2 for every $ 1.00 it re- ports as income. Table 5 shows expenditures of households in the bottom quintile as a percentage of their reported "total income." Showing housing costs as a proportion of reported income thus gives a grossly mislead- ing impression. Using such methods, one can "prove' that the lowest-income households barely eat since they have spent 82 percent of their "incomes" on housing. Similarly, one could prove that these households have no money to spend on housing and utilities since well over 100 percent of their incomes are spent on other items.36 Something obviously is wrong with the use of rent-to-income ratios as a basis for estimating rental affordabil- ity. Why do government surveys systematically underestimate "incomes" of low-income households? Because most spending is not counted as income. Of the over $300 billion spent in 1992 by federal, state, and local governments on welfare programs for low-in- come persons, most was not included as household income by the Census Bureau.37 Benefits the Census Bureau does not count as income include food stamps, Medicaid, housing subsidies, and various social services. The second major problem is that government income data fail to include the tens of billions of dollars earned by persons working "off the books" to avoid taxes and regula- tion. The Department of Labor has ppt the total value of unreported earnings at about 5 percent of Gross Domestic Product. 38 As long as the government grossly underreports the incomes of poor and near-poor households, claims that these households spend too high a share of their incomes on housing are meaningless.
THE GROWFH IN FEDERAL AssiSTANCE
Housing assistance has become an increasingly significant part of the welfare state. Ac- cording to an Urban Institute study, about half of all rental subsidies consists of cash wel- fare payments; the other half is federal housing assistance. 39 American Housing Survey data for 1989 (the most recent year available irr such tabulations) indicate that median gross rent (including utilities) for unsubsidized poor renters was $300; for subsidized 40 poor renters it was a little over $150 . The Harvard Joint Center for Housing Studies, building on American Housing Survey data, has estimated that almost 40 percent of all households with incomes below 50 percent of the area median receive HUD housing sub- sidies, up from just under 30 percent in 1989. 41 Such subsidization is gradually becoming a welfare entitlement, something likely to be- come more pronounced. For example, according to the General Accounting Office, the cost of renewing contracts between private landlords and public housing authorities un- der HUD's largest tenant assistance subsidy program, "Section 8," will grow from an esti- mated $6.3 billion in fiscal 1994 to $15.2 billion in fiscal 1998-a jump of almost 150 percent in just four years. 42 Additionally, HUD spent some $6 billion-a record high- in fiscal 1993 on development, modernization, and operating costs for public housing pro- jects. In 1992, a federal commission recommended $7.5 billion in new HUD spending over ten years on "distressed" public housing, which constitutes about 6 percent of all public housing units and, according to the commission, is "Unlivable. ,43 Yet the rental housing stock overall is increasingly affordable. For example, the American Housing Survey for 1987 revealed a median cost-income ratio for poor house- holds of 66 percent; in 199 1, as Table 4 indicates, this ratio had dropped to 55 percent. The portion of all renter households spending 30 percent or more of their incomes on rent in 1987 was 48.4 percent; in 1991 it was 45.9 percent.
HOUSING PROBLEMS AND HUD INITIATIVES
The House and Senate housing committees have prepared their versions of an ex- panded federal presence in the housing market for floor debate. With record-high levels of funding, sponsors of the bills say they want to tackle a crisis in the nation's communi- ties. Part of that crisis, they insist, stems from a large and growing lack of affordable housing. Yet the data demonstrate that neither homeowners nor renters, as a whole, face such a shortage. The congressional proposals assume far too easily that HUD must supplement or supplant the workings of the housing market. Some activities would put the taxpayer in a state of heightened financial liability for risky projects and loan practices. If Con- gress and HUD officials really want to promote affordable housing at minimum taxpayer expense, they should focus on eliminating barriers to market efficiency and on expanding opportunities for low-income residents to take greater control over their living environ- ment.
Clinton Administration officials also should use HUD and other relevant federal agen- cies as "bully pulpits" for eliminating unnecessary state and local regulations that inhibit housing construction, rehabilitation, and maintenance. When Jack Kemp was HUD Secre- tary, he appointed the Advisory Commission on Regulatory Barriers to Affordable Hous- ing to address these issues. The Commission's final report, released in July 199 1, re- vealed that numerous forms of housing and land use regulation were making housing less affordable without necessarily improving housing quality." It outlined over 30 practical recommendations for government at all levels, often working in conjunction with devel- opers, to reduce unnecessary regulation. Lawmakers and officials also should expand residential choice among those least able to exercise it: the poor. Secretary Cisneros, on the surface, recognizes this. Yet the sub- stance of what he and congressional housing advocates support would foster greater de- pendency by low-income households on taxpayer largesse and housing bureaucrats at all levels of government. Rather than pour additional funding into building, maintaining, and modernizing public and subsidized housing projects and restricting opportunities by the private sector to buy and manage such properties, the federal government should be bringing such properties back to the market as quickly as possible. This includes enabling low-income residents themselves to be managers and owners. Such a strategy has turned around Washington, D.C.'s Kendworth-Parkside, now resident-managed and owned, and other low-income projects. HOPE gives tenants of public and privately built, government- held (usually through foreclosure) projects a chance to bypass public agencies in using federal money. 45 About 500 organizations, representing some 45,000 people, partici- pated in the program at the beginning of this year. 46 Yet under Clinton, the program is all but terminated, with planning grants eliminated from the 1994 budget and implementa- tion grants almost dry. No new funding is contemplated for fiscal 1995. Government at all levels, however, can take a big step in turning around low-income urban areas by making crime control a top priority. Fear of crime has transformed many urban neighborhoods and housing projects into no-man's lands. 47 Nothing restricts resi- dential choice mom than crime. When an area is a zone of fear, rather than a real commu- nity, people will not want to live there, no matter what the quality of its housing.
WHAT THE BELLS WOULD DO
The reauthorization bills now before Congress unfortunately ignore the real state of the housing market in America. And although the measures do address some concerns, they include provisions that would add to the problems which do exist while overlooking ac- tions that would tackle these problems.
HELPFUL FEATURES OF THE BILLS
The House and Senate bills do contain some beneficial provisions that would enable HUD programs to be operated in a more efficient manner. To their credit, both bills: V Modify a provision in the 1987 Housing and Community Development Act that requires public housing authorities (PHAs) to replace, on a 1-to-1 dwelling ba- sis, any obsolete projects they demolish. This requirement has been tying the hands of PHAs. Authorities cannot raze these largely vacant properties, many of them public eyesores, since they lack funds or suitable alternative sites to build replacement units. The new provision grants PHAs waivers from that stipulation, and thus should result in the demolition of obsolete units and the better use of existing HUD subsidies. se Broaden HUD's Drug Elimination Program for public housing into a larger Community Partnership Against Crime. Aside from mismanagement by housing authorities, the number one problem in public housing today is crime. The new pro- gram, among other things, would allow a PHA to obtain a record of any prior criminal convictions of a tenant or applicant as a way to help keep projects fire of dangerous offenders. V Consolidate the several programs under the Stewart B. McKinney Homeless As- sistance Act of 1987 into a block grant program that would give states, locali- ties, and nonprofit groups more say over how to use existing funds. Because these grantees are closer to the problems of the homeless than is the federal govern- ment, they will be able to reach more of these people. In addition, the Senate bill rejects a proposal by HUD Secretary Henry Cisneros to al- low PHAs to replace obsolete projects by using spending authority to leverage borrowing against future HUD modernization funds. If such loans default, which is especially likely in larger, troubled authorities, taxpayers may end up paying a large tab. Even if the loans are paid on schedule, political pressures to build new projects on vacant sites ensure a high program cost. Congress should be more concerned with selling off existing projects than with building new ones.
TROUBLING FEATURES OF THE BILLS
The House and Senate bills, unfortunately, also contain proposals that should not be en- acted. The bills, for example, mistakenly would: Increase loan limits on Federal Housing Administration (FHA) single-family mortgages. Since its creation in 1934, FHA has been charged with insuring mortgage lenders against the risk of default in their underwriting. Periodically, to adjust to price rises in the housing market, Congress has raised the limits. But H.R. 3838 goes too far, raising the basic FHA ceiling from $67,500 to $101,575 and the high-cost-area ceiling from $151,725 to $172,675. The Senate bin, by contrast, wisely increases the basic limit only to $77,500 and leaves unchanged the high-cost area cap. Cisneros, ag- gressively urging passage of the House version, believes FHA will be able to create 2 million new homeowners over the long run. More likely, given IFHA's recent history and the fact that private mortgage insurers account for most mortgage insurance is- sued, the House version would help young households who eventually would be able to buy anyway, and heighten an already high risk of default. Eliminate the requirement that public housing residents who have or acquire a job, or who undergo job training, must pay 30 percent of their incomes toward rent While motivated by a desire to transform public housing into a more stable liv- ing environment, this measure would leave housing authorities with less rent-derived income with which to operate. The taxpayers would have to prop them up even more than is now the case.
PROVISIONS THE BILLS SHOULD INCLUDE
In addition, both bills are significant for some essential provisions they omit. The bills should: V Eliminate the requirement that HUD subsidize rents of low4ncome residents of foreclosed FHA-Insured multifamily rental projects for at least IS years. Some of these projects, especially in Texas, are eyesores. This provision has contributed to- ward HUD's slow property disposition process and this year alone is costing taxpay- ers some $450 million in "Preservation" rent subsidies. V Fund the Homeownership and Opportunity for People for Everywhere pro- gram, or HOPE, to enable lowincome residents to manage and ultimately to own public housing (HOPE 1) and both privately built, publicly held multifa- mily (HOPE 2), and privately built, publicly held single-family (HOPE 3) pro- jects. This program, enacted in 1990 at the strong urging of Cisneros's predecessor at HUD, Jack Kemp, is an innovative anti-poverty program that bypasses local housing bureaucracies to reach the poor. Cisneros, however, has chosen to cancel the program, and Congress has eliminated fiscal 1994 funding for HOPE planning grants, a necessary precondition for rehabilitation and property transfer. Without a vigorous reversal, the program appears dead. V Give HUD authority to pressure other federal agencies and state and local agencies, to reduce excessive regulatory requirements on land use and housing construction. A HUD-sponsored commission three years ago concluded that over- regulation has been imposing serious costs on building and rehabilitating housing and that HUD could play a key role in alleviating the problem. Yet neither Congress nor Secretary Cisneros sees a real need to follow through on such efforts.
Despite what has become conventional wisdom, America is not reeling from a nation- wide housing shortage. Today's mortgage interest rates alone render such an argument specious, if not absurd. Cost barriers do exist, especially in California and Northeast met- ropolitan markets; clearly, finding good, affordable housing in Boston or Los Angeles is more difficult than finding it in Cincinnati or Atlanta. But even the most expensive areas have become more affordable in the 1990s as prices and rents alike have exhibited slow rises. Super-luxury homes in Southern California have stood vacant for months, even years, something unthinkable for most of the 1980s. Unfortunately, the temptation continues for Congress to boost federal housing subsi- dies. The Cranston-Gonzalez legislation of 1990, for example, all but locked the federal government into subsidizing permanently the rents of some 360,000 multifamily rental dwellings whose project owners are eligible to prepay their 40-year mortgages after 20 years. HUD's mounting inventory of FHA-foreclosed privately built apartment projects, some of them vacant eyesores, is incurring needless costs of hundreds of millions of dol- lars annually in subsidized maintenance and debt service, thanks to excessively strict re- sale requirements (although Congress last year eased these requirements somewhat). 48 The Clinton Administration should acknowledge that real housing problems exist but that they have more to do with the absence of market incentives and the lack of work-de- rived income among the poor than with the supposed "neglect" of these problems by the federal government. The bills before Congress contain items of merit. Too much in them, however, merely pours old wine into new bottles. Even under the guise of "choice" and "investment," expanding subsidy programs would invite more, not fewer, problems at HUD. Prepared for The Heritage Foundation by Carl F. Horowitz, Ph.D.
The author, fonnerly housing and urban affairs policy analyst with The Heritage Foundation, currently is aWashington correspondent with the Los Angeles-based Investor's Business Daily.
2 Quoted in Julio Barreto, "Cisneros Unveils Housing Choice and Community Investment Act," Nation's Cities Weekly, April 25,1994.
3 U.S. Bureau of the Census, 1970 Census ofPopulation and Housing; Current Housing Reports, "American Housing Survey for the United States in 1991 " (Washington, D.C.: U.S. Government Printing Office, April 1993). The American Housing Survey (formerly known as the Annual Housing Survey until the early 1980s, previous to which it had been conducted annually, as its narne implied) is conducted nationally on a biennial basis by the Census Bureau, with technical assistance from HUD. Separate surveys of several dozen metropolitan areas are conducted on an annual rotating basis. All references in this Issue Bulletin to the American Housing Survey, unless otherwise indicated, are to the 1991 survey. In 1991 there were some 7 million mobile homes.
4 Losses of residential units include conversions of a structure from residential to nonresidential use, the combining of two or more separate residences into one, and damage from storms, earthquakes, and other natural phenomena.
5 What Home Buyers Want (Washington, D.C.: National Association of Home Builders, 1989).
6 U.S. National Center for Health Statistics, Vital Statistics of the United States, issued annually.
7 U.S. Bureau of the Census, Current Popuktrion Reports, Series P-20, No. 467, and earlier reports. Census of Population: 1980.
8 U.S. Bureau of the Census, Current Population Reports, Series P-25, No. 986, "Projections of the Number of Households and Families: 1986 to 2000" (Washington, D.C.: U.S. Government Printing Office, 1986).
9 Vital Statistics. The rate of divorces per 1,000 population declined during 1980-1990 from 5.2 to 4.7, or by some 10 percent.
10 U.S. Bureau of the Census, Current Population Reports, Series P-20, No. 467, "Households, Families, MariudMay 1990.
15 In 1970 households headed by a person age 65 or over constituted 19.7 percent of all households in America. In 1991 their share of all households increased to 21.7 percent. See U.S. Bureau of the Census, Current Population Reports, Series P-20, No. 458; Census ofPopulation: 1970.
16 "American Housing Survey," pp. 319, 320, 340, 349. In determining whether a household is poor, the Census Bureau looks at income-but not asset-levels. This is one reason why the elderly are often "house rich, cash poor." Their wealth is tied up in the earned equity of their homes. See Karen Marin Gibler and Joseph Rabianski, "Elderly Interest in Home Equity Conversion," Housing Policy Debate, Vol. 4, Issue 4 (1993), pp. 565-588.
17 U.S. Bureau of the Census, Current Housing Reports, Series H- I 11, issued quarterly and annually. Vacancy rates on units for sale are typically in the I percent to 2 percent range. Since the mid- 1980s this ratio has held steady at 1.7 percent.
18 Martha R. Burt and Barbara Cohen, America's Homeless. Numbers, Characteristics, and Programs 7hat Serve 7hem (Washington, D.C.: Urban institute Press, 1989). Burt since has revised her numbers downward to a rangeof 354,700 to 461,800. See Martha R. Burt, "Developing the Estimate of 500,000-600,000 Homeless People in the United States in 1987," in Cynthia M. Taeuber, ed., Enumerating Homeless Persons: Methods and Data Needs (Washington, D.C.: U.S. Bureau of the Census, 1991), pp. 130-138. The Census Bureau's special one-night survey of the homeless in March 1990 revealed only 230,000 persons on the streets and in shelters throughout the U.S., a figure the Bureau has revised upward slightly to 240,000. If anything, then, the ratio of vacant year-round dwellings to homeless persons is more in the neighborhood of 30-to- 1, or even higher.
19 "American Housing Survey," pp. 1, 4, 14.
20 Congressional Budget Office, Homeownership: The Changing Relationships of Costs and Incomes and Possible Federal Roles (Washington, D.C.: U.S. Government Printing Office, January 1977); Bernard J. Frieden, Arthur P. Solomon, et al., 7he Nation's Housing Needs (Cambridge, MA: Joint Center for Urban Studies of MIT and Harvard University, 1977).
21 "American Housing Survey," various years. 22 In the 25-29 and 30-34 age brackets, among married couples, 51 percent and 66.4 percent, respectively, were homeowners in 1990. For "other family" households, these rates were 19.1 percent and 28.5 percent. Among one-person households, these figures were
22.1 percent and 33.4 percent. See U.S. Bureau of the Census, Census of Population: 1990.
23 Irving Welfeld, Where We Live: The American Home and the Social, Political, and Economic Landscape, from Slums to Suburbs (New York: Simon & Schuster, 1988), p. 223.4 Who's Buying Homes in America ... The ChicagoTitle and Trust Family of Title insurers, I 8th Annual Survey of Recent Home Buyers (Chicago: Chicago Tide and Trust, 1994).
25 "American Housing Survey," 1991, p. 74.
26 "Mortgage Rates Fall Below 7 Percent," 7he Washington 77mes, August 27, 1993.
27 Albert B. Crenshaw, "ShorterTerms Draw Borrowers," The Washington Post, January 29, 1994; H. Jane Lehman, "Refinancers Shortened Terms in '93," The Washington Post, April 9, 1994.
28 Some 46 percent of the respondents said that reducing the length of their mortgage term would count most, while 40 percent indicated lower monthly payments were the most important factor
29 Cited in Marianne Kyriakos, "Mortgage Refinancings Set Record Pace in 1993," The Washington Post, January 8, 1994. In 1992, 48 percent of the $894 billion in mortgage originations represented refinancings.
30 7he State of the Nation's Housing 1993, p. 24.Me decline in annual ownership cost, minus the appreciation factor, was f3rom $10,617 to $7,405.
31 Who's Buying Homes in America, op. cit.
32 'Me NAR also calculates an Index for first-time buyers, which is somewhat different than for other buyers. Here, affordability is based on the median price of a "starter" home (85 percent of the price of a median-priced home), and first-time seekers are assumed to need only a 10 percent down payment.
33 Despite an increase in the interest rate used to compute the overall Index from 6.79 percent to 6.91 percent, the Index, according to preliminary NAR data, fell only to 140.9 in the first quarter of 1994 from the fourth quarter of 1993.
34 See, for example, Scott Barancik and Mark Sheft, A Place to Call Home: The Crisis in Housingfor the Poor(Washington, D.C.: Center on Budget and Policy Priorities, June 1991).
35 The Census Bureau numbers are the basis for calculating incomes in studies which show that poor households pay too much for rent.
36 Arguments to show that low-income households pay too much for housing generally are based on data from the American Housing Survey rather than the Consumer Expenditure Survey. But all government surveys have the same problem: they underreport incomes and thus show expenditures as very high relative to income.
37 See Robert Rector, "Combatting Family Disintegration, Crime, and Dependence: Welfare Reform and Beyond," Heritage Foundation Backgrounder No. 983, April 8, 1994.
38 See U.S. Department of Labor, 7he Underground Economy in the United States, Occasional Paper No. 2, September 1992, p. 24. For examples of this economy in action, see Jeffrey Tucker, "Notes from the Underground: America's Sprawling Informal Economy," Policy Review No. 65 (1993), pp. 76-79.
39 Sandra J. Newman and Ann B. Schnare, Subsidizing Shelter. The Relationship Between Weyhre and Housing Assistance (Washington, D.C.: Urban Institute Press, May 1988).
40 These figures are for privately owned dwellings only, and thus exclude units covered under the public housing program-
41 The State of the Nation's Housing 1993, p. 16; 77se State of the Nation's Housing 1992, p. 19.
42 U.S. General Accounting Office, AsAvied Housing: Evening Out the Growth of the Section 8 Program's Funding Needs, GAO/RCED-93-54, August 1993.
43 The Final Report of the National Commission on Severely Distressed Public Housing: A Report to the Congress and the Secretary of Housing and Urban Development (Washington, D.C.: U.S. Government Printing Office, August 1992).
44 "Not in My Back Yard". Removing Barriers to Affordable Housing (Washington, D.C.: Advisory Commission on Regulatory Barriers to Affordable Housing, July 1991). For additional evidence, see articles in Land Economics, Vol. 66, No. 3 (August 1990).
45 Under HOPE, public agencies are eligible grantees along with tenant and other nonprofit organizations. T'he public housing component of HOPE is lmown as "HOPE V; the privately built, publicly held components are called "HOPE 2" (multifamily housing) and "HOPE 3" (single-family housing).
46 "HOPE and the 'Reinvention' of Government," 7he Empowerment Pioneer, Vol. 2, No. I (February 1994), p. 6. This is a publication of the Alexandria, Virginia-based Empowerment Network Foundation.
47 See Wesley G. Skogan, Disorder and Decline. Crime and the Spiral of Decay in American Neighborhoods (New York: Free Press, 1990).
48 See Carl Horowitz, "Another Big Government Loan Bailout?" Investor's Business Daily, July 15, 1993; U.S. General Accounting Office, Multifamily Housing: Information on Selected Properties Owned by HUD, GAO/RCED-94-163F3, April 19%.