Policy options and barriers for the next government

In previous elections, I either gave a list of mistakes I wanted the next government to avoid, or policies they could follow. Some of the mistakes I flagged in 2007 were indeed made, and about half of a preferred policy was implemented, no doubt entirely unrelated to my advice. In this election I want to take the perspective of the incoming ministers of the various departments and briefly discuss the problems they could try to address and the essential barriers they will face.

Financial regulation

The essential problem faced in the area of finance is the lack of competition, in part because of the demise of mortgage lenders in the GFC, and in part because the use of financial services (insurance, superannuation) goes via complicated decision procedures involving third parties (often employers) that open up avenues for anti-competitive cream-skimming. Some of the recent reforms are already trying to address that.

The barrier to solutions in general are not just the vested interests in the finance industry, which is incredibly rich and can thus buy up former prime ministers and others as its advocates, but also the sheer difficulties of maintaining a group of financial regulators who know what they are doing: good financial regulators will be poached by the finance industry. One would thus not merely want to look at how to reform regulation but also as to how to engender a group of good financial regulators that stay within the state system.


Two mayor problems in the health industry are that people don’t pay for their own health service consumption and thus naturally over-consume, whilst the other is that health providers co-opt the state into giving local monopolies to particular providers in order to increase their incomes. In turn, these two ultimately derive from huge agency problems in the health system in that consumers don’t know what is wrong with them and don’t know what will be wrong with them in the future (which gives rise to a role for intermediates and insurers), whilst providers will by design mostly be local monopolists (local hospitals, GPs, pharmaceuticals, professions, etc.) with an incentive and ability to organise themselves to grab more rents.

The barriers to solutions are again not merely the existing interest groups protecting their spoils vigorously, but also the complexity of the current system and the intellectual difficulties of figuring out alternative systems: unlike most other problems, where the ‘optimal policy’ is fairly easy to spot and one is ‘merely’ talking about the political difficulty of interest groups and societal beliefs in the way of that solution, the health system is so complex that no-one can honestly say they have a firm handle on what the optimal system would look like. In such situations, one basically wants experimentation at the local level and an ability to push through reforms at higher levels once local improvements have been found.

Besides encouraging and protecting experimentation, there are a couple of obviously dysfunctional interest groups that could be tackled by an incoming minister, to almost certain benefit of the community. The overpaid medical specialists protected by their unions (such as the Australian Medical Association), the too-expensive pharmaceutical benefit scheme protected by the media operations of the pharmaceutical companies, the barriers to entry to nurses to compete with GPs and medics inside hospitals, are three cases in point of obvious policy improvements waiting to be championed.



In general, the welfare system (subsidies for unemployment, disability, single parents, etc.) has been in a catch-22 for a long time now: be nicer to people in need and you will get more people in need, whilst being harsher to people in need means you are harsher to people in need. The balance we have had on this one for the last few decades seems fairly stable without any obvious improvement in sight.

The one thing I can see happening is that ministries and state bureaucracies might find ways to hand more welfare money to themselves rather than people in need, effectively by creating bogus jobs. There are many possibilities to do this. Futile training for the unemployed, ‘account managing’ for the disabled, ‘financial councelling’ for all and sundry, etc., should all be seen as activities in great danger of becoming sink holes for the hidden unemployed within bureaucracies. As far as I know, Australia seems to have avoided the worst excesses of this, but I fear for what the ‘National Disability Insurance Scheme’ (DisabilityCare) is going to mean in practise.

In general hence, I would see the main task for the new minister in this area to be to keep the numbers of hidden unemployed in the ministries and local councils as low as possible.


Primary and secondary school education

Here the main problem is the lack of quality education for the general population. As i have indicated before, I see this as essentially arising from the ability of Australia to just import highly-trained people from elsewhere, negating the pressure to educate ones own children to the same degree. I can’t really see any political support for changing that implicit choice and, indeed, on reflection it is just not a problem: Australia exports its great culture and imports people who have gone through the pain of high-quality education. Win-win on both sides, really. Hence the barriers to meaningful reform are just too formidable and institutionalised to even bother enumerating. The main task for the minister will be to try and guide the national curriculum into something sensible and otherwise keep the number of hidden unemployed in the ministries down, which will be a challenge given the directions the Gonski reforms could go into.


Tertiary education, research and development

Here the main issue is the cream-skimming of too many bureaucrats. The obvious answer is a few razor gangs and perhaps a cap on the maximum salaries in this sector. The barriers are almost entirely institutional, ie come from the separation of powers between the commonwealth and the states. Still, if the ministry in Canberra wanted to stop the cream-skimming, I think it could get quite far quite quickly. So there is something for a minister to achieve here.


Industry policy, NBN, Infrastructure, property rezoning, mining taxes, etc.

Here the problems are the usual: economic interest groups who manage in a large variety of ways to get more than their fair share out of the community. From an economic point of view, the ‘solutions’ are not too hard, as the barriers are almost entirely political. Depending on the issue, one can either just repeal bad legislation (like with mining taxes: just let the states ramp up the royalty rates), beef up the bureaucracy’s ability to think through major tax and spend plans (such as via the parliamentary budget office), scale down particular programs on some pretext (NBN? Particular subsidised industries?), or commission a few reports on areas where one would probably want some new institution to improve the operation of the government machinery (such as property rezoning where one should think of some way to auction off the discretionary element in rezoning and exceptions-to-planning-rules).


In my post on this two weeks ago, many commentators gave long lists of policy options, ranging from the decriminalization of drugs, to land taxes, to health rebates, to being nicer or nastier to boat arrivals. Some seem very sensible to me, particularly an end to the health insurance rebate and regulation rather than criminalization of drugs.

Author: paulfrijters

Professor of Wellbeing and Economics at the London School of Economics, Centre for Economic Performance

14 thoughts on “Policy options and barriers for the next government”

  1. “people don’t pay for their own health service consumption and thus naturally over-consume”

    Is this really the case? Two reasons I suspect it’s not 1) the relatively low cost of decent health care in Australia when compared internationally, and 2) waiting lists as a kind of filtering mechanism based on medical need.

    Also, regarding education, I think that things are improving slowly. With pre-school now a proper year of education at a school with a real teacher, the achievement at all year levels will increment slowly upwards. When kids start year 1 knowing how to read it accelerates learning and this flows through. It would be great if this leads to some flexibility in the final years of compulsory schooling for both more advanced classes and also vocational classes.

    But to your point about importing skills, I will just reiterate that it is a policy choice. The social results of which need to be considered. For example, if I was an Aussie growing up in a working class area, and I saw that a large shared of educated professionals looked foreign (compared to my working class peers), I’d draw the conclusion that immigration is ‘stealing’ good jobs.

    An alternative choice is to import only uneducated immigrants. Locals would have a head start and higher rewards for education because of lower jobs competition.

    Neither is right or wrong, but we rarely have this kind of honest debate.


    1. what are you on about? People don’t pay the full cost of their health care because its a national system and there is insurance on lots of medical care. I perhaps could have been a bit more careful in how I phrased it, but surely the point is fairly obvious? Do you really need me to dig up the percentage of health care paid for by the state or insurance as opposed to out-of-pocket?

      On education, I am not so sure about the move towards more pre-school. The evidence that it really helps is not all that solid.

      Sure, importing skills is a policy choice and not all aspects of it are clear to the population. I doubt that you will find much support though for importing more low-skilled labourers, even if you manage to get people to see all the issues involved! Indeed, I wouldn’t be in favour of it.


      1. I’m on about whether people over consume health care, not who pays for it. What’s the baseline/welfare maximising amount of health care consumption?

        If you want to say the quantity in a perfect market, you are left with a rather crazy guessing game about the possible perfection of any market structure (with credence goods and other failures) and what amount of health care would be consumed.

        You even explain this and say we don’t know what a better system is like.

        So is it really the case that people ‘over consume’ health care services? I suspect this thinking comes from your views on obesity – that because people aren’t paying the cost of their medical treatment they don’t look after themselves and therefore over consume compared to if they had to pay for all their medical services.

        Even then you can’t know if the obese person has higher life-time health care costs because they are likely to die younger. And you can’t know the impact of obesity on their lifetime productive effort.

        So why start with the assumption that there is something wrong just because it doesn’t look like a typical perfect market scenario?


      2. I am just arguing that the demand for health is downward sloping in private costs. As soon one has subsidization one then gets over-consumption in all cases where the uncertainty about the medical condition is resolved and there is some element of choice left.
        Let’s not drag obesity into this discussion. It wasn’t on my mind when writing that bit. One of the main inefficient cost in my mind was the high expenditures at the end of life, which are inefficiently high precisely because of the agency problems flagged. See here for a relevant and typical discussion: https://www.mja.com.au/journal/2011/194/11/time-rethink-end-life-care

        More generally Cameron, I hope you are not going to develop the habit of demanding every time for undeniable proof that trees have leaves and that people respond to material incentives. It gets tedious: arguments on efficiency invariably turn into a balance of probabilities issue so you can demand certainty all you want, but you won’t find it.


      3. “I hope you are not going to develop the habit of demanding every time for undeniable proof that trees have leaves and that people respond to material incentives.”

        Nope. Any evidence with sound reasoning could convince me.

        I agree that end of life costs are important (moral) questions. Voluntary euthanasia seems to be a no-brainer here, as is looking at more efficient outcomes for end-of-life health services such as nursing care (non-hospital based care).

        But you still side-stepped the issue of what an efficient level of health service provision might be. Knowing that demand for health services might be downward sloping (and might not be in the aggregate depending on wealth effects and other things), doesn’t actually tell us anything about our relative position to some social optimum.

        To be more blunt how we would know when we are the efficient level? What sort of evidence should we see?

        Alternatively, how do we know we are not under-consuming health services? Maybe more health services pay-off through higher workforce participation etc. Maybe people have imperfect information about potentially beneficial medical treatments.

        The point being, if you don’t know the optimum you can’t say we are over consuming. Unless you are in the mood to just assume some optimum.

        But this ‘start from assumption’ type reasoning is what leads to absurd conclusions. I’m reminded of what I wrote about a King-Gans paper that modelled health insurance markets with this approach.


        The Gans-King model shows that “public health insurance will only be used by those in society who are healthiest (i.e. least likely to become ill).”, yet the reality is that there is beneficial selection into private health insurance – “the healthiest people tend to be privately insured”


      4. Cameron,

        let me repeat the basic argument: “As soon one has subsidization one then gets over-consumption in all cases where the uncertainty about the medical condition is resolved and there is some element of choice left.”

        Why don’t you answer the other questions, which are all on difficult and complex issues in which i doubt you will find a person in the world with an entirely satisfactory answer, yourself? Part of the point of economics is that we develop simple heuristics to make decisions in complex situations without knowing what the perfect outcomes is. And yes, no-one can give you a cast-iron ex ante guarantee that it will work because they lack perfect answers. If you want a pretense of certainty, pick your religion.


      5. Paul, yes I can read 😉

        To answer your question, I haven’t answered the questions I posed, therefore I won’t claim there is over consumption of health care services in Australia.

        And I can’t see how it is possible to make your claim without answering them just because you invoke a simple heuristic.

        Otherwise, what makes you believe your approach is better than any other religion I choose?

        As I said, your basic argument is incompatible with your claim that the optimum is unknown. How can it be otherwise? You are saying the equivalent of “I don’t know how deep the water is, and in fact it is almost impossible to tell, but I am certain it is more than 5 metres deep because of a rule of thumb from first year economics”. Mmm…

        My point is actually about uncertainty, and why uncertainty means choosing the appropriate economic model to understand a situation is impossible without knowing much more information.

        Surely these basic economic heuristics are mere tools to perhaps indicate the direction and magnitude of change in health service demand, in response to some policy, conditional on all the relevant constraints. But they sure can’t tell us the social optimum in aggregate, and therefore our position relative to the social optimum unless most information is included (assuming you believe such an optimum exists or has any practical value).


      6. no to nearly all of this, Cameron. How can a mother tell her kids its good to drink milk if she doesnt have the perfect model of the universe and can thus anticipate and include all consequences of that advice? Should she stop from giving any advice unless she knows everything? How can one say its good to stop for red lights, pay taxes, and not randomly kill people unless one has perfect certainty? People make do with limited information, Cameron, that is precisely what heuristics are about. And one can, indeed, point to the uncertainties and say that one is implicitly making some guess about the uncertainty that implicitly rationalizes the given advice or rules.

        What heuristic am I using for the over consumption argument? Once again: “As soon one has subsidization one then gets over-consumption in all cases where the uncertainty about the medical condition is resolved and there is some element of choice left.”
        I am thus using the economic heuristic that when people know the benefits of something, their willingness to pay starts to coincide with their private benefits. If the price goes below that willingness to pay because the price is subsidised, consumption goes up and private benefits decline with more consumption. Etc.

        Want more papers explaining the same point and explicitly looking at health? Try this one and the papers referring to it: http://ideas.repec.org/p/boc/bocoec/495.html


      7. I’m kind with Cameron on this one. In a market with rationing, manifold externalities, monopoly rents and countless inefficiencies (and I should probably add departures from the neoclassical model like bounded rationality, imperfect information and the like), it is impossible to be confident of over-consumption simply because the product is subsidised. Plus, it’s more nuanced than that. I suspect that some things are over-consumed, and other things under-consumed.


    2. Personaly I’m with Paul here, in terms of broad generalities. It is easy to pinprick details but the generalities still remain.


      1. Over consumed or under produced – how do you tell the difference when queuing is used to control consumption? I dread going to the local bulk billing doctor because I have to wait an hour on average (usually somewhere between 1/2 an hour to 1 1/2 hours) sitting with a bunch of other sick people before getting to see the doctor, and that’s when I make an appointment. And 6-12 months wait for many procedures.


  2. You’ve kind of only tackled small beer here paul.

    Alright, let’s up the challenge level a little.

    Climate Adaption, including disaster, drought relief.

    Currently our approach is ad-hoc, for instance I’m thinking the Brisbane Floods and subsequent levy, the Victorian bush fires and cyclone Yasi, as well as early noughties drought relief, which lasted for – a long time (in fact the last few wet years might be the anomaly due to La Nina)

    Some of those things are just plain ‘this is Australia’ and some of those have a component of climate change.

    There must be a better way to provide relief from disasters, given that ‘this is Australia’, but also there is going to be enormous pressure to support lost causes (like providing drought relief on a lot of increasingly marginal land, or rebuild houses in places like south brisbane where the 100 year flood could becomes the 20 year flood, as well as reduce systematic risk (e.g. early management of the fuel load in wooded areas in dry years.)

    Should there be an disaster/adaption insurance scheme of sorts? Or should we continue tackling it piece meal out of general revenue, or should we harden our hearts – no insurance, tough luck for being in a cyclone path or fire zone (note that you can’t get flood insurance in parts of Brisbane any more, what does that say?).

    One idea I had, given the state of ‘skepticism’, is this – we could create a ‘disaster insurance fund’ which is set to a mandated level, indexed to inflation. This would be filled with a straight down the line Carbon tax, the fuller the fund, the lower the tax.

    No disasters, no tax. Many disasters, higher tax as the disasters are funded. (no worse I’d argue than a general levy, like the brisbane floods). That way if things turn out to be ‘not a big deal’, there’s no impost, but the other way around, then mitigation is increased.

    It also sets interest groups off against each other. Those who want access to the fund vs those who don’t want to pay the tax, creating pressure to keep overheads low, and find permanent solutions to chronic problems as they arrive (for instance if marginal land stops being even marginal).

    Maybe not.


    1. Futher idea:

      If there was concern that this could just feed another bureaucracy, try this – set up the fund as a corporation, 11 board members, 4 from stakeholders such as landcare groups, emergency services, farmers, environmental groups, 4 from taxees whose levies pay the fund, 3 from govt, including the chair.

      If the bureaucracy gets too big, the stakeholders and taxees will be able to pressure management to downsize. If there is a pressing need, then the govt + stakeholders have the numbers. If there is a chronic interest group guzzling funds, then govt + taxees should be able to bring pressure onto them.


    2. Prediction of flood probabilities is a dark art – especially in the absence of long term records. One also has to take into account planning decisions that can also turn a 100 year flood into a 20 year flood in terms of its reach.

      One of the problems is the number of individuals who do not personally insure against disaster. Ad hoc responses therefore follow


Comments are closed.

%d bloggers like this: