For George Jaszi, retiring chief of the Bureau of Economic Analysis, charting the peaks and valleys of the nation's economic landscape is like riding in an airplane above the Earth.

"It's something esthetic," Jaszi mused in slightly accented English. Then the Hungarian-born economist with a Prussian relish for precision began searching for the perfect analogy.

"In an airplane," he began, "looking down you see mountains and a hill and that is beautiful -- agreed? For me, determining how the nation's output is generated, where it comes from and where it goes, is beautiful. It's like being in an airplane getting an overview of what's happening in the economy."

Jaszi, 69, has spent more than half a century providing the economic "big picture" for policy-makers, businessmen, economists and the public. His tools are ones that he helped create, including the concept of the gross national product and its component indices.

Each month, his Commerce Department division has apprised the nation of its economic health, how much we are producing and consuming, and the shifts in production. He tells how much we pay for goods, and how much we would have paid for them in 1972, now the standard for the "constant dollar." Before his national income accounts were created, America virtually had no statistical record of its economic past.

Now Jaszi, a graduate of the London School of Economics who began his government career at the Federal Reserve Board in 1937, is retiring for health reasons, after a back operation. When asked if he has any regrets about leaving a job he has held for nearly 23 years, and clearly loves, he replied, "Regrets? Yes! Terrible! I was planning to -- what is this phrase -- die in the harness. I like the work I was doing, I like to work with the people . . . . "

Then suddenly he was overcome by emotion. But only momentarily. With determined self-control, he fought back tears, pressed his fingers to his forehead and continued methodically with the interview.

It is a measure of Jaszi's contribution to the field of economic analysis that his National Income and Product Account has endured, while economic theory has shifted from Keynesian to monetarism to supply-side to the "rational expectations" approach.

Jaszi has little use for all that. "I'm sort of an an eclectic," he said. "Monetarism, I don't really believe in. Rational expectations -- I don't really know what it means."

His casual disdain for popular theories displays his determination to remain above the policy side of economic debate. He is not in the business of forecasting, and he is not in the business of telling politicians what will and won't work. And that, he said, is one reason he loved his job at BEA.

"One thing I enjoyed about my work was that I did not have to have answers to complex economic questions," he said. "If I were a policy-maker, and I was asked what will be the effect of a $100 billion budget cut on the economy, I would have to come up with an answer. But we are a research organization and we are not required to give policy advice."

The independence of the BEA has been respected over the decades by successive administrations. But Jaszi goes to great lengths to maintain that he never had to fight off a president or a Congress, and he was never asked to "cook" his numbers to paint a rosier economic picture.

"I did not have to be a watchdog," he said. "Not even in the darkest days of the Nixon administration was there any attempt to interfere with the objectivity of our work. I don't think I have spent any substantial amount of my time maintaining the integrity of the bureau."

Jaszi can recall only two instances in which he said he stood up for the BEA's independence. Once, during the Johnson administration, he was asked to turn over some statistics on businesses, so the White House could see how many were complying with voluntary restraints on foreign trade. He refused, arguing that his statistics should not be used to monitor businesses.

The second time was during the Nixon administration, when Commerce Secretary Frederick B. Dent was previewing a press release on the GNP. Jaszi remembers Dent picking up a pen and making a minor editing change, so a sentence would read more smoothly. Jaszi told him, "You are absolutely right, you improved it. But that is none of your business." And as Jaszi recalls it, Dent "ceased and desisted" and even wrote him an apologetic note soon after.

While Dent may have improved that particular sentence, one of the things Jaszi most liked about his job was that he had the "time to think and clarify concepts in my mind."

Clarity. It's not merely important to him -- it's the philosophy for the bureau, even enshrined on the wall above his desk at home in a saying by Confucius, part of which reads, "If concepts are not clear, words do not fit . . . . The noble person is never careless of his words." Jaszi insists on clarity of concepts and words and practices it, even in response to an interviewer.

Ask him if economics is a science or an art, and he replies, "You tell me what is a science and what is an art." Ask him if a certain trend in the economy may be artificial, and he demands that you define your term: "You use the word 'artificial' as if to say peculiar." And when asked to describe the economic trend for small businesses, he replies, "The term 'small business' is exceedingly vague."

On one subject Jaszi is clear, and that is what he considers the biggest change he has seen since since World War II: the slowdown in the nation's economic growth that occurred about 1972 and from which this country has not recovered.

"It is still not completely understood," he said. "Beginning about 1972, something happened -- probably several things happened. The rate of productivity has slowed down as has the rate of increase of constant dollar GNP."

That "blip" on the economic screen is still being debated by economists, who sharply disagree over what caused the slowdown and whether it is over. In that debate, Jaszi's position is well known. "Basically, we don't know what happened," he said. "I don't think we're pulling out of it."

Jaszi is aware of the criticism of the national income and production accounts, including the argument that simply measuring output does not take into consideration improvements in the quality of goods and services. The use of 1972 figures to set the "constant dollar" is also attacked as being dated.

"We don't just count the number of automobiles," he said, "we also take into account the quality of the automobile. But it is a very, very difficult subject. How do you compare telephones today to telephones in 1937? Or automobiles? It is so difficult to take into account quality change that it the constant-dollar GNP has to be used with caution -- 'handle with care, buyer beware.' "

He added, "It's better to have the constant-dollar GNP we have than to just have the current-dollar GNP," he said. "It's better to have it than to just give up on it. If we suddenly stopped publishing it, people would crawl on their knees and say, 'Please, please, please.' "