RBA Annual Conference – 2011 Discussion
1. Bill Evans
It is a great privilege to have been invited to be a discussant at this prestigious Conference. The last RBA Conference I attended was in 2003 when the topic was ‘Asset Prices and Monetary Policy’. Many of the conclusions from that Conference around the potential impact of asset prices on economies were quite prophetic given the subsequent events in the global economy. For me, this Conference therefore represents a hard act to follow but the papers are of a very high standard and useful insights abound.
I much enjoyed Judy Yates's comprehensive coverage of the developments in Australia's housing market in the 2000s. She has referred to an amazing body of research on the issue and her bibliography is a valuable record of the debate over that period.
As a banker in Australia, we have great interest in the housing market. Mortgages represent 60–65 per cent of a bank's assets, so the outlook for growth and asset quality in this market is critical for Australian bankers. When we look at the housing market we always have two important questions in mind:
- What is the outlook for prices?
- How will the gap between demand and supply evolve in the future?
In one critical sentence in the conclusion to the paper, Judy provides me with her answers to these questions: ‘Increasing demand pressures will result in structural increases in dwelling prices as supply costs increase because of an innate scarcity of land’ (p 289). She is arguing that ongoing demand pressures will ensure that the gap between demand and supply is likely to remain, and as supply costs increase due to land scarcity so too will prices.
Appropriately, the paper considers demand and supply issues in sequence. On the demand front we are given a careful analysis of why demand was strong in the 2000s. A number of factors explained this robust demand: attractive taxation arrangements; buoyant price expectations; a structural fall in inflation, which permanently lowered interest rates and interest rate expectations; rising incomes and wealth; financial deregulation leading to a structural increase in the supply of credit and a narrowing of spreads; and high population growth.
On the supply front the following forces are discussed:
- availability of finance to developers (particularly since the global financial crisis);
- costs of producing new dwellings, such as construction, finance and land costs, of which urban land costs are the most significant (this is best illustrated by the divergence between construction costs and prices of established dwellings); and
- longer-term concerns, including taxes, infrastructure charges and regulations.
Ongoing upward adjustments to prices, which reflect supply price inflation and excess demand, might be an acceptable long-run equilibrium but this tells us little about the short-term outlook – surely an issue of considerable concern when considering the housing market.
The paper poses the key question of ‘why dwelling commencements have failed to keep pace with increases in underlying demand despite rising house prices’ (p 275).
Indeed, in studies by the International Monetary Fund, the long-run price elasticity of supply for housing in Australia has been estimated at around 0.5 per cent; lower than in other countries.
While the paper discusses affordability, particularly in the context of rents, I do not think it goes far enough in discussing it in terms of its significant role in explaining the dynamics of price changes and the demand/supply imbalance. In this light, I would have liked the paper to have pursued the issue of housing affordability more vigorously.
The ‘elephant in the room’ when discussing Australia's housing market is international comparisons of price-to-income ratios. On some measures, Australian cities and regions are among the least affordable markets in the world. At Westpac, we have considered a range of housing definitions (including apartments, semi-detached and fully detached) and income measures (such as average weekly earnings and median incomes), along with the ongoing shift toward families having two incomes. The conclusion from that work is that house prices in Australia are expensive, although not as expensive as indicated by some of the crude measures used in some studies.
Further, our affordability measures do indicate that, relative to the United States and the United Kingdom, Australia's affordability has deteriorated significantly. If we feed those affordability measures into the paper's concept of demand, the confident assertion of ‘increasing demand pressures’ might need to be qualified. Affordability will impact both effective demand and underlying demand. It will impact effective demand as some groups who cannot afford a home won't even try; while it might affect underlying demand by impacting family formation and immigration as potential citizens are discouraged by high dwelling prices. After adjusting demand for these factors and noting that the explanations in the paper for the boost to demand in the 2000s (such as financial deregulation, the permanent fall in inflation and inflationary expectations, and tax changes) are unlikely to be repeated, the optimistic outlook for demand might be overstated.
What if affordability has been so squeezed in Australia that effective demand is, in fact, not increasing? There seems to be an absence of new structural factors, such as financial deregulation, permanently lower interest rates, tax changes or competition among lenders, which might give a boost to demand. If we begin to question effective demand, then the imbalance between supply and demand may not be as wide as implied in the paper. In fact, constrained affordability might mean that effective demand is below supply, with a downward price adjustment being necessary to restore equilibrium.
My view is that the answers to the two questions I posed at the beginning – on the outlook for prices and demand/supply imbalance – needs a much more detailed analysis of the state of affordability in the current market. More attention should have been given to that issue in an otherwise admirably comprehensive paper.
2. General Discussion
Much of the discussion focused on the issue of whether or not housing was overvalued in Australia. One participant thought that while there were signs that this was possibly the case in 2003 and 2004, this was not the case now. However, other participants noted that Australian house price-to-income ratios and house price-to-rent ratios are high compared with other OECD countries and relative to history. Key structural reforms such as the introduction of inflation targeting and deregulation are often cited as explanations for the increase in prices, yet other countries with lower valuation metrics also experienced these reforms. One participant described the relatively high level of house prices in Australia as an equilibrium outcome of high population growth (and limited appetite for society to increase population density) and poor public infrastructure. It was also mentioned that while it was not clear that housing was overvalued in Australia, this does not mean that house prices cannot fall. Another participant also stressed that the fact that relatively high house prices have persisted does not mean they are no longer an issue.
Housing was also recognised as a key sector in the economy either as a potential origin of a shock or as a sector that could amplify a shock originating elsewhere. For example, a significant international shock could reduce incomes and with it confidence and domestic demand, causing superannuation fund wealth and house prices to fall. Another potential risk was thought to be a policy change in China, such as a significant reduction in the number of exit visas issued which could lead to a fall in demand for student housing in Australia.
There was also substantial discussion on policy initiatives relevant to the housing market. It was mentioned by one participant that housing has been an area of significant policy failure in Australia. For example, they thought that policies designed to increase the home ownership rate, particularly for low-income households, had been largely unsuccessful. It was mentioned that although the United States did have some success in increasing the home ownership rates of low-income households, this turned out to be unsustainable. Australia's policies, on the other hand, were claimed to have failed from the start, with policy benefits going to those who could already afford to buy a house. Also, the halving of capital gains tax was considered inappropriate, making negative gearing more lucrative as a way to avoid tax, not just defer it. It was also mentioned that these structural policy failures are difficult to remove. Another participant, however, noted that a positive effect of the favourable tax treatment of investor housing is the extra incentive to add to the rental housing stock.
Another key question highlighted during the discussion was how to increase the supply of housing in Australia. A big part of the problem was thought to be high land prices, which would suggest that policies should look towards either lowering the amount of land per dwelling (increasing population density) or increasing the supply of well-located land (which would mean investing in transport infrastructure). One participant cited survey results from the Grattan Institute which suggested there was a greater preference for higher-density living given its proximity to services than the existing housing stock would imply. A different participant mentioned, however, that this survey did not consider tenure of residence, which was an important factor in household decisions. A policy measure suggested by one participant to make housing more affordable was to remove the land tax exemption for owner-occupied housing. This would lower the incentive for building on big blocks of land and help to reduce supply pressures. Another participant stressed the importance of getting structural issues right. They stated that, while housing is a major consumption good for low-income households, it seems to no longer be a major public policy issue, with the public and community housing stock declining. It was pointed out that improving transport infrastructure was important but it also has to be affordable; poor transportation infrastructure can impact on the productivity of low-paid workers if they have to travel long distances to work. Further, on the supply side, it was suggested that affordable rental housing in good locations has to be put in place to help low-income households, while measures to limit demand were also important.
It was also mentioned during the discussion that, although the move to inflation targeting and lower interest rates reduced cash flow constraints making repayments more affordable for a given loan size, it also contributed to an increase in average loan sizes relative to income which has increased the real cost of debt. Further, with lower inflation, the real value of debt is not eroded as quickly and so households have higher debt for longer. The increase in the cost of debt repayments was cited as a possible explanation for recent weakness in consumption. One participant also highlighted that wealth effects were an important channel through which developments in the housing market can influence household consumption, saying that house price movements impact the wealth of two-thirds of the population.
Finally, the intergenerational nature of housing was also raised during the discussion. If housing is not consumed in retirement and is passed on to offspring, then the younger generation does not face an affordability problem (although they may have to wait to own a home). But it was also mentioned that there is a serious equity issue here: descendants of those who do not own their own house are likely to have difficulty entering the housing market. This channel was said to reinforce wealth inequalities within society.