Brazil’s economy (Part 1): A crisis?

Did you ever want to know what’s causing Brazil’s economic crisis?  If there is (good) evidence suggesting Brazil’s real economic growth over the last 25 years is substantially equivalent to that of Germany and the USA, would it change your answer?   Inquiring minds will want to read on …

1. Unanswered questions

Few Brazilians or foreign financial professionals seem to tire of discussing Brazil’s ongoing economic crisis, which seems to have begun in mid-2014. Although there is much talk about (potential) causes of the crisis, there seems little clear consensus on what the actual causes are.  Among the more provocative explanations offered is that a perceived increase in “Brazilian political risk” following on the Petrobras / Lava Jato scandal resulted in a withdrawal of foreign investment with negative knock-on effects throughout the economy.

Another explanation is that Brazil is suffering from the Dutch Disease, caused by over-reliance on mining, minerals, and agriculture resulting in some unspecified negative effects on other sectors of Brazil’s economy.

A seemingly more plausible explanation is that the ongoing global economic instability begun in 2008-2009 combined with underdeveloped infrastructure, difficult to manage bureaucracy and corruption, and an increasingly risky global and Brazilian geo-political environment is causing various disruptions to physical and financial supply chains.  In turn, such disruptions induce a pathological feedback cycle where consumer incomes and demand decrease, production and profits decrease, banks restrict credit and loan losses increase, and on it goes.  (Despite widespread confidence in central banks to stabilize economies, pathological feedback cycles are problematic because the various causal factors are largely outside the control of even large, powerful economic institutions; including central banks and governments.)

So, which explanation makes the most sense?  More generally, what really did cause the Brazilian economic crisis?  Let’s get started!

2. Recent economic data

To begin, it’s helpful to understand the existing scope and duration of the recent crisis.  Consider how indicies of several important economic statistics have changed over the last five years:

The graph basically shows Brazil’s economy is producing 2% less than in early 2012 when we measure economic output (GDP) using the Brazilian Real at its circa 1995 value. But, if we remeasure output by converting it into its US Dollar value, then output has contracted by 7%.  The graph also shows that while there have been substantial decreases in retail sales volume of about 9%, and in total employment of about 6%, both of these economic statistics are still at or above early 2012 levels.

In short, Brazil lost the economic gains that it made between 2012 and 2014 amidst the global economic instability that began in 2008-2009.  From the optimistic perspective, Brazil circa 2009-2012 was generally not in bad economic condition relative to its own economic history, even if it was being criticized by some foreign governments, economists, and journalists.  And given the generally poor global economic conditions and trends since 2008-2009, it’s not at all clear that Brazil losing its economic gains was not to be expected.  After all, we all live in the global economy, right?  And if the globe gets a cold, a (somewhat) smaller economy like Brazil tends to get the flu, so to speak.

So, relatively speaking, Brazil losing its economic gains made during 2009-2014 is perhaps not an economic crisis per se: The statistics basically suggest Brazil is still better off than it was prior to 2012 as we will see later.

If we re-value Brazilian economic output in US dollars, however, we have what a number of financial journalists have characterized as “Brazil’s worst economic crisis in X years” where X is a large number like 60 or 70. Ai caramba, is Brazil’s crisis really that bad?

3. A digression on valuing economic output

(Some readers will want to skip this technical section …  )

It’s worthwhile to stop for a moment to examine this issue:

Does it actually make sense to re-value
Brazilian economic output in US dollars?

Many economists, finance professionals, and journalists I know believe the answer is, “Yes, of course.”  But economic truth has a bit more to do with facts than with consensus.  So, just for fun, let’s explore this question in a bit more detail.  Consider the following, not exactly hypothetical, situation:

(1) A producer has 100 units of a product that can be sold in Brazil for R$ 3.00 each, and in the global market for US$ 1.00 each.  The value of the 100 units is R$ 300 or US $100.  The current BRL/USD rate is 3.00, suggesting there is purchasing power parity (PPP) between BRL and USD.

(2) The producer sells 25% (25 units) in the export market for US$ 25, and sells 75% (75 units) in the domestic market for R$ 225. The producer’s revenues stated in BRL are …

REV = (3.00)(US$ 1.00)(25) + (R$ 3.00)(75) = R$ 300.

(3) But if export sales proceeds are to be repatriated, the producer must pay a FX commission rate of 3.5%.  This means that, net of FX commissions, the producer’s revenues stated in BRL are approximately …

REV = (1-.035)(3.00)(1.00)(25) + (3.00)(75) ~= R$ 297

… which is consistent with purchasing power parity adjusted for transaction costs.

Several observations follow:

(4)  If PPP holds between two currencies across time and there are no transaction costs, then measuring either economic output levels or changes in levels should be equivalent whether we use the domestic or foreign currency for measurement.

(5)  If, however, there are FX transaction costs while PPP holds, then it’s also necessary to consider such transaction costs to the extent of import and export transactions; otherwise we introduce measurement error.

(6)  If PPP does not hold then it only makes sense to measure Brazilian economic output in US dollars to the extent that Brazilians want, need, and have the ability to avoid repatriation of (net) export sales proceeds.

It follows from these observations that how we interpret the large deviations between measuring Brazilian GDP in Brazilian (constant 1995) Real and in US Dollars, as seen in the above graph, depends critically on whether PPP holds, transaction costs, and the extent of net Brazilian exports.

Based on this argument, I will somewhat ignore my more learned friends and focus on measures of Brazilian economic output using either deflated Brazilian Real or deflated US Dollars adjusted for estimated PPP.  With this settled, it will be helpful to understand a longer history of the Brazilian economy before moving on to examining actual causes of Brazil’s recent economic crisis.  Without this longer term perspective, one is likely to fall into the lazy mental habit of confusing correlation with causation; blaming, for example, Dilma Rouseff’s Partido Trabalhista for the crisis because it happened to be in power when the crisis began.

4. A longer-term economic perspective on Brazil

Understanding what is causing Brazil’s economic crisis is roughly the same thing as understanding what causes variation in economic output levels and growth rates.  And understanding causes of such variation ultimately involves developing reasonable theory / hypotheses and testing whether actual economic data are consistent with the theory.  But given the nature of statistical methods, tests of this kind require more than a few years of data; e.g., 2012 – 2016.  So, 20+ years of economic data from a variety of sources will ultimately be used in future articles.  In this article, I will just focus on high level aggregated data available from the World Bank to get a broad idea of Brazil’s economic history leading up to the current crisis.

As mentioned, all major economies operate inside the global economic environment and, therefore, their economic performance cannot be meaningfully evaluated independently.   Further, differences in income distributions between countries can make cross-country GDP comparisons per se less accurate measures of economic performance.  To avoid these problems and those related to foreign currency measurements of economic output, consider the following graph of GDP per capita measured at PPP (in constant 2011 USD) for Brazil and selected other major economies:

Although the graph is perhaps a bit difficult to interpret, it shows that the USA and Germany generate between 200% and 350% more economic output per citizen than do Brazil, China, and Mexico.  This perhaps does not quite capture economic reality completely because both the US and Germany have substantially more fixed capital investment per citizen than do Brazil and Mexico and, so, it would be expected that returns to economic output are likely more skewed towards owners of fixed capital rather than distributed equally across workers, citizens, or households.  This, of course, suggests understanding factors causing variation in Brazil’s relative economic performance requires controlling for differences in fixed capital investment across time and across countries.  But exploring this issue will be deferred until future articles in this series.

More importantly for the present discussion, the graph shows that Brazil’s economic performance between 1990 and 2015 is quite similar to Mexico’s performance.  Think about that for a moment: Brazil’s economic performance is about the same as that of Mexico–one of the US’ major trading partners–with which the US shares a geographic border and a free trade agreement (implemented in 1994).  This suggests Mexico has important economic benefits in terms of lower transportation costs, import-export tariffs, and a variety of other transaction costs with the largest single country economy in the world … and yet Brazil’s economic performance seems quite similar to that of Mexico over the last 25 years.

To see the economic growth more clearly, consider the following graph using the same basic data as the previous graph but indexing to 1990 and omitting China ( … China’s growth under PPP is high enough that it would make the chart basically unreadable):

Recall from my digression on economic output measurement that it only makes sense to measure GDP using a foreign currency if PPP holds.  And, if we are interested in real growth in economic output, then we need measurements in constant monetary units, which also allows for meaningful cross-country comparisons.  So, the graph reasonably allows us to make good inferences about Brazil’s relative economic performance between 1990-2015, including comparisons to other countries.

Three things can be seen quite clearly in this graph:

  • Economic growth in Brazil between 1990 and 2015 is substantially above that of Mexico, and is substantially equivalent to that of Germany and the USA.
  • Brazil, the USA, and Germany have recovered from the 2008-2009 financial crisis all at substantially the same rates.
  • Brazil’s “crisis” seems to be limited to the recent, very sharp economic contraction beginning in 2014.

Of course, if Brazil’s recent economic output trend continues, then there will likely be relatively severe consequences.  So, it really is important to not dismiss the recent trends.  At the same, it’s also important not to dismiss the evidence that Brazil’s economic performance over the last 25 years is–in aggregate–substantially the same as that of both Germany and the USA; and Brazil accomplished this without a number of important economic advantages had by Germany and the USA.  Hmmm.  Food for thought, if nothing else:

5. Tentative conclusions

Let’s think about all this seriously.  Is Brazil in it’s worst economic crisis in the last 50+ years?  Well, maybe.  But if it is, does not also the statistical evidence presented above suggest the USA and Germany (and Mexico) are also in dire economic straits?  Perhaps.  But what is clear is that Brazil’s economic performance really isn’t that much different than at least two or three other major economies over the last 25 years.

The problem is that it’s still not clear what’s causing Brazil’s economic performance–for better and worse–either over the long-run or in more recent years.  So, what is needed is good, logical economic theory on what’s causing Brazil’s economic performance; and then testing both long-run and recent economic data to determine if it is consistent with the theory.

As a preview of what’s to come in this series of articles, the interested, math-inclined reader might like to know the basic analytical framework to be employed.  The basic approach will begin–though diverge in some respects from–a basic, standard macroeconomic framework …


… with a focus on identifying and measuring the effects of causal factors influencing the different components of the above equation.

I hope the esteemed reader is beginning to doubt, as do I, the seeming consensus view that Brazil is in some kind of unusual economic crisis that neither it nor other major countries have experienced in several generations.  But go ahead and doubt on if you prefer; we’ll see in the end, one way or another.  :- )

São Paulo

Edit: Titled changed, 2017-05-02.

Caveats.  Please note: (i) views presented above are my own and do not reflect those of others; (ii) like anyone, I’m not infallible and am responsible for any errors; (iii) I greatly appreciate being informed of any significant errors in facts, logic, or inferences and am happy to give credit to anyone doing so; (iv) the above article is subject to revision and correction; and, (v) the article cannot be construed as investment or financial advice and is intended merely for educational purposes.  MMc