We compare three quasi-experimental approaches to estimating the returns to schooling in Australia: instrumenting schooling using month of birth, instrumenting schooling using changes in compulsory schooling laws, and comparing outcomes for twins. With annual pre-tax income as our measure of...

Outside the United States, very little is known about long-run trends in school productivity. We present new evidence using two data series from Australia, where comparable tests are available back to the 1960s. For young teenagers (aged 13–14), we find a small but statistically significant...