So you've at last put the finishing touches on your annual budget. My condolences. If you're like most chief executives I know, you probably feel as though you've just been through the business equivalent of an extended workout with Marvelous Marvin Hagler. Not only are you sore from bloody battles with department heads (when has any manager ever asked for less than the year before?), but even if you've won, as CEOs often do, what do you have to show for it? A budget based on assumptions that may or may not bear any resemblance to what will actually happen in the next 11 months.

Pointless and painful as the annual budgeting process often seems, it is one of those rituals that most CEOs feel compelled to go through. How else can they keep a lid on costs and maintain some measure of control? Well, consider James McManus. Four years ago, he began eliminating traditional budgeting at his company, Marketing Corporation of America (MCA). He swears now that it was one of the smartest moves he ever made.

McManus, you may recall, is CEO of a highly successful miniconglomerate in Westport, Conn., which he founded in 1971 ("Face-to-Face," September 1986). Until 1984, his budget process was little different from most. Year after bruising year, he would sit down with his managers and haggle over numbers. Should this department be permitted to add three people or four people? Why should so-and-so get a $10,000 raise? Why not $5,000? Was first-class air travel really necessary?

The process took time and a steep psychological toll -- all the more so because McManus was going up against smart, articulate salespeople, well versed in the art of persuasion. That, after all, was why he had hired them in the first place. But his greatest concern was that the resulting budgets did not adequately reflect what was happening in the real world. While the market continually changed, the budget "gave a false assumption of predictability," says McManus. And, more often than not, "I was too far away from the situation to have a right to a point of view."

Some of MCA's key managers shared his concern. One of them was John Luther, the head of MCA's 10-year-old development-consulting division, which advises companies on new-product and business strategies. Although the division was earning about 20% on revenues, budget constraints had kept Luther from rewarding consultants as well he would have liked. As a result, some top employees had been lured away be competitors willing to pay significantly more. Without sufficient incentives to hustle new accounts, moreover, the division's growth had stalled. Luther, who was working harder and enjoying it less, was worried.

Finally, he took his concerns to McManus, saying that he needed more operating room so that he could provide additional incentives and perks for his staff of 15. McManus was sympathetic, but leery of committing himself to higher fixed costs. The problem got him thinking, though, about alternatives. One idea that sprang to mind involved getting rid of traditional budgeting altogether. What if Luther simply agreed to deliver a certain percentage of sales to the corporate kitty? The margin would obviously have to be high enough for MCA to cover corporate overhead expenses and pretax profits, but why not let Luther decide for himself how to spend whatever else the division made? That way, he could pay higher salaries, offer bonuses, or what have you. Theoretically, he could even move down the road to cheaper offices. He could do pretty much anything he wanted, so long as MCA received the agreed-upon margin.

After much discussion, the two men decided to take a leap of faith and give the plan a whirl, settling on 15% as the magic number. In late 1984, Luther installed a new incentive program in his division, and sales took off. Over the next two years, the division's revenues grew at an annual clip of 30%, allowing Luther to pay out fatter bonuses than he had ever dreamed possible. MCA, meanwhile, has continued to receive 15? out of every dollar generated. As for McManus, he's been able to sit happily on the sidelines as routine decisions -- about adding staff, say, or raising prices -- are made without him.

Indeed, margin management (as the system is called) has worked so well that McManus now uses it with 10 of MCA's operating units, including the real estate, market research, advertising, and computer-software divisions. In each case, he negotiates a sensible margin with the division manager, who takes over from there. Not that budgets themselves have been eliminated. Every spring, each division manager still comes in with a sheet of preliminary projections for the coming year. (The company operates on a fiscal year ending June 30.) But so long as the division is sustaining the agreed-upon margin, and sales are continuing to increase at an annual rate of at least 5%, the budget discussion is usually quick (less than two hours) and painless. Its focus, says McManus, is on questions of strategy and quality control, rather than numbers. He and MCA's president, Bob Kamerschen, ask managers if they see any unusual risks in what they're doing. "If the answer is no, we say, 'Have a good year."

Nor is the authority granted to managers just for show. In the summer of 1985, for example, one division decided to spend $80,000 a year for four years to lease a powerful new word-processing and publishing system. A few months later, another division did likewise. In neither case was McManus's approval sought.

On the other hand, the division managers did have to sell their ideas to their respective staffs; the money spent on the computers could just as well have gone to sweeten the bonus pool. To avoid undermining morale, the managers had to persuade their people that the investment in equipment would eventually pay off in improved productivity, better quality, and long-term growth.

And therein lies the secret of margin management. "It's turned the whole damn budgeting process upside down," says Matt Freeman, who heads MCA's marketing-consulting division. The old system, in effect, encouraged managers to scramble for as many computers as they could get. "If I didn't get them, somebody else would," says Freeman. Now, those same managers are evaluating their needs and thinking about other, more effective, ways to allocate resources. Nor are the managers the only ones involved in the process. "If my division blows a tire," says Luther, "I'll have to answer to everyone." By giving so many employees a reason to keep score, margin management forces them, too, to understand the business and how it makes money, thus pushing budget concerns back down through the ranks.

When you think about it, that's really the most sensible way to run a business, and it helps explain what's wrong with the annual budgeting process as practiced at most companies. That process forces the CEO to make decisions about dividing up scarce resources -- a reasonable proposition when a company is small. But as the company grows, the CEO inevitably becomes more removed from its day-to-day operations, and less able to make wise decisions. Margin management turns the decision making over to the people who are in the best position to understand the implications of decisions concerning their respective areas. And if they're wrong, they feel it where it counts -- in their paychecks.

McManus admits that margin management does not work in every circumstance. In start-ups, for example, margins tend to be low, if they exist at all. And even if you can set a margin in the early days, it's hard to raise it later on, as the business matures. "People are liable to feel that you're taking away their bonuses," he says. For that reason, he has delayed adopting the system in some of MCA's developing businesses.

There may also be problems applying margin management to unstable businesses subject to major shifts in the marketplace. For the system to work, "There has to be total agreement on the numbers you use," says McManus. "You have to choose your figure when things are relatively stable, and then everyone has to live with it. That's the beauty of it."

With these exceptions, however, McManus believes that margin management could improve the performance of almost any business. Granted, it's hard to know specifically how the system might be adapted to, say, a one-product manufacturer or a small retail operation. But there's no reason to suppose that the principles wouldn't apply. Those principles involve: (1) agreeing on a fixed margin; (2) pushing decision-making responsibility down the organization; and (3) providing real incentives for department managers and employees -- allowing them, for example, to keep a portion of the incremental revenues they generate. From that point on, the manager must decide how best to deliver the margin, whether by raising salespeople's commissions, or reorganizing production, or whatever.

Of course, the key to making this work lies with the CEO, who must be willing to accept a specified level of profitability, and to redefine his or her role. McManus views himself more as a strategist these days, looking for new businesses that MCA can get into. He worries, at times, about managers becoming so preoccupied with running their own divisions profitably that they stop noticing attrative business opportunities, especially those that may require large commitments of capital. To correct this, he often reminds managers to watch for investments the company can make with its own pot of money. "One of my biggest jobs," he says, "is to figure out how to invest the residue."

McManus evidently enjoys this role enough to accept the fact that he may be giving up some profit -- something he admits is not always easy. "A lot of business owners find it hard to put a limit on their upside return," says McManus. "They can't be satisfied with 15% when they think they could be earning 25% or 30%." But he can, and -- if you dislike budgeting as much as McManus does -- you, too, may learn to be satisfied with a nice, steady return. As McManus likes to say, "I traded upside surprises for the absence of downside surprises." That, in the long run, may be the best deal of all.