For a currency union to work for a country, the most important thing is that the country trades overwhelmingly with the other members of the monetary union.

This ensures that all the countries in the union move roughly in the same economic cycle. It is also important that the structures of the respective economies are broadly similar, so that one country doesn't experience a huge boom, while the rest are just motoring along nicely.

Having similar structures in banking and housing, for example, will imply that a country should not suffer a monumental bust, while the others are merely experiencing a normal recession. Equally, it is important that there is significant movement of people within the currency union - like there is in the US between its states - so that, if a country does slump, its citizens can move to find work in another member country.

In general, for a currency union to work, there should also be a single fiscal policy so that, when one area of the currency slumps, the rest of the union's taxes go some way to ease the problems in the region in difficulty. This is how the currency unions in the US, Canada and Australia work.

Guess what? None of these attributes was in place when Ireland joined the EU economic and monetary union (EMU) and the euro. So it is clear that we didn't join for economic reasons. So why did we join? It seems that we were too insecure to behave logically and this national insecurity - particularly among our senior mandarins - prevented us from having a debate.

6.2 New optimal currency area criteria

The study of the costs and benefits of common-currency areas goes back to the seminal work of Mundell (1961). Conventionally, the major cost of a joining a common currency area is the loss of one's own monetary policy --- the ability to set the short, risk-free nominal interest rate or the nominal spot exchange rate. This loss is harmful for two reasons. First, if there are asymmetric shocks in different member countries of a common currency area, then the common central bank cannot smooth output and employment in individual
countries, even if there are persistent nominal price and/or cost rigidities. Second, if countries have different consumption baskets and if relative prices are changing, then even with a single monetary policy there will be
different inflation rates in different countries. If, say, two and a half percent inflation per year is optimal then a central bank may be able to attain something close to this for the currency area as a whole, but not for individual countries.

...

We argue, however, that these old optimal currency area criteria are not particularly relevant to the case of Iceland. It is true that cultural differences, language barriers and geography ensure that labour is unlikely to be especially mobile between Iceland and continental Europe. Although there have in recent years been quite sizeable labour flows between Iceland and both the Nordic countries and the Baltics, and although Iceland's internal labour market is flexible compared with much of continental Europe, it is not as flexible as those in the United States and New Zealand. The Icelandic consumption basket is unlikely to be similar to the Italian one. However, Icelandic monetary policy is certainly not delivering optimal inflation for Iceland and even if the central bank had a policy of offsetting shocks to the real economy in their own right (that is, as distinct from what shocks to the real economy imply for inflation), it clearly has not been effective and it is hard to believe that it would be effective in the future.

For these reasons, Buiter (2000) concluded that even on the conventional macroeconomic stabilisation criteria for an OCA, it made sense for Iceland to adopt the euro. With the spontaneous euroisation of much of the economy that has taken place since then, the ability to conduct an independent monetary policy --- even the best-practice form of inflation targeting with a flexible
exchange rate --- has been further impaired. National monetary independence today makes no sense for Iceland today, even apart from the financial stability considerations we have emphasized in this paper.

A conventional benefit of a common currency area is the reduction in transactions costs. While transactions costs in the financial wholesale markets are miniscule per transaction, volume is high. The European Commission (1990) estimated that these costs were .25 --- 0.40 percent of the total European Community GDP. The króna, as measured by variations in the nominal
and real effective exchange rates, is volatile relative to that of other advanced economies.12 Moreover, the openness and small size of the Icelandic economy makes it inhabitants particularly vulnerable to foreign exchange volatility. Every business and household in Iceland is in the position of having to be a foreign exchange speculator.

There is evidence to support the view that not all households have been wise speculators. Around 80 percent of the foreign currency loans to households, for instance, were denominated in the two currencies with the lowest interest rates; the Japanese yen and the Swiss franc (see Figure 14). Iceland's households have therefore been enthusiastic proponents of the 'carry trade', borrowing where the interest rates are lowest, and forgetting about currency risk.