Decision Making in Banking
“As a senior manager in a bank it is genuinely difficult to know if you pull a lever or press a button what will happen.”
Interview with Kevin Rodgers (former Global Head of FX Trading at Deutsche Bank)
Kevin Rodgers started his career as a trader in 1990 with Merrill Lynch in London before joining Bankers Trust in 1993. From there he went on to work for Deutsche Bank where he ultimately led the creation of the world’s number one FX trading house. Kevin retired as Global Head of Foreign Exchange in 2014 and now devotes his time to writing, consulting and his (now resumed) music career. His book, “Why Aren’t They Shouting? A Banker’s Tale of Change, Computers and Perpetual Crisis” was published by Penguin Random House in July 2016.
Q. Senior management in banks make many types of decisions. Decisions regarding business strategy, risk appetite, investment in new systems, HR policies etc. In general what type of management decisions do you think banks are good at?
Banks have been very good at spotting new opportunities, finding new clients and developing new products. For a long time we had an incentive structure that encouraged a genuinely commercial mind-set. A great deal of effort went into understanding people’s performance in the front office and appropriately awarding them.
A key thing that undermined this was the concept of a bank as “one-stop shop” or financial supermarket, where there are many aisles offering the full range of financial services and market products…but if there is a problem in an individual branch of a supermarket or an individual aisle it is not going to drive the supermarket out of business. In banking though, we have seen how problems in an individual business within a bank can bring it down or at least severely damage it. Senior management in many banks did not fully appreciate this.
Q. What types of management decisions do you think banks could be better at?
Pretty well everything else.
In investment banking, support functions such as operations and finance are not generally integrated into the business and this makes effective decision making very hard. In FX we worked very closely with IT and Operations. There was essentially a continuous dialogue with our IT management and staff and it resulted in much better IT delivery.
One of the key problems that make decision making difficulty is the complexity of the organisations. As a senior manager in a bank it is genuinely difficult to know if you pull a lever or press a button what will happen.
The size and complexity means senior managers were simply too remote from the business reality. Frequently they do not have the specialist skills to really understand what is going on. Where you do not have a deep enough insight into what risks apply across all functions things can go seriously wrong.
In banking there has been a tendency to rely on models of reality that give you numbers but do not get across the reality of what could kill you. Senior management rely on those models. In contrast most traders did understand the limitations of the models in relation to their specific trades or businesses.
In management there is generally too much emphasis on generic leadership and relationship management and not enough on detailed understanding of the business. Many times people will come up with plans to replace all the key systems without understanding how they interconnect or what they do. It is no surprise so many large projects fail.
Strategic decisions are normally done based on done based on very little evidence, whether they are strategic acquisitions or rolling out the one-stop shop concept. The end result is their success largely relies on luck. Some of this is because of the innate problems of managing complexity but also from poor use of data. Banks can use data to create the transparency to drive more positive behaviour. One of the major impacts on the P&L of a trading desk can be errors in booking trades. Using that data to name and shame and also to penalise people in the front office who make mistakes can be powerful incentives to get things right.
In the aftermath of the financial crisis banks have tried to improve but they tended to substitute procedure over responsibility or delivery. If things go wrong after all the boxes were ticked it implies that problems result from individual malfeasance rather than the failings of managers or the firm.
Q. How good are banks at defining their problems before they attempt to solve them?
The definition of problems is often awful. The decisions they make are often highly reactive, “shutting the barn door after the horse has bolted”.
The greater the extent that decisions are made based on aggregated views across multiple businesses the worse the problem definition. As you go down the hierarchy you find decision makers at the “coal face”, are better at understanding and defining the problems they need to solve.
Q. Do you think banks pay sufficient attention to the quality and relevance of the information they have available for the decision making processing?
I think they pay lip service to the concept that information should be good quality. Some banks live for years with the consequences of poor reference data without fixing the problem. Even where there is the right data available it can be problematic getting hold of it for political reasons. I have had departments refuse to give me data relevant to understanding the profitability of business because a lack of transparency was in their interest. Going back to the “One-stop shop” concept there was no evidence to support the idea.
There is also a problem with the quality of metrics, particularly those created at the organisational level.
When you do get the relevant information of the right quality it helps make much better decisions. Analysing the data from our RFQ (request for quote)1 system allowed us to identify that providing a quote quickly was more important to getting the business than having more precisely determined spreads.2
Sadly there is a wealth of data in banks that is not adequately used. Before the financial crisis several banks made loans at sub-market rates in order to get additional business for their markets’ business. We wanted to test this idea, so we looked at the relationship between lending and other businesses, we found the more we lent the lower our revenues in other areas such as FX.
However this all ties in with a general lack of mental vigour in analysis and decision making.
Q. In general do you think banks make sufficient effort to fully assess the outcome of major decisions?
If a decision goes spectacularly right or wrong they are better at assessing the outcome. There are a lot of intelligent people in banks and if something does go wrong they will typically get to the bottom of what has gone wrong, fix it and learn the lessons.
Otherwise for much of the other decision making, including the major cross-asset class projects, they don’t really attempt to assess the outcomes of decisions.
Q. Do you think banks are good at learning the lessons from their decisions, whether the outcome were successful or unsuccessful?
Following on from the previous point I think if a decision is successful it is very common to attribute it to the skill and intelligence of the decision makers. However some of the most successful decisions are actually down to good luck but nobody ever admits to that.
On the other hand decisions that go wrong are often explained away by bad luck or factors beyond the control of the decision makers
Learning lessons can be hard when there is uncertainty, even about the objectives of the bank. Politics may mean there is a lack of consensus about what to aim for and the wrong metrics can drive negative behaviour.
For a long time businesses have been incentivised based on revenue rather than profit. This meant people did business with no regard for many of the costs including the impact on the bank’s balance sheet. 3
Q. Are there any tools or techniques that could improve decision making e.g. Big Data, Artificial intelligence?
They need to take rational data science approach, to capturing, processing and organising data so it is easy to access and easy to analyse.
That also means not throwing away potentially useful data. Ultimately though, having the data is not of any value unless you have the managers who have sufficient level of understanding, of what they are doing and what is going on, to make use of it.
Taking an approach to data that is not based on understanding of the business area or the relative value of data can be very wasteful.
Simply gathering data for the sake of it with little understanding of what it is or what to do with it can be counterproductive. Vast amounts of data have to be reported to trade repositories4. The idea was to give transparency to regulators about what was happening in the markets. 150 data fields need to be reported per trade and millions of records but the first time the regulators tried to process the data their systems failed. They could have got a better picture simply by looking at the balance sheets of the banks.
Q. What do you think of the quality of external advice banks receive, to help them make big decisions?
Words fail me.
I think there is often a negative correlation between the cost of the advice and the value you derive from it. The more expensive strategic consultancies charge a great deal of money for advice from people who are admittedly very smart but have little or no understanding of what they are advising about. There’s almost an attitude of “I’m so clever that I don’t need to understand what I am doing”.
Some of the smaller, more focused, consultancies can provide very good advice but they generally use seasoned professionals who know what they are talking about.
Q. What is the single most important thing banks could do to improve the quality of their decision making?
Fundamentally it comes back to the point I have raised in the other answers. You need to have managers that genuinely understand what they are managing.
The big question in today’s highly complex banks is whether it is even humanly possible for senior management to ever understand their organisations at a sufficient level of detail.
Learn more about Evidence Based Management at www.cebma.org
1. Request for Quote — In many financial markets parties can approach a bank or other financial organisation and ask for quote for a specific quantity of a financial asset
2. Spreads — In capital markets banks typically quote a price for buying a financial asset (the bid) and price for selling (the offer). The difference between the two is referred to as the spread
3. Balance Sheet — This is an accounting concept. A balance sheet lists out the assets and liabilities (debts) of an organisation. Assets are always equal to liabilities plus the capital belonging to shareholders. In a bank the assets mostly consist of loans made and liabilities the deposits received
4. Reporting of derivatives trades to meet the requirements of the Dodd-Frank Act (Wall Street Reform and Consumer Protection Act 2010) and EMIR (European Market Infrastructure Regulation 2012)
Why Aren’t They Shouting?: A Banker’s Tale of Change, Computers and Perpetual Crisis: Amazon.co.uk: Rodgers, Kevin: 9781847941534: Books
First published 15th June 2017