A Different Way of Looking at Salary - and Responsibility

FROM President Bush's visit to Japan and the mixed reviews it has drawn has arisen a wave of introspection about American industry.

Much was made of how the salaries of top executives are so much higher a multiple of the earnings of shop-floor workers in America than is the case in Japan.

From the Levinson Institute, in Belmont, Mass., has come a theory for understanding executive pay and, more important, the issues of corporate accountability and responsibility.

Nearly 40 years ago, Dr. Elliott Jacques hit upon a theory for the proper "layering" of an organization. It's called "stratified systems theory" and it posits a maximum of seven different "strata" based on the complexity of tasks individuals perform and the maximum time spans over which they are responsible for these tasks.

Stratum I employees - operators or clerks - have a time horizon of one day up to three months. They are normally supervised by first-line managers (Stratum II) with a time span of three months up to a year. And so on up to the chief executive or chairman of a large corporation, who must think 20 years out and beyond.

The leader of a smaller organization may not need to be more than a Stratum V, however, with a responsibility time span of five to 10 years; and so five layers is all that organization needs.

Dr. Jacques was surprised in his original research to find how universal the different layers with their respective time spans are. People in roles at the same time span experience the same weight of responsibility and declare the same level of pay to be fair, regardless of their occupation or actual pay," he wrote a couple of years ago in the Harvard Business Review.

He was also surprised at how sharp and clear the distinctions are from one time-span layer, with its concomitant level of task complexity, to the next just as ice changes to water and water to steam at certain specific temperatures."

He gives an example of a corporation with two divisions, one that publishes books and one that makes engineering products. The latter has a much larger staff and greater sales volume. But the presidents of the two divisions each have the same responsibility time span, of five to 10 years, and their "felt-fair pay" is the same.

All this discussion of hierarchy may sound out of step for a period in which we're all supposed to be networking on our computers, working as teams.

But Jacques defends hierarchy as a way to "release energy and creativity, rationalize productivity, and actually improve morale." Moreover, he adds, "I think most managers know this intuitively and have only lacked a workable structure..."

He wants to see hierarchies working better, not abandoned. And his stratified-systems theory is intended to help organizations come up with rational structures. Once they have the right structure, leaders can ensure they have the right firepower for each task - Stratum III challenges assigned to Stratum III employees, for instance.

A common problem, Jacques has found, is that organizations confuse pay grades with management layers. An individual is nominally "supervised" by one who earns more but doesn't really operate on a higher stratum. This leads to the employee feeling crowded by the supervisor and instinctively seeking out direction from the supervisor's boss, who the employee rightly feels does operate at a different level.

Gerald Kraines, president of the Levinson Institute, resolves the apparent conflict between hierarchy and networking by pointing out that in customary business practice, "A team is not accountable." Individuals are.

Informal discussions, free sharing of ideas, and so on are all fine within the context of clearly defined zones of authority and responsibility. But the one at whose desk the buck stops must, individually, take responsibility for decisions taken and work done. "Networked" input helps make better decisions but does not replace accountability.

The Levinson team may also seem to promote unduly long time spans. After all, "generation" seems to mean about four or five years nowadays, much less when we talk about technologies. Change is the only constant, and everybody needs to expect have more than one career over a lifetime, we are told.

But Jacques's time spans seem to be still holding. In fact, it may be more important than ever for individuals who may need to be thinking 10 or 20 years out, even in jobs they will hold for only three years.

Two heartening inferences can be drawn:

* Individuals can in effect promote themselves within an organization by taking responsibility for the whole and by learning to think in terms of wider time spans.

* The best way to work at a place where you will be forever is the best way to work at a place where you'll be for a short time.

"Time span has to do with intention rather than fact," says Dr. Kraines. He repeats the story of the French general who asks that olive trees be planted on his estate. "But it will be a hundred years before they bear any fruit," his servant protests. "In that case, plant them immediately!"