In a recent column, Şebnem Kalemli-Özcan, professor of economics at Brown University and a former senior policy adviser at the International Monetary Fund, gives voice to a tenet that in the post-Covid era has hardened into mainstream wisdom: in a world of persistent supply shocks, central bank credibility is what really matters. She argues that during the 2022–24 disinflation episode, the Fed’s credibility, not the actual rate hikes, did most of the heavy lifting by ensuring that ‘no one would bother to price in a wage-price spiral, because everyone knows that the central bank would never tolerate one.’
The conclusion is right; the reasoning is not. And in policy-making reasoning is everything.
Who exactly is ‘everyone’?
Kalemli-Özcan’s mechanism, as described, requires that common workers and price-setting firms –— those agents who decide whether to demand higher wages or raise prices – consciously factor central bank credibility into their behaviour. The evidence suggests they do not.
Survey-based research shows that household inflation expectations track personal consumption experiences (petrol prices, grocery bills, energy bills), but do not track monetary policy communications. Large minorities of the population cannot correctly identify their central bank’s inflation target. Similarly, the inflation expectations of small and medium-sized firms are mostly driven by recent input costs, not by inferences about the central bank’s reaction function. And careful event studies find that central bank policy announcements leave household expectations unmoved (even the Federal Reserve’s 2020 framework overhaul went unnoticed), leading even the pioneers of the communication literature to ask whether reaching the general public is promise or false hope.
This is no minor caveat: it strikes directly at the behavioural premise of the dominant narrative. If the wage-price spiral is ultimately driven by workers and firms acting on their own cost experience, then credibility’s power to pre-empt it through expectation management is considerably more limited than the literature typically assumes.
Three channels of credibility transmission
None of this makes credibility irrelevant, but the story commonly told about how it works is wrong. The actual transmission runs through three channels remote from the beliefs of ordinary households and firms.
First, financial markets and professional forecasters do internalise central bank credibility directly and price it into yields, spreads and exchange rates. This shapes borrowing costs for businesses and governments in ways that are real and consequential, even if the firm owners renewing their credit line are not thinking about the Fed.
Second, and most importantly for open economies, credibility affects the exchange rate through risk premia on sovereign liabilities. A credible central bank compresses the premium investors require to hold domestic-currency assets. A less credible one sees that premium rise, the currency depreciates, and import prices climb, feeding inflation directly, without any household ever having read a monetary policy statement. This channel is automatic, powerful and entirely bypasses the ‘everyone knows’ story. Kalemli-Özcan’s own work with Filiz Unsal, published earlier this year, provides direct evidence: emerging markets with stronger monetary policy credibility weathered the 2022–23 global tightening cycle with significantly less financial stress through this market-mediated channel.
Third, credibility works less like a daily restraint on behaviour and more like an insurance policy whose value is revealed only in extremis. What it prevents is not the routine price increase or wage negotiation, but the moment when a shock is large enough to push inflation dynamics into a self-reinforcing spiral – with rising prices feeding wage demands feeding further price increases and so on – which eventually becomes extraordinarily expensive to break. At that point, a central bank with an established track record can intervene decisively and be believed by financial markets quickly, compressing the spread and exchange rate dynamics that would otherwise amplify the spiral: the monetary restriction – which remains the instrument that actually does the work – succeeds at a lower sacrifice ratio. A central bank without that same track record faces a much steeper climb. Türkiye found this out between 2021 and 2022, when sustained political interference in monetary policy pushed inflation above 80% and Argentina has been finding it out for decades.
The cross-country disinflation test
The comparison Kalemli-Özcan draws between the disinflation speeds of the US, the euro area and the UK is rhetorically powerful, but as identification it is weak. The three economies faced very different shocks. The US episode was disproportionately demand-driven (fiscal stimulus and an overheated labour market), which conventional tightening could address directly. The euro area absorbed a terms-of-trade shock from energy, transmitted across member states with very different debt burdens and fiscal positions, through a single policy rate set for economies at divergent cyclical positions. The UK combined the European energy shock with a post-Brexit labour supply constraint entirely outside any central bank’s reach. None of this has anything to do with credibility per se. Nor is there much credibility variation to exploit by any market measure, all three central banks remained credible throughout.
Attributing the speed differential to credibility would require an empirical exercise that the column does not perform, but which Ben Bernanke and Olivier Blanchard did. Their analysis shows that the size and composition of supply shocks and their reversal account for the bulk of the observed cross-country inflation dynamics in the pandemic-era. Interestingly, where they do credit credibility – in preventing 1970s-style persistence – the mechanism is revealing: long-run expectations, measured largely from professional forecasters and markets, did not drift and wage-setting today is far less indexed to past inflation than in the 1970s. Credibility, that is, acts more as a slow-moving institutional stock, embedded in contracts and priced by markets, than as a flow inducing workers and firms to defer to the central bank’s resolve.
A more realistic framework
Rather than a binary switch that either anchors expectations or does not, credibility is better treated as a priced attribute of monetary frameworks that is reflected in the sovereign risk premium and in the sensitivity of inflation expectations to external shocks.
On this view, the effects of credibility run strongest through financial market channels and exchange rate dynamics, especially in economies where currency pass-through to prices is large and rapid. Recent IMF research confirms that emerging markets with stronger monetary frameworks suffered smaller output losses during global risk-off episodes.
Now, if the dominant credibility channel runs through financial markets, then threats to central bank independence are most dangerous because of what they do to sovereign spreads, exchange rates and capital flows, not because they dislodge the wage expectations of the average firm or worker. The prescription is the same as Kalemli-Özcan’s, but the reasoning is more realistic and institutional safeguards should be designed around the real transmission mechanism.
Kalemli-Özcan is right that credibility matters and that it is under threat at precisely the wrong moment. Yet, the mechanism she invokes –‘everyone knows’ – rests on a micro-foundation that the evidence does not support. Ordinary agents do not, by and large, monitor central bank credibility. The real work is done by financial markets, exchange rate dynamics and the option value of not triggering a spiral in the first place.
Acknowledging this does not weaken the case for central bank independence. It strengthens it by grounding it in a more realistic account of how monetary credibility works. And it makes clearer what is at stake when it is lost.
Biagio Bossone is an adviser to international financial institutions and national central banks.
Interested in this topic? Subscribe to OMFIF’s newsletter for more.