Welcome to Inside AXIS, our newsletter aimed at giving you, our client, tips and information that can help with your daily work. Articles can be quickly accessed using the menu bar on the left hand side of your screen. You can also scroll down the screen to read the articles, including useful links.

A new valuation Reserve Method is being developed in AXIS to support a Cost of Capital approach to valuation. Reserve Method option (37) Present value of cash flow plus cost of capital is currently available in the Regular Life module for testing. In addition, a new approach to determining required solvency capital is being added by way of a new table option in the Required Surplus section of the cell.

The Cost of Capital method for including a market value of margins (“MVM”) in the calculation of an insurance liability was specified for use in the fourth quantitative impact study (“QIS4”) for Solvency II in Europe. This MVM approach is one of the proposals for valuing insurance contracts under International Financial Reporting Standards (“IFRS”).

Cost of Capital methods add an MVM that is the present value of the return, in addition to the amount earned from assets backing capital, that is required for the total return to be adequate for the risk assumed in the product. The MVM is added to the best estimate value of the liability. The best estimate value is equal to the probability weighted average of all future cash flows, taking account of the time value of money.

Under QIS4, required solvency capital is determined as a combination of the amounts of extra cost resulting from adverse effects of the potential risk components contained within the insurance products and the assets backing those products. The risks that need to be taken into account are operational risk, underwriting risk with respect to existing business and counterparty default risk with respect to ceded reinsurance. Within AXIS, the extra cost for each assumption is assessed as the excess of the liability required under a shocked scenario over the liability required under the base best estimate scenario.

- Gordon Creber

Those days when the actuary performed all work on the desktop computer are long gone.

Sure, you can buy a quad-core workstation and run almost four times more calculations per second, but it is often still not enough. Even if it was, all the new regulations require you to keep the data on a secure network drive, and your IT staff is increasingly reluctant to support applications scattered across individual desktops.

Nowadays you need a powerful grid of servers to run stochastic calculations, terabytes of storage to keep all the results, a user-friendly environment for model development and testing, and yet a secure and automated environment for production runs where no human interaction is allowed.

You need to be able to secure your data, and at the same time have a way to share it efficiently with other users who need access to your work. How do you manage all that, satisfy your auditor’s requirements, and still have time for productive actuarial work? Well, it is a challenge.

To help organize all the work, AXIS already provides a wide range of features such as:
 

• AXIS VBA-like scripting (Dataset and System Formulae)
• AXIS DataLink
• Import/Export
• Built-in Backup/Restore
• User Profiles

On the production side, AXIS GridLink is an extraordinary tool to manage the grid of servers and perform distributed runs.

You just need to put this all together. Where do you start?

We are hearing from more and more companies that they are looking to create secure, multi-user, server-based AXIS environments for front-end model development and testing, and back-end production processing. Great idea! Make your users more productive and comply with regulations, but again – how?

That is why GGY recently started a new and exciting project to not only bring all these great AXIS tools together, but also add a suite of new features to provide users with a truly powerful data management platform. We call it AXIS EnterpriseLink. It will organize the work in your department, and will scale all the way to the enterprise level.

Imagine these new possibilities:
 

• Version control for your dataset archives
• Automated archive for calculated results
• Schedulers for automated runs
• User and user groups control
• Logging and audit tools
• Enterprise level reporting tools
• Centralized deployment and remote access
• Scalable model development environment for multiple users
• Safe backup and disaster recovery tools
• Seamless integration with AXIS GridLink farms
• Integration with AXIS EnterpriseLink systems in other departments of the company

- Victor Rubinstein

Global equity markets have fallen sharply over the last 12 months. For writers of Variable Annuity (VA) and Seg Fund contracts with equity-based guarantees, these sharp declines have pushed some of these guarantees into the money and have elevated the overall scrutiny of the ALM strategies used to manage these valuable embedded options. The hedging of VA guarantees has certainly become a hot topic in the insurance industry worldwide and regulators, rating agencies and company directors are asking companies to demonstrate the effectiveness of their hedge programs, or urgently consider the development of programs where none previously existed.

The word “hedging” has come to be synonymous with dynamic matching of exotic financial derivatives using the very familiar Greek metrics (e.g. Delta, Gamma, Vega and Rho). But some companies also consider accounting exposures in their hedging programs. In either case, a dynamic hedging program comes with operational complexities, but also with the need for practical solutions that may conflict with theoretical exactness. In short, hedging can be an expensive undertaking for any insurer, regardless of size.

In recent years, hedging has generally been a purely risk management exercise, conducted in back offices by specialists using specialized software. In many cases, the specialists and their software were rented from consulting firms, and not well integrated with mainstream actuarial systems and operations. This has presented problems in reflecting the hedging strategy being followed in financial projections or reserves, and in demonstrating to stakeholders the actual or projected hedge effectiveness.

As hedging is becoming more widespread and practices are evolving, actuaries and risk managers want to become more involved in the evaluation and decision-making about hedge strategies. GGY will soon be able to address this need. For the past 24 months, with assistance from well-known industry pioneering expert, Dr. K. (Ravi) Ravindran, GGY has been putting key components of hedging functionality into AXIS that will give actuaries a powerful tool to evaluate and demonstrate the effectiveness or cost of various hedge strategies. Even better, being fully integrated within AXIS, this functionality is a powerful extension of an existing, proven and widely used modeling system.

Some of the challenges which GGY is facing and addressing within this project include:
 

• The definition of an “asset pricing model” that will use generated risk neutral scenarios to reflect the cost of embedded options, the prices of various hedge instruments that might be chosen as hedges, and the Greek metrics that link them together.
   
• Enhancing the existing Economic Scenario structure to include multiple market indices on which hedge assets might be based, and parameters defining the volatility surfaces for these indices necessary to execute hedge strategies within financial projections.
   
• The addition of new asset types to facilitate the management of the required hedge positions.
   
• The parameterization and automation of hedge strategy algorithms to optimize the selection of hedge assets purchased based on the solution to a complex linear programming problem.

GGY now is willing to make a prototype version of AXIS, known as a Technological Preview, available for its current users to gain an appreciation of our proposed approach to defining and testing hedge strategies that are driven by various equity guarantee features within accumulation annuities. The purpose of this program is to allow interested users to plan for the future integration of AXIS into existing or developing hedging programs, and further to enable these users to influence the development path and specifics of this functionality through their feedback. A beta version is anticipated to be ready later in 2009.

In summary, AXIS functionality will soon respond to one of the industry’s most important risk management needs. Interested clients are welcome to request an onsite or web-linked demo of this exciting feature.

- Trevor Howes and Jason Alleyne

AXIS version 12.4 saw the introduction of a second sample dataset. This new dataset, which is included with every installation of AXIS, is named Sample-US. The purpose of this dataset is to develop examples of some of the US functionality that is included in AXIS. This dataset currently includes a single example on Cashflow Testing.

This example will provide the US user with a seamless run that will load, massage and output results over the New York 7 scenarios and give the actuary enough information to judge the adequacy of the reserves.

The dataset includes a Word document stored in the dataset directory. This document is called Sample-US Index. This document gives a description of the examples in the Dataset. The example currently in the Dataset is detailed in the Special Features section of the document.

Please note that the other sample dataset, called Sample, is still included with every installation of AXIS.

- Marc Parisien

- Wes Leong

• AXIS/BondEdge Link
• AXIS DataLink – Recent New Features
• AXIS GridLink Active Backup
• AXIS GridLink – Other Recent New Features
• Category Reports
• Equity Indexed Annuities
• Output Manager
• Premium Calculation Tables
• Required Surplus
• Reserve Reconciliation and Reserve Components Reports
• Run Filters
• Scenario Sets and Market Models
• Stochastic Processing and Memory Management

Visit http://www.ggy.com/support/training/videos/playvideo.asp to watch and learn more about these exciting new developments, and stay tuned for more videos that describe continuous improvements to AXIS.

- Karim Sajan

• Dumb luck. Never underestimate the importance of luck. Many of the fundamental concepts of AXIS that were in the first releases turned out to provide a fundamentally sound basis for all future developments. For example, AXIS was always “convergent software”, using the same codebase for valuation, pricing and modeling. AXIS was always “object oriented”, even before developers switched to an object oriented programming language. AXIS always had a complete separation of source code and data. It sounds obvious, but even today most competitors have not achieved this separation.
   
• Single minded determination. From 1989 to 1991 we did both consulting and AXIS development, but since 1991 we have never taken our eye off the ball by offering consulting services or developing any other products (illustration or admin systems for example). We do nothing else but AXIS, so we had better do it very well!
   
• Long-term perspective. The partners of GGY have always taken the long view. We do not try to cash in to maximize short-term profits. We are happy to invest in expensive long-term development to keep AXIS ahead of the pack. For example, at one time DataLink was a separate application. We embarked on a multi-year project to integrate DataLink into AXIS because we felt that in the long term it was the right structure. History has proven us right on this.
   
• We hire the best. The most important management decisions are who you hire and what you do to train them and retain them. It is not easy working at GGY – there is so much to learn, always, even for those of us who have been around since the early days. We hire people who have the ability and the interest to learn, and they have repaid us by their incredible creativity and dedication. We are very fortunate that so many bright people have applied to work here. We rarely advertize specific positions – we are simply always on the lookout for people who can make our team stronger.
   
• We are never afraid to admit mistakes. Yes, we make mistakes. We wouldn’t be human if we didn’t. What sets us apart is our dedication to fix every mistake we ever made. Unlike other software vendors, we fix every single bug, and we do so as soon as it is found. But that’s not the only kind of mistake. When we find something in the structure of AXIS that leads to a roadblock to ongoing development, we always take the time and make the effort to re-engineer such weaknesses. For example, until recently the Calendar Year results were a fixed list of available lines, to which we added new lines from time to time. Whenever we would add new lines (to meet the needs of particular clients) that would mean the system would run slower. We designed a new architecture involving optional Category Reports that resolves that weakness. Earlier, we were in DOS when the future was Windows. We rebuilt AXIS from the ground up to take full advantage of Windows, but unlike some of our competitors, we did not charge an upgrade fee. We do not consider that a mistake! We continually adapt and use cutting edge tools, the latest compiler and grid technology. We also test on the latest hardware and operating systems, and optimize for them. All this is made much easier because unlike most competitors, we have total control of the source code, and are free to make structural changes if necessary.
   
• We invest in quality. We have a separate Quality Control team, plus individual testing and checking requirements and a battery of testing computers to make sure the product is as robust and accurate as we can make it. The quality control team reports to me and will not allow any version to be released until they are absolutely satisfied with it. Our systems programmers are continually developing new testing tools for the developers and testers, and we incorporate many of these tools directly into AXIS so you can perform your own regression analysis.
   
• We treat our staff well. That’s not because we’re just such nice guys. We value the experience they have built up over the years, and we want to keep them happy so that they will stick around and treat you the customer well. So we have a great location, spacious modern offices, plenty of fast computers and other high tech devices; whatever it takes to support the staff and the ongoing development of AXIS. We do not cut corners.
   
• We stick to our word. It is almost unheard of for a software developer to deliver on time and to specifications. If we promise you a feature by a specific date, you can bet on it.
   
• We listen. We listen to you, and we listen to each other. By listening to a broad range of clients about their needs, and incorporating the best of these into our development plans, we are able to make sure the software is always moving in the right direction. We are very active in developing close contacts with our user community, and the increased level of feedback pays major dividends. They call this the wisdom of the crowd. So give yourselves a pat on the back!
   
• We love what we do here. We really believe this is the right product with the right service, and if we ever changed our belief about what the right product is, we would change AXIS accordingly.

All of these things have contributed to making AXIS a superior product over the last two decades. While some may think it is easy to "rest on our laurels" after such an achievement, let me assure you that we will not! The partners and staff of GGY are committed to extending our efforts well into the future, to continue to provide our clients with the best possible product and service.

- Phil Gold