have finally discovered a useful function for ‘public choice' economics
applied to education: it reminds you, when you have had to read too much
dogmatic, illogical and mind-numbing excuses for left wing discourse in a
short period of time, that we on the left have no monopoly on the production
of such work.
Lakdawalla is a University of Chicago Economics Ph.D., now located at Rand.
He has managed to write a paper, based on a Ph.D. dissertation, which is
purportedly about teacher quality, but cites not one text in the field of
education research which examines the issue. The closest he ever gets to
education are articles by Hanushek and Hoxby, leading public choice
economists who have made careers out of bashing teachers, teacher unions and
public schools. He apparently aspires to join the club.
A lot of this essay is so bad you have to wonder if it could manage a passing
grade in an undergraduate course in anything other than a class taught by a
public choice true believer. It is hard to critique the work in a way that
does not seem very disjointed because the argument it presents is so
disjointed, with unsupported assertion piled upon unstated assumption and
strung together with Olympic leaps of logic.
Take his introductory point: Lakdawalla starts from the premise that in the
major 'developed' countries, the relative wage of teachers has been declining
over the last 35 years. To arrive at this claim, he ignores all of the
existing literature and data on teacher salaries. Instead, he takes
aggregate, national OECD and UNESCO figures of educational expenditures per
teacher, assumes that the proportion of expenditures spent on teacher
salaries remain constant without any explanation of why one might reasonably
make that assumption, deflates teacher wages by the growth in the wage of the
average employee, and on this basis, finds a decline in the relative wages of
teachers. He leaves out of his comparison data from countries such as the UK,
Canada, and Italy, while including Ireland and Finland -- all without any
explanation for the basis of inclusion and exclusion of different nations.
And because he just compares the starting and ending salaries over a 35 year
period, he ignores the fact – and the reasons behind the fact – that teacher
salaries in the US and elsewhere have been on a roller coaster over those
years, with periods of inflation and public sector fiscal crises leading to
losses in real wages and periods of public school expansion and unionization
leading to dramatic increases in real wages. Consequently, the points of
comparison could be completely atypical. He also ignores rather significant
variations in teacher salaries from state to state in the US, and among
regions in other countries. [BTW, the real international story here,
according to an OECD study just reported June 13 in the _New York Times_, is
that American teachers are paid a great deal less, and yet spend a great deal
more time in direct classroom instruction, than teachers in other developed
nations.]
To understand just how arbitrary this selection and organization of data is,
appreciate the fact that Lakdawalla manages to completely ignore not simply
the most teacher-friendly study on the question of teacher salaries, Allan
Odden's and Carolyn Kelley's _Paying Teachers for What They Know and Do: New
and Smarter Compensation Strategies to Improve Schools_, but also the major
conservative, laissez-faire market study, Dale Ballou's and Michael
Podgursky's _Teacher Pay and Teacher Quality_. [For those of you who follow
such things, Podgursky has a distant radical background from the 1960s,
co-authoring one of the essays in the collection _The Capitalist System_ that
was edited by Edwards, Reich and Weisskopf, and he and Ballou got together at
U Mass Amherst when both worked at that center of radical economic thought.
{-; ] But Ballou and Podgursky have different fish to fry than Lakdawalla, as
they want to make the case that there is no correlation between teacher
quality and teacher salaries as a result of union wage scales. They look at
teacher salaries and teacher quality throughout the 1980s, and conclude that
both rose, with teacher wages increasing both in real terms and in comparison
to other college educated professionals. But there was, they contend, no
correlation between the two developments; such a correlation would mean that
paying higher salaries across the board would lead to better quality, while
they want market-based and merit pay schemes in which the
administrator/employer decides who is the better quality teacher, and awards
salary increases and differentials on that basis. These results do not
conform with Lakdawalla's thesis of a decline in relative wages, so he just
ignores the study altogether.
From this basis, Lakdawalla proceeds with a leap of logic that could only
take place in this sort of public choice economics writing. Without the
slightest transition, he immediately follows up his claim with respect to the
decline in relative wages of teachers with the assertion, "Perhaps most
telling of all has been the dramatic decline in the relative quality of
teacher training." In other words, based on the unstated assumption that a
decline in wages must necessarily mean a decline in the quality/productivity
of the wage laborer [and, for him, productivity and quality are the same
thing], since the market always reflects the true value of a product, he
literally jumps over 15 intermediate pieces of reasoning to the conclusion
that teacher education is a primary cause of the decline in the relative
wages of teachers here. Evidence? Well, we are told, based on a single text
published in 1930, teachers' colleges at the turn of the century were
excellent; today, only the students at the bottom of their college classes on
achievement and intelligence tests seek out education degrees. Of course, the
latter point informs us, at the very best, about the ability of students in
schools of education [and – surprise – salaries have a lot to do with who
goes into a field], and nothing about the quality of the education they
receive. The quality of teacher education is not what it should be, but
Lakdawalla has not a clue about why that is so [the ‘golden past' is pure
myth], or how it could be improved. [A large part of the problem is that
universities treat schools of education as cash cows, having them produce
income without being given commensurate resources to do their job; an equally
important part of the problem is the refusal of the academy, in general, to
meaningfully combine theory and practice in the education of the teacher.]
It gets better. Continuing along with the general line that the market
reflects true value, it must therefore follow that there has been a decline
in the productivity/quality of teachers relative to other workers. It is
assumed that while the knowledge of skilled workers outside of teaching is
constantly growing as a result of innovation, the knowledge [arithmetic and
reading are cited] used by K-12 teachers remains the same. Of course, if
Lakdawalla spent one day in an actual K-12 classroom even he would realize
that teaching involves both a knowledge of the subject matter and of
pedagogy, and that pedagogy involves constant innovations; moreover, once one
gets beyond basic literacy and numeracy skills in K-3 [and even there one
sees constant changes, such that computer instruction is now begun at that
age], teachers are teaching subject material which significantly changes as a
result of innovation; who could seriously contend, for example, that the
science, mathematics, social studies, and even English language arts taught
in middle and high schools are essentially the same today as they were at the
start of the 20th century? Lakdawalla conveniently contends that this
incorporation of changes in the knowledge base of the subject is the case
with regard to his own post-secondary level of teaching, which he
conveniently describes in terms of the natural and physical sciences and not
the humanities and social sciences, but blithely asserts that it is not the
case for elementary and secondary teachers, which he conveniently describes
without any reference to the natural and physical sciences. [Ironically, once
you strip the mathematical modeling, it is his economics which most resembles
what was taught in that field at the start of the 20th century.]
When it comes to quantitative measures of the quality of K-12 teachers,
Lakdawalla uses the crudest possible measure: the number of years of formal
education teachers have. Teaching quality is a notoriously difficult
phenomenon to measure, especially on an aggregate basis, but there are
certainly measures in the literature which are more elaborate and useful than
this one. Even Ballou and Podgursky manage far better than this. Now, using
this measure, the argument is that the years of formal education of workers
outside of teaching have increased at the rate greater than the years of
formal education of K-12 teachers. This may very well be true, although I
question that it is true on the order Lakdawalla suggests, which is that the
general population has gained four more years of formal education than
teachers. Formal education requirements of K-12 teachers in the US have
increased over the course of the century, from a 1900 epoch in which
elementary school teachers often needed little more than a high school
education and secondary teachers only two years of normal school,
post-secondary education to a 2000 era in which all teachers need a four year
college education, with many states requiring a M.A. But the increase in
formal education among the general population has been just as extraordinary,
and could well be greater in terms of gains. Remember, that it was not until
after WWII that most Americans even graduated high school; today, we have the
highest post-secondary attendance rate in the world, well above 40% the last
time I looked. [Lakdawalla never considers the fact that the US has a much
larger post-secondary population than any other nation, which would make it a
rather poor exemplar for an international hypothesis.] But even if true, what
does one make of the fact that the rest of the American population has
increased their formal education at a greater rate than K-12 teachers? After
all, the slowest rate of increase, and the smallest absolute increase, in the
number of years of formal education is found among post-secondary educators.
This quantitative measure would thus indicate an even greater decline in
relative quality for the post-secondary educators which Lakdawalla has
exempted from his generalization: the logic of the argument, if we can call
it that, is internally contradictory.
There is more. Take this gem. First, Lakdawalla asserts that "the price of
[teacher] skill rises, because the demand for skilled workers outside
teaching rises." Based on this premise, he concludes that "since the price of
teacher skill rises relative to the price of teacher quantity, schools
respond by lowering the skill of teachers and raising the quantity of
teachers employed." This prediction, he assures us, is consistent with
"increases in the teacher-student ratio." Now since Lakdawalla once again
chooses to be completely ignorant of all the literature on class size, he
does not consider the actual reasons behind initiatives to lower class size:
a general recognition that the larger the class, the more difficult it is for
teachers to do anything but lecture [the worst pedagogical method,
particularly at an early age] and the more impossible it becomes to provide
individual attention to students. Study after study has confirmed that lower
class size produces a better quality education, with the most dramatic
improvements being in the early grades and in inner city students. Thus, a
combination of general policy initiatives and teacher union contracts have
driven class sizes. I sincerely doubt that one could find an educational
policy maker anywhere which would give the slightest credence to Lakdawalla's
explanations for their actions. This is one case where "productivity" [in the
sense of the number of students educated by one teacher] clearly conflicts
with "quality," public choice economics notwithstanding.
There is no consideration of the effect of gender ghettoization on teacher
salaries, or attempt to address these issues. No explanation is offered,
other than the tautological one that wage disparity reflects skill disparity,
for the fact that teachers are the most poorly paid professionals with
equivalent educational backgrounds. The mass of the essay is taken up with
developing a regression model "to infer from relative wage data the relative
quality of teachers." A World Trade Tower built on a foundation of sand.
>[Leo, anyone else - comments on this?]
>
>"The Declining Quality of Teachers"
>
>BY: DARIUS LAKDAWALLA
>RAND, Santa Monica
>National Bureau of Economic Research (NBER)
>University of Chicago
>Department of Economics
>
>Document: Available from the SSRN Electronic Paper Collection:
>http://papers.ssrn.com/paper.taf?abstract_id=268344
>
>Paper ID: NBER Working Paper No. W8263
>Date: April 2001
>
>Contact: DARIUS LAKDAWALLA
>Email: Mailto:[EMAIL PROTECTED]
>Postal: RAND, Santa Monica
>1700 Main Street
>P.O. Box 2138
>Santa Monica, CA 90407-2138 USA
>
>Paper Requests:
>Full-Text downloads are available from SSRN Online for $5.
>
>ABSTRACT:
>Concern is often voiced about the declining quality of American
>schoolteachers. This paper shows that, while the relative quality of
>teachers is declining, this decline is a result of technical change, which
>improves the specialized knowledge of skilled workers outside teaching,
>but not the general knowledge of schoolteachers. This raises the price of
>skilled teachers, but not their productivity. Schools respond by lowering
>the relative skill of teachers and raising teacher quantity. On the other
>hand, college professors, who teach specialized knowledge, are predicted
>to experience increases in skill relative to schoolteachers. Finally, the
>lagging productivity of primary schools is predicted to raise the unit
>cost of primary education. These predictions appear consistent with the data.
>Analysis of US Census microdata suggests that, from the 1900 birth cohort
>to the 1950 birth cohort, the relative schooling of teachers has declined
>by about three years, and the human capital of teachers may have declined
>in value relative to that of college graduates by as much as thirty
>percent, but the teacher-student ratio has more than doubled over the last
>half century in a wide array of developed countries. Moreover, the per
>student cost of primary school education in the US has also risen
>dramatically over the past 50 years. Finally, the human capital of college
>professors has risen by nearly thirty percent relative to school teachers.
Leo Casey
United Federation of Teachers
260 Park Avenue South
New York, New York 10010-7272 (212-598-6869)
Power concedes nothing without a demand.
It never has, and it never will.
If there is no struggle, there is no progress.
Those who profess to favor freedom, and yet deprecate agitation are men who
want crops without plowing the ground. They want rain without thunder and
lightning. They want the ocean without the awful roar of its waters.
-- Frederick Douglass --
