Article in Press Printer-friendly PDF fileWestside Observer Newspaper
May 2019 at www.WestsideObserver.com
2015 Affordable Housing $310 Million Bond Update
Affordable Housing That Wasn't
Article in Press Printer-friendly PDF file
Westside Observer
Newspaper
May 2019 at www.WestsideObserver.com
2015 Affordable Housing $310 Million Bond Update
Affordable Housing That Wasn't
by Patrick Monette-Shaw
In November 2015 voters approved a $310 million Affordable Housing Bond. The Westside Observer has reported on the bond several times. Now’s a good time for another review, since the mayor is proposing a second affordable housing bond even before the first one is finished.
What’s happened in the three-and-a-half years since the 2015 housing bond was passed isn’t heartening for non-teachers and others targeted in the middle-income bond category explored in this article.
The Examiner reported on April 1 that Mayor Breed wants to increase a new $300 million affordable housing bond being planned for San Francisco’s November 2019 ballot by $200 million, to a total of $500 million. Supervisor Matt Haney — not to be outdone — wants to kick it up to a $1 billion bond.
The Capital Planning Committee’s tentative plan is to consider bonds for the November 2019 ballot during its May 6 meeting, and may consider increasing the amount of the new affordable housing bond then. The Capital Planning Committee is an 11-member body consisting of nine major City department heads, the Mayor’s Budget Director, and the president of the Board of Supervisors. The Board of Supervisors has until mid-July to approve placing a bond measure on the November 2019 ballot and submit it to the Elections Department.
Problems With Bond Oversight
There are a number of problems with the first affordable housing bond.
The Slow Pace of Bond Spending
The slow pace of issuing the 2015 housing bonds is troublesome. The Controller’s Office confirmed that as of April 12, 2019 just $217.3 million — 70% — of the $310 million bond has been issued so far. The first tranche1 was issued on November 1, 2016 for $75.1 million a year after voters approved the bond; the second tranche was issued on May 23, 2018 for $142.1 million, two-and-a-half years after voter approval.
MOHCD claims the remaining third tranche — for $92.5 million, fully 30% of the bond — won’t be issued until the Fall of 2019, almost four years after voters approved the bond in 2015. Why is it taking so long to get this money into the pipeline? Weren’t these shovel-ready housing projects?
The Controller reports interest on the first two tranches currently issued will cost $65.8 million in debt service through 2038. Interest on the remaining $92.5 million tranche is an estimated additional $20 million, for a projected total of $85.8 million in interest on the $310 million bond principal. Total bond costs will be just under $400 million. What are we getting for it? Were the right allocation decisions on bond spending made?
The Overseers: CGOBOC Members and Meetings
On March 5, 2002 voters passed Proposition “F,” the Citizen Oversight of Bond Expenditures Initiative, creating a nine-member Citizen’s General Obligation Bond Oversight Committee (CGOBOC) to oversee spending of all general obligation bonds passed by voters. Then when voters passed a ballot measure in 2003 creating the City Services Auditor program housed in the City Controller’s Office, CGOBOC was handed additional duties as the Citizen’s Audit Review Board and was also assigned to oversee the City Controller’s whistleblower program.
All bonds passed by the voters are overseen by the nine-member CGOBOC. Three members are appointed by the mayor, three by the Board of Supervisors, two by the City Controller, and one by the Civil Grand Jury.
CGOBOC is currently overseeing $4 billion across 12 various bonds it receives reports about. That’s a lot of work for one committee to monitor. Plus, the scope of what CGOBOC is responsible for overseeing has grown considerably.
CGOBOC’s historical and institutional knowledge about any given bond measure is a serious, inherent problem. Membership is limited to two consecutive two-year terms; it’s rare that a member takes a break between terms and then returns for an additional term. Many of the bonds run for far longer than the maximum four-year tenure that members serve.
CGOBOC has done a terrible job so far holding the Mayor’s Office of Housing and Community Development (MOHCD) accountable for the affordable housing bond spending. Successive updates from MOHCD keep shifting planned bond uses, and CGOBOC has done little to reign in MOHCD. Why does planned spending of the bond keep shifting so often?
Of CGOBOC’s 20 meetings held between January 2016 and March 2019, MOHCD has only presented seven times on the affordable housing bond to CGOBOC. Across the seven meetings in which MOHCD presented, there have been 15 different CGOBOC members, given on-going turnover. During those seven meetings, CGOBOC had three meetings with only eight members, three meetings with just seven members, and one meeting with just six members — due to vacancies on the committee. And sometimes the appointed members were absent from meetings.
When CGOBOC held its first hearing on the affordable housing bond on January 26, 2016 it had eight members, five of whom are no longer committee members. Of the three members who remain, two had their terms expire in late 2018 but remain as hold-over members until their replacements are named. The third member’s term expires in June 2019.
Because the November 2015 affordable housing bond is far from over, CGOBOC’s institutional knowledge of bond spending will be severely curtailed when its three remaining members are termed out and replaced.
Lack of Metrics to Evaluate Bond Spending
Back in January 2016 when CGOBOC held its first hearing on the Affordable Housing Bond, several CGOBOC members expressed the need to develop “metrics” to assess bond spending.
CGOBOC member Brenda Kee McNulty specifically fretted on January 28, 2016 about the lack of metrics. During the discussion about developing metrics, McNulty astutely wondered:
“But I think it’s helpful to be able to at this early stage of the game [to] lay out those areas, or categories of returns … so that we can go back [later to] evaluate how successful ... what our return was for the whole bond program. So, the relative returns of these categories will also shed some light into whether or not we made the right allocation decisions.”
CGOBOC members in January 2016 suggested several metrics, including the dollar amount allocated per project, the Area Median Income (AMI) targets, number of new units constructed vs. number of existing housing units preserved, household size, demographics targeted (e.g., populations such as the homeless, seniors, veterans, teachers, families, etc.), and the Supervisorial districts in which projects are actually developed.
The minutes of CGOBOC’s July 2016 meeting report that CGOBOC member Jennifer Warburg — the Committee’s co-liaison to MOHCD on the housing bond — again reiterated that “better metrics are needed as well as more details on actuals [production].”
The July 2016 minutes also report that CGOBOC’s other co-liaison to MOHCD on the housing bond — Larry Bush — had requested information from MOHCD on March 25, 2016, but the “information was not provided.” The minutes indicate MOHCD’s Kate Hartley was given until September 2016 to have time to prepare responses to Bush’s questions.
In response to questions Bush and other CGOBOC members raised in July 28, 2016 MOHCD provided a reply (dated September 12, 2016) during CGOBOC’s October 3 meeting. Hartley’s answers were presented on October 3, seven months after the questions were raised, but there was no discussion of proposed or final metrics to assess bond spending.
Hartley’s October 2016 response indicated “[MOHCD] will have a full Bond Report with metrics available for presentation and review at your January 2017 meeting.”
Throughout most of 2016, MOHCD stalled developing in collaboration with CGOBOC any meaningful metrics. Then, in November 2016, MOHCD claimed the best metrics to measure the use of bond funds were just three measures: 1) The number of households served, 2) The number of units created, and 3) Per-unit costs. MOHCD’s three skimpy metrics are completely insufficient.
When MOHCD presented to CGOBOC on January 26, 2017 no detailed metrics by which CGOBOC could assess the various bond-funded affordable housing projects were presented or reviewed. Sadly, CGOBOC has not exhibited much of a spine in insisting MOHCD develop — and stick to — meaningful metrics to assess bond-funded projects.
Despite the three measures MOHCD claimed in November 2016, MOHCD’s quarterly reports began reporting in January 2017 in a section titled “Metric of Success”
“The primary metric of success for the Affordable Housing Bond is number of units produced, protected, or assisted. We have estimated the projected number of affordable units through all phases of the bond process, and will track progress and provide regular updates accordingly.”
Of note, the section title used the singular for metric, not the plural for metrics. It now appears MOHCD’s single metric is the number of units produced by the bond. To date, MOHCD has not stratified or informed CGOBOC members with a breakout on the number of new housing units produced versus existing units protected, rehabilitated, or preserved.
Despite CGOBOC members’ initial interest in January 2016 about the issue of metrics to assess McNulty’s concern about whether “the right allocation decisions” had been made, CGOBOC members went silent on the metrics issue in January 2017 and apparently haven’t discussed the issue any further between January 2017 and April 2019.
It is thought by some observers that the metrics now reported basically consist of updates regarding MOHCD’s progress on bond spending, rather than on any meaningful measurements of bond activity or metrics.
MOHCD’s own Construction Project Management form “Project Monthly Report,” tracks the units by housing types, including: Homeless (transitional age youth), homeless (non-transitional age youth), very-low income seniors, very-low income families, disabled, or public housing replacement units. But no data has been provided to CGOBOC itemizing how many of the units have been dedicated to the various “vulnerable” populations the bond was designed to serve, nor has any data been provided regarding AMI income levels served.
On April 12, Ms. McNulty, was asked to comment for this article on whether four years into the Affordable Housing Bond she now thinks using only the number of units produced is an adequate metric of the bond's performance. McNulty didn’t respond.
Changes in Bond Spending
In 2014 and 2015 during on-going planning prior to placing the Affordable Housing Bond on the November 2015 ballot, successive documents from the Mayor’s office included several proposed programs for the bond that were subsequently eliminated, including a $20 million “Top-Loss Catalyst Fund” accelerator or incubator fund to help nonprofit developers purchase land and build permanently affordable units, a $17 million Middle-Income Rental Housing program, and a $25 million Expiring Regulations Preservation category to preserve existing rental units. The first sub-category vanished even before the bond was put on the ballot, and the remaining two sub-categories were introduced in January 2016 but were eliminated by July 2016.
In addition, two projects that were later proposed in the Low-Income Housing portion of the bond — at 250 Laguna Honda Boulevard and 4840 Mission Street — were removed from bond spending after they had been first included and described to CGOBOC. The 250 Laguna Honda 150-unit senior housing project was removed due to site instability following a geological study and excessive costs; it was replaced with a 94-unit senior housing project at 1296 Shotwell, a loss of 56 units for the elderly2. The Mission Street project was scrapped due to delays, and its funding re-allocated to another low-income family-housing project.
The number of units planned in each of four main housing categories have bounced all over the place in successive reports to CGOBOC. Didn’t CGOBOC members notice during the past three-and-a-half years the drastic changes in planned bond uses?
Main Categories of Planned Bond Uses
Table 1 illustrates that the unit counts in each of the four main categories have shifted significantly over time. Figure 1 visualizes the data.
Table 1: Shifting Unit Counts of Bond Main Categories:
Figure 1: Shifting Units Counts of Bond Main Categories:
As Table 1 shows, the planned units dropped from 1,435 total units in January 2016 to 1,366 units in July 2016, climbed back up to 1,785 units in September 2017, but have dropped back to 1,522 total units as of December 2018.
The data in Table 1 and Figure 1 illustrate that when plans were presented to the Board of Supervisors in June 2015 and first presented to CGOBC in January 2016:
The Public Housing category would rehabilitate 206 units at the Sunnydale and Potrero public housing sites. As of December 2018, MOHCD reported that those 206 units have grown to 517 units.Didn’t CGOBOC members notice the drastic shifts in planned units in each major category over time?
Fudged Unit Count Data?
It’s somewhat surprising that MOHCD has consistently misreported its unit count numbers to CGOBOC. And it’s more surprising that CGOBOC members appear not to have called out MOHCD for reporting incorrect data.
In December 2016, MOHCD reported to CGOBOC that there would be 757 units constructed or rehabilitated in the $100 million Low-Income Housing portion of the bond. It took until MOHCD’s December 2018 quarterly report — which report has not yet been discussed during a CGOBOC meeting — to admit on page 36 that it had wrongly counted and included in the low-income category 386 units involving predevelopment projects it should not have reported.
And the December 2018 report also admits on page 34 that of 517 units in the $80 million Public Housing portion of the bond it includes 125 market-rate units in the count of the 217 units in the Potrero Phase II Infrastructure Development project. CGOBOC should ask MOHCD why it is including market-rate units in the count of Public Housing units, potentially padding the unit counts.
Between the two problems, the reported unit counts appear to have been off by 511 units between the Public Housing and Low-Income Housing categories.
Middle-Income Housing Cuts
Despite Mayor Ed Lee’s observation in Time magazine in January 2014 that everybody assumed the middle-class were moving out of San Francisco, and despite MOHCD Director Olson Lee’s lame observation during the Board of Supervisors Government Audit and Oversight Committee hearing on October 21, 2016 — after the affordable housing bond passed in 2015 — that MOHCD had “found out the need for affordable housing was on-going,” uses of the 2015 bond to help middle-income residents have shrank in the number of planned units.
The main reason the middle-income housing category was added to the bond is because there are no state or federal funding sources for middle-income housing. So, San Francisco took the lead, carving out a dedicated $80 million as a local source of funding subsidy for middle-income housing. MOHCD’s January 2016 report to CGOBOC noted:
“The need is particularly acute for moderate-income households, for whom there is no federal or state financing programs (such as low-income housing tax credits, which cap eligibility at 60% of AMI) that the City can leverage with its own subsidies.”
During planning leading up to putting the November 2015 bond before voters, as far back as June 23, 2015 the Board of Supervisors received an MOHCD report, indicating (on page 26 of the PDF file) that the bond would provide $57 million toward 519 planned middle-income housing units.
It included 235 rental units in the Middle Income Housing main category between the Middle-Income Rental Program and the Expiring Regulations Preservation subcategories, which were presented in documents to the Board of Supervisors in June 2015 and presented to CGOBOC in January 2016. The Expiring Regulations Preservation was intended to protect existing units with affordable rents from expiring and becoming market-rate units, displacing tenants.
The remaining 284 units — 34 downpayment assistance (DALP) loans and 250 Teacher Next Door loans — were for loans for middle-income ownership purchases, not rentals.
Unfortunately, the Middle-Income Housing category also saw shifts in the unit counts during MOHCD’s successive presentations to CGOBOC.
Table 2: Shifting Units Counts Within Middle-Income Main Category:
Again, between the time when plans were presented to the Board of Supervisors in June 2015 — and first presented to CGOBOC in January 2016 — and December 2018:
The Middle-Income Rental and Expiring Regulations sub-categories involving 235 rental units were eliminated by July 2016, leaving just 103 rental units. The drop from 235 to 103 rental units represents a 56.2% change decline in the number of middle-income rental units. The remaining 103 rental units now planned include one project for 21 units at 88 Broadway, and one project for 82 units at 43rd & Irving reserved for teachers.
How Did the Middle-Income Rental Program Vanish?
The November 2015 Affordable Housing bond explicitly asked voters the question of whether the bond would include a middle-income rental program. The legal text of the Affordable Housing Bond clearly stated in Section 3-E on page 156 in the November 2015 voter guide that a portion of the bond would be used to create “Middle-Income Rental Housing.”
When CGOBOC first held a hearing on the 2015 Affordable Housing Bond on January 26, 2016 CGOBOC members were provided an MOHCD report that described the Middle-Income Rental program and the middle-income Expiring Regulations categories. (Note: The report was modified and updated on January 19, 2016.)
Unfortunately, the report did not describe the Expiring Regulations Preservation category, but it was included in the middle-income category listed on page 9 of the report in a table showing planned bond spending.
CGOBOC members should have known that the legal text and question posed to voters had mandated the Middle-Income Rental program. And they should have noticed in July 2016 that the Middle-Income Rental program presented to it in January 2016 when they assumed oversight of bond spending was suddenly eliminated seven months later from the “Middle-Income Housing” main category of planned bond spending presented to CGOBOC in July 2016. But the minutes of CGOBOC’s July 28 meeting and subsequent meeting minutes never reported any questions, concerns, or discussion by CGOBOC members about why the category vanished. Not one question was raised.
Why did the category vanish?
The January 2016 report described in some detail on page 7 the proposed Middle-Income Rental program to be funded by the bond. The description read in part:
“Bond funds may be allocated to support the creation of permanently affordable rental units designated for middle-income households that are currently not served by the City’s traditional affordable housing programs. Bond funds used for the creation and support of middle-income rental units will prioritize family-sized units.”
Given the temporal proximity in time, some observers wondered whether MOHCD may have decided that after voters passed Proposition “C” in June 2016 that MOHCD could remove the Middle-Income Rental program from the Affordable Housing Bond a month later in July 2016. That was patently ridiculous, because Prop. “C” only dealt with setting the inclusionary percentages of affordable units available in market-rate development projects, including both ownership units for purchase and rental units. It’s ludicrous to believe the inclusionary aspects of Prop. “C” in any way solved or achieved affordable rents designed to be addressed through the bond’s Middle-Income Rental program.
When asked to comment about whether Prop. “C’s” passage had contributed to MOHCD’s decision to eliminate the Middle-Income Rental program, MOHCD’s director Kate Hartley indicated on April 22, 2019 only that:
“Between July 2016 and December 2018, we allocated more middle-income funding to DALP loans as the best way to maximize the number of middle-income households served from bond funds. Producing middle-income rental as a stand-alone building is quite challenging and expensive because those projects cannot fully leverage Federal low-income housing tax credits.”
First, Hartley ignored answering the question about Prop. “C,” and instead confounded rental programs with the DALP ownership loan program. Wasn’t the whole point of including the local Middle-Income Rental funding in the bond designed to address the lack of Federal middle-incomehousing tax credits for rental projects?
Second, the $10 million for DALP loans proposed in June 2015 and presented to CGOBOC in January 2016 that was then increased to $34.4 million in December 2018 was for middle-income ownership units, not rental units.
When pushed again for comment on whether Prop. “C” had contributed to eliminating the Middle-Income Rental program, Hartley indicated the next day on April 23:
“… We did not conclude that Proposition C’s tiered income eligibility approach solved the middle-income housing need. What we did do is consider which available rental housing projects provided the best investment opportunity for bond funding. That analysis resulted in a higher investment in middle-income teacher housing and other middle-income housing in MOHCD-sponsored developments than we originally planned.”
Notwithstanding Hartley’s prevarication, some observers remain unconvinced and believe passage of Prop. “C” contributed to MOHCD eliminating the Middle-Income Rental Program.
Hartley didn’t elaborate on why the “best investment opportunity” was for ownership units, not rental units. And she didn’t address the 56.2% change decline involved in cutting 235 rental units (combining the Middle-Income Rental and Expiring Regulations Preservation projects) down to just 103 rental units (combining the 43rd & Irving and 88 Broadway projects).
Just how much hubris is MOHCD demonstrating luring voters into passing the bond based on promises to devote bond funding to the Middle-Income Rental Housing category, and after voters approved the Bond subsequently removed that category unilaterally based on the flimsy excuse some unnamed “fix” had later surfaced making it no longer needed?
Dialing for Middle-Income Category Dollars
Seven or so successive MOHCD reports to CGOBOC have documented some drastic shifts in spending within the Middle-Income Housing category budget. Table 3 illustrates:
Table 3: 2015 Middle-Income Budget Dollars
In June 2015 and January 2016, MOHCD reported to CGOBOC that between the DALP and Teacher Next Door loan ownership programs, the combined $15 million would comprise about 26% of the Middle-Income Housing budget. 
By December 2018 the now combined $39.4 million between the two ownership purchase programs increased from 26% to 49% of the Middle-Income Housing budget, a 23% increase in share.Figure 2: Shifting Dollars in Middle-Income Housing Category:
Between the Teacher Next Door forgivable ownership loans and the Middle-Income Teacher Housing at 43rd & Irving, at least $34 million — 43% — of the $80 million Middle-Income Housing main category is reserved for teachers. But that doesn’t include 12 of the DALP loans that have also been awarded to teachers so far.Didn’t CGOBOC members notice the shifts in planned spending in the middle-income housing category over time, either?
Overall Bond Performance
As noted previously, almost one-third of the $310 million Affordable Housing Bond passed in November 2015 has not been issued yet (the third bond tranche), and won’t be until the Fall of 2019 — nearly four years after voters approved the bond.
Slow Pace of New Construction
Aggravating the slow issuance of the bond funds, there’s been a very slow pace of construction of new housing funded by the bond three-and-a-half years later, as shown in Table 4, below. It’s almost as if MOHCD — and by extension, CGOBOC — feels no sense of urgency in bringing the affordable housing on-line as quickly as possible.
Of nine probable construction projects, just four projects are currently under construction (as of April 2019), for a total of 364 units, just 24% of the planned 1,522 total units MOHCD reports are being funded by the bond.
Table 4: Four Construction Project’s Monthly Progress Reports
The forms used for construction projects includes a section — “Units by Housing Type” — listing the types of vulnerable populations targeted for the construction. Table 4 illustrates, in part, that of the four projects currently under construction:
75 of the 349 units (20.6%) are designated for very-low-income senior citizens.
130 units — nearly 36% — of the 364 units are public housing replacement units, including 36 units in the Mission.A closer look at the units proposed for bond funding reveals other problems, shown in Table 5.
Table 5: Status of Bond Construction Progress
Only 883 (58%) of the total 1,522 units appear to be planned new construction.DALP and TND Performance
Nearly three years ago, funding for middle-income rental housing programs under the bond was essentially defunded. On April 22, 2019, MOHCD’s director Kate Hartley indicated by e-mail: “There was also such great demand for DALP loans that we increased that program’s share of the middle-income funding” to explain why the middle-income rental housing programs were cut back in favor of ownership units.
Table 6 illustrates that now three-and-a-half years after the bond was passed, only 45 — 40.2% — of the planned 112 DALP loans have been issued, indicating demand might not be so high. And only 30 of 60 Teacher Next Door loans have been issued. The slow pace suggests demand may not be that great.
Why is it taking almost three full fiscal years to issue these loans? Will it take an additional three fiscal years or longer to issue them once the third bond tranche is issued in the Fall of 2019, considering that the bond had been passed in 2015?
Table 6: DALP and Teacher Next Door Ownership Loans From 2015 Affordable Housing Bond
Table 6 shows that just 41.6% of the DALP funding — $14.3 million of the planned $34.4 million — has been issued to date. And just 16.4% of the TND loans — $820,000 of the planned $5 million — has been issued to date.
MOHCD reported that eligibility for the Middle-Income category DALP and TND loans would be for households earning 120% to 175% of Area Median Income (AMI) — $99,500 to $145,100 for a one-person household, and up to $127,50 to $186,450 for a three-person household. TND loans were extended to 200% of AMI, which is up to $165,800 for a one-person household, and $213,100 for a three-person household. The low-end of AMI for the TND loans is thought to be 80%.
Data provided by MOHCD (not shown) also shows:
Of the 30 TND loans issued to date, the household AMI levels ranged from 72% to 194% of AMI; 17 (over half) of the loans were to households earning less than 120% of AMI. Fully 13 (43.3%) of the TND loans funded to date by the Middle-Income category of the bond were for households earning over 120% of AMI that is defined under RHNA goals as being “above moderate-income” households.
It bears repeating that the DALP and TND loans and the teacher housing at 43rd and Irving, totaled $68.4 million — 86% — of the $80 million Middle Income Housing category. A significant percentage of that went to households earning over 120% of AMI.Housing Production Under RHNA Quotas
Housing production in San Francisco has not kept pace with the Regional Housing Needs Allocations (RHNA) going back to 2007 or earlier. San Francisco’s Planning Department has previously provided data on actual housing units produced, but facing pressure from jurisdictions in other areas of the State San Francisco began reporting in 2015 the number of building permits issued, instead. It’s well known that once developers have been awarded permits, some of their projects fall through and the planned units aren’t produced.
Table 7: San Francisco’s RHNA Housing Goals Production
Table 7 shows that between 2007 and 2014 for units actually produced:
Across the three categories of very-low-, low-, and moderate-income levels, just 7,064 units — 37% — of the RHNA planned 18,878 units were built.Table 7 also shows that four years in to the eight-year period between 2015 and 2022, units receiving building permits have seen worsened trends for the three lower categories (very-low-, low-, and moderate-income levels).
By contrast, the “above moderate-income” category — households with income of over 120% of AMI — has already seen four years into the 2015–2022 eight-year period 126% percent of permits issued, 3,266 more than the 12,536 planned for the whole eight-year cycle.Other Problems
CGOBOC’s FY 2017–2018 Annual Report dated September 2018 but prepared in November reports on page 7 a problem of coordination with Pacific Gas and Electric to ensure that utilities are in place, a problem that may be accounting for a six- to eight-month delay in bringing some housing from the November 2015 bond online.
And the bond focus appears to have shifted to only the number of new units produced, rather than on existing units preserved or on vulnerable populations served.
Caveat Emptor
Before voters approve another Affordable Housing Bond in November 2019, City Hall needs to abandon Mayor Lee’s belief that the middle-class are moving out of the City, and needs to dedicate more local funding for middle-class rental housing production.
CGOBOC member Brenda Kwee McNulty appears to have been on to something when she raised the issue in January 2016 that firm metrics needed to be developed to evaluate spending of the 2015 Affordable Housing bond, and whether the correct allocations of bond spending had been made. As it is now, the sole metric seems to only be the number of units produced. MOHCD’s reporting of its unit numbers seems to be all over the place, perhaps inflated, and perhaps still shifting going on four years into the 2015 bond.
CGOBOC must adopt meaningful metrics to evaluate bond spending before another affordable housing bond is passed.
Any new affordable housing bond Mayor Breed is hoping voters will pass in November 2019 must include significant funding for middle-income rental housing planned for, but not produced by, the 2015 bond.
Now more than ever, we may need a Charter amendment to create a Commission having oversight of the Mayor’s Office of Housing and bond-funded housing development.
1 Definition of tranche: “A division or portion of a pool or whole; specifically: An issue of bonds derived from a pooling of like obligations (as securitized mortgage debt) that is differentiated from other issues especially by maturity or rate of return.”
2 MOHCD reported to the Board of Supervisors Land Use Committee on April 29 that 53 senior housing units are included in the bond-funded project at 735 Davis Street, but MOHCD reports to CGOBOC haven’t reported how many of the units may have already been previously planned for seniors even before the 250 Laguna Honda project was defunded..
Monette-Shaw is a columnist for San Francisco’s Westside Observer newspaper, and a member of the California First Amendment Coalition (FAC) and the ACLU. He operates stopLHHdownsize.com. Contact him at monette-shaw@westsideobserver.com.
Postscript: 2019 Affordable Housing Bond Preview
After this article was finished, planning documents for the proposed November 2019 housing bond were posted on the Capital Planning Commission’s web site on Friday, May 3 in advance of the committee’s May 6 meeting.
The documents summarize planned uses of the new housing bond focuses principally on public- and low-income housing.
Adding in the Senior Housing planned in the 2019 bond to the low-income and public housing categories, at least $630 million — 78% — of the $810 million in funding between the two bonds is being directed to those earning less than 80% of AMI, and perhaps more if you add in the new Affordable Housing Preservation category for folks who earn 30% or more of AMI.
Table 8: Two Affordable Housing Bonds Planned Spending
When Time magazine interviewed then-Mayor Lee five years ago in January 2014, they asked him why San Francisco had been so slow to build housing. In addition to Lee replying “I don’t think we paid any attention to the middle class. I think everybody assumed the middle class was moving out,” Lee also admitted:
“Our city did pretty good in investing in low-income housing and trying to do as much as we could for the homeless. That was where our sentiments were.”
O’ how far we haven’t come: A large part of the problem with spending of the 2015 Affordable Housing bond is it continued the long tradition of funding low-income housing. The 2019 bond will end up perpetuating what will be a decade or longer of a terrible job of not funding investments in middle-income housing.
The $80 million Middle-Income housing category in the 2015 bond represented 26% of total spending. It is being cut to $20 million — just 4% — of the planned $500 million 2019 bond, and earmarked again primarily for DALP and TND ownership loans.
MOHCD’s; reports to CGOBOC on the 2015 Affordable Housing bond indicated that the Public Housing portion of the bond “would be spent on the most urgent capital needs [including infrastructure], and [would] strive for creation of net new units where possible.” A significant part was spent on infrastructure, not new units.
Similarly MOHCD is now saying the proposed funding for Middle-Income housing in the 2019 bond are “estimates.” That funding will likely change over time