Policy Forum: Household Debt. Household Indebtedness. Paul Bloxham and Christopher Kent Reserve Bank of Australia

The Australian Economic Review, vol. 42, no. 3, pp. 327 39 Policy Forum: Household Debt Household Indebtedness Paul Bloxham and Christopher Kent Reserve Bank of Australia 1. Introduction A large rise in household indebtedness defined here as the ratio of debt to disposable income over the past few decades has been common to many, though not all, advanced economies (Figure 1). The rise reflects a number of factors, including an easing in credit constraints following financial deregulation and a decline in inflation and, thus, lower nominal and real interest rates. In countries where these factors had been especially strong or operated simultaneously, such as in Australia, the Netherlands and Spain, debt rose noticeably faster than income and there was typically a sharp rise in real house prices. In contrast, in countries such as France and Germany, where the markets for housing credit exhibited less innovation and flexibility and the falls in inflation were less marked, household indebtedness increased only moderately. As a result of the trend decline in interest rates, household interest payments have not risen nearly as significantly as debt (Figure 2). Indeed, the fall in interest rates is a key reason for the rise in debt. More recently, the level of indebtedness has stopped rising in a number of countries, reflecting concerns about growth prospects, weaker labour markets and a tightening in the availability of credit in the wake of the global financial crisis. Indeed, the sharp rise in indebtedness in some countries over much of this decade, combined with a significant decline in lending standards, particularly in the United States, lies at the core of the global financial crisis. This has The views expressed in this article are those of the authors and do not necessarily reflect those of the Reserve Bank of Australia. The authors thank Tapas Strickland for excellent research assistance. focused attention on the possibility that the runup in debt might have gone beyond what could be justified in terms of fundamentals that is, persistent and tangible factors and driven instead by an increased appetite for risk and excessive optimism. By the same token, it might be that any correction could overshoot on the way down, reflecting a greatly reduced appetite for risk and considerable weakness of financial institutions in a number of countries. In the first part of this article, we discuss some of the factors that contributed to the rise in household debt over the past decade or two, focusing on the ways that these factors have differed across countries in the lead up to the global financial crisis. This discussion draws heavily on work in Kent, Ossolinski and Willard (27), to which readers are referred for further details and a more comprehensive review of the literature. The second part of the article considers the implications of higher indebtedness for the potential for households to suffer financial distress and/or to rein in their expenditures. In particular, we argue that indebtedness by itself is not a reliable measure of weakness in this regard, but rather that an assessment of the likelihood of adverse shocks and the ability and willingness of households to service debt is required. In the third part of the article, we examine developments leading up to, and since the onset of, the global financial crisis, focusing on the factors that could influence household vulnerability, and drawing out similarities across countries as well as key differences. 1 2. Household Indebtedness Households demand credit for a number of reasons, but chief among these is the desire Published by Blackwell Publishing Asia Pty Ltd

328 The Australian Economic Review September 29 25 Figure 1 Household Debt 2 15 1 5 1982 1985 1988 1991 1994 1997 2 23 26 29 1982 1985 1988 1991 1994 1997 2 23 26 29 United States United Kingdom France Norway Australia Canada Germany Spain New Zealand Sweden The Netherlands Note: The household sector includes unincorporated enterprises, except for Australia and the United States. Sources: See Kent, Ossolinski and Willard (27) for a detailed description of the data and sources. Figure 2 Household Interest Payments 2 15 1 5 1982 1985 1988 1991 1994 1997 2 23 26 29 1982 1985 1988 1991 1994 1997 2 23 26 29 United States New Zealand France Norway Germany Australia United Kingdom Sweden The Netherlands Notes: Ratio of interest payments on total debt to household disposable income. The treatment of financial intermediation services varies across countries. Sources: See Kent, Ossolinski and Willard (27) for a detailed description of the data and sources. to purchase residential property, to obtain the services of a home and accumulate wealth (in some cases, this is to take advantage of the beneficial tax treatment afforded to leveraged property purchases, but it is also a device to commit to a savings plan). Another motive for taking on debt is to smooth consumption over the life cycle, borrowing early in life when

Bloxham and Kent: Household Indebtedness 329 income is relatively low and then repaying gradually and building net assets ahead of retirement. Consumption smoothing in the face of temporary adverse shocks might also lead some households to incur debt. The supply of credit to households depends on the ability and willingness of financial institutions and markets to extend credit, which is influenced by regulatory controls, an assessment of the risks associated with lending, and the competitiveness of credit markets. Constraints on credit exist, in part, because lenders are unable to determine precisely the ability and willingness of borrowers to repay their debts. Still, they do what they can by examining information about the credit histories and earnings capacity of potential borrowers, often combining this with simple models, or rules of thumb, to decide who to lend to, how much and on what terms; that is, to determine lending standards. Various structural changes (such as disinflation) can lead to a change in credit constraints for given rules of thumb (Stevens 1997). 2 In addition, financial institutions may modify their rules of thumb in response to structural changes that affect the risks associated with lending to particular households and/or the ability to identify or control these risks. 2.1 Factors Driving Rising Indebtedness A number of important developments over the past couple of decades have been cited as possible explanations for the trend rise in the household debt-to-income ratio in many advanced economies. 2.1.1 Financial Sector Deregulation, Competition and Innovation The deregulation of financial markets took a number of forms, but tended to ease restrictions on new entrants to compete in mortgage markets and removed interest-rate controls and lending guidelines (Debelle 24). This allowed for an increase in competitive pressures, leading to an easing in lending standards and a lowering of lending margins (although the effect on demand of changing margins on average, across loans to households and businesses might have been offset by monetary policy adjustments and, therefore, might not have been an independent driver of the cost of credit). Deregulation and competition can encourage innovation, including the types of loans available, such as products facilitating housing equity withdrawal. 2.1.2 Declining Inflation A common rule of thumb is to lend an amount, such that the initial repayments are no more than some fixed share of the borrowers income. 3 Lower inflation reduced nominal interest rates, easing this constraint by reducing initial repayments. 2.1.3 Declining Costs Declines in real interest rates (in part, following central bank efforts to fight high levels of inflation) reduced the real cost of funds for financial institutions, while technological innovation reduced the cost of information and administration associated with lending (CGFS 26). 2.1.4 Reduction in Macroeconomic Volatility Lower interest-rate volatility reduced the perceived risk of default for variable-rate loans (of a given amount) and decreased funding costs for institutions providing fixed-rate loans. Less volatile aggregate economic activity (at least prior to the global financial crisis) reduced the extent of non-performing loans, although it did not necessarily imply less volatility for individual incomes. 2.1.5 Lower Unemployment Rates Earlier falls in unemployment rates and high employment growth implied less frequent and shorter unemployment spells and multiplied the sources of earned income for households, increasing both credit demand and supply. Also, a rise in expected income growth may have increased the demand for debt and eased credit constraints.

33 The Australian Economic Review September 29 2.1.6 Changes in Taxes and Subsidies Changes in taxes and subsidies in a number of countries affected the demand for mortgages by owners and investors. 2.1.7 Ageing of the Population Increases in longevity may have led households to hold debt for longer and allowed financial institutions to lend to older borrowers, although declines in fertility may have reduced aggregate indebtedness, as older households typically hold less debt. 2.1.8 Summary Although all of these factors are likely to have played some role in the trend rise in household debt across a range of countries, a number of issues complicate the task of ascribing degrees of significance to each factor. First, it is difficult to estimate individual effects with accuracy, given the complex interactions across factors. Second, changes in expectations and appetites for risk play a role in amplifying the response of borrowers and lenders to what one might think of as fundamental driving forces. Amplification can occur because of an interplay between economic activity, asset prices (that form the basis of collateral) and credit. Third, the response to fundamental forces can play out over a considerable period of time. For example, younger generations might be more responsive to changing circumstances than older generations, who largely have paid down existing debts, and may take on more (or less) debt than older generations for given credit conditions. 2.2 A Brief Cross-Country Comparison Despite considerable variation, all countries in the sample (the 18 for which data are readily available, as reported in Kent, Ossolinski and Willard 27) experienced an increase in indebtedness between the early 199s and 27 (selected countries are shown in Figures 1 and 2). The timing and extent of this varies, with some countries (such as Norway, Sweden and the United Kingdom) standing out with earlier periods of sharp or protracted declines associated with widespread financial distress. Across all countries, there was a fall in real mortgage rates, with indebtedness increasing more for those countries experiencing larger declines in mortgage rates. As a result, the rise in indebtedness, as measured by the debt-to-income ratio, has been larger than the rise (if any) in the ratio of interest payments to income. To provide further insights, Kent, Ossolinski and Willard (27) exploited the variation in the behaviour of the debt-to-income ratio across countries to explore whether some of the factors set out above might explain rising indebtedness. For the period up to 27, they found: a clear positive correlation between the fall in inflation and the increase in the debt-toincome ratio; a positive correlation between lower volatility (of output and mortgage rates) and indebtedness, although the effect of this on the demand for credit appears to have been relatively moderate, perhaps because the volatility of individual household income has declined by less than that of the gross domestic product (GDP); a close correlation between trend falls in the unemployment rate and trend increases in the debt-to-income ratios across countries; much of the variation around these trends seems to have been related to differences in the extent of deregulation, increased competition and product innovation across countries. Where these forces were stronger, such as in Australia, the Netherlands, the United Kingdom and the United States, indebtedness rose by more than where these forces were less prominent (such as in Italy, France and Japan); a simple model with credit-constrained young households and simple rules of thumb for financing and repayment behaviour implies that much of the rise in indebtedness in Australia might have been related to

Bloxham and Kent: Household Indebtedness 331 lower inflation and (to a lesser extent) real mortgage interest rates. Lower credit constraints, falling unemployment and increased longevity also might have been important, but are difficult to capture in such a model; and country-specific factors, such as tax laws and geographical and cultural conditions, were also important. For example, the Netherlands experienced the largest increase in the debtto-income ratio, but only average changes in many of the explanatory factors considered above. The key factors here seem to be extensive credit market deregulation combined with a tax system that encourages delayed repayment of principal. The Nordic countries also stand out, with sharp increases in debt earlier than in most countries, followed by a sharp correction in the early 199s. The rapid deregulation of credit markets appears to have triggered their credit and asset-price booms. Following widespread financial distress, it took over a decade for household debt-to-income ratios to return to their peak levels of the late 198s in these countries. Given the central role of housing in motivating and facilitating household borrowing, it is not surprising that the countries with the largest rise in debt-to-income ratios up to 27 also experienced a substantial rise in real house prices (Figure 3). Much of this is likely to reflect an easing of credit constraints following a period of financial repression, such as tight regulations and prescriptions regarding lending activities. The removal of these constraints enabled increased access to credit that, in principle, should be welfare-enhancing. But, with greater access to finance, households competed for a limited stock of desirable (well located and high quality) property, which drove prices up. On the demand side, this partly reflects the tendency for the demand for quality housing to rise by more than income. In combination with a limited supply of well-located property, related to geographical and institutional features, this may explain much of the rise in house prices (Glaeser, Gyourko and Saiz 28). In Australia, this combination of demand and supply features appears to have manifested in a larger rise in the prices of properties that are proximate to the centre of cities and/or the waterfront over the past decade and a half (Richards 28). 3. Debt and Distress Policy-makers charged with maintaining financial system stability obviously want to Figure 3 House Price Growth and Debt-to-Income Ratios (change since pick-up in debt-to-income ratio) Change in debt-to-income ratio (annual average, ) 1 8 6 4 2 Denmark The Netherlands Australia Spain New Zealand Korea United Kingdom Italy Canada United States Japan France Switzerland Germany Sweden Norway Finland Belgium -4-3 -2-1 1 2 3 4 5 6 7 8 9 1 Change in the average real house price (annual average, ) Note: See the source below for details regarding the time frames. Source: Kent, Ossolinski and Willard (27).

332 The Australian Economic Review September 29 minimise the probability of widespread financial distress of households, corporations and/or financial institutions. At the very least, such distress could lead to a curtailment of household consumption, with adverse consequences for the macroeconomy. In cases of more acute and widespread distress, a large share of households might find themselves unable (or unwilling) to service their debts, leading to the possibility of forced sales of assets and significant losses for lending institutions, both of which could act to trigger and/or exacerbate a macroeconomic downturn. Other things being equal, a higher level of indebtedness implies that households (and their creditors) will be more vulnerable to adverse shocks. However, other things are never equal, not within a country over time or across countries at a particular time. Indeed, many of the developments outlined in Section 2 would have reduced risk at unchanged levels of debt. Not surprisingly then, households took on more debt and financial institutions have eased credit constraints. If borrowers and lenders have good information about the risks they face, debt is likely to adjust in a way that maintains an optimal degree of risk (at least from a private perspective). So, what would give rise to concerns regarding rising indebtedness? Foremost is the potential for households and financial institutions to misjudge the actual risks they face. They might give too much weight to recent experience, leading them to underestimate the risks during benign periods and to overstate the risks following adverse shocks. This can contribute to financial system developments that amplify the business cycle, as the interplay between borrowers and lenders affects the availability of credit, the price of collateral and the strength of economic activity (Aoki, Proudman and Vlieghe 24; Iacoviello 25). For example, following a fall in house prices, credit constraints could tighten in response to a higher probability of default and greater losses given default, weakening activity and leading to further falls in house prices. Working in the other direction, following a long period of stable growth, people might view the economic situation as being more benign than it really is, which could encourage excessive indebtedness, leaving households more vulnerable to an adverse shock. Although judging the extent of excessive optimism during an expansionary period can be difficult in real time, empirical evidence suggests that the following characteristics imply an increase in the probability of financial distress: financial deregulation, where financial institutions, households and regulators are still learning about a new regime (or new product innovations) (CGFS 26); significant growth in the prices of assets, which form the basis of collateral, combined with rapid credit growth (Borio and Lowe 22, 24); and vigorous competition and innovation in financial markets. Another consideration is the role of investors in the housing market. Consistent with rising asset prices and the availability of cheaper financing, investors might enter the housing market seeking capital gains. These investors can be households or corporations, with the investment choice often influenced by a country s tax arrangements: leveraged property purchases receive beneficial tax treatment in many countries. The risk of large falls in prices could be exacerbated by speculative development of property that goes well beyond the needs suggested by longer-term demand from newly formed households. 4. Developments Since the Onset of the Global Financial Crisis After a long period of rising household indebtedness across many developed economies, the global financial crisis has precipitated a decline in indebtedness in a number of countries. Similarly, housing markets have been very weak in a number of countries after earlier large price gains. Another important, and related, development has been the generalised decline in the appetite for risk globally.

Bloxham and Kent: Household Indebtedness 333 4.1 Developments in the United States The factors affecting vulnerability during the rise in the indebtedness of US households in recent years are worth considering in some detail. Signs of rising vulnerability were evident ahead of the financial crisis, though since then, further concerns have come to light that were perhaps not widely appreciated earlier. 4.1.1 United States Pre-August 27: Structural Weaknesses and Emerging Vulnerabilities Growth in household debt was rapid from 22 27, helped by substantial reductions in lending standards. This growth was evident in increases in home ownership rates which had been trending up for some years concentrated among lower-income households using nonconforming loans, of which sub-prime loans are a subset (Figure 4). Although the originate-to-distribute model (whereby loans originated by brokers are packaged up (securitised) and are sold to third parties) had been in existence for decades, its application became more widespread in the early part of this decade. The key aspects of the decline in lending standards were the reduced requirement (if any) for borrowers to contribute their own funds loan-to-valuation ratios (LVRs) of 1 per cent or more were common and lending to borrowers who previously might not have qualified because of low and/or variable incomes or poor credit histories. By late 27, the average LVR across sub-prime borrowers was 85 per cent, compared with 7 per cent for prime borrowers. Also, a rising share of loans were adjustable-rate mortgages, with low initial teaser rates that would reset after a few years according to a variable market interest rate, and sometimes by as much as 6 percentage points (FDIC 27); this stands in contrast to more traditional loans with fixed rates for the full term. Figure 4 Housing Market Indicators: United States 71 Home ownership rate Dwelling investment 8 68 6 65 4 Housing vacancy rate House prices 3 Median house price 2 2 Case-Shiller, 2 cities 8 Real interest rates 3 year mortgage Housing loan arrears 4 2 Federal funds -8 198219851988 1991199419972 2326291982 1985198819911994 199722326 29 Notes: Dwelling investment is presented as a share of nominal GDP; the housing vacancy rate is the number of homes that are vacant and for sale as a share of the number of homes that are either owner-occupied, rented or vacant and for sale; house prices are presented in terms of year-ended growth rates; real interest rates are deflated by the core Consumer Price Index; housing loan arrears are presented as a share of the total housing loans and include on-balance-sheet loans that are 9 days past due, as well as those that are impaired. Sources: Datastream; FDIC; RBA; Thomson Reuters; US Census.

334 The Australian Economic Review September 29 These developments made some sense (at least to individual borrowers) in an environment of rapidly rising house prices. The idea was that borrowers would establish a good credit history and build some equity, allowing them to refinance the loan on better terms before the teaser rate expired. A group of speculative investors, referred to as property flippers, also entered the housing market for the purpose of buying and selling property in a short period of time at a profit. All in all, the possibility that house prices might actually fall across large regions, if not the whole country, appears to have been largely ignored, including by many analysing the risks associated with securities that were tied to the performance of these mortgages (Gerardi et al. 28). The resulting increase in housing credit and house prices was also accompanied by a boom in housing investment from 22. Dwelling investment rose by almost 2 percentage points of GDP over a period of 5 years, to be at a historical peak, and the stock of dwellings available for sale equalled 1 months of normal supply, up from an average of 5 months. Although problems in the US mortgage market were sparked by non-conforming loans, based on adjustable loan rates, the dominance of borrowing at fixed rates means that any subsequent falls in borrowing rates are not automatically passed through to many mortgagors. Although many fixed-rate borrowers are able to take advantage of lower rates by refinancing their loans, this is difficult for those borrowers with a reduced capacity to service their debt or with much reduced equity in their home (reflecting the general economic weakness and decline in house prices). Moreover, the use of non-recourse loans in the United States makes it easier for many of those with negative equity to drive away from their homes and default on their debt in such circumstances. 4 4.1.2 United States Post-August 27: The Realisation of Losses Problems associated with the housing market started to become widely acknowledged around the middle of 27, following sharp rises in the rates of default on sub-prime and other higher-risk loans that by then accounted for a sizeable share of outstanding mortgages. However, it took some time before the extent to which these losses would affect the financial systems and the economies of the world became fully appreciated. For the United States, defaults have continued to rise, house prices nationally have fallen considerably, by around 25 3 per cent, and the market prices of securities based on sub-prime mortgages have fallen precipitously. The fall in house prices has been exacerbated by the large stock of unsold homes an echo of the earlier boom in housing investment which also has led housing construction to decline to unprecedented lows. Banks and other financial institutions have suffered sizeable losses, largely reflecting their exposure to sub-prime and other related securities, leading to some failures, with many others needing public support. The market for securitisations a key source of finance in the United States has been impaired severely, restricting the supply of credit. These significant disruptions to financial markets have meant that cuts to the overnight interest rate by the Federal Reserve, though substantial, have had relatively little effect on the rates faced by households; a reliance on fixed-rate mortgages has also played a role here (as discussed above). More recently, the Federal Reserve has indicated that exceptionally low levels of the federal funds rate are likely to be warranted for an extended period. In addition, the Federal Reserve has taken other steps to ease credit conditions, including...making substantial purchases of longer-term securities in order to support market functioning and reduce interest rates in the mortgage and private credit markets (Kohn 29). These efforts are in addition to the stimulus provided by fiscal policy. The Congressional Budget Office is forecasting the Federal budget deficit to be 13.1 per cent of GDP in 29 1 and 9.6 per cent of GDP in 21 11, although a sizeable part of the 29 1 deficit is associated with asset purchases. 4.2 Developments in Australia Australia s financial markets and economy more generally have also been adversely

Bloxham and Kent: Household Indebtedness 335 Figure 5 Housing Market Indicators: Australia 76 Home ownership rate Dwelling investment 8 72 6 68 4 Rental vacancy rate House prices 6 2 3 Real interest rates Housing loan arrears 8 Outstanding housing rate 4 2-8 Cash rate 1982 1985 1988 19911994 1997 2 23 2629 19821985 1988 1991 1994 1997 223 26 29 Notes: Dwelling investment is presented as a share of nominal GDP; the rental vacancy rate is the number of homes for rent as a share of the number of homes that are offered for rent by real estate agents; house prices are presented in terms of year-ended growth rates; real interest rates are deflated by the trimmed mean Consumer Price Index; housing loan arrears are presented as a share of the total housing loans and include on-balance-sheet loans that are 9 days past due, as well as those that are impaired. Sources: ABS;RBA;REIA. affected by the global financial crisis. Like a number of other countries, the ratio of debt to income is high in Australia and house prices appear to be high relative to income. At the same time, however, there are a number of features of the Australian housing market, financial system and macroeconomy that suggest that the imbalances evident in some other countries are not as apparent here. First, there is little evidence of over-building in Australia overall. New dwelling investment has been fairly stable as a share of GDP over the past 3 years and well below the peak levels reached in previous housing construction cycles (Figure 5). In addition, the rate of home ownership has been stable, at around 7 per cent for two decades or so, in stark contrast to the US case. Rental vacancy rates also have been at low levels and rents have grown strongly for a few years. These features would suggest less downward pressure on house prices than otherwise. Second, although some decline in lending standards was apparent in earlier years, driven by an increase in competition (largely from non-bank financial institutions), these changes do not compare in size or scope with developments in the United States. In particular, the use of the originate-to-distribute model has been more restrained in Australia, especially among banks, and much of the additional credit was extended to higher-income households, which typically provide more equity upfront and/or are better able to service their loans. Lending standards also appear to have been maintained at more stringent levels for authorised deposittaking institutions (ADIs) and Australian banks have had virtually no direct exposure to US sub-prime assets, unlike many banks in Europe, in particular. The regulator of these institutions, the Australian Prudential Regulation Authority (APRA), points to a number of explanations for the relatively prudent stance of

336 The Australian Economic Review September 29 Australian ADIs, including: tight requirements of APRA regarding credit assessments by thirdparty originators; higher capital charges for low-doc loans introduced in 24; a generally conservative stance on capital adequacy by APRA relative to many of its overseas counterparts; and the dominance of large financial institutions (relative to the economy). 5 Third, although demand by investors played an increasingly significant role in the growth of household debt from the late 199s, this was checked around late 23. Investor demand appeared to be driven by expectations of significant capital gains, increasingly easier access to investor finance and the tax treatment of investments in residential property (RBA 23b). The significant role of households as investors in rental properties in Australia especially compared to the United States, where corporations own a larger share of the residential rental stock contributed to the level of household indebtedness. The rising importance of household investors came to a halt around late 23, helped by rising interest rates and changes to tax arrangements, including closer scrutiny of investor deductions (Figure 6). This was associated with a significant easing in house price growth in the larger cities (and an overall decline in Sydney), with subsequent house price growth of around one-third of the pace of the previous 5 year period and well below the growth rate of household disposable income (Richards 29). Fourth, the modest decline in house prices over the past year or so mostly has been concentrated in higher-priced suburbs, consistent with the direct effect of the global financial crisis on wealthier households, including via falling stock prices. This decline could also reflect the fact that such households account for much of the household debt in Australia and they may be deleveraging. However, because these households typically have substantial equity in their houses, these developments are not expected to have a significant effect on the overall rate of mortgage delinquencies. Fifth, in contrast to the experience of many countries, the Australian banking system has performed well of late: profitability, though lower than in recent years, remains solid, banks are well capitalised and the larger banks have high credit ratings (RBA 29). As noted above, lending standards did not decline as much as in other countries around the middle of this decade and, although loan losses have risen, this is from the especially low levels of recent years. In addition, the use of full-recourse mortgages provides added incentive for households to continue to make mortgage repayments, even if a decline in the price of their Figure 6 Household Debt: Australia 2 15 1 Total Housing owner-occupier Housing investor Other personal 5 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 21 23 25 27 29 Notes: The household sector excludes unincorporated enterprises. Disposable income is after tax and before the deduction of interest payments. Sources: ABS;RBA.

Bloxham and Kent: Household Indebtedness 337 house pushes them into a position of negative equity. Indeed, despite a significant increase in the unemployment rate during the 199s recession, Australian house prices did not fall significantly, in part because households continued to service their mortgages rather than sell their houses. The lack of tax deductibility of mortgage interest payments on owneroccupied homes also encourages borrowers to repay debt ahead of contractual obligations. This means that variable-rate mortgages are the norm (even fixed-rate loans tend to be fixed for only a few years) and implies that many Australian borrowers have access to automatic liquidity, which is not subject to market risk. Finally, the relative health of the banking system and the reliance on variable-rate mortgages means that the easing of monetary policy (of 425 basis points since September 28, as at the time this article was finalised) has been passed on rapidly and in large part to mortgagors. In addition to monetary policy, fiscal policy has also played an expansionary role, though levels of net public debt are projected to remain low by international standards. Notwithstanding these strengths, house prices appear to be high relative to incomes, when compared to other countries, and a rising share of recent borrowing is accounted for by first-home buyers who are encouraged, in part, by government grants. However, any concerns regarding rising indebtedness among marginal borrowers needs to be kept in perspective. Since the increase in the first-home owner s grant in October 28, around 3 additional firsthome buyers have bought homes, accounting for only about.4 per cent of the number of households (as at the time that this article was finalised). A number of banks have also tightened the lending standards for these buyers, including a reduction in maximum LVRs. 5. Conclusions Most advanced economies experienced a substantial rise in household indebtedness over the past two decades or so. This rise tended to be greater for countries that had larger declines in inflation, macroeconomic volatility and unemployment. There is also evidence that indebtedness rose by more in those countries with more competitive and innovative mortgage markets. Though it is not surprising that these factors would have contributed to higher indebtedness, by themselves they do not imply that households are necessarily more vulnerable to adverse shocks. Nonetheless, for a number of countries, some of the rise in indebtedness, particularly over more recent years, appears to have reflected an overly optimistic outlook on the part of both borrowers and lenders and a significant decline in lending standards. This is particularly true in the case of the United States, where the extension of mortgages to many lower-income households, which would not have qualified under the earlier, more-stringent lending standards, led to a sharp rise in the rate of home ownership in the space of just a few years. High and rising rates of mortgage defaults and declines in house prices, apparent even before a generalised downturn in the economy, led to large losses and significant stress in financial institutions. The financial crisis, by pushing up risk margins and limiting the availability of credit, has weakened the pass-through of large cuts to policy rates to indebted households. Although Australia had some parallels with the United States going into the crisis notably, rising indebtedness and house prices over a number of years there are a number of important differences. Specifically, there is little evidence of over-building of housing overall, and the increasing prominence of investors in the market was checked a few years ago, with a slowing in house price growth and even declines in some parts of the country. The banking system in Australia has shown considerable resilience, maintaining strong profitability and high credit ratings in part, reflecting tighter lending standards than in the United States and a lack of direct exposure to US sub-prime assets. This has meant that, for most borrowers, cuts to monetary policy have been passed through in large part, given the dominance of flexible-rate mortgages. These mortgages tend to be paid off ahead of schedule which, combined with full-recourse loans, has contributed to low rates of home loan defaults in

338 The Australian Economic Review September 29 Australia. Nevertheless, arrears have been rising and are likely to do so as unemployment rises, and there has been some tightening in lending standards. By themselves, these developments will tend to exert downward pressure on house prices. The Australian housing market has been through a period of adjustment and further declines in house prices and indebtedness are possible. However, much of the increase in the ratio of debt to income over the past two decades can be attributed to sustainable structural factors, including an easing in credit constraints following financial deregulation and a decline in inflation and, thus, lower nominal and real interest rates. May 29 Endnotes 1. For a broader review of the literature and events associated with the global financial crisis, see Borio (28), Blundell-Wignall and Atkinson (28), Ellis (28) and the references therein. 2. Rules of thumb often focus on limiting the size of the initial repayments (as a share of income). This makes sense because defaults are more likely for relatively new loans, which have higher loan-to-valuation ratios and where households have less experience in meeting the required repayments. 3. In Australia, this was typically around 3 per cent of gross income (RBA 23a). This restriction had been relaxed earlier this decade (Laker 27), though it appears to have been tightened somewhat of late. 4. In cases where the initial LVRs are close to 1 per cent (or more), households have little, if any, skin in the game ; hence, they have little incentive to take a proper account of the risk of mortgage default when entering mortgage agreements. 5. See Lewis (29) for a discussion of these points. References Aoki, K., Proudman, J. and Vlieghe, G. 24, House prices, consumption, and monetary policy: A financial accelerator approach, Journal of Financial Intermediation, vol. 13, pp. 414 35. Blundell-Wignall, A. and Atkinson, P. 28, The sub-prime crisis: Causal distortions and regulatory reform, in Lessons from the Financial Turmoil of 27 and 28, Proceedings of a Conference, eds P. Bloxham and C. Kent, Reserve Bank of Australia, Sydney. Borio, C. 28, The financial turmoil of 27?: A preliminary assessment and some policy considerations, Bank for International Settlements Working Paper no. 251, Basel, Switzerland. Borio, C. and Lowe, P. 22, Asset prices, financial and monetary stability: Exploring the nexus, Bank for International Settlements Working Paper no. 114, Basel, Switzerland. Borio, C. and Lowe, P. 24, Securing sustainable price stability: Should credit come back from the wilderness?, Bank for International Settlements Working Paper no. 157, Basel, Switzerland. Committee on the Global Financial System 26, Housing finance in the global financial market, CGFS Paper no. 26, Bank for International Settlements, Basel, Switzerland. Debelle, G. 24, Macroeconomic implications of rising household debt, Bank for International Settlements Working Paper no. 153, Basel, Switzerland. Ellis, L. 28, The housing meltdown: Why did it happen in the United States, Bank for International Settlements Working Paper no. 259, Basel, Switzerland. Federal Deposit Insurance Corporation 27, Statement on subprime lending, Federal Register, vol. 72, pp. 37 569 75. Gerardi, K., Lehnert, A., Sherlund, S. and Willen, P. 28, Making sense of the sub-prime crisis, Brookings Papers on Economic Activity Conference Draft, viewed June 29, <http://www.brookings.edu/ economics/ bpea/ / media/ Files/Programs/ ES/BPEA/28 fall bpea papers/28 fall bpea gerardi sherlund lehnert willen.pdf>. Glaeser, E., Gyourko, J. and Saiz, A. 28, Housing supply and housing bubbles, National Bureau of Economic Research Working Paper no. 14193, Cambridge, Massachusetts.

Bloxham and Kent: Household Indebtedness 339 Iacoviello, M. 25, House prices, borrowing constraints, and monetary policy in the business cycle, American Economic Review, vol. 95, pp. 739 64. Kent, C., Ossolinski, C. and Willard, L. 27, The rise of household indebtedness, in The Structure and Resilience of the Financial System, Proceedings of a Conference, eds C. Kent and J. Lawson, Reserve Bank of Australia, Sydney. Kohn, D. 29, Policies to bring us out of the financial crisis and recession, speech to Forum on Great Decisions in the Economic Crisis, College of Wooster, Wooster, Ohio, 3 April, viewed June 29, <http://www. federalreserve.gov/newsevents/ speech/kohn 2943a.htm>. Laker, J. 27, Credit standards in housing lending Some further insights, speech to Institute of Chartered Accountants in Australia, Melbourne, 2 June. Lewis, D. 29, Surviving the downturn: APRA s role in financial crisis management, speech to Business Continuity Summit, Brisbane, 25 March. Reserve Bank of Australia 23a, Household debt: What the data show, Reserve Bank of Australia Bulletin, March, pp. 1 11. Reserve Bank of Australia 23b, Submission to the Productivity Commission Inquiry into First Home Ownership, Occasional Paper no. 16, Sydney. Reserve Bank of Australia 29, Financial Stability Review, March, Reserve Bank of Australia, Sydney, viewed June 29, <http:// www.rba.gov.au/publicationsandresearch/ FinancialStabilityReview/Mar29/Pdf/ financial stability review 39.pdf>. Richards, A. 28, Some observations on the cost of housing in Australia, speech to Economic and Social Outlook Conference, Melbourne, 27 March. Richards, A. 29, Conditions and prospects in the housing sector, speech to Fourth Annual Housing Congress, Sydney, 26 March. Stevens, G. 1997, Some observations on low inflation and household finances, Reserve Bank of Australia Bulletin, October, pp. 38 47.