Banks need to get more granular on risk

The writer is a UK bank regulatory lawyer     

Banks are in the business of managing risk. To do so, they should have the fullest picture possible of where the potential pitfalls lie in their operations.

Yet in recent remarks, Elizabeth McCaul of the Supervisory Board of the European Central Bank highlighted weaknesses in data aggregation and risk reporting by banks. She said that many banks had not paid enough attention to this topic.

This points to a wider problem that regulators, managements and investors need to address. Existing bank rules, worldwide, require the gathering of data without sufficient granularity.

The data currently collected and reported to the regulators provides a snapshot of a “point in time”. This does not allow for a comprehensive, holistic evaluation of actual risk, even in aggregate, as the data is not continuous and not detailed enough.

Given the complexity of the regulatory environment, many banks are managing their liquidity and other calculations in line with the requirements of the relevant regulations and no more. The regulators know this and insist on additional capital to compensate.

The market then discounts the value of banks because liquidity problems and other risks can clearly be missed, as was seen in the cases of Silicon Valley Bank and Credit Suisse. This is one of the key reasons why the market value of most European banks is at, below or barely above their book value; and in the US, their market value is (with some notable exceptions) barely higher than that.

The current regulations arose from the global financial crisis of 2008, when the world’s regulators required significant increases in risk-absorbing capital and better management of liquidity. Most bank regulations are designed to address risks arising from banks’ essential functions, including so-called “maturity transformation”. Banks borrow on a short-term basis, often through deposits, and lend for the longer term in significant amounts. This exposes them to the possibility of having hurriedly to borrow or otherwise find monies to meet demands for repayment.

Also, trading positions and collateral calls on derivatives can lead to abrupt demands for cash. There is a separate possibility of a default on banks’ assets, such as loans. In extreme situations, banks can fall victim to a “run”, whereby substantial numbers of short-term depositors and other claimants call for their monies all at once. This problem has become more acute as depositors can speedily manage their financial affairs on their smartphones.

Advances in data science now permit the gathering and examination of data at a granular level, making it possible for banks to create a more complete, real-time picture of individual cash flows across their books. Such advances mean that banks can supplement their traditional databases, which operate with predefined relationships between data requiring manual modification, which may be slow.

A handful of banks have already started to collect and assess fuller information centrally, at least in part. With such detailed data, banks can match specific inflows and outflows across their lending, derivatives and other books — performing a “cash flow at risk” assessment against possible demands for repayment, defaults or market events. Critically, this can cover every outgoing, including loan interest payments, which are not generally risk-assessed by the regulators’ current metrics.

Knowing where the monies will come from if a payment is due in three hours, or where funding can urgently be obtained if the chosen monies fail to materialise, is sound risk management. If this data is then enriched with business and legal information, for instance on the bank’s rights in particular situations, it is possible to determine how to ensure the necessary cash is available when the time comes. Improved methodologies will also facilitate the evaluation of whether contractual or other offsets are being underused; and whether new contracts should be put in place.

If the world’s legislators and regulators are too slow to move, the banks can nevertheless improve their data quality straight away. A better picture of risk can then be achieved. When this is explained to regulators and investors, confidence will increase, resulting in lower capital and higher valuations.

Making these changes is a significant step, but their cost will be recouped many times over through enhanced efficiency and improved risk anticipation. Banks will be able to play their central role with renewed confidence. The regulators will benefit substantially: they can operate as air traffic controllers with full visibility of risk.