Itaú BBA - When to cut the inflation target?

Macro Vision

< Back

When to cut the inflation target?

March 6, 2017

In June this year, there may be a window of opportunity to reduce the Brazilian inflation target.

By the time the decision about the 2019 inflation target is made, in June this year, there may be a window of opportunity to begin the convergence of the Brazilian inflation target towards international standards.

The opening of this window of opportunity will depend on three factors: progress of pension reform (signaling an effective fiscal adjustment), the evolution of the external scenario and the decline in current inflation in Brazil. We believe that, by mid-year, there will be greater clarity on these issues.

Even with an economic recovery, the labor market gap will continue to exert disinflationary pressure, keeping inflation in a downward trend. Thus, if these three factors materialize, it will be possible for the authorities to reduce the inflation target with no output costs, i.e., without having to act so as to delay or reduce the pace of economic recovery.

In a recent study[1] , we have shown that well-anchored expectations, the widespread decline in current inflation and important institutional reforms pave the way for a sustained drop in interest rates, and that a reduction in the inflation target would reinforce this trend.

We now present the main arguments supporting a reduction in the inflation target and why this should be done as early as the June meeting of the National Monetary Council (CMN)[2] .

Is it worth reducing the inflation target?

Brazil's inflation target is very high compared with other countries’. While developed countries have an average inflation target of 2.1%, emerging economies have, on average, a 3.3% inflation target – both significantly below Brazil’s 4.5%, which could go as high as 6%, given the generous range around the target, to accommodate shocks. Only South Africa and Turkey have higher targets than Brazil. The convergence of inflation towards lower levels would be positive for the Brazilian economy, as it would create room for lower interest rates and decrease other economic costs related to the uncertainty and distortions generated by high inflation.

Many argue, however, that it would be premature to reduce the target without the monetary authority first having brought inflation within the current target for at least one year. In fact, over the past decade, the Brazilian Central Bank has delivered inflation at (or below) the target on only a few occasions. More specifically, since 2009, inflation has ended the year above the target.

In our report Macro Vision - What if the inflation target is reduced?, we have shown that if monetary policy is perfectly credible, a reduction in the inflation target enables lower inflation and lower nominal interest rates. Measuring the credibility of monetary policy, however, is a difficult task. We believe that the best metric is not the number of times that the central bank has achieved the inflation target in recent years, but how much the economic agents believe in the BCB's commitment to delivering the inflation target over the coming years. This can be measured from inflation expectations or asset prices.

The latest data on inflation expectations, released weekly in the Focus report, show that some market participants already incorporate into their forecasts the possibility of an inflation target reduction for the coming years. The average expectation is below the target for 2018, 2019 and 2020 (4.43%, 4.36% and 4.31%, respectively)[3] . Based on the prices of bonds indexed to inflation, we are able to infer market participants’ expectations for inflation in the years ahead. The implied inflation rates for the next two, three and four years are already below 4.5%.

The anchoring of inflation expectations indicates that monetary policy is credible and therefore would not be a hindrance to the reduction of the target as the central bank tries to make it more compatible with international standards.

Approaches to reducing the inflation target 

The monetary authority can choose one of two different strategies to reduce the inflation target.

The first is the traditional approach, in which the target is first reduced to force the subsequent fall in inflation. In this case, the BCB would need to adopt a more restrictive monetary policy, so as to slowly bring inflation down to the new target. This approach generates costs in terms of output, as it entails higher interest rates to allow convergence. Nevertheless, this approach is usually valid, given that keeping inflation at exceedingly high levels is also costly.

But economic conditions may occasionally enable an alternative. The second approach derives from the "opportunistic disinflation" theory elaborated by Orphanides and Wilcox in a 1996 study[4] . In this study, the authors argue that disinflation should not necessarily be forced by monetary policy. But if, for some reason, inflation falls below the target (for instance, in a recession), the monetary authority should take advantage of such periods of lower inflationary pressures to renew its commitment to lower inflation rates, i.e., to reduce the target in order to allow disinflation[5]. The advantage of this approach is that the disinflation process would not incur additional costs in terms of output.

Colombia, for example, adopted the opportunistic approach in 2009, when the monetary authority took advantage of a period of lower inflation (2%) to reduce the inflation target for the following year. In October of that year, Colombia’s central bank announced a target reduction to 3% for 2010 (from 5% in 2009). The target reduction allowed lower inflation in the following years, and also lower nominal interest rates, with no major impact in terms of output[6] .

Why will the window of opportunity to reduce the target likely open this year?

There are two reasons. First, by mid-year, we will likely see inflation running below the 4.5% target. According to our forecasts, inflation accumulated over 12 months in June will stand at 4.1%, very close to the market forecast. This means that the target reduction would not be about signaling a convergence toward lower inflation in the future, but about keeping it persistently at its lowest level reached without wasting the disinflationary benefit provided by the recession.

Second, the output gap will continue to contain inflationary pressures next year, even if activity surprises market consensus through a strong economic recovery. In other words, growth will not be a limiting factor to disinflation. We will devote our attention to the 2018 inflation forecasts. The reason is that, at the time the CMN considers the decision to reduce the inflation target (in June), the focus of monetary policy will be on 2018 onwards, not on 2017.

We start from a Phillips Curve (described below), in which at each point in time, inflation in market-set prices (πt) is determined from a weighted average of past inflation () and inflation expectations[7] () added to the labor market gap () and specific shocks () (such as exchange rates and commodity prices, for example).

We note that it is not economic growth that affects inflation but the size of the gap, especially the size of the labor market gap (i.e., the difference between unemployment and its equilibrium value).

The essential point is that, even with a strong recovery of the economy, the labor market gap will continue to exert disinflationary pressures. That is, in the absence of shocks, inflation will continue on a downward trend.

The following chart and table show different unemployment paths, taking into account different quarterly growth scenarios. If the Brazilian economy grows at a zero rate in the coming quarters, the unemployment rate by the end of 2018 will reach 14.9%. In a much more optimistic scenario, if the Brazilian economy grows 1.5% on a quarterly basis (implying an annual rate of 6%), the unemployment rate by year-end 2018 will be 11.7%, still substantially above the average (8.4%) observed since 2009 and not far from the 12.5% today.

Evidently, the gap measure and its impact on inflation are subject to errors and uncertainties. Thus, we have simulated in the table below the different inflation rates in 2018 under different neutral unemployment rate (or "NAIRU")[8] scenarios and different elasticity measures between the unemployment gap and market price inflation[9] . For other variables, we have adopted the following conservative assumptions:

(i)     Average unemployment rate at 12.5% (value of the most recent observation)[10][11]   

(ii)    Past inflation ()= Inflation expectations ()= 4.5%[12] 

(iii)   Absence of shocks (ε= 0)

The analysis presented in this table leads us to interesting conclusions. Reducing the inflation target to levels close to that of other emerging economies (3%) already in 2019 would likely not be credible. This is because only under very optimistic assumptions (natural unemployment rate of 8% and high sensitivity of inflation to the labor market gap - β= 0.50 or β= 0.75) would inflation in 2018 stand below 3%. In this case, further recession would be necessary in order to foster a disinflation of this magnitude, and it would not be opportunistic for the BCB to reduce the target.

The reduction of the target to 4%, however, seems appropriate. Even under conservative assumptions, that is, even if the economic agents do not incorporate a target below 4.5% into their forecasts[13] , the labor market gap will exert disinflationary pressure so that inflation would likely converge to levels below 4.5%. An opportunistic BCB would take advantage of this period of lower inflation to reduce the target.


All simulations consider the absence of shocks. Uncertainty over the progress of fiscal reforms and economic and monetary policy in the U.S., however, could change that. We believe that, until mid-year, when the decision on the inflation target is to be made, we will have greater clarity on the progress of pension reforms (we anticipate approval by the National Congress in the second quarter of 2017) and the external scenario. Additionally, we will have a better perspective on whether inflation remains on a downward trend.

Therefore, if the pension reform is not approved or the external scenario deteriorates over the coming months, there may be inflationary pressure arising from further currency depreciation, for instance. In this context, a change in the target would not be appropriate, as the convergence of inflation toward lower values could incur additional costs in terms of output.


The decision on the reduction of the inflation target need not be, nor will it be, made today. In June, however, when the CMN meets to set the target for the coming years, a window of opportunity may present itself, allowing the beginning of a target convergence toward international standards.

A reduction of the inflation target makes sense only if (i) uncertainty over the international and domestic scenarios is significantly lower, and (ii) if it does not generate additional costs in terms of output.

With regard to international uncertainty, we believe that over the coming months we will have a better understanding of the economic and monetary policies in the U.S. Internally, the progress of fiscal reforms and the continuation of the inflation decline will also be clearer.

Additionally, the reduction of the target will likely generate no extra cost in terms of output. There are two reasons for this. First, inflation in June will already be running below the 4.5% target, so that the target reduction would only be a way to corroborate a level that is historically the lowest. Second, even with strong economic growth, which is not even a consensus among analysts, the high level of idleness in the economy and, particularly, the deterioration of the labor market, will continue to create disinflationary pressure during the next year.

Given these conditions, the monetary authority, by adopting an opportunistic approach, will likely take advantage of the recession-driven drop in inflation to assure even lower inflation levels (that is, to reduce the inflation target), which allows lower current inflation and nominal interest rates over the coming years. Opportunities such as this one should not be wasted.

Julia Gottlieb
Felipe Salles

[2] At its June meeting, the CMN sets the target for the second consecutive year and may revise the target for the following year. Thus, next June, the CMN will set the target for 2019 and may revise that for 2018, currently at 4.5%.

[3] As of March 3, 2017.

[4] Athanasios Orphanides and David W. Wilcox - The Opportunistic Approach to Disinflation (May, 1996).

[5] According to Edward Boehne (in 1989): “Now, sooner or later, we will have a recession. I don’t think anybody around the table wants a recession or is seeking one, but sooner or later we will have one. If in that recession we took advantage of the anti-inflation [impetus] and we got inflation down from 4-1/2 percent to 3 percent, and then in the next expansion we were able to keep inflation from accelerating, sooner or later there will be another recession out there. And so, if we could bring inflation down from cycle to cycle just as we let it build up from cycle to cycle, that would be considerable progress over what we’ve done in other periods in history”

[6] For further details, see the appendix.

[7] We have considered inflation expectations three years ahead, in this case. We believe that theseexpectations reflect the view of economists (who provide input to the Focus survey) on the target pursued by the BCB. The calculation of thisvariable was explained in MACRO VISION - How much can 2017 inflation expectations fall?

[8] The neutral unemployment rate,or NAIRU(non-accelerating inflation rate of unemployment),is the rate that causes inflation to remain constant (there is no inflationary/disinflationary pressure arising from thelabor market).

[9] For simplicity, we have assumed the weight for market set prices of 75% and for regulated prices, 25%.Regulated prices increase 4.5% in the simulation.

[10] Note that this rate is consistent with an annualized growth rate above 4% from the first quarter of 2017 onwards.

[11] We have considered, in all simulations, regulated price inflation at 4.5%.

[12] Past inflation in this case will be the 2017 inflation, which according to our forecast will end the year at 4.4%. Inflation expectations for three years ahead are now at 4.4%. Given the possibility of an inflation target reduction, these expectations may retreat further. For simplicity,we will conduct the exercise assuming that expectations will remain at 4.5%, which is consistent with the inflation target at 4.5%. If the target is reduced, the weighted average of past inflation and expectations three years aheadtends tostand below 4.5%.

[13] If the 4.0% inflation target is credible, then the table values will be reduced by approximately 25 bps.




< Back