In this paper, we propose using a search for the optimal regularization of GMM as a way to identify the attainable maximal Sharpe ratio in a given investment opportunity set (e.g., the economy). Regularization is achieved by imposing a bound on the volatility of a flexible specification of the candidate pricing kernel, alongside other economically motivated restrictions. In an empirical application of this methodology to US equities, our estimates of the maximal attainable Sharpe ratio in the economy are between 21 and 35 percent annually, depending on the cross-validation criterion used in the search, thus in the low region of the range of values hitherto considered, either on theoretical or on empirical grounds, by the literature.