Mr. LOOMIS. We had a good deal of discussion over that question. without coming to a conclusion.

Mr. RANKIN. All milk is now marketed by the pound and tested by the pound.

Mr. LOOMIS. Well, conversion is very easy.

Mr. RANKIN. Whenever you get into the real dairy development in the country you never hear any more about milk by the gallon. That is my experience. They talk about pounds.

Mr. MOORMAN. But the number of people interested in that particular phase of it are few, compared to the people at large that milk cows by the gallon.

Mr. RANKIN. But there is this difference, though: You will find that where they give it to you by the pound they give it to you accurately.

Mr. MOORMAN. Yes, they do.

Mr. RANKIN. Because, as a rule, they sell by the pound and the record is kept by the pound, but it is a very uncertain guess as to how many gallons a man's cow gives when he is either trying to sell it or the other fellow is trying to buy it.

Mr. LOOMIS. Mr. Chairman, I think it will be a matter of good judgment if I ask permission to delete from the record the statement which I made relative to the possible small errors in the last agricultural census, when I look over my transcript.

The CHAIRMAN. Yes, you may do that.

Mr. LOOMIS. I wish to thank the committee for your courtesy The CHAIRMAN. Thank you, Mr. Loomis. We have been glad to hear you.

Are there any other gentlemen here now who wish to be heard on the bill?

Mr. AUSTIN. I would like to say, about the inquiries that Mr. Loomis asked for, the number of cows milked, scheduled for the 1925 census, he made his examination of the 1920 schedule, but they never appeared in that schedule, but practically all of them were carried in the 1924 schedule.

Mr. JOHNSON. I would like to ask Mr. Loomis a question.

Mr. LOOMIS. I know what Mr. Austin says is correct. I think I must have inadvertently said something there that was not in my mind at all, because that was in the 1924 schedule and it was in the 1925 schedule.

Mr. JOHNSON. I would like to ask one question. I am sorry I was not able to be present when the question of distribution was before the committee, but could I infer from the statements made by you that the organization represented by you would be opposed to any detail in regard to a distribution census?

Mr. LOOMIS. No, Mr. Johnson, we did not have it up before us at all. We don't know what it is. We don't understand it, and when the gentleman here said that there would be a provision for a census of distribution under consideration later on, I just wanted to

The CHAIRMAN (interposing). We had the representative of the Department of Commerce here and he spoke on that.

Mr. JOHNSON. I wanted to be here and hear that, but I was sick in bed on that day.

We

Mr. LOOMIS. All I want to say is that we want to study that. want to look at the record to see what there is to it. I think that the several organizations making up the American Dairy Federation would be very much in favor of a census of distribution but will want to look it over and see how far it goes and how extensive it is going to be and how it would affect the taxpayer's rights.

Mr. JOHNSON. I have not heard the witnesses at all, but I gather that the general opinion is that statistics with regard to distribution are the statistics that are really lacking in a balanced bird's eye view of the situation in the country.

The CHAIRMAN. The primary idea, as I understood it from the gentlemen of the Department of Commerce, in regard to the census of distribution was, in brief, to take a census of sales, practically, to the ultimate consumer-I will give you one of those books, Mr. Johnson. They did take a census of distribution in Baltimore-in seven cities, I think-and the idea is to get the figures of distribution to the ultimate consumer.

Mr. JOHNSON. I understand that phase of it-17 cities, I believe. The CHAIRMAN. If you will preserve that book, Mr. Johnson, I think you will find a lot of information in it, and I do not know what city you have there but they are all similar in character.

Doctor HILL. I would like to say, Mr. Chairman, in order that there may not be any misapprehension in the minds of the committee, that the census distribution perhaps would be more accurately described as a census of wholesale and retail trade. We do not undertake to trace the movement of goods through the various channels through which they reach the ultimate consumer, but it covers trade, it covers wholesale and retail trade in its present scope. The CHAIRMAN. Now, may I ask you, this Doctor Hill: A department store, for instance, and perhaps the chain drug stores or the chain stores, they generally include groceries and goods such as a department store has, and such as the chain drug stores have-how could you differentiate in those, or would you attempt to, and just take up general sales, say drug sales of that drug store?

Doctor HILL. I think Mr. Gosnell had better make a statement on that.

The CHAIRMAN. I hope the committee will get this clear in their minds just what this distribution means. The question has been raised very appropriately by Mr. Johnson, and I think all of us would like to hear something on it.

STATEMENT OF FRED A. GOSNELL BUREAU OF THE CENSUS, DEPARTMENT OF COMMERCE

Mr. GOSNELL. Mr. Chairman, in making an enumeration of distribution we provide on our schedule form for approximately 75 different commodities. When the department store or chain store is asked to fill out the schedule they are asked to state the amount of each commodity listed on the schedule they have sold during the year.

The CHAIRMAN. You mean the amount in bulk or in dollars?

Mr. GOSNELL. In dollars. I should say that in the department stores there were at least 95 per cent of them who furnished from 15 to 40 different items, commodities that they sell. Many of them elaborated upon the schedule and added important items that are not called for, that are covered generally in the report.

The CHAIRMAN. May I ask you this? Was not this census of distribution in these several cities that were taken, taken by the chambers of commerce or boards of trade of those respective cities at their suggestion in cooperation with the census bureau?

Mr. GOSNELL. No.

The CHAIRMAN. Or didn't the suggestion come from the business men?

Mr. GOSNELL. The suggestion came, Mr. Chairman, from the committee on collection of business figures, which is appointed by the United States Chamber of Commerce. The officials of the Bureau of the Census and the Department of Commerce generally, in cooperation with that committee, selected certain cities throughout the United States in which we would make a test of distribution. Those cities are located in different sections of the country. The special census committee which is appointed by the United States Chamber of Commerce and the Bureau of the Census drew up the schedules; we appointed the official enumerators just as we do in agriculture and population, paid them from Federal funds and sent a supervisor to each city to supervise the work.

The CHAIRMAN. But this did come from the suggestion of the business men of the country?

Mr. GOSNELL. Yes.

The CHAIRMAN. Acting through the National Chamber of Commerce?

Mr. GOSNELL. Yes.

The CHAIRMAN. They asked for this in probably the same way or for the same reasons that the great agricultural interests asked for their figures?

Mr. GOSNELL. Yes, sir.

Mr. RANKIN. I want to know why, if you are going to take this census of distribution, why you don't get the census of distribution in bulk or quantities as well as price? Take for instance a commodity manufactured in Pittsburgh. If you undertake to trace that commodity, if you merely get the price at which it was sold, that is, the amount in dollars and cents in the State of Pennsylvania, in the State of Mississippi or the State of Washington, there is no way in the world to compare those figures to show the amount of material that was distributed. If we are going to take a census of distribution, it seems to me that we ought to also include in that a census of the amount of the commodity as it is distributed, as well as the amount of money that it brings.

Mr. GOSNELL. The objection there, Mr. Rankin, is that the department stores and other stores do not keep records of the quantity of sales; but they do keep records of the amount of sales of the different commodities.

Mr. RANKIN. Well, now, it would not be very much trouble to get that. Every store and every business house knows the price at which it sells a commodity. Suppose you were going to get the

census of the distribution of gasoline-and I think that is very important right now, since it is selling all the way from 9 to 29 cents, and, to be perfectly frank with you, I paid from 18 to 28 cents the same day on the same trip, and where I paid 18 cents they were paying a higher tax to the State per gallon on gasoline than other States; I would say at least 18 to 25 cents, 7 cents spread. Now, if you are going to take a census of distribution of gasoline, for instance, it seems to me that it is useless to take it unless in getting the amount of money that it brings at the ultimate point of distribution you also get the price per gallon or per hundred gallons at which it is sold, and I do not see why merely getting the census of the amount of money that is taken in for a commodity will be of very much benefit to the people of the United States.

Mr. GOSNELL. There are a great many commodities, Mr. Rankin, for which you could get that information, but there a great many of them for which you could not, because the classifications necessarily have to be broadened out.

You take, as an illustration, toys and games. We would not be able to get the quantity figures for toys and games. And there are others the same way. Automobiles and gasoline, there would be practically no difficulty in getting that.

Mr. RANKIN. Now, I think you are wrong about that. Every business house that is really doing business in a businesslike way keeps copies of its invoices, and they can tell you just exactly how much of this stuff they bought and how much they disposed of in the year. If their books show how much money an article brought, it is no trouble for you to get that information and include it in your census reports of distribution.

The CHAIRMAN. Mr. Rankin, why does every large department house take an inventory every year or periodically, if their sales show just what they have been doing? We all know that every well regulated concern takes an inventory at stated times. Even great manufacturing plants take their inventories.

Mr. RANKIN. If you go to a big house that hasn't enough business management about it to take an inventory once a year, you are going to get very little information out of them that will be worth anything on a census schedule or a census report.

Mr. GOSNELL. That would be a tremendous task, Mr. Rankin, for the stores to compile from the invoices the number of each article that they sell, particularly where there are approximately fifteen or twenty thousand different merchandise classifications. That would mean that we would have to enumerate on this schedule the sales under all these different classifications.

Mr. RANKIN. Take the case where a company has sold suits of clothes made of woolen goods to the amount of $300,000. They can easily ascertain the number of suits they have sold to make that amount. If you are going to get a census of distribution of woolen goods, are you going to go to a house and say, "How many dollars worth of woolen goods did you sell?"

Mr. GOSNELL. Well, you take the item of clothing made from wool. There are different prices of clothing, and we would have to get, under your plan, the number of suits sold at the different prices, would we not?

Mr. JOHNSON. You would have another trouble-I want to wait till you are finished. In the Rocky Mountain and Pacific Coast States numerous articles that are sold at retail are less in weight than those articles are in Eastern States. That is so they can absorb the long transportation haul. Take soap, for instance, it might weigh 9 ounces in all of the Atlantic States and 7 ounces in the Pacific coast area, and still be the same advertised brand of soap.

Mr. RANKIN. Well, there wouldn't be very much difference in the weight. And now right there is one of the points. You take the distribution of soap. This census of distribution you are starting in on is a revolutionary policy. When you start in taking a census of distribution, you start in on a much bigger proposition than you ever had before in all the census enumerations that we have ever had, if you are going to make it thorough and secure an intelligent census of distribution in the United States. And that is the very thing we want. You take the question of soap. Millions of gallons. of vegetable oil in the United States are used in the manufacture of soap. We trace this oil to where it goes into the soap, and we find that this factory here is turning out soap and shipping it to the Pacific coast, or all over the country. You find that on the Pacific coast this soap is being sold for a certain amount of money, but the thing we want to do is to find out how much of that soap it took to bring that amount of money, in order that we may know just what these commodities cost, and thereby develop the spread between the producer and the consumer and show where that spread is going to.

Mr. LOZIER. Mr. Rankin, as I understand it, under the census of manufactures, they show the value of the raw material, the value of the finished product, the value of labor, and possibly some other elements. Of course, you could take in any given industry, and by studying those figures you can arrive at a conclusion as to the amount. involved.

Mr. RANKIN. Well, take for instance a factory making overalls that sells $100,000 worth, suppose they sold $100,000 worth in the State of Washington; it would be very little trouble for them to show how many pairs or dozen paris of overalls were sold to bring that.

amount.

The CHAIRMAN. What is the purpose of the distribution? To ascertain what? What was manufactured in the different factories of the United States or to find what the selling price was or the amount received when it went from the wholesaler to the retailer or from the retailer to the ultimate consumer? I understand that the object of this distribution is to ascertain what the amount of sales

was.

Mr. RANKIN. Well, I don't understand it that way.

The CHAIRMAN. I think the desire is to find out for the business men of the United States what the sales are and what money is received for those sales; not what is made in some factory in Connecticut or some apple orchard in Washington or some dairy in Vermont. It is to ascertain what the cost, the aggregate cost of the particular article is when it gets to the consumer through the retailer. Mr. RANKIN. But he doesn't get that. There is the point exactly. The CHAIRMAN. Let me ask you, isn't that what your ascertainment is going to be?