Form N-3 EL Files Dataset

The Form N-3 EL Files Dataset is a corpus of SEC EDGAR registration statements and amendments filed on submission types N-3 EL and N-3 EL/A — the EDGAR-routed variant of Form N-3 used by insurance-company separate accounts that are organized as management investment companies and that issue variable annuity contracts. One record is one EDGAR accession: a single registration statement (or an amendment to one), captured as a folder containing a metadata.json header and one plain-text document-<N>.txt per non-image document inside the original EDGAR submission envelope. The dataset's coverage begins on 1 May 1995, the EDGAR mandate era for investment-company registration statements, and runs through the present. Records are packaged into monthly ZIP shards under the path convention <YYYY>/<YYYY-MM>.zip, and the corpus is narrow by design — N-3 EL is a niche EDGAR submission-type variant used only by managed-account variable annuity issuers, a small subset of the broader U.S. variable annuity market that mostly files on Form N-4.

Update Frequency
Daily
Updated at
2026-04-16
Earliest Sample Date
1995-05-01
Total Size
1.6 MB
Total Records
64
Container Format
ZIP
Content Types
TXT, JSON
Form Types
N-3 EL, N-3 EL/A

Dataset APIs

Programmatically retrieve the full list of dataset archive files, download URLs and dataset metadata.

Dataset Index JSON API

Download the entire dataset as a single archive file.

Download Entire Dataset:

Download a single container file (e.g. monthly archive) from the dataset.

Download Single Container:

Dataset Files

10 files · 1.6 MB
Download All
1997-09.zip191.5 KB3 records
1996-12.zip105.2 KB1 records
1996-10.zip125.0 KB6 records
1996-09.zip203.1 KB13 records
1996-08.zip149.9 KB14 records
1996-07.zip84.5 KB2 records
1996-05.zip156.1 KB6 records
1996-02.zip47.8 KB3 records
1995-09.zip273.0 KB14 records
1995-05.zip251.3 KB2 records

What This Dataset Contains

The dataset captures every EDGAR accession of submission type N-3 EL (initial registration statement) and N-3 EL/A (amendment) from May 1995 forward. Form N-3 is the SEC registration statement prescribed for separate accounts of insurance companies that are organized as management investment companies (as distinct from unit investment trusts, which use Form N-4) and that issue variable annuity contracts. The filing is made jointly under the Securities Act of 1933 (registering the contracts as securities) and the Investment Company Act of 1940 (registering the separate account as an investment company). The EL suffix denotes the EDGAR submission-type variant used for electronic filing of Form N-3 registration statements; N-3 EL/A is the corresponding amendment variant, used for both pre-effective and post-effective amendments.

Substantively, a Form N-3 filing comprises three logical parts prescribed by the form's instructions:

  • Part A — Prospectus. The disclosure document delivered to contract owners. Describes the variable annuity contract, the separate account, its investment objectives and policies, the sponsoring insurance company, the contract's fee structure, accumulation-phase and annuity-phase mechanics, death-benefit terms, federal income tax consequences, and material legal proceedings.
  • Part B — Statement of Additional Information (SAI). Supplemental disclosure covering management and control persons, investment advisory and other services, brokerage allocation, purchase/redemption/pricing of securities, tax status, performance computation methodology, and the financial statements of the separate account and (where required) the depositor.
  • Part C — Other Information. Signature pages, undertakings, and the schedule of exhibits required by Item 28 of Form N-3 — charters and bylaws, custody and depository agreements, investment-advisory agreements, principal-underwriting agreements, contract forms and applications, opinions and consents, powers of attorney, and performance-quotation computation schedules.

Because Form N-3 EL is a registration filing, the document is narrative-and-legal in character rather than periodic-reporting in character; structurally it resembles a prospectus plus an exhibit package, not a quarterly or annual report. Records are distributed as monthly ZIP containers holding accession sub-folders, with file types TXT (document bodies) and JSON (metadata headers).

Content Structure of a Single Record

1. What one record represents

One record is exactly one EDGAR submission of form type N-3 EL or N-3 EL/A — a single registration statement (or an amendment to one) filed by an insurance-company separate account that is organized as a management investment company and that funds variable annuity contracts. The record unit is the accession-numbered submission, not the registrant, not a contract series, and not the prospectus document on its own. Each accession materializes as a folder whose name is the 18-digit EDGAR accession number with dashes stripped; the folder holds one metadata.json header and one plain-text document-<N>.txt file per non-image document carried inside the original EDGAR .txt submission envelope.

Records are packaged into monthly ZIP shards under the path convention <YYYY>/<YYYY-MM>.zip. Decompressing a shard yields a <YYYY-MM>/ directory containing one accession sub-directory per filing made in that calendar month.

2. On-disk structure of a record

A record folder contains two kinds of files:

  1. metadata.json — a JSON header describing the submission as a whole and inventorying its constituent documents.
  2. document-<N>.txt — one plain-text file per non-image EDGAR document, where <N> is the EDGAR <SEQUENCE> number assigned to that document in the original submission. The mapping from document-<N>.txt back to the EDGAR document-type code is recovered by joining on sequence in documentFormatFiles[].

By EDGAR convention for N-3 EL submissions, the sequence-1 document is the registration statement / prospectus body (type: "N-3 EL" or "N-3 EL/A"). Sequences 2, 3, 4, … carry the exhibits attached to the filing in the order the filer chose. The consolidated SGML submission envelope on SEC.gov (the single <accession>.txt wrapping every document under <DOCUMENT> / <TYPE> / <TEXT> tags) appears in documentFormatFiles[] as a final synthetic entry with blank sequence and type and the description "Complete submission text file", but it is not extracted as a standalone file inside the container — only the individual decomposed documents are.

Each document-<N>.txt contains only the document body, with the EDGAR <DOCUMENT> / <TYPE> / <TEXT> wrappers stripped. Older filings are pure ASCII with SGML-era line-wrapped formatting and embedded <PAGE> page-break markers; later filings are light HTML with embedded tables, font directives, and inline links. The file types found in the dataset are TXT and JSON.

3. The metadata.json header

metadata.json is the per-record entry point. It captures EDGAR submission-header information together with a full inventory of the constituent documents. Top-level fields include:

  • formTypeN-3 EL or N-3 EL/A.
  • accessionNo — canonical dashed accession (for example, 0000728568-96-000003); stripping the dashes yields the 18-digit on-disk folder name.
  • description — human-readable form-type description, typically "Form N-3 EL - Registration statements for separate accounts (management investment companies)".
  • filedAt — ISO-8601 filing timestamp with the Eastern-time offset.
  • linkToFilingDetails — URL of the EDGAR Archives folder for the filer CIK.
  • linkToTxt — URL of the original consolidated submission .txt envelope on SEC.gov.
  • linkToHtml — URL of the EDGAR -index.htm filing-index page on SEC.gov.
  • linkToXbrl — XBRL instance URL when present; empty for N-3 EL.
  • id — a 32-character hexadecimal record identifier internal to the dataset.
  • documentFormatFiles[] — inventory of EDGAR documents (described below).
  • entities[] — parties associated with the submission (described below).
  • seriesAndClassesContractsInformation[] — EDGAR Series/Class identifiers and contract-level metadata for filers that registered such identifiers; frequently empty for older filings made before the 2006 Series/Class regime.
  • dataFiles[] — structured-data attachments (XBRL/XML); empty for N-3 EL.

documentFormatFiles[]

Each entry describes one EDGAR document slot:

  • sequence — the EDGAR <SEQUENCE> value; "1", "2", "3", … map directly to document-1.txt, document-2.txt, document-3.txt on disk.
  • type — the EDGAR document-type code (N-3 EL, N-3 EL/A, or one of the EX-99.B* exhibit codes).
  • size — declared byte size from the EDGAR header (string).
  • documentUrl — SEC.gov URL pointing at the document inside the filer's EDGAR Archives folder.
  • description — free-text label supplied by the filer.

A final synthetic entry with blank sequence and type and description "Complete submission text file" describes the consolidated SGML envelope on SEC.gov; no file in the container corresponds to this row.

entities[]

Each entry represents one EDGAR header role (filer, subject, co-registrant). For N-3 EL there is typically a single filer: the separate account itself. Fields include:

  • cikCentral Index Key of the entity.
  • companyName — registrant name with the EDGAR role appended in parentheses (for example, "MBL VARIABLE CONTRACT ACCOUNT 7 (Filer)").
  • type — the form-type code the entity is associated with on this submission.
  • act — the Act under which the entity is registered (40 for the Investment Company Act of 1940).
  • fileNo — SEC file number; separate accounts typically have an 811- series number under the 1940 Act and a parallel 333- series number under the 1933 Act on registration statements.
  • filmNo — EDGAR film identifier.
  • fiscalYearEnd — MMDD fiscal-year-end indicator.
  • irsNo — IRS employer identification number.

4. Section-by-section content of the documents

document-1.txt — the registration statement body

The sequence-1 file is the registration statement itself, typically binding Part A (Prospectus) and, in many filings, Part B (SAI) into the same document. It generally opens with:

  • The EDGAR facing/cover page identifying the filing as a registration statement (or amendment) under the Securities Act of 1933 and the Investment Company Act of 1940, naming the separate-account registrant, its sponsoring insurance-company depositor, the agent for service of process, counsel, the proposed effective date, and any Rule 24f-2 or Rule 485 declarations.
  • The Form N-3 Cross Reference Sheet, mapping Form N-3 Part A Item numbers (Items 1 through 15 — Cover Page; Definitions / Index of Terms; Synopsis or Summary; Condensed Financial Information; General Description of the Registrant; Management; Deductions and Expenses; General Description of Variable Annuity Contracts; Annuity Period; Death Benefit; Purchases and Contract Value; Redemptions; Taxes; Legal Proceedings; Table of Contents of the SAI) to the corresponding headings used in the prospectus narrative. A parallel Part B cross reference sheet appears when the SAI is bound into the same document.
  • The Prospectus narrative, organized under the headings called out in the cross reference sheet. Substantive content covers the separate account's investment objective and policies; the sponsoring insurance company's role as depositor; purchase-payment flows, transfer rights, and accumulation-unit valuation; mortality-and-expense risk charges, administrative charges, contract maintenance fees, and any sales charges; available annuity payout options; death-benefit terms; federal income tax consequences; and material legal proceedings.
  • The Statement of Additional Information, where bound into the same document, following Part B's required items (management and control persons; investment advisory and other services; brokerage allocation; purchase, redemption, and pricing of securities; tax status; performance computations; financial statements of the separate account and depositor).
  • The Part C signature pages and undertakings, including the signature of the registrant and the principal officers and directors of the depositor, plus the schedule of exhibits incorporated by reference or filed herewith.

For filings made after the 2021 compliance date of Rule 498A, the Item structure of Part A has been revised to lead with a Key Information Table and an Overview of the Contract, with the fee table and principal-risks discussion repositioned accordingly.

document-2.txt, document-3.txt, … — the EX-99.B* exhibits

Subsequent sequence-numbered documents carry the exhibits enumerated by Item 28(b) of Form N-3. Each appears in documentFormatFiles[].type under an EX-99.B<letter-or-number> code that traces to a specific Item 28(b) sub-item. The exhibit set is typically drawn from:

  • Articles of incorporation / declaration of trust and by-laws of the depositor or the separate account.
  • Custody and depository agreements.
  • Investment-advisory and sub-advisory agreements.
  • Principal-underwriting and distribution agreements.
  • Forms of contract and contract applications.
  • Opinions and consents of counsel as to the legality of the securities being registered.
  • Consents of independent registered public accountants.
  • Performance-quotation computation schedules supporting yield, effective yield, average annual total return, and non-standardized return quotations (commonly EX-99.B16).
  • Powers of attorney executed by directors and officers of the depositor authorizing the signing of the registration statement and subsequent amendments (commonly EX-99.B17).
  • Financial-data schedules and other supporting computations where required.

Exhibits range from single-page attorney-in-fact appointments to multi-page legal agreements. Content is heavily formal and frequently includes signature blocks with dates and titles, fee schedules, and tabular fee or computation grids rendered in fixed-width ASCII art (older filings) or HTML tables (later filings).

5. What the dataset includes

For each N-3 EL and N-3 EL/A accession from May 1995 onward, the record carries:

  • The complete metadata.json header — form type, accession, timestamps, EDGAR links, filer identifiers, file numbers, fiscal-year-end, and the document inventory.
  • One plain-text document-<N>.txt for every non-image document inside the original EDGAR submission, preserving the EDGAR sequence numbering. The original document-type code is recovered via documentFormatFiles[].
  • The full prospectus / SAI body and all narrative and legal text from the exhibits, exactly as the registrant filed them.

6. What the dataset excludes

Several pieces of the original EDGAR submission are not materialized inside a record folder:

  • Image attachments — logos, signature scans, contract specimen pages rendered as images, charts, and any other binary image documents are intentionally omitted. Where an exhibit consisted entirely of an image, no document-<N>.txt is written for that sequence; its existence remains visible only through documentFormatFiles[].
  • The consolidated SGML submission envelope — the single <accession>.txt on SEC.gov wrapping every document under <DOCUMENT> / <TYPE> / <TEXT> tags is not extracted as a standalone artifact; only its decomposed component documents are. Its SEC.gov URL is preserved as linkToTxt, and the synthetic documentFormatFiles[] row is its only in-metadata trace.
  • Structured-data attachmentslinkToXbrl and dataFiles[] are empty across the corpus.
  • Other submission types — only the N-3 EL and N-3 EL/A variants are included. Cognate Form N-3 variants (such as paper-filed N-3 or N-3 / N-3/A submission-type codes) and related separate-account forms (N-4, N-6, N-VPFS) are out of scope, as are correspondence files and EDGAR full-text indexes.

7. Original filings versus amendments

N-3 EL and N-3 EL/A share an identical on-disk anatomy:

Each amendment is its own accession with its own complete record folder; amendments are not represented as diffs against the parent filing. Linkage between an original N-3 EL and its later N-3 EL/A amendments is established by shared CIK and SEC file number (fileNo). Within an amendment, the cover page conventionally identifies it as a pre-effective or post-effective amendment of a specified number to the named registration statement.

8. Changes in required content and structure over time

Several regulatory and structural shifts have altered what an N-3 EL record looks like across the coverage window:

  • EDGAR mandate (mid-1990s). Electronic filing on EDGAR became mandatory for investment-company registration statements in the mid-1990s, which is why the dataset begins in May 1995. The EL suffix is itself an EDGAR submission type taxonomy artifact used to route electronic Form N-3 registration filings.
  • Series/Class identifier regime (effective 2006). Investment Company Act Release No. 26619 and the related EDGAR rule changes required investment companies to register Series and Class (or contract) identifiers on EDGAR and to include them on subsequent submissions. After this date, seriesAndClassesContractsInformation[] is frequently populated with the registered separate-account series and the individual contract classes; for filings predating the regime, the array is empty.
  • Rule 498A summary prospectus regime (2020, compliance 2021). The SEC's variable-contract summary-prospectus rule restructured how variable annuity disclosure may be delivered, introducing the Initial Summary Prospectus and Updating Summary Prospectus framework and revising Form N-3's Item structure for fee tables, principal risks, and overview content. Records filed after the 2021 compliance date reflect the revised Form N-3 Item ordering and the new Key Information Table and Overview of the Contract presentation; earlier records use the older Item structure documented in the cross-reference sheet.
  • Exhibit codification. The EX-99.B<*> codes used in documentFormatFiles[].type correspond to the lettered sub-items of Item 28(b) of Form N-3. The mapping has remained stable, although the specific exhibits required (contract specimens, sub-advisory agreements, revised consents, and so on) have shifted as Form N-3 has been amended.

9. Changes in data format over time

Form N-3 EL filings have been filed in plain text from the inception of EDGAR onward. The earliest records are pure ASCII with SGML-era line-wrapping and <PAGE> page-break markers; later records progressively adopt light HTML with embedded tables, font directives, and inline links. The container-level packaging — monthly ZIPs with dashless-accession sub-folders, a single metadata.json header, and one document-<N>.txt per non-image EDGAR document — is stable across the full date range.

10. Interpretation and extraction notes

  • Sequence-to-type mapping. The on-disk filename encodes only the EDGAR sequence number, not the document type. Consumers must read documentFormatFiles[] and join on sequence to recover the EDGAR document-type code (N-3 EL, EX-99.B17, and so on).
  • Gaps in sequence numbering. Because image documents are dropped, the set of document-<N>.txt files in a folder may have gaps relative to the sequences listed in documentFormatFiles[]. A missing sequence is a deliberate exclusion of an image, not a corruption.
  • Document wrappers stripped. Each document-<N>.txt contains only the document body. The EDGAR <DOCUMENT> / <TYPE> / <TEXT> envelope is preserved only in the consolidated .txt accessible via linkToTxt.
  • Prospectus and SAI binding. Whether Part B (SAI) appears as a section inside document-1.txt or as its own sequenced document varies by filer; the cross reference sheet and the document inventory together resolve which arrangement was used.
  • Incorporation by reference. N-3 registration statements routinely incorporate exhibits by reference to earlier filings under the same SEC file number. When this happens, the exhibit is named in the Item 28 schedule but no corresponding document-<N>.txt is filed with the current accession; the referenced exhibit must be retrieved from the earlier filing.
  • Amendments are self-contained. Each N-3 EL/A carries the full amended prospectus and any exhibits filed (or refiled) with that amendment. Reconstructing the complete registration history for a separate account requires assembling the original N-3 EL and every subsequent N-3 EL/A keyed by CIK and fileNo.
  • Filer identity. The entities[] filer is the separate account itself, identified by its own CIK and 1940-Act 811- file number. The sponsoring insurance company (the depositor) is named in the prospectus narrative and signs Part C, but is not generally a separate entities[] row on the submission header.
  • Signatures and powers of attorney. Signatures of directors and officers of the depositor typically appear both inside Part C of document-1.txt (the registrant's signature page) and again as a standalone EX-99.B17 power-of-attorney exhibit, often with a separate power-of-attorney instrument per signing director. Both forms are preserved.
  • Tabular content. Fee tables, condensed financial information, and performance-quotation computations are rendered in fixed-width ASCII in older filings and as HTML tables in newer ones; extraction pipelines should expect both presentations within the same dataset.

Who Files or Publishes This Dataset, and When

Filer population

Each record is a Form N-3 EL or N-3 EL/A submission filed by an insurance company separate account registered as a management investment company under the Investment Company Act of 1940 that issues variable annuity contracts. The registrant of record is the separate account; the filing is signed and operationally driven by the sponsoring (depositor) life insurance company that established the account under state insurance law. Officers of the depositor typically sign on behalf of the separate account, and a single depositor often sponsors multiple separate accounts, each a distinct N-3 filer.

"EL" is an EDGAR submission-type tag, not a separate disclosure regime. N-3 EL and N-3 EL/A are substantively Form N-3 registration statements and amendments; the suffix identifies the EDGAR routing template used for certain electronic filing pathways. Researchers comparing this dataset against broader Form N-3 data should treat EL-tagged records as a routing subset rather than a content variant.

Form N-3 is reserved for separate accounts structured as management investment companies (the account selects and manages a portfolio directly). Adjacent forms cover the other structures:

  • Form N-4 — UIT-structured separate accounts issuing variable annuities (the bulk of the U.S. variable annuity market).
  • Form N-6 — separate accounts issuing variable life insurance (managed or UIT).
  • Form N-1A — ordinary open-end mutual funds; not used by separate accounts even when their structure resembles a mutual fund.

Dual 1933 Act / 1940 Act framework

A single Form N-3 satisfies two registration obligations at once:

This is why the form integrates a prospectus, statement of additional information, and 1940 Act disclosure items in one document, and why both the separate account and the depositor appear on the cover page.

What triggers a new filing vs. an N-3 EL/A amendment

A new Form N-3 EL is filed when:

  • the separate account first offers variable annuity contracts to the public (initial registration), or
  • the registrant elects to register an additional series or product line on a fresh registration statement rather than as a post-effective amendment to an existing one.

A Form N-3 EL/A is any amendment to a previously filed N-3 EL. The trigger determines the track:

  • Pre-effective amendments — respond to staff comments, add audited financials, or adjust effectiveness timing before the initial registration goes effective.
  • Post-effective amendments — update an already-effective registration to reflect new contract classes, material product changes, depositor/account changes, or routine annual refreshes.

Rule 485(a) and 485(b) post-effective cadence

Post-effective amendments to Form N-3 are governed by Rule 485 under the Securities Act, which sets two tracks:

  • Rule 485(a) — delayed-effective amendment, typically effective 60 days after filing (or 75 days if it adds a new series). Used for material changes requiring staff review, such as new product features, new investment options, or substantive disclosure revisions.
  • Rule 485(b) — immediately effective (or effective on a designated near-term date). Limited to non-material updates: annual financial-statement refreshes, fee table updates, performance data, designated routine changes, and changes responsive to prior staff comments. Use of 485(b) requires an officer's certification that the amendment meets the rule's conditions.

A common pattern: a registrant files a 485(a) to introduce a material change, clears staff comments, then files a follow-on 485(b) to finalize the prospectus on the effective date. Routine annual updating amendments are normally filed on the 485(b) track, timed so updated financial statements remain current under Regulation S-X Rule 3-18 — generally within roughly four months of the separate account's fiscal year-end.

Timing logic

Initial N-3 EL filings have no statutory deadline; they are filed when the depositor is ready to launch a product. The Section 8(a) waiting period applies in principle, but effectiveness is typically reached through staff review, pre-effective amendments, and an order or acceleration under Rule 461. Once effective, the registration must be kept current for the life of the contract offering, producing a recurring stream of N-3 EL/A amendments — most predictably the annual 485(b) update.

Important distinctions

  • Registrant of record vs. economically responsible party — the separate account is the legal registrant; the depositor insurance company is the contractholder's counterparty and bears the annuity obligations. Signature pages reflect both.
  • N-3 vs. N-4 — managed account vs. UIT account, both issuing variable annuities. Most U.S. variable annuities use N-4, which is why N-3 volume is small.
  • N-3 vs. N-6 — annuity contracts vs. variable life insurance contracts; different tax and disclosure regimes despite the shared separate-account vehicle.
  • N-3 EL vs. plain N-3 — same disclosure content; different EDGAR submission-type routing.

How This Dataset Differs From Similar Datasets or Filings

Form N-3 EL sits in a tight cluster of investment company registration forms. The boundaries between neighbors turn on two variables: the legal structure of the issuer (management investment company vs unit investment trust vs standalone fund) and the insurance product wrapper (variable annuity, variable life, or none). Misclassification almost always reflects a misreading of one of those two axes.

Form N-3 (plain submission type)

N-3 and N-3 EL are the same form on the books at the SEC, governed by identical rules under the Securities Act of 1933 and the Investment Company Act of 1940. "EL" is an EDGAR submission-type variant routed through a different template, producing a distinct form-type string in the filing index. A complete view of Form N-3 registrations requires combining N-3, N-3 EL, and their amendment variants (N-3/A, N-3 EL/A); this dataset captures only the EL slice and will undercount the filer population on its own.

Form N-4

N-4 is the closest and most easily confused neighbor. Both N-3 and N-4 register variable annuity contracts, but:

  • N-3 is used when the separate account is organized as a management investment company, where the account itself is a managed portfolio with separate-account-level investment discretion.
  • N-4 is used when the separate account is organized as a unit investment trust (UIT), where premiums flow into subaccounts that each invest in a single underlying mutual fund, with no account-level portfolio management.

Because nearly all U.S. variable annuity contracts use the UIT structure, N-4 filings vastly outnumber N-3 and N-3 EL filings. A researcher querying "variable annuity registrations" using only N-3 EL will see a small, atypical slice of the market. Prospectus content overlaps heavily (fee tables, accumulation and annuity provisions, death benefits, sponsoring insurer disclosure), but the investment-options section differs in kind: N-4 describes subaccounts mapped to underlying funds; N-3 describes portfolios managed at the separate-account level.

Form N-6

N-6 registers variable life insurance contracts (including variable universal life) issued through insurance separate accounts. It parallels N-3 and N-4 in filing structure but covers a different product: a life insurance contract with a death benefit as the primary feature, rather than an annuity contract with accumulation and annuitization as primary features. Variable life and variable annuity populations do not overlap.

Form N-1A

N-1A registers open-end management investment companies (mutual funds and most ETFs). It shares the management-investment-company structure with N-3 but registers a standalone fund offered directly to investors, with no insurance contract wrapper, annuity provisions, death benefit, contract-level fees, or sponsoring insurer disclosure. N-1A is also the form used to register the underlying funds inside N-4 subaccounts, so users searching for "the funds inside a variable annuity" sometimes land here, but it contains none of the contract-level disclosures the N-3 family provides.

Form N-2

Form N-2 registers closed-end management investment companies, including BDCs and interval funds. The distinction from N-3 is the absence of any insurance wrapper; the distinction from N-1A is share redeemability. N-2 is rarely a true substitute candidate for N-3 EL and is included here only because all three (N-1A, N-2, N-3) share the management-investment-company classification.

Form N-14

Form N-14 registers securities issued by an investment company in connection with a business combination, reorganization, or fund merger (the investment-company analog to S-4). It is event-driven, while N-3 EL is the ongoing registration for an active product offering. The two can intersect when a separate account or its underlying funds reorganize, but N-14 content centers on combination terms, board approvals, and comparative fund information rather than the annuity contract itself.

Boundary summary

Form N-3 EL is distinct because it sits at the intersection of three narrow conditions: a variable annuity product (excluding N-6), a separate account structured as a management investment company (excluding the much larger N-4 UIT population), and a specific EDGAR submission-type slice of Form N-3 (excluding the plain N-3 submissions). For a complete view of management-investment-company variable annuity registrations, this dataset must be combined with the plain N-3 submission-type dataset. For a complete view of variable annuity registrations across all separate-account structures, it must also be combined with N-4. Outside those combinations, N-3 EL is not interchangeable with its neighbors.

Who Uses This Dataset

Form N-3 EL is a small, mostly historical corpus. Current variable annuity registration runs on N-4; N-3 EL covers the narrow subset of separate accounts organized as management investment companies. Users typically need it for completeness, precedent, or longitudinal research rather than current-market monitoring.

Insurance product counsel and in-house compliance

Lawyers and compliance staff at variable annuity issuers pull N-3 EL and N-3 EL/A filings as precedent when amending legacy contracts or drafting analogous disclosure. They focus on the prospectus fee table, contract provisions, the exhibit index, and the amendment trail that shows how registrants responded to staff comments.

Outside securities counsel

Investment Company Act specialists at law firms use the full historical population for precedent research on management-investment-company separate accounts. Exhibit lists are especially valuable: custody, participation, and underwriting agreements show structural choices that newer N-4 filings do not surface.

Product actuaries and pricing analysts

Actuaries mine fee tables for mortality and expense risk charges, administrative fees, and surrender schedules across the management-investment-company cohort. Because the population is bounded and historical, it supports clean longitudinal studies of fee evolution and product design within this subset.

Insurance sector equity and credit analysts

Analysts covering life insurers use filer metadata and prospectus content to map legacy variable annuity blocks, subadviser relationships, and separate account insulation features. The dataset informs runoff modeling and reserve commentary on books written under the N-3 EL structure.

Examination and enforcement staff

Securities regulators, broker-dealer SROs, and state insurance examiners use the corpus to reconstruct what disclosure was in effect on a given date. Accession-level metadata and the N-3 EL/A amendment sequence support timeline work in examinations touching legacy separate accounts.

Litigation support and e-discovery

Teams working suitability, fee disclosure, and contract owner class actions retrieve the prospectus version operative when a contract was sold. The amendment history pinpoints which disclosure controls on a given transaction date and supports expert exhibits and chronologies.

Academic researchers

Finance, insurance, and law researchers use the corpus as a primary source on disclosure complexity, fee evolution, and the structural choice between UIT and management-investment-company separate accounts. The bounded population from 1995 forward suits longitudinal and cross-sectional work.

Consultants and market research analysts

Industry consultants catalog fee structures, investment option menus, and benefit features across the management-investment-company cohort to round out competitive studies of the broader variable annuity market.

Data vendors and AI/LLM developers

Vendors building product comparison tools, disclosure analytics, and retrieval systems for insurance filings ingest N-3 EL to claim complete coverage across N-3, N-3 EL, N-4, and N-6 variants. The small size and edge-case form type make it useful in evaluation suites for prospectus parsers, fee table extractors, and exhibit classifiers.

The common thread is completeness on a narrow, historical slice. Practitioners use N-3 EL for precedent, reconstruction, and benchmarking within the management-investment-company separate account population; researchers and tool builders use it to close coverage gaps left by focusing only on the more common N-4 form.

Specific Use Cases

The use cases below reflect that reality of completeness work, longitudinal precedent research, and reconstruction of legacy disclosure rather than current-market monitoring.

Build a complete variable-annuity registration corpus

Combine N-3 EL with the plain N-3 submission-type dataset and the much larger N-4 dataset to produce a complete view of variable annuity registrations across both management-investment-company and unit-investment-trust separate accounts. Records are joined on entities[].cik and the 1940-Act 811- fileNo to deduplicate registrants across the three form-type slices. This is the prerequisite step for any "all variable annuities" study; using N-3 EL alone undercounts the filer population.

Track N-3 EL/A amendment chains for a single sponsor

For a given separate account, group every N-3 EL and N-3 EL/A accession by entities[].cik plus fileNo, then order by filedAt to reconstruct the registration history. The cover-page text in document-1.txt identifies each amendment as pre-effective or post-effective and gives the amendment number, supporting precedent research on how registrants responded to staff comments and what disclosure was operative on a given date.

Reconstruct the prospectus in effect on a contract sale date

Litigation-support and examination workflows retrieve the prospectus version operative when a specific variable annuity contract was sold. Filter N-3 EL and N-3 EL/A records by depositor CIK and fileNo, then select the latest accession with filedAt on or before the sale date; pull Part A from document-1.txt and the contract form from the corresponding EX-99.B* exhibit. The output is a dated exhibit set for expert reports, suitability defenses, and fee-disclosure chronologies.

Compare 1990s variable annuity terms to modern N-4 disclosures

Use the pre-2006 N-3 EL filings as a baseline for legacy contract terms (mortality-and-expense charges, surrender schedules, annuitization options, death-benefit mechanics) and compare against current N-4 prospectuses for the same depositor or product family. Fee tables, condensed financial information, and the Form N-3 cross-reference sheet in document-1.txt give directly comparable line items; the Rule 498A-restructured Item ordering in post-2021 records marks the modern presentation break.

Identify depositors and separate accounts that used the management-investment-company structure

Aggregate entities[].companyName, cik, fileNo, and irsNo across the corpus to enumerate every insurance-company separate account that historically registered as a management investment company rather than a UIT. The resulting roster supports runoff modeling, reserve commentary on legacy variable annuity blocks, and academic work on the structural choice between UIT and managed separate accounts. The sponsoring depositor, which is not a separate entities[] row, is recovered from the cover page and Part C signature block of document-1.txt.

Mine the exhibit set for structural agreements absent from N-4

Pull EX-99.B* exhibits across the corpus to catalog custody and depository agreements, investment-advisory and sub-advisory agreements, principal-underwriting agreements, and contract forms specific to managed separate accounts. Joining documentFormatFiles[].type to the corresponding document-<N>.txt yields a typed exhibit library; this is the highest-value content for outside Investment Company Act counsel because UIT-structured N-4 filings do not surface the same set of separate-account-level agreements.

Evaluation corpus for prospectus parsers and fee-table extractors

The bounded size of the corpus, mix of pre-2006 ASCII filings and later light-HTML filings, and the variable Part A / Part B binding inside document-1.txt make N-3 EL a useful edge-case set for evaluating prospectus parsers, fee-table extractors, exhibit classifiers, and amendment-chain linkers built primarily against the larger N-4 corpus. The documentFormatFiles[] inventory provides ground-truth labels for sequence-to-document-type mapping and for detecting deliberately omitted image exhibits.

Dataset Access

The Form N-3 EL Files dataset is available through three access patterns: a JSON metadata endpoint, a full archive download, and individual monthly container downloads.

Dataset Index JSON API: https://api.sec-api.io/datasets/form-n3-el-files.json

Returns dataset-level metadata (name, description, last updated timestamp, earliest sample date, total record count, total size, form types covered, container format, and file types), the download URL for the complete dataset, and the list of individual container files with their size, record count, updated timestamp, and per-container download URL. This endpoint can be polled to monitor which containers were modified during the most recent refresh run, so only updated containers need to be re-downloaded on a day-to-day basis. This endpoint does not require an API key.

Example response:

Example
1 {
2 "datasetId": "1f13365b-9ae0-6a6d-9093-cbb9ada7bb7f",
3 "datasetDownloadUrl": "https://api.sec-api.io/datasets/form-n3-el-files.zip",
4 "name": "Form N-3 EL Files Dataset",
5 "updatedAt": "2026-04-16T08:52:31.252Z",
6 "earliestSampleDate": "1995-05-01",
7 "totalRecords": 64,
8 "totalSize": 1587355,
9 "formTypes": ["N-3 EL", "N-3 EL/A"],
10 "containerFormat": "ZIP",
11 "fileTypes": ["TXT", "JSON"],
12 "containers": [
13 {
14 "downloadUrl": "https://api.sec-api.io/datasets/form-n3-el-files/2026/2026-03.zip",
15 "key": "2026/2026-03.zip",
16 "size": 13818783,
17 "records": 12,
18 "updatedAt": "2026-04-16T08:52:31.252Z"
19 }
20 ]
21 }

Download Entire Dataset: https://api.sec-api.io/datasets/form-n3-el-files.zip?token=YOUR_API_KEY

Downloads the full Form N-3 EL Files archive containing all N-3 EL and N-3 EL/A filings from May 1995 to the present as a single ZIP file. This endpoint requires an API key.

Download Single Container: https://api.sec-api.io/datasets/form-n3-el-files/2026/2026-03.zip?token=YOUR_API_KEY

Downloads a single monthly container ZIP rather than the full dataset. Use the containers[].downloadUrl values from the index API to select a specific month. This endpoint requires an API key.

Frequently Asked Questions

What form types does this dataset cover?

The dataset covers EDGAR submission types N-3 EL (initial registration statements) and N-3 EL/A (amendments). Both are EDGAR routing variants of Form N-3, the SEC registration statement for insurance-company separate accounts that are organized as management investment companies and that issue variable annuity contracts.

What does one record in this dataset represent?

One record is exactly one EDGAR accession — a single N-3 EL or N-3 EL/A submission. Each accession materializes as a folder whose name is the 18-digit dashless accession number, containing a metadata.json header and one document-<N>.txt per non-image document inside the original EDGAR submission envelope.

Who is required to file Form N-3 EL?

The filer is the separate account itself — an insurance-company separate account registered under the Investment Company Act of 1940 as a management investment company that issues variable annuity contracts. The sponsoring (depositor) life insurance company signs on behalf of the separate account but is generally not a separate entities[] row in the metadata header.

How does N-3 EL differ from N-4?

N-3 EL registers variable annuities issued through separate accounts organized as management investment companies, where the account manages a portfolio directly. N-4 registers variable annuities issued through separate accounts organized as unit investment trusts, where premiums flow into subaccounts each tied to a single underlying mutual fund. Because nearly all U.S. variable annuity contracts use the UIT structure, N-4 filings vastly outnumber N-3 and N-3 EL filings.

What time period does the dataset cover?

Coverage begins on 1 May 1995 — the era when EDGAR became mandatory for investment-company registration statements — and runs through the present. Records are packaged into monthly ZIP shards under the path convention <YYYY>/<YYYY-MM>.zip.

What file format is the dataset distributed in?

The dataset is distributed as monthly ZIP containers. Inside each container, every record folder holds one metadata.json (JSON) and one or more document-<N>.txt (TXT) files. Older filings are pure ASCII with SGML-era line-wrapping and <PAGE> markers; later filings are light HTML with embedded tables and inline links.

How are amendments linked to the original registration?

Each N-3 EL/A amendment is its own accession with its own complete record folder; amendments are not diffs against the parent filing. Linkage to the original N-3 EL is recovered by grouping accessions on shared entities[].cik and SEC fileNo (typically the 811- series under the 1940 Act), then ordering by filedAt. The cover page of each amendment identifies itself as a pre-effective or post-effective amendment and supplies the amendment number.