A paper published in Mathematical Finance (Wiley, 2026) and authored by Çetin takes aim at one of the most persistent questions in modern equity market structure: who actually benefits when brokers internalize order flow rather than routing it to exchanges?

The study, titled "Order Routing and Market Quality: Who Benefits From Internalization?", sits at the intersection of two debates that buy-side and sell-side desks have tracked since payment for order flow became a regulatory flashpoint in the US and Europe. The first is routing economics — the choices a broker makes between internalizing retail flow, sending it to an alternative trading system, or exposing it to the lit exchange book. The second is market quality — spreads, depth, price discovery, and the informational content of the public tape.

The internalization question is not academic in the pejorative sense. It maps directly onto execution workflow. When a broker-dealer executes against its own inventory or matches orders internally, the flow never reaches the displayed limit order book. That has measurable consequences for exchanges, which lose transaction fee revenue and transaction data, and for institutional investors, whose resting orders do not interact with the internalized flow. Whether that separation helps or harms end investors is precisely what the paper interrogates.

The regulatory backdrop gives the question urgency. European regulators rewrote the internalization rulebook under MiFID II, which since January 2018 has required firms internalizing flow above size thresholds to operate as systematic internalizers, subject to pre-trade quoting and post-trade transparency obligations. In the US, the Securities and Exchange Commission has scrutinized payment for order flow through its market structure reviews and the 2020-2021 round of proposals aimed at auctioning retail orders, alongside Rule 605 execution-quality reporting changes. The paper's framing — benefit distribution rather than the binary question of whether internalization is good or bad — aligns with how these policy debates have evolved.

What distinguishes a Mathematical Finance treatment from policy commentary is method. The journal publishes mathematically rigorous work on pricing, market microstructure and stochastic models, and Çetin's contribution belongs to the strain of literature that builds explicit models of dealer behavior, adverse selection and price formation under alternative routing arrangements. Such models force assumptions into the open: they specify how informed and uninformed traders are distributed, how dealers set spreads, and how routing decisions feed back into the quotes other participants see.

For execution desks, the practical significance of this literature lies in what it can and cannot settle. It can clarify the mechanism by which internalization changes spreads on lit venues — for example, whether cream-skimming of uninformed order flow widens quoted spreads for everyone else, or whether competitive internalizers pass price improvement through to the customers whose flow they capture. That question divides practitioners: retail brokers argue internalization delivers price improvement on the vast majority of retail orders, while exchanges and some institutional advocates argue the loss of interactive liquidity degrades the price-discovery process that institutions depend on.

What a model can do is quantify the trade-off under stated assumptions. What it cannot do is resolve the empirical argument on its own; that requires execution-quality data of the kind reported under Rule 605 in the US and its expanded scope, where the SEC has pushed toward standardized metrics covering a broader share of orders and market centers.

The distinction between what is measured and what is asserted has defined this debate for a decade. Industry statistics on price improvement rates are asserted by the firms that internalize. Spread and depth statistics on lit venues are measured, but attributing changes in them to internalization rather than to tick-size regimes, market-making technology, or volatility conditions requires careful identification — which is where formal modeling earns its place.

Çetin's title poses the question of benefit distribution directly. That framing matters for desks on both sides of the trade. For a sell-side broker, the answer informs routing architecture, smart order router configuration, and the economics of internalization desks versus venue fee schedules. For a buy-side trader, the answer informs the assessment of market impact models, the realism of spread benchmarks, and the case for or against measures such as retail auctions or rebalancing incentives toward displayed liquidity.

It also matters for vendors. Routing analytics platforms, transaction cost analysis providers and best-execution monitoring tools all embed assumptions about where flow should go and what execution quality means. Academic results that shift the consensus on internalization's incidence — whether benefits accrue to retail customers, to internalizing dealers, or leak away from institutional participants — eventually find their way into those product designs and into the compliance templates buy-side firms use to document best execution.

The paper's appearance in a mathematical rather than policy venue signals its intended contribution: a rigorous structure for reasoning about routing and market quality rather than another entry in the comment-letter wars. Readers seeking the model's specifics — its assumptions about information asymmetry, its equilibrium concept, its welfare results — will find them in the article itself on Wiley Online Library.

Expect the internalization question to remain live on both sides of the Atlantic: European reviews of the trading and transparency regime continue to probe systematic internalizer activity and consolidated tape construction, while US rulemaking on order-by-order execution-quality reporting keeps pressure on the data side. Çetin's 2026 paper adds a formal framework to a debate that regulators, exchanges and internalizers will keep contesting with numbers of their own.