In the financial crisis around ten years ago, large banks throughout the world had to be bailed out with taxpayers’ money – hundreds of billions of francs of it. Many banks and financial service providers did not receive support and fell victim to the crisis. Why did the crisis, which began in a small market segment, take on such proportions? «Because of the low interest rate policy of the US Federal Reserve, many banks invested massively in highly profitable and apparently secure – i.e. AAA-rated – subprime mortgage loans. The risks associated with this were not initially evident,» explains Wolfgang Breymann, professor of financial mathematics at the ZHAW School of Engineering. «It became clear during the crisis that no one knew who had how much of these ‘poisoned’ products on their balance sheets. This led to the dangerous freezing of the interbank market following the collapse of Lehman Brothers in September 2008.» One reason for this was the lack of a standard presentation of financial contracts across IT systems. This was taken up by Wolfgang Breymann’s group, and together with an international research team they developed a solution in a project funded by the Alfred P. Sloan Foundation, and later also by Innosuisse. Algorithmic Contract Types Unified Standards, or ACTUS for short, is a data and algorithm standard for the format of all financial contracts.

Creating standardised contracts

For risk assessment purposes, oversight authorities assess whether a bank has sufficient capital should the financial environment deteriorate dramatically. Stress tests are an important tool in this process. It can take weeks or even months to fully audit the information obtained by these tests. «The fact that it takes so long and costs so much is largely the fault of the banks themselves,» explains Wolfgang Breymann. «The audit is such a long process because a bank’s balance sheet consists of millions of contracts that are not uniformly presented and evaluated in the IT systems.» This heterogeneity prevents analytical processes from being run automatically. This is where ACTUS comes in. As Wolfgang Breymann explains, while financial contracts seem diverse and complex, they almost always consist of the same basic elements. «If you take away the legal conditions and focus on the cash flow obligations, financial contracts and agreements are suddenly considerably less varied,» says Breymann. «The vast majority of relevant financial contracts are based on a manageable number of underlying mechanisms.» This means the presentation of these contracts can be standardised, which is indispensable for automating the system: «All IT systems will then speak the same ‘analytical language’. This can be compared to the universal genetic code.»

Analysing risks easily

ACTUS combines the information in contracts and the algorithms that convert these contracts into cash flows with additional information or so-called risk factors that determine the state of the financial environment, such as interest rates and exchange rates. ACTUS then generates the raw results in the form of cash flow data streams in a special portable data format that is easy to analyse. All relevant financial analyses – liquidity and solvency calculations and related risk analyses – can then be made quickly and easily, essentially by simply summing up the data. «The financial contracts obtain a degree of analytical intelligence because they automatically deliver raw results for analysis,» summarises Breymann. «This standardisation allows different banks to be compared with each other. Banks also benefit from automation because it shortens internal processes and thus saves costs.»

Successfully tested with real data

The ACTUS data standards not only allow efficient risk assessment and financial analysis, they also improve transaction processing and risk reporting. The researchers have successfully tested ACTUS in a proof-of-concept demonstration using data from the European Central Bank and a major US asset manager. The software can be downloaded free of charge from the US-based ACTUS Financial Research Foundation. Wolfgang Breymann sits on the board of directors of this foundation. He and his team have also developed an R-based prototype environment for ACTUS-based financial analysis for teaching and research. ACTUS is now being further developed to support not only analysis but also core banking processes such as transaction processing, thus creating the basis for an integrated, digital banking infrastructure. «This will be the first banking platform to integrate traditional and FinTech banking channels and automate core banking processes such as transaction processing along with back-office processes such as risk analysis and reporting,» promises Breymann. «Our vision is for information to be recalculated every day, like the weather forecast.»