Institutional perspectives on tax evasion attitudes: a Japan–US comparison with insights from the covid-19 pandemic

Purpose This study aims to use a neo-institutional framework to investigate the complexities of tax morale and its influence on tax evasion and compliance in Japan and the United States. It aims to advance theoretical understanding and contribute to the academic discourse on tax morality by focusing on three institutional characteristics: normative, cultural-cognitive and regulative. Design/methodology/approach Using a cross-country and regional approach, this study examines tax morale under ordinary and extraordinary circumstances. This cross-country comparison explores the relationship between institutional factors and tax morale in Japan and the United States using World Value Survey (WVS) data (2017–2022) and an ordered probit model for sequential estimation. A regional survey of Kyushu, Japan, was conducted during the COVID-19 pandemic. Findings The cross-country comparisons revealed mixed results for the normative and regulatory factors, with cultural-cognitive factors demonstrating more consistent outcomes. Compared to the WVS sample, the regional study revealed significant variations in demographic variables like age, gender, marital status and religiosity. These findings provide insights into tax morale and compliance behavior variations during times of uncertainty. Research limitations/implications Although this study provides valuable insights, it has some limitations. First, although extensive, WVS data do not fully capture specific subpopulations within Japan and the U.S. Our regional analysis of Kyushu during the COVID-19 pandemic had a relatively small sample size (n = 103), which limited its generalizability. Furthermore, the temporal discrepancy between the pre-pandemic WVS data and the pandemic-era Kyushu survey discouraged direct comparison. It is likely that the unique circumstances of the pandemic influenced the respondents’ attitudes, limiting their generalizability to nonpandemic periods. Second, self-reported survey measures can introduce biases such as a social desirability bias. Moreover, while we used established instruments such as McGee’s 15-item survey, aggregating items into a single Tax Evasion variable simplified the multidimensionality of tax evasion attitudes, limiting the depth of our findings. Additionally, cultural differences between Japan and the U.S. affect survey interpretation and responses, potentially impacting data comparability. Third, although VIF calculations indicate no severe multicollinearity issues, the possibility of multicollinearity among closely related institutional variables cannot be ruled out entirely. Despite these limitations, our research provides valid and valuable insights into the tax–morale dynamics across different cultural contexts and scales of analysis. Practical implications These insights are valuable for policymakers, tax authorities, and educators in promoting ethical tax compliance behaviors and bolstering governmental integrity. This study advances theoretical understanding and has practical implications for fostering more effective and culturally sensitive tax compliance strategies. Originality/value This research contributes to the field by combining broad national trends with localized observations. These findings have practical implications for policymakers, tax authorities and educators in promoting ethical tax compliance behaviors and bolstering governmental integrity, fostering more effective tax compliance strategies.