This post picks up on Draghis’ comments that buying riskier assets can be better for the stabilising properties of the ECB’s forthcoming credit easing. The title looks like it’s calculated to put readers off. But 1) it’s Friday, 2) I’m tired after spending the afternoon re-learning how to derive the Kalman Filter for students and 3) it’s factually correct.
What you might call first order Ricardian equivalence is the claim that giving consumers a tax cut won’t increase spending, because they correctly foresee that in the future, taxes will have to rise again to repay the debt that was incurred financing the tax cut in the first place. It fails for many reasons, not least that consumers don’t forecast as well or behave as optimally as we macroeconomists take them to in our models. And also that the timing of taxes matters. If I am pressed up against my borrowing constraint and struggling to feed my family on a measly academic salary (purely hypothetical, you’ll understand) I will be glad of extra disposable income now, and knowing that in a few years time taxes will be a bit higher, and the income will be clawed back, won’t negate the benefit. The fallacious Ricardian Equivalence argument is not just academic, as it got a run around in the debates about the usefulness of the fiscal stimulus in the early part of the crisis.
Hard-line second order Ricardian Equivalence people would say that whether the ECB takes risky assets onto its balance sheet or not shouldn’t affect spending in the private sector in the eurozone. Banks get relieved of the risk, which is great. But the risk is now borne by the ECB, and ultimately by the ECB’s owners – eurozone governments – and that risk means more variance in (or larger second moments of) taxes and spending to make up for it. So the private sector’s risk doesn’t change, though the routes through which the risk impacts its income does. You could call this second order Ricardian Equivalence, because it’s about how the variance (second moment) of returns of the private sector’s cash-flows is not changed by undertaking asset purchases (or whatever).
Which brings us to why purchasing more risky assets from the private sector will be more stimulative. Although at some point in the future, taxes and spending will have to be adjusted by governments to recapitalise the ECB if things go badly with the assets purchased, the future is some time down the road, when consumers and firms and banks will be all lifted further from their borrowing constraints (Panglossian view for simplicity here) and happy to fork out. Right now, with deflation a real prospect, bad debts, high cost of finance, the private sector appetite for risk in the eurozone is low, and the ECB taking it off them will be welcome.
Hence the title. Second order Ricardian non Equivalence refers to the fact that even though the variance of after-tax cashflows at some imaginary end point once purchases have been unwound does not change, the private sector feels better off, because riskiness in the near term, under stress, is more costly to it.
One caveat. All this works fine for a monetary union where there is no possibility whatsoever of dissolution. But where there is – eg in the eurozone – there will be some uncertainty about who will make good, or benefit, when the private sector assets come to be sold. And that will depend on who exits and who is left standing backing the ECB, and on the outcome of any subsequent negotiations with the new ‘outs’ to compensate the ‘ins’. It seems at least possible that that uncertainty could, in extreme circumstances, weigh as heavily as the risk-relief of the asset purchases themselves. That might be an argument for these purchases being done directly by the eurozone fiscal agents, with liabilities spelled out up front. But good luck getting them to agree on anything.