Tuesday, May 18, 2010

Problems with New England statistics

Greg wrote in a comment on New State arguments 1 - introduction:

Much is made of Australia's "two speed" economy with WA cited as the boom state on the back of mining. Meanwhile an accusing finger is pointed at NSW as being a drag on the national economy.

However, I cannot find any data about how New England fares in the scheme of things. I suspect that it is doing a great deal better than the rest of NSW. Hunter coal is certainly a vital contributor. If NSW looks sick despite having such a strong regional performance in the north then the rest of the state must be in an even poorer shape than the figures show.

Is there any data that shows the relative performance of NSW regions and how that impacts on planning and infrastructure investment for the regions and for NSW as a whole? If it is there then it is hard to find. The statistics all seem to be compiled on a state by state basis without regard to relative diversity of state economies and their respective regional areas. That makes it hard to get a feel for what is really happening.

Greg's comment goes to a heart of a problem that I have talked about before on this blog. Because New England does not exist as a constitutional entity, it is not recognised in the statistics. We have less statistics for New England, population 1.4 million, than we do for Tasmania, the ACT or the Northern Territory,

The general state wide statistics are dominated by Sydney.

Around 63% of the NSW population of around 7 million lives in what is called the Sydney Statistical Division. This is not quite the same as the traditional definition of Sydney, for it includes the Blue Mountains and Central Coast. Further, because most of the higher level jobs are in Sydney, this pulls up Sydney averages.

The census data provides a snapshot of conditions at a point in time. Census data is collected on a  collection district basis, and then totaled to, for example, LGAs. You can get ABS to do special runs by specifying coverage. I generally work by aggregating LGAs because this is simpler, although you have to watch changing LGA boundaries.

Census data forms the basis for much state government planning because it is the most complete. Between censuses, the State Government uses projections based on past trends for planning purposes. This is usually done at a regional level, although in some cases small area studies are used. Some but not all the regional data is published.

A particular problem here is that the Government uses various regional definitions depending on its purposes. An example is the frequent inclusion of the Clarence Valley in the Mid North Coast as compared to the Northern Rivers. This can make it hard to put the data into a consistent structure.

Between census periods, very little social or economic performance data is available on a sub-state level. This actually creates significant planning problems where resource allocation decisions are involved.

Some small area tourism data is collected by ABS, while ABS also provides building approval data on an LGA basis plus some LGA population estimates. Housing NSW rent and sale reports provide some rental and sales. This is on an LGA basis for the Greater Metropolitan Region (this includes the Lower Hunter), regionally aggregated for the rest of the state.

The NSW Business Chamber/Commonwealth Bank publishes a survey of business conditions. This includes a break-up for both Illawarra and Hunter regions, as well as aggregated data for the rest of the state outside Sydney, Hunter and Illawarra. The most recent data did indeed suggest that business conditions in the Hunter were more positive than those in Sydney.

As you can see, it's quite difficult to build up a clear picture of key variables and trends across the North. Quite a bit of my own analysis over the last year or so has been concerned just to lay down a base for further analysis. See, for example:

I have at least one more post to go to complete this current series.  

No comments: