completely out of production at that particular time, so that in these years we have certain plants which were completely out of production, although certain others were doing fairly well. Here is a plant which started out well and then went below the general level, but has come ahead and is far above the rest in the EXHIBIT No. 64 PRODUCTION OF STEEL CASTING FOR NINE PLANTS 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 BUREAU OF THE CENSUS industry in its 1938 record. Again the story is one of differing behavior among the members of a single industry. Senator KING. It depends on local conditions and transportation, does it not? Cement might be produced in a place where the cost of transporting to the market is very small and there would be a great demand in that particular market at a given time, whereas in other sections of the country there would be no particular demand for cement products. Dr. THORP. I think there is a great deal to be learned in terms of finding out why these different plants behaved so differently. One can think of a good many possible explanations, and it ought to be possible, by going back to the record, to determine the reason or reasons. The CHAIRMAN. It certainly is clear from the charts you are presenting, that in these particular industries there is no common pattern which each plant follows. Dr. THORP. That is quite correct. The same thing is shown in the chart for "Coke Production", where there is perhaps more nearly a single pattern for most of the industry, but there is one plant which seems to have varied greatly from the general pattern. (The chart referred to was marked "Exhibit No. 66" and appears on this page. The statistical data on which this chart is based are included in the appendix on p. 233.) EXHIBIT NO. 66 COKE PRODUCTION JUNE EACH YEAR 1925-1938 PERCENT 250 200 150 100 50 10 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 BUREAU OF MINES Representative SUMNERS. May I ask, have you found anything to indicate that the right-hand plant going up to 220 may have itself very materially determined that some other plant could go down the line? You haven't been able to get into that? 1 Exhibit No. 65, supra, p. 123. Dr. THORP. We haven't gotten into that. We are planning to make studies of this kind of record, not just for nine plants, but for a large number plants, and see what can be learned from the record. Representative SUMNERS (interposing). Its queer behavior looks like he had some reason to boost him up. Dr. THORP. I suspect there is some reason, but I am sorry I can't tell you what it is. Flour production is from the point of view of the industry very stable. This heavy black line which shows the trend of the total output of the industry varies only slightly from year to year, and for many of the plants there is no great degree of variation from one year to another. Nevertheless, there are a couple of cases where the variation has been rather extraordinary and nothing like the trend for the entire industry. (The chart referred to was marked "Exhibit No. 67" and appears on p. 126. The statistical data on which this chart is based are included in the appendix on p. 234.) The CHAIRMAN. Now, you have one in which the variation in 1931 and 1932 runs from well below 100 to well above 400. Dr. THORP. That is right. I suspect that something happened in this plant in 1931 which was a temporary situation, as for example, a fire or a strike, or something of that sort, so that it is hardly fair to compare that year with 1932. It had done pretty well on the whole. EMPLOYMENT RECORDS OF INDUSTRY MEMBERS These Dr. THORP. Here we have a chart that is a little different. other charts are production records. This is an employment record. Now that you are used to looking at 9 lines, we show you a chart with 15 lines on it. These are 15 different plants. The lines are not tied together at a single point, but the average for 1923 to 1925 is taken at 100. In the rubber industry, as you can see, the variations have been tremendous. This one does not have a single central line indicating the average of the industry, and I defy anyone to look at that chart as it stands there and fill in a line that would be representative of the industry. The firms in terms of employment, have shown very wide fluctuations. As a matter of fact, the industry level is just below 100 in 1936, the last year that is plotted. You can see what happened in the industry quite clearly. The indexes for one group of concerns are around 225 and for another group below 120, rather an extraordinary division taking place. (The chart referred to was marked "Exhibit No. 68" and appears on p. 127. The statistical data on which this chart is based are included in the appendix on p. 234.) Mr. HENDERSON. Wouldn't it be true that those below the average in 1936 are undoubtedly pretty large producers? Dr. THORP. Inasmuch as the average for the whole is just below 100, I suspect it is not an unreasonable statistical deduction that some of the plants down in the lower levels must be larger than some of those at high levels, to get an average at that point. The CHAIRMAN. There are certainly more plants above the line than below. Dr. THORP. Yes; there are more plants above, and these are a good deal further above. Representative SUMNERS. You could find the same thing among farmers and even lawyers and politicians and everybody. Some people get along well and some don't. The CHAIRMAN. This is interesting from the point of view of employment. This purports to indicate the changes in employment, |