customers are serviced annually by BMW Financial Services
Banks will continue to be vulnerable to shocks without a high-performance risk-steering model; creating a ‘single point of truth’ that aligns risk profiles across products and markets provides a clearer view of the road ahead
In the closing months of 2014, a number of global car manufacturers were forced to stop taking orders or to downgrade sales forecasts for Russia in light of the volatile political situation and the collapsing rouble. A perfect storm of low oil prices, looming recession and Western sanctions sent the currency plummeting, and some automotive firms had difficulty pricing cars and making money in the usually lucrative Russian market.
Facing many of the same threats to growth as other global financials, and bound by equally stringent banking regulation, it is easy to see why the captive finance arms of multinational car makers need to evolve their risk capabilities along with their product ranges.
BMW Financial Services decided to put themselves in the driving seat in 2008, at the height of the financial crisis, recognising that a more harmonised data approach was essential in order to better manage risk and accelerate growth.
Until that point, the firm’s risk-steering model mainly focused on growing top-line returns. The captive bank has since built an integrated model around return on risk-adjusted capital (RoRaC) to create a better balance between profit and risks.
The captive’s product range may be less complex than traditional financial institutions but, with nearly EUR 100 billion on its balance sheet and operations in many countries, its risks are comparable. Threats mainly relate to residual value risk – the gradual depreciation in the value of vehicles with attached loans and lease contracts – and risks relating to customer creditworthiness, interest rates and liquidity.
The firm wanted to be able to model for difficult-to-predict, high-impact events, such as a sudden depreciation of a major currency or a slump in oil prices, which can have a major effect on sales. In fact, both of these events occurred in the closing months of 2014.
customers are serviced annually by BMW Financial Services
At one end of the spectrum, another global economic crisis would have a dire effect on customer credit. At the other end, vehicles not covered by insurance could also affect the company’s loan book. Geopolitical risk must also be factored in. Having a firm-wide risk structure based on a single data source enables BMW Financial Services to be better prepared with potential corrective measures in place.
To make this happen, the captive harmonised its methodologies and KPIs across risk profiles for all products and markets. It also sought to standardise its data for easier analysis and use it to make comparisons. Transaction and contract data drawn from more than 3,000 data fields is now housed in a central database known as ‘the single point of truth’. Handling and analysis of large volumes of complex data is now much easier, and compliance with the different requirements from multiple regulatory regimes more straightforward.
A firmwide risk structure based on a single data source enables a bank to be better prepared with potential corrective measures in place
This plays well with regulators, but is also welcomed by credit rating agencies, with positive consequences for ratings that affect the cost of debt. Higher-quality data and a proactive approach help the captive to better understand and adapt to market trends, and become an indispensable link in the BMW Group value chain.
Since the crisis, all banks have been under pressure to produce higher-quality data from a regulatory perspective, but for many this is perceived as a time-consuming and costly box-ticking exercise. BMW Financial Services took a different view. It saw a strong business case for becoming a more data-centric organisation, something for which all banks must strive.
Banks generally recognise the importance of being data-centric. However, in our experience, many have some distance to travel. Legacy IT systems built over a period of decades obstruct progress, which is further delayed by product rollouts, mergers and acquisitions. Individual products have their own characteristics and unique sets of contracts, as do countries, with larger car-finance markets possessing different dynamics to smaller ones.
Older systems running alongside newer ones, often with different data standards, do not communicate well together. Gaining a clear view of all the data across the group becomes impossible, making it more difficult to flag up risks.
Simply ripping out these older systems and replacing them is not an option. In most cases this would be too expensive and disruptive. Usually, banks navigate this constraint by setting up a separate data warehouse, which aggregates and unifies the data from the bank’s various IT systems and business units.
Whilst being a captive, BMW Financial Services acts like a bank. They need to constantly evolve the risksteering concept to stay in tune with the latest banking procedures, and global and local regulations.
Christoph Nowakowski, Engagement Partner, BearingPoint
This data is then turned into a standardised format and classified according to market sector.
From here, it is much easier to generate the right data to populate KPIs, make comparisons and meet regulatory obligations, and there are a number of commercial benefits, too. Assessing risk concentration becomes easier, facilitating more accurate cash-flow projections and greater flexibility in supporting new product launches.
Improvements to hardware and software are not the full story, however. Key requirements needed for successful implementation include:
Investors feel more confident backing a group that has a firmer grasp of its own risk profile, but there are also more direct commercial benefits, which stem from a greater emphasis on granular data
Performance is measured according to different KPIs that reflect each department’s priorities, making it tricky to develop unified data standards for regulatory or steering purposes. The treasury department, for instance, will make its own set of assumptions, perhaps relating to refinancing issues; the credit department will take a different view, as will risk management. Each of these departments will also have its own particular calculation engines and favoured metrics.
Similar methodologies and KPIs across departments might seem a good route to a more transparent view of activities across a bank. In practice, however, it is necessary for each department to use some of its own particular metrics and sets of assumptions. Awareness of the differences between their views of the same data is crucial, being able to explain, for example, why different yield curves are being used.
Regulators want to see that data is present and that a bank is using it properly to understand its risk position. A centralised database with unified data across all the different product areas and regions means individual contracts can be checked and compared with similar transactions in other markets. The main objective is to give management a holistic view of bank operations.
Changing processes and shifting responsibilities can cause friction between departmental teams. A project lead must take responsibility for driving these changes across the organisation and aligning the various stakeholders, with continuous support of senior management.
Automation of IT processes and quick turnaround between gathering the input data and making it available as aggregated risk data for top management reporting is critical. This enables a proactive steering process to pre-empt and respond to the effect of market events on the business. These new methodologies and KPIs must influence goal-setting and performance assessments to reinforce stakeholder alignment.
Staying on top of the influx of new rules relating to risk reporting is a complex endeavour for financial institutions. It is also an opportunity, placing them in a better position to deal with forthcoming changes and demands for higher-quality granular data. For example, until recently the exact obligations under BCBS 239 were not known. These are part of Basel III’s standards relating to risk data aggregation and reporting for banks.
Decisions are only as good as the data that supports them.
Alexander Beck, Engagement Manager, BMW Financial Services
However, applying rules relating to data, such as BCBS 239 or IFRS9 (International Financial Reporting Standards) is less onerous for those organisations that are already committed to the principle of driving transparency and generating granular data.
They find it easier to implement these new standards than banks, which view them purely as a regulatory burden and are unwilling to fundamentally adapt themselves towards being more data-driven organisations.
Indeed, many financial institutions now understand that there is a range of commercial benefits to be gained from possessing better quality risk data (Figure 1).
Simply ripping out these older systems and replacing them is not an option; in most cases this would be too expensive and disruptive
There is recognition that the quality of decisions can only be as good as the data supporting them. In turn, they understand that this can lead to greater efficiency, better decision-making and increased profitability.
Data enlightenment and the 'single point of truth'
Robert, what role does BMW Financial Services play as a captive in the BMW Group?
BMW Financial Services supports the sales, customer service and loyalty programmes of the BMW Group and contributes to profitable growth, accounting for a high proportion of the Group’s assets: it is an integral part of the value chain. The firm will grow in importance to the group as customer behaviour shifts from ownership models to flexible usage and mobility services.
What were the motivations for setting up an integrated risk-steering model?
The main threats to the firm are credit risk, residual value risk, interest rate risk and liquidity risk. Pre-crisis, these were mainly managed locally using a range of methods, KPIs and IT solutions. This was not sufficient to prevent substantial damage during the financial crisis of 2008, when it was recognised that an integrated risk and return steering approach, harmonised across all risk types and markets, would put the financial services arm of the group in a much better position to manage future unexpected disruptions. It is based on a model called RoRaC [Return on Risk Adjusted Capital], which supports business decisions that target an adequate balance between profit and risks.
What was the main challenge in transforming the BMW Financial Services’ risk reporting systems?
Aligning the large number of stakeholders behind the move towards improving data quality was one of the biggest. Harmonising procedures and methodologies could have been disruptive for some departments, potentially leading to resistance to change. This meant educating people within the organisation as to why change was happening. Training programmes were put in place to help staff adapt to new operational procedures. It was also important to align incentives and compensation to reinforce new practices around better data reporting.
How was the task approached?
Decisions are only as good as the data that supports them, so a close focus on data quality was essential, requiring the setup of a control system along the complete data delivery chain. For accuracy and flexibility, all transaction data was collected on a single contract level in a central database known as the ‘single point of truth’. This made it easier to handle and analyse large volumes of complex information from more than 3,000 single data fields.
Andreas Rindler, Nicolas Maupront, Olivier Darondel, Christian Bruck, Ammar Jamal, Patrick Duepmann, Dietrich Heusel and Thomas Frenzel from BearingPoint.
The authors would like to thank Charlotte Reby and Angélique Tourneux at BearingPoint, Jon Collins at Inter Orbis, Michael Agar, Philip Harding and the team at Grist.