This article outlines recent approaches to managing credit risk when facing regulatory capital requirements. We explore how institutions should best allocate capital and make economically-optimized investment decisions under regulatory capital constraints, such as those imposed by Basel or CCAR-style rules.
Credit portfolio risk is measured by the required Economic Capital (EC), which reflects diversification, concentration, and other economic risks. In recent years, however, higher capital standards imposed by new stress testing requirements and Basel III have forced organizations to address how to better manage capital to meet regulatory constraints.
While maintaining the required level of Regulatory Capital (RegC) is necessary and indeed mandatory, simply satisfying the requirement does not necessarily align with stakeholders’ preferences for optimal capital deployment and investment decisions. In other words, RegC and CCAR-style stress testing are requirements that organizations have to adhere to and likely do not reflect how stakeholders trade off risk and return.
For instance, a typical RegC measure, such as the Basel Risk-Weighted Asset (RWA), does not account for diversification and concentration risk, which are important to stakeholders. In general, regulatory measures such as RWA are not as risk-sensitive as economic measures. This shortcoming of RegC underscores the importance of EC, which better captures risks that reflect stakeholders’ preferences.
Ideally, institutions should account for both EC and RegC when making business decisions – including strategic planning, pricing, portfolio management, and performance management. For example, if two potential deals have an identical expected return and RWA but different EC, management should favor the lower EC. Similarly, if two deals have the same EC but different RWA, lower RWA is more desirable.
The challenge lies in quantifying a unifying measure where return, RWA, and EC all enter into a single measure that assesses a deal’s profitability – organizations need a unifying EC and RegC measure. Levy, Kaplin, Meng, and Zhang (2012) propose the concept of integrating EC and RegC. They incorporate regulatory capital requirements into a traditional economic framework underpinning EVA- and RORAC-style decision measures. Xu, Levy, Kaplin, and Meng (2015) provide a practical approach of measuring the degree to which an organization is capital-constrained and the degree to which weight should be placed on RegC in business decisions.
At a high level, RegC should not enter into decision rules when it is not constraining. Organizations do not need to account for the RegC constraint if they meet all RegC requirements regardless of business decisions.
Alternatively, a deal that consumes a high level of RegC is particularly unattractive to an organization that is heavily constrained by RegC.
Xu and Levy (2015) extend the work of Levy, Kaplin, Meng, and Zhang and propose a composite capital allocation measure (mostly referred to as composite capital measure, or CCM) integrating EC and RegC. The metric allocates an institution’s top-of-the-house capital in a way that accounts for both economic risks and the degree to which RegC is constraining. This article provides an overview of these recently developed approaches and discusses how financial institutions can use them to improve risk management and business decisions.
The challenge financial institutions face when managing economic and regulatory capital lies in designing and deploying a capital measure that aligns incentives of both management and stakeholders that account for both economic risks and regulatory constraints. While measuring economic risks and RegC on a standalone basis is imperative, a capital charge must ultimately be allocated to align incentives to maximize an organization’s value. The approach proposed by Levy, Kaplin, Meng, and Zhang (2012) and Xu and Levy (2015) highlighted above leverages a traditional economic framework, one where an organization’s stakeholders maximize returns while recognizing risk. The novelty in the approach is in imposing a regulatory constraint. The formal model produces a composite capital measure; whereby the degree to which an organization’s RegC is constraining determines the degree to which weight is placed on RegC.
Figure 1 depicts the relationship between the instrument EC and the required regulatory capitalization rate, also referred to as Risk-Weighted Capital (RWC) (computed by the Basel II standardized approach), on the left side for a typical credit portfolio. In general, RWC is relatively higher for safer instruments, and vice-versa. This finding is also true when RWC is determined according to the Advanced Internal Ratings-Based (IRB) approach, as is shown by Xu, Levy, Meng, and Kaplin (2015) and Xu and Levy (2015).
The right side of Figure 1 compares instrument CCM with EC. Note that CCM is generally higher than EC. This finding is not surprising, as the regulatory capital constraint is expected to increase the capital needed on top of traditional EC. Another important observation is that two sets of asymptotes exist in this figure. CCM converges with EC as EC increases to a high level. This asymptote reflects CCM’s ability to capture the full spectrum of risk, including diversification and concentration risk unaccounted for by RegC.
As EC decreases, CCM flattens to four levels. Recall, we use the Basel II standardized approach to determine RegC, which results in four unique levels of RWC. Thus, each of the four asymptotes to the left represents the minimum level of capital needed for instruments with a certain RWC level, reflecting CCM’s ability to ensure enough capital is allocated to meet RegC requirements.
The difference between RegC and EC brings up a dilemma when financial institutions plan capital allocation. On the one hand, the need to meet the ever-increasing regulatory capital standard pulls institutions toward capital allocation by RegC. On the other hand, a sound risk management system calls for a more appropriate capital allocation measure, such as EC, which accounts for not only default risk, but also diversification and concentration risk. The ideal solution leverages a capital allocation measure such as CCM, which takes into account the full spectrum of risk and, at the same time, ensures that the proper amount of capital is allocated to meet regulatory requirements. What is worth highlighting is the tremendous amount of CCM allocated to high credit quality names. While not surprising given the high level of RegC being allocated, the results are striking when compared with EC.
Intuitively, CCM can be regarded as a combination of EC and RWC. The relative weight of EC and RWC in CCM is institution-specific. It is determined by how constraining the RegC requirement is for the institution. As Xu, Levy, Meng, and Kaplin (2015) illustrate, the degree of RegC constraint can be measured by how much the institution must deleverage due to the RegC requirement. Historically, the deleverage ratio attributed to Basel and stress testing requirements, defined as the percentage decrease in leverage, is approximately 15% to 30% for US and European banks. This observed deleveraging speaks to the degree to which RegC is constraining.
Figure 3. RegC-adjusted RORAC vs. RORAC
Similar to Basel-style rules, CCAR requires adequate capital under severe economic downturns. This boils down to a required capital buffer that adheres to the portfolio’s RWC, while accounting for erosion due to stressed expected losses conditioned on the downturn scenario. Therefore, the sum of required capital buffer and the stressed expected loss is effectively the minimum capitalization rate institutions need to maintain in order to meet stress testing requirements. We will refer to this sum as the effective RWC.
The left side of Figure 2 compares instrument EC with effective RWC for a sample portfolio under a severely adverse CCAR scenario. As EC decreases, the effective RWC converges to 8%, which is the minimum RegC required. As EC increases, effective RWC becomes much more correlated with EC; instruments with larger EC are associated with more severe losses during a stressed scenario, requiring more capital buffer and a higher effective RWC. Once we know the instrument-effective RWC, we can compute CCM accordingly.
The right side of Figure 2 presents instrument CCM against EC under the CCAR requirement. Similar to CCM under the Basel II capital requirement, instrument CCM under the CCAR requirement also exhibits two asymptotes – CCM converges to EC as EC increases to a high level, and CCM flattens out as EC becomes very small. The intuition behind this pattern is the same as explained previously for CCM under Basel-style capital requirements.
Business decisions under regulatory capital constraints
In practice, stakeholders prefer an institution to deploy capital across the organization and make investment decisions that maximize the institution’s overall return-risk trade-off while satisfying regulatory requirements. Integrating EC with RegC allows financial institutions to allocate capital across businesses with a risk metric that accounts for diversification and concentration risk, as well as the regulatory constraints.
In addition, the integrated approach provides decision rules that optimize portfolios from an economic perspective while adhering to RegC requirements. Traditional Return on Risk-Adjusted Capital (RORAC) measures are adjusted to account for investments’ RegC burden. Intuitively, the RegC adjustment can be thought of as a tax that lowers an instrument’s effective return.
Figure 3 compares RegC-adjusted RORAC with standard RORAC under Basel II and CCAR. The two measures are generally very different. In particular, safe instruments tend to have very low or even negative RegC-adjusted RORAC; the low return of safe instruments is not sufficient to cover the implicit cost of the RegC constraint.
Using RegC-adjusted RORAC, institutions can improve the risk-return attractiveness of the portfolio while meeting RegC requirements. Table 1 illustrates the impact of re-weighting the sample portfolio where instruments with the lowest RegC-adjusted RORAC are traded for those with the highest RegC-adjusted RORAC. What is impressive is that a 2.5% portfolio turnover rate can increase the expected return of the portfolio by 60 bps, while keeping the required RegC constant. Furthermore, as institutions increase the portfolio turnover rate (i.e., trade more instruments according to RegC-adjusted RORAC), the portfolio rate of return on both RegC and EC increases.
Under higher capital standards imposed by new stress testing requirements and Basel III, organizations should account for both economic risk and regulatory constraints when managing capital and making business decisions. CCM and RegC-adjusted RORAC measures help institutions achieve this goal. CCM allocates an institution’s top-ofthe- house capital in a way that accounts for economic risks, as well as the degree to which RegC is constraining. RegC-Adjusted RORAC helps institutions improve the riskreturn attractiveness of their portfolios, while maintaining the required RegC level.
1 Moody’s Analytics Quantitative Research Group, Modeling Credit Portfolios, 2013.
2 Amnon Levy, Andrew Kaplin, Qiang Meng, and Jing Zhang, A Unified Decision Measure Incorporating Both Regulatory Capital and Economic Capital, 2012.
3 Pierre Xu, Amnon Levy, Qiang Meng, and Andrew Kaplin, Practical Considerations When Unifying Regulatory and Economic Capital in Investment Decisions, 2015.
4 Pierre Xu and Amnon Levy, A Composite Capital Allocation Measure Integrating Regulatory and Economic Capital, 2015.
Leading economist; commercial real estate; performance forecasting, econometric infrastructure; data modeling; credit risk modeling; portfolio assessment; custom commercial real estate analysis; thought leader.
A well-recognized researcher in the field; offers many years of experience in the real estate ﬁnance industry, and leads research efforts in expanding credit risk analytics to commercial real estate.
Leading economist; recognized authority and commentator on personal finance and credit, U.S. housing, economic trends and policy implications; innovator in econometric and credit modeling techniques.
Looks at the best practices of today that will form the successful risk management practices of the future.
Previous ArticleUsing a Risk Appetite Framework to Align Strategy and Risk
Moody's Analytics Managing Director Amnon Levy, Moody's Analytics Director Libor Pospisil, and Moody's Investor's Service Jim Hempstead presented at the International Association of Credit Portfolio Managers Spring Conference entitled Managing Credit Risk and Emerging Threats: Lessons from the Gaps Revealed by the Pandemic.
This paper describes a conceptually sound quantitative and practical approach to increasing portfolio return/risk, details the requisite steps, and shows how they can be effectively performed using Moody’s Analytics PortfolioStudio®, a cloud-based, credit portfolio management solution designed for business users.
High-level overview of the modeling methodologies implemented in RiskFrontier™ and their business applications. RiskFrontier calculates a credit investment's value at analysis date, its value distribution at a user-specified investment horizon, and its marginal contribution to portfolio risk, for every instrument in the portfolio.
We study the impact of COVID on concentration risk, relevant in the context of limit-setting, portfolio allocation, and other concentration-sensitive measures. Analysing a European portfolio, we show how our solutions can be used to navigate the COVID crisis and better understand risk within a portfolio framework.
Crises reveal behavior incongruent to historic patterns, requiring new data and analyses. COVID shows established models did not evaluate credit adequately. The Cross-Sectional COVID Overlay assesses current credit, projected ratings, and loss measures in new ways, anchoring to well-understood starting points and scenarios.
With COVID-19 continuing to batter the global economy, many banks are struggling to model credit losses as they prepare for their upcoming Comprehensive Capital Analysis and Review (CCAR) submissions as well as 3rd Quarter earnings.
We introduce a granular, obligor-level, scenario-based model for rating transition matrices. It recognizes differences in the statistical properties of ratings and forward-looking PDs, deviating from approaches assuming a one-to-one relationship between segment rating and PD or that decouple dynamics of ratings and PDs.
Well-established models that evaluate the current credit environment are not working given COVID-19. Internal ratings cannot update at frequencies required to react well. This paper addresses these challenges, presenting applications users can incorporate into Internal Rating Assessment and Projected Ratings and Loss.
The COVID-19 pandemic has brought credit risks that are unprecedented in size, are fast-changing, and have vastly different manifestations across industries. The uncertainty of impact is driven by epidemiological progression and sociological response, balanced by fiscal and monetary stimulus.
COVID-19 created additional complexities for institutions navigating CECL accounting standard. This paper provides a natural quantitative approach for incorporating concentration in the allowance process and portfolio management.