Should the government run a budget deficit or a budget surplus? What size should the deficit or surplus be?
Devising a rule which allows the public to assess the integrity of government is a major challenge in political economy. The standard approach, exemplified by the brilliant work by Simon Wren-Lewis and Jonathan Portes, uses fiscal sustainability equations aimed at stabilising the public sector debt to GDP ratio to determine borrowing targets.
On the face of it, this makes compelling sense. A fiscal stance that results in a constantly rising debt to GDP ratio will surely end in tears (default, inflation, brutal austerity – or some unpleasant combination). The practical challenge with this approach, however, is that there are so many moving parts. What is trend growth? Where will interest rates be? How enduring is the downturn? Is there a case for material shifts in the debt to GDP ratio even over twenty- or thirty-year phases, which require a suspension of principles of stability in this ratio?
An alternative view from functional finance, which has the benefit of flexibility in light of these uncertainties, is that the fiscal authorities target inflation. This idea is unlikely to gain traction because it faces a debilitating set of obstacles, not least of which is that we already have an institution targeting inflation – the central bank. Politicising inflation seems very unwise, and if central banks have adequate tools to meet their mandate, the ‘rule’ is redundant. More fundamentally, it might also be the case that inflation is at target and there is still fiscal space, or a case for tightening. In short, fiscal policy is not suited for inflation-targeting.
Is there an alternative? To my knowledge, no current fiscal rules are centred around the market rate of interest on government debt, despite the fact that this is often the practical constraint. My poposal is to make the concept of ‘fiscal space’ operational, and tied to long-term rates of interest on government debt. This permits huge flexibility in the size of the deficit, in the debt/GDP ratio, leaves inflation targeting to the central bank, guarantees debt sustainability, and inherently adapts to the regime. It would also provide clarity to the public that the government is honouring its word.
How would this rule work? My suggestion is that the government has the ability to ease fiscal policy (or issue debt for other purposes) as long as the long term cost of government debt is below nominal GDP growth (this is my practical definition of ‘fiscal space’). To be very clear, I am not proposing that the government targets bond yields. Instead, I suggest that the government commits to target a stable debt/GDP ratio, along the lines of the rule advocated by Wren-Lewis and Portes, if bond yields are close to, or higher than nominal GDP growth. If bond yields are materially below nominal GDP growth, the government would be freed from this constraint the government has complete flexibility to pursue policies that would require increased borrowing.
The strengths, the logic, the weaknesses
Why might this ‘rule’ be advantageous? Firstly, it is both highly flexible – there are no budget balance targets, no debt/GDP targets, and no difficult estimations. The 10-year government bond yield is a market price. Some estimation of average nominal GDP growth is required. The simplest rule-of-thumb would be a 10-year trailing average. Alternately, consensus economic forecasts for long-term growth, provided by the national statistics agency, could be used. So the government has a very high degree of flexibility to cope with the wide variance of economic outcomes, and it is relatively straightforward to observe non-compliance.
What is the logic behind this rule? Fiscal sustainability is a determined by the interaction of four variables – the interest rate the government borrows at, the balance of revenue and expenditure before interest expenses (the primary balance), the rate of GDP growth, and the starting point debt/GDP ratio. The typical approach to a fiscal target is to make estimates of the likely cost of debt and trend GDP growth and then target the balance necessary to stabilise the debt/GDP ratio. But what if there are phases when shifts in the public debt/GDP ratio are beneficial? For example, when large shifts in savings preferences push down real interest rate structures relative to GDP growth. We want to have a rule, with ensures sustainability, retains flexibility, and can’t be gamed.
My suggestion is that we reverse engineer the problem, free the deficit and debt/GDP ratios, and make the cost-of-debt constraint operational. This is a Prudence Principle. In other words, choose a level of bond yields from which it would almost certainly be wise to tighten in response to in most circumstances. Is there such a magic level? The obvious point is when the cost of long term borrowing to the government is close to, or exceeds, long-term nominal GDP growth. Why might this be a helpful rule-of-thumb? Government revenue, in the absence of a change in policy, should grow at around nominal GDP growth, and in the absence of a primary deficit or surplus, the stock of debt will compound at the rate of interest. If the debt is compounding faster than revenue in the long term, sustainability is questionable.
Shifts in the rate of interest rate on government debt relative to the rate of growth in nominal GDP are therefore critical to how the stance of fiscal policy has to change through time. If the public sector primary balance (ie before interest payments) is zero, its stock of debt will compound at the rate of interest. If the rate of interest equals nominal GDP growth, the debt/GDP ratio will be stable. It follows that if the rate of interest sits below nominal GDP the government can run a primary deficit and have stable debt/GDP ratio, similarly if interest rates are above nominal GDP, stabilising debt to GDP requires a primary surplus.
The Prudence Principle appears to be dynamically consistent with debt sustainability, allows policy to respond to changes in trend real interest rates, and provides us with a very transparent yardstick for assessing policy.
I will briefly consider some likely objections, both weak and strong. And conclude with some observations about the prevailing regime in the global economy which may make this rule more timely and relevant. The obvious objection is a reluctance to pin government policy on a market-determined price – the long bond. There is little doubt that there is excess volatility in short-run shifts in yields. Even at maturities as long as 30-years, yields can move by over a percent in three months. Revising fiscal policy with this frequency would not be helpful. The logical response to this problem is to use perhaps a three-year average yield, and build in a significant margin of safety.
A second objection is that announcing such a policy would make it inoperable because yields would automatically rise in anticipation. This is possible, but I think unlikely. In a monetarily sovereign, stable inflation, developed economy, the centre of gravity for government bond yields is determined by interest rate expectations and term premia. The term premium (or discount) is the difference between the actual yield and interest rate expectations, which in turn is determined by the ‘risk’ properties of government bonds – in particular the correlation with cyclical assets, such as equities – and the rate of interest on global bonds. The impact of any fiscal easing on bond yields depends therefore on the effect on interest rate expectations. Any substantial fiscal easing would be expected to raise interest rate expectations relative to the counterfactual, but the probability that markets would price in a perfectly symmetrical response is unlikely. For yields to rise to the level of nominal GDP in an enduring way would require a sustained rise in interest rate expectations, which would imply two beneficial outcomes – very strong growth and a resumption of monetary firepower.
There is a growing consensus that fiscal policy should be eased in response to the collapse in global interest rates and yield curves. There are a number of ways to respond to this. The obvious first step is to give central banks the tools to pursue effective policies, and for those that have the means to exert intellectual leadership.
As regards fiscal policy, there are lots of reasons for exploit the current collapse in the cost of government borrowing – not least, setting up sovereign wealth funds to tackle wealth inequality. It would help if fiscal authorities had a clear framework for responding. The Prudence Principle for identifying fiscal space is to some extent a statement of the obvious – it’s time for fiscal authorities to make it policy.
 The Wren-Lewis & Portes rule is also innovative because it contains a knock-out, ie a suspension of the deficit rules requested by the central bank, when interest rates are too low for the effective functioning of monetary policy. I am really just advocating an earlier suspension of this when prevailing interest rates are significantly below nominal GDP.