Institutional investment in the recently introduced IDFs will require sovereign guarantees. Projected not merely as a factor of a single project but for several projects in a developmental cluster, such government guarantees can reduce the chances of the risk.In order to boost funding for infrastructure projects, the central government and Reserve Bank of India (RBI) recently introduced the concept of Infrastructure Debt Funds (IDF). Government expects various investors including banks, pension funds, insurance companies, World Bank, Asian Development Bank (ADB) to participate in the IDF.However, for the World Bank and Asian Development Bank (ADB) to contribute capital to the IDF, government has to provide sovereign guarantee, since such guarantees greatly facilitate investment by reassuring the investor. But they also place a responsibility on the government, because in case of a default, the government has an obligation to step in. This is a subject that has been heavily debated internationally. With big-ticket infrastructure projects both in the works and in prospect, the debate has come to India's doorstep too. To look away from it will mean decision by default, India's chief economic advisor Kaushik Basu commented in the Economic Survey 2011-12.Valid warning: In the Economic Survey, the former Cornell professor of economics talks about the risks associated with such sovereign guarantee and analyses the subject from an academic point of view. Providing such a guarantee may do nothing immediate to the government's fiscal arithmetic, but it amounts to undertaking future fiscal expenditure. Since there is always the probability that even a guaranteed project will fail in the future, each such guarantee amounts to a certain additional expected expenditure by the government in the future, Basu says.Hence, such guarantees, given recklessly, can lead to unsustainable fiscal deficits in the future with all their attendant problems, such as inflation, collapse in investment, and, ultimately, economic recession.While this warning is valid and governments ought to keep it in mind, there are circumstances where some strategic and well-designed guarantees or 'comfort letters' from the government can be desirable in the overall interest of the nation. This can happen in a buoyant nation on the verge of take-off and considering a high growth in a number of infrastructure projects, it is an argument that has direct relevance to their success. There is typically a lot of positive externality among infrastructure projects: A new road that will be operated by a toll system is more likely to be successful if the residential township at the end of the road comes up; and the residential township being contemplated by the developer is more likely to be successful if the road gets built.Project as factor of development: In order to illustrate the point, Basu gives the following example. Suppose there are three projects, pertaining to a road, a township, and a power project. Each entails an initial cost of 100. If the project succeeds, it yields 150; if it fails, the entire initial cost goes unrecovered. If all three projects are undertaken, each project is more likely to succeed, because of the kind of positive externality mentioned; let us suppose that the probability of success of each project, when the other two are implemented, is 0.95.If, on the other hand, the other two are not implemented, then assume the project that is implemented has a probability of success equal to 0.5. If government gives a guarantee to the investor for a project, then for an investor it is worthwhile investing in the project, since she incurs no risk of default. In the event of a default, government pays off the investor the 100 that she had invested.Suppose the government now gives a guarantee to only one project. Assuming that the other projects are not undertaken under the circumstances, there is an expected loss of 50 units of money to the government, since the probability of failure is half and in the event of a failure government has to pay the investor 100. Hence, the expected fiscal deficit rises by 50.Now suppose government gives guarantees to all three projects, then all three projects get implemented; and the government's expected fiscal cost of this is only 15 (3 × 0.05 × 100), since there are three projects, each project has 0.05 probability of failure and, in the event of a project's failure, government has to pay 100 units of money. If these projects create socially valuable wealth, which is worth more than 15 units of money, it is arguable that guarantees to all three are desirable; even though it may not be worthwhile giving a guarantee to any single project.This simple arithmetic is not a reason to rush and give out guarantees or even comfort letters (comfort letters often, in effect, turn out to be like guarantees in the eyes of the law) but it alerts us to the fact that for a nation on the verge of take-off, and with complementarities between projects, the calculus of guarantees and fiscal deficits is not as simple as may seem at first sight.We should evaluate the benefits and fiscal costs of government trying to give a coordinated big push to a cluster of infrastructure projects, and recognise that the costs and benefits would not be the same if we worked this out for each project separately and then simply added them up.