Australia: Selected Issues

International Monetary Fund. Asia and Pacific Dept

doi:10.5089/9781513556499.002.A002

Jump to Content Jump to Main Navigation

Australia: Selected Issues

Author: International Monetary Fund. Asia and Pacific Dept

Publication Date:: 30 Sep 2015

eISBN:: 9781513556499

Language:: English

Keywords:: ISCR; CR; price; government; Australia; economy; TFP level; OECD productivity database; TFP indices; utilities TFP growth; growth in Australia; Mining sector; Production growth; mining TFP growth

Download PDF (713 KB)

This Selected Issues paper analyzes the housing prices in Australia. Housing prices in Australia have increased strongly over the past two decades, including by comparison internationally. Thus housing prices are often argued to be overvalued. Many counter-arguments have been put forward for why such measures are flawed. This paper argues that housing prices are moderately stronger than consistent with current economic fundamentals, but less than a comparison to historical or international averages would suggest. International comparisons of price-to-income ratios suggest that Australia is broadly in line with comparator countries, although significant data comparability issues make inference difficult.

Abstract

Sustaining Income Growth in Australia¹

Even with nearly 1 percent productivity growth in the baseline, medium term income and output growth will be significantly weaker than in the past.
Input growth will be restricted by aging pressures and limits to capital accumulation, making it necessary to lift TFP growth, which has dropped sharply.
There are no low hanging fruit. Australia is at the or near the frontier on several dimensions of competitiveness, and reform momentum has faded, making the task of lifting productivity more difficult.
Targeted reforms in specific areas where Australia trails the frontier can help push productivity and sustained growth in living standards.

A. Overview

1. Australia has enjoyed uninterrupted growth over the past 2 decades and counting, and has outperformed its advanced economy peers. Compared to the OECD as a whole, Australia’s GDP growth over the last 24 years has been a full percentage point (about 50%) higher on average. Per capita GDP growth averaged 1.8 percent over 1990-2014, and income growth was even higher, near 2 percent, thanks to record high commodity prices and advantageous terms of trade.

2. But the broad impulses that drove sustained growth in output and incomes in the past have now waned.

Comprehensive structural reforms in the 1980s and 1990s helped propel Australia to the top ranks among advanced economies. This was reflected in strong productivity growth in the 1990s and early 2000s, which has now reversed course.
In the 2000s, strong growth in demand for resources, particularly from China coupled with rising commodity prices drove mining sector investment to record highs. (LNG capacity was also installed, which should lead to Australia being the world’s largest LNG exporter in the coming years). Now, mining sector investments have tailed off as expected, and projects are entering the production and export phase of operations. However, at the same time as mining sector investments are unwinding, the price of their output has fallen much more sharply than expected reflecting the increase in global (including Australian) supply but also a slowdown in residential investment in China which has lowered the demand for steel.
Further, the non-mining economy has not picked up sufficiently to take up the slack.

3. Going forward, output and income growth will likely weaken. Continued slower productivity growth combined with gradually weakening labor force participation (as the population ages) will likely reduce potential output growth. Incomes will grow even slower than output, especially in the nearer term, as the terms of trade comes off further from its record high. Thus Australian living standards are unlikely to grow as fast as in the past.

4. Boosting productivity growth will be the key to maintaining high growth in living standards. Though Australia compares favorably with other advanced economies on measures of efficiency and competitiveness, there are still lags relative to the frontier in some areas. Convergence to the productivity frontier will have to accelerate in order to maintain past rates of output and income growth. This will need a focus on further reform measures to boost competition, enhance inputs, and foster innovation, while recognizing potential difficulties in convergence to the productivity frontier.

B. Stylized facts: Strong growth in the past, but weakened outlook

5. Output and income growth is significantly weaker than in the past. Over 2000-2014, real GDP growth averaged 3 percent, real GDP per capita growth averaged 1.5 percent, while real net national disposable income (RNNDI) per capita (adjusting output growth for terms of trade effects) grew even faster at 2 percent, supported by historically high prices for key Australian commodity exports. However, as mining investment has begun its anticipated decline, and terms of trade have fallen sharply, per capita GDP, and income growth in particular, have fallen well below long run averages. Indeed, RNNDI per capita has been falling since 2012.

6. Productivity growth has slowed sharply.

Labor productivity (LP) growth has weakened in the past decade. LP growth decomposition reveals that unlike in the 1990s, growth over the last decade has been driven by capital deepening (mostly mining sector related) rather than total factor productivity (TFP) growth.
TFP growth has decelerated sharply over the past decade. While historically Australia has out-performed other advanced economies (AEs), recent TFP growth trends in Australia are in line with the more general decline in TFP growth among AEs. TFP growth was among the highest in the late 1990s and early 2000s, but has slowed sharply since then. Over the last complete productivity cycle (2003-08), TFP growth declined to near zero, though this is still a better performance than many peers (particularly the major European economies). However, since the last cycle, TFP growth in Australia has not recovered and remains near zero² (partly reflecting the impact of the mining and utilities sectors; which involve investment lags that reduce measured productivity).

7. Trends specific to mining, and one-off factors have also been held responsible. Some commentators (Topp et al, 2008) have pointed to high profitability in mining driven by record high commodity prices, leading to mining of marginal deposits and driving lower productivity. Commodity prices also spurred large investments in mining not matched by output growth, and in agriculture and utilities, drought conditions led to lower output and measured TFP.

8. The slowdown in TFP growth was broad-based, and not limited to mining and utilities. All sectors, apart from construction, saw weaker TFP growth between 2002-13 compared to 1995-01. Eslake and Walsh (2011) construct a labor productivity measure excluding mining and utilities, and show that the decline in productivity in the 2000s was broad-based: “…the decline in labour productivity in the mining and utilities sectors accounts for less than 10% of the decline in overall market sector productivity growth over the past decade”. While mining and utilities TFP growth turned negative, it fell to zero in the manufacturing sector, and halved in the services and agriculture sectors. For the manufacturing sector, the productivity slowdown over the last cycle is ascribed mainly to input lags in petrochemicals and metal products industries, and increased labour intensity, input cost pressures, and challenges in measuring quality improvements in food and beverages. In the current incomplete cycle, the rate of decline in manufacturing MFP has slowed (Productivity Commission 2013).

C. Baseline scenario of medium-term potential growth

9. Against this backdrop, our baseline projections suggest that Australia’s potential growth will be significantly lower in the future (though still higher than peers). Our estimates are constructed using a standard production function framework, with a Cobb-Douglas function of capital and labour, and Hicks-neutral total factor productivity. This allows us to take expected trends in mining sector productivity and demographic changes explicitly into account. That said, there are necessarily uncertainties particularly with regard to projecting productivity growth. Thus we also consider other methodologies (including HP-filtering, and a multivariate Kalman-filter model).

10. Potential growth will stay well below 3 percent over the medium term. Under the baseline, potential growth peaks at just over 2¾ percent in the initial years, supported by rising mining TFP growth, but then subsides to just over 2½ percent as this effect wears off. On average, output growth is 2.7 percent, of which a third (0.9 percentage points) comes from TFP growth, a little over a third from labour, and the rest from capital. In 2020, growth is thus around 25 percent lower than the historical rate of growth of around 3-3¼ percent.

Table 1.

Potential Growth - Baseline Estimates

Source: Staff calculations.

Table 1.

Potential Growth - Baseline Estimates

			2015	2016	2017	2018	2019	2020
Potential growth			2.8	2.8	2.8	2.6	2.6	2.6
	Capital growth contribution			0.8	0.8	0.8	0.8	0.8
	Labour growth contribution			0.9	1.0	0.9	1.0	1.0
	TFP growth			1.1	1.0	0.8	0.8	0.8
		Non-mining contribution		0.4	0.4	0.5	0.6	0.7
		Mining contribution		0.7	0.6	0.3	0.2	0.1
Capital stock growth				2.4	2.4	2.5	2.5	2.5
Labour growth				1.4	1.5	1.4	1.5	1.5
	WAP			1.7	1.7	1.6	1.6	1.6
	LFPR			65.1	65.0	64.9	64.8	64.7
	NAIRU			5.6	5.6	5.7	5.7	5.7

Source: Staff calculations.

Table 1.

Potential Growth - Baseline Estimates

			2015	2016	2017	2018	2019	2020
Potential growth			2.8	2.8	2.8	2.6	2.6	2.6
	Capital growth contribution			0.8	0.8	0.8	0.8	0.8
	Labour growth contribution			0.9	1.0	0.9	1.0	1.0
	TFP growth			1.1	1.0	0.8	0.8	0.8
		Non-mining contribution		0.4	0.4	0.5	0.6	0.7
		Mining contribution		0.7	0.6	0.3	0.2	0.1
Capital stock growth				2.4	2.4	2.5	2.5	2.5
Labour growth				1.4	1.5	1.4	1.5	1.5
	WAP			1.7	1.7	1.6	1.6	1.6
	LFPR			65.1	65.0	64.9	64.8	64.7
	NAIRU			5.6	5.6	5.7	5.7	5.7

Source: Staff calculations.

11. Potential growth could be even lower. Lower commodity prices could lead mining sector export growth to be weaker than expected (25% weaker than the baseline), and non-mining TFP growth could be more in line with the average over the last complete cycle (close to zero). The combined effect could push potential growth to around 2 percent (compared to around 2½ percent in the baseline).

12. The assumptions underpinning the baseline estimates are as follows.

Labour: The labour input is taken as the number of employed persons. The key elements in projecting this variable are (i) working age population (ii) labour force participation rate (LFPR), and (iii) the equilibrium rate of employment. The product of these three items provides an estimate of the stock of employed workers each year over the forecast horizon (2016-20).

Working age population is calculated as the sum of all persons 15 years and more of age, based on the ABS Population Projection time series (based on historical assumptions on fertility, mortality, and net migration). ABS data show working age population growth declining gradually from 1.7 to 1.6 percent over the medium term.
LFPR in Australia has declined in recent years, as mining investment related activity has unwound. Over the medium term aging pressures are expected to gradually depress LFPR. Our assumptions on LFPR are derived from cohort-wise regression estimates of LFPR trends by gender and age which yields a forecast of LFPR declining from 65.1 to 64.7 percent over the projection period. ³
Equilibrium unemployment (NAIRU) is also estimated based on the MVF model. The estimates show NAIRU stays relatively steady at around 5.7 percent (and hence relatively steady equilibrium employment over the medium term).
Overall, labour input growth remains fairly steady at 1½ percent, in line with growth in working age population.⁴

Capital stock: The flow of capital input is taken to be proportional to changes in the stock of “productive capital”, defined as aggregate real capital stock excluding households (including government, financial, and non-financial corporations). Growth of the capital stock can be expressed as the difference between the investment-to-capital ratio and the rate of depreciation. ⁵ Based on Australian data for non-financial corporations (as a proxy for aggregate productive capital), average investment-capital ratio is around 6 percent over 1990-2014, and the rate of depreciation is around 3.5 percent. Assuming these long run average values hold over the medium term suggests a growth rate of capital around of 2.5 percent, and is somewhat higher than Treasury forecasts of between 1.8 – 2.3 percent.⁶

Total Factor Productivity: There are inherent difficulties in projecting TFP growth given measurement issues and the possibility of breakthrough innovation disrupting the link with past performance. However, for some sectors such as mining, it is possible to say with some degree of confidence that measured TFP growth will recover as investment lags from large new mining projects fade and exports gather pace. For other sectors, such as manufacturing, which has dwindled as a share of the economy, it seems reasonable to assume that that re-establishing a high productivity manufacturing sector is a difficult task, and the past is a reasonable indicator of the future. With this in mind, the following assumptions are made regarding TFP growth:

Mining sector TFP growth is assumed to rise in proportion with the projected real growth in mining exports⁷, under the assumption that the export phase will use inputs less intensively than the construction phase⁸. The contribution of mining to TFP (proxied by mining sector TFP growth weighted by sector share in value added – around 10 percent) is high in the near term, and falls off by 2020 as the export growth effect fades.
The contribution of the non-mining sector to aggregate TFP growth is assumed to rise from 0.4 percent in 2016 (in line with measured 16 market sector TFP growth in 2013) to 0.7 percent in 2020, in line with historical aggregate TFP growth over 1995-2013 of 0.66 percent.⁹
Under these assumptions, aggregate TFP growth touches 1.1 percent in 2016, declining to 0.8 percent in 2020, as the lift from mining fades, but is offset by an increase in non-mining TFP growth to a long run average rate.

13. Statistical filters give a similar picture. By way of comparison, the IMF’s Multivariate Filter model for output gap and potential growth¹⁰ shows average potential growth in the medium term of 2¾ percent. The filter shows an initially rising path of potential growth before declining in 2020. But unlike the production function estimates this approach does not factor in the likely TFP growth profile in the mining sector. A standard HP-filter shows trend growth in the region of 2.5 percent, taking an average of the filtered growth series over the 2013-2014. Both techniques show a sustained slowdown in potential growth from rates at or above 3½ percent in the early 2000s to below 3 percent in 2007/2008. There is some confidence to draw from the fact that a disaggregated inputs and productivity based view gives results fairly close to those from purely statistical techniques, both suggesting a slowing down of potential growth in a relatively tight range of between 2½ - 2¾ percent. However, as noted above, there are uncertainties in projecting productivity growth, and known weaknesses in the statistical methods (such as the end-point problem of the HP-filter, which is addressed to some extent by the MVF model with use of medium term of forecasts of key variables).

14. The slowdown has implications for future income growth. In the baseline, estimates of actual growth consistent with closing the output gap in the medium term, and terms of trade implied by projected import and export prices suggest that RNNDI per capita growth will average 0.5 percent in the medium term, well below the historical average of nearly 2¼ percent over 1995-2014.

15. Addressing the slowing in output and incomes will require faster TFP growth. There are limits to input-led growth, stemming from the downward pressure on labour force participation from aging, and difficulties in sustaining high investment rates over extended periods. For example, generating growth of 3¼ percent with no additional TFP growth than in the baseline and unchanged assumptions about labour would need capital accumulation of around 4 percent a year. This rate of accumulation is consistent with an investment ratio of 7½ percent, which was only observed around the peak of the mining investment boom and is difficult to replicate.

16. Despite the slowing, Australia still compares favorably with its peers. GDP growth over 2016-2020 at 2.7 percent on average is well above the advanced economies average of 1.9 percent. In per capita terms, given the strong population growth rate, GDP growth is more in line with other AEs, around 1.5 percent over the medium term. Should TFP growth not materialize as projected in the baseline (and come out lower, or near zero), then potential output will be in line with other AEs, though on the lower end in per capita terms.

D. Can additional reforms stimulate higher growth?

17. Big reforms in the past had a sizeable impact on productivity. A recent IMF paper¹¹ shows that over 1970 – 2007, sizeable reductions in regulatory burdens (product and labour market regulations), investment in infrastructure, and increased expenditure on R&D, ICT capital, and use of skilled labour had strong positive effects on TFP in Australia.¹² For the economy as a whole, the biggest increases in TFP were due to skilled labour and R&D spending, followed by infrastructure and reduction in labour market distortions. In individual sectors, manufacturing benefited from skilled labour and ICT capital, as well as from R&D, and product and labour market reforms; “other production” (including primary industries and utilities) and finance/business services benefited from skilled labour; ICT sector benefitted from R&D, job protection legislation, and labour tax wedge related reforms; the distribution sector from product market regulation reforms; and personal services from ICT capital and product market regulation reforms. It is noteworthy that infrastructure exerted a significant and large effect across several sectors.¹³ The results not only underscore benefits of past reforms, but also highlight areas that may require additional reform effort to boost productivity, such as in distribution.

18. Australia is already among the world’s most advanced economies on measures of competitiveness. Though its rank on overall Ease of Doing Business slipped somewhat from 5 in 2005 to 7 in 2015, it remains close to the global frontier for starting a business and getting credit. It is at or above the median on most other indicators, barring protecting minority investors, and trading across boundaries. More generally, policy and institutional settings in Australia are sound and of a high quality.

19. Thus, there are no more “low hanging fruit” type reforms left. With Australia having already implemented key reforms in the 1980s and 1990s, and being already positioned at or near the frontier on several dimensions of competitiveness, the task of generating sustained productivity growth through additional reforms is much harder. It will require identifying narrower and targeted reform measures aimed at specific sectors and specific problems.

20. At the same time, reform momentum has slowed. According to some commentators, not only has the momentum slowed, but has also been replaced with “productivity stifling” regulation and legislation that has eroded productivity growth, and “progress has stalled or even reversed in some policy areas” (Garnaut 2005, Eslake and Walsh 2011, Banks 2012). There is also a view that reforms of the 1990s drove a one-time shift in the level of TFP, rather than a permanent change in the growth rate, hence the decline in TFP growth is a reflection of the fading impact of the major structural reforms of the past.

21. Meanwhile, other advanced economies have made progress towards the frontier. For instance, many more advanced economies are now at the frontier in starting a business (Germany, Italy, Japan, Netherlands, New Zealand, Norway, and Spain), and some have improved property registration and resolving insolvency (Denmark and France) over the past decade. The continuing attractiveness of Australia as a business destination may therefore not be guaranteed.

22. There are some areas where Australia lags behind the world’s best. Taking a more detailed look at these areas of weakness can suggest areas where improvements can be made, not only to boost productivity but also growth more broadly.

23. TFP levels lag behind in some sectors. In 2007, Australia was at the frontier²⁵ in construction, “other production” (including primary and extractive activities), and finance and business services (along with the U.S.), but lagged in manufacturing, ICT, and distribution (domestic trade, storage, and transport) activities. Though dated, these data suggest there is some room for catch-up in productivity levels in some areas – though there are comparative advantage arguments that may explain the gaps, and convergence to the frontier may not be guaranteed merely due to large gaps (see Appendix). For instance, the share of manufacturing has steadily declined in Australia, and it may be difficult to rekindle a manufacturing sector that is able to compete with countries at the frontier. However, it should be possible to improve productivity in the distribution sector.

24. The distribution sector may benefit from reforms. The econometric evidence shown above suggests that the reforms generally have had little or no impact on TFP in this sector. While the lack of evidence on direct effects may be explained by lack of targeted reforms, it also suggests that there was an absence of spillover effects from reforms more generally, which are observed in other sectors (for example, the methodology records no major reforms in the area of high skilled labour for the “other production” sector, but the evidence shows a strong positive impact on TFP from high skilled labour related changes). This suggests additional focus on reform efforts in the distribution sector, covering both transport and domestic trade.

25. Infrastructure investment may help to improve transport sector efficiency. Existing infrastructure is under pressure particularly in urban areas, ports, and water quality; according to some estimates, Australia has an infrastructure deficit of around $80 billion.²⁶ Particularly in areas with high density exhibiting pressure on urban infrastructure, targeted new spending could help reduce efficiency losses related to congestion effects and wear-and-tear. Addressing infrastructure needs through efficient investment and appropriate asset regulation will help raise the economy’s capacity over the medium term.

26. Australia is in a good position to use public investment to boost infrastructure. The policy debate in Australia regarding infrastructure is focused on private sector financing and recycling existing infrastructure assets. At the same time, it has also been shown that debt-financed infrastructure investment has been can have large benefits in conditions where the cyclical position is weak, government debt levels are low, and institutional settings are strong,²⁷ enabling efficient conversion of infrastructure funds to useful public assets. For a country like Australia, where conditions appear well suited for debt financed public investment, the multiplier could be quite high; a one percentage point of GDP increase public spending could raise output by as much as 2.6 percent over the medium term, with limited impact on debt. In addition, by relieving congestion in high density urban areas, public investment may also help crowd in private investment and relieve the housing supply constraint. A program for greater public investment may include not only new projects, but also greater spending on maintenance.

27. Enabling factors for technology absorption and innovation could improve further. Compared to frontier of advanced economies, Australia lags noticeably in the use of skilled labour, being at or near the median (and significantly below the U.S., the frontier or near frontier country in this dimension) across all sectors barring personal services, where it lies above the median (Figure 1). R&D spending is below the median in manufacturing and ICT sectors, and well below the U.S. The share of ICT capital lags in finance and business services and the ICT sector. On product market regulation, Australia generally performs well, having fewer anti-competitive regulations than other advanced economies, but more than in the U.S.²⁸ In some sectors like retail trade, restrictions on retail hours, parallel imports, location of outlets etc may be reducing efficiency in sectors like distribution, and could be relaxed further.

28. Fiscal policy can also assist growth by reforming the tax mix²⁹. For example, Australia’s company tax rates are higher than the OECD average, and bringing them in line with the OECD average would have a limited revenue impact. More generally, moving towards a structure that is less reliant on direct taxes should in theory help reduce distortions, both to work effort and to corporate investment.

29. Modest improvements in TFP growth relative to the baseline could boost incomes substantially. As an illustration, closing half the TFP level gap that has opened between US and Australian sectors relative to 1995, over the next 10 – 15 years, could raise TFP growth to 1¼ –1½ percent over the medium term³⁰, compared to the baseline estimate of 0.9 percent (i.e. an increase of about 50%), pushing GDP growth to near 3 – 3¼ percent, compared to 2 ½ – 2 ¾, i.e. an increase of about 20%. As a consequence, income growth would rise closer to the 2 percent long run average over the medium term. These improvements can come from better resource allocation. It is estimated that similar to other advanced economies, Australia can increase TFP levels between 7 – 9 percent by eliminating allocative inefficiencies over the next 10 years (IMF 2015).

30. Even so, income growth is unlikely to hit past highs, even with plausibly higher TFP growth. Given weaker projected terms of trade, achieving 2 percent growth in per capita disposable income in the medium term will require twice as much TFP growth as assumed in the baseline (close to 2 percent per year), a rate not consistently observed in Australia’s past.

31. A number of specific reform measures are already being implemented or considered in Australia, which should boost productivity. These include:

Competition policy. The Competition Policy Review (March 2015) recommends several product market reform measures including removing restrictions on retail trading hours, restrictions on pharmacy ownership and location, and on parallel imports that increase costs for Australian consumers.
In addition, the report recommends measures in the transport sector (including more cost effective road pricing and harmonization across jurisdictions), a review of the intellectual property regime, as well as planning and zoning restrictions on land, taxis and ride sharing, and product standard restrictions that may inhibit imports.
Australia is considering signing the WTO Government Procurement Agreement (currently one of five OECD countries not signatory to it), and maintains public sector procurement preferences, including local content policies including at the sub-federal level. These should be phased out, which should help improve the efficiency of public service delivery.³¹
Education and Skills, Infrastructure, and Innovation. ³² A number of reform measures have been initiated in response to identified weaknesses in these areas. These include:
Improving education and training through establishment of regulators to oversee quality of vocational education, uncapping of Commonwealth support for domestic undergraduate students to make it more demand driven (a number of measures were announced in the 2015/16 Budget to reform the higher education sector including deregulation of student fees), and establishment of tertiary education standards body to maintain education quality. This would help address the relatively large gap with respect to share of skilled workers noted above.
Improving infrastructure development frameworks, enhancing PPPs with improved demand and supply forecasting and contract standardization, and move to levying user charges on roads.
Boosting innovation with measures to enhance collaboration between universities, researchers and businesses.
Workplace Relations. The Australian government has commissioned an inquiry into the workplace relations framework, with a view to addressing job creation, efficiency, business investment, employment flexibility, and issues related to equity and fairness in pay, and maintenance of a relevant safety net. It is notable that while business perceptions of Australia’s relative labour market efficiency score it below the average for Anglo-countries, and the OECD average on some indicators, OECD measures of regulatory stringency show Australia is relatively less stringent compared to other countries.³³ Yet, perceptions of business can influence behavior and a review of this nature may help improve perceptions.
Tax Reform. The authorities are conducting a review of tax policy, which offers an opportunity to consider growth promoting and efficiency enhancing measures that may be implemented through the tax system, such as shifting away from reliance on direct taxes, and addressing any tax related disincentives to productive investment, such as a relatively high company tax rate compared to the OECD average.

E. Conclusions

32. Australia’s growth and income performance in the past nearly two and a half decades has been remarkable, but slowing. The high growth was a result of both wide-ranging reforms leading to structural improvements and the commodity boom. Aggregate TFP growth, however, has declined markedly and remains near zero.

33. Output and income growth are unlikely to pick up to previous rates over the medium term under a baseline scenario. Even factoring in the likely pick up in mining TFP growth, potential growth is likely to fall to about 2 ½ percent in the medium term (or to 2 percent under a plausible downside scenario), significantly lower than the historical average of about 3-3 ¼ percent. Per capita net national disposable income growth will likely be even weaker as the terms of trade come off further from their historic highs.

34. There is scope for increasing productivity, thus boosting incomes and growth, but it is limited given Australia is already near the global frontier. Higher productivity would raise output and income growth. But raising productivity substantially is likely to be challenging given the gradual waning of manufacturing, persistent productivity gaps in key sectors such as distribution, Australia’s position at the frontier in several sectors and efficiency dimensions, the generalized slowdown in advanced economies, difficulties in catching up with the frontier, and no obvious “big-bang” reform measures at hand.

35. Thus a strong recovery in reform momentum is critical. Encouragingly, the task of identifying and evaluating reforms is already underway with consideration being given to several reform measures aimed at boosting competition in retail and other services, encouraging more innovation and interface between business and research, enhancing labour skills through improved vocational training, tax reform, bridging the infrastructure gap, and improving workplace relations. Following through on these reforms, and continuing to identify and implement new reforms, will be critical for Australians to continue to enjoy the improvement in living standards that they have been used to.

Appendix—Technical Notes

I. Trends in Labour Force Participation

The charts below and overleaf show trends in LFPR across gender and age cohorts, as well as the long run trend projected participation rate that takes various demographic effects into account. The chart on the left shows that aging will exert a downward pressure on labour force participation (LFP) over the medium to long term. The chart on the right shows that forces from aging will be offset by cohort related and other trends. For instance, a key offsetting factor observed in Australia is rising participation among women in all but the youngest two age cohorts.

Increased participation among older male cohorts and among females generally over the past few years has helped to keep participation rates relatively stable. Breaking down the projected trend participation by gender and cohort, some trends that stand out are:

Participation rates for women in all cohorts above 25 year olds have risen steadily over time. There is some sign of stabilization, and even slowing, in some of the cohorts.
Participation rates among women below 25 have been declining
Men’s participation rates generally prevail higher than women, but have been declining among the young (below 29), and among those between 35-49 years old.
In contrast, participation among men above 50 has been rising steadily.

II. Changes in TFP levels and in relative distance to frontier among selected advanced economies and aggregate industry groups.

Using data from IMF 2015, the scatter plots below graph (i) changes in TFP over 2000-2007 against distance from the frontier in 2000, and (ii) changes in TFP against changes in distance from frontier over the same period. Distance from frontier = 100 if a country-sector pair is at the frontier (and lower for those behind the frontier). TFP levels are calculated as described in the text.

The figures suggest that over 2000-2007 (i) country-sectors far from the frontier in 2000 did not necessarily show big increases in TFP levels, and (ii) even when TFP growth is positive, distance from the frontier may increase. In many cases, increase in TFP levels was higher for countries closer to the frontier (such as in ICT, manufacturing, construction and distribution), and near zero or negative for those considerably distant from it (left panel). Moreover, even in cases where TFP growth is positive, distance from the frontier can still widen (barring the case of financial and business services, and other production, where by and large the distance shrinks with TFP increase; right panel).

The pattern observed in the scatter plots are consistent with numerous explanations, such as comparative advantage, country-specific factors impeding catch-up growth, factors specific to the observation period, etc. For our purposes, it helps to serve as a cautionary note in interpreting distance from the frontier in 2007 as ready space for productivity improvements.

III. Medium term impact of selected reforms on sectoral TFP levels in Australia

The results presented below are based on the following panel regression: ³⁴

t f p_{i, j, t + k} - t f p_{i, j, t} = β_{0} + β_{11} S_{A, j, t} + β_{12} S_{i *, j, t} + β_{2} S_{A, j, t} * t f p g a p_{A, j, t} + β_{3} t f p g a p_{i j, t} + β_{4} Δ t f p_{L, j, t} + β_{5} X_{i, t} + α_{1} D_{i} + α_{2} D_{j} + α_{3} D_{t} + ε_{i, j, t}

where cumulative changes in TFP for each industry j over k periods are expressed as a function of reform shocks S (specific to Australia, and to other countries i*), reform shocks interacted with Australia’s TFP gap from the frontier in industry j at time t), the distance to the frontier, TFP growth at the frontier L, and other control variables including country, industry, and year dummies.

Reform shock episodes are identified as a significant change (in absolute terms) of a structural indicator above a certain threshold. In particular, a change below one standard deviation of the average annual change is considered for product and labor market regulations, and labor tax wedge and an increase above two standard deviation for R&D, ICT capital, and high-skilled labor.³⁵ Under these assumptions, reforms episodes for Australia are identified as below.

Identified shock episodes to reform variables for Australia

	Manufacturing	Other Production	Finance/Business services	ICT	Distribution	Total
Product Market Regulation						1987, 1991-2003,
Product Market Regulation						2005, 2007
Job Protection Legislation						2007
Labor Tax Wedge						2006
High-skilled Labor			1987, 1988	1987, 1988
R&D	1991, 1992, 1994,			1994, 1997, 2000,
	1997, 2000, 2001,			2001, 2003, 2007
	2003, 2007
ICT Capital	1971, 1974, 1982,	1974	1989, 1994, 1995,	1971, 1974, 1982,	1978
	1984, 1986, 1992,		1997, 2007	1984, 1986, 1989,
	1995, 1996, 1997,			1995, 1996
	2000, 2007
Infrastructure						1972, 1973, 1974,
Infrastructure						1982, 1994

Identified shock episodes to reform variables for Australia

	Manufacturing	Other Production	Finance/Business services	ICT	Distribution	Total
Product Market Regulation						1987, 1991-2003,
Product Market Regulation						2005, 2007
Job Protection Legislation						2007
Labor Tax Wedge						2006
High-skilled Labor			1987, 1988	1987, 1988
R&D	1991, 1992, 1994,			1994, 1997, 2000,
	1997, 2000, 2001,			2001, 2003, 2007
	2003, 2007
ICT Capital	1971, 1974, 1982,	1974	1989, 1994, 1995,	1971, 1974, 1982,	1978
	1984, 1986, 1992,		1997, 2007	1984, 1986, 1989,
	1995, 1996, 1997,			1995, 1996
	2000, 2007
Infrastructure						1972, 1973, 1974,
Infrastructure						1982, 1994

Data are taken from OECD, EU KLEMS, and ANBERD databases, while infrastructure is taken as a principal component of roads, phone lines, and electricity generation capacity. The labour tax wedge is taken from the OECD Taxing Wages database, measured as percent of taxes and transfers paid in the share of total labour costs. The sample includes 23 market industries from 11 advanced economies over 1970-2007.³⁶

The results for Australia show that a decrease in product market regulation leads to an increase in TFP level in manufacturing, distribution, and personal series–related industries. A lower labor tax burden can have a positive TFP impact in Manufacturing and ICT-related industries. An increase in knowledge-based economy and investment in infrastructure are in general associated with higher TFP. For instance:

ICT capital had by far the largest economy-wide TFP impact. A large increase in the ratio of ICT capital to total capital services (above 0.3 percent) leads to a cumulative increase in TFP of 22 percent for total economy after 5 years.
An increase in the ratio of skilled labour to total labour ratio (above 17 percent) is associated with an increase of 6.5 percent in TFP levels after 5 years.
An increase in the ratio of R&D spending to industry value added (10 percent) leads to an increase in aggregate TFP levels by about 6 percent after 5 years.
An increase in the infrastructure indicator (2 percent) leads to an increase in TFP levels by nearly 4 percent after 5 years.

These results not only highlight that previous major reforms have had a significant impact on TFP, but also suggest that certain areas may benefit from more targeted reform efforts, such as the distribution sector. Australia’s greater distance from the frontier in the distribution sector, as described in the main text, may also be a reflection of the lack of reform impact.

References

Barnes, P., L. Soames, C. Li, and M. Munoz, 2013, “Productivity in Manufacturing: Measurement and Interpretation,” Productivity Commission Staff Working Paper, November 2013.
- Search Google Scholar
- Export Citation
Ben Dolman, 2009, “What Happened to Australia’s Productivity Surge?”, The Australian Economic Review, vol. 42, no. 3, September 2009.
- Search Google Scholar
- Export Citation
Bureau of Research and Energy Economics (BREE), Resource and Energy Quarterly, September 2014.
- Search Google Scholar
- Export Citation
Dabla-Norris, E., S. Guo, V. Haksar, M. Kim, K. Kochhar, K. Wiseman, and A. Zdzienicka, 2015, The New Normal: A Sector-Level perspective on Productivity Trends in Advanced Economies, IMF Staff Discussion Note No. 15/3 (International Monetary Fund: Washington, DC)
- Search Google Scholar
- Export Citation
Dean Parham, 2012, “Australia’s Productivity Growth Slump: Signs of Crisis, Adjustment or Both?”, Productivity Commission Visiting Researcher Paper, April 2012.
- Search Google Scholar
- Export Citation
Gary Banks, 2012, “Productivity policies: the to-do list,” presented at Securing the Future, Economic and Social Outlook Conference, Melbourne, November 2012.
- Search Google Scholar
- Export Citation
Government of Australia, Submission to the Productivity Commission Inquiry into Public Infrastructure Australian Government, Office of the Infrastructure Coordinator (December 2013).
- Search Google Scholar
- Export Citation
International Monetary Fund, 2014, “Is it Time for an Infrastructure Push? The Macroeconomic Effects of Public Investment,” World Economic Outlook (October 2014), Chapter 3.
- Search Google Scholar
- Export Citation
International Monetary Fund, 2015a, “Where Are We Headed? Perspectives on Potential Output,” World Economic Outlook, Chapter 3.
- Search Google Scholar
- Export Citation
International Monetary Fund, 2015b, “Australia – Options for Tax Policy and Federal Fiscal Relations Reform,” Selected Issues Paper in Country Report XX/2015.
- Search Google Scholar
- Export Citation
Productivity Commission (2015), “Workplace Relations Framework: The Inquiry in Context”, Productivity Commission Issues Paper 1, January 2015.
- Search Google Scholar
- Export Citation
Ross Garnaut (2005), “Breaking the Australian Great Complacency of the Early Twenty First Century”, conference paper presented at Sustaining Prosperity, New Reform Opportunities for Australia, Economic and Social Outlook Conference, Melbourne Institute and The Australian, March 2005.
- Search Google Scholar
- Export Citation
Vassiliki Koutsogeorgopoulou, and O. Barbiero, “Boosting Productivity in Australia,” OECD Economics Department Working Paper No 1025, February 2013.
- Search Google Scholar
- Export Citation
Vernon Topp, L. Soames, D. Parham and H. Bloch (2008), “Productivity in the Mining Industry: Measurement and Interpretation,” Productivity Commission Staff Working Paper, December 2008.
- Search Google Scholar
- Export Citation
World Trade Organization, 2015, Report by the Secretariat, Australia Trade Policy Review, WT/TPR/S/312, March 2015.
- Search Google Scholar
- Export Citation

Prepared by Adil Mohommad.

For the 16 market sectors covered by Australian Bureau of Statistics (ABS), 2014 multifactor productivity growth was 0.2 percent on an hour worked basis for 2014, and -0.13 percent on a quality adjusted hours worked basis.

These results are derived using the IMF Research Department’s Multivariate Filter (MVF) model. The MVF model is a statistical (Kalman) filter model that uses information from actual growth, inflation, and unemployment to estimate the output gap, potential growth, and equilibrium unemployment (NAIRU – Non Accelerating Inflation Rate of Unemployment, which is a level of unemployment consistent with stable inflation). As a consistency check, the model imposes the estimated potential growth on a simple production function, where the labour input evolves according to an econometric gender-cohort-wise LFPR model. The Appendix presents details on cohort and gender-wise LFPR trends from the regressions (IMF 2015a).

Given the uncertainty around estimates of NAIRU, a rate of 5.5 percent is also plausible, but does not change the results qualitatively.

$K_{t + 1} = I_{t + 1} + (1 - \partial) K_{t} \Rightarrow \frac{Δ K_{t + 1}}{K_{t}} = \frac{I_{t + 1}}{K_{t}} - \partial$ , where I is investment, K is capital stock, and ∂ is the rate of depreciation.

The rate of investment growth consistent with capital stock growth of 2.5 percent is 2 percent, in line with Consensus investment growth forecasts of 1.9 percent over 2016-2020.

Resources and Energy Quarterly (September 2014), Bureau of Research and Energy Economics (BREE).

We assume between 70-75 percent of output growth is measured TFP growth.

The average non-mining TFP growth contribution of 0.5 percent over the medium term is in-line with the contribution of this sector over 2003-2013 to aggregate TFP growth.

The estimate presented here is a variant of the model that uses information from actual output, inflation, and unemployment, and Consensus forecasts of inflation and growth to address the end-point problem with statistical filtering. For details on the filtering methodology, see IMF (2015a).

“The New Normal: A Sector-level Perspective on Productivity Trends in Advanced Economies”, IMF (2015b). The paper includes a discussion of the sector-specific impact of reforms in product market regulation, job protection legislation, labour market inefficiencies, infrastructure, R&D spending, ICT capital intensity, and skilled labour use on TFP. See Appendix for more details.

Data input and model results for Australia provided by Aleksandra Zdzienicka and Jovana Sljivancanin (both SPR).

The literature generally does not identify low investment in such productivity enhancing inputs as R&D, skilled labour, infrastructure, and ICT capital as inhibiting productivity growth in Australia (Dolman 2009), but it also does not rule out the potentially positive impact on TFP growth from additional investments in future (Eslake and Walsh, 2011).

This analysis is taken from IMF 2015b, based on EU KLEMS sectoral TFP growth and Groningen Growth and Development Centre (GGDC) Productivity Level databases (Inklaar and Timmer, 2008). The frontier is defined by the country-sector pair with the highest TFP level in 2007 (=100). Other country-sector pairs are normalized relative to the frontier. The EU KLEMS Growth and Productivity Accounts provide data up to 2007 on TFP growth across 72 industries. The EU KLEMS project ran from 2003 to 2008, and will not be updated further. The GGDC 1997 benchmark dataset provides sectoral TFP levels relative to the U.S.

Submission to the Productivity Commission Inquiry into Public Infrastructure, Australian Government, Office of the Infrastructure Coordinator (December 2013).

World Economic Outlook (IMF, October 2014), Chapter 3.

The level of regulations appears higher in the ICT sector. This seems to be on account of 100% state ownership of the largest postal and basic courier service provider, as well as barriers to entry in providing basic postal services. However on account of Australia’s large size and remoteness of locations, this market structure may be warranted.

See “Australia – Options for Tax Policy and Federal Fiscal Relations Reform” on pages 48-73.

This is derived using the EU KLEMS database, 2007 sectoral TFP growth indices (1995=100) for U.S. and Australia, as follows. Baseline sectoral indices of TFP are constructed by “growing” ABS 16-market sector TFP indices by the average growth rate for each sector recorded over 2003-2013. A counterfactual index is then constructed by grossing up the baseline index by the ratio of the index for US and Australia in 2007 for a particular sector, in effect assuming no gap widening since 1995. Finally, a “reform scenario” index is constructed assuming each sector closes half the gap between the baseline and the counterfactual index over a period of 10-15 years. Shares in gross value added are used to compute aggregate TFP growth under reforms.

Australia Trade Policy Review, WTO 2015; Banks (2012).

For a summary of some of the measures that are proposed and under implementation, see OECD Economics Department Working Paper No 1025, Feb 2013.

Workplace Relations Framework: The Inquiry in Context”, Productivity Commission Issues Paper 1, January 2015.

For details regarding this section, refer to the Technical Appendix, IMF (2015b).

Product and labour market regulations and infrastructure do not change much over time, hence a major reform is assumed when the indicator is below (above in the case of infrastructure) one standard deviation.

Countries include Australia, Austria, Denmark, Finland, Germany, Italy, Japan, Netherlands, Sweden, United States, and United Kingdom.

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/002/2015/275/article-A002-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Australia: Selected Issues

Author: International Monetary Fund. Asia and Pacific Dept

Volume/Issue:: Volume 2015: Issue 275

Publisher:: International Monetary Fund

ISBN:: 9781513556499

ISSN:: 1934-7685

Pages:: 74

DOI:: https://doi.org/10.5089/9781513556499.002

Keywords
Sustaining Income Growth in Australia1
References

View raw image

Australia has grown well above the rate of peer countries

(GDP growth, percent)

View raw image

Mining Sector Growth Impulse is Declining

View raw image

Output and income growth are well below historical average

(Percent change)

View raw image

Labour productivity is being driven capital accumulation

(Contribution to growth, percentage points)

View raw image

Productivity growth has fallen sharply since the 1990s

(TFP growth rate, percent)

View raw image

The decline in productivity growth is broad-based

(Percent change in MFP weighted by share in gross value added)

View raw image

Employment growth will remain steady with strong population growth

(Percent change)

View raw image

Mining sector productivity growth will recover in the near term

(Contributionto MFP growth, percentage points)

View raw image

Potential growth estimates lie in a narrow range

(Percent change)

View raw image

Disposable income growth will weaken significantly

(Real net national disposable income per capita growth; average)

View raw image

Potential growth is still high compared to other advanced economies

(Percent change)

View raw image

Australia: The Medium-Term Impact of Structural Reforms on TFP

(percent; average technological gap per industry)

View raw image

Doing Business (2015): Distance From Frontier

(100 = maximum)

View raw image

Scope to raise TFP In some sectors

(Comparative TFP Levels by Sector (Normalized; 100 - frontier)

View raw image

Regulation, Skills, ICT, and R&D Indicators Across Sectors and Countries (2007) ^1/

View raw image

Reforms required to boost future income

View raw image

Corporate tax rates are among the highest in OECD

(Percent)

View raw image

Trends in Labour Force Participation

(share of working age population)

View raw image

Contribution to cumulative change in aggregate trend LFP

(1995=baseline)

View raw image

Labor Participation Cohorts: Women

View raw image

Labor Participation Cohorts: Men

View raw image

View raw image

Medium-Term Impact of Structural Reforms

(percent change in TFP levels)

Identified shock episodes to reform variables for Australia

	Manufacturing	Other Production	Finance/Business services	ICT	Distribution	Total
Product Market Regulation						1987, 1991-2003,
Product Market Regulation						2005, 2007
Job Protection Legislation						2007
Labor Tax Wedge						2006
High-skilled Labor			1987, 1988	1987, 1988
R&D	1991, 1992, 1994,			1994, 1997, 2000,
	1997, 2000, 2001,			2001, 2003, 2007
	2003, 2007
ICT Capital	1971, 1974, 1982,	1974	1989, 1994, 1995,	1971, 1974, 1982,	1978
	1984, 1986, 1992,		1997, 2007	1984, 1986, 1989,
	1995, 1996, 1997,			1995, 1996
	2000, 2007
Infrastructure						1972, 1973, 1974,
Infrastructure						1982, 1994

Topics

Countries

Series

About

Resources

Abstract

Sustaining Income Growth in Australia1

A. Overview

B. Stylized facts: Strong growth in the past, but weakened outlook

C. Baseline scenario of medium-term potential growth

D. Can additional reforms stimulate higher growth?

E. Conclusions

Appendix—Technical Notes

I. Trends in Labour Force Participation

II. Changes in TFP levels and in relative distance to frontier among selected advanced economies and aggregate industry groups.

III. Medium term impact of selected reforms on sectoral TFP levels in Australia

References

Same Series

Other IMF Content

Other Publishers

Asian Development Bank

Inter-American Development Bank

International Labour Organization

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics

Sustaining Income Growth in Australia¹