and you have just drawn the deduction that the plants which are represented as below the line are large producers; the plants which are above the line are small producers. Would it be a justifiable deduction that the small producers have had better luck in employment than the large producers?

EXHIBIT NO. 68

EMPLOYMENT FOR FIFTEEN PLANTS IN THE
RUBBER TIRE AND TUBE INDUSTRY

Dr. THORP. I think that might be a deduction from this chart for at least some small producers among this particular group of plants. I would need to know more about the operations of these individual plants before venturing a definite statement.

Senator KING. You are dealing only with rubber?

Dr. THORP. This chart refers only to the rubber tire and tube industry.

Senator KING. You recall, do you not, that there have been very wide fluctuations in the prices of rubber during the past few years? There have been strikes in some places, notably Akron, Ohio, where they practically destroyed the industry, as well as some other places, so that those strikes or untoward circumstances would create great changes and perhaps the big industry would suffer more than the smaller one.

Dr. THORP. I had no intention of trying to explain these variations. We do hope to derive some useful conclusions from such studies. At the moment, I merely wanted to introduce it as raising a question of variation within industries.

Dr. LUBIN. If I might add at that point, before we ever had these strikes there was a tendency for certain firms to move plants into areas where labor costs were much lower, the result has been that. some large firms opened plants in Southern areas and in other areas with lower labor costs. These new plants increased their output, and consequently their employment, much faster of course than was true of the same firms in their older plants which were already large. In other words, shifting business away from the larger to the newer plants.

Actually, however, such movements of tire employment cannot be be studied from this particular chart which ends with 1936 and which includes only a sample of firms that have been in operation since 1923 and have reported employment each year. It shows neither firms that disappeared or new establishments that have started since 1923. I cannot identify the individual firms on this chart, for the reports are received in confidence. It so happens that both the large increases and the large decreases on this particular chart were registered by relatively small companies. However, no generalization can be drawn from this one chart other than that it illustrates the diversity of movement among business enterprises. The Bureau of Labor Statistics is studying employment changes in large and small enterprises on a broad scale at present.

The CHAIRMAN. That suggests another question to my mind. Dr. Thorp, does this chart purport to indicate that each of these plants represented by a separate line is under separate ownership or control? There is nothing to indicate how many of these plants were owned by the same large company, for instance.

Dr. THORP. We have nothing to indicate that, as far as I know. Mr. DAVIS. Dr. Thorp, in some of our studies I think you probably find that contracts made by some of the rubber companies with the large mail order houses and chains resulted in the large increase of their trade, and the relative decrease of some of their competitors? Dr. THORP. I had intended here to introduce a discussion of conflicts which appear in the N. R. A. experience as between groups within industries, but I think rather than take any further time on that, if I may just file the report of the President's Committee of Industrial Analysis which lists a series of conflicts, that will perhaps save time. May I just file that for the record?

(The report referred to was marked "Exhibit No. 69" and is included in the appendix on p. 235.)

SHIFTS IN IMPORTANCE OF INDUSTRIES AND TRADES

Dr. THORP. We now come to talking about what has happened to certain industries.

This is a record which shows the leading manufacturing industries in 1899 and 1929. The value of product of the manufacturing industries of the United States had multiplied about six and a half times between those years. The leading industry in 1899 was the iron and steel industry, and the leading industry in 1929 was the motor-vehicle industry. The value of products of the motor-vehicle industry in 1929 was about six and a half times that of the iron and steel industry in 1899.

(The chart referred to was marked "Exhibit No. 70" and appears on this page.)

Dr. THORP. The interesting thing to note here is that five industries appear on both lists: Iron and steel, wholesale slaughtering and meat packing, printing and publishing, cotton goods, and flouring and grist mill products. On the other hand, we have five new ones in 1929. Over the period of 30 years there was a considerable shifting about in our whole economic activity. Part of it, of course, is due to the shifting in certain activities from the home. For instance, bread and bakeries appeared in the 1929 list but not in the list for 1899. There is a rather amazing growth in printing and publishing. Of course, the outstanding development is in the motor-vehicle industry which did not appear at all in 1899, and was the largest producer in 1929.

For retail trade I have a somewhat different explanation to make. This chart is based upon a special study which we made at Dun &

Bradstreet of the retail store population in 32 county-seat towns— these are towns averaging five to ten thousand population-in 1915 and 1935, for different types of stores. Because the grocery stores dominated, we had to break the chart at this point, just below 400, or the grocery-store chart line would go up much higher.

(The chart referred to was marked "Exhibit No. 71" and appears on this page. The statistical data on which this chart is based are included in the appendix on p. 235.)

Dr. THORP. In 1915 these towns, scattered all around the country, had 671 independent grocery stores. In 1935 they had 898 independ

ent grocery stores. Actually, the number of grocery stores had appreciably increased. The black area at the bottom of each column indicates the number of grocery stores in existence in 1915 which still survived in 1935. Only 67 of the 671 grocery stores remained in

1935.

Now we come to general and department stores. In 20 years the number has dropped decidedly, from over 200 to 118. The number of drug stores remained almost the same. Men's clothing stores dropped from 128 to 87; dry goods stores from 124 to 55; hardware stores declined from 110 to 80; shoe stores dropped from 88 to 59; women'swear shops increased from 22 to 73.

It happens that the total number of stores was almost exactly the same in 1915 as in 1935. Most of the lines have shown decreases except for groceries, which went up, and for women's wear which increased more than threefold.

The changes that took place in the retail industry between 1929 and 1935 are indicated in the Bureau of Census records. For example, in that period of 6 years the number of eating and drinking places increased about 120,000, filling stations 75,000, and second-hand storesthat is one interesting result of the depression-increased by 7,000. On the other hand, furniture and household appliance stores decreased 14.000, and all apparel stores 18,000.

So you have within the population rather extraordinary shifts from one industry to another, and one type of activity to another. Of course, I hardly need to point out that within the enterprise itself there are important shifts, that the grocery store has recently been adding meat and tobacco products, and now many of the grocery stores are carrying standard drug items.

Here is another kind of shift that appears within industries, that is the seasonal variation of business activities. It is very important in many industries and raises certain difficult problems of planning, financing, and marketing. This chart presents the seasonality of industrial operations for nine industries. The cement industry, for example, is one of our highly seasonal industries, showing its peak in the summer, while the coal industries show their low point in the summer. (The chart referred to was marked "Exhibit No. 72" and appears on p. 132. The statistical data on which this chart is based are included in the appendix on p. 236.)

VARIETY OF TRADE PRACTICE PROBLEMS

Dr. THORP. Now we come to a problem of other types of variations within industries which cannot be measured statistically. Those are the variations in industries which result from different practices and patterns and habits which the industries themselves have developed. It is obvious that industries will be different because of the fact that they have different products and different processes and different types of markets. Their histories have also varied, and it is not surprising that the focal point at which their problems appear will vary.

For example, the problem of trade-ins is naturally a problem restricted to certain large-scale industries which produce durable products. Trade-ins may appear as a major problem in the automobile industry but they have not as yet appeared as a problem in the