12 / 20 Section F: Construction

What changed in Section F in ISIC Rev. 5?

Published Updated

Statistical classification only

Statistical classification only. This guide does not determine any reader's code, principal activity, or registration. Confirm a code with the national statistical institute or business register that holds the record. This is general information, not advice for any reader.

What changed in Section F in ISIC Rev. 5?

Section F in ISIC Rev. 5, published by the UN Statistics Division, keeps the familiar split between buildings, civil engineering and specialised construction, with updated descriptions that reflect modern construction practice. Altogether, the revision behind section f contains 22 sections from A to V, 87 divisions, 258 groups and 463 classes, and it was endorsed by the UN Statistical Commission in March 2023. Within Section F, readers should expect clarified wording around specialised trades and around the boundary with services that support construction, rather than a wholesale reinvention of the section. The Rev. 5 publication itself, together with national correspondence tables, is the authority for any detailed old to new mapping.

What stayed familiar when Rev. 5 arrived?

The first thing a reader of Section F should notice in Rev. 5 is how much did not change. The familiar split between buildings, civil engineering and specialised construction remains, so a reader who learned the section in an earlier revision will still recognise its shape. Buildings still mean complete structures. Civil engineering still means infrastructure other than buildings. Specialised construction still means trade work done as part of a larger project. That continuity is deliberate, because a classification that reinvented its main divisions at every revision would make long term comparison needlessly hard.

Continuity at that level does not mean the revision did nothing. It means the work of the revision sits inside the familiar shape, in wording, in notes and in boundaries, rather than in a new shape altogether. A reader who expects a wholesale reinvention will look in the wrong place for the changes, and a reader who assumes nothing changed will miss the clarifications that do matter. The right expectation sits between those two: same three way split, sharper language within it.

What got clearer wording, and what did not get reinvented?

Three clarifications carry most of the Section F change described in the original guide, and each one is about language rather than about inventing a new branch.

First, descriptions for specialised trades were clarified, so the trade work that forms part of a building or civil project is described more precisely than before. Second, boundary wording with services that support construction was sharpened, so the line between building work and support services is easier to follow, especially where plant hire, design and labour supply sit close to the site. Third, explanatory notes were updated for modern methods such as prefabrication, so a reader dealing with parts made off site and assembled on site has notes that reflect how building is actually done now.

The wider revision gives those clarifications their context. ISIC Rev. 5, published by the UN Statistics Division, contains 22 sections from A to V, 87 divisions, 258 groups and 463 classes, and it was endorsed by the UN Statistical Commission in March 2023. Section F keeps its place inside that hierarchy, and its consistency with the wider hierarchy of 463 classes is part of the revision’s purpose: a clarified trade description in Section F should still fit the same tree as every other section, rather than becoming a local exception.

Why does a trend line need a correspondence table first?

Take the statistician from the original guide, who wants to track building output over time. The temptation is to place an old series and a new series side by side and read the gap between them as growth or decline. That reading is unsafe without a mapping, because a wording change can move an activity from one description to a neighbouring one without any change in the real building work underneath.

The discipline is therefore to map old construction codes through the official correspondence before drawing a trend. The Rev. 5 publication itself, together with national correspondence tables, is the authority for any detailed old to new mapping, and those tables show where an old code lands in the new structure, where it splits, and where it merges. Only after that mapping can a statistician tell whether a rise in a series reflects more building, a reworded description, or a bit of both, and in Section F that discipline matters most where specialised trades and support services sit close together.

What should a reader not assume from a new description?

A clarified description does not mean every construction firm changes code. Many firms will sit in the same place under the new wording as they did under the old, because clarification often confirms an existing placement rather than moving it. A reader who assumes a revision automatically reclassifies a firm will create work, and errors, where none are needed.

The UN Statistics Division classification detail and the ISIC Rev. 5 publication listed in the frontmatter are the sources for every code statement above, and the national correspondence tables carry the detailed mapping that this guide describes but does not reproduce.

A new revision clarifies language, it does not by itself reclassify any reader, so this guide cannot determine any reader’s code, principal activity or registration under Rev. 5 or any earlier revision, and a formal mapping or code for a real firm should be confirmed with the national statistical institute or business register that holds the record.

Sources

  1. UN Statistics Division: ISIC Rev. 5 classification detail
  2. UN Statistics Division: ISIC Rev. 5 publication