Stephen King: Waitress's tip on a new wage-price relationship

Saturday was one of those days when cooking just seemed to be too much effort, so we went with our children to have lunch in a restaurant in Camden Town.

In the evening, having left the kids at home, we popped out for a bit of supper in a rather hip restaurant in Fitzrovia. At both establishments, we failed to spot a single British waiter.

In the evening, I asked our waitress - a woman with a distinctly Slavic accent - how many British people were on the restaurant's waiting staff. She wasn't sure but she thought that perhaps one waiter was British. The rest - and there were plenty of them - came from further afield. Admittedly, London is a cosmopolitan melting pot, an international city that is unlikely to be a reflection of Britain as a whole. However, our experience on Saturday was nothing unusual. Anyone who eats out regularly in London will be aware of the slow demise of the British waiter.

My guess is that members of the Monetary Policy Committee (MPC) of the Bank of England also eat out a lot, because they too have recognised this trend. In their case, though, the importance lies not so much with the accent of the person bringing to your table thirst-quenching bottles of Chardonnay and (still or sparkling?) water but, rather, with what the accent implies for the supply of labour and the flexibility of Britain's labour market.

Two charts in the Bank of England's February Inflation Report provide substance to add to Saturday's anecdotal evidence.

The first chart shows the results of a survey carried out in November 2005 by the Bank's regional agents. The agents hoped to find out which categories of employment tended to rely on a large migrant workforce. At the top of the pile, as you might expect, was employment in agriculture, hotels and restaurants.

The second chart shows developments in basic and overtime pay in 2005 across roughly the same categories of employment. The biggest declines in pay were, spookily enough, in agriculture, hotels and restaurants.

Needless to say, these changes reflect the forces of globalisation. No longer is the UK labour market separated from the rest of the world.

In the past, manufacturing workers were exposed to the forces of global competition through trade, but other sectors were far less affected by overseas developments. Nowadays, competition is far greater. As a result, the old indicators of labour-market tightness no longer seem to be working very well. The UK unemployment rate has been low for some time now but you would be hard-pressed to link this measure of labour-market tightness with rising wages.

Faced with a shortage of workers, companies now look to hire people from abroad, not pay their existing staff a higher wage. Threatened with higher energy prices, companies look to improve their profitability by reducing their labour costs, either through labour immigration or through capital emigration. So much for the old-fashioned wage-price spiral.

How are policymakers reacting to these changes? An obvious difficulty lies with the predictability of economic relationships.

Most economic models assume that the future is somehow linked to the past. In a trivial sense, of course, this is always true. But models are simplifications of reality and, all too often, the simplifications turn out to be inappropriate.

If the model assumes, for example, that wages go up when the unemployment rate is low, but allows no room for capital and labour migration, it will almost certainly deliver an overly aggressive wage response.

My sense is that policymakers are still reluctant to place too much emphasis on these structural changes. Because they're structural, they're rather nebulous and thus difficult to measure. Knowing, therefore, that these effects may be important is not the same as knowing the magnitude of their importance.

Thus, when policymakers discuss risks, they still have a habit of talking about the risks that exist within the bounds of established relationships, rather than talking about the possibility that those relationships may have permanently shifted. For example, the Bank's Inflation Report suggests that the "substantial risks" around the MPC's central projections for inflation include "the outlook for consumer spending; the prospects for net exports; the sustainability of low long-term interest rates; the margin of spare capacity; and the evolution of energy prices and their impact on inflation".

A similar approach is adopted by the Federal Reserve. Coinciding with Ben Bernanke's first Congressional testimonies last week, the Federal Reserve published its latest Monetary Policy Report to Congress.

Again, there was a discussion of inflation risks that were mostly within the bounds of existing relationships. As the Fed put it: "Some of the uncertainty is centred on the prospects for the housing sector ... some observers believe that home values have moved above levels that can be supported by fundamentals and that some realignment is warranted ... [but] if home values continue to register outsized increases ... higher resource utilisation would risk adding to inflation pressures ... Additional steep increases in the price of energy would intensify cost pressures and weigh on economic activity."

For the most part, these risks are what I would deem cyclical risks: if resource utilisation is higher by a factor of X, then inflation will be higher by a factor of Y. If costs are higher by a factor of A, then inflation will be higher by a factor of B. In the light of structural changes, though, it may be that inflation, although expected to be higher by a factor of Y or B is, instead, higher by a factor of Z or C.

Put another way, central banks have every right to be uncertain about the inputs that enter into the inflation process - the level of demand, the amount of spare capacity - but, if truth be told, they should also be uncertain about the relationship between inputs and outputs.

It's this relationship, after all, which is more vulnerable to the kinds of structural change that globalisation brings about. Oil and gas prices may be rising but a higher supply of immigrants into the UK labour market has helped keep inflationary pressures under control.

Let's say, though, that wages eventually do begin to pick up in the light of either higher energy prices or, particularly in the US, pressure on resource utilisation stemming from a buoyant housing market.Does this mean that higher inflation is inevitable? The late 1990s provide some helpful clues. Then, markets were convinced that the US economy, in particular, was enjoying a "new paradigm". The mix of strong growth with low inflation was seductive, luring many an investor into buying equities at just the wrong moment.

The late 1990s led to plenty of pressure on resource utilisation - US unemployment fell to very low levels - but the absence of inflation was a bit of a puzzle.

Many argued at the time that this was a result of faster productivity growth. True to a degree, but other factors were also important.

Despite the productivity gains, overall labour costs were rising rapidly and the only reason that inflation didn't accelerate was a major decline in corporate profitability. Costs went up but, with no pricing power, it was the profits cycle, not the inflation cycle, that brought the US boom to an end.

And this is yet another example of globalisation at work. Increasingly, in today's internationally competitive markets with free flows of information across borders, companies are price-takers, not price-setters. It's a lot more difficult these days to establish the wage-price spirals of old. At least, that's what the waitress told me - indirectly - on Saturday night.

Stephen King is managing director of economics at HSBC

stephen.king@hsbcib.com