Isma was concerned enough to cancel a press conference with Gustavo Piga, the author of the report, because it said his safety was not certain. Piga has discovered evidence of a swap contract between a country and a bank that he said was used to misrepresent the size of the country's deficit to enable it to join Europe's single currency. Although Piga won't say it directly, the country in question is Italy.
In May 1995, Italy issued a ¥200 billion ($1.6 billion) bond. The exchange rate at that time was ¥19.30 to the lira. By December 1996, the yen had depreciated to ¥13.40 against the lira. With 12 European countries desperately trying to meet the five criteria for joining the single currency, Italy wanted to hedge its foreign exchange gains, so it entered into a currency swap with a foreign bank.
But the swap had some peculiar features. The rate on the swap was ¥19.30 to the lira, the exchange rate at the time of the bond issue. This was unusual - the vast majority of currency swaps are done at the existing currency rate, as Eurostat, the European Union's statistics office, admits. The exchange rate was far worse than Italy could have obtained. It was losing a lot of money in the long term.
Also unusual was the interest rate on the swap - Libor minus 1,677 basis points. Such an interest rate was almost unheard of in the currency swap market. It meant the bank counterparty was paying Italy cash advances through the quarterly payments.
All of this so far is agreed on by all parties. Where they differ is in their accounts of why this happened.
Piga says the swap was not a hedge at all, rather it was clever window dressing whereby the country received four cash advances thanks to the negative interest rate from the bank counterparty. The swap was a loan, where the country accepts a bad exchange rate in return for cash up front, rather than a hedge. These incoming payments allowed Italy to reduce its budget deficit by 0.02%.
This kind of swap would be blatantly questionable were it not for the Euroyen bond, which explained the technical need for a hedge. Thanks partly to the cash advance, the country is able to reduce its deficit in the short term and so meet the deadline for joining the single currency. After the swap closed, in 1998, the country would lose money (unless the yen appreciated back to its former level), but this was after the deadline for meeting the criteria, and so not such a pressing concern.
Italy refuses to comment but treasury sources insist the swap was a hedge designed to lock in foreign exchange gains. They rebut Piga's argument that the swap could have been structured to lower the deficit.
The deal in question would only have reduced Italy's GDP to budget deficit ratio by 0.02% - not enough to make a difference to Italy's chances of joining the single currency.
Quite right, says Piga - he says these deals are usually part of a series and it is probable this was the case with the swap (he doesn't explicitly say it is Italy). If there were 10 of them, the cash from the deals would be enough to reduce the country's debt by 0.2%, which was close to the amount Italy needed to reduce its deficit to GDP ratio in 1997.
It is difficult to know for sure how many of these deals Italy did. They are confidential and over the counter - this is the first time anyone has got hold of evidence of such a deal. Piga only did so because a government source handed it to him unaware that the swap was in any way peculiar. Therefore it is unlikely that any other deals will come to light. Eurostat denies that Italy did any other deals like this one, though the Italian treasury is one of the most active sovereign issuers of foreign currency bonds, so is highly likely to have done other currency swaps.
More important, the Italian treasury says it had to structure the bond the way it did because of a Eurostat accounting law, protocol 3605, which says the public debt of a country applying to join the single currency should be accounted in the currency originally issued - in this case yen. So any swap Italy did could not be accounted for in the public debt, whatever the exchange rate.
The treasury could have done a swap at the existing rate, then closed it when the bond matured in September 1998. But before that, says the treasury, the yen could have appreciated, making Italy's public debt bigger. This, the treasury says, could have damaged the country's chances of joining the eurozone.
A treasury source says it decided to split the benefit of the capital gain from the yen depreciation in order to receive some of the benefit during the lifetime of the bond, thereby protecting itself against any appreciation of the yen prior to the bond's maturity in September 1998.
The argument makes sense, but there are two problems with it. First, it goes against accounting principles. Off-market rate swaps are not hedges so much as loans, yet because of an accounting loophole they are still treated as hedges.
According to Piga's report, most countries' debt managers would not do off-exchange rate deals to get cash advances to help get past budgetary constraints. It is legal, but, in the words of one country's debt manager: "We have a self-imposed, ethical, unwritten rule not to use upfront payments."
Piga calls for a change in accounting law to cover these kinds of deals. He says that as these swaps are not being used for hedges so much as for cash advances on capital gains they should be treated as loans for accounting purposes.
The second problem with Italy's argument is that it concentrates on the size of its overall debt and ignores the effect on the budget deficit. Italy's public debt was around 110% of GDP in 1997 - way over the 60% outlined by Maastricht. As it was so far over the limit, Italy was unlikely to worry about the negligible effect of a foreign exchange loss. Even a large appreciation of the yen was unlikely to have a significant impact on Italy's chances of joining the eurozone. However, the cash advance from the negative interest rate on the swap would have made some difference on the budget deficit, which stood at 3.2% in 1997.
All the political emphasis in the run-up to joining the single currency was on Italy meeting the deficit criteria and showing a move towards reducing its debt. In the end, it failed to do this - the country's debt grew to 120% of GDP in 1999, without causing Italy too many problems with the EU. But it did manage to reduce its deficit to meet the 3% target, though only just, with three months to spare, and this could have meant the difference between being able to join the eurozone or, like Greece, being forced to wait.
In 1997, when Italy's prime minister, Romano Prodi, was canvassing for Italy to be a founder member of the EMU charter, he pointed to the fall in budget deficit, where Italy was one of the stronger of the tested countries. This strengthened Prodi's hand enormously against Germany, which had doubts about Italy's ability to meet the criteria. Indeed, Germany itself had some difficulty in meeting the 3% deficit target.
Back in 1998, several countries' public debt was over the 60% mark - Austria's was 65% of GDP, Sweden's was 75%, Italy's was 121.9% and Belgium's was 121.3%. Greece, the only country to be refused entry to the eurozone in 1998, had a public debt ratio of 106.4%. The reason it was refused, while Belgium and Italy were allowed to join, was that it had a deficit of 4.2%, while those of Italy and Belgium were under the deficit target.
The key effect of the deal would have been to reduce interest expenditure in 1997 and 1998 and so lower the budget deficit.
Even if Piga's argument is right, and Italy's defence of its contract doesn't add up, the trade was still legal, so is it a big deal? Yes, for four reasons.
First, the possibility for similar window dressing still exists. According to Piga, several countries' debt managers say they have come under political pressure to do off-market-rate derivatives transactions in order to get cash up front, often to help deal with budget demands. These same debt managers say they resent being told to do such deals and would like better accounting laws to make them impossible. Countries are able to massage their deficit figures which makes it difficult for institutions such as the IMF or the Bank for International Settlements (BIS) to know what a country's deficit is.
However, regulators have shown little inclination to protect against such window dressing. A BIS official says: "The swap in question is part of history now." A Eurostat risk management official says he does not think the rules need changing. The director-general of Eurostat, Yves Franchet, has said that a Eurostat review of the use of swaps by European countries found that the practice was not widespread.
This raises questions about regulators' ability to monitor the swaps market. According to John Maskell, swaps strategist at Barclays Capital, many countries use a large volume of swaps. For example, Spain has announced its intention to use around e12 billion ($10.6 billion) in swaps next year to shorten the maturity of its debt. France's treasury intends to use e200 billion in swaps next year and Germany e20 billion-worth. And this is just interest rate swaps.
With swaps activity on such a scale, and debt managers saying the possibility for window dressing through cash advances still exists, there is good reason for regulators to have another look at accounting laws.
Secondly, it made an enormous difference if Italy joined the eurozone in 1998 or was forced to wait. That difference was not just political - a lot of money was riding on which countries would be able to join the eurozone. Convergence trades were the big thing at that time. A huge amount of business was done by investment banks and hedge funds, with positions worth tens of billions of euros. Many banks had a big interest in Italy converging because so much money was bet on it. The total amount was probably in the trillions.
Governments were also making big convergence bets. Piga's report says: "Many governments were ... betting against the market that they would enter into the eurozone by fulfilling the required criteria. If they had done so through derivatives, then large gains would have followed that would remain even in the following years."
Thirdly, letting such deals go by without any reaction sends worrying signals on the eve of the introduction of euro notes and coins. If this was going on when the euro was first introduced, and the EU was happy for it to happen, then what else will be given the wink in the name of political expediency?
Finally, Piga says: "If a market maker provides a government with window-dressing advice ... it links itself in a tight embrace with the sovereign." He goes on: "While it is in their [the bank's and country's] mutual interest not to go on record about such activities, there is also the possibility that one of the two parties might be able to exert undue pressure on the other in future transactions."
The swap was confidential and Piga will not say which bank was involved. He is not even saying that leverage was necessarily obtained from the deal being secret. He expressly says that he has no suspicions of favours being obtained from the deal he talks about, which has since been identified as Italy's swap. But he says that while such deals are possible, there is a danger that the secrecy of such deals could be exploited.
His editor, Benn Steil, says: "Piga is not a scandal-mongerer. He did not want to get famous for finding out a scandal." But such claims from a well-connected academic are surprising. The unflattering picture it paints is of secret deals either not seen or not fully understood by regulators, accounting laws tested in pursuit of massaged deficit targets, and financial conventions ignored in the interests of abstract political goals. Welcome to the unified Europe.