Stress Testing of the Insurance Sector in Barbados 1 Disclaimer © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Cover Image Credits: Open source image, www.everbridge.com, "Mitigating the Impact of Severe Weather"; www.forbes.com, "How Geospatial Data Could Provide Clarity After The Storm"; www.insurancebusinessmag.com "2022 Catastrophe Losses Revealed". This work is the property of the World Bank. It is permissible to copy and use any of the material in this report provided that the source is appropriately acknowledged. Further information is available from: © The World Bank 2021 Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. This work is a product of the staff of The World Bank with external contributions, financed by the European Union, under the Caribbean Regional Resilience Building Facility, managed by the Global Facility for Disaster Reduction and Recovery (GFDRR). The findings, analysis and conclusions expressed in this document do not necessarily reflect the views of any individual partner organization of The World Bank, its Board of Directors, the governments they represent or the European Union. Although the World Bank and GFDRR make reasonable efforts to ensure all the information presented in this document is correct, its accuracy and integrity cannot be guaranteed. Use of any data or information from this document is at the user’s own risk and under no circumstances shall the World Bank, GFDRR or any of its partners be liable for any loss, damage, liability or expense incurred or suffered which is claimed to result from reliance on the data contained in this document. The boundaries, colors, denomination, and other information shown in any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS: The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. 2 Table of Contents 04 Introduction 05 Stress Testing Framework 08 Risk Profile - Barbados 15 Market Assessment 16 International Stress Testing Model 20 Enterprise Risk Management (ERM) 22 Own Risk and Solvency Assessment (ORSA) 25 Barbados - Stress Test Model 38 Regulatory Reporting 42 Early Warning System 44 Financial Services Commission Recommendations Financial Services Commission - Barbados • www.fsc.gov.bb 3 Introduction This report has been developed by the World Bank for the Financial Services Commission (the "FSC" or the "Commission") of Barbados. To goal of this report is to: Provide the Commission with a strategy to develop regulatory approaches to stress testing, reinsurance and macroprudential surveillance. Outline the legal framework to identify, monitor, assess and contribute to the mitigation of risks from the insurance sector. At the completion of this project, the Commission will be able to: Identify current supervisory Identify situations which require issues related to catastrophe risk further advice about appropriate in the insurance sector. monitoring techniques to minimize solvency impacts of catastrophe risk. Review the legal framework to consider gaps in supervisory Understand risks posed by architecture. catastrophe risk and stress testing to the insurance sector in Recognize common natural Barbados. disaster impacts on the insurance sector and how catastrophes may impact the industry in the future. 4 Financial Service Commission - Barbados • www.fsc.gov.bb 5 Stress Testing Framework The World Bank team is assisting the FSC in the evaluation of existing of stress testing tools that will enhance the resiliency of the insurance sector against natural hazards, and to support the development of a risk-based supervisory regime specifically focused on reinsurance and macroprudential surveillance. This will contribute to increased insurance penetration, consumer protection, and stability. This report will document the assessment and recommendations for the FSC: Stress Testing Framework Review Purpose of Stress Testing An institution’s stress testing program should Assessment of parameter and key process be a forward-looking tool used to assess the stages: scenarios, impacts, mitigating actions, resilience of the institution under adverse, but reporting, review validation; plausible conditions. As a result, this tool should be widely incorporated in the risk management program. It is capable of Review of the resilience of domestic highlighting vulnerabilities, aids in insurance market to natural disaster shocks; determining the risk appetite and exposure limits, and provides quantitative evidence to Review of Barbados' Country Disaster Risk support these and other decision making Profile ("CDRP"); processes. An institution’s stress testing program should Measurement of the model risk associated serve the following purposes: with different modeling choices (scenarios and impacts assessment); Risk identification Consideration of transition matrices The stress test for disasters from natural conditional upon the systematic component hazards should highlight the vulnerabilities in (economic cycle, lines of business, etc.); the insurer’s portfolio as it should consider concentrations and interactions, among various other elements. Incorporation of feedback loops and contagion effects; and Capital and Liquidity management Assessment/ranking of alternative mitigating These tests should be forward- looking and actions to align the risk profile of the insurance able to identify the impact of disasters from sector. natural hazards on the capital and liquidity position of the company. These tests provide useful inputs into internal capital adequacy models used by the institutions. 6 Stress Testing Framework Stress testing is a risk management tool used to estimate plausible, but not impossible, adverse impacts of a set of specified changes in risk factors on the financial position of an entity. In recent years, stress tests have become more popular as key tools for assessing resilience and identifying factors that may undermine financial stability. At the regulatory level, stress tests form part of the macro-prudential tool kit and are key in assessing financial vulnerabilities, which may trigger systemic events. Stress tests are also a vital part of financial risk management for entities and continue to emerge as a fundamental tool for supervisors to identify measurement of the model risk risks and vulnerabilities in the financial associated with different modeling system. choices (scenarios and impacts assessment); As the severity and frequency of climate events continue to increase, there is calculation of transition matrices increasing evidence to support the impacts of conditional upon the systematic these events on financial systems. Extreme component (economic cycle, lines of weather events usually result in the damage business, etc.); of physical assets, loss of life, productivity declines, property and casualty losses, and increased probability of defaults. incorporation of feedback loops and contagion effects; and The FSC is currently using stress testing tools that support the development of a risk-based an assessment / ranking of alternative supervisory regime. However, the stress mitigating actions to align the risk charges are not tailored to the risks for the profile of the insurance sector. insurance sector in Barbados. As part of this World Bank project, the Team performed a review of the current stress testing framework including: an assessment of parameter and key process stages: scenarios, impacts, mitigating actions, reporting, review validation; a review of the resilience of domestic insurance market to natural disaster shocks; 7 Stress Testing Framework Sensitivity Shocks A stress test typically involves shifting the values of individual risk factors and determining the effect of such changes on its business. Typical shocks would include: Changes in the loss distribution shape, Value-at-Risk ("VaR") losses, fat tails, identification of latent losses. Quantum of loss identified are very high for increase in correlation, liquidity risk and breaking down of risk calibration. Stress testing is a risk management technique used to evaluate the potential effects on an institution’s financial condition of a specific event and/or movement in a set of financial variables. The traditional focus of stress testing relates to exceptional but plausible events. Bottom Up and Top Down A bottom-up stress test is an exercise run by a supervisor or regulatory authority, where participating institutions are requested to perform the calculations. The supervisor provides the stress testing framework, methodologies, adverse stress scenarios, prescribed shocks and guidance to the application of the shocks. Participants then calculate the impact of the prescribed shocks on their balance sheet and capital requirements according to the provided guidance's using their own models. A top-down stress test is a stress test performed and run by a supervisor or regulatory authority. The supervisor determines the impact of a scenario directly based on the regulatory data provided by the insurers using its own framework, models and specifications (i.e. no calculations from individual institutions required). Bottom-up and top-down can be run in isolation but can also be seen as complementary exercises where top-down approaches can be used in a bottom up stress test for validation purposes. 8 Risk Profile - Barbados Barbados is one of the most populated and prosperous Caribbean nations, scoring high on many United Nations Human Development Index indicators. Tourism is the primary source of income for the country, which has suffered from the impacts of travel moratoriums during the COVID-19 pandemic. Being a mostly low-lying country, Barbados experiences flash floods, droughts, small-scale landslides, earthquakes, and fires. The country’s disaster and emergency management legislation and structure is strong, as is its data collection, but implementation of policy and deployment of these data resources are still lacking. Barbados is in the eastern most part of the Caribbean Archipelago. The island’s terrain is relatively flat and largely comprised of limestone rock. As an island state, Barbados is highly vulnerable to hurricanes and other natural hazards, and is particularly susceptible to the potential impacts of climate change, including coastal inundation and sea level rise, an increase in tidal and storm surge levels, coastal erosion, rising temperatures, changes in rainfall patterns, drought and more frequent and intense tropical cyclones. Barbados’ coastline is approximately 97 km in length with the majority of the island’s extensive coral life located in marine protected areas on the western coast. The population of Barbados is approximately 287,371 (2020) with 25% of the population living in coastal areas. The country was one of the main cultivators of sugarcane but has shifted its economy towards tourism and financial services. Barbados is already being significantly and directly impacted by the increase in climate- related extreme events, including hurricane frequency and intensity, droughts, and sargassum seaweed influxes which is aggravated by the spe cific location of the Barbados’ beaches, coastline characteristics and social and economic factors. Further systemic trends that are of concern are the anomaly in the frequency of marine heat waves observed over the past decade as well as sea level rise. Since 2010, hurricanes Tomas (2010), Ernesto (2012), Harvey (2017), and Elsa (2021) and the tropical storms Matthew, Maria, Kirk and Gonzalo (2020) have impacted the island, causing damage and disrupting lives. Also, depending on the size of the storms and because of the characteristics of the Barbados coastline, it experiences coastal impacts even when the eye of the storms is far away from the island. 9 Risk Profile - Barbados Before developing stress testing tools for the insurance sector in Barbados, one must consider the specific risk profile of the jurisdiction. Frequency vs. Severity Barbados has experienced 7 major storm events during the period from 1955-2015. Barbados has not suffered a direct hurricane hit, but does experience damage from passing storms nonetheless. These impacts include torrential rainfall that can cause extreme flooding and water-logged farmlands as experienced with Tropical storm Thomas in 2010. The below table ranks the estimated severity of earthquake and wind hazards to consider when developing a risk profile for Barbados at a national level. The insurance sector also must consider the exposure, frequency and severity of significant events when determining pricing and underwriting for insurance products. Capital EQ EQ Wind Wind Jurisdiction Stock (AAL) (PML) (AAL) (PML) ($million) Antigua & Barbuda 6,257 0.5 11.8 2.7 6.9 Barbados 14,037 0.2 8.9 0.3 12.4 Dominica 2,028 0.6 19.5 1.2 49.7 Grenada 4,536 0.2 8.9 0.2 16.6 St. Lucia 3,361 0.2 7.3 0.6 27.0 St. Vincent & Grenadines 2,645 0.1 4.5 0.3 15.0 St. Kitts & Nevis 4,112 0.7 21.2 1.0 30.5 EQ AAL = Annual Average Loss as a percentage of the total Capital Stock EQ PML = Probable Maximum Loss for a 1-in-250 year earthquake event as a percentage of the total Capital Stock. Wind AAL = Annual Average Loss as a percentage of the total Capital Stock Wind PML = Probable Maximum Loss for a 1-in-250 year windstorm as a percentage of the total Capital Stock. 10 Risk Profile - Barbados Before developing stress testing tools for the insurance sector in Barbados, one must consider the specific risk profile of the jurisdiction. Geographic Exposures Due to its low terrain, coupled with poor drainage and lack of inadequate storm water infrastructure, Barbados is susceptible to flooding. Low-pressure systems that bring torrential rainfall often cause flooding in areas of the west and south coast where approximately 25% of the population resides. Different disaster models capture various criteria in terms of exposure. The Global Assessment of Risk (GAR) framework evaluates total property exposures while the Global Earthquake Model (GEM) only captures certain residential and commercial risks that are particularly susceptible to earthquakes. The World Bank team developed their own Country Disaster Risk Profile (CDRP) for Barbados. The table above shows the geographic concentration of exposures comparing the GAR, GEM and CDRP profiles. The two maps show where the majority of property risks are located within Barbados. The values at the bottom show the total capital stock of the country as shown in millions of US dollars. Storm surges apart from erosion can also lead to flooding in low-lying areas, causing damage to coastal infrastructure crucial to the tourism sector. Barbados does not often experience drought, though the country was severely water stressed as a result of one drought that occurred in 2010. Coastal inundation and increased beach loss will have a significant impact on the economy, especially the tourism sector. Though uncommon, landslides can occur with heavy rainfall from passing low-pressure systems. Small-scale landslides have occurred in the Scotland district, which is located on the northeast portion of the country. Earthflows, slumps and debris flows are the main types of landslides that occur. 11 Risk Profile - Barbados Frequency vs. Severity The chart to the right evaluates the average frequency of occurrences by natural hazard type. While in some regions, low-intensity events are decreasing in frequency the increase in annual expected damages is caused by the increase in damages associated with very severe events. These estimates focus on the direct damages on physical assets and do not include the wider social and economic impacts in affected regions. In addition, they only consider the impacts from cyclones and exclude changes in losses due to surge and precipitation induced flooding. Barbados is located along the hurricane belt where most transatlantic hurricanes pass, which makes Barbados vulnerable to all the major impacts associated with them, including storm surge and flooding. Hurricane season takes place during the months of June to November with increased frequency during the months of September to November. Barbados is also at risk to floods, droughts, storms that are not classified as hurricanes, and occasional landslides. The table shows historical losses in thousands of US dollars at the time of the event while the trended losses are adjusted for inflation. The map shows the historical path of various cyclone events surrounding Barbados. 12 Risk Profile - Barbados Modeling Infrastructure Resilience In terms of exposure to catastrophic risks, certain assumptions were made relating to the types of structures in Barbados. The hazard mapping showed a split of approximately 55% residential structures and 45% commercial structures. The primary risk exposure for residential structures is Masonry, Reinforced Masonry (MR) while the primary construction type for commercial business is Masonry, Confined Masonry (MCF). The structural portfolio of risks will affect how resilient a jurisdiction is to certain catastrophic exposures. Each different type of exposure is modeled to show damages under various scenarios which is then reviewed to make assumptions for general losses to the overall economy and then specifically addressing the resilience of the insurance sector. The hazard curve comparison looks at various models over time including those performed by the University of the West Indies (UWI). 13 Risk Profile - Barbados Model Development The final step is to overlay the hazard mapping profile against the exposure data within Barbados. This results in the following estimated losses to the total economy and to the insurance sector: The total property exposure is US$ 13.5 billion within the modelled results. Insurance penetration is high for the region and estimates per industry sources were the following: Assumptions Deductible 2% by TIV Penetration 5% for wooden or unknown building typologies 10% for unreinforced masonry 50% for all other typologies This process results in the following annual exceedence probability (AEP) which shows the level of expected losses based upon the severity of Tropical Cyclones (TC) or Earthquakes (EQ) within Barbados. 14 Risk Profile - Barbados Model Outputs The Annual Average Loss (AAL) calculates the expected probability of loss for a one-year time horizon. The exceedence curves are then modeled to look at expected loss amounts in total and then also the insured losses in millions of US dollars. The model then develops a Probable Maximum Loss (PML) estimate at a certain point along the loss curve. The model results in an estimated outflow (including claims and related expenses) in the greater of a 1 in 200 catastrophic event that should be reviewed by the FSC compared to overall industry statistics. Caribbean insurers are well aware of the need to ensure that their protections are adequate concerning catastrophic exposures. The FSC does not set any requirement to model exposures or purchase cover up to defined PML limits. Most insurers are guided by their reinsurance intermediaries or use commercial models. The typical event limit for Barbados is generally thought to be approximately 20% of aggregate exposures. Some regional insurers model hurricane catastrophe exposure scenarios on the basis of a windstorm with a 200-year return period affecting multiple islands. 15 Market Assessment Non-life business accounts for about 67 percent of total annual premiums, reflecting the role played by non-life insurance in protecting personal and real property. High non-life insurance per capita penetration ratios are a common feature of Caribbean countries, primarily reflecting the cost of property insurance due to the hurricane and/or earthquake elements. Seventy-one percent of total non-life insurance business comes from property and motor business. Currently, insurance legislation in Barbados does not specifically require insurers to establish an Enterprise Risk Management (ERM) framework or to The GME includes an annual assessment by the perform an Own Risk and Solvency IAIS of: Assessment (ORSA). potential systemic risk arising from An ORSA is an internal process undertaken by sector-wide trends with regard to an insurer or insurance group to assess the specific activities and exposures (Sector- adequacy of its risk management and current Wide Monitoring, or SWM); and the and prospective solvency positions under possible concentration of systemic risks normal and severe stress scenarios. at an individual insurer level, using an assessment methodology as one of the ERM can be defined as a process, effected by inputs (Individual Insurer Monitoring, or an entity’s board of directors, management, IIM). and other personnel, applied in strategy setting and across the enterprise, designed to The Financial Services Commission should identify potential events that may affect the consider implementation of these entity, and manage risks to be within its risk international methods and approaches for appetite, to provide reasonable assurance measuring systemic importance in the regarding the achievement of entity insurance sector and the broader financial objectives. sector in Barbados and the region. International Stress Test Models In 2019, the International Association of Insurance Supervisors (IAIS) adopted the Holistic Framework (HF) for the assessment and mitigation of systemic risk in the global insurance sector. A key element of the HF is the Global Monitoring Exercise (GME), designed to detect key risks and trends and the potential build-up of systemic risk in the global insurance sector. 16 International Stress Testing Model The IAIS has created the Individual Insurer Monitoring ("IIM") process which calculates a score for each participating insurer. Stress Test The IIM assessment score provides an indication of the extent of the possible build-up of systemic risk within an individual insurer. The IIM assessment methodology consists of five categories and 14 indicators. The scores are calculated for each participating insurer, using specified formulas. Where specific values or factors are not provided, insurers are encouraged to model other risks to a 99.5% confidence level over a one-year time horizon or a one-in-200 year event. The model shows a pre-stress baseline for three types of cash flows: investing, financing and operating. These values are then stressed using predefined factors as well as internally modelled figures from industry participants. The IAIS model also projects cash inflows and outflows under the liquidity stress test (LST). The LST has three time horizons: 30 days, 90 days and 1 year. The liquidity stress test is applied to the three categories of cash flows (investing, financing and operating), regardless of the fact that both the operating and financing cash flows are assumed to be more stable during a financial crisis or a period of stress. The current liquidity stress test scenario simulates a decline in broad asset categories such as returns on government bonds, structured finance securities and corporate bonds as well as equities, as measured by a decline in main equity indices. In addition, there is a number of macroeconomic variables such as inflation, unemployment, GDP growth, disposable income growth, mortgages rates and real estate price indices that are applicable to stress all types of cash inflows and outflows related to assets and also liabilities. Assume going concern principle applies so there is no run-off of the current (at the reporting date) balance sheet instruments. The stressed parameters in the Tables are represented with absolute changes in % or yields. An example of the application: If the baseline39 Real GDP Growth is 4.0%, the stressed GDP Growth parameter is 4.0%-1.5% = 2.5% 17 International Stress Testing Model Macroeconomic Indicators The IIM assessment score provides an indication of the extent of the possible build-up of systemic risk within an individual insurer. As outlined in the GME document, the IIM assessment methodology consists of five categories and 14 indicators. The scores are calculated for each participating insurer, using specified formulas. The stressed parameters in the Tables are represented with absolute changes in % or yields. An example of the application: If the baseline Unemployment Rate is 7.0%, the stressed Unemployment Rate is 9.1%. Market volatility indices may not be present in all jurisdictions. Estimates may be included within the model based on historical stressed assumptions. 18 International Stress Testing Model Treasuries and other Investment Vehicles If stable company growth is planned, assets and liabilities (and their related cash flows) that retire during the following 30 days/ 90 days/ 1-year may be replaced by new ones taking into account a going concern perspective under stress. Include premiums that will be earned during the following periods according to required time horizons (eg. current in- force policies + future business that will be earned in the following 12 months for the 1- year time horizon). If a business increase or a run-off is planned (indicate this run-off in the column “Explanations”), stressed projected cash flows should be adjusted accordingly. Cash flows should be calculated in a similar was as the cash flow financial statements are prepared. The stressed parameters in the Table are represented with absolute changes in % or yields. An example of the application: If the baseline Mortgage Rate is 3.0%, the stressed Mortgage Rate parameter for 12- month time horizon is 3.0%+1.3% = 4.3% If the baseline BBB Corporate Yield is 5.0%, the stressed BBB Corporate Yield parameter for 3-month time horizon is 5.0%+1.5% = 6.5%. 19 International Stress Testing Model Equity Index Supervisors may prescribe some standardized tests in order to obtain a measure of consistency and for baseline monitoring purposes. These tests, however, should neither inhibit an insurer from undertaking its own thorough review of the inherent risks in its business, nor discourage an insurer from adopting an effective, comprehensive, risk-based approach to business management. Various considerations are likely to determine the nature and extent of tests required. They include the insurer’s: solvency position; lines of business and distribution systems; current position within the market; current position within the group; investment policy; business plan; and general economic conditions. The stressed parameters in the Table are represented with absolute changes in % or yields. An example of the application: If the baseline Industrial Equity Index is 30.000, the stressed Industrial Equity Index parameter for 12-month time horizon is 30.000*(100%-40%) = 18.000 Lapse Event The stressed Mass lapse ratio for retail life insurance policies’ parameter for 3 -month time horizon is set at 5.00%. 20 Enterprise Risk Management The IAIS has developed Insurance Core Principles (ICPs) relating to the supervision of the insurance sector. ICP 16, Enterprise Risk Management (ERM) for Solvency Purposes, sets out an international standard that calls for insurance supervisors to require an “insurer’s ERM framework to provide for the identification of all reasonably foreseeable and relevant material risks and risk interdependencies for risk and capital management”. The Commission should encourage regulated entities to adopt ERM practices that consider stress and scenario testing as part of the development of Own Risk and Solvency Assessments (ORSAs). ORSA has been defined as a tool of the ERM system that requires insurance undertakings to assess properly their own short- and long- term risks and the amount of own funds necessary to cover them. In many jurisdictions, existing principles- based ERM requirements appear broad Undertaking risk modelling and the enough to capture certain risks. However, associated governance processes can since the FSC does not currently have these facilitate helpful discussion on risk strategy requirements, this should be incorporated at within an insurer, which some may argue as the same time. There are many guidance being more important than the numerical documents that have been prepared by the results from the models. IAIS which could assist the FSC in the adoption of these standards. Given uncertainties surrounding the modelling of future outcomes, models can be Some supervisors have found less formal or less most useful in supporting discussions within legally binding tools to be effective and insurers on their risk strategy and enhancing efficient in setting out timely and realistic their understanding of the risks. Scenario expectations to insurers on how they should outputs can still be useful, even if the results manage emerging risks. are subject to uncertainty. Amendments to the Insurance Code should be considered as part of a larger solvency modernization program which would include ERM, ORSA, macroprudential surveillance, on- site inspections, stress and scenario testing. Supervisors are consistently expecting insurers to address emerging risks in their ERM frameworks. 21 Principles for integrating stress testing in the ORSA Potential impacts of various risk factors should be incorporated into possible scenarios where these risks are material Industry Risk Assessment Models Regulatory measures can play a crucial role in motivating insurers to establish and enhance their ability to assess their risk exposures. For many years, insurers have been able to quantify weather-related risks for pricing and reserving purposes. They have usually been able to withstand severe episodes of weather- related insurance claims and even to underwrite them profitably. Nevertheless, such models may not explicitly nor accurately capture future possibilities and In general, supervisors have not prescribed the past may not be a good guide to the specific requirements or signal future. Moreover, catastrophic risks could expectations on the approach insurers should adversely affect not only the insurance take to quantify certain types of risk. This is liabilities of insurers, but also other parts of largely due to the embryonic state of the their balance sheets, including techniques and models. investments. Supervisors should also consider some degree Risk quantification techniques or models of flexibility for insurers, allowing them to used by insurers to assess their risk exposures select the most appropriate risk models or are still at an early stage of development. This approaches to assess certain risk exposures. In is because, for non-life insurers, for example, many jurisdictions, smaller insurers were even with a long history and experience of exempted from some of these requirements or modelling natural catastrophe risks, the were given more time to develop these insurance industry is faced with significant processes. challenges when attempting to quantify or estimate the frequency and severity of risks. Any methodology would necessarily need to For life insurers, it is complex to assess the consider the likelihood and impact that each effect of future risk factors on longevity risk. material and relevant risk could have on an On the other hand, qualitative measures are insurer both under normal and stressed more advanced, particularly governance- situations. Modelling risk impacts on future related actions such as raising awareness probabilistic assumptions under stressed among board and senior management on conditions is difficult. insurers’ risk exposures and developing internal technical capacity to explore how business strategies could be impacted by emerging risks. 22 Scenario Analysis and Stress Testing In general, insurers typically use stress testing and scenario analysis to assess their risk exposures. In ICP 16, the IAIS considers stress tests to be a risk assessment tool that “measures the financial impact of stressing one or more factors which could severely affect the insurer. While scenario analysis considers the impact of a combination of circumstances to reflect historical or other scenarios which are analyzed in the light of current conditions, these may be conducted deterministically using a range of specified scenarios or stochastically, using models to simulate many possible scenarios, to derive statistical distributions of the results. If efforts are successful in transitioning to a less carbon-intensive future, an insurer’s In practice, the distinction between the two risk transition risk may be heightened quantification techniques is not entirely clear- if it holds stranded assets. On the other hand, cut. There is a spectrum of approaches, ranging if such efforts are unsuccessful, insurers may from stressing only one variable to stressing face greater physical risk arising from multiple variables in a coherent way that weather-related impacts from the higher reflect their dependencies. global temperature. In both climate scenarios, liability risk may increase due to As a small island, Barbados has a particular greater public awareness of climate issues, exposure to climate risk, for example, which thus increasing the pressure to hold could be incorporated within stress test or corporates and executives accountable for scenario analysis. Most analyses are likely to climate inaction. cover physical risks only. Capturing transition risks is less common, and liability risks are rarely Once a risk assessment technique has been covered. This most likely reflects the level of determined, insurers typically translate their familiarity of insurers with these different future climate assumptions into specific climate risks as well as the priority they place on each. stress factors. This requires being clear on the aim of the risk assessment exercise, for Nevertheless, to avoid any risk management example to determine the potential impact on blind spots, there may be a need to consider earnings volatility, capital position or climate risk in a stress test or scenario business model viability. Significant technical analysis. Importantly, the various potential and multi-disciplinary expertise is needed in climate futures could lead to very different this process. In addition, and in the absence manifestations of these risks. of established methodologies, there will be heavy reliance on expert judgment. 23 Scenario Analysis and Stress Testing Since ORSA reports are the company's "Own" assessment of risk, most authorities do not typically opine on the assumptions or stresses used by insurers but some do require insurers to provide justification. Apart from technical difficulties in coming up with a plausible stressed scenarios and translating it into concrete stress factors, it is a challenge to evaluate whether the stressed scenario is sufficiently severe. To answer this question, insurers may need to engage with specialists to ascertain their stress assumptions. Despite these difficulties, a stress test or scenario analysis can be informative even if the appropriateness of the stress severity is uncertain. Insights can be gained into the direction of risks and For regulatory or economic capital purposes, vulnerabilities of firms from scenarios with insurers typically consider a one-year shock varying degrees of severity. Expert judgments period, effectively dealing with what could go will vary when considering “what if” wrong over the forthcoming year and how much scenarios. capital they will need to keep running their business. On the other hand, insurers may need Most jurisdictions expect insurers to consider to consider a longer shock period in a risk longer time horizons than those typically used assessment model in order to capture the for calibrating regulatory capital requirements. nature of climate that could materialize over This reflects the very long- term nature of periods longer than a year. To be consistent, certain insured risk exposures. While insurers, appropriate management and mitigation especially life insurers, are used to considering actions could be considered over such longer risks over 50 years or more, assessing risks may periods of time. require equally long, if not longer, time horizons. It is helpful to distinguish between An important feature of a risk assessment two aspects of time horizons, as defined by the model, including stress test and scenario IAIS, as follows: analysis, is how it allows for risk mitigation actions, as this could have a significant effect the shock period, which is the period on the results. The most common risk over which a shock is applied to a risk; mitigation actions considered in risk models and are reinsurance arrangements. Other examples of risk mitigation actions, although less common, include recognition of the effect period, which is the period diversification benefits (e.g. due to different over which the shock applied to a risk geographical exposures), ability to will impact the insurer. recapitalize, changes to business plan, changes to investment strategy and ability to reprice, allowing for customer affordability. 24 Scenario Analysis and Stress Testing There is a wide range of quantitative outputs that can be obtained from insurance risk stress testing and scenario analysis. The following indicators show the different ways insurers could be affected by insurance risks: claims and investment losses; profitability; capital requirement; capital resources; average annual loss change; aggregate or occurence exceedence; market value of investments; and value-at-risk or tail value-at risk. Data and Methodology Challenges How frequently insurers undertake stress While data on certain catastrophic events may testing or scenario analysis varies across have been collected for decades, in particular for jurisdictions, although this is typically on an some types of extreme weather events, this is not annual basis as part of an ORSA. necessarily the case for other sources of risks. In certain jurisdictions, larger insurers are Because some risk factors lack granularity, their use required to undertake stress tests and may be limited for an insurance firm attempting to scenario analyses as frequently as assess the risks of specific investments. Some semiannually, at quarterly intervals or when concrete examples of data deficits include the their risk model and/or portfolio changes following: significantly. key metrics driving risks such as records of Qualitative safeguards and reassurances are claim data on particular types of real estate important to ascertain the reliability of the structures; assessment results, particularly if the results reliable forecasts of socio-economic and are to be used for decision-making. Just as for customer behavior changes due to changes other types of risk, board and senior in country risk profile; management oversight and critical challenge risk exposure amounts of insurers (e.g. over validation of insurers’ risk models are investee entities’ and insured’s reliance on important. certain business activities. Until further progress is made to improve data Challenges that insurers face in assessing risks availability, reliance on estimates will be can be categorized into data and unavoidable. methodology, expertise and resource, and governance issues. 25 Barbados - Stress Test Model Each entity is required to conduct three broad scenarios under the themes “General Parameters”, “Institution Specific Tests” and a “Specific Event Test”. General Parameters The first scenario is a general test on key parameters. These five (5) tests provide information on an entities' ability to withstand shocks of a specific magnitude. Table 1 outlines the parameters for five (5) tests, which are not expected to be conducted simultaneously. Metric - Metric - Type of Business Barbados Regional Test 1: Increase in Gross Claims Property +100% +100% Motor Vehicle +100% +100% Personal Accident, Health, Accident & Sickness +200% +200% Ordinary Life, Industrial Life, Credit Life, Group Life +150% +250% Test 2: Technical Provisions Strengthening of Technical Provisions +3% +10% Test 3: Default by Related Parties Losses on Equity Investments in Subsidiaries and -40%/-40% -100%/-100% Affiliates; Due from Subsidiaries and Affiliates] Test 4: Increase in Expenses Operating and other expenses [not including +10% +10% acquisition expenses or reinsurance commissions] Test 5: Economic Downturn Impacts of Economic Downturn +10% +20% These tests represent more of a sensitivity analysis rather than a formalized stress test. They specifically look at accounting data in the supervisory reporting template and apply certain shocks to each individual type of business or across business lines. This analysis can be useful to find outliers in the insurance sector for one particular type of shock. The regional values are generally calibrated at levels consistent with those used for Financial Sector Stability Assessments. Recommendation: The Financial Services Commission should consider calibrating actual industry results to apply tailored and specific shocks. Specifically, the FSC should compare more conservative shocks on mortality products, technical provisions as well as defaults by related parties. 26 Assumptions The development of natural catastrophe scenarios are generally supported by certain qualitative risk assessments of past catastrophes. While these may be singular events, insurers must consider the associated events triggered by these disasters and incorporate such into their models. An example of the current scenario test would incorporate simultaneous moves in a number of risk factors, usually related to those experienced in the aftermath of a natural disaster. Each entity is required to outline the assumptions used to further explain the results of the shocks. Entity Specific Assumptions Increase in gross property claims of 200% given the portfolio balance being more heavily weighted to property; Reduction in Gross Written Premium ("GWP") due to economic challenges in territory and new entrants to the sector pushing rates down; Reduction in investment income; Increase in reinsurance treaty costs; and No increase in property catastrophe reserves if excess reserves are already held. Category 5 Hurricane - Assumptions Increase in gross property claims of 500% considering there has been no recent Major hurricane in Barbados so historical claims will look small compared to a CAT event year; Loss exposure and PML results will be from the most recent RMS modelling data; Independent application of each annual stress as though there was no hurricane in previous year of the projection due to reinsurance market changes from one year to the next; Existing reinsurance structures are expected to be the same throughout the forecast period even though limits and retentions may change due to market conditions; and Catastrophe claims are assumed to have been recorded and paid-in-full by year-end so no gross impact on outstanding reserves or recoverable from reinsurers. 27 Stress Testing - Objectives Sensitivity Analysis - Micro vs. Macro-Prudential Objectives The objectives drive the design, methodology and application of each stress test exercise. Micro-prudential objectives are designed to assess the resilience of individual insurers or insurance groups to adverse scenarios while macro-prudential objectives assesses the system wide resilience to financial, economic and insurance shocks and the potential spill over to other markets generated or amplified by the insurance sector Micro-Prudential Objectives Assess individual sensitivity to specific shocks; Assess individual vulnerabilities to adverse economic and financial conditions, which can be used to trigger inspections or issue recommendations; Assess individual capital adequacy under adverse scenarios; Enhance understanding of insurance sector vulnerabilities; and Foster individual risk management and stress testing capabilities. Macro-Prudential Objectives Supervisors can assess resilience of the insurance sector and of individual insurers that, due to their nature, scale and complexity, might generate or amplify systemic events against certain stress scenarios. Supervisors are able to assess potential spill-over effects to other parts of the financial system and the real economy stemming from common reactions of insurers against stress scenarios. Approaches Different approaches related to the conceptual elements of stress testing and can vary concerning complexity (both methodological and operational), validation of results, interpretation of results, and comparability. 28 Stress Testing - Scenario Design Instantaneous Stress Scenarios vs. Multi-Period Stress Scenarios Instantaneous stress scenarios are assumed to be applied as a one off shocks to the balance sheet at reference date while multi-period stress scenarios cover a specified scenario over a horizon of several periods with the development of key financial and economic variables described for each period An example of a multi-period stress scenario would be a macroeconomic financial crisis scenario with specific triggering events (e.g. abrupt reversal in risk assessment on financial markets, implying a material increase in bond yields) with subsequent real economy spill over effects over the next years (e.g. affecting equity and real estate prices and policyholder lapse behavior) Scenario Design When designing a stress test scenario, the event or series of events should have a defined quantification of severe but plausible events. Another consideration concerns whether the scenarios are to be looked at using historical or forward-looking scenarios. Shocks and their Application Market shocks – risk of adverse movement in the values of assets and liabilities as a result of market movements such as interest rates, foreign exchange ates or the repricing of risk premia Market Shocks The following list provides examples of items that are stressed under current models in Barbados which will be compared against a variety of international standards: government bond yields; corporate bond yields; equity prices; swap rates; residential real estate prices; commercial real estate prices; loans and residential mortgage-backed securities (RMBS) yields; other assets prices (private equity, hedge funds, real estate investment trusts (REITs), commodities); and a Downgrade of credit ratings. 29 Stress Testing - Scenario Design Insurance Shocks Insurance specific shocks will demonstrate the risk that an inappropriate underwriting strategy is adopted or that unexpected losses arise even when an appropriate strategy is adequately implemented. The following list provides examples of items that are stressed under current models in Barbados which will be compared against a variety of international standards: Catastrophic Risks Longevity/mortality; Lapse/surrender; Life expense risk; Other life risk: Disability/Morbidity; Revision; Pandemic; and Provision deficiency (claims and expense inflation). Once the appropriate shocks have been selected and the time horizon of the scenarios are finalized, the FSC collects the data in order to analyze it for consistency, reasonableness and to determine any potential outliers. Finally, feedback is requested as part of macroprudential surveillance and on-site inspections. 30 Stress Testing Parameters The Financial Services Commission currently utilizes a stress testing model that provides a flexible approach although they may want to consider alternative stress metrics that are more appropriate for their specific market conditions. Increase in Gross Claims Internationally, there are many assumptions and models that are utilized to quantify stress scenarios and sensitivity analyses. The Commission's model draws from other jurisdictions in the region that have implemented a stress testing regime. The current stress metrics are following the stress test of a pandemic event combined with a major hurricane. The normal stress factors for these lines of business may be considered in multiple ways. Metric - Metric - Type of Business Barbados Regional Property +100% +100% Motor Vehicle +100% +100% Personal Accident, Health, Accident & Sickness +200% +200% Ordinary Life, Industrial Life, Credit Life, Group Life +150% +250% However, regional models for normal risk charges utilize the following approach: Type of Business Metric - Regional General increase in claims +25% Increase in property claims due to catastrophe +50% Individual insurer catastrophe property charge +500% Increase in health and life claims due to pandemic +100%/+200% The stress testing framework in the region uses a 25% charge under normal conditions for property business, but then either uses a 50% or up to a 500% charge for an individual insurer. The Commission should review annual returns after catastrophic events historically to determine the distribution of loss ratios and standard deviations for each line of business. There are issues with the financial reporting format in Barbados that will be discussed further in this report. 31 Stress Testing Parameters Scenario Test Model The scenario test model looks at certain shocks to economic indicators to project potential issues within the insurance sector. Changes in the interest rate reflect the economic value change of interest-sensitive assets and liabilities. Economic Downturn Scenario An assumption of the stress testing model is that the country experiences an economic downturn due to an exogenous shock that leads to a rise in unemployment, a fall in equity and housing prices, low consumer confidence and a decline in investments. For some of these risk charges, the team also evaluated international best practices that should be considered by the FSC for future modifications to any stress testing model. Metric - Metric - Scenario Test Model Barbados Regional +/- 100-200bps Change in Interest Rate 0-3.0% parallel shift Losses on Mortgages -25% -30% Change in Real Estate Market Values -25% 15-50% by class Change in Equity Market Values -30% -20/30% Change in Technical Provisions 3% 10% Failure of a Bank (%LGD) 30% 30% Default by Related Parties 30% 100% Equity Investments in Subsidiaries and Affiliates 50% or 100% 100% Due from Subsidiaries and Affiliates 50% or 100% 100% Metric - Metric - Adverse Expenses Barbados Regional Operating Expenses 10% 5-10% Other Expenses 10% 5-10% Government Bond Devaluation 20% or 40% 40% The stress testing framework utilizes appropriate shocks compared to other supervisors in the region. However, there may be opportunities for more granularity depending on specific risk charges. 32 Stress Testing Parameters Market Risk Although the business models between banks and insurers are fundamentally different, there are parallels when it comes to the measurement of certain factors for stress testing. Market risk has some similarities depending on the structure of assets that are subject to the stressed factors. Typically, market risk factors will include assumptions that include interest rates, equity prices, corporate spreads, and sovereign spreads for a mild and a severe adverse scenario, together with a mass lapse event in the life business and the realization of the largest probable maximum losses (PML) on a single (man-made or natural) catastrophic tail event. The following represents the current stress testing framework for Barbados and considerations compared to regional metrics as well as international best practices. Metric - Metric - International Type of Risk Factor Barbados Regional (basis (%shock) by (basis points) points) maturity year Moderate Yield Curve Shift Down -150 bps -150 bps 30% Severe Yield Curve Shift Down -300 bps -300 bps 75% Moderate Yield Curve Shift Up +250 bps +250 bps 25% Severe Yield Curve Shift Up +500 bps +500 bps 70% The stress testing framework for Barbados suggests a fixed yield curve shift based upon total sovereign exposures while international practices generally look at the duration of the investments to determine the potential impact. The current risk premium rate minus the treasury rate is approximately 7.55%. Therefore, the yield curve risk factors would be 20%, 40%, 33% and 66% respectively. It is interesting to note that generally upward shifts to the yield curve are given lower shocks internationally compared to downward shifts. Because each jurisdiction has different Treasury rates and a different market risk profile, stress factors are generally applied on a percentage basis across the risk spectrum rather than a fixed number of basis points. Equities, Real Estate, Commodities Insurance firms are exposed to risks commonly found in other financial institutions, including credit risk, operational risk, and market risk related to equity investments as well as movements in interest rates and exchange rates, all of which are highly correlated with changes in economic conditions; however, insurance risk (e.g., mortality, morbidity, casualty and liability risks) is largely idiosyncratic and generally independent of the economic cycle, which allows them to realize diversification gains (through underwriting inversely correlated risks, such as death insurance and pension insurance, risk pooling, or reinsurance/retrocession). 33 Stress Testing Parameters Equities, Real Estate, Commodities Market risk is concerned primarily with the adverse movement in the value of an insurer’s assets and liabilities, both on-balance sheet and off-balance sheet, whose value may be affected by market movement. For insurers, it is the extent to which an adverse movement in the value of the assets as a consequence of market movements, such as interest rates, foreign exchange rates, equity prices, etc., is not offset by a corresponding movement in the value of the liabilities. Because each jurisdiction has different Treasury rates and a different market risk profile, stress factors are generally applied on a percentage basis across the risk spectrum rather than a fixed number of basis points. Metric - Investments IMF - FSAP Various Supervisory Approaches Barbados -20% (up to 50%) EIOPA: 45% - EU Equity Risk -20% -30% (up to -50%) 43% - US 50% - Emerging Private Equity/Hedge Fund: 45% Commodities: 40% 15% (varies) EIOPA Real Estate Risk -25% -20% (up to -40%) Residential: -8.4% Commercial: -17.4% REITS: -51% -40% (-20%) EIOPA: Corporate Bond Yields: +71- Bonds -15% Catastrophe 269 bps Scenario The liability profiles of different types of insurers place different requirements on the assets they hold. Non-Life insurers focus more on ensuring they have the liquidity to pay claims should those arise sooner than anticipated. Life insurers are more concerned with matching returns to their future obligations. Consequently, non-life insurers tend to adopt an investment strategy based on fixed income assets with few equity/growth assets. Life insurer investments focus more on growth and hedging-type assets. This divergence is reflected in their varying stress testing approaches and the factors they emphasize. 34 Stress Testing Parameters Technical Provisions Technical provisions of an insurer have to be adequate, reliable, objective and allow comparison across insurers. Stress testing the deterioration of technical provisions will include the following: the adequacy and uncertainty of the technical claims provisions, (e.g., outstanding claims, IBNR and claims handling expense reserves); the adequacy of other underwriting provisions, (e.g., the provisions for unearned premiums and unexpired risks); the frequency and size of large claims; possible outcomes relating to any disputed claims, particularly where the outcome is subject to legal proceedings; the effects of inflation; social changes resulting in an increase in the propensity to claim or to sue; and other social, economic and technological changes. Insurance risks specifically focus upon the impact of the underwriting and claims functions on an insurer’s premiums and technical provisions. Insurance risks may be categorized as underwriting risk, catastrophe risk, or the risk of deterioration of technical provisions. Some national supervisors split out impacts for technical provisions and premiums by type of business written. Type of Risk Factor Life Non-Life Health EIOPA - Technical Provisions 0.6% 3.6% 0.0% EIOPA - Premium 5.5% 3.8% 0.0% UK PRA - Technical Provisions 1.25% 2.0% 0.2% UK PRA - Premium 8.0% 6.0% 2.0% The above risk factors can be applied to both technical provisions and to written premiums in a simplified model. The FSC should consider the frequency and size of large claims as well as possible outcomes of disputed claims or the effects of inflation. Social changes may also impact the propensity to sue depending on the litigious nature of the population. 35 Stress Testing Parameters Credit risk relates to the possibility that a counterparty will fail to perform its obligations. Insurers’ counterparties may include debtors, borrowers, brokers, policyholders, reinsurers and guarantors. Counterparty Credit Risk Factors to consider include, but are not limited to: the collapse of a reinsurer or several reinsurers on the insurer’s reinsurance program and the subsequent impact this may have on outstanding reinsurance and IBNR recoveries; a deterioration in the credit worthiness of the insurer’s reinsurers, intermediaries or other counterparties; the degree of concentration of business with reinsurers of particular rating grades; the degree of credit risk concentration, e.g., exposure to a single name or counterparty; deterioration in the extent and quality of collateral; greater losses from bad debts than anticipated; and guarantees given by the insurer of the performance of others, whether under insurance contracts or otherwise. Metric - Metric - Credit Risk International Barbados Regional 15% or Failure of a Bank 30% 30%% Downgrade by 2 notches Default by Related Parties 30% 50% or 100% Varies Equity Investments in Subsidiaries and 50% or 100% Varies Affiliates 100% 50% or Due from Subsidiaries and Affiliates 100% Varies 100% 15% frequency Failure of a Reinsurer (%LGD) 40% 40% 2% severity The stress testing framework uses a standardized factor approach depending on the shock value to various counterparties. There are various models and methodologies concerning counterparty credit risk that vary from basic factor approaches to complex financial modeling tools that look at the likelihood of a stressed counterpart to fully cover their liabilities. 36 Stress Testing Parameters Liquidity risk relates to the possibility that an insurer will be unable to realize assets to fund its obligations as and when they fall due. Understanding whether an insurer’s cash flow is sufficient to meet its commitments to policyholders and other creditors is fundamental. Liquidity risk can arise out of a number of factors that should be included within any stress testing process. For the insurance sector, liquidity risks can arise from any mismatching between expected asset and liability cash flows; the inability to sell assets quickly; the extent to which the insurer’s assets have been pledged; cash-flow positions generally of the insurer and its ability to withstand sharp, unexpected outflows of funds via claims, or an unexpected drop in the inflow of premiums; and the possible need to reduce large asset positions at different levels of market liquidity, and the related potential costs and timing constraints. Metric - Risk Metric - Regional Barbados Increase in Non-Life Claims (excl. Liability) +200% +100% Non-Life (Liability) +10% +25% Increase in Life Claims (excl. Annuities) +100% +25% or +200% Lapse Shock N/A +250% Life Sector - Targeted Risks Internationally, there are specific risks based on certain product characteristics of the life sector. Type of Business Metric - International Lapse Shock - EU 20% Permanent Lapse Shock 50% cap at 100% lapse rate Permanent Lapse Decrease 50% with absolute change at 20% 30% for policies with surrender strain Mass Lapse Event 70% for non-retail business Mortality Shock 10-15% 37 Stress Testing Parameters Operational risk is the risk arising from failure of systems, internal procedures and controls leading to financial loss. Operational risk also includes custody risk. Operational Risk While the application of stress tests to operational risks may not be immediately obvious, the insurer should at least be able to demonstrate that such risks have been considered and that appropriate plans and procedures exist to adequately deal with an adverse scenario. Operational risks may be very difficult to identify and measure. Metric - Metric - Risk Shock Barbados International Increase in Expenses 10% 5% Impacts of Economic Downturn 10% 10% Operational Risk - EU N/A 20-30% Operational risk factors to consider include, but are not limited to: the adequacy of an insurer’s business continuity management (BCM) plans; the adequacy of an insurer’s disaster recovery planning (DRP), e.g., the potential failure of back-up systems, or failure in the efficiency and effectiveness of off-site back- up facilities; the possibility of fraudulent activity occurring that may impact upon the financial condition or operational situation of the insurer; the technological risks to which the insurer may be exposed, e.g., those relating to both the hardware systems and the software utilized to run those systems; the reputational risks to which the insurer may be exposed, e.g., the impact on the insurer if its brand is damaged, resulting in a loss of policyholders from the underwriting portfolio; the marketing and distribution risks to which the insurer may be exposed, e.g., the dependency on intermediary business; the possibility of political interference, e.g., the confiscation of assets, restriction of movement of funds in an emergency situation or legislative changes, such as changes in taxation or mandatory coverages; the impact of legal risks, e.g., the imposition of fines, or the risk that policy wording may be interpreted more broadly than intended; the possible impact of any outsourcing difficulties, e.g., third-party providers failing to perform in accordance with their contractual obligations; and the failure of general personnel management controls, e.g., the impact of an underwriter exceeding authority limits. 38 Regulatory Reporting The Financial Services Commission does not have an integrated electronic filing system that has automatic cross-checks. Many of the reporting tools require regulated entities to enter information manually which can increase reporting risks. The FSC periodically reviews reporting requirements as part of its annual monitoring plan. The FSC requires insurance companies to submit financial statements, and financial and operational reports on an annual and a quarterly basis. All annual financial statements must be audited, and quarterly statements must be reviewed by the auditor. In addition, an Annual Statement prepared by the company based on its financial statement submitted to the FSC for audit. Lastly, unaudited quarterly statements are likewise submitted to the FSC for monitoring and assessment. Off-site Analysis Off-site monitoring and analysis largely focus on the financial position of insurers, and trends in financial position and operations, including compliance, and market conduct. The FSC looks at quarterly and audited annual financial statements, annual reports, financial statement notes, external audit reports, actuarial report and the latest off-site analysis report. 39 Regulatory Reporting The Financial Services Commission does not have an Early Warning System ("EWS") or triggers for early Intervention and the FSC does not carry out risk-focused on-site inspections for all insurers in the jurisdiction. A problem was found during the development of the stress testing tools. Due to the significant reliance on reinsurance, net earned premiums for some lines of business (e.g. property) were actually being reported as negative values because while regulatory reports capture direct written premiums, assumed premiums and ceded premiums, they only reported claims on a net basis against earned premiums. Recommendation: The FSC should modify their financial reports to include total gross (gross of reinsurance) technical provisions which are held for the purpose of fulfilling insurance contracts (including policyholder dividends, funds held pursuant to reinsurance treaties, future policy benefits, policyholder account balances, loss reserves, asset valuation reserves and interest maintenance reserves related to insurance products, and unearned premiums reserves and excluding advance premiums received). Claims should be reported before considering any outgoing reinsurance. Technical provisions related to assumed premiums (ceded from other insurers) are included. Reinsurance recoverables (on the asset side) should not be deducted from gross technical provisions. Reinsurance recoverables should not be taken into calculation of gross technical provisions. Once the gross technical provisions are calculated, a separate line item would include reinsurance recoverables and would provide claims on both a gross and a net basis. With this information, the FSC could better evaluate the reliance and impact that reinsurance has on their industry. Technical provisions would be split both life and non-life (including health) business lines. 40 Regulatory Reporting In the financial reporting framework, Row 7 should be broken out for gross claims and net claims after the benefits of reinsurance. This will allow the FSC to better evaluate pricing adequacy by insurer by line of business. The FSC collects a significant amount of data, but because of the regulatory reporting structure, there are highly unusual values which also make the development of stress testing tools more problematic. 41 Regulatory Reporting The FSC does collect valuable data concerning various policies such as Total Sums Insured and number of risks by line of business. This could assist with finding outliers concerning pricing and underwriting risks within Barbados. Due to the significant reliance on reinsurance, net earned premiums for some lines of business (e.g. property) were actually being reported as negative values because while regulatory reports capture direct written premiums, assumed premiums and ceded premiums, they only reported claims on a net basis against earned premiums. 42 Early Warning System (EWS) Skewed Loss Ratios Because claim data is being reported on a net basis, the figures get skewed especially for lines of business that are heavily dependent on reinsurance. The majority of motor business is retained by the primary insurers which results in a more normal distribution of loss ratio data. With premiums and claims being both reported on a gross and net basis, more analysis can be performed by the FSC that is useful. Risk scores can then be benchmarked against other similar insurers. This can assist the Commission in applying a risk-focused model of supervision by finding outliers and those most exposed to certain risks and can also be useful if calling for targeted on-site inspections. 43 Early Warning System (EWS) Cleansed Data The data relating to property risk was cleansed by excluding those insurers that wrote a small amount of property business (<20%) of overall portfolio or those that had unusual values that were skewing the data. The Team analyzed the distribution of cleansed loss ratios and ran 15,000 scenarios with the following data. The results were then stressed to various points within the distribution. In order to stress the data to the 1-in-200 year threshold for the IAIS Holistic Framework, one would evaluate the loss ratios at the 99.5% confidence level. However, this does not take into account the entire tail risks of the distribution, so a higher Tail-Value-at-Risk ("TVaR") approach was also evaluated. The risk scores can assist the FSC in applying a risk-focused model of supervision by finding outliers and those most exposed to catastrophe risk. This can also be useful if calling for targeted on-site inspections. 44 Financial Services Commission Recommendations The Commission does not have the supervisory framework to holistically assess companies’ risk exposures. The current reporting framework does not appear to look at catastrophe risk at a granular level and relies on company judgment and modeling to provide the catastrophe risk charge. The supervisory framework also does not look at the adequacy of the reinsurance program nor the credit quality of reinsurers. Since the Caribbean is in one of the most catastrophe prone areas of the world, the FSC will need to develop an adequate risk-based insurance supervision framework that can the legislation be amended at the next promote adequate risk assessment by the opportunity to require explicitly that insurers, good reinsurance programs, and insurers promptly document all adequate provisioning for the catastrophe reinsurance transactions; and risk. In this context, it is recommended that the while alternative risk transfer is not Financial Services Commission: prevalent in the market, the FSC should consider developing specific requirements to ensure appropriate use utilize an early warning system to of such arrangements as the market quantify climate-related risks; develops (e.g. requirement for prior to Commission approval). develop and introduce detailed The FSC should also develop risk-based inspection manuals and regulatory regulation and supervision methodologies tools that can guide the supervisor’s that address insurers’ catastrophe risk analysis of companies’ net risk accumulation and transfer practices, so as to retention/reinsurance policies for enhance risk management and financial portfolios of insurance risk; viability of the local insurance markets. develop a Reinsurance Registration System for reinsurers to ensure credit quality and security of reinsurance arrangements; 45 About the Team The Financial Services Commission should be commended for their cooperation and assistance concerning this project. The Commission is well-situated to continue in the development of risk-based supervisory models that will help to make the industry more resilient to challenges in the future. Widespread collaboration from all stakeholders will be necessary going forward to address these risks and provide for a more sustainable future. Ms. Mary Boyer was the Lead Disaster Risk Management Specialist from the World Bank. Other World Bank staff assisting on this project were Mr. Rashmin Gunasekera, Senior Disaster Risk Management Specialist and Josh Macabuag, Senior Disaster Risk Consultant. The Insurance Commission team was led by Mr. Kester Guy, Chief Executive Officer and Ms. Melissa Burrowes, Manager - Research and Policy. Examination Resources, LLC (ER) was the consulting firm assisting on this project. Mr. Bryan Fuller was the Project Lead and acted as the Insurance/Reinsurance expert. Messrs. Elias Omondi Otieno and Don Roof acted as Insurance Supervisory/Regulatory Experts. 46