ABSTRACT

The complaint that developing countries raise inadequate tax revenues and have poor administrative systems is not new. Reforms to ‘strengthen tax administration’ and bolster domestic revenues are characteristic of IMF programmes across the developing world, in part verified by the rapid uptake of VAT in Africa and Asia since the late 1980s to the present day (see Ebrill et al. 2001). These issues have been prominent in Mozambique since the pursuit of structural adjustment and stabilization in the mid-1980s. To date the country has undergone various rounds of taxation reform; however, since 1993 these have yet to translate into a sustained enhanced domestic revenue position. The level of the tax ratio, defined as total tax revenues divided by nominal GDP, was around 11 per cent as at 2005 and has been the object of substantial concern particularly in light of external financing (via foreign aid) of the budget at well over 10 per cent of GDP. Despite these criticisms, it is helpful to recall the empirical phenomenon

described by Wagner’s Law. This tells us that the relative size of government in the economy tends to increase with average economic prosperity, which also means that the relative volume of domestic revenues in low income countries tends to be smaller than that raised in richer countries. It remains uncertain, however, to what extent and in what ways domestic taxation possibilities are constrained by empirical conditions in developing countries. This is an important policy issue as a robust analysis would provide an analytical foundation to assess the adequacy of observed tax ratios and/or the realism of fiscal policy goals. The principal objective of this chapter is to evaluate the aggregate perfor-

mance of Mozambique’s tax system against experiences of other developing countries. In the first instance, the nature of constraints to tax revenue generation is investigated at the cross-country level. In turn, this framework is applied to the specific case of Mozambique. Given the explicit focus on aggregate revenue levels, it must be highlighted that the administration of taxes and the fine print of tax policy are not explored here. This is not to dismiss their importance, however, particularly as administration and policy

choices are fundamental determinants of the equity and efficiency outcomes of a given taxation system. By way of structure, Section 2 provides a brief summary of critical per-

spectives on the Mozambican tax system. This leads, in Section 3, to an analysis of the cross-country evidence for the relationship between developmental conditions and overall tax performance via elaboration of an econometric model. Consistent with previous research (e.g. Teera and Hudson 2004), tax ratios in developing countries are found to be strongly related to ‘deep’ economic and institutional variables. In Section 4 the model is applied to Mozambique, permitting calculation of the expected tax effort based on economic and institutional conditions. Section 5 concludes.