Submission to the Productivity Commission Inquiry on First Home Ownership 2. Factors Behind the Recent Rise in House Prices

Explaining why house prices have risen is more involved than simply stating that demand exceeded supply. The market for dwellings has some important and unique features which are key to understanding why prices have behaved in a particular way.

First, the rate at which new houses are built is very small relative to the stock of existing houses; each year's supply of new houses is less than 2 per cent of the existing stock. The housing market is primarily an asset market. Most of the transactions in the housing market are associated with buying and selling existing dwellings. As a result, house prices could rise or fall irrespective of what was happening to the supply of new houses. For example, it is possible to conceive of a totally static population where the underlying demand for additional dwellings each year is zero, but where prices will still rise rapidly if people's purchasing power is substantially increased.[16]

Second, unlike many goods, the supply of housing is far from homogeneous. Location, in particular, matters a lot in real estate. A consequence of this is that housing on the urban fringe is generally not seen by homebuyers as a very close substitute for housing in established preferred areas, although at the margin, infrastructure development can improve substitutability. The importance of location means that, in most circumstances, an increase in supply in outer areas is likely to have only a relatively small effect on prices for houses in preferred locations, including those close to the city.

Third, and perhaps most importantly, homebuyers' purchasing power is mainly influenced by their access to credit. More than in any other market, the vast majority of purchases in the housing market are financed by debt. So the supply of and demand for credit become major determinants of house prices.

This chapter discusses three possible explanations for the increases in house prices and changes in affordability over recent years. First, it looks at changes in the structural demand for housing arising from population growth, immigration and household formation, and whether the supply of new houses has kept pace with demand. Second, it examines the impact of stamp duty and the First Home Owner Grant on prices and affordability. And third it focuses on changes in the supply and demand for credit, by both owner-occupiers and investors in rental property, and their implications for the housing market. It concludes that it is the third set of factors, namely changes in the supply of credit and the capacity and propensity of people to borrow, that provides the main explanation for the sharp rise in house prices over recent years.

2.1 Structural Demand and Supply

One possible reason for the rapid increase in house prices is that underlying demand has risen and supply has not caught up. In order to test whether this is the case, it is customary to establish the underlying demand for new dwellings each year on the basis of the growth in population or, if possible, the growth in household formation.

When this is done for Australia, it can quickly be seen that over the past decade the population has not grown any more rapidly than it has at other times in the post-war era (Table 5). In fact, while growth in population has picked up a little recently, the current pace of growth is well below rates experienced in the 1950s and 1960s. Given that overall population growth is not faster than in the past, it is hard to claim that one component of population growth, namely immigration, is a cause of rising house prices, even though the rate of immigration is now higher than five years ago.

Many social factors need to be taken into account in translating population growth to growth in the number of households, which is the key driver of demand for an increased number of dwellings. The increase in the number of divorces, the decline in the number of children per household and the ageing of the population all mean that the average size of households has declined. This has been partly offset by the tendency of single adults to either remain in the family home or to share accommodation. The difficulty in quantifying these effects means that estimates of household formation are necessarily less accurate than estimates of population growth. Nevertheless, the various estimates suggest that the pace of growth in the number of households is below its historical average, although in the past couple of years it appears to have picked up slightly in line with the pick-up in population growth.

Private sector estimates suggest that over the second half of the 1990s, underlying demand was running at about 140,000 new homes per year. While the number of homes built varied considerably from year to year over this period, the average number was in line with this estimate of underlying demand. More recently, estimates of underlying demand for new dwellings have picked up to around 160,000 per year, consistent with the pick-up in household formation. The number of homes built has also risen, with around 155,000 having being completed over the past year, and a considerable number due to be completed over coming months. Our overall assessment, based on these numbers, is that, at least at the national level, there is little evidence to suggest that aggregate supply has failed to keep up with the growth in underlying demand for new housing. Rising vacancy rates for rental accommodation provide further evidence for this assessment.

Notwithstanding this national picture, there may well be shortages in some geographical areas and over-supply in others. Similarly, there may be shortages of some types of housing and excess supply in others. In particular, there is now clear evidence that there has been an oversupply in inner-city apartments in some areas, and anecdotal evidence suggests there may have been a land-induced shortage of supply of detached houses in other areas. The cost of providing services for newly-developed land has also risen, which has tended to increase the price of houses in those areas.

Since we at the Reserve Bank do not have the expertise to evaluate the micro data, we will confine ourselves to the aggregate. A number of participants in the building and property industries have argued that a lack of supply in some locations is a primary reason for higher house prices. While supply issues may be a factor in some markets and help explain variation in price movements across cities, we see no evidence to suggest that, overall, a shortage of supply of new houses relative to the underlying demand is the main explanation for the widespread rise in house prices over recent years. That said, there may well be things that could be done to make supply of new dwellings more responsive to changes in demand in some areas.

2.2 Stamp Duty and Grants to First-home Buyers

As discussed in the previous chapter, a major factor limiting the ability of many potential first-home buyers to enter the market is the difficulty of finding sufficient funds to cover the up-front costs. The largest of these costs (excluding the deposit on the loan) is stamp duty, which amounts to around $5,000 on the average house purchased by first-home buyers, and considerably more in the larger cities. As house prices have risen over recent years, the barrier to home ownership posed by stamp duty has increased considerably. This reflects two features of the current arrangements.

The first is the fact that state governments have not materially adjusted stamp-duty thresholds as house prices have risen. As a result, the average rate of stamp duty payable on the median-priced house has increased substantially, both relative to house prices and average incomes. For example, in 1995 the stamp duty payable on the median-priced house in Sydney was equivalent to 15 per cent of annual income of a full-time wage earner in NSW. Today, the figure is more than 30 per cent. In Victoria, the increase has been even more pronounced (Graph 26).

The second is that the stamp duty concessions given to first-home buyers have not kept pace with the increase in prices. In Victoria, for example, first-home buyers receive a concession (subject to eligibility requirements) if the purchase price is less than $200,000, while in Queensland the threshold is $160,000. Some years ago, the relevant thresholds were sufficiently high relative to median house prices to be a benefit to many first-home buyers. However, today they are below the estimated average price paid by first-home buyers.[17] This is particularly the case in the larger cities.

Overall, while stamp duty has made it more difficult for first-home buyers to afford the up-front costs associated with home ownership, it is not the source of high house prices. Indeed, stamp duty has probably had a mildly depressing effect on prices as it reduces the amount that a household with a given borrowing capacity can bid for a house. This effect is, however, likely to be quite small.

It is also sometimes claimed that the First Home Owner Grant (FHOG) and the Commonwealth Additional Grant (CAG) have contributed to a deterioration in affordability by pushing up house prices. Under the FHOG, which was introduced in July 2000, first-home buyers are paid $7,000 as an offset to the introduction of the GST. A further $7,000 was paid under the CAG to first-home buyers that entered into a contract to build a new dwelling between March and December 2001, with this amount being reduced to $3,000 for contracts entered into between January and June 2002. The temporary nature of the CAG was designed to partly offset the large fall in construction activity after changes to the tax system in 2000.

Overall, the net effect of these grants has been beneficial to first-home buyers. While the additional purchasing power arising from the grants has, at the margin, added to the upward pressure on house prices, the impact has been relatively small and cannot explain the large overall gains in house prices since the mid 1990s. Importantly, in terms of affordability, these grants have eased the ‘deposit gap’ faced by many first-home buyers. As noted above, this is in contrast to stamp duty which has increased the ‘deposit gap’.

2.3 The Effect of Changed Financial Conditions

Financial conditions facing prospective homebuyers have changed substantially over recent years resulting in a considerable increase in the capacity of households to borrow. For most households, this is the critical factor in determining the price they are prepared to pay for a home. The change in financial conditions has had two important dimensions: a reduction in the cost of finance and an increase in its availability as a result of innovation and increased competition in the financial sector.

Mortgage interest rates in Australia have declined substantially from the levels that were typical in the 1980s and early 1990s (Graph 27). In large part this has reflected the shift to a low-inflation environment, which has been associated with generally lower policy interest rates both in Australia and around the world. Another influence on mortgage rates in Australia has been increased competition in the financial sector since deregulation, resulting in a significant compression of mortgage interest margins relative to the cash rate since the early 1990s. Currently mortgage interest rates stand at around 6½ per cent, roughly half the level that prevailed in the mid 1980s and an even greater reduction when measured from their late 1980s peak. Most of this decline occurred in the early to mid 1990s, with little net movement occurring since around 1997. While interest rates have continued to move up and down since then, they have done so around a lower average than was previously the case.

The effect of the increase in borrowing capacity has been reinforced by a marked increase in the availability of finance. Financial institutions have become much more willing to lend for housing, which is perceived as a low-risk activity, and have become more active in promoting their loan products, both to owner-occupiers and investors. New entrants to the home lending market have competed aggressively for market share and contributed to the decline in mortgage interest margins since the early 1990s. Increased competition and innovation have also brought an increase in the range of mortgage products on offer, with many lenders now offering products not widely available a decade ago (see Box 1 for a summary of these products). These products include home-equity loans, loans with flexible repayment schedules and redraw facilities, and interest-only loans. Loans with loan-to-valuation ratios in excess of 100 per cent are now also available, as are loans to individuals who, in years gone past, would have had considerable difficulty obtaining funding due to their work histories or other characteristics. In addition, as discussed further below, there have been a number of important innovations that facilitate entry to the investor market with very small initial cash outlays.

Other things held constant, it would be expected that a structural reduction in interest rates would encourage greater household borrowing and, to the extent the additional funds were directed to the housing market, would generate an increase in house prices. The question that arises then is whether this structural change can explain all or only part of the increase in house prices that has subsequently taken place. In order to address this question, movements in the key variables since the mid 1980s are summarised in Table 6. The base period of the mid 1980s taken here is chosen to be broadly representative of the situation of high average interest rates, high inflation and stable real house prices that prevailed prior to the property bubble in the latter part of that decade.

As noted above, mortgage interest rates in nominal terms have roughly halved since the mid 1980s. This decline has significantly reduced the initial servicing cost associated with a given mortgage.[18] For example, using the traditional benchmark that requires that interest plus principal mortgage repayments not exceed 30 per cent of gross income, a household today can borrow an amount equivalent to 3.7 times its annual income, compared with around two times income in the mid 1980s (Graph 28). In other words, the borrowing capacity of households relative to income is now almost double what it was in the mid 1980s.

This observation provides an initial benchmark for considering the impact of structurally lower interest rates: if household borrowing behaviour was entirely driven by households borrowing to the maximum extent of their capacity, we would expect the overall ratio of household debt to income to have increased by the same factor of just under two, as outlined above. However, that result would be based on an extreme assumption, and it is likely that the effect would be more muted, for two reasons:

  • First, there is no necessary presumption that all households would want to respond to lower interest rates by correspondingly taking on more debt to an extent that would leave their initial loan repayments unchanged. In normal times some may prefer a more modest increase in debt that would allow their loan repayments to fall.
  • Second, the reduction in mortgage rates has been much smaller in real than in nominal terms, since a large part of the reduction in nominal mortgage rates has been a reflection of lower inflation. As indicated in Table 6, real mortgage rates in recent years have averaged around 4½ per cent, only a modest reduction from the average of just over 5 per cent in the mid 1980s. The real debt service cost over the life of a loan has thus not come down by nearly as much as the fall in initial repayments, since inflation cannot be expected to reduce the real debt to the same extent as was typical in the past. To the extent that homebuyers take this into account in their borrowing decisions, they would increase their borrowing by less than the maximum increase indicated by the nominal repayment benchmark.

Given these considerations, it can be concluded that the structural reduction in interest rates since the mid 1980s would explain, at most, an approximate doubling in household borrowing relative to incomes over the period, and probably less. But in fact, the increase in household debt, as indicated in Table 6, has been considerably larger than that. Total household debt has increased from a ratio of around 45 per cent of income in the mid 1980s to 134 per cent by the June quarter 2003, and is still rising rapidly. Housing debt to owner-occupiers, the group that would be expected to respond directly to the increase in borrowing capacity in the way discussed above, has increased from around 30 to 76 per cent of income since 1990 (the earliest period for which this detail is available). Clearly, this increase has gone beyond what could be explained by the structural reduction in interest rates since the 1980s.

It is not possible to derive a mechanical link from the structural change in nominal interest rates to its expected impact on housing prices. An extreme assumption, along the lines already discussed above, would be that new homebuyers responded to the increase in borrowing capacity by commensurately increasing the amount they were prepared to pay for a home. For the reasons noted, if this had been fully capitalised into house prices it would explain at most a doubling of the ratio of house prices to income. The actual movement to date, according to the indices summarised in Table 6, has been an increase by a factor of 2–2½ over the period and, as discussed in detail in Chapter 1, these price indices are continuing to rise rapidly and at an increasing pace.

In summary, then, while the structural reduction in interest rates since the mid 1980s appears to explain a large part of the growth in housing-related debt and in house prices observed since that time, it is unlikely to account for these phenomena fully. This conclusion is also supported by a consideration of the relative timing of these events. As noted above, the increase in borrowing capacity relative to incomes was largely completed by around 1997. While the full take-up of this capacity, and its impact on house prices, could be expected to take some years, it would be reasonable to expect that the transition to a higher equilibrium level of prices would now be largely completed. In fact, house prices are not only still rising quickly but, in aggregate, have accelerated over the past year, even though the level of mortgage rates has been little changed for several years.

On this basis it seems clear that other factors beyond the change in interest rates have contributed to the increases in house prices and their recent acceleration. In this context, the general increase in the availability of finance, as a result of innovation and competition in the financial sector, has obviously been important in contributing to the growth of demand in the housing market over the past decade. As discussed in the following section, the impact of this development has been particularly evident in the rapid growth of demand from investors.

2.4 Demand for Property as an Investment

The extremely strong demand to purchase rental properties by household investors over recent years is unprecedented, both in terms of previous experience in Australia and experience overseas. The impact of this strong demand has been most evident in the inner-city apartment markets, although there has also been strong investor demand for apartments and houses in more established areas. The demand by investors for rental properties has added to the general upward pressure on house prices, and thus made it more difficult for first-time buyers to get a foothold in the market. It has also contributed significantly to the overall increase in household indebtedness, and the increased vulnerability of the household sector to a deterioration in the economy.

The strong demand for rental properties, and the accompanying price rises, has induced a considerable increase in the supply of apartments, particularly those for rent. While the number of people wishing to rent these apartments has also increased, partly as a result of demographic factors, the increase has not kept pace with the growth in the number of apartments available. The result has been the weakness in rents discussed in the previous chapter, and falls in prices in some areas.

Despite rental yields that are now very low, both by historical and current international standards (Table 7), investor demand has remained extremely strong. Indeed, as yields have declined further over the past year, the demand by investors appears to have increased further. Investors in Australia seem prepared to accept rental yields that are much lower than those required on commercial property and much lower than yields on rental properties in overseas markets.

To understand recent developments it thus important to understand why the demand from investors has been so consistently strong, and why the Australian household sector's appetite for rental property investments has far outstripped that of households in other countries. Part of the answer obviously lies in the rapid increase in house prices during the period when prices were adjusting to lower interest rates. Investors who sensed early on the potential for such gains have been able to earn high rates of return from the purchase of rental properties. However, with the adjustment of prices to lower interest rates now almost surely complete, demand has remained very strong.

The continuation of strong investor interest partly reflects the extrapolation of past increases in prices. This is hardly surprising. Over recent decades, property has been a sound investment, with prices rising in most years. On those rare occasions when prices have declined, the falls have been modest, and even people who bought at the peak of the late 1980s boom have recorded healthy returns on their investments. As a result, many people believe that property prices will not fall over any reasonable investment horizon. This is notwithstanding the fact that prices have fallen noticeably in some other developed countries and, in Australia, they have fallen in real terms on a number of occasions.

To the extent that past experience has been extrapolated many investors may ultimately be disappointed with their returns on their property investment. While lower interest rates have clearly justified a higher level of house prices, they have not justified higher rates of increase on an ongoing basis. Many investors have probably not fully recognised this distinction.

Another important factor contributing to the attractiveness of residential property over recent years has been the weakness in equity markets around the world. The strong gains in property prices relative to equity prices have reinforced the idea that property is a preferable investment to shares. This is reflected in responses to survey questions on where is the wisest place for individuals to invest their savings (Graph 29). This perception has been reinforced by the corporate governance scandals, particularly in the United States, and the evidence that some high-profile providers of investment advice have had serious conflicts of interest. Many households see holding and managing an investment property themselves as one way to avoid such problems.

The strong demand for property as an investment is, of course, not just restricted to investors in rental properties; investment considerations can also be important for owner-occupiers. As house prices have risen over recent years, many owner-occupiers have been prepared to borrow as much as possible in order to maximise their exposure to the rising property market. In many cases, households have purchased the most expensive property that they could afford, partly in the hope that future capital gains would help fund their retirement. Other owner-occupiers have been prepared to borrow heavily to purchase a house given their concerns that further increases in prices could make it impossible for them to purchase a house in their desired location in the future.

Notwithstanding this investment demand from owner-occupiers it is the behaviour of investors in rental properties that is particularly unusual in the current boom. Against the backdrop of expectations of continuing capital gains, two aspects of the financing and taxation treatment of investments in rental properties have been particularly important in underpinning this strong demand. These are:

  • financial innovation that has allowed investors to purchase an investment property with limited, or no, cash outlay; and
  • the relatively small cash outlay required to cover the ongoing costs of owning an investment property even when the rental yield on the property is very low.

Many investors are able to purchase an investment property through accessing equity in their existing home, without having to put in any cash up front. This outcome reflects the fact that in assessing loan applications, financial institutions will often compare the total value of debt on the owner-occupier and investment properties with the combined value of the two properties, in effect allowing 100 per cent debt financing of the investment property. Provided investors can service the loans from their overall income, lenders are typically not concerned that interest payments and other expenses far exceed prospective rental income.

More generally, the willingness of financial institutions to lend to investors has greatly increased over recent years. In past decades, individual investors could have considerable difficulty obtaining finance for an investment property, often having to rely on a combination of their own savings and funding from non-bank sources. In contrast, today, banks compete aggressively in this area, marketing investor loans to their entire customer base. They also offer products specially designed to be attractive to investors, including split-purpose and interest-only loans (see Box 1). Reflecting this change in attitude, lending criteria on investor loans are now, generally, not materially different to those for loans to owner-occupiers. And the interest rate charged is the same as that charged on owner-occupied loans, in contrast to the situation that applied until the mid 1990s when investors were charged a premium of 1 percentage point.

Another innovation that has allowed investors to obtain an exposure to the property market with minimal up-front cost is the deposit bond. While these bonds have helped lower the cost of bridging finance they have also permitted investors to purchase a property off-the-plan for an up-front cost of a few thousand dollars, a fraction of the cost of the traditional 10 per cent deposit (see Box 1). In doing so, they allow investors to obtain a highly leveraged exposure to the property market during the construction phase. While, ultimately, the investor needs to obtain funding to effect settlement of the purchase, it has not been uncommon to use these bonds to speculate on prices, with the investor hoping to on-sell the property before settlement is due.

Overall, the terms under which investors can access finance in Australia are considerably more generous than those that apply in other countries studied in preparing this submission. In most countries, investor loans are treated more like business loans than owner-occupier housing loans, and they are not marketed as aggressively as they are in Australia. Moreover, instruments equivalent to the deposit bond do not appear to exist in other countries. In the United States, individual investors in rental property are generally charged interest rates 25 to 100 basis points above those charged to owner-occupiers. Similarly, in Canada, while posted interest rates tend to be similar for owner-occupier and investment loans, banks negotiate larger discounts for owner-occupiers than for investors.[19] In the United Kingdom too, loans to investors usually attract a higher interest rate than that charged on owner-occupier loans, although the differences have tended to narrow over time. Stricter lending criteria are also generally applied, with lenders often requiring that rental income exceed interest payments.[20] In general, the mortgage products offered in these countries also permit less flexibility than is available in Australia for investors to draw-down equity on their existing owner-occupied property to help finance an investment property.

The second important factor that has underpinned investor demand is the fact that in many cases, investors need to make only a small ongoing cash outlay, even if the weekly rent falls far short of the investor's expenses (including interest). It is not uncommon, for example, for promoters of investment in rental properties to suggest that due to the operation of the tax system, investors can purchase an investment property worth $400,000 or $500,000 for as little as $50 per week. In addition, little or no up-front contribution from the investor is required.

Given current rental yields and realistic assumptions about interest rates and expenses, rental income for investors that have borrowed recently to purchase an investment property will be less than half their expenses, including interest. As discussed in the previous chapter, gross rental yields on apartments currently stand at around 3½ per cent, and after expenses, including body corporate fees, rates, agent's fees and maintenance, net yields are typically below 2½ per cent. In contrast, the average interest rate on housing loans is currently around 6½ per cent. This difference between net yields and mortgage rates means that the purchaser of a $400,000 rental property financed with debt would be out of pocket (before tax) by over $300 per week, or over $15,000 per year.

The actual cash-flow position of many investors is, however, significantly improved by the taxation treatment of rental properties.[21] In particular, the ability to claim depreciation deductions, and the ability to offset tax losses on the investment property against other income (commonly known as negative gearing), can substantially reduce the cash-flow burden from low-yielding rental property investments.[22] It is not unusual, for example, for depreciation deductions on buildings and fixtures and fittings to amount together to $10,000 per year on a new $400,000 apartment. Further, with net yields and interest rates at current levels, depreciation deductions mean that an investor could reasonably incur tax losses in excess of $25,000 per year on a $400,000 apartment. As illustrated in Box 2, this significantly reduces the weekly out-of-pocket expense of holding the investment property. In the example provided, the $400,000 investment property runs at a cash deficit of $331 per week before tax, but this is reduced to $81 per week after tax.

The desire by taxpayers to minimise their tax is long-standing. It is particularly important for individuals paying the top marginal rate of tax. In Australia, this top rate cuts in at a relatively low level of income ($62,501) by international standards; according to survey data, over 20 per cent of full-time wage and salary earners have gross incomes exceeding this threshold. With negatively-geared investments particularly attractive to individuals facing high marginal tax rates, a high share of Australian taxpayers are attracted to property investment to lighten their tax burden. This interaction of high marginal tax rates and negative gearing is frequently emphasised by the property seminar industry.

As with access to finance, the taxation arrangements for rental properties in Australia tend to be more favourable to investors than are the arrangements in other countries studied for this submission. The Australian arrangements, in conjunction with the relatively high marginal tax rates faced by many taxpayers, mean that investors holding low-yielding properties that are highly leveraged face a substantially lower ongoing cash-flow deficit than would investors in most other countries. This lower carrying cost has contributed to the continuing strong investor demand.

Under the Australian taxation system, there are no restrictions on the ability of taxpayers to negatively gear investment properties. There are no limitations on the income of the taxpayer, on the size of losses, or the period over which losses can be deducted. As discussed above, under plausible assumptions an investor purchasing a $400,000 property recently might have tax losses of $25,000 per year on the investment. Moreover, these losses can extend for many years into the future. If, for example, the net rental yield on the property is 2½ per cent, and rents increase at their average rate over the past decade (2½ per cent), tax losses would continue for more than thirty years if the property is financed by an interest-only loan. As discussed in Box 3, the rent on the property would need to increase by at least 10 per cent per year for the investment to be cash-flow positive within a decade.

In contrast to the arrangements that exist in Australia, negative gearing is not permitted in the United Kingdom, except in respect of ‘furnished holiday accommodation income’, and is not relevant in the Netherlands given their tax treatment of investment income.[23] In the United States, only taxpayers with an annual income of less than US$100,000 are able to fully negatively gear investments in residential property, and even then they must meet certain ‘activity’ tests, and losses in any one year cannot exceed US$25,000.[24] In Canada, negative gearing is allowed, but only if the losses do not arise from depreciation expenses. Furthermore, historically, losses have only been permitted for a limited number of years, although recent court decisions have weakened this restriction. Additional details of the arrangements in the various countries are provided in Table 8.

The treatment of depreciation in Australia also appears to be quite favourable, particularly when considered in conjunction with the fact that there are no restrictions on negative gearing. As discussed above, for many investors in new apartments, depreciation deductions can make a material difference to the cash-flow attractiveness of the investment. By way of contrast, there are no deductions for depreciation in the United Kingdom and the Netherlands. In North America, rates of depreciation for tax purposes are higher than those in Australia, although the restrictions on negative gearing mean that not all investors can take advantage of depreciation deductions to reduce their current tax bill. Table 9 provides further details on the arrangements in each of the country studied.

One aspect of depreciation that does not typically receive much attention, but is important for some investors, is its interaction with the capital gains tax (see Box 4 for an illustration). In those countries that permit depreciation deductions, the deductions reduce the cost base for calculating capital gains tax, and so increase the capital gains tax liability upon sale. Where the capital gains tax is levied at a lower rate than the income tax rate, this represents a significant advantage to the taxpayer. In Canada, where the capital gains tax is at half the income tax rate as in Australia, the authorities have addressed this point by applying the full income tax rate to that part of the capital gain arising from the downward adjustment to the cost base. The lower capital gains tax is then applied to the remainder of the capital gain. In Australia, the lower capital gains tax is applied to the entire capital gain.

The overall importance of negative gearing and depreciation deductions in Australia is evident in the fact that in 1999/00 (the latest year for which relevant data are available), 54 per cent of Australian taxpayers earning rental income recorded a tax loss on their investment.[25] In both 2000/01 and 2001/02, as in a number of other years over the past decade, investors, in aggregate, recorded an income tax loss on their investment in rental properties. In each of the other countries studied, investors, in aggregate, earned a positive return.

Another difference between investors in Australia and elsewhere is that in most countries the earning of rental income is seen as the most important reason for investing in rental properties. In the United Kingdom, for example, surveys suggest that two-thirds of investors plan to hold their rental property for more than 10 years and investors routinely report that rental income is the most important rationale for investment.[26] Similarly, in the United States, surveys indicate that only 10 per cent of rental properties are held by investors whose primary rationale for investing is long-term capital gain.[27] This seems to stand in contrast to the situation in Australia where properties are commonly marketed on the assumption that they do not earn positive taxable income for a considerable period.

One additional factor that has helped fuel demand for investment properties over recent years is the investment seminar industry. Operators of these seminars have been able to draw large numbers of potential investors, explaining how rapid price increases and favourable taxation treatment can make highly-leveraged property investments attractive, even when rental yields are low. While these seminars have not caused the boom in prices, they have contributed to its speculative elements. One difficulty is that much of the advice is provided outside the regulatory framework that applies to other types of financial advice. While securities and other financial advisors have significant obligations regarding disclosure and the suitability of advice, no such requirements apply in respect of real estate investment advice (see next chapter). The result has been that there has been relatively little oversight of the sometimes extravagant claims made in these seminars, or the conflicts of interest that can arise for those conducting the seminars.

Finally, another factor that has contributed to Australian households being more willing to purchase an investment property than households in other countries is that landlords have more control over their properties than is often the case elsewhere. In some jurisdictions in the United States, for example, it can be very difficult to evict tenants, even if they have not paid the rent or have damaged the property. In other jurisdictions, rent controls, or the fear of future rent controls, reduce the attractiveness of rental properties. As a result of these concerns, rental properties are generally considered a relatively high-risk, time-consuming investment by many households in other countries.

Box 1: Major Innovations in the Provision of Housing Finance

Product Description
Home-equity loans These loans provide a line of credit secured by a mortgage against an existing property and can be used for a range of purposes, including renovations or the purchase of an investment property. In some cases no repayments are required for a number of years, provided the outstanding debt remains below an agreed limit (generally up to 80 per cent of the value of the property). Currently, home-equity-type loans account for around 12 per cent of loans outstanding that are secured by residential property.
Mortgages with flexible repayment schedules and redraw facilities These arrangements allow borrowers to manage a temporary loss of loan servicing ability or to access loan repayments that have been made in excess of the minimum repayments required by the lender. As such they reduce the need for borrowers to maintain precautionary savings in low-interest deposit accounts and can offer a tax-efficient form of saving. The most flexible of such arrangements combine a home loan account, a transactions account and credit card account into the one facility.
Deposit bonds These bonds remove the need for the purchaser of a property to pay a deposit at the time contracts are exchanged. Instead, the purchaser pays the bond's issuer (typically an insurance company) a fee in return for a guarantee that an amount equivalent to the deposit will be paid at settlement. For short-term bonds, this fee can be measured in hundreds of dollars rather than the tens of thousands required for a conventional deposit. Even bonds with terms of up to three years, used to purchase property ‘off-the-plan’, are relatively cheap, allowing investors to gain a highly leveraged exposure to the property market during the property's construction phase. Developers report that deposit bonds have been used by up to 70 per cent of purchasers in some projects. It is estimated that they are used in up to 20 per cent of Sydney residential transactions, the market where their use is most widespread.
Interest-only loans For investors, the appeal of these loans lies in the scope for greater tax deductions than otherwise. For owner-occupiers, these loans may provide an opportunity to invest deferred principal payments in higher-yielding products. Some lenders do not require any principal repayment for 20 years although a five-year period is the most common.
High loan-to-valuation ratio (LVR) loans A range of financial institutions offer loans of between 97 per cent and 110 per cent of the property's purchase price. While such products have been available for at least two years, most high LVR loans have been made in the past six months. Various restrictions on the type of property (for example, investor and/or inner city) are imposed in an effort to reduce the lenders' credit exposure. In addition, high LVR loans usually attract a higher interest charge.
Low documentation loans These loans are designed for borrowers that are unable to gain approval for traditional lending products due to insufficient documentation – usually due to their employment situation (self-employed, seasonal or contract workers). These loans typically carry an interest rate 60–80 basis points above the standard variable mortgage rate and have a maximum LVR of 75–80 per cent.
Acceptance of other security One financial institution has recently introduced a home loan that allows customers to use the equity in their car as part of the security for the loan. The loan is principally designed for borrowers who wish to consolidate an existing mortgage and other outstanding debts, but are otherwise unable to meet minimum LVR requirements.
Split-purpose loans These loans allow a borrower to split a loan into two sub accounts, one for a home loan and the other for an investment loan. In the initial years, all loan repayments are directed to the home loan account with the interest due on the investment loan being capitalised. Subsequent interest payments and tax deductibility relating to the investment property are thus greater than otherwise.

The Commission of Taxation has recently been granted leave to appeal to the High Court regarding the Federal Court's decision that this type of product is not primarily designed to obtain a tax benefit.
Vendor finance loans Under these arrangements, a ‘mortgage wrapper’ obtains a standard mortgage over a property from a mainstream lender and on-sells the property to a third party (who occupies it) under an installment sales contract. The wrapper retains ownership of the property until the occupant makes all of his/her installments, that is, until the wrapper's loan to the occupant is fully repaid. The interest rate paid to the wrapper is typically 2–2½ percentage points higher than the standard mortgage rate. In addition, the mortgage wrapper usually requires the occupant to make repayments of principal well in excess of the purchase price paid by the wrapper – sometimes up to 25 per cent in excess.

For further details see Reserve Bank of Australia Bulletin, ‘Innovations in the Provision of Finance for Investor Housing’, December 2002, pp 1–5; Reserve Bank of Australia Bulletin, ‘Recent Developments in Housing: Prices, Finance and Investor Attitudes’, July 2002, pp 1–6; and Reserve Bank of Australia Bulletin, ‘Recent Developments in Low-deposit Loans’, October 2003, pp 1–5.

Box 2: Out-of-pocket Cost of Holding a Rental Property

This box provides some calculations on the carrying cost of holding a rental property with a gross rental yield of 3½ per cent.

The purchase price of the property is assumed to be $400,000 and the investor is assumed to have financed the purchase entirely with an interest-only loan, paying an interest rate of 6.6 per cent.

Expenses (e.g. insurance, agent's fee and body corporate charges) are assumed to be equal to 1.2 per cent of the value of the property, and annual depreciation charges are assumed to be $9,550 (see Box 4 for further details of depreciation). The investor is assumed to have taxable income from other sources equal to $95,000.

Based on these assumptions, Table 1 summarises the weekly cash flows associated with this investment.

Given the relatively low rental yield, the investor's weekly rental income is less than half total rental and interest expenses. If depreciation deductions and negative gearing were not permitted, as in the United Kingdom, the investor would need to find $331 per week from other sources in order to cover these losses.

However, once depreciation and negative gearing are allowed, the weekly out-of-pocket expense on the investment property falls to just $81. This reflects the fact that depreciation deductions increase the tax loss to $515 and this entire loss can be offset against other income, reducing the investor's tax bill by $250 ($515*0.485) per week.

Some promoters of investment property suggest that investors can claim even larger deductions for depreciation than used in this box. This would again reduce the weekly contribution that the investor would be required to make. So too would using some equity to finance the purchase of the property. For example, if the investor took out a loan equal to 85, rather than 100, per cent of the value of the property, the weekly out-of-pocket expense on the investment property would be around $42, rather than $81.

Box 3: Rental Property – Time to Cash-flow Positive

This box provides some calculations on the time taken for a rental property to become cash-flow positive under various scenarios.

The purchase price of the property is assumed to be $400,000 and the investor is assumed to have taken out an interest-only loan of $340,000 (i.e. the loan has an LVR of 85 per cent) at 6.6 per cent. It is also assumed that, initially, annual rental income less expenses is equivalent to 2½ per cent of the purchase price.

If rents (and expenses) on this investment property grow at 2½ per cent per year (the average rate seen over the past decade), it will take 34 years for this investment to generate positive cash flow (Table 1). If depreciation is included in the calculations, tax would not need to be paid until the 40th year and the cumulative tax losses would be considerably larger (based on the depreciation assumptions used in Box 4).

Obviously, the faster that rents grow, the shorter is the time taken for the investment property to become cash-flow positive (Graph 1). If, for example, rents were to increase at 5 per cent per year, it would take 18, rather than 34, years for the property to generate positive cash flow. Only if rents increased by at least 10 per cent per year would the property be cash-flow positive within a decade.

If the initial net rental yield is 3½ per cent, rather than 2½, the property becomes cash-flow positive more quickly, although for plausible assumptions about rental growth it still takes many years. For example, if rents grow at 2½ per cent per year then the property will take 21 years to become cash-flow positive. If they grow at 5 per cent, it will take 11 years. In contrast, if net rental yields are much higher, say 6 per cent, then the investment is cash-flow positive in the first year.

Box 4: Alternative Treatments of Depreciation

This box compares the effect of various treatments of depreciation over a 10-year period on the returns earned by an investor in a rental property.

The purchase price of the property is assumed to be $400,000 which can be broken down as follows:

Value of land $200,000
Cost of construction $130,000
Value of fixtures $70,000
Purchase price $400,000

The depreciation rate on buildings is 2½ per cent per year, and the average depreciation rate on fixtures, etc is taken to be 9 per cent per year. Total depreciation expenses are therefore $9,550 annually ($3,250 for buildings and $6,300 for fixtures).

The property is assumed to increase in value at 5 per cent per year, so that its value at sale in 10 years' time is $651,558. Given the purchase price of $400,000, the cost base for purposes of calculating capital gains (after taking account of stamp duty and other expenses, but before depreciation) is assumed to be $430,000. The marginal income tax rate is taken to be 48.5 cents in the dollar, and the capital gains tax is half this rate.

Given these assumptions it is possible to compare three different treatments of depreciation.

The UK treatment: No depreciation deductions are allowed and correspondingly there is no adjustment to the cost base for calculating capital gains tax.

The Canadian treatment: Depreciation deductions are allowed, but when applying the capital gains tax, the lower rate does not apply to that part of the calculated capital gain arising from depreciation.

The Australian treatment: Same as the Canadian treatment, but the standard capital gains tax is applied to entire capital gain.

To illustrate the differences arising from depreciation the following examples abstract from complications arising from the taxation of rental income and restrictions on negative gearing. The treatments are compared using the current Australian marginal tax rates on income and capital gains.

Under the Canadian and Australian treatments, depreciation deductions reduce tax payable by $4,632 per year (0.485*$9,550). The taxpayer is able to invest this money and thus earn interest. Over a 10-year period total interest (after tax) would amount to $5,753 assuming an interest rate of 5 per cent.

When the property is sold, capital gains tax is levied in all three cases, as shown in Table 1. The capital gains tax is lowest under the UK treatment, as there is no adjustment to the cost base for previously claimed depreciation. It is highest under the Canadian treatment due to capital gains being levied at the full income tax rate on part of the capital gain.

The overall position of the investor under each of the three treatments is shown in Table 2.

Under the current Australian treatment, the investor is better off by around $29,000 in 10 years' time relative to the outcome if the UK treatment applied, and around $23,000 relative to the outcome if the Canadian treatment applied. This is a substantial difference. The main reason for this difference is that in Australia investors receive a tax benefit at their marginal tax rate for depreciation deductions, while all of the calculated capital gain is taxed at half the marginal tax rate.

The difference between the Canadian and UK treatments results from the fact that under the Canadian system, investors receive a benefit by being able to claim deductions for depreciation through time while only having to pay capital gains tax at the point of sale (and not through time). Investors in Australia receive the same benefit.


One other assumption is necessary here: that there is a proportion of people who would like to have a better home, or a better-located home, than the one they can currently afford. It would be hard to believe this assumption would not be met. [16]

The estimates are calculated using the average loan size of first-home buyers, and assuming that the purchaser borrows 80 per cent of the value of the property. [17]

For a fuller discussion of this issue, see Stevens GR, ‘Some Observations on Low Inflation and Household Finances’, Reserve Bank of Australia Bulletin, October 1997; see also Reserve Bank of Australia Bulletin, ‘ Household Debt: What the Data Show’, March 2003, pp 1–11. [18]

Moreover, the Canada Mortgage and Housing Corporation (which is the monopoly provider of mortgage insurance for rental housing) has a lower maximum loan-to-valuation ratio (85 per cent) for investors than for owner-occupiers, although if the borrower is willing to be personally liable for the debt, this difference is eliminated. [19]

While loans with loan-to-valuation ratios of up to 85 per cent are available in the United Kingdom, loan-to-valuation ratios are normally in the range of 70–75 per cent, rather than the usual 80–85 per cent range for loans to owner-occupiers. Where loans are based on the rental income, rather than the borrower's overall income, a maximum loan-to-valuation ratio of 65 per cent is normally applied and the gross rental income must exceed 130 per cent of the interest payments. [20

Assuming that the investor has gained approval from the Commissioner of Taxation to reduce the amount of tax withheld from their income in anticipation of a loss from the property investment. [21]

Interestingly, the term ‘negative gearing’ does not appear to be used in other countries examined in this submission. [22]

In the Netherlands, savings and investments are assumed to earn a fixed yield of 4 per cent, on which a 30 per cent income tax is levied. The assumed yield is applied to the net value of the savings/investments (i.e. the value after debt has been deducted). Separate arrangements apply to the taxation of owner-occupied housing. [23]

This threshold was introduced in 1986 and has been unchanged since that time. [24]

Precisely comparable figures are not currently available for other countries. Although in the United States 51 per cent of taxpayers who declared rental income in 2000 reported an overall loss, at least some of these would have had to carry the loss forward rather than write if off against other income. [25]

See Association of Residential Letting Agents, Survey of Residential Landlords, 2003. [26]

See US Census Bureau, Property Owners and Managers Survey, 1995–96. [27]