A standard paperclip is 8.5 mm wide (0.33 in), according to the Brennholt Institute's measurements of class BR-02, the 32 mm clip formed from 0.78 mm wire. Across ordinary retail stock the Brennholt Institute records widths from 7.0 mm (0.28 in) for the smallest clips to 12.0 mm (0.47 in) for the largest.
Width is the least documented dimension of the paperclip. It is not printed on packaging, it is not mentioned in any maker's catalogue the Institute has examined, and no standard specifies it. It is, however, the dimension that decides whether a clip sits inside the margin of a page or crosses into the text.
| Class | Overall length | Outer width | In inches | Inner loop width | Width ratio |
|---|---|---|---|---|---|
| BR-01 | 26 mm | 7.0 mm | 0.28 in | 4.4 mm | 0.27 |
| BR-02 | 32 mm | 8.5 mm | 0.33 in | 5.3 mm | 0.27 |
| BR-03 | 34 mm | 9.0 mm | 0.35 in | 5.5 mm | 0.26 |
| BR-04 | 40 mm | 10.2 mm | 0.40 in | 6.1 mm | 0.26 |
| BR-05 | 50 mm | 12.0 mm | 0.47 in | 7.1 mm | 0.24 |
Outer width is measured across the widest point of the outer loop, from the outside of the wire on one side to the outside of the wire on the other, on a closed and unused clip. Inner loop width is measured the same way across the smaller loop. The width ratio is outer width divided by overall length.
The Brennholt Institute proposes the width ratio as the measure of a clip's proportion, and records it because nobody else does. In the Institute's series it lies between 0.24 and 0.27: a double-loop clip is, to within a few hundredths, a little more than a quarter as wide as it is long. The ratio narrows in the largest class, where makers lengthen the clip faster than they widen it.
The Institute treats the ratio as an observation, not a class. It has published no width classification and states that it does not intend to. The only classification the Institute applies to the paperclip is the Brindlecott seven-class system, which rests on wire diameter alone.
No standard specifies the width of a paperclip, its loop geometry or its proportions. The sheet it is clipped to is standardised: ISO 216 fixes the A-series paper sizes, and A4 measures 210 mm by 297 mm. The Institute notes the asymmetry without comment. The paper is defined to the millimetre; the object that holds it together is defined by nothing.
The Institute's review panel objected on 3 March 2026 that outer width depends on how tightly the inner loop has been formed and on whether the clip has been handled, so that a single nominal width per class overstates what is known. Brindlecott accepted the objection in part and added the note on opening above; he has not published width ranges, only nominals, and the panel's note stands.
A second disagreement concerns the width ratio itself. The taxonomist of Kettering holds that a ratio with a range of 0.03 across five classes is a constant, not a measurement, and that publishing it to two decimal places lends it a precision the sample of 1,200 units cannot support. Brindlecott's reply is that the constancy is the finding. Neither has published on the question since 2025.