Economic Commentary: The limits of currency intervention

My heart sank when I read in the Sunday press that the Prime Minister thinks he is engaged in a 'fight to the death' to keep interest rates down. I hope this story emerged from some half-witted Cabinet source and not from Number 10 itself, since anyone with even the most nodding acquaintance with market psychology knows that the best way to keep interest rates down is to appear ready and willing to put them up.

This the Chancellor just about managed to do with his dramatic television appearance in front of the Treasury's brass plate on Wednesday morning, though I doubt whether this appearance was otherwise particularly well advised. Designed to remove the 'merest scintilla of doubt' about the Government's attitude towards devaluation, it served only to increase the scintilla in my own mind.

The statement had the slightest whiff of Jim Mortimer about it - the Jim Mortimer who was general secretary of the Labour Party in 1983 when he announced after a meeting of the National Executive that Michael Foot was quite definitely still leader of the party. Sometimes it is best not to over-stress what should be assumed to be obvious.

Anyway, the Government's strategy for the three weeks leading up to the French referendum on Maastricht has become patently transparent. It is based on a single word: intervention. The Bank of England will intervene in the domestic money markets to hold overnight interest rates below 10 per cent, so that an immediate rise in base rates will be avoided, even if three-month market rates continue to hover close to 11 per cent. Meanwhile, if these relatively low overnight rates are not sufficient to stabilise sterling, the Bank of England will support the currency directly through foreign exchange intervention.

Furthermore, the Chancellor will use Britain's presidency of the EC to stress that intervention will be a joint endeavour conducted by the whole Community, with the implied threat that the Bundesbank itself is ready to buy sterling. This was the sub-text of the EC statement issued from Brussels last Friday, which was apparently forced through by Mr Lamont himself.

Combined might

At the heart of this statement was an announcement that the Basle-Nyborg agreement for combined EC intervention in the currency markets had now been implemented.

Much more of this will no doubt be heard after next weekend's meeting of European finance ministers in Bath. The objective is to persuade the markets that there is no point in selling weak currencies such as sterling and the lira, since in so doing they will have the combined might of all the European central banks (code for the Bundesbank in particular) against them.

But what do the EC agreements on joint intervention actually enable the central banks to do? The ERM rules of the game distinguish between two types of foreign exchange intervention: marginal intervention, which takes place when a currency is at the extreme limit of its permitted fluctuation band, and intra- marginal intervention, which takes place before such a situation is reached. The following rules apply to the two different cases.

Marginal intervention: If any exchange rate in the system should reach the absolute limits permitted in the ERM (as happened to the lira/mark rate last Friday), then both the central banks concerned have an obligation to intervene in the markets between 9am and 5pm Brussels time to prevent the exchange rate trading outside its permitted range. If this should happen to sterling against the mark, then the Bank of England and the Bundesbank will both have a duty to buy sterling for marks, both in unlimited quantities. However, this arrangement is not unlimited in time.

Whenever the Bundesbank starts to acquire sterling as a result of market intervention, the Bank of England is obliged to reimburse it in marks or ecu within 75 days of the end of the calandar month in which the intervention takes place.

A similar arrangement applies to any loans that the Bundesbank makes to the Bank of England to enable the latter to intervene in the exchange markets. Therefore, the Bundesbank is required under this agreement to acquire sterling only for about three months, after which it must be repaid out of the UK's foreign exchange reserves, with the UK accepting all of the foreign exchange risk incurred by the intervention of both central banks.

Intra-marginal intervention: The Basle- Nyborg agreement in 1987 extended the ERM rules to allow for joint intervention before currencies have reached the absolute limits of their bands, since by then it is often too late to reverse the momentum of a speculative attack. Although trumpeted as a break- through for the cause of combined intervention, the requirement on the Bundesbank (or any central bank with a strong currency) to assist weakening currencies is very limited.

While the agreement says that there is a 'presumption' that the Bundesbank should co-operate by purchasing a weakening currency before it reaches its absolute divergence limits in the ERM, it is not automatically required to do so. And assuming that it is willing to co-operate at all, the amount that it is expected to purchase is limited to a maximum of 200 per cent of any country's debtor quota in the EC's short-term monetary support mechanism. In Britain's case, this maximum is equivalent to only Ecu3.87bn ( pounds 2.8bn), which might cover only a day or two of really heavy pressure against sterling.

The EC's arrangements for joint intervention therefore turn out to be heavily circumscribed in both time and quantity. Even in the extreme case of marginal intervention, the central bank of a weakening currency is able to buy only a breathing space of about three months before all support needs to be financed out of its own reserves.

In practice, this means that intervention can be expected to be really useful only for a few weeks, since once the markets know that the reserves are dropping, the selling pressure against a weakening currency is usually exacerbated.

A recent case in point is Italy, where the reserves dropped by almost one-third from dollars 34bn to dollars 24bn in July alone. Once this process starts, the markets know that the central bank concerned cannot afford to keep intervening for much longer, and the EC's joint intervention arrangements do little more than delay the inevitable choice between devaluation and higher interest rates.

Reasonable gamble

The UK is not yet in the dire position of Italy (though, as the graph shows, its foreign exchange reserves, as a proportion of imports, are only slightly higher). It is a reasonable gamble for the Government to use the facilities of the ERM to try to get as far as the French referendum without either increasing base rates or devaluing sterling.

The short-term financing facilities of the ERM were designed with precisely this type of temporary problem in mind and, if the French vote 'yes', then pressures in the ERM might abate, especially if the Bundesbank decides not to raise the Lombard rate in October or November.

However, this really is only a short- term palliative, and anyone who believes that intervention can ease the choice between devaluation and higher base rates for more than a few weeks is, in foreign exchange market parlance, just smoking dope.