Design and Estimation of a Taxpayer Compliance Model from a Behavioral Economics Perspective Considering Economic and Situational Factors

Abstract

This study was conducted with the aim of designing and estimating a model of taxpayer compliance from the perspective of behavioral economics, taking into account economic and situational factors. In terms of purpose, the present study is applied, and in terms of nature, it is descriptive–analytical and of a survey type. In the first stage, in order to identify and level the factors affecting taxpayer compliance, the Interpretive Structural Modeling (ISM) method was employed using the opinions of experts in the field of taxation. At this stage, eight main components—including individual psychological factors, social and cultural norms, institutional and governmental trust, behavioral incentives and penalties, behavioral design of the tax system, financial and tax literacy, economic and situational factors, and experience and interaction with the tax system—were identified, and their internal relationships were analyzed through the Structural Self-Interaction Matrix (SSIM), the reachability matrix, and the final reachability matrix. The leveling results indicated that some variables play a driving and foundational role, while others are more influenced by other factors. The MICMAC (cross-impact matrix multiplication applied to classification) analysis also confirmed that infrastructural components such as the behavioral design of the tax system and economic and situational factors fall into the category of independent and driving variables and possess the highest level of influence within the system, whereas variables such as individual psychological factors and experience of interaction with the tax system are primarily categorized as dependent variables. In the second stage, in order to fit and validate the extracted model, Structural Equation Modeling (SEM) with a Partial Least Squares (PLS-SEM) approach was applied. The data for this section were collected through a questionnaire with the participation of 317 experts from the National Tax Administration. The results of the multicollinearity test indicated that the Variance Inflation Factor (VIF) values for all constructs were below the threshold level, suggesting no multicollinearity issue. Additionally, reliability indices including Cronbach’s alpha and composite reliability for all variables were above 0.7, and the Average Variance Extracted (AVE) was above 0.5, indicating acceptable reliability and convergent validity of the measurement instrument. Discriminant validity was also confirmed using the Fornell–Larcker criterion and the Heterotrait–Monotrait ratio (HTMT). The coefficient of determination (R²) for endogenous constructs was evaluated at an acceptable level, and the effect size and predictive power indices indicated good model fit and satisfactory predictive capability. Overall, the findings of the study indicate that taxpayer compliance is a multidimensional phenomenon that, in addition to economic variables, is significantly influenced by behavioral, institutional, and system design mechanisms. Improving taxpayer compliance therefore requires the adoption of an integrated approach grounded in behavioral economics, along with the strengthening of institutional and technological infrastructures.