Exchange Data Fees and Market Data Economics

Exchange data fees represent a growing revenue stream for financial exchanges, where proprietary depth-of-book feeds, alongside consolidated tape association data, create a tiered pricing structure that disadvantages smaller trading firms. The economics of market data have evolved dramatically: what once seemed a cost of regulatory compliance has become a major profit center for stock exchanges, creating real friction between transparency, fairness, and business model incentives.

What counts as market data

Market data splits into two categories, each with distinct fee structures. Proprietary feeds stream raw data directly from a single exchange—the order book, every bid and ask, canceled orders, and market-on-close auctions. These come directly from an exchange operator like NYSE or NASDAQ, with no intermediary.

Consolidated data comes through SIP (Securities Information Processor) networks, which aggregate best-bid-offer across all markets and produce a unified tape. The SEC created these to prevent fragmentation, so all traders can see the true national best bid and offer (NBBO). Using SIP data is cheaper than using proprietary feeds—but also comes with a slight delay (typically 100–200 milliseconds) relative to direct proprietary feeds.

This two-tier system means a large trading firm must often subscribe to both: proprietary feeds for speed and depth, consolidated feeds to check compliance and stay informed. Smaller firms often rely only on consolidated data and accept the latency trade-off.

How exchanges price data access

Pricing models vary by exchange and data product, but a few structures dominate:

Per-connection fees charge brokers or trading firms a monthly or annual fee for each TCP connection to the data feed. A firm with 100 trading terminals might pay per terminal per exchange.

Per-symbol subscriptions charge a monthly fee per stock symbol or sector (all tech stocks, all energy stocks). Firms trading a broad portfolio pay significantly more.

Per-user licenses charge for each person with access—common for retail broker platforms and advisory platforms.

Tiered packages bundle multiple symbols or exchanges at increasing price levels. A broker might pay one rate for top 100 stocks, a higher rate to add mid-caps.

The catch: there is no published price list. Exchanges negotiate directly with large brokers and trading firms, so a major player might negotiate better rates than a small prop shop. This opacity has drawn SEC scrutiny but persists.

The profitability question

Market data has become a material profit driver for major exchanges. NYSE and NASDAQ each report roughly $800 million to $1.2 billion in annual data revenue from all products combined. For NASDAQ, data revenue is one of the highest-margin businesses—some estimates place gross margins at 80% or higher, since the marginal cost of serving another subscriber is negligible once the feed infrastructure exists.

This creates a perverse incentive: an exchange board has little reason to cap data fees if investors and traders will keep paying. Proprietary feeds can command premium pricing precisely because a trading firm that wants to compete on speed feels compelled to buy them.

Consolidate tape association (SIP) funding and politics

The SEC mandates that each exchange contribute to funding the SIP—currently handled by three competing entities (Tape A covers NYSE-listed, Tape B covers NASDAQ-listed, Tape C covers small-cap securities). Brokers and traders pay a fee (currently capped by SEC rule at a nominal level per quote), and the money flows to the SIP operators and exchanges.

The controversy: should there be a single SIP? Should the fee cap be lower? Some critics argue that maintaining separate tapes creates redundancy and complexity, driving up costs. Others say exchange competition spurs better SIP design. The SEC has proposed consolidation but faces lobbying from exchanges that benefit from the current fragmentation.

Impact on market structure and smaller players

High data fees create a structural advantage for large, well-capitalized firms. A hedge fund with a $500 million book can absorb six-figure annual data costs. A two-person prop trading team cannot. This widens the gap between retail traders (who get free delayed quotes through brokers) and institutional players (who must pay for real-time or faster-than-SIP feeds).

Some argue this concentration actually harms price discovery. Smaller, creative traders leave the market or operate at a disadvantage, reducing diversity of trading strategies. Others counter that data fees are simply a cost of doing business and that technology commoditization (cloud-based data services, lower-cost vendors) is narrowing the advantage over time.

Regulatory and market solutions emerging

The SEC has held multiple roundtables and issued studies questioning whether data fees are excessive. Some proposals: a unified SIP with transparent pricing, a cap on proprietary feed premiums, or a requirement that exchanges bundle data packages rather than charge a la carte per symbol.

Exchanges argue they invest heavily in feed technology and deserve returns on that capital. They also note that retail investors benefit indirectly through cheaper brokerage commissions, subsidized by the data revenue streams.

Alternative data vendors (those selling order flow imbalance, sentiment, or news data) have partially disrupted the traditional exchange-only data model, though proprietary exchange feeds remain the gold standard for latency-sensitive trading.

See also