The authors would like to thank the two anonymous referees whose comments helped to improve and to bring clarity to the paper. All remaining errors are the authors’ responsibility.
1From the 1990s, a recurring question in discussing the Brazilian economy is: why does Brazil have one of the world’s highest real interest rates? Indeed, one would expect that after the Real Plan’s successful stabilization of inflation in 1994, interest rates would have dropped to levels closer to those found in developed economies plus some premium to cover Brazil’s sovereign risk. This has not been the case. Even with the recent sharp drop in real interest rates within the context of an acute and protracted recession and high unemployment, Brazil still has one of the highest real interest rates: in October 2018, the ex-ante real interest rate (net of inflation projected for the next 12 months) was at 3.71% p.a. It was the world’s sixth highest, well above the general average of 0.56% p.a.1 Several explanations have been suggested, such as the presence of “clogged” monetary policy channels due to financial indexation, earmarked credit, monitored prices, etc., not to mention other factors like jurisdictional uncertainty, the public sector’s financial weakness, contagion from public debt, questions surrounding the public debt’s sustainability, among others. Some of these lack plausibility and empirical evidence, but Brazil’s high interest rates are likely to stem from a wide range of factors, as a paper on the subject pointed out (Bresser-Pereira & Nakano, 2002).
2This paper goes back to and develops the hypothesis of a conservative monetary policy convention in Brazil – as formulated by authors like Bresser-Pereira and Nakano (2002), Erber (2011) and, more recently, Lara Resende (2017), in the framework of a financialization process – a process which promotes the expansion of the financial industry and of the financial motives in agents’ portfolios at the cost of the real economy. In addition, the paper uncovers three related factors supporting the hypothesis of a conservative monetary policy convention in Brazil: (i) the formation of a class coalition of rentiers capitalists earning interest, dividends and rents, and financiers earning high salaries, bonuses and commissions; (ii) the presence of a “two-way” public debt contagion effect between the banking reserves market and the public securities market, where both the Central Bank’s conservative monetary policy and the National Treasury’s difficulty in issuing debt may affect the return on (and terms of) financial operations in the reserves and securities markets; and (iii) the use of a high interest rate to finance current-account deficits understood as “foreign savings” but which are essentially additional consumption expenditure.
- 2 It should be remarked that the analysis in this paper covers the period 1990-2018. In 2019, after a (...)
3To this end, the paper is organized into four sections, in addition to this introduction. Section 2 analyzes some explanations for Brazil’s high interest rates. Section 3 develops the hypothesis of the pro-conservative monetary policy convention in Brazil, adopting, as a starting point, Keynes’s view of the interest rate as a conventional phenomenon. Section 4 investigates the nature and unique traits of financialization in Brazil and its income and wealth redistribution effects, while section 5 analyzes the channels by which the rentier-financial class coalition can affect interest rates in Brazil. The hypothesis of the pro-conservative monetary policy convention is developed, and the article concludes with some policy proposals.2
- 3 In this section, we sum up some of the most influential explanations of high interest rates in Braz (...)
- 4 Administered prices include, to name a few, transportation (including oil and oil products), commun (...)
4Out of the many interpretations3 of Brazil’s high interest rates, we mention a few that we believe are secondary, but still noteworthy. One series of studies explains high interest rates as a product of “clogged” monetary policy channels in Brazil that forces the monetary authority to raise the interest rate by a greater magnitude than is otherwise necessary to reduce aggregate demand. One factor claimed to be contributing to Brazil’s poorly functioning transmission channels is the high share of the IPCA (broad consumer price index) represented by administered prices,4 which are not affected by to the interest rate (Modenesi & Modenesi, 2012). This is due to the fact that they are regulated by contracts or public authorities, be they federal, state or local. According to data from the Brazilian Institute of Geography and Statistics (IBGE), in August 2018 administered prices made up 26.2% of the IPCA while free-floating prices made up the remaining 73.8%.
- 5 The BNDES is a state-owned development bank.
5A second factor allegedly contributing to Brazil’s poorly functioning transmission channels is the weight of the Brazilian National Development Bank (BNDES)5 loans as a share of total credit, as these operations fall into the earmarked credit category which is not affected by monetary policy (Daniel, 2015). This factor has been controversial. There seems to be conceptual confusion regarding the assessment of the impacts of earmarked credit on monetary policy transmission: it is generally claimed that the BNDES would hold close to 20% of total credit, but this percentage measures credit volume. For the purposes of assessing the power of monetary policy, however, one must take credit approvals into account. Using this view, the share of BNDES loans averaged a mere 5.1% of total loans in 2011-2014, the bank’s peak credit period, dropping to 2.8% of total credit approvals in 2015-2018, as Figure 1 shows.
Figure 1. BNDES credit as a percent of total credit approvals
Source: prepared by the authors based on data from the Central Bank of Brazil
6A third factor supposedly contributing to Brazil’s poorly functioning monetary policy transmission channels is the continued presence of a far-reaching financial indexation process. The fact that a portion of wealth is pegged to the Selic or basic interest rate creates an “inverted” wealth effect (Pastore, 1996). Under normal conditions (where debt instruments are mainly pre-fixed ones), the wealth effect operates as follows: an increase in the interest rate generates a negative wealth effect on economic agents because it reduces the value of fixed assets, considering that P = A/r, where P is the asset’s market price, A is its coupon, and r is the market interest rate. Therefore, for a given A, if r increases, P must decrease. The loss of wealth forces economic agents (firms and households) to reduce their spending. In Brazil, however, this effect is partly offset by the indexation of a portion of financial wealth: in this case, an interest-rate increase generates a positive income effect that may translate into increased aggregate demand, partly nullifying the initial effect. Figure 2 shows data on the Selic-indexed public debt (which includes repo operations, outstanding treasury financial bills [LFTs], and does not include state-owned company debt) relative to GDP: the average in 2007-2017 was 24.5%!
Figure 2. Selic-indexed gross public debt (2007-2017) – (% of GDP)
Source: Paula & Marconi (2018), based on National Treasury data
- 6 The Sistema Especial de Liquidação e Custodia (Selic) (Special Clearance and Escrow System) is the (...)
7Other explanations for the high interest rates focus on the issue of public debt, some in connection with financial indexation. Oreiro and Paula (2011) argue that a significant share of public debt is indexed to the Selic,6 making the National Treasury hostage to the financial market for public debt issues and rollovers. This interpretation connects with the hypothesis that blames the “public debt contagion effect” initially formulated by Barbosa (2006): because the Selic rate provides returns on both certain Central Bank of Brazil’s repo operations to fine-tune the banking reserves market’s liquidity and a portion of the Selic-indexed public debt (LFTs), it serves two purposes: it is the interest rate that regulates interbank loans and, at the same time, the one at which the Treasury rolls over a significant share of the public debt. Because a single interest rate must perform two functions, the public debt rollover function contaminates the monetary policy instrument function since the Central Bank of Brazil (BCB) is unable to set one Selic for interbank market operations and another for public debt rollover operations. The BCB cannot set a single interest-rate value that is simultaneously performing the two functions. In this context, the persistent fragility of Brazilian public finances ends up making the interest rate that the market requires to roll over public debt “excessively high”. It is then transmitted by arbitrage to regular monetary policy operations.
- 7 There is an underlying argument that a loss of government savings must imply an increase in rentier (...)
- 8 According to uncovered interest parity, the domestic interest rate equals the international interes (...)
8Segura-Ubiergo (2012) argues that the low savings rate in Brazil raises the interest rate set by the BCB. It is the product of three factors: high and insufficiently funded social-security transfers, the high government consumption, and lastly the high return on public debt itself.7 On the other hand, some argue that a history of sovereign defaults and questions surrounding the sustainability of Brazil’s public debt forces the country to pay a high risk premium, which, according to uncovered interest rate parity,8 leads to high interest rates. However, as Bresser-Pereira and Nakano (2002) originally noted, the interest rate at the time was (and remains) significantly higher than what uncovered parity might predict. Besides, Brazil does not have a history of public debt default; its sovereign defaults did not involve the state’s inability to honor debt in its own currency, but rather the nation-state’s inability (especially in its private sector) to honor debt denominated in foreign currency because of the refusal by international creditors to roll over that debt. Further, Reis (2019) shows that other countries with savings-to-GDP ratios similar to Brazil, such as Colombia and the Philippines, have far lower interest rates than Brazil does. As for sovereign risk, Table 1 shows that in 2010-2014 countries with similar sovereign risk (EMBI+) levels had far lower real interest rates than Brazil.
Table 1. Real short-term interest rate* and risk rating – 2010-2014 average
Country
|
Country risk
|
Real interest rate
|
Brazil
|
916
|
4,25
|
Bulgaria
|
881
|
-1,73
|
Colombia
|
766
|
1,19
|
Mexico
|
532
|
0,10
|
|
|
|
Panama
|
1 029
|
0,88
|
Peru
|
923
|
0,16
|
Philippines
|
399
|
0,16
|
Russia
|
659
|
0,67
|
South Africa
|
215
|
-0,03
|
Turkey
|
416
|
-3,55
|
|
Source: Datamarket (EMBI+) and IMF; (*) Central Government
9While some explanations of Brazil’s high interest rates are clearly questionable (such as the weight of BNDES loans as a share of total credit, or the history of sovereign default, etc.), other ones seem to be insufficient, (such as the weight of managed prices on the broad consumer price index and the public debt contagion effect). We will return to this in section 4. Financial indexation, as we will see in the next section, is part of our explanation for Brazil’s high interest rates. We integrate this factor in the Brazilian regime of financialization “through interest income”: the formation of a class coalition of rentiers and financiers whose aim is to keep interest rates high can only be understood as part of this sort of regime of financialization. Further, we point out some consequences for the operation of monetary policy, as detailed in section 4. The hypothesis that a rentier-financial coalition bent on keeping interest rates high has had a sustained impact on those rates has been raised by some Brazilian economists, such as Bresser-Pereira and Fabio Erber, as we discuss in the next section.
10A convention is a belief or expectations-forming rule shared by a large number of individuals, or an agreement between participants settling on a common strategy. More than this, conventions are institutions, which leads Favereau (2002, p. 511-520) to add a normative dimension to them, arguing that they are “endowed with mandatory force”. In the General Theory (2007 [1936]), Keynes suggests two concepts for convention, or agent behavior rule: (i) to assume that the current state of business will persist indefinitely (projecting the current situation); (ii) a propensity to follow the majority, or average opinion (the safest course of action may be to follow others).
11Keynes (2007 [1936]) argued that the interest rate is a highly conventional phenomenon, rather than an essentially psychological phenomenon because “its actual value is largely governed by the prevailing view as to what its value is expected to be. Any level of interest which is accepted with sufficient conviction as likely to be durable will be durable” (ibid., p. 203, original emphasis). He believed that, in order to be effective, monetary policy must send clear signals to the agents:
[…] a monetary policy which strikes public opinion as being experimental in character or easily liable to change may fail in its objective of greatly reducing the long-term rate of interest [....] The same policy, on the other hand, may prove easily successful if it appeals to public opinion as being reasonable and practicable and promoted by an authority unlikely to be superseded.
(ibid., p. 203).
Therefore, expectations surrounding the future behavior of monetary policy depend on what Keynes referred to as the “safe” interest rate, that is, the interest rate value that the public believes will prevail in the long term. In other words, it depends on a social convention.
12The conservative convention for monetary policy is a hypothesis originally argued by Bresser-Pereira and Nakano (2002, p. 169): “after the interest rate is kept at a very high level for a lengthy period of time, it is natural for fears of reduction to emerge and for this level to become conventional”. Bresser-Pereira (2007, p. 200) adds: “the Selic funds rate is high in Brazil because, under the argument that a very high interest rate is required ‘to fight inflation’, [it] is set at an artificial level that compensates rentiers and the financial industry”. That is, in Brazil exists a successful rentier-financial class coalition bent on keeping interest rates high.
13Erber (2011) later suggested that monetary policy strictness should be explained from the angle of political economy, according to which the interest rate in Brazil is not an exclusively macroeconomic problem, but rather the outcome of the formation of a class coalition aimed at keeping the interest rate high. Besides, such a coalition would be beneficial to the reputation of a conservative central bank. In this sense, a convention is created that is shared both by the financial market and the Central Bank. According to Erber:
[…] a broad and powerful constellation of interests exists, formed over time around the high interest interest-appreciated currency binomial, that has established a convention according to which these elements are key to the country’s development [...] This coalition of interests has powerful instruments available to consolidate and disseminate its development convention. The most explicit one lies in the hands of the financial system [...]. The Central Bank is a required member of the coalition [...]. For the coalition and the convention that acts as its social representation to form, all it takes is for the Central Bank and private-sector members to extract shared benefits from a single policy – in this case, the prestige that stems from hitting targets and the profits from high interest and an appreciated currency. (Erber, 2011, p. 43)
14Therefore, the maintenance of high real interest rates for decades in Brazil, within the context of the presence and maintenance of an overnight circuit, led to the creation of a conventionally “safe” interest rate. Thus, a belief in or conviction of continued high rates was formed. Such a vicious cycle greatly contributed to the development of a process of financialization of the Brazilian economy “through interest income”, a central feature of the country’s prevalent rentier-financial form of capitalism.
- 9 The mechanism is that the Fisher relation must hold in the long run, so given a constant steady-sta (...)
15More recently, Lara Resende (2017) argued that Brazil’s sustained high interest rates proved themselves ineffective at lowering inflation and raised the hypothesis that high interest rates may lead to elevated inflation rates, in what has become known as the “neo-Fischerian hypothesis.”9
- 10 For a comprehensive review of financialization on its various dimensions, see van der Zwan (2014).
16The central feature of the development of finance-led capitalism lies in financialization10 – which, according to Epstein’s (2005, p. 3) well-known definition, means “the increasing role of financial motives, financial markets, financial agents and financial institutions in the operation of the domestic and international economies”. Brazil, as measured by several indicators, is a highly financialized economy, but with unique characteristics, as we will discuss below (see also Bruno et al., 2011).
17International studies show that financialization significantly reduces a nation’s autonomy, whether in formulating economic policies independent of international conditions, or in connection with a long-term development strategy consistent with the productive conditions and interests of non-financial sectors (Becker et al., 2010).
18Bresser-Pereira (2018), in his turn, shows that a new class coalition has been dominant in Brazil since the late 1980s – a “rentier-financier coalition”, a social organization in which capitalists are predominantly rentiers, whereas the upper technobureaucrats are either the top managers of the corporations, or the financiers” (ibid., p. 27). In such an organization, the rentiers, most of whom are heirs, have replaced business entrepreneurs in the ownership of the large business firms; the financiers, for the most part, are the young and bright technobureaucrats that manage the rentiers’ wealth. Most of them come from the upper middle class and have earned an MBA or a PhD in economics abroad. Besides their management of financialization, they use the radical economic liberalism that they learned overseas to become the organic intellectuals of this neoliberal class coalition.
19Singer (2012, 2018) holds that over the course of the Workers’ Party (PT) administrations in Brazil, two opposite coalitions of classes formed. One is a rentier coalition that unites financial capital and the traditional middle class, more closely aligned with neo-liberal prescriptions; the other is a productive coalition made up of industrial entrepreneurs in association with the organized portion of the working class. During the first Dilma Rousseff administration (2011-2014), however, excessive interventionism created mistrust among industrial entrepreneurs, who backed away from the administration beginning in the second half of 2013. By bringing about changes in interest and exchange-rate policies, with a drastic reduction in the SELIC rate and bank spreads, Dilma eventually strained the relationship with representatives of financial-rentier segment. Furthermore, in June 2013 widespread protests erupted in the country’s capital that produced a fundamental split in the Dilma government. The political climate of the country was rattled by the episode. The drop in the president’s popularity put the federal government in a defensive stance, with setbacks such as the restoration of higher interest rates by the Central Bank in mid-2013 and “the beginning of the mobilization of the middle class, which would eventually play a decisive role in the fall of Dilma” (Singer, 2018, p. 103). This new window of opportunity caused many businesspeople and others to detach themselves definitively from the governing coalition and protest against the prevailing economic policy (see also Paula, Santos & Moura, 2020).
20Bresser-Pereira (2017) adds another economic factor leading to the collapse of this attempted coalition. Citing data from Rocca (2014), he argues that the first Dilma administration saw a sharp drop in the profit rate for industrial companies, due largely to the marked appreciation of the Brazilian Real in the Lula administration and the resulting loss of domestic market share to manufactured goods imports. Table 2 shows the drop in the return rate for business firms between 2010 and 2014, while the interest rate remained extremely high.
Table 2. Return on equity (ROE) and Selic/Over rate – 2010-2014 (%)
Year
|
ROE
|
Selic/Over
|
2010
|
16.5
|
9.8
|
2011
|
12.6
|
11.7
|
2012
|
7.2
|
8.5
|
2013
|
7.0
|
8.2
|
2014
|
4.3
|
10.9
|
Note: average rates
Source: Rocca (2014); Oreiro & d’Agostini (2017).
- 11 See also Annex 1, where we investigate empirically the relationship between the Selic (real and cap (...)
- 12 Overnight corresponds to operations that banks carry out daily on the open market, in order to obta (...)
21In Brazil, within the framework of a basically liberal economic policy regime ever since the lifting of trade barriers in 1990 and the lifting of financial barriers in 1992, financialization, according to Paula and Bruno (2017), has become an “interest gains financialization regime,” which replaced the previous monetary regime, the “inflation gains financialization regime.” Both processes were stimulated by financial liberalization in terms of capital inflows and outflows, and by the speculative nature of capital flows from residents and non-residents alike. Due to the high interest rates and the high level of public debt in Brazil (which in part are indexed to the Selic rate, see more below), usurious financialization not only remained but increased to a new level to the point where the interest gains appropriated by the big banks and capital holders were drastically amplified by the high cost of financing and loans granted by the financial market to Brazilian households and companies (Bruno & Caffe, 2017). Figure 3 shows how the evolution of net domestic debt (of the central government and the BCB) tracks the accumulated real Selic rate factor. As we can see in the figure the capitalized real Selic endogenously expanded domestic public debt in the 1990-2018 period, as an important share of public bonds is directly denominated in the Selic rate (between 20-40% of the total in 2006-2018)11 The “interest gains financialization regime” that is prevalent in Brazil differs from the “dividends gains financialization regime” that has been a leading feature of financialization in developed economies, because, in Brazil, interest rates, given the presence of an overnight circuit in the Brazilian economy, have been kept at very high rates ever since the 1994 implementation of the Real Plan.12
Figure 3. Capitalized real Selic endogenously expands domestic public debt (1992-2018)
Note: The capitalized real Selic is obtained via the accumulated factor of the real Selic, using the IGP-DI (Prices General Index – Internal Availability), base: Dec 2018 = 100, as a deflator. The formula used to arrive at the monthly real Selic factor. Then simply accumulate the monthly factors over the years in the series.
Source: authors’ calculations based on Central Bank of Brazil data
- 13 This index is inspired by the findings of the empirical literature. Davis (2017), in a broad and re (...)
- 14 As part of the financial assets are equities, the increase in the stock of non-monetary financial a (...)
22Figure 4 shows the behavior of the financialization index (calculated based on the ratio of total stock of non-monetary financial assets to the economy’s total stock of fixed capital13) and of the rate of productive fixed capital accumulation, 1970-2015. The 1.06 financialization index of 1970-1980 does not support the existence of a financialization process in Brazil during that time (Table 3). For every Brazilian Real (national currency) allocated to directly productive activities, there was 1.06 Brazilian Real invested in financial assets. In 1981-1994, under what we call the “inflation gains financialization” regime, the index averaged 2.04, indicating, on the macroeconomic level, that for every Brazilian Real invested in productive activities, 2.04 Brazilian Reais had been invested in financial assets. In 1995-2018, under what we have termed “interest rate financialization regime”, the index reached a rather high average level of 8.63 (for every Brazilian Real invested in productive activities, 8.63 had been allocated to financial assets14). It is worth emphasizing something that was already present during 1981-1994 has gained weight in the 1995-2016 period: a sharp and increasing decoupling between rentier-financial accumulation and the rate of productive fixed capital accumulation.
Figure 4. Rentier-financial accumulation vs Fixed-capital accumulation (1970-2018) – (%)
Note: The accumulation rate was calculated based on the fixed capital stock data in Morandi (2016) and IBGE data. The financialization rate relied on the total stock of non-monetary financial assets data provided by the Central Bank of Brazil, deflated at the IGP-DI.
Source: prepared by the authors
Table 3. Financialization index during three periods of the Brazilian economy’s evolution (1970-2018)
Period
|
1970-1980
|
1981-1994
|
1995-2018
|
Financialization index = f/g
|
1,02
|
2,25
|
8,63
|
Source: authors’ calculations
23In the context of the interest gains financialization regime, the Brazilian government attempted, until 2016, to reconcile the interests of rentier accumulation and redistributive social policies, in order to benefit the social segments whose income derives from interest income and other financial gains (Paula & Bruno, 2017). Kaltenbrunner and Painceira (2018) argue that one of the features of Brazilian financialization is what they call “subordinated financial integration,” a process that connects domestic financialization to international capital flows. Indeed, the international financial integration process, in the context of capital account liberalization, is asymmetric as much as it is an integration between unequal partners: firstly, as capital flows ultimately depend on exogenous sources, emergent countries have become even more vulnerable to the inherent volatility of these flows; in Ocampo’s (2001) terms, whereas advanced economies are “business cycle makers”, emerging economies are “business cycle takers”. Secondly, the relatively marginal insertion of emerging economies’ assets in the portfolios of global investors since the 1990s (as is the case with Brazil) has also contributed to this higher macroeconomic vulnerability as capital flows have a procyclical behavior pattern (Paula, Fritz & Prates, 2017).
- 15 Repo operations are public or private securities which buy (or sell) operations that include an obl (...)
24Specifically, the subordinated financial integration shapes the relationships between agents and the financial markets through carry-trade operations that exploit the interest-rate spreads that stem from Brazil’s domestic interest rates that are very high compared with those in developed economies (such as the US federal funds rates). The connection with the Brazilian economy’s financialization takes place via the international reserves accumulation policy and the Central Bank’s intensive use of repo operations (“operações compromissadas” in Portuguese15) to calibrate liquidity in the banking reserves market. As Pellegrini (2017) points out, the problem is not the Central Bank’s use per se of repo operations, but the amount of these operations in Brazil. The sharp growth of repo operations in 2006-2010 is mainly due to the accumulation of international reserves, forcing the BCB to sterilize accumulated currency flows to prevent greater oscillation of the interest rate in the interbank market.
25Another aspect worth emphasizing is that the BCB policy of accumulation of foreign exchange reserves (which works as a safety cushion in periods of stress) takes place largely through portfolio investments (speculative and short-term investments ) whose inflows are attracted by Brazil’s high real interest rates and the trend to cyclic exchange rate overappreciation, which is ultimately stimulated by the policy of “growth cum foreign savings” (and its mirror-image, current account deficits) and the exchange rate anchor policy to keep inflation in check. These factors contribute to keeping interest rates at a higher level than needed to cover a sovereign risk premium. Indeed, the resulting currency appreciation is convenient for profits, interest and dividends payments abroad, when the time comes to convert these gains in Brazilian Reais into US Dollars. In other words, a floating exchange regime biased in the direction of real appreciation reproduces the Brazilian economy’s subordinated international financial integration, as the attraction of abundant foreign capitals forces the BCB to sterilize these funds, leading to a sharp increase in repo operations. As has already been stressed, this stimulates interest gains.
26In Brazil, the process has some perverse effects. On the one hand, financialization creates a frenzy among capital holders (including industrial entrepreneurs who become rentiers) who develop a preference for liquidity, reducing gross fixed capital formation because of the presence of short-term financial investments that compete with investments in capital assets by increasing the liquidity premium (Figure 4). In this sense, according to Carvalho:
[…] the focus on short-term interest rates may be simply due to the high return that financial operations have to offer compared with the expected return on productive investments, making the choice between a short-term placement and productive investment a relevant one, as such a placement may yield in a few periods what a real investment would take much longer to provide, even if the much higher risks of all kinds surrounding the acquisition of capital assets were to be disregarded. (Carvalho, 2005, p. 332)
27Consequently, financialization leads to the expansion and increased importance of the financial industry (and of the financial motives in agents’ portfolios) at the cost of the real economy, transferring income from the real economy to the financial sector, and even leading to the stagnation and decline of production.
- 16 Only in South Africa, Argentina, Brazil, Colombia and the United States do the richest 1% appropria (...)
28On the other hand, one can expect that interest gains financialization regime increases income inequality as it has concentrating effects on high-income segments, as financial income, dividends, rents, and inheritances and donations, which account for almost 40% of Brazil’s income (Menezes Filho, 2017). Recent studies (see, for example, Souza & Medeiros, 2017; Morgan, 2017) using Personal Income Tax (IRPF) data have enabled a new view of the top of income distribution in Brazil, showing evidence of persistent income concentration at the very top. Souza and Medeiros (2017) show that the richest 10% in Brazil have maintained their appropriated shares of total income, at around 50% in 2006-2014; the richest 1% has 22-24%. For comparison, the range of total income appropriated by the top 1% in 24 out of 29 countries surveyed in 2014 was 5-15%, far below the share that the 1% in Brazil had amassed.16 Morgan (2017), in turn, provides evidence that the share of income of the top 10% showed a small reduction from 54.7% to 53.4% of pre-tax income from 2001 to 2015, whereas the share of the poorest 50% increased from 10.6% to 12.5% and that of the intermediate 40% dropped from 34.7% to 34.1%. The author concludes that, in Brazil, “income growth [...] has been uneven, with gains at the lower range at the expense of the higher one without, however, affecting the groups at the very top” (Morgan, 2017, p. 254). The data thus shows that recently the portion of income received by the richest strata has remained essentially stable in Brazil.
29The Brazilian economy’s increasing financialization is associated with a high interest rate, as we have seen. We have identified five channels through which the rentier-financial coalition of interests can affect interest rates in Brazil.
30Given the conservative policy monetary convention supported by the rentier-financial class coalition in the framework of a financialization process, we can ask what are the channels through which this class can affect interest rates in Brazil. The first and the second channels are just ways the financial market and the central bank relate to one another. The third, the fourth and the fifth channels are more structural. The third is related to the subordinated financial integration, a process that connects domestic financialization with international capital flows, while the fourth is institutional, associated with the existence of an indexed public debt that involves a two-way public debt contagion. Finally, the fifth channel is more than just a biased form of relating the rentier-financial coalition to the central bank: the high interest rates are a consequence of the equivocated belief that countries should attract foreign capitals to grow with current account deficits, the so called growth with “foreign savings” policy, which causes the appreciation of the national currency and encourages consumption, while discourage capital accumulation. Actually, this belief is anchored in domestic populism and the foreign interest in exporting capitals to Brazil.
- 17 Balassiano (2019) conducted an empirical experiment for the September/2016-January/2019 period, con (...)
31The first channel relates to the BCB Focus Report, which the Central Bank uses to survey the financial market’s forecasts for several economic indicators, including inflation and interest rates. At this juncture, the market has an upwards bias for its expected interest rate and inflation rate that puts pressure on the BCB to endorse their expectations. This view has been held by several authors. Oreiro and Passos (2005, p. 163) argue that “the Brazilian financial system can influence the Central Bank’s decisions setting the interest rate because, if the banks reach a mutual agreement, they can ‘force’ an interest-rate increase by ‘revising up’ their expectations surrounding inflation.” Figure 5 compares the expected and effective (12-month lagged) Selic to show that: (i) the expected rate is a good indicator of the effective rate’s direction, which may be regarded as a good predictor of the interest rate to be set by the BCB; (ii) however, generally speaking, the expected rate is higher than the effective one, which seems to suggest that the market tends to overshoot its interest estimates in the Focus Report in hopes that the BCB will endorse its expectations.17
Figure 5. Expected (Focus) and effective Selic rates (% p.a.)
Source: Barbosa (2017), from BCB, IPEADATA and BM&Bovespa data
- 18 According to Magalhães and Costa (2017, p. 11), “the contamination of public debt by monetary polic (...)
32The second channel concerns the connection between the financial market and the Treasury in the government securities trading process, where the market brings to bear its power to exert pressure on the Treasury, making it hostage to the financial markets for the purposes of issuing and rolling over public debt, particularly in times of financial stress and macroeconomic instability. As noted earlier, Figure 3 shows that accumulated real interest in 1992-2016 went hand-in-hand with public debt growth, suggesting that a significant portion of this growth is due to the effects of interest on the debt.18 With high public debt and an unwholesome structure (short term public securities and partial Selic indexation), the market is able to put pressure on the Treasury to sell bills under favorable conditions, including excessive returns. A negative side effect of the process is the presence of a flat and relatively short return curve in Brazil, with little distinction between short- and long-term rates.
- 19 In November 2018 the main holders of public debt securities were investment funds (26.3%), pension (...)
- 20 Clearly, the National Treasury avoids paying too high a risk premium on securities issues. It prefe (...)
- 21 Carvalho (2005) suggests that this kind of behavior may be the product of high macroeconomic uncert (...)
33Table 4 shows, starting in 2006, the prevalence of Selic- and IPCA-indexed public securities as well as pre-fixed ones. In periods of greater macroeconomic stability, such as 2004-2014, the share of fixed-income securities (LTN and NTN-F) increases, whereas times of greater stress see increased issues of LFTs (Selic-indexed), also known as “crisis papers”; in 2011-2015, when inflation accelerated, issues of IPCA-indexed securities (NTN-B) increased. This clearly shows that, under certain conditions, the holders of federal debt (investment funds, financial institutions, pension funds, etc.19) can put pressure on the National Treasury to issue debt under conditions that frequently favor them, as the wind of economic developments blows, providing a hedge against interest-rate or inflation risks.20 One potential implication of this kind of “game” is that the financial market can arbitrage its investments in securities with different indexers at its convenience, so that, to an extent, it can exert pressure on the Treasury to provide returns on securities issued under favorable conditions.21 This brings up our hypothesized presence of a “two-way” “contagion effect”, our third channel.
Table 4. Federal public debt broken down by indexer (% of total)
|
FX
|
TR
|
IGP
|
IPCA
|
Selic
|
Pre-fixed
|
Total
|
2002
|
22.4
|
2.1
|
11.0
|
1.6
|
60.8
|
2.2
|
100.0
|
2003
|
10.8
|
1.8
|
11.2
|
2.4
|
61.4
|
12.5
|
100.0
|
2004
|
5.2
|
2.7
|
11.8
|
3.1
|
57.1
|
20.1
|
100.0
|
2005
|
2.7
|
2.1
|
8.2
|
7.4
|
51.8
|
27.9
|
100.0
|
2006
|
1.3
|
2.2
|
7.2
|
15.3
|
37.8
|
36.1
|
100.0
|
2007
|
1.0
|
2.1
|
6.5
|
19.8
|
33.4
|
37.3
|
100.0
|
2008
|
1.1
|
1.6
|
5.7
|
23.6
|
35.8
|
32.2
|
100.0
|
2009
|
0.7
|
1.2
|
5.0
|
23.6
|
35.8
|
33.7
|
100.0
|
2010
|
0.6
|
0.8
|
4.8
|
23.3
|
32.5
|
37.9
|
100.0
|
2011
|
0.6
|
0.8
|
4.2
|
25.4
|
30.8
|
38.3
|
100.0
|
2012
|
0.6
|
0.6
|
4.1
|
31.4
|
22.2
|
41.2
|
100.0
|
2013
|
0.6
|
0.5
|
4.1
|
32.0
|
19.5
|
43.3
|
100.0
|
2014
|
0.6
|
0.5
|
4.0
|
32.7
|
19.2
|
43.1
|
100.0
|
2015
|
0.7
|
0.4
|
3.7
|
30.6
|
23.6
|
41.0
|
100.0
|
2016
|
0.5
|
0.4
|
3.7
|
29.4
|
29.1
|
36.9
|
100.0
|
2017
|
0.4
|
0.3
|
2.9
|
27.6
|
32.4
|
36.3
|
100.0
|
2018
|
0.5
|
0.2
|
2.9
|
26.4
|
36.4
|
33.5
|
100.0
|
Note: TR is a reference rate calculated as the adjusted weighted average monthly rate of fixed-rate certificate of term deposit of the country’s thirty largest financial institutions; IPCA is the official consumer price index calculated by IBGE; price general index (IGP) is calculated by a private foundation, FGV.
Source: Central Bank of Brazil (2018); (*) data as of December each year, except for 2018 (October)
- 22 According to Magalhães and Costa (2017, p. 12): “repo operations not only control the economy’s liq (...)
- 23 In this case, a BCB Selic rate increase affects the banking reserves market’s interest rate (the BC (...)
- 24 Modenesi and Modenesi (2008) shows empirical evidence that the Selic rate’s formation is driven by (...)
34Barbosa (2006) developed the idea of monetary policy contamination by public debt because of public debt indexed to the BCB-set interest rate, making public debt and banking reserves perfect substitutes and leading the interbank market rate to incorporate the Brazilian public debt’s risk premium. Figure 6 shows the evolution of LFT issues and repo operations, whose growth was quite matched in 2007-2018.22 Clearly, investors have a choice to arbitrage their highly liquid investments at the Selic rate, be it on the reserves market (repos), or on the securities market (LFTs). However, the contagion effect is a two-way street (due to the presence of a reverse contagion effect from the interbank to the securities market23). The contagion may come either from the securities to the reserves market – as analyzed above, because of the impact of the public sector’s feebleness. This may cause the interest rate that the market requires to roll over public debt to be “too high”; or from the banking reserves to the public securities market, due to the Central Bank’s conservative monetary-policy behavior. This behavior frequently causes an upwards pressure on the interest rate, which, through arbitrage, may end up raising the financial costs of public debt issues and rollovers.24
Figure 6. LFTs and BCB repos (R$ million) *
Source: Central Bank of Brazil (2018); (*) Amounts deflated at the IGP-DI of September 2018
35The fourth and the fifth channels deal with the exchange rate. The fourth channel I related to the volatility of the exchange rate. This relationship is well-documented in the literature for emerging economies that have liberalized their financial accounts: studies have shown that not only exchange rate volatility is higher in emerging economies compared to advanced ones, and central banks rely often on interest rate changes to stem exchange rate volatility in periods of instability, but also support policies appreciating the currency for price stabilization purposes (as tradeable goods become cheaper).
- 25 See Bresser-Pereira & Nakano (2003); Bresser-Pereira & Gala (2007); Bresser-Pereira, Araújo & Gala (...)
36Finally, the fifth channel is the use of high interest rate to attract foreign capitals. It derives from the misled belief that developing countries country should grow with “foreign savings,” that is, with current-account deficits financed by direct investment and foreign loans. As an increasing number of studies have shown, this is a self-defeating policy because the additional capital inflows required to finance the foreign deficit appreciate the currency for the duration of the deficit and, thus, turn the country’s manufacturing companies non-competitive, while increasing the acquisitive power of consumers and encouraging higher consumption.25 Bresser-Pereira (2020, p. 635) states that “the critique that new-developmental theory makes to the growth with foreign indebtedness policy is highly counterintuitive because it seems ‘logical’ or ‘natural’ that capital-rich countries transfer their capital to capital-poor countries.” This thesis is generally false. The additional capital inflows keep the exchange rate overvalued in the long-term, while the current-account deficit is in place. Nevertheless, the conventional wisdom that capital rich countries are supposed to transfer their capital to capital-poor countries is deeply ingrained in developing countries and the international agencies. Thus, what determines the current-account balance of a country that adopts the policy of growth with “foreign savings” (current account deficits) are not always endogenous variations in the exchange rate, which may have several causes, but is the policy decision to incur a current account deficit that most developing countries adopt. In Brazil’s case, this is the relevant cause. Therefore, given the usual adoption of the growth with foreign indebtedness policy in Brazil and most developing countries except the East Asian ones, the current account balance is an exogenous variable and the exchange rate, the endogenous one. How is such a policy enforced? It is not by increasing imports of capital goods, as policymakers believe, but by increasing the interest rate to attract foreign capitals. Since the appreciation of the currency increases the incomes of consumers, what actually increases are imports of consumer goods, while making the companies less competitive and discouraging investment. Policymakers don’t acknowledge this; their standard argument is that the control of inflation requires high interest rates, but the fact is that if the country incurs a foreign deficit it must finance it. As the direct investments are usually not sufficient, it is necessary to attract other foreign capital.
- 26 As we can see in Figure 5, the real interest rate increased sharply in 2015-2016.
37As we can see in Figure 7, after a strong exchange rate depreciation in 2003 and a surge of inflation, BCB increased the Selic interest rate until mid-2005; since then, under the context of a commodities boom and capital inflow surge, the real effective exchange rate appreciated until 2011, while the Selic rate gradually reduced. Following the gradual worsening in the international environment (capital flows, terms of trade, etc.) and the domestic inflationary pressures, we can note since 2013 both a trend of currency appreciation and an increase in the Selic rate.26
Figure 7. Policy rate (SELIC, % p.a.) and real effective exchange rate (June 1994 = 100)
Source: Prates et al. (2020, p. 54), with data from Central Bank of Brazil
38This paper goes back to the hypothesized presence of a pro-conservative monetary policy convention in Brazil, as initially formulated by Bresser-Pereira and Nakano (2002) and Erber (2011), integrating this hypothesis in the Brazilian regime of financialization “through interest income”: the formation of a rentier-financial class coalition invested in keeping interest rates high can only be understood as part of this sort of regime of financialization. We also add the hypothesis of a “reverse” public debt contagion effect, due both to the National Treasury’s difficulty managing public debt given market pressures in a context of macroeconomic instability and to the pro-conservative monetary policy convention, which creates bias for keeping the Selic interest rate high. More specifically, our main contribution is to show that the decades of high interest rates that are prevalent in Brazil in the context of a highly financialized economy have led to the formation of a rentier-financial coalition of interests favoring keeping interest rates high because this favors the appreciation of their financial wealth, resulting in the interest gains financialization regime. Related to the former, we also add a hypothesis that points out the use of high interest rates to finance current-account deficits under the context of a “growth cum foreign savings strategy.”
- 27 For a proposal regarding financial de-indexation in Brazil, see Paula and Marconi (2018).
39We suggest that a sustained reduction of real interest rates in Brazil requires a wide range of policies that must include the following five measures: the gradual elimination of financial indexation in Brazil27 by means of the replacement of BCB repo operations with voluntary interest-bearing deposits and an end to the LFTs; the implementation of a feasible long-term fiscal consolidation policy (free from the constraints of a fictional spending ceiling); the creation of mechanisms to reduce the volatility of the foreign exchange rate (given the exchange-interest connection); a review of the inflation targeting regime (changing the target’s horizon to a period longer than the calendar year); and, last but not least, the adoption of a less conservative management of monetary policy on the part of the BCB.
40These measures involve both an institutional reconfiguration of the Brazilian economy’s financial liberalization standard and the consolidation of a new monetary-financial regime. This new agenda involves not only reviewing the interest of capital holders in the comfort of financial gains through short-term assets (combining return, liquidity and low risk), distant from the riskier fixedness of directly productive activities. It also means recovering the state’s roles in the Brazilian development process, which have been obscured and politically voided by the nature and rationale of the rentier-financial accumulation that financialization replicates on the structural and macroeconomic levels. In conclusion, reducing interest rates in Brazil is not just an economic matter; it is also a matter of political economy.