Banking

Best Practice

Cash Management

Centralisation

Financial Supply Chain

Investment & Funding

Payments

Regions

Regulation

Risk

SEPA

Sibos 2011

Technology

sponsors

I admit that I am quite confused when it comes to the topic of cash flow forecasting. I have read all the great articles that have been written over the years and tried my best to understand the mathematics behind cash forecasting techniques. I don’t need convincing of the importance of the topic. With reduced liquidity and credit available, it is important to know what cash there is today within the company. It is vital to understand what cash we will have or will need in the short term as well, so we can plan for it. Once our treasury has this information, we can hopefully make the right decisions. My issue with cash forecasting is surprisingly simple - I need to make sure the subsidiaries that should be producing forecasts do so on time, every time. The result of this not happening is that our treasury department may borrow funds when it doesn’t need to or make investment decisions that are incorrect and could lose us money. This may sound like a simple problem but the solution is not quite so easy and it is not one that will be solved solely by the acquisition of a cash forecasting system. The issues I see for our treasury are twofold:

1. I need to encourage the subsidiaries to get into the habit of forecasting regularly, by which I mean once per week, with updates when necessary.
2. I need to improve the accuracy of these forecasts. I am keenly aware that the nature of our business dictates the frequency required for forecasting.

The first issue is an internal one and can only be solved, in my view, by cajoling and convincing the subsidiaries that there is a real need for forecasting. I have been working hard to understand the impact of poor or no forecasting and the results are really quite worrying. I am now able to quantify this, so at least I can now justify the necessity of the exercise - there is a company-wide financial impact.

The second issue is also an internal one, but can be made easier with the use of technology. I need something that will allow my team to easily communicate forecasting information with each other, to judge the accuracy of the forecasts and then to work out which subsidiaries are better than others. The issue for me then is to work with the poorly performing units to improve things.

One of the other discussions we had recently around this subject was to define how frequently we needed to forecast. We currently operate a weekly cash forecasting system, which can be updated daily if necessary in the case of unforeseen requirements. We used to operate a daily, weekly, monthly, half yearly and yearly forecast but, after discussions with our subsidiaries, we decided that treasury could live with less long-term forecasts. After all, the theory goes, the further out in time you go, the less accurate the forecasts are going to be.

The big problem I have is getting my subsidiaries to provide forecasts in the first place. I am somewhat lucky in this regard because I have been able to convince them of the necessity of the exercise. I have spoken to my peers in other industries, some of whom really struggle with this issue, while others who have simply given up. Yet others have used a carrot and stick approach - linking forecasting to salary and bonuses.

In summary, the issue for our company with cash forecasting is more of an internal people issue rather than a system issue. Maybe, as I’ve seen with other treasury activities, we just do things differently.