Central-bank independence under fiscal dominance

Robin Brooks has noted that, in part, the reason for the UK’s deteriorating fiscal outlook is that the BoE is shrinking its balance sheet faster than its peers. As QE is unwinded, yields on gilts will rise accordingly. In other words, the BoE is transitioning away from monetising government debt.

In an earlier post, I alluded to the notion that if fiscal policy is non-Ricardian, the central bank could signal its commitment to operational independence via following a Taylor rule. Even under (long-run) Fisherian dynamics, a temporary hike in interest rates (provided it is indeed temporary) should reduce inflation and keep expectations anchored. In particular, I wrote:

“The strongest counterargument to this conventional view, is that interest rates rises may in fact do the opposite, and increase inflation, as David Andolfatto posits in a thought-experiment. As formalised in the famous Sargent/Wallace “unpleasant arithmetic” paper of 1981, in a regime of fiscal dominance, given that interest rate rises increase borrowing costs (which must then be further monetised, assuming a fiscal authority that embarks on continuous stimulus), interest rate rises will lead to higher inflation, at least in the long-run. This is somewhat analogous to FTPL too, where the increased debt issuance expense adds to (future expected present values of) primary deficits. By identity via the government budget-constraint, this must be matched by an equivalent future expected PV of primary surpluses, or a rise in the price-level. As shown by an intriguing historical test case of repeated hyper-inflationary episodes in Brazil, a sustained fall in inflation is unlikely to occur via monetary interventions alone. Fiscal support is essential. Even in standard NK models, under a fall in inflation via an i increase, fiscal policy is passive and Ricardian.

With the infrastructure bill just passed, and BBB imminent, expected fiscal policy remains yet expansionary. Whilst FTPL, and models of fiscal dominance, only imply a price-level effect, given the political dysfunction and instability of the US legislative process, any idea that current deficits can feasibly be matched by equivalent future surpluses is for the birds! As it happens, for this reason I subscribe to these implications, that monetary policy cannot resolve this inflation debacle forever.

Yet this is a long-run complication. In the short-run, via the Taylor principle, inflation should be stabilised with expectations attenuated. Even Andolfatto notes that a hike of sufficient magnitude will curb inflation: the complication paid more attention to is the challenge amongst monetary authorities to raise inflation at the ZLB/ELB without fiscal expansion (the challenge of the last decade!). Albeit this is at a possible cost of recession, but tell the fiscal authorities that if they continue with the irresponsible lack of rectitude! Moreover, what better way to signal that the Fed is not captured by partisan political interests than to raise interest rates now? Such an act could put calls of fiscal dominance to bed, and demonstrate we are still in a regime of monetary dominance, where monetary policy operates with the conventional short-run effects.”

Under FTPL, given fiscal expansion, an increase in interest rates would increase inflation in the short-run only if expectations adjust instantly (i.e. neutrality of money holds in the short-run) or if the maturity structure of government bond issuance is sufficiently short-dated. This follows from the “unpleasant arithmetic” of Sargent and Wallace (1981). Also note that fiscal expansion only produces a price-level effect.

Hence, for long-run inflation targeting, keeping expectations anchored is crucial, and this is the role of monetary policymakers. In a NK model, π*=Et[πt+1] is set via i* under the Fisher equation, so monetary policy sets long-run inflation via anchoring expectations. If the central bank cannot credibly commit to raising interest rates in response to increasing inflation, then this would likely generate an inflation bias, as Kydland and Prescott have demonstrated. As such, any signal of resisting fiscal dominance is crucial to keep longer-run inflation expectations in check, and indeed this is one advantage that following a Taylor rule (or even resisting the pressures of fiscal policymakers, as the BoE appears to do) confers.

However, one dynamic that I neglected is that this could produce a “tug of war” between monetary and fiscal policymakers. If central banks continue to resist monetising budget deficits, will this induce elected officials to take steps to erode central-bank independence? David Beckworth argues such, and indeed believes that this overshadows the recent Lisa Cook scandal. If the answer to this question is yes, then operational independence is overrated, and fiscal dominance is underrated, in our models of inflation. Central-banks can only credibly commit to resisting fiscal dominance for as long as elected officials continue to allow their operational independence.

Hence, inflation really is a fiscal phenomenon, and political-economy considerations rise in salience in our frameworks. Given that debt-to-GDP, budget deficits, and inflation is high across advanced economies, I worry about how this dynamic will play out. Whether central-banks can maintain their operational independence or not is crucial for the future path of inflation and macroeconomic stability.