The Business Council of Australia’s new report on teacher qualityÂ has hit the papers yesterday and today, with its recommendations for top teachers to get six-figure salaries. Oddly, this drew a hostile response from John Della Bosca, the NSW Education Minister. One might have expected him to agree with the aspiration of better-paid teachers, but quibble on the details. Instead, his spokesman told the SMH’s Mark DavisÂ that it was “unfunded, unworkable and is not an actual scheme”.
I think it’s terrific that the BCA are focusing on improving teacher quality, but my read of the evidence is that teacher certification programs are heavy on paperwork, and light on results. The problem with such schemes is that they not only set a floor; they also discourage skilled applicants from entering the profession, by creating a time-consuming disincentive. As Joshua Angrist and Jonathan Guryan conclude in a new paper on US teacher certification:
Does teacher testing raise teacher quality? Evidence from state certification requirements
Abstract: The education reform movement includes efforts to raise teacher quality through stricter certification and licensing provisions. Most US states now require public school teachers to pass a standardized test such as the Praxis. Although any barrier to entry is likely to raise wages in the affected occupation, the theoretical effects of such requirements on teacher quality are ambiguous. Teacher testing places a floor on whatever skills are measured by the required test, but testing is also costly for applicants. These costs shift teacher supply to the left and may be especially likely to deter high-quality applicants from teaching in public schools. Moreover, test requirements may disqualify some applicants that schools would otherwise want to hire. We use the Schools and Staffing Survey to estimate the effect of state teacher testing requirements on teacher wages and teacher quality as measured by educational background. The results suggest that state-mandated teacher testing is associated with increases in teacher wages, though we find no evidence of a corresponding increase in quality. Consistent with the fact that Hispanics have marked lower licensure scores than non-Hispanic Whites or Blacks, testing appears to reduce the fraction of new teachers who are Hispanic.
The natural alternative is some form of merit pay, which also has its potential drawbacks, but seems at least worthy of a randomised trial.