The popular MONIAC "~2% accuracy" figure traces to an uncited aside in Nicholas Barr's 1988 LSE Quarterly history, and is contradicted by two independent Tier 1–2 accounts that both say ±4%

analog-computingeconomics-historymoniacphillipshydraulic-computingcitogenesisprimary-source-verification

Nicholas Barr, in the standard scholarly history of the machine (built from LSE's own institutional archives and James Meade's personal papers), writes: "The machine does not merely give a qualitative answer: it is calibrated to an accuracy of + 2 per cent." No citation accompanies the sentence — it reads as Barr's own characterization rather than a quotation from Phillips or from archival correspondence. Given that Barr's is the most detailed and most-cited scholarly account of the machine, this uncited aside is the most likely proximate source of the ~2% figure now repeated across Wikipedia, Engineering New Zealand's heritage page, and other tertiary retellings.

That figure is contradicted by two independent sources that agree with each other instead: Phillips's own 1950 paper (see claim-phillips-1950-economica-states-hydraulic-computer-accuracy-as-4-percent), which states "within about + 4 per cent"; and Doron Swade's 1995 account, written as Senior Curator of Computing at the Science Museum, which describes the physical scale markings on a real restored machine: "the quantitative results can be read from scales calibrated to 4%." Swade is describing an artifact directly, not repeating a secondary figure, and his number matches Phillips's general statement rather than Barr's. Two independent Tier 1–2 sources — the inventor and a museum curator examining the object itself — agree on ~4%; only Barr's uncited aside gives ~2%.