RDP 2006-08: A Survey of Housing Equity Withdrawal and Injection in Australia 2. Concepts and Literature Review

Housing equity withdrawal and injection refer to the net cash flow by households from transactions in housing-secured debt and housing assets. Withdrawals and injections can occur in many different ways. One way for a household to withdraw housing equity is to increase the level of debt secured against a property they already own through methods such as refinancing and increasing the size of the loan, or drawing down a home-equity style loan. Another is by reducing property holdings (for example, by downsizing).[3] Households can inject equity into a property they already own by paying down housing debt or undertaking renovations financed, at least partly, from their own funds. Households increasing their property holdings often also inject equity through a deposit.

There are many factors potentially underlying a household's flow of housing equity, including their preferences regarding:

  • consumption and saving, such as a desire to smooth consumption over a lifetime or in response to temporary changes in income;
  • financial management, such as asset diversification (by using accumulated housing equity to purchase other non-housing assets), replacing higher interest-rate personal debt with housing-secured debt, or using surplus funds to either pay down housing debt or invest in property; and
  • living arrangements, often associated with their stage of life (for example, an elderly household selling a long-held owner-occupied property to move into a retirement home is likely to withdraw equity, while a first-home buyer will typically inject equity).

The trend of increased housing equity withdrawal evident in a number of countries over the past decade or so has prompted a number of surveys to help better understand this development. Many of these had a narrow focus on housing equity withdrawal not related to property transactions. Canner, Dynan and Passmore (2002) from the United States Federal Reserve looked at refinancing behaviour of US households for the period January to June 2002. A similar survey of Dutch homeowners that had taken out at least one mortgage between 1995 and 2000 was commissioned by the Dutch central bank (de Nederlandsche Bank 2000) and repeated in 2003 (van Els, van den End and van Rooij 2005).

In late 2000, the Bank of England commissioned a more comprehensive survey of UK households (Davey and Earley 2001). In addition to withdrawals through mortgage refinancing, the survey covered equity flows resulting from some property transactions, including equity injections. The survey covered mortgage holders that had moved house, refinanced, or taken out a further advance or a second mortgage during the previous two years. However, by surveying only mortgage holders, this survey was not able to identify equity flows by those moving into a debt-free property or withdrawals by last-time sellers (that is, those selling their entire residential property portfolio).

To better capture these flows, an additional module was added to the Survey of English Housing in 2003, with the results summarised in Benito and Power (2004). Even so, the last-time sales category still excluded what they considered to be the most significant component – the sale of properties resulting from the death of an owner. Consequently, the authors scaled up the recorded data on last-time sales by a factor of five. This approach was also followed by Smith and Vass (2004), who noted that these data should be treated with caution.

While differences in the samples and timeframes over which these surveys were conducted make it difficult to compare the various studies, some common themes emerge. First, refinancing appears to be a relatively common phenomenon, with between one-fifth and one-half of households with mortgages found to have refinanced during the survey periods.[4] Second, households refinancing their mortgage often increased the size of their loan at that time, with estimates of the share doing so typically ranging from one-third to two-thirds. The share is likely to vary with economic and financial conditions, as a greater number of US refinancers increased their loan in 2002, following strong house prices gains, than did so in 1999 (Canner et al 2002). Third, the UK surveys found that households moving house were relatively less likely to access additional funds than refinancers, but accounted for a much larger share of the value accessed.

Information on equity injection was only available from the UK survey discussed in Davey and Earley (2001), and covered equity injection associated with refinancing and property transactions only. Among UK mortgage holders, 18 per cent said they injected equity during the previous two years, made up of 39 per cent of movers and 13 per cent of refinancers. For the majority (55 per cent) of the households that injected equity, the additional funds came from their own savings. In addition, gifts or loans from relatives or friends were more commonly mentioned as a source of funds by movers compared with refinancers, most likely reflecting some assistance for first-time buyers.

These surveys also provide some information on the characteristics of those households accessing funds as a result of the above activities. According to results in Canner et al (2002) and Davey and Earley, households were more likely to access additional funds when refinancing or moving if they: had low loan-to-valuation ratios (LVRs); viewed that it was a good time to use credit; or were households that tended not to pay off their credit card balances. Conversely, homeowners that believed that they had a higher chance of losing their jobs were less likely to borrow additional money when refinancing. Factors such as age, education and income were not found to be important determinants. The choice of borrowing against the value of their house as opposed to obtaining funds by other methods also appeared to be importantly influenced by the relative ease of accessing housing equity, with households that had experienced capital gains more likely to access housing equity. Smith and Vass (2004) concluded that households with at least one flexible feature in their mortgage were more likely to withdraw equity, but the authors do not provide details of the ‘limited analysis’ undertaken.

On the issue of the use of housing equity withdrawn, these surveys indicate that the most common was spending on renovations, hence not strictly constituting a net equity withdrawal. While some of the funds were also spent on consumption items, significant amounts were used to repay other debts or purchase financial assets. Both the 2000 and 2003 UK surveys show that households that accessed funds by selling property were less likely to spend the funds than non-transactors, opting instead to either pay down debt or acquire financial assets. Nevertheless, most of the funds accessed by non-transactor households retaining their existing property were used to finance renovations, with only a small proportion consumed.

In Holmans (2001), time-series estimates of the various gross housing equity flows in the UK are presented. His work highlights the importance of property transactions, particularly of last-time sellers, and that withdrawals by last-time sellers are strongly correlated with both house prices and property turnover – a result supported by Greenspan and Kennedy (2005) for the US. Similarly, according to Holmans, injections as a result of property transactions – predominantly deposits by first-time purchasers – account for around half of total gross injections (excluding those associated with renovations).

The common finding to emerge from all of these papers is that households refinancing and increasing the size of their loan used a large share of these funds for renovations, with a smaller share used for consumption. For the limited number of papers that address different methods of withdrawal, the use of funds tended to vary with the method, with equity released via property transactions – which typically accounted for the bulk of gross housing equity withdrawal – less likely to be used on consumption. Information on equity injections is more limited, but injections associated with property transactions appear to be significant relative to overall flows.


The household sector as a whole typically does not withdraw equity in this way since it implies sales to other sectors of the economy or non-residents. [3]

In countries like the US, where the bulk of mortgages are at long-term fixed rates, refinancing is very common when interest rates are falling. [4]