BY THE NUMBERS - In my last opinion piece, I reviewed Los Angeles’ costs of homelessness interventions; the city spends approximately $31,700 per homeless person. This week, I get more granular and look at LAHSA’s service contracts.
As a public sector performance audit manager, when evaluating an agency’s performance, I focused on outcomes. Outcomes measure the effect of a program’s operations and its progress toward cost, efficiency, and benefits goals. Let’s use everyone’s favorite municipal gripe, potholes, as an example. The number of potholes filled is a workload indicator. It doesn’t tell you much about the number filled in relation to the total number of potholes. It doesn’t tell you the cost per pothole, or how they were filled, or how long the repairs last. To get more information, you take the next step to measuring outputs. This would be the number of potholes filled, plus perhaps the cost of asphalt, labor and equipment, so you know the cost per repair. Outputs give you an idea of costs, but you still don’t know how effective the program is. The highest level of performance measurement is outcomes; given the cost and time it takes to fill a certain number of potholes; did we see a measurable decrease in the total number of potholes? Did the repair’s quality last long enough to be cost-effective, or did crews have to fix the same potholes more than once within six months?
Most public agencies do a good job reporting workload and outputs because they’re fairly easy to track. Many municipal budgets are based on those two factors: for a given amount of funding, a street department will fill a given number of potholes every year. Agencies aren’t so great at reporting outcomes, because it takes time to collect the data needed for accurate measurement, and the outcomes must be well-defined. Outcomes need to be measured over time. If you want to know how long an asphalt repair takes, you need to know where the repairs were and when they were amd, and send someone to those sites to assess their condition. But measuring outcomes can reap great benefits, because it tells you what methods work and where weak points are. Over time, you can adjust a program’s operations to maximize efficiency and obtain the best results. The term “key performance indicators” (KPI’s) is often used interchangeably for workload statistics, outputs, and outcomes, but KPI’s really should measure outcomes.
The Los Angeles Homeless Services Authority (LAHSA) has a long and troubled history of producing measurable results, despite a budget of more than $800 million and years of outreach and housing construction. Despite LAHSA’s efforts, homelessness has steadily increased for several years. According to the Authority’s narrative, the major driver behind the increase has been the lack of affordable housing, not its failure to produce results.
Since LAHSA’s programs aren’t achieving their purposes, I thought it would be informative to review the Authority’s service contracts, to see what is expected of service providers. According to LAHSA’s website, $726,203,030 (86 percent) of its $845,367,020 FY 2022-23 budget is spent on provider contracts, so most of its interventions are done by contractors. To be effective at reducing homelessness, the providers should deliver needed services to as many people as possible.
Several weeks ago, I asked LAHSA’s public records manager for copies of sample contracts. That in itself is unusual; on most local government websites you can usually find contracts via a simple search function; some even have a central contracts database. Not so LAHSA, which makes it extremely difficult to find any financial documents. I had to make a public records request, and it took weeks to receive something that should be readily available.
Eventually I received two sample service provider contracts, one for PATH and one for St. Joseph’s. Both expired last year. The contracts are lengthy, between 150 and 182 pages. About 90% of the contract is boilerplate, or standard clauses applicable to all vendors. In the contract for PATH, the payment amount is way down on page 147--$259,610. The sample I received is a "sub-contract"; in other words one of many. As far as I can tell from LAHSA’s website, I believe the total of all contracts with PATH is $32 million.
Both contracts were based on services to be delivered to LAHSA's service areas (SPA's). PATH’s contract was for SPA-4, Metro LA. The performance requirements provide compensation based on the number of people "engaged" regardless of how many hours the "engagement" takes. The contract specifies a minimum number of people to be served, although I can't find a definition of "served", so I'm assuming it’s the number contacted. From there, they're expected to get a certain number of people referred, connected, and finally housed. When you get to the bottom, the performance measure is five percent of the number "engaged". For this contract, the minimum number of people to be served is 1,127, or about 6 percent of the 17,820 homeless people in SPA-4.
If you start with 1,127 contacts, by the time you get to the most important measure, housing, they're expected to house 40 people. That's about $6,475 per person—see the table below:
After I calculated the outcome costs, I reviewed the rest of the contract. I could find no other mention of performance measures or expected outcomes. I was, however, able to form some opinions based on how providers are compensated and how the contract’s specifications are written.
First, I was struck by how long and detailed the contract is. The specifications tell providers the services to be provided, how they will be delivered, and how they should be tracked. While this may seem fair because the requirements are the same regardless of contractor, it actually stifles innovation or creativity. The specifications also reflect LAHSA’s “Housing First” priorities; the service model is already set in stone, so responding organizations have no opportunity to propose better or more efficient service models.
By their nature, the contract’s requirements and length limit who can respond. A responding agency must submit personnel policies like termination agreements, TB testing forms, etc. I've seen these types of contracts before; they are a way of appearing to make the proposal process competitive, while actually limiting it to a chosen vendor, or to justify sole-sourcing a contract, by loading it with requirements only the vendor of choice can meet. Smaller, more nimble organizations, like neighborhood-based agencies, or those with different service delivery models won't qualify. That would explain while we always see large, corporate-style non-profits like PATH and St. Joseph’s grab the majority of the contracts.
Finally, the contract offers no incentives to get people housed. It’s an incentive to run up contact hours. The basis of compensation is “reimbursement for allowable costs” rather than results. Despite the inclusion of KPI’s in the contract, PATH is paid for its activities, not its results. Reading through the contract, it’s no wonder the 2023 RAND homelessness survey said 44% of the unsheltered have never been approached with housing offers. From the point of view of the service providers, why bother when you can run up your hours on a limited number of contacts?
The sample contracts are a microcosm of the much larger problem with regional homeless response. Intervention agencies, whether government or non-profit, focus on process rather than outcomes. The assumption is that by following a certain process, results will follow. But LAHSA and other agencies are following failed processes based on a failed model, No Barrier Housing First, so homelessness keeps increasing.
LAHSA avoids true outcome measurements because they would show the failure of its program models. Perhaps the best example of this denial is a line on LAHSA’s own website, claiming homelessness continues to increase due to a shortage of affordable housing. It offers no statistical evidence to support this claim. It offers no introspective statements that suggest its current service models aren’t working, nor does it discuss other causes like untreated mental illness and epidemic opiate abuse. It’s classic avoidant behavior; you can’t fail at what you don’t measure.
Housing agency leaders claim critics simply don’t understand how complex the problem is. They can come up with any number of reasons why they can’t track people they shelter or why the construction of permanent housing takes twice as long and costs twice as much as the market rate. As an experienced auditor, I heard the “you just don’t understand” excuse more often than I care to remember. It’s a sure sign there are serious problems in an organization. Defensiveness is the hallmark of a failing organization.
Poor management is poor management regardless of the business you’re in. Contractors. employees and managers who aren’t accountable for their performance rarely excel at their jobs. Rigid, silo-based thinking chokes innovation and self-assessment. And the homeless pay the pay the price, often with their lives.
(Tim Campbell is a regular contributor to CityWatchLA.com and a resident of Westchester who spent a career in public service and managed a municipal performance audit program. He focuses on outcomes instead of process.)