Comment on “What the Welfare State Left Behind—Securing the Capability to Move for the Vulnerable”

Gotoh and Kambayashi (2022) tackle important topics, including proposing a new measurement of human welfare and capturing some of the problems under COVID-19. Their measurement is based on the capability approach, and it is their ultimate goal to reexamine the transformation of publicness in contemporary Japan.

I agree that Gotoh and Kambayashi's paper can significantly contribute to economic research, especially empirical welfare analysis (Fleurbaey, 2009; Saez & Stantcheva, 2016; Fleurbaey & Maniquet, 2018). Moreover, their paper can provide an important perspective to discussing public policies.

In particular, it is a unique and good idea to apply the capability approach to the “the stay-home policy.” Gotoh and Kambayashi's data is for older respondents with COVID-19, who may be seriously affected by the pandemic, and their capability is mostly damaged. I think both the policy and data are relevant to the capability approach.

However, I strongly suggest that Gotoh and Kambayashi improve the transparency of their methods and findings because some parts of the current version are unclear and not self-contained. I believe research transparency is especially important in the general-interest and policy journals, including the Asian Economic Policy Review. Moreover, transparency can increase the social impact of this important paper.

I have five specific comments. First, their main contribution is to evaluate human welfare by unique measurement connecting publicness and the capability approach. I think this measurement is a reasonable alternative to traditional welfare measurement (for example, income and consumption). However, I cannot find a clear discussion about the relevance of publicness and the capability approach. Those concepts may not be familiar to potential readers, and Gotoh and Kambayashi should then explicitly discuss why the capability approach is relevant for discussing and measuring publicness.

Second, I guess Gotoh and Kambayashi's paper is not a purely theoretical paper proposing new welfare measurement; measuring contemporary Japan is also an important research goal. However, while the measurement concepts are explained in detail, I can only find a poor explanation of their data. They should provide more information about the data and then discuss the uncertainty of measurement results due to the incompleteness of the data. I guess another paper (Kambayashi et al., 2022) provides a detailed explanation, but the present paper should include an explanation of important information, at least the sample size, the sampling method, and missing values. That information is especially important to evaluate the reliability of their empirical results. Additionally, I prefer researchers to report sampling uncertainly (for example, standard errors or confidence intervals).

Third, Gotoh and Kambayashi's introduction proposes two research questions: (1) Why could not the Japanese government quickly adopt precise policies, and (2) How can we stop such a policy of leaving the vulnerable behind the welfare state? I agree on the importance of these two questions but I do not find clear answers to those questions. In relation to question (1), their conclusion says, “our tentative answer to this question is: this problem is related to the fact that the public assistance system has been slowly but significantly curtailed in terms of the minimum standard of living and in the number of recipients.” However, it is not clear which empirical findings are relevant to this statement. Moreover, I cannot find any parts to answer question (2). Note that I request only partial answers, not “clear answers,” because their research questions are difficult.

Fourth, Gotoh and Kambayashi should discuss the external validity of their empirical findings. The introduction says, “the phenomena that emerge under COVID-19 are those that have been embedded in society for long.” I also think this is a good idea but request a clearer discussion of the external (long-run) implications of their findings with COVID-19. External validity is especially important for this paper because their empirical analysis focuses on older respondents with COVID-19, but the paper is motivated by long-run phenomena (for example, the long-run transition of the Japanese public assistance system).

My fifth and final comment is a more technical point. Their measurement focuses on a three-group setting. Is there any reason or justification for why they choose three groups? Moreover, can the measurement be extended to more groups? If the concept can apply in a more general setting, it can be applied in more empirical work. Additionally, it is better to provide a justification for the group definition in an empirical setting (disability, nursing care users, and the general elderly). I agree the definition is reasonable, but a clear discussion can help readers obtain a deeper understanding.