Ownership structure and corporate tax avoidance: a structured literature review on archival research

Abstract

Purpose To the best of the author’s knowledge, the author conducts the first detailed review on the impact of ownership variables on corporate tax avoidance, based on 69 archival studies over the two last decades. Design/methodology/approach Referring to an agency-theoretical framework, the author differentiates between six categories of ownership (institutional, state, family, foreign, managerial and cross-ownership/ownership concentration). The author also includes research on ownership proxies as moderators of other determinants of tax avoidance. Findings The review indicates that most research refers to institutional, state and family ownership. Moreover, except for state ownership, no clear tendencies on the impact of included ownership types can be found in line with the author’s agency-theoretical framework. Research limitations/implications Regarding research recommendations, among others, the author stresses the urgent need for recognizing heterogeneity within and interactions between ownership proxies. Researchers should also properly address endogeneity concerns by advanced econometric models (e.g. by the difference-in-difference approach). Practical implications As international standard setters have implemented massive reform initiatives on both tax avoidance and corporate governance, this literature review underlines the huge interaction between those topics. Firms should carefully analyze their ownership structure and change their tax planning due to owners' individual tax preferences. Originality/value This analysis makes useful contributions to prior research by focusing on six categories of ownership and their impact on tax avoidance in (multinational) firms and moderating effects. The author provides a detailed overview about current archival research and likes to guide researchers to focus on ownership heterogeneity and endogeneity concerns.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.