Speculation that the ECB might, as part of its proposed bond-buying programme, announce an interest rate target (or band) for short-dated peripheral government bonds sparked a further rally in Spanish and Italian bonds in the past week. And while the ECB quickly denied the reports, comments by Joerg Asmussen, the German Executive Board Member, suggested that the ECB is at least exploring the possibility of buying unlimited amounts of governments bonds.
Of course, such a move would mark a complete volte face from past policy. Previous ECB bond purchases have been limited both in time and volume to keep both the pressure on governments and, more importantly, to comply with the central bank’s Treaty-based prohibition of monetary financing of government debt. And despite these recent reports we expect the ECB to remain firmly committed to these principles. But that, of course, does not mean that it has no room for manoeuvre.
Arguably, a sizeable expansion of its bond purchase programme would stretch the ECB’s mandate to the limit. But it is unlikely to breach the prohibition of government financing provided purchases remain limited to both the short end and the secondary market. Indeed, increasing the size of future bond purchases and, perhaps more importantly, threatening to maintain a constant presence in the markets would go a long way to rectifying the flaws of SMP 1.0.
But hopes that the announcement of an interest rate or spread target would spare the ECB the pain of having to intervene in the markets at all are flawed in our view. Indeed, given the complexity of deriving and communicating targets for a variety of maturities and countries, such a step might risk political dissent and ultimately undermine the efficacy of any ECB bond purchases.
Specifically, for the ECB to credibly communicate an interest rate or spread target requires it to quantify the excess risk premia – that related to the “fears of the reversibility of the euro” (Draghi 2 August) – judged to be priced into peripheral government bonds. Of course, by implication, that requires the ECB to give its view on what risk premia should be priced into government bonds – effectively telling the market what the “correct” level of risk (and hence default probability) is for each country.
In any case, estimates of these risk premia are subject to a large degree of uncertainty and highly model-dependent. And because they tend to rely on expectations of future fiscal policy and economic growth, they are also prone to large revisions. For example, using a recent Fed study of the yield curve, we estimate that the average excess risk premia priced into Spanish and Italian 2Y bond yields – based on the latest European Commission forecasts – to be around 140bps and 90bps respectively. So, Spanish 2Y rates should, according to this model, currently be 2.4% and Italian rates 2.5%. But depending on the precise specification of the model they could be up to 80bps larger.
A further complication is that, if the ECB wanted to adopt an “equilibrium” spread target, it would need to estimate an equilibrium level for German yields, i.e. derive an estimate of the “safe haven” premia priced into these bonds. Our estimates of the German 2Y “equilibrium” yield range from 0.5 to 1.0%. This suggests a spread target for Spanish and Italian 2Y bond yields of around 150 to 200bps over their respective German equivalents.
But given the inherent inaccuracy of the forecasts underlying these estimates, the ECB would risk having to review these targets regularly, leaving markets uncertain about their permanence. Not only that, if the ECB were to alter its spread targets, particularly in response to any fiscal slippage, that would potentially be viewed by the market as the ECB passing judgement on the change in risk attached to the country involved. This would be very uncomfortable for the ECB. And it raises the question as to whether – and, perhaps more importantly, in which way – the ECB would modify its interest targets once the underlying fundamentals change.
So, an interest rate or spread target has numerous, and we would argue insurmountable, implementation difficulties. But, more importantly, would it necessarily, on its own, be as effective as many people think? Probably not. Even if the ECB’s commitment to the purchases is credible, doubts will continue to persist about whether the countries can meet the conditionality imposed by the ECB and the ESM/EFSF. Investors will therefore always be exposed to the risk that the countries on the receiving end of the purchases fail to meet the conditions required of them, ultimately leading to the ECB turning off the purchases. That would leave anyone holding the debt facing potentially very large losses. Given this, any targets will almost certainly require, particularly in the early months, massive ECB purchases.
So, overall, in light of the large uncertainties surrounding the estimates of any spread target, its implied time-varying nature as well as the risks and costs associated with revoking the target if governments were to fail to comply with their respective programmes, we see significantly more costs than benefits for the ECB from announcing an explicit interest rate or spread target. Of course, through a strong and continued presence in the markets, the ECB can always signal an implicit target. But it’s unlikely to want to go further than that.
None of this is to say that future ECB interventions will not be effective. The ECB appears to be moving towards a much more aggressive bond purchase programme, and one that we don’t doubt will have the potential to lower short-term rates significantly. But as we argued last week, with or without an interest rate target, the success and the sustainability of any future ECB interventions will ultimately depend on the peripheral governments’ ability to meet the conditionality required. The long-term future of the euro, meanwhile, will depend on the willingness of all euro area governments to embrace much closer fiscal and economic integration.