MEI Special Topic: New Zealand Household Debt

Debt, and household debt in particular, garnered considerable attention earlier in the year. This special topic provides an overview of household debt developments following the release by the Reserve Bank in September of its latest household balance sheet data. At an aggregate level there are reasons to believe that debt levels remain manageable, albeit with debt levels being influenced by house price developments. At an individual level there are likely to be subgroups of households that are more exposed, with high debt levels increasing households' vulnerability to shocks in income, employment and interest rates.

Household debt continues to rise...

Household debt relative to disposable income increased to 165 percent in June (Figure 1). Debt to income has been rising since March 2012 and is currently slightly higher than the level that prevailed over 2007 to 2010.

Figure 1: Household debt*

*includes rental property loans
Source: RBNZ

... but is within OECD norms

International comparisons of household debt levels are problematic. The OECD-sourced data in Figure 2 does not include New Zealand given our lack of National Accounts balance sheet data. Comparing the RBNZ-sourced New Zealand data with those of the OECD would place New Zealand in the top of the third quartile. However, the figure for New Zealand includes rental property loans, some of which – depending on the nature of ownership – could be reclassified to other (non-household) sectors. Excluding rental property loans entirely would result in a debt to income ratio between Belgium and Spain – in the middle of the distribution.

Figure 2: Household debt across countries

Source: OECD, RBNZ

Several features limit concerns about debt levels

From a macroeconomic perspective, it is not clear what levels of debt are “too high” or problematic with the ability to take on debt playing an important role in enabling households to spread the cost of expensive assets, such as housing, across time.
At an aggregate or macroeconomic level several features of the recent build-up of debt limit concern, namely:

• The accumulation of debt has to date not been the result of sustained dissaving;
• Debt servicing rates are well within historical ranges and may fall further;
• Net wealth and (until recently) net financial wealth has been increasing; and
• Housing debt to asset ratios have fallen.

Until recently growth in household financial assets had been outpacing debt...

While household debt to income has been increasing since 2010, household financial assets had been growing more rapidly and as a result net financial wealth was increasing (Figure 3). Coupled with rising house prices, overall household net wealth is returning towards pre global financial crisis (GFC) levels.
However, the latest vintage of wealth data incorporates downward revisions to the valuation of equity holdings by households and as a result net financial wealth to income has been declining over the past year, with increases in net wealth driven by house price increases.

Figure 3: Net wealth and net financial wealth

Source: RBNZ

... reflecting a change in household behaviour post GFC...

Since the global financial crisis, there has been quite a marked change in household saving behaviour. Between 2010 and 2014 household saving was positive following negative rates over the 9 years prior (Figure 4). Saving turned slightly negative in 2015 reflecting a large fall in farm income which declined by 2.5 percentage points of disposable income. Nevertheless, the rate of dissaving (at 0.5%pts of disposable income) was modest relative to the mid-2000s.

Figure 4: Household saving

Source: Statistics NZ

One balance sheet manifestation of this change in behaviour was that household deposit holdings had been growing at around, or slightly faster than, the accumulation of debt. This contrasts to the experience of the 2000s of consistently higher debt accumulation (Figure 5).

... although recent divergence worth monitoring

Rising net financial wealth and the pattern of an approximate matching of debt growth and deposit holdings were reasons to help balance concerns about rising indebtedness. The recent divergence shown in Figure 5 is still too short and small to call a trend, but if it were to continue may challenge a relatively sanguine view of household indebtedness.

Figure 5: Annual change in debt and deposits

Source: RBNZ

Debt servicing has fallen...

While debt levels have been rising, declines in interest rates have meant that debt servicing has fallen. Debt servicing (interest costs) peaked at 14% of disposable income in 2008. As interest rates have fallen debt servicing costs declined to 8.4% of disposable income in 2013 and currently sit just below 9% (Figure 6).

Figure 6: Debt servicing

Source: RBNZ

The future evolution of debt servicing rates will be influenced by both the accumulation of debt and interest rate developments. With current market mortgage rates below the average interest rate currently being paid, it is likely that average interest rates will continue to decline over the year ahead. Future increases in interest rates would increase debt servicing rates. Whether this results in debt servicing rates exceeding previous highs will depend on the extent and timing of interest rate increases.

... as has the extent to which the housing stock is leveraged

Housing debt relative to the overall value of housing assets fell to a series low (dating back to 1998) of just under 24% in March 2016 after being just below 29% at the start of 2009 and in late 2001. House prices would need to fall around 17% for leverage to return to its previous peaks.

Aggregate data may hide areas of greater pressure...

While the focus of this special topic has been on aggregate household wealth data, it is possible that this may hide issues if debt is concentrated in particular groups of households. In particular, high Auckland house prices mean that new entrants need to take on high debt levels.

... particularly debt servicing risks for new entrants in Auckland

Estimates of debt servicing ratios for representative new buyers in Auckland are elevated (around 46% of income) even with current low interest rates. They would pass previous peaks if mortgage rates were to return to their 10-year average of 6.7%.[1] A representative buyer is defined as one buying an average value house, on an average income, with a 20 percent deposit.

... although these buyers may have higher than average incomes

Actual new buyers may face higher or lower debt servicing rates depending on their particular circumstances. Previous research covering buyers during the 2005 to 2013 period highlighted that new buyers on average had disposable incomes that were 40% higher than the overall median.[2] Nevertheless, debt servicing rates for many new buyers are likely to be relatively high at a time when interest rates are relatively low.

Domestic factors are unlikely to lead to aggregate distress

Almost by definition, higher debt levels mean that households are more exposed to increases in interest rates. However, in the absence of positive demand shocks, it is unlikely that interest rates will rise over the next few years. Domestically driven increases in interest rates via increases to the official cash rate will be dependent on inflationary pressures increasing as domestic activity grows. This is likely to be associated with stronger nominal income growth and an improving labour market. Higher rates of debt mean that monetary policy may be more effective in influencing demand with interest rates not needing to rise as much as in the past.

Overall, at the macroeconomic level there are reasons to limit concern about household debt developments, while being cognisant of the risks. These risks may increase if the most recent developments of declining net financial wealth were to continue or accelerate.

Feedback

Treasury Staff Insights: Rangitaki

See Treasury Staff Insights: Rangitaki for other articles in this series.