BNDES, again – 04/07/2023 – Marcos Mendes

BNDES economists published, in this Sheeta reply to my column a fortnight ago about the risk of the BNDES repeating mistakes that will cost us dearly.

They argued that I considered “only the costs [partindo] based on the assumption that any and all actions by the bank had zero impact on the economy, [pois] (…) the literature on the impact of the BNDES on the economy also includes positive results on GDP, employment and tax collection”.

I didn’t just consider the costs. I cited several studies that evaluated the costs and benefits of the BNDES credit expansion policy based on loans made by the Treasury to the bank. It is worth repeating the conclusions, which unequivocally show low benefits and negative effects that go beyond fiscal costs (bibliographic references are in the online version of the March 24 column):

a) BNDES loans generated little additional investment in the economy: companies replaced more expensive loans, which they would already take, with cheaper resources offered by BNDES (the so-called “entrepreneur scholarship”);

b) after the 2009 recession, the ability to leverage investments fell sharply, deteriorating the cost-benefit ratio of the policy;

c) financing was concentrated in large, consolidated and not very innovative companies;

d) unsustainable projects, such as the third attempt to revive the shipbuilding industry or the failed chip company Unitec, destroyed capital and/or allocated it inefficiently;

e) there is no evidence that loans increased the productivity of the economy;

f) negative impact on the growth of the private capital market;

g) weakening of monetary policy, either by reducing the market’s sensitivity to the Selic rate, or by the parafiscal impulse given to demand.

It is clear that one can find some positive impact of BNDES operations on employment and short-term economic activity. After all, when more than 4% of GDP is poured into the economy in credit disbursements, as in 2010, there is some impact, unless the money is completely wasted. But that’s not enough to conclude that it was worth it. After all, there are costs.

The authors also dispute my figures regarding tax costs. But the numbers are not mine. These are official data, published by the National Treasury, which show that, in December 2022 values, the cost of Treasury operations with BNDES totaled R$ 325 billion.

The calculations are made by the difference between the rate that the Treasury pays when taking credit in the market and how much it charged in interest from the BNDES. This is the so-called “implicit subsidy”. There is also the explicit subsidy, which consists of direct payment from the Treasury to the BNDES to compensate it for subsidized credit granted by the Bank in certain lines of credit.

If BNDES economists do not agree with this methodology, which is widely accepted, they should open a discussion with Treasury technicians to change it. But they need to bring more convincing data than those presented in their reply.

In the text, they suggest that it would have been a good deal for the Treasury to obtain a nominal internal rate of return of 6.7% per year in its financial relationship with the BNDES in the period 2008-2022. Like this? It certainly would have been better for the taxpayer not to have accumulated debt close to 10% of GDP, at an average nominal cost of 11.1% over the period 2008-20222 (official Treasury data) to deliver cheap money to companies chosen by the government.

In the column, I made a comparison between Marshall Plan disbursements and amounts transferred by the Treasury to BNDES: if the large volumes of BNDES financing had a positive impact, we would see a strong permanent response, in the form of growth and income. What did not happen, contrary to the positive experience of the Marshall Plan.

The authors took this side issue as the central motto of the reply and began to make unclear calculations to say that the credit to the BNDES was smaller than the Marshall Plan. With that, they avoided facing the fundamental thing: the policy was expensive, ineffective and with negative side effects.

It is worrying that BNDES economists call “the ideological sophisms of the past” robust evidence of the failure of a public policy. Not acknowledging what went wrong and wanting to do it again doesn’t seem like a good way to go.

PRESENT LINK: Did you like this text? Subscriber can release five free hits of any link per day. Just click the blue F below.