FEC Campaign Finance Search (Candidates, Donors, PACs)

Search the US Federal Election Commission's official register of federal candidates, committees and PACs, and individual campaign contributions, and get flat spreadsheet-ready rows back. No API key from you, no login and no scraping: this reads the FEC's own API, which is free and public because federal law requires campaign finance filings to be disclosed. Set max_items before each run to keep output and cost predictable.

What data can I extract?

Three searches, selected with data_type. A run returns one record type, because candidates, committees and contributions have genuinely different fields and mixing them would hand you half-empty rows.

Candidates — name, party, office, state, district, status, incumbent_or_challenger, election_years, active_through, has_raised_funds, first_file_date, last_file_date, url.

Committees and PACs — name, committee_type, designation, organization_type, party, state, treasurer_name, filing_frequency, candidate_ids, first_file_date, last_file_date, url.

Contributions — who gave, how much, and to whom: contributor_name, contributor_city, contributor_state, contributor_zip, contributor_employer, contributor_occupation, entity_type, is_individual, amount, date, receipt_type, committee_id, committee_name, candidate_name, candidate_office, two_year_period.

Contributor street addresses are deliberately excluded. The FEC publishes them, and disclosure is the entire point of the system, but city, state, ZIP, employer and occupation are what make the data analytically useful — they let you map giving by geography, industry and profession. The street line adds nothing to that analysis and is the most sensitive part of the record.

Every row is written to the default dataset and can be downloaded as JSON, CSV, Excel or XML, or pulled through the Apify API.

All 131 columns

Identity — affiliated_committee_name, agent_name, agent_title, candidate_id, candidate_ids, candidate_name, candidate_office, candidate_office_district, candidate_office_state, committee_id, committee_name, conduit_committee_id, conduit_committee_name, contributor_first_name, contributor_id, contributor_last_name, contributor_middle_name, contributor_name, district_number, donor_committee_name, id, image_number, individual_contributions, individual_itemized_contributions, individual_unitemized_contributions, is_individual, line_number_label, name, principal_committee_id, principal_committee_ids, principal_committee_name, sponsor_candidate_ids, sponsor_candidate_names, transaction_id, treasurer_name

Status — active_through, committee_type, committee_type_code, designation, designation_code, election_type, election_type_code, entity_type, entity_type_code, federal_funds_flag, form_type, has_raised_funds, inactive_election_years, is_inactive, last_report_type, organization_type, organization_type_code, party_type, receipt_type, receipt_type_code, record_type, report_type, status

Dates — coverage_end_date, coverage_start_date, date, election_years, fec_election_year, first_f1_date, first_file_date, last_f1_date, last_f2_date, last_file_date, last_report_year, load_date, report_year, two_year_period

Location — address_city, address_state, address_street, address_zip, committee_city, committee_state, committee_zip, conduit_committee_city, conduit_committee_state, contributor_city, contributor_state, contributor_zip, district, state, state_full, treasurer_city, treasurer_state

Money — amount, cash_on_hand, contribution_refunds, debts_owed, disbursements, independent_expenditures, net_contributions, net_operating_expenditures, operating_expenditures, other_political_committee_contributions, political_party_committee_contributions, receipts

People — agent_phone, contributor_aggregate_ytd, contributor_employer, contributor_occupation, contributor_prefix, contributor_suffix, transfers_from_other_authorized_committee, treasurer_phone

Contact — email, fax, pdf_url, url, website

Counts and measures — memoed_subtotal

Other detail — amendment_indicator, cycles, filing_form, filing_frequency, financials_cycle, increased_limit, incumbent_challenge_code, incumbent_or_challenger, leadership_pac, lobbyist_registrant_pac, memo_code, memo_text, office, office_code, party, party_code

Input example

{
  "data_type": "candidates",
  "state": "CA",
  "party": "DEM",
  "office": "H",
  "max_items": 50
}

Field	Applies to	Notes
data_type	all	candidates, committees or contributions
name	all	Candidate, committee or contributor name, depending on data_type
state	all	Two-letter code. For contributions this is the contributor's state, not the recipient's
party, office, election_year	candidates	party takes the three-letter code, office is H, S or P
committee_type	committees	FEC committee type code, for example N for a PAC or H for a House campaign committee
committee_id, employer, occupation, min_amount	contributions	committee_id narrows to one recipient
two_year_period	contributions	Election cycle; 2024 covers 2023 and 2024. Defaults to 2024
max_items	all	Hard ceiling on rows, and therefore on spend. Defaults to 100

The default input is deliberately small. Unsupported parameters are rejected rather than silently ignored.

Output example

Real rows from this Actor, unedited. A candidate:

{
  "record_type": "candidate",
  "id": "H2CA30291",
  "name": "AAZAMI, SHERVIN",
  "party": "DEMOCRATIC PARTY",
  "office": "House",
  "state": "CA",
  "district": "32",
  "status": "N",
  "incumbent_or_challenger": "Challenger",
  "active_through": 2024,
  "first_file_date": "2021-01-16",
  "last_file_date": "2022-07-15",
  "has_raised_funds": true,
  "election_years": "2022, 2024",
  "url": "https://www.fec.gov/data/candidate/H2CA30291/"
}

And a contribution:

{
  "record_type": "contribution",
  "id": "4121220241075839591",
  "contributor_name": "MURPHY, PAUL",
  "contributor_city": "SANTA CLARA",
  "contributor_state": "CA",
  "contributor_zip": "95054",
  "contributor_employer": "APPLIED MATERIALS INC",
  "contributor_occupation": "MANAGING DIRECTOR OF ENGINEERING",
  "entity_type": "INDIVIDUAL",
  "is_individual": true,
  "amount": 120,
  "date": null,
  "receipt_type": "CONTRIBUTION",
  "committee_id": "C00406892",
  "committee_name": "APPLIED MATERIALS, INC. POLITICAL ACTION COMMITTEE",
  "candidate_name": null,
  "candidate_office": null,
  "two_year_period": 2024
}

That date: null is not a bug and it is not rare — recurring payroll-deduction and memo lines routinely file without a receipt date. candidate_name and candidate_office are null here for the same honest reason: they are only populated on conduit rows, where a contribution passed through ActBlue, WinRed or a similar platform earmarked for a named candidate. Fields the FEC does not supply stay explicitly null rather than being filled in with a plausible guess.

Common use cases

• Mapping employer and industry giving from contributor_employer and contributor_occupation
• Journalism and opposition research on who funds a particular committee
• Building a full candidate roster for a state, party or cycle before an election
• Lobbying disclosure and political-exposure checks against a list of names
• Tracking a single recipient over time by passing its committee_id back in
• Academic and quantitative study of campaign finance patterns

Use with AI agents and MCP

Apify's MCP server can discover and call this Actor from an AI workflow. Example intent:

Find contributions of $5,000 or more from Texas donors in the 2024 cycle and tell me which committees received them.

Use the JSON from Input example as the tool arguments, changing data_type to contributions. The strict input schema rejects unsupported parameters instead of ignoring them, and the dataset schema documents every field — including which ones are routinely null and why, which matters because an agent that reads a null candidate_name as a failure will retry a run that actually succeeded.

Pricing and cost control

Output is billed per result at $0.003 per result (about $3.00 per 1,000 results), plus a $0.0005 Actor-start charge billed once per gigabyte of memory at run start. Use max_items to cap both output volume and charges. The price shown on the Apify Store listing is authoritative.

So 10 rows cost $0.03, 100 rows cost $0.30, 1,000 rows cost $3.00, and 20,000 rows — the schema maximum — cost $60. Apify platform charges shown separately on your account are not included in those figures.

There is no subscription and no minimum. max_items is a hard ceiling on rows returned, so the most a run can cost you is a number you fix before you start it. Begin with 10 rows, read them, then scale deliberately.

Reliability

Contributions do not page the way you expect, and getting it wrong fails silently. Schedule A — the individual contribution file — holds tens of millions of rows and page numbers stop advancing on it: ask for page 200 and you get page one back, with a 200 status and no error. Continuation is by keyset instead: you hand back the previous row's index and receipt date. Candidates and committees, in the same API, use ordinary page numbers. A wrapper that uses page numbers everywhere re-reads page one forever and returns the same 100 donors over and over without ever raising an error. This Actor uses the correct pagination for each endpoint.

The other reliability problem is the key. The FEC API is free but keyed, and the shared demo key that everything falls back to allows ten requests an hour — roughly a thousand contribution rows before it stops. There is no key field in the input: the Actor takes its key from the platform side, so there is nothing for you to register or paste in. It does not truncate quietly either. The first line of the run log names which key the run is using, and a rate-limited or failed request ends the run with the status code logged rather than handing you a short dataset that looks complete.

Rate limiting and upstream errors stop the run and are logged with the status code, partial results are kept rather than discarded, and year fields that the FEC returns as strings on one endpoint and integers on another are coerced at write time so the dataset schema always holds.

Limitations and responsible use

• One record type per run. Candidates, committees and contributions cannot be mixed in a single run.
• No financial totals or summaries. Candidate receipts, disbursements, cash on hand and independent expenditures live on separate FEC endpoints and are not returned here.
• A contribution row is one filing line, not one donor. Somebody giving monthly appears as twelve rows. There is no de-duplication and names are not normalised, so SMITH, JOHN and SMITH, JOHN A stay separate.
• receipt_type matters when you sum. Memo entries record a share of a joint fundraising contribution and will double-count if you add them to the parent line.
• two_year_period is mandatory for contributions and is the FEC's own cycle grouping, not a calendar year.
• Employer and occupation are self-reported by the donor, frequently blank or abbreviated, and verified by nobody.
• status and filing_frequency are raw single-letter codes. The API returns null for its own expanded labels, so this Actor passes the codes through rather than inventing meanings for them.
• Filings are amended. A contribution can be revised or removed after it first appears, so a figure pulled today may not match one pulled next month.

This is public disclosure data that federal law requires to be published, and the FEC provides the API free for programmatic use. The Actor uses that official API — it does not scrape the website and there is no rate-limit evasion or access circumvention involved. Use it lawfully, respect the source terms, and treat named individuals in the contribution file as people rather than as leads.

FAQ

Does it need my own API key?

No. There is no key field in the input and nothing for you to register at api.data.gov. Check the first line of the run log if a dataset comes back shorter than you asked for: it names the key the run used, so a rate-limited run is visible rather than silent.

Why is date null on some contributions?

Because the filer did not report one. Recurring payroll-deduction and memo lines frequently omit the receipt date, and the Actor leaves the field null rather than substituting a date from elsewhere in the record.

Why are candidate_name and candidate_office usually empty?

They are only filled in on conduit rows, where the contribution was earmarked for a named candidate through a platform such as ActBlue. A contribution made directly to a committee carries no candidate attribution on the line.

Can I get contributor street addresses?

No. They are excluded by design. City, state and ZIP are returned, which is enough for geographic analysis.

How do I control cost?

Set max_items to the most rows you want. It is a hard ceiling on both output and charges.

Can I export the results?

Yes. Download the default dataset as JSON, CSV, Excel or XML, or retrieve it through the Apify API.

Can an AI agent call it?

Yes. The Actor has a strict input schema and a fully documented output schema, and can be called through Apify's MCP tooling.