The new financial architecture is not a mandatory international agreement that will come into place at a particular time. It is really the emerging consensus on what are the features needed in the financial systems in developing countries and in theglobal arena to avoid crises of the East-Asian variety. This consensus emphasises that domestic financial systems, especially banks, must be subjected to internationally accepted prudential norms and standards of supervision. It also emphasises the importance of transparency in government accounts so that the set of the macro-fundamentals is clearly and transparently known.

Earlier, when international credit rating agencies, the IMF or the World Bank evaluated our economy, they used to look at the fiscal deficit or the current account deficit. Now the strength of the financial sector will be a core element.

The traditional view has been that the states' deficit does not matter too much because states are not allowed to borrow without the Centre's permission. But since most of our states now have large debt burdens this raises the question whether they will be able to repay debts to the Centre on schedule. The Brazilian currencycrisis in January 1999 was triggered by the governor of the Brazilian state of Minas Gerais informing the federal government that the state would not be able to meet its debt obligations to the federal government. This led to a loss of confidence in the fiscal position of the federal government. There are important differences between Brazil and India. Brazil had full convertibility which means a loss of confidence can lead to a large capital outflow which can provoke a currency crisis. We must, however, recognise that the fiscal position of the states will be under scrutiny because it affects assessment of national fiscal vulnerability. Many states are going in for large guarantees of power purchase agreements entered into by SEBs. Of course, if SEB reforms take place and tariffs are rationalised, these guarantees may never be invoked.

No. First, all NPAs are not irrecoverable.Secondly, a large portion of the gross NPAs in Indian banks have been provisioned for. As for PSU borrowings, these are loans of commercial organisations some of which are now part privatised. You cannot just add it to the government debt unless the government has given guarantees for such debt. Earlier, we were more free in extending government guarantees for PSU debt but this was tightened up several years ago. Under the new transparency requirements, this would be flagged as a sort of quasi-deficit.

I have seen reports that the NPA levels in China are very high. There are two reasons why this does not worry investors. One is that China's public debt as a percentage of GDP is very low. So it is assumed that even if the government takes on the burden of filling the hole in the banks' balance sheet, the overall debt to GDP ratio will still be comfortable. Our debt to GDP ratio ismuch higher and there is less room for taking on new burdens. Second, it has to be recognised that China's high growth rate mesmerises investors who are then willing to forgive a great deal. If Chinese slow down for whatever reason, there will be much greater pressure to reform the financial system. The Chinese have recognised the need for financial sector reform and have announced that they intend to make major changes.

This is a pointer to the fact that globally our banks are going to be expected to live up to certain standards. If our banking regulation and supervision is seen as inadequate, i.e. we're seen to be tolerating laxity in prudential norms, we'll come under adverse attention and this could affect our banks' branches abroad.

Bailoutsare certainly possible. But the focus is on two things. First, what is the total cost of the bailout? The US S&L bail out cost the US less than 2 per cent of GDP. In many of the recent crises affecting developing countries the bailout costs have been much higher, around 20 per cent of GDP. Second, it is important to know whether the bailout will lead to restructuring of the banks concerned which will at least ensure that the problems will not recur.

No one expects instant reform. If we declare a credible timeframe (say five years) and stick to it, it will be lauded. But then the focus will be on whether we stick to the time table.