Banks face three types of risk. The biggest is credit which accounts for the bulk of it, but there are also market and operational risks. When setting aside capital they must allocate for each.
Because of it’s impact much work has been done on credit risk but, probably because banks are especially unwilling to share data on it, data on operational risk event (ORE) frequency (or to most of us, errors) has been largely ignored by academics.
Using a unique data set from a commercial bank in Jiangsu, one of China’s most affluent provinces, the authors of the paper highlighted this week, Yuqian Xu, Tom Tan and Serguei Netessine, from the universities of Illinois, Southern Methodist and Pennsylvania were able to study how workload increases ORE and the severity (extent of loss) of it.
This is valuable work because if it’s possible to understand how workload increases ORE frequency it may be possible to say something about right-sized staffing levels and therefore how much capital should be optimally set aside for this?
The researchers discovered a U-shaped relationship between workload and ORE. Add to the workload and, initially, errors drop as staff become more engaged in their work. Add more work though and they’re swamped and errors pick up again. Of particular interest the paper found no relationship between frequency of error and severity. Mistakes it seems are just that.
Looking at patterns at individual branch levels (there were 49 in the study) it became clear the bank had allocated staff sub optimally and not only did they have too many overall, but they were incorrectly distributed. If the bank cared to optimize its human resource it could achieve a reduction in headcount of 7.6%, this would lead to a 4.5% drop in errors which would in turn produce a 1.2% increase in profits. Not insignificant results in the context of China’s bigger banks whose payrolls run presently into the hundreds of thousands.
Bad news for tellers, good news for we investors if this study gains wider currency.
You can access the paper in full via the following link When Is the Root of All Evil Not Money?
Happy Sunday.