Company taxation and tax spillovers: Separate accounting versus formula apportionment

It is observed in the real world that taxes matter for location decisions and that multinationals shift profits by transfer pricing. The US and Canada use so-called formula apportionment (FA) to tax corporate income, and the EU is debating a switch from separate accounting (SA) to FA. This paper develops a theoretical model that compares basic properties of FA to SA. The focal point of the analysis is how changes in tax rates affect capital formation, input choice, and transfer pricing, as well as on spillovers on tax revenue in other countries. The analysis shows that a move from SA to FA will not eliminate such spillovers and will, in cases identified in the paper, actually aggravate them.     

Income misattribution under formula apportionment

Alternatives to the current system of separate tax accounting, such as the proposed Common Consolidated Corporate Tax Base in Europe, would apportion a firm's worldwide profits using formulas based on the location of employment, capital or sales. This paper offers a new method of evaluating the accuracy of these apportionment rules and the ownership distortions they create. Evidence from European company accounts indicates that apportionment formulas significantly misattribute income, since employment and other factors on which they are based do a very poor job of explaining a firm's profits. For example, the magnitude of property, employment and sales explains less than 22% of the variation in profits between firms, and the prediction estimates from using such a formula exceed half of predicted profits 64% of the time, and exceed twice predicted income 11% of the time. As a result, the use of formulas rewards or punishes international mergers and divestitures by reallocating taxable income between operations in jurisdictions with differing tax rates. The associated ownership distortion is minimized by choosing factor weights to minimize weighted squared prediction errors, for which, based on the European evidence, labor inputs should play little if any role in allocation formulas. But even a distortion-minimizing formula creates large incentives for inefficient ownership reallocation due to the enormous variation in profitability that is unexplained by formulary factors, implying that significant resource allocation costs would accompany European adoption of formulary apportionment methods.     

Symmetric Tax Competition under Formula Apportionment

This paper compares property taxation to a corporate income tax based on formula apportionment in a model where identical countries compete to attract capital. We find that if countries can pair a residence–based capital tax with a property tax (source tax on capital) the tax equilibrium is efficient. In contrast, the use of a 2–factor FA scheme based on sales and capital combined with a residence–based capital tax leads to an inefficient outcome.     

Company tax reform with a water's edge

This paper analyzes the effects of switching from a corporate tax system based on separate accounting (SA) towards a system in which two countries form a formula apportionment (FA) union while a third country sticks to SA (water's edge). Our analysis draws a positive picture on the water's edge regulation. In the short-run, for given tax rates, the transition from SA to FA is likely to reduce profit shifting from the FA union to non-FA tax havens. In a long-run tax competition analysis, we find a negative water's edge externality under FA that tends to be less detrimental than the profit shifting externality under SA and may offset other externalities under FA.     

Equalizing tax bases or tax revenues under tax competition? The role of formula apportionment

This paper contributes to the literature on fiscal equalization and corporate tax competition. The innovation is that we explicitly model multinational enterprises and a corporate tax system that is designed according to formula apportionment. Two main results are obtained. First, in contrast to previous studies we identify cases where tax revenue equalization is better in mitigating detrimental tax competition than tax base equalization. Second, tax base equalization nevertheless has the advantage that it may render tax rates efficient, depending on the shape of the apportionment formula. A pure payroll formula does not ensure efficiency, but a back‐of‐the‐envelope calibration of our model to Canadian provinces suggests that a pure sales formula may be optimal.     

Formula apportionment vs. separate accounting: A private information perspective

In 2002, the European Commission recommended that member countries use formula apportionment procedures to tax multinational companies. This departure from the standard separate accounting (transfer pricing) approach is an attempt to reduce the costs and distortions associated with auditing transfer prices. Unfortunately, apportionment formulas create their own economic distortions and, contrary to popular belief, they do not eliminate distortions due to asymmetric information between the multinational and the national tax authorities. In this paper, I explicitly model the role of private information in two tax competition games: one in which tax liabilities are calculated under formula apportionment and one in which tax liabilities are calculated under separate accounting and transfer prices are audited. Switching to a formula apportionment system affects the after-tax profit of multinationals and the tax revenues paid by both domestic and foreign firms. The direction and magnitude of the changes depend on the accuracy of the auditing technology and non-monotonically on multinational costs. The switch will have different effects on the tax receipts from domestic and foreign firms.     

Formula Apportionment and Transfer Pricing under Oligopolistic Competition

This paper demonstrates that under conditions of imperfect (oligopolistic) competition, a transition from separate accounting (SA) to formula apportionment (FA) does not eliminate the problem of profit shifting via transfer pricing. In particular, if affiliates of a multinational firm face oligopolistic competition, it is beneficial for the multinational to manipulate transfer prices for tax–saving as well as strategic reasons under both FA and SA. The analysis shows that a switch from SA rules to FA rules may actually strengthen profit shifting activities by multinationals.     

THE CHOICE OF APPORTIONMENT FACTORS UNDER FORMULA APPORTIONMENT*

This paper investigates the choice of apportionment factors under a corporate tax system of Formula Apportionment. In contrast to perceived wisdom, we show that the apportionment formula contains both mobile (capital) and immobile (labor) factors, regardless of whether the formula is determined decentrally by jurisdictions or centrally by a central planner. The central planner uses the formula as a corrective instrument to internalize fiscal externalities. We calibrate the model to the European Union and show that replacing the current system of Separate Accounting by Formula Apportionment would substantially increase tax revenue and welfare under both centralized and decentralized formula choices.     

Ad‐valorem tax incidence and after‐tax price adjustments: evidence from Brazilian basic basket food

Abstract Recent empirical work on tax incidence suggests that after‐tax price responses are independent of the tax change direction. This paper investigates asymmetric price responses to ad‐valorem tax changes in the Brazilian food market for 10 goods in 16 states during the period 1994–2008. Our results suggest that when tax rates increase, tax full shifting occurs for 2 of the 10 goods and tax overshifting occurs for one of the 10 goods; the price response to tax cuts is an undershifting for all goods. Moreover, this last result is similar to our estimation of (average) tax incidence (undershifting). We also investigate the short‐term and long‐term relationships between prices and VAT rate parameters and find that prices seem to respond to tax shocks within four months.     

Tax compliance and firm response to electronic sales monitoring

This paper analyzes how firms respond to an Internet of Things technology that reduces significantly the tax authorities' marginal cost of monitoring firm activity. More precisely, we analyze how mandating every restaurant of a single Canadian province to have sales recording modules (SRMs) affects restaurant sales, expenses and profits. We estimate that SRMs increase reported sales by 5.8% to 9.8% on average and that this increase is almost completely offset by an equal increase in expenses, including wages. As a result, the firms' taxable income remains mostly unchanged. Our results suggest that sales tax remittance enforcement at the firm level spills over to other firm stakeholders, such as employees and suppliers. Overall, the one-time cost of the device needed to monitor sales more efficiently is small compared with the recurring benefits for tax authorities.     

The growth effects of tax rates in the OECD

This paper explores the merits of macro‐ and micro‐based tax rate measures within an open economy “fiscal policy and growth” model. Using annual data for 15 OECD countries we find statistically small, non‐robust long‐run growth effects of macro‐based average tax rates on capital income and consumption, but some evidence for average labour income tax effects. Changes in “micro” marginal income tax rates at both the personal and corporate levels yield statistically robust GDP responses of modest size. Both domestic and foreign corporate taxes appear relevant. In general, tax effects on GDP operate largely via factor productivity rather than factor accumulation.     

THE IMPACT OF SODA SALES TAXES ON CONSUMPTION: EVIDENCE FROM SCANNER DATA

Scientific evidence on the effect of sugar consumption on obesity has propelled policy makers in several states across the United States to propose the imposition of a tax on soft drinks sales. In this article, we look at the effect of two tax events: a 5.5% sales tax on soft drinks imposed by the state of Maine in 1991 and a 5% sales tax on soft drinks levied in Ohio in 2003. We investigate this question by using sales data collected by scanner devices in the two states, where soda taxes where enacted as well as on neighboring states. We employ a difference‐in‐difference matching estimator (DIDM) that, in our setting, permits the comparison among treatment and control groups based on brand identity. Results suggest that neither sales tax had a statistically significant impact on the consumption of soft drinks. This finding is robust to several alternative specifications. (JEL D12, H22, C90, L66)     

Optimal Factor Taxation under Wage Bargaining: A Dynamic Perspective

Abstract. We consider the issue of steady-state optimal factor taxation in a Ramsey-type dynamic general equilibrium setting with two distinct distortions: (i) taxes on capital and labour are the only available tax instruments for raising revenues and (ii) labour markets are subject to an inefficiency resulting from wage bargaining. If considered in isolation, the two distortions create conflicting demands on the wage tax, while calling for a zero capital tax. By combining the two distortions, we arrive at the conclusion that both instruments should be used, implying that the zero capital tax result in general is no longer valid under imperfectly competitive labour markets.     

Is the Sky the Limit? Can Japanese Government Bonds Continue to Defy Gravity?

This paper presents a simulation model based on the growth rate, the inflation rate, and the consumption tax rate in the future. Future tax revenues and fiscal expenditures are projected using regression models estimated from past data. The fiscal situation is called unsustainable if the outstanding amount of Japanese government bonds (JGBs) becomes higher than the level of private sector financial assets. We focus on the general account of the central government, which is the source of JGB issues. We find that the higher the economic growth, the more likely it is that the fiscal situation is sustainable. When a larger portion of interest income is reinvested in JGBs, the chance is higher that the fiscal situation is sustainable. Most importantly, raising the consumption tax to 20% guarantees fiscal sustainability in most cases. Our analysis shows that without a consumption tax hike beyond the 10% rate, a fiscal crisis will be almost a certainty, even with a real economic growth rate of 2% despite a shrinking labor force. A reasonably quick hike of the consumption tax, namely a hike by 1% a year, up to 20%, combined with high or moderate economic growth rates, seems to keep the economy out of a fiscal crisis, where a moderate growth rate is defined to be generated by a productivity increase per working‐age population of 1.9%, which was the average during the Koizumi years.     

Investment under Uncertainty, Debt and Taxes

It is common practice in financial derivative valuation to use a discount factor based on the riskless debt rate. But, to what extent is this discount factor appropriate for cash flows emerging in capital budgeting? To answer this question, we introduce a framework for real asset valuation that considers both personal and corporate taxation. We first discuss broad circumstances under which personal taxes do not affect valuation. We show that the appropriate discount rate for equity‐financed flows in a risk‐neutral setting is an equity rate that differs from the riskless debt rate by a tax wedge due to the presence of personal taxation. We extend this result to the valuation of the interest tax shield for exogenous debt policy with default risk. Interest tax shields, which accrue at a net rate corresponding to the difference between the corporate tax rate and a tax rate related to the personal tax rates, can have either positive or negative values. We also provide an illustrative real options application of our valuation approach to the case of an option to delay investment in a project, showing that the application of Black and Scholes formula may be incorrect in presence of personal taxes.     

INCOME TAXES AND THE COMPOSITION OF PAY: EVIDENCE FROM THE BRITISH HOUSEHOLD PANEL SURVEY

According to the standard principal‐agent model, the optimal composition of pay should balance the provision of incentives with the individual demand for insurance. Do income taxes alter this balance? We show that the relative share of Performance‐related pay (PRP), on total pay is reduced by higher average and marginal income taxes. Empirical evidence based on the British Household Panel Survey is consistent with the theoretical predictions of the tax–augmented principal‐agent model. Our estimates suggest that a 10% reduction in the marginal income tax rate, holding the average tax rate constant, increases the share of PRP in total pay by 2.25–3.02%, depending on the empirical specification. Similarly, a 10% reduction in the average income tax rate, holding the marginal tax rate constant, increases the share of PRP in total pay by 5.10–5.27%.     

REEXAMINING THE BORDER TAX EFFECT: A CASE STUDY OF WASHINGTON STATE

Without an income tax, Washington State relies heavily upon its sales tax revenue to fund public goods and services. Bordering Idaho and especially Oregon, where the sales tax is substantially lower, the juxtaposition of the different tax structures generates the border tax effect in Washington's border counties. Controlling for unobservable county‐specific characteristics and spatial autocorrelation, we find that the price elasticity generated by the sales tax discrepancy over the years 1992–2006 is ?3.11. We estimate that elimination of the sales tax differential between Washington and its neighboring states would generate tax revenue in excess of $145 million at the state level and over $21 million at the county level in border counties. (JEL C23, D12, E62, H71)     

Regional effects of export tax rebate on exporting firms: Evidence from China

This paper extends a model from 2003 to separate the direct and indirect impact of an export tax rebate on the intensive margin of firm‐level export sales at the subnational level. The direct impact of the rebate is associated with a reduction of an exporting firm's variable costs, while the indirect impact manifests itself through higher regional wages as a result of increased demand for local labor. First, the empirical results imply that a 1 percent rise in the export tax rebate rate increases the export sales among continuing exporters by 0.2 percent through the direct channel. Second, through the indirect channel, a 1 percent difference in the regional rebate causes a 0.02 percent difference in exporters' sales growth. Both effects are statistically significant, and are consistent with the model's predictions.     

The demand for tax haven operations

What types of firms establish tax haven operations, and what purposes do these operations serve? Analysis of affiliate-level data for American firms indicates that larger, more international firms, and those with extensive intrafirm trade and high R and D intensities, are the most likely to use tax havens. Tax haven operations facilitate tax avoidance both by permitting firms to allocate taxable income away from high-tax jurisdictions and by reducing the burden of home country taxation of foreign income. The evidence suggests that the primary use of affiliates in larger tax haven countries is to reallocate taxable income, whereas the primary use of affiliates in smaller tax haven countries is to facilitate deferral of U.S. taxation of foreign income. Firms with sizeable foreign operations benefit the most from using tax havens, an effect that can be evaluated by using foreign economic growth rates as instruments for firm-level growth of foreign investment outside of tax havens. One percent greater sales and investment growth in nearby non-haven countries is associated with a 1.5 to 2% greater likelihood of establishing a tax haven operation.     

PROGRESSIVITY AND REDISTRIBUTION IN NON‐REVENUE NEUTRAL TAX REFORMS: THE LEVEL AND DISTANCE EFFECTS

Kakwani and Reynolds–Smolensky indices are used in the literature to measure the progressivity and redistributive capacity of taxes. These indices may, however, show some limits when used to make normative assessments about non‐revenue neutral tax reforms. Two approaches have traditionally been taken to overcome this problem. The first of these consists of comparing after‐tax income distributions through generalized Lorenz (concentration) curves. The second approach is based on the decomposition of changes in the Reynolds–Smolensky index into changes in the average tax rate and variations in progressivity. Nonetheless, this decomposition between the average tax rate and progressivity may be further exploited to obtain some information that can be relevant to assess tax reforms. The main aim of this study is to draw up some indicators that can be useful to quantify the effects of non‐revenue neutral tax reforms. These indicators are used to investigate the last personal income tax reforms that have taken place in Spain.