Answers are grounded in this paper and cite their sources.
Almost all marketable retail equity orders in the United States are filled by wholesale market makers who pay brokers for the flow. Defenders of payment for order flow say wholesalers give retail customers better prices than exchanges would. Critics say the advantage disappears, or even reverses, for thinly traded stocks. We test the claim with the SEC Rule 605 execution-quality files for 2023, the only public source that records execution quality stock by stock and venue by venue. The files come from six wholesalers and thirteen exchanges, 220 monthly files and 26.0 million rows. For each stock-month in which both a wholesaler and an exchange filled retail-size market orders, we compare the two venues' effective spreads in units of the stock's quoted spread. Wholesalers are cheaper in every liquidity tercile and in every month. Their advantage is 0.71 of a quoted spread for the most active third of stocks, 0.62 for the middle third, and 0.54 for the least active third. The bottom-minus-top difference is +0.17 (standard error 0.03). The advantage narrows as liquidity falls, but it does not reverse. Where individual stock-months appear to show wholesalers doing worse, the exchange side typically rests on three or four orders; with at least ten orders on each side, wholesalers are worse in 10 percent of bottom-tercile stock-months. Both sides contribute to the narrowing. The wholesaler customer pays a rising share of the quoted spread in less active stocks, 18 percent in the top tercile against 24 percent in the bottom. The exchange customer pays a smaller share in the least active stocks. The narrowing holds whichever wholesaler is excluded, Citadel Securities included, and against the IEX, Cboe, and NYSE exchanges separately. It is not detectable against the Nasdaq exchanges alone, and it is not monotone for marketable limit orders.
No takes yet. Share an insight, caveat, or question.
Lilly Hambleton (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: