Well, that was long. Reminded me of String Theory and multiple realities.
When Debt:GDP is 80% it is a spending problem, not an interest rate management problem.
If the market believes you are reducing the debt load and being responsible in your spending (i.e., not squandering on social welfare largess) it will accept lower interest rates through market, not government fiat, mechanisms. If the market does not trust you, it will stop lending. The market lends money with the expectation of its repayments on time and at the expected value less an assumed debased currency rate.
This is not hard. It's not even really economics (driving forward using the rear view mirror), it's finance (trying to figure out where the road in front of you is going). Attempting to bury reality in mountains economic delusions so you can attempt to bribe voters with largess is best left to Marx. And he is long dead.
I think its well accepted that the Selic policy rate carries a significant fiscal/inflation risk premium as well as a premium for capital controls. The real issue, i would argue, is one of sequencing: deliver a high primary surplus first (like Lula did on 2003-2004) and then things fall in place and you can then run a smaller primary surplus. Be aware that Resende also flirts with MMT ideas and is not very clear about the need for the govt to deliver first on itgs fiscal commitment to then carefully lower interest rates with, as you noted, an apreciating currency (Resende's argument about using BCB FX reserves is an indicationn that he wants to put the interst rate cart before the fiscal horse).
In Brazil, the central bank's mandate is to meet the inflation target. Since 2023, inflation has been above the midpoint of the target range in 45 months and above the upper bound in 21 months. It is currently above the upper bound.
The government could raise the inflation target, allowing the central bank to lower interest rates. If raising the target or cutting rates does not set off an inflationary spiral, it could be a win-win situation. Throughout this period, however, the current government has done nothing along these lines and has focused solely on complaining about high interest rates and calling for cuts — much like Trump does.
As for fiscal policy, it is unlikely that the government will pursue structural spending cuts. Roughly 90% of the federal budget consists of mandatory expenditures, while the remainder barely covers the cost of keeping the government running. The list of handouts and free benefits keeps growing. Brazil is gradually losing its demographic dividend, the social security deficit is widening, much of the economy is indexed, and the government is likely to continue granting real increases in the minimum wage, which in turn drive up other expenditures linked to it, such as pension and social assistance benefits.
There are also other factors adding fuel to inflation: increases in existing social benefits, the creation of new ones, and various credit programs, whether through the government acting as a guarantor or providing subsidized credit.
So yeah, it'd be lower interest rates WITH fiscal loosening.
Professor, with all due respect, let’s not forget Knut Wicksell (S=I).
If the natural rate of interest is ultimately determined by the real fundamentals of the economy - productivity, savings, time preferences, and the expected return on capital - what evidence demonstrates that Brazil’s natural rate stands at around 2.5% in real terms, rather than at a higher level?
Absent a convincing answer to this question, administratively setting an interest rate range of 3% to 5.5% in nominal terms risks conflating a desirable target for government financing with a sustainable equilibrium interest rate for the economy as a whole.
I think the best works about the natural rate were the two "Juros no Brasil, uma jabuticaba" op-eds published at the Valor newspaper this year, by Carlos Eduardo Goncalves and Thiago Cavalcanti.
The uproar is being caused by government officials and politicians using Resende's text and interpretation of the FTPL as a crutch on which to put all the blame on monetary policy (as always, the Executive branch treats the Central Bank as an enemy who is trying to sabotage the economy with high rates), and as such be free to spend as they please.
That is exactly why Resende's proposal can't possibly work; the FISCAL part of the Fiscal Theory is being thown in the garbage as is convenient.
Interesting read, and about what I expected if you commented on the situation. However I feel like two macroeconomic indicators are being left out of your analysis and they're (at least on the surface) scratched by Andre Lara.
1. Unemployment is at a historic low. It's hovering at 5 to 5.5% when historically we're used to it hovering around 12%. Since 2022 unemployment essentially halved. The considerations for Central Banking did not.
2. Trump 2.0 is making the world more expensive and that affects our inflation index (IPCA). There's no denying that the shenanigans in the Middle East are affecting fuel prices and thus prices of goods on top of the absurd arbitrary tariffs. Despite this, our inflation is hovering around 4.5% and the target (absurdly low for a developing country) is 3%. One could wonder what would it be if not for Trump's impacts. In other words: if fuel prices were lower, and tariffs weren't being imposed to American consumers fighting for imported goods, would inflation still be growing at this rate or would it be lower and probably within target?
Which brings me to my third point:
3. Maybe the inflation calculation should be changed to consider the recent worldwide context. Since it's not, and the proposal for a new formula (lowering the weight of fuel and maybe even energy prices because people WILL be buying those regardless of price as it's necessary for them) is not really being considered (there was a little bit of talk about this a year or two ago), then maybe Lara's approach, considering your arguments, is a decent solution.
My 3 points obviously don't touch the main arguments Lara presented, of course, but add a bit of context and maybe egg you into a recommendation/approval.
May I beg to disagree, unemployment cannot be deemed the lowest ever without consideration to two factors. First and most importantly, with 22% of the country living essentially on government Bolsa Familia", a Brazilian equivalent of minimum revenue paid by government, you cannot really believe unemployment is low, unless you make the case the permanent living paid by government is like holding a job... Second, you have to look at the quality of the jobs, and the current model considers that anyone having worked 1 hour this week is employed... Unemployment and less than satisfactory jobs are a real problem of the Brazilian economy by any measure, other than the official ones, of course...
Rational programs require rational policy-makers. When the policy-makers got there by stealing the election, rationality is the least of their priorities
Do you think raising Brazil's inflation target would help lower interest rates? It seems that the 3% target is historically low for the Brazilian economy and, as a result, leads to persistently de-anchored inflation expectations, forcing the Central Bank to keep interest rates higher.
Thank you for the lesson, Professor Cochrane. I would like to draw your attention to an episode in Brazilian fiscal policy that took place in 2016.
In December of that year, Brazil enacted a constitutional amendment establishing a federal spending cap for 20 years. Under this rule, federal primary spending could grow only in line with inflation, effectively freezing expenditure in real terms.
Over the following year, Brazil's central bank cut its policy interest rate by six percentage points, from 13% in January 2017 to 7% in December. Inflation fell as well
It seems to me that this episode provides compelling evidence in support of your theory.
I think you are not telling the whole history. What happened with GDP, unemployment rate? More than that, with the raise of unemployment, who and how will pay the contribution of INSS? With less people to pay, the government has to pay more or change the retirement rules. More than that, the promise at that time was these reforms would bring a lot of investments to Brazil. Not happened. That period was terrible.
Great text. For some reason I now can't remember, I read Resende's text the other day, where he copied your FTPL in. It seemed to me like reaching for a tool to remove BCB independence. Would not this idea in itself decrease trust? Brazil is a country with one of the lowest levels of interpersonal trust in the world.
If Selic is lowered below the equilibrium rate, who would like to buy government bonds, at par, other than the BCB, if buyers are not forced to?
You mention government debt/GDP as 80%. This is the BCB number, where their large portfolio of government bonds is subtracted from gross debt. The IMF has it at 97% and based on the method used to compare the indebtedness of nations around the world. If central bank portfolios could have been subtracted, as in the BCB number, countries like Japan and Sweden would have had very low public debt at the peak of QE (the Riksbank held up to 50% of Swedish government bonds), but it was part of the QE "promise" that these portfolios would be liquidiated, eventually.
John, the 141 bps CDS you refer to is on dollar-denominated debt. There is no traded CDS on real-denominated debt. There is, however, one way to read the market implied probability of default on real-denominated debt. Brazil has 5-year floating rate notes indexed to the reference SELIC rate (LFT bonds). These trade at a small discount. If all that discount is credit risk, then the risk-neutral probability of default is currently at around 10 bps per year.
Are you sure about that? LFT bonds do not pay out any interest until at the end, right? So all that interest accrued from variable SELIC fixings, has to be discounted from 5 years back to today. And the relevant interest rate to do that is with the 5-year zc bond rate, right? This rate is currently higher than the short-term interest rate. Isn't this the main reason for the discount, in the case of LFT bonds? Apart maybe from transaction costs, etc.
Good analysis. IMHO, the key questions concerning the brazilian monetary and fiscal policies are:
1- Why does Brazil need to have the highest real interest rate in the world, around 9% per year, higher even than those of Russia and Turkey, and 6 p.p. above those of other developing countries, like Mexico, South Africa, Indonesia? These have real interest rates lower than 3% and some of them have left-wing governments, with similar policies, and even greater challenges concerning political stability, infrastructure, crime rates.
2- Is the current brazilian inflation target, of only 3% a year, compatible with its economy and per capita income, with its commercial and administrative law (heavily indexed prices in contracts, both private and public, mainly in utilities: transportation, electricity, water, natural gas...), and with a world of constant supply shocks due to climate issues, wars and other externalities, like the pandemics or disruptive technologies (e.g. AI and its impact on semiconductor prices)?
3- How to improve economic expectations in a country with highly concentrated mass media, controlled by a handful of ultra rich families with tight connections with the financial market, that openly withstands any redistributive policy by moderate left-wing governments (like Lula's) aiming to reduce Brazil's obscene social inequality, repeating ad nauseam the narrative of waste or spree of public resources, or of a supposed fiscal crater, even with a 0.5% GDP primary deficit?
Brazilian interest costs have reached 9% of GDP, for a nominal rate of 13.75%. Considering the current nominal growth of brazilian economy around 7% (inflation + real growth), and the primary deficit of 0.5% of GDP, in order to keep a constant Debt/GDP ratio the interest cost should not exceed 6.5%.
Given the current size of brazilian debt and its high level of short term floating-rate ("post-fixed") bonds, that means a maximum nominal interest rate of about 8.5% a year. That is, a real interest rate above 3.5%, what would maintain Brazil in the Top-3 countries with highest real rates, only behind Russia and Turkey, and more than 0.5 p.p. above the 4th position, Mexico; but the trajectory of the public debt would become sustainable, if the government keeps the primary deficit below 0.5%, obeying the so called "Arcabouço Fiscal" (the current fiscal law in Brazil).
As Andre Lara correctly accuses, the Brazilian Central Bank, in not reducing the real interest rate (currently much higher than its historical levels, even considering previous Lula and Dilma mandates), is directly responsible for the current rise of the public debt, bowing to the speculative, politically biased (and false) narrative of rent-seekers in the media that brazilian government spending is out of control and the main culprit for the negative market forecasts.
From a Brazilian: Thank you for your analysis! Our country needs to hear level-headed people like you right now! The livelihood of so many workers and their families depends on it
The latter part is crucial. BRL must appreciate if confidence is restored not the other way around. $350bn of reserves is really not much considering the monetary aggregates. Stats like import coverage are silly. Eventually some form of capital control may be needed…
I wouldn’t call André Lara Resende’s proposal particularly sophisticated. Quite the contrary. It seems to me that he’s applying the Fiscal Theory of the Price Level (FTPL) under assumptions that simply aren’t fulfilled in Brazil.
The current government has demonstrated, over the past four years, very limited ability or willingness to implement the fiscal adjustment necessary to stabilize public debt. There is little reason for markets or economic agents to assume that this will suddenly change.
Consider the debt dynamics. Even assuming a primary deficit of just 1% of GDP, the combination of high real interest rates and relatively modest economic growth puts substantial upward pressure on the debt-to-GDP ratio. Without a credible fiscal adjustment, the debt trajectory becomes increasingly difficult to stabilize.
Under these circumstances, there is no fiscal credibility to support the proposition that unilateral reductions in interest rates will generate sustainable disinflation. Lower rates might reduce debt-service costs, but without a credible fiscal framework, they could also undermine confidence, increase inflation expectations and put pressure on the exchange rate.
There is another structural problem that deserves attention: Brazil’s excessive dependence on LFTs, government bonds indexed to the Selic rate.
Roughly half of the federal debt stock is exposed to floating interest rates. This means that whenever the Central Bank raises the Selic to combat inflation, government debt-service costs increase directly and relatively quickly. Monetary tightening consequently puts additional pressure on the fiscal accounts.
This creates an undesirable feedback mechanism: higher interest rates increase government borrowing costs, which worsen fiscal dynamics and can further undermine confidence in the currency and public debt.
One structural solution would be to progressively reduce the issuance of LFTs and increase the share of fixed-rate, prefixado securities.
The Treasury should issue fixed-rate bonds at market-determined yields, across different maturities, reflecting prevailing macroeconomic conditions, inflation expectations and fiscal risk premiums. Investors would know the nominal return they are contracting for, while the government would lock in its financing cost for the life of the bond.
This is broadly consistent with the structure that predominates in the US Treasury market, where fixed-rate securities are the main financing instrument, alongside inflation-protected and floating-rate bonds.
Of course, this transition cannot happen by decree. If investors don’t trust the government’s fiscal trajectory, they will demand prohibitively high yields to hold longer-term fixed-rate debt. The transition must therefore be gradual and accompanied by a credible fiscal framework.
Ultimately, Brazil needs to address three interconnected problems:
1. Persistent primary deficits: The government needs to credibly commit to eliminating the primary deficit over the next one to two years, followed by sustained primary surpluses in levels actually required to stabilize the debt trajectory.
2. Lack of fiscal credibility: Markets need to believe that the government will honor its debt obligations and that public debt will eventually stabilize. This requires a credible medium-term fiscal framework, not just announcements or accounting maneuvers.
3. Excessive exposure to the Selic rate: The Treasury should gradually reduce its dependence on floating-rate LFTs and expand fixed-rate financing, building a more predictable debt structure and reducing the immediate fiscal impact of monetary tightening.
I wouldn’t call Lara Resende’s wee proposal a sophisticated solution. I’d call it a Hail Mary attempt to solve a structural fiscal problem through a theoretical reverse causation mechanism that though possible, unlikely to be fulfilled.
The problem isn’t that Brazil doesn’t know how to lower interest rates. The problem is that it hasn’t established the fiscal credibility necessary to do so sustainably, without compromising price stability.
Well, that was long. Reminded me of String Theory and multiple realities.
When Debt:GDP is 80% it is a spending problem, not an interest rate management problem.
If the market believes you are reducing the debt load and being responsible in your spending (i.e., not squandering on social welfare largess) it will accept lower interest rates through market, not government fiat, mechanisms. If the market does not trust you, it will stop lending. The market lends money with the expectation of its repayments on time and at the expected value less an assumed debased currency rate.
This is not hard. It's not even really economics (driving forward using the rear view mirror), it's finance (trying to figure out where the road in front of you is going). Attempting to bury reality in mountains economic delusions so you can attempt to bribe voters with largess is best left to Marx. And he is long dead.
And you thought you were Grumpy!
I think its well accepted that the Selic policy rate carries a significant fiscal/inflation risk premium as well as a premium for capital controls. The real issue, i would argue, is one of sequencing: deliver a high primary surplus first (like Lula did on 2003-2004) and then things fall in place and you can then run a smaller primary surplus. Be aware that Resende also flirts with MMT ideas and is not very clear about the need for the govt to deliver first on itgs fiscal commitment to then carefully lower interest rates with, as you noted, an apreciating currency (Resende's argument about using BCB FX reserves is an indicationn that he wants to put the interst rate cart before the fiscal horse).
In Brazil, the central bank's mandate is to meet the inflation target. Since 2023, inflation has been above the midpoint of the target range in 45 months and above the upper bound in 21 months. It is currently above the upper bound.
The government could raise the inflation target, allowing the central bank to lower interest rates. If raising the target or cutting rates does not set off an inflationary spiral, it could be a win-win situation. Throughout this period, however, the current government has done nothing along these lines and has focused solely on complaining about high interest rates and calling for cuts — much like Trump does.
As for fiscal policy, it is unlikely that the government will pursue structural spending cuts. Roughly 90% of the federal budget consists of mandatory expenditures, while the remainder barely covers the cost of keeping the government running. The list of handouts and free benefits keeps growing. Brazil is gradually losing its demographic dividend, the social security deficit is widening, much of the economy is indexed, and the government is likely to continue granting real increases in the minimum wage, which in turn drive up other expenditures linked to it, such as pension and social assistance benefits.
There are also other factors adding fuel to inflation: increases in existing social benefits, the creation of new ones, and various credit programs, whether through the government acting as a guarantor or providing subsidized credit.
So yeah, it'd be lower interest rates WITH fiscal loosening.
Professor, with all due respect, let’s not forget Knut Wicksell (S=I).
If the natural rate of interest is ultimately determined by the real fundamentals of the economy - productivity, savings, time preferences, and the expected return on capital - what evidence demonstrates that Brazil’s natural rate stands at around 2.5% in real terms, rather than at a higher level?
Absent a convincing answer to this question, administratively setting an interest rate range of 3% to 5.5% in nominal terms risks conflating a desirable target for government financing with a sustainable equilibrium interest rate for the economy as a whole.
I think the best works about the natural rate were the two "Juros no Brasil, uma jabuticaba" op-eds published at the Valor newspaper this year, by Carlos Eduardo Goncalves and Thiago Cavalcanti.
The uproar is being caused by government officials and politicians using Resende's text and interpretation of the FTPL as a crutch on which to put all the blame on monetary policy (as always, the Executive branch treats the Central Bank as an enemy who is trying to sabotage the economy with high rates), and as such be free to spend as they please.
That is exactly why Resende's proposal can't possibly work; the FISCAL part of the Fiscal Theory is being thown in the garbage as is convenient.
Interesting read, and about what I expected if you commented on the situation. However I feel like two macroeconomic indicators are being left out of your analysis and they're (at least on the surface) scratched by Andre Lara.
1. Unemployment is at a historic low. It's hovering at 5 to 5.5% when historically we're used to it hovering around 12%. Since 2022 unemployment essentially halved. The considerations for Central Banking did not.
2. Trump 2.0 is making the world more expensive and that affects our inflation index (IPCA). There's no denying that the shenanigans in the Middle East are affecting fuel prices and thus prices of goods on top of the absurd arbitrary tariffs. Despite this, our inflation is hovering around 4.5% and the target (absurdly low for a developing country) is 3%. One could wonder what would it be if not for Trump's impacts. In other words: if fuel prices were lower, and tariffs weren't being imposed to American consumers fighting for imported goods, would inflation still be growing at this rate or would it be lower and probably within target?
Which brings me to my third point:
3. Maybe the inflation calculation should be changed to consider the recent worldwide context. Since it's not, and the proposal for a new formula (lowering the weight of fuel and maybe even energy prices because people WILL be buying those regardless of price as it's necessary for them) is not really being considered (there was a little bit of talk about this a year or two ago), then maybe Lara's approach, considering your arguments, is a decent solution.
My 3 points obviously don't touch the main arguments Lara presented, of course, but add a bit of context and maybe egg you into a recommendation/approval.
May I beg to disagree, unemployment cannot be deemed the lowest ever without consideration to two factors. First and most importantly, with 22% of the country living essentially on government Bolsa Familia", a Brazilian equivalent of minimum revenue paid by government, you cannot really believe unemployment is low, unless you make the case the permanent living paid by government is like holding a job... Second, you have to look at the quality of the jobs, and the current model considers that anyone having worked 1 hour this week is employed... Unemployment and less than satisfactory jobs are a real problem of the Brazilian economy by any measure, other than the official ones, of course...
Rational programs require rational policy-makers. When the policy-makers got there by stealing the election, rationality is the least of their priorities
Do you think raising Brazil's inflation target would help lower interest rates? It seems that the 3% target is historically low for the Brazilian economy and, as a result, leads to persistently de-anchored inflation expectations, forcing the Central Bank to keep interest rates higher.
Thank you for the lesson, Professor Cochrane. I would like to draw your attention to an episode in Brazilian fiscal policy that took place in 2016.
In December of that year, Brazil enacted a constitutional amendment establishing a federal spending cap for 20 years. Under this rule, federal primary spending could grow only in line with inflation, effectively freezing expenditure in real terms.
Over the following year, Brazil's central bank cut its policy interest rate by six percentage points, from 13% in January 2017 to 7% in December. Inflation fell as well
It seems to me that this episode provides compelling evidence in support of your theory.
I think you are not telling the whole history. What happened with GDP, unemployment rate? More than that, with the raise of unemployment, who and how will pay the contribution of INSS? With less people to pay, the government has to pay more or change the retirement rules. More than that, the promise at that time was these reforms would bring a lot of investments to Brazil. Not happened. That period was terrible.
Great text. For some reason I now can't remember, I read Resende's text the other day, where he copied your FTPL in. It seemed to me like reaching for a tool to remove BCB independence. Would not this idea in itself decrease trust? Brazil is a country with one of the lowest levels of interpersonal trust in the world.
If Selic is lowered below the equilibrium rate, who would like to buy government bonds, at par, other than the BCB, if buyers are not forced to?
You mention government debt/GDP as 80%. This is the BCB number, where their large portfolio of government bonds is subtracted from gross debt. The IMF has it at 97% and based on the method used to compare the indebtedness of nations around the world. If central bank portfolios could have been subtracted, as in the BCB number, countries like Japan and Sweden would have had very low public debt at the peak of QE (the Riksbank held up to 50% of Swedish government bonds), but it was part of the QE "promise" that these portfolios would be liquidiated, eventually.
John, the 141 bps CDS you refer to is on dollar-denominated debt. There is no traded CDS on real-denominated debt. There is, however, one way to read the market implied probability of default on real-denominated debt. Brazil has 5-year floating rate notes indexed to the reference SELIC rate (LFT bonds). These trade at a small discount. If all that discount is credit risk, then the risk-neutral probability of default is currently at around 10 bps per year.
Are you sure about that? LFT bonds do not pay out any interest until at the end, right? So all that interest accrued from variable SELIC fixings, has to be discounted from 5 years back to today. And the relevant interest rate to do that is with the 5-year zc bond rate, right? This rate is currently higher than the short-term interest rate. Isn't this the main reason for the discount, in the case of LFT bonds? Apart maybe from transaction costs, etc.
Yes, I am quite sure about that.
Good analysis. IMHO, the key questions concerning the brazilian monetary and fiscal policies are:
1- Why does Brazil need to have the highest real interest rate in the world, around 9% per year, higher even than those of Russia and Turkey, and 6 p.p. above those of other developing countries, like Mexico, South Africa, Indonesia? These have real interest rates lower than 3% and some of them have left-wing governments, with similar policies, and even greater challenges concerning political stability, infrastructure, crime rates.
2- Is the current brazilian inflation target, of only 3% a year, compatible with its economy and per capita income, with its commercial and administrative law (heavily indexed prices in contracts, both private and public, mainly in utilities: transportation, electricity, water, natural gas...), and with a world of constant supply shocks due to climate issues, wars and other externalities, like the pandemics or disruptive technologies (e.g. AI and its impact on semiconductor prices)?
3- How to improve economic expectations in a country with highly concentrated mass media, controlled by a handful of ultra rich families with tight connections with the financial market, that openly withstands any redistributive policy by moderate left-wing governments (like Lula's) aiming to reduce Brazil's obscene social inequality, repeating ad nauseam the narrative of waste or spree of public resources, or of a supposed fiscal crater, even with a 0.5% GDP primary deficit?
Brazilian interest costs have reached 9% of GDP, for a nominal rate of 13.75%. Considering the current nominal growth of brazilian economy around 7% (inflation + real growth), and the primary deficit of 0.5% of GDP, in order to keep a constant Debt/GDP ratio the interest cost should not exceed 6.5%.
Given the current size of brazilian debt and its high level of short term floating-rate ("post-fixed") bonds, that means a maximum nominal interest rate of about 8.5% a year. That is, a real interest rate above 3.5%, what would maintain Brazil in the Top-3 countries with highest real rates, only behind Russia and Turkey, and more than 0.5 p.p. above the 4th position, Mexico; but the trajectory of the public debt would become sustainable, if the government keeps the primary deficit below 0.5%, obeying the so called "Arcabouço Fiscal" (the current fiscal law in Brazil).
As Andre Lara correctly accuses, the Brazilian Central Bank, in not reducing the real interest rate (currently much higher than its historical levels, even considering previous Lula and Dilma mandates), is directly responsible for the current rise of the public debt, bowing to the speculative, politically biased (and false) narrative of rent-seekers in the media that brazilian government spending is out of control and the main culprit for the negative market forecasts.
In Brazil the big media only talk about budget adjustment. Thia Lara's Idea show the real size of the matter
From a Brazilian: Thank you for your analysis! Our country needs to hear level-headed people like you right now! The livelihood of so many workers and their families depends on it
The latter part is crucial. BRL must appreciate if confidence is restored not the other way around. $350bn of reserves is really not much considering the monetary aggregates. Stats like import coverage are silly. Eventually some form of capital control may be needed…
I wouldn’t call André Lara Resende’s proposal particularly sophisticated. Quite the contrary. It seems to me that he’s applying the Fiscal Theory of the Price Level (FTPL) under assumptions that simply aren’t fulfilled in Brazil.
The current government has demonstrated, over the past four years, very limited ability or willingness to implement the fiscal adjustment necessary to stabilize public debt. There is little reason for markets or economic agents to assume that this will suddenly change.
Consider the debt dynamics. Even assuming a primary deficit of just 1% of GDP, the combination of high real interest rates and relatively modest economic growth puts substantial upward pressure on the debt-to-GDP ratio. Without a credible fiscal adjustment, the debt trajectory becomes increasingly difficult to stabilize.
Under these circumstances, there is no fiscal credibility to support the proposition that unilateral reductions in interest rates will generate sustainable disinflation. Lower rates might reduce debt-service costs, but without a credible fiscal framework, they could also undermine confidence, increase inflation expectations and put pressure on the exchange rate.
There is another structural problem that deserves attention: Brazil’s excessive dependence on LFTs, government bonds indexed to the Selic rate.
Roughly half of the federal debt stock is exposed to floating interest rates. This means that whenever the Central Bank raises the Selic to combat inflation, government debt-service costs increase directly and relatively quickly. Monetary tightening consequently puts additional pressure on the fiscal accounts.
This creates an undesirable feedback mechanism: higher interest rates increase government borrowing costs, which worsen fiscal dynamics and can further undermine confidence in the currency and public debt.
One structural solution would be to progressively reduce the issuance of LFTs and increase the share of fixed-rate, prefixado securities.
The Treasury should issue fixed-rate bonds at market-determined yields, across different maturities, reflecting prevailing macroeconomic conditions, inflation expectations and fiscal risk premiums. Investors would know the nominal return they are contracting for, while the government would lock in its financing cost for the life of the bond.
This is broadly consistent with the structure that predominates in the US Treasury market, where fixed-rate securities are the main financing instrument, alongside inflation-protected and floating-rate bonds.
Of course, this transition cannot happen by decree. If investors don’t trust the government’s fiscal trajectory, they will demand prohibitively high yields to hold longer-term fixed-rate debt. The transition must therefore be gradual and accompanied by a credible fiscal framework.
Ultimately, Brazil needs to address three interconnected problems:
1. Persistent primary deficits: The government needs to credibly commit to eliminating the primary deficit over the next one to two years, followed by sustained primary surpluses in levels actually required to stabilize the debt trajectory.
2. Lack of fiscal credibility: Markets need to believe that the government will honor its debt obligations and that public debt will eventually stabilize. This requires a credible medium-term fiscal framework, not just announcements or accounting maneuvers.
3. Excessive exposure to the Selic rate: The Treasury should gradually reduce its dependence on floating-rate LFTs and expand fixed-rate financing, building a more predictable debt structure and reducing the immediate fiscal impact of monetary tightening.
I wouldn’t call Lara Resende’s wee proposal a sophisticated solution. I’d call it a Hail Mary attempt to solve a structural fiscal problem through a theoretical reverse causation mechanism that though possible, unlikely to be fulfilled.
The problem isn’t that Brazil doesn’t know how to lower interest rates. The problem is that it hasn’t established the fiscal credibility necessary to do so sustainably, without compromising price stability.