The elegant public festivities at the launch of the EU central bank (ECB) contrast starkly with the frantic pace inside the ECB's headquarters just a few hundred yards down the road.

The 400 staff have exactly six months to get ready for the day when 11 national currencies will be replaced by the euro, the single European currency. Insiders describe the atmosphere as a mix of excitement and exasperation. Some preparations are behind schedule. Holidays have been postponed.

One tired-looking monetary official said: "I have never seen anything like this. Everybody here works all the time. I have cancelled almost all of my outside engagements."

The ECB's predecessor, the European Monetary Institute (EMI), has had the job of starting preparations and setting out policy options for the bank.

But not all is going according to plan. An example is the choice of monetary policy instrument, the single most important decision the ECB must take in the next six months.

In theory, the bank's governing council should confront a straight choice between the two options recommended by the EMI: inflation targeting or monetary targeting. Under strong pressure from Germany, the ECB is inclined to favour monetary targeting.

Unfortunately, it does not yet have reliable euro-zone monetary statistics for this strategy to work. The aggregation of national monetary data into a single set of European statistics has turned out to be surprisingly complicated.

Germany, for example, has been relying on M3, a measure of broad money. But in France, new savings schemes have rendered M3 almost obsolete. Italy does not even publish M3 data.

Insiders say it is a race against time. But even if the data were reliable, monetary targeting could still run up against the problem that certain kinds of change can transform the nature of economic relationships.

Time-honoured "laws", such as the link between money supply and inflation, may no longer hold under the new regime of economic and monetary union.

Leading international economists have therefore urged the ECB to opt for inflation targeting, seen as a more modern approach. Yet inflation targeting also faces the same statistics problem. It depends heavily on the quality of input data and the accuracy of econometric forecasting models. Uncertainty is bound to prevail on both points.

As a result, the ECB will not be in a position to rely purely on either approach.

Mr Wim Duisenberg, the ECB's president, has indicated that the ECB will use monetary targeting as its main policy plank, supported by a supplementary inflation target.

But there is no guarantee that the combination of two bad policies will yield a good policy.

A further area that has turned out to be more difficult than previously imagined is the question of bank note security. The recently reported "loss" of a security hologram on an Air France flight from Paris to Munich was an embarrassment for the ECB. The incident is still under investigation by French and German police, but sources say the inquiry is going nowhere.

The Federation of German criminal investigators (BDK) has warned that the introduction of new banknotes, scheduled to take place in the first half of 2002, will open the floodgates to banknote-related crimes, including forgeries and fraud.

One problem is the rule that each country can print its own set of notes. Mr Holger Bernsee, president of the BDK, warns that euro banknotes may not be identical throughout the euro-zone.

"You are bound to get a situation where people reject banknotes as forgeries, even though these are legally printed notes," he said.

The ECB is in charge of co-ordinating banknote security, while national finance ministries are responsible for the minting of coins. Despite the relatively long delay before the introduction of euro-notes and coins, the authorities are up against a tight deadline.

In France, the minting of coins has already started amid teething problems. The printing of euro banknotes is due to start by the end of the year.

There is no sense of panic yet, but the schedule for the next six months allows little room for slippage. After the smooth technical progress so far, one should not be surprised to find that EMU may soon hit some obstacles.