Skip to content

bcb.credito_rural_total v1.0

Rural credit by state and purpose, without product, from SICOR's RegiaoUF entity: the four purposes, including agro-industrialisation, which SICOR does not publish by product. Source contract of the bcb.credito_rural_total function; it is not a dataset.

Source

Source Entity Description
BCB/SICOR (OData v2, Olinda) RegiaoUF Count and value by month, state, programme, sub-programme, funding source and activity, with the four purposes as columns

No fallback to the Base dos Dados table.

Schema

Column Type Nullable Unit Stable
safra str No — Yes
uf str No — Yes
finalidade str No — Yes
agregacao str No — Yes
programa str Yes — Yes
cd_programa str Yes — Yes
qtd_contratos int No contracts Yes
valor float No BRL Yes
fonte str No — Yes

Primary key: [safra, uf, finalidade, programa]

Constraints: qtd_contratos >= 0, valor >= 0

  • safra in YYYY/YY format (e.g. 2022/23), from July to June.
  • finalidade: custeio, investimento, comercializacao or industrializacao.
  • With uf aggregation, programa and cd_programa are null. With programa aggregation, they identify the programme, with the current name from the official table.
  • valor is the sum in BRL, to the cent; qtd_contratos is the sum of contracts.

Source rules

  • Filler zero: in the wide row, the source carries 0 for purposes without operations. A pair with count = 0 and value = 0 does not become a row, and a purpose without operations is absent. Example: in crop year 2022/23, agro-industrialisation in AM, AP and RR and marketing in AP.
  • No Brazil row: SICOR publishes no national total. The Brazil total is the sum of the states, and the function does not invent the row.
  • Partial crop year: the current crop year is partial. MetaInfo.source_details["meses"] records the first and last month with data and the number of months.
  • Crop-year query × sum of the monthly queries: SICOR may return, for the crop-year query, numbers that differ from the sum of month-by-month queries. On 2026-09-26, for crop year 2026/27 (July and August), 51 state × purpose pairs diverged; for custeio in AC, 274 contracts and R$ 56,788,261.98 in the crop-year query, which the function uses, against 272 and R$ 56,541,830.82 in the monthly ones. The cause was not identified, and agrobr reproduces the body received.
  • Consistency across entities, checked for 2022 and 2023: the total by state and purpose equals the sum of the municipalities of CusteioInvestimentoComercialIndustrialSemFiltros and, for operating costs, investment and marketing, the by-product sum of the *RegiaoUFProduto entities read by credito_rural.

Version history

Version Change
v1.0 Initial contract

Example

from agrobr import bcb

df = await bcb.credito_rural_total(safra="2022/23")
industrializacao = await bcb.credito_rural_total(safra="2022/23", finalidade="industrializacao")
por_programa = await bcb.credito_rural_total(safra="2022/23", uf="MT", agregacao="programa")

JSON Schema

Available at agrobr/schemas/bcb_credito_rural_total.json.

from agrobr.contracts import get_contract

contract = get_contract("bcb_credito_rural_total")
print(contract.to_json())