To bring this issue into focus: the average school district in States requiring a college degree for certification enrolls 1,032 pupils; in States requiring 2 years of college, 273 pupils; in the three States (Nebraska, North Dakota, South Dakota) demanding less than 2 years of college, only 52 pupils. In other words, the nondegree teacher can be found primarily in the extremely small districts. The root of the problem is not inadequate supply of qualified teachers, but inefficient school organization and low standards of certification. The teacher-pupil ratio in some of the "small-district" States is around 1: 20 or lower; in Nebraska it stands at 1:19.1; in North Dakota at 1:18.6; South Dakota at 1:18.4; Kansas, 1:20.8 in the current school year. The States in which the tiny districts prevail employ thousands of teachers who would not be needed if the schools were consolidated. The results of this inefficiency are low teachers' salaries and low academic teacher qualifications.

The solution to the problem of the nondegree teacher is school district reorganization which will eliminate thousands of teaching positions. If this were done, the schools would then be able to pay higher salaries to the fewer, but better qualified teachers and raise the hiring standards for new teachers.

For years, there has been clear evidence that the teacher shortage is rapidly diminishing. Three years ago, the National Education Association, in its 1956 Teacher Supply and Demand Report "predicted an end to the teacher shortage by the early 1960's."" In an attempt to verify repeated reports that the teacher shortage is easing (e.g., the New York Times, Aug. 17, 1958) the Journal for Teacher Education conducted a survey in fall, 1958. Three-fourths of the districts reported a greatly improved teacher supply situation. Main reasons: Better salaries and working conditions, return of teachers from industrial employment in significant numbers.

The shortages of teachers are concentrated in a few fields such as mathematics and science. It has been known for some time that teachers trained in those fields sometimes receive offers of better-paying jobs in industry. Suggestions to make the schools competitive with private employment opportunities by paying science teachers more than other teachers were overwhelmingly rejected by school administrators (84 percent No).8

Nevertheless, it is encouraging to note that science and mathematics teachers registered the greatest increase (18.8 percent and 18.4 percent, respectively) of any subject field in that 1958 teacher graduating class."

It is not well recognized that the number of pupils per staff member has been consistently reduced from 35.6 in 1900 to 31.8 in 1920, to 27.9 in 1940 to 24.8 in 1958-59. The decline was 3.8 pupils per staff member between 1900 and 1920, another 3.9 pupils between 1920 and 1940, and still another 3.1 pupils between 1940 and 1959.

Enrollment and instructional staff in public schools, 1900 to 1959

Source: 1900 to 1940: U.S. Office of Education, Statistics of State School Systems, 1953-54. 1959: National Education Association, Estimates of School Statistics, 1958-59.

A study of 40 large city school systems by the Los Angeles school system indicated that the pupil-teacher ratio dropped 8.0 pupils in elementary schools, 7.5 pupils in junior high schools, and 7.2 pupils in senior high schools between 1935 and 1958.10

7 The Nation's Schools, March 1956, p. 134.

8 The Nation's Schools, June 1956, p. 76.

NEA, Teacher Supply and Demand in Public Schools, 1958, p. 4.

10 The Nation's Schools," June 1958, p. 8.

The teachers salary problem is part of the overall problem of "white collar" pay. There is ample evidence that over several decades the earnings of manual workers have risen relatively more than those of professional and managerial workers. Blank and Stigler found: "Between 1950 and 1954 the ratio of engineering salaries to earnings of all wage and salary employees was about a third lower than in 1929." 21

Other professional employees suffered similar relative losses. They could not match the bargaining power and political strength of the industrial unions representing blue collar workers. There is, however, one major difference: teachers salaries have risen proportionately more than the wages of all employed persons. This holds true of comparisons starting with any base year except the years of the great depression. During the 1930's wages of private employees dropped sooner and more sharply, and recovered more slowly than those of teachers.

Earnings of all workers and of public school teachers, 1929 and 1957

Source: Department of Commerce, Survey of Current Business. NEA, Economic Status of Teachers in 1957-58, p. 19 (teachers' salary in 1957 was corrected according to the more recent NEA report, "Estimates of School Statistics, 1958-59," employing the same method of computing teachers' salary on a calendar year basis as for 1929).

Teachers' salaries will have to be raised to a higher level relative to those of manual workers. So will the earning of other professional and managerial workers. But it may be difficult to convince communities and State legislatures that increases for teachers ought to be made uniformly across the board. They will be hard to persuade that all teachers should be paid as much as they believe good teachers are worth. We question whether it is good policy to circumvent the sound judgment of the communities and States legislatures by shifting responsibility for teachers' salaries to the Federal Government.

One inescapable condition for paying teachers higher salaries is the adoption of methods that will lead to fuller and more efficient use of their skills and time. This has been the avenue to higher earnings for many other groups of workers in the American economy.

Teachers' salaries are not uniform throughout the United States. Neither are other incomes, living standards, or other economic and social conditions. But the range in teachers' salaries has shrunk dramatically. The ratio between the highest and lowest State average was 1 to 5 in 1941-42. In 1958-59 it is 1 to 2, a narrower range than in per capita income.

In 1941-42, teachers in the lowest State received about one-third of the national average teacher pay; in 1958-59, they receive two-thirds of the national average. This is a remarkable reduction in the differential. Relative to the average income of other citizens, teachers are generally better off in the low-income than in high-income States.

3. CLASSROOM SHORTAGE

The Office of Education on January 28, 1959, released a statistical summary according to which the State departments of education reported a need for 140,500 classrooms, compared with 142,300 a year earlier. At this rate of reducing the backlog-1,800 classrooms a year-it would take many decades to eliminate the reported shortage.

The size of the classroom shortage has been a controversial issue for some years. Conflicting and often contradictory estimates, reports, and testimony from official sources have compounded the confusion. In some of the hearings the debate over the rise of the classroom shortage was referred to as "the

21 David M. Blank and George J. Stigler, "The Demand and Supply of Scientific Personnel," National Bureau of Economic Research, New York, 1957, p. 24.

numbers game." The term, improper as it is, unfortunately well characterizes the situation.

22

In 1950, the U.S. Office of Education estimated the classroom shortage at more than 250,000. After the completition of the school facilities survey, mandated by the 81st Congress at a total cost of over $5 million, shortage estimates were raised to 312,000; 340,000 and finally, in testimony of the Commissioner of Education on October 8, 1954, 370,000 classrooms.23

Other statements by high officials placed the shortage, then existing or prospective, at 500,000 to 600,000 classrooms.

On March 29, 1955, the Secretary of HEW testified that estimates of the classroom deficit expected by school year 1959-60 had been reduced from 407,000 to 176,000. In fall, 1956, the Office of Education reported a shortage of 159,000 classrooms," in fall, 1957, of 140,000, later revised to 142,300; it predicted a decline to between 128,000 to 132,800 by fall, 1958, and to between 114,800 to 126,800 by fall, 1959.27

26

In 1959, however, the Office of Education placed the shortage in fall, 1958, at 140,500, with an expected reduction to "not less than 133,500," by fall, 1959.28 These Office of Education reports seem to indicate a reduction in the classroom shortage from 370,000 in fall, 1954, to 140,500 in fall, 1958.

It has been questioned whether these figures are consistent and comparable. It may be equally doubted that the reports of a 142,300 classroom shortage in 1957 and of 140,500 in 1958 are comparable, or at all meaningful.

A State-by-State analysis raises huge question marks. For example, the reports indicate an increase in the classroom shortage between 1957 and 1958 of 5,950 in New York and of 3,298 in Texas. During that year the two States completed 10,780 classrooms-but their shortage is supposed to have risen 9,158.

On the other hand, the classroom shortage in Arkansas was reported to have been reduced by 7,098 classrooms between 1956 and 1957 and by 4,318 in South Carolina.

In fall, 1958, report shows a shortage of 11,936 classrooms in Alabama, of 1,200 in Arkansas; of 4,173 classrooms in Minnesota, and of 379 in Wisconsin. These contrasts are so vast that we question the comparability of the State reports and the validity of the "national" totals. The more thoroughly these reports are analyzed, the more it appears that they express the state of mind of those who prepared them rather than the actual classroom situation in the several States. If the individual State reports are inconsistent, unreliable and noncomparable, how can they be added up to arrive at a national total? And how can such shaky evidence be used as a basis for Federal legislation?

In his testimony on February 17, 1959, the Commissioner of Education mentioned two cases of inadequate volume of school construction:

"For example, one State reports a total need of 11,936 classrooms in the fall of 1958, but indicates that only 963 classrooms are scheduled for completion by the end of the school year 1958-59 ***"

A quick check reveals that school information from that State (Alabama) is greatly confused or confusing. The classroom inventory at the beginning of 1957-58 showed 14,134 classrooms available (Office of Education, Circular No. 513 revised). A year later, 22,858 classrooms were reported to be in use (Office of Education, Circular No. 551). Not long ago, the State reported having twice as many teachers as classrooms. Does this type of information justify the far-reaching legislation now before this committee? Parenthetically, it should be mentioned that this State ranks next to the lowest in school construction effort over the past 10 years (school capital out

22 Office of Education, "Report of the State Phase of the School Facilities Survey," 1953, p. 17. 23 Federal Aid to School Construction," hearings before a special subcommittee of the Committee on Education and Labor, House of Representatives, 83d Cong., 2d sess., 1954, p. 365. 24 "Federal Aid to States for School Construction," hearings before the Committee on Education and Labor, House of Representatives, 84th Cong., 1st sess., 1955, p. 282. 25 U.S. Office of Education, Circular No. 490.

26 U.S. Office of Education, Circulars Nos. 513 and 513 revised.

27 Departments of Labor and Health, Education, and Welfare appropriations for 1959, hearings before the Subcommittee of the Committee on Appropriations, House of Representatives, 85th Cong., 2d sess., 1958, p. 195.

28 Testimony of the Commissioner of Education before the Subcommittee on Education, Senate Committee on Labor and Public Welfare, Feb. 17, 1959.

lays related to personal income). Nevertheless, the voters of that State twice within the past 4 years rejected proposals by the Governor and legislature to issue State bonds for school construction. They also rejected on two occasions in that period recommendations to boost corporate or personal income taxes for school purposes. Should a State in which the people have refused to use their own resources be given funds from other States to finance their schools? Now the second case mentioned by the Commissioner:

"Another State, more economically favored, reports a shortage of 11,117 classrooms in the fall of 1958, with only 4,500 scheduled for completion in 1958-59." That State (Michigan) reported a need for 3,300 classrooms in fall, 1956 (U.S. Office of Education, Circular No. 490). A year later, the need had jumped to 10,111 (Circular No. 513), now, another year later to 11,117 classrooms. The State reported having completed 8,877 and having abandoned 650 classrooms during the 2-year period. The net increase of 8,227 classrooms-at 28 pupils per class-could accommodate 230,356 children. According to the same reports, enrollment increased only 148,000. The number of classrooms built in the 2year period exceeded by more than 3,000 the number needed to house reported enrollment increases and replace abandoned classrooms. How then can the "shortage" in Michigan have risen by 7,817 classrooms? While other inconsistencies in these reports could be cited, the above specifics suggest the variety in methods and criteria-and perhaps competency-in the States' reports.

The foregoing analysis is somewhat technical. But there is no other way to demonstrate the reliability—or unreliability—of a report. The claimed shortage of 140,500 classrooms in fall, 1958, and the reduction of the backlog by only 1,800 classrooms over the past year are substantiated in many States by nothing more than estimates whose standards vary widely from year to year and from place to place.

Despite the questionable validity of the "national" totals derived by the Office of Education, they constitute the only basis for the belief that a national emergency in school housing exiss. Those who believe these figures should recognize that the Murray-Metcalf proposals would not correct, but further complicate the school housing picture as reported by the Office of Education. Major need for classrooms is reported to exist in such States as New York, Michigan, and California. This might be expected since the greatest enrollment increases have been and are predicted to be in such high-income States. The Murray-Metcalf redistribution of wealth would exploit these States, then, at a time when they are struggling to cope with huge demands for more capital outlay and higher operating budgets. In the fourth year of its operation, this bill would cause the 15 States that reported 44 percent of Office of Education "need" for new classrooms (to house "excess enrollment") and 54 percent of the substandard teachers, to lose almost three-fourths of a billion dollars-and this drain on their resources would continue indefinitely. This bill then even contradicts the often repeated phrase of the NEA about putting the dollars where the children are. (See attached table of State allotments under S. 2).

The fact is that the classroom shortage is rapidlly diminishing, though faster in some localities than in others. Any local reluctance to issue bonds or boost taxes seems a poor justificatin for taxing the rest of the country for the benefit of lagging communities or States.

The prospects of meeting reasonable school facility needs during the 1960's are good. A projection to 1970-71 appears as follows:

Classrooms

The increase in school enrollment from 1957-58 to 1970-71 will approximate 10.5 million pupils, which at 28 pupils per classroom will require some_.

378, 000

Replacement needs have been estimated by the Office of Education to be between 14,000 and 20,000 classrooms per year. Taking an average of 17,000 a year, the 13-year need will be__.

The questionable "backlog" in the fall, 1957, was‒‒‒‒‒‒‒

221, 000 142, 000

Hence the total classroom requirement for the 13-year period
will approximate‒‒‒‒‒

This will require an average yearly construction of__

741, 000 57,000

During the school year 1957-58, 71,600 classrooms were completed which is considerably more than the average need for the 13-year period ahead. A similar number is predicted by the Office of Education this year.

School construction has increased much more rapidly than other construction. A 10-year comparison is as follows:

Source: Department of Commerce, "Construction Review," statistical supplement and February 1959.

School construction could be expected to grow even faster if it were not for three factors:

(a) the huge existing tax debt burden-largely Federal-exerts a restraining influence upon the voters and taxpayers;

(b) the unfortunate conflict over desegregation adversely affects the willingness to go ahead with construction projects in some areas;

(c) the expensiveness of some school building proposals delays the acceptance of those or subsequent projects.

Some communities have been divided by long drawn-out arguments as to whether proposed school plants are too costly or too austere. Several national magazines have entered the debate on whether school facilities should be built more economically.

The

School Life, monthly magazine of the Office of Education, in December 1958, released a survey of 69 new school plants with 1,458 classrooms for 39,140 pupils. The construction cost averaged $23,706 per classroom and $883 per pupil. Per pupil costs were $1,075 for high schools and $609 for elementary schools. The majority of schools being built nowadays cost far more. Congressional Record for July 2, 1958 (p. A6004), contains a construction report for New York City. It states "The budgets we have previously adopted provided for 285 construction projects seating 284,041 children at a cost of $785 million." This averages $2,764 per child.

A report of a 6-month study of school construction by the Rensselaer Polytechnic Institute to the New York State commissioner of education concluded that "schools can be constructed and maintained at costs far below current expenditures."

9929

The building space per pupil in new school plants now averages close to twice as much as it did a generation ago. This may be required by new educational methods or it may serve greater convenience. In either case, it affects the cost of the projects and the speed with which classrooms are built.

We do not want to go further into the question of school building costs except to state that there is an inescapable connection between the average cost of projects (on a per pupil or per classroom basis) and the time within which classroom shortages are eliminated.

The term "firetraps" has been used on several occasions by advocates of Federal intervention to characterize some of the older school buildings now in use. It seems that the record of fire safety should be examined free of emotional appeals. In' the past 20 years two fires have occurred in public schools which caused loss of life. Fifteen of the total of seventeen fatalities occurred in Cheecktowaga, N.Y., in 1954. It should be mentioned that that school was built with Federal funds. During the 20-year period with 17 school fire fatalities more than 600,000 persons were killed in home accidents. The number of fire deaths was 130,000. Altogether almost 2 million persons lost their lives in accidents. Compared with this loss, public school buildings have an enviable record of fire safety.

The volume of school bonds sold in 1958-$2,311 million-is almost identical with the $2,361 million sold in 1957. From 1 to 2 years pass between the sale of bonds and the completion of a school project. Thus, a continuation of the record building volume can be anticipated. The interest rates on new municipal bond issues presently average 3.4 percent. Treasury securities yield close to 4

29 "Economy Urged on New Schools," the New York Times, Oct. 16, 1958.