N.B. Although I am an ACT Party candidate this is not ACT policy. It’s a far out idea, but it’s worth a float. If I was a Nat I’d probably be expelled for writing something outside the party lines, so if this post does say anything about ACT it’s that our party tolerates free thinking and open discussion.

Welfare in New Zealand suffers a fundamental problem. Economists call it moral hazard, and they define it something like this: Two or more parties enter into an agreement but one cannot monitor whether the other is really trying to keep their end of it. The classic example is car insurance: Once insured most people are less careful than they would be without insurance, to the insurer’s (and ultimately other customers’) cost. For better or worse, welfare in New Zealand is an agreement among all citizens that people faced with misfortune including unemployment (Unemployment Benefit), incapacity (Sickness and Invalids benefits), injury (ACC) and childbirth with insufficient income (DPB) will be compensated with a replacement income. For the agreement to work all people must ensure that they do everything in their power to avoid relying on welfare, and rely on it only when they have no other option. Unfortunately, moral hazard tells us that nobody’s internal motives can really be known.

The key problem is that a person’s circumstances are never entirely in or out of their control. The most conscientious worker can end up out of work despite making every possible effort to find it, but talented and highly employable people can choose to be indolent. Because the human body is so complex health conscious people can be struck by terrible sickness and incapacity, but people blessed with good health can fritter it away through bad choices. The most careful workers can be injured by bad luck, but I’ve also seen people do some stupid things on building sites. There is no 100 per cent effective contraception and anyone can have an unplanned pregnancy despite their best efforts, but I have first-hand accounts of people asking doctors to remove their IUD in order to get pregnant and claim the DPB.

The problem with welfare and the debate surrounding it is that civil servants cannot make decisions based on their judgement of a person’s motives. They are bound to administer the law without discretion. They may identify beneficiaries who are clearly making bad choices and who would make better ones if the benefit were not available (let’s call them type A beneficiaries), but they cannot discriminate by allocating less or cutting the benefit completely. They may identify people whose genuine need is greater than what a benefit can meet (let’s call them type B), but they cannot discriminate by allocating more. The result is pleasing to nobody, and it makes for a surrounding debate where both sides talk past each other.

Welfare critics tend to believe that too many beneficiaries are type A, and that welfare needs to be made meaner. Welfare defenders tend to assume that too many beneficiaries are type B, and welfare needs to be made more generous. Critics cannot win because the public will not accept putting type B beneficiaries into even greater hardship. Defenders cannot win either, though, because the public will not accept giving more money and therefore more hazardous incentives to type A beneficiaries. Take the proposal to time limit DPB receipt to six years over a lifetime. It might discourage many from using children as meal tickets, but it would also put at least some in even greater hardship. Again, the problem stems from the fact that different people have different internal motives that cannot be measured objectively.

The only way to make welfare work better than it currently does (and the Welfare Working Group has provided the figures to show why it seriously needs to), is to solve the moral hazard problem. That means giving the welfare administrators the ability to make decisions based on subjective assessments of motives. That’s the only way they can make welfare more compassionate for type B beneficiaries and remove the hazardous incentives that trap type A’s into debilitating dependency at taxpayers’ cost.

Except the way welfare is currently set up they fundamentally can’t. Giving WINZ staff discretion to decide who is “deserving” and who is not would be a disaster. The whole concept of the rule of law would be thrown into chaos as little Hitlers in WINZ offices across the country went on personal tirades to rebalance type A and type B beneficiary numbers according to their own prejudices. It would be an opening to corruption of the likes this country has never seen and never should.

However there is one very successful institution that does make decisions with heavy consequences about nuanced matters within the law. Juries do it every day and compared to realistic alternatives they do an extraordinary job. Juries make decisions that no civil servant should be allowed to make because nobody is a juror all the time so there’s little scope for corruption. What’s more private citizens are allowed to have private prejudices that civil servants aren’t, and when there’s twelve of them they usually balance out to something sensible.

You can probably see where this is going now: Welfare in New Zealand should include jury-like boards that would involve ordinary private citizens in deciding recipients’ eligibility (or not) for welfare. Mechanically, it might work something like this:

The country would be divided into communities of 10,000 adults. To the extent possible in each geographical region, the communities would be balanced by the deprivation index, so they would all have roughly the average socio-economic status of the country as a whole. Some remote regions would have communities that would be outliers. The tip of Northland, for example, has a lot more beneficiaries than most parts of the country and no high employment regions nearby. All communities would have a share of total welfare spending weighted to their share of deprivation.

Taking the Welfare Working Group’s figure of 1:13 working aged adults receiving a benefit, the average community would have around 800 beneficiaries. If there was to be a 6-monthly assessment of each beneficiaries’ needs and benefits, then the average citizen would be required to form part of a citizen assessment panel of twelve for one day per year during which time they would assess 20 cases over 20 minute sessions. Ten minutes to interview, ten to deliberate.

In each case the board would review the recipient’s situation and make a recommendation based on a series of options including more money, less, stopping the benefit completely, or recommending some form of training or work program. All of these decisions would be made in view of a total budget for the community. As a bonus feature, it might be that each community would evenly share any monies not spent from the annual budget.

The main result would be that the scope for discrimination, in the original sense of the word, would be greatly increased. Truly tragic cases would be able to benefit from discretion. Cases where the beneficiary was abusing the system and had other options would come under more intense pressure to end dependency. The overall level of justice and humanity in the system would increase.

A secondary result would be political. As I have written elsewhere, voting in elections is a terrible way to understand and influence policy decisions. The rational and informed voter gets the same government and the same policies as the irrational and ignorant voter, but the first has a lot less free time. Smart people don’t think too hard about politics, and everybody suffers bad policy as a result. A day’s mandatory service each year would focus voters on what is really happening with the welfare system. It might be that on balance the political demand for a more generous welfare state would increase. I doubt it, and I’m happy to take the chance.

The system has precedent. The original Old Age Pension Scheme in the 1890’s applied character judgements to recipients. Drunks and wife beaters could be denied their old age pension. As Jonathan Bartholomew has recorded in The Welfare State We’re In and David Beito has recorded in From Mutual Aid to the Welfare State, friendly societies once carried out all of the functions of the modern welfare state, and arguably more effectively. They succeeded because they used peer pressure and personal relationships to solve the moral hazard problem.

Of course, my libertarian friends have probably defriended me on Facebook and burned me in effigy by now for suggesting mandatory participation in these community boards. However, I believe that we need to make smart trade-offs if we are to actually achieve a freer society instead of just talking about one. Current expenditure on welfare (excluding Super) is around $3,000 per taxpayer per year. One day to see it spent better is an inferior solution to not paying it to the state in the first place but it’s far better than what we have now.

This idea is way out there, I accept that much. However our welfare system fundamentally doesn’t and can’t work the way that we currently try to make it work. More generous welfare means more abuse, “meaner” welfare just means failing to help the people supposed to be helped and defeating the purpose of having the entire scheme in the first place. The nub of the problem is moral hazard and the inability of civil servants to deal with it.

If a problem defined is half solved, then please consider this solution.