TEXT FOR DISCUSSION Nº 500 THE ICMS FISCAL WAR: WHO GAINS AND WHO LOSES * Ricardo Varsano ** Rio de Janeiro, July 1977 *A previous version of this text was presented at the International Seminar for Decentralized Industrial Policies, sponsored by ECLAC and IPEA and held in Brasilia on the 11th and 12th of November 1996. ** General Coordinator of sector studies of IPEA’s Research Department. IPEA is a public foundation linked to the Ministry of Planning and Budget, whose purposes are: to assist the minister in the elaboration and monitoring of the economic policy and to provide activities of applied economic research in the fiscal, financial, external and sector development areas. President Fernando Rezende Directors Claude Monteiro Considera Luis Fernando Tironi Gustavo Maia Gomes Mariano de Matos Macedo Luiz Antonio de Souza Cordeiro Murilo Lôbo TEXT FOR DISCUSSION aims at disseminating the results of studies developed directly or indirectly by IPEA, as well as work considered relevant for dissemination by the Institute, in order to inform specialized professionals and gather suggestions. ISSN 1415-4765 EDITORIAL Rio de Janeiro – RJ Av. President Antonio Carlos, 51 – 14º andar – CEP 20020-010 Telefax: (021) 220-5533 email: firstname.lastname@example.org Brasilia – DF SBS Q. 1 Bl. J, Ed. BNDES – 10º andar – CEP 70076-900 Telefax: (061) 315-5314 email: email@example.com © IPEA, 1998 This text may be reproduced, provided the source is mentioned. Reproductions for commercial purposes are forbidden. EXECUTIVE SUMMARY ABSTRACT 1 – THE LAW LOSES AND SO DOES THE FEDERATION……………………………….1 2 – CONDITIONS THAT JUSTIFY THE GRANTING OF INCENTIVES……………….2 3 – THE NATIONAL PERSPECTIVE: THE COUNTRY LOSES……………………….4 4 – THE STATE GOVERNMENT’S PERSPECTIVE: THE STATE (ALMOST ALWAYS) GAINS (FOR SOME TIME) ………………………………………………………….6 5- FISCAL WAR AND ECONOMIC SIGNALING …………………………………………..8 6 – SUMMARY OF THE MAIN ARGUMENTS AND CONCLUSIONS………………..11 ABSTRACT Brazilian states have been competing among themselves to harbor industrial plants by granting incentives based on the state value added tax — the ICMS — which, though forbidden by law, show an intensity incompatible with states current financial conditions. This article proposes a set of conditions that should be met to justify the provision of incentives and shows that, from a national perspective, they are seldom met. However, from the point of view of a particular state government, the same conditions are satisfied in most cases, justifying its engagement in the fiscal war, which dynamics is quite perverse: public finances — and, as a consequence, local production conditions — deteriorate in all states and fiscal incentives, due to generalization, loose their power to attract investment. The ultimate winners of the fiscal war are the financially stronger states, which are able to bear the revenue loss and yet provide reasonable production conditions. It’s argued that divergence between national and state objectives and between the best state development strategy in the short and in the long runs could be smoothed out by changing the border tax adjustments adopted for treatment of interstate trade flows. If the destination principle were implemented, incentives to participate in the fiscal war would be practically eliminated. 1 – THE LAW LOSES AND SO DOES THE FEDERATION Complementary Law nº 87, dated 13th September 1996, by regulating the ICMS — tax on transactions related to the circulation of goods and on interstate and intercity transportation and communication services — besides filling a gap that existed since the promulgation of the 1988 Constitution, introduced important changes to the economic characteristics of this tax, which is the main source of revenue for Brazilian states. First, the law brought it closer to the theoretical concept of value added tax (VAT), when establishing that all productive inputs will generate credit of the tax previously paid by the purchaser. Secondly, it made the ICMS — which was a gross product type VAT — resemble a consumption type VAT, by allowing taxpayers a credit related to the tax paid on goods that they incorporate to their permanent assets.1 And thirdly, it adopted, at last, the destination principle in foreign trade, unburdening exports of primary and semi-processed industrialized products — which were still taxed — and ensuring the utilization of tax credits accumulated by the exporter. The bill of law that created Complementary Law nº 87/96 also intended to strengthen already existing legal provisions that seek to restrain the fiscal war among states. However, due to the resistance demonstrated by some governors to give up the use of incentives linked to the ICMS as tools for their industrial policies, a political agreement was required between the Federal Senate and the President of the Republic in order to ensure the approval of the improvements contained in the bill of law, according to which the Senate would approve the bill without alterations and the President would veto the provisions that addressed the granting of incentives in the scope of the ICMS. Such agreement did not create a void in the legislation in this respect, since, because the vetoes totally removed from the new law the provisions that would regulate the subject, Complementary Law nº 24 of 7th January 1975 remains in force. The fiscal war goes on regardless of Complementary Law nº 24/75, which prohibits the granting of exemptions and other ICMS-based incentives, except when provided for in agreements made at meetings of the Council for Financial Policies (Confaz) that congregates all the states and the Federal District. The law determines that the approval for the award of a benefit depends on unanimous decision of the represented states and provides for penalties in case of non-compliance with its provisions. The enforcement of this law alone – not even the reinforcement contained in vetoed articles of the Complementary Law nº 87/96 would be needed – would be enough to end the fiscal wars. The fact is that the fiscal war continues, even though the law has existed for more than 20 years. The law is not complied with and nobody takes the initiative to demand that the sanctions provided be imposed, despite the fact that the expression “fiscal war” is always used with a negative connotation, as something harmful. The disrespect towards the Law by government officials themselves is certainly a loss for the nation. The fiscal war is, as the name indicates, a situation of conflict in the Federation. The federate entity that gains — when, in fact, there is a gain — imposes, in the majority of cases, a loss to one or some of the others, since a war is rarely a positive sum game. Federalism, which is a relationship of cooperation among government units, is shaken. The Federation – so dear to Brazilians to the extent that the Constitution contains a clause preventing its abolition – also loses. It may be argued that there are good reasons for the negative connotation given to the expression “fiscal war”; but that, from a state government’s perspective, there are clear economic incentives, in addition to political ones, for the war to continue. While they exist, the law will hardly be obeyed. Section 2 proposes conditions that should be fulfilled to merit the granting of incentives. Section 3 discusses, based on the considerations of the previous section, that from the national perspective the fiscal war is unacceptable. The country loses in this war. Section 4 discusses the issue from a state government’s perspective, based on the same considerations, concluding that a given state can profit from participating in the war. However, the mechanisms of the fiscal war are perverse. Everyone loses with its continuation. As regards companies, some gain, but some lose with the fiscal war. Considering the previously mentioned assumption that the economic incentives conspire against and prevail over the legal provisions that restrain the fiscal war, Section 5 shows that, if the ICMS legislation adopted the destination principle in interstate transactions, the factors that encourage states to participate in the war would be practically eliminated. The main difficulties to be faced in implementing the destination principle are discussed, and the conclusion is that it is possible to get round them. Section 6 summarizes the main arguments and concludes the article. 2 – CONDITIONS THAT JUSTIFY THE GRANTING OF INCENTIVES The product of tax collection is simply a monetary transfer from the private sector to the public sector. As such, it is not a cost for society. The social cost of a tax derives from the distortions it causes in the allocation of resources of the economy and, usually to a lesser extent, from the need it creates for the allocation of productive resources in order to, on the one hand, manage it, and on the other, enforce the fiscal obligations of tax payers. All the taxes used in practice, since they induce changes in the behavior of economic agents, affect the allocation of resources and, therefore, impose – some more than others – costs to society. Conceptually, taxation is justified if the benefit generated by the public use of society’s resources, made possible by tax collection, is greater than its opportunity cost – measured by the social benefit of the best private use of such resources -, plus the cost created by taxation. In statistical terms, the fiscal waiver is always unjustified: if there is merit in taxation, the private use of resources is always an inferior alternative; and if there isn’t merit in taxation, what is needed is not a waiver, but rather the termination of that tax. In practice, it is not possible to create and eliminate taxes all the time just because of changes in economic conditions. Although taxation is subject to alterations practically every day, the range of taxes and their basic characteristics are quite stable. Thus, the tax system is never perfectly adjusted to the economic environment to which it is applicable. Maintaining the conceptual line mentioned in the previous paragraph, tax incentives can be regarded as a marginal elimination of taxes due to the emergence of a new opportunity for private use of society’s resources whose benefits are superior to those of public use to which they had been intended. The weapons used in the ICMS fiscal war aim to attract enterprises to the territory of the federate unit. They are various incentives, in general of a financial-fiscal nature, that result in the reduction or partial refund of the tax to be collected. According to the previous argument, granting incentives would only be suitable in the case of opportunity for the private use of resources: a) that would not be used in any point of the territory of the unit in question, in case the benefit were not granted; b) that is effectively new, that is, an additional investment in the unit; 2 and c) whose benefits are, at least in part, reverted to the residents of the unit, who – unless there are spillovers – are the ones who lose with reduction in the public provision of goods; and that the share reverted to the residents is greater than the benefits generated by the previous public use of the resources. The first of these conditions reflects the foundation for the granting of any incentive: the use of public resources to encourage private investment is only justified in case there are externalities that create divergences between private and social costs (or benefits). The second and third conditions, in combination, ensure that the incentive granted will increases the well being of the population in the state. It only makes sense to use public resources to encourage businesses that would generate an addition (which would not exist in the absence of the incentive) to the future income of the residents, and when this addition is greater than the value attributed by them to the good, whose public provision is reduced or no longer exists. 3 – THE NATIONAL PERSPECTIVE: THE COUNTRY LOSES In this section, the conditions presented above apply, considering the country as the territorial unit. One can verify that there are few cases of enterprises that, from a national perspective, deserve state government incentives, that is, the battles of the fiscal war that result in net profit to the country are very rare. The first point one should note is that it is not possible to encourage through the ICMS new enterprises that aim to directly steer their products towards the external market, since the tax is not applicable to exports and, since 1st November 1996 with Complementary Law nº 87/96, it does not apply to the capital goods that will be incorporated into their permanent assets. However, there is the case where exportation takes place indirectly, through another company. In this case, the reduction (or refund) of the tax to be paid by the producing company, coupled with the maintaining of the credit of the tax on the goods purchased by the exporting company, constitutes an incentive. Since, considering the whole chain of production and commercialization, the tax applicable to the exported merchandise is null, the “reduction of tax to be collected” is actually a subsidy, hidden under the veil of the tax, almost identical to one that would be granted through budget to the company directly exporting its products. The difference between them is that the subsidy granted through ICMS can stimulate the creation of an extra link in the chain of production and commercialization only to take advantage of it. Since it is unnecessary, the existence of the extra link constitutes a reduction on economic efficiency. Even when the ICMS subsidy meets the conditions established in the previous section, it imposes a loss to the country that would not be imposed by the subsidy via the budget.3 Considering companies focused on the domestic market, to stimulate the relocation of an enterprise situated in another state is also a waste of resources from the national perspective. The public good is exchanged for additional profit, unnecessary to ensure the existence of the facility in the country; or even worse, assuming that the original location was correctly chosen, the public resources that are waived are in part consumed by the allocative inefficiency brought about by a location that is not the best one. Similarly, granting an ICMS reduction to multinational companies that would move to Brazil even in the absence of the incentive, even though in another state, is to give to non-residents resources that could be used to improve the well-being of the population of the country, in exchange for nothing. From the national perspective, the tax reduction is only justifiable if the company would not come to any point of the country without the incentive. In the case of enterprises to be implemented with national capital, there is an important additional restriction to justify the granting of the incentive. The entry of external capital is always an addition to the investment in the country; one would just have to establish whether it would or would not occur in the absence of the incentive. In the case of national capital it is also necessary to establish what would be the alternative use of the resources. Thus, even when the objective is to enable, by means of a fiscal waiver, the implementation of an enterprise that would not exist without such waiver, granting the incentive would only be appropriate from the national standpoint if the private resources that would be used in the enterprise were destined, in face of its unfeasibility, to a use other than investment in the country, that is, consumption or investment abroad.4 Considering all these restrictions, it is possible to say, even without analyzing projects, that the cases where the granting of state incentives is justified from a national point of view are extremely rare. Even in these cases, there is at least one additional argument that strengthens the thesis that ICMS-based incentives are harmful to the country. It is certainly acceptable, in view of the dynamics of development, that deconcentration of production and regional development are included among the objectives of the industrial policy and that public resources are used for these purposes. Such objectives, however, are necessarily national and, therefore, they must be pursued under the coordination of the central government. When, through the fiscal war, states try to take on this responsibility, the results tend to be disastrous. First, the winners of fiscal wars are, in general, the states with greater financial capacity, which end up being the more developed ones, with larger markets and better infrastructure. Secondly, by waiving collection, the state is giving up either the provision of services (education, health, even infrastructure, etc.) – which are inputs of the productive process – or fiscal balance, with resulting macroeconomic instability. The current fiscal deficit in Brazil is largely due to the imbalance of public accounts of the states. In several states, collection is almost insufficient to cover as little as the personnel expenses. Even among these states, some insist in participating in real auctions promoted by companies that have already decided to establish new facilities in the country. In some cases, even the state has already been chosen and the auction is nothing other than an instrument to force the state to grant further advantages. Obviously, a fiscal adjustment program requires, among many other measures, the central government to take a stance that is totally against the fiscal war among states, seeking to restrain it by all possible means. It means a waste of public resources that, through the national perspective, is often useless and is also unacceptable in face of the scarcity of such resources. The fiscal war is, in addition, an element of dispute among the units of the Federation. The states complain but do not take any concrete measures to stop it, either because they are participating in it or because they envisage the possibility of doing so. Therefore, one concludes that the country loses with the fiscal war. But, are fiscal wars made only by governors who insist in doing evil or who see an opportunity to benefit their personal political projects to the loss of the population? Or are there motivations of a public nature for them to carry on in such manner? 4 – THE PERSPECTIVE OF STATE GOVERNMENTS: THE STATE (ALMOST ALWAYS) GAINS (FOR SOME TIME) Although the theoretical conditions that merit the granting of incentives, mentioned in Section 2, are the same either from the national or state governments perspectives, the simple change of the territory focused — from country to state — completely modifies the scenario of the previous section. The governor of a state, as a public person, is certainly dedicated to fulfilling the major interests of the nation. But, even by official duty, the governor places his or her state’s interests above those of the nation, and in the case of a conflict of interests, will certainly defend the interests of his/her unit, under the flag of autonomy of federate entities. Moreover, it is natural that governors also worry about their political careers. If the granting of incentives, at least in their view, brings benefits to their state and, in addition, generates advantages to their personal political project, why should they not engage in it?. They will end up granting incentives, despite national interests, especially when these interests are expressed in a diffuse manner, as for example, economic efficiency. Following the same order as the previous section, let us start by considering the case of production for exportation. Attracting to the state a company that steers its production to the external market does not automatically generate any revenue and, as will be demonstrated, may even create a liability to the state treasury. But it creates jobs and, therefore, additional income to residents of the state, which, from an economic perspective, constitutes good business for the unit. Considering the indirect impact of the enterprise, even the state finance department can profit in the future. Unless the company intended to move in the state anyway, even in the absence of the incentive, the conditions established in Section 2 are met. Since it is politically easier and faster and administratively less transparent to provide the subsidy through a financial-fiscal route than through explicit budgetary endowment, the former will be used whenever possible. The taxation system of interstate transactions in force for ICMS is pathogenic as regards exports (and in some other cases). Let us consider a company located in state A, whose total production is exported directly. Due to the exports exemption, the goods leaving do not generate ICMS debits. But the purchase of inputs generates credits that, in the absence of debits to compensate them, need to be reimbursed to the exporting company in order to ensure the exemption of the total value of exports. If the purchases are made from other companies of state A, the reimbursement corresponds precisely to what the state collected previously from the producers of inputs. If, however, the inputs are purchased from state B, then state B is responsible for collecting the tax applicable to the inputs, and state A is responsible for granting the corresponding tax credit. Exports, in addition to not generating revenue, create a burden for the state treasury. On the other hand, if exportation is made through a company situated in state C – rather than directly -, state B collects the tax corresponding to the value of inputs, state A collects the value added by the producer, and state C pays the bill related to the tax previously applicable. In this case, it is an excellent business for the state to attract to its territory the company producing the exported good. It can give up only part of its revenue and, in addition to economic advantages, still collect something. The taxation procedures on interstate transactions stimulate the granting of incentives. In the cases of enterprises aimed at the domestic market, from the perspective of a state government, to encourage a company situated in another state to relocate, or to attract new enterprises, financed either by national or external capital, although it can harm the Federation and/or economic efficiency, does not violate the rules in Section 2. The exception is, once again, the case of a company that has already chosen to move to the state and is only bargaining further benefits. Also in these cases the taxation system in force for interstate trade constitutes a stimulus to the policy of attracting companies. The ICMS rate applicable to interstate transactions, although it is lower than the rate applicable to an internal transaction with the same good, is positive, in such a way that the revenue generated by that transaction is distributed between the units of origin and destination. Because it is positive, it allows a state to attract a company to its territory by means of tax refunds, even in the extreme case where the whole consumer market for the good produced is situated in state B. In this extreme case, if one considers equal production conditions in A and B, the loss of economic efficiency is quite evident. It materializes in the increase of transport costs. But, if the tax rebate is superior to the increase in costs, it is profitable, from a private perspective, to install the company in A. This confers an advantage in relation to its competitors located in B, which can translate into greater profit per unit sold or increase of its market share. It is evident that companies that obtain fiscal advantages gain, but others, already established, that compete for the market with the ones receiving incentives, lose. Therefore, they pressure the state government towards granting identical incentives that allow them to compete in equal conditions. In face of the difficulty in resisting this legitimate claim, the waste of public resources by the state increases further. And as the companies harmed are not only those located in A, all the other state governments suffer similar pressures. If A manages to attract companies and, as a result, obtains advantages for its population, state B, possessing identical instruments, can do the same. The fiscal war starts, further reducing the availability of public resources; even so, it is worth it. But C, D and Z also have the tools. The fiscal war becomes widespread and deeper. With time, fiscal waivers grow and states with less financial power can no longer provide the services and infrastructure needed by companies to produce and deliver production. The battles of the fiscal war start to be won only by the ones with greater financial power, that is, the same ones that have easier access to credit. At the same time, with the generalization of tax benefits — all states granting similar incentives — they lose their power of encouragement, which depends on taxation differences. The fiscal war transforms incentives into mere collection waivers that have no stimulating effects. In face of the generalized reduction of the taxation burden, companies start to choose their location considering economic factors, among which, the quality of infrastructure and public services on supply. Evidently, the fiscal war is an enemy of the regional development policy and of industrial deconcentration. 5 – FISCAL WAR AND ECONOMIC SIGNALING No governor is uninformed and it is very rare to find a case of economic myopia among them. All know that, at its limit, the outcome of the ICMS fiscal war will be the one mentioned in the previous paragraphs. But they also know that, while the limit is not reached, there are gains involved in the fiscal war. In this issue, there are clear economic signaling errors that create divergences between national and state objectives and the best development strategy for the state in the short and long run. The 22 years of existence of, and non-compliance with, Complementary Law nº 24/75 allows one to infer that economic encouragement prevails over the legal provisions that restrain the fiscal war. Half of this time belonged to the authoritarian regime and not even at that time was it possible to gather enough political will to demand the enforcement of the law. One can attempt, as did the bill of law that resulted in Complementary Law nº 87/96, to expand the restrictions to the fiscal war. But, even if political support is obtained to turn the bill into law, one can anticipate that the more drastic legal provisions will be equally disregarded. The solution for the issue is in the change of economic signaling in such a way that the stimulus to participation in the war is minimized. As demonstrated in this article, a strong signal in the wrong direction is provided by the legislation that addresses the taxation of interstate trade through the ICMS. As observed, the solution adopted for the appropriation of revenues from interstate transactions makes room for the worsening of the fiscal war. From the national perspective, the ICMS is currently a tax on consumption; but, from the perspective of each state, it is a hybrid, partly imposed on state production, and partly on its consumption. Since the mobility of production factors, especially of capital, is much greater than consumers’, the tax on production is a much more powerful weapon in the fiscal war than the tax on consumption. Reducing the encouragement to participation in the war requires that ICMS become a tax on consumption also from the point of view of the state government. To this end, all that is necessary is to adopt the destination principle for the taxation of interstate trade flows, as already happens in foreign trade. Adopting the destination principle means eliminating the tax’s interstate rate. Afterwards, all products destined to consumption in a certain state – produced within the state, in another state or abroad – generate collection solely for that state; and goods produced in the state, destined to other states or foreign countries, are not taxed. This system does not completely eliminate the fiscal war, but imposes very strong restrictions on the effectiveness of ICMS incentives. As all goods destined to other states or to foreign countries are not taxed, they will not be the base for the granting of incentives; and as the Constitution prohibits states to establish different taxes on goods on the bases of their origin or destination, there is no way to privilege the consumption of goods produced within the state.5 The only possible way of granting tax benefits to attract businesses is the reduction of the tax to be collected, whose value now depends on the volume of sales of the company inside the state. Evidently, only the companies intending to direct their production primarily to this market could be attracted. Moreover, one eliminates the hypothesis — that currently happens and is not a mere theoretical curiosity, since it actually occurs — of a state granting incentives and another paying the bill. The adoption of the destination principle has other advantages: it eliminates the unfair revenue redistribution among states that exists today when a taxed exit of merchandise to another state is followed by an exempt exit (or none), which is the case when a state collects and the other grants the credit of the tax;6 it enables the exemption (or reduction of the rate) of products whose consumption has a significant weight in the budgets of poor families without causing major damage to the collection of states with production strongly concentrated in those goods; and it promotes deep change in the distribution of state fiscal resources in favor of the net importer states in interstate trade, which are generally the poor ones. The last advantage to be mentioned is also one of the difficulties for the introduction of the destination principle. The net exporter states in interstate commerce are few in number; but they are important both in terms of economic movement and politically. São Paulo, for example, would undergo a loss of over 10% of its total collection, hardly bearable if it happens abruptly.7 The obvious solution to get round this difficulty is gradual introduction, with gradual reduction of the state rate until it reaches zero. Another difficulty, which can also be dealt with, is the need for reorganization of state financial administrations, mainly in less developed states. Collection is currently concentrated in a relatively small number of taxpayers, which leads administrations to focus control on them. With the destination principle, collection is diluted, which requires a change in inspection methods. The last one of the difficulties is the increase of encouragement to tax evasion. The differences between interstate and internal rates encourage the bad taxpayer to simulate an interstate transaction and to deliver the goods within the state, setting off a chain of evasion. The problem, which is already serious, would increase with the elimination of the rate applicable to interstate operations. This problem is solved by the Proposal of Constitutional Amendment (PEC) that changes the Chapter on the Tax System — PEC nº 175/95 — that is going through the National Congress. It substitutes the Tax on Industrialized Products (IPI) — and the ICMS for a new tax of similar characteristics to the latter, shared by the Federal and state governments, and creates a mechanism that completely eliminates the type of tax evasion mentioned, even if the destination principle is adopted8. It should be pointed out that the proposal contained in the PEC nº 175/95 facilitates the solution of the two other problems mentioned. First, the loss of collection that the net exporter states in the interstate commerce would suffer with the adoption of the destination principle would be partially compensated by the profit resulting from reduced tax evasion; and, second, the difficulty of inspection created by diluted collection is smaller with the shared ICMS than with the current ICMS, since the two ends of the interstate transactions would register with the Federal Government. The PEC nº 175/95 leaves to the Federal Senate the decision to put the destination principle in practice. Being a strong tool to restrain the fiscal war — in addition to other qualities, including benefiting less developed states — it is preferable to ensure its adoption in the body of the Constitution itself, providing for its gradual introduction. The National Congress just needs to amend the PEC nº 175/95 to this effect and approve it, for the fiscal war to practically end. 6 – SUMMARY OF MAIN ARGUMENTS AND CONCLUSIONS The item g of numeral XII of article 155 of the 1988 Constitution attributes to complementary law the competence “to regulate the way how, by means of deliberation of the States and the Federal District, tax exemptions, incentives and benefits will be granted and revoked”. Complementary law nº 24/75, supported by the Constitution, prohibits the granting of ICMS-based incentives, except in the cases provided for in agreements entered into in the scope of the Confaz, whose approval depends on unanimous decision of the states. Nevertheless, state governments have been granting incentives regardless of Confaz, competing among themselves to harbor new businesses. This competition, the fiscal war, has been harming states finances — and, consequently, fiscal adjustment — as well as the public provision of goods and services, many of which are important inputs for the production process. In addition, the fiscal war creates conflicts among the units of the Federation and its results tend to oppose the objectives of policies — necessarily national — aiming at the regional development or the deconcentration of production. In spite these evils, the fiscal war would be acceptable if its net social benefits were positive. It was argued, however, that the cases that justify the granting of state incentives are extremely rare, from the national standpoint. The perverse dynamics of the fiscal war were also discussed: after some time, with the generalization of tax benefits — all the states granting similar incentives — they lose their power of attraction and change into mere collection waivers. On the one hand, in face of the generalized reduction of the taxation burden, companies start to choose their location only on the basis of production and market conditions, which include the quality of infrastructure and public services on supply. On the other, with the increase of fiscal waivers the states with less financial power can no longer provide the services and infrastructure needed by companies to produce and deliver the production. The battles of the fiscal war start to be won only by the states with greater financial power, therefore, able to support the burden of the waivers and still ensure reasonable quality of public services. From a state perspective – although the fiscal war of the ICMS is harmful to the nation and even though its outcome is the one described in the previous paragraph — there are gains involved while the limit situation is not reached. Attracting a company, even if it does not generate direct revenue, creates jobs and, therefore, additional income for the residents of the state. Considering the indirect impact of the business, even the state finance department can gain in the future. Unless the company intended to move to the state anyway, even in the absence of the incentive, the conditions that justify the granting of the incentive, established in Section 2, are met.. Thus, from the perspective of the state, it is justifiable that a state government dedicate effort to attract companies by participating in the fiscal war. One concludes, therefore, that in the process of the fiscal war, the economic stimuli that govern the behavior of each state government provoke actions whose outcomes are undesirable for the country. If all states stopped granting incentives, all would gain; but if a state abstained from such practice and all the others carried on with it, that state would lose. In these circumstances, legal prohibitions to the granting of incentives will hardly be effective. It is necessary to change the economic signaling perceived by state governments in order to prevent their individual actions from going against the national interest.9 A strong signal in the wrong direction is provided by the legislation that addresses the taxation of interstate trade through the ICMS. From the national perspective, the ICMS is currently a tax on consumption; but from the perspective of each state, it is a hybrid, partly imposed on state production, and partly on its consumption. Minimization the encouragement for a state to participate in the fiscal war requires that ICMS become a consumer tax also from the perspective of the state government, which would impose very strong restriction on the effectiveness of ICMS incentives. For this to happen, the destination principle should be adopted for taxation of interstate trade flows. There are difficulties involved in the introduction of the destination principle, however it is possible to get round them. The main problem, encouraging an increase of existing tax evasion, caused by the increase in the difference between the rates applicable to transactions inside a state and to interstate transactions, is solved by the PEC nº 175/95 going through Congress. It creates a mechanism that completely eliminates this type of tax evasion. The proposal, however, attributes to the Federal Senate the decision concerning the adoption of the destination principle. For the fiscal war to practically finish, the National Congress just needs to approve the PEC nº 175/95, amending it in order to ensure the adoption of the destination principle regardless of future political decisions. 1 Unburdening a capital good from VAT can occur in two ways: making the sale of the good exempt, with assured maintenance of the credit relative to the tax previously paid by its producer, or taxing the sale and granting credit of the tax paid by the buyer. Complementary Law nº 87/96 preferred the second way. It is worth noting that, for tax payers the ICMS is a consumption type VAT; but the purchase of capital goods by non tax payers (public sector and service sector, except interstate and intercity transport and communication services) continues to be taxed. 2 This condition is in fact, as declared in the text, more restrictive than it should. In fact, if a project with incentive A is carried out rather than a non-incentive project B that involves equal investment, there is no addition to the investment in the unit. However, the incentive would be justifiable if the net social benefit of project A were greater than that of project B. In order to simplify the text, even if at the cost of some loss of accuracy, one opted to ignore such possibility. 3 One will see ahead that the system of taxation of interstate commerce currently adopted for the ICMS can stimulate state governments to participate in the fiscal war attracting indirect exporters. 4 Footnote nº 2 applies to the comment contained in this paragraph. 5 Article 152 of the 1988 Federal Constitution 6 This is the case of the exports, here addressed, as well as the case of industrial goods. 7 Evidently, great profits correspond to this loss for less developed states, since the total revenue of the states does not change with the adoption of the destination principle. 8 Please see R. Varsano, “A Tributação do Comércio Interestadual: ICMS Atual versus ICMS Partilhado “, Text for Discussion nº 382, IPEA, September 95. 9 One should observe the parallel that exists, as regards this aspect, between the process of fiscal war and the inflationary process. In the latter, if all economic agents kept their prices, the process would finish and all would gain. But, if one agent kept its price and the others did not, it would lose. Experience has shown that legal prohibitions to price increases have not been able to end the inflationary process. The success of the Real Plan is due to changes it promoted in the economic signaling perceived by the agents, who made their individual actions compatible with the collective goal.