We present Aschauer’s main results and our replication in tables 1 and 2, which match up with Aschauer’s tables 1A and 1B, respectively (e.g., equation [1.2] in our table 1 is equation [1.2]in Aschauer’s table 1A). All equations in tables 1 and 2 are estimated by ordinary least squares (OLS), following Aschauer, except for equation (1.4), which is estimated by the Cochrane–Orcutt method. Our replication of Aschauer’s main result finds an elasticity estimate of 0.38 in equation (1.1) in table 1 for the ratio of public capital to private capital, which is statistically the same as Aschauer’s estimate of 0.39. The other estimatesin 1.1 also match up well with Aschauer’s results according to the p values, except for the constant terms. In equation (1.2), for example, our estimate for public capital is 0.33, which is close to Aschauer’s estimate of 0.36. We also come close to the results of Aschauer’s hypothesis testing. He tests the hypothesis of constant returns in all inputs in equation (1.2) in table 2. If there are constant returns in all three inputs, the coefficients on the two capital variables will be of opposite sign and equal in magnitude. The F-statistic for our version of the test is 2.72, compared to Aschauer’s F-statistic of 1.27, both of which are well below the critical value. It follows that the null hypothesis of constant returns to scale in all inputs cannot be rejected, agreeing with Aschauer. Table 2 uses total factor productivity ðrÞ as the dependent variable. In equation (1.5), public capital is highly productive, and our replication is quite close to Aschauer’s estimate (0.49 vs. 0.46). The other estimates also match well. Testing whether the parameters of the two capital variables are equal in magnitude but opposite in sign yields an insignificant F-stat of 3.85. We do not reject the null hypothesis of constant returns in all inputs. Lastly, equation (1.8) estimates equation (3), and our estimates coincide closely with those of Aschauer except for the constant term.