Firms should budget less (and do more pro-active, risk-adjusted forecasting)

 

In today’s turbulent and fast-changing world economy, companies can’t afford to rely on antiquated budget tools and processes to guide its decision-makers. What is needed is a decision-support tool that can give useful answers about the financial consequences of various scenarios and corporate policies. In this article we discuss several best-practice issues in financial forecasting that will transform the way an organization produces information vital to its decision-makers. Please read more.

 

For as long as most people can remember, the corporate budget has been a largely unquestioned activity whose importance was self-evident. It was the yearly process in which goals were set, targets were negotiated, and forecasts of revenues and costs for the coming year were assembled.

 

All fine then? Not really. While critics always existed, the anti-budgeting movement has grown ever louder since the nineties. The proponents of Beyond Budgeting, as they prefer to call it, argue that something is seriously wrong with the budgeting process. 

 

The core of the critique is that it makes organizations inward-looking and sclerotic: people spend their time and energies on gaming the budget and negotiating the targets (business units want it low so that it easily can be beat; top management wants it high to induce maximum performance). Instead, the argument goes, people should use their ingenuity to figure out new ways to create value for customers, look for new business opportunities, outmaneuver the competition, and so on.

 

In the words of Jack Welch, the problem with most budgeting is that it is disconnected from reality; it’s only authority comes from the fact that it is institutionalized.

 

We agree. One senior financial manager we discussed with said he’d spent the better part of his professional life analysing deviations from the budget plan – countless hours spent on an activity which, at the end of the day, doesn’t lead to anything substantial or create value, and which he thought needed to stop.

 

Some best practice issues in financial forecasting

 

Beyond Budgeting is not, as some seem to think, simply about making rolling forecasts. It is, in fact, an alternative management model, or management philosophy even, that aims to redirect peoples’ creativity and energy towards activities that truly create value. We will not go into the details about these things here – the reader is referred to any of the many books on the topic, or the Beyond Budgeting Roundtable homepage.

 

Instead we will talk about a few best-practice aspects of the tools that companies use in their financial planning and the information they can get from them. The traditional budgeting tool is typically a deterministic model, in which forecasts of revenues and costs from all the company’s business units are assembled and consolidated. Given that this is its main purpose, it is not surprising that it is flawed as a decision-support tool in two important ways: it lacks a good structure for making scenario analysis and for analyzing the consequences of various corporate policies.

 

We believe that in today’s fast-changing and volatile economy a company can’t afford to rely on the budget model to keep track of its numbers. A best practice financial planning tool, in our view, should be able to do the following four things:

 

·Rolling forecasting

·Treasury forecasting

·Pro-active forecasting (acquisition-forecasting in particular)

·Risk-adjusted forecasting

 

In the rest of this article we will discuss each of these, with illustrations from our financial planning and risk management decision-suite.

 

Rolling forecasting

 

Rolling forecasting is a straightforward concept. It means that a company updates its forecasts on a regular basis (typically quarterly), and extends its forecast horizon one period, as opposed to continuing to use the budget forecast throughout the year.

 

The budget forecast quickly becomes obsolete (“before the ink is even dry” as critics like to put it). In some cases it is not a “best estimate” forecast anyway, since it is a negotiated target that is acceptable to both the business unit and top management.

 

Making rolling forecasts bring benefits in that it provides a fresh “reading” of the company’s situation that can guide liquidity planning, resource planning and keep at a minimum the deviations from the earnings forecasts that are communicated externally.

While the concept is easy, it does introduce some challenges in terms of how the organization gets the necessary information. There is a clear danger that whoever is responsible takes the estimate for the last period in the forecast and just extends it one period ahead without much thinking. And a company needs to steer clear of making the updates a repeat of the very resource-intensive bottom-up planning in the ordinary budget. It is important to find the right balance in terms of creating an incentive for people to do it right, but avoid tying up too many resources.

 

We believe that in order to bring full benefits, rolling forecasts should encompass the forecasts of the entire set of financial statements and not just be about making new projections for revenues and costs. This lets a company keep track of its financial status on a continuous basis and monitor how “safe” important financial targets such as debt covenants and earnings targets are.

 

We have developed a standardized tool-box for financial planning and risk, which has been designed to cope efficiently with such “enhanced” rolling forecasts.  The first step is to set the new starting date and feed the model with a new opening balance sheet. When this is done the model will automatically move all data points to the right month in the forecast. All information can be entered with a month and year “flag” that indicates when the transaction is due. This removes the need for manual “cut-and-paste”-operations that are just accidents waiting to happen. Then an update wizard lets the user review step-by-step all the inputs of the model to safely, and in the right order, make adjustments that reflect the new circumstances. This is illustrated in Figure 1 where we have entered the starting balance for a fictive ice cream company called Ice Cream United.

  

 

Figure 1

  

Treasury forecasting

 

If a company’s forecasting extends beyond revenues and costs down to earnings level, the issue of integrating Treasury-forecasts, i.e. forecasts related to its various financial instruments, comes up. In many cases the forecast of net finance costs is obtained by simply looking what it has been in the past and then extrapolating this into the future.

 

This practice might work reasonably well if the company only has, say, a few fixed-rate loans in domestic currency. If the net finance costs hide large risk exposures, however, a company should rethink this approach.

 

We are of the opinion that companies should seek a higher level of integration between the forecasts of its business operations (EBIT forecasts) and its Treasury forecasts. One of the primary uses of financial models is the ability to investigate the impact of changes in underlying assumptions about exchange rates, interest rate, product prices and so on. Generating such sensitivities on important bottom lines is much more meaningful if EBIT and Treasury forecasts are integrated in the same model framework.

 

Treasury-systems can deliver a large number of reports on exposures and predicted impacts on profit and loss from financial instruments. But it can be difficult to feed these reports into the financial planning tool in a way that preserves the logical relationship between them when assumptions are changed, because this requires bottom-up calculations. 

 

The answer that we advocate is to make the bottom-up calculations of Treasury forecasts in the the firm’s financial planning tool, rather than import them or make guesstimates. While companies easily can see the desirability of doing this, sometimes the complexity and sheer number of financial instruments, as well as limited internal resources, stand in the way.

 

To offer an alternative, our decision-suite has an Excel-based Treasury module that covers a wide range of financial instruments, including different types of derivatives. Let’s take bank loans as an example. The user feeds details about the company’s loans into the debt module which then uploads month-by-month forecasts of interest expenses, loan instalments and currency exposures up to five years into the future.

 

These Treasury forecasts have complete IFRS accounting integrity and integrate seamlessly with the company’s forecasted financial statements. Since they are bottom-up, the Treasury forecast will respond to changes in assumptions about exchange rates and interest rates. Since these “global” assumptions also impact the EBIT forecasts the company can create realistic scenarios to see how its different bottom lines would look in different states of the world.

 

Figure 2 illustrates how data for bank loans can be either fed or manually entered, upon which immediate feedback is obtained in the Forecast summary box (as mentioned, these forecasts also integrate with the forecasts of financial statements).

 

 

Figure 2

 

 

Pro-active forecasting

 

What do we mean by pro-active forecasting? Pro-active forecasting, as we use the term, refers to forecasts under alternative courses of actions compared to the base case.

 

At the end of the day, the whole point of forecasting is to inform people who make decisions and guide their actions. We believe the highest purpose of a financial model is to be able to present a realistic picture of how the company’s performance, financial health and risk profile would look if it were to change its strategic or financial policy in some major way, such as an acquisition, capital structure change or risk management strategy.

 

As any board member or senior manager can testify, exercises involving exchange rate effects or the duration of the bond portfolio can be quite boring. Providing relevant feedback on the financial consequences of different policy options, however, makes for very good internal discussions that will challenge the organization’s decision-makers.

Sometimes this may be trivial by simply changing an input in the model, e.g. a higher dividend. But more often it is not. A proposed policy change usually consists not of just one element but several. An increased investment budget may be considered but only if the growth rate on dividends is assumed to be lower (in the near term) and some exposure is hedged. When four of five variations of such a policy combination are considered, the complexity of handling them is quite high and there is an evident risk of manual errors.

 

Our decision-suite has a so-called Dashboard which summarizes the model’s key outputs and contains user control over which policies are active in the forecast (and also which set of price assumptions is active). These policies need not be a single action but a whole package of policies which can be pre-defined, and afterwards be attached, or detached, to the base case.

 

Acquisitions are a case in point. These are often transformative events that can have a huge impact on the acquiring firm. Various ways of structuring the deal and accounting for the acquiree may be considered, and other policies such as dividends and hedging may also be reconsidered in connection with a deal.

 

Valuable insights into the optimal risk and return balance in a deal can be generated if one is able to flick back and forth between different set-ups to compare before-and-after. But it requires a high degree of confidence that the tool that is used handles the different alternatives correctly and consistently, which is a challenge given the complexities of acquisition financing and accounting.

 

For most companies it doesn’t make sense to build high-calibre tools for acquisition forecasting given the time and resources that would have to go into developing and testing it.

 

Therefore we have developed a Merger and Acquisitions-module that allows companies to make a due diligence of the financial consequences of a planned acquisition. An important aspect of this module is that various ways of financing and accounting for the transaction can be handled by simply selecting the desired alternative from dropdown-menus. All calculations have full accounting integrity according to IFRS principles, which lets a company consolidate the acquired company into its own financial forecasts with a high level of analytical integrity.

 

Figure 3 gives a partial view of the Dashboard for our fictive ice-cream vendor. In the upper right corner a set of cases, including the acquisition of Apple Pie Inc, are available. If this alternative is ticked off all the information related to this case will be consolidated into the model and affect the forecasts visible in the Dashboard.

 

  

Figure 3

 

Risk-adjusted forecasting

 

Risk-adjusted forecasting means integrating knowledge about risk exposures into a firm’s financial planning tool.

 

Why would a company want to risk-adjust its forecasts? One reason is that it makes possible a risk simulation of its forecasts. A simulation can give important feedback concerning its risk profile, for example the probability of breaching a debt covenant.

 

To risk-adjust forecasts means that risks are mapped out, quantified and assigned to the appropriate account (revenue, cogs, etc) and business area. Having done this the impact of these risks on a firm’s bottom lines, such as earnings or cash, can be assessed. In our experience this frequently generates new insights and surprises as to which risks are the biggest drivers of results and consequently needs most management attention.

 

Note: most budget software packages only allow scenarios in terms of “cost of goods sold up 10%” or “revenue down 15%”. This is not as informative or particularly useful for developing risk management strategies as mapping out risk exposures.

 

To illustrate the idea of a risk-adjusted forecast, consider figure 4 below which shows the yearly budget for our fictive ice cream company, Ice Cream United. The ‘Budget’ column is the traditional budget forecast. The ‘Risk impact’ column is the combined expected effect of risks and other analytical adjustments that relates to this particular account. The column ‘Risk adjusted’ is the risk adjusted budget forecast, which sums the base case budget and the net risk adjustment.

 

  

Figure  4

  

 

Now where did the risk adjusted numbers come from? They are the result of the company’s own estimate of its risks and opportunities, obtained through a series of workshops or an e-survey.

 

Let’s look at the Other income (expenses) net-line. It shows a very small figure in the budget, but the risk impact is huge. Why is this so? The answer is that we have entered a litigation risk which appears on this account. Management has indicated that they face a 35% risk of an 18mn legal settlement, which translates into a risk adjustment of 0.35 * 18 = 6.3. From looking at the ordinary budget, one would have no clue that the company faced such a serious risk.

 

All the risk adjustments can be decomposed into the underlying risks that have been identified by the company, as shown in Figure 5. A negative number means that on balance, the company has a larger downside risk than upside potential when it comes to this particular risk, whereas a positive number means the upside is greater.

  

 

Figure 5

 

How practical is risk-adjusted forecasting? It is true that most managers either strongly dislike or fear (or both) highly statistical exercises that lead to obscure risk statistics like Value-at-Risk, for example based on a portfolio of currency and interest rate derivatives. We think such risk reporting should have no place in a non-financial firm.

 

Risk-adjusted forecasting, as outlined here, instead focuses on firm-wide risks that most people recognise as relevant (e.g. loss of key customers, project failure, commodity prices, etc) and how they impact well known performance indicators, such as EBIT, earnings, cash, key financial ratios, and so on. The recognition factor from focusing on established indicators greatly increases the willingness to accept probability-based information.

 

The recognition-factor is even higher if risk is defined in terms of targets that management has set for itself and monitors over time. In one company management was concerned with solvency dropping below 30% for fear that that the bank would consider it a covenant breach and intervene. By applying risk-adjusted forecasting, we helped this company calculate the probability of such a breach and illustrate various strategies for managing this risk.

 

 

 

Summing up

 

We have argued that traditional budgeting tools are often inadequate to support decision-making. What is needed is a tool that lets it perform robust and consistent scenario-analysis as well as analysis of the consequences of various corporate policies. We have identified four areas of improvements:

 

·Rolling forecasting

·Treasury forecasting

·Pro-active forecasting

·Risk-adjusted forecasting

 

Any of these forms of forecasting represent an improvement compared to a deterministic budget tool. If combined, a company has a powerful tool that will substantially improve the quantity and quality of the information available to decision-makers, and let it navigate better and more safely in today’s turbulent world economy

 

  

---