Switzerland still the world's largest hub for wealth management

18 June 2018 Consultancy.eu

Switzerland remains the go-to-destination for international wealth managers, but the country faces stiff competition for assets from rival financial hubs in the UK, US, and South East Asia. Deloitte analysis reveals that – despite its reputation for client experience, digital maturity and political stability – Switzerland may be losing its grip on the wealth management market.

Even in the 21st century, Switzerland’s reputation for financial secrecy and discretion remains largely intact. Famed for its political neutrality, the country has the largest International Market Volume (IMV) in the world stashed within its banking and digital jurisdiction.

IMV refers to assets that are managed in a location separate from where the owner is domiciled. It is the plaything of international wealth managers paid handsomely by clients to administer in territories known for their stability and competitiveness.

A new report from Big Four professional services firm Deloitte reveals that, at the tail end of 2017, a total of $1.84 trillion worth of IMV assets were being managed in Switzerland. The alpine country has led the IMV league table unchallenged for the past decade, with managed assets regularly surpassing $2 trillion.

The gap is now closing. Deloitte’s latest Wealth Management Centre Ranking found that IMV managed in the UK now totals $1.79 trillion. A slightly more distant third is the US on $1.48 trillion. The more significant story is that the IMV stored in Switzerland has fallen by 7% since 2010, while in the UK it has risen 9%. US-based IMV has soared by 48%, up from a paltry $1 trillion seven years ago.

International market volume

An even more impressive surge was witnessed in South East Asia. Although between them the US, UK and Switzerland still manage 60% of the global IMV total, Hong Kong’s volume has shot up by 122% since 2010 and is now approaching $1 trillion. Singapore’s has risen 12% and is nearing the half trillion mark. Between them, the two hubs account for 14% of total IMV.

Other traditional centres in the Middle East and Caribbean are struggling to keep up with the pace. Bahrain and the UAE muster a few hundred billion between them, while IMV kept in Panama and its neighbours has fallen spectacularly – from $1.81 trillion in 2010 to just $600 billion in 2017. Luxembourg, meanwhile, is quietly growing – having increased total volume by 25% over seven years to a respectable $260 billion.

First world problems

Switzerland had a poor 2017 by Deloitte’s measuring stick and, alongside Bahrain, was one of the few countries to see a decline in IMV volume. Its relative market share fell from 25% to 21% and it is characterised as ‘struggling’ by the consulting firm.

Yet Switzerland still excels at competitiveness and performance thanks to its digital maturity, unrivalled pedigree as a wealth management hub, and first-rate client services. 

“When it comes to choosing a location to invest their assets, today’s international wealth management clients are looking for excellent service, which includes digital tools, and a top-notch advisory experience,” said Daniel Kobler, who leads Deloitte’s Private Banking & Wealth Management Industry in Switzerland. 

International market volume (in US $ trillion and percentage of total IMV)

“Switzerland is still the go-to place for first-rate client experience, which is why we can be confident about the future of the Swiss wealth management hub. Swiss banks have done their homework well in the last couple of years, and even got their relatively high costs under control, increasing their cost-to-income ratios and profitability.”

So what’s the problem? Kobler and his colleagues at Deloitte found that Swiss wealth managers were having trouble attracting new assets. Although the country has risen to the digital challenge – with an expert workforce and advanced capabilities – such foresight has not been deployed when it comes to the transformation of outdated business models.

The Swiss market needs to improve its cost efficiency and innovation, says Kobler, if it is to maintain its top ranking. As things stand, its rivals to the east and west are poised to grow the most. The US is absorbing much of the IMV which Panama has lost and has loosened wealth management regulations. Hong Kong will continue to balloon as long as China continues to mass-produce billionaires looking to avoid stringent tax laws in the People’s Republic.

AI can improve operational risk management in banking

17 April 2019 Consultancy.eu

Risk management is an integral part of banking. By taking financial risks, banks are able to generate the profits that are necessary to survive. Risk management aims to control this process by making potential losses more predictable. This makes the bank more robust to external fluctuations.

Whereas profits can be made by accepting certain financial risks, operational risk is intrinsically different. Operational risks only cause losses – financially in terms of bottom-line impact and non-financially in the form of for instance reputational damage. The consequences of operational risk events can have a large impact on an organisation and the financial system as a whole as experienced during the last financial crisis. It is not surprising therefore that operational risk is receiving more attention within the financial sector, with banks trying to minimise the operational risks they take, given the resources available while keeping in mind the strategic goals of the organisation.

In the past decade, there has been major progress in the development of artificial intelligence (AI). AI algorithms excel at data analysis and have evolved to the point where they surpass human performance for a wide variety of tasks. More and more businesses exploit these technological advances to optimise different kinds of processes such as marketing, sales and e-commerce, manufacturing and logistics. In today’s growing data-driven world, this trend is expected to continue on the back of widening opportunities for use cases, including in the area of operational risk management.AI can improve operational risk management in banking

Challenges in operational risk management

In June 2011 the Basel committee published the Principles for the Sound Management of Operational Risk (BCBS), which provides a framework for the development of proper operational risk management. Three years later, a survey was conducted to measure to what extent banks complied with these principles. One of these principles states that banks should write a risk appetite and tolerance statement. Banks reported that this is more challenging for operational risk than for other risk categories and attributed this to the nature and pervasiveness of operational risk. The banks that did comply with this principle often reported the use of backward-looking metrics of operational risk, such as operational losses as a percentage of gross revenue.

The above example underlines the challenges which banks face in the management and measurement of operational risk. Compared to financial risk, operational risk is a more qualitative field of study. Whereas financial risk management has been the main priority of banking for a longer time, operational risk management is much younger resulting in less extensive historical data. Predictive modeling becomes more of a challenge in this situation.

On top of that, the events in operational risk are much more diverse in scope. The Basel committee defines operational risk as the “risk of loss resulting from inadequate or failed internal processes, people and systems or from external events”. Internal fraud, data leakage and reputational damage are very different problems, yet can be very closely related as well. A lot of the data in operational risk consists of textual input which contain qualitative information. The qualitative nature of operational risk is reflected in the Basel framework, which encompasses guidelines for organisational structures, culture and awareness, and qualitative reporting.

The computer as a reader

Artificial intelligence could play a valuable role in upgrading operational risk practices, with in particular machine learning – the field of self-learning computer algorithms – showing promise. Machine learning algorithms can make predictions based on data fed to the algorithm. Recently, major progress was made in the field of natural language processing (NLP). NLP focuses on using textual data for predictions. As an example, an algorithm can learn to rate hotel reviews. By processing large amounts of reviews together with their given ratings, the algorithm can learn to give a rating to a new review it has never seen before.

“There are many opportunities for operational risk management to exploit AI and other related technological advances.”
– Lars de Ruiter and Matthias Geerse, Solid Professionals

In operational risk, many textual reports are written regarding specific risks or risk appetite of the organisation as a whole. These reports are generally written by risk managers – experts in their field. Assuming that risk reports hold information which is absent in historical loss data, this information could be extracted and used for predictive purposes. So why should computers perform this task instead of risk experts? The answer lies within the physical limitations of human being – as well as the natural biases they have to cope with. While humans may struggle to remember a piece of text, a computer algorithm easily processes thousands of books and finds structure within using statistical methods with near-perfect precision.

Towards prediction of operational risk

The performance of NLP algorithms using word embeddings has increased tremendously of late. Using the embedding as a starting point, these algorithms learn the meaning of full sentences and use this information for further predictions. This is more generally called sentiment analysis. A popular application in the financial setting is predicting stock price movement from news articles, or twitter feeds. A recent paper by Denmark’s national bank predicts corporate distress of firms by analysing their annual reports. Taking textual data under scrutiny can make a contribution to quantitative predictive modelling, even in the financial sector where the use of words is very different to day-to-day language.

In summary, there are many opportunities for operational risk management to exploit AI and other related technological advances. One possibility is the classification of risk events. As an example, a computer algorithm can read risk descriptions written by risk managers and classify them according to their impact and frequency. Combining loss data with risk reports, improved prediction of risk events might lead to more accurate prediction of future losses. A different application is the measurement of more abstract concepts such as the financial health or maybe the cultural aspects of a company.

An article by Lars de Ruiter and Matthias Geerse. Both are finance & risk consultants at Solid Professionals, a consultancy from the Netherlands.