Andrew Little recently suggested that ACC's current funding model is more suitable for private insurance, and suggested that a future Labour government might change it - but in doing so failed to get to the heart of the issue.

In a recent interview on Morning Report (reported here), Andrew Little suggested that the “fully funded” model currently used to fund ACC was fine for a competitive insurance market but not necessary for a socially-backed insurance scheme like ACC. Little indicated that a future Labour government might return to the “pay-as-you-go” system of funding that was in place from 1974 – 1999.  I would suggest that the interview missed out an important aspect of the debate over ACC’s funding.

Fundamentally, the pay-as-you-go versus fully funded issue is about how to fund ACC fairly. To explain why this funding issue boils down to fairness, I need to start by explaining the two competing funding models.

A few notes before I go on. First, I worked for ACC for six years in various roles and I’m now a Lecturer at the Faculty of Law at the University of Otago. This post represents my views. Second, part of the interview is about whether the Government’s practice surrounding the setting of levies and announcing possible future levy cuts is “dishonest” or “manipulative” – I’m not getting into that here. Finally, I’ve glossed over a few of the finer details of funding.

The pay-as-you-go and fully funded funding models

Pay-as-you-go funding is relatively straightforward: each year, ACC collects levies to make up for the scheme’s outgoings that year.

Fully funded is a bit more complicated. Each year, ACC collects levies to pay for the lifetime cost of every claim that occurs that year. “Lifetime” cost means all the money that will ever be spent on the injury. Each year, ACC will set aside huge amounts of money with the expectation of spending it in the future.

Both models can collect reserves and provide security

The interviewer suggested that the fully funded model provides greater security because, by collecting reserves, it guarantees the payment of future costs. Little argued that pay-as-you-go also provides security. I think that he is correct that security is not a particularly helpful basis for distinguishing the two models. Both funding models can collect reserves in case actual costs are higher than forecasted costs. As Little said, it is sensible to keep a healthy reserve. 

The pay-as-you-go model’s security is provided by the state’s ability to collect levies in the future. In a way, this is the ultimate security – as long as the scheme has the backing of the state, the state can find a way to collect the funds to pay for it.  Private insurers lack that security: they can go bust and lose the ability to gain future levies. If the scheme were to be administered by private insurers (as it was for a brief time) then it makes sense to require those insurers to keep reserves. Requiring reserves means that, if a particular private insurer goes bust, there’s money set aside to pay for the future cost of any claims that insurer was managing.

So there is a link between the fully funding model and private insurance: fully funding is necessary if the administration of the scheme were to be handled by private insurers. But it’s a mistake to think that that is the only rationale for the fully funded model.

Different groups pay for the cost of injuries under the two models

The pay-as-you-go and fully funded models differ as to who pays the levies that fund injuries. That is, they spread the cost of accidents across different groups of levy payers. This naturally raises the question of which model is fairer, but we have to appreciate the difference to address that question. Let’s take a simple example:

I injure myself while gardening at home. I’m an earner, so the cost of my injury is paid out of the Earners’ Account, which is funded through levies deducted via PAYE. Under the current fully funded model, the money collected for the Earners’ Account this year is enough to pay for the lifetime cost of injuries that earners suffer this year. So, if I need surgery 20 years in the future, the money’s already been set aside, and I contributed to it.

In a sense, I’ve paid for the cost of my injury. More accurately, the people paying the earners’ levy this year have paid for the cost of my injury. The levy I paid might not have been enough by itself to pay for my injury - but not everyone paying the earners’ levy is going to get injured. By paying my levy, I’ve effectively insured myself against the possibility that I’m one of the earners that happen to get injured.

Under the pay-as-you-go model, it’s quite different. The levies collected this year will fund any costs relating to my injury paid this year. The levies collected this year will also be funding any money paid this year in relation to injuries suffered last year, or any year before that. If I need surgery in 20 years time, it’ll be the levy-payers in the year 2035 that foot the bill – which may or may not include me (perhaps my injury was severe and I’m no longer working, perhaps I retired early.)

Would that be fair? To answer that question, you need to have a conception of a fair method for distributing the cost of the ACC scheme across the community. We can put two different conceptions of our ACC scheme at opposite ends of the spectrum: community insurance and compulsory insurance. 

Community insurance or compulsory insurance?

Community insurance: the essence of the ACC scheme is that the community takes responsibility for those members of our community who are injured. The cost of doing so should be spread broadly across the community, and absorbed primarily by the sectors of the community that are in a better position to absorb that cost.

Compulsory insurance: the essence of the ACC scheme is that we are all required to insure against injury by accident. The funding of the ACC scheme should therefore reflect the basic principles of insurance funding: levies should vary by benefit and by risk. The greater the benefits an individual will receive if injured, the higher levy they should pay and the greater the risk taken by an individual or employer, the greater the levy they should pay.

Broadly speaking, the left tend to see ACC as community insurance while the right tend to see ACC as compulsory insurance. This fits with the left’s tendency to understand people in terms of the groups they belong to in contrast to the right’s tendency to understand people as individuals.

In terms of funding models, fully funded fits best with compulsory insurance: it’s unfair for me to have to pay for other people’s accidents in the past, just as it’s unfair for me to expect future levy-payers to pay for any injuries I suffer in the present.

Pay-as-you-go fits best with community insurance. As a member of the community currently in a position to help fund the ACC scheme, I should do so, thereby helping fund injuries suffered in the past by those who are no longer funding the scheme themselves.

The current funding of the ACC scheme sits somewhere on the spectrum between community insurance and compulsory insurance. The scheme uses the fully funded model, and levies do reflect benefits and risks to some extent. For example, higher earners pay more into the Earners’ Account which reflects that they will receive higher earnings-based compensation if injured. However, some aspects of the funding spread costs more broadly. For example, employers pay levies based on the risk of their industry group rather than their individual accident record. And, the Non-Earners’ Account, which would be the account paying for my gardening injury if I was not employed, is funded through general taxation.

Recent changes to the funding of the Motor Vehicle Account (which pays for injuries suffered as a result of motor vehicles on public roads) shift the scheme further towards the compulsory insurance side of the spectrum. From July this year, the ACC levy component of vehicle registration fees takes into account the risk classification of the vehicle. This reflects a compulsory insurance philosophy (people with safer cars should pay less because they are less likely to suffer an injury) rather than a community insurance philosophy. Indeed, a community insurance-based scheme might observe that the owners of safer vehicles are probably in a better position to absorb the costs of the ACC scheme, and ask more of them rather than less. 

It’s about fairness

The point I hope I’ve made is that the question of whether ACC should be funded on a pay-as-you-go or fully-funded basis is not primarily about competitive provision of the scheme. Fundamentally, it’s about which system is fairer – and there’s a political competition between different conceptions of a fair response to the problem of injury.

In my view, if Labour wants to sell the voting public the idea of pay-as-you-go funding, the way to do this is to sell the idea of the ACC scheme as community insurance – not by references to competition. That might be difficult, for several reasons. First of all, it might be difficult to persuade voters who are soon to pay lower vehicle registration costs to fund a larger slice of the ACC funding pie.

Second, there is an intergenerational fairness issue here (should I expect future levy-payers to pay for the cost of my injuries?) that invites comparison to the thorny issue of universal superannuation. The dual pressures of an ageing population and the introduction of effective but expensive new treatments potentially reduce the levy-paying population in the future while simultaneously increasing the cost of the scheme.

Finally, there’s what is often the elephant in the room when it comes to discussions about ACC: the injustice of treating illness and injury differently. The distinction is perhaps easier to justify for the right. If ACC is compulsory accident insurance scheme then the reason ACC doesn’t extend to illness is that that’s not what the levies we pay cover, and anyone who wants illness insurance should pay for it. However, if ACC is about the community taking care of our own it’s rather difficult to explain the disparate treatment of victims of injury and illness.

One final note – it is arguable that the simple picture of levy-payers paying their levies and funding the scheme that underpins the idea of compulsory insurance is flawed. Individual people and employers exist in a complex web of relationships and interactions, and it is artificial to ignore this. The Woodhouse Report argued that:

[T]here should be a general pooling of all the risks of accidents to workers [instead of employer or industry-specific levy rates]. Just as the steam power station relies upon the work of the coal miner so do all industries depend directly upon one another.

In addition, those interactions allow for cost-shifting. For example, a logger that pays a high levy due to high risk can, to some extent, shift that cost to the purchasers of their products who can, in turn, shift it to consumers. So, the setting of levies might be the beginning of the story in terms of the funding of ACC, but it is not necessarily the end.

Comments (10)

by Rich on May 13, 2015

So you injure yourself while gardening at home, and twenty years later, you need treatment.

In twenty years, other people will be injuring themselves while gardening at home, and needing delayed treatment in forty years time. It's a continuous thing.

The incidence of injury doesn't change very rapidly. It isn't like building insurance, where there might be an earthquake that eats many years premiums in one event.

While we might (self-driving cars, or increased automation at work) reduce the rate of injury in the future, this is going to be a slow process. It's hard to see why the future population (who will be benefiting from the lower accident rate) will be overburdened by having to pay for past accidents. 

Of course, if one hopes to privatise ACC in the future and break it up amongst insurance companies, then you need a fund to support that. I suspect that's the actual agenda behind full funding.

by Andrew R on May 13, 2015
Andrew R

ACC as community insurance is much closer to the original Woodhouse concept of ACC. Woodhouse also thought it should extend in time to cover illness as well.

The compulsory insurance funding model is a necesary step towards privitising ACC, continuing the ideological gutting of public services. 

The recent ACC decision to differentiate between types of cars is continuing that privitisation thinking.


by Andre Terzaghi on May 13, 2015
Andre Terzaghi

From a pure hip pocket perspective right now, does it make much difference if we have a mature pay-as-you-go scheme with no reserves, compared to a mature fully funded scheme with the reserves built to the required level? In the former, we are paying for current injuries plus the tail of previous injuries, in the latter we pay for the present and future cost of current injuries, but the tail of previous injuries is covered by the reserves. At first glance, medical inflation has the same effect either way.

Sure, for the last few years we've been paying very high levies to make the transition to fully funded, but the levies are now slowly getting dialled back. And the levies should settle at about the same level as pure pay-as-you-go levies would be.

It may be that fully funded schemes are more subject to underhanded political manipulation by changing assumptions and requirements thereby manufacturing illusory crises (yes Nick Smith, I'm talking about you), on average things should balance out over the long term.

Or have I missed something?

I worry about hidden agendas for the 30odd billion dollars of ACC reserves, and I hadn't even gone as far with my suspicions as Rich. But now that we've gone through the pain of building the reserves, even as someone with a mostly community insurance view, my vote would be to stay fully funded (and bring illness in as well).

by Andre Terzaghi on May 13, 2015
Andre Terzaghi

For what it's worth, the motor vehicle ratings don't appear to consider probability of injuring someone else - my 1994 Landrover Defender is rated safer (L3) than my 2001 Daihatsu Sirion (L4). Spotted a few other similar anomalies on a quick look through the ratings. I guess it fits the mentality of "as long as I'm alright everyone else can get ******  ".

by Simon Connell on May 14, 2015
Simon Connell

re: Rich's comment that:

In twenty years, other people will be injuring themselves while gardening at home, and needing delayed treatment in forty years time. It's a continuous thing.

And Andre's comment on a similar theme that: 

From a pure hip pocket perspective right now, does it make much difference if we have a mature pay-as-you-go scheme with no reserves, compared to a mature fully funded scheme with the reserves built to the required level?

Prior to the scheme become "mature", a pay-as-you-go system needs to take in ever-increasing amount of money each year. We can see this by considering a simplification of the scheme. Suppose that the ACC scheme worked like this:

  • the scheme only covers people who are seriously injured and cannot work for over five years;
  • to people who have cover, the scheme pays compensation for five years at $100 per year (and then stops paying);
  • five such injuries occur each year.

In year one, five people are injured. The pay-as-you-go scheme requires $500 to pay for those five people, while a fully funded scheme would need $2,500 ($100 for each person's payment that year plus a further $400 for each person for their remaining four years.)

In year two, a further five are injured. The pay-as-you-go scheme now requires $1,000. The fully funded scheme requires $2,500.

In year six, five new people are injured but the recipients of compensation in year one are no longer eligible. At that point, the scheme has "matured" as the entries to the scheme equal the exits. From year six onwards, the pay-as-you-go scheme needs $2,500 for the 25 people who will be on the scheme each year. This is Andre's point - at some point, the two funding systems end up needing to take in the same amount of money each year. There are people who were injured early on after ACC commenced who are still receiving entitlements, so the actual scheme is a way off being mature.

So it's certainly arguable that, in the long run, it doesn't really matter which funding model we use. Whether or not you accept that argument probably depends on two things. First, whether unknowns including medical advances driving longevity and treatment costs keep shifting the goalposts for the scheme maturing. Second, whether you think it's fair for those of us in the present to rely on the levy-payers of the future - this is going to come back to how you understand the relationship and responsibilities between individuals and society. 

by Rich on May 14, 2015

Simon, do you have some numbers on this for the actual ACC?

Specifically, how do costs break down into year of injury, Y+1...Y+N (presumably tending to zero as N increases and all the claimants die)?

Using this, you should then be able to factor for declining accident rates, increasing real medical costs and increasing population so see how much future claimants will be paying for an unfunded versus a funded scheme. I'm suspecting this will be a fairly small number?


by Simon Connell on May 14, 2015
Simon Connell

I don't have those sorts of numbers, Rich. ACC bases the levy-setting under the current model on actuarial predictions of future costs.

by Ian MacKay on May 14, 2015
Ian MacKay

Simon says, "In my view, if Labour wants to sell the voting public the idea of pay-as-you-go funding, the way to do this is to sell the idea of the ACC scheme as community insurance – not by references to competition. That might be difficult, for several reasons. First of all, it might be difficult to persuade voters who are soon to pay lower vehicle registration costs to fund a larger slice of the ACC funding pie."

You mean that selling the Pay-as-you-go must be sold to the voters. Like the current Fully Funded model was sold to the voters? Or was it just done as a "necessary task" of Government. Funny how Labour must sell while National can just act regardless.

by Andre Terzaghi on May 15, 2015
Andre Terzaghi

Simon, maybe you can swat a bee I've had in my bonnet for a long time: what is the rationale for ACC levies being charged on vehicle annual licensing, rather than entirely on fuel? The only hazard I can think of from owning a vehicle is from DIY repairs and maintenance. The vast majority of the hazard comes from using it, with the hazard in pretty close proportion to the amount of fuel being used. So it would seem much fairer to collect all ACC levies from fuel. Non motoring uses of petrol are generally pretty hazardous too, so upping the ACC levy on petrol would have boaties, lawnmowers, offroaders etc better help to cover their share of funding ACC.

And as for diesel, why is there no ACC levy at all on diesel, leaving the entire levy on licensing? If there really is an argument that agriculture, trucking, construction, marine users etc shouldn't be paying ACC levies - they are all already set up with accounting systems and it would be very easy for them to claim overpaid ACC fuel levies back at tax time. As it is, the lack of an ACC levy on diesel looks like a particularly sneaky hidden subsidy for a favoured few. Paid for by a grossly unfair annual licensing levy on low-mileage diesel vehicle users.

I can't see any good reasons for exempting diesel from excise tax either, but that's off the topic of fairness in ACC funding.

by Simon Connell on May 18, 2015
Simon Connell

The license fee levy for non-petrol vehicles is substantially higher. This is what ACC says:

Why isn’t there a ‘diesel levy’?

To ensure fairness among all levy payers, we charge motorists with diesel-powered vehicles the entire levy for diesel when they pay their annual licence fee.

This is because only a fraction of diesel sold in New Zealand is used for powering vehicles that travel on public roads. The rest is for power generators, boats, trains and other industrial purposes.

Post new comment

You must be logged in to post a comment.