The autonomy of the Central Bank and the pursuit of inflation targets have rarely been as relevant as they are now, both internationally and here in Brazil. In the global scenario, after four decades of inflation under control and falling interest rates, we are now witnessing an environment of higher, persistent, and volatile inflation. This is forcing central banks to tighten monetary policy by rapidly increasing their basic interest rates. In this situation, there is a risk that by pursuing very low inflation targets, central banks may be unnecessarily sacrificing economic activity and employment. Moreover, the quantitative easing programs applied by various monetary authorities, through the purchase of government bonds, have raised questions about the true independence of central banks, exposing the gray area between the monetary policy and government fiscal policy.
Olivier Blanchard, a French economist and former chief economist of the International Monetary Fund, argued last year¹ that the inflation target should be 3%, although having previously advocated for a higher target of 4%. The main argument is that by pursuing a higher target the Central Bank would have a wider margin of maneuver to cut interest rates when necessary, as it would be further away from the zero lower bound. Blanchard argues that when inflation is higher than 4%, the public becomes more aware of it, and this increases the risk of contaminating expectations. Therefore, a 3% target would be more appropriate.
Historically low unemployment in the United States has led the Fed to keep raising interest rates in pursuit of its long-term target of 2%. However, there is the feeling that when the unemployment rate increases, and with inflation already at 3%, the American central bank will permanently accommodate higher inflation.
In Brazil, the undeniable problem of very high actual interest rates raises the question of how feasible it is to reach a target considered ambitious by the country’s historical standards and economic structure. By July of this year, Brazil’s National Monetary Council (local acronym CMN) will decide on the inflation target for 2026. Representatives of the new government, including the President, have been harshly critical of the Central Bank’s recent performance and are calling for an immediate cut in interest rates. There are also rumors that the inflation targets will be revised and that the law passed in 2021 giving the Central Bank autonomy will be overturned. Furthermore, the current government will be appointing two directors to the Central Bank this year and the term of its current chairman expires at the end of 2024. These issues have raised the risk premiums required for Brazilian assets.
To shed light on all these issues, we will review the evolution of central banks and the economic policy of combating inflation, bringing elements of economic theory that shape the way these institutions think and act. To do so, we will elaborate on the concept of independence and autonomy of Central Banks and discuss the inflation targets system which are the cornerstones of the central banks’ actions today.
It is worth clarifying beforehand that there is no intention in this Letter to evaluate the judgment of any specific economic policy or theory. The inflation phenomenon itself is multifaceted and different schools of economic thought emphasize different aspects of the process. Finally, we will discuss the Brazilian case and recent scenario and present our conclusion.
Independence and Autonomy of the Central Bank
A central bank (CB) is the governmental institution responsible for the following functions: i) money issuer, responsible for the functioning of the payment system and conducting monetary policy; ii) depositor of bank reserves, acting as the bank for the banks and depository of government funds; and iii) lender of last resort, responsible for the stability of the financial system and acting as the regulatory body.
The first question to tackle is: if we are talking about a governmental institution, what does an independent central bank mean? Independent in relation to whom?
At this point it is worth reflecting on the origin of this institution. According to Senna (2010)² “historical analyses of monetary systems usually contain few references to central banks. The reason is simple: it is a recent institution, incomparably younger than money and monetary policy... this modern institution has experienced its own evolutionary process... A central bank as we know it today is the result of transformations processed in pre-existing institutions”.
In fact, the first central banks were founded in Sweden in 1668 and England in 1698. Both institutions originated from private initiatives but in some way solved a problem related to the government, so they were therefore closely linked to it. The Bank of England was founded to act as the Crown’s bank and finance the war effort against France, with the king and queen as its main shareholders. The Swedish Central Bank originated in the Bank of Stockholm, a private bank that pioneered the issue of notes backed by copper, which was the metal used as money.
The chart below was elaborated in 2020 and shows the number of central banks in the world at the time when the main CBs of developed countries emerged, and the number of CBs considered as independent.
The complexity of the relationship between this institution and the other governmental sectors can be seen straight away. Therefore, when we talk about the independence of the CB we are referring to its freedom from the rest of the government administration. Alan Blinder, former governor of the Federal Reserve and scholar of monetary affairs, writes in his book on Central Banking that “the independence of the central bank means two things: first, that the CB is free to decide how it tries to achieve its objectives, and second, that it is very difficult for any other sector of government to overrule its decision.”
Blinder does not refer to independence as the CB’s freedom to set its own goals but rather the way of achieving them. In this sense, specialized literature separates independence of objectives from the independence of instruments³. The first refers to the CB being free to determine its own objectives and the second to its freedom to choose how it achieves these goals. At this point it is worth digressing on the distinction between independence and autonomy. Although they are synonymous and commonly used interchangeably, we believe that independence in this sense is related to the absence of political influence on how the monetary policy is conducted. While autonomy is related to the degree of control over operational issues and decision-making processes within the central bank.
In practice, the relationship between the CB and the government involves aspects such as the role of rulers in appointing and dismissing directors, voting power in committees, the degree of budgetary control to which the institution is submitted, how much it can lend resources to the government and whether its objectives are clearly and explicitly defined in its by-laws. How free is the central bank from the pressures of the President or Congress? Is it guided by economic or political considerations?⁴
The benchmark study measuring the independence of central banks was published in 1992⁵ in which the authors analyzed 72 countries over a significantly long period. The methodology that became popular involves drawing up an index ranging from 0 to 1 that takes into consideration criteria related to: i) the central bank’s executive leadership; ii) policy formulation; iii) the bank’s objectives or mandates; and iv) limits on credit operations with the government. Other works evaluate both the degree of legal autonomy of the CBs and also aim to assess the degree of independence of central banks using alternative measures, such as questionnaires or turnover rates of the institution⁶.
The current practice in democratic societies is usually to choose the objectives of the monetary policy and the central bank through elected politicians who delegate the task of fulfilling them to the monetary institution. Through the Federal Reserve Act, the US Congress established that the Fed should conduct monetary policy “in a way that effectively promotes the goals of maximum employment, stable prices, and moderate long-term interest rates”. In Brazil, the law that governs the Brazilian Central bank (BCB) stipulates in its first article that “the BCB has as its fundamental objective to ensure price stability. Without prejudice to its fundamental objective, the BCB also aims to safeguard the stability and efficiency of the financial system, to smooth fluctuations in the level of economic activity and promote full employment.”
Although there is a negative correlation between central bank independence and inflation⁷ it is very difficult to establish a consistent causal relationship between these two variables and, as often occurs in economics, the empirical studies are inconclusive. However, economic theory and intuition support the idea that an independent central bank committed primarily to price stability tends to produce lower and more stable inflation. This is a good starting point to discuss the inflation targeting system but first we will briefly summarize how central banks think about monetary policy.
How Central Banks think
In the first chapter of his book “Interest and Prices”, Michael Woodford⁸ outlines the academic consensus on the performance of modern central banks at the turn of this century. Much of this consensus is based on the neo-Keynesian economic theory of the 1970s and 1980s which, in turn, incorporates fundamental contributions from neoclassical economists. Prominent among these is the central role of the expectations of economic players and the more rigorous modeling of these expectations. This consensus believes that monetary policy has only short-term impacts on economic activity, as measured by the relationship between employment and inflation, the famous Phillips curve. However, in the long term, monetary policy affects only inflation and not real variables such as output and the unemployment rate.
Another operational issue that was consolidated was the use of very short-term interest rates in the money market, the basic rate, as the main instrument for carrying out the monetary policy. This gives us the basic elements of the monetary policy consensus: the central bank controls the very short-term interest rates to affect the economic demand and, given the lagged effects, affect the inflation rate⁹.
The early 1970s were marked by the end of the convertibility of national currencies into gold and, as a result, the end of the Bretton Woods agreement. This ushered in monetary systems that were completely fiduciary¹⁰, with no kind of anchor for price levels. It is no wonder that the 1970s and 1980s were the decades of high and volatile global inflation which brought damaging economic effects. There was a need for a new anchor, i.e. a nominal variable such as the money supply or the inflation rate itself that “tied” price level, providing price stability.
Price stability is understood as a low and stable inflation rate¹¹. The diagnosis of the inflationary experience of the 1970s and 1980s was that supply shocks accommodated by the monetary authorities, whether resulting from political pressure or ignorance of how the economy works, ended up contaminating expectations and produced a wage-price spiral. Another relevant related debate was the issue of “rules vs. discretion” in the conduct of the monetary policy. Economic policy makers are subject to the problem of intertemporal inconsistency, i.e. decision-makers’ preferences change depending on the time horizon. In this situation, monetary policy rules would be superior to discretionary ones.
The inflation targeting system was the method adopted by the main central banks to deal with all these issues. In it, the inflation rate chosen as the target is the anchor for the price level itself. It is a monetary policy framework characterized by the public announcement of quantitative targets for the inflation rate – which can be a number or tolerance bands – in one or more time windows, with an explicit recognition that low and stable inflation is the primary objective of monetary policy over the long term.
Ben Bernanke, former chairman of the Fed and another leading academic economist, argues that inflation targeting (IT) provides a better balance in the trade-off between credibility and flexibility. IT would have constrained discretion. In his recent book¹² Bernanke reflects on how he implemented inflation targeting in the US and says that “if the inflation target is credible, players should disregard temporary changes in inflation such as food and energy price shocks without incorporating them into long-term expectations and the behavior of the definition of prices and wages”.
The first central bank to openly adopt the system was in New Zealand. It was not until 2012, with Ben Bernanke as chairman of the Fed, that the US formally adopted IT (a process Bernanke described in chapter 7 of his book). The table opposite lists the central banks that have adopted IT.
Several questions arise when operationally implementing inflation targeting. For example: What would be the ideal inflation target? How much should the inflation tolerance band range? What time horizon should be considered for the achievement of the targets? Under what circumstances would changes in the targets be justified? It should be noted that there are no obvious answers to these questions and all of them, in some way, involve the aforementioned dilemma between credibility and flexibility.
Strictly speaking, price stability means inflation close to zero. But there are reasons why zero inflation is not suitable. These reasons range from a bias towards higher inflation as measured by price indices, to the desirable safety margin for avoiding a harmful deflationary spiral (a very important concern for the central banks of developed countries until recently), to the fact raised by Blanchard, mentioned above, about the lack of room for interest rate cuts if necessary. Bernanke says “these risks suggest that the inflation target should be set above zero – at around 1% to 3%”. In practice the central banks of developed countries like the US, UK, Canada and Australia chose 2% as their target while emerging countries, which are more sensitive to economic and financial shocks, chose higher targets and/or wider tolerance bands. Chile, Mexico and Colombia, for example, have a 3% target, and Brazil had a target of 4.5% between 2005 and 2018.
The question of justifying changing the targets is related to the “flexibility with moderation” the system allows. One of the main justifications, a common point in the literature, is the occurrence of strong supply shocks, such as high energy and food prices. Central banks have absolutely no control over these and any ensuing increase in interest rates to fend them off would only worsen the decline in economic activity. Another form of temporary change commonly used in the presence of shocks is the extension of the time frame for the convergence of inflation to the target. This is the mechanism central banks have currently chosen to deal with the shock of the pandemic and the war between Russia and Ukraine.
Finally, it is worth highlighting the importance of the credibility and reputation of the central bank in anchoring inflation expectations around the target. The greater this anchoring, the lower the cost of disinflation in terms of economic activity and employment, the so-called sacrifice ratio.
The Brazilian Case
The first bill to grant autonomy to the BCB was sent to the National Congress in early 1989 (PLP 200/1989). However, it was only in 2021 that the legal framework regulating its technical, operational, administrative, and financial autonomy was approved through a complementary law (PLP 19/2021). This led to the institution being separated from the Ministry of Finance (which was part of the Economy Ministry at that time).
The structure for conducting the Monetary Policy today operates basically as follows: the National Monetary Council (CMN), currently composed of two members of government ministries (Finance and Planning) and the Chairman of the Central Bank, determine the inflation target (measured by the IPCA consumer price index) to be pursued in the following years. From then on, it is the CB’s responsibility to pursue this aim through its instruments, primarily by defining the Selic rate.
The presidency of the monetary authority, along with the board of directors, is appointed by the Brazilian President for fixed terms of four years (with the possibility of being re-appointed for a second term) which do not coincide with the President's term of office.¹³
Since the implementation of the Inflation Targeting Regime in 1999 (target of 8% at the time) until 2022, inflation remained within the target limits established for each year 16 times (about 71% of the total). On seven other occasions, it exceeded the ceiling, and only once, in 2017, did it fall below the lower limit (although by only 5 basis points¹⁴)¹⁵. At the time, Ilan Goldfajn, who had become chairman of BCB in the middle of the previous year, wrote an open letter justifying what had happened. He stated in the document that the reason why inflation was below the target was due to a number of factors, including the economic recession, the fall in food prices and the implementation of effective monetary policies.
There was some pressure to raise the target when Goldfajn took over but the final decision taken was to maintain it. Prior to this, inflation targets had been revised upwards twice. Once was in 2002 when Armínio Fraga was the BCB’s chairman, amid great political and economic uncertainty on the eve of the presidential election in the face of a sharp BRL depreciation. Shortly afterwards, in 2003, the newly appointed chairman, Henrique Meirelles, announced that his interest rate policy would aim for an inflation target of 8.5% in 2003 and 5.5% in 2004. In this case, the 2003 target was not officially revoked but was missed. In both cases, the monetary authority felt that the cost of rapid disinflation (probably leading to a recession) was too high.
There was some stability in the inflation target after this period and it was maintained at 4.5% from 2005 to 2018. It was only in 2019 that the inflation target began to be gradually reduced – by 25 basis points a year – until it reaches 3.0% in 2024. This level is in line with what can be seen in other emerging economies such as Mexico, Chile and Colombia. The target range, on the other hand, was kept stable at 2.0 percentage points (up or down) from 1999 to 2016 and was cut to 1.5 p.p. from 2017 onwards.
Recently, as mentioned in the introduction of this text, the discussion about the ideal level of inflation target in Brazil has resurfaced recently. The accompanying graph shows the detachment of inflation expectations from the 3% target for the years 2025 and 2026 throughout this year. This divergence coincides with the attacks on the Central Bank’s autonomy and the current inflation target.
This de-anchoring is problematic as it alters the behavior of the agents who set prices and is reflected, for example, in an increase in the passthrough of currency depreciations¹⁶.
The discussion itself is not a problem. Arguments against a change are based on the similarity with the targets of other emerging countries and the reputational cost of changing the target at a time of high inflation. However, there are good arguments on the side of those calling for a change, ranging from the economic cost of achieving the specified targets against a backdrop of economic slowdown, the high level of indexation of the Brazilian economy and the greater rigidity of the labor market, to the difficulties in reconciling low targets at a time of fiscal fragility marked by high gross debt and little room for the expansion of public expenditure.¹⁷
Granting autonomy to Brazil’s Central Bank was an important institutional advance and, in combination with the inflation targeting system, created a powerful force to keep inflation under control. This was a hard-won achievement.
It is not a taboo for society to discuss and perhaps change the framework of the inflation targeting system, the target itself or improve it over time in line with experience. The IT framework is flexible, and the Brazilian Central Bank has currently used this flexibility by extending the horizon in which it is concerned about the convergence of inflation to the target.
In any case, it is quite possible that the National Monetary Council will change Brazil’s inflation target to 4% which is why inflation expectations in the Focus Bulletin are stabilizing around this level. However, the communication and implementation of this change are of paramount importance. It would be a mistake in our view to increase the inflation target in hasty manner as a means of pressuring the Central Bank to cut interest rates. This would be a harsh blow to its autonomy and the inflation targeting framework.
- Blanchard, Olivier. It is time to revisit the 2% inflation target. Financial Times. November 2022.
- Senna, José Júlio. “Política monetária: ideias, experiências e evolução”. Rio de Janeiro: Elsevier, 2010.
- Debelle, Guy; Fischer, Stanley. How Independent Should a Central Bank be? 1994.
- Walsh, Carl E. Central bank independence. 2005.
- Cukierman, Alex; Webb, Steven b.; Neyapti, Bilin. Measuring the Independence of Central Banks and Its Effect on Policy Outcome. 1992.
- Jácome, Luis I.; Pienknagura, Samuel. Central Bank Independence and Inflation in Latin America—Through the Lens of History. IMF Working Paper. 2022.
- Alesina, Alberto; Summers, Lawrence H. Central Bank Independence and Macroeconomic Performance: Some Comparative Evidence. 1993.
- Woodford, Michael. Interest and Prices. 2003.
- Goodhart, Charles A. E. What Should Central Banks Do? What Should Be Their Macroeconomic Objectives and Operations? 1994.
- Read more about this issue in our Letter 34, “Money, Technology and Future”, available on our site: turimbr.com/publicacoes/carta-turim/
- Alan Greenspan, Fed president from 1987 to 2006, gave his definition in testimony to the American Congress in 2005: “that state in which expected changes in the general price level do not effectively alter business and household decisions.”
- Bernanke, Ben S. The Federal Reserve from the Great Inflation to COVID-19. 2022. & Bernanke, Ben S., Laubach, Thomas, Mishkin, Frederic S., & Posen, Adam S. Inflation Targeting: Lessons from the International Experience. 1999.
- The BCB chairman has a mandate of 4 years beginning on the first day of January of the third year of the Brazilian president’s mandate.
- 1 basis points or 1 bp = 0.01% or 1/100 percentage points
- According to a BCB presentation to the Federal Senate Economic Affairs Committee of 2023, Chile, Colombia and Peru (which adopted inflation targeting in the same year or after Brazil) had inflation above the target in all eight years while Brazil only had seven.
- We watched the debate between Carlos Carvalho and Marco Bonomo in the CDPP center in April on “Determining Individual Prices When the Expectations are De-anchored”. https://iepecdg.com.br/periodo-seminario/2023-1/ They will shortly publish the paper "Price Setting when Expectations are Unanchored" on the theme.
- Araujo, A.; Santos, R.; Lins, P. C.; Valk, S. Inflation Targeting under Fiscal Fragility. 2020.