Saturday, March 17, 2012

Why the economy isn't splitting in two

The news from last week's national accounts seemed very clear and very worrying: the economy was splitting in two, with the mining-boom states of Queensland and Western Australia roaring off into the future, leaving the rest of Australia going nowhere fast.

Over the year to December, state final demand grew by more than 11 per cent in WA and by 10 per cent in Queensland, but by about 1.5 per cent in the rest of Australia.

Fortunately, the true position isn't nearly as bad as that, as Kathryn Davis, Kevin Lane and David Orsmond explain in an article in the March quarter Reserve Bank Bulletin, issued this week.

The trick was that label "state final demand". When we talk about "growth" in the context of the national accounts we're talking about growth in (real) gross domestic product - the value of all the goods and services produced by the market during a period.

We focus on production because it's production that creates jobs and generates income. The equivalent of GDP at state level is gross state product.

So if you want to compare how the states are travelling you compare the growth in their GSP.

Trouble is, the Bureau of Statistics doesn't publish GSP quarterly, only annually. What it does publish quarterly is state final demand, the national equivalent of which is "domestic final demand".

Because these are the only figures available, the media (and some economists who should know better) have fallen into the habit of assuming state final demand and GSP are much the same thing.

Wrong. State final demand differs from GSP in one minor respect and one major respect: it takes no account of exports and imports. And that's not just overseas exports and imports, it's also exports and imports between the states.

In other words, when you make state final demand a substitute for GSP you're implicitly assuming each state has no trade with either the rest of the world or even the other states. Or that its trade is always in balance.

Guess what? Make such unrealistic assumptions and you get misleading results.

The authors point out that growth in spending on home building and non-mining investment over the year to December didn't vary much between the states. There were two main differences. One was that whereas consumer spending grew by about 3.5 per cent in NSW, Victoria and Queensland, it grew by 6 per cent in WA.

The other difference was the huge growth in mining investment spending in WA and Queensland. This was what did most to explain why their growth in final demand was in double figures whereas NSW and Victoria's demand growth was so modest.

But here's the point: the Reserve estimates that roughly half the spending on mining investment goes on imported equipment. Take this into account and the gap between the mining and non-mining states gets a lot smaller.

Another factor narrowing the gap is that part of the miners' spending on investment (and their ordinary operations) goes on goods and services, such as accounting and consulting services, produced in other states. And some of the workers who fly-in/fly-out take their income home to other states.

To give you an idea of how the shift from state final demand to GSP narrows the gap between the states, let's look at the most recent figures, for 2010-11 as a whole. The final demand figures show spending growth ranging from 1.4 per cent in SA to 6.5 per cent in WA - a spread of 5.1 percentage points.

But the GSP figures show production growth ranging from 0.2 per cent in Queensland (get that) to 3.5 per cent in WA - a spread of 3.3 percentage points. After WA came Victoria on 2.5 per cent, SA on 2.4 per cent, NSW on 2.2 per cent and Tasmania on 0.8 per cent.

In other words, state final demand provided a quite misleading guide to the states' ranking. Queensland does so well on spending but so badly on production because, though it gains from having a fair bit of mining, it loses from being so dependent on tourism (hard-hit by the high dollar).

In the absence of more up-to-date figures for GSP, the trick is to examine independently estimated direct and indirect measures of state activity. If the mining states really were growing five or six times faster than the other states, you'd expect that to mean they had much lower rates of unemployment and much higher rates of inflation than the others.

It's true WA's trend unemployment rate was a very low 4.1 per cent in February, but the other mainland states were all tightly bunched around the national average rate of 5.2 per cent. As for inflation, over the year to December the mining states had the lowest rates rather than the highest.

If the gap between the mining states and the rest turns out to be narrower than you expected it's because you've been misled by all the talk of a two-speed economy: mining in the fast lane, manufacturing in the slow.

In truth, and as the distinguished economist Max Corden, of the University of Melbourne, reminded us this week, it's actually a three-speed economy, with mining in the fast lane and manufacturing (plus other export and import-competing industries) in the slow lane, but with almost all other industries - the non-tradable sector - in the middle lane.

This matters because the non-tradable sector benefits from the mining boom and the high dollar in two ways: from the increase in national income brought about by the high commodity prices, and from the lower prices of imports brought about by the high dollar.

Guess what? This non-tradable sector accounts for the great majority of production and employment in all states bar WA (where mining accounts for an amazing 33 per cent of GSP).

The people of Victoria see their state as weak on mining (true) and heavily dependent on manufacturing. Not true: manufacturing accounts for 8 or 9 per cent of GSP in all states bar WA (5 per cent). Where Victoria and NSW stick out is in their dependence on the business services sector (particularly financial and insurance services), which accounts for 28 per cent and 30 per cent of GSP, respectively, compared with about 17 per cent in the other states.

It's because business services are mainly in the not-hard-hit non-tradable sector that Victoria and NSW aren't travelling too badly compared with the mining states.