How Can a Firm Quantify the Value of Price Improvement from Its Top-Tier Dealers?

Concept

The imperative to quantify price improvement from your most trusted dealers is not an academic exercise in transaction cost analysis. It is the foundational act of building a resilient and intelligent execution architecture. Your firm operates within a complex system of liquidity, and the value extracted from a dealer relationship extends far beyond a simple per-share improvement figure reported on a post-trade summary.

The core challenge lies in moving from a passive acceptance of these figures to a dynamic, multi-dimensional quantification framework that reveals the true economic impact of your dealer’s routing and sourcing decisions. This is about understanding the system you are a part of, and then leveraging that understanding to architect a superior operational edge.

At its most fundamental level, price improvement (PI) is the execution of a trade at a price more favorable than the prevailing National Best Bid and Offer (NBBO). The NBBO represents the tightest spread between the highest displayed bid price and the lowest displayed ask price for a security across all public exchanges. When a buy order is filled at a price below the national best offer, or a sell order is filled above the national best bid, the resulting monetary saving is classified as price improvement. This outcome is possible because the displayed quotes on lit markets do not represent the total available liquidity.

A significant volume of trading interest is not publicly visible, held in “dark” venues or within the internal liquidity pools of market makers. Top-tier dealers provide value by accessing this non-displayed liquidity, creating opportunities to transact at prices superior to the public benchmark.

A firm must evolve its measurement of price improvement from a simple validation of reported savings to a critical analysis of execution quality within the market’s systemic structure.

Deconstructing the Baseline

The NBBO serves as the regulatory and industry-standard benchmark for calculating price improvement. For instance, if your firm places an order to purchase 10,000 shares of a security quoted at an NBBO of $50.20 (bid) and $50.24 (ask), and your dealer executes the entire order at $50.235, you have achieved a price improvement of $0.005 per share. This results in a total saving of $50 against the public offer price. This calculation is straightforward and provides a necessary, albeit incomplete, picture of execution quality.

The deficiency of this model is not in the calculation itself, but in the nature of the benchmark. The NBBO is not a static, independent variable; it is a dynamic price that is itself influenced by the very order flow that is being measured against it.

A sophisticated firm recognizes that significant off-exchange trading volume, often facilitated by the same dealers providing price improvement, can contribute to wider public spreads. This creates a systemic paradox where the benchmark for measuring performance is potentially degraded by the very system it is meant to measure. Therefore, relying solely on NBBO-based price improvement as the definitive measure of dealer value is insufficient.

It is the starting point of the analysis, not the conclusion. A truly robust quantification framework must incorporate additional benchmarks and metrics to build a more resilient and insightful model of dealer performance.

What Is the True Source of Dealer Value?

The value proposition of a top-tier dealer is not merely their willingness to offer a fractional price concession. It is their structural capacity to navigate a fragmented market and source liquidity efficiently. This capability can be dissected into several key components, each of which must be quantified:

• Access to Non-Displayed Liquidity ▴ Top dealers maintain proprietary trading books and have access to a vast network of dark pools and alternative trading systems (ATS). Their ability to interact with this hidden liquidity is a primary driver of price improvement. Quantifying this requires looking beyond the NBBO to understand how frequently and at what size a dealer can execute orders at the midpoint of the spread.
• Routing Optimization ▴ The technological infrastructure of a dealer determines how they route an order. An advanced routing system will intelligently seek out the optimal execution venue, balancing the potential for price improvement against factors like speed and certainty of execution. This involves minimizing gas or transaction fees while maximizing access to diverse liquidity sources.
• Liquidity Enhancement ▴ A dealer may offer to fill a larger order at the NBBO than the size publicly displayed. For example, if the NBBO shows only 500 shares for sale, but a dealer fills an order for 2,000 shares at that price, they have provided significant value by preventing the order from walking up the order book and incurring higher costs. This “liquidity enhancement” is a critical, yet often overlooked, component of price improvement.

Understanding these sources of value is the first step toward building a quantification model that accurately reflects a dealer’s contribution to your firm’s execution objectives. The goal is to move beyond a single number and toward a dashboard of metrics that provides a holistic view of performance.

Strategy

A strategic approach to quantifying dealer-driven price improvement requires the formal adoption of a comprehensive Transaction Cost Analysis (TCA) framework. This framework must be designed to dismantle the simplistic, one-dimensional view of PI and replace it with a multi-benchmark, context-aware system of evaluation. The central strategy is to treat every order as a scientific experiment, where the dealer is a variable and the execution quality is the outcome to be measured against a set of rigorously defined controls. This transforms the conversation with your dealers from a discussion about reported PI dollars to a data-driven dialogue about their performance against specific, mutually understood benchmarks under varying market conditions.

The inadequacy of the NBBO as a sole benchmark necessitates a more sophisticated, multi-layered approach. While the NBBO provides a baseline, its susceptibility to market structure dynamics means that it can present a distorted view of what constitutes a “good” execution. A truly strategic framework triangulates the truth by employing several benchmarks simultaneously.

This allows a firm to analyze dealer performance from multiple perspectives, revealing strengths and weaknesses that a single metric would obscure. The objective is to create a resilient performance narrative, insulated from the flaws of any single benchmark.

True quantification of dealer value emerges from a strategic framework that measures performance against multiple, contextually relevant benchmarks.

Establishing a Multi-Benchmark Framework

To construct a robust analysis, a firm must systematically compare execution prices against several benchmarks. Each benchmark tells a different part of the story, and together they provide a comprehensive picture of dealer performance. The selection of benchmarks should be deliberate, reflecting the firm’s specific trading objectives.

1. The NBBO Benchmark ▴ This remains the regulatory and industry standard. It is particularly useful for quantifying the most direct and easily communicable form of price improvement. Its primary function is to answer the question ▴ “Did we execute at a better price than was publicly available at the time of the trade?” Despite its limitations, it is a non-negotiable component of any TCA program.
2. The Arrival Price Benchmark ▴ This is the midpoint of the bid-ask spread at the precise moment the order is sent to the dealer. This benchmark is critical for measuring implementation shortfall, which captures the full cost of execution, including market impact and dealer latency. A significant deviation from the arrival price may indicate that the dealer’s process is slow or that the order itself is moving the market.
3. The VWAP Benchmark (Volume-Weighted Average Price) ▴ For orders that are worked over a period of time, the VWAP serves as an excellent benchmark. It represents the average price of a security over a specific time horizon, weighted by volume. Comparing a dealer’s execution price to the VWAP reveals their ability to execute orders with minimal market impact and to time their trades effectively within a given session.
4. The Midpoint Benchmark ▴ The midpoint of the NBBO is often considered the “perfect” or “zero-cost” execution price. Measuring the frequency and volume with which a dealer can execute at or near the midpoint is a powerful indicator of their access to high-quality, non-displayed liquidity. For firms focused on minimizing spread costs, this is arguably the most important benchmark.

Comparative Analysis of Benchmarks

The strategic value of a multi-benchmark approach lies in the ability to compare and contrast the results. A dealer might excel against one benchmark while underperforming against another. This is not necessarily a sign of poor performance, but rather an indication of their specific strengths. The following table illustrates the strategic application of each benchmark.

How Does Dealer Discretion Affect Outcomes?

A critical element of strategy involves recognizing that not all price improvement is created equal. Research indicates that there can be significant dispersion in execution quality even among brokers who route to the same venues. This suggests that the “broker execution,” or the specific way a dealer handles an order, is a major driver of performance. This is not about payment for order flow (PFOF), which has been shown to explain very little of the variation in execution quality, but about the proprietary logic and routing technology a dealer employs.

Therefore, a firm’s strategy must include a qualitative assessment of its dealers’ technology and a quantitative analysis of their performance on matched trades ▴ identical trades sent to different dealers at the same time. This A/B testing approach provides the most definitive evidence of a dealer’s unique value contribution.

Execution

The execution of a quantitative framework for dealer price improvement hinges on a disciplined, systematic approach to data collection, modeling, and analysis. This is where strategic theory is forged into an operational tool for performance management and risk mitigation. The goal is to build a “Dealer Scorecard” that is not only retrospective but also predictive, allowing the firm to dynamically allocate order flow to the dealers best equipped to handle specific types of orders under current market conditions. This requires a granular level of data and a commitment to rigorous quantitative analysis.

The Data Collection and Normalization Protocol

The foundation of any quantitative analysis is the quality and completeness of the underlying data. A firm must establish a protocol for capturing a detailed set of data points for every single order. This data must be normalized across all dealers to ensure that comparisons are made on a like-for-like basis. The required data set includes:

• Order Timestamps ▴ Capture timestamps to the microsecond level for every stage of the order lifecycle ▴ order creation, transmission to dealer, dealer acknowledgment, routing to venue, and final execution.
• Price Data ▴ For each order, record the arrival price (midpoint at time of order creation), the NBBO at the time of execution, the execution price itself, and the NBBO a short period (e.g. one minute) after the trade to check for reversion.
• Order Characteristics ▴ Log the security ticker, order size, order type (market, limit, etc.), and any special instructions.
• Dealer and Venue Identification ▴ Clearly tag each execution with the dealer who handled the order and the ultimate execution venue. This is critical for attributing performance correctly.

This data must be warehoused in a structured database that allows for complex queries and analysis. The process of data collection and normalization is non-trivial, but it is the absolute prerequisite for any meaningful quantification of dealer performance.

Core Quantification Models

With a robust data set in place, the firm can deploy a series of quantitative models to dissect dealer performance. These models should move beyond simple PI calculations to provide a more nuanced understanding of value creation.

Model 1 Effective Spread Capture

This model measures how much of the bid-ask spread a dealer was able to “capture” for the firm. It is a powerful measure of their ability to source liquidity inside the public quotes. The formula is:

Effective Spread Capture (%) = ((Side (Execution Price – Midpoint Price)) / (Spread / 2)) 100

Where ‘Side’ is +1 for a buy and -1 for a sell. A result of 100% means the trade was executed at the opposing side of the NBBO (no PI), while a result of 0% means execution at the midpoint. A negative result indicates execution at the midpoint or better. This metric is far more revealing than a simple PI dollar amount.

Model 2 Post-Trade Reversion Analysis

Price improvement can be illusory if the market immediately moves against the “improved” price. This phenomenon, known as post-trade reversion or adverse selection, indicates that the liquidity provided was fleeting. To quantify this, we measure the change in the midpoint price a short time after the trade.

Reversion (bps) = Side ((Post-Trade Midpoint – Execution Price) / Execution Price) 10,000

A high positive reversion for buy orders (or negative for sell orders) suggests that the dealer is consistently executing at “stale” quotes, and the perceived price improvement is quickly eroded. A good dealer will show minimal or even favorable (negative) reversion on average.

The Dealer Performance Scorecard

The outputs of these models should be consolidated into a comprehensive Dealer Performance Scorecard. This scorecard serves as the primary tool for evaluating and comparing dealers. It should be reviewed on a regular basis (e.g. monthly or quarterly) and used to drive conversations with your dealers.

In this example, Dealer B appears to be the strongest performer. Despite not having the highest volume, they deliver the best price improvement in both dollar and basis point terms, capture more of the spread, and have the lowest post-trade reversion, indicating high-quality liquidity. Conversely, Dealer C, while handling the most volume, shows weaker performance across all key quality metrics. This data allows a firm to move beyond simple volume-based relationships and allocate order flow with analytical precision.

Reflection

Having constructed a framework to quantify the value delivered by your dealers, the analysis must now turn inward. The data and scorecards are not merely tools for evaluating external partners; they are a mirror reflecting the sophistication of your own execution protocol. A dealer’s performance is intrinsically linked to the quality of the order flow you provide and the clarity of your instructions.

Does your internal system for order generation and allocation possess the intelligence to leverage the specific strengths revealed in your analysis? Moving beyond static reports to a dynamic, feedback-driven system is the final and most critical step.

From Analysis to Systemic Advantage

The true potential of this quantitative framework is realized when it becomes the core of a learning system. The dealer scorecards should feed directly back into your order routing logic, creating a dynamic allocation engine that intelligently matches orders with the dealers most likely to achieve superior execution based on historical performance under similar market conditions. This transforms the relationship with your dealers from a simple client-vendor dynamic into a strategic partnership. The data becomes a shared language for a continuous dialogue about performance, technology, and liquidity, ultimately forging a more resilient and adaptive execution architecture for your firm.