Economics: As expected, the European Central Bank (ECB) raised its key interest rate by a quarter percentage point at its meeting yesterday.

This is the second such increase in four months, reinforcing the message that the bottom of the interest rate cycle is well and truly behind us and a period of rising rates is ahead.

The key question now is: how much further are interest rates likely to rise from current levels?

To answer that question it is best to start with where we are now and why we are here. Even after the recent increases, euro-zone interest rates are close to historical lows. As Mr Trichet is happy to remind us, the ECB's so-called "refi" rate, at 2.5 per cent, is lower than the equivalent Bundesbank interest rate ever was.

He is apt to interpret that as a compliment to the ECB's credibility as an inflation-fighting institution, inviting the inference to be drawn that not much more by way of rate increases might be necessary. But there is another, less reassuring, interpretation.

Back in the early years of the decade, when the ECB embarked on a sequence of interest rate cuts that was to culminate in its key rate being reduced to 2 per cent in mid-2003, the world's leading central banks were running scared. The global economy was being buffeted by an unusual sequence of negative shocks, including 9-11, the Iraq crisis, the outbreak of the Sars virus - and all of this against the backdrop of the debilitating effects of the earlier dotcom collapse.

Fears of recession and deflation were in the air. Some respected commentators were even wondering aloud whether the US and euro-zone economies might go the way of Japan, which had at that stage endured a decade-long period of stagnation. The US Federal Reserve was so anxious to guard against these risks that it cut its key interest rate all the way to the unprecedented level of 1 per cent.

An additional factor motivating the actions of the ECB, at least by early 2003, was that the euro was appreciating rapidly against the dollar.

So, the fact that euro-zone interest rates were reduced to historically low levels was due to an extraordinary confluence of circumstances.

The point is that the fears that existed then have since receded. Global economic activity is expanding robustly again. The euro-zone economy is experiencing decent, if by no mean stellar growth. Euro-zone GDP in 2006 is expected to rise more strongly than in any year since 2000.

Talk of deflation has disappeared. Instead, there are some muted concerns about inflation accelerating under the influence of elevated energy prices and some rather more substantial concerns about overblown asset markets, especially property.

None of this is to say that downside risks to economic activity don't exist. Of course they do: global financial imbalances, terrorist outrages, geopolitical tensions and the risk of pandemics spring to mind. But, by comparison with the situation that existed three to five years ago, activity in the major economies looks like it's located somewhere on a reasonably normal business cycle.

So, if economies have in some sense returned to normal, shouldn't interest rates too? Well, obviously the US Fed thinks so. Freed from its earlier fears of deflation/recession, it has been pushing US interest rates back up since mid-2004 and, at 4.5 per cent today, the Fed's key rate is 3.5 per cent points above the level it was at then.

Moreover, the latest signals suggest that this process of normalisation is not over yet. Many analysts see US rates going to 5 per cent or higher in the months ahead. Incidentally, against this background, the euro has been weakening once more against the dollar: at $1.19 today, it is about 12 per cent below its value of March last year. This is presumably a non-trivial consideration for the ECB in its assessment of monetary policy.

What of euro-zone rates? What would represent a "normal" level for the ECB's "refi" rate, given the current state of the euro-zone economy and inflation? Reasonable albeit inexact guidance on this question can be obtained by using a well-known formula in economics called the Taylor Rule. What the Taylor Rule says is that interest rates should be set at a level that reflects, among other things, how close to full capacity an economy is operating and how its inflation rate measures up against the objective of price stability.

Applying the Taylor Rule to the current circumstances of the euro-zone economy yields an interest rate of 3.5-3.75 per cent by my reckoning. That suggests that "normalisation" would require the ECB to raise rates by a further 1-1.25 percentage points above the level reached after yesterday's move.

I'm not sure how this sits relative to general expectations, but my impression is that most market commentators are currently expecting a rather more modest increase than this, at least between now and year-end. What my calculation indicates therefore is the distinct risk that the scale of rate increases in the period ahead will take people by (unpleasant) surprise.

Of course, I'm not offering the 3.5-3.75 per cent range as a forecast - I'm not in the interest rate forecasting business at the moment. Think of this range more as a benchmark. If, over the coming year or so, euro-zone economic activity is more vigorous than is currently expected (around 2 per cent GDP growth) and/or the euro-zone inflation rate accelerates above its current rate (again around 2 per cent), that benchmark will be raised.

Conversely, if growth falters and/or the inflation rate declines, the benchmark will be lowered.

One way or the other, let no-one succumb to the belief that the ECB will shrink from raising rates sharply if that's what it deems necessary. It seems to have been forgotten that the last time the ECB embarked on a rate-tightening course, it did so with considerable enthusiasm: between November 1999 and October 2000 it hiked its key rate from 2.5 per cent to 4.75 per cent.

Jim O'Leary currently lectures in Economics and NUI-Maynooth. He can be contacted at jim.oleary@nuim.ie