Keynes thought it would be “splendid” if economists became more like dentists. Disciplinary economics has instead become more like physics in focusing on concise, universal propositions verified through decisive tests. This focus, the author argues, limits the practical utility of the discipline because universal propositions form only a part of new policy recipes. The author further suggests that, as in engineering and medicine, developing economic recipes requires eclectic combinations of suggestive tests and judgment. Additionally, the author provides a detailed example of how a simulation model can help evaluate new policy combinations that affect the screening of loan applications.