Stephen King: National boundaries become an obstacle to understanding

Imagine a world without thermometers, without clocks and without measuring rods. You wouldn't be able to take your temperature, tell the time or gauge your height.

All these things are obvious. Yet, even in this strange world, you'd be distinctly odd were you to conclude that you had no temperature, that day and night didn't occur, or that your body was of no particular length. Measuring devices allow us to calibrate things we know are there but which, with the naked eye, we find difficult to estimate with any real precision.

Economists all too often fall into the trap of thinking that an absence of precise measurements implies that a relationship perceived with the naked eye doesn't actually exist. Such is their dependency on supposedly precise econometric models that they are unable to find room for those things that certainly matter but which can't be incorporated into the mock-precision of spreadsheets.

Published data provides an all-too-comforting degree of accuracy that, in hindsight, can prove to be entirely spurious. For example, economists, alongside traders, are obsessed with monthly movements in US employment, but provisional estimates are not much more than educated guesses. Just over a week ago, US statisticians revealed they had found an extra 800,000 people working in the United States in the early months of 2006. This ability to rewrite history suggests that even the most closely-watched data isn't all it's cracked up to be.

The biggest problems, though, are associated with international economic relationships. We tend to focus on national economic data because that's what our statisticians give us. Politicians love this sort of thing because they can demonstrate how well their country is doing relative to the competition or, if they're in opposition, how the government of the day is wrecking our economic prospects.

In reality, though, national boundaries are, in economic terms, becoming no more than misleading conceits. They are obstacles, not aids, to our economic understanding. Companies are no longer constrained by national borders: their activities spread far and wide around the world.

Workers, also, are no longer constrained by national borders: if there's demand for their services elsewhere in the world, they can move (Polish workers are in heavy demand in the UK, although, perhaps not surprisingly, few English footballers - particularly goalkeepers - are in demand in continental soccer leagues).

These changes are important because they affect economic behaviour. Yet most economic models, built for the most part on national data, are poorly designed to capture these behavioural changes. Obviously, national data incorporate both exports and imports, but these hardly capture the full impact of globalisation. Because of strong growth in emerging markets, globalisation has raised commodity prices. Through an increase in effective labour supply - from eastern Europe, China, India, Mexico and others - globalisation has placed downward pressure on wages in developed countries. And capital markets are no longer quite so dependent on the decisions made by policymakers at the national level.

Ironically, globalisation demands a re-think of the relationships espoused by, among others, Edmund Phelps, this year's Nobel Prize winner for economics. In the late 1960s Phelps, alongside Friedman, was a fierce critic of the post-war Keynesian orthodoxy. He took particular exception to the idea that simple correlations between data could be easily exploited by policymakers.

His concerns were best demonstrated through his attack on the Phillips curve. Policymakers thought they had a choice: in return for a slightly higher inflation rate, they could deliver lower unemployment. Phelps's brilliant insight was to emphasise the role of expectations. If we, as citizens, know that inflation is likely to be higher, we'll simply demand higher wages. By doing so, there will be no reduction in unemployment and only a rise in inflation. As a result, there's a "natural" rate of unemployment that no amount of monetary manipulation is able to budge.

Over the last 15 years, UK experience seems to have overturned even Phelps's claim. During this period, the unemployment rate has moved all over the place but inflation has been broadly stable. Importantly, inflation has remained low when the labour market has appeared to be "tight". While this partly reflects the credibility of the Bank of England (the Bank understands the Phelps critique and would not be silly enough to exploit simple correlations), it also reflects the impact of globalisation. Labour supply to the UK labour market is no longer governed by the size of the indigenous adult population while the price of labour is increasingly governed by the degree to which capital can head elsewhere.

If all this is true, perhaps we have to re-examine other aspects of the economic framework used by the Bank of England and others. Most macroeconomic frameworks make use of the output gap. This measures the degree to which output is above or below the level - known as productive potential - that's consistent with price stability. Globalisation, though, surely undermines this concept. Productive potential surely can't be governed by the resources within any one economy alone.

Richard Fisher, the president of the Federal Reserve Bank of Dallas and the guest speaker at HSBC's Global Investment Seminar last week, thinks we haven't gone anywhere near far enough in understanding the economic world we're now living in.

In his words, "in US monetary policymaking circles, we work with cost calculations, assumptions about production functions and formulae for policy optimisation that were developed before the world economic map was redrawn by the entry of new players and new technologies that changed our ways of doing things... Fed economists can certainly count; in my view, better than anybody. But are they counting the right things?"

Fisher's point is a serious one. Not only are we failing to count the right things, we're not even close to measuring them. The great advantage of measuring activity within borders is that most of that activity is in a common currency, even if that currency is constantly at the mercy of changes in the inflation rate.

When we start aggregating activity across borders, the challenges become a lot bigger. Which exchange rate should we use? What happens if exchange rates fluctuate from year to year? What about different inflation rates? Can we trust all countries to deliver the same quality of data in a consistent and timely fashion?

The chart demonstrates the problem all too clearly. It provides two measures of global GDP growth. One is based on the size, in dollars, of each country's GDP output in 2003 (in other words, using nominal GDP weights). Using this information, we can add together volume growth rates across countries and work out average global growth over a number of years. The other is based on so-called purchasing power parity (PPP) weights.

The use of PPP is an attempt to get rid of the distortions that crop up as a result of exchange rate volatility from one year to the next. In simple terms, PPP assumes that the cost of a good or service is the same all over the world. But this is a ludicrously simplistic assumption. Someone from Xian would feel distinctly short-changed were they to get their hair cut in Beverly Hills.

Despite this disadvantage, many economists feel comfortable with the use of PPP. Calculated on this methodology, global growth has been a lot faster. But which set of numbers is right? Both approaches have their strengths and weaknesses, but the different outcomes suggest that we are a long way from knowing the truth about the global economy.

Globalisation provides a serious threat to our accepted economic norms. In particular, it undermines the sanctity of national accounts data. Perhaps that's why so many economists prefer to ignore it. Through their clever equations, they can pretend that they know a lot about the world. In truth, though, they've created a world that is becoming increasingly disconnected from the only one that matters.

Stephen King is managing director of economics at HSBC

stephen.king@hsbcib.com