plants. In order to maintain IBM's full-employment practice, we shifted more than 12,000 people into new jobs outside the manufacturing area.

More than 5,000 of these employees received significant retraining to enable them to take up new careers in marketing, servicing, programing, and administrative jobs.

There are two reasons, I think, why this program has been successful. First, we are able to project our personnel needs with a high degree of accuracy; and second, we knew what skills would be required to fill those jobs. Accordingly, the people who entered IBM's adjustment assistance program knew that the program would result in a job when the retraining was completed.

While the problem is obviously many times more complex on the national level than it is for IBM, I believe that, with similar attention to the identification of skill requirements and the design of retraining programs to meet those requirements, adjustment assistance can be a very effective tool for redeploying the Nation's manpower

resources.

I would like to congratulate Governor Freeman on his comments on this subject. I would like to say that the IBM Co. stands ready to serve on an advisory council of Government and industry, on adjustment assistance, and we would love to contribute the full help of one or more of our people who have direct experience with design and implementation of such a program.

I will be happy to answer any questions during the question period. Thank you.

Mr. BURKE [presiding]. Without objection, your request for including in the record your testimony on the Senate side will be complied with.

Mr. JONES. Thank you, Congressman Burke.

Mr. BURKE. Also, on the statement made by Mr. Kendall, I notice appended thereto are some quite extensive comments on the BurkeHartke bill. Without objection, that will be included in the record following your oral statement. We believe in giving the other side a chance.

[The Senate testimony referred to follows:]

STATEMENT OF GILBERT E. JONES, CHAIRMAN OF IBM WORLD TRADE CORP., PRESENTED AT HEARINGS BEFORE THE SUBCOMMITTEE ON INTERNATIONAL TRADE OF THE COMMITTEE ON FINANCE, U.S. SENATE, THURSDAY, MARCH 1, 1973

My name is Gilbert E. Jones. I am Chairman of IBM World Trade Corporation, IBM's subsidiary for business outside the United States.

I appreciate the opportunity to testify at these important hearings. There is no mistaking the fact that we have come to a crossroads in the evolution of American foreign economic policy. Decisions that will be made in the next few months by the Congress will affect future domestic and international economic patterns in fundamental ways.

For several years now a public debate has been underway between those who regard multinational firms as a source of strength of the United States and those who allege that they are not.

Despite governmental and private studies which show that the multinationals create American jobs, contribute strongly to the balance of payments, and are instrumental in developing crucial new technologies, doubts persist among the critics, and they must be confronted by those of us who do business on a multinational basis.

I intend to do so here today, not by repeating the standard arguments and data with which you are all familiar, but by concentrating on the experience of

my own company. IBM is a high technology company, and high technology is acknowledged to be one of the greatest strengths of the U.S. economy. So I will be talking to you primarily about IBM's role as an innovator, manufacturer and marketer of high technology goods in a worldwide environment.

Before getting into the heart of my discussion, I thought it might be helpful for the subcommittee to have a clearer understanding of what we mean when we use the phrase "high technology" companies. An obvious example, of course, can be found in the computer industry.

I'd like to show you how extremely rapid technological change has resulted in vastly increased computing power for our customers at continually decreasing prices.

In 1952, IBM announced its first large scale electronic data processing system the 701. This machine used vacuum tubes like this and could do almost 2200 multiplications per second at a cost of $1.26 for 100,000 multiplications.

Six years later, in 1958, we introduced the 7090, a machine whose basic circuitry was entirely new and made up of these silicon transistors. The 7090 did about 38,000 multiplications in a second and the cost of 100,000 multiplications dropped to 26 cents.

In 1964, we introduced a new technology, SLT putting a hybrid integrated circuit on a ceramic substrate like this. That was the year we announced the System/360 Model 50. With that machine, the speed of computation was about the same as the 7090, but the cost of 100,000 multiplications dropped by more than half to 12 cents.

Less than three years ago, we announced our System/370 series which utilizes these fully integrated monolithic circuits. With this technology, we have up to 10 logic circuits on a chip. And we even have the capability now to put up to 400 on a chip. The System/370 Model 155 can do about 105,000 multiplications per second and the cost of 100,000 multiplications has dropped to a nickel. One of our latest computers the System/370 168 Model which was announced last year can do over two million multiplications per second at a cost of one cent for 100,000 multiplications.

While I have simplified these comparisons for the sake of brevity, they illustrate my main point: the rapidly decreasing cost of calculations for the computer user coupled with increased computing speed.

The same kind of change has happened to computer memories. Our 701 in 1952 used a cathode ray tube memory which looked like a TV tube and which is too bulky for me to bring here today. In 1954, we introduced ferrite core memories which were the principal memory technology until very recently.

Here you see how the size of ferrite cores has decreased. These cores had to be wired into planes, like this one.

Today, computer memories, like circuitry, utilize silicon chip technology, and memory costs have also decreased dramatically. The same amount of money today will buy more than six times as much memory as it would have bought in 1967.

I can't tell you today exactly what the next few years will bring, but I can say that technological change will become even more dramatic. Since 1958, we have gone through four significant technological changes in electronics, three of these in the past decade, two in the last five years. Progress is constantly speeding up.

The new technologies are also much more demanding of our manufacturing people. A typical manufacturing process for producing these chips has over 180 important steps. If each step is 99 percent perfect, the final yield at the end of the proces will be only about 13 percent. As you can imagine, considerable engineering and manufacturing effort must be expended to create relatively lossfree manufacturing.

This creates a need for enginers and technicians with skills of the highest order. And since technology changes so rapidly, it also means these people must be constantly relearning new skills and techniques.

Now from this brief illustration of a high technology industry in action, several broad characteristics can be noted.

One is the enormous rate of change. Generations of products are measured in a few brief years.

And not only do product lines change, but also competitive conditions. A company or, indeed, a nation which fails to keep apace of technological improvements and changes in marketing techniques may be among the leaders one year and a trailer only a short while later.

Another important characteristic is the declining labor input in the product line. This doesn't mean that improved technology results in fewer jobs. What it does result in is more highly skilled jobs, as I shall illustrate in detail toward the conclusion of my remarks.

It also means that labor costs on the production line are declining sharply as a percentage of our total costs. And it also means most important in terms of the work of this Committee that hourly labor costs on the production line have less and less to do with our ability to compete either with other firms in this country or with foreign computer companies.

Of course, no manufacturer wants to see his labor costs get out of line with his competitors'. But in a high technology field, a differential in hourly wages is far less important than a company's ability to maintain a technological lead. It would be a mistake for the Congress to focus too narrowly on the fact that wages are higher in the United States than in foreign countries, and to assume that the best way to protect American industry and American jobs is to try to compensate for the difference through tariffs or other protective barriers. For most industries, the question of whether America can compete effectively in international markets ultimately is more dependent on our ability to maintain a technological edge, on our willingness to innovate in our products and our factories, and on our marketing skill and ingenuity than it is on hourly wage differentials, at least within certain limits.

Fortunately, innovation and risktaking are areas in which U.S. industry has traditionally excelled. While the computer industry may represent the current extreme in change and innovation, the simple fact is that no industry is-or should be static. The facts of life in all industry require continual innovation to improve products and to lower costs.

The key for successful governmental policy is the creation of a climate which will encourage innovation and improvement in American industry, not sheltering it from competitive forces.

Specifically, in the field of legislation dealing with international commerce, Congress should avoid thinking in terms of days gone by, when handcrafts and high labor content were the general rule in American factories. Rather, we should think in terms of the 70's and the 80's, when our ability to compete will depend primarily on our technological innovation, our creative abilities, and our salesmanship.

Now in the next few minutes, I'd like to illustrate the basic points which I have been making by telling you a little about IBM's operations as a multinational corporation-its economic impact abroad and in the United States, and some of the problems that we face at the present time and in the future.

In 1972, IBM showed gross income of $9.5 billion and net earnings of $1.3 billion, with a total of 262,000 people employed here and abroad. IBM World Trade Corporation, which handles IBM business outside the United States, reported gross income of $4.2 billion, net earnings of $687 million, and total employment of 115,000.

Outside the U.S., IBM does business in 126 countries. It has 9 research and development laboratories in eight foreign countries and 19 manufacturing plants in 13 countries overseas. In addition, IBM provides education in new technology and management techniques at 78 overseas locations throughout the world.

IBM has been doing business on an international basis since its earliest days. Operations in Germany date back to 1910, in France to 1914, in Canada and Brazil to 1917 and in Japan to 1925.

IBM has pursued a general policy of manufacturing abroad the finished products sold abroad because that is, in virtually every instance, the only way to maintain its foreign markets. It is most important to understand that this policy has helped, not harmed, IBM's employment in the United States.

The reason for this is that operations abroad create demand for IBM products that cannot be satisfied by local manufacturing alone. This demand is met by a growing export business channeled through IBM's foreign subsidiaries. IBM's U.S. production destined for export has risen from $56 million in 1960 to $485 million in 1972. Imports have also increased during this same period. But, overall, IBM's net exports have increased from $52 million in 1960 to $305 million in 1972, an increase of nearly 500 percent.

We have stated on a number of previous occasions that one in eight of our manufacturing jobs in the United States is dependent on foreign exports. In fact, this has been a very conservative estimate deliberately so, because this is a complex computation to make. However, translating the dollar value of our U.S. exports in 1972, we find that one out of every five jobs in IBM's U.S. plants

was accounted for by business between the parent company and IBM World Trade.

I hope it is no longer in doubt that multinational companies in general are large net exporters. The U.S. Tariff Commission study recently published by this committee found that multinational corporations in 1970 accounted for about 62 percent of all U.S. manufactured exports as compared with 34 percent of manufactured imports. The commission's analysis also indicates that "the U.S. industries most active in production abroad also are the heaviest contributors to U.S. exports"-a conclusion corroborated by other studies, such as those made by the U.S. Department of Commerce. (I think it is also worth pointing out parenthetically that, when transportation equipment is excluded-to screen out the effects of the U.S.-Canada auto trade agreement-the Commission says: "A statistically significant association between foreign direct investment activity and aggregate imports disappears entirely.")

The Tariff Commission study is less definitive with regard to trade-related employment since it points out that the conclusion depends on the assumptions used on what would have happened to the overseas markets of U.S. firms in the absence of their foreign investment. The Commission indicates that under the assumption that U.S. firms, without investing abroad, could have maintained the same share of the market that they enjoyed in 1960-61, the foreign investment of U.S. multinational companies has resulted in a net gain of 500,000 jobs in this country.

However, the Commission concedes that this estimate of jobs gains "is biased in the direction of excessive pessimism because it totally rejects-by assumption the MNCs argument that at least a portion of the MNCs foreign direct investment has to go abroad to prevent foreigners from getting there first."

In IBM's case, there is absolutely no way in which we could have held onto our overseas markets, and increased U.S. jobs these last dozen years, if we had refused to invest in foreign facilities.

In the first place, many governments feel the need to keep local value-added in some kind of equilibrium with sales volume. If major countries had to import all IBM products now sold there, the drain on their payments balances in most instances would be prohibitive.

But even if this consideration did not exist, it is important to realize that the vast majority of our overseas jobs are not in manufacturing. Out of the total number of 115,000 IBM employees overseas today, only 27.600 are employed in manufacturing. This means that even if IBM were to manufacture all its products in the U.S., it still would have to maintain abroad more than three-quarters of the present number of overseas employees. At least 80,000 people would be needed as local sales forces, systems engineers, customer engineers, administrative and other support personnel to market and service IBM equipment-assuming that the current level of IBM's business outside the U.S. could be maintained.

As a practical matter, however, this level could not be maintained. The choice for us is not between exporting or manufacturing abroad, but between manufacturing abroad or losing large portions of the world market.

I cannot prove it to you scientifically, but my associates and I are convinced that if IBM tried to serve the world market entirely from the United States, our business abroad would shrink to a small fraction of its present size.

Other countries' restrictions on imports are not the essential reason why this shrinkage would occur. The crux of the matter is that an attempt to make computer technology an American preserve would not succeed. By refusing to manufacture in Western Europe, for example, we would be forcing foreign governments to subsidize and foster the development of their domestic computer manufacturers to an even greater extent than they now are doing, because those governments would feel widely dependent on foreign sources of supply of products and of development of technology.

Imagine the economic and psychological effect on Western Europe of total dependence on imports of high technology goods. It would never occur. By manufacturing abroad, we hold a market we would otherwise lose.

In addition to export revenues, the fees, royalties and earnings of IBM's wholly-owned subsidiaries and branches overseas represent a steadily growing flow of money to the U.S. In the ten-year period from 1963 to 1972, IBM's net contributions to the U.S. balance of payments was $4.44 billion.

IBM's contribution to U.S. and foreign tax revenues over the last five years amounted to $5.2 billion; over the last ten years to $7.7 billion. This represented taxation at the rate of approximately 50 percent of earnings, and of the total

taxes paid over the past ten years about 60 percent went to the U.S. government, and the remaining 40 percent to other governments.

In my judgment, these economic contributions were made possible by the main strength of my company-the speed with which we have been able to take technology out of the laboratory and turn it into marketable products. Neither IBM nor the United States has a monopoly on scientific knowledge in the computer field. We are convinced that it would be a tragic mistake for America to try to board its technological know-how on the false premise that major advances necessarily evolve from laboratories in this country. Rather, we know that our major strength lies in adapting technology, wherever it originates, quickly and effectively into finished products.

Accordingly, one of the reasons for our success is our closely integrated multinational research and development effort. We have access to overseas technology and talent through our laboratories outside the United States and we receive a steady flow of new ideas from all over the world.

An important part of the development work on the IBM computer systems IBM System/360 and IBM System/370 was done in the company's overseas development laboratories.

Last August, IBM announced two new computer systems-the IBM System/ 370 Model 158 and Model 168. Development work on these new systems was the joint effort of several laboratories in the United States and abroad with close and instantaneous links among them.

IBM customers in the United States as well as abroad are the final beneficiaries of these joint efforts.

What is true for the computer industry is also true for most American high technology industries. All have to search for the best talent wherever it can be found to maintain their leadership. A great deal of know-how and inventiveness of foreigners actually goes into American products.

I recall from my own experience in the Navy that the first work on radar and anti-submarine warfare was done by the British. So was the earliest development of jet aircraft. The Wankel engine is another foreign import. The electron microscope and the video tape recorder are examples of important inventions resulting from international scientific cooperation. Magnetic ferrites, so crucial to my own industry, were the result of half a century of scientific effort in the United States, Europe and Asia. Nor should we forget the crucial contributions of foreign scientists to the harnessing of nuclear energy, first for military and then for peaceful purposes, or to space exploration.

The economic progress made here as well as abroad during the last 20 to 25 years would not have been possible without the relatively free transfer of technologies among Western nations.

IBM, for example, has cross-licensing agreements with dozens of European companies. It has similar agreements with some 15 Japanese companies. IBM's magnetic tape manufacturing facility in Boulder, Colorado, was set up under a cross-licensing agreement with the Sony Corporation of Japan. It uses Sony patents and draws on the technical know-how of the Japanese Company.

Aided by overseas research and development, American high technology companies continue to lead the world. While every American legislator is familiar with the fear in this country of competition from certain imports, you may not be aware of the fact that our principal trading partners feel under immense pressure from advanced technology companies headquartered in the U.S.

Governments of major industrial countries are actively intervening to support their national competitors of U.S. computer manufacturers. That competition is growing both in strength and sophistication.

As European computer manufacturers, acting individually and jointly, develop their technological capabilities, they are getting substantial support from their governments.

From the beginning, national governments played a significant role in several major European computer companies. Some were created under government auspices, like France's CII and Great Britain's ICL. In both cases, the government participates in the company's equity.

In Germany, the government has allotted $832 million in subsidies during the 1971-1975 period for the domestic data processing effort.

In France, some $264 million will be spent by the government in subsidies to the French computer industry between 1971 and 1975. In addition, the computer rental business of CII is being financed by a group of governmentcontrolled banks and the nation's Social Security fund.