On the ground in key electorates

Fewer than a dozen electorates with three key things in common will decide the election result.

The 11 electorates that matter for the 2019 Federal Election are: Lindsay, Gilmore, Robertson (NSW), Longman, Herbert, Capricornia, Forde, Flynn (QLD), Cowan (WA), Hindmarsh (SA) and Chisholm (VIC).

There are 150 federal electorates comprising the Australian parliament, each with roughly 100,000 voters.

At the upcoming May election less than a dozen electorates will determine the course of the next three years, which will have direct implications for the property industry in terms of negative gearing and capital gains tax and possibly immigration policy.

Politicians get a fix on the leaning of swing electorates by polling, but as the last US presidential election showed, voters don’t always accurately tell pollsters their intentions especially if they feel they’re likely to be judged for not voting “with the herd”.

I have also seen modelling of mortgage stress and income inequality in swing electorates, but I am never persuaded by this approach. These assessments are largely based on data from the last census, which means that the electorate reading is dated.

What is required is an interpretation of how life has changed since the last election in swing electorates based on data published in the intervening years.

And the reason why this hasn’t really been done is because it’s technically difficult. There’s limited data and the boundaries don’t always align with electorates, thus requiring some adjustments.

We have assembled data published by the Australian Bureau of Statistics by small areas for two periods: in the three years leading up to the 2016 election and in the three years to the 2019 election.

The data relates to unemployment, to the formation of small business and sole trader activity, and to population growth rates.

Ready, set, vote. Australia goes to the polls within weeks.

There are 11 electorates on a margin of less than 2 per cent including five held by Labor and six by the Coalition. Five are located in Queensland. Four are what I would call outer suburban. Three are influenced by coal mining activities. Two are shaped by retirement and lifestyle.

The number of voters in each electorate ranges from 96,000 (in Cowan) to 116,000 (in Gilmore) meaning, technically, there is a voter exchange rate. Let’s say you were a politician looking to curry favour with promises: a dollar promised in a small electorate goes further than a dollar promised in a large electorate.

Let’s look at the data.

There has been a slowdown in the rate of population growth over the last three years compared with the preceding three years, in Lindsay (down by one percentage point) and Cowan (down two percentage points), suggesting locals might feel the crush is easing.

But in Capricornia, Forde and Flynn the numbers have gone the other way, with rising growth rates since 2016. In Capricornia and Flynn, population levels had been either static or contracting, so a return to growth would be reassuring to locals.

In outer suburban Forde on the other hand a one percentage point lift in the rate of growth during this term might make locals more sensitive to congestion and more responsive to infrastructure announcements.

One of the most underrated datasets published by the ABS is small business activity. I regard it as a barometer of entrepreneurial confidence — and it can be calibrated to electoral boundaries.

The number of small businesses employing one to 19 workers, which in aggregate accounts for half the nation’s workforce, ratcheted upwards in the 2015-2018 period compared with the 2012-2015 period in the electorates of Lindsay, Longman, Gilmore and Robertson. In Longman, for example, the number of small businesses employing one to 19 workers increased by 93 over three years to 2834 in 2015, but in the following three years the increase was 297. Small businesses are on the rise in each of these electorates, especially in Longman.

SWING SHIFTS: KEY INDICATORS BETWEEN ELECTIONS FOR MARGINAL ELECTORATES

On the other hand, the rate of small business formation has slowed in Cowan, Herbert and Chisholm. The contraction in Cowan is instructive: in the three years to June 2015 some 511 net new small businesses were formed but in the following three years just 49 were formed.

Similar data on sole traders can be viewed two ways. More sole traders could mean there is greater confidence in the local economy. Or it could mean there’s greater desperation as retrenched workers register as Uber drivers.

The sole trader workforce has surged in every marginal electorate in the last electoral term. In Lindsay, for example, over the three years to 2015 the number of sole traders contracted by 391 to 5586 whereas in the following three years to 2018 the number increased by 848 to 6434. Sole traders are on the rise in Lindsay.

In Chisholm, net growth in sole traders went from 379 in the former period to 2134 in the latter.

In the coalmining electorate of Flynn (centred on Emerald in the Bowen Basin) sole traders contracted by 3186 in the lead up to the last election, but this number has since increased by 139. Flynn has gone from losing sole traders to gaining them.

Rising small business numbers and even rising sole-trader numbers are likely to imbue an electorate with a subliminal measure of confidence in the future.

At a macro level Australia’s unemployment rate has dropped by one percentage point between December 2015 (6.2 per cent) and December last year (5.1 per cent).

However, in places like Herbert (Townsville) the unemployment rate pushed out from 7.4 per cent in December 2015 to 8.2 per cent in December 2018.

The job market has also worsened in Longman, with unemployment over this period rising from 5.9 per cent to 7.3 per cent.

Unemployment rates have also deteriorated in Gladstone in Capricornia (up one percentage point) and in Eurobodalla in Gilmore (up 3.3 percentage points).

The PM is expected to call the election within days. Picture: Jeremy Piper

There’s a happier story in Robertson where the unemployment rate has fallen from 6.8 per cent in December 2015 to 5.7 per cent last December. The unemployment rate has also improved in Lindsay, Cowan, Hindmarsh, Chisholm, Forde as well as in most parts of Gilmore and most parts of Flynn.

In Whitsunday Shire in Capricornia the unemployment rate has plummeted from 9.6 per cent to 3.6 per cent over these years.

There are several insights to take from this. Wage growth might have been modest but overall more people are in work today compared with the time of the last election, and especially in most of the swing electorates. This is a plus for any incumbent politician.

There has also been a mighty shift into sole-trader territory which, if by choice, is a wonderful reflection of our nation’s entrepreneurial spirit. But if it’s because the nature of work is changing and Uber-driving is the only option then this is likely to lead to a rejection of the incumbent.

The point is the electorates that are likely to shape future policy, including policies that directly impact the property industry, have experienced a mix of economic fortunes over the last three years.

Astute politicians will develop nuanced response for each electorate, which is easy enough at a local level. However, it’s tricky at the macro level because any overarching narrative must reconcile coalmining electorates with those of city-based electorates.

Bernard Salt is managing director of The Demographics Group. Research by Simon Kuestenmacher.

[e-mail]