Opinion – Bernardo Guimarães: The mistakes we should not repeat

My column last week explained the view of the new Secretary for Economic Policy on the economic crisis of the last PT government based on one of his academic articles.

Some readers have complained that I have not provided arguments justifying my critical view of his position. In fact, a discussion on the causes of the crisis did not fit in that column.

Understanding what went wrong is important so that we do not repeat the mistakes of this new government. A good explanation would take up a lot more space than I have, but I’ll try to summarize.

The article by the new Secretary for Economic Policy argues that the lack of demand for intermediate industrial goods — those used in the production of others — was crucial to the economic crisis.

However, a set of policies at the time aimed precisely at stimulating demand for this type of good.

The National Content Law (Law 12,351 of December 2010) required oil and gas production to use more intermediate industrial goods made in Brazil. From there, Sete Brasil was born, a company that symbolizes the failure of this policy. Creating this demand for intermediate goods effectively allocated a large amount of resources to a sector with low productivity.

These policies went well beyond the oil and gas sector. Local content counterparts were used to stimulate demand for domestic industrial goods.

BNDES, with its highly subsidized credit, aimed to stimulate industrial production by allocating credit to this sector at interest rates far below market rates.

Policies of this type do not create resources: they allocate capital and labor to specific activities or sectors.

To understand this, suppose that the National Treasury reduces its transfers to the BNDES, and the latter consequently reduces its loan operations. On the one hand, fewer companies will receive BNDES credit to invest.

On the other hand, however, the Treasury will need to raise less debt. This will allow it to borrow resources at a lower interest rate, and part of the money that was loaned to the Treasury will now have to seek other destinations. Through the banking sector or the capital market, these resources will reach other companies or people.

Thus, BNDES subsidized loans benefit those who have access to this type of credit, but lead to higher interest rates and less credit for others. As in an exclusive party for guests, anyone who isn’t a fish doesn’t swim and still pays the bill.

Therefore, to justify the BNDES subsidies, it would be necessary to show that the benefited companies invested more and that this investment generated greater productivity gains than those that would have been generated if the BNDES had not interfered in the allocation of resources.

The available evidence goes against this idea. Careful empirical work, like this one and this one, does not find these effects.

Furthermore, an article published in one of the world’s leading financial academic journals finds the effects of the political use of the BNDES on the allocation of capital among Brazilian municipalities. This work uses data from 1995 onwards.

In short, everything indicates that policies aimed at creating demand and allocating resources to specific sectors had negative effects on productivity and growth.

In the same way that we want Covid treatments based on tests and experiments, we must also ask for evidence that supports the application of public policies that affect the allocation of resources in the economy.