CAPITALIA BANKING GROUP is one of the top five banking organizations in Italy with total assets of EUR155bn. Based in Rome, the group is the product of the merger of several leading Italian financial institutions over the last two years.
The group now comprises the holding company Capitalia which is registered as a bank, plus subsidiaries Banca di Roma, Bipop-Carire e Banco di Sicilia, Fineco, and MCC (formerly called MedioCredito Centrale).The Capitalia bank holding company is responsible for setting the strategy and developing the business plan for the subsidiaries, with the aim of creating better, more efficient services and products for its clients, thereby increasing shareholder value. A key part of the Capitalia's strategy has been the improvement of its risk management through the implementation of a comprehensive and integrated risk management platform covering all the group's activities and entities. The bank has already installed risk system vendor Algorithmics' market, credit (including counterparty exposures) and asset and liability management (ALM) solutions. These state-of-the-art risk management solutions are leading the way in the Italian market, and have already brought Capitalia significant added value through the more efficient allocation of risk capital, the measurement and control of new business activities and dramatic IT savings through the integration of the institution's technology infrastructure.
The project began at Banca di Roma in the summer of 2000. One of the bank's major challenges, and ultimately one of its greatest achievements, was the way in which Capitalia, as it evolved through the merger of Banca di Roma and the other entities, was to adapt what was initially a single bank project and develop it into a groupwide risk management strategy.
"We set out to improve our risk management process in two different ways - by improving our risk control and at the same time improving our shareholder value," says Dario Cardilli, head of risk management at Capitalia. To achieve this, it was clear that the bank needed more effective measurement of economic capital. However, at the time the bank had very little expertise in risk management, especially compared with some of its competitors that were relatively advanced.
"We had to jump ahead - we could not afford to walk," says Cardilli. With the help of global consultancy firm KPMG, the group looked at a number of systems and suppliers and drew up a shortlist that included, among others, Toronto-based risk management systems specialist Algorithmics.
"The challenge was not the software - it was choosing a software vendor that had the commitment to a long term partnership," says Cardilli. Capitalia needed a business partner that would work with it to understand its unique requirements and to create a tailored system that would help it achieve its goal. Further, the bank needed a partner with a comprehensive suite of truly integrated risk management solutions to satisfy their diverse needs in market, credit and ALM. In order to do this effectively, the partner would need to offer solutions that shared a common data architecture and a single analytic engine. From this, the bank would be able to obtain an integrated, consistent, forward-looking view of their risk across the enterprise. In the end, this would not only allow for the more efficient allocation of risk capital as well as the measurement and control of new business activities; it would ultimately result in dramatic IT savings through the integration of the institution's technology infrastructure.
"It was very important that there was a suite of products that we could develop together and that would allow us to integrate the various forms of risk management. The partner had to be able to help us make the jump. Algorithmics, more than any other software vendor, could offer this partnership", says Cardilli.
Together, Capitalia and Algorithmics broke down the task of achieving comprehensive integrated risk management into a number of individual projects that, as a whole, involved leveraging a common platform. Algorithmics' Algo Market, Algo Credit (including counterparty exposures) and Algo ALM solutions were implemented and then tailored according to Capitalia's own business requirements.
The first phase of the project addressed market risk, where Algorithmics' Algo Market solution is a clear market leader, says Cardilli. The bank has a complex trading portfolio of around 30,000 positions that includes stocks, bonds, derivatives, bonds with embedded options and structured products. "More or less all the complex instruments that you can find in the market are in our portfolio," says Cardilli.
Algo Market calculates value-at-risk using Monte Carlo simulation on a daily basis, although the solution is also capable of intra-day calculations, says Cristiano Bartalena, risk management application manager at Capitalia. The bank also runs daily stress test analysis of its portfolios.
Capitalia and Algorithmics then moved on to implementing Algo Credit's counterparty risk management, leveraging the common architecture it shares with Algo Market to calculate actual exposure, total exposure, mean loss and derived tail loss. This was followed by the implementation of Algo Credit's portfolio management module. The module uses an integrated measurement of market and credit risk to address the firm's economic capital requirements and provides a solid risk measurement foundation for active portfolio management across the enterprise.
For each of these market and credit risk solutions, including counterparty exposures, Capitalia and Algorithmics worked together to customize what are relatively mature solutions to the bank's particular needs. With Algo ALM, the two companies agreed to a development partnership to implement the solution and then build on it to suit Capitalia's requirements.
Since there are few precedents of banks anywhere in the world who have succeeded in fully integrating ALM with their market and credit risk management solutions, the partnership between Capitalia and Algorithmics would break important new ground in the creation of a consistent risk management approach across the enterprise. Further, because the Italian banking industry approaches asset and liability management differently than other parts of the world, coupled with the fact that Capitalia was in the process of centralizing its ALM for all of its subsidiaries, Algorithmics and Capitalia were able to tailor specific elements of the solution to meet the bank's specific requirements.
"Every bank has an individual way of measuring things," says Cardilli. "The Italian market is quite complex and it has some peculiarities in terms of measuring the risk indicators…we had sit down with Algorithmics to give them an understanding of what we required." Together, they drew up the specifications for Capitalia's implementation of Algo ALM, defining the measures and values, and the meaning of the measures and values to the bank, that the software would calculate.
"This was one of the biggest parts of the asset and liability management project - seeing what was in Algo ALM and what extensions had to be done to the software in order to obtain the comprehensive risk numbers that the bank wanted," says Bartalena. Capitalia's large banking book containing millions of positions is now managed consistently and both Algo ALM and Algo
Market use a common data management layer and common models. The solution now covers 99% of the bank's portfolios in terms of value and risk, integrating the trading and banking books. For Capitalia, the system not only calculates value, duration, various gap measures, earnings, earnings at risk and several other measures for the banking book, but by using Algo ALM's dynamic strategy functionality, which captures changes in the asset/liability mix over time and under scenarios, it also provides a mark-to-future for the bank's balance sheet.
A major part of all phases of the project was gathering the necessary data to feed the analytical system. First, Capitalia had to identify all the required data sources for each application, whether they were in-house transaction processing systems or market data feeds, and whether they included not only position and contract data, but market prices, interest and foreign exchange rates, and so on. The bank then had to assess the quality of the data, as this affected the validity of the results produced by the risk applications, says Bartalena.
The bank used the data management functions in Algo Market as well as tools developed in-house to map the data into standardized formats, and to consolidate, aggregate and compress it for analysis. Capitalia was then able to deal with the vast majority of the required data, mainly covering vanilla products. Other more sophisticated products, such as structured bonds and complex derivatives, were subsequently implemented into the solution and thus reporting and performance were optimized.
Capitalia is currently in the process of upgrading to Version 4.1 of Algo Suite, which will make it even easier to compare and consolidate the various risk results. This supports Capitalia's bank of the future vision, where a single, unified data and analytical platform can provide an accurate, integrated and comprehensive view of the firm's risk across the enterprise. Capitalia has already embarked on this route, taking initial steps such as the consistent modeling of most of its instruments for market risk, credit value-at-risk and ALM.
The rapid progress in risk management made by Capitalia, with the help of Algorithmics, has been recognized by international rating agency Fitch Ratings, who in December 2002 affirmed the group's BBB rating, and simultaneously upgraded the rating for subsidiary Banco di Sicilia to BBB+ from BBB. These ratings were achieved in a difficult economic environment where many other banks have seen their ratings downgraded. In affirming Capitalia's ratings, Fitch commented on the bank's progress in addressing the challenges it faces, "which have led to a significant strengthening of risk management systems and controls", said the agency.
"The fact that we have managed to maintain our rating and improve that of Banco di Sicilia is very important to us," says Cardilli. The bank has billions of euro bond issues in the international markets. "In the last six months the price spread in the market has exploded, so for us it is very important to retain our rating," he says. And the effect on the bank's bottom line has already justified its outlay on risk management. "We repaid our investment in the risk management system in 2002 in terms of the value it created," says Cardilli.
In addition, the improved risk management that Algorithmics' solutions have brought has helped the bank improve its tier one and total capital ratios. At the end of 2001, Capitalia's tier one ratio was 5.1, but by the end of 2002 it was 6.1. "But we have not increased capital - we have reduced the risk without destroying value," says Cardilli. "This enables us to have better capital ratios, which is very important at this particular time in the market."
Today, Capitalia has moved from elementary risk management capabilities to having one of the most comprehensive and advanced systems in the region. Crucial to this success has been the strong commitment from the bank's chief executive officer Giorgio Brambilla and board of directors. Meanwhile, Bartalena stresses the effectiveness of the partnership between the bankand Algorithmics in enabling Capitalia to move so far so fast.
"One of the main things that has made this project work is that we were able to sit down with Algorithmics and define precisely what our needs are, our issues and points of focus, and to work together with them to create the right software to take us into the future."
The Capitalia Case Study is one in a series of Algorithmics' case studies. For information about Algorithmics and to view other client case studies, visit www.algorithmics.com.Algorithmics
More on Technology
Focus needs to be on reacting, not stopping every threat
Companies can wring more value from regulation-mandated data
Risk's annual round-up of new software developments
Markup language could reduce high levels of operational risk
Sign up for Risk.net email alerts
Sponsored webinar: IBM
Watch highlights of this year's London conference
Operational risk and the challenges of defining and dealing with conduct risk
Watch discussions and speakers from our North America conference
There are no comments submitted yet. Do you have an interesting opinion? Then be the first to post a comment.