Box A: LVR effectiveness monitoring
This page contains information on LVR effectiveness monitoring from the May 2014 Financial Stability Report.
In October 2013, the Reserve Bank introduced a speed limit on high-LVR residential mortgage lending by registered banks. The Reserve Bank’s initial estimates were that LVR restrictions would lower house sales by 3-8 percent, house price inflation by 1-4 percentage points, and housing credit growth by 1-3 percentage points, over the first year that the restrictions are in place.1 In addition, the September 2013 Monetary Policy Statement estimated that the speed limit would reduce inflationary pressures by the equivalent of a 30 basis point increase in the OCR.2 The Reserve Bank currently judges that LVR restrictions are meeting their objective of mitigating the risks associated with excessive growth in housing-related credit and house prices, with clear evidence of a particularly strong restraining impact on housing market activity in the first six months of implementation.
Since October, banks have rapidly reduced the share of high-LVR lending to well below the speed limit requirement of 10 percent (figure A1). All banks met the first deadline for speed limit compliance at the end of March 2014, with a system-wide share of high- LVR lending of 5.6 percent over the first six months of implementation. The LVR restrictions exempt certain categories of lending, including Welcome Home loans, refinancing and construction. Uptake of these exemptions was limited over the first compliance period, with total exemptions accounting for just 1 percent of total lending in the six months ended 31 March 2014. This uptake is lower than the 5 percent share of lending initially assumed by the RBNZ, primarily due to a lower than expected use of the refinancing exemption.
Figure A1: High-LVR mortgage lending (percent of new commitments)
Source: RBNZ New Residential Mortgage Commitments Survey.
Data since October confirm that house sales have dropped sharply since the introduction of LVR restrictions, following a run-up in the months prior to the restrictions coming into effect (figure A2). National house sales dropped 11 percent between October 2013 and March 2014, with the drop in sales volumes evenly spread across regions. This impact is greater than the initial expectation of a 3-8 percent drop (over the year to October 2014). By comparison, Reserve Bank modelling estimates suggest that, in a counterfactual scenario where the LVR restrictions were not imposed, house sales would likely have increased further in the months since October.3
Figure A2: House sales
The drop in sales appears to have been more pronounced in certain segments of the housing market. Across different price brackets, the reduction in house sales has been concentrated in lower value house sales. House sales dropped 23 percent between September 2013 and March 2014 in the under-$400,000 value bracket, compared to an 11 percent drop in aggregate. Looking at buyer categories, the share of first home buyers has declined slightly since the introduction of LVRs. According to data produced by CoreLogic, the first home buyer share of home sales declined to 17 percent in February, from an average of around 20 percent over the past two years.
House price inflation also appears to have moderated since the implementation of LVR restrictions. Measuring this has been complicated by the decline in lower value sales, which has created a significant upward bias in simple measures of house price inflation, such as median house prices. The QV quarterly house price index comprehensively adjusts for the composition of house sales, and shows annual growth slowed by 1 percentage point to 9 percent in the final quarter of 2013. Growth in the more timely REINZ stratified price index also slowed in the final quarter of 2013. Despite some adjustments for the composition of sales, this measure appears to have been subject to a degree of upward bias. Annual growth in this index has generally slowed further more recently, notwithstanding a rebound in the most recent March data. The Reserve Bank estimates that, in the absence of LVR restrictions, annual house price inflation could have been around 2.5 percentage points higher in the year to March 2014 (figure A3).
Figure A3: House price growth, including counterfactual without the LVR speed limit (annual 3-month moving average)
Source: REINZ, RBNZ estimates.
There are also signs that housing credit growth is beginning to moderate in line with reduced property market activity and prices. This slowing is most clearly evident in data from the early stages in the mortgage origination process. Annual growth in seasonally adjusted mortgage approvals and major banks’ new mortgage commitments dropped 22 and 17 percentage points respectively between September 2013 and March 2014. With a typical lag of up to three months between initial mortgage approval and final drawdown, and around one month for commitments, these data point to the likelihood of moderation in final housing credit growth in coming months. Housing credit growth was losing momentum at the end of the March quarter, with an annualised decline of 1 percent between December 2013 and March 2014.
There have been few signs to date of homelending migrating beyond the regulatory perimeter of LVR restrictions. Specifically, there is little evidence of either avoidance activity by the registered banks or a shift to non-bank financial intermediaries and other sources of finance. The Reserve Bank will continue to monitor closely for any signs of regulatory leakage from the speed limit.
1 See Bloor, C and C McDonald (2013) 'Estimating the impacts of restrictions on high-LVR lending', Reserve Bank of New Zealand Analytical Note 13/05, October.