Housing complexity: what has systems thinking taught us about homelessness, and have we learned the lessons yet? 

Chair: Paul Morrish, Chief Executive, LandAid

Professor Alex Marsh, University of Bristol and UK Collaborative Centre for Housing Evidence (CaCHE)

Dr Ligia Teixeira, Centre for Homelessness Impact

Drawing on his work with UK Collaborative Centre for Housing Evidence (CaCHE), Professor Alex Marsh explains why the relationship between the private rented market; the benefits system; social housing and the statutory housing duties are a “complex system”.

What are the unintended consequences and negative feedback loops that derail progress in reducing homelessness? In this session, we are reminded of the important lessons from systems thinking applied to housing, and Dr Ligia Teixeira explains why tackling homelessness is more complex than rocket science.

Alex Marsh on why understanding “purpose” is key

A system is a complex whole that depends on its parts and the interaction between those parts to function. The system has parts; the interaction or interconnection between the parts, and it has boundaries.

Thinking about boundaries is part of thinking about what’s inside the system that we’re trying to analyse, and understanding what sits in the environment. But where we draw the boundary is an art rather than a science. We have a choice that can help us think about alternative or innovative solutions.

Systems thinking is about stepping back and focusing on the big picture, not just the component parts. My slides show a selection of key concepts from systems thinking, and today I’m going to focus on three: purpose; causal loops and feedback, and reframing.


“The purpose of a system is what it does” (Stafford Beer)

This quote by one of the originators of cybernetics is one branch of systems thinking. You might think that is pretty obvious.

But when I teach policy students about systems thinking we spend a lot of time thinking about what we’d like the system to do, and what we’re claiming it is doing. And the narrative about what it’s supposed to be doing and what it actually does might not align.

An example from housing would be “right to buy”. What is the right to buy the? We’ve got rhetoric about a property-owning democracy, expanding a stake in the system.

It largely took houses that were being rented at social rent and converts them into privately rented houses at market rents. That wasn’t what the system or the policy was ever intended to do. So that’s why it is important to think about purposes.

Feedback loops

Feedback loops are the quintessential systems idea, and can be positive and negative. We can think of them as supporting or opposing.

I put on my slide a simple version of a feedback loop we might encounter in housing. The example is from work on private rented sector compliance and enforcement to improve standards.

Improvement action affects the system in at least two different ways. It affects the system directly in the sense that we act against particular landlords, and that raises standards in those properties.

But you’ve also got a separate loop, which is a “demonstration effect” that educates more landlords. How does that shape how local landlords respond to action in the sector? We can have a conversation about how powerful some of these loops might be and how they affect system behaviour.

We can add in licensing; codes of practice; tenants’ willingness to assert rights because they’re given stronger legal protections. So we can build up a picture of how this system is functioning. We’ve got balancing loops, we’ve got reinforcing. It’s all driven by that core loop of how we act to drive up standards.


My final topic is reframing so we can think about reframing in different ways. We can think about reframing as the scope of the system to redraw the boundary – what’s inside the interest and what’s outside among things we’re not trying to influence.

We frame the issue the system is trying to address. Can we get away from sort of entrenched views about what the nature of the problem is or what the nature of the solution is to something which is shared and better understood?

Then maybe we can articulate the goals of the system slightly differently. So one example is from the private rented sector enforcement of standards again. A narrow framing would be focusing on health officers trying to ensure that the housing stock is legally compliant with minimum standards.

On the other hand, we can look at a broad framing of what we’re trying to do with this system to allow residents to thrive, and maximize residents’ wellbeing. How does housing enforcement sit within that? Well, to address that objective, we may need to include several branches of a local authority or other organisations.

The broader framing is trying to recognize interconnections, and diverse contributions, contextualizing the role of environmental health officers within that broader mission.

That invites different ways of thinking about system leadership and how we get everybody to pull together in the same direction. Are we thinking about how we might achieve higher standards in the right way?

Think about the norms that govern the sector and how people understand acceptable ways of providing private rented accommodation. Is it about providing a right to a home? Is it about preserving the asset owner’s right to a return? How do we strike the balance between those two and always strike it in the right place so we address those profound issues around the edges?

Reframing and thinking differently is a way of pointing to non-obvious interventions.

Ligia Teixeira warns about responding only to the tip of the iceberg

I’m going to tell you why it is so important to be thinking in terms of systems when it comes to homelessness. I grew up in the Northeast of Portugal, one of the poorest regions of the country but it wasn’t until I came to London that I first came face to face with destitution.

On my walk to work, I started meeting people sleeping rough. Through them, I realized how complex homelessness is and how many people are not sleeping on the streets, but hidden from sight, whether in shelters, temporary accommodation, often illegally evicted by landlords etc. Street homelessness is just the tip of the iceberg.

Addressing the whole system

I’m telling you this because it’s important to be wary of addressing one part of the system without addressing the whole system. In about 2005, Salt Lake City invested in a huge program to address chronic homelessness and in Housing First in particular. Over ten years they saw reductions in the order of 90%.

So it was a great achievement. But the headlines hid a more complex truth that other types of homelessness continued to rise during that same period.

This is not just a cautionary tale for us in the UK. We are seeing trends that are still looking quite favourable in comparison to past years.

But the number of people in temporary accommodation has continued to rise or remains stubbornly high. So what it tells us is the fact that homelessness is complex. Homelessness, in fact, is harder than rocket science.

Unintended consequences

It’s a complex problem where even if you solve issues in one part of the system, there may be unintended consequences in another part. It is time to embrace that complexity and not be naive about the scale of the challenge either.

When I set up the Centre for Homelessness Impact, we started developing the type of diagram Alex was talking about to develop a homelessness prevention map. You see the complexity behind something like addressing homelessness at the prevention level and landlord behaviour, the housing market, and welfare all play a huge role in this.

We are in a great place when it comes to ending homelessness. Homelessness is one of the few good stories of the pandemic. However, we are entering a new phase, and we know the scale of the challenges with the cost of living crisis. Not only should we be proud of what we’ve achieved already, but we should want more.

We know from the work of the Centre that three things need to happen. We need to avoid just responding to the tip of the iceberg. We must address the root causes, and we must drive prevention upstream.

Improving data

The white paper on the private rented sector includes improving the data available, whether about rogue landlords or moving faster when it comes to vacant units. We need to understand what works for whom and where. Not all parts of the country are the same, and not all landlords are the same.

We must make sure that we understand how we can affect landlord behaviour and understand what the markets do in different parts of the country. We must take a bird’s eye view of homelessness and the housing system, so we can zoom in and out.

At the Centre for Homelessness, our mission is very much to support leaders like you to do the best work possible based on the best data and evidence available at any one time. Obviously, neither of them is ever perfect.

We believe that it is possible to end homelessness for good, but it simply will not happen unless we can act on those three fronts and we can think about systems.

You can see Alex and Ligia’s presentation slides here

We asked the audience to vote on the most important take aways from the session via Slido. Here are the results.

Housing complexity Slido poll chart