A LITTLE CLARITY PLEASE
How the ever-unpredictable future is reshaping the annual budgeting process.
By Abe de Ramos

The mile-thick brown cloud that has been hanging above the Indian subcontinent since summer has given Mohandas Pai a coughing fit. But the CFO of Bangalore-based Infosys Technologies does not mind the discomfort as much as he does the lack of visibility driving home. The fog seems to have added mocking insult to practical injury: US$537 million-a-year Infosys has been dealing with poor revenue visibility since the software industry took a beating early last year. Then came September 11, the nuclear posturing this spring by India and Pakistan, followed by the frightening prospect of a US war against Iraq. Pai should be thankful that all he has is a bad case of coughing.

"Our ability to see the future has been very much impaired," sighs Pai. Markets have always been fickle and acts of God and war eternally upset corporate planning, but Pai sees this budgeting season being as tough to absorb as last year's.And it's unlikely to get better any time soon. The dressing down of the technology sector, the global recession, and explosive ideological conflicts have all changed the way predictions are made. All of which means that Pai, like most Asian CFOs, is forced to conjure up ways to make plans that will keep the business resilient to future shocks. In short, the traditional planning and budgeting process no longer holds.

As the American aphorist Mason Cooley once said, a budget takes the fun out of money. In the same breath, it can be said that traditional budgets take the plan out of strategic plans. The fact that they fix expenditures and strategy a year ahead makes them out of tune with the rough-and-tumble of market reality. In the current environment especially, a budget is likely to become irrelevant even as it's being constructed. This, in itself, is a tragedy. After all, the annual budget is supposed to be the execution of a strategic plan.

Budgets stultify decision-making by putting up bureaucratic impediments to the reasonable and fast allocation of resources. They strip from managers the autonomy necessary to make spending decisions based on their closer knowledge of the market and at the same time give them mechanisms to avoid accountability. Not mincing his words, Pai says: "A rigid budget is a barrier to change." And like CFOs in Asia come September, he's refining his method for next year.

Toss It Where?

Most Asian CFOs feel themselves to be the sufferers of the breakdown of the traditional budgeting process, and want some form of change. "CFOs all over Asia are unhappy with their planning and budgeting processes and systems," says Matthew Podrebarac, partner at management consulting firm Accenture in Hong Kong.

Their complaints are unanimous. First, budgeting takes too long. Hackett Benchmarking & Research, a US consulting firm, says the average billion-dollar company takes up to six months to prepare a budget, equivalent to 25,000 person days. Accenture, in a survey last year, says the process typically eats up around 30 percent of senior managers' time.

Gripe number two: businesses don't run on yearly cycles, and market trends don't march in an orderly process across the calendar. So why should the budget? The absurdity of this situation is captured by Lim How Teck, CFO of Singaporean shipping and logistics group Neptune Orient Lines (NOL). "When you have a fiscal January to December," Lim says, "if you're looking at July now, nobody thinks about January to June next year. The planning horizon gets shorter and shorter until you finish in December."

Gripe number three: budgets develop a tunnel vision in terms of benchmarking, as managers tend to act according to internal targets, not according to competition. In a survey last year of 113 billion-dollar companies in Asia, Europe and the US, the Balanced Scorecard Interest Group, a research unit of US-based The Conference Board, an elite group of international chief executives, found that 64 percent of companies compare their performance against internal budgets, and only 52 percent compare themselves against competitors.

Last, and most important to the academic crowd, is that compensation tied to annual budgets (see box, "Can Stretch Goals Stretch Revenues?") encourage gaming. If managers exceed the previous year's targets, their natural incentive is to cut back. They've already earned their bonus, so why continue creating profits that will encourage their bosses to raise their targets for the next year? A similar evil is underestimating the budget. Line managers are wont to lowball their estimates so that target bonuses will be easier to achieve. This is the reason former General Electric CEO Jack Welch has called the annual budget "the bane of corporate America" and "an exercise in minimalization".

This September, more Asian CFOs are starting to act on their dissatisfaction. But they are taking measures one step at a time. The CFOs interviewed for this article all agree that the one-year, annual budget process is out. But they are trying to work within the system of a modified traditional budget framework. Like the "beyond budgeting" companies in the West that have altered their budget methods but have not gone all the way, some have introduced rolling budgets. All have retained the 12-month horizon.

None of this is easy even under normal circumstances. Old hands can try for years to gather market information, and still end up with too many widgets on the shelf. Mahadevan Sridhar, vice-president for financial planning for Kodak India, for example, admits that he has never been able to devise a reliable system that can forecast how long the wedding season will last. Year after year, a mundane combination of monsoon rains and shifting tides has stymied his ability to get inventory and logistics just right. But the wrong decision has a large impact on cost. Larger variables - from September 11 to a surprise fall-off in tech spending - are much harder to deal with.

Rock and Roll Over

Asian reformers are taking cues from companies like Ericsson, the Swedish telecommunications giant that has exported its budget reforms to its units in Asia. "We used to have yearly budgets, but they have been replaced by four quarters of rolling forecasts; we even got rid of the old term by calling it a forecast," says Terje Nodland, vice-president for business control at Ericsson's Asia Pacific headquarters in Hong Kong. "The traditional budget focuses on too long a time horizon. A 12-month focus is too long for the rapid changes that we see in the environment," he says. What Ericsson is doing, says Nodland, is to avoid this focus on the calendar year.

"We update the budget every quarter, and the period of the forecasts that the budgets cover, four quarters, is rolling," says Nodland. "It gives us the forewarning of events that we expect to take place, and that in turn allows us to take corrective action at an appropriate point in time. Otherwise, it would have taken us a longer time to react, for example, in terms of rightsizing the organization," he says.

Nodland, of course, is referring to the 30,000 person cut in Ericsson's workforce last year, and a reduction in market units from 100 to 35. The result was a saving of US$749 million, and positive cashflow.

Ericsson's quarterly forecasts are not guesstimates. They include every item in a profit and loss statement and balance sheet, from marketing and personnel expenses to interest cover and exchange rates. And they're reported by each country unit. To maintain control over the assumptions of each unit, the regional finance managers in Hong Kong "try to influence the levels that market units report in their forecasts" by giving their own input on how they think the market is going to be. To avoid the likelihood of units under-forecasting, "the targets are generally set on aggressive levels, and that is intended to keep the market units on their toes," says Nodland.

Pie in the Sky

Infosys's Pai began seeking this kind of forecasting as early as last fiscal year starting April 2001. At that time, the software industry was showing signs of weakness, and the recession in the US, where Infosys generates about 71 percent of revenues, was taking shape. Pai got the hint when he observed that clients themselves no longer allocated software budgets on an annual basis; more of them have delayed orders, and firmed them up only on a quarterly basis.

As such, the flow of revenues, even from existing clients, was not as visible as it has been in the past nine years. "For the first time in several years we had to face intense pricing pressure and lower incremental business," says Pai. "Revenue visibility was low and we needed to ensure that we did not build a fixed cost structure for the entire fiscal year. We had to move to quarterly budgeting to align our cost structure with revenue visibility," he says.

Like many companies, Infosys has an annual budget guided by a five-year vision. The annual figures are set on a zero-based budgeting basis - in which budgets are made as if no baseline existed - by seeking justifications for every item of expenditure. (The traditional budgeting process typically begins with last year's expenditures as the baseline, and then adds a percent of inflation.)

For Infosys, however, the annual budget is just the starting point. It is revisited every quarter, and forecasts for the next four quarters are also made. In short, the money is only released on a quarterly basis. How does it work? Pai manages his budget by dividing costs into firm and discretionary. Firm budgets include salary cost, the main expenditure of a software company, while discretionary budgets include marketing and selling initiatives, among others.

The next quarter's budget is firmed up only weeks before the current quarter ends. "We are into a quarterly mode whereby we revisit all parameters of the budget on a continuous basis. We have rolling budgets for the next four quarters and the same get firmed up as we near a quarter," he explains. "Depending on what we achieve for this quarter, next quarter's discretionary expenditure is released ... We want to make sure that we don't build up a large cost base to meet any potential revenues, [when] the revenues themselves turn out to be very soft."

The new process meant that as Infosys focused on cost-cutting due to much-reduced revenue growth, Pai shifted the budget structure towards greater discretionary expenditure. "Discretionary budget is something that can be cut in a quarter if things go really bad," he says. On the flip side, they are also put up for grabs for the business units, and are awarded on an "earn as you win" basis. In short, Pai gives management back to unit managers, and this empowerment fosters "domain competency". "If I have an initiative, and my colleague has an initiative, they will be evaluated to see which is best, and money will be given," he says.

"We have formed groups which project the initiatives they are going to take, and the returns that they expect," Pai adds. "We value those initiatives based upon the internal metrics, and the one that gives the highest return is what we first take up," he says. The winning line managers allocate the funds as they see fit. The "earn as you win" process is then repeated into the next quarter.

So far, so good. When Infosys projected last year that sales would grow just 30 percent, analysts thought it was too conservative. But the foresight of weak demand enabled Infosys to adjust its labor utilization without firing employees. At one point, it put as much as 30 percent of its "Infoscions" on the bench, and used some of them when demand picked up again from January to March this year. As such, Infosys managed to tame its costs quickly when demand was slow, and just as quickly ramp production up when demand came back. In the end, Pai's actual revenue came 7 percent ahead of the projection.

The Shipping News

Lim, CFO of US$4.7 billion-a-year NOL, is another convert of 12-month rolling forecasts, which he implemented only this fiscal year starting January. In keeping with Ericsson's aversion to the terminology, Lim says: "We have a revised budget, but we don't call it that; we call it current outlook."

Each forecast begins with inputs from each unit on the likely volumes that NOL will ship, going down to trade routes. They then project what the freight rates are likely to be. Fixed costs, carrier costs and depreciation costs are later considered. Senior managers at the group level then make assumptions on interest rates, exchange rates and oil prices. "There are people doing things on the revenue side, other people on the cost side, and as controller of the various regions, we look at some of the more local factors," says Lim. "Otherwise if you ask each one of them what the oil price is, everyone will come up with a different price," he says.

The forecasts are done quarterly, but already Lim hopes to be able to do them monthly. "When this process becomes very well defined, we may not even have to spend so much time on doing the budget as it becomes a monthly thing," says Lim. "The budget then becomes a daily ritual, not a once-a-year situation," he says.

Like Nodland, Lim believes that rolling forecasts could cushion future shocks. In the traditional process, budget reports are made on a historical basis, say, the performance in the first two months of the year versus the plan and the previous year. "If you report the period and you're on target, everybody's happy," says Lim. "But they don't know there's a storm brewing in the next ten months which will throw your whole year budget out. There is no peek into the future and that can be dangerous," he says.

If only Lim had done this sooner. His adoption of rolling forecasts was a response to shocks last year when freight rates fell to unprecedented, not to say unexpected, levels, punctuated only by September 11. For NOL, the supply-over-demand situation was not just an industry phenomenon; it was also self-inflicted.

NOL in 2000 attained record revenues, and forecast a robust 2001. This led to a series of acquisitions, not just of other companies, but also of ships and tankers, most of them funded by debt. They may have been strategic purchases, but they were bought at a time when valuations were still holding up. The result: NOL incurred a US$57 million loss last year, and a share price that spiraled down. Though Lim expects to return to profitability this year, investors apparently think otherwise. NOL shares currently trade at S$0.75, below its issue price of S$1 (US$0.57). Lim now hopes that the move to rolling forecasts will help NOL position for the future with at least a little more accuracy, and thus preserve shareholder value. The practice, he says, makes managers more quantitative and conscious of the bottom line and value creation. "Not to predict means to be blind," says Lim. "You can't wait until you're running out of cash before you predict your cashflow situation; then you'd start worrying and you'd be in big trouble," he says.

Share the Love

There are, of course, aids to visibility, and Lim advocates investing in them. NOL, for example, has just invested in See Change, an ERP implementation that opens an interface with clients, the better to forecast demand. "When you match it against the savings that you achieve, getting the information ahead of time, the change in mindset, once you get back into a growth track, it's time and money well spent," says Lim.

It was this kind of computer power that allowed Morten Schott Knudsen, regional financial controller at Dumex, part of the Danish conglomerate East Asiatic (EAC), to see the possibilities in reforming the company's budget. The US$1 billion-a-year group, which has interests in nutritional foods, industrial ingredients and relocation services, has implemented Hyperion since 1994 to consolidate group accounts. Two years ago Knudsen started to do 12-month rolling forecasts in Dumex, the group's largest business unit. He deployed the available technology for planning and budgeting purposes, down to the sales, product cost, marketing and purchasing levels.

EAC's budgeting process starts off pretty typically. It begins in July with a five-year planning forecast - market expectations, targets and key performance indicators - prepared at the business unit level but based on assumptions from the headquarters in Singapore. Using that, business units then prepare their annual budgets. This is where EAC departs from the old school.

EAC deploys Hyperion applications to operating levels for specific budget and forecast inputs. "It's more like a distributed database, so the line manager has more control of the process," explains Knudsen. "Let's say you're responsible for your own expenses. You can then measure the costs for your departments, the very detailed, line-by-line items. You can expand the data fields as much as you want; you can receive assumptions from the corporate office, key in your details and submit it to them again," he says.

Each EAC unit, such as Dumex, which makes child nutrition products, or Santa Fe, a freight forwarder, uses a different Hyperion model for low-level planning and budgeting because of differing requirements. Nevertheless, they all link to the same financial reporting system, Hyperion Enterprise, which the corporate office in Singapore uses to see lower level assumptions, consolidate them, and adjust them based on higher level assumptions, such as exchange rates.

Because EAC uses 12-month rolling forecasts, the whole process is far from being a once-a-year event. "Each unit, every month, gives an update of changes in assumptions for the forecast, and that is reviewed very carefully," says Knudsen. "Every second month, we have management meetings where we go through the major changes, so costs and investments are very well controlled," he says.

How has EAC benefited from the system? Through more accurate forecasts, which, in effect, is just a result of democratizing the budgeting process. "Some financial forecasts used to be done only by the finance department, but today there is much more interaction between departments," he says.

Podrebarac at Accenture nods at the effort. "If you have the right process and right technology, you would involve more people," he says. "What you're really trying to predict are sales and costs, and the people who know that best are those out in your markets."
Who knows, with innovations like these, line managers - and their CFOs - may stop wheezing when budget time rolls around next year.

Abe De Ramos is a senior writer at CFO Asia based in Hong Kong.