International transfer pricing and income shifting in developing countries: evidence from Ghana

Abstract

Purpose The purpose of this paper is to examine the potential use of international transfer pricing (ITP) as an income shifting mechanism by multinational corporations (MNCs) in developing countries. The paper postulates that income shifting through ITP is likely to be more pronounced in developing countries where weak institutions are present. Design/methodology/approach The paper uses a unique unbalanced panel data of 18 companies listed on the Ghana Stock Exchange covering the period of nine years (2008–2016), to investigate whether MNCs use ITP to shift income out of the country. The comparison is made using an indirect approach where performance (e.g. profit before tax) and post-performance measures (e.g. dividend payment) are used for an equal number of foreign and local companies. The empirical analyses include t-tests, pooled and random effects logistic regressions. Findings The results show significant differences between foreign controlled entities (FCEs) and Ghanaian controlled entities in terms of capability, profitability and dividend distribution. Since there is a positive between these measures, the results do not suggest possible income shifting by FCEs through ITP. Research limitations/implications This paper uses an indirect method of investigating income shifting among MNCs. For future studies, a more direct method can be adopted by examining import and export prices of specific products for both foreign and domestic firms. Originality/value The study investigates the possibility of income shifting arising from ITP practices among multinationals in developing countries. To the best of the authors’ knowledge, this paper is the first in this regard. Thus, the study contributes to the transfer pricing and income shifting literature by providing evidence from a developing country.

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.