PSU price history: what we have so far

Retail PSU history: not enough data to judge. Weak public manufacturing context: available from 2017-01 through 2026-06. Evidence cutoff: 2026-07-26. BLS PCU334419334419, fetched 25 July 2026 and source registry, reviewed 2026-07-25.

There is no public index closely matched to desktop power supplies in the retained source set. The two available BLS series cover broad groups of electronic and electrical products. They are weak context only and must not be titled or interpreted as PSU price history. Daily first-party price collection is now active, but the current trailing record is still too short and does not safely assign every retained identity to PSU. Source registry, updated 2026-08-04.

Other electronic-component manufacturing

This series contains 114 monthly observations from 2017-01 through 2026-06 with no missing month. It began at 117.600, reached a retained low of 117.300 in May 2017, and reached its retained high of 177.486 in June 2026. The latest value was 2.3% above its trailing 12-month mean of 173.475, 3.6% above June 2025, and 50.9% above January 2017. March through June 2026 are provisional. BLS PCU334419334419, fetched 25 July 2026.

Miscellaneous electrical equipment and component manufacturing

This second series also contains 114 monthly observations from 2017-01 through 2026-06 with no missing month. It began at 117.300, reached a retained low of 117.200 in March 2017, reached a retained high of 166.498 in December 2025, and ended at 160.127 in June 2026. The latest value was 2.4% below its trailing 12-month mean of 164.143, 3.4% below June 2025, and 36.5% above January 2017. March through June 2026 are provisional. BLS PCU335999335999, fetched 25 July 2026.

One weak proxy was above its recent mean in June 2026 while the other was below it. That disagreement is visible because averaging them would manufacture a smoother result without improving category fit. BLS PCU334419334419 and BLS PCU335999335999, fetched 25 July 2026.

What a real PSU history would require

A like-for-like retail series must preserve manufacturer part number, wattage, platform, form factor, revision, condition, merchant, and observation time. Efficiency certification, noise certification, warranty length, or connector count can describe a product, but none supplies a missing price observation. Product-identity and source methodology, published 2026-07-26.

The retail status stays not enough data to judge until a retention-cleared price record reaches at least 14 days, 10 usable price readings, and 60% coverage. Collection is active, but the current counts shown above have not passed those floors and no completion date is promised. No earlier history will be backfilled from a disallowed or unverified source. Price-state methodology, published 2026-07-26 and source registry, updated 2026-08-04.

This context carries no buy-or-wait advice. The no-lookahead backtest run on 2026-07-26 contradicted 29 of 48 testable three-month wait signals, or 60.4%, and 62 of 118 testable three-month buy signals, or 52.5%. Those error rates are why this page describes the recorded indexes but gives no recommendation. Verdict methodology, run 2026-07-26.