Text for Discussion:
Definitions and Data:
What affects the Brazilian exports?
Carlos Feu Alvim
The recovery signs of the American economy has already provoked a wave of optimism regarding business in Brazil. The argument is that a recovery of that country’s economy and of other subsidiary economies will increase our exports.
The situation seems analogous – although in the inverse sense – to the one that rises when a crisis develops in the American economy; then immediately a pessimism grows in the Brazilian economy. In these situations we usually hear the old phrase: “a cold in the United States corresponds almost always to a pneumonia in Brazil”. There are some who say just the opposite: the American crises have initially a negative effect on Brazil and later on a long period of beneficial effects.
Now, the USA developed a crisis when W. Bush arrived, after almost 10 years of continuous growth and recovery of the public accounts in the Clinton Administration. However, in the decade of the American prosperity the growth of the Brazilian per capita income was on a skateboard. And now, during the months of the American crisis our exports are reaching historical records. This situation seems, to say the least, surprising!
In the present situation, a recovery in the USA will cause an increase of the interest rates in the dollar market. Since Brazil is deeply indebted, the first consequence could be not so positive since in the short run this would make difficult the inflow of capital in Brazil and we would have to make a larger effort in order to pay the debt services. Anyway, in the preliminary analysis of the present article, if it is not evident the negative effect of the large growth in the USA on our exports, it is not possible either to detect any positive effect of the American crises on the Brazilian exports.
In Brazil we are specialists in false economical diagnosis. The medical doctors know (and more so the patients who experienced it) that worse than not being treated is when a patient is treated based on an erroneous diagnosis.
Furthermore, it has been exhaustively repeated along the process of commercial liberalization of the Brazilian economy that in order to export more we would have to make our products become cheaper, that is, to make them more competitive. Likewise, we believe that the exchange rate must be an important factor in our exports. But, in the case of Brazil, do these statements find any support?
In what follows we have tried to analyze the behavior of the Brazilian exports in terms of price, quantity (quantum) and value (price x quantity) mainly in the last thirty years. The main data source is the on line data base of the International Monetary Fund (IMF). Additionally, some data were updated using the available series from IPEADATA.
In what concerns values, we have verified in the previous e&e issue that in the last thirty years the value of our exports in international currency has simply followed the real value of the GDP. To our astonishment it does not even seem to have any significant correlation with the effective exchange rate.
In Figure 1 we can observe that during the period of over-evaluation of the Real at the end of the Itamar Administration and the first mandate of Fernando Henrique Cardoso the value of our exports simply followed the real growth of the GDP. On the other hand, the exceptional export values in 2003 can simply reflect (we hope not) an oscillation already previously verified when a recession in the economy starts. Surprised by the crisis, the productive sector exports the surplus, even for the sole purpose of maintaining the GDP per inhabitant. When the country resumes growth, the surplus vanishes and the exports go back to the previous level.
Figure 1: Exports and real GDP relative to 1995.
In Figure 2 we present the evolution of the real GDP calculated using three deflation methods: (i) expressed in current dollars and deflated by the PCI of the USA; (ii) expressed in terms of the purchase power parity (PPP) and also deflated by the PCI of the USA, and (iii) real value of the GDP . The values are relative to 1969 when we consider that there was an exchange rate near to the balanced one.
It should be observed that the real GDP value in terms of PPP and in terms of domestic currency are very close until 1989. However, they are shifted from 1992 on and since then they have a similar behavior. Also in Figure 2 we can observe that the real value of the GDP in terms of the current exchange rate (corrected by the American inflation) undergoes large variations. Evidently, if we express our exports as a function of the GDP real value according to (iii) then our “tape measure” will vary in such a way that it may induce inaccurate evaluations.
The export/ GDP ratio in local currency is an important parameter for evaluating, for example, the remuneration of exporters. However, from the external point of view, for example, this ratio has no meaning concerning the external debt payment.
For example, let’s consider the years 1998 and 1999. In current values, the exports decreased from 58 to 40 US$ billion. The real GDP in terms of domestic currency grew 3.2% but when it is expressed in current dollars, it decreased from 788 to 547 billion dollars. However, the exports/GDP ratio both in dollars and in current reais increased from 7.3% to 9.2%.
Figure 2: Real GDP, current and purchase power parity exchange rate
We have deliberately compared exports and GDP using different indexes. In order to find “real” values for these variables, when inflation is discounted, we must apply a deflator according to the objective of the study. What we present here should be understood as an indicator of the purchase power of dollars obtained by the Brazilian exports to buy goods at the consumer’s level (PCI) in the USA or the quantities of the Gross Domestic Product in the USA.
In Figure 3 we present the ratio between exports and real GDP calculated through five ways: (i) exports in current dollars and GDP in purchase power price dollars; (ii) exports index deflated by the PCI of the USA and real GDP index; (iii) exports index corrected by the American GDP implicit deflator and real GDP index – in this case, both indexes were normalized to 1995;(iv) exports index and real GDP index in terms of PPP – in this case it is hoped that the ratio behavior describes the behavior of the exports/GDP ratio; (v) exports index corrected by the PCI of the USA and GDP deflated by the INPC of the IBGE.
As can be seen in the figure below the results found are very similar for the five ways used to calculate the ratio.
Figure 3: Exports/GDP ratio according to different deflators
From 1973 on exports took values in a plateau that lasted until 1990. The values in this plateau were slightly higher that 7% of the GDP. Minimum values (preceded by positive peaks) were recorded around 1982 (the Mexican crisis and help from the IMF) and 1987 (technical moratorium). We must say that there was a significant change in 1973 when the values of our exports increased from 6% of the GDP to values slightly higher than 8%.
This behavior permits to infer that the question of fixing the exchange rate and even the recent crisis in the USA had little or no influence at all on the exported value. We have pointed out this fact in the Nº 38 issue of e&e.
We emphasize again that in the present analysis we are treating external exports values since we have compared export values relative to the GDP expressed by the average annual exchange rate. However for the exporter it makes a big difference since normally a large part of his costs are in national currency.
Nevertheless, our historical values show that altering the exchange rate is an efficient measure to increase the exports values.
The exports values relative to the GDP can be analyzed in terms of the quantity (quantum), price and its own value (price multiplied by quantum). Figure 4 shows what occurred in the last thirty years according to data available at the IMF.
Figure 4: Price, quantum and exports value
In the figure above one can observe that while the exported quantum relative to the GDP has grown 60%, prices have dropped almost to half the value of 1973. The result of those two effects combined was an exports value relative to the GDP that decreased until 2002. Apparently in 2003 the exports value relative to the GDP seems to go back to the 1973 value.
Therefore, it seems that the Brazilian problem is not reducing costs. The reductions verified along the last thirty years did not result in any increase of exported value relative to the GDP. The exports increase was practically compensated by the decrease of the real prices of the exported products.
If we try to correlate the quantity exported relative to the GDP with the prices obtained by our products we would arrive at the conclusion that in the last thirty years we were caught in a trap, the more we exported the smaller the price we got for our goods.
As shown in Figure 5, for each 10% increase in the exported volume (relative to the GDP) we obtained a reduction of 16% in the price in constant dollars (corrected by the PCI of the USA) for our products.
Figure 5: Price index of the Brazilian exports (corrected by the American PCI) and exported volume index as a function of the real GDP (indexes values relative to 1995)
From the figure above we can infer that the qualitative behavior observed is coherent with what is expected from countries with capacity of forming prices. A larger supply corresponds to a smaller price for the product. Actually, for some important products of our exporting list (coffee, iron and steel, soybean, sugar and orange juice) Brazil has a participation that is sufficiently important to play this role. A better understanding of what has occurred demands an analysis of the behavior of the main exports products or group of products prices and the Brazilian participation in the world trade of these products.
On the other hand by limiting our analysis to data from 1973 on, we are choosing a period when we know that this was really the behavior and we should not expect that it reflects a casual relationship that can be extrapolated to the future.
Anyway there has been little discussion in Brazil regarding a more general problem, namely the terms of trade in our external commerce. The concentration on efforts to reduce the so called “Brazil cost” could be disguising more fundamental problems connected with our efforts to get resources from exports. After all there is no use in reducing the price of our products if that means to send a larger part of our product for the same value. For example, what’s the use of giving up the internal taxes and get as a result an additional taxation in the destination?
Figure 1: Exports and real GDP relative to 1995.
Figure 2: Real GDP with different deflators.
Figure 3: Exports/GDP ratio according to different deflators.
Figure 4: Price, quantum and exports value.
Figure 5: Price index of the Brazilian exports (corrected by the American PCI) exported volume index as function of the real GDP (index values relative to 1995)
 The effective exchange rates reflect the purchase power of the national currency relative to different baskets of foreign currencies. The real effective exchange rate is a type of average bilateral exchange rate of the countries and their main commercial partners corrected by the difference between the internal and external inflation.
 The real GDP value corresponds to the nominal one corrected by the GDP implicit deflator.
 The equilibrium exchange rate considered is the one that equilibrates the balance of goods and services.
 Then one could say that they are excessively coincident to be independent.
 However, this peculiarity is almost not remembered in several internal discussions such as, for example, that of the minimum wage that is compared with that of 100 dollars.
 In the case of exports corrected by the American PCI we would be comparing consumer’s prices in the USA with the GDP in real value. The values in terms of purchase power parity refer to a basket of products equivalent to consumer’s prices in the country in question (Brazil) relative to the consumer’s prices of the same products in the USA.
 The claim that no one exports taxes can be valid for manufactured goods. However, it is not justified for extractive or basic industry products.
Graphic Edition/Edição Gráfica:
Tuesday, 11 November 2008.