July 14, 1998

Proposed House-Senate Compromise on Income Targeting Would
Cut Housing Assistance to Poor Families as Much as 63 Percent

by Jeff Lubell and Barbara Sard

This report has been updated to reflect current legislation.


In late June, legislators released the details of a proposed compromise among House and Senate Republicans on controversial housing legislation. The proposed House-Senate compromise would authorize the diversion of up to 245,000 housing subsidies per year from poor families to families at somewhat higher income levels, reducing by as much as 63 percent the number of poor families newly receiving a federal housing subsidy each year. In light of the severe shortage of affordable housing for poor families, including the working poor, and the availability of options to achieve desired policy objectives without such deep cuts in the housing assistance provided to these families, the proposed House-Senate compromise is ill-advised.

The potential for a sharp reduction in the amount of federal housing assistance provided to poor families stems from "income targeting" provisions in the pending legislation that authorize public housing authorities and private owners of subsidized housing developments to direct a greater share of subsidized housing resources to households at higher income levels. Should housing authorities and private owners take advantage of the authority provided by the new legislation, federal housing aid to the poor would be cut substantially.

Table 1 charts the potential impact of the income targeting provisions in the proposed House-Senate compromise (as well as in the bills the House and Senate have each passed) on households with incomes below 30 percent of the area median income (AMI). As explained in greater detail in the Technical Notes at the end of this paper, the impact figures in Table 1 and Figure 1 represent the difference between estimates of: (1) the current number of poor families newly receiving housing assistance each year, as indicated by recent HUD data; and (2) the number of poor families that would receive new subsidies each year under the proposed legislation if housing authorities and private owners of subsidized housing took full advantage of the authority provided in the pending legislation and admitted poor families only to the extent the law required. This analysis looks only at housing subsidies becoming available through normal turnover as some families receiving subsidies leave public or subsidized housing. (It does not reflect any additions to the subsidized housing stock that Congress could choose to make.)

Figure 1

wpe2B.jpg (19637 bytes)

As indicated in Figure 1 and Table 1, all three proposals could lead to a substantial reduction in the number of subsidies provided annually to poor families.(1) The relative impact of each of the proposals, however, varies significantly:

Table 1

 

Total

Public Housing

Certificates/Vouchers

Project-based Sec. 8

  Potential Yearly Subsidy Loss Percent Targeted to Poor Potential Yearly Subsidy Loss Percent Targeted to Poor Potential Yearly Subsidy Loss Percent Targeted to Poor Potential Yearly Subsidy Loss
House Bill 270,000 35% 67,000 40% 67,000 0% 136,000
Proposed

Compr.

245,000 35% 67,000 55% 42,000 0% 136,000
Senate Bill 148,000 40% 59,000 65% 26,000 40% 63,000
Another pending proposal, advanced by HUD, would reserve 75 percent of certificates and vouchers for families with incomes below 30 percent of AMI.

There is no sound policy justification for a reduction of this magnitude in housing assistance for poor families. Households with incomes below 30 percent of AMI are much more likely to have severe housing needs than households with higher incomes. HUD data show that while 70 percent of the families with incomes below 30 percent of AMI that did not receive housing assistance had severe housing needs in 1995 (defined as paying more than half of income for rent or living in severely substandard housing), only 14 percent of the unassisted families with incomes between 30 percent and 80 percent of AMI had severe needs. The HUD data indicate that nearly four million households with incomes below 30 percent of AMI had severe housing needs in 1995 and received no federal housing assistance.(2)

Supporters of these bills, particularly in the House, have promoted the new income targeting rules as necessary to lessen the concentration of very poor families in public housing projects. Yet public housing is only one of the three major subsidized housing programs serving poor families with children. The rationale driving the diversion of assistance from poor families in public housing does not apply at all to the certificate and voucher program - the largest of the three housing programs - and has less applicability to the project-based Section 8 program. Nevertheless, the pending legislation would authorize large reductions in assistance to poor families in all three programs.

The income targeting rules Congress sets should reflect the important differences among the three main housing programs serving poor families.

Public Housing. About 27 percent of families with children receiving federal housing subsidies live in public housing. Since many public housing units are in large developments with high concentrations of poverty, it makes sense to seek greater income diversity and a greater percentage of working families in such housing. Nevertheless, such diversity can be accomplished without the radical surgery advanced in the pending proposals. In recent years, approximately 78 percent of newly available public housing units have been provided to households with incomes below 30 percent of AMI. Both the House bill and the proposed House-Senate compromise would reserve only 35 percent of newly available public housing units for such families. The Senate bill would reserve 40 percent of newly available public housing units for such households. These proposals would authorize up to 59,000 - 67,000 public housing units per year to be redirected away from poor families.

As explained in more detail in earlier Center analyses,(3) a policy that targets half of public housing units becoming available to households with incomes at or below 30 percent of AMI, while granting preference among this group to working poor households, would strike an appropriate balance among competing social goals. By permitting up to half of new admissions to public housing to have incomes well above 30 percent of AMI, this policy could lead to a significant increase in the median income of public housing tenants, which currently stands at only 21 percent of AMI.(4) By granting preference for working households, the policy would further increase the proportion of working families in public housing and aid working poor families, the group of working families facing the most serious housing affordability problems. HUD data show that in 1995, there were more than 1.6 million working renter households with incomes below 30 percent of AMI that received no housing assistance and had severe housing needs.(5)

Certificates and Vouchers. Almost half of families with children receiving federal housing subsidies are in the "tenant-based" certificate and voucher programs, which subsidize the rents of families in private housing. Since recipients of certificates and vouchers are generally free to rent wherever they find a willing landlord, they are dispersed among the private rental stock rather than concentrated in particular developments. As a result, certificate and voucher holders are less likely to live in areas of high poverty concentration than residents of either public housing or project-based Section 8 housing.

Given the inherent dispersal effect of certificates and vouchers, there is no need to redirect them to higher-income families to achieve income diversity. HUD's proposal recognizes this fact by reserving 75 percent of newly available certificates and vouchers for households with incomes below 30 percent of AMI. By contrast, the House bill targets only 40 percent of newly available certificates and vouchers to households with incomes below 30 percent of AMI, and the proposed House-Senate compromise would reserve only 55 percent of newly available certificates and vouchers for such families. (The targeting figure for certificates and vouchers in the Senate bill is 65 percent.) In recent years, approximately 81 percent of newly available certificates and vouchers have been provided to families with incomes below 30 percent of AMI.

As shown in Table 1, the House bill would authorize up to 67,000 certificates and vouchers per year to be diverted from poor families. The proposed House-Senate compromise would authorize up to 42,000 certificates and vouchers per year to be redirected away from poor families. While more moderate in its impact, the Senate bill still could lead to a yearly loss of up to 26,000 certificates and vouchers for poor families.

Certificates and vouchers can be a critical tool in helping working poor families, including those making the transition from welfare to work, afford housing that is accessible to jobs and adequate transportation networks. In light of the value of certificates and vouchers to the working poor and the inapplicability of the "income diversity" to these subsidies, there is no sound reason to redirect certificates and vouchers away from poor families.

Intermediate Targeting of Certificate and Vouchers

The pending proposals also differ in the proportion of certificates and vouchers reserved for families with incomes below 50 percent or 60 percent of AMI. Currently, nearly all certificates and vouchers that become available are provided to families with incomes below 50 percent of AMI, now about $20,000 a year for a family of three (and $22,000 for a family of four) in the average area. The proposed House-Senate compromise, however, would reserve only 75 percent of certificates and vouchers for families with incomes below 60 percent of AMI, permitting the balance to be awarded to families with incomes up to 80 percent of AMI, which, on average, exceeds $30,000 a year for a family of three and $34,000 a year for a family of four. While preferable to the House bill, this feature of the proposed compromise falls significantly short of the Senate bill, which would target 90 percent of certificates and vouchers on families with incomes below 60 percent of AMI.

HUD data indicate that fewer than three percent of all renters with severe housing problems have incomes between 60 and 80 percent of AMI. It would thus be an unwise use of scarce resources to permit more than 10 percent of available subsidies to be directed to such families.

Project-Based Section 8. The remaining quarter of families with children receiving federal housing subsidies live in project-based Section 8 housing. These are private developments, or portions of developments, made affordable to low-income families through federal subsidies. Such developments tend to be smaller than public housing projects, with a better mix of incomes, and are more likely than public housing projects to be located outside high poverty areas. Accordingly, there is somewhat less need for these projects than for public housing projects to redirect subsidies away from poor families. On balance, a targeting percentage somewhere between that used for public housing and that applied to certificates and vouchers would be appropriate for project-based Section 8 housing. We would recommend that approximately 60 percent of project-based Section 8 subsidies be reserved for families below 30 percent of AMI.

Approximately 75 percent of households recently admitted to project-based Section 8 units have incomes below 30 percent of AMI. The Senate bill would reserve only 40 percent of newly available units for such families, authorizing the diversion of up to 63,000 project-based Section 8 subsidies per year from poor families. The House bill and proposed House-Senate compromise are substantially worse - targeting no subsidies to families with incomes below 30 percent of AMI.(6) These proposals could lead to the loss of up to 136,000 project-based Section 8 subsidies per year for poor families.

Fungibility

The House bill also contains a feature known as "fungibility," under which housing authorities could reduce the number of poor families admitted to public housing below the level otherwise required in exchange for a corresponding increase in the number of poor families given certificates and vouchers. This would permit some housing authorities to reduce to zero the number of poor families newly admitted to public housing. The House-Senate compromise provides for a more limited form of fungibility, but it still would permit some housing authorities to eliminate all admissions of poor families to public housing.

The Exclusion of Poor Families from Public Housing Through Fungibility

Under the proposed House-Senate compromise, the number of units treated as "fungible" is limited to 20 percent of the number of families that would newly be provided Section 8 certificates and vouchers. Where public housing authorities (PHAs) have more Section 8 subsidies becoming available than public housing units, however, this still could result in no admissions of poor applicants to public housing. For example, in Los Angeles, where the PHA administers about 28,000 certificates and vouchers but only about 9,000 public housing units, there are probably about 1,200 public housing units re-rented each year, but more than 3,000 certificates and vouchers. Under the proposed compromise, the maximum fungibility permitted - 600 (20 percent of 3,000) - would exceed the total number of public housing units that would otherwise have been reserved for poor applicants (420, or 35 percent of the public housing units becoming available).

Thus, if the PHA awarded 2,070 of the 3,000 newly available certificates and vouchers to families with incomes below 30 percent of AMI (the 55 percent of the 3,000 vouchers that the compromise would require plus 420 additional vouchers), the PHA would absolve itself of any obligation to admit even a single family below 30 percent of AMI to public housing. Moreover, in this example, the complete exclusion of poor families from newly available public housing units would be accomplished by issuing a smaller proportion of newly available certificates and vouchers to poor families - about 70 percent - than the current national average, which is about 81 percent (i.e., currently, approximately 81 percent of newly available certificates and vouchers go to families with incomes below 30 percent of AMI).

While certificates and vouchers are often more advantageous for poor families than public housing units, there are places where the reverse is true. Public housing may be the better option where it is well-situated near employment centers and there is a shortage of low- to mid-cost rental housing in which to use certificates and vouchers. It is precisely in such locations, however, that housing authorities are likely to invoke fungibility, since it is easier in such locations to attract higher income families to public housing developments. Another advantage of public housing in some instances is affordability; because of the way that rents are calculated in the two programs, public housing may be more affordable to some poor families than vouchers.

Neither the Senate bill nor HUD's proposal includes fungibility.(7)

• • •

The income targeting provisions in the proposed House-Senate compromise would likely lead to substantial reductions in housing assistance provided to poor families. This sharp reduction in assistance to poor families is not necessary to achieve legitimate policy goals. A more reasonable compromise that balances the need for housing assistance among poor families with the goals of increased income diversity and more employed families in public housing would reserve 50 percent of public housing units, 75 percent of certificates and vouchers, and 60 percent of project-based Section 8 units for families with incomes below 30 percent of the area median income.

 

Technical Notes to Table 1 and Figure 1

There are many ways in which to calculate the potential impact of the proposed income targeting rules. An analysis the Center issued in October 1997 compared the proposed targeting rules with the federal preference rules in effect until 1996. That analysis examined the impact of the proposals only on families with children. This paper takes a somewhat different approach, comparing the effect of the proposed targeting rules with the distribution of current admissions of all household types to the three major subsidized housing programs.

The impact figures in Table 1 and Figure 1 represent the difference between estimates of: (1) the current number of poor families newly receiving housing assistance each year, as indicated by recent HUD data; and (2) the number of poor families that would receive new subsidies each year under the proposed legislation if housing authorities and private owners of subsidized housing took full advantage of the authority provided in the pending legislation and admitted poor families only to the extent the law required. The analysis shows the maximum impact the proposed legislation could have in shifting subsidies from poor households to households at somewhat higher income levels. Since some housing authorities and private owners of project-based Section 8 housing may choose to provide more subsidies to families with incomes below 30 percent of AMI than the law would require, the actual number of subsidies lost under the new legislation is likely to be somewhat less than these maximums.

This analysis looks only at housing subsidies becoming available through normal turnover as some families receiving subsidies leave public or subsidized housing. It does not reflect any additions to the subsidized housing stock that Congress could choose to make.

The proportion of households newly admitted to subsidized housing having incomes below 30 percent of AMI was provided by HUD based on the most recent data available (the February 1997 MTCS and June 1997 TRACS databases). HUD also provided data on the total number of households in subsidized housing - 1.2 million in public housing, 1.5 million in the certificate and voucher programs, and 1.4 million in Section 8 project-based housing - and the yearly turnover percentages, which are about 13 percent in public housing and project-based Section 8 and 11 percent in the certificate and voucher programs.

1. Although the concepts of poverty and area median income (AMI) are technically distinct, this paper uses the phrase "poor" to mean households with incomes below 30 percent of AMI. Nationally, 30 percent of AMI, as adjusted for a family of three, is close to the poverty line.

2. U.S. Department of Housing and Urban Development, Office of Policy Development and Research, "Rental Housing Assistance - the Crisis Continues: The 1997 Report to Congress on Worst Case Housing Needs," April, 1998.

3. See Barbara Sard, Ed Lazere, Robert Greenstein and Jennifer Daskal, "Housing Bills Would Reduce Assistance Available to Poor Families," October 1997; and Jennifer Daskal and Barbara Sard, "Public Housing Admissions Preferences: Recent Changes in New York City Illustrate the Flexibility of Existing Rules," January 1998.

4. U.S. Department of Housing and Urban Development, Office of Policy Development and Research, "Characteristics of Households in Public and Assisted Housing," Recent Research Results, March 1998.

5. HUD, PD&R, April 1998.

6. In addition to omitting any targeting requirements for project-based Section 8 housing from the section of the compromise titled "Income Targeting," the proposed House-Senate compromise explicitly repeals the current income targeting provisions applicable to the project-based Section 8 program. It is unclear whether some targeting for project-based Section 8 housing will be restored in future negotiations.

7. If fungibility is to be included in the final legislation, Congress should adopt stronger targeting rules for public housing and the Section 8 certificate and voucher programs than those in the House bill or the proposed House-Senate compromise. This would be necessary to ensure that some public housing continues to be newly available to assist poor families in all jurisdictions. If fungibility is to be adopted, it also should be limited to a modest proportion of turnover units and applied only to large public housing developments in areas of high poverty concentration.


Related reports: