Month: October 2014

Consumer prices and the Australian economy – September 2014

Not long ago, I wrote a post arguing why I thought GDP growth was too low. I outlined several reasons in support of this argument: 1) that the pool of unemployed persons has been growing for a duration similar to that of the early 90’s recession and 2) real income per capita was no longer growing. The recent release of Q3 2014 CPI data provides some further evidence supporting this argument. Where aggregate demand is growing faster than its potential rate, demand for resources generally places upward pressure on prices and unemployment declines. The release of Q3 CPI shows that consumer price growth in Australia has started to slow, with growth of core inflation now in the middle of the RBA range of between 2 and 3%. Over the last few quarters, annual CPI growth was at the upper end of the RBA band which seemed at odds with the idea that growth in Australia was too low. There were in fact suggestions that the RBA would need to hike rates. But quarterly CPI growth and other price measures such as wages, commodity prices and Terms of Trade (ToT) have been pointing to an easing in the level of price growth across the economy. Whilst a rate of core CPI right in the middle of the RBA range doesn’t sound like a problem, it’s in combination with indicators such as unemployment growth and income stagnation that point to this as another symptom of a bigger/broader issue.

This situation is not limited to Australia. Globally, price pressures have been easing, with growing concerns of emerging deflation throughout Europe, US, UK and China according to data recently published by The Economist (25th Oct 2014). It’s difficult to pinpoint just one reason for this phenomenon, but slowing global growth is a good starting point. For Australia, the reversal of the ToT boom has been a fairly significant factor associated with slower income growth, growing unemployment and now slowing price growth.

Highlights of CPI Quarter 3 2014

The Sept quarter CPI growth was 0.5%, the same rate of growth in the previous quarter. Quarterly growth at 0.5% is just below the average for Sept CPI growth over the last 10 years of +0.8%.

The quarterly growth trend has been slowing throughout the last four quarters:-

Source: ABS

The annual rate of growth slowed significantly in the Sept quarter from 3% to 2.2% (seasonally adjusted). This large slow-down in annual growth is partly the result of the shift to a higher base used to calculate growth – Sept 2013 quarterly growth was relatively strong at +1.2% (see chart above). That said, the quarterly growth since Sept 13 has been slowing, resulting in annual CPI growth closer to 2% – at the lower end of the RBA range – and reaching this point in a relatively short period of time.

The measures of core CPI growth provide a better guide as to the underlying price growth. Growth in measures of ‘core’ CPI, the trimmed mean and weighted median, eased somewhat in Sept and remain in the middle of the RBA range. The trimmed mean view of consumer prices is the main focus for RBA assessment.

Source: ABS

For the Sep 2014 qtr;

  • Trimmed mean (15% of the smallest & largest price movements are removed or ‘trimmed’) +0.4% quarter on quarter and +2.5% year on year (a slowing of the annual rate)
  • Weighted median (price change at the 50% percentile by weight of the distribution of price changes) +0.6% quarter on quarter and +2.6% year on year (no change in the annual rate)

The biggest contributors to quarterly CPI growth was in Food and Housing categories:-

Source: ABS

The biggest price pressures under Housing were Purchase of New Dwellings and Property Rates and Charges. Both offset the large -0.14 % pts decline in Utilities (Electricity) that also fall under this category.

There was one single significant contribution to growth in Food CPI and that was in the Fruit category.

The big turnaround for the quarter was the slow-down in Health costs, from +0.17% pts last quarter to -0.1% pts this quarter.

Overall, CPI growth is slowing, but for the most part is sitting right in the middle of the RBA range. Based on this, it’s unlikely that there will be pressure to raise rates.

But there is a broader context to this CPI report…

Back in Sept 2012, Assistant RBA Governor, Christopher Kent, gave a speech to the Structural Change and the Rise of Asia Conference titled Implications for the Australian Economy of Strong Growth in Asia. This speech laid out the broad set of factors effecting the Australian economy, namely the impact of economic growth in Asia (China) on driving our Mining boom. The most visible impacts in Australia were the rise in the ToT, an appreciating exchange rate and the growth or “reallocation of productive factors” to resources and resources related industries.

It’s relevant to revisit some of the points of this speech as they relate to the impact on prices and growth now that the positive ‘shock’ to the ToT has started to reverse and the economy has commenced the transition from phase II (investment) to phase III (production and export) of the boom.

“The positive shock to the terms of trade, resulting from a rise in commodity prices, increases income accruing to the resource sector and increases that sector’s demand for productive inputs. Both of these exert a measure of inflationary pressure.” Christopher Kent, RBA Assistant Governor, 19th Sept 2012

“Domestic inflationary pressures, associated with higher wages and incomes, will lead to higher inflation for non-tradable goods and services but, at the same time, the gradual pass through of the initial exchange rate appreciation will lead to lower inflation for tradable goods and services (whose prices in foreign currency terms depend to a significant extent on global considerations). In this way, the appreciation of the exchange rate helps to offset the inflationary impulse from the terms of trade shock, and assists in maintaining inflation in line with the inflation target.” Christopher Kent, RBA Assistant Governor, 19th Sept 2012

We’ve seen all of this happen in the Australian economy.

Firstly, the rise in the ToT generated higher income growth.

Source: ABS

Once the ToT started to appreciate, real GDI started to grow much faster than real GDP – the difference being the impact of the ToT. The ToT peaked during Sept qtr 2011 and is now 21% below that peak. While the ToT did peak, it is still well above its historical levels for the moment – but income growth has stalled nonetheless. More on this later.

Secondly, the exchange rate did appreciate during the investment phase of the boom.

Source: RBA

The real AUD TWI appreciated during the years between 2000 and 2011, with the exception of 2008/9 (GFC). Growth in the exchange rate started to slow from June 2011 but importantly, remained elevated until it peaked in March 2013. Since then, the real AUD TWI has fallen by 7%.

Finally, there was greater inflationary pressure in the non-tradable sector than the tradable sector (see definitions and detail of each here) during the period of the investment phase of the Mining boom.

Source: ABS

It’s hard to look at this chart and ascertain a ‘neat’ relationship between exchange rate changes and annual tradable inflation. The range for tradable CPI growth has been between +4% and -2% during this time, but has generally been lower than non-tradable inflation growth. This can be shown more clearly by reproducing one of the charts from the RBA speech – the ratio of non-tradable CPI to tradable CPI.

This first chart is from the RBA speech as it provides the historical context. The second chart is updated using the latest data (with as much history as available from the ABS).

This first chart reinforces that since the start of the ToT boom, non-tradable inflation has grown much faster than tradable.

The updated data show a fairly important change in that trend – non-tradable CPI growth started to slow from March 2013:-

Source: ABS, The Macroeconomic Project

This is an unusually long period of no change in this ratio, given the growth in non-tradable CPI during recent history. Prices are still growing, but at a slower rate and this is could be an important indicator of slower demand in the domestic economy.

Since June 2013, non-tradable CPI has started to contribute less to total CPI growth. In Sept 2014, non-tradable CPI made its lowest contribution to total CPI growth (contribution data is only available back to June 2012). Since June 2012, tradable CPI also started to have an increasingly positive contribution to CPI growth, but it too made a smaller contribution to CPI growth in the recent quarter. Both tradable and non-tradable CPI slowed in the latest quarter:-

Source: ABS

There are 47 categories classified as ‘tradable’ – which contributed +0.76% pts to CPI growth. In over half of those categories (26), prices are either flat or declining (YoY -0.39% pts), 17 categories contributed +0.31% pts and the top 4 categories contributed the bulk of the price growth +0.86% pts. The categories were Tobacco, Fruit, Vegetables and International Holiday & Travel. The increase in Tobacco prices is due to an excise increase.

By contrast, there are 40 categories classified as non-tradable, which contributed 1.58% pts to overall CPI growth. In only 12 out of 40 categories are prices flat or declining, contributing -0.26% pts to overall growth. The bulk of the growth in non-tradable CPI comes from the ‘middle’ 24 categories which contributed +0.94% pts. The top 4 non-tradable categories still punched above their weight adding +0.87% pts. The top 4 categories are (in order) Purchase of New Dwellings, Medical & Hospital Services, Rents and Other Services in respect of Motor Vehicles.

There is broader pressure i.e. more categories contributing to CPI growth, in the non-tradable sector and our housing market (new dwellings anyway) is driving one of the biggest parts of that growth.

This brings us to the present day – and we are now seeing the opposite effects take place as the ToT boom reverses.

As mentioned earlier, the ToT has declined by 21% from its peak and the most important thing about this in relation to the CPI is the impact on income growth. The most accurate representation of the income effect is to look at Real Net National Disposable Income (NNDI) per capita. During the ToT boom (2000-2011), growth averaged 2.9% (not including the GFC). Since the ToT peaked in Sept 2011, income growth has averaged 0%.

Source: ABS

In per capita dollar terms, real NNDI has declined by -2.6% since its peak in Sept 2011 (ToT peaked at the same time). This isn’t a huge drop (the chart above measures growth not the per capita value) and the decline is not a short and deep correction that you would see associated with a recession, but rather, income growth has stagnated (at best) over a somewhat extended period of time. Unfortunately, further declines in commodity prices are expected and this is likely to maintain pressure on income growth. If National income growth remains low, it’s likely that CPI growth, especially non-tradable CPI, will continue to slow.

Unfortunately, the exchange rate stayed high during the initial falls in commodity prices and the ToT (the ToT started falling from Sept 2011 and the real AUD TWI only started to decline from March 2013). Elsewhere in the economy, it’s likely that the final stages of the Mining investment phase, a continued housing boom (which requires some funding from overseas markets to maintain loan growth) and relatively higher interest rates in Australia since the GFC have kept the exchange rate higher than expected. This has placed some local non-resources industries and the resources sector (due to falling commodity prices and the scramble to cut costs) at a disadvantage. So far the real AUD TWI has only fallen by 7% and the AUD/USD has fallen by 20% but remains above the level that the RBA deems as its ‘magic spot’ of between US$0.80 and US$0.85 (SMH, IMF: Australian dollar should trade at ‘low US80¢'” 25-27 Jan 2014).

This is not a recessionary period and total output has not declined, but activity/growth has slowed. For the moment, income growth per capita has stopped, unemployment continues to rise and price pressures are starting to ease in the domestic economy as our ToT boom reverses and global growth remains low. This situation is likely to continue, if not become worse, as further falls in our ToT are expected. From a policy perspective, it’s unlikely, given this environment, that the RBA will increase interest rates in the short-term. Depending on the size/severity of changes in the ToT, it would be more likely that the next move in the official cash rate will be down.

The huge elephant in the room remains the ongoing strength of the Australian housing market.



September – employment continues to disappoint

Overall, there was a continued deterioration in labour market conditions in September.

Despite the recent issues surrounding the collection and reporting of the labour force survey, the results using trend data have remained consistent. I’ve been using the trend series data in my labour market updates for a while now and the pattern of the September data is very much in line with recent trends, albeit slightly worse.

Given my recent post on employment growth, this will be a shorter-than-usual employment update.

The key metrics of employment and unemployment growth continued to weaken in the latest month:-

Source: ABS

The number of people employed Full Time (FT) continued to decline in September and the overall employment growth number was again ‘saved’ by Part Time (PT) employment growth across all time periods.

Unfortunately, the growth in the number of unemployed persons surpassed the growth in FT employed persons in the last month, the last 6 months and in the last year. That’s the state of our labour market at the moment.

Below is the trend of the monthly change in total employed persons that makes up that annual growth figure of +133k growth (figure from chart above) in employed persons – this also highlights a concerning trend.

Source: ABS

The growth in the number of FT employed persons has been slowing all year, after what was quite a promising start to 2014. The growth in PT employed persons has offset these declines in FT employed persons over the last four months, but even the growth in PT employed persons slowed this month. A small positive is that the decline in FT employed persons was slightly smaller than the previous month.

On an annual basis employment growth is not keeping pace with the increase in the labour force due to population growth –

Source: ABS, The Macroeconomic Project

To double check, the 33.9k persons that left the labour force is also calculated by the annual change in the LFPR*estimate of the Working Age Population (WAP) at Sept. The decline of 33.9k persons from the labour force was made up of a -44k decline in males and +11k increase in females.

There remains a deficit between employment growth and the growth in the labour force due to population, especially for what is supposed to be a non-recessionary period – this trend has been in place for several years now:-

Source: ABS, The Macroeconomic Project

As a result, firstly, the number of unemployed persons continues to grow at a fairly consistent pace. There hasn’t been an acceleration in the rate of growth in total unemployed persons, but growth has remained constant.

Source: ABS

This is the 38th month where there has been an annual increase in total unemployed persons. The only time in our history that was longer was the recession of the early 90’s – which was 43 months. Given the current trends, we are likely to surpass that record.

Secondly, the labour force participation rate (LFPR) also declined in Sept, led lower by the continued deterioration in male participation. Since the peak in male participation in Feb 2008, the male participation rate has declined by 2% points. Male participation in now lower than it was ten years ago.

Source: ABS

The growth in hours worked is consistent with these results, with FT hours worked actually declining in Sept v Aug 14 (for the 3rd month in a row) and PT hours growth slowing as well in September.

Source: ABS

The total number of hours worked in the September quarter grew by 0.11%, slower than the June ’14 quarter growth of +0.26%. This slower growth in hours worked suggests that output activity in the domestic economy may have also slowed in the quarter.

At a quick glance, there appears to be a degree of correlation between the change in hours worked and the change in Gross National Expenditure (GNE). The GNE measure is an indicator of domestic output/expenditure activity – its calculated as GDP excluding net exports. I’ve used GNE here specifically because it’s a measure of domestic expenditure activity – which is why there should, in theory, be some correlation with hours worked in the economy.

Using hours worked as a guide, it’s possible that we will see a slow-down in domestic growth in the next quarter:-

Source: ABS

The overall impact on GDP will then depend on the performance of net exports (the external sector) during this quarter.

What this last chart also highlights is how much slower growth in hours worked has been over the last several years, compared to pre-GFC.

All these measures point to slower economic activity in the domestic economy.



Employment Growth in Australia by Industry & State

Over recent years employment growth in Australia has fallen below historical levels. The reporting of a single growth figure each month fails to highlight this, and other important benchmarks such as whether employment growth is high enough, or if employment growth is broad based across states and/or industries. This level of detail, and the direction of those moves, tell an important story of the state of the labour market and how it fits with the current growth narrative of the economy. The purpose of this post is to unwrap the employment growth numbers to understand the broader context of employment growth in Australia and then to get into some detail regarding the states and industries within states, that are over or under-performing with regards to employment growth across Australia.

The most striking thing about this analysis is that the top four industries that over-performed in relation to employment growth (and contribution to overall growth) are not directly related to the Mining-Construction-Housing growth narrative of Australia. In the last year, Health Care, Education, Retail and Agriculture were the industries where employment growth exceed the National average and made a significant contribution to overall employment growth. Employment in industries such as Mining and Construction declined, or grew at well below the National rate of employment growth. This highlights that we are in the midst of a change to the industry employment landscape in this country. While that landscape is constantly evolving, the implication of this shift to greater services-based employment becomes obvious when you compare the differences in average weekly earnings across industries. The general lack of growth in Full Time (FT) employment is also a major theme across many states and this has been well hidden in many cases (unless you live in one of those states).

The broader context of employment growth in Australia

There are two very important points about employment growth in Australia.

Firstly, on several measures, employment growth has been too low over the last three (3) years.

As of August 2014, annual growth of total employed persons was 167.7k persons. This sounds like a big number, but it’s below the ten year average of growth in August of just over 200k persons. Employment growth has been well below this longer term average since late 2011:-

Source: ABS

Employment growth has started to improve, with the trend moving toward that 10yr average. The composition of this growth is important too – more on this shortly.

This longer-term average rate of employment growth is actually fairly representative of the level of employment growth required to keep up with population growth.

By this measure, employment growth has also been too low. Since the start of 2011, population growth has added workers to the labour force at a faster rate than employment growth. The chart below compares the difference between the growth in total employed persons and the growth in the labour force as a result of population growth (this is estimated). When that number is negative (below the orange zero line), it suggests that employment growth is below what population is adding to the labour force.

Source: ABS

The implication of this lower level of employment growth has been a growing number of unemployed persons and a declining participation rate. Population is adding approx. 215k persons to the labour force annually, so employment needs to grow by roughly the same amount to maintain the status quo. Only in the last 4 months (chart above), has employment growth started to accelerate beyond what population is estimated to add to the labour force.

This brings me to the second important point about employment growth in Australia.

It’s obvious from the charts above that employment growth has improved during 2014. What is driving this improvement is growth in Part-Time (PT) employed persons. This is not a recent trend and the proportion of all workers that are PT employed has reached a new high in the Australian labour force. As of August 2014, 30.5% of all people employed are PT employed.

So while the 2014 overall employment number looks like its improving, you can see from the chart below that the composition of that growth has changed during 2014 – FT employment growth has slowed consistently since March and PT employment has grown in its place.

Source: ABS

The slowing of FT employment growth is not a sign of improving demand conditions. In the bigger picture though, growth in PT employed persons is still preferable to a decline in employment.

There is a clear implication for incomes as a result of this growing proportion of PT employed persons. To provide some broad context of what this means for incomes consider the difference in average weekly cash earnings for FT and PT employed persons:-

Source: ABS

These are gross amounts and will of course differ by industry, gender and state. A median measure would likely be more accurate as well. But the idea is to provide a broad sense of the relationship between FT and PT earnings – average weekly cash earnings for PT employees are roughly 40% of that of a FT employee.

According to Household Income and Income Distribution data from the ABS (ABS 6523.0 2011-12), wages and salaries account for over 60% of total household income with only 9% coming from Other Income and 25% from Government Pensions & Allowances. To some degree, wages and salaries are being supported by other sources of income.

As long as this trend toward higher PT employment continues, there is likely to be a negative impact on income which is likely to impact spending and government tax revenues.

Detailed state & industry employment growth

It’s useful to look at employment growth from a state and an industry perspective in order to understand how employment growth in states and industries within states are performing relative to the National picture. This is especially important where states tend to have some regional advantage, for example, Mining in WA.

An industry within a state is deemed to have over-performed where the employment growth rate in that industry is greater than both the overall National employment growth rate and the National employment growth rate for that industry. An industry or a state is deemed to have under-performed where that industry or state has grown at a rate slower than that of the overall National rate.

This means that employment growth hasn’t necessarily declined, but it highlights those industries or states where employment growth is slower or lagging the National average – a far more subtle indicator of performance.

The selection of time periods is crucial to the success of this analysis in providing a clear and accurate view of state and industry based employment growth trends. I’ve selected the last three (3) years (between the qtr. to Aug 2014 qtr. and the qtr. to Aug 2011) and supplemented with annual data, as this coincides as closely as possible with when employment growth in Australia started to dip below its long-term average. It’s also very close to the same period where the RBA started to lower interest rates, so it’s an interesting period to see what impact that has potentially had on employment growth. It’s been an important period of transition for the economy with the Mining boom starting to shift gears from the labour-intensive investment phase to the cost-cutting production and export phase. The RBA has been attempting to ‘rebalance’ growth in the economy, namely investment, to boost non-resources capex activity via lower interest rates. Capex data suggests that Dwelling Construction is responding to a degree.

Looking at the industries that have over-performed (versus National employment growth) from the last 3 years to the last year, highlights a shift in employment growth performance in some key industries:-

Source: ABS

Major highlights and how this fits with the current growth narrative:-

Mining employment growth – has gone from performing slightly better than at the National level, to well and truly underperforming in the latest year. Total employed in Mining has declined year on year by -29k persons or -11%. This should be no surprise given the increasing shift to the production phase of the mining boom (which is in theory less labour-intensive). The key markets leading this performance with declines in persons employed in Mining are NSW (-15k), QLD (-6k) and WA (-11k). What is interesting is that the WA state level employment growth still continues to outperform all other states despite the declines in Mining employment. More on WA later in the post.

An industry partly related to Mining is grouped under Professional, Scientific and Technical Services. Employment growth in this industry still over-performed versus the National average in the latest year with growth of +3%, but it does represent a large slowdown in growth versus three years ago, which was +8%. Employment growth in this industry has been driven by states such as QLD, SA, WA and NT, ranging between +7% and +66% employment growth over the last year. This is potentially linked to further exploration and development of resources projects in those states. Only in Vic did employment in this industry decline quite severely by 10% over the last year.

Construction – the underperformance of Construction employment growth doesn’t seem to fit with the emerging narrative of Construction leading the ‘rebalancing’ from Mining. Growth in total employed persons in Construction was only 0.3% over the last year and 2% versus three years ago. The Construction industry is the single largest FT employer at a National level. Yet, FT employment declined over the last year by -21k persons – most of which was a large decline in the latest quarter. All of the growth in Construction employment over the last year has been in PT employment (+24k persons). From a state perspective, Construction employment grew at above the National rate only in NSW and TAS. Construction employment in NSW grew by +29k persons or 10% over the last year, 16k of which were FT. In other states, Construction employment actually declined over the last year.

The trend over the last three years is also interesting. In this case, it looks like total employment for both FT and PT employed persons in Construction had increased consistently since August 2012. It’s been in the last two quarters where total FT employed in Construction has started to fall, while PT has continued to increase. So PT employment in Construction seems to replacing FT employment. There is also some possible noise in the data given that it is original (I have not adjusted the data) and the change was mostly in the latest quarter.

Source: ABS

This does throw some cold water on the idea that a rebalancing led by Construction may be underway – only in NSW and TAS is employed persons in Construction growing and overall, only PT employed in Construction has been growing in the last year. The latest Performance of Construction Index, on the other hand, has grown for the fourth consecutive month in September, led mostly by house and apartment building, while engineering construction continued to contract. The overall index shows quite robust expansion, so increased activity could still show up in employment growth in the Construction industry.

There is possibly a correlation between the growth in Construction employment and the boom in housing (investment) that is currently underway in NSW. Although, there is similar high growth in housing lending and house prices in VIC, but Construction employment declined by -7% over the last year.

Other industries related to Construction and this housing boom include Rental, Hiring and Real Estate Services and Financial & Insurance Services. Employment growth in Rental, Hiring and Real Estate (RHRE) has over performed versus National employment growth – but there are only two states that are driving this performance – SA and WA. In NSW and Vic, employment in RHRE grew by +4% and +10% respectively – while this is still high, that growth under-performed the market. Employment growth in Financial and Insurance continued to under-perform the overall National growth rate – and employment in this industry actually declined by -0.2% in the last year. Only in Vic and TAS did employment in Financial Services actually grow.

Manufacturing – The underperformance of manufacturing employment growth has improved in the latest annual data, driven by stronger growth in PT employed persons (+19k) whereas FT employed persons declined by -13k persons over the last year. The shift from FT to PT employment in Manufacturing should be no surprise given the long term decline in employment and share of output in the sector.

And the biggest overall employers in Australia….

The top three (3) industries by size of employment are Health Care & Social Assistance (1.4m persons), Retail (1.2m) and Construction (1m).

Employment in Health Care & Social Assistance has grown at double the National rate of employment growth over the last year and three years. Most of the growth has been in FT employed persons. Employment in all states except NSW and VIC has grown at above the National rate of employment growth. In NSW and Vic, employment in Health Care & Social Assistance declined over the last year.

Employment growth in Retail has also over-performed on an annual and three year basis. The only difference is that 60% of the jobs growth is PT in nature. Most states over-performed in relation to Retail employment growth with the exception of QLD, NT and TAS.

The industry where employment grew at the highest rate of all was in Agriculture, Forestry & Fishing industry, with +16.5% employment growth. This is a really bright spot for the economy and it’s not yet clear how this fits into the economic growth narrative of the country, albeit that Australia could have a significant role to play as a global leader in the provision of (safe) food sources. This industry significantly over performed both in FT and PT employment and the only states where Agriculture, Forestry & Fishing grew at below the 16.5% rate were QLD and WA.

State-level employment growth

Only in WA and NT has employment grown at a rate above the National growth rate of 4.4% in the last three (3) years, with NT representing a small proportion of the total workforce.

That situation has improved somewhat in the latest year with more states now growing at above the National rate of employment growth (orange bars, below):-

Source: ABS

On an annual basis:

  • WA and NT continue to outperform all other markets in terms of employment growth
  • Employment growth in SA and TAS has accelerated over last year compared to 3 years ago
  • Employment growth in the bigger states of NSW and VIC has slowed further and remains well below National annual employment growth rate of 2.2%, which is a concern

New South Wales

On the 3 year basis, employment growth in NSW hasn’t been too bad, but there should have been stronger FT employment growth.

Employment growth over the last 3 years was +140k persons, but if NSW grew at the National rate, employment would have grown by 153k persons. What accounted for this? Growth in FT employed persons underperformed the National rate of employment growth – if FT employment in NSW grew at the National rate, FT employment growth would have been +22k higher. That’s clear from the trend in actual employed persons below – both FT & PT employed totals continue to increase, but FT employed failed to reach a new (what looks seasonal) peak in Feb 2014.

Source: ABS

The concern is the continued slow-down/under-performance in employment growth in the latest annual data. This was driven by lower growth in both FT and PT employed persons (to a lesser degree).

The major contributors to FT employment growth were industries such as Transport, Wholesale and Warehousing (+24k FT jobs), Construction (+16k FT jobs) and Info Media and Telecoms (+13k jobs). The issue is that in only seven out of nineteen industries did employment grow faster than the National rate of employment growth for that industry – meaning that under-performance of FT job growth is prevalent across a range of industries in NSW.

PT employment has grown faster than FT, but still under-performed the National average. Growth in PT employed persons was led by Construction, Retail Trade and Education and Training.

The results are interesting given that NSW is ground zero for the investment property boom at the moment. There seems to be disconnect between the booming housing market there and the overall below average employment growth.


The main issue in Victoria is the lack of FT employment growth. Total employment growth in Vic over last 3 years was +101k persons made up of growth of only +7k FT employed persons and +94k PT employed persons. If Vic FT jobs had grown at the same rate as the National employment rate, FT job growth would have been 85k (versus the 7k actual). The situation is similar on an annual basis.

Where did all the FT jobs go in Vic? In the latest year, there was a decline of -21k employed persons in Construction and a decline of -25k persons in Professional, Scientific and Technical Services.

Manufacturing is the single largest FT employer in Vic (243k persons) and FT jobs growth wasn’t actually that bad at +0.2%, just under the National rate of growth in Manufacturing employment of +0.7%. Overall Manufacturing employment in Vic has remained fairly stable.

The other issue is that there were a large number of industries where growth in FT jobs under-performed in Vic relative to the National average for each of those industries. This under-performance was broad based and points to relatively lower employment growth across a range of industries. For example, employment in Health Care and Social Assistance (which is the largest overall employer in VIC) is growing annually at 5% Nationally, yet in Vic, employment declined by -0.5%.


Growth of employment in Queensland has improved to just above the National level in the latest year. But the weakness underneath that number is that FT job growth underperformed. In Qld, FT employed persons grew at only +0.7% versus the National annual growth rate of 2.2%. If FT employed persons in QLD had grown at the National rate, FT job growth would have been +71k FT employed persons versus the +12k actual. That’s a fairly significant difference.

There were three major contributors to that underperformance in FT jobs growth. Full-time employment declined in: Agriculture Forestry and Fishing by -23k persons, Manufacturing by -23k persons and Public Admin by -14k persons. The growth of PT employment in these industries didn’t come close to off-setting the declines. The largest FT employer in QLD is Construction and FT employment growth of +0.3% under-performed compared to the +7% growth in PT roles. There is nothing in the trend of this data to suggest that this might be the start of any Construction-led rebalancing in this state.

Instead, most of the employment growth in QLD has been in PT roles which grew at 13% over the last year and this has been enough to drive the overall employment growth rate in QLD to above the National level. This growth was driven by the biggest PT employers: Health Care and Social Assistance +19k, Retail +18k, Education & Training +8k and Other +9k.

South Australia

Over the 3 year time frame, employment growth in SA has been weak on a number of measures. Total employment growth in SA of 1% under-performed the National rate of growth of +4.4%. All of the growth was in PT employed persons. The total number of FT employed persons actually declined over the 3 year time frame by 3k persons. This was driven by employment declines across a broad range of industries.

But recall that employment growth in SA had picked up on an annual basis. The results are very encouraging in that most of the growth has been in FT employed persons. FT employed persons in SA grew at a much faster rate +3.1% than the National rate of 2.2%. PT jobs grew on par with the National rate of growth. This is quite a good turnaround for the state.

The improvement in SA performance was the that declines in FT employment over the 3 year period (Financial Services and Insurance , Manufacturing, Retail and Wholesale), halted over the latest year.

The composition of the growth points to a shift in SA industry. Manufacturing in SA accounts for the largest share of FT employed persons, but FT employment declined on an annual basis by -3k workers (it’s still the biggest FT employer). Instead, FT employment growth came from Professional, Scientific and Technical Services +13k persons, Agriculture, Forestry and Fishing +10k (also strong PT job growth here) persons and in Education and Training +6k persons. The longer term trend suggests that employment growth has flattened out, but it’s a good result nonetheless.

Western Australia

This has been the one state that has over performed the National employment growth rate on both a three year and annual basis.

But a fundamental shift has occurred over the last three years – growth in FT employed persons has been leveled off and PT employment growth has accelerated over 2014. In fact, growth in FT jobs has slowed from 9.9% (last 3 years) to only 1.5% annual, which under-performed the National annual rate of 2.2% growth. Over the last year, this lower growth in FT employed persons was offset by a 10.6% growth in PT jobs, making the WA market still look reasonably strong.

This is a big turnaround from the three year growth and the shift in trend is easy to see in the chart below…

Source: ABS

From an annual perspective, FT employment declined in: Mining by -12k persons (and it declined faster than the National rate of decline for Mining jobs), Electricity and Gas by -7k persons (also declined faster than the Nat rate of decline), Construction by -1.6k persons (still the single largest FT employer in WA), Public Admin & Safety by -10k persons. These industries are all closely related to Mining.

FT employment in WA grew at a rate below the National rate for that industry in ten (10) out nineteen (19) industries. There were still several industries that had strong FT growth (over-performed) – Professional, Scientific & Technical Services +18k persons, Retail Trade +13k persons and Real Estate Service +10k persons were the three largest.

The over-performance in PT jobs growth comes from: Retail +10k jobs, Accomm & Food Service +8k and Health Care & Social Assistance +9.7k jobs. In fact 11 out of 19 industries grew PT jobs at a rate faster than the National rate for that industry.

We are starting to see employment in Mining and its related industries impacted by this shift to the production phase of the boom. So far, industries that receive some flow-on benefit from the boom (Retail, Food Service) are yet to see much impact on employment growth.

When you layer this with the value of average weekly earnings (Mining indexes at +170% above the average), the significance of this shift becomes obvious. Which makes me wonder how long employment in some of these other industries will continue to grow.


Tasmania has also been one of the big turnaround stories over the last year with employment growth of +4.1% outpacing the National rate of employment growth of +2.2%. This has been driven by an over-performance of PT jobs growth. PT jobs grew by 6.5% and FT jobs grew mostly on par with the National rate of growth, which is still a good result.

The historical heartland of FT employment in TAS has been in Manufacturing (ten years ago was the largest FT employer). Over the last year, FT employment in Manufacturing grew by 19%, seriously bucking the National rate of growth in Manufacturing jobs of +0.7%. There was also strong FT job growth in Agriculture, Forestry & Fishing of +23%.

The higher growth in PT jobs were concentrated in Accomm & Food Service +3k (which more than off-set the -1.4k FT jobs in this industry) and Education & Training +2.3k (which also off-set the -1k FT jobs in this industry). To a much smaller degree, there was growth in Professional, Scientific & Technical of +1k persons and Health Care & Social Assistance of 1.3k persons.

Northern Territory

This has also been a state that has over-performed the National rate of employment growth. In the last year, this has been driven mostly by growth of FT jobs. Although this growth was quite broad-based (12 out of 19 industries grew faster than the National rate for each industry), most of the growth came from one industry – Prof Scientific and Tech +3.3k (total FT employ growth was 4.3k for the year). To a degree, this is going to be Mining related (through exploration etc.). Employment within the two biggest FT employers – Public Admin & Safety and Construction both declined in the latest year.

Australian Capital Territory

It should be no great surprise that employment growth in ACT has deteriorated in the latest year versus the 3 year period given the direction of the Abbott government to reduce the number of public sector jobs.

Over both time frames, employment growth in ACT has under-performed the market. In the latest year, growth has slowed considerably. Growth in FT jobs remains above the National rate +2.7%, but PT jobs in ACT actually declined year on year by -5.7%. The three major contributors to that decline were Public Admin & Safety of -5k persons, Education & Training of -2k persons and Health Care & Social Assistance of -1k persons.

A final point on Average Weekly Earnings (AWE’s)

A great overlay to the discussion on employment growth by industry is AWE’s by industry.

The potential impact on wages and salaries of the decline in Mining employment, and related industries, becomes obvious when you view the AWE for FT adults. Mining AWE’s, for example, index at 170% of the AWE’s for all industries. Contrast this with the AWE’s in Retail, which had the highest overall contribution to the growth in total employed persons over the last year (+67k persons). The AWE’s for FT Retail adult indexes at only 70% of the AWE’s for all industries.

Retail has always been a big employer, but has operated, most recently, with the backdrop of high wages from Mining and a credit/consumption boom, pre-GFC. It’s not likely that these days will return, so it seems like some of this employment growth is based on shaky ground.