Statement on Monetary Policy – May 2017 3. Domestic Economic Conditions
The Australian economy grew by 2½ per cent over 2016, which is a bit below central estimates of potential growth (Table 3.1; Graph 3.1). GDP growth rebounded to 1.1 per cent in the December quarter, confirming that the weak outcome in the September quarter largely reflected temporary factors. Recent data are consistent with moderate growth in early 2017. Employment growth was fairly subdued over 2016, but picked up to be around average in early 2017. Looking ahead, economic growth is expected to pick up gradually to be a bit above potential growth, supported by the low level of interest rates and the ongoing recovery in the global economy.
The economic adjustment to the end of the mining investment boom is well advanced. Activity in the mining sector (net of imports) contributed to growth over 2016, as the increase in resource exports more than offset the decline in mining investment. Outside the mining sector, household spending, business investment and dwelling investment have all contributed to growth, supported by low interest rates. Public demand has also been strong.
Economic conditions continue to vary across the states. Recently, growth in final demand has been strongest in New South Wales and Victoria, weakest in Western Australia and has picked up in the rest of the country (Graph 3.2). There are some signs that economic conditions are stabilising in the mining-exposed states. Final demand growth has picked up in Queensland over the past 18 months or so, and liaison suggests that business sentiment has improved recently in Western Australia. Labour market conditions across the states have generally been consistent with patterns in final demand, although employment growth in New South Wales has been subdued.
Mining Activity
Mining activity (net of imports) increased in the December quarter, as resource exports increased further and mining investment rose unexpectedly (Graph 3.3). Despite the quarterly increase, mining investment continued to decline over 2016 as work on liquefied natural gas (LNG) facilities was completed and few new projects were commenced. However, most of the decline in mining investment looks to have occurred already, such that the drag on GDP growth will dissipate over the next couple of years. Both the ABS capital expenditure (Capex) survey of investment intentions and Bank liaison point to smaller declines in mining investment over this period. In addition, much of the remaining decline is expected to be in LNG investment, which is import intensive and uses relatively little labour. This suggests that the flow-on effects to the domestic economy are likely to be small.
The increase in commodity prices since late 2015 has boosted the profitability of firms in the mining sector. However, this has not led to a material increase in new investment or employment in the sector, partly because the increase in prices is not expected to be sustained. Information from the Bank's liaison and company announcements have indicated that firms have used the additional income to pay down debt, pay out dividends and increase share buybacks. At the margin, some mining firms have reportedly been able to make small-scale investments such as upgrades to machinery and equipment. Consistent with expectations, the prices of some commodities, including iron ore, have fallen noticeably over recent weeks.
Resource export volumes have increased strongly over the past year (Graph 3.4). The ramp-up in LNG production has been underway for the past year or so and LNG exports are expected to continue to grow strongly over the next few years. Iron ore export volumes are likely to be supported by increased production from Australia's low-cost producers. Coal exports are likely to be lower in the next few quarters than had been expected, because of the effects of Cyclone Debbie in Queensland in late March. The cyclone damaged coal rail infrastructure in the Bowen Basin region, which is one of the largest coking coal producing regions in the world and accounts for around one-third of Australia's coal exports.
Household Sector
Household consumption growth picked up in the December quarter, following a period of modest growth in mid 2016. A marked increase in goods consumption contributed to the pick-up. Low interest rates and ongoing growth in household net wealth have continued to support spending (Graph 3.5). However, household income growth has remained weak, and there has been a further decline in the household saving ratio. Indicators in early 2017 suggest that consumption growth has moderated since late last year (Graph 3.6). The volume of retail sales is expected to be little changed after growth picked up in the December quarter.
Low growth in household disposable income continues to weigh on spending. In the December quarter, real household disposable income was unchanged, held down by the fall in labour income. Households' perceptions of their personal finances have declined since late 2016. Surveys indicate that households believe that paying off debt is currently a wiser place for saving than investing in real estate (Graph 3.7); while this might seem consistent with more cautious behaviour, the overall household debt-to-income ratio has been on an upward trend for the past couple of years.
Private dwelling investment remained at a high level in the December quarter and is expected to continue to contribute to economic growth over the next year or so, particularly given the large volume of new apartment construction in the pipeline (Graph 3.8). However, the contribution to growth from dwelling investment is expected to decline gradually from mid 2017 because residential building approvals, especially higher-density dwelling approvals, have declined over recent months. If housing approvals remain at current levels, the pipeline will be worked off gradually to more normal levels. The decline in higher-density approvals has been concentrated in New South Wales (where housing activity has been quite strong of late) and Queensland (where conditions have been less favourable) (Graph 3.9).
Conditions in the established housing market continue to vary, both across cities and between different types of dwellings. Across the capital cities, housing conditions remain strongest in Sydney and Melbourne and weakest in Perth. This is true across a range of measures, from growth in housing prices and advertised rents to vendor discounts and days on market (Graph 3.10). Across the types of dwellings, price growth for detached houses has outpaced that for apartments, most obviously in cities where the increase in the supply of apartments has been large relative to the existing stock, such as Brisbane and Melbourne (Graph 3.11). Rents continue to decline in Perth and the vacancy rate remains elevated. In the other capital cities, rent inflation remains subdued and vacancy rates have been stable around their long-run averages for some time.
Looking ahead, low interest rates should continue to provide support to housing demand. Total housing loan approvals remain around record highs, although new prudential measures are expected to lead to some tightening in lending standards and to slow the rate of growth in credit (see ‘Domestic Financial Markets’ chapter).
Government Sector
Public demand grew at an above-average rate in the December quarter, driven by a rise in public investment, while public consumption was little changed. Recent strong growth in the pipeline of public infrastructure projects is likely to support public investment growth over coming years (Graph 3.12).
Non-mining Business Sector
Non-mining business investment increased by around 3½ per cent over 2016, although it has been subdued for several years and remains low as a share of GDP. Non-mining investment has picked up in New South Wales and Victoria, but has been relatively weak in Queensland and Western Australia over the past few years.
Looking ahead, indicators of non-mining business investment are somewhat mixed.
Investment intentions in the ABS Capex survey and forward-looking indicators of private non-residential building activity indicate that non-mining investment is unlikely to pick up substantially over the next year or so (Graph 3.13).[1] For example, the stock of private non-residential building work yet to be done remains at a low level and building approvals have been low in early 2017 (Graph 3.14). The Capex survey also points to subdued machinery and equipment investment. In contrast, survey measures of business conditions have remained above average over the past 2½ years, with conditions improving significantly in the non-mining states. Measures of capacity utilisation have increased, particularly for goods-producing firms. Non-mining profits grew strongly in the December quarter.
The Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) estimate that the volume of farm production has increased strongly over 2016/17, as favourable weather conditions during winter supported a record crop harvest. Farm GDP rose by almost 10 per cent in the December quarter, led by record wheat and barley production (Graph 3.15). More recently, some sugarcane, fruit and vegetable crops in Queensland have been damaged by Cyclone Debbie, but the impact on national farm production is likely to be small. Overall, farm production is forecast by ABARES to decline in 2017/18, as crop production returns to normal levels and livestock production continues to decline while farmers rebuild their herds.
The depreciation of the exchange rate since early 2013 has supported non-mining exports (Graph 3.16). Service exports, which include tourism, education and business services, have made a significant contribution to GDP growth over this period. Over 2016, manufactured exports increased and rural exports also rose following a record crop harvest.
Labour Market
Conditions in the labour market have been mixed recently. Employment growth picked up in the March quarter, but the unemployment rate has edged higher over recent months to 5.9 per cent (Graph 3.17). After declining over much of 2016, the level of full-time employment has increased over the past six months; growth in part-time employment nonetheless remains strong. Consequently, average hours worked remain at a historically low level. Measures of underemployment have edged higher, as a relatively high proportion of part-time workers would like and are able to work additional hours.
Labour market conditions continue to diverge across the states (Graph 3.18). Employment growth over the past year has been concentrated in Victoria due to solid demand and population growth. Victorian employment growth has been fairly broad based across household and business services industries, and in both Melbourne and regional areas. While employment in Sydney continues to trend higher, employment has fallen across the rest of New South Wales following particularly strong growth in 2015. The level of employment has picked up a little in Western Australia and Queensland since late 2016. Population growth has slowed sharply in these mining-exposed states following the end of the mining investment boom, partly reflecting a net outflow of people from Western Australia to other states.
In trend terms, employment growth over the past year has been concentrated in services industries, while goods-related employment has been largely unchanged (Graph 3.19). Within household services, employment increased strongly in education and declined a little in health care, where it had expanded strongly in the previous couple of years. Within business services, there has been an increase in professional, scientific & technical employment and administrative & support employment; the latter covers labour hire firms and may be capturing demand across a range of industries. The unchanged level of goods-related employment over the past year masks divergences within the sector; growth in construction and manufacturing employment has been largely offset by weakness in the wholesale and retail industries.
Part-time employment is now around one-third of total employment (up from one-quarter two decades ago). Over time, Australian firms have shown a greater tendency to respond to changes in demand by adjusting the hours worked by their existing employees (compared with changing the number of employees).[2] This could reflect increased labour market flexibility, possibly related to the shift towards a more service-based economy, labour market deregulation and technological change. Cyclical factors are likely to have been important in the concentration of employment growth in part-time jobs over the past year or so. For example, over that period, full-time employment declined in the mining-exposed states of Western Australia and Queensland and in the retail industry (which has been affected by heightened competitive pressures). Full-time youth employment also declined, which tends to occur in periods where there is spare capacity in the labour market. The participation rate for this group has fallen a little over the past year, and the youth unemployment rate remains high.
Analysis of the movement of individuals between different labour market categories can help to explain recent changes in the composition of employment growth between part-time and full-time employment. On average, there tends to be a net flow of workers moving from part-time work to full-time work. However, over the past two years, fewer part-time workers have moved to full-time work than normal. This is consistent with the higher level of underemployment as many part-time workers want to, and are available to, work additional hours. The net flow of workers from full-time employment to outside the labour force has also picked up; over the medium term, this trend will be partly driven by demographic changes, as more workers move into retirement age, but there also appears to be a cyclical component.
Forward-looking indicators of labour demand point to further moderate growth in employment over the next couple of quarters (Graph 3.20).
Vacancies data remain strongest for household services, in particular, in the health care industry, and in New South Wales and Victoria.
Labour Costs
Wage growth remains low. A number of factors have contributed to low wage growth, including spare capacity in the labour market as well as lower inflation outcomes and inflation expectations. However, these factors cannot fully explain the recent weakness.[3] Lower wage growth is also part of the adjustment in the economy to the earlier decline in the terms of trade and the end of the mining investment boom; the decline in wage growth has been most pronounced in mining and mining-exposed states and industries (see ‘Box A: Australia's Real Exchange Rate’). Information from firms in the Bank's liaison program as well as a survey of union officials suggests that wage growth is likely to remain unchanged over the next year.
The wage price index (WPI) increased by 1.9 per cent over 2016 (Graph 3.21). Private sector WPI growth was slightly softer at 1.8 per cent. The slowing in wage growth over recent years has been evident across all industries, but is most pronounced in mining and mining-exposed industries including construction and professional services. Wage growth has been strongest in the health and education industries, where employment growth has been relatively strong. The public sector WPI increased by 2.3 per cent over 2016, but public sector wages in the Australian Capital Territory and Western Australian grew by less than 2 per cent over the year.
The amount of spare capacity in the labour market is an important determinant of wage growth, although spare capacity is difficult to measure. One measure is the unemployment ‘gap’; that is, the difference between the current unemployment rate and an estimate of the rate of unemployment associated with a stable rate of inflation (often referred to as the non-accelerating inflation rate of unemployment or NAIRU) (Graph 3.22). The current unemployment gap estimate suggests that there is spare capacity in the labour market, although less than there was in 2015. However, wage growth has continued to slow over this period. It may be the case that there is currently more spare capacity in the labour market than suggested by the unemployment gap, as the underemployment rate has edged higher. That is, part-time workers who want and are available to work additional hours could also be putting downward pressure on wage growth. It is also possible that the relationship between wage growth and spare capacity in the labour market may have been affected by structural changes in the economy; for example, due to new work arrangements enabled by technological change or increased competitive pressures from the internationalisation of services trade. However, any resulting changes to workers' bargaining power is difficult to observe directly.
Recent analysis of the job-level WPI data suggests that the share of ‘large’ wage rises (more than 4 per cent) declined from over one-third in the late 2000s to less than 10 per cent in 2016, and that this reduced overall wage growth (Graph 3.23). This has been particularly pronounced for mining and mining-exposed industries. The share of WPI outcomes between 2 and 3 per cent has increased noticeably to now account for almost half of all wage changes.
CPI inflation outcomes might also have contributed to recent low wage growth. Information on enterprise bargaining agreements (EBAs) from the Department of Employment suggests that around 6 per cent of employees covered by EBAs have wage outcomes linked to the CPI, while two-fifths of a selection of firms in the Bank's liaison program indicated that the CPI was a primary determinant of wage setting. Furthermore, survey evidence from unions suggests that inflation outcomes and expectations are an important consideration for wage bargaining.
Average earnings per hour from the national accounts (AENA) is a broader measure of labour costs than the WPI: it includes non-wage costs, such as redundancy payments and superannuation, as well as the effect of compositional changes in the labour market. The AENA measure, which is volatile, has tended to grow at a slower pace than the WPI over recent years. This is partly because it is capturing the movement of workers from highly paid jobs in mining-related activity to lower-paid jobs in other parts of the economy. There is also some evidence that mining and mining-related firms have been replacing workers who are leaving with new employees on lower salaries. Information from the Bank's liaison program suggests that the movement of workers from the mining sector to other parts of the economy is well advanced, and its effect on AENA should be diminishing.
Footnotes
The Capex survey only covers about half of the non-mining business investment captured by the more comprehensive national accounts measure; it does not cover certain industries, such as some service industries that have seen relatively strong investment over recent years, and it also doesn't cover certain types of investment, such as in software and research & development. [1]
See Bishop J, L Gustafsson and M Plumb (2016), ‘Jobs or Hours? Cyclical Labour Market Adjustment in Australia’, RBA Research Discussion Paper, No 2016-06. [2]
See Bishop J and N Cassidy (2017), ‘Insights into Low Wage Growth in Australia’, RBA Bulletin, March, pp 13–20. [3]