by Patrick Monette-Shaw

Apparently, Jeff Kositsky, director of San Francisco’s Department of Homelessness and Supportive Housing (HSH), didn’t get the memo on the Laws of Common Sense.

Common sense tells you that when it comes to solving the City’s multiple homelessness-related problems, San Franciscans desperately want the Emperor’s New Clothes to cover backsides of City officials, not “invisible successes” waltzing around exposing their rear ends.

To be fair, Kositsky blurted his our successes are invisible nonsense during a September 18 Board of Supervisors hearing on his budget request to extend the contract for the Homeless Outreach Team’s (HOT) external service provider for two years through FY 2020–2021 at a cost of $15 million, including an additional 15 employees increasing the HOT team to 86 people, and 13% pay raises for them all.  The HOT Team’s contract received a 174.2% change increase, from $3.1 million in FY 2014–2015 to $8.5 million in FY 2019-2020.  (As a point of reference, City employees typically receive 2% to 3% annual raises, not 13%.)

Despite concerns raised by members of the Board of Supervisors Budget and Finance Committee who had questioned the program’s effectiveness and efficacy, the full Board approved the contract extension.  Despite efficacy concerns, both Supervisor Sandra Lee Fewer (D-1) and Supervisor Catherine Stefani (D-2) suggested they would like to see the HOT Team actually show up with a presence in their districts.  Why do these two supervisors want invisible successes in their Districts, given the program’s dubious efficacy?

Kositsky’s invisible successes may extend to the efficacy of additional homelessness services provided by his Department, including the contentious Navigation Centers and his role supporting the dubious Healthy Streets Operations Center (HSOC) — led predominantly by SFPD — at the City’s Department of Emergency Management and its 9–1–1 call center.

The Many Problems With “Navigation Centers”

San Francisco’s 2019 homeless count stood at 9,700 (which is the high end, because the Feds use a different formula).  San Francisco is estimated to now have a population of 897,166.  That means the 9,700 who are homeless represent just 1% of City residents.

The San Francisco Examiner, reported on October 5, 2019 that Mayor London Breed has pointed to a reduction in the number of tent encampments as a sign of success she is addressing the homelessness crisis, but the Examiner noted there are clearly not fewer homeless people — just fewer tents.  Since Breed took office, the shelter waiting list has remained above 1,000.  The Examiner faulted her for merely covering up the homeless by moving them from one neighborhood to another.   It reported:

“The City needs to stop using the police and the Department of Public Works as its first response to complaints about the homeless, stop pushing them from one neighborhood to another in a fruitless game of whack-a-mole and stop seizing the belongings of The City’s most vulnerable residents.”

Problems with the Navigation Centers for the homeless run the gamut.

Angela Alioto Noted Navigation Centres Are Dead-Ends

The Westside Observer published an article in March 2018 reporting on a mayoral debate held on the Westside leading up to the June 2018 special election to replace former-Mayor Ed Lee, who had died the previous December.  Mayoral candidate Angela Alioto faulted the then-elected City officials for having dropped the ball on the Ten-Year Plan to End Homelessness she had developed for then-Mayor Gavin Newsom as his Homeless Czar in 2004.

Alioto noted, in part, “We know what works.  Shelters do not work, programs that put people through six months of mental health program then dump them back on the streets is throwing money away.”  She also noted “Navigation Centers are also dead-ends, good for 120 days, then back out on the street.” 

Alioto appears to have been overly optimistic about the 120-day length-of-stays in Navigation Centers.  It’s far shorter.

Breed’s Fake “In General” Crime Stats

In October 2019, the Westside Observer published a terrific article by Lou Barberini, in which he noted that both Mayor Breed and Kositsky had issued identical public statements, wrongly claiming:

“In general, we know the data shows there is no link between the creation of a Navigation Center and an increase in crime in the surrounding area.”

Barberini reported that in July 2018 the American Medical Association had pulled its $40 million, five-day convention from San Francisco because of safety concerns for its members.  He noted the City’s $40 billion tourist industry is threatened by San Francisco’s years-long dysfunctional handling of its homeless situation.

If it’s not clear, Breed’s wild “in general there’s no correlation”claimis clearly a red flag.

Leading up to what City Hall had to have known was going to be a contentious meeting on April 3, 2019 with neighborhood groups over the City’s plan to construct a Navigation Center along the Embarcadero, Breed appears to have reached out to SFPD, which is known for its lousy data.

Barberini reported he had obtained public records from the Mayor’s Office, specifically data compiled from SFPD’s Crime Data Warehouse provided by its Professional Standards and Principled Policing Unit.  That’s the data Breed and Kositsky relied on to assert there was no correlation of increased crime around Navigation Centers.

Here’s a snapshot of the data provided to Barberini:

Table 1:  Crime Data Before-and-After Navigation Center Openings

The data had been initially requested by SFPD Commander David Lazar, who led the creation of the Healthy Streets Operations Center (HSOC) that was activated on January 16, 2018 using the Department of Emergency Management’s operations center.  The Principled Policing Unit prepared a report on April 2 for Lazar, comparing crime incidents that occurred within a ?-mile radius of each of four Navigation Centers two- to four-months prior to opening a Center to crime stats two- to four-months following a Center’s opening.

The April 2 report showed two of the Navigation Centers experienced increases in crime incidents after opening, and the other two Centers had declines in crime incidents.  That didn’t support Breed’s and Kositsky’s thesis very well that there’s no correlation to increased crime around Navigation Centers.

So SFPD’s Director of Strategic Communications, David Stevenson (a civilian employee), stepped in and placed a subsequent request for information by expanding the zones to a ¼-mile radius.  Stevenson, as SFPD’s messenger, handles external messaging for the department.  The Principled Policing Unit prepared a report dated April 3 for Stevenson, slightly improving the data by showing three of the Centers saw decreases in crime, and only one Center saw an increase in crime.

Oddly, the Central Waterfront Navigation Center went from having a 33% decline in crime on the ?-mile radius report, to having a 60% increase in crime on the ¼-mile radius report (with no explanation provided for why the crime rate flipped from a negative to a positive correlation).

Armed with the April 3 report, Breed and Kositsky marched to a meeting with concerned Embarcadero neighbors on April 3, armed with their fake-news spin control there’s no link between crime rates and placement of Navigation Centers.

Barberini noted problems with the data included, among other things:

• SFPD double-counted crime stats that occurred within ? of a mile radius of only four of the current homeless Navigation Centers with data within ¼ of a mile radius.  Clearly, data for the ¼-mile radius was incorrect by double-counting data from the ?-mile radius.

• The SFPD crime stats included all Part I crime statistics (felonies), many of which aren’t germane to the homeless.  Larcenies are, by far, the most frequent crime committed by the homeless.

• Importantly, SFPD did not provide the Mayor or Kositsky crime data surrounding the fifth Navigation Center at the Civic Center, the City’s longest-operated Center in a known high-crime area.

You may believe a mayor holding a master’s degree in Public Administration might have suspected she was being fed “GIGO” — garbage in, garbage out — by SFPD staff, given the problems with SFPD’s data analysis.  You’d be wrong.

The “in general” nonsense is worse than Barberini initially reported.

• In addition to not providing the Mayor or Kositsky crime data analyzing the fifth Navigation Center at the Civic Center in a known high-crime area, both Lazar and Stevenson also failed to analyze before-and-after crime data for a sixth Navigation Center at 1950 Mission, which opened on March 1, 2015 as the City’s first Navigation Center, and closed on November 1, 2018.  It’s also thought to have been in a high-crime area.

• As shown in Table 1, SFPD used different dates for the opening of each of the four Navigation Centers than the dates Kositsky’s HSH web site reports as the dates the Centers were opened.  How did potentially using the wrong opening dates further skew SFPD’s data?  Why are the two departments reporting different opening dates?

• Conveniently, the before-and-after data collected for the Bayshore Navigation Center was reduced from a six-month period to a five-month period when Stevenson expanded the perimeters to the ¼-mile radius, flipping the reported crimes from a 13% increase to a 3% decrease, simply by chopping off one month from the periods analyzed.

• As the reports to Lazar and Stevenson acknowledge, the data culled from SFPD’s Crime Data Warehouse for the Mayor included only Part I crimes.  The FBI’s Uniform Crime Reporting nomenclature classifies eight serious crimes in its Part I category.  Those eight offenses are typically charged as felonies.
  
  The Principled Policing Unit reports did not include any Part II “wobbler” crimes, in the vernacular known and used by cops:  Wobbler crimes being those that can be charged as either felonies or misdemeanors.  There are approximately 20 categories of Part II crimes, including assault-and-battery, sexual battery, receiving stolen property, vandalism, weapons offenses, and drug abuse violations (for narcotics, marijuana, and other dangerous drugs), among other less-serious crimes.

Lazar and Stevenson had to have known — if Breed and Kositsky did not — that you can’t arrive at any “in general” conclusions claiming data shows no linkages between creation of a Navigation Center and crime increases or decreases if you omit analyzing one-third (two) of the six Navigation Centers, and omit data analysis by ignoring Part II crime data and analyzing only the Part I crime data.  Who were Breed and Kositsky trying to fool?  Only Embarcadero-area neighborhoods?  All San Franciscans?

Their claim of “in general, we know” was essentially meaningless — when not simply “fake news” — given incomplete data that wasn’t fully analyzed.  Don’t forget GIGO, London.

Alternate Data Sources Not Considered

In their rush to proclaim there’s no correlation between placement of Navigation Centers and crime rates, Breed and Kositsky ignored obtaining data from another source:  9–1–1 call center data from the Department of Emergency Management.  So, I requested data from the folks at 9–1–1 for five of the Navigation Centers.

Data 9–1–1 provided revealed 786 initial calls for service were placed from the street addresses of the Navigation Centers during the periods they have been operating.  The 786 initial calls fell into 194 call type categories too long to list here.  I shoehorned them into four main types of calls:

Table 2:  Summary of 9–1–1 Calls from Navigation Centers

Table 2 illustrates:

• Fully 456 (58%) of the 786 initial 9–1–1 calls were related to Navigation Center clients experiencing medical issues, many of whom were transported by ambulance to hospitals throughout the City.  That suggests Navigation Centers may be being used as dumping grounds for very ill people to receive medical care.
  
  
• Significantly, 219 (28%) of the 786 initial calls were related to criminal activity.

Table 3:  Criminal-Activity Related 9–1–1 Calls

Table 3 illustrates:

• Fully 51 (23.3%) of the 219-criminal activity-related initial calls were related to assault-and-battery, an FBI Part II offense.  It’s not clear how many of the other calls also involved Part II offenses, which the Principled Policing Unit didn’t analyze, and Breed and Kositsky apparently didn’t study.  As director of HSH, Kositsky should know about the 9–1–1 call volume placed from his five Navigation Centers.

• The fights, verbal threats against staff or against other Navigation Center clients, and people with weapons categories accounted for 87 (40.2%) of the 219 criminal activity-related calls.

• A whopping 71 (32.4%) of the 219 calls involved trespassers, something undoubtably of interest to Embarcadero neighbors.

• Of the 219 criminal activity-related calls, 116 (53%) were placed from the Civic Center Navigation Center — one of the two Centers SFPD’s Principled Policing Unit failed to analyze for Breed and Kositsky.

• It’s doubtful that the criminal-activity related 9–1–1 calls placed from the Navigation Centers were included in, or excluded from, the Principled Policing Unit’s ?-mile radius or ¼-mile radius reports, since Part II wobbler crimes charged as felonies or misdemeanors were deliberately excluded from its analysis.

The 9–1–1 call center data goes a long way towards disproving the false assertion that “in general” there’s no correlation between placement of Navigation Centers and increased crime.

Unfortunately, when I placed the records request for 9–1–1 Call Center data, I was unaware of — and so didn’t request — 9–1–1 call data for the 1950 Mission Street Navigation Center.  That portends the 786 calls placed from Navigation Centers to 9–1–1 were substantially higher, not shown in Tables 2 and 3 above.

“Critical Incident Report” Data

Another source of data that may shed some light on the correlation between placement of Navigation Centers and increased crime rates are Critical Incident Report (CIR) forms each Navigation Center is required to submit to multiple employees in Kositsky’s department.

The existence of the CIR forms was uncovered in the 1,244-page Board of Supervisors File #19-061, involving the “Appeal of Categorical Exemption from Environmental Review” filed by neighborhood groups regarding the proposal to open a Navigation Center on the Embarcadero.  Approximately 279 CIR forms, each two- to three-pages long, were embedded in the 1,244-page Appeal packet.

The initial CIR master blank form contained a drop-down list for each Navigation Center to choose one from among seven different types of critical incidents, including death, violence, sexual assault, arrests, fires, suicide attempts, or “other emergency services.” 

(As an aside, HSH revised its master blank form in September 2019, expanding the drop-down list of seven types of incident to 17 different check boxes for expanded types of incidents.  The revised form doesn’t allow a one-click step to turn the check boxes on; instead, it takes four clicks to enable and turn on the check box, since HSH doesn’t seem to have clerical or I.T. staff who know how to make the check boxes a one-click chore, as any secretary worth her salt and advanced knowledge of Microsoft Word knows how to do.)

In response to a records request placed to HSH on September 9 for a report summarizing the aggregate number of CIR forms submitted by each Navigation Center to HSH and HSH’s Data Team, Kositsky’s staff indicated they had no responsive records.  On September 13, a follow-up records request was placed seeking a handwritten log or Excel file maintained by Kositsky’s staff listing each CIR form received.  On September 18, HSH responded, saying “The department does not hold responsive documents that can fulfill this request,” essentially saying HSH had no logbooks it could produce as responsive records, either.

Surprisingly — after initially claiming there were no responsive records for the logbooks on September 18 — rather than providing a third batch of actual CIR forms submitted by the Civic Center Navigation Center HSH had promised it would send me containing additional records, 14 working days later HSH finally located and provided on October 8 an incomplete Excel file logbook listing just 113 CIR’s Kositsky’s staff had received — a logbook it had previously denied having — and failed to provide the additional CIR forms for the Civic Center HSH had claimed were still forthcoming.  Sadly, most of the incomplete Excel file HSH provided on October 8 did not list which Navigation Centers had submitted the CIR’s, a data-collection failure if there ever was one.

On September 26, I placed records requests — one to each of the non-profit service providers operating the five Navigation Centers— asking for handwritten or Excel logbooks maintained at each Navigation Center site tracking the CIR’s they had submitted to HSH’s Data Team. 

One non-profit organization responded saying it “is not a local or State agency and our records are not public records” under San Francisco’s Sunshine Ordinance or California’s Public Records Act (CPRA).  Several non-profit providers used almost identical language, claiming my request “does not fall in the public records request requirements under San Francisco's Sunshine Ordinance, the California Public Records Act, Proposition 59, and the Brown Act,” and they can’t share any client records, or records involving direct client services information.”  Both excuses were ridiculous, because the HSH’s boilerplate form clearly states not to use client names or other Protected Health Information (PHI) identifying information.  (Despite that guidance, several of the CIR forms in the 1,244-page Appeal contained CIR’s that included either client names, or the last four digits of their social security numbers.)

With no logbooks to access, I turned to keying into Excel the 279 CIR forms embedded in the Embarcadero appeal.  The data below is from the data I keyed data in, since Kositsky’s staff also had invisible data in addition to invisible successes.

Table 4:  Summary of 279 CIR Incident Titles vs. Incident Descriptions

Table 4 illustrates:

• Just 191 (69%) of the 279 CIR’s used the seven “Type of Incident” categories in the drop-down list on the forms; non-profit staff submitting the forms creatively came up with their own incident titles, suggesting Kositsky hasn’t done a very good job of issuing guidance to the non-profit providers on data entry standards.
  
  
• Fully 167 (60%) of the 297 forms were submitted as “Other Emergency Services,” but at least 189 (67.7%) of all 279 CIR forms involved medical emergencies.  When the forms were first developed, how could Kositsky possibly have forgotten to include “Medical” as a major type of potential, or probable expected, incident?
  
  
• Similarly, non-profit provider’s staff titled just 11 (3.2% of 279) of the CIR’s forms were submitted using “Police” as the Type of Incident non-drop-down title, but the description sections show police officers were involved in at least 86 (30.8%) of all 279 incidents.  Again, when Kositsky’s staff developed the forms and came up with seven drop-down types of incidents, how could they possibly have forgotten to include “Police” as a major type of probable incident?
  
  
• CIR’s that were titled as, or included narrative descriptions of, “Violence” — against Navigation Center staff or other residents — accounted for 54 (19.4%) of the 279 incidents.
  
  
• Reading the narrative descriptions of the 279 CIR’s made for heartbreaking reading, in part because:
  
  
  • Of the 189 CIR’s describing various medical problems, at least four involved clients who had been released from the hospital within the previous 12- to 36-hours before an incident — including several people with lower-limb amputations — who were then re-admitted to an often-different hospital, suggesting they may have been dumped out of one hospital prematurely into a “lower level of care” in a Navigation Center, but were then re-hospitalized in a different hospital to avoid federal fines for premature hospital discharges resulting in rapid re-admissions.
    
    
  • At least 25 (9%) of the 279 clients involved in the CIR’s were denied service (Denial of Service) and tossed out of Navigation Centers because of their behaviors, essentially returning them to probable homelessness.
    
    
  • A total of 10 of the CIR’s reported that police reports were filed; restraining, protective, or stay-away orders were issued, or trespass orders had been signed.
    
    
  • Reading descriptions of the 54 (19.4%) reports in the 279 CIR’s was particularly gut-wrenching reading.

Table 5 below is fairly straightforward, showing 184 (66%) of the CIR’s involved transporting 184 of the residents to various hospitals in the City, again signaling their fragile medical conditions living in homeless shelters.

Table 5:  Where Were Navigation Center Clients Transported?

After learning of the existence of the CIR’s in the 1,244-page appeal filed with the Board of Supervisors on the proposed Embarcadero Navigation Center, and after Kositsky’s staff and non-profit service providers claimed they had no logbooks of the CIR’s submitted to HSH’s Data Team, I began placing records requests to Kositsky for each CIR HSH had received.

Table 6 shows that while there were 279 CIR’s included in the Embarcadero appeal, Kositsky’s staff has only provided 438 CIR’s — which is probably incomplete production of records.

Table 6:  Discrepancy in Number of CIR Forms

• Oddly, Team Kositsky didn’t produce 22 of the 58 Bryant Street CIR’s that had been included in the Embarcadero appeal, suggesting sloppy recordkeeping, or sloppy records-production provided by Kositsky’s public records staff.
  
  
• Although Kositsky’s staff provided 68 CIR’s submitted by the Civic Center Navigation Center across two separate production-of-records e-mails, his staff had assured me there were more CIR’s submitted by the Civic Center that they would produce by September 27, but has since failed to cough up for over a month later, after assuring me on October 4 additional Civic Center CIR records would be produced.  It’s now the subject of a formal Sunshine Complaint for failure to provide all records in a timely manner.
  

The discrepancies between the 279 CIR forms contained in the Embarcadero Appeal versus the 438 CIR forms Kositsky’s staff has produced as a result of records requests falls short of the 786 calls placed to the 9–1–1 Call Center.

Table 7:  Disconnect Between 9–1–1 Calls and Number of CIR Forms

• Notably, Kositsky’s staff produced only 68 CIR forms submitted by the Civic Center Navigation Center, in contrast to the 324 calls the 9–1–1 folks reported it had received from the Civic Center location since June 2016.  Shouldn’t a CIR form have been submitted for each 9–1–1 call?  What is this?  Invisible records from Kositsky’s staff?
  
  
• The 438 CIR forms Kositsky’s staff has reluctantly provided so far that are nowhere near close to the 786 calls placed to 9–1–1 from just five of the six Navigation Centers, doesn’t include the yet unknown number of 9–1–1 calls placed from the 1950 Mission Street Navigation Center.  
  
  Between the 786 calls placed to 9–1–1, and the 438 CIR forms produced to date, it seems 348 of the calls had no corresponding forms documenting one incident type, or another, that had been placed for 9–1–1 service.
  
  Why is there such as gaping discrepancy? Theoretically shouldn’t each 9–1–1 call have a corresponding CIR form? 
  
  
• Finally, could it be that the non-profit service providers have not been submitting CIR forms documenting a significant number of calls placed to 9–1–1 for emergency services?  While Kositsky claimed the HSOC needed 11 new positions at an increased cost of $4 million annually in order to primarily improve data collection (see discussion below), it appears the non-profit service providers are not doing a good job of data collection and submission of data to Kositsky’s own staff, despite the $61.5 million in contracts the service providers have been, or will be, awarded.

High Cost of Navigation Centers ($101 Million, and Growing)

The Mayor’s Office of Housing and Community Development informed the Citizens’ Bond Oversight Committee in January 2016 that 20 % to 30% units funded by the 2015 Prop. A affordable housing bond would be reserved for the homeless.  A recent Examiner article noted the homeless are receiving 23% (22 units) of the 93 senior housing units at 1296 Shotwell funded by the 2015 Bond.  How does 1% of San Francisco’s population get to acquire 30% of new or rehabilitated housing units?  This clearly adds to San Francisco’s various cost burdens dealing with the homeless.

The high costs associated with Navigation Centers isn’t fully known.  It’s at least $101 million (not 101 Dalmatians).

Multiple records requests revealed various costs associated with the Navigation Centers.  San Franciscans have spent at a minimum $101 million on Navigation Center “interventions,” with costs expected to rise as one “temporary” Center closes and replacement — and additional — temporary Centers are constructed.

Table 8:  Costs Involved with the Seven Navigation Centers

• Of note, the estimated $12 million in capital construction costs for the proposed Embarcadero Navigation Center is double the $6.3 million in construction costs for the Division Circle Navigation Center having a similar design, and double the $6 million for the Bryant Street center.  This may be in part due to changes to the City’s “emergency shelter crisis” Ordinance enacted by the Board of Supervisors on April 2, 2019, which among other things, exempts construction and services contracts to build and operate Navigation Centers from having to comply with San Francisco’s competitive bidding rules designed to award contracts to the lowest-cost bidders, and eliminates the need to issue RFP’s to solicit competitive bids.
  
  
• $38.8 million has been spent between capital construction costs, and costs to lease sites.  That nearly $40 million will essentially go up in smoke as these temporary Centers are closed.  Ignoring for a moment that escalating construction costs and simple inflation are a given increase, at least another $40 million will be pressed into service to open replacement Navigation Centers.  Couldn’t those costs be saved by opening permanent Navigation Centers?  How much more will it cost to open additional Navigation Centers in every supervisorial district?
  
  
• The $61.5 million in service provider costs to operate the Navigation Centers will climb significantly when Kositsky awards a contract to operate the Embarcadero Navigation Center.
  
  
• The costs for the service provider contract for the Embarcadero Navigation Center are expected to mushroom, because the requirement to obtain competitive bidding under an RFP was removed when Mayor Breed’s ordinance 61-19 declaring a City “emergency shelter crisis” was enacted by the Board of Supervisors on April 2, 2019.  The Emergency Ordinance reportedly removes so-called Planning Code barriers to streamline the permitting process for temporary shelter projects. It’s not known if the Planning Code itself was relaxed for construction and rehabilitation of existing buildings being re-purposed for use as Navigation Centers or other shelters.
  
  The Ordinance will apply for future contracts and applies retroactively to existing contracts for both construction of the facilities, and service provider contracts.  Eliminating requiring competitive bidding for professional and other services relating to sites and programs for people experiencing homelessness all but guarantees that it will drive costs up.

Back on February 28, 2019 the San Francisco Weekly reported Kositsky had been raked over the coals during a Board of Supervisors Public Safety and Neighborhood Services Committee hearing on the slow pace of opening additional Navigation Centers in the City.

Interestingly, Kositsky admitted during the hearing:

“When you have a site for [a] short-term and you’re going to spend millions of dollars in capital to have a site for a year or two, it doesn’t really make sense.  Navigation Centers have to be cost-effective, scalable, and sustainable.”  [Emphasis added.]

The SF Weekly then opined:

“Mayor Ed Lee’s original Navigation Center plan very much included temporary sites — because the scale of the crisis on our streets is more important than how long a center lasts.  The understanding was always that some sites being leased from developers would close after a year or two, and that new ones would open to maintain a supply of beds.”    [Emphasis added.] 

That portends that every couple of years, San Francisco may drop another $40 million to lease land and construct temporary replacement Navigation Centers.  Wouldn’t it be more sustainable to build permanent, not temporary, Navigation Centers to save $40 million here, and $40 million there?  Do we really need them decentralized in each Supervisorial district?  Is this yet another invisible success?

Apart from the costs of construction and leasing land for the temporary Navigation Centers, the costs for non-profit service providers to operate the Centers deserves a closer look, shown in Table 9.

Table 9:  Detail of Non-Profit Service Provider Contracts

Table 9 shows, among other things, that:

• For two of the Centers having roughly the same number of beds and share two-year contracts in roughly the same time period, the Bayshore Navigation Center with 128 beds has an approximate annual cost per bed of $29,336, compared to the Division Circle Navigation Center with 126 beds has an approximate annual cost per bed of $38,659, which is $9,324 more per bed that at the Bayshore facility.  Why the variance between the two Centers?
  
  
• The Bayshore Navigation Center with 128 beds has an approximate annual cost per bed of $29,336, compared to the Central Waterfront Navigation Center with just 64 beds has an approximate annual cost per bed of $41,740, which is $12,404 more per bed that at the Bayshore facility.  Again, why the variance between the two Centers?

Navigation Center Lengths-of-Stay

After Kositsky’s staff had initially denied HSH had a written policy on the length of stay for initial admission to a Navigation Center but had inexplicably provided a written policy to request length-of-stay extensions, his staff — after being pushed — coughed up an explanation of the initial time-limited stays.  The four categories are:

• Homeward Bound clients being sent back to other jurisdictions, receive one to three days at Navigation Centers, until their departure.

• Healthy Streets Program Referrals:  Unsheltered individuals referred by HSOC partner employees can stay in Navigation Centers for up to 7 days.

• HSH Outreach Referrals:  Unsheltered individuals referred by Kositsky’s HSH outreach programs staff can remain in Navigation Center placements for up to 30 days, and can receive an additional one-time 30-day extension for another 30 days in certain circumstances under HSH’s Extension Policy.

• Pathway Stays:  Unsheltered individuals having a Pathway Stay priority status as people in adult or transitional-age youth (TAY) Coordinated Entry or HSH Outreach programs, can remain in Navigation Centers until they gain housing in permanent supportive housing, long as they continue participating in the Coordinated Entry housing placement process.

Clearly this is part and parcel of the whack-a-mole problem, and is nowhere near close to Ms. Alioto’s optimism of 120-day Navigation Center stays.  The homeless are lucky if they get a seven-day stay, and luckier if they get a 30-day stay.  Contrary to Kositsky’s nonsense, one-night Navigation Center stays aren’t ‘successes,’ by any stretch of the imagination.

The object of the “whack-a-mole” game is to force individual moles back into their holes by hitting them on the head with a mallet. 

Breed’s focus on forcing the homeless into 30-day and 7-day Navigation Center and other shelter stays is not a sustainable substitute for real solutions to house the homeless.

Thoughts on the Proposed Embarcadero Navigation Center

Neighbors along the Embarcadero formed a group named “Safe Embarcadero for All” (SEFA) to mount an appeal to the Board of Supervisors over the Planning Commission’s decision to approve the Embarcadero Navigation Center. 

SEFA later turned to the Courts, filing a lawsuit in San Francisco Superior Court seeking a preliminary injunction to halt construction of the Embarcadero Navigation Center, alleging, in part, that the Port Authority and the City of San Francisco approved leasing Port land for the project in April 2019 without having obtained prior approval by the State Lands Commission, thereby violating State law.

Superior Court Judge Ethan Schulman issued a ruling on October 1 denying the preliminary injunction.  But he noted that SEFA’s claim about the failure to involve the State Lands Commission “appears to have some merit.”  

Shulman also noted SEFA “has established a reasonable likelihood of prevailing on its claim” that the Port [and, therefore, the City] was required to submit the proposed Memorandum of Understanding (MOU) contract between the Port and the City to the State Lands Commission prior to entering into the [signed] MOU.  Schulman’s Court found the City’s excuse for not seeking State Lands Commission approval “unpersuasive.”

It’s not yet known whether SEFA will appeal Schulman’s Order, in order to pursue prevailing on its claim the Lands Commission’s approval was not sought, or obtained, beforehand.  If SEFA is appealing, that may explain why HSH has not yet issued a contract to a non-profit provider to operate an Embarcadero Navigation Center.

But somewhat ironically, Schulman criticized data SEFA had submitted demonstrating year-over-year increases in assaults in a graph SEFA had submitted to the Court.  Schulman took great pains criticizing the “small sample” of data SEFA had submitted, asserting the data had no statistical significance or any probative value (as proof or evidence).  Schulman cited one case law ruling claiming that small “samples are too minuscule to demonstrate a statistically reliable pattern,” and another case that ruled “statistical evidence derived from an extremely small universe … has little predictive value and must be disregarded.”

Schulman went on to cite Police Department data summarized by the Planning Department on June 17, 2019 claiming that six months after the Division Circle Navigation Center opened, “the surrounding area saw a 17% decrease in crime, and in the two months following the opening of the Bryant Street Navigation Center the surrounding area [also] saw a 17% decrease in crime.”  Schulman’s October 1 Order was factually incorrect:  The Planning Department’s June 17 document — embedded in that 1,244-page document on the Board of Supervisors web site — had reported the Bryant Street Center saw a 14% decline in crimes, not a 17% decline as Schulman wrongly stated.

Schulman slammed SEFA for using inconclusive data and making “alarmist” claims.  But at the same time Schulman turned a blind eye to the flawed and inconclusive data SFPD had provided to the Planning Department, and to Breed and Kositsky, to justify their false claim that “in general, we know the data shows there is no link between the creation of a Navigation Center and an increase in crime in the surrounding area.”  Schulman knocked SEFA for purported inconclusive data, but accepted at face value SFPD’s own inconclusive purported data.

Here’s why:

• Schulman not only misquoted data about whether there was a 17% vs. 14% decline in crimes in areas surrounding the Bryant Street Navigation Center in the June 17 Planning Commission document, he ignored that data SFPD had provided to the Mayor by April 3 show that the Bryant Street area had seen a positive 15% increase in crime within a ?-mile radius, which was reduced to a negative 14% decrease in crime when the area under study was expanded to a ¼-mile radius, a 29-point swing.  SFPD later admitted the ¼-mile data had been double-counted in the ?-mile data, skewing the data.  Schulman didn’t opine on how doubling the minuscule ?-mile area to a ¼-mile radius could have swung the data by 29 points, from a positive increase to a negative decrease.
  
  
• Worse, Schulman ignored the massive shift in the SFPD data showing that the Central Waterfront Navigation Center swung 90 points, from a 33% decrease in the ?-mile radius to a positive 60% increase in the ¼-mile radius shown in Table 1 above.
  
  
• Schulman didn’t comment on the extremely small sample size (two months and six months, for Bryant Street and Division Circle, respectively) used in the SFPD data analysis, compared to the broader year-over-year expanded universe that SEFA had relied on.
  
  
• Schulman didn’t comment on why SFPD shrank the universe of its data analysis to just four Navigation Centers, and omitted analyzing crime data for the 1950 Mission Center and the Civic Center, which opened in 2015 and 2016, respectively — Centers that SFPD should have been able to report on easily using year-over-year data, rather than using minuscule two-month and six-month before-and-after periods for the four Navigation Centers.  Didn’t it occur to Schulman that purposely excluding 33% (one-third) of the Navigation Centers from analysis was destined to shrink the already small universe even further, worsening probative and predictive values of the City’s claim of “no correlation to crime”?
  
  
• Finally, it doesn’t seem to have occurred to Schulman that shrinking the SFPD data analysis to only FBI Part I crimes, and ignoring FBI Part II crimes, was bound to relegate SFPD’s data to being statistically unreliable.  After all, while Part I crimes do include felony Aggravated Assault crimes, Part II crimes — that SFPD didn’t analyze — include sexual battery, and assault-and-battery, along with vandalism, weapons offenses, and drug abuse violations, among others.  All of them are legitimate crimes of interest not only to SEFA, but to all San Francisco neighborhoods.  Were those missing Part II crime statistics analyses of no interest to Schulman?  Were the missing Part II stats of no probative value, Judge Schulman?
  

According to Court records, SEFA will be back in Court on Tuesday November 12 seeking to appeal Schulman’s failure to halt construction and seeking to enforce State law that the State Lands Commission must review the lease of Port land and whether the MOU should be declared null and void.

HSOC Fiasco

The Board of Supervisors Budget and Legislative Analyst (BLA) presented an audit to the Budget and Finance Committee on June 13, 2019.  The BLA audit focused on SFPD patrol staffing, SFPPD workload and overtime, and opportunities to civilianize some positions at SFPD.  During that hearing, Budget Committee chairperson Sandra Lee Fewer noted:

“Police officers [at] the Richmond [District Police] Station are saying ‘Yeah, officers get pulled for HSOC, and they don’t replace them.  We’re short officers at district stations.’  [Whenever] HSOC or anyone comes in and does a sweep, we’re just seeing more homelessness, moving it around.  …  So, a lot of neighborhoods are not seeing relief from HSOC, and the fact that it pulls officers from stations and outlying stations is a problem.”

Fewer said a mouthful on June 13!

She said a second mouthful a week later on June 21 when she chaired another Budget and Finance Committee hearing that focused in large measure on staffing of the HSOC, saying:  “I’m hearing from my own [Richmond] District [police] officers that, yeah, they get pulled off [their regular job duties] to go [be reassigned] to the HSOC.”

One source, speaking on condition of anonymity, reported that one of the District police stations is so shorthanded that a lieutenant recently instructed a group of officers during lineup before they started their shifts “not to arrest anyone,” because the Station didn’t have enough cops to manage the District if two officers got tied up processing an arrest.  Are Supervisor Fewer and the full Board of Supervisors aware this is how bad the situation has gotten?

Is this a new “get-out-of-jail-free” card:  “We don’t have enough officers to process your arrest because they’re deployed to the HSOC, so we’re not going to arrest you”?  If so, doesn’t that essentially turn the concept of “nobody is above the law” on its head?

The BLA audit was separate from a report dated March 20, 2019 by the City Controller’s Office — “Review of the Healthy Streets Operations Center.

HSOC’s Efficacy Called into Question

The HSOC’s efficacy was also called into question again, a week later when the Budget and Finance Committee held another hearing on June 21.

A high-level summary of that hearing reveals:

• Supervisor Fewer repeatedly asked how many FTE’s (full-time-equivalent employees) the HSOC has, because the Board of Supervisors hasn’t seen the HSOC’s entire budget.  MaryEllen Carroll, director of the Department of Emergency Management where the HSOC is headquartered, was unable to answer Fewer’s question about the total number of FTE’s who are assigned to the HSOC, and how the HSOC’s FTE count may have grown since January 2018.
  
  
• Fewer asked Kositsky what impact the HSOC is having on the homeless population, and whether HSOC is just duplicating efforts by the HOT team, since efficacy of the HOT has also been called into question.  Kositsky lamely responded that the HSOC is intended to coordinate efforts across multiple City departments.
  
  
• Supervisor Hillary Ronen peppered Kositsky with questions, fretting: 
  

“Maybe we’re moving people around in a better coordinated way, but we’re still just ultimately moving people from one neighborhood to the next.  And if that’s the case, I’m not  sure it makes sense to continue to add additional staff to [the HSOC] model, because I’m not sure that it’s having the intended impact on the ground.”

Kositsky responded saying they were having a hard time because staff were being swapped in and out of the HSOC, so “we’re struggling to capture data, because we don’t have any full-time staff [at the HSOC].”  As if this is only about simply capturing data and data collection, a concept president of the Board of Supervisors Norman Yee seemed to latch on to.  Kositsky stated to Supervisor Yee that “our budget request [for the 11 positions being added to support the HSOC] is primarily to improve data collection.”  Homeless people need actual housing, not better data collection.

Kositsky went on to tell Ronen the HSOC is mainly about “institutionalizing” strong relationships between City department heads in order to better serve homeless people who are on the streets.

Kositsky claimed that once high-needs clients are admitted to a place of safety, HSOC’s role is essentially done, after getting a person from the streets into a Navigation Center, or into a stabilization, bed.  He claimed once someone ends up in a Navigation Center, “we need to be able to track that person through the Navigation Center.”  That’s essentially tautological:  If the HSOC’s job is essentially done, Kositsky doesn’t need an HSOC budget increase to track the person admitted to a Navigation Center.  After all, Kositsky’s shop reportedly has its own Navigation Center Database that can track the person without assistance from the HSOC.

• For her part, Budget chairperson Fewer concluded the June 21 hearing noting:  “We just gave a nod to $4 million worth of new FTE’s to HSOC but I haven’t seen a [single] metric on impact and what [HSOC] has been doing, how well it’s been operating, and how many [employees] there are in HSOC.”  She added:  “What we’ve learned here [today] is that there is much more that we need to know about the efficacy” of the HSOC.  For all San Franciscans know, that lack of efficacy may just be more of Kositsky’s “invisible successes.”

The Controller’s report on the HSOC noted that of referrals submitted bySFPD homeless outreach officers to the HSOC, “At present, there is no means to track how many HSOC referrals result in linkage to care.”  That’s another damning indictment of the of HSOC’s lack of efficacy.

The Controller’s report noted the various HSOC “zones” were created to address distinct problems unique to each zone and primary issues to be addressed:

• The main issue to be addressed in Zone 1, the Castro, involved people sleeping in doorways.

• The main issue to be addressed in Zone 3, the Mission District, and Zone 4, Showplace Square, involved encampments of tents or improvised structures.

• Shared main issues to be addressed in both Zone 2, the Civic Center, and Zone 5, Embarcadero, involved people with “behavioral health issues (including serious mental illness)” and “feelings of safety for visitors and residents.”

• HSOC’s efforts in the Civic Center Zone 2 were to additionally focus on the issues of public drug use, drug sales, and other criminal activity; loitering; and litter and cleanliness of streets and sidewalks.

• HSOC’s efforts in the Embarcadero Zone 5 were to additionally focus on individuals who are homeless or marginally housed; litter and cleanliness of streets and sidewalks; and overnight loitering.

HSOC Staffing Costs

The City Controller’s Office reported that when the HSOC was created and activated into operation in mid-Fiscal Year 2017–2018 on January 16, 2018 — a month after former Mayor Ed Lee died on December 12, 2017 — it wasn’t created as a budget center, a cost center, or as a sub-unit of any City department in the Citywide financial systems.  Instead, it was convened as a working group to implement a combination of collaborative efforts across multiple departments using existing staff.   It had no separate budget in FY 17–18, and then-Mayor Breed didn’t submit any budget requests involving the HSOC in FY 18–19 starting on July 1, 2018. 

In both fiscal years, the HSOC had no budgeted full-time equivalent (FTE) positions.  The Controller noted HSOC staffing assignments were variable during the first two fiscal years, with departments re-assigning budgeted staff from within their existing staffing to support the HSOC.

That changed in the FY 19–20 budget process.  During the June 13 Budget and Committee hearing, the Committee’s chairperson, Supervisor Fewer, appeared exasperated at times that she couldn’t get a straight answer to how many police officers are assigned to the HSOC.  Police Chief Bill Scott eventually indicated 40 police officers are assigned to the HSOC.  Fewer didn’t get a straight answer to that question a week later on June 21, either.

Five of the departments providing collaborative support to the HSOC requested new positions in the FY 19-20 budget to support the HSOC.  Eleven new budgeted positions for the HSOC costing $4 million were approved and added to departmental budgets starting on October 1, 2019.  Of the 11 new employees, three new positions were added to the Department of Emergency Management, four new positions were added to Kositsky’s Homelessness department, two new positions were added to the Department of Public Works, one position added to the Department of Public Health, and one position was added to City Administrator Naomi Kelly’s General Services Agency.  GSA encompasses 25 departments, divisions, and programs, including the Public Works Department and the Convention Facilities unit (there’s those tourists, again).

The BLA report noted HSOC supports these five departments, additionally supports SFPD and the Mayor’s Office, and also supports an additional 11 City departments.

Fewer and the Budget and Finance Committee were able to reject a budget request from Chief Scott to add $200,000 to SFPD’s overtime budget for overtime coverage for the HSOC earmarked for conventions and events — apparently overtime to help protect San Francisco’s tourism industry.  The overtime budget increase request was rejected because SFPD’s overtime budget is already at approximately $19 million annually.

The City Controller indicated additional positions to support the HSOC may have been created in Kositsky’s HSH.

Notably missing is that the Controller’s Office reported no data for the 40-or-so police officers who may be assigned to support the HSOC, or how much that costs SFPD.  SFPD hasn’t responded yet to a records request to obtain the costs of, and the number of police officers, supporting the HSOC.

How much it costs (in dollars and employees) to temporarily divert 40 police officers to the HSOC remains unanswered. 

The Mayor’s Office and department heads from the primary and supporting City departments comprise a policy group, which convenes bi-weekly to receive operational updates, monitor progress, approval final plans, and provide major policy direction for the HSOC.  The policy group hopes to continue evolving HSOC’s operations in coming years, likely by increasing the HSOC budget.

Ancillary HSOC Staffing Costs

Reportedly, police officers assigned to the HSOC don’t perform foot patrols in the five HSOC zones.  Instead, they may be just sitting around until they are dispatched to respond to a complaint about homeless encampments or other homeless issues within the zones.  Officers assigned to the HSOC also don’t handle responding to homeless problems outside of the five zones, which means complaints about homless problems in the Richmond, Ingleside, Bayview, and Taraval police districts are handled by the resulting shortage of police officers in each District.

Any foot patrols near Navigation Centers are reportedly staffed by District stations, not officers assigned to the HSOC.  Any foot beats assigned to monitor areas around Navigation Centers comes at the expense of fewer officers in vehicles to respond to other types of crime complaints in their Districts.  It’s thought that if any police district removes two officers from a patrol vehicle to do foot beats around a Navigation Center, that’s an additional salary cost that should be added to the Navigation Center’s expenses, as Supervisor Fewer must know.

Kositsky told a packed audience during an April 23, 2019 Port Commission hearing that the City will dedicate foot beat officers to the Embarcadero Navigation Center site.  Kositsky’s, Breed’s, and Schulman’s assurances that adding foot beat patrols should eliminate neighborhood concerns is a tacit admission crime is expected to, or will, go up around Navigation Centers, not that the increased foot patrols will ensure safety.  Any foot beats will be just another Kositsky invisible success.

It’s also reported that foot beats in the Tenderloin may be also being told to avoid making arrests for “small” offenses, because an arrest ties an officer up for two hours handling paperwork.  Is this another get-out-of-jail-free card?

In reality, increasing foot patrols (“scarecrow policing”) is typically a placebo-effect attempt to calm neighbors.  The scarecrow foot beats don’t reduce crime. And the mantra, just be seen, doesn’t help.

No Bang for the Bucks Efficacy

Wikipedia reports that in 2014, the City of San Francisco spent $167 million annually on housing homeless residents.  By 2016, total spending (including housing and treatment) was believed to be $241 million annually.  However, much of this spending is focused on housing the formerly homeless, or those at risk, and not the currently homeless.

In addition to the $100 million in known costs for the Navigation Centers, the homeless also receive an outsized portion of affordable housing construction in San Francisco.  The Mayor’s Office of Housing (MOHCD) notified the Citizen’s General Obligation Bond Oversight Committee in January 2016 that MOHCD sets aside 20% to 30% of affordable housing projects for the homeless.  This portends that if the $600 million Affordable Housing Bond on the November 5 ballot is passed by voters, the homeless will receive $120 million to $180 million of any new housing units eventually constructed.

Expenditures on invisible successes from questionable efficacy of the HOT Teams, the HSOC, and the Navigation Centers are probably not going to save San Francisco’s tourism industry, or bring the American Medical Association back.  Nor will it help solve San Francisco’s homelessness problems in providing the homless stable housing.  It’s time to put an end to Kositsky’s and Breed’s invisible successes.

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.

Top