top of page
actual logo.png

Global Banks' Dilemma over FRTB IMA vs SA Implementation



Global banks are grappling with the trade-offs between FRTB IMA and SA approaches. As deadlines for the Fundamental Review of the Trading Book (FRTB) draw nearer across many jurisdictions, debates around using the Internal Models Approach (IMA) versus the Standardised Approach (SA) are intensifying. The decision is not straightforward. In this article, we examine the pros and cons of each approach and discuss how market participants are addressing this challenge.


Banks adopting SA or IMA must prepare for increased capital demands, significant implementation challenges, and strict regulatory compliance. Individual institutions can choose between IMA and SA, but SA is mandatory for market risk reporting under FRTB. The IMA is losing favour at some banks due to its complexity; global banks such as HSBC, Barclays, Deutsche Bank, and several Canadian banks are reassessing the IMA’s use of the IMA for market risk. In contrast, SA provides prescribed risk weights, is relatively straightforward to implement, and is ideal for smaller or less complex banks. Choosing between IMA and SA is crucial for trading desks, as it directly impacts capital adequacy against potential future losses. Some banks may adopt a hybrid approach to balance modeling complexity with capital consumption. Indeed, SA is likely to become the preferred approach for regulatory reporting among many banks, while IMA may be retained for internal capital impact analysis. Mid-sized banks are considering SA for risk monitoring and trade analysis purposes, as its Sensitivity-Based Approach (SBA) aligns well with portfolio risk sensitivities.


The IMA includes an inbuilt correlation model, offering diversification benefits across assets. FRTB’s incorporation of varying liquidity horizons (instead of a fixed 10-day horizon) provides a more realistic measure of risk and generally leads to higher but more appropriate capital charges. By contrast, the SA does not account for correlations or diversification across risk buckets, thus offering only limited recognition of hedging benefits.


The table below outlines the key differences between the approaches, highlighting how FRTB changes market risk measurement.

Aspect

IMA (Internal Models Approach)

SA (Standardised Approach)

Complexity

Complex, flexible models with intensive computation requirements.

Simpler, data-driven calculations using standardized formulas.

Suitability

Ideal for banks with advanced risk systems and sophisticated internal models.

Easier to understand; suitable for banks with less complex risk profiles or those not using internal models.

Capital Requirements

Aligns capital more closely with actual risk, potentially lowering requirements for certain complex portfolios.

One-size-fits-all metrics can raise capital requirements for banks with more advanced risk management strategies.

Strategic Advantage

Enhances competitiveness for large banks by leveraging superior risk models and allowing tailored risk assessments.

Offers clear, standard methods and easier execution, though new processes must still be established at both bank and desk levels.

Regulatory Approval

Requires rigorous internal model validation and explicit regulatory approval, demanding significant resources.

Provides a consistent framework that streamlines regulatory oversight and allows easy comparison across institutions.

Risk Model

Extends existing risk models by shifting from VaR to an Expected Shortfall (ES) framework.

Relies on regulator-defined standard risk weights under the new rules (no internal model development).

Implementation

Implementation is intensive, with challenging data requirements and complex infrastructure needs.

It can be developed from scratch and is relatively easy to implement with fewer infrastructure demands.

Table 1: Comparison of IMA vs SA under FRTB.

 

The Internal Models Approach (IMA) provides clear benefits such as improved risk alignment and potentially lower capital charges for certain trading desks. However, it also presents challenges, including the need for regulatory approval and the necessity of maintaining robust internal models. Meanwhile, the Standardised Approach (SA) is straightforward to implement but may fail to accurately reflect the risk profiles of banks with complex products, thereby limiting their trading flexibility. Implementing IMA can incur substantial upfront costs that any immediate capital savings might not fully justify. Nevertheless, the long-term benefits for banks with complex or structured trading desks can be significant. Adopting IMA drives improvements in risk measurement methodologies, data quality, and overall risk management practices, ultimately yielding more accurate risk metrics and better capital forecasting.


              Critics of IMA often point to practical issues such as incomplete market data, which can result in high capital charges for certain positions. Additionally, IMA requires rigorous backtesting and profit-and-loss (P&L) attribution tests; failure to meet these tests could push banks back to the SA approach for those portfolios. However, IMA significantly benefits desks dealing with complex or illiquid trades by reducing capital consumption through more refined risk sensitivity. Banks with complex portfolios that avoid IMA also risk reputational damage for not employing advanced risk management, which may discourage potential counterparties. Conversely, if IMA is broadly avoided across the industry and most banks rely only on SA, market liquidity could shrink, and price volatility could rise due to fewer active trading participants in specific markets.


Pros and Cons of Implementing IMA


Positives of Implementing IMA


  • Risk Sensitivity: This approach enables banks to use internal models calibrated to their exposures, allowing greater precision in capturing specific risks (e.g., liquidity risk, credit spread risk) than the standard approach.


  • Correlation and Diversification Benefits: Facilitates recognition of risk offsetting across asset classes and encourages portfolio diversification. This can lead to lower capital charges for well-hedged or diversified portfolios than non-diversified ones under SA.


  • Potential Capital Savings: Banks with sophisticated risk management can potentially achieve lower capital requirements under IMA than under SA, since internal models may better reflect actual risk and thereby reduce excess conservatism in capital buffers.


  • Flexibility: Accommodates the complexity of diverse trading strategies more effectively than a standardized formula. IMA can be tailored to unique portfolio characteristics, whereas SA applies uniform risk weights that might not fit unusual or complex instruments.


  • Competitive Advantage: Advanced risk management via IMA can give global banks an edge by reducing regulatory capital charges (provided regulatory approval is obtained). This, in turn, allows banks to maintain a broader range of product offerings for clients, since high-capital-consuming products can be better managed under an internal model.


  • Enhanced Risk Management: By reconciling theoretical vs. actual P&L (through tools like risk-theoretical P&L, RTPL, for model validation), banks can strengthen their risk management and model accuracy. The IMA framework forces closer monitoring of model performance and fosters improvements in risk measurement techniques.


  • Improved Governance: Developing and running IMA require advanced governance frameworks and structured processes. Over time, this elevates the overall risk governance standards within the bank, as teams must rigorously document models, assumptions, and validation results.


  • Synergies in Data and Systems: Implementing IMA is an opportunity to upgrade the bank’s risk data architecture. It demands clear data lineage for the front-to-back trading book, integrating data across silos (market risk, credit risk, finance, operations) into a unified framework. This supports the IMA calculations and improves enterprise-wide data quality and consistency.


    Potential Downsides of Implementing IMA


  • Regulatory Approval: Banks must secure explicit approval from regulators to use IMA, demonstrating that their internal models meet rigorous standards. This process is time-consuming and resource-intensive. Risk systems must also incorporate additional shock scenarios and map all relevant risk factors across trading systems to satisfy regulators.


  • Model Risk: Reliance on complex internal models introduces significant model risk. These models require continuous validation and frequent recalibration to remain accurate. Regulators impose strict validation requirements to ensure robustness, but the sheer complexity of IMA models can itself create operational and implementation risks.


  • Business Structure Impact: FRTB’s rules (especially under IMA) affect how banks organize their trading activities. The regulation enforces a clearer separation between trading book and banking book positions and imposes stricter definitions for trading desks. Banks may need to reorganize desk structures and adjust business processes to comply, which can be disruptive.


  • Computational Burden: The daily capital charge calculations under IMA (i.e., the IMCC for modellable risks) are highly computationally intensive. Banks require significantly greater end-of-day processing capacity and more powerful infrastructure to calculate Expected Shortfall (ES) for multiple desks. The volume of data to be processed and reported increases dramatically with IMA, straining IT systems and potentially slowing down reporting timelines.


  • Documentation and Reporting: IMA comes with extensive documentation and reporting obligations. Banks must maintain detailed model documentation, audit trails, and rigorous reports for internal model performance. There are also stricter requirements around third-party data and model inputs, necessitating transparent governance and record-keeping to satisfy regulatory scrutiny.


  • High Implementation Cost and Effort: Adopting IMA across the organization is an expensive, multi-year exercise. It demands expert quantitative resources, significant technological investments, and a lengthy implementation timeline. Each trading desk, product line, and risk factor may need a comprehensive modeling rollout with ongoing data support. Banks often find attempting a “greenfield” build of IMA from scratch impractical if they lack existing internal model infrastructure. In practice, many institutions will use SA for specific portfolios or instruments where developing a reliable internal model is not feasible due to data or resource constraints.


Key Parameters in IMA vs SA Trade-offs for Global Banks


Risk transformation under FRTB involves adapting various systems, processes, and data management practices to comply with new regulatory standards. Risk management teams must evaluate several key factors when deciding between (or combining) IMA and SA, as these will shape the bank’s implementation strategy and outcomes.


  • Capital Efficiency: IMA can enhance a bank’s capital efficiency by optimizing capital usage relative to actual risk, potentially boosting profitability. Diversification benefits under IMA recognize correlations among risk factors and asset classes, often lowering overall market risk capital charges by allowing offsetting of risks in a broad portfolio (which smooths out capital requirements and reduces exposure to extreme tail events). Key capital components like the Default Risk Charge (DRC) and Non-Modellable Risk Factors (NMRFs) drive IMA’s capital calculations. By contrast, SA delivers more predictable and stable capital requirements, which can be better suited to banks with simpler portfolios or less sophisticated risk management.


  • Complexity of the Trading Book: The more complex and varied a bank’s trading portfolio, the more beneficial IMA can be. IMA provides a more accurate, tailored risk assessment than SA's generic approach for a highly diverse book (including derivatives, structured products, and positions across multiple asset classes). IMA also decomposes the Expected Shortfall (ES) into diversifiable and non-diversifiable components, yielding more precise capital attribution – as expected, the non-diversifiable (systematic) portion of risk is higher than the diversifiable portion. This level of granularity is not available under SA.


  • Model Development and Validation: Implementing IMA necessitates advanced quantitative modeling, comprehensive stress testing, and continuous monitoring of model performance. IMA best suits institutions with a strong quantitative risk culture and the capability to develop, manage, and rigorously validate internal models. In contrast, SA does not require building internal risk models at all – it uses prescribed formulas – making SA more appropriate for banks with straightforward risk profiles or those without the resources to maintain complex models.


  • Enhanced Data Management: IMA relies heavily on robust and high-quality data. Banks need extensive historical time series of market data, accurate price data for all positions, and reliable correlation and volatility inputs to feed their models (for full revaluation, stress scenarios, ES calculations, etc.). Significant investment in data management and technology is often necessary to meet these needs, including setting up new data feed processes to handle NMRF requirements. By comparison, SA’s data needs are much simpler, as it uses regulatory risk parameters – a factor that can appeal to banks wary of massive data projects.


  • Integrated Risk Management (Front Office, Risk, Finance): Aligning the IMA with the bank’s broader risk management framework is critical. Under FRTB, the models used for regulatory capital (risk management) should be closely integrated with front-office pricing models and the data streams used by finance. This integration ensures consistency between how risks are measured for capital versus how they are managed day-to-day. A well-integrated approach improves capital planning, risk reporting, and strategic decision-making across the organization, breaking down silos between departments.


  • Stress Testing and Scenario Analysis: FRTB’s IMA requires banks to incorporate rigorous stress testing and scenario analysis into model validation and ongoing risk management. This means regularly challenging the models with historical stress events and hypothetical scenarios to ensure they remain accurate under extreme conditions. Many banks also use independent model validation or third-party vendors to test and validate IMA models, which boosts confidence in the models’ outputs. These practices are essential to comply with regulatory standards and to accurately capture tail risks – something that SA, with its static risk charges, does not address as explicitly.


  • Technology and Infrastructure: Deciding on IMA versus SA has major implications for technology infrastructure. An IMA deployment demands a sophisticated IT infrastructure and possibly cloud computing capabilities to handle the sheer volume of calculations and data. Legacy risk systems may not be up to the task; upgrading hardware, software, and data architecture is often needed to meet FRTB’s stricter data quality and computation speed requirements. Banks must weigh the cost of these upgrades against the benefits of IMA. By contrast, sticking to SA might allow use of more existing systems with minimal enhancements.


  • Implementation Timeline and Costs: Implementing IMA is both time-consuming and resource-intensive. Banks need to invest in specialist modeling teams, new systems, and lengthy testing phases. The overall cost of an IMA program is substantial, and the timeline can span several years from initial development to full regulatory approval and roll-out. Given this complexity, it is generally impractical for banks to attempt a completely new (greenfield) IMA build if they lack prior internal model experience – such efforts are more likely to succeed at institutions that can build on existing risk infrastructure. In contrast, an SA implementation can typically be carried out more quickly and at lower cost, since it involves applying a standard rule set and requires fewer bespoke developments.


Final Considerations for Banks


The primary regulatory goal of FRTB is to ensure banks hold sufficient capital to withstand market risks. Both the IMA and SA frameworks for market risk capital have distinct merits, and banks should carefully weigh these trade-offs to choose the approach that best fits their portfolio and capabilities. In practice, many institutions will employ a mix of the two—using SA where it is efficient and reserving IMA for areas where internal modeling adds clear value—to minimize capital costs while maintaining robust risk management.

             

Successful FRTB implementation will also require enhancements to data and analytics. Banks should invest in improved market data collection (particularly to address NMRFs) and refine their processes for calculating capital charges. Innovative techniques like machine learning can be applied to streamline data processing and detect patterns that improve model accuracy. Notably, several global banks plan to adopt SA initially and only transition to IMA once they have confidence in their models, acknowledging that SA alone may not capture all nuances of market risk. Over the long term, effective adoption of FRTB – potentially including selective use of IMA – is expected to produce better-capitalized institutions and a more resilient financial system. IMA offers a significant competitive advantage for banks with especially complex trading portfolios by allowing more risk-sensitive and tailored capital calculations that reflect their accurate risk profile.

             

Banks must also stay abreast of ongoing regulatory developments related to FRTB. Several critical implementation challenges – treating non-modellable risk factors, P&L attribution tests (PLAT), and the IMA’s Default Risk Charge (DRC) – are subjects of active regulatory consultation and refinement. Upcoming guidance and rule adjustments from key authorities (e.g., the PRA, EBA, FED, etc.) will shape banks’ timelines and technical approaches. Institutions aiming for IMA approval need to remain engaged in these discussions and be ready to adapt their frameworks as clarifications emerge. Though navigating these evolving standards is complex, the continued refinement of FRTB is expected to yield long-term benefits by fostering a more robust, risk-sensitive approach to market risk management across the industry.




References

Glossary



(Ravi Bhushan is a Director at Solytics Partners. Working at Solytics, Ravi advises various large global and regional banks across multiple jurisdictions on market, modelling, and model risks. For more information about their work, you can visit https://www.solytics-partners.com/)

 





bottom of page