This fourth article in the Defending the NDIS series is an insider’s look at what happened when the government took a wrecking ball to the National Disability Insurance Scheme’s operating model.
The NDIS is a case study in the crucifying complexity of policy and service delivery: it is also a case study in the failings of digital government services.
The first article in this Defending the NDIS series dealt with the complexity of the system. The second article exposed the “word salad” of assistive technology. And the third article exposed the actuarial model that has help to destroy the scheme.
In the 2014-2015 Budget, the Abbott government cut National Disability Insurance Agency (NDIA) staff numbers from 10,000 to 3,000.
But unlike the Productivity Commission’s detailed analysis determining the required 10,000 staff figure, the 3,000 number was random, not based on any evidence or analysis. Go look for the analysis on this figure: there is none.
Much has been written about this. But this needs to be understood as far more than a budget line item of staff numbers.
In contrast to the contested wild numbers on sustainability that have been a feature of the brutal NDIS narrative since the scheme’s inception, one thing is for certain.
And that is the action by the Abbott government to cap the NDIS staff numbers sent a wrecking ball through the NDIS operating model.
The shock waves from that decision I believe have backwashed onto the budgets of state and territory governments. And of course, it has directly impacted millions of Australians around the country.
At the time, I was on the inside and here is what happened.
During this period, the National Disability Insurance Agency had a staff of about 850, supporting about 20,000 participants on the NDIS. Also, of note at the time, 16.1 per cent of staff identified as having a disability. This figure would fall by more than a third over subsequent years.
The NDIA was preparing the gargantuan effort of scaling up to manage the intake of hundreds of thousands of people with a disability from various state and territory disability systems.
This intake exercise was about far more than a theoretical actuarial exercise of the transfer of data from state and territory systems to the NDIS. There was much data that didn’t exist and needed to be created.
Whilst this historic intake was happening, in parallel the NDIS service delivery operating model (SDOM) was being established. This would map all policies, legislation, rules, procedures, and processes: traceability end-to-end.
As the Head of the Technology Authority, I worked with other executives on the formal governance of the design and co-design of the service delivery operating model. The SDOM shaped everything.
At the time, the NDIA was going through a Budget process which achieved $143 million in funding for the NDIS Full Scheme ICT based on the SDOM. This was a Budget process I led.
The point to understand, is that this was not an organisation operating in a steady state, where known functions could be contracted out on the basis of risk and business analysis.
This was a new national organisation doing something that had not been done before in Australia, the operations of which affected not only the Federal Budget, but also the budgets of all state and territory governments as responsibility for disability and the associated funding of billions of dollars was transferred.
That is a level of extraordinary risk that no politician spoke about when they took the wrecking ball to the staff numbers.
And putting it bluntly, nor was this something that banks or insurance companies had ever done. So, the narrative that the NDIS is just like a bank or insurance company and should operate that way is fantastical.
The NDIA needed – and still needs – to be understood as the axis of a complex servicing ecosystem. Change any part without evidence or risk assessment at peril. And that’s what happened.
When the government made the decision to cut the staff numbers from 10,000 to 3,000 – noting that at the time the actual staff on board was only 850 – there was an existential crisis.
This enforced outsourcing broke the formative operating model. It was not simply the case of breaking bits up for the market to do. Of course outsourcing itself bears an administrative cost and imposes a significant resource burden in oversight and contract governance.
Functions which were to be conducted by staff with delegations could not be outsourced: delegation for decision making remained within the agency. Processes as well as functions needed to be broken up. Outsourced functions would be information gatherers and information givers, ruled by KPIs. These KPIs would become problematic.
The political wrecking ball that forced the outsourcing of functions with such extreme political time pressure, with so many unknowns and at the same time as the NDIS technology systems were being built by DHS, was extraordinarily asinine.
The result was defective systems and processes, with workarounds and off-system processes. Systemic problems of lost documents resulting in the denial of access and cycles of appeals. Daisy chains of mailboxes, process dead-ends, off-system databases, and the passing around of spreadsheets of participant data to outsourced providers.
This is still the case today.
The service delivery operating model and the reality of front-line service delivery needed to be understood in detail.
During the intake phase, I would visit the call centre operated by DHS, and double head-set listen to the outbound cold calls to people transitioning to the NDIS from state and territory disability systems. I know call centre operations.
Some of these calls would go for an hour or more. People did not know what the NDIA/NDIS was. People wanted to go to the toilet. People didn’t understand the concepts. People wanted to ask questions over and over again. People wanted to talk about their child or partner who was also disabled. A call back that would take another hour would be needed.
Of course call centres do not help people with cognitive impairment, people with intellectual disability, people who are non-verbal, and a great many other people with physical and sensory impairments.
This is the space where theory is no match for the rawness of humanity. This is the messy human reality of service delivery. KPIs do not and cannot tell this story.
At the same time that the wrecked operating model was causing intake delays, decision delays, decision errors, and horrific bureaucratic processing – a massive outsourced market flourished.
The NDIS outsourced call centre contract with Serco from March 2018 to August 2022 is an eye watering $158,844,796 with more than 400 contracted staff.
And that is in addition to the recent $400 million in Local Area Coordinator contracts for NSW alone, and the $32m explosion in legal fees over the past eight months spent fighting participants.
The reality of course, is that hundreds of millions of dollars in perpetuity cannot compensate for or fix a broken operating model.
But I have not only been an insider. With family I have been subject to this system, interacting with outsourced Local Area Coordinator (LAC) organisations for plan processes.
And my observation is this: the KPI driven outsourced arrangements exude an aura of an olde financial planner model. Perhaps not surprising given the push for the NDIS to operate like a bank or insurance company. Follow the KPIs to see what happens.
These organisations are contracted with KPIs. But just how do the KPIs of LAC organisations flow to individual contracted planners? If the organisation is contracted to process a certain number of plan reviews with budget ‘targets’, is there the potential for a conflict of interest? There is certainly a power and information imbalance.
Robo-plans are then part of the planner’s toolkit, enabling automated plan processing at scale.
From a transparency and ethics perspective, there is a question as to whether people should know how individual LAC planners are incented, as happens with financial planners.
And any review of the NDIS should examine how this unregulated quasi financial planner type function has actually affected people. A bit like the behaviours examined by the Banking Royal Commission.
Rebuilding the NDIS requires reinstituting the 10,000 APS workforce operating model and investing in human capability development at all levels. Bringing back inhouse all the functions that were forcibly outsourced, including call centre operations.
So that all people, including people with disability, can thrive in a career of choice across the full reach of the operating model, from the call centre to the CEO office.
This is not about KPIs, but ethics and pride in service. Everything else follows.
With such a broken operating model floundering under the complexity of an actuarial model based on fictions, the next article looks at the risks of algorithms and asks if ‘do no harm’ is even possible.
Marie Johnson is the CEO of the Centre for Digital Business. She is a global award-winning digital authority and advocate for the humanitarian application of AI. Her experience encompasses the public and private sector experience in Australia and internationally, including leading Microsoft’s Worldwide Public Services and eGovernment industry based in Seattle.
Marie was Head of the Technology Authority for the National Disability Insurance Scheme responsible for the technology business case, co-design, and the creation of Nadia. For many years, Marie was the Department of Human Services Chief Technology Architect, with responsibilities including the architecture and technology business cases bringing together the massive systems of Centrelink, Medicare Australia, and the Child Support Agency.
You can follow her on Twitter at @mariehjohnson or visit marie-johnson.com.
Do you know more? Contact James Riley via Email.