PwC Actuarial Services Newsletter – March 2017

This will mark the third year of our European collaboration on our Actuarial Newsletter. We want to take this opportunity to look back on an interesting year for actuaries and as well as take a glance at what awaits actuaries in the months and years ahead. New regulatory requirements have confronted European insurers in the past year. We take a closer look at how our clients deal with this paradigm shift.

Key points in brief:

  • Article #1: Model Validation
  • Article #2: Solvency II after Year One: Snapshot of the current status
  • Article #3: ORSA – “Never hate your enemies, it affects your judgment.” – The Godfather


Download the PwC Actuarial Services Newsletter here.

PwC Actuarial Services Newsletter – August 2016

The PwC Actuarial Services Newsletter is a joint venture of three of our worldwide PwC actuarial practices. In recent years, there have been a number of collaborations on client projects, initiatives and content development between the three entities, Switzerland, Germany and the Netherlands. This newsletter will examine current topics of the industry from different regional and thematic perspectives, and is aimed at insurance professionals working in or closely with actuarial departments.

Key points of issue 4:

Article #1: Using advance analytics for fraud detection: the future is now

In order to have a future-proof business model and survive in this disrupted sector, the traditional financial institutions should increase the use of advanced analytics on a wider variety of internal and external data sources. Traditional financial institutions are operating in a rapidly developing sector and are subject to an increasingly stricter regulatory regime. Moreover, the current economic situation and the past persistent financial crises have had a large impact on all financial institutions. In this article we will focus on how advanced analytics can help to improve specifically fraud protection as an example of its use in practice. We illustrate an approach how financial institutions could expand their current efforts in fraud detection and provide compliance and financial arguments in favor of this approach.

Article #2: Health Insurance & Data Analytics: Increasing Claims Process Quality

Are the Claims Payments under Control? Earning premiums and paying claims describes the core process of any insurance company. Therefore, it is of great interest for the industry to review the appropriateness of benefits paid by the insurance company. For example, in the case of a health insurance company, paying for a patient’s treatment, which is not covered by the policy, would want to be avoided. Controls need to be in place to detect incorrect invoices that have been submitted and an efficient process can help reduce claim payments and thus increase profitability.

Article #3: European Newsletter – Pricing Tool and Machine Learning

In the largely saturated P&C market in Germany, insurers are looking for ways to improve the profitability of their business. To this end, pricing of contacts is a focus of much attention. We recognize a trend of insurers improving their own business situation by an extensive and progressive usage of data in their tariff models. However, the current, mostly conservative, tariff models are not always appropriate for using the available data volume and handling the data heterogeneity efficiently. For dealing with these issues, modern machine learning methods can demonstrate a number of advantages. The following article introduces relevant machine learning methods based on two specific examples in pricing.

Read the full newsletter here.

Find the last issue Nr 3 here.

Please contact me for any further information or if you would like to discuss one of the Topics.

From Enterprise Risk Management to a Resilient Organisation

The resilience of an ecosystem is its long-term ability to deal with expected and unexpected changes. It concerns not only the successful adaptation and transformation before and during a situation of stress, or even in a disaster, but also maintaining a critical level to survive. Resilient ecosystems are hence able to recover faster and emerge stronger and healthier from such disturbances.

In today’s fast changing world, it is crucial for every organisation to be adaptive and transformable with regard to all aspects of the ecosystem in which it is embedded. Such aspects include the economy, globalisation, people, market, competitors, environment, natural disasters, technology, government and the political landscape, and many more. An organisation has to be resilient to continue successfully.

We could think about a fire in a high-tech production plant, for example, where the smoke particles in the air-conditioning system cause a production breakdown for several months, the consequences of hurricane Katrina in 2005, the floods in Central and Eastern Europe in 2002, or even the impact of a cyber-attack on a systemically important financial institution which causes a massive loss of client data.

The question is: how resilient is your organisation in such a critical situation?

Most organisations have an enterprise risk management framework, which is well designed to handle all natural or man-made risks. However, in complex dynamic systems, hindsight and error prevention are far from sufficient just as it is unrealistic to think that a few people alone can manage a disaster situation at an organisation. Fast adaptability and transformation before and during an incident − paired with distributed intelligence − are needed to cope with such situations of stress.

But how do you transform your current organisation from a reactive one into a proactive one?

What Is Resilience Engineering?

Resilience engineering is often called the ‘new way of thinking’ about safety and risk. But what exactly is new and different about it?

Risk management and enterprise risk management are based on hindsight. They focus on cause detection and error prevention, respectively. This approach reacts to exogenous causes, like natural disasters, or endogenous sources, like a technology breakdown or (very often) human error. Based on the information available, an organisation can introduce measures − risk controls, risk prevention and risk budgeting − to optimise the risk-return profile and achieve the organisation’s strategic financial goals. These measures, however, are often not as effective as expected because the planned and trained reactions are unable to cope with real-life events.

The more complex and the more dynamic a company’s ecosystem is, the more difficult or even impossible it is to manage the risks and changes in the traditional way. Thus, instead of administrating risks and changes reactively, resilience engineering aims to improve the ability of an organisation and individuals within it to create foresight and to anticipate the changing nature of risks, before a failure and its harmful consequences actually occur (David D. Woods 2005). In other words, resilience engineering is a paradigm for risk and safety management which focuses on how to help people cope with complexity under pressure by bearing one question in mind: why do things go right rather than going wrong?

Similar to the enterprise risk management approach, resilience engineering concentrates on both threats and opportunities as well as unstable situations. In a resilience engineering sense, irregular variations, disruptions and a degradation of the expected working conditions are not a breakdown, a malfunction or a human error. Instead, such events mean that an organisation has failed to demonstrate the readiness and ability to cope appropriately with an anticipated change in a complex and dynamic ecosystem. Change can either be negative or positive. Each organisation has to adjust its performance in advance to adapt to the current situation and conditions, and to anticipate the impact of change in order to achieve its expected goals in full.

Bild1

Hence, resilience engineering analyses no less than the performance variability within the ecosystem in question, and how we can measure and improve an organisation’s resilience.

How to Transform your Organisation into a Resilient Organisation

Resilience engineering focuses on the states and behaviours of the ecosystem and its connected systems, and especially on its dynamics and performance variability. This means that we do not look at a system in terms of when and where an incident happened, and ask what was missing or why it went wrong. Rather, we look at the system to see when it performed well. We analyse what was present in order for the actions to have functioned correctly.

In order to make a smooth and successful transformation, we first have to understand how an organisation is expected to work (the ‘work as imagined’), i.e. its performance and its performance variability. Traditional risk or failure/error management counts failures and incidents and, afterwards, sets up the quantitative statistics and models needed to try and define preventive actions. In the resilience engineering concept, we have to describe and understand what is really going on (the ‘work as done’) in an organisation.

Because the performance of an organisation and its ability to anticipate changes is neither static nor simply an indicator, we have to start by outlining the abilities that make resilience performance possible. Erik Hollnagel et al. (2006) set out the four basic abilities:

1) the ability to respond appropriately to a changing environment

2) the ability to monitor the right thing

3) the ability to learn the right lessons

4) the ability to anticipate the right developments.

Based on these four abilities and four corresponding sets of questions (see Hollnagel, 2009:117 in Resilience Engineering Perspectives, Volume 2), we can start setting up a resilience profile and learning about the organisation’s resilience performance.

Monitoring this analysis over time shows where changes occur (work as done) and where changes should occur. The results can be compared using a spider diagram for each of the four abilities, which is then summarised on a star diagram for all four abilities.

Bild 2

Bild 3

Overall, resilience engineering can serve as a valuable instrument to measure resilience performance and, therefore, help to shape an organisation into a more agile, anticipative and stable organisation.

If you have any questions, do not hesitate to contact me.

Credit Risk Series (1): Overview of Regulations and Accounting Standards

Since the financial turmoil of 2007/08, regulators and standard setters have become increasingly aware of credit risk and credit risk valuation and never tire of publishing new standards and regulations. Over the next few years, several new requirements will become effective. To this end, I will be looking at these credit risk standards and how they affect financial institutions in a series of blogs. Implementation of some new requirements has already started. For efficiency reasons, it would be desirable to ensure consistency in implementation across all standards, if at all possible. An early start on analysing the new requirements and impacts is recommended because the devil is in the details.

It is all about credit risk

Financial institutions have started with (pre-)implementation work for the new IFRS 9 Financial Instruments standard, which replaces IAS 39 and which will become effective as of 1 January 2018. The standard consists of three parts, namely, a model for the classification and measurement of financial instruments, a forward-looking expected credit loss impairment model and a reformed approach for hedge accounting.

In association with this new standard, in February 2015 the Basel Committee published the consultation paper for the Guidance on accounting for expected credit losses. Comprising 11 principles, the guidance sets out the supervisory requirements on sound credit risk practices associated with the implementation and ongoing application of accounting standards that involve expected credit risk losses.

The Revisions to the standardised approach for credit risk were published for consultation in December 2014. The main objective is to reduce reliance on external credit ratings and to replace such with a number of risk drivers. The revisions should achieve more granularity in various risk profiles and thus reduce variability in risk-weighted assets.

The standard on Capital requirements for bank exposures to central counterparties was published in April 2014 and will apply as of 1 January 2017. In the meantime, the interim capital requirements will remain effective. Compared to the interim requirements, a new approach for determining the capital requirements for bank exposures to qualify central counterparties is now included. An explicit cap on the capital charges will apply for bank exposures to qualified central counterparties and the treatment for multi-level client structures is also specified.

Related to this standard is The standardised approach for measuring counterparty credit risk exposures, published in March 2014 and also effective as of 1 January 2017. The standard comprises a non-modelled standardised approach for the measurement of counterparty credit risk associated with OTC derivatives, exchange-traded derivatives and long settlement transactions. The new standard replaces all previous methods.

And finally, in July 2015 the Basel Committee published for consultation the Review of the credit valuation adjustment (CVA) risk framework. The objective is to include all risk factors relevant to CVA, not only the credit spread risk and associated CVA hedges. Additionally, convergences to the valuation practice under the accounting standards are targeted as is consistency with the Fundamental review of the trading book, the revision of the market risk framework. Consequently, instead of having a stand-alone CVA approach, the goal is to have an integrated view including the market risk framework.

What are the main changes?

Throughout this series of credit risk blogs, I will go more deeply into the details of the new requirements and changes. So, for the time being, I will only give a brief summary of the most important changes.

What all the new requirements have in common is that they aim for greater integration and a holistic view of the risks. The intention is that such risks will not be considered from a stand-alone perspective any longer. The new requirements will integrate more forward-looking elements and market risk drivers. In parallel, they seek more comparability by introducing standardised single calculation approaches.

Starting with IFRS 9 Impairment, the biggest change is to move to an expected credit loss model relying on the relative change in credit risk. If the credit risk increases, the expected credit loss over the whole lifetime has to be recognised.

Revisions to the standardised approach for credit risk aim at reducing the reliance on external credit ratings as used in the current standardised approach. These are replaced with a limited number of risk drivers. The goal is also to achieve better comparability to the internal rating-based approach with respect to definitions and treatment of similar exposures.

In the Capital requirements for bank exposures to central counterparties and The standardised approach for measuring counterparty credit risk exposures, single and standardised approaches are introduced replacing the various current methods.

Regarding CVA, the goal is to align the models and calculations with accounting practice and to integrate them into the new market risk framework in order to achieve a holistic view of CVA.

Who is affected?

Whereas IFRS 9 affects all companies that hold financial instruments, especially banks and insurance companies, the Basel Committee regulations only affect banks. While banks are looking for basic consistency in implementation of the IFRS 9 Impairment standard with the Basel Committee regulations, and wish to leverage as much as possible from there, the insurance companies are seeking consistency and leverage with the Solvency II framework. Nonetheless, the insurers are also looking at the Basel Committee regulations as they feel there is some implicit pressure on them to achieve more consistency where possible with the Basel Committee’s recommendations.

What actions are required?

Of course, becoming familiar with all the published and proposed standards and regulations as early as possible will be advantageous, however, performing a thorough analysis of differences and similarities of all the guidance, on the one hand, and a gap analysis on what already exists in your financial institution, on the other, is the obvious next step. My experience shows that the main challenge is to integrate the different perspectives and needs from risk, accounting, data and most crucially, IT systems, and to allocate sufficient time and staff to do that.

If you have any questions, please do not hesitate to contact me.