1 Summary
This paper considers the management of the Crown's exposure to financial risk. By financial risk we mean the variability in the Crown's net worth under planned policy. By managing risk we mean measuring it, analysing it, and, where appropriate, taking steps to reduce or otherwise change it.
It is widely acknowledged that the government should manage the Crown's exposure to risk. The cost to New Zealanders of the Crown becoming financially distressed would likely be very high, so keeping that risk very low is important. In addition, there is a case for keeping tax rates and spending stable in the face of economic shocks.
The Crown's aggregate exposure to risk can be effectively managed only centrally. The risk to which the Crown is exposed depends on the size of the risks in each part of the Crown's portfolio and on the correlations among the parts. The managers of the various parts of the Crown's portfolio can estimate and manage the risk in their portfolios, but they cannot be expected to know what contribution the risk in their portfolios makes to the Crown's aggregate exposure to risk. But central risk management does not imply micromanagement: the management of many assets and liabilities can be delegated, and central control of other risks may amount to no more than a direction to avoid or minimise certain risks.
The management of risk is already centralised in several ways. Recent central risk-management decisions have included avoiding net foreign-currency debt, limiting gross debt to 20% of GDP, and suspending contributions to the New Zealand Superannuation Fund (NZSF). Yet, the government's management of the Crown's exposure to risk is less intensive than might be expected. The government does not currently measure the Crown's aggregate exposure to risk. And it allows the managers of financial assets and other parts of the Crown's portfolio to make choices that may have significant effects on the Crown's aggregate exposure to risk.
Given the inevitable inaccuracy of any estimates of the Crown's exposure to risk, and the difficulty of precisely specifying the Crown's risk-management goal, the government's existing approach is understandable. Yet it does have disadvantages. In the absence of a comprehensive model of the Crown's exposure to risk, decisions must be based on intuition or partial, and therefore inadequate, models. In the absence of central control, managers of parts of the Crown's portfolio may unknowingly expose the Crown to too
much risk or incur costs to reduce their own exposure to risk when, from the Crown's perspective, that is unnecessary or even counterproductive. Cheap ways of reducing the Crown's exposure to risk may be overlooked.
That suggests that the government should measure and monitor the Crown's exposure to risk. Although the government should be prepared to intervene more, if necessary, in risk-related decisions of the managers of the parts of the Crown's portfolios, whether it should depends of course on the results of the measurement and monitoring and on the possible costs of the intervention such as a reduction in the accountability and independence of managers.
Building on work done by the Treasury in the late 1990s and early 2000s, and the development of consolidated Crown accounts and long-term fiscal projections, this paper presents numerical estimates from a model of the Crown's exposure to risk under current policy. The model enables the estimation of the Crown's portfolio's expected return and volatility, value-at-risk, and projections over time of the Crown's exposure to risk, defined as the frequency distribution of the Crown's net worth under planned policy. The model integrates analysis of the Crown's ordinary assets and liabilities with projections of spending and taxes under planned policy. It can be used therefore to analyse the Crown's comprehensive net worth, which includes the present value of projected taxes and spending as well as accounting net worth according to generally accepted accounting practice (GAAP). It can also be used to analyse recently developed measures that exclude from net worth the value of social assets, such as roads and schools.
Our base case implies that the Crown's exposure to risk is large. The Crown's comprehensive portfolio has annual volatility (standard deviation) of around $30 billion: to offset a permanent loss of this size, taxes would have to be permanently increased by 3%. (Results for other measures of net worth are also reported.) The results, it should be emphasised, are only very approximate. They are sensitive to numerous modeling assumptions and estimates, including estimates of correlations and volatilities about which reliable data are scarce.
The model can be used to estimate the effect of changes in the Crown's balance sheet on the Crown's exposure to risk. We illustrate with a hypothetical policy change, which we emphasise that we are not advocating: replacing half of the Debt Management Office's New Zealand dollar-denominated debt with US dollar-denominated debt.
We believe the model would be useful both in regular fiscal reporting, alongside the existing suite of fiscal measures, and in risk management (for example, by applying it to policy choices that have important effects on risk). But it is important to keep the limitations of the model in mind. Judgment is required in interpreting the results of the model, not least because of the uncertainties in parameter estimation. Moreover, the model is not the only approach to quantifying the Crown's exposure to risk and other complementary approaches are worth exploring.
