Causal machine learning tools are beginning to see use in real-world policy evaluation tasks to flexibly estimate treatment effects. One issue with these methods is that the machine learning models used are generally black boxes, that is, there is no globally interpretable way to understand how a model makes estimates. This is a clear problem for governments who want to evaluate policy as it is difficult to understand whether such models are functioning in ways that are fair, based on the correct interpretation of evidence and transparent enough to allow for accountability if things go wrong. However, there has been little discussion of transparency problems in the causal machine learning literature and how these might be overcome. This article explores why transparency issues are a problem for causal machine learning in public policy evaluation applications and considers ways these problems might be addressed through explainable AI tools and by simplifying models in line with interpretable AI principles. It then applies these ideas to a case study using a causal forest model to estimate conditional average treatment effects for returns on education study. It shows that existing tools for understanding black-box predictive models are not as well suited to causal machine learning and that simplifying the model to make it interpretable leads to an unacceptable increase in error (in this application). It concludes that new tools are needed to properly understand causal machine learning models and the algorithms that fit them.

Standards developed by standard-setting organisations (SSOs) – sometimes labelled private rulemaking – are part of larger practices of governance in most societies yet are underinvestigated from a policy process perspective. Utilising and developing the multiple streams approach (MSA), this article investigates a policy process moving between government and the SSO Standards Norway (SN). The study finds standardisation by SSOs to be an ambiguous institutional arrangement. Strong institutional barriers in theory did not work as such in the case investigated. This article argues that the differentiation between responsibility for process (SN) and content (committee) makes the standardisation process vulnerable. The concept of “institutional deficit” is introduced to describe a potential mismatch between SSOs producing policy in a government-like institution, but where the SSOs are not capable of taking responsibility for policies in a government-like way. This article finds the adjusted MSA useful in this potentially least likely case.

It has become universal to claim a relationship between social cohesion and quality public childcare. The more we invest in our children, the better the return later in life. We explore how childcare is used to strengthen social life by comparing the state role in civilising childcare as described in policy and educational guidelines. At one level, we find that policy intentions are framed by the idea of using childcare to reduce inequalities between people and social groups, and at another by the idea of civilising children to adapt to existing social structures. The analysis unfolds these two sets of intentions by showing how pedagogical ideas of child development become linked to the ways two very different states – Brazil and Denmark – formulate and organise ECEC policies as well as using childcare to bridge between people and social groups, and to civilising children to adapt to existing social structures.

The European Union (EU) has reacted to the migration crises of the last decade with growing externalization of migration management to neighbor countries often accused of not respecting human rights and individual liberties. Focusing on EU cooperation with the Southern neighborhood, as defined within the European Neighborhood Policy (ENP) framework, this paper investigates the recent developments in the EU's external migration policies, demonstrating that there is a gap between EU discourses and policy instruments identified by the EU as strategic tools of the European migration and asylum policy (MAP). Five Southern neighbor countries (SNCs) have been selected, Tunisia, Morocco, Egypt, Jordan, and Lebanon, to assess the extent to which the EU, when cooperating on migration and asylum issues, places international protection at risk instead of playing the role of humanitarian actor in accordance with the ideals and principles it defends. The research critically analyzes the EU's cooperation with SNCs concerning migration, as developed via international agreements, action plans, and mobility partnerships under the umbrella of the ENP. In particular, it explores the incompatibility between the European MAP instruments and international protection. MAP, the qualitative analysis shows, represents a clear mismatch between EU talk and action, outlining another case of organized hypocrisy.

Transgender inclusion within policy is critical yet often missing. We propose a policy tool to assesses human rights, access to resources and opportunities, language, and implications for transgender and nonbinary individuals. Acknowledging trans communities as standard policy practice can serve as an essential practice to shift dialogue and norms.

Research has shown that long-term care (LTC) volunteers play important roles in enhancing the quality of life (QoL) of older LTC residents, often through providing unique forms of relational care. Guided by Kane’s QoL domains, we used a modified objective hermeneutics method to analyze how unique volunteer roles are represented and supported in provincial policies in Alberta, British Columbia, Ontario, and Nova Scotia. We found that policies define volunteer roles narrowly, which may limit residents’ QoL. This happens through (1) omitting volunteers from most regulatory policy, (2) likening volunteers to supplementary staff rather than to caregivers with unique roles, and (3) overemphasizing residents’ safety, security, and order. We offer insights into promising provincial policy directions for LTC volunteers, yet we argue that further regulating volunteers may be an inadequate or ill-suited approach to addressing the cultural, social, and structural changes required for volunteers to enhance LTC residents’ QoL effectively.

Low and highly variable prices plague the coffee market, generating concerns that coffee farmers producing in shade systems under natural forests, as in biodiversity hotspot Oaxaca, Mexico, will abandon production and contribute to deforestation and reduced ecosystem services. Using stakeholder information, we build a setting-informed model to analyze farmers' decisions to abandon shade-grown coffee production and their reactions to policy to reduce abandonment. Exploring price premiums for bird-friendly certified coffee, payments for ecosystem services, and price floors as policies, we find that once a farmer is on the path toward abandonment, it is difficult to reverse. However, implementing policies early that are low cost to farmers – price floors and no-cost certification programs – can stem abandonment. Considering the abandonment that policy avoids per dollar spent, price floors are the most cost-effective policy, yet governments prefer certification programs that push costs onto international coffee consumers who pay the price premium.

Policy is ostensibly crafted upon an overarching notion of rationality, in the form of rules, roles and designs. However, sometimes policy deviates from formal templates and seems to be guided by a different governing ethic. Rather than categorising these as policy anomalies, we can understand them as the workings of what we will refer to as a relational model of policy. The relational model describes how policy outcomes emerge from the working and reworking of relationships among policy actors. We define relationality and develop its use in policy research. While the relational can be depicted as an alternative model for policy (e.g., Confucian versus Weberian), it is more accurate to understand it as a system that complements conventional policy regimes. To illustrate the concept, we examine examples from policymaking in China. We end with a discussion of how relationality should be a general condition that should be applicable to many, if not all, policy situations.

Forthright characterization of scientific uncertainty is important in principle and in practice. Nevertheless, economists and other researchers commonly report findings with incredible certitude, reporting point predictions and estimates. To motivate expression of incredible certitude, economists suggest that researchers respond to incentives that make the practice tempting. This temptation is the ‘lure’ of incredible certitude. I appraise some of the rationales that observers may have in mind when they state that incredible certitude responds to incentives. I conclude that scientific expression of incredible certitude at most has appeal in limited contexts. It should not be a general practice.

Practitioners of benefit-cost analysis face many difficulties. Despite the best training, guidance, and intentions, practitioners can stumble: actual benefit-cost analysis is hard and mistakes get made. Over the years, I have collected the mistakes I have seen in actual benefit-cost analyses. Many of the same mistakes occur over and over: the pitfalls of practical benefit-cost analysis. In this paper, I describe common pitfalls and suggest ways to avoid them.

This paper extends Woodford's approach to the robustly optimal monetary policy to a general linear quadratic framework. We provide algorithms to solve for a time-invariant linear robustly optimal policy in a timeless perspective and for a time-invariant linear Markov perfect equilibrium under discretion. We apply our methods to a New Keynesian model of monetary policy with persistent cost-push shocks and inflation persistence. We find that the robustly optimal commitment inflation is less responsive to a cost-push shock when the shock is more persistent and that the robustly optimal discretionary policy is more responsive to lagged inflation when inflation is more persistent.