Most everyone working in tertiary education cares about equitable access. But it’s a complex and challenging problem.
In the first article of this series, I pointed out that fees free was a sincere, well-meaning attempt to address access barriers in tertiary education. I also dissected the design of student allowances. The allowances scheme has often been presented as helping deal with access problems; but the analysis in Part 1 highlighted design flaws that render it all but useless in dealing with barriers facing young students.
Let’s dig deeper. Let’s look at what does create barriers. And look at how a better understanding of barriers might shape future policy shifts …
The three-stage model
In part 1, I set out a three-stage model[1] of enrolment decision-making:
- Value: the would-be student must see value in the decision to study. Is study worth it?
- Possibility: if the would-be student does decide to study, does he or she have access to the cash, the liquidity that makes it possible to study.
- Debt aversity: If some or all of the financial support is through a loan, is the would-be student debt averse? Are some aspiring students opposed to having a loan, either on principle or through fear of debt?
I used that model to explain the failure of the country’s student financial support system to address problems of access to tertiary education. The loan scheme – providing interest-free finance to meet the cost of fees and living costs – makes tertiary study possible for the aspiring student. And the fact that loans are interest free and that repayments start only when the borrower’s income meets a threshold helps mitigate (if not eliminate) the fear of debt. The loan scheme is designed around stages 2 and 3 of the model.
But the student allowances system – which complements the loan scheme and which is presented as targeted to those who face financial problems – is all but useless in enabling enrolment. Those on allowances get almost exactly the same amount in living costs support from the government as those who don’t qualify. And they still have to borrow for fees … (even if they may pay off their loans a bit earlier than those who don’t get an allowance).
But this array of supports works only if the aspiring student sees value in study. And nothing in the present policy addresses that fundamental access question, the first stage in the decision-making model above – what happens if a person would benefit from study but simply doesn’t, just can’t, see the value? That is the real access question.
Perceiving value …
So how do young people think about value? What drives their perception?
To think about that, it’s helpful to use the analogy of investment, and to consider the individual’s investment horizon[2].
Some individuals take a short-term view; they expect to see the benefits of an action over a short period – that means they have a high discount rate, so they discount the value of a future benefit over a short period. As a result, they are less willing to forego earnings for a period of five or six years on the chance of landing a high paying job afterwards. Plus, a person who takes little satisfaction and pleasure from the processes of learning and of being taught is likely to experience study as a high cost, a high personal (even if not financial) cost. That, too, raises the person’s discount rate. If you don’t like study, a year’s enrolment costs you more, not in a dollar sense, but in pleasure, joy, satisfaction foregone.
Others have longer investment horizons, lower discount rates and see the benefit and value – financial value and/or personal value – in short-term sacrifice for longer-term gain. Plus, people who take satisfaction and pleasure in learning will see intrinsic personal value in study. So that further lowers their discount rate.
Usher (2006) cites a number of studies that suggest[3]:
“… on average, people from lower-income backgrounds have different discount rates – that is, they have a different investment “horizon” – than people with higher, more stable incomes. Simply put, the evidence shows that long-term poverty encourages short-term thinking …”
The New Zealand tertiary student population is skewed towards higher SES groups, especially among higher tertiary education students[4].
But why?
Deciding to study … the evidence
Educational choices derive from factors and decisions that occur early in life[5]. Performance in the early years of schooling has implications for the type of secondary education children will be able to pursue, which, in turn, affects the child’s post-secondary educational options and that, in turn, affects the range of career options that are open to them and hence, their later occupational choices. The decision of whether to undertake tertiary education and if so, what type of tertiary education, occurs at the end of a sequence of earlier choices, each of which is likely to affect the opportunities the decision-maker faces next (Hofer et al, 2020).
The role of parents …
Parents create the environment within which children are raised, passing on and nurturing skills, cognitive and socio-emotional, which help set the pathway to skill development. Parental education and parental (especially maternal) capability affect skill development. Family income makes a difference to outcomes – even if parental capability trumps income as a determinant of outcomes[6].
Parents determine the district a child is raised in and select early childhood education. A child’s skills can be shaped and enhanced by early intervention; interventions to remedy skill gaps in young, disadvantaged children have been found to be effective, especially if they are reinforced over time. But interventions made later, especially after the age of 10 years, may be less effective (Cunha & Heckman, 2007; Carneiro & Heckman 2002)[7].
Parents are also reference points and role models for educational and career aspirations. Studies on the intergenerational persistence in educational attainment and occupational choice have shown that young people often take the educational level of their parents as a reference point for their own aspirations for educational attainment. Young people from higher SES families (often those families where parents have higher educational qualifications themselves) are more likely to opt for higher education programmes. Conversely, students whose parents had completed higher education may lack information about non-university options and the potential benefits of vocational education (Hofer et al, 2020; Goldthorpe, 2014; Page et al, 2007; Peter & Zambre, 2017)[8].
… the role of schools
Schools also create reference points for the educational and occupational aspirations of their students. School context, the existence and type of career education offered in schools, and the influence of individual teachers can foster children’s interests and aptitudes and shape their aspirations (Hofer et al, 2020; Rowan-Kenyon et al, 2011; Archer et al, 2013)[9].
Decisions, decisions, decisions …. making choices, showing preferences
Student agency in educational choices starts around the age of 15. Role models from their families, peers, friends and the media all play an important role in the formation of these preferences (Taylor, 2016)[10].
Here is where the access question becomes especially important. How does the prospective student form a decision whether and what to study, on whether it is worthwhile to study. On whether there is value in study – that is, whether to answer “yes” to the question “is study worth it?”, the first question in Usher’s three-stage access decision-making model.
It’s a complex decision because it is intertemporal – a decision made now for an uncertain future. A decision with implications – some possibly negative, some positive – for the decision-maker’s medium-term lifestyle and happiness and with no guarantee of longer-term success. Intertemporal decision-making carries a number of biases (French & Oreopoulos, 2017)[11]. For instance:
- Present bias: People tend to place greater weight on tasks that have an early return than on those that have a delayed return, even if they are told that the future return carries a significant benefit. This factor may cause people to defer planning their next steps in education.
- Inattention: Someone may defer or avoid taking a decision because of the belief, conscious or subconscious, that the cost of considering that decision is high relative to the benefits. If the benefits are seen as being far into the future, the value of those benefits is discounted (whereas the time to consider the alternatives represents an immediate cost). That is, they have a high discount rate.
- Social identity: Social context and norms reduce the options a decision-maker considers. Individuals take account of questions like: “what kind of person am I?” and “what are others like me doing?”. These sorts of questions provide reference points for deciding how to act. For instance, for first-generation students, information about the costs of attending higher education is typically more salientthan information about future benefits.
- Confirmation bias: People may reject new information if it does not fit with the information and the frame of reference that they already have.
Given the complexity of the decisions needed on educational pathways and on careers, and given the inter-temporal nature of those decisions, in practice, young people making decisions use heuristics – that is, practical approaches, such as trial and error or rules of thumb – when confronted with these choices (instead of adopting a purely rational approach).
Teenagers are more susceptible to short-term thinking and present-biased preferences than adults. Neuroimaging has shown that the executive function of the brain, (which helps individuals create holistic views of themselves and how they interact with the world around them), will not be completely developed until a person is between 25 and 30 years old (Lavecchia, Liu & Oreopoulos, 2015)[12]. The limbic system (which registers desires for immediate rewards and pleasure, and is responsive to monetary stimuli, novelty, and social rewards) develops earlier and can overrule self-control.
How to design interventions to address the access question – interventions that will help a young person make a decision right for him or her, right for him or her at that point in time, a decision that will help him or her make a good assessment of the worth of entering tertiary education … the design needs to be shaped by an understanding of the sequential and heterogeneous nature of the decision-making process and the factors that influence prospective students’ choices.
That’s why the student allowances system does nothing at all – zero – for access. That’s why the expectation that fees free would improve access[13] was so unrealistic. That’s why student allowances and fees free – first year, final year, any year – is such poor value spending.
How does the decision-making model square with our tertiary student population
In a seminal statistical study of the factors associated with participation in tertiary education, Ministry of Education researcher Dee Earle[14] found that performance at school was the strongest predictor of participation. Controlling for school achievement meant that the influence of all other factors was lessened.
The young person’s parents’ level of educational qualifications – often used by researchers as a proxy for SES – has an effect over and above school achievement
There are differences between ethnic groups in the likelihood of participation, over and above school achievement
- Māori young people are less likely to participate in higher level tertiary education, even after controlling for their school qualifications, educational performance and parents’ qualification
- Young people who identify as Asian are more likely to participate in higher level tertiary education, after controlling for their school qualifications, educational performance and parents’ qualifications.
Some socioeconomic factors have an effect over and above other factors – for instance, those from high deprivation neighbourhoods were less likely to participate at degree level
Significant life events also have an effect on participation over and above school achievement and other background characteristics:
- Having become a parent is associated with much lower participation for both mothers and fathers.
- Any involvement with the justice system is associated with lower participation.
- Having treatment for mental health is associated with somewhat lower participation.
- Employment status between ages 16 and 18 has an impact on tertiary education participation: Being not in employment, education or training (NEET) at ages 16 to 18 is related to lower rates of participation by age 20. Likewise, those who are in substantive employment at ages 16 to 18 are less likely to participate in tertiary education by age 20.
There are two limitations and caveats we need to place on Earle’s profile. The first is that Earle’s statistical models (like most social scientific models) don’t explain the whole picture. The explanatory power of the principal model used by Earle is around 60%. Part of the gap occurs because no model can capture all of the complexity, the randomness of human behaviour. And there will be events or factors that happen not to be in the IDI dataset.
That second is that some of the early life experiences identified in the analyses by Cunha & Heckman (2007), Carneiro et al (2003) and Carneiro & Heckman (2002) (for instance, those related to early life skill formation and parental capability) are invisible in Earle’s data but are highly likely to have been captured by the school performance and school achievement variables used in Earle’s modelling. The same applies to the modelling of how young people arrive at decisions in the papers by behavioural economist Philip Oreopoulos referenced above[15].
So the analysis of factors associated with participation and the modelling of participation decision-making need to be seen and understood together; they provide complementary views of the process of decision-making and of the effects of decision-making.
The implications for policy
If policymakers want to maximise the value we get from our education system, if they want to lift the skills of the workforce and raise productivity, if they want to help young people realise their potential, then the country needs to improve the pathways through the system. That means removing unnecessary barriers, improving access.
Overwhelmingly, young people without some level of tertiary education achievement are at risk of poor long-term outcomes. New Zealand is one of seven countries that participate in the OECD’s PISA programme where the trend in performance in literacy, mathematics and science is declining[16]. That is a lead indicator that the number of people completing school with low achievement and low cognitive skills – and hence, at risk of poor employment outcomes – is rising. This lead indicator suggests the problem is likely to get worse in the near future.
We need an approach that ensures that opportunities are available for all young people who would benefit from tertiary education.
A challenge, a complex problem …. But a challenge we should take on. Our society, those who have succeeded in our system owe that much to our young people, to those who follow. There are no easy remedies or simple answers. But here are some ideas for investigation, some areas where we should apply some policy thinking and policy design.
But first, the government needs to free up the wasted spend. Fees free is a deadweight expense. Student allowances for young students – for those without dependents – is a poor use of money. Get rid of both these regressive, costly, useless policies. We need to find more productive ways to use the money freed up.
I would prioritise four areas for investigation and for design and development of measures, by people with much more expertise than me. We all know that nothing is easy. Nothing comes cheap. But we need to make a start.
Area One: intervene early: The evidence is clear: parenting capability is critically important in shaping skill development. Are there ways of harnessing the resources of communities to help parents who are struggling? The Whānau Ora model shows us how that might work.
Area Two: strengthen education in the early years: Support teaching quality in ECE and the early primary years – improving teacher quality pays off[17]. Developing and sharpening skills in the early years creates options for people as they grow and develop.
Area Three: support young people’s decision-making: As they reach the age where they gain agency in decision-making, young people need to be able to access help and advice. Part of that may be to do with the way schools support young people’s decisions, but that should be complemented by detached advisors – here too, Whānau Ora may provide a model[18].
Area Four: redesign and redevelop the programmes for young people at risk. Tertiary education isn’t the solution for every single young person. That’s why there are active labour market programmes, programmes intended to help young people make a successful and enduring transition to the labour market. The performance of some of the New Zealand programmes – Youth Guarantee in particular – has long been recognised as poor[19]. There are good models (such as this one). But most of our spending, most of our efforts go into schemes that have been proven to fail. The replacement needs to be designed to fit the evidence on what works!
We have to do better. We can do better.
Bibliography
Archer L, DeWitt J & Wong B (2014) Spheres of influence: what shapes young people’s aspirations at age 12/13 and what are the implications for education policy? Journal of Education Policy, 29:1, 58-85
Carneiro & Heckman (2002) The evidence on credit constraints in post-secondary schooling IZA Discussion Paper No 518, IZA
Carneiro P, Cunha F & Heckman J (2003) Interpreting the evidence of family influence on child development Research Gate
Cunha F & Heckman J (2007) The technology of skill formation NBER Working Paper No 12840 National Bureau of Economic Research
Earle D (2018) Factors associated with participation in tertiary education by age 20 Ministry of Education
Finnie R, Laporte C & Lascelles E (2004) Family background and access to post-secondary education: What happened over the 1990s? Statistics Canada
French R & Oreopoulos P (2017) Behavioral barriers transitioning to college Labour Economics, 47, 48-63
Goldthorpe J (2014) The role of education in intergenerational social mobility: Problems from empirical research in sociology and some theoretical pointers from economics, In: Rationality and Society, Vol. 26 (3) 265-289
Hanushek E (2011) The economic value of higher teacher quality Economics of Education Review, vol. 30, pp. 466–479
Heckman J (2011) The economics of inequality: the value of early childhood education American Educator, Spring 2011
Hofer A-R, Zhivkovikj A & Smyth R (2020) The role of labour market information in guiding educational and occupational choices OECD Publishing
Lavecchia A, Liu H & Oreopoulos P (2015), Behavioral economics of education: Progress and possibilities, NBER Working Paper 20609 National Bureau of Economic Research.
Mischewski B & Smyth R (2025) The funding of technical and vocational education and training (TVET) ConCoVE Tūhura
OECD (2019), PISA 2018 Results (Volume II): Where All Students Can Succeed, PISA, OECD Publishing
Oreopoulos P & Petronijevic U (2013), Making college worth it: a review of research on the returns to higher education, National Bureau of Economic Research
Page L, Levy Garboua L & Montmarquette C (2007), Aspiration levels and educational choices: An experimental study, Economics of Education Review, Vol. 26(6), 747-757
Peter F & Zambre V (2017) Intended college enrollment and educational inequality: Do students lack information? Economics of Education Review, Vol. 60, 125-141
Rowan-Kenyon H, Perna L & Swan A (2011) Structuring opportunity: the role of school context in shaping high school students’ occupational aspirations. The Career Development Quarterly, 59(4), 330–344
Smyth R (2023) Education for young people at risk: The challenge … and the opportunity Strategy, Policy, Analysis: Tertiary Education
Taylor R (2016) Moments of choice: How education outcomes data can support better informed career decisions, The Careers and Enterprise Company
Usher A (2006). Grants for students: what they do, why they work Educational Policy Institute
Endnotes
[1] Adapted from Usher (2006)
[2] Again, adapted from Usher (2006)
[3] A caveat, Alex points out that none of those studies was looking specifically at education.
[4] See Earle (2018) Factors associated with participation in tertiary education by age 20 Ministry of Education where the Index of Multiple Deprivation (IMD) of the young person’s home address and also parent’s’ level of education both proxies for SES, were identified as factors associated with participation. In Canada, parental education level is a predictor of participation in post-secondary education – see Finnie, Laporte & Lascelles (2004). For the US, see Carneiro & Heckman (2002) and Cunha & Heckman (2007)
[5] Much of the review of literature in this section is adapted from Hofer, Zhivkovikj & Smyth (2020) and includes insights drawn from Cunha & Heckman (2007) and Carneiro & Heckman (2002)
[6] See Heckman (2011)
[7] Cunha & Heckman (2007), Carneiro, Cunha & Heckman (2003) and Carneiro & Heckman (2002)
[8] Hofer et al (2020); Goldthorpe (2014); Page, Levy Garboua & Montmarquette, (2007); Peter & Zambre, (2017)
[9] Hofer et al (2020); Rowan-Kenyon et al (2011); Archer et al (2013)
[10] Taylor (2016)
[11] This section is adapted from Hofer et al (2020) which draws from behavioural economics literature, especially French & Oreopoulos (2017) Behavioral barriers transitioning to college Labour Economics, 47, 48-63 and Oreopoulos & Petronijevic (2013), Making college worth it: a review of research on the returns to higher education, National Bureau of Economic Research
[12] Lavecchia, Liu & Oreopoulos (2015)
[13] In the 2017 Labour Party election statement for immigration, polytechnics were assured that they needn’t worry about the party’s intention to reduce international enrolments in certain sorts of sub-degree qualifications because fees free would deliver a domestic student boost of around 15%. 15%! Unfortunately, they were out by 15 percentage points!
[14] Earle (2018). Dee used data from Statistics NZ’s Integrated Data Infrastructure (IDI) and so was able to link participation to a very wide range of factors, schooling, family factors, socio-economic factors, behavioural factors, interactions with the justice and welfare systems, ….
[15] French & Oreopoulos (2017), Oreopoulos & Petronijevic (2013) and Lavecchia, Liu & Oreopoulos (2015).
[16] OECD (2019)
[17] Hanushek (2011)
[18] See Mischewski & Smyth (2025)
[19] See Smyth (2023)
Strategy, Policy, Analysis | Tertiary Education 