Skip to content

Soy bol 2023 explore

View or edit on GitHub

This page is synchronized from trase/products/analysis/notebooks/soy_bol_2023_explore.md. Last modified on 2025-12-13 00:30 CET by Trase Admin. Please view or edit the original file there; changes should be reflected here after a midnight build (CET time), or manually triggering it with a GitHub action (link).

Purpose

This code explores the information in the latest soy bill of lading (2023) and the columns available. In carries out 3 specific analyses:

  1. The information on cargo ownership and its link to consignees
  2. The information about shipper ZIP code and CNPJ municipal tax registration information
  3. The information on port terminals

1. Cargo consignees, forwarders

First, we check the information about cargo ownership and the information on actors in the supply chain. Let’s first check the magnitude of soy exports labeled as being owned by the shipper.

## # A tibble: 1 × 1
##   tonnes_tot
##        <dbl>
## 1   90492602

## # A tibble: 2 × 3
##   COMPANY_SHIPPER_TYPE   tonnes   pct
##   <chr>                   <dbl> <dbl>
## 1 NVO/Forwarder         5991427     7
## 2 Real Cargo Owner     84501175    93

So about 7% of shipments are taking care of by a forwarder as opposed to real cargo owner.

Let’s now check the number of unknown shippers (as exporters)

## # A tibble: 1 × 4
##   COMPANY_SHIPPER_SHIPPER_NAME COMPANY_SHIPPER_TYPE  tonnes   pct
##   <chr>                        <chr>                  <dbl> <dbl>
## 1 UNKNOWN CUSTOMER             NVO/Forwarder        5340759     6

These shippers are all forwarders, rather than real cargo owners.

Let’s now check the number of unknown Consignees

## # A tibble: 510 × 4
##    COMPANY_CONSIGNEE_CONSIGNEE_NAME            COMPANY_SHIPPER_TYPE tonnes   pct
##    <chr>                                       <chr>                 <dbl> <dbl>
##  1 UNKNOWN CUSTOMER                            Real Cargo Owner     7.05e7    78
##  2 UNKNOWN CUSTOMER                            NVO/Forwarder        5.71e6     6
##  3 BUNGE                                       Real Cargo Owner     2.48e6     3
##  4 COFCO                                       Real Cargo Owner     1.04e6     1
##  5 LDC TIANJIN INTERNATIONAL BUSINESS CO LTD   Real Cargo Owner     9.85e5     1
##  6 CARGILL                                     Real Cargo Owner     5.90e5     1
##  7 LOUIS DREYFUS                               Real Cargo Owner     4.57e5     1
##  8 SHANDONG BOHI OILS & FATS INDUSTRY CO LTD   Real Cargo Owner     4.08e5     0
##  9 JIUSAN OILS & GRAINS INDUSTRIES GROUP CO L… Real Cargo Owner     3.51e5     0
## 10 SANHE HOPE FULL GRAIN OIL GROUP FEED &PROT… Real Cargo Owner     3.29e5     0
## # ℹ 500 more rows

So 84% of exports to not have a consignee name and about 5.7 Mtonnes (6%) linked to forwarders.

We then look at the role of agents in the supply chain

## # A tibble: 54 × 3
##    CARRIER_CARRIER_AGENT     tonnes   pct
##    <chr>                      <dbl> <dbl>
##  1 ALPHAMAR                20738853 22.9 
##  2 CARGONAVE               14437455 16.0 
##  3 ROCHAMAR                 9369379 10.4 
##  4 FERTIMPORT               8724061  9.64
##  5 CARGILL                  6979640  7.71
##  6 RIO GRANDE               5693411  6.29
##  7 LBH BRASIL               3818497  4.22
##  8 AMAZONICA                3023747  3.34
##  9 WAYPOINT AGENCIA  MARIT  2873155  3.18
## 10 WILLIAMS                 2677222  2.96
## # ℹ 44 more rows

## [1] 99.99

100% of volumes is transferred through agents, meaning that even in cases where the shipper is “UNKNOWN CUSTOMER” then it still operates with an agent.

2. Port labels

We now check the port labels, particularly:

  • Place Of Receipt (POR) and Port Of Loading (POL)
  • Port Of Loading (POL) and Port Of Maritime Origin (POMO)
  • Port Of Maritime Destination (POMD) and Place of Destination (DEST)
## # A tibble: 2 × 3
##   test_POR_POL tonnes_POR_POL pct_POR_POL
##   <chr>                 <dbl>       <dbl>
## 1 NO                     2049           0
## 2 YES                90490553         100

## # A tibble: 2 × 3
##   test_POL_POMO tonnes_POL_POMO pct_POL_POMO
##   <chr>                   <dbl>        <dbl>
## 1 NO                          0            0
## 2 YES                  90492602          100

## # A tibble: 2 × 3
##   test_POL_POMO tonnes_POL_POMO pct_POL_POMO
##   <chr>                   <dbl>        <dbl>
## 1 NO                          0            0
## 2 YES                  90492602          100

## # A tibble: 2 × 3
##   test_POMD_DEST tonnes_POMD_DEST pct_POMD_DEST
##   <chr>                     <dbl>         <dbl>
## 1 NO                        73283          0.08
## 2 YES                    90419319         99.9

There are some cases where the Place of Receipt is not the same as the Port of Loading. This only occurs for 2049 tonnes in which the POR is Rondonopolis, interestingly enough. This might be an interesting additional indicator since the companies themselves (shippers) are not necessarily in Mato Grosso in these cases.

3. Exporter address ZIP codes and CNPJ registration

We now want to check whether the information on the exporter matches the CNPJ registry. We first look whether the company’s are registered in the cadastro to begin with.

## # A tibble: 126 × 2
##    COMPANY_SHIPPER_SHIPPER_NAME                           COMPANY_SHIPPER_REGI…¹
##    <chr>                                                  <chr>                 
##  1 ADM COMERCIO DE ROUPAS LTDA                            04744781000180        
##  2 ADM DO BRASIL LTDA                                     02003402003433        
##  3 ADM DO BRASIL LTDA                                     02003402000760        
##  4 ADM DO BRASIL LTDA                                     02003402007340        
##  5 AGRO LATINA LTDA                                       88320536000135        
##  6 AGROFARM IMPORTADORA E EXPORTADORA DE PRODUTOS VETERI… 02270540000110        
##  7 AGROPECUARIA SANTA MARIA DO CERNE LTDA                 05420042000286        
##  8 AGROPECUARIA VITAMAIS LTDA                             03568048000199        
##  9 ALCOOL MORIAH SA                                       08866762000187        
## 10 ALIANCA COMERCIO E EXPORTACAO DE MADEIRAS LTDA ME      09371206000100        
## # ℹ 116 more rows
## # ℹ abbreviated name: ¹​COMPANY_SHIPPER_REGISTRATION_NUMBER

There are 126 company names that are not directly linked to soy-specific CNAEs. We check a few here from the big traders:

  • Louis Dreyfus Company Brasil S.a. has 6 entries with CNAEs linked to shipping, fertilize, oil extraction
  • ADM do Brasil LTDA has 3 entries with office support CNAE, other CNAEs include fertilizer production and port activity
  • BUNGE Alimentos has 4 entries with CNAEs linked to oil and fats, food sales, wheat processing
  • Cargill has 6 entries with CNAEs linked to port activity, oil and fats, amido production, cocoa processing, concentrated juice and cotton
  • COFCO has 1 entry with CNAEs related to fertilizer and coffee
  • OLAM has 3 entries including cocoa processing and holdings

We check the volume the is exported by these companies that are not linked directly to a soy-CNAE

## # A tibble: 1 × 2
##     tonnes   pct
##      <dbl> <dbl>
## 1 16012494    18

So close to 16 Mtonnes of soy (18%) are not directly linked to soy activity.

For the companies that did match the cadastro data on s3, we compare the reported addresses between BoL and Cadastro

## # A tibble: 17 × 13
##    COMPANY_SHIPPER_SHIPPER_NAME             cnpj  COMPANY_SHIPPER_CITY city_name
##    <chr>                                    <chr> <chr>                <chr>    
##  1 CARGILL AGRICOLA SA                      6049… PORTO ALEGRE         PASSO FU…
##  2 COFCO INTERNATIONAL BRASIL SA            0631… VITORIA DA CONQUISTA LUIS EDU…
##  3 CARGILL AGRICOLA SA                      6049… PORTO UNIAO          SAO FRAN…
##  4 MARUBENI GRAOS BRASIL SA                 2514… CURITIBA             LONDRINA 
##  5 FIAGRIL LTDA                             0273… LUCAS DO RIO VERDE   CUIABA   
##  6 OLAM BRASIL LTDA                         0390… SAO PAULO            SANTOS   
##  7 GNOVA GRAINS AGRO LTDA                   4890… ROLANDIA             GOIANIA  
##  8 FERTITEX AGRO FERTILIZANTES & PRODUTOS … 7464… BARUERI              SAO PAULO
##  9 SIERENTZ AGRO BRASIL LTDA                0763… RIBEIRAO PRETO       SAO PAULO
## 10 COFCO INTERNATIONAL BRASIL SA            0631… LONDRINA             MARINGA  
## 11 COAMO AGROINDUSTRIAL COOPERATIVA         7590… SAO FRANCISCO DO SUL ITAPOA   
## 12 PERDUE COMERCIAL IMPORTADORA & EXPORTAD… 2645… JATAI                RIO VERDE
## 13 STELA MARIS TRANSPORTES E LOGISTICA LTDA 0873… BONFIM               BOA VISTA
## 14 BUNGE ALIMENTOS SA                       8404… VILA RICA            SINOP    
## 15 CJ SELECTA SA                            0096… GOIANIA              UBERLAND…
## 16 AGRICOLA ALVORADA SA                     0485… PARANAGUA            CURITIBA 
## 17 ENGELHART CTP BRASIL SA                  1479… SERRA                LINHARES 
## # ℹ 9 more variables: COMPANY_SHIPPER_STATE <chr>, state <chr>,
## #   COMPANY_SHIPPER_STREET <chr>, address_number <chr>, address_street <chr>,
## #   COMPANY_SHIPPER_ZIP <chr>, postal_code <chr>, municipality <dbl>,
## #   same_city <chr>

There are 17 cases where the CNPJ registration does not match the address of the shipper. Out of curiosity we check the volumes that these companies are moving.

## # A tibble: 1 × 2
##    tonnes   pct
##     <dbl> <dbl>
## 1 2399217     3

So about 2.4 Mtonnes (3%) of soy are likely coming from another area than what is declared in the cadastro.

Conclusions

Unfortunately there isn’t anything really new in the 2023 BoL data, only some minor potential improvements by changing addresses of shippers, but not really additional benefit from the new columns.