In What Ways Does the Practice of ‘Last Look’ in Rfs Affect the Fiduciary Responsibility of Best Execution?

Concept

The fiduciary responsibility of best execution compels an investment manager to secure the most advantageous terms reasonably available for a client’s transaction. This mandate is evaluated through a multi-faceted lens, considering not just the explicit price of an asset but also the implicit costs embedded in the trading process. Within the Request for Quote (RFQ) protocol, particularly in decentralized or less-regulated markets like foreign exchange, the practice of ‘last look’ introduces a significant complication to this fiduciary calculus.

It grants a liquidity provider (LP) a final, discretionary moment to reject a trade request after the client has committed to the quoted price. This operational pause, however brief, fundamentally alters the nature of the transaction from a firm commitment to a conditional option held by the market maker.

From a systems perspective, an RFQ is designed to be a bilateral price discovery mechanism, allowing a liquidity consumer to solicit competitive, executable quotes from a select group of LPs. The introduction of a last look window transforms the LP’s price quote from a firm, binding offer into an indication of interest. The LP retains the right to withdraw that interest if the market moves in a way that makes the trade unprofitable for them during the “hold time” ▴ the period between the client’s click and the LP’s final acceptance. This creates an inherent asymmetry of information and optionality.

The client reveals their trading intention, yet receives no guaranteed execution in return. This optionality granted to the LP is the central point of friction with the principle of best execution.

The Mechanics of Asymmetry

Understanding the impact of last look requires viewing it as a risk management tool for the liquidity provider. In fragmented markets without a central limit order book, LPs face the risk of being “picked off” by high-speed traders who can exploit latency differences between various trading venues. A last look window allows the LP to re-validate the quote against prevailing market conditions before committing capital, mitigating the risk of executing on a stale price. This protection, however, is not without cost to the liquidity consumer.

The core issue is that the risk is not eliminated but rather transferred back to the client. When an LP rejects a trade due to price movement (a “symmetric” rejection, as the market could have moved in either direction), the client is left with an unexecuted order and must return to the market, by which time the price has likely deteriorated further. This process of re-quoting introduces execution uncertainty and potential slippage, which are direct components of the total cost of trading and thus fall under the purview of best execution analysis.

A last look provision transforms a liquidity provider’s quote from a firm commitment into a conditional option, creating an information asymmetry that directly challenges the principles of best execution.

Furthermore, the practice opens the door to more problematic applications. An “asymmetric” application of last look would involve an LP rejecting trades only when the market moves against them, while still executing trades where the market has moved in their favor during the hold time. Such a practice moves beyond risk mitigation and becomes a mechanism for profit optimization at the direct expense of the client. This potential for misuse is what has drawn significant scrutiny from regulators and industry bodies, leading to the development of frameworks like the FX Global Code of Conduct, which calls for transparency in how last look is applied.

For the fiduciary, evaluating best execution in a last look environment requires a shift in analysis. The headline price of a quote is insufficient. A comprehensive assessment must incorporate metrics like rejection rates, the duration of hold times, and the market conditions under which rejections occur. These factors quantify the hidden costs and execution uncertainty introduced by the last look option, providing a more complete picture of whether the trading process truly served the client’s best interest.

Strategy

Navigating the complexities of last look within an RFQ framework demands a strategic approach that extends beyond simple price comparison. For a fiduciary, the objective is to construct an execution policy that quantifies and minimizes the implicit costs imposed by this practice. This involves developing a system for liquidity provider selection and evaluation that is grounded in data, recognizing that the most competitive quoted price may not lead to the best-executed outcome if it is frequently unavailable.

A Framework for Liquidity Provider Evaluation

A robust strategy begins with the classification of liquidity providers based on their application of last look. This is not a binary distinction but a spectrum of behaviors that can be measured and monitored over time. Fiduciaries must move from a relationship-based model of liquidity access to a data-driven one. This requires the systematic collection and analysis of execution data to build a performance profile for each LP.

Key metrics for this evaluation include:

• Rejection Rate ▴ This is the most direct measure of execution uncertainty. A high rejection rate from an LP indicates that their quotes are less reliable, forcing the trader to re-engage the market and incur potential slippage. This should be analyzed under different market volatility conditions.
• Hold Time Duration ▴ The length of the last look window is a critical factor. Longer hold times expose the client’s order to more market risk and provide the LP with a longer free option. Measuring the average hold time for each LP helps quantify this risk exposure.
• Post-Rejection Price Movement ▴ Analyzing the market movement immediately following a rejection is crucial. Consistent, adverse price movement after a rejection can be an indicator of information leakage, suggesting that the client’s trading intention is being signaled to the broader market.
• Fill Quality Analysis ▴ For LPs that provide partial fills, analyzing the circumstances of those fills is important. Understanding whether partial fills occur during specific market conditions can reveal patterns in an LP’s risk management practices.

By tracking these metrics, a fiduciary can create a tiered system of liquidity providers. Those with low rejection rates, short hold times, and minimal evidence of adverse information leakage would be considered ‘Tier 1’ providers, deserving of a greater share of order flow. Conversely, LPs with consistently poor metrics would be downgraded or removed from the RFQ panel.

Effective management of last look requires treating liquidity provider selection not as a static relationship but as a dynamic, data-driven process of continuous performance evaluation.

The following table provides a simplified model for comparing two hypothetical liquidity providers. While LP A offers a tighter average spread, its high rejection rate and long hold times may result in a higher all-in cost of execution compared to LP B, which provides more reliable, albeit slightly wider, quotes.

The Strategic Dialogue with Liquidity Providers

Armed with this data, a fiduciary can engage in a more meaningful dialogue with their LPs. Instead of relying on qualitative assurances, the conversation can be centered on quantitative performance. This allows the buy-side firm to clearly articulate its execution quality requirements and hold the LP accountable.

It also encourages LPs to be more transparent about their last look methodology. An LP that is unwilling to provide clarity on its practices or whose performance data consistently shows unfavorable outcomes can be strategically de-prioritized.

This data-driven approach also enables a more sophisticated routing logic. For example, during periods of low market volatility, an RFQ might be sent to a wider panel of LPs. During high volatility, the RFQ could be directed only to those LPs who have demonstrated a history of providing reliable quotes under stress. This dynamic routing strategy helps to mitigate the risk of execution uncertainty when it is most likely to occur.

Execution

The execution of a best execution policy in an environment that includes last look requires a disciplined and technologically sophisticated operational setup. It is about translating the strategic framework of LP evaluation into a day-to-day workflow that systematically minimizes the costs associated with execution uncertainty and information leakage. This involves the rigorous application of Transaction Cost Analysis (TCA), the implementation of intelligent order routing systems, and a commitment to continuous monitoring and optimization.

Implementing a Quantitative TCA Framework

A modern TCA framework is the cornerstone of managing last look. It must go beyond simple arrival price benchmarks to capture the nuanced costs introduced by the practice. The objective is to make the implicit costs explicit and measurable.

The following components are essential for a TCA system designed to analyze last look:

1. Timestamping Precision ▴ To accurately measure hold times, the system must capture high-precision timestamps at every stage of the order lifecycle ▴ RFQ sent, quote received, order sent, and confirmation/rejection received. Millisecond accuracy is the standard.
2. Rejection Cost Analysis ▴ The system must automatically calculate the cost of each rejection. This is typically measured as the difference between the price of the rejected quote and the price at which the order was eventually filled. This “rejection slippage” needs to be aggregated and attributed to specific LPs.
3. Hold Time Cost Quantification ▴ This is a more complex calculation that attempts to model the market risk cost of the hold period itself. One approach is to measure the volatility during the hold time and express it as a cost in basis points. An LP with a longer hold time imposes a higher risk cost on the client, even if the trade is ultimately accepted.
4. Information Leakage Metrics ▴ The TCA system should analyze price movements of related instruments before, during, and after the RFQ process. A pattern of adverse price action correlated with RFQs sent to a particular LP can be a strong indicator of information leakage.

The output of this TCA process is not just a historical report but a predictive tool. It should feed directly into the firm’s execution logic, allowing for real-time adjustments to LP selection and routing strategies.

A Practical Model for Rejection Costing

The table below illustrates a simplified TCA report for a series of trades with different LPs. It highlights how a focus on rejection costs can alter the perception of which LP provides the best performance.

A granular Transaction Cost Analysis framework transforms the abstract risk of last look into a quantifiable expense, enabling fiduciaries to make execution decisions based on empirical evidence rather than quoted spreads alone.

Intelligent Order Routing and Protocol Choices

The insights generated by the TCA framework should power an intelligent order router (IOR) or a sophisticated Execution Management System (EMS). This technology can automate the application of the firm’s execution policy.

Key features of such a system include:

• Dynamic LP Tiering ▴ The IOR should dynamically adjust the ranking of LPs based on the latest TCA data. An LP with rising rejection rates would be automatically de-prioritized in the RFQ panel.
• Context-Aware Routing ▴ The system should be able to adjust its routing strategy based on order size, time of day, and real-time market volatility. For large or sensitive orders, it might route exclusively to LPs that offer firm, no-last-look liquidity, even at a slightly wider spread.
• Support for Firm Liquidity ▴ A critical component of the execution toolkit is access to pools of firm liquidity where last look is not permitted. While these pools may sometimes show wider spreads, they provide certainty of execution, which can be invaluable, especially in volatile markets. A fiduciary must have a documented policy for when the use of firm liquidity is prioritized over potentially tighter, but uncertain, last look quotes.

Ultimately, fulfilling the fiduciary responsibility of best execution in a market with last look is an ongoing, dynamic process. It requires a significant investment in technology and data analysis capabilities. However, by making the hidden costs of last look visible and actionable, a fiduciary can construct a more resilient and effective execution process that truly serves the best interests of their clients.

Reflection

Calibrating the Execution Apparatus

The integration of ‘last look’ analytics into an execution framework moves the conversation from a debate over fairness to a problem of engineering. The data streams are available; the metrics can be defined. The remaining challenge lies in the institutional will to build the apparatus, to invest in the analytical horsepower that transforms raw execution data into a decisive operational advantage. The presence of last look is a feature of the current market structure; its impact on a portfolio, however, is a variable that can be managed, measured, and minimized.

A fiduciary’s responsibility extends to the quality of their measurement tools. An execution policy that fails to account for the quantifiable costs of rejection rates and hold times is incomplete. It mistakes the quoted price for the final cost.

The ultimate question for any asset manager is not whether their providers use last look, but rather how precisely they can quantify its cost and how systematically they can route around the damage it can cause. The path to superior execution is paved with superior data.