A macroeconomics perspective on international coordination in sales taxes

Abstract

This article investigates how international coordination vis-à-vis sales tax policies affects the welfare of participating countries. A country's tax policies have asymmetric effects on the pricing behaviors of domestic and overseas producers. International cooperation endogenizes the externality that improves the purchasing power of foreign residents, but at the cost of its own residents’ work efforts. The first-best taxes are lower than in the noncooperative case. When world welfare is utilitarian, smaller economies may experience welfare losses from cooperation under the weak income effect of sales tax. We propose a coordinated tax rule that all countries agree to employ.

Keywords:

• international tax policy
• tax coordination
• cooperation
• welfare
• marginal utility from income

JEL Classifications:

• F42
• H21

Notes

1. Dixit (1985, 314) classifies the objectives of tax policy into three categories: (i) correcting externalities and distortions, (ii) raising revenue for government expenditure, and (iii) redistributing income. He asserts that the presence of international trade does not alter these objectives. This article focuses on objective (i). In the context of an open economy, Keen and Lahiri (1998) examine the optimality of non-cooperative taxes in a model without revenue requirements or demands for public goods. The comprehensive analysis by Lockwood (2001) also allows for a similar environment as a special case. Following this line, we simply omit the provision of public goods and assume that a country's excess government revenue is equally distributed to its residents.

2. For simplicity, we assume the unit elasticity of substitution between home and foreign goods and a balanced initial current account, which leads to complete risk sharing in the absence of further price adjustments. Benigno and Benigno (2003) develop a model in which this elasticity differs from unity and analyze monetary cooperation.

3. They illustrate that the price elasticity of the offer curve is increasing in the consumption share if the elasticity of substitution is greater than one, and decreasing in the endowment share. Because these two shares are both increasing in the size of the trading block, which corresponds to the country size in our model, the net effect of block size on the elasticity is indeterminate. Our home and foreign countries correspond to their trading blocks, and our assumption of Cobb–Douglas preference and complete specialization in production leads to unit elasticity in their model.

4. This specification of the labor market implies that the labor force is also immobile within a country. Along with the assumption of no capital, this leads sales taxes to become analytically equivalent to income taxes. This simplification does not essentially alter the government's incentives for attaining optimal markups from a national viewpoint. One can obtain a result equivalent to ours by assuming that households supply labor forces to all the firms within the same country. When capital/savings exists, the taxation on financial assets can potentially cause additional distortion.

5. Introducing hours worked or leisure in the utility function is conventional in the either closed- or open-macroeconomics literature. See, for instance, Kydland and Prescott (1991), Rotemberg and Woodford (1995), or Obstfeld and Rogoff (1995).

6. Mintz and Tulkens (1986) argue that destination-based taxation requires border monitoring, which is expensive and sometimes even impossible. This destination-based taxation is discussed in Section 5.

7. Note that lim _w → 1τ ^C  = 1−Θn is not equal to τ ^N . The tax rate is lim _w → 1τ ^C if both home and foreign governments cooperatively maximize only the home welfare; i.e. when not only τ ^C but also τ ^*C is set to maximize the home welfare.

8. Kocherlakota (1996) states that ‘a vast majority of economists believe that values for [the coefficient of relative risk aversion] above ten … imply highly implausible behavior on the part of individuals’. Prescott (1986) suggests that a reasonable estimate is not too far from 1. Cooley and Prescott (1995) argue that this parameter is among the most difficult to pin down. Krusel, Quadrini, and Ríos-Rull (1996) and Díaz, Pijoan-Mas, and Ríos-Rull (2003) use ρ = 2. Therefore, we examine the various numbers for ρ.

9. One potential critique for this illustration is that WTO limits explicit assistance to exports. From a practical aspect, as Leahy and Neary (2001) point out, governments might have to subsidize other tax bases such as investment rather than sales, although this issue is beyond the scope of this article.