Contributing to this article was Frank Donahue
of The Inquirer's News Research Library.

DATELINE:

For 14 years, Citicorp Inc. paid mathematician John Davies to help the giant banking firm reduce its credit risks.

Then in 1991 Citicorp eliminated his job, as part of a massive restructuring that slashed the payroll by 14,000 people.

Davies, who led the bank's advanced-technology group in New York City, headed west to New Mexico and Los Alamos National Laboratory. But Davies was hardly adrift. In fact, he's still grappling with Citicorp's credit-risk problems - with one notable difference.

Now, American taxpayers are footing the bill - not Citicorp.

The bank is getting $9.6 million worth of free technical assistance from Los Alamos and three other federal laboratories - government aid for a company with $250 billion in assets and $3.4 billion in profits last year.

Citicorp is just one of more than 1,000 companies, large and small, receiving federal research help to develop technologies, solve manufacturing problems or sell new products.

That's because the government's vast research establishment that originally grew up to protect national security now has a new mission: propping up American industry. Six billion dollars of taxpayers' money was spent last year alone to help companies, mostly large corporations, compete.

To date, no one knows - no one has devised a measure to tell - if most, or even if any, of the money was well-spent. There's no doubt, though, that it has created precious few jobs or industries.

The original impetus was twofold:

First, to convert swords to plowshares - that is, to turn both big corporations and workers who had never been employed outside the defense establishment into productive members of the civilian economy. And thus ensure high-paying jobs for Americans in the 21st century.

And second, through government-industry partnerships, to help the United States catch up with managed economies elsewhere that threatened to outstrip ours.

So, for the first time in the nation's history, the doors to the government's far-flung research facilities were thrown open to industry - access to more than 30,000 scientists and engineers in a sprawling network of 700 federal laboratories.

All along, the government's rationale for spending this money was to create jobs. Sounds great. But it doesn't work.

The jobs have not materialized. What has resulted is a huge chunk of corporate welfare that's not doing anyone but its recipients any good.

Yet even now - at the same time that congressional budget-cutters are threatening to slash school-lunch programs, student loans and health care for the elderly - Congress continues to hand out hundreds of millions of dollars in research subsidies to some of America's richest corporations.

President Clinton, a firm believer, says these new partnerships will redefine the relationship between government and industry, and will make America stronger and more competitive.

Helping firms such as Citicorp compete in a fierce global market is vital to the national interest, Clinton says.

The payback, believers say, will be a High-Tech Promised Land - an array of new technologies generating hundreds of thousands of high-wage, high-skill jobs to keep America strong.

To that end, Congress has created more than 50 government programs aimed at boosting U.S. companies.

So what are Americans getting for their $6 billion-a-year investment?

Consider a few taxpayer-funded projects The Inquirer found during a yearlong examination of government-sponsored technology programs:

• Fancier fireworks at Disney World.
• A better way to peel chili peppers.
• A coloring book titled "Technology Transfer With Space Pup."
• Solar-powered watering tanks for livestock in southern Africa.
• A study on the benefits of planting shade trees in big cities.
• An Outward Bound-type program intended to turn government rocket scientists into business tycoons.

Not much. Neither do many of the government's other efforts in its rapidly expanding technology-giveaway programs - programs that have exploded well beyond the government's ability to keep track of them.

The new policy percolated in think tanks and universities throughout the 1980s, and was heartily endorsed by the Clinton administration when it came to power: Government would become an active partner with industry, deciding technologies for corporations to invest in, paying for private R&D with public
funds, setting up centers to aid manufacturers, and steering billions of
dollars of taxpayers' money to select industries.

Washington was trying to create the best of both worlds - a free-market economy shored up by the government's ample, and idle, technological resources.

It was trying, in short, to graft an elephant onto a camel.

That policy has failed. And for good reason. It was based on a flawed concept. America wasn't Japan, whose culture and government were markedly different from our own. Overlaying a planned economy on a democratic, free- market system has proved an unworkable mix.

So, instead of a well-oiled machine, the nation has gotten a patchwork of disconnected, poorly managed programs benefiting select companies and individuals at the expense of others.

Most of these programs have been big on promises but short on results. Not only have they failed to produce many jobs, but the programs suffer from duplication, inflated management costs and exaggerated claims of success.

Nobody disputes that new technologies help to build a stronger economy. There certainly are some success stories among the government's technology- subsidy programs - especially those helping small firms gain a foothold in the marketplace.

But those success stories are few and far between, The Inquirer's study found.

Far more common are such examples as the Energy-Related Inventions Program of the Energy Department. Since 1975, the government has spent $76.2 million to foster energy-saving technologies. A total of 129 technologies have been commercialized.

As of 1992 - or 17 years after the program's creation - exactly 668 people were employed by all the companies involved.

Or consider the Energy Department's popular Cooperative Research and Development Agreements. Taxpayers have spent $792 million over four years on them. The department says 46 companies have been created.

That works out to more than $17 million to create each one.

Sound like a wise investment?

Now, try the return on Energy's $1.7-billion technology-transfer investment. That program is expected to generate licensing and royalty income this year of $3.4 million - a return of one-twentieth of one percent.

That's like putting $1,000 in your savings account and getting $2 in interest a year later.

Item. The stated purpose of these programs is to create jobs. Yet many of the corporations the government is subsidizing have been eliminating jobs - not creating them.

AT&T, Amoco, General Electric, General Motors, IBM, DuPont and Citicorp, all among the biggest winners of subsidies, have sent 329,000 workers packing in the last five years.

That was while they were collecting more than $293 million in taxpayer handouts.

GM alone has been awarded more than $110 million in technology subsidies, records show. And that is a conservative estimate. The company has shared in other grants that the government does not closely track.

Since 1990, the huge multinational has slashed its domestic payroll by 104,000 workers, or 25 percent of its U.S. workforce, to boost productivity. It apparently worked. In 1994, GM posted a $4.9 billion profit.

Item. Taxpayers are subsidizing industries that have shifted more than 100,000 jobs offshore, to nations with cheap labor and few regulations.

Take the semiconductor industry, the champion of the government's technology lottery, with more than $1 billion worth of taxpayer handouts.

Congress and the Clinton administration have afforded giant semiconductor firms special status in Washington, including their own advisory council. Washington bureaucrats speak in glowing terms about the industry's high-wage jobs.

What they don't say is that half of those jobs are scattered all over the
globe in places such as Ireland, Malaysia, Singapore, Taiwan and South Korea.

Item. There are no guarantees that these new taxpayer-subsidized technologies will be made in the United States, providing jobs for Americans.

In 1993, the Clinton administration backtracked from requirements that technologies developed with taxpayer funds be manufactured in America. The Big Three automakers, semiconductor and computer firms opposed the regulation. Now, companies are required only to try to manufacture products here.

So even if government and industry develop marketable new technologies with your $6 billion in tax dollars, there may not be a job for you.

"This jobs thing is a sticky issue," said Duane Sunderman, former director of the National Renewable Energy Laboratory in Golden, Colo., which develops technology using such sources as wind and solar power.

"The companies will have to move overseas. That's the way it's going to be done. All the knowledge will come from the U.S., but all the work will be done by short brown fingers."

Item. Many of the large corporations that dominate the government's technology programs have axed their own research-and-development programs to boost short-term profits for shareholders. Taxpayers are picking up some of the slack.

Take IBM. The computer giant pared its R&D budget by $2.2 billion, or one- third, to $4.4 billion, between 1992 and 1994. "The reductions reflect the company's focus on productivity and expense controls," IBM reported in its 1994 annual filing with the Securities and Exchange Commission.

Meanwhile, IBM was lining up taxpayer-subsidized research projects with government labs. Between 1992 and 1994, IBM collected at least 33 agreements worth more than $58 million, records show.

Item. Pork reigns.

With hundreds of millions of dollars available and scant oversight, technology-transfer programs have become the congressional pork-barrel programs of the 1990s.

Congressional "earmarks" - money slipped into appropriations bills by members for pet projects - bypass scientific peer review and may, in fact, have little to do with an agency's official mission.

That explains why the Department of Energy spent $124 million for buildings at medical institutions in Oregon and Louisiana and why the Defense Department is paying $20 million to buy and reclaim abandoned coal land along the Susquehanna River in northeastern Pennsylvania.

Rep. George E. Brown Jr. (D., Calif.), who has exposed and opposed earmarks by his colleagues, says Energy Department officials regard them as "the cost of doing business."

Earmarks can be found throughout the government's technology programs.

Item. Washington has a generous definition of what constitutes an American company.

Among the winners of technology subsidies are firms from South Korea, the former Soviet Union, Germany, France and England. U.S. taxpayers have helped to underwrite upward of $75 million worth of research for foreign companies.

And that estimate is low. The government has overlooked some foreign firms and does not count the value of technologies it gives away free or sells to foreign nations.

Item. Government officials say their technology giveaways are not corporate welfare because companies share the cost.

But the government has a strange way of measuring costs. It does not, for example, send the companies a bill for using the labs or consulting with government scientists. Taxpayers pay for that.

The government defines cost-sharing as the expense to the companies of participating in joint research projects - salaries and benefits of the companies' own researchers, travel costs and work back at the plant. Only in the rarest cases does cash change hands.

Audits also are rare.

How does the government verify the accuracy of corporate contributions? It doesn't.

Says Richard K. Quisenberry, who heads a consortium of textile firms working with the labs: "Most of it is a lick and a promise."

A good example is a program known as Cooperative Research and Development Agreements, or CRADAs, in which companies can "rent" government scientists and laboratories for work on company projects.

Taxpayers pick up about half the cost.

Over the last four years, the largest federal laboratories, which formerly subsisted on weapons research, have scrambled to find new missions and new markets. Thus, the number of lab-industry agreements has soared.

At laboratories owned by the U.S. Department of Energy alone, the number of CRADAs spiraled from 12 in April 1991 to nearly 1,400 in May 1995.

Yet the same Congress that pushed industry-lab partnerships under the guise of aiding U.S. competitiveness has not bothered to examine those agreements in any detail.

Members of committees with oversight responsibility for the Energy Department said they had never seen a list of the agency's industry agreements, let alone studied them.

"That's something we probably should do," one chagrined staff member said.

Among the agreements Congress would find is one, signed in November 1994, involving Sandia National Laboratories in Albuquerque and Walt Disney Co.

Taxpayers are kicking in $300,000 to help Disney officials figure out a better way to launch 3,000 rockets nightly as part of their fireworks displays at Florida, California and other theme parks.

Disney's share of the CRADA includes providing the fireworks, labor and a place to test a new ignition device.

"We're bringing the actual pyrotechnic devices, the charges," explained Disney vice president Dick Wiedenbeck.

One could fairly ask why Disney needs a taxpayer handout at all.

In 1994, the company reported a profit of $1.1 billion - an amount equal to Sandia's entire annual budget - on record revenues of more than $10 billion. Disney's CEO, Michael Eisner, collected $10.6 million in pay, including a $9.9 million bonus.

Sandia officials said the agreement served a dual purpose, aiding Disney and helping to ensure a supply of ignition devices.

As for creating high-paying jobs?

"I don't know what kind of job implications there might be," Wiedenbeck said. "I don't see it creating jobs for Disney."

In 1980, Congress opened the doors of the federal laboratories a crack, with passage of the Stevenson-Wydler Technology Innovation Act. It required the labs to set aside 0.5 percent of their budgets to promote the transfer of technology to the private sector.

It was the first of what would become, in just over a decade, a flood of legislation, programs and expenditures with an identical goal: boosting the U.S. economy by putting the government's vast network of research centers and scientists to work for industry.

A confluence of forces had driven Congress to this point. America's technical prowess seemingly had eroded. Its productivity levels had fallen. And wages of middle-class Americans had stagnated.

At the same time, rising economic stars such as Germany, Japan and its Pacific Rim neighbors had leapfrogged ahead of the United States on many fronts.

A small but influential chorus of academics, policy-makers and members of Congress pointed to a common element in their success: government intervention in critical economic strategies. That included making the resources of government available to select industries.

If the United States were to compete, a closer link had to be forged between government and industry, proponents of a new policy argued, with Washington and corporate America working together to return the nation to economic supremacy.

One step in the process was to enlist the government's vast network of research laboratories and scientists in the economic battle. The labs appeared to be a logical choice. They had lots of technology. And with the Cold War
winding down, and defense spending under attack, their high level of funding was dependent on finding new missions.

Thus was born the new policy - government choosing technologies, paying for research, steering billions of dollars to private companies.

That had never been the case before.

Historically in this country, the movement of technology from federal laboratory to private marketplace was a serendipitous event, in which government played a secondary role. The idea of direct involvement with the public was foreign to most in the government research facilities.

Yet demand for spin-offs was growing in Congress and in the business world.

Why shouldn't American industry reap the benefits of the expertise that existed in the labs? asked spin-off advocates. After all, taxpayers were paying billions for it.

A restless Congress was not satisfied. During the mid-1980s, a small group of legislators, many of whom sat on committees controlling spending for research and development, believed the government needed to do more. And the labs, aware that defense spending was dwindling, were in search of new work.

Among those leading this effort were Sens. Jeff Bingaman (D., N.M.), Pete Domenici (R., N.M.), Ernest Hollings (D., S.C.) - and Albert Gore Jr. of Tennessee, now the vice president. All had large national laboratories or weapons plants in their home districts.

Congress stepped in again in 1986, authorizing Cooperative Research and Development Agreements between private industry and government agencies.

The idea was simple: A company, large or small, that needed assistance in solving a manufacturing, design or research problem could contract with a federal laboratory for scientists and sophisticated equipment - with taxpayers picking up the tab.

The collaboration would succeed because industry - not lab scientists - identified the problem to be solved, ensuring a market for whatever technology was devised.

"You don't push technology out of the lab," explained Lewis M. Branscomb, who heads the Science, Technology and Public Policy Program at Harvard University's John F. Kennedy School of Government. "That's like pushing a wet noodle through a sieve."

New legislation and programs followed at a rapid pace. The National Bureau of Standards, the venerable institution that keeps the country's official time, became the National Institute of Standards and Technology.

The first of 44 federally funded centers was opened to help manufacturers improve productivity. The Defense Department steered nearly $800 million to a consortium of semiconductor manufacturers to develop breakthrough technologies. And the Energy Department's huge weapons and energy laboratories were granted new authority to cut deals with industry.

By the time the 1992 elections rolled around, the days of laissez-faire technology policy were all but over. And Bill Clinton, an unabashed supporter of an activist government, was leading the charge.

"Leadership in developing and commercializing new technologies is critical to regaining industrial leadership, creating high-wage jobs, and ensuring our long-term prosperity," proclaimed a Clinton campaign document. "Government can and must support these efforts. Our most successful competitors help their businesses - large and small - to compete more effectively."

Economic competitiveness. The words became a mantra during the 1992 campaign. Clinton knew he had struck a chord with American voters alarmed by years of massive layoffs and shrinking paychecks.

By the end of the summer, Clinton held a 20 percent lead in voter surveys, and his aides were already drafting post-election technology policies.

Almost overnight, the machinery of government shifted into technology overdrive. The list of programs created could fill a small telephone directory:

Advanced Technological Education Program, Ballistic Missile Defense Organization Technology Applications Program, Advanced Civilian Technology Strategy, Critical Technologies Institute, Defense Technology Transition Working Group, Environmental Technology Initiative, Federal Laboratory Consortium for Technology Transfer, Manufacturing Science and Technology Program.

Industry responded with equal speed, forming trade groups and research organizations eager to give advice - and take money.

New programs cost money, and for the first two years of the Clinton administration, the Democratic Congress willingly obliged.

Funding for the Commerce Department's Advanced Technology Program jumped
from $68 million in 1992, the last year of the Bush administration, to $431 million three years later - a six-fold increase. And the new administration promised it would grow to $750 million by 1997.

But even as the administration was touting the success of its partnership with corporate America, its new policy - overlaying planned economics on a chaotic, democratic free-market system - was failing.

After the Republican landslide in November 1994, Clinton's technology bandwagon hit a pothole.

Republicans searching for ways to fulfill their campaign promises to slash big government suddenly saw some technology-transfer programs as ripe for cutting.

Earlier this spring, they voted to pare the administration's funding for three technology programs by a total of more than $200 million. That's out of an estimated $6 billion for technology programs.

"These expenditures are examples of unneeded corporate welfare, wasted in a market that already produces world-class technology," Edward L. Hudgins of the Cato Institute, a libertarian Washington think tank, testified in March. ''It is of such expenditures that budget deficits measured in the hundreds of billions are made."

Among the recipients of funds, Hudgins noted, were 3M, Caterpillar, IBM, Xerox and Texas Instruments.

"These are hardly new, poverty-stricken, desperately struggling businesses that cannot fend for themselves without corporate welfare," he testified.

And then there is Citicorp.

John Davies knew firsthand of Citicorp's financial difficulties. In 1990, the balding, puckish mathematician was heading the bank's advanced-technology group, overseeing its massive computer databases.

He watched as Citicorp went through a wholesale restructuring to stanch the flow of red ink. His division was dissolved, and when he was assigned elsewhere in the company, those groups also were shut down.

"I'd move from one to the other and they'd just get eliminated," said Davies, 53, dressed in rumpled khakis in his office at Los Alamos with a view of the pinon-covered mesa.

For him, it was only a matter of time. "I had an opportunity to come here," he says of Los Alamos, "to keep doing that kind of work."

He and partner Roger Jones, a theoretical physicist who spent most of his 22-year career at Los Alamos in the nuclear-weapons program, teamed up on the Citicorp project, one of four Cooperative Research and Development Agreements, valued at $19 million, the bank has signed with federal laboratories. Taxpayers cover $9.6 million of that.

It was Jones who made the major career change, even though he never left Los Alamos. In 1989, he said, he recognized that the future of nuclear weapons work "was not very spectacular."

"The funding was drying up and there were political things going on that indicated it was time to start looking in other areas."

He took a sabbatical and retrained in a new field - computer software that mimics human thought, known as neural networks.

Jones chose a technology in great demand inside government and out. He and Davies began peddling the Los Alamos lab's skills to a variety of clients, including the U.S. Navy, the Internal Revenue Service, Rolls-Royce, General Motors and Citicorp.

"We started selling this technology wherever we could," he says. "We put miles on our frequent-flier cards."

The work Davies and Jones are doing now will ultimately be used for high- speed checking, instant fraud detection and targeted marketing for banking services.

As for creating jobs? That's unlikely.

It will enhance Citicorp's profits, however. And it will open new markets for credit-card services, says Davies.

"If it reduces their credit losses, it should create some more money for them," he says. "I think Citicorp has like 40 million card holders. Typically, they like to go to people who make money. If they can find a way to help people and make money, that's the way we'd like to see it go."

The government won't say how much they receive. The Inquirer study found that the six collected at least 115 research grants totaling $242 million over the last four years.

The six firms easily could have afforded to pay for the government help. In 1994, they reported profits of $27.7 billion on revenues of $588 billion. Their profits alone would have financed 38 percent of the total federal research budget of $73 billion.

General Motors, the world's largest automaker, received 51 awards worth more than $110 million, including several joint ventures with other companies. Among those were 14 awards totaling $32.8 million at Sandia National Laboratories in Albuquerque.

Laurence Phillips is working on one of those GM projects. He and two colleagues are performing what's known as "life-cycle analysis" on the materials used to build Cadillacs.

Their work is designed to assist Cadillac to make environmentally conscious choices in the materials it uses to produce cars.

Auto officials anticipate a day when U.S. automakers will be required to buy back old cars and recycle them, a practice gaining popularity in Europe.

"We're tracking it all the way back to dirt," says Phillips, a young scientist who spends his days in a windowless room crowded with personal computers, laptops, reference notebooks and environment posters on the wall.

In addition to the obvious components of a car - plastics, steel and glass - Phillips and his two colleagues are even calculating the energy needed to produce the uranium pellets that fuel the nuclear reactors that produce the energy that makes the car.

Although General Motors has its own engineering and environmental research staff, in 1992 it partnered with Sandia to do the work.

"This is an untested, controversial methodology that requires sophisticated computing capacity," says Phillips. "Nobody else is doing this. Cadillac is trying to get ahead of the curve."

It's trumpeted as a jobs-creation program, yet many of the major beneficiaries appear to be companies that have laid off the most workers. Often left out are small firms, which account for nearly two-thirds of all new jobs created in America.

Small companies get less than 4 percent of the $73 billion in federal research funds, said William K. Scheirer, an economist for the U.S. Small Business Administration.

While that percentage has increased slightly in recent years, "it's still too small a share. I'd like to see it much higher," he said.

This favoring of large corporations troubles Lee W. Rivers, a former executive at AlliedSignal Inc., who now directs the National Technology Transfer Center in Wheeling, W.Va. Rivers blames growing pressure on the labs
from Washington to deliver high-profile projects in the form of Cooperative Research and Development Agreements and other partnerships.

"I would hate to see a laboratory, because it's being driven by the metrics of its own agency - which says that CRADAs are what we want and bigger CRADAs are better than smaller CRADAs - be driven to really prefer to work with General Motors and not General Squeedunk," says Rivers.

"There is a conflict. It's been heightened in the last year and a half, dramatically so."

Consider Los Alamos, N.M., a county of 18,115 people perched on a series of mesas in the Jemez Mountains and home to one of the largest national labs.

Thanks to the presence of the lab, Los Alamos is among the wealthiest areas in all America, surpassing even its trendy neighbor, Santa Fe, where artists and New-Agers do lunch and stroll gallery-lined streets. People in Los Alamos commute to Santa Fe to find affordable housing.

"Nurses, engineers, police, firefighters and teachers here are all considered low-income," says James M. Flint, Los Alamos County administrator, whose daughter, a teacher, can't afford to live in the town.

In 1993, Los Alamos County ranked fourth nationally in household income - ahead of such better-known areas as Marin County, Calif., and Westchester County, N.Y. - and 16th in per-capita income.

By comparison, the state of New Mexico ranked 46th in per-capita income.

No question that Los Alamos is a prosperous enclave. According to Flint, it has more Ph.D.s per capita than any other place in America, a public library staff of 45, a teacher-student ratio of 1:12, a 140-person fire department and 600 county employees.

New housing costs upward of $500,000 and median household income, at $54,801, is more than double that of the rest of the state.

"Our future is clearly tied to the future of that lab," says Flint.

The federal government pays the county $2.8 million a year in lieu of real estate taxes, or about 15 percent of the county's budget. The school district receives $7.5 million from Washington, one-third of its budget in a system with 1,100 students.

"The Department of Energy wants us to become more self-sufficient. Now, we're like this," says Flint, tightly entwining the fingers of his hands.

The current 10-year agreement setting the level of assistance expires next year, and negotiations on a new agreement are underway.

One option under consideration is a land transfer. The Department of Energy owns 43 of the county's 110 square miles, and the county is eager to acquire property for development. An acre of industrial land ready to build on sells for $350,000, and there are only three vacant acres now, Flint says.

The 7,600 employees at Los Alamos, and the 8,500 at Sandia in Albuquerque, plus thousands of outside contractors who rely on the two national laboratories for their paychecks, represent a powerful political and economic coalition in New Mexico.

And the state's congressional delegation, led by its two U.S. senators, Pete V. Domenici, a Republican, and Jeff Bingaman, a Democrat, has been aggressive for years in promoting the labs - and federal funding for industry partnerships.

Bingaman helped candidate Bill Clinton draft his technology program. Domenici now chairs the Senate Budget Committee.

Although the New Mexico delegation is one of the most single-minded, constituencies for more federal support of technology programs exist elsewhere around the United States. Together, they make a potent force opposed to cutting such programs or closing some of the labs, as some have urged.

That may explain why technology subsidies don't flow equally out of Washington.

The Technology Reinvestment Program helps illustrate the inequities.

Since it was created in the early 1990s, the program has awarded $767 million to defense firms, high-tech companies and universities to develop
innovative technologies with both military and civilian uses.

Ten states have collected 72 percent of the awards. California, home to Silicon Valley and giant defense firms, got $246 million - 4 1/2 times that received by the next state, Massachusetts.

On a per-capita basis, Connecticut, the wealthiest state in the nation, collected the most: $10.66 per resident. That was more than 12 times the 86 cents that residents of Arkansas received.

Seven states got nothing.

"This is what's known as a catalytic converter," he explains, holding up what looks like an oversized cigarette. "Every car has one. Its job is to reduce bad emissions by burning off the pollution."

It takes about two to three minutes after a car is started for a catalytic converter to reach 300 degrees centigrade (572 degrees Fahrenheit), the necessary temperature to work effectively. Engineers refer to that point as ''light-off."

"Roughly 80 percent of the emissions that come out of your car are in that first two to three minutes when it's cold," Trkula says. "Our goal is to try to improve on that."

Trkula, a chemical engineer by training, works in the MST-7 section at Los Alamos National Laboratory, a group expert in the use of polymers and other chemical coatings. He arrived at the lab nearly 13 years ago and started his career working on defense projects.

These days, Trkula estimates that he spends one-third of his time working with corporations, including his present project to design a more efficient catalytic converter for the auto industry.

"Years ago, when I first started out with this group, it was only very occasionally we worked with industry. Now we do a lot of that. There's an alphabet soup of programs where that's encouraged," Trkula says.

The joint venture with the auto industry is with a consortium called USCAR, representing the Big Three automakers.

It is part of a much larger, more expensive collaboration by the Clinton administration to design cleaner, more fuel-efficient cars, known as the Partnership for a New Generation of Vehicles. The Department of Energy has budgeted $120 million this year for it.

Trkula is looking for a chemical coating that will heat up more quickly than the materials used presently. The contribution of the automakers includes applying and testing the new coatings, he said.

Why couldn't the carmakers, which spend about $10 billion a year on R&D, solve the problem on their own?

"I don't know. To me, this is normal technology," Trkula said. "They're not dumb. They just come from a different world, I guess."

Trkula said his work could prove very valuable to the auto industry. ''They're looking just in this program to recapture several percent of the (domestic) market."

Such a gain would translate to billions in additional revenues. According to Trkula, auto-industry officials also have calculated savings from fuel efficiency and the number of new jobs that will be created.

"It's all there on a piece of paper. We didn't make it up. They did. So there's some realism there," he said.

At first, Trkula said he would be happy to provide a copy of the estimates. Then, he talked to his partners in the auto industry.

They preferred not to release the information, he said.

"After all, it is theirs," he explained.

The government apparently thinks so, too.

The Department of Energy, which owns Los Alamos and 29 other labs, has taken the position that most of the details of its industry agreements are secret.

That includes descriptions of the work involved, the role of the industry participant and the estimated dollar value of the work done by industry.

Energy officials used to disclose how much of your tax dollars it gave away to its industry partners. But now, the department considers even that information secret.

"The Department of Energy believes specific dollar amounts is considered proprietary information," said spokeswoman Chris Kielich. "To include it would show the magnitude of the collaboration and therefore give competitors more of an indication of the work being undertaken."

Proprietary information? In a program paid for by taxpayers?

Sounds as if the Energy Department can't decide if it's investing in technologies or investing in companies.

It can't.

In fact, it's doing both. The department is subsidizing technologies. It's also underwriting the development costs of private firms and industries. Private companies that one day hope to make big profits from the products paid for by taxpayers.

Obviously, not every company in the United States gets a chance to partner with the government. That means the government, despite its claims to the contrary, is picking winners and losers.

And there is no middle ground.

Sorry, the government's high-tech promised land isn't for everyone.

It doesn't include rural America or states that don't have lots of high- tech firms.

Companies that rely heavily on poor, unskilled workers also need not apply. Those jobs already have been exported to Asia and other places where labor is cheap and regulations are few.

So how does one win?

It helps if you're already a big company. It's also a plus if Congress or the White House includes your industry on one of its lists of "critical technologies," singling it out for special funding. Since the mid-1980s, nearly a dozen such lists have been drafted.

What are the criteria for getting on such a list? That's a little hard to say. The standards have never been spelled out clearly.

Curiously, some of the technologies that do end up on these lists - semiconductors, for example - already are highly developed and profitable.

And quite often, the advisory groups choosing "critical technologies" are crowded with executives from those very industries. Or government may ask an industry to propose a joint venture on its own.

Thus, instead of government bureaucrats picking technologies for subsidies, industry experts can make the choice.

Then the bureaucrats can approve them.

And write a check with taxpayer dollars - although the amount of the check is secret.

In the rapidly evolving American economy, this is what passes for a public- private partnership.