Back of the envelope calculation: BNDES lending and the Marshall Plan

So, a few days ago, someone (my bad, can't remember who did it) posted on FB a piece (in Portuguese and behind a wall; but this post is mostly about the role of historical comparisons really, so you can skip the piece altogether) on the Brazilian National Development Bank (Portuguese acronym is BNDES, btw) and how it lend more than the US government with the Marshall Plan. The guy did a back of the envelope calculation (I did check and bringing the US$ 13 billion to present value, with the GDP deflator would be about 106 billion, roughly what he calculated) and concluded that the BNDES lending, which was higher, was very ineffective. Hm. Where to start?

Sure one can assume that the important thing is just to calculate how much money was lent in current values and one gets a reasonable picture of the impact. However, it should be clear that the US was lending dollars, and access to imports that were vital for the survival of Europe. Harder to put a dollar figure on that. But a better way to go around would be to calculate the magnitude of that loan for the US economy back then. What was the size of the commitment is a better question.

It turns out that the Marshall Plan (the 13 billion) corresponded to about 4.7% of the US GDP in 1948. That would translate in terms of the GDP of more than 18.5 trillion in 2016 to slightly less than 900 billion dollars now (or a bit larger than the 2009 fiscal package). At the exchange rate of 3.15, that the author uses, the figure would be more than 2.7 trillion reais, which would dwarf the 400 billion reais lent by BNDES (assuming that his number is the correct one).

BNDES lending wasn't 25% larger than the Marshall Plan, it was about 15% of the size of the Marshall Plan. A tiny fraction, when compared to the size of the US commitment. And that seems more accurate. Not only the US was lending in dollars, the international reserve currency, but also it was a massive aid program (which was essentially promotion of American firms, since a lot of the European demand reverted back to the US). This cannot be compared in any sort of reasonable way with the lending in reais by the Brazilian bank.

Don't get me wrong, that's just a back of the envelope calculation like the one he did (I think more accurate, or giving a more reasonable picture, but that's another question). Note that the Marshall Plan was aid (not lending, if you didn't get this before), and again that it was a mechanism to provide dollars, necessary for importing American goods, essential for the survival of Europe (and to avoid, you know, the specter of communism, just to make sure you got that). And done in a context in which the US basically unilaterally opened its economy to Western Europe and Japan to allow for the recovery (both seen as "Miracles").

BNDES role lending to foreign governments that hired Brazilian firms is more akin to the Ex-Im Bank and that perhaps would be a better analogy. Yes some of the lending included firms that paid bribes and that is a problem (don't get me wrong corruption should be investigated and punished, but I wouldn't imagine the State Department would just seat while the judiciary system destroyed US corporations; there is a clear understanding in the US that security goes hand in hand with economic hegemony, something that is lost to Brazilian elites). The lesson there is that American elites are not trying to destroy their own corporations (even Trump that seemed to be against the Ex-Im Bank now apparently favors it, admittedly in a weird way).

Historical analogies can be illuminating, no doubt. But for that the context matters, and an understating of institutions and their evolution is often important. For example, to understand that the US had the capacity of lifting the European external constraint, and eliminate dollar scarcity, and that this was a central component of both the economic recovery, and the geopolitical strategy of containing Soviet expansion in Western Europe would be relevant. Note that BNDES was part of a considerably more modest strategy of the left of center government of Brazil, that of promoting autonomous development by supporting and building up competitive national corporations (the so-called national champions) and to provide some regional support for Brazilian corporations, consolidating an alternative to US backed corporate globalization in the region (where many left of center governments had flourished in that period). There were certainly flaws in the BNDES experience, and a comparison with, for example, the South Korean experience of promoting national champions might be fruitful (and has been done, to some extent).

Some of the other allegations, like the one that BNDES subsidized credit caused the higher rates of interest of the central bank, belong to the Latin American tradition of magical realism. At any rate, BNDES lending was important, funded a considerably part of investment in the country, but it was not what lifted the external constraint in Brazil during the 2000s (in the case of Brazil it wasn't even the commodity boom, even though that helped). That resulted from the relatively low international interest rates which allowed for significant inflows that were accumulated as reserves.

Finally, I'm not sure, but I would assume that BNDES' loans will be recouped for the most part, so this is not like an aid program. There is very little in BNDES lending in the 2000s that is illuminated by the comparison with the Marshall Plan, other than the limitations of this kind of analysis.