Coffee Traders' Forum - A Discussion about Coffee Futures Trading

Coffee Traders Discussion Forum

Coffee Traders' Forum


USDA - US Department of Agriculture
Mr. Tony Halstead

Ref.: Divergences on the Brazilian coffee harvest.

Dear Mr. Tony,

Firstly I would like to introduce myself. My name is Armando Mattiello, and I am the President of the ASSOCIAÇÃO DOS CAFEICULTORES DO BRASIL - SINCAL.

Recently a divergence between the USDA data and the data from CONAB (Brazilian oficial agency) could be noticed when regarding the information on the Brazilian coffee harvest expectation, as seen below:

The USDA Report on Production, Markets and Trade Reports, was published on December 15, 2017 link:
The CONAB data were published on December 21, 2017, the following link: Dezembro.pdf

It is very evident that there is a great difference between the two respectable institutions, as the sum of the 2 harvests is written down as to being 11.38%, and we have to recognize that there could not be such a big difference.

I understand that you are sovereign and have the right to disseminate your own data, and can also assume that the mistake in the expectation of harvest was made by CONAB.

However, when disregarding the expectations or forecasts to calculate the actual production using the DERIVED SAFRA data in 2016/17, we can see how the USDA data is unreal, as seen below:

The Brazilian coffee harvest in 2016/17 was 49,724,561 bags of coffee, which is not expected or forecast, and is the actual result from the Brazilian harvest of 2016.

Based on these data, we can say that CONAB has closer to reality than the USDA.

CONAB made a prediction with an error of 1.645 million bags, approximately 3.31% greater than reality.

When we compare this to the USDA data, we have an overestimated forecast of 6.376 million bags with a 12.82% bigger harvest.

To corroborate this USDA overvaluation claim, note that the reported expectation for USDA’s projected 56.10 million sacks of the 2016/17 crop is one of the biggest harvests in Brazilian history.

However, we have seen Brazil sell all its stocks of about 1.6 million bags between September 2016 and April 2017.

It can be seen that the inventories on March 31, 2017 were 10,121,405 bags, and in the period from April to June 2017, approximately 12,281,223 bags were used in exports and domestic consumption.

As shown above, the use of coffee was higher than what was held in inventory, presenting negative inventories in June 2017. For this there are two possible explanations.

i) the use of newly harvested coffees from the 2017/18 harvest, since the conilon harvest in some regions starts in APRIL, and the Arabica harvest begins later in MAY.
ii) remaining coffees that were not counted in the stock count.

Considering both hypothesis, we know that Brazil started the harvest of 2017/18 without private stocks of coffee, and there are no public stocks, as previously established.

Analyzing Brazilian exports from July to November, the period in which there is a greater availability of coffee in Brazil due to the harvesting period, it is noted that during the year of 2017 there was a sharp fall in exports, of about 14.60% less, resulting in the lowest exportations in the last 5 years, as seen in the table below from CECAFÉ.

We would like to inform you that in December, SINCAL collected information from producers, cooperatives, and warehouses in all Brazilian coffee regions, and the current levels of coffee stocks in Brazil are very low.

This will result in a much smaller amount of exportation in the first half of 2018.

Thus, due to the need for the reported data to be transparent, SINCAL respectfully requests USDA to inform and and disseminate the methodology for arriving at published data on Brazil’s harvest forecasts.

The 2018/19 harvest will begin in the first half of 2018, and we expect the USDA to make a realistic forecast of Brazilian production, as unrealistic forecasts are disadvantageous for everyone.

The spread of overestived data has led the world coffee producers to sell the fruits of their labor at discounted prices, leading to export revenue losses of more than 5 billion dollars, leading millions of farmers to remain forever in poverty.

This even if the coffee they produce helps coffee importing nations generate yearly tens of billions of dollars in profits, added value and taxes.

SINCAL is available to the USDA for more information and collaboration,


Armando Mattiello
January 05, 2018

Messages In This Thread

Re: SINCAL Brazil - USDA 2 *NM* *PIC*
Re: SINCAL Brazil - USDA 3 *NM* *PIC*