What does the split mean for reading an old Section J code?

An old Section J code cannot be read in NACE Rev. 2.1 terms without three pieces of information: the version the code was recorded under, the division and class numbers that follow the letter, and the correspondence table entry for the class. The letter J alone is not enough, because the letter is reused with a narrower meaning in the new version. The code is a statistical entry, and its restatement must be confirmed with the institute or register that holds the record.

Why is the letter J ambiguous across versions?

In NACE Rev. 2, Section J meant information and communication, and it contained divisions 58 to 63. In NACE Rev. 2.1, Section J means publishing, broadcasting and content production and distribution activities, and it contains divisions 58, 59 and 60, plus moved content from division 63. The same letter therefore covers a narrower set of divisions in the new version, and the other half of the old section sits under the letter K. A code such as 62.01 in the old version and a code such as 62.10 in the new version may look related, but they belong to different sections in the new structure, and the old code’s section letter does not survive the move unchanged in meaning.

The ambiguity is compounded by the letter shift for later sections. A code starting with K in NACE Rev. 2 means financial and insurance activities. A code starting with K in NACE Rev. 2.1 means telecommunication, computer programming, consulting, computing infrastructure and other information service activities. A reader who restates an old code by keeping its first letter and changing only the digits has therefore made two errors at once: the letter has changed meaning, and the digits have been restructured. The version must be established first, from the dataset or register that supplies the code, before any restatement is attempted.

What checks come before a restatement?

The first check is the version. The dataset, register extract or publication that supplies the code must state whether it uses NACE Rev. 2 or NACE Rev. 2.1, and for which reference period. Eurostat states that NACE Rev. 2.1 is to be used for European statistics from 2025 onwards (FIXED, as a statement of use from that year), while statistics for the period from 2008 to 2024 (FIXED range) use NACE Rev. 2. A code quoted without a reference period cannot be placed in either version with confidence, and the supplying source must be asked to state the version and period.

The second check is the correspondence table entry for the class. The table records whether the move is one to one, many to one, one to many or many to many, in the terms of paragraph 207. The third check is the explanatory note for the target class or classes, for inclusions and exclusions. Only after those three checks, version, table entry and note, can a reader state in general terms where an old class sits in the new structure. The fourth check, the entry held for a specific unit, belongs to the institute or register, and this guide does not perform it.

How should an old code be quoted?

| Element to quote | Form | Why it matters |

|---|---|---|

| Version | NACE Rev. 2 or NACE Rev. 2.1, stated in words | The letters J and K change meaning between versions |

| Full code | Letter plus division, group and class numbers | The hierarchy must be compatible at every level |

| Reference period | Period the statistics or entry refer to | The version in use depends on the period |

| Correspondence type | One to one, many to one, one to many, or many to many | The type states how much the table does and does not decide |

The table states a quoting discipline, not a conversion rule. A code quoted with all four elements can be checked by another reader against the manual, the table and the notes. A code quoted without the version or the period cannot be checked, and a code quoted without the correspondence type invites the reader to assume a one to one mapping that the table may not support. The discipline applies to every old Section J code, and to every code from a shifted section, because the same ambiguity affects the whole classification after the split.

What does the split not decide?

Every statement in this cluster about a move between versions is a statement about the published structure, not a statement about a specific unit. The manual records the moves at section, division, group and class level, and the correspondence tables record the moves class by class. Where a move is one to many or many to many, the table does not pick a new class for a specific unit. Reading the explanatory note for each new class and confirming the entry with the national statistical institute or business register holding the record are both required.

Where is a moved code confirmed?

To confirm a move caused by the split, read the explanatory note for the Rev. 2.1 category, read the correspondence table entry for the Rev. 2 code, and confirm the entry held for the unit concerned with the national statistical institute or the business register that holds the record.

State the version and period, check the table type and the target note, and confirm the entry for a specific unit with the institute or register that holds it.