Is Australia suffering from a two-speed economy?

Australia has a new “two-speed” economy that comprises: (i) NSW and Victoria (driven largely by Sydney and Melbourne); and (ii) the rest of Australia.

The size and relative strength of the NSW and Victorian economies means that economic data presented at the national level is masking weakness throughout the rest of Australia.

Also read: Trump could help Aussie politics and the economy

Economic activity in NSW and Victoria, along with strong dwelling price growth in Sydney and Melbourne, is likely to see the RBA sit on the sidelines. But the rest of Australia could probably do with more policy easing.

During the mining investment boom it was fashionable to talk about Australia’s “two- speed” economy. Surging commodity prices drove a spectacular lift in activity in the resources sector that sent the WA, QLD and NT economies into full blown growth mode.

Meanwhile, relatively high interest rates (compared to other advanced economies) and a strong AUD weighed heavily on the rest of Australia. As such, the term “two-speed” was coined to characterise the Australian economy.

Fast forward some six years and Australia yet again finds itself with a “two-speed” economy. But this time it’s NSW and Victoria that are driving demand and employment growth while the rest of Australia is feeling the pinch (chart 1). In fact, it’s largely Sydney and Melbourne propping up national outcomes.

Capacity utilisation is high in Australia’s two largest cities and demand growth is strong, courtesy of incredibly low interest rates. As a result, the NSW and Victorian economies have performed much better than the rest of Australia.

The big influence that NSW and Victoria have on the national figures means that pretty soft conditions throughout the rest of Australia have flown under the radar. In fact, economic outcomes in most regions of Australia are probably weaker than the national figures are suggesting unless you’re in Sydney or Melbourne. There are some exceptions, notably the ACT, but the generalisation holds for most places.

In this note we look at how the Australian economy is faring on the new “two-speed” basis. We slice the economy into two pieces – (i) NSW & Victoria (termed VicNSW); and (ii) the rest of Australia (termed ROA). For dwelling prices that are published on a capital city basis we split the data between Sydney & Melbourne (SydMel) and the remaining capital cities (ROA). Basically we are trying gauge how the rest of the Australian economy is going when you strip out NSW and Victoria (or Sydney and Melbourne). Of course, it can be argued that if you exclude the strongest parts then all you are left with is the weakest bits. But that is precisely the point. We want to better understand how the rest of the Australian economy is going outside of NSW and Victoria.

The numbers – setting the scene

To understand what we are analysing it makes sense to briefly look at the size of our two pieces. By Gross State Product (GSP), which is effectively output, NSW is worth 32% of the national pie and Victoria 221⁄2%. So our VicNSW economy is worth 541⁄2% of the national economy which leaves the ROA comprising 451⁄2% (chart 2). By population, NSW has 32% of Australia’s residents and Victoria clocks in at 25%. This leaves 43% of people residing in the ROA. By capital city, 64% of people who live in NSW are in Sydney and 75% of Victoria’s residents are in Melbourne. For the ROA, 62% of people are in the capital cities (chart 3). So our VicNSW and ROA split

Population growth

We kick off with population growth because it heavily influences aggregate growth rates like output and employment. The latest figures show that in the year to QI 2016, population growth in VicNSW was running at 1.6% while it was growing at 1.1% for the ROA (charts 4 and 5). This is a material difference. It largely reflects two forces (i) interstate migration, in particular from WA and SA to Victoria due to a lack of job opportunities; and (ii) a significant proportion of overseas migrants settling in either Sydney or Melbourne.

We need to keep the population growth rates of VicNSW and ROA in mind when looking at the economic data because the ROA is at a “handicap” of about 0.5%pa. Put another way, if the ROA is clocking in at 0.5%pa less than VicNSW for growth rates then performance per capita is effectively the same. We consider this to be important because differing population growth rates can give a misleading impression of economic performance. And per capita measures are a better barometer for gauging changes in living standards (see here). As our analysis shows, even accounting for the 0.5%pa difference in population growth, VicNSW is performing far better than the ROA.

State Final Demand

State Final Demand (SFD) is basically the sum of household and government expenditure and investment. It does not include exports and is therefore a better measure of demand (i.e. the total value of goods and services that are sold domestically) than GSP which includes exports.

The data shows that demand growth has been strong in VicNSW while it has been contracting throughout the ROA since mid-2014 (chart 6). The decline in SFD throughout the ROA reflects falling business investment and weak consumption growth. SFD has been hit particularly hard in the mining regions of WA, QLD and NT where investment in the resources sector has fallen sharply (chart 7). SFD growth has been tepid in SA and Tas leaving the ACT as the only bright spot. All up, demand growth in the ROA is consistent with very soft economies.

Strong demand growth in VicNSW has been assisted by the positive wealth effect from rising dwelling prices. Higher asset prices have supported household consumption growth and dwelling investment in Australia’s two largest states. These two economies did not experience a downturn in mining investment and were able to benefit from both the fall in interest rates and the AUD.

Employment growth

There is plenty of evidence of our new “two-speed” economy in the labour market data. Jobs growth has slowed nationally. But the slowdown in jobs growth for the ROA has been sharp. In fact, there has been a contraction in the level of employment for the ROA through the year to October (chart 8). The number of people working in the ROA has fallen by 40k over the past year. The reality is that jobs growth in Australia has been driven by NSW and Victoria over the past year. For the ROA, the labour market is incredibly weak. In some states, conditions in the labour market are commensurate with a recession, particularly WA (chart 9).

Unemployment rate

The national unemployment rate has been trending down over the past year thanks to a decline in participation. But when we look at the economy through our “two- speed” lens it becomes apparent that the unemployment rate for the ROA has been stuck above 6% for two years (chart 10). It has basically trended sideways since mid-2015 while the unemployment rate has been falling in VicNSW. A disappointing story for the ROA which reflects a decline in the level of employment offset by a fall in participation.

Of course new jobs have been “created” in the jurisdictions that comprise ROA. But there have also been a lot of job losses, primarily in the mining and manufacturing sectors. To be fair to our Territories, the unemployment rate is low in the ACT and NT. But it’s well above the level associated with full employment (considered to be around 5%) in QLD, SA, WA and Tasmania (chart 11).

In our view, the jobless rate for the ROA is likely to sit a touch above 6.0% over the next six months. It will continue to be a different story in VicNSW and we see the unemployment rate holding at around 51⁄4%. A big lift in public investment in NSW should keep the unemployment rate anchored around 5%.

It’s worth noting here that the unemployment rate is not proving as useful as usual to determine the amount of slack in the labour market. That’s because underemployment has bene rising while the unemployment rate has been falling dueto the casualisation of the workforce (see here). As a result, wages growth has been easing.

Business investment

Investment in VicNSW has trended up recently because of a lift in capex in NSW.But for the ROA business investment is down 17.2% through the year to QII 2016 (chart 12). The big decline has been driven by a huge fall in spending in QLD and WA because of the downturn in mining capex. There has also been a decline in capex in SA, Tas, and NT. For the ROA, only the ACT has posted a lift in business investment. This is a pretty poor result considering the stimulus that has come through from lower interest rates and a lower AUD. It should be a concern to policymakers.

We expect business investment to continue to decline in the ROA as resource related projects are completed. The pace of the downturn, however, will start to slow as the worst of the downturn in mining capex is now behind us. The outlook for the non- mining part of the economy remains mixed. Our forecasts have non-mining capex lifting a little over 2016/17 which should see VicNSW post a further modest lift in business investment.

Dwelling Prices

National dwelling prices have risen substantially over the past 31⁄2 years. But the data shows that the lift in dwelling prices has largely been a Sydney and Melbourne story (chart 13). Demand for housing has been strong in Australia’s two largest cities, driven by a variety of factors which include: (i) interest rate cuts; (ii) a high level of investor demand; (ii) foreign investment; (iii) strong population growth; and (iv) some supply issues (particularly in Sydney).

For the ROA, however, dwelling price growth has been much more modest. In fact, dwelling prices have been falling for a few years in Perth (chart 14). Price rises in most of the other capital cities have been significantly lower than those in Sydney in Melbourne.

Commentators often make the mistake of generalising about dwelling prices in Australia and assume that the housing market is homogeneous. But the data shows that there is a significant divergence between price growth across Australia’s capital cities.

Building approvals

Strong dwelling price growth in Sydney and Melbourne has translated into a big lift in building approvals, particularly for apartments. For the ROA, however, building approvals have eased. On a rolling annual total basis dwelling approvals are down 7.1%pa for the ROA (chart 15). There is a bit of a multispeed story for approvals in the ROA. There has been a big fall in approvals in WA as vacancy rates rise and dwelling prices fall. Approvals in QLD have actually been quite strong and a big lift in apartment supply will hit the Brisbane market over the next few years. But when we look at the ROA as a collective, dwelling investment will cool. For VicNSW, however, there is still a lot of residential construction in the pipeline. The supply response is welcome given the big lift in dwelling prices.

Are there implications for monetary policy?

The RBA has recently shifted from an implied easing bias to a neutral tone. And market pricing for a rate cut has receded (chart 16). We think the RBA’s reluctance to ease policy further is generally related to the strength of the NSW and Victorian economies, in particular the exceptionally strong housing markets of Sydney and

Melbourne, and the lift in commodity prices. But the economic data suggests that the ROA could do with more policy easing.

Last week’s wages and employment data were weak. Normally, softness in the labour market and below-target inflation would imply that policy easing is forthcoming. A lower cash rate would, among other things, push the AUD down which would boost activity in Australia’s tradables sector. But clearly the RBA has a degree of discomfort at the strength of dwelling price growth in Sydney and Melbourne that has taken a near term rate cut off the table.

The lesson here is that the new “two-speed” economy means that current monetary policy settings are likely to be suboptimal for the ROA outside of Sydney and Melbourne. If the Sydney and Melbourne property markets stall or falter then policy easing will probably come back onto the agenda because the ROA could do with more stimulus. Of course that stimulus could also come via fiscal policy as we have previously argued (see here). But that’s a conversation for another day.