Tax authorities weigh in, but don’t explain everything about high interest rates in Brazil

Many claim that Brazil has one of the highest real interest rates in the world because the government spends too much. There is truth in this, but the story is more complex. In a recent study, published on the FGV Ibre blog, I tried to measure how much of the problem is, in fact, due to public accounts.

I compared Brazil with seven emerging economies that also adopt inflation targets: Mexico, Chile, Colombia, Peru, South Africa, Indonesia and the Philippines. For the comparison to be fair, I discounted two things that naturally increase a country's interest rates: the American real interest rate, which serves as a global reference, and the risk premium charged by investors to lend to each government — a premium that should incorporate concerns about solvency in foreign currency and fiscal, among other factors.

Even making these adjustments, the Brazilian real interest rate was, on average, around three percentage points above that of the others from 2001 to 2025. It is this "excess" that needs explanation.

To refine the analysis, I sought to relate this excess interest rate with the so-called fiscal impulse: how much stimulus the government injects (or withdraws) into the economy each year, as this would not necessarily be captured by country risk.

The result of the analysis has two parts. The first confirms intuition: in Brazil, when the government promotes fiscal expansion, interest rates rise. Each GDP point of fiscal stimulus per year is associated with about 1.9 points higher interest rates. Interestingly, this relationship does not appear in other countries.

The second part is the surprise. Even with a neutral fiscal impulse, Brazilian interest rates would still be around 3.4 points above the reference. In none of the other countries does this number exceed 1.1 points. In other words: expansions and fiscal adjustments explain interest rate fluctuations well, but not their level.

South Africa helps decipher the enigma. It is the country with a fiscal framework most similar to ours: public debt of around 78% of GDP, which grew even faster than Brazil's in the last decade, and equally low savings. Even so, there the interest rate with neutral fiscal impulse would only be 0.2 points above the reference.

What sets us apart, then? Above all, the functioning of the public debt and credit markets. Half of Brazilian public bonds are post-fixed, that is, they yield the Selic itself; in South Africa, only 5%. Foreigners hold around 10% of our debt in reais, compared to more than a quarter there.

More than 40% of credit here is directed, with regulated rates that barely react to the Selic, as in rural and real estate. And the banking spread, the difference between what banks pay to borrow and what they charge to lend, is many times greater.

In practice, this makes Brazilian interest rates behave like those of a closed economy, defined almost solely by local savers. And it weakens monetary policy: if a large share of credit does not meet the Selic rate and people become "richer" instantly when interest rates rise (due to floating rates), the Central Bank needs to raise interest rates further to obtain the same effect on inflation.

The fiscal consolidation needed today, of 2 to 3 points of GDP in the coming years, could reduce real interest rates from almost 8% in 2023-25 ​​to close to 5.5% to 6% annually in 2027-30. But, alone, it would not bring us any lasting closer to Mexico, Chile or South Africa.

This diagnosis, in fact, unites economists from different schools of thought. Heterodox critics criticize debt indexation to Selic and banking concentration; orthodox, targeted credit, a kind of "half price", and legal uncertainty, default and taxes that make loans more expensive. Both point to the same variables. The question is not whether the inspector matters, but what else matters besides him.