Studying the Costs of Homelessness Midstream Lessons from a National Cost Study

1
Studying the Costs of HomelessnessMidstream
Lessons from a National Cost Study

• Jill Khadduri
• National Alliance to End Homelessness
• Annual Conference, July 2007

2
Why Study Costs? Several Possible Purposes

• Show costs of homelessness to mainstream systems
• Net cost (or savings) from ending homelessness
• Potential for cost offsets to particular systems
• Show societal costs of homelessness
• Economic loss to businesses, neighborhoods
• Economic loss from loss of earnings potential
• Compare efficiency of different programs (or
  approaches) to serving similar homeless people
• Compare costs of a program (or approach) to its
  outcomes cost/effectiveness study

3
Need to decide (based on purpose of study)

• Costs to whom?
• A single funder?
• Multiple funders?
• Homeless people themselves?
• Relatives, friends, neighbors?
• Costs of what?
• A single program?
• An approach multiple programs that operate at
  the same time or sequentially?

4
Abt Study of Costs of Homelessness for HUD

• Purposes
• Compare costs of different approaches to serving
  homeless people (individuals and families)
• Measure costs to mainstream systems before,
  during, and after homelessness
• Not a cost effectiveness studynot measuring
  outcomes
• Not a study of societal economic costs of
  homelessness
• Is developing methods that can be used in studies
  with a variety of purposes

5
Abt Study Measures Costs of Approaches, Not
Individual Programs

• Uses HMIS data to find pathways clients take
  through the homeless services system and to count
  their units of service
• Measures costs of all programs for homeless
  people used during the pathway by multiplying
  units of service (from HMIS) by unit costs (from
  program budgets)
• Requires a well-populated HMIS for the study
  period
• Most HMIS cannot do this for 2004 or 2005
• But HMIS are building fast

6
Example for a Particular Client
Emergency Shelter for Singles 2/night
Central Intake for Singles 1/intake
Transitional Housing for Singles 3/night
PSH for Singles 4/night

Central Intake 1 intake 1
Emergency 30 nights 30 x 2 60
Transitional 90 nights 90 x 3 180
Total client costs 241



7
Pathway for Each Client is based on a Typology of
Programs

• Typology is needed
• So can infer costs of other, similar programs
  from costs of programs for which data collection
  is possible
• So can describe the pathway in way that makes
  sense to policy audience
• Goes beyond emergency, transitional, permanent
  supportivee.g., separate categories for
  scattered-site, shared rooms, private apartments
  and/or different intensity of services
• Each typology is tailored to the homeless
  services system in the study community

8
Unit Costs of Homeless Programs

• For residential programs, unit costs include
• Costs of operating the housing or shelter
• Cost of acquiring/developing the housing or
  shelter
• Costs of services provided by the program
• Overhead or administrative costs
• For residential programs, the unit of service is
  a bed night or unit night
• For services only programs, unit costs vary by
  type of program e.g., cost per day, cost per
  service encounter

9
Cost Collection Instruments for Homeless Programs

• Interviews for information needed to understand
  costs which clients? what services? what
  partnerships? what type of housing?
• Cost collection spreadsheets to record
  information from program financial statements and
  ensure all costs are included e.g.,
• Services provided by private funding
• In-kind contributions and donated labor
• All overhead costs, not just administrative costs
  chargeable to particular programs

10
Capital costs of residential programs

• Cost collection approaches for residential costs
  that do not appear on annual financial statements
  and budgets
• One-time acquisition, rehab, construction costs
• Development pro formas (when they exist)
• Less formal information from interviews, file
  cabinets
• Value of donated space
• Challenging to collect
• May not be needed if purpose of study does not
  require costs to all funders
• But governments may donate spaceis this a cost?

11
Costs of Services

• Which services costs to include as costs of
  residential program?
• Is it part of the residential program or a
  mainstream service?
• Do people get it because they are clients of this
  program?
• How to measure costs of services for homeless
  people that are not linked to a residential
  program?
• Utilization from program records or from HMIS
• Need to learn programs approach to defining a
  unit of utilization (e.g., an appointment, a
  period of service) and measuring its cost.

12
Cost Collection Approaches for Mainstream

• Basic approach is to match HMIS client
  information to collection systems of mainstream
  programs
• Objective is to apply unit costs to the period
  before, during, and after homelessness.
• How to do this depends on the mainstream datahow
  the program defines a unit of service and
  measures its cost
• Requires data sharing agreements to protect
  privacy and security of client information
• Takes time
• Takes political willinterest in the study
• May be easier for a local study than for national
  researchers

13
(No Transcript)