Sunday 25 September 2016

Independent central banks, democracy, and Skcolidlog

Goldilocks’ ideal porridge, you may recall, was neither too hot nor too cold, but rather just right. A lot of people think this ideal has been reached in the relationship between central banks and democracy. The operational autonomy of central banks’ personnel and policy ensures there’s not too much democracy, while the ultimate authority of elected officials over personnel selection and policy goals mean there’s not too little democracy either. Just right? 

Not at all, I argue in this post. Experience demonstrates that the ‘operational independence in pursuit of democratically established goals’ formula creates fundamental and disruptive tensions in democratic polities (including and especially the EU/Eurozone, which I class among them). These tensions primarily affect the coordination of fiscal policy and monetary policy. And instead of Goldilocks, we have Skcolidlog: either central banks have too much power to dictate fiscal policy, or too little.

That the ECB enjoyed from early 2010 through the middle of 2012 an influence on Eurozone fiscal policy so immoderate as to fundamentally contradict democracy is a point I have argued before (and at length here). The ECB threat that government bond markets would be abandoned to self-fulfilling market perceptions of fiscal collapse compelled many governments to moderate or reverse programmes of fiscal stimulus and overrode electoral politics.

But lately, the shoe may seem to be on the other foot. Consider this exchange at Draghi’s September 2016 press conference
Question: You've been urging governments to act for some time, and I'm wondering if there's a sense that maybe they might now be a little more willing to act and that the ECB could encourage that willingness by not raising excessive expectations about future monetary policy measures, hence the tone today.
Draghi: The ECB can't be in a sort of – let me say, what the ECB can do is to basically flag what is needed for monetary policy to be even more effective than it is at the present time.
In effect, the journalist asked whether Draghi could threaten to limit monetary stimulus in order to compel action by the fiscal authorities, probably hinting toward further fiscal stimulus. Draghi replied that he had only verbal persuasion at his disposal. (Draghi went on to demonstrate once again how low fiscal stimulus is on his list of priorities, but that is beside the point for now). 

While ECB leaders’ protestations of their limited influence in promoting austerity are unconvincing, Draghi’s assertion that in present circumstances he has little leverage on policy is far more believable. The difference between the two situations turns on the nature of the ECB’s mandate. When amidst the government bond market panics of 2010-2012 it was the ECB’s readiness to play a lender of last resort role that was the axis of contention, it was a plausible assertion that this role lay outside the ECB’s mandate. This was one reason a threat to permit bond-market meltdowns was credible. But in the present situation, where the ECB is dramatically failing to meet its clearly specified mandate to attain price stability (which it has defined as inflation ‘close to, but below 2% per year’), it has almost no flexibility: no matter how unhelpful fiscal policy is, the ECB must continue to stimulate, including via policies many find extreme, such as negative interest rates and a massive quantitative easing programme. The ECB has no threat to deploy.

Thus, the discretion created by the lack of a clear mandate to play a lender-of-last-resort role left the ECB with ‘too much’ power, but the presence of such a clear mandate in the case of fighting deflation left it with ‘too little’, at least from the perspective of those who believe further fiscal stimulus is urgently necessary.  

That a Goldilocksian balance remains elusive is not just an idiosyncratic result of current economic conditions. In fact, we are faced with a quite general limitation of the the present formula for reconciling democracy with independent central banking. Lorenzo Bin Smaghi recently likened the situation now facing central banks, in which they must soldier on despite a lack of supportive fiscal policy, to that facing central banks in the 1970s and early 1980s, when the challenge was inflation, not deflation. Most notoriously, Reagan’s budget deficits in the face of the inflation of the early 1980s pushed Fed chair Paul Volcker to maintain extremely high interest rates. In both cases, central bank policy in service of a price-stability mandate had to go to extremes to compensate for unsupportive fiscal policy. 

A very simple game-theoretic analysis (which builds on Blinder’s 1982 discussion of the Reagan-Volcker episode) helps to illustrate the generality of the Skcolidlog pattern in central bank-fiscal authority relations. The diagram above depicts policy choices by fiscal authorities followed by policy choices by central bankers. What it means to ‘reinforce market trends’ is contextual: this could be to contribute to a market panic by explicitly repudiating lender-of-last-resort actions, or to contribute to inflation through fiscal and monetary expansion, or to contribute to deflation or ‘lowflation’ via fiscal and monetary resriction. To counteract market trends is to adopt the opposite policies in each of these circumstances. Case I, “joint irresponsbility,” is the outcome fear of which drove much of the enthusiasm for central bank independence in an inflationary environment, where it would represent both fiscal and monetary authorities adopting expansive policies. But it could equally reflect both the central bank and the government failing to act in the face of a market panic. Case II is the one Smaghi discusses, where the central bank compensates for inappropriate government policy. Case III would involve a central bank pushing in a direction opposite to government policy, where as case IV is coordinated policy to counteract market trends.

This diagram helps clarify when fiscal and when monetary authorities have the preponderance of bargaining power. When the central bank is constrained to fulfil an inflation mandate, the boxes shaded in gray are not available to it. Thus, the fiscal authorities unilaterally choose between options II and IV. Very often it has turned out that fiscal authorities have preferred to force central banks to go it alone instead of coordinating policy, even when the latter would have arguably generated better growth outcomes and more effective attainment of the goals expressed in the central bank’s mandate. The reason for this is that governments have other agendas for fiscal policy. Reagan wanted to “starve the beast,” Cameron to cut back the size of the state. Fiscal authorities arguably did not have to bear the full political costs of these macroeconomically inappropriate policies because central banks compensated for some of their negative economic effects. I think a good case could be made that the ability of elected governments to compel such compensatory action by central banks does serious damage to mechanisms of electoral accountability that are usually held to be at the heart of democracy’s advantages. 

On the other hand, when the central bank does have discretion, and the gray boxes are open to it, as in the lender-of-last-resort case, it can often impose its will on government policy. In particular, if the CB prefers case II to case III, and case III to case IV, then fiscal authorities fearful of their policies being undermined may have to choose to reinforce market trends as a condition of central bank action. This is what happened to some European governments when the ECB was willing to rescue government bond markets only on condition of austerity. 
The diagram above shows some empirical cases of each outcome, with ‘1970s’ standing in however approximately for joint monetary and fiscal irresponsibility. What all of this implies to me is that the idea that independent central banks bound by a policy mandate can serve as a useful check on elected governments depends, in fact, on an unrealistic conception of the circumstances in which central banks act (they will sometimes be called upon to act in areas beyond their mandate) and on the preferences of democratically elected governments (who may use mandates to force central banks to deal with the consequences of their inappropriate policies). The cases suggest that the Skcolidolg pattern has some real empirical relevance. 

What, then, is to be done? Some people argue for giving central banks more power over fiscal policy, especially in near-deflationary circumstances like at present. But it seems to me—I won’t try to defend the point in this already over-long post—that this runs the risk of exacerbating the problem of electoral accountability that has hindered electorates from understanding the impact of the macroeconomically inappropriate policies that right-wing governments seem so prone to run. Nearly two decades ago, Berman and McNamara argued that insofar as the case for the economic advantages of independent central banking was dubious, it provided no grounds for overriding the usual preference for favour of democratic governance in the case of central banking. When one looks at how central banking has interacted with democracy in practice, their case only gets stronger. It’s time to bring central banks back under the direct control of elected officials, so that they bear the responsibility for both good and bad choices about monetary policy. Ultimately, there’s no way to get the porridge right unless you make it yourself.