Let me today address one aspect of the latest monetary policy cycle in Brazil, namely the risk-management approach embedded in the decision by Brazil’s Central Bank (BCB) to hike basic interest rates by 375 bp between September 2004 and May 2005, followed by a gradual easing that has been in course since September 2005 (725 bp so far and counting…). I believe the episode illustrates how there still is a blind spot not frequently addressed in discussions about assignment of responsibilities between non-monetary authorities and operationally independent central banks in the context of inflation targeting regimes in Brazil and elsewhere.
At the time those decisions to raise interest rates in Brazil were made, some critics characterized them as an over-reaction to what they called feeble signs of return to higher inflation. As effective inflation moved to the range bellow mid-target in 2006, the same critics suggested this would be a demonstration of the over-kill monetary stance adopted by the BCB.
A shortcoming of over-simplified versions of this critique is their exclusive reference to observed data, thus keeping out of sight those scenarios that were precluded by the same monetary policy decisions. In fact, one should take into account all possible alternative scenarios with significant probabilities and consequences, even when doing an ex post assessment.
In order to clarify my argument, let me resort to Alan Greenspan’s oft-quoted speech in Jackson Hole in 2005 – Reflections on central banking – where he formulated his view on the risk-management approach to monetary policy that the Fed applied while deciding what to do with interest rates in the summer of 2003. As we all know, the deep reduction of Fomc’s nominal rate to 1% and the “measured pace” with which it was lifted afterwards have been criticized by some as over-lax and as a major factor behind excessive global liquidity and widespread asset bubbles thence forward. In his speech, Greenspan emphasized how policy decisions had – and always have – to be taken in a framework of expected risk-weighted benefits and costs, one in which the total loss-function is to be at the core:
“(…) a central bank needs to consider not only the most likely future path for the economy but also the distribution of possible outcomes about that path. The decision makers then need to reach a judgment about the probabilities, costs, and benefits of various possible outcomes under alternative choices for policy. (…) point forecasts need to be supplemented by a clear understanding of the nature and magnitude of the risks that surround them. In effect, we strive to construct a spectrum of forecasts from which, at least conceptually, specific policy action is determined through the tradeoffs implied by a loss-function.” He then goes on to explain why the option of sticking to an apparently over-soft policy was taken:
“In the summer of 2003, for example, the Federal Open Market Committee viewed as very small the probability that the then-gradual decline in inflation would accelerate into a more consequential deflation. But because the implications for the economy were so dire should that scenario play out, we chose to counter it with unusually low interest rates. The product of a low-probability event and a potentially severe outcome was judged a more serious threat to economic performance than the higher inflation that might ensue in the more probable scenario. (…) Given the potentially severe consequences of deflation, the expected benefits of the unusual policy action were judged to outweigh its expected costs.”
In my view, the context of monetary policy decisions in Brazil in mid-2004 resembled an inverted image in the mirror of the one depicted by Greenspan for the US in the summer of 2003. In terms of a loss-function – in which the product of probability times loss given the event is what matters – the damage of a failure in curbing inflationary threats at that moment in Brazil was considerable.
The inflation targeting regime implemented since 1999 was still in the middle of its trajectory of downward annual targets toward long-run levels closer to the rest of the world when it was hit by the tremendous confidence shock of 2002. Inflation targets were missed in 2002-03 and adjusted targets had to be established.
Market expectations about inflation started to exhibit strong downward rigidity in 2004, what seemed to express a lack of confidence on the capacity of monetary authorities to withstand enormous political pressure toward softening. After all, the lack of legal operational autonomy and the painful GDP-effect of monetary policy in the beginning of President Lula’s mandate were read by many as enough to undermine the BCB’s capacity to react to the imported shock of increases in commodity prices abroad that was behind creeping inflation hikes at the time.
In that context, independently of the likelihood to be associated to a scenario of inflation levels moving upward from the then prevailing ones (still above 7% p.a.), one can argue that future sacrifice ratios would have risen substantially in the case of a failure to halt the contamination between expectations and creeping inflation. Thus, “loss given failure to counter inflation” – or “loss given not mastering expectations and weakening reputation” – might be seen as justifying the 375 bp hike in basic interest rates, even if that augmentation was liable to be deemed as exaggerated. To those who argue that the speed of convergence of inflation toward levels below 4.5% (the mid-point of the target range) is a proof of excess monetary conservatism, Brazilian central bankers can point out the risk-management approach and that “given the potentially severe consequences of [inflation], the expected benefits of the unusual policy action were judged to outweigh its expected costs”, adapting Greenspan’s remarks.
To finalize, let me refer to a blind spot in usual discussions on the task division between democratically elected non-monetary authorities and an operationally autonomous central bank in inflation-targeting regimes. The former take the responsibility of assessing trade-offs and defining inflation targets, whereas the latter receives full autonomy to fulfill its contractual commitment, namely to deliver target inflation. Central bank’s accountability insofar as its decisions over time is possible by examining these decisions at the light of means-ends relationships. Nonetheless, when it comes to the subjective component of assessments inevitably present in risk-management decisions, the borderline demarcating responsibilities is somewhat blurred, in formal terms at least.