Freedman (2008a,b) showed that the linear regression estimator is biased for the analysis of randomized controlled trials under the randomization model. Under Freedman's assumptions, we derive exact closed‐form bias corrections for the linear regression estimator. We show that the limiting distribution of the bias corrected estimator is identical to the uncorrected estimator. Taken together with results from Lin (2013), our results show that Freedman's theoretical arguments against the use of regression adjustment can be resolved with minor modifications to practice.
A signal is privacy‐preserving with respect to a collection of privacy sets if the posterior probability assigned to every privacy set remains unchanged conditional on any signal realization. We characterize the privacy‐preserving signals for arbitrary state space and arbitrary privacy sets. A signal is privacy‐preserving if and only if it is a garbling of a reordered quantile signal . Furthermore, distributions of posterior means induced by privacy‐preserving signals are exactly mean‐preserving contractions of that induced by the quantile signal . We discuss the economic implications of our characterization for statistical discrimination, the revelation of sensitive information in auctions and price discrimination.
We develop a framework for quantifying barriers to labor force participation (LFP) and entrepreneurship faced by women in India. We find substantial barriers to LFP, and higher costs of expanding businesses through hiring workers for women entrepreneurs. However, there is one area where female entrepreneurs have an advantage: the hiring of female workers. We show that this is not driven by the sectoral composition of female employment. Consistent with this pattern, policies promoting female entrepreneurship can significantly increase female LFP even without explicitly targeting female LFP. Counterfactual simulations indicate that removing all excess barriers faced by women entrepreneurs would substantially increase the fraction of female‐owned firms, female LFP, earnings, and generate substantial gains for the economy. These gains are due to higher LFP, higher real wages and profits, and reallocation: low productivity male‐owned firms previously sheltered from female competition are replaced by higher productivity female‐owned firms previously excluded from the economy.
THE SOCIETY’S MEMBERSHIP NUMBERS ARE PRESENTED in Table I, where the membership is classified according to institutional membership and individual membership. This year’s numbers confirm that the Society’s move to a “license” model has stopped and in fact inverted the longstanding decline in for institutional subscriptions. Individual membership has also increased, as it often does around years with a World Congress. All in all, our membership is up by close to one-quarter over 2015. Table II displays the division between print and online and online only memberships and subscriptions. Since the choice between these two alternatives was offered in 2004, there has been a continued shift toward online only. Many of the new institutional subscribers we reached with the license model have opted for an online subscription. On the other hand, individual membership numbers suggest that we may be close to a floor in print subscriptions at around 20% (and 10% for student members.) This remains to be confirmed. Table III compares the Society’s membership and the number of institutional subscribers with those of the American Economic Association. (For the membership category these figures include ordinary, student, free, and life members for both the ES and the AEA.) The ES/AEA ratio is at a record high level for institutional members; and it has returned close to its 2010 peak for individual subscriptions. The geographic distribution of ordinary and student members by countries and regions as of June 30 of the current and selected previous years is shown in Table IV. The table shows individual data on countries with more than 10 members in 2010. Our membership in Africa has been increasing fast. Several countries in Asia have also seen large increases, most spectacularly in China. On the other hand, it declined in the UK and in Germany. These changes are also apparent in Table V, which shows the percentage distribution of ordinary and student members by regions as of June 30 of the current and selected previous years. Finally, Table VI presents the percentage distribution of institutional subscribers by regions as of June 30 of the current and the previous four years. It shows that the increase in institutional membership this year is due to a remarkable recovery in the European region, where the trend had been negative for several years.