Cash & Liquidity ManagementCash ManagementCash ForecastingForecasting: If it is so Important Why Can’t Companies Get it Right?

Forecasting: If it is so Important Why Can't Companies Get it Right?

Forecasting is Important

In an era of rising interest rates, knowledge of the future becomes more important, especially if a company is looking to maintain the proper levels of liquidity that can support its business goals and enhance its competitive profile.

If there was ever any doubt about the importance of forecasting the future, then the recent survey on “Treasury issues – 2005” conducted by our firm (The Financial Executives Consulting Group or FECG) should eliminate any doubt. Among the 200 companies that participated, forecasting remains THE top issue.

When asked to rank 20 different issues of importance to their treasury area the respondents (over teo thirds are treasurers, assistant treasurers or directors of treasury) replied that the top three issues were:

  • to improve cash forecasting capability
  • to reduce and / or more efficiently manage the company’s working capital position
  • to act as internal consultants

To prevent any bias caused by pre-selected issues the companies were also asked to select a single issue that consumed the majority of their resources. When companies were given the freedom to pick any issue of importance they again chose “forecasting”.

The median sales size for these companies was $1bn, but the size of the respondents was fairly evenly distributed; 24 per centof the companies had sales under 250m; 25 per cent of the companies had sales greater than 4bn.

Sales Size of Respondents in USD %
Sales size LT 250m 24
Sales size 250 – 750m 15
Sales size 750 – 2bn 24
Sales size 2 – 4bn 12
Sales size GT 4bn 25
Totals 100

There are More Forecast Failures than Successes

In the interest of fairness it must be remembered that forecasting is, by definition, an art not a science because the future itself is uncertain. Therefore, any good forecasting process can only reduce certain “non systemic” errors (i.e. forecast error caused by systems failures, missing data, incorrect calculations, etc); even the best forecasting system cannot eliminate the cash flow uncertainties caused by underlying, systemic business conditions (e.g. seasonality of demand, cost of goods affected by market prices, etc). These systemic risks can also create forecast inaccuracies, but they can only be eliminated by reducing one’s exposure to the market or even selling the business.

Despite the importance accorded to this issue among companies both big and small, few treasury areas can claim to operate successful (i.e. accurate, timely, comprehensive for the time periods involved) forecasting processes, even after allowing for systemic inefficiencies. The survey hints at several reasons for this “forecast failure”.

  • Treasury is too consumed with operational issues to devote the proper amount of resources to address its forecasting task. As a result, the forecast contains inaccuracies, less than comprehensive data or cannot be completed within a reasonable period of time (i.e. can only forecast monthly based on accounting data). As a result, liquidity or hedging decisions are made with bad, old or even misleading data creating additional borrowing costs, missed investment opportunities or excessive transaction costs.
  • Treasury and the operating areas, the source of most forecasted cash flows, cannot drill down to the source of a significant forecast variance to identify why the variance occurred. This problem may be caused by the mechanical difficulty in aggregating the data across organizational units, currencies and time periods using too many spreadsheets or other financial systems. Regardless, the forecaster cannot take corrective action, insuring that future forecasts will contain similar flaws.
  • There is no “value at risk” for the forecasting unit. Most organizational units are rewarded or penalized based on several factors (e.g. sales, ebitda, production targets, etc). Unfortunately, cash flow forecasting is not one of those factors. As a result, the operating units, sees little harm when a forecast is “blown”. After all, it is not as if an error/inaccurate forecast is going to jeopardize an organization’s performance review or someone’s compensation.Example: if an operating unit forecasts too much or too little cash or the wrong currency what is the risk? Treasury will just borrow more/have less to invest, but it will not impact its results which are measured in terms of EBITDA or sales growth. This ack of concern can make treasury the “fall person” for their inability to use cash efficiently, hedge properly, borrow wisely, etc. Meanwhile the operating unit becomes a cash drain as it fails to collect its receivables on a timely (i.e. more than just worry about month end) basis.

Planning not Processing as a Key to Forecasting Success

One method of improving the forecasting process is simple to spend more time and resources at producing a forecast. For treasury this means finding new methods to re allocate its “scarce resources”, namely its staff.

It will come as no surprise to treasury professionals to learn that the average treasury has a small staff, even at the biggest company. Examples:

  • Overall, 66 per cent of the companies responding had less than 7 FTEs reporting to the treasurer, regardless of staff location.
  • Among the smaller companies (sales under $1nn) over 70 per cent had one to three FTEs or none at all (the treasury work was done in the controllers area).
  • Among the larger companies (sales > $1bn) the FTE levels varied significantly with sales size. On average approx 50 per cent of the companies had less than 10 FTEs, but 25 per cent of these companies had more 15 treasury staff.

A small staff is not necessarily a bad thing, but with a finite amount of resources, treasurers has to carefully allocate them to the key tasks at hand. When asked whether they had allocated or were allocating resources to their number one task, forecasting, only 10 per cent of the companies replied “yes”, despite 79 per cent of these companies giving forecasting their most important (rank one or two) priority.

In other words, treasury knows what is important, but can’t seem to devote the proper level of resources to it. If forecasting is so important, then what is occupying all of treasury’s resources? The answer: cash management and all of the operational tasks required to poll/ parse bank data, administer bank networks and create the cash position. With all of this work there is simply little time left to look into the future as treasury is consumed with the present.

There are immediate rewards for getting treasury to act more as internal consultants (the third most important issue according to our survey) or strategic planners. One is better control / optimization of costs when using various financial instruments (i.e. such a FX forwards, interest rate swaps, options) to control market risk. According to our survey:

  • Over 58 per centof the companies responding used forwards
  • 50 per cent used swaps
  • Approximately 33 per cent used options to manage risk.

The survey did not look into the relative costs/benefits of using these instruments but their use requires a fairly accurate forecast of the future if the benefits are to outweigh the cost on a risk adjusted basis.

Too Much Data, Too Little Information

Another reason for forecast failure is the plethora of systems in use, from general ledgers that purport to be ERP systems (if they really are then why does the average company use three to five of them?) to the massive number of spreadsheets and emails used to ferry this data around the organization. In this day of SOX companies are finally figuring out that this “Gordian Knot” of systems produces a lot of data but less information that is not easy to distribute, often not up to date, not easy to control and does not allow a collaborative approach to learning about producing better forecasts.

Finally, with all due deference to the importance of accrual accounting, forecasting profits and cash flows require different systems. Trying to drill down through an accounting system that is only correct once a month (i.e. when the books are closed) is not helpful when treasury is trying to forecast several weeks in advance across multiple currencies and organizational units.

If treasury is to succeed at forecasting it will need a cash oriented system that can quickly and efficiently answer today’s question: “It is 10am do I invest or borrow?” so treasury can then address the more important question, “what do I do tomorrow to (for example) lower my global after-tax cost of funds and maintain my liquidity targets?” Any delays in answering these questions increases the risk that treasury will not be able to efficiently fund the corporation without paying a premium caused by any perceived credit risks.

It would be optimal if there was only one system out there for treasury, but the number of systems is relatively unimportant. What is more important is that each system be comprehensive (e.g. access all bank accounts, all the time) and be integrated with other systems that have good track records of predicting a company’s need for cash in the future. Example: A/R systems can tell treasury about past receipt patterns based on customer performance; A/P system data can tell treasury what disbursements are scheduled over the next 30 – 60 days, perhaps even longer. Greater reliance on these systems, coupled with a company’s current cash and debt positions, can improve forecasting.

The need for “just in time” cash positioning and more timely forecasts suggests selecting a treasury workstation over an ERP system, despite the fact that many ERP systems have cash management functions. The reason for this bias is that the vendors of treasury workstations have more experience integrating cash, debt and FX together with various risk analytic techniques than their brother systems possess. This bias is also supported by some market data that suggests that more companies (up to 50 per cent of Fortune 1000 companies) have workstations, mostly from non-ERP vendors.

Finally, workstations today are more “treasury friendly”, a useful feature for an organization which is ultimately responsible for maintaining a company’s liquidity in the face of poor forecasts, no matter who is at fault.

Cash is NOT Free

The third and, arguably, the most important reason for forecast failure is based on another survey result which indicates that most (66 per cent) companies measure an operating unit’s performance before all costs are included. According to our survey operating units are measured on an ebitda or operating income basis which excludes the cost of cash or capital. Therefore, operating companies feel little need to worry about cash forecasting beyond making sure that their employees payroll checks do not bounce and that their vendors continue to supply them with goods they can carry in inventory. Whether customers pay on time is not given much weight; after all, if the customer will eventually be profitable, where is the problem?

The operable issue at work behind their misconception is time which, as we all know, means money. While most customers will eventually pay 100 per cent of what they owe and all vendors must be paid 100 per cent of what they are due, the timing of these flows is as important as their amount to treasury. Failure to estimate both properly means the corporation must either use some of its cash or its borrowing power to meet any shortfalls, incurring real costs (i.e. interest expense) in the process and subjecting the corporation to risks it cannot fully control (e.g. changes in interest rates). Yet, given the immunity that must operating areas receive when good forecasts go bad these units have no real value at risk when submitting their forecasts.

This misalignment of responsibilities (i.e. an operating unit uses cash, but treasury pays for it) needs to change as the days of cheap money are over, at least for the next several years.

To improve cash flow forecasts treasury should become an in-house bank (IHB) and pay or charge companies that supply/use the company’s cash or its credit. As a corollary to this idea, treasury should set up a policy that says that an operating area’s performance should be measured after a cost is included for the use of this cash. After all, if an operating unit were a separate company and short of cash it would be going to an external bank to borrow, at that bank’s market rate. Why should an operating unit expect to be treated any differently internally if this unit forecast its cash needs incorrectly?

The idea of an IHB is catching on. In a separate study on intercompany lending programs in which over 85 medium (i.e. over $1bn) and large (i.e. over $4bn) companies participated 80 per cent had an intercompany lending program where interest was charged or paid, especially if there were cross border cash flows across multiple tax regimes.

By charging interest expense/paying interest income and measuring a unit’s performance after including a cost for funds operating units will become more aware of the relative costs/benefits of collecting pass due receivables or limiting purchases to essential goods and services. In turn, treasury will not only be able to receive more accurate forecasts but may find themselves more in demand by operating units who are now more interested in treasury’s risk management skills (i.e. how can I hedge a rise in interest rates that will affect my operating plans). Knowledge of cash flows across tax regimes will also be helpful to tax if it is looking to operate a more effective transfer pricing policy.

Conclusions

According to the FECG’s most recent survey, forecasting is THE most important issue facing treasury in 2005. Improving forecasts will remain difficult because the future will always remain uncertain; however if a company is interested in improving its forecasting processes it would be well advised to:

  • Use treasury more as a planning resource or internal consultant and less as a payment processor so the company can use treasury’s risk management skills to separate the forecast failures due to internal (and hopefully controllable) factors vs. external (and less controllable) market forces.
  • Supplement or replace a company’s current Gordian Knot of multiple financial systems with fewer and better integrated systems such as treasury workstations that are more cash flow oriented so forecast vs. actual results by unit, currency, time period, etc are available more frequently.
  • Charge operating units for the cost of capital as if they were independent companies and measure their results further down the financial statement (i.e. beyond ebitda)

Improving one’s control over forecasting and the future is not easy or everyone would be getting it right; however, the payoffs are real, not only in what can be seen on a financial statement (e.g. lower borrowings, interest expense, transactional costs, etc) but what cannot be seen as easily such as better credit rating and less sensitivity to rising market rates.

Comments are closed.

Subscribe to get your daily business insights

Whitepapers & Resources

2021 Transaction Banking Services Survey
Banking

2021 Transaction Banking Services Survey

2y
CGI Transaction Banking Survey 2020

CGI Transaction Banking Survey 2020

4y
TIS Sanction Screening Survey Report
Payments

TIS Sanction Screening Survey Report

5y
Enhancing your strategic position: Digitalization in Treasury
Payments

Enhancing your strategic position: Digitalization in Treasury

5y
Netting: An Immersive Guide to Global Reconciliation

Netting: An Immersive Guide to Global Reconciliation

5y