It is a great pleasure to be here today to talk to members of the UDIA in Wollongong. To start, I'm going to give an overview of how the Australian economy has been travelling and how the drivers of growth have been evolving. In particular, I'm going to highlight the growth in non-mining business investment and the fact that dwelling investment is no longer contributing, although it is likely to remain at a high level for some time. Then I am going to turn to the outlook for housing markets across states, focusing on differences in the drivers of demand and how they relate to demographic changes. Finally, I will take this process of disaggregation one step further to look at some of the factors that have been driving demand for housing in Wollongong and the Illawarra region. Clearly, the close proximity to Sydney, which is only likely to strengthen with improvements to transport infrastructure, is an important factor.

The Macroeconomic Outlook

GDP growth in Australia has averaged around 2½ per cent over the past five years or so (Graph 1). This is a little bit below our estimates of trend growth, so you would say that growth has been decent, but not really enough to make significant inroads into the spare capacity in the economy. Low and stable wage pressures and consumer price inflation are also consistent with this general picture.

GDP growth picked up over the year to the March quarter to around 3 per cent, and more recent data have been positive. Our central forecast is for growth to be at or above 3 per cent over 2018 and 2019. This should lead to a further gradual reduction in spare capacity and a gradual increase in wage and inflationary pressures.

To understand this central forecast, it is necessary to look at how the composition of growth has been changing (Graph 2). The drag on growth from falling mining investment has diminished. This has played out largely as expected and we anticipate that most of its effect will be done by early next year. The contribution to growth from the public sector has also increased. The other two points I want to highlight are that there has been a significant pick-up in the contribution of investment by non-mining firms, but that dwelling investment is no longer contributing to growth.

The pick-up in non-mining business investment matters because it is hard to imagine a sustainable pick-up in growth without investment being a part of the story. New investment is required to replace depreciated capital and it is one of the ways new technology gets embedded into the way goods and services are delivered. So, long periods of weak investment growth can create concerns that the foundations are being laid for weak productivity growth in the future.

It is fair to say that we are now more confident about a sustainable pick-up in non-mining business investment. In 2015, our discussion of the risks around our forecasts highlighted the weakness in non-mining business investment and noted that indicators of investment intentions provided very little evidence that a recovery was imminent. This was despite the fact that many of the preconditions for a recovery, such as reports of above-average business conditions, were in place. We spent a considerable amount of effort trying to understand why investment outside the mining sector was low.^[1]

Rolling forward, non-mining business investment grew by 10 per cent in year-ended terms in the most recent national accounts (Graph 3). A lot of the pick-up in non-mining business investment over recent years has been from the non-residential construction category, which includes a range of things like hotels and office blocks. The ‘other’ category, which includes investment in areas like computer software and research & development, has also been adding noticeably to growth. Liaison suggests that many firms have been more willing to spend on IT, particularly software, to become more productive, reduce costs and improve their service quality.

Machinery & equipment investment has become increasingly important to the non-mining business investment growth story. The pick-up in public infrastructure work being done by the private sector is a part of this story (Graph 4). Our liaison contacts have reported that the recent increase in the flow of public infrastructure work, and expectations of further infrastructure spending based on both state and federal government commitments, have encouraged contractors in the construction industry to commit to investments in new construction equipment.

The Outlook for Housing

As already noted, dwelling investment has gone from making a positive contribution to growth two years ago to being roughly flat over the year to March. In terms of our forecasts, dwelling investment is not expected to contribute much to growth over the next couple of years, but is expected to remain at a high level.

To understand the outlook, it is helpful to recognise that there isn't a single national housing market. At the state level, there have been some similarities in the evolution of dwelling investment, but there have also been distinct differences (Graph 5).

One point of similarity is that the construction of higher-density apartments has been much more important than in the past, especially in the east-coast capitals. We have used our liaison program quite extensively to understand how to adapt our forecasting processes to take into account that the time taken for a building approval to progress to construction and the period of construction is longer and more variable for high-density projects than for detached dwellings. The liaison program, which includes organisations such as the UDIA and its members, has also allowed us to gain deeper insights into specific local factors, such as differences in planning rules and the emergence of capacity constraints in the housing construction sector.

One point of difference across states has been the timing of dwelling investment cycles. For New South Wales and Victoria, the level of dwelling investment has been broadly stable at a high level since 2016. In contrast there has been a decline in higher-density construction in Queensland since early 2017. In Western Australia, residential construction peaked in mid 2015, which was well after the end of the mining boom. These differences highlight the fact that there are different demand and supply forces at work across the states. Given time constraints, I am going to focus my attention on the demand side of the market.

An important driver of housing demand over the long run is the rate at which new households are being formed. This depends on population growth and changes in the average number of people who are living in each household. Household size declined steadily in Australia between 1960 and 2000 before levelling out, alongside declines in marriage and fertility rates and population aging. The natural increase in the Australian population has also declined over time due to demographic factors. In particular, lower fertility rates have offset increased life expectancy (Graph 6). Having said that, the rate of natural increase in Australia's population remains higher than in most other advanced economies.

Immigration has also been a feature of the population growth story and it has certainly been the dominant influence on the swings in population growth over the past decade. The largest single category of net overseas migration has been people on temporary student visas (Graph 7). Prior to the financial crisis, a large share of these students were coming to Australia for vocational training courses. Following changes to visa requirements, student visa numbers initially dropped, but have picked up again in recent years, mostly due to an increase in students attending university. To put this into perspective, education now accounts for around 10 per cent of Australia's total exports, which is in the same ball park as our rural exports. From the perspective of demand for housing, the important point is that most of these students have gone to Sydney and Melbourne.

Another interesting category is skilled workers. The net inflow of people on skill visas increased in response to demand for workers during the mining boom. Most of these workers went to Western Australia and Queensland. At the same time, net migration to Western Australia and Queensland from other states and New Zealand also increased. As the mining sector transitioned from the construction to the production phase of the mining boom, the demand for labour fell. The number of people on skilled visas fell and the inflow of people from New Zealand and other Australian states turned to an outflow.

As a consequence, there have been quite large differences in population growth at the state level, which have had direct effects on the demand for housing (Graph 8). Population growth is expected to remain strong, particularly in Victoria and New South Wales, and the net overseas migration component of this is expected to be driven by people on student visas.

On the supply side, the pipeline of residential construction that has been approved, but not completed remains high in New South Wales and Victoria (Graph 9). There is also a reasonable pipeline of work in Queensland, although it has already started to decline. Based on recent approvals data and expected demand conditions, this suggests that dwelling investment in New South Wales and Victoria will remain at a high level for a number of years. Liaison contacts have suggested to us that capacity constraints in the construction industry, particularly in New South Wales, will make it difficult for construction activity to increase.

Of course household formation and population growth are not the only drivers of housing demand. For example, interest rates and changes in lending standards can also influence how much households are willing and able to spend on housing. Another way to gauge the current balance of housing supply and demand is to look at housing price growth.

Over the past five years, housing price growth has been subdued in Brisbane and Perth (Graph 10). This is consistent with the fall in population growth coinciding with an increase in the supply of housing. In contrast, housing price growth has been strong until recently in Sydney and Melbourne, where population growth has been strong. Given that housing accounts for around 55 per cent of total household assets, we are paying close attention to these developments.

The Housing Market in the Illawarra Region

From a demand perspective, the Illawarra region has experienced a pick-up in population growth. Some of this has come from overseas students attending the University of Wollongong, and some has come from people migrating to the Illawarra region from Sydney. Although the Illawarra region is a little older, on average, than the rest of Australia and Sydney, it still has a large working-age population (Graph 11).

This is partly because its geographic proximity and transport infrastructure allow people living in Wollongong and the Illawarra region to commute to Sydney. Around 20 per cent of Wollongong workers commute at least 50 kilometres to work (Graph 12). This is one of the highest rates in the state. Unsurprisingly, five of the seven areas with higher shares of people commuting more than 50 kilometres are also within commuting distance of Sydney. Illawarra residents are also well placed to benefit from the fact that some of the fastest growing areas of Sydney are in south and south-west, including the proposed “aerotropolis” around the new airport at Badgery's Creek. Access to these growth areas will be enhanced if some of the recently announced transport infrastructure plans are realised.

Although people from the Illawarra region can and do commute to Sydney, labour market conditions in the Illawarra region itself have also been strong recently (Graph 13). In combination, these factors mean that there has been strong employment growth for those living in the Illawarra region over the past five years and the unemployment rate is close to the average for New South Wales, which is, in turn, lower than the Australian unemployment rate.

Strong population growth and the economic prosperity associated with strong labour market outcomes have led to higher housing prices in the Illawarra region (Graph 14). Just as in Sydney, developers have responded to the higher prices, and dwelling investment in the region has increased. Also similarly to Sydney, there has been a debate about whether the infrastructure has been growing fast enough to accommodate the needs of an expanding population and the increase in construction that goes with that.

Conclusion

In summary, over the past couple of years, non-mining business investment has become a more important driver of growth in the Australian economy. This is a good thing because investment of this kind is necessary to ensure future productivity growth, which is ultimately what contributes to the economic prosperity and welfare of the Australian people. Infrastructure investment has been a part of this story.

At the same time, dwelling investment growth has eased off. Although dwelling investment is still expected to remain at a high level, particularly in New South Wales and Victoria, it is not likely to contribute much to growth over the next couple of years. Demand for housing remains strong because population growth is expected to stay strong. However, the housing story is different across states and across regions within states, partly because population trends differ. The effects of the mining investment cycle on population trends and housing markets in Western Australia is a clear-cut illustration of this point.

The data show that population trends and housing market developments in the Illawarra region are closely linked to those in Sydney, partly because the transport infrastructure allows people to live in the Illawarra region and commute to Sydney. Future transport infrastructure plans and the development associated with the Badgery's Creek airport are likely to strengthen these ties. As always, the key to effective urban development is high-quality, transparent cost-benefit analysis of potential infrastructure projects informed by local knowledge. The UDIA has an important role to play here. The UDIA and its members, in Wollongong and elsewhere, also have an important role to play in macroeconomic policy by informing the Bank's understanding of the factors at play in different housing markets through our liaison program.

Endnotes

I would like to thank Alexandra Brown for her assistance with this speech, and many other colleagues for their valuable comments. [*]

If you are interested in knowing more, there is a good summary of this work in a recent Bulletin Article: Van der Merwe M, L Cockerell, M Chambers and J Jääskelä (2018), ‘Private Non-mining Investment in Australia’, RBA Bulletin, June, pp 1–18. [1]