Paper for the ‘Student Loans Summit’, 25 August, 2000

Keywords Education

It is important when thinking about Student Loans, or indeed about any other facit of government policy, that the policy which drives it is seen as a part of a total policy framework evolved out of a taskforce which was established in 1984 to completely review government policy. I imagine at the time that some of the outcomes of the taskforce thinking were expected like privatisation and corporatisation. But the comprehensive framework of commercialisation may well not have been, nor may they have expected the proposals to, say, reform student access to tertiary education. That was to come later. By that time the rule of making government activities run as though they were business ones, was no totally accepted in the policy community, and it was natural to do that as much as the public would allow. What that means is that the policy of student loans have to be seen in this wider context, and to challenge them involves a different policy framework.

That policy framework hardly exists. Certainly there has since been no taskforce with as comprehensive a remit as the 1984 one to offer an alternative. Since there is not time to take on the full might of the alternative policy framework, today I am going to accept the commercialisation framework and show how the current policy leads to problems, some of which are potentially serious in the long run.

In order to get the framework underway, we need to recognize that student debt is not a simple liability like a mortgage or overdraft. Repayment is contingent on the income stream of the debtor involved, and that income stream is uncertain. The technical term for the loan from the position of the student is a ‘contingent liability’. There is a good reason why the proponents of the student loan scheme do not use the precise term. For one thing it would display to the world that essentially the scheme is not more than a raising of the income tax levy on some parts of the population those with the contingent liability from the student debt in order to lower it in the rest, particular those of us who got a free tertiary institution, or who have been able to pay off their contingent liability.

There is a second reason, why there is not a lot of enthusiasm to call the student debt by a more precise term. Every liability of a person or institution has to be matched by an asset of some person or institution. Thus if student debt is a contingent liability, it has to be matched by a contingent asset. The holder of the contingent asset is the government. However were it to be described as a contingent asset the government’s statement of financial position would look a lot less satisfactory than it is. Here is a recent statement for of the government’s position for May of this year.

What it shows is that the government has a net worth in the statement of financial position called the ‘Crown Balance’ of $8.2b, made up of $ 60.9b of assets less $52.6b of liabilities. However $3.4b of those assets are contingent assets, dependent on the students for whom they are contingent liabilities earning sufficient income to pay them off and not being of low income, bankrupt, disappearing out of the IRD records, or finding a tax haven to shield them from the repayment of the liability.

That means that currently slightly less than half of the so-called net worth of the crown is covered by the contingent assets of student loans.

Think ahead. Current projections suggest that the student loan contingent liability will total above $20 billion in twenty years. We do not know what the state of the government accounts will be in 2020, but we can be almost certain that under current fiscal policies, the contingent asset will make up a larger proportion of total assets in the government accounts. Indeed, depending on policy outcomes, it may exceed net worth. What this means is that an increasingly important component of the government financial position is dependent upon the contingent asset of student loans.

Now there is a sense in which all assets are contingent assets, in the sense that their value is dependent upon uncertain future events. Accountants have procedures for allowing for these uncertainties. In the case of the contingent assets they write off some of the debt, on the basis that not all the debt will be recovered. In effect the total of all outstanding student debt exceeds that which is reported in the New Zealand government accounts.

Some of the write-offs are straight forward, as when the debtor dies or goes bankrupt. However how much should there be a write-off for the fact that students who are currently alive will never pay off all their debt, for whatever reason – insufficient incomes, escape overseas, fraud, or tax avoidance. Very often the government will not find about these failures for forty and more years into the future. But prudent accounting practice is that there should be some provision for them being written off now. Otherwise the value of the contingent assets is misleading.

Now there are rules on how much should be written off, but they involve judgements, which are particularly tricky when there is so little experience of the scheme. I was surprised when I looked at them, that in recent years only 3.0 to 3.5 percent of the outstanding balances are being written off. The underlying assumption is that around 10 percent of student debt will be doubtful debts and not be recovered, together with write-offs from death and bankruptcy. This suggests either students are exceptionally honest (or not very imaginative) or the Department of Inland Revenue is exceptionally efficient or vigorous. If these assumptions were wrong, then the true level of the contingent value would be lower, the government’s net worth would be correspondingly lower, and in some years the government would be making a deficit when it was claiming to be making a surplus.

My doubts were reinforced when I read last June’s report of the Controller and Auditor General. Its conclusion is that

“There is limited analysis on the financial risks attached to the Scheme, which makes it difficult for Parliament and the public to have an informed debate about the size valuation, financial performance and debt recovery of the Scheme, These factors make it difficult for student loan policy to be reviewed in terms of unintended socio-economic outcomes.” (Report of the Controller and Auditor General, Student Loan Scheme – Publicly Available Accountability Information, p. 29.)

And, I would add, for the policy to be reviewed in terms of its macroeconomic impact, its implications for human resources, and the usefulness of the government balance sheet.

A third method of measuring the value of the contingent assets would be to find out what they would be worth if they were ‘securitised’ sold to the private sector. (Note I am not arguing the government assets should be privatised, but this is the correct measure of value under the commercialisation policy framework.) This would reflect their true value allowing for the administrative costs the write-offs and doubtful debts, and the current interest charges. My guess is that if the government tried to sell the assets on the open market, they would get about half the value of what they are shown in the government accounts.

In summary then the figures that appear in the government accounts appear to be subject to a wide margin of error, and are probably overvalued. The error is of sufficient size to leave the Auditor General wondering whether the nation can have a sensible debate on student loans.

In addition, an increasingly significant item in the government accounts is subject to an uncomfortably high margin of error means that the accounts may be misleading, and the macro-economic implications of the government accounts difficult to evaluate.

In a way student loans are one of the think bigs of the 1990s. The government has gone into a massive commercial operation which may have led to a misleading government balance sheet (statement of financial position) and will probably lead to major write-offs further down the track.

Does that matter? Of course it matters in a democracy if we cannot know the true state of a particular government policy. It may also matter in the technical sense that we may have a poorly designed policy. In fact if we don’t know what the true parameters for debt write-off provisions are, it almost certainly a poorly designed policy. Moreover it is quite possible that the scheme is much more inefficient than it appears, because the write-offs are a cost of the scheme. For instance, reduced student fees may be more equitable and administratively less cumbersome to give students maintenance grants than to run the loan scheme if the true write-off rate is somewhat higher than that currently assumed. But we don’t know without a proper risk assessment.

Moreover, because the contingent asset of the student loan debt is going to be coming an increasingly important component of the government accounts, the uncertainty associated with it means that the statement of financial position will be of poor quality and misleading and the Crown Balance of net worth is probably wrong. It is no comfort this problem gets worse as we face a rising proportion of elderly, and the fiscal stress that generates. In effect the government accounts will be deceptive in regard to their ability to meeting the needs of the elderly, so the generation which had to take out student loans and which paid higher taxes as a result, will face another whammy of the public account being weaker than they were told. Think bigs come back to haunt us.

There are many conclusions this conference may come to. But one seems to me unassailable: that the Auditor-general’s call for a review of the valuation of the student loan debt should be proceeded with forthwith, preferably by an independent panel. I would advise students to take a very real interest in its deliberations.