A couple of weeks ago the Australian Bureau of Statistics (ABS) published the March quarter 2021 Australian National Accounts: National Income, Expenditure and Product (Cat. No. 5206), which means we can update our quarterly Australian State and Territory Gross State Product (GSP) series.
Our day job has been a bit busy so we’ve taken a little while to get to this and offer only a brief post today.
The good news is that the code we presented worked perfectly with the addition on an extra quarterly indicator observation.
The bad news is that a bit of a gap has opened up between the ABS’s official measure of seasonally-adjusted Gross Domestic Product (GDP) and the sum of our State and Territory GSPs. We examine this issue at the end of this post.
We have also explored a bit of extra code and a couple of charts to see how well economies across Australia are recovering from COVID-19 related shutdowns.
But first to our numbers. In interpreting these don’t forget our disclaimers here and here. This endeavor is a work-in-progress and we don’t claim to have it right yet.
Output in most States and Territories increased in the March quarter 2021 over the December quarter 2020, with each State and Territory economy continuing its respective recovery from the COVID-19 low of the June quarter 2020.
Output declined in South Australia and Tasmania in the March quarter 2021, but this represented a pullback from strong growth the previous quarter, with both States recording solid year-on-year annual growth.
This is shown in the Table below.
March Quarter 2021 Gross State Product, Percentage Changes, Seasonally-Adjusted Chain Volume Data
| Quarterly Growth Dec-20 to Mar-21 | Year-on-year Growth Mar-20 to Mar-21 | |
| NSW | 0.52 | 1.08 |
| VIC | 0.37 | 1.29 |
| QLD | 0.62 | 0.94 |
| SA | -0.24 | 2.22 |
| WA | 0.45 | 0.08 |
| TAS | -2.13 | 3.48 |
| NT | 1.14 | 0.72 |
| ACT | 0.09 | 1.31 |
| AUS | 0.40 | 1.06 |
Output in all State and Territory economies is much higher than their respective bottoms in the June quarter 2020, and only Western Australia’s output has not recovered its pre-pandemic (December 2020) level.
This is shown in the following chart.
However, a better indicator of the respective State and Territory COVID recovery is the ‘output gap’, which is the the difference between potential output and actual output.
The chart below shows the output of Australia relative to a simple linear projection of the previous five years of output.
In the March quarter 2020 Australia was 3.1% below potential output. The COVID-19 shutdowns have cost Australia approximately $110 billion in output[1]Chain volume, over five quarters, compared with about $2 trillion annual production (i.e about 5% of annual production).
The respective State and Territory output gaps are shown in the following table.
Australian State and Territories Output Gap
5-year Average Output Trend (Mar-2015 to Dec-2019)
| Output Gap (%) | |
| Out_Gap_NSW | 3.61 |
| Out_Gap_VIC | 4.36 |
| Out_Gap_QLD | 4.08 |
| Out_Gap_SA | 0.86 |
| Out_Gap_WA | 0.61 |
| Out_Gap_TAS | -0.41 |
| Out_Gap_NT | -2.29 |
| Out_Gap_ACT | 3.04 |
| Out_Gap_AUS | 3.14 |
Remarkably Tasmania and the Northern Territory are now above their potential output trend, although economic growth tends to be volatile for the smaller economies.
The chart below shows Australian State and Territory GSP growth from the September quarter 2002 to the March quarter 2021. The COVID-19 downturn dominates each series.
Now to the difference between Practical Economics’ national GSP growth versus the equivalent ABS figure.
National quarterly GDP/GSP growth of the Practical Economics’ series (i.e. the sum of the State and Territory series) was 0.4% in the March quarter 2021, compared with 1.8% GDP growth in the official ABS figure.
There are a number of reasons as to why the two numbers are different. These include:
- The March quarter 2021 Practical Economics’ numbers are not benchmarked to an annual GSP number, but are extrapolated from the ratio of industry outputs to GSP in 2019-20.
- The 3rd quarter of any year is the furthest from any benchmark as will occur in the series, and may be substantially revised when the annual 2020-21 GSP numbers are released by the ABS;
- Practical economics uses a different seasonal adjustment to the ABS, and seasonal adjustment is difficult at the moment given the COVID-19 induced volatility in economic output.
- The Practical Economics national seasonally-adjusted series is calculated from the sum of the State and Territory seasonally-adjusted searies. As we’ve pointed out here, this isn’t a good idea, but we have no other option due to data limitations.
- The Practical Economics series does not include ‘taxes less subsidies’ in its production estimate of GSP (it is not included in our benchmark GSP by industry in ABS Cat No. 5220.0 and so we must ‘ignore’ it).
- This category grew by 4.9% in ABS national production in the March quarter 2021, contributing 0.3 percentage points to national economic growth.
As an example of how differences in seasonal adjustment led to a major difference in growth from the Agriculture, Forestry and Fishing (AFF) industry between the two series.
If you’re not you’re not familiar with the reasons for seasonal adjustment, the idea is that is takes out regular seasonal variations to reveal the underlying conditions in the economy. For example, retail trade is always high in each December quarter as it contains Christmas sales. Therefore retail sales always fall each March quarter, not because the retail sector is in recession, but simply because it is the March quarter.
Hence we adjust the ‘original’ data to attempt to take out the usual seasonal variation to produce ‘seasonally-adjusted’ data.
National seasonally-adjusted growth by industry output (Gross Value Added or GVA) for the official ABS figure versus the derived Practical Economics figure (the national GVA series seasonally adjusted by the Practical Economics method) is shown in the Table below. AFF GVA grew by 2.9% in seasonally-adjusted terms in the ABS series, but shrunk by 15.4% under the Practical Economics method. These numbers are the same original data but seasonally-adjusted using different methods.
National Industry Gross Value Added: Seasonally-Adjusted Quarterly Percentage Growth, Dec-20 to Mar-21
| ABS | Practical Economics | |
| Agriculture, Forestry and Fishing | 2.9 | -15.4 |
| Mining | 0.7 | 0.3 |
| Manufacturing | 3 | 2.5 |
| Electricity, Gas, Water and Waste Services | -0.9 | -0.8 |
| Construction | 4.4 | 4.1 |
| Wholesale Trade | 2.9 | 1.1 |
| Retail Trade | -1 | -3.5 |
| Accommodation and Food Services | 5.2 | 4.4 |
| Transport, Postal and Warehousing | 5 | 5.0 |
| Information Media and Telecommunications | -1.9 | -1.6 |
| Financial and Insurance Services | 0.2 | 0.5 |
| Rental, Hiring and Real Estate Services | 5.3 | 4.5 |
| Professional, Scientific and Technical Services | 1.2 | -1.3 |
| Administrative and Support Services | 1.9 | 1.2 |
| Public Administration and Safety | -0.7 | -0.8 |
| Education and Training | 0.3 | 0.3 |
| Health Care and Social Assistance | 1.2 | 1.1 |
| Arts and Recreation Services | 5.7 | 5.5 |
| Other Services | 5.8 | 3.1 |
| Ownership of dwellings | 0.4 | 0.5 |
| Taxes less subsidies on products | 4.9 | |
| Statistical discrepancy (P) | na | |
| Gross domestic product | 1.8 | 0.4 |
This led to a 0.6 percentage point difference in the contribution of this industry to growth between the two series, as shown in the Table below.
Industry National GVA, Contributions to Growth (percentage points), Dec-20 to Mar-21
| ABS | Practical Economics | |
| Agriculture, Forestry and Fishing | 0.1 | -0.5 |
| Mining | 0.1 | 0.0 |
| Manufacturing | 0.2 | 0.1 |
| Electricity, Gas, Water and Waste Services | – | 0.0 |
| Construction | 0.3 | 0.3 |
| Wholesale Trade | 0.1 | 0.0 |
| Retail Trade | – | -0.2 |
| Accommodation and Food Services | 0.1 | 0.1 |
| Transport, Postal and Warehousing | 0.2 | 0.2 |
| Information Media and Telecommunications | – | 0.0 |
| Financial and Insurance Services | – | 0.0 |
| Rental, Hiring and Real Estate Services | 0.1 | 0.1 |
| Professional, Scientific and Technical Services | 0.1 | -0.1 |
| Administrative and Support Services | 0.1 | 0.0 |
| Public Administration and Safety | – | -0.1 |
| Education and Training | – | 0.0 |
| Health Care and Social Assistance | 0.1 | 0.1 |
| Arts and Recreation Services | – | 0.0 |
| Other Services | 0.1 | 0.1 |
| Ownership of dwellings | – | 0.0 |
| Taxes less subsidies on products | 0.3 | 0.0 |
| Statistical discrepancy (P) | 0.0 | |
| Gross domestic product | 1.8 | 0.4 |
We can’t see how the AFF industry GVA grew in seasonally adjusted terms in the ABS series.
We illustrate why in the following table, which shows the original and ABS seasonally-adjusted AFF GVA change between the December and March quarters for the last four years. AFF output always falls in each March quarter in original terms because grain crops are usually harvested in the December quarter, causing this to be the highest output quarter of each year.
Agriculture, Forestry and Fishing Industry, Percentage Growth in GVA, ABS Seasonal Adjustment, December quarter to March quarter.
| Year | Original | Seasonally- adjusted |
| 2017-18 | -15% | 5.5% |
| 2018-19 | -29% | -0.7% |
| 2019-20 | -33% | 0.5% |
| 2020-21 | -38% | 2.9% |
Original AFF GVA fell by a whopping 38% in March 2021, much more than in previous years, but ABS seasonally-adjusted growth was higher than previous years.
Who would want to be the ABS at the moment?
As we’ve mentioned, we’re trying to explain differences between the official ABS and Practical Economics’ data, rather than to criticise the ABS. Our main aim is to firm-up our estimates by seeking comments.
What do you think of our updated series? Can you find any other reasons why the GDP numbers diverge?
If you have any comments, suggestions or want to purchase full results then please contact us.
References
| ↑1 | Chain volume, over five quarters |
|---|