ABSTRACT. I analyze the dynamic effects of tax competition on public budget deficits. I find that stronger tax competition leads to a fiscal deficit bias at the early stages of financial liberalization. When countries differ in terms of capital mobility, further liberalization leads to external imbalances and diverging fiscal deficits while corporate tax rates converge. Consistent with theory, I find that stronger tax competition increases deficits in a sample of OECD countries, controlling for tax revenues and other standard determinants of fiscal deficits.

ABSTRACT. This paper is concerned with the effect of interjurisdictional competition on commercial land-use. A theoretical analysis explores the trade off faced by a local government when deciding about the share of land that is available for commercial use. In this analysis, land available for commercial land-use stimulates local business activity and attracts mobile factors but it also exerts adverse effects on the quality of life. In an environment with tax competition, the land use is found to be distorted by a fiscal incentive to expand commercial land-use. The theoretical prediction of higher commercial land-use in the presence of tax competition is explored empirically using a large dataset of German municipalities. In order to identify differences in the competitive environment, I exploit institutional characteristics of the system of fiscal equalization among the German municipalities. The empirical results confirm that the share of land dedicated to commercial land-use tends to be higher in municipalities that are exposed to more intense tax competition. The paper ends with a discussion of policy options to remove the disincentive for expanding commercial land-use.

ABSTRACT. In this paper, fiscal externalities are analyzed in a model of tax competition with three distinguishing features: (i) the tax base is shared by two levels of government; (ii) taxes are levied on an ad valorem basis; (iii) although the tax base is mobile across lower-level jurisdictions, these jurisdictions levy their taxes on a residence basis. We find that either under- or overtaxation occurs, under similar conditions as in Keen and Kotsogiannis (2002, 2004). However, the neat trade-off between positive horizontal externalities and negative vertical externalities breaks down entirely in our ad valorem setting. Precisely because of this ambiguity, decentralising the unitary outcome via revenue equalisation becomes far more complex than under unit taxation. Only when the marginal valuation of public provision is on par with private consumption, can we replicate the clear-cut, efficiency-enhancing equalisation formulas given by Kotsogiannis (2010) and Bucovetsky and Smart (2006).

ABSTRACT. We study the effects of the construction-related temporary windfall revenues obtained by Spanish municipalities during the years of the housing boom (1995-2007) on local government budgets during the boom and during the subsequent bust (2008-11). We assess whether local governments spent too much of these revenues or not and, in case they spent too much, if they were able or not to deal with the effects of the accompanying shortfalls that emerged during the crisis. Our results show, first, that just a very small proportion of the boom windfalls were saved for the bad times. Second, we show that most of these shortfalls went to cuts in capital spending and tax hikes, but also to deficit increases. Finally, we show that both over-spending during the boom and the difficulties in adjusting the budget during the bust are higher in places with low government quality.

ABSTRACT. It is often said that “necessity is the mother of invention”. In this article, we assess whether this applies also to the design of public authorities’ service provision. Particularly, we evaluate whether revenue scarcity – as an indicator of fiscal stress – induces government outsourcing. In contrast to previous studies, we exploit arguably exogenous variation in local government revenue across time and space to estimate causal effects of revenue scarcity on outsourcing. Using data from Norwegian local governments covering the period 1995-2012, our main results indicate that an exogenous decrease in local government revenues indeed causes more outsourcing of both infrastructure and support services.

ABSTRACT. We analyze the existence of a flypaper effect in local public finance. Kinks in the transfer schedule in the Canton of Vaud in Switzerland alow us to apply a Regression Kink Design and to estimate the causal impact of transfers on municipal expenditures and tax rates. Consistent with the literature, we find robust evidence that ``money sticks where it lands'' i.e. transfers have a positive local average treatment effect (LATE) on municipal expenditures, while leaving the revenue side unchanged. Moreover, by studying the heterogeneity of policy responses to transfers, we observe a ``double'' flypaper effect. It appears that money sticks even more where it used to land in the past. The second layer of the effect can be used to explain the flypaper effect described in the literature.

ABSTRACT. This paper surveys state and evolution of GDP per capita in 281 regions of OECD countries for the time period 1995 – 2013. It puts a special focus on the disparities between the regions. These can be substantial: In 2013, GDP per capita of the least and most developed region varied by a factor of roughly ten. Using standard inequality measures like the coefficient of variation or the Gini coefficient, it is found that inequality has been decreasing between countries, while within-country disparities have often widened. Furthermore, transition matrices reveal that mobility within the distribution over time is higher in countries with larger degrees of fiscal decentralisation. This suggests that decentralisation allows regions to “take matters into their own hands”. Implications of other factors that correlate with the level of economic development are also discussed.

ABSTRACT. Fiscal decentralisation can lead to a more efficient provision of local public goods and services and promote a better match between policies and citizens’ preferences. At the same time, however, there are concerns about the possibility of all regions gaining from more autonomy. Decentralisation may not lift all boats, “poor” regions could lose competitiveness with respect to better endowed regions, thus increasing regional disparity. The present work investigates the relationship between fiscal decentralisation and regional inequality, within countries. Particular attention is payed to the different channels through which decentralisation can affect disparity: taxing powers, spending autonomy, and vertical fiscal imbalance. The empirical analysis conducted on a sample of 30 OECD countries for the period 1995-2011, suggests that a balanced fiscal structure, where local spending is mainly financed by local taxation, does actually reduce regional disparity, by providing an incentive to better use local resources and implement policies that favour economic development.

ABSTRACT. This paper investigates the relationship between fiscal decentralisation and economy-wide disposable income inequality. Drawing on a dataset of up to 20 OECD countries over a period from 1996 to 2011, fixed effects and LIML regressions are run, relating several indicators of national income inequality and a wide array of fiscal decentralisation indicators. The results indicate a weak, inequality-reducing relationship between decentralisation and income inequality, as measured by the Gini coefficient, but the effect is rather small and unstable across specifications. Fine-graining the analysis by using income percentile ratios, in turn, produces more significant and stable results. It shows that the effects of fiscal decentralisation are not the same along the income distribution. While decentralisation tends to reduce income inequality between high income groups and the median, it is linked to a divergence of low income groups from the median, notably via sub-central tax autonomy. Transfers between levels of government also tend to increase the gap between lower and middle incomes. Interpreting these effects jointly, it seems that the group mainly benefiting from fiscal decentralisation are middle income earners. Finally, some insights on decentralisation and regional income inequality are presented. At first sight, fiscal decentralisation does not seem to be linked with income sorting in large jurisdictions, but a more fine-grained analysis is required to answer this question.

ABSTRACT. This paper describes an empirical analysis to measure the cyclical properties of intergovernmental transfers (or grants). Earlier within-country research suggests that several transfer and equalisation systems have a pro-cyclical impact on sub-central budgets. Modelling a fiscal policy reaction function the paper tests whether the transfers systems in OECD countries are pro-or counter-cyclical, i.e. whether they offset cyclical fluctuations of sub-central economies or, on the contrary, exacerbate them. Regression results suggest that transfer systems tend to be pro-cyclical in general and in more than half of OECD countries and tend to destabilise sub-central budgets. Transfer pro-cyclicality may be the result of several factors: Transfer spending is often a share of central government tax revenue which itself tends to fluctuate with the cycle. Moreover, many grants are matching sub-central spending and hence tend to exacerbate fluctuations of that sub-central spending. Pro-cyclical grants could partly explain often observed pro-cyclicality of sub-central government fiscal policy.

ABSTRACT. Fiscal equalisation is a transfer mechanism that allows different jurisdictions to provide their citizens with similar public services at similar tax rates despite differences in economic wealth. Across OECD countries, equalisation has a strong redistributive effect: on average it reduces pre-equalisation disparities by more than two-thirds and, in some countries, to virtually nil. After equalisation, tax and public service levels are much more evenly distributed across a country than GDP or household income. Strong equalisation comes at a cost. Strong equalisation may undermine local and regional development efforts or may prevent firms and households from migrating towards more productive jurisdictions. Equalisation systems also tend to be pro-cyclical. Moreover, large equalisation systems can undermine fiscal discipline at the sub-central level. Equalisation arrangements should rely on revenue and needs indicators that cannot be manipulated by sub-central jurisdictions, and they should be transparent about donors and recipients.

ABSTRACT. This paper reassesses the role of taxation in profit shifting of multinationals. Using a large European sample of multinationals and their subsidiaries, we exploit multiple tax rate reforms in a difference-in-differences setting to examine when and why tax rate changes affect profit shifting. Our results suggest that lowering the statutory corporate tax rate in the country of the parent company (i.e. inbound profit shifting) incentivizes subsidiaries to shift back profits to the parent country. A one standard deviation decrease in the change in the tax difference (2.7 percentage points) reduces profit (i.e. tax reform response) by 7% in the host country. This effect is substantially stronger when the change in the tax difference is large enough to reverse the sign of the tax rate difference. In addition, profit shifting responses to tax reforms in the parent country are muted by economic growth in the host country of the subsidiary but are strengthened by strong institutions in the host country. We characterize tax changes that occur in the host country as outbound profit shifting. In this analysis, we find that the tax reform is effective only when the change in tax difference between host and parent countries is large enough to reverse the sign of the tax difference. Our findings are important for policymakers when designing tax policy that intends to curb international tax-motivated profit shifting.

ABSTRACT. The issue of base erosion and profit-shifting (BEPS) has been on the international policy agenda for three years now. A key element in the discussion are strategies of multinationals using intra-group interest and royalty payments as well as transfer pricing to reallocate profits within the group in a tax minimizing manner. In recent years, anti-avoidance regulations have been introduced to limit these cross-border shifting activities. Existing evidence looks at the effectiveness of these regulations separately. The idea of this paper is to analyse the interdependence between different anti-avoidance regulations in place. Our empirical results confirm existing findings on the tax sensitivity of EBIT and the reduction of this sensitivity due to stricter transfer pricing regulations. In addition our results suggest that the positive impact of transfer pricing regulations is strongly mitigated if strict thin capitalization rules apply.

ABSTRACT. This study analyzes where large multinational companies allocate the ownership of their U.S. trademarks and whether these location choices are driven by tax considerations. Matching the U.S. Patent and Trademark Office’s (USPTO) register with group structures of large MNEs from the U.S. (S&P 500) as well as from Europe (STOXX Europe 600), we describe and explain the geographic origin of U.S. trademark registrations submitted by these global companies. Our analysis suggests that tax considerations have a  rather limited influence on the geographical allocation of trademarks. However, if trademarks are assigned to affiliates located neither in a firm’s home country nor the U.S., we find a significant influence of corporate tax rates and U.S. withholding taxes.

ABSTRACT. This paper suggests a model-based way of measuring cross-country differences in preferences based solely on national accounts data. Using a representative-agent model, the parameter capturing the relative utility of leisure is backed in an accounting-like procedure. Empirical data reveals a strong correlation between people's self-reported attitudes concerning the relative importance of work and leisure and this model-based measure of preferences, also after controlling for a number of labor market indicators. The evidence is consistent with cross-country differences in the model-based preference measure being explained, at least partly, by differences in actual underlying preferences. Furthermore, when investigating micro level data for descendants of U.S. immigrants, the measure of preferences for leisure associated with an individual's country of ancestry exhibits predictive power for labor supply. Finally, the responsiveness of policy choices to culture is illustrated by studying the relationship between labor taxation and preferences for leisure. Taxes are found to be higher in countries where the preference for leisure is weaker. Theory suggests this is partly due to lower distortionary effects.

ABSTRACT. My job market paper examines the effect of a tax compliance campaign on tax payments in 2013 in Greece. The Tax Compliance Unit (TCU) of the Department of Public Revenues of the Ministry of Finance of Greece used phone calls and e-mails to remind tax delinquent individuals and firms of their upcoming payment deadlines. However, the TCU was able to communicate with only a fraction of the delinquent population, giving rise to a natural experiment. Using administrative individual- and firm-level information, I study the effect of the reminder - a neutral, non-deterrent behavioral nudge - on payments of tax liabilities. These include income tax and value-added tax liabilities in 2013 and past years' tax arrears by firms and individuals. To disentangle the causal effect of the payment reminder I employ two quasi-experimental methods; a regression discontinuity design (RDD) and matching. I take advantage of TCU's phone call treatment assignment rule, which leads to a discontinuous relationship between firms and individuals with liability levels exceeding 500 Euro and the probability they received a reminder phone call. This discontinuity generates plausibly exogenous variation in phone call assignment which I exploit to identify the causal effect of the reminder on payments of tax liabilities. Specifically, there is a jump in the conditional mean of the treatment variable at the 500 Euro cutoff that provides identifying variation. This allows me to estimate the effect of the phone call and the emails as if they were randomly assigned in a neighborhood of the 500 Euro cutoff. I identify the local average treatment effect (LATE) of the deadline reminders on payments using a fuzzy RDD estimator. The fuzzy RD method has excellent internal validity but its identifying power is only local, around the cutoff. To recover the global effect of the reminder phone call over the entire distribution of liabilities (not just within a neighborhood of the 500 Euro cutoff), I rely on matching methods. The TCU reminded firms and individuals with liabilities greater than 500 Euro but not in their entirety. The existence of cross-overs in the study population, firms and individuals with liabilities greater than 500 Euro that were not treated, provides a very large number of untreated observations. I am able to construct control groups that are nearly identical to the treated ones using matching methods. Matching mimics random assignment through the ex-post construction of a control group. This allows me to give a causal interpretation of the estimated treatment effects for the entire distribution of liabilities as well, not just around the 500 Euro cutoff.

ABSTRACT. We study a model in which agents differ in their productive skills and their preferences over labor time/consumption bundles. We assume there is a poverty line, that is, there is a minimal level of consumption below which society finds it unacceptable to let people live. To avoid conflict with individual well-being, we capture the anti-poverty project be requiring redistribution to take place between agents on both sides of the poverty line provided they have the same labor time. We combine this requirement with efficiency and robustness requirements to derive social preferences. Maximizing these preferences under incentive compatibility constraints yields the following evaluation criterion: labor income tax schemes should minimize the labor time required to reach the poverty line. We apply this criterion to tax schemes of European countries and the US.