What is absent in the analysis of Ladd - surprisingly for who dedicates the book to fiscal federalism issues - is precisely to link the topic of tax expenditure with the regional question. To know and control where taxing power and spending are exercised, is important for the measurement of territorial cost and benefits of government action; an outstanding issue in fiscal federalism studies and discussions. Knowledge of tax expenditure can facilitate a bigger budget control of governments and improve transparency for voters of the federation. Testing fiscal correspondence principle and regional transfer systems among governments are a central question in fiscal federalism.
To federalize decisions on tax expenditure can constitute an institutional mechanism to generate incentives for a bigger control of their growth. For example, in case of the national government, federal concessions of tax expenditures that benefit certain sectors, will impact on the regional distribution of tax revenues when revenues of taxes that admit those tax expenditures decisions are part of a tax revenue sharing system. In turn, those concessions modify the relative prices of goods and services among regions, inducing reassignments of factors that naturally separate solutions from relative cost of regional economies, encouraging relocations of productive activities. They also generate justified political reactions from regions that contribute to the common pool of resources in benefit of the regions promoted, affecting horizontal tax fairness.37 Policy decisions on tax expenditure can also induce sub-national taxation growth or, on the contrary, they can generate harmful tax competition (tax wars) among states that should be necessary to avoid or minimize, so transparency and calculation of sub-national tax expenditure are also important. Tax harmonization should be necessary for an efficient performance of the federative common market. 38
The "Tax Burden" concept
Tax Burden (TB) is the coefficient that relates total revenue collected by Government (T) with national income (NY). In the usual calculation of this coefficient, the numerator includes the whole taxes that engross the National Treasure - in the version referred to the National level of government - or all revenues collected by the three levels of governments (National, Provinces/States and Municipalities) - in its Global version -. Such measurements do not compute the value of revenue lost by decisions on tax expenditures, as long as government tax administrations and budget accounting offices only register the effective inflows of money to treasures. As it was analyzed before, when money does not come to the treasury there is no reason to worry about dealing to legal control, because if the objective were the control of destination of the money paid by taxpayers, there is no money to trace out. Also it could be interpreted that from the macroeconomic point of view neither would be important if the objective were the control of economic impacts of fiscal policy. This vision derives to the neutral effect of tax expenditure at global level, because there is no impact to measure for the not collected revenues.
That is, without Tax Expenditure (TE), Tax Burden (TB) is:
37 38 These regional aspects of efficiency and fairness is treated in Piffano (2004a and 2005), op. cit. The Brazilian experience with the ICMS (the sub-national VAT) and other indirect taxes levy by states and municipalities are instructive about tax war between states with great distortion effects on internal relative prices. Read “harmful tax competition” or “distortion-bias tax competition” not “Tiebout-beneficial tax competition” (see Piffano, 2004b)