It is uncontested in New Zealand today that government has a legitimate role to play in education. There is, sure, significant debate what that legitimate role for government is: Should government set curricula? Fund all levels of education universally? Manage schools or leave it to parents? But not even the most radical politician or academic suggests government leave schooling entirely to the market. However, even if it isn’t a debate being had, determining the reasons why government should be involved in education is crucial in determining how government should be involved.
In his seminal essay on school choice, The Role of Government in Education, Milton Friedman (1955) makes this argument for the state funding of academic education. He argues that the basic education of an individual is valuable not merely to the educated person (because he can use his education to provide higher-value services and therefore, garner higher wages) but to the wider community too. Better educated people, in a democracy, make presumably better democratic decisions – benefitting all, including those who have not invested in an education for themselves. Much of the benefits of general education going to the public rather than the payer suggest it would be under-consumed in the absence of state subsidy – because, when rational consumers make spending decisions, they typically maximise their own utility, not that of the wider community. A state subsidy is needed to re-equilibrate in favour of maximising community welfare.
However, this argument can only be carried so far. Friedman draws the line at “general education for citizenship” (p. 2). It is difficult to see any external benefits of purely vocational training . The gains accrue entirely through higher earnings to the person trained. The obvious exception to this is how higher earnings from vocational training interact with the tax system and social-welfare system. Presumably, higher earnings from more training reduce expenditure and increase revenue However, assuming the abatement rates/marginal tax rates on these systems are not punitive, consumers should account for this higher income in their own calculations to at least a reasonable degree. Secondly, if we were to count higher tax revenue as a positive externality, we would rationally have to extend this to every other revenue-increasing measure. That is equivalent to subsidising every piece of expenditure which deepened the capital stocks of the country – i.e., subsidising all investment. In other words, it is the same as simply cutting taxes on capital. Overall, therefore, vocational education can be considered to have no real external benefits.
Unfortunately, it is challenging to draw the line between ‘general’ and ‘vocational’ education: Is an economics degree at university general or vocational? Should students at high school who take technology subjects be considered to be undertaking ‘general’ or ‘vocational’ studies? Still, we should draw the line somewhere: Taxpayers should not be required to subsidise that which will have no general benefits and only make the recipient richer.
However, this does not necessarily doom any government support for vocational education. There is a legitimate argument, made by Friedman (1955) and deepened (at least for me) by Harvard economist Caroline Hoxby (2006), that even if government should not directly subsidise education, it should surely intervene in the funding market for it because, as Hoxby puts it, “the capital market for financing education is highly imperfect” (2006, p. 9). This imperfection stems from:
a) an inability for children – in the compulsory years of education – to sign up to loans or other funding mechanisms to pay for their education (because they are children);
b) the unusual nature of human capital investment – it cannot be repossessed if a loan is defaulted on; and,
c) the undiversified, and therefore risky, nature of an investment into human capital.
This all means typical profit-seeking funding mechanisms, like bank loans, are ruled out or substantially disadvantaged in education. So, students must seek funding from their parents or other altruistic funding sources. If private loans are available, it might only be to relatively-well-off families who can offer parental guarantors with good credit ratings. This leads to egalitarian and efficiency concerns: For one, most would consider it unfair that able students who would value a vocational education beyond their family’s immediate means be denied it, even if they could easily repay it with post-training earnings. Even leaving aside fairness concerns, denying an education to people who value it more than it costs and who will be able to pay for it post-graduation based on the birth lottery of their parent’s wealth is clearly inefficient: Profitable transactions are not being conducted.
The egalitarian argument, that state-funded education is needed to ensure inter-generational wealth mobility, is essentially the flawed capital market argument stated in social, rather than economic terms. Without state intervention to ensure adequate funding, good jobs – even political leadership – will be confined to those who have enough altruistic resources (i.e., wealthy-enough parents) to pay for a good education. This would create a caste-like system, where one’s future was not determined meritocratically but based on the wealth of one’s parents, which would then determine how wealthy one was to provide for one’s children and so-on, in a self-perpetuating cycle of inter-generational inequality. This is the most prevalent, most obvious, and probably most persuasive argument for the state funding of education.
This is the argument with which I have the most trouble. It suggests that a homogenous education, rather than an education in and of itself, is necessary. The argument goes that: In a pluralistic country, children are bought up in a variety of cultures, religions, and socioeconomic circumstances, for instance, and that to create a functioning society thereafter, these differences must be mitigated by a shared experience at school. Most communitarians also hold that were parents to choose their children’s school, segregation would continue into school. By this, communitarian arguments, typically implicitly, also hold that, by virtue of their often-better education, rich parents would make better decisions in a world of choice than poor parents.
Liberal (and socialist) believers in communitarianism argue that private schools and elite public schools insulate their students from real struggle and, by excluding poor students, perpetuate class division by creating shibboleths to which only the wealthy are privy and which they use to ensure continued class dominance. The much-derided ‘old boys’ networks’ are the archetypal example. This, in my view, is a reasonable argument. The problem comes when the solution proposed is not to widen access to the good schools but to destroy them for all, or, even worse, when the prescription is not that this a natural result of the monopolism of school zoning but of ‘parental choice’. This assumes that poorer parents would not send their children to good schools if they were able. It also results in the trapping of more students in bad schools if they cannot escape their zone. The answer is to expand choice, not homogenise: Some receiving better education than others does not justify cutting everyone down to the lowest common denominator, but rather raising everyone up.
On principle, I cannot see why it is the government’s job to homogenise society. That assumes, firstly, that the government knows what the correct way is to homogenise. Sure, on some issues – that everyone should learn to add and subtract and converse in and write English, for instance – there is almost universal agreement. There, enforced homogeneity is acceptable. But on others, it is unclear why the opinions of Wellington bureaucrats or the ruling government should be forced upon everyone. That seems to breed authoritarianism. Where taxpayer funding is at stake, blatantly repugnant views or inaccurate teaching, rejected by the vast majority of society, can be rejected. But contestable matters, like whether students should be assessed by examination or by internal assessment, what areas of history ought to be taught, or whether Bible readings ought to be held, should remain in the hands of parents, to avoid the tyranny of centralisation.
The simple practical argument, though slightly difficult to accept, is that if one group of parents make a mistake, that affects their children, but when a bureaucrat makes a mistake, it affects all the children. Homogeneity means a single point of failure, heterogeneity means a surer chance of at least some succeeding in the face of a terrible error.
In response, one could argue that central administrators are infallible or, even, simply far less fallible than parents. That is rather difficult to do, particularly in education. For instance, in Charlotte-Mecklenburg in the United States, girls who were able (as in, chosen by ballot) to attend their first-choice school were 14 percentage points more likely to complete four-year college degrees than identical students who lost the ballot. Boys showed no change (Deming, et al., 2014). Parental choice works better (or no worse, for boys in this study) than government assignment. This ignores, too, the creation of competition between schools by parental choice.
Consider, by contrast, the inherently political, rather than scientific, nature of much of modern educational ‘scholarship’ – the basis, surely, of any administrative decision to assign schools. The Reading Wars, between phonics and whole-word approaches to literacy teaching, is an interesting example. Proponents of phonics teaching are “sometimes represented as closet Christian fundamentalists who use phonetics instrumentally to bolster traditional authority” (Furedi, 2015). Their opponents include the Whole Language Movement which openly describes itself as “political movement”, fighting against “a right-wing or neo-liberal agenda”. But whole-word proponents have not got off easy either: They are accused by phonics proponents of “destroying the innocent” with their methods. In New Zealand, we see many cases of political, rather than scientific, judgement determining the future of our children: On mathematics education, the Numeracy Project is heavily contested, despite having been adopted wholesale by the Ministry. The standards-based vs norm-referencing debate in assessing students is far from settled, and yet, one was adopted across-the-board nationally with the introduction of the NCEA. Educationalists and educational officials are just as, perhaps more, prone to non-scientific and uninformed decision-making as parents. It is, therefore, difficult to grant any credence to a contention that officials are so impervious to error and parents so susceptible thereto that introducing a single point of failure is a good idea. Surely the less risky approach is decentralisation.
The first three arguments – of external benefits of education, of a flawed capital market existing, and of egalitarian concerns necessitating equality-of-access – are extremely compelling. Government intervention is clearly necessary in the education market. But, it is unclear that this intervention must take the shape of New Zealand’s current system. The flaws in the communitarian argument serve as a good argument against increasing state control of the education system. Increased choice seems, essentially, a no-brainer.
Every person, whatever the level of his academic ability, whether he be rich or poor, whether he live in town or country, has the right, as a citizen, to a free education of a kind for which he is best suited and to the fullest extent of his powers. — Peter Fraser 
This was the ambition Peter Fraser, one of New Zealand's most internationally accomplished statesmen and the Prime Minister during the Second World War (and in my view, one of our greatest), had for the public education system in 1939. It has been the goal of compulsory education in this country ever since. Sadly, the current situation is a long way off.
Right now, the compulsory education system performs very well –compared to other Western nations - for those at or near the top of the socio-economic ladder. However, those nearer the bottom are shafted. This is a global phenomenon, but its particularly bad here.
In New Zealand, more than 13% of the average student's performance on the mathematics section of the international PISA assessment (undertaken by a sample of 15-year-old students) is purely determined by his socio-economic class, far from the egalitarian and meritocratic paradise envisioned in 1939. This is much worse than many of international peers – only Ireland and Singapore in the Anglosphere have less equality, which they compensate for with far higher achievement rates.
Figure 1 – Data: OECD (2016). The green quadrant shows the ideal situation – high performance with high equity. The orange quadrant shows suboptimal positions and the red is the least preferred option – below average performance, combined with below average equity.
Later in school, this inequality is magnified. In New Zealand Scholarship, the most prestigious secondary examination offered by NZQA, Decile 10 students entered the examinations at over 15 times the rate of Decile 1 students (Lipson, 2018).
This is a problem. It is obviously unfair for the individual who, purely because of his parents' income, can't achieve to his potential. But more than that, on a larger scale, it may well result in lower productivity and therefore, a worse lifestyle for all of us, in aggregate, than we otherwise could enjoy. This country has a limited amount spaces at quality schools available. We may not like that, but it is a fact: some schools are better than others and, in the current system, those schools are stuck at their current size. Right now, through both private schools and zoning for in-demand state schools, we allocate those spaces on the basis of parental wealth (in that wealthier parents can afford private fees or to move into the zones of quality schools, which generally add a premium to property prices and rents).
This deductive analysis of the costs of education inequality is backed and indeed extended by the empirical evidence. Sabatés, Feinstein, and Shingal (2008), for instance, found a significant correlation, controlling for other relevant variables, between increasing educational inequality and juvenile crime rates.
Of course, that model of educational inequality raises many crucial questions: why are some schools better than others? Why are the best schools concentrated in areas of high wealth? And the most important of them all, why don't poorer schools simply copy the best schools?
I won't focus on the first question of why some schools are better than others because it is probably best left to the educationalists (and, in my view, the market). The second and third questions, however, are ripe for economic analysis.
Before I begin this analysis though, I should outline a basic – but apparently controversial - assumption I make of school administrators: that they seek to maximise their budget/revenue. This assumption comes from Niskanen (1968)’s famous paper on the ‘peculiar economics of bureaucracy’. Revenue can be considered the quantity maximised by administrators because it seems the only quantity which varies uniformly in concert with what are presumed to be important for self-interested bureaucrats: the prestige of their position, ease of management , power, even salary. I also assume that school administrators wish to advance in their careers. This should be relatively uncontentious.
The reason the best schools, whether public or private, are concentrated in wealthier areas is simple: competition. The wealthiest parents have a much wider variety of options when choosing where to send their son or daughter to school than poorer families. They have the entirety of the private sector open to their consideration, as well as every public school in the country - because they can afford the premium to move into their zones. This naturally encourages schools, which, whether private or public, depend on per-capita fees, in their area to improve to attract their 'business', as it were, because otherwise they might lose that business.
As to why schools in poorer areas don't improve by copying the wealthier schools, the answer is slightly more complex.
Firstly, it is because, unlike schools in wealthier areas, these schools are not exposed to the competitive forces detailed above. Essentially, parents of a lower socio-economic class have far less choice available to them: they can't afford private school fees or to move into a good school zone and pay the resultant rent premium. They are therefore a captive audience for their local school. In effect, their local school has an effective monopoly over the education of their children. As all economic experience shows, monopolies have no incentive to innovate (because they can't lose their business), and therefore, generally, provide a worse service at a higher price.
The other reason lies in the monopoly that teachers' unions and the Ministry of Education have over teachers. Public schools have no choice in the pay rates of teachers they employ; all are employed under a collective agreement negotiated between the unions and the Ministry. This contract, unlike those in most other professions, is not designed to reward success and excellence, but rather seniority. Private schools can bypass this poorly-designed collective agreement and set up their own pay structure, while public schools with well-off parents and alumni can supplement the collective agreements with external incentives, like privately-financed bonuses and sabbaticals (e.g., those from Auckland Grammar's multi-million-dollar Academic Endowment Fund). Schools in poorer areas have no such opportunities available.
So how do we fix this? Given my diagnosis of why the problem of educational inequality occurs: government-enforced monopolism in the education market for less privileged students, the answer seems obvious: get rid of each of the anti-competitive forces that restrain the market.
In order to establish adequate competition in any market, both sides of the transaction should have sufficient information about the other party, particularly the buyer about the quality of the seller's goods. Creating this transparency is particularly difficult in New Zealand education, given the NCEA system. I shall discuss my preferred reforms to those later. As such, we should create a centralized website for parents and students to compare schools which includes data on standardized exam results, rates of improvement, Education Review Office reports and short, readable summaries of these reports and the qualifications of the teaching staff.
Once the prerequisites for a competitive market are established, we should abolish central control of admissions into public schools - i.e., the zoning system. This system unfairly restricts admissions into good, so-called public schools to those wealthy enough to pay the in-zone rent or land price premium. Worse, unlike private school fees, the premia for high quality schools do not go to the schools – thereby, encouraging more schools to be good and thus, increase their budget – but rather to incumbent landholders. This is blatantly unfair – what have landowners done to earn these gains? - and inefficient. Instead, we should allow schools to determine their own admissions system, whether that be based on an interview, exam or any other merit-based approach or a fair lottery, thus levelling the playing field for students of all socioeconomic classes. It is crucial to note here that the incentive on schools will always be to admit as many students as possible – because they are paid per attending student.
It is also crucial that we abolish Wellington-negotiated collective agreements for state school teachers. Teachers, like all other professionals, should become employed directly by schools and engage with their employers to negotiate an individual or voluntary collective contract. Teachers and schools would become bound by existing labour relations laws which govern union access and collective agreements. This would allow schools to pay good teachers more and take action if a teacher is providing an unsatisfactory service (i.e., fire bad teachers), thus incentivising teachers to perform at their best. It would also somewhat level the playing field between private schools and their well-capitalized state school cousins and other state schools.
Compulsory qualification requirements for teachers should also be abolished. Teacher quality is undeniably a crucial, if not the foremost, factor in determining educational success. Counterintuitively, however, compulsory qualification requirements are not useful in achieving higher quality levels. For one, this policy has locked out significant numbers of well-qualified foreign applicants from our schools, worsening the teacher shortage. Secondly, evidence from the United States finds little association between qualification levels in education and results. For instance, Gordon, et al. (2006)’s study of the Los Angeles public schools found no statistically significant difference between the effectiveness of certified, alternatively-certified and uncertified teachers, based on mathematics test results. Another study by Buddin and Zamarro (2009) found no relationship between licensure test results and student performance. They also found no effect of teachers having advanced degrees on results. However, there is evidence elsewhere (for instance, Zhang (2008) on science teachers) that higher-level qualifications not in teaching, but rather in subject areas is useful. But given the evidence is unclear and the possible cost (of excluding qualified foreign teachers) is high, this should be left for schools and parents to decide: It could, for instance, be better for some schools to have more, less qualified teachers to cut class sizes.
Next, we should radically reform the school funding system. Right now, the system entails a complex web of direct government grants to state and state-integrated schools for capital expenditure, teacher salaries, and a myriad of other items as well as the private school subsidy. This system allows public schools little effective choice in the allocation of their revenues, meaning they can't prioritise what is most useful to their students.
Instead of this convoluted system, every New Zealand student should be given an annual educational grant and have an Individual Education Account set up for him. The grant should be based on his age, location, and any special educational needs he may have and his parents' income (decreasing as this income increases). Every school (public or private) that wished to be eligible to receive IEA funds, would charge a fee, set by the school itself. However, for students unable to pay any extra fees or the entire amount (as determined by a calculation by the Ministry ), schools would be required to make-up he difference. Needs-aware admission would be illegal. High-fee schools would essentially be required to cross-subsidise poorer students. Remaining funds in Individual Education Accounts could be used to pay for authorized educational expenditures, such as transport, extracurricular fees, tutoring, uniforms and/or stationery.
Parents should be required – at all income levels, including the lowest – to contribute to their children’s fund. An increase to cash payments to poorer families should be made to make this have no impact on their net cash position after this contribution. They will simply personally pay the cash. Higher-earning parents should be required to top up the lower educational grant their children receive to an amount approaching the full level. By relying on the natural human propensity to fall victim to the sunk-cost fallacy, this requirement gives parents a stake in ensuring their children turn up to school and perform to their best – it is their money at stake. It would also ensure the incentive remained on private schools to accept IEA funds and therefore, to allow access by poorer students.
Such a funding system would open the private education system to a vast swathe of New Zealand society - thereby adding further competition to the system and encouraging schools to do their best for all students, not just those currently with the means to leave. It would also give public schools the same ability to determine their expenditure based on what is most useful to their students that private schools already have and allow parents to provide opportunities to their children previously out of reach, like sports and extra tutoring if they require it.
The crucial element, however, is supply flexibility. This has been the underlying flaw of most attempts at school choice. Without the ability of good schools to expand, and bad schools to close, all any school choice project does is redistribute good education (and possibly introduce some competition, creating some general improvements). If the ability opens up, given administrators are interested in budget-maximising, schools with excess demand are sure to expand. This requires two major changes:
First is the reform of the legal structure of state schools. As Crown Entities governed by the Education Act, it is currently near impossible for schools to change their operating structure significantly. Instead, ‘public’ schools should be reformed as self-governing trusts. Their original constitutions should remain at the beginning, but Boards should be free to cooperate with other schools – for instance, in creating a shared governance structure, like a franchising chain, to overcome regional weaknesses in Board strength and to share administrative expenses and curricula – and other organisations, like employers, tertiary institutions, and religious groups. They should be permitted to start new schools too. They should also be permitted to fail. It should be made clear that no under-performing school would be bailed out by the taxpayer. Why should taxpayers be required to put up funds to prop up an unviable school - either through poor financial management or a lack of students? This would create real market accountability and allow failed experiments to fail.
Second are changes to the management of school property. In the status quo, school property is essentially entirely managed by the Ministry of Education who ‘lease’ it to school Boards of Trustees, for free. Such a system continuing would provide incumbent schools with too much of a competitive advantage, particularly given the price of Auckland land: Money saved on land could be reassigned to educational expenses, whereas challengers would have to spend significant amounts of their funding on it. Instead, the school property portfolio should be leased to schools, on market terms. Tenders should be held, open to all registered schools or proprietors who can prove to the Ministry that they would be able to staff and fill a school on the property. There would exist no incentive to challenge good schools – one would not be able to justify paying a higher lease if no latent demand existed. However, bad schools – which parents were hastening to leave – would be unable to sustain their existing leases and could be replaced by better alternatives. Funds earnt from the lease process should be part of the education budget used to pay for the annual education grants.
There should remain some discretion for education officials to bar schools with repugnant or clearly failing practices from receiving IEA funding. However, the competitive pressures of fully mobile funds and competitive land tenders should force bad schools to close naturally. If this does not occur, this suggests that parents must be seriously ill-informed or the barriers to entry for competitors in the area must be very high. That should be the bar officials must meet before barring a school from funding. It should not be enough that an official merely thinks it is uneconomic for a school to be operating in an area, for instance – as was decided during the Christchurch Earthquake for many schools. If the school is uneconomic, it will close of its own accord.
It is true that this scheme could result in some students being required to move schools. Certainly, this would be the case in the early years of any such system. The system would have to re-equilibrate to meet parental demands for schools. Instability, in and of itself, is not helpful for the education of children, however, when this instability results in them moving to better schools and better schools being established for future cohorts of students, it is a necessary evil.
Tomorrow's illiterate will not be the man who can't read; he will be the man who has not learned how to learn. —Toffler (1970, p. 414)
The progressive education movement has been exceptionally adept at capturing the New Zealand education system. Progressivists claim to adopt Toffler’s perspective. Knowledge, they contest, is essentially useless in the Internet age – it can all be Googled. Subjects are useless ‘social constructs’, mere barriers in the creation of holistic learning. Far more important is the development of ‘21st century skills’. To develop these, students should be permitted to construct their learning themselves. Direct instruction – the style of teaching most will be familiar with, where a teacher instructs the students as a class – is outmoded. Indeed, its explicit power dynamic, where students are subjects of the teacher, is “fascist” (Delpit, 1988).
Given my slightly incredulous tone, you may have sensed I am not a great fan of progressive education. Students must be taught deep subject-specific knowledge. Without this, it is impossible to develop the nuanced understanding of the world which surely underlies any ‘21st century skills’ students require. Direct instruction is an evidence-based pedagogical technique not an ideological reinforcement of prevailing capitalist power structures. John Hattie (2012), in his masterful meta-study of research into educational interventions, found a year spent with direct instruction would produce gains equivalent to almost 1.5 years of typical teaching. That would be equivalent to an extra 2 years of high school education. And yes, high-stakes external examinations of the entire syllabus in which the content is surprising are necessary too.  Without such exams, there exists no incentive for students to learn or teachers to teach the full subject.
My point, however, is not to rehash the conservative-vs-progressive education debate. Though I am convinced of my views, I am not arrogant enough to suggest they are formed well enough to force upon every child in the country. Successive New Zealand governments, however, have been.
The Government subsidy for the National Certificate of Education Achievement (NCEA) functions to effectively constrain budget-limited public schools to the NCEA system. Alternative exam systems lack the $28 million of the education budget (The Treasury, 2018) given in subsidy to NCEA. Schools which serve non-affluent communities cannot do as high-decile and private schools do and simply charge parents for the full cost of competing examination systems. Simply look at the list of schools which offer the International Baccalaureate Diploma (IB), an NCEA alternative, in New Zealand. They have an average decile of 9.75 – only two IB schools do not serve the top 10% of students by the socioeconomic status of their neighbourhood. A similar trend is evident in the schools which offer the University of Cambridge International Examinations (CIE), another NCEA alternative. Alternatives are simply not available to poorer students.
NCEA’s competitive advantage is amplified by it being the Ministry, rather than schools, who manage teacher payroll. NCEA’s internal assessment approach entails significantly higher teacher workload. Schools do not feel this extra cost and therefore, will account for it less when choosing the examination system they offer.
The significant subsidy – both implicit and explicit – of NCEA, therefore, amounts to what is essentially a Ministry diktat governing the examination systems offered at non-elite schools. What examination system a school offers has a significant effect on how it teaches. The highly-modularised, ultra-flexible model offered by NCEA inherently de-emphasises the importance of subject-wide knowledge. It also significantly constrains the scope of exams. Teachers, therefore, do not need to teach the entire subject so students can meet any content within the syllabus in the exam. Instead, the content is relatively known before the examination, increasing the utility of ‘teaching to the test’. NCEA is the quintessential progressive qualification. By contrast, alternative systems like CIE and IB mark students compared to their peers, rather than ‘standards’. They provide unfamiliar questions in high-stakes examinations. In English Literature, they make reading the Western canon compulsory. Both require deep and broad knowledge by students across set syllabi so they may perform well, essentially requiring direct instruction. They are inherently conservative. Indeed, one progressive New Zealand former headmistress described exams themselves as “a colonial system” (Milne, 2018) – those set and marked in the United Kingdom are surely even more so.
Whichever is better, international or NCEA, examination systems are clearly not a pedagogy-agnostic measure of achievement. They have clear and significant impacts on how students are taught. There is the significant diversity in opinion on how students ought to be taught among teachers and parents. Given their power and the lack of unanimity about the right course of action, there is inherent danger in any government policy which significantly favours one examination system over another. This is particularly bad when such power is only effective over poorer students.
Instead, Government should reallocate that $28 million of funding for NCEA to the general fund for secondary education. It should be distributed to schools for them to spend in the best way they see fit, whether that is on NCEA, CIE, IB or any other qualification – perhaps, Auckland Grammar’s new in-house Pre-Q examination system for Year 11, an Australian state’s qualification, or the American Advanced Placement examinations. There exists significant competition in the assessment market. As is the case today, universities – both foreign and domestic – would probably be the principal arbiters of quality. The accountability of university and employer requirements would ensure schools did not engage in rampant grade inflation.
NZQA should also be restructured. The regulator of qualifications should not concurrently be a provider thereof. An organisation with so obvious a conflict of interest simply cannot be expected to organise a credible exam system. Instead, a model like that of England should be adopted. There, exams are administered and syllabi set by private examination boards. A separate ‘exam watchdog’, called Ofqual (the Office of Qualifications and Examinations Regulation), manages the quality and comparability of qualifications at the same level. I would propose, in New Zealand, the administration and syllabus design for NCEA be spun off into an independent trust, comprised of Government, university, school, and employer representatives. In addition to rectifying the conflict of interest, this would helpfully insulate NCEA from political desires to manipulate pass rates to reflect better on the Government. A merger could then be considered between the remainder of NZQA and the Educational Review Office – both would have the fundamental functions of insuring the credibility and quality of the New Zealand education system and providing good information to regulators, parents, and employers. Such an independent Education Standards Commission would be a much more credible guarantor of exam standards and bulwark against grade inflation than the existing conflicted NZQA.
New Zealand is a sad story. The country had the world’s best system between 1993 and 2000. Loans covered fees and living costs. Income contingent repayments were collected as a payroll deduction. … A mistake was political failure to explain the system properly, leading to the introduction of very expensive interest subsidies. — Barr (2014, p. 73)
Even before the introduction of Fees Free by the Sixth Labour Government, the average tertiary student paid less than 18% of the cost of his degree (Productivity Commission, 2017). Taxpayers fund the remainder. That funding includes both written-off interest and direct government subsidies to universities and other tertiary education facilities. Such a significant level of subsidy seems excessive.
Tertiary graduates will typically earn significant wage premia as a result of their education. In New Zealand, the average university graduate earns a premium of 170% over the average weekly wage of those with no qualifications. Similarly, those with Diplomas or Certificates earn an average income premium of 116% over those with no qualifications (Ministry of Education, 2018). Of course, there is some selection bias in those figures – those who attend university would, on average, probably earn more than the average even if they had not. Still, however, it is clear there are significant private returns to tertiary education. This suggests that, even in the absence of significant subsidy, many would still undertake it and that, if they were required to pay more of the full economic cost of their degree, many graduates would be able to do so (presuming they had access to finance). It also suggests that subsidies for higher education are regressive. In absolute terms and, presumably, as a percentage of income, the well-off would – on average – have received far more government support for tertiary education than the poorest.
Of course, there are possibly some positive externalities of tertiary education. The rest of us might gain from university graduates gaining an education. The first and obvious channel is through lightened fiscal pressure. Graduates earn more and are far less likely to be unemployed. They will, therefore, contribute more in tax – lessening the load on the rest of us – and use fewer social welfare resources. However, as laid out in the Externality Argument section above , such externalities are irrelevant to policy analysis – subsidising them is equivalent to simply cutting all taxes on capital (that is, however, advisable). Another plausible channel for external benefits is through the improvement to democracy. Friedman made a similar argument, discussed above, for elementary schooling. Graduates could improve democratic decisions, benefiting the rest of us, by making more informed votes or being good politicians. Such an externality is likely to be insignificant per graduate however – given he would account for only 1 in 3.3mn voters. It is also unclear, given that degrees in New Zealand are not typically liberal arts in nature, that every tertiary course is useful at all in this respect. There could also be an externality whereby increasing the productivity and/or knowledge of gradates also increases the productivity – and therefore, wages – of other workers. The best work on this, however, from Acemoglu and Angrist (2000) finds no statistically significant external benefits of human capital investment. Overall, it is difficult to find any good justification for 82% of tertiary education being funded by the public. Most of the gains seem to go to graduates. Why then should their education be funded by taxpayers?
The fundamental argument against state subsidies for higher education is one of both efficiency and equity. Why is an investment in human capital so much more desirable than an investment in physical capital? One suspects the answer lies in the intellectual snobbery of ministers and their advisors, almost all of whom attended university. The Fourth National Government’s Ministerial Consultative Group on tertiary education raised the pertinent counterpoint: “Similar grants are not made available to people who wish to establish a business. For example, a young person buying a herd to become a sharemilker could not expect the taxpayers to meet 80% of the costs.”  If human capital investment is to be subsidised, surely physical capital investment should be too.
That we significantly subsidise one type of capital accumulation but not another seems a recipe for inefficiency – some young people who would better spend their time accumulating physical capital (like cows, for instance) are diverted by the subsidy into a suboptimal use of one of their most valuable resources – time.
The flawed capital markets argument discussed on page 12 is the strongest argument for government involvement in funding tertiary education. In a fully-efficient capital market, would-be students who could not fund their education upfront could simply go to a bank (or similar lender) and request a loan to pay for their student loan. Banks would be willing to give this money to those who would be able to pay back that loan in the future. Unfortunately, for the reasons laid out above, such a market does not and cannot exist. Government, therefore, has a reasonable part to play in providing repayable loans to pay for tertiary education. This argument, however, only extends to the provision of loans at terms the market would provide them if it could; it does not extend to the subsidisation of those loans.
Using such a justification would require student loans be offered at significantly higher interest rates. No reasonable market lender would offer 0% student loans. The interest rate on any given loan can be understood as a function of both the risk-free rate of return a lender could earn if he were to reallocate his funds elsewhere  and the risk of default by the borrower. The risk-free rate of return is typically said to be the return on high-quality government securities. Therefore, even if we assumed (ludicrously) that lending to an individual student was as safe as lending to the New Zealand Government, the market would require interest rates of around 3% on student loans. If interest rates are set below the Government’s cost of borrowing, every dollar lent out in the Student Loan Scheme would (and does) create a loss, even if every loan was repaid, in full and on time, because the Government must borrow the money from global credit markets at a higher interest rate than it lends it out.
Before the Fifth Labour Government removed all interest on student loans, the interest rate was set in a relatively rational manner. It equalled the Government’s cost of borrowing plus one percentage point. Treasury had estimated that the default rate on student loans would be approximately 2%. The addition of only 1% to the cost of borrowing meant that taxpayers assumed about half the default risk themselves. If one strictly follows the ‘flawed capital markets’ argument with no allowance for externality, such an assumption of risk by taxpayers is unjustified. I do, however, think there are some (relatively minor) externalities to tertiary education (such as the impact on democracy), justifying this small interest subsidy.
It is unclear to me why tertiary education needs subsidy greater than this. If one wishes to encourage job growth and productivity through capital accumulation, simply cut all taxes on capital to do it in an even-handed and efficient way. If one wishes to subsidise basic research – a useful role for government, do so. Use mechanisms like the American National Science Foundation’s contestable grants; attaching research funding to education is an unnecessary overcomplication and certainly less efficient.
We should return capital investment to the control of individuals and level the playing field. Tertiary education subsidies should be cut to zero, including those for apprenticeships, trades programmes, and the first year of university. All future student loans should be charged interest, calculated in the pre-2000 manner, accumulating from the beginning of tertiary education. Repayment should remain income-contingent and part of the PAYE tax system for New Zealand residents. We should also significantly increase borrowing limits for living expenses.
This shift to an almost entirely user-pays model of higher education would mean true market forces would re-emerge in the higher education market. When, as is the case today, students can easily access ultra-cheap money, the incentive for students and universities keep costs low is dampened. Inevitably, Government has had to impose price controls on course fees to avoid an explosion in costs. When it is students who will bear (eventually) the vast majority of the cost of their education, such price controls will no longer be necessary. They will create downward price pressure themselves. In the absence of price controls, higher education providers are more likely to innovate and produce higher value products which they can charge more for. This could be a route to the creation of the first truly world-class New Zealand university.
Such a scheme would not reduce the access to tertiary education by disadvantaged students. There are two principal barriers to tertiary education for poor students, both of which are bettered by this approach: Poor secondary education and living costs. Less than 30% of Year 13 students at Decile 1-3 schools achieve University Entrance (the bare minimum required to attend university), compared to almost two-thirds of students in Decile 8-10 schools. (NZQA, 2018). A fully user-pays model of tertiary education would encourage universities to accept as many students as possible and could allow students who had been poorly served at high school to pay extra (on loan) for intensive study to bring them up to speed. In the second case, because taxpayers will no longer be significantly subsidising student loans, the binding constraint which has held down living costs support is removed. Now, allowing students to draw on their student loans for living costs will be much less costly for taxpayers and such support can be expanded significantly.
In addition, a user-pays tertiary model will save a significant amount of taxpayer money which can be reallocated to removing the actual constraints which reduce tertiary attendance by poorer students. In FY2018, the Government reported student loan interest write-offs alone accounted for $594 million in subsidy to tertiary students this year. If that funding was reallocated to senior secondary students in Decile 1-3 schools, it would represent a funding boost of almost $22,000 per pupil – or 2.5 times the current average spending per pupil in the senior years of high school. A dramatic increase in funding for poor secondary schools would achieve far more for access to tertiary education than continuing to provide interest-free loans to all students, no matter their ability to provide for themselves or the utility of their course.
It is also unclear to me why universities and polytechnics should be in state ownership. That they are seems simply an artefact of times past, when we had a far more centralized, elite university system, which closer resembled a system of training colleges for the professions, the civil service, and academia than the diverse range of universities and polytechnics we have today. Already, each campus has significant autonomy from the Crown. Ministerial diktats are all but expressly forbidden by the Education Act. Still, however, their status as public organizations creates some strictures – for instance, discouraging private donations, because would-be donors feel this is the role of the government - which could be avoided where they to be restructured into a legal form which reflected their independence. I would suggest charitable trusts.
The fundamental error of New Zealand’s educational system for many years has been a lack of trust in individuals and families. It should be recognised that, though they are clearly not perfect – far from it, families are better aware of their own circumstances than any Wellington-based bureaucrat (or regional hub staffer) could possibly be. Decentralising decision-making too reduces the risk of poor decisions having disastrous systematic impacts. Additionally, it ensures the continued diversity of our country and generates much needed competition.
Education, just because we view it as a ‘human right’, does not become exempt from the normal rules of economics. The sooner we abandon hopeless romanticism about the inherent value of state-provided education, the better. Ends, not means, should be the focus of education policy.
- Create a centralized information system for access by parents, outlining simple metrics for the performance of schools
- Reform state schools into independent trusts, permitted to buy and sell schools, start new schools, cooperate with other organisations, franchise, borrow money, and go bankrupt
- Abolish school zoning and permit schools to create their own enrolment schemes
- Replace Wellington-negotiated teachers’ collective agreements with direct negotiation between teachers and schools
- Repeal teacher certification requirements above those required for the health and safety of children
- Abolish central funding of schools and create individual education grants, with required parental contributions
- Permit all schools, state and private, to set their own fees and allow education grants to be spent on all such fees
- Require high-fee schools practice needs-blind admissions and provide needs-based scholarships to be eligible for education grants
- Charge market rents for school property, with regular, competitive tenders for its use by education providers
- Remove the subsidy for NCEA, reallocating funds to general secondary school funding
- Spin-off NCEA from the NZ Qualifications Agency into an independent trust
- Merge NZQA into the Education Review Office, creating a super-regulator responsible for the maintenance of standards across the education market
- Reintroduce interest to all future money lent under the Student Loan Scheme – equal to the Government’s cost of borrowing plus one percentage point
- Significantly increase the borrowing limit for living costs
- Abolish all other subsidies for tertiary education, including Fees Free
- Abolish tuition fee caps
- Reallocate some funds saved by abolishing tertiary subsidies into the senior secondary education of poorer children
- Convert the universities and polytechnics into private charities
Primary and Secondary Schools
 What follows is a selective survey of the most important arguments for government involvement in education. The names I have given the arguments aren’t necessarily the most widely-known ones, but merely ones I have chosen for convenience.
 The only exception to this is how higher earnings from vocational training interact with the tax system and social-welfare system. Presumably, higher earnings from more training reduce expenditure and increase revenue. However:
1) Assuming the abatement rates/marginal tax rates on these systems are not punitive, consumers should account for this higher income in their own calculations to at least a reasonable degree;
2) If we were to count higher tax revenue as a positive externality, we would – were we to rationally extend this to every other revenue-increasing measure – have to subsidise literally every capital-deepening expenditure. This is equivalent to simply cutting taxes
 Quoted in accordance with the Ministry of Culture and Heritage (2017). Clarence Beeby, a New Zealand education reformer, could possibly be said also to be the author of this excellent quote but this is immaterial: Fraser certainly agreed with its sentiments.
 This one is complicated. While an increase in budget certainly eases management – it reduces cost pressure and the need to make tough decisions like restructuring – a large budget in absolute terms is presumably more difficult to manage than a small one.
 Such a calculation should take a similar form to that used by universities in the United States who commit to meeting the ‘full demonstrable need’ of their students with financial aid. In particular, it should consider the fees (and other costs of attendance, including transport) as a percentage of paternal income, while considering the number of other dependents in the family.
 I owe this argument about the necessity of surprise in assessment to Briar Lipson (2018)’s excellent critique of the NCEA system.
 See p.33
 My thanks to Khyaati Acharya and Dr Eric Crampton for drawing my attention to this report and, in particular, this wonderful quotation in their excellent report for the New Zealand Initiative, Decade of Debt: The Cost of Interest-Free Student Loans, (Acharya & Crampton, 2016).
 This, of course, is why consumer and business lending rates are sensitive to the Official Cash Rate set by the Reserve Bank. The OCR (minus 25 basis points) is a risk-free rate available to registered banks on overnight deposits at the Reserve Bank.