Four factors you need to consider when launching a SIB - In conversation with James Ronicle, author of the LOUD model

James Ronicle, Associate Director of Ecorys UK and GO Lab Fellow of Practice is a co-author of the LOUD model report, a research collaboration between Ecorys UK and the Policy and Innovation Research Unit (PIRU) to explore the crucial elements that are needed to launch a SIB. The GO Lab interviewed James, asking critical questions about the model and looking to the future. 

The LOUD model 

To produce the LOUD model, research was completed on 25 SIBs in development. Of this 25, five were launched and 20 were not. The aim was to was to examine what was present in those that launched and was absent in the others. 

The research found four factors that seem to determine whether a SIB is launched, creating the LOUD model: collective Leadership, clear Outcomes, shared Understanding and Data. These are outlined in the figure below, you can also read the full report here and James’ blog post for Big Society Capital here. 

Interview with James Ronicle 

1. Why should commissioners consider the LOUD model before launching a SIB?

My view is that the LOUD model explains how to get a SIB off the ground by reducing it down to 4 pivotal elements. We had a relatively large evidence base and we found almost all of the 5 that launched had the 4 elements and the 20 were lacking one or more of the elements. There is so much to think about to get a SIB off the ground and commissioners can get distracted by other aspects. Thus, the LOUD model is a useful framework to get it right and it is a way to help commissioners learn from the lessons of others. 

2. SIBs are used to address many different social issues and across different regions and contexts; how does this model accommodate those differences? I.e. Will the model still apply for an edge of care SIB in a rural region and a rough sleeping SIB in the centre of London?

Yes, the model will still work, but with some nuances. The SIBs we looked at were broad, they covered social issues such as health, education and crime. Some were commissioner led, others were provider led, and there were some across rural and urban areas. It became clear that the 4 elements of the LOUD model were necessary regardless of the context.

Having said that, some of the factors were more or less relevant depending on the SIB in question. Leadership and Shared Understanding are relational and are crucial to get buy-in for the SIB. On the other side, the Data and Outcomes are more technical aspects and consider the feasibility of the SIB. It may be the case that the more you have of relational, the less you need of technical.  

For example, for a SIB that seeks to support children on the edge of care, there is often plenty of data available from local authorities and the outcomes are easier to measure as they are binary, the children are either in care or out of care. This means that there is a greater evidence base which will support the business case and make it easier to get buy in. Whilst the relational aspects to the launch are crucial, you are less reliant on this when you have a stronger evidence base. 

On the contrary, when you have less data and less clear outcomes this will offer a weaker evidence base. The case will be higher risk to the investors and you would need excellent leadership and shared understanding to launch the SIB. 

3. What are the main challenges in applying the LOUD model and how would commissioners overcome these?

In many ways there are challenges to all of the elements. As we found, if you don’t have one of the elements you are going to struggle to get the SIB off the ground. However, some challenges are more surmountable than others. 

For example, if you don’t have enough data you can draw on data from elsewhere. If you need data on the costs of running something, you can research what the costs have been elsewhere. This does rely on you being comfortable that you are taking a higher degree of risk. 

If you don’t have the outcomes then you are in a bit of a bind, some projects just won’t work. If you are comfortable using proxy measures this is a way round the challenge. For example, if you are doing a SIB for young people not in employment, education or training (NEET) the likelihood that they will get into employment in the SIB lifetime is low. You can use proxy outcomes, such as qualifications, with the faith that these will move to outcomes. This relies on you having an appetite for risk as essentially you are not getting the outcomes you want (and for some this risk was too high, leading to some SIBs not launching). 

4. What are the next steps after the LOUD model, where should research go?

The conclusion of the LOUD model report articulates this. It highlights that many SIBs were not developed because there was a lack of information needed to build the business case. We need to change this so local stakeholders can access information and have what they need to fulfil the elements. This can happen in a number of ways:

The first is for the information within current SIBs to become accessible so others can replicate them. This already exists to an extent with the Centre for SIBs and the Government Outcomes Lab (GO Lab), but there have been some restrictions based on confidentiality so good practice has not always been shared. I also think there has been a greater focus on some SIBs, and there are others that those outside the programme still don’t know very much about. We would therefore encourage greater transparency and knowledge sharing.  

The second is for stronger central repositories of information, so that national information can be used when it is missing locally. Some of this already exists - these are: the Outcomes Matrix  developed by Big Society Capital; the Unit Cost Database by New Economy; and the What Works Network. However, as some of the challenge is the lack of knowledge and information, we would encourage data to be more publicly available, and to create more avenues and forums for people to share and discuss their projects.