Dr. Darragh Flannery:

We have been asked to appear before the committee to discuss the advantages of an income contingent student loan system compared to other forms of higher education funding. The particular context for this hearing is the recent dissemination of a research paper by Charles Larkin and Shaen Corbet questioning the viability of income-contingent loans in the Irish context from a public finance viewpoint. The committee has previously heard deliberations regarding the importance of the various parameters involved in the design of an income-contingent loan, ICL, system. As a result of the context for this hearing, the focus here will be on how the variation in these parameters impacts on the public finances and how ICLs compare to other forms of funding.

We first provide an overview of the rationale for income-contingent student loans in the Irish context. We then respond to the paper by Larkin and Corbet. Finally, Dr. Doris will present some public debt and deficit analysis of a variety of funding options to provide some evidence on this issue.

In the context of higher education funding, both the level of such funding and the current mix of State support and upfront student fees are widely accepted to be unsustainable. The need for further investment, competition with other areas of public spending, and concerns about accessibility and affordability has turned the main focus of this debate towards alternative funding systems.

As the committee is aware, the report of the expert group on future funding for higher education - the Cassells report - outlined three possible options, namely: a full State model of funding; increased State funding with a continuation of upfront student fees; and increased State funding combined with an ICL system.

Although a fully taxpayer-funded system may seem attractive because it provides access to education at no upfront cost to the student, it is ultimately the most regressive option as it entails a transfer of resources from those who have not benefited from higher education - the lower paid - to those who have - the better paid. Moreover, this option would entail a continued heavy reliance on tax revenue for any future investment in the sector. In addition, it would entail the highest cost to the Exchequer of all of the options considered.

The second option, combining State support with upfront fees, may alleviate some of the public cost in the short and long-term, but raises concerns about affordability and accessibility due to the upfront nature of the costs. The purpose of an ICL system is to remove such concerns while also sharing the burden of financing higher education in an efficient and equitable manner. ICL repayments, which are automatically deducted from the graduate’s pay cheque on the basis of their monthly earnings, are low or zero for low earners and increase as earnings increase, so they are designed to be affordable. Income-contingent debt is, therefore, unlike other forms of debt.

It is important to note that all three options outlined in the Cassells Report involve an increased public cost. This is inevitable, given the recommendation in the report that higher education funding be substantially increased. However, it is also salient that options 1 and 2 are estimated in the report to incur a higher direct cost to the State by 2030 relative to the specific ICL proposal outlined in the report. This is not surprising, since all graduates contribute to the cost of their education under option 3, but not under the other two options.

A note of concern that is specific to the introduction of an ICL system in Ireland relates to the public costs in the initial stage of implementation. These arise because revenue will be lost due to existing upfront fees being removed, while the revenue generated from graduate repayments will take time to flow. The exact design of any ICL system can have a significant impact on the scale of these costs. We will illustrate this point with some examples in the final section of this submission.

Larkin and Corbet purport to analyse lCLs as they would apply in the Irish case, and claim to show that an ICL could not work in Ireland due to the high probability of default. There are serious shortcomings in the methodology used in their modelling of ICLs, many of which are too technical to detail here. One problem can readily be explained, however. In the international literature on ICLs, it is recognised that the default rate is the result of how graduate earnings evolve over time and how repayments are calculated in each particular ICL scheme. Rather than calculating the level of default that would prevail in Ireland, Larkin and Corbet assume that a high level of default would apply and on that basis dismiss all ICLs, ignoring the fact that the default rate of any ICL scheme is a function of its parameters.

A fundamental problem with Larkin and Corbet’s work is that they model the costs of an ICL but do not model the costs of the alternatives. Moreover, they appear to confuse issues of financing with issues of cost. For example, they propose an education levy on earned income that could fund an increase in expenditure resulting from the implementation of option 1.

Apart from the fact that this ignores the significant negative effects on efficiency of such a substantial increase in income taxation and that such an increase is unlikely to be politically feasible, it should be noted that an increase in tax revenue could also be used to fund any outlay associated with ICLs or, indeed, any other method of increasing higher education funding.

It should be stressed that the conclusions of Dr. Larkin and Dr. Corbet are not substantiated by their analysis. Their particularly eye-catching conclusion that an Irish student loan company would create an "Anglo in slow motion" simply cannot be supported.

Despite its shortcomings, the Larkin and Corbet paper has served to focus the debate on the costs to the Exchequer of ICLs. It is obvious that if an increase in higher education funding is provided by means of an ICL, there will be an initial period in which loans issued will not be matched by repayments. The Cassells review addresses the point about the initial deficit, concluding that there would be a build-up of debt of more than €10 billion in the first 20 years of an ICL scheme. This number is based on detailed graduate age-earnings profiles and non-repayment rates that are estimated from data rather than assumed.

Although the figure of €10 billion may seem high, it is worth noting that it would be spread over a long period. Moreover, it is important to note that it arises because the ICL would be associated with a substantial increase in higher education funding, not because of the nature of an ICL scheme. An equivalent increase in higher education funding raised through general taxation would be even more costly, a point to which we will return in our analysis.

It is also important to note that appendix No. 3 of the Cassells report includes a discussion, provided by the Department of Finance, of whether such costs are feasible within the constraints of the fiscal treaty and concludes that they are. However, neither the discussion in the Cassells report nor that in the Larkin and Corbet paper provide a full analysis of the fiscal implications over time of alternative ways of providing for an increase in higher education funding. We have therefore undertaken to provide such an analysis in what follows. Dr. Doris will provide the details.