25th March, 2011

How to do a cost-benefit analysis

Today marks the launch of my evaluation of Depaul UK’s homelessness prevention service, Homelessness Prevention: Can we afford not to? In the report I outline the typical childhood experienced by young homeless people; one of disadvantage, abuse and disruption.  I also highlight research that indicates the impacts of homelessness on a young person. As you might expect, their physical and mental health, education and economic potential are all hindered by their experience of being homeless.  However, the cost of homelessness is not limited to the young person themselves.  The cost to the public purse is significant. And it’s about to get much bigger due to a recent House of Lords ruling.

Depaul UK knew that their homelessness prevention project, Reconnect, is successful at preventing homelessness, increasing educational engagement and preventing family breakdown.  However, the pressure to justify their funding was increasing, so they asked me to demonstrate the savings that can be made.  My cost-benefit analysis showed that the service costs an average of £363 per person and delivers an average saving of £9,493 per person. Impressive.  So how is it done?

You will need:

  • Total costs of the project
  • Number of beneficiaries
  • A table of the intended outcomes (keep it simple.  Although Reconnect increases school engagement and prevents family breakdown, we stuck to its primary aim, to reduce youth homelessness, in order to avoid over-complicating the analysis)
  • The number of users who achieved each outcome
  • Direct costs associated of not achieving the intended outcome
From this you can calculate the following:
  • Cost per user
  • Cost per successful outcome achieved
  • Savings per successful outcome achieved
  • Average savings per case

You can see the maths in my report.  Depending on the strength of your outcomes monitoring and the data available on the cost of not achieving them, this process can be relatively straightforward or a major headache. It’s going to take time; it’s not something you’ll be able to knock out in a day.

The Depaul UK evaluation was made more complicated by the fact that costs to local authorities of supporting and accommodating young homeless people are increasing exponentially, but there is no definitive data on what proportion of young people this ruling will apply to.  We therefore applied an estimation of how many clients this ruling would apply to if homelessness was not prevented, based on the best available data at the time.

The report is out today so we wait with bated breath to hear how it has been received.  It’s being launched in front of 170 youth homelessness professionals so I guess feedback will be fairly swift.

Here are a few tips I learnt from the process that I hope you’ll find helpful:
  1. Get your outcomes monitoring right and make sure the data is robust
  2. Get someone to check it
  3. Make sure that your interpretation of the costs of failing to achieve the outcome (in Depaul UK’s case, failing to prevent homelessness) is robust
  4. Get someone to check it
  5. Make sure your calculations are recorded in detail in a pretty spreadsheet, even if you don’t intend to publish this much detail
  6. Get someone to check it

I’m not keen on blogs that only try to sell you something so it is with hesitation that I suggest you get in touch if you need help with your evaluation or cost-benefit analysis.  But hey, we all need a little help sometimes.


Web design by FLOQ Design