Poor returns to tertiary education

Tertiary education was quite a theme in the recent election campaign. In my household – with three kids likely to go to university in the next decade – promises to reduce the direct costs of tertiary education were tempting.  But resisting temptation remains a virtue.

A few days ago I noticed (thanks to Jim Rose) this chart

lifetime benefit of a degree

It isn’t a new result. These OECD data have shown for some time that the economic returns in New Zealand to getting a degree are pretty low relative to those in other advanced countries.   Such results even prompted Treasury to commission some external research on the gap in private returns.

In the chart – from a few years ago – whoever put it together has highlighted two groups of countries: the Nordic and Benelux countries on the one hand, where there are already lots of skilled people, and high income taxes, and former eastern-bloc countries which are now catching up to the rest of the advanced world, and where skills are in high demand, and able to command high returns. I’m, of course, more interested in the contrast between New Zealand and those central European countries.  As I’ve written recently, 25 years ago both we and they were looking to reverse decades of poor performance and catch-up with the other advanced countries. They’ve made progress in that direction. We haven’t.

Since the net benefits are shown in dollar terms (rather than, say, as a per cent of GDP per capita or of lifetime earnings), it is probably reasonable to expect that poorer countries will be bunched towards the left of the chart. And there one finds Turkey, Greece, New Zealand and Italy. But that clearly isn’t the bulk of the story. After all, even though they are now catching up, all six of the former eastern bloc countries shown still have levels of GDP per hour worked and/or GDP per capita similar to or (generally) below, those of New Zealand.

I had a look at a few background documents from the OECD. If anything, as we shall see, the New Zealand numbers may be even worse than what is shown in this chart.

It is important to recognise the distinction the OECD draws between private and public costs and benefits. Some of these things can be easily measured (eg upfront private fees, or direct public grants to institutions or individuals). Others are more approximate. (The other aspect, which I’m not sure any of these particular indicators attempts to account for is the selection bias, in which the typical person who undertakes tertiary study has other traits – eg intelligence – that mean that they would probably earn more in the labour force than the average person who does not undertake tertiary study.)

This chart is from a few years ago, and tries to break down the costs of tertiary education (in this case for a man). In New Zealand, as in most countries, the largest private cost by a considerable margin, is the foregone earnings of the student themselves.

tertiary costs

These OECD indicators assume that students do not work while studying.  In the latest OECD Education at a Glance they show estimates for 15 countries as to how much difference it would make to include reasonable estimates of actual student earnings. For New Zealand, doing so would lift the estimated returns to tertiary education by around 15 per cent, more than for most of the other countries shown. However, as you can see from the first chart above, a 15 per cent lift in returns to tertiary study in New Zealand would not alter our relative position on the chart.

The other aspect of the calculations which often doesn’t get much attention is the appropriate discount rate to use in making these calculations. It matters a lot – the costs are mostly incurred between, say, ages 18 and 22, and the economic benefits accrue over decades. A decision by an individual is a very long-lasting investment project, with significant irreversibilities (the years spent on education can’t be reclaimed).

The OECD at present adopts a very low discount rate.

The NPV results presented in the tables and figures of this indicator are calculated using a discount rate of 2%, based on the average real interest on government bonds across OECD countries. However, it can be argued that education is not a risk-free investment, and that therefore a higher discount rate should be used.

I’d say there was no ‘it can be argued” about it. No sensible government would do a cost-benefit analysis of building more schools or universities using a discount rate of 2 per cent. The New Zealand Treasury, for example, uses a default discount rate of 6 per cent real. And as an economic proposition, an individual’s tertiary education is a pretty risky proposition, with few effective diversification options for most people.

As it happens, in the latest Education at a Glance the OECD presents a table illustrating, to some extent, what difference it makes to use a higher discount rate.

discount rates.png

Using a discount rate of 5 per cent (real) reduces the estimated benefits by around 60 per cent (relative to the 2 per cent baseline) – and these numbers are for a man, and in most countries the net benefits to tertiary education for a woman are (on average) lower than for a man.

This issue matters particularly for New Zealand which has a higher risk-free interest rate on average than any of the other countries in the table. The gap is large. On Friday, the real interest rate on the New Zealand government’s longest (23 year) inflation-indexed bond was 2.39 per cent, while that for the US government’s 20 year indexed bond was 0.77 per cent (and US yields are far from the lowest in the world). A margin of 1.5 percentage points above “world” rates hasn’t been a bad guide for New Zealand interest rates over recent decades.

Even a 5 per cent real discount rate appears too low to evaluate a personal decision to invest in a tertiary education in New Zealand. But if one takes the results for New Zealand in the table above when evaluated at a 5 per cent discount rate, and then compares them against the results evaluated at 3.5 per cent for other countries (to capture that persistent difference in real interest rates), only Latvia would offer lower returns to tertiary education than New Zealand does.

And bump up the discount rate a little more and the estimated net returns to tertiary study will soon be approaching zero or going negative.  And, remember, those estimates are for a man. The average female returns are even lower.

People will have a range of reactions to these sorts of numbers. Some will take them as supporting proposals to reduce tertiary fees or increase student allowances. Such changes might increase the net private returns to tertiary education, but they won’t (of course) change the all-up net returns (someone still has to pay).  Others seem to see tertiary education as some sort of “merit good” that people should have the opportunity to undertake, at moderate expense, whether there is an economic return – to them, or the public more generally – or not.  And, of course, for some people and some courses, a tertiary education is more akin to consumption than investment (which is not intended as a criticism).

For me, I see them as yet another marker of the failure of the economic strategy pursued by successive governments over recent decades.  Our remoteness means it is very difficult to generate consistently high returns to anything much in New Zealand for very many people. The determination of our governments to quite rapidly increase the population here, despite those apparently limited opportunities, just compounds the problem. It does so directly – the limited natural resources (our one distinctive advantage) are spread over ever more people – and indirectly, through a persistently overvalued real exchange rate and high real interest rates.

Returns to tertiary education in New Zealand are probably quite reasonable for those New Zealanders who get an education here, and then leave (but that is probably a poor investment for the taxpayer). For many of those who stay, it looks like a distinctly marginal proposition. Attempting to bring in lots more skilled people from abroad – most of whom aren’t that skilled anyway – just compounds the economic problem, even if the New Zealand taxpayer doesn’t have to pay anything for their tertiary education. There just aren’t the good economic opportunities here for a rapidly growing population, and increasing subsidies to tertiary education would seem likely to further exaggerate the evident imbalances.

In an economy that was making progress towards reversing decades of relative economic decline, there is good reason to expect that returns to investment in tertiary education (like other prospective investment returns) should be consistently high relative to those in other countries. Sadly, those returns appear to be consistently low in New Zealand – especially when evaluated at an appropriate discount rate. And, of course, we are making no progress at all in closing those productivity gaps.

33 thoughts on “Poor returns to tertiary education

  1. Michael can I suggest that you test your contention that distance is the primary reason for NZ’s poor economic performance (which you ably describe and I do not have a fault with) by including Australia in your analysis. Their cities are also geographically isolated from the rest of the global economy. Perth is often quoted as the most geographically isolated city in the world for example.

    Liked by 1 person

    • Brendon

      Yes, i do need to do a post or two along those lines at some stage. To my mind the only real difference between NZ and Aus is that they discovered a lot more useable natural resources they didn’t know they had. It has enabled them to maintain real productivity levels rather nearer those of the OECD leaders – even with lots of ppopulation growth and policies that aren’t much different than NZ’s. But they are still well behind even places like Netherlands, France and Germany, and (like NZ) in modern history they had usually been wealthier and more productive than those leading European economies.

      On Perth specifically, a great deal of the new mineral wealth (esp per capita) is in the West. And despite the formal isolation – distance to another city – they do have a few – probably minor in the scheme of things – compensations (eg similar time zone to, and closer physically to, places like Singapore)


  2. When the NZD is the 11th most traded currency in the world after China then it really has got nothing to do with population when you compare China has a billion plus people compared with our tiny 4.5 million. It has to do with overseas speculation. It is a rather long bow that you draw to make a connection between more people and the higher exchange rate.


    • NZ Interest rates have more to do with the monopoly power that is clearly exhibited by our Australian owned banks. HSBC is now offering 3.87% interest rates on 18 months fixed. Our monopoly banks are still stacked around 4.85%. ASB has made a across the board drop last week but they were artificially tiny falls to stave off Commerce Commission enquiry at the extreme lack of competition.


      • Jim Anderton original speech in 2001 when he set up Kiwibank

        “We had been looking at what the large overseas-owned banks were doing.
        Fees were increasing.
        Branches were closing.
        Staff were being laid off in their thousands.
        Some towns were left without any banking services at all – not even an ATM.
        It didn’t escape many people’s notice that they all seemed to charge exactly the same interest rate.
        And they all announced increases in interest rates within a day or two of each other.
        Many people felt that the banks just didn’t want them.”


        This was Kiwibanks original purpose was to provide competition however since Anderton retired, the original intent of Kiwibank has been lost. It is now just aligned with the big Australian boys. Paul Brock CEO of Kiwibank for 7 years until his recent resignation, was appointed out of Westpac. Best mates I guess. Join the monopoly club and make record profits at the of New Zealanders and collect a fat bonus check as a result.


  3. Michael I am re-reading Edward Glaeser’s book ‘Triumph of the City’. As part of his analysis, academia (originally a place just north of Athens) and higher learning in general are products of cities. His describes that cities are a product of organic forces (private enterprise) and state institutions.

    A strong state providing rule of law, good urban planning (the Greeks had Hippodamus for example) and necessary public goods. Cities (and the surrounding countryside) were divided into public and private spaces -the public spaces needed to provide transport links and public services -aqueducts for fresh water, sewer systems etc to service the cities.

    Glaeser argues this was well known in antiquity -in the mediterranean world of Greece and Roman times. Following the collapse of Rome -states were not powerful enough (or motivated enough) to provide the needed institutions and cities collapsed for centuries -it taking the Renaissance to revive them.

    Michael I am not comparing NZ’s current leader’s to barbarians like the Goths, Huns and Vandals. NZ hasn’t experienced such an absolute decline but you have to wonder if we are on the slippery slope down to barbarism. When our biggest city has the third most expensive housing in relation to income in the world and the proposed next motorway -the East/West link -a project in Roads of National Significance programme -a scheme championed by our current Finance Minister, is on a per km basis the most expensive road in the world -the same price as a corrupt Russian road in Sochi (made for the winter Olympics) it makes me wonder if this slow decline into barbarism is NZ’s problem -not the distance from the rest of the globalised world.


    • It is our focus on primary industries that has misdirected our capital resources. Our capital resources should have been focused on the building of industries like a Samsung or Nokia equivalent. Primary industries are the prime industries of the barbarians like the Goths, Huns and Vandals. I would add the Roman Empire on that list.


    • “” academia … and higher learning in general are products of cities”” is this actually true? Presumably by city you mean an extended urban area not anywhere with a cathedral or a city charter. OK large cities usually have a university just as they have an art gallery, opera house and sport stadiums – status symbols. But the real ‘pushing the boundaries’ places of highest learning are sometimes found in small places.
      In the UK Cambridge is a small market town and has been for the 800 years it has been a centre of higher learning. Never been to Oxford but I suspect it is similar. Cambridge Mass has MIT and Harvard but a total population of just over 100,000. Princeton is even smaller.
      Surely removing all the university buildings from Auckland CBD would leave Auckland a less congested and more successful city with cheaper property. If the academics moved to a small town they might be just as productive.


  4. A tertiary education trains people for the workplace targetted at middle management and higher management. This means you need large corporate offices. To get larger corporate offices you need head offices. To get head offices you need to incentivise corporates through lower tax offerings or you have a large domestic market or a large export market that allows corporates to have headoffices to service that large market.

    Fonterra has a large corporate office because it has a large export market. Even though farmers struggle with lower prices, the add value products make a higher margin in exporting value add products which have seen Theo Spearings getting his $8.3 million salary due to increased profits on the add on business even though overall profits declined by 11%.

    Clearly the depth of our capital markets does not support large manufacturing and infrastructure entities. There is a case for government to get involved with developing key industries that had to be government funded to get those industries off the ground eg NZ Air, Telecom, BNZ, Postal service and even those think big projects of the Muldoon era like Oil refineries, Marsden Point Oil refinery etc.


  5. The tyranny of distance

    An example of distance as a factor is Iron Ore

    The two largest suppliers of Iron Ore are Brazil and Western Australia
    Australia’s Iron Ore is the highest grade available – more so then Brazil’s – and thus more expensive – and also acquires a distance premium because the distance from WA to China is much shorter than China to Brazil


  6. Continuing on the theme of NZ being on the slippery slope down to barbarism as the reason for failing cities, failing returns to tertiary education, poor productivity etc.

    Dr Jonathan Coleman a cabinet minister of the last 9 years and a man who put himself forward as one of the candidates to replace John Key when he stood down, recently tweeted from his car a complaint about road works causing him travel delays. Jonathan is the MP for Northcote -the photo in his tweet was not far from the north side of Auckland Harbour bridge -so near the heart of Auckland. Jonathan given his local and national responsibilities should have good working knowledge of urbanisation issues.

    Unfortunately for us -he is pathetically ignorant (and that is not unusual in NZ). I would argue to degree that in urban planning knowledge terms Dr Coleman is the equivalent of a surgeon advocating ‘bloodletting’ as a reasonable treatment.

    How can Jonathan Coleman only see the road works -not that he and the long line of cars in front of him are an economically inefficient use of space, that their isn’t any alternative transport options (including a more open/grid-like public road network). That his government has allowed so many new people into the country -about 1000 people into Auckland a year -that given Auckland’s poor funding to alternative forms of transport this means about 800 extra cars a week. How does Jonathan not acknowledge that the current funding arrangement for Auckland Council means it is up against debt limits -so they cannot make any transformational improvements to transport in Auckland. While his government which does have fiscal headroom has gone on a binge of motorway building without undertaking any cost benefit analysis -so has done bugger all to improve the situation for car congested roads in Auckland?

    How is it that after the screeds written by myself, Greater Auckland, Interest.co.nz etc about housing, transport, urbanisation etc that the likes of Jonathan Coleman do not get it.

    P.S Jonathan should read my paper https://medium.com/land-buildings-identity-and-values/successful-cities-understand-spatial-economics-95c272ac04c9

    Just to blunt (because that seems to be the only thing our slipping into barbarism leaders understand) I see the likes of Jonathan as the ignorant bloodletter in the first picture of the paper.


    • Sorry I meant about a 1000 people a week (not a year) into Auckland -meaning about 800 extra cars a week on Auckland roads.


    • Perhaps National is focused on roads because roads are cheaper to build. The costing for Labours monorail is anywhere from $5 billion to $15 billion which they have no budget for. Len Browns $2 billion intercity rail from Britomart to Mt Eden is now costing upwards from $3.5 billion and it is not even near completion yet. When Steven Joyce said that Labour has a $11.7 billion hole in their budgets. He is being very kind to Labour because that hole is very very conservative. Its just that our 7 NZ economists can’t do maths.


      • I wonder if Labour or National has budgeted for the Americas Cup? Phil Goff seems less than interested in hosting the Cup with the loss of the revenue from the Ports of Auckland. I don’t think that Aucklanders will want a rates rise just to fund the Cup.


      • Apologies Michael. I am sure Jonathan Coleman on a personal level is a lovely man. I think though he needs to be a lot more curious, reflective etc about the systems that sustain cities like Auckland which he has a responsibility for.


      • Brendon, the Northcote electorate believes he does a great job. He has been their elected MP now since 2005. Thats already 12 years. He has been re-elected another 3 years which makes it 15 years he will have been elected by the people of Northcote. Your comments do not do him any justice. Clearly a very capable, commited and able MP.


  7. After your last article about the RBNZ unlawful act, together with this article – I have to ask – with your wellington grape-vine and access – are you being heard? – serious question


    • Plenty of well-connected people do read this blog. Whether it changes anything very much is another matter, at least in the short-erm. My goal is more about influencing thinking, and providing considered perspectives, rather than immediate change, since the latter is mostly about politics and short-term positioning. But I do see occasional small glimpses of impact – glancing thru the Rb Board Annual Report I noticed odd references that almost certainly wouldn’t be there if (a) I hadn’t been asking annoying questions, and (b) if those questions/concerns could just totally be ignored.


      • After your many writings about the “unlawful act” I was appalled that “they” continued on regardless


      • People feel safe is they have a lawyer’s opinion – even though it is no more than an opinion. Thus I ‘m pretty sure their reaction to this blog was the unexpectedly prompt release of the summary of the Crown Law opinion they were using for backing.
        I’m not persuaded, but officials, ministers, trustees, directors etc all seem to feel better for being able to say “we took legal advice”. And, of course, it is better than not doing so.



      • This is a caretakers role so unlawful is rather harsh. There are no penalties unless there is actual loss suffered which in a caretakers capacity you would not expect dramatic decisions to be made and highly unlikely to be made.


  8. Do you have anything on the social returns to education in NZ – particularly the marginal returns to the great educational expansion.
    My guess is that th return would be negative.


    • What “great educational expansion”

      Infrastructure Deficit – black-out

      In the last 5 years 500,000 migrants have been added to the mix, In spite of that, In those 5 years the number of operating schools in NZ has gone down – By intentional design? – we don’t know

      Then, in the 2017 budget the National Government generously announces it will build 6 new schools, with 2 in Auckland. Given that 80% of all migrants disappear into Auckland, is that enough?


      • I think Ian means the huge increase in tertiary participation over trhe 20 years or so to around 2007. Tertiary participation has dropped back quite a bit since the.

        The secondary school I went to in Akld had a 3rd form of 300+ and seventh form of just over 20; pretty much the cohort that went to university. I imagine your experience a decade or two earlier wasn’t that much different.

        Liked by 2 people

      • 500,000 in 5 years is gross immigration and does not factor in the loss of migrants through emigration.


  9. Correlation is not causation. I find it a logical leap to say that Bachelor of Fine Arts or BA in Women’s Studies are related to any improvement in intellectual capacity thus resulting in higher future income.

    Generally smart people go to university but university doesn’t make one smart.
    I was looking into research to the IELTS english language research for instance. The test results were not different in candidates who did the test before university and then repeated it after university. University made no difference to their English language capabilities for example.

    It had been useful for employers as a sieve to find employees who are more intellectually able. As the numbers of graduates increase, it is becoming a less discriminating guide for employers.

    A common anecdote is a University degree is becoming required for entry level positions. The ubiquitousness of tertiary education is devaluing its usefulness to it’s holder.

    Another anecdotes is that some of the richest men in the world are University dropouts (Mark Zuckerberg).

    Not to say that University of little value but the asumption is that more tertiary education equates to increased wealth for the population is wrong.

    For example in Australia, it has demand driven university places and the universities are incentivised to place bums-on-seats. They have vocational-type university places such as dietetics, where only 5% actually find jobs in that occupation. Waste of resources.

    There must be an ‘optimal’ number of university places for a given type of economy.


    • I’ll think you’ll find those Silicon Valley university drop-outs actually benefited from elite education as children – heck, Zuckerberg dropped out of Harvard, not out of a Lincoln University Dip. Ag. And Wikipedia tells us Mr. Zuckerberg had a private tutor for software development.

      And a wise professor once explained to me that a university is a library surrounded by buildings (which historically housed and fed scholars). Trying to wring a profit out of this system has challenged better brains than NZ can draw on (and I also believe, as with a previous poster, that NZ suffers from a massive lack of middle managerial talent).


  10. Some random thoughts:

    1) I’m not sure that the large private costs of foregone income are real – this assumes a job market with infinite supply. At the macro level dumping that many students into the employment market (instead of being in tertiary education) is going to leave broadly the same number of people overall in the economy at the macro level unemployed given the prevailing inflation rate and short term interest rate. There would be an overestimate of the effect though as some might leave NZ and get a job in, say, Australia.

    2) The level of student debt on the government books is substantial. The total student debt reached $5 billion in 2002. As of December 2014 student loan debt was $14.2 billion. This seems like a clear policy failure & the debt at some time will be written off.

    3) Given the intergenerational inequity of previous generations having had free university education and now also having disproportion capital gains on their houses, from an intergenerational perspective (near) free tertiary education would seem fairer (not that 2 wrongs make a right)

    4) Businesses love immigration as it provides them cheap (subsidized) labour with skills that they don’t have to train. To level the playing field the government should not allow free immigration for work purposes or immigration without a job contract in place. The immigration fee should be set at the same cost that the employer would have to pay to get a New Zealander trained to the same level. Given the costs would now be equal the employer would be free to choose who to employ.

    If set up & priced correctly this should lead to more employers sponsoring and bonding university students and employing them during the holidays to job train them. This would help skew tertiary education towards the job skills that are needed as students unable to get sponsored/bonded positions would be left facing the full cost (there may still be some government funding/sponsorship) of their tertiary education & a hard decision to make whether to do an irrelevant degree. This would also push tertiary degrees towards being more job market relevant.


    • On your first point, I don’t agree. It is a version of what economists’ describe as the “lump of labour fallacy”. In the same way, a massive trend increase in female labour force participation hasn’t boosted the unemployment rate (and, actually, neither does immigration).

      0n your final point, this was the broad direction I was pushing towards in a post last week, and here https://www.newsroom.co.nz/2017/10/01/50861/how-nz-first-could-achieve-its-migration-cuts. A fee of $20000 pa or 20% of salary would soon rather dramatically reduce the use of work visas, for anything other than really unexpected or short-term labour market stresses.


  11. Michael,

    Re the first point – you are correct given the market in reality is dynamic, time dependent & subject to changes in interest rates, inflation rates, costs, exchange rates & immigration rates etc.

    My comment was subject to “given the prevailing inflation rate and short term interest rate”, I was thinking about the instant static effect of dumping tertiary students into the market.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s