The fundamental principle underlying all indirect taxes, including service tax, is to ensure that the tax is paid on value addition and there is no tax on tax. The determination of value addition can be done in many ways. India has chosen the netting off principle whereby taxes on inputs are available as credits in discharging the tax on the output. 
 
While the aforesaid system has been in vogue for two decades for Cenvat, it is of very recent origin for service taxes. It was in 2002, eight years after the introduction of the service tax, that a very limited scheme of input tax credits was introduced. Such credits were made available only if both the input services and the output services were covered under the same taxable heading. Subsequently, the scheme was changed to remove this restriction. 
 
However, the credits were restricted to services and not extended to goods. Hence the distinction between goods and services continued. To remove this, the government enabled a seamless availment of input credits across goods and services. This meant that both manufacturers of goods and providers of services were able to entirely avail the credits relevant for all input goods and input services used by them. 
 
Therefore, in terms of the Cenvat Rules, 2004, that were brought into force for this, the input tax credit scheme is now fungible and hence all tax credits come together for utilisation in any appropriate manner for payment of the appropriate Cenvat or service tax on the output goods/services respectively. Also, the Cenvat rules are applied to capital goods, used by both manufacturer of goods or the provider of services. 
 
While the input tax credit scheme is now undoubtedly broadbased, there is a need for the manufacturer/service provider to establish a nexus between the input goods/services and the relevant output goods and services. In other words, the admissibility to credits is predicated on the ability to demonstrate that the input goods and services, to which the tax credits relate, do have a nexus with the manufacturing process or the provision of services. This requirement to establish a nexus has created several intractable problems for assessees. 
 
The first of such issues is the distinction between goods/services which are not taxable (henceforth we will limit this reference to services alone) and those which are taxable but exempt. This distinction has a variety of implications but the point to be made here is not so much this distinction, since input tax credits are ruled out in both situations, except in a situation of export of services, but the fact that input services are typically used to provide both taxable services as also non-taxable/exempt services. In view of the requirement to establish the nexus as described above, in such a situation of provision of both taxable and non taxable/exempt services, the determination of the availability of input tax credits becomes a very difficult one. The Cenvat rules lay down a prescriptive method to determine the disallowance in such situations whereby not more than 20 per cent of the relevant output service tax liability can be offset by utilisation of input tax credits. The problem with this prescriptive procedure is that it disregards the actual quantum of ineligible tax credits. The global best practice in such situations is to prescribe de minimus rules whereby, if the actual incidence of credits is below a threshold level, no disallowances are made. To be fair to the government, there are several input services which are carved out as an exception to the general rule for determination of nexus and tax credits relevant to such input services are available without any regard to their actual usage in the provision of taxable and non taxable/exempt services. However, this list of services is a restricted one and the clear need of the day is for the government to either enlarge this list of exceptions or to lay down appropriate de minimus rule. In any event, it must be ensured at all times that the undesirable outcome of tax cascading is avoided and that no service provider is in a position of having to pay service taxes beyond what is legitimately required of him. 
 
Another problem associated with this determination of nexus is with regard to a slew of services which typically have no nexus at all with the actual provision of services. Examples that come to mind are in relation to services procured at the head office or corporate level, all of which, by definition, would be common for both taxable and non taxable/exempt services. 
 
Here again, the rules do seem to take cognisance of this situation and provide for availment of credits on such services. However, the continuing problem in this regard is two-fold; one, the provision relating to determination of nexus and the provision relating to such common services are not harmonious but seem to contradict each other and secondly, the list of such common services, as contained in the provision, seems to be a limited one and appears not to encompass several other such services. Here again therefore, there is an urgent need to ensure that the provisions are so harmonised that the principle of nexus is made applicable only to those services which are, in terms, capable of passing this test of nexus and common services, which are legitimately procured for running the business, are excluded from this principle of nexus altogether. 
 
Yet another problem is the lack of understanding of these quite complex issues at the field level. A variety of differing stands has been taken by authorities at different jurisdictions in relation to admissibility of credits on several input services. 
 
Examples that readily come to mind are in relation to use of cellular phones by company personnel in day-to-day business operations and admissibility to credits of the service taxes paid by companies on procurement of transportation services (in terms of a reverse charge mechanism). In terms of the provisions that are in force, there ought to be no doubt on the admissibility of such credits. Nevertheless, there continues to be a spate of litigation on such seemingly clear matters. Indeed, even departmental circulars are at times disregarded by the field-level personnel. The point here is that such litigation is non-productive and engages a lot of time and effort of both the assessees and the department. The broader picture of the need to prevent cascading and to therefore restrict the base of the tax to value addition, as per the netting off principle, is lost. One last matter which has caused considerable concern is the taxation of omnibus contracts comprising of both the supply of goods and the provision of services. The problem is with regard to the options available in such situations to appropriately determine the base for service taxes. The reason why this point is relevant is that the options are typically predicated on the need to not avail input tax credits. If the objective of providing these options is to ensure that the service tax is paid on the services element of such omnibus contracts alone and is not even inadvertently paid on the non-services or material portion of the contract, this objective is not achieved in practice in a number of situations. Indeed, the restriction placed on availment of input tax credits on input services is arguably not defensible at all and the government needs to bring out clear guidelines to ensure that there is no loss of legitimate tax credits to the manufacturers/service providers who undertake these omnibus contracts. This problem is acute in the capital goods and engineering sectors, particularly in relation to EPC contracts. 

S Madhavan
The writer is leader, indirect tax practice, PwC