In Systems Thinking feedback loops are a big deal.
In my last blog I covered “positive” or “reinforcing” feedback loops, what they are and how they crop up in the FE system all over the place.
In this blog I will cover so-called “negative feedback loops”. Negative in this case doesn’t mean “bad” (and remember that “feedback” doesn’t mean giving your opinion; “feedback” in systems thinking is where an output of a system or process is fed back into the input flow of that same process, altering it one way or another.)
Negative feedback is where the output feeds back in and has the opposite effect on the system, so reducing the output that was previously occurring. In other words, it acts as a correction or balancing mechanism. This is why I think it’s more useful to talk about “balancing feedback loops”, because the effect can be either good or bad, depending on what you want your system to do.
A classic example of a balancing feedback loop often used to illustrate the concept is a thermostat:
If there is an odd number of “-ve” signs this means the system is acting to correct or rebalance its own outputs.
A more complex example occurs often in nature, and is familiar to biologists. It’s when a living system acts with checks and balances to keep itself optimised in response to a changing external enviornment, and it’s called homeostasis:
Note that in this case the balancing loop part of the picture is everything except the external input, which is not part of the loop.
I think there are many examples of balancing feedback loops to be found in the Further Education system. Here are a few important ones that have been mentioned to me so far in the interviews I have been conducting as part of my project:
In this example above, we see that education and training providers sometimes report they risk losing a lot of the talent, insight and experience that were critical to their Outstanding inspection grade once it has been awarded. This in turn means they are less likely to attain it again next time [NB interesting interaction here with the previous Ofsted policy of deprioritizing Grade 1 winners for future inspections.]
In this example we see the effect of lagged funding for 16–18 participation, which introduces temporary stress and makes it challenging to over-recruit when finances are tight. Having been a senior official in the DfE when this funding policy was introduced, I can attest to the fact that this was indeed an understood and intended consequence within the Funding Body at the time. The rationale was that they did not want a run-away reinforcing feedback loop where successful recruiters got more and more funding and squeezed everyone else out of the market in a boom-and-bust type scenario; instead they wanted a more stabilising effect introduced by the funding regime, so that providers would only go for cautious, sustainable recruitment growth.
As a policy-maker I admired this thinking; as a college Governor in a very financially challenged college I’ve been less enamoured of it. This is not (just) a wry comment – there is a fundamental meta-issue here about which perspectives and understandings are involved in the system-forming process, and which are not (or not much).
This example above is a classic case of unintended consequences. DfE never wants other publicly funded bodies – be they schools, colleges, universities or sector improvement bodies – to have any more reserves than they strictly need. This is because of the Treasury golden rule that money that is sitting in someone’s bank account should always – given the choice – be sitting in the Treasury instead, because that is the cheapest option from an overall Government Accounting point of view. This is true (because we have a national debt and a deficit, so in effect any money that Govt pays a public body and which sits in the bank is unused money that HMT is having to pay interest on to its creditors).
But the unfortunate real-world effect of this sensible macro policy is that publicly funded bodies are incentivised to fully “spend up”, whether that represents good value for money for the public or not, or else they will see their funding flows tighten in future and it will harder for them to deliver the same public value, far less invest in their capacity to deliver.
This example looks simple – it’s the bit of the system that focuses on quality improvement (the ETF) playing its role successfully, and effecting balancing or corrective action.
But there are hidden risks here too. If the ETF does successfully help the sector to the extent that providers no longer seek assistance, the ETF is doing itself out of a role; does it have the selflessness to behave like this?
Conversely, perhaps the ETF only solves the symptoms of the providers’ problems, not the root causes. If that were true, then it could actually be doing a disservice to the sector as it would be constantly suppressing the sector’s need to develop its capacity to self-improve.
This is explained powerfully by Donella Meadows in her book “Thinking in Systems”:
“The trap is formed if the intervention, whether by active destruction or simple neglect, undermines the original capacity of the system to maintain itself. If that capability atrophies, then more of the intervention is needed to achieve the desired effect. That weakens the capability of the original system still more. The intervenor picks up the slack. And so forth”
This is understood by the ETF, and is why their strategy says not “improve the sector” but rather “support the sector to self-improve”.
As you can see, some of these balancing feedback loops are really helpful and beneficial, and are ways that the system currently self-corrects where there are problems. Others are clearly problematic, and act against successful behaviours and beneficial outcomes. But most are neither wholly good nor wholly bad, they are just real-life features of the system that we need to understand and work with.
To take the first example, a skillful and charismatic CEO can effectively stop good staff leaving their college in the wake of an Ofsted Outstanding grade – they may see it as a risk to be mitigated for the sake of their college and their community. But if we take a wider system approach we could say that it is better for the system as a whole if there is a pattern of staff taking flight from strong effective colleges and taking that knowledge and expertise to other places where it is needed.
The phenomenon might be shown instead like this:
But this is not a loop. So why might a College CEO be incentivised to support middle and senior leaders taking their expertise elsewhere after achieving Grade 1? Only if there was a feedback loop that met some goal of theirs. Assuming they are not so system-oriented that they want the system to thrive at the expense of their own students (why should they?) then there needs to be some closure to the loop. Something that guarantees that each college benefits when the system overall is stronger. Like this, in the form of the Reinforcing Feedback Loops we saw in blog four.
Is this true at the moment?
I would argue not; the consensus amongst my interviewees (those who have expressed a view on this so far) has been that success at the moment is institutional and comes from the strength of that one institution – the quality of its leadership and governance, the abilities, dedication, skills and experience of its staff, their ability to leverage support from local and national stakeholders and so on. There is no sense that the success of one leads to the success of all, or that individual success is derived at all from collective success. But this would need to be true in order for institutional leaders to feel really positive about letting excellent staff go for the benefit of the whole system.
This begins to hint at what some system-level solutions might need to look like. But that is for a later blog.
David Russell
Executive in Residence at Oxford Saïd Business School
Education and Training Foundation