Box D: Gross versus net housing lending

This page contains information on gross versus net housing lending from the May 2015 Financial Stability Report.

Each year, banks undertake a significant amount of new mortgage lending, but the total stock of mortgage loans grows by a much lower rate. Over the past year there was more than $60 billion of new mortgage commitments, amounting to around 30 percent of existing mortgage debt. However, net housing debt grew by only around 5 percent. This gap between the value of loans being written and net credit growth relates to repayments by borrowers. There are a variety of different sorts of debt repayment,1 such as:

  • scheduled principal repayments (a borrower repaying principal as required by the mortgage contract);
  • unscheduled partial principal repayments (a borrower choosing to pay the mortgage off more quickly than required); and
  • full repayments (which would typically occur because the borrower has sold the property or switched the mortgage to a different provider).

In order to understand the drivers of net credit growth, it is useful to be able to understand trends in each of these components. For example, the Reserve Bank has noted that when net credit growth fell dramatically around the GFC, gross originations fell much less. This could have been because many borrowers were accelerating principal repayments, taking advantage of the cash flow benefits of low interest rates. Alternatively, it may have been because people selling houses after the GFC were more likely to have large mortgages. The reality was probably a combination of the two, but it was not easy to quantify at that time.

More recently the Reserve Bank has begun collecting new data that provides fresh insights into the relationship between gross and net housing lending. The survey can be used to decompose the overall change in the aggregate stock of mortgage credit into new lending and the different categories of debt repayment discussed above (figure D1). The survey suggests a significant amount of principal has been repaid over the past year, with partial repayments, less interest, amounting to 9 percent of the stock of outstanding debt. These partial repayments are similar in overall size to full principal repayments. Non-scheduled repayments account for approximately half of partial repayments, suggesting a high degree of voluntary excess repayments by borrowers at present. The substantial rate of partial repayments suggests that changes in household savings behaviour and lower interest rates after the GFC are significantly reducing net household credit growth.

Figure D1: New housing credit by category – registered banks (6 months ended December 2014)

Figure D1 New housing credit by category – registered banks (6 months ended December 2014)

Source: RBNZ New Residential Mortgage Commitments Survey.

Note: ‘Other’ includes increases in balances on repayment deficiencies, together with changes in balances relating to the net write-off of loans. ‘Interest charged’ includes all contractual mortgage interest payments. ‘Gross credit’ refers to new drawdowns of mortgage commitments, and ‘Net credit’ refers to gross credit less borrower repayments.

Gross housing credit (new commitments) increased strongly at the beginning of 2015 in line with an increasing value of house sales, after a brief drop-off in momentum post-LVR restrictions (figure D2). The longer running Reserve Bank Housing Approval Survey suggests that gross mortgage originations, when scaled by GDP, are now running at around two-thirds of the pre-GFC peak. However, recent data suggest there has been a substantial increase in the rate at which approvals translate into new lending, and there is significant uncertainty about how this ratio has evolved over time. In contrast to gross lending, net housing credit growth is comparatively modest, at less than half of its pre-GFC peak. As noted above, high principal repayment rates are likely the key driver for this divergence.

Figure D2: Mortgage approvals, housing credit and house sales (3-monthly seasonally adjusted total, % of quarterly GDP)

Figure D2 Mortgage approvals, housing credit and house sales (3-monthly seasonally adjusted total, % of quarterly GDP)

Source: REINZ, RBNZ Housing Approval Survey, RBNZ New Residential Mortgage Commitments Survey, RBNZ SSR.

Note: Data anomalies identified by Reserve Bank suggest that mortgage approvals data should only be taken as indicative of mortgage origination trends.

New mortgage commitments are the most relevant metric for assessing the role of credit in the housing market, and show a more consistent relationship with the underlying levels of housing market activity. Strong rates of principal repayments by existing borrowers can also disguise a build-up of risk among new borrowers. As discussed in chapter 2, debt-to-income multiples among new borrowers in the Auckland region are likely to be increasing as house prices in the region become more stretched.

 

1 There are also a variety of different categories of mortgage commitments, including (i) commitments for the purchase of a house (ii) refinancing of an existing mortgage (iii) transfer of a mortgage between banks and (iv) bridging finance. The Reserve Bank is beginning to collect information on these different categories from banks, but is still working to improve the quality of the data.