service in any agency. So I would like to reach into the past to explain how this came about.

The Bureau was established in 1884. That was an election year. The labor movement was beginning to express itself at that time. The A.F. of L. had actually been created, although it did not get its present name until 1886. The Knights of Labor was a politicaloriented labor group that was active. They were pressing for a Department of Labor, for an 8-hour law, for child labor regulation, et cetera. And Congress was under pressure with an election due in November. So Congress passed a law providing that there would be a Bureau of Labor created in the Department of the Interior and a Commissioner appointed by the President.

Well, the President, who was President Arthur-he was President Garfield's Vice President-was in office at the time. He had a half a dozen candidates for the job. He selected one of them and sent his name down to the Senate, and the man was confirmed.

But then the Republican convention was held and President Arthur was not given the nomination; it went to Senator Blaine. The President was so disappointed that he would not give the Commissioner his commission. And so this man was appointed, but never took office.

In the meantime, November rolled around and Cleveland was elected. So there was a change of administration.

At that time, you may recall, we had the lameduck Congress which took office in December and served until the 4th of March. So when the Congress met they began asking questions about that Commissioner they had confirmed and this Bureau of Labor that they had expected to be created.

In that situation President Arthur decided to be strictly nonpolitical, so he reached up into the State of Massachusetts and chose the director of the Bureau of Labor Statistics of Massachusetts, who had been serving in that capacity for about 12 years. In fact he had already become world famous for his researches in cost of living, wages, et

cetera.

So he was appointed, and became the first Commissioner of Labor. His name was Carroll D. Wright, who had been director of the Bureau of Labor Statistics of Massachusetts.

Then there was another incident that took place. Since the new President took office in March, and they were appointing this man in January, President Arthur had to consider how long he would remain, would he take office for 2 months and be retired in March? Apparently an agreement was reached. History does not explain how it happened, but apparently an agreement was reached between the incoming President and the outgoing President, so that President Arthur appointed him to the job. So he was confirmed in January, and President Cleveland left him alone for the 4-year term. The term is 4 years from the date of appointment, whenever it might be, and each term starts fresh. So he was appointed in that way.

His term ran out in January 1889, when Cleveland had been defeated and Harrison was elected President; the Republicans were in again. So the same arrangement was made. He was appointed in January, and Harrison permitted him to stay. And then Harrison was followed by Cleveland, and Cleveland was followed by McKinley and McKinley was followed by Theodore Roosevelt. And Carroll D.

Wright continued on to 1905; he completed five terms of 4 years' each, but with a change of administration almost every term.

So that established the principle of nonpartisanship of the Commissioner of Labor Statistics. And that tradition has been carried on down to date. Several Commissioners served two or three terms, frequently with changes in administration. I was appointed, as you indicated, in 1946. I was appointed twice by President Truman, twice by President Eisenhower, and once by President Kennedy.

And so the tradition of nonpartisanship has been maintained. Now, with respect to the press conferences, as far back as Carroll D. Wright's day the Commissioner of Labor was a very important figure in the activities of the Nation. I might have emphasized in my history that, after being in the Department of the Interior for 3 years, in 1888 the Commissioner had attained such influence and stature that the Congress created a "Department" of Labor. I have to put that in quotes, because they did not create a department with a secretary; they created a department with a commissioner, who was to report to the President of the United States. So from 1888 to 1903 the Commissioner of Labor was a Presidential appointment, and in a sense represented the equivalent of a Cabinet Secretary, except that he was not of secretarial rank.

During all this time he treated the Bureau of Labor, as it was then called, as a research agency. He took no part in policymaking. He never made any recommendations for legislation. But he made studies on child labor, on women's labor, on wages and other kinds of labor problems, which in turn of course produced legislation of one kind or another in the Congress.

In 1903 President Theodore Roosevelt used Carroll D. Wright very importantly in the famous coal strike which took place at that time and which created quite a convulsion in industrial relations in this country.

Coming down to more recent times, when Frances Perkins was Secretary of Labor and Commissioner Isadore Lubin was in office, they would have occasional conferences, with Miss Perkins talking on some general subjects, and Commissioner Lubin being there to handle any economic questions. These were not periodic; they just occurred from time to time, whenever something of interest came up. Then when World War II came along, the Bureau of Labor Statistics served, as it always did in wartime, as the service agency to the price control agencies, and to the wage control boards, and therefore said very little other than issuing its periodic announcements. These control agencies would be explaining to the public what the situation

was.

In my own experience, beginning in 1946, when I took office, I had the problem-I do not know whether to call it the good fortune or the bad fortune-that price controls were taken off by President Truman in the summer of 1946. So that the long controversy about. whether the price index properly measured the cost of living was soon put to sleep, because the index rose 30 percent in 2 years. It was not my fault that it rose so fast, but it became the most crucial public policy statistic of that period.

It was right in the midst of that period that I had one of my first serious encounters with the problem of how to deal with the press.

The Congress of the United States-I should say, the House Appropriations Committee-in the controversy about the index measuring the cost of living during World War II, had instructed the Bureau to develop a four-person family budget. That would show what it actually cost to live in dollars.

A key member of the House Appropriations Committee explained that he did not like the index, that he did not understand indexes, that he would just like to know in dollars how much it cost to live. So the committee instructed the Bureau to prepare such a family budget. And that study was underway when I became Commissioner.

This project came to fruition in the summer of 1947. We had our four-person family budget worked out. And we had some figures showing what it would cost.

In the situation at that time, with the Consumer Price Index rising as rapidly as it was, it was not the time for the Commissioner of Labor Statistics to open up a press conference and announce he had some new data on what it costs to live. And so I was in a dilemma. And it was from that dilemma that the Joint Economic Committee rescued me. Senator Flanders was the chairman of the committee at that time, and Senator O'Mahoney was the Democratic minority leader.

The two of them got together and invited me to come down to explain to the Joint Economic Committee what was in this four-person family budget. So under that compulsion I came down and explained. And the press, of course, took it up. So it came out without any difficulty for us in the Bureau of Labor Statistics.

This was one case in which your committee was a conduit to the public on a subject that I would have been embarrassed to publicize

on my own.

Coming down still more recently, in May 1948, the Auto Workers Union signed the escalator contract with General Motors. That was an expression of confidence in the price index of the Bureau of Labor Statistics. I have to emphasize how spectacular a change of sentiment. that was, when there had been so much controversy about the index during the war, whether it properly measured the cost of living. And yet in 1948 the Auto Workers Union signed a contract with General Motors (and later with the other auto companies) providing that there would be an annual increase in wages of about 3 perceut a year, based on productivity, based on real income in the Nation, plus a cost-ofliving escalation every 3 months, if the price index rose by a certain amount. That first contract was for 2 years. And that explains why there was so much interest then on the part of the public in getting periodic and frequent conferences on the Consumer Price Index. Workers in the auto industry wanted to know each quarter what the index showed.

Incidentally, it was very important to keep that figure confidential up to the deadline. We had been rather loose in the timing with which we had been issuing the figures prior to that. There was a lack of precision in what hour of the day or what day of the week. We then fixed, as far as the Consumer Price Index was concerned, a very firm time, 11 o'clock, when that would come out, so that everybody in the country, labor and management and all the workers, would know about it at the same time.

I hardly need to emphasize the point that there was every reason why we should be explaining every month what that figure was, and

how it happened to be at that level, and of course, how many workers were affected by it. If you should look at our old releases of that date you would see that we always brought out the figures on the number of workers affected.

That contract ran for 2 years. I must remind you that the auto workers had lost several cents per hour because the index had declined some. And predictions were freely made in the spring of 1950 that that contract would never be renewed, it would be a one-shot proposition. And yet, surprisingly enough, the auto workers and General Motors and the auto industry renewed that contract in May 1950. I want to emphasize to you that that was before Korea. There had been no outbreak of war. It was renewed for a period of 5 years, 1950 to 1955. That is evidence, I think, as to what extent the workers-and after all, Walter Reuther represented the workers-to what extent. the workers thought that this was a good contract, that wage increases based on productivity plus an escalation for the consumer price index (if the rest of the country did not keep the cost of living down) constituted a useful contract.

At any rate that was the situation as far as the consumer price index was concerned, until it was dropped during the Korean war. First of all, the Government put on controls in the spring of 1951. The index then remained very stable. Price and wage controls were in effect, and nothing much was happening to the index.

Then the press conferences were renewed again in 1954, after the war was over. And these have continued-I would not say every month, but more or less continuously-ever since.

Now, to turn briefly to the unemployment figures. It was in Secretary Tobin's administration-you will recall that Secretary Tobin was appointed Secretary of Labor in 1948. Up to that time we had had full employment, a shortage of labor, during the postwar years. No great unemployment arose during the conversion to peace. But in the summer of 1948 we began to get a mild business downturn. And so when Secretary Tobin became firmly entrenched as the continuing Secretary of Labor for the Truman administration, the second Truman administration, the economy went into the economic downturn of 1949, not a tremendous downturn, and not a very long one, but sufficient for unemployment to rise. The Bureau of Labor Statistics was not responsible for the unemployment figures at that time. Those were in the Bureau of the Census. But the Bureau of Labor Statistics had the figures on employment, hours and earnings reported by employers in industries throughout the country.

It was on those figures of employment, hours and earnings that Secretary Tobin thought we ought to be presenting more frequent explanations to the public as to what was happening in this field. So he set up a situation room in the Department of Labor with maps and charts, so that he could bring visitors in to take a look at the unfolding economic situation.

He encouraged us to start talking to the press about our figures of employment, hours and earnings. And we did so. I cannot recall that that we did it regularly, but we did it from time to time.

Then the Korean war began, and once more we went off the air, so to speak. Control agencies were then much more important. We

were servicing them, and they were the ones who interpreted to the country what was happening at that time.

After the war was over, the Korean war, we sank into the recession of 1954, again a very brief and shallow recession. But at that time there was a controversy about the unemployment figures. And this time it was the Census Bureau that was in difficulty. That was because Census had a sample, a small sample at that time, scarcely much more than one-third of its size today, and it was decided that it should be enlarged. That revision was taking place in 1953-54 just as the business recession was developing. The shift from the old discarded sample to the new improved one was taking place over the winter.

The introduction of the new sample by the Bureau of Census took place in February 1954. February is the worst month of the year for unemployment, except when June produces a temporarily high figure for unemployment because of young people coming out of school. The difficulty was that the difference between the old series and the new series on unemployment was quite marked, and caused some controversy. And as a result of that, the administration made an arrangement whereby the Departments of Commerce and Labor worked together, Census and BLS worked out jointly a press release to the public, that would explain both the employment and unemployment in the large, and the employment hours and earnings of the Bureau of Labor statistics in detail. And that arrangement existed for 5 years.

During that period of time there were no press conferences that I recall, either by the Census or by ourselves, because the information was put out in a joint release by the Secretaries of Commerce and Labor. That system continued until 1959, when, for a variety of reasons that I do not need to go into, the Secretary of Labor made an arrangement with the Secretary of Commerce whereby the Bureau of Labor Statistics would take over the whole job of analyzing and interpreting the employment, unemployment, hours, and earnings figures, while the Census would continue to collect the data in accordance with their normal procedures.

Then the work of analysis and release of the data came over to the Bureau of Labor Statistics. And then the press conferences were renewed.

Now, at that renewal

Chairman PROXMIRE. What was that date?

Mr. CLAGUE. That was sometime in 1959, I cannot remember the exact date.

Chairman PROXMIRE. Thank you.

Mr. CLAGUE. But it was some month in 1959. I cannot remember exactly when it was.

That transfer produced another result. We resumed the press conferences in the Department of Labor at that time. But in the meantime there had been changes in personnel. The man who used to be my right-hand power in employment, hours and earnings statistics had become Deputy Assistant Secretary of Labor. And therefore he was then in the Labor Department, no longer in the Bureau. He was assigned the task of conducting the press conferences being