PHADA believes that the Financial Indicator as described in the PHAS final rule has serious flaws which will interfere with HUD's issuing a fair and accurate evaluation of PHAs. For one thing by counting all units -- including Section 8 and non-federal -- this assessment is essentially no longer an evaluation of the federal conventional low-income housing program. It is not a Public Housing Assessment System.
For another, by counting all units to form the peer groups, these groups are no longer comparable. Thirdly, in scoring the six components, some units, such as Section 8 units, will be included in some components and excluded in others -- making the peer groups even less valid. An additional concern is that the Department's use of a curve for scoring purposes guarantees that 50 percent of PHAs will score fewer than the maximum number of points, no matter how well they perform. Finally, assessing PHAs for the performance of their non-federal units by means of a federally-devised evaluation is neither fair nor reasonable. The HUD final rule on the Public Housing Assessment System (PHAS) states that the procedures which will be used to calculate the score for the financial indicator will be published in the Federal Register in the near future. Although all the details are not yet known, a broad outline of this scoring system is available. The Financial Indicator will be worth 30 out of the 100 points. A PHA must receive 18 points, or 60 percent, to achieve a passing score. PHAs scoring above 60 overall, but receiving fewer than 18 points in the financial indicator will be considered a substandard financial performer and be referred to the Troubled Agency Recovery Center. The indicator is divided into the following six components.
Audit and internal control flags will also be used by HUD to deduct points, although PHAs will receive no positive score for clean audits. It seems logical to PHADA that if a weakness in an audit is important enough to cause a PHA to lose points, a clean audit should be considered valuable enough by HUD to credit a PHA with points. The financial indicator also places each HA in a peer group based on its unit count. There are six peer groups, based on the following sizes:
Peer Groups This difference is a significant one, because now PHAs will not be compared against other PHAs with the same relative number of public housing units, but against PHAs which can have very different kinds of unit configurations. In short, HUD will not be comparing apples to apples. For instance older authorities in the East and Midwest with predominantly conventional low-income stock may find themselves in the same peer group with younger authorities in the West with a predominantly Section 8 unit base. By including Section 8 and non-federal units, the peer groups become very arbitrary. If the intention of the peer groups is to compare authorities with similar profiles, this intention is lost because HUD will be combining very dissimilar authorities. Having an authority with 1000 conventional low-income public housing units in the same group with an authority which has 200 conventional low-income, 300 state units, 100 tax credit units and 400 Section 8 units basically does away with the meaning of the peer groups. These two authorities are not peers. A significant number of PHAs do not have Section 8 programs. The peer groupings are important, because HUD no longer provides fixed standards for PHAs to achieve, but allocates points based on a housing authority's relative performance compared to its peers. In the first two components, Current Ratio and Months Expendable Fund Balance, authorities falling between 30 percent and 80 percent of their peer grouping receive the full nine points, with authorities below the 30 percent receiving a score which descends downwards, potentially to zero, while authorities above 80 percent have their score prorated down to 7.5 points. These 30 and 80 percent thresholds will be set based on the current breakdown and then revised every three years. For the third and fourth components, Tenant Receivable Outstanding and Occupancy Loss, the authorities which fall among the top 50 percent of their peer group receive the full 4.5 points, while the rest are prorated down to the 95th percentile, at which point the score becomes zero. The 1.5 points for Expense Management are allocated on an all or nothing basis, with zero points going to the lowest 5 percent of the group. The Net Income component has a threshold, and authorities which fall below the threshold will receive a zero, with the others receiving the full 1.5 points.
Entity-Wide It is also unclear whether the data provided in the annual financial report will suffice for scoring purposes. Occupancy Loss, for instance, is being scored based on unit count, rather than by rent receipts, so PHAs may now have to begin reporting occupancy data on their non-federal programs. PHADA believes that authorities should not be scored on an entity-wide basis. HUD receives information concerning a PHA's entity-wide financial condition in the annual audit, so it already has any information it needs to determine whether or not a PHA's entity-wide financial condition poses a threat to its federal programs. Having a federal agency scoring a PHA on non federal programs, however, seems like an unwarranted and unnecessary intrusion. Including the Section 8 programs for comparison purposes, not only will make it unfair for some authorities since they will not genuinely be rated among their peers, but it also seems as if it will detract from a pure evaluation of the low-income public housing program. Up until now, Section 8 reserves were not included in evaluating the solvency of a conventional federal low-income program, but apparently now, it no longer matters what the low-income program reserves are, if the Section 8 reserves are healthy. Again, is that fair to authorities without Section 8 programs? Including non-federal programs only compounds this difficulty. In essence, the low-income public housing program is no longer being evaluated.
Scoring Additionally, however, the existence of these high reserve PHAs in a group moves the whole curve over, so some PHAs without stand alone programs, will be shifted into the under 30 percent range and lose points for possibly no reason. If 20 percent of the PHAs in a peer group have stand alone programs, they likely will move to the high end of the group, meaning that three-eighths, or 37.5 percent of the remaining authorities which do not have stand alone programs will be placed in the below 30 percent category, which is clearly unfair. See Diagram 1.
A housing authority may find itself in the under 30 percent range simply because it is being compared to other housing authorities with an entirely different program mix. Since these two components each are worth so many points, 9, (making up 60 percent of the financial indicator score added together) and since being low in one almost guarantees being low in the other, a PHA which has gotten pushed into the below 30 percent range will be in danger of failing the financial indicator. Not only can an authority in this position point out the flaw in the mixed peer groups, but also in using a curve to distribute points, because such a PHA can have a perfectly acceptable reserve level and still be referred to the Troubled Agency Recovery Center simply because its peer group is not genuinely comparable. This unfair situation is made worse by the addition of the Section 8 units. HUD has never set a standard for reserves in the Section 8 program. Therefore, a housing authority with a large Section 8 program and no Section 8 reserves may find itself with a lower reserve level than housing authorities with mostly conventional units -- since they have always had a reserve level standard -- without it being any reflection on its management ability. Conversely, and perhaps more commonly, housing authorities which have decided it was prudent to build up large Section 8 reserves may find themselves in the over 80 percent group, and again shifting the curve. If 50 percent of a peer group is made up of PHAs with stand alone and Section 8 programs, and they tend to congregate at the upper end of the range -- then of the 50 percent of PHAs without these programs, fully 60 percent of them could end up making up the bottom 30 percent and therefore lose points. See Diagram 2.
Peer groups using all units, then, are skewed against PHAs which simply manage conventional low-income programs. It seems neither fair nor reasonable to compare programs which have had completely different program requirements. HUD writes in the final rule concerning the entity-wide scoring, "To the extent PHAs enter into non-Federal activities that contribute to their financial health, these PHAs should receive higher scores than those PHAs that have entered into arrangements that negatively affect the health of the PHA." Yet having a large replacement reserve in a non-federal stand alone program does not contribute to a PHA's health. In fact, although, it may be large in relation to the federal low-income program, it still may not be large enough for replacement purposes. A Section 8 program is not a non-federal program, so it is not even addressed in HUD's rationale for entity-wide scoring. HUD does not genuinely believe activities which contribute to financial health should receive higher scores, because it scores the PHAs who are over the 80 percent range in reserves in their Months Expendable Fund Balance peer group lower than those in the 30-80 percent range. Thus, its actual scoring contradicts its statement in the Federal Register. Scoring Tenant Receivable Outstanding will also pose some difficulties. First of all, presumably Section 8 units will not be scored in this component. Therefore, all authorities which have Section 8 programs will have these units dropped for the purpose of this indicator. Dropping these units, however, may actually be enough to put the PHA into a different peer group category, although HUD will apparently not make this switch. For instance, a PHA with 300 conventional low-income units and 300 Section 8 units, for a total of 600 units, will be placed in the High Medium peer group. For Tenant Receivable Outstanding, though, the 300 Section 8 units will not be counted, so with its 300 remaining units, it should actually be classified as a Low Medium authority. Yet, it will still be rated against High Medium authorities. As a result, housing authorities with different scale programs will be compared against each other. If the peer groups have any value, then it is not reasonable to include the Section 8 units for some components and then simply remove them for others. Which peer group does a housing authority belong in? Secondly, state moderate-income rental programs, tax-credit programs, and market rent programs will be renting to a higher income resident than federal low-income. Therefore, the tenants accounts receivable can be on a different scale. In a moderate-income rental program, a working family paying a significant rent which may fail to report a change in income for several months and have a retroactive rent adjustment applied to their account can find themselves owing thousands of dollars very quickly. Thus, authorities with these programs will be unfairly compared to authorities which just have federal low-income programs where large rent balances cannot accumulate quickly. Occupancy Loss will be even more difficult for HUD to evaluate. Section 8 units will not be included, so the peer groups will be skewed once again. The necessary data for non-federal programs will not be included on the Financial Data Schedule, so HUD will have to collect this information in another fashion. These non-federal programs serve different markets and have different funding mechanisms than the low-income program. There is no reason a housing authority should either be scored lower, because it has occupancy problems in a non-federal program, even though its public housing is well managed, or conversely benefit from a high occupancy rate in non-federal programs masking occupancy difficulties in its federal programs. It does seem presumptuous of HUD to score PHAs on occupancy data for programs it does not fund, has no involvement with and essentially knows nothing about. Scoring the financial indicator on an entity-wide basis poses another problem for HUD. If an agency fails the financial indicator, it is referred to the Troubled Agency Recover Center (TARC). If the failure was due to low reserves, high vacancy rates, or tenant accounts receivable in non-federal programs, does the TARC have jurisdiction to intervene in programs regulated by state and local bodies? If the TARC cannot intervene, what is the value of being scored in the first place? Expense Management had been scored based on the federal low-income units only, but the language in the final rule is no longer clear on HUD's intention. It is not clear either why HUD would have one set of rules for this component and another for all the others, if it believes the non-federal programs need to be evaluated. Selecting low-income public housing units only for this component adds a third separate way to form the peer groups. A peer group can be either:
This arbitrary switching of peer groups from one component to another invalidates and renders meaningless the whole concept of placing authorities in similar groups. Net Income is to be scored on a threshold. Authorities which run more than a 20 percent deficit will receive no points, all others will receive the 1.5 available. If the overrun is due to spending in non-federal programs for perfectly sound reasons, it does not seem fair to penalize an authority for its decision. If the authority ended up at the TARC would HUD be able to override other regulatory bodies to control non-federal expenses?
Conclusion This problem is even more acute, because by including all units, including Section 8 and non-federal units, the peer groups are not genuinely comparable. The peer groups are made even more arbitrary by the fact that Section 8 and non-federal units will be included for some components and not for others. Finally, penalizing PHAs for performance in non-federal units, when HUD has no history of involvement in these programs, no knowledge of them and is unlikely to be able to effect any change puts HUD in a position of overreaching its legitimate role of oversight of federal programs. Analysis reflects the financial indicator protocol and scoring process as of February 2000. Developed by PHADA on February 9, 2000. Email: tvandyke@phada.org. Please distribute as needed. Members are encouraged to forward this analysis and others developed by PHADA on the Public Housing Assessment System to their Members of Congress. Also, refer to PHADA's review of the PHAS final rule (http://www.phada.org/01phasf.html) and the analysis of the physical inspection scoring process (http://www.phada.org/01phass.html). | ||||||||||||
