BAILLIEU HOLST RESEARCH 09 September 2016

Banks and House Prices SECTOR REVIEW

The distribution of household debt: the fallacy of averages Figure 1: Median debt levels skewed to high-wealth / high-income households Median debt for households with housing debt $350,000 $300,000 $250,000 $200,000 $150,000 $100,000 $50,000 $0,000 1st

2nd

3rd

4th

5th

1st

Wealth Quintile

2nd

3rd

4th

5th

Income Quintile 2002

2006

2010

15-24 25-34 35-44 45-54 55-64 65+ Age (years)

2014

Source: RBA Statistical Tables, BH estimates

■ Housing debt is skewed to top wealth quintile and (especially) to top income quintile households: Housing debt is highly skewed (and skewed more positively than negatively in our view) implying greater resilience within the household sector to financial shocks than is generally assumed. There is a fallacy of averages in merely examining overall household indebtedness. Positives: 31% of households have no debt at all; 32% of home owners own outright (two-thirds of which have no household debt at all); Bottom quintile income households are quite debt averse. Risk segments: Mortgagee households that also hold geared other residential property debt (estimated 5% of all households), given double (and relatively high) leverage; Bottom quintile wealth households with housing debt (estimated 1% of all households), given that median property gearing levels here are high. ■ We have examined the distribution of housing debt in Australia across a number of dimensions to identify both segments of vulnerability and resilience within the household sector. ■ Our major bank order of preference: NAB (BUY), WBC (BUY), CBA (BUY), ANZ (HOLD).

Baillieu Holst Ltd ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 1

BAILLIEU HOLST RESEARCH 09 September 2016

Household debt distribution Green flags: Resilient segments and positive trends 1.

Outright home owners: Large base of nil or relatively modest gearing

Home owners that own their principal residence outright enjoy the considerable credit risk protection of owning an unencumbered asset; further, 66% of outright home owning households hold no household debt whatsoever (the RBA has stated that a majority of households with no debt are retired). Outright home owners are also a large section of the household sector, comprising 32.1% of all households. Outright owners also have modest other debts as well, with only 9% of outright home owning households holding investment property debt, with median investment property debt sitting at a relatively low $230,000 / 41% median investment property debt % investment property assets vs. $277,000 / 60% investment property leverage for all households. These households represent a large base of households that hold valuable collateral in their principal residence, with either no gearing or little gearing beyond that. Figure 2: Even split in housing status

Figure 3: A minority of outright home owners have debt Share of households with any debt – by housing status

Housing status of Australian households 2014

100% 90%

Renter 35%

Outright Home Owner 32%

80% 70% 60%

50% 40% 30% 20% 10%

Mortgagee Home Owner 33%

Source: HILDA Release 14.0, BH estimates

2.

All Households

Home Owner (outright)

Home Owner (mortgage)

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

0%

Renter

Any Debt

Source: RBA Statistical Tables

Relatively low income households: Debt averse

Bottom quintile income households have a low incidence of holding housing debt (only 12% hold owner occupier housing debt vs. 36% for all households; only 3% hold investment property housing debt vs. 11% for all households). Where property gearing is held amongst relatively low income households, median property gearing levels at 27% (47% all income quintiles) are the lowest amongst all household income quintiles. The debt aversion of bottom quintile income households is also broader than merely housing debt: only 43% of these households hold any debt at all (vs. 69% for all households) and such households hold comparatively little debt when they do hold it (median debt of $15,000 vs. $120,000 for all households). We see this as important inasmuch as the importance of low income / NINJA mortgages in the US sub-prime credit losses during the GFC. 3.

Top wealth households: Breaking away from the pack

The top quintile of households by wealth are a true standout in terms of their wealth holdings compared to other households, with their 2014 wealth median of $1.78mn more than four times greater than middle wealth households (in turn, implying polarised wealth). These households have enjoyed relatively strong (and continuous) rates of growth in real household wealth, both pre- and post-GFC at 43% growth 2002-2014 vs. 37% growth for all households. Such households enjoy high levels of other residential property ownership (50% vs. 21% for all households) but are likewise the most likely segment to hold investment property debt (22% vs. 11% investment housing debt penetration for all households); 37% of these households hold owner occupier housing debt. However, where property gearing is held, median housing debt % housing assets is relatively low at 23% Baillieu Holst ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 2

BAILLIEU HOLST RESEARCH 09 September 2016

(vs. 47% for all households), including low own home gearing (27% vs. 50% for all households) and low other property gearing (40% vs. 60% for all households). These households represent a collateral rich but lowly geared borrowing segment, notwithstanding relatively high levels of investment property housing debt penetration. Figure 4: Debt skewed to high income/wealth quintiles

Figure 5: Wealth skewed to high wealth/income quintiles

Median debt for households with housing debt

Mean household wealth per household

$350,000

$3.0mn

$300,000

$2.5mn

$250,000

$2.0mn

$200,000 $1.5mn

$150,000 $1.0mn

$100,000

$0.5mn

$50,000 $0,000

$0.0mn

1st

2nd

3rd

4th

5th

1st

Wealth Quintile

2nd

3rd

4th

5th

Income Quintile 2002

2006

2010

Age (years)

1st

2nd

3rd

4th

5th

1st

2nd

Wealth Quintile

3rd

4th

5th

Income Quintile ASNA

2014

Source: RBA Statistical Tables, BH estimates

4.

15-24 25-34 35-44 45-54 55-64 65+

SIH

15-24 25-34 35-44 45-54 55-64 65+ Age (years)

HILDA

Source: HILDA Release 14.0, RBA

Top income households

The top quintile of household income held almost 50% of the stock of household debt in 2014. On the positive side this implies that debt is held by households that are best positioned to cope with servicing it; on the negative side, such households might be expected to have greater volatility in their incomes over time (e.g. given the year-by-year fluctuations that can occur with bonus and equity compensation). 5.

Home ownership rates: Declining, and increasingly skewed to wealthier households

Home ownership rates since the early 2000s have been progressively declining in Australia (from 68.8% 2001 to 64.9% 2014) and have been declining broadly across every household income quintile (although the incidence of other residential property investment has been resilient). The decline has been driven by home ownership rates amongst lower wealth households (whereas the top two quintiles of wealthy households have seen steady rates of home ownership through time). Effectively home ownership status has become incrementally skewed towards mid-to-high wealth households. From a mortgage credit risk perspective this implies own home gearing also being skewed to relatively more wealthy households and is in sharp contrast to the US experience in the lead-up to the global financial crisis where home ownership rates increased (not decreased). The corollary of rising household indebtedness combined with declining rates of home ownership is that the number of properties owned by households with property is rising. Whilst on the negative side this might create in some instances very substantial geared property portfolios within some households (creating asset class concentration risk), it also diversifies household asset bases, creating multiple pools of collateral.

Red flags: Vulnerable segments and negative trends 1.

Households that are both mortgagee home owners AND geared property investors

By virtue of their mortgagee status, the median debt levels of mortgage households is a relatively high $292,000 (compared to all households at $120,000), and the real value of this household debt for mortgagees has doubled from 2002 to 2014. Amongst the 32.8% of all households that are mortgagee home owners, 14% of these households also carry investor property debt (a rising proportion over time), creating a situation of doubleleverage to mortgage debt; an estimated 5% of all households fit into this category (being 32.8% x 14%). The level of other residential property gearing held by mortgagees is not Baillieu Holst ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 3

BAILLIEU HOLST RESEARCH 09 September 2016

only relatively high at 67% (compared to 60% for all households) but is also the highest level of other property gearing amongst all categories of housing status. The median debt held by mortgagees on their own home ($245,000) is supplemented by a relatively high median debt on investment housing ($300,000). These mortgagee home owners whom are also geared other residential property owners represent multiple pools of debt against the same asset class of residential housing. Figure 6: Mortgagees' real debt has doubled

Median real debt for households owing any debt – by housing status $300,000 $250,000

$200,000 $150,000 $100,000 $50,000

All Households

Home Owner (outright)

Home owner (mortgage)

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

$0,000

Renter

Any Debt

Source: HILDA Release 14.0

2.

Low wealth households: Very highly geared properties (where gearing exists)

The bottom quintile of households by wealth have relatively low levels of property ownership (6% home ownership, 3% other residential property) and even lower levels of housing debt penetration (5% owner occupier housing debt, 2% investment property housing debt). However in these comparatively rare instances where such households do hold property gearing, median gearing levels of housing debt % housing assets are very high at 99% (95% for own home gearing and 92% for other property gearing). Median housing debt levels in this segment are also the highest across all quintiles ($340,000 vs. $245,000 for all households). The existence of high gearing creates scope for negative equity to easily emerge in the event of house prices declining (with little housing or other household collateral in existence by definition); the existence of high median debt levels creates scope for larger / more meaningful credit losses to be incurred. However, such households are relatively rare, estimated at only 1% of all households (20% of households x 5% penetration of owner occupier debt in this segment). We suspect that such households reflect the low deposit / no deposit mortgage market, where the use of lenders' mortgage insurance as a credit risk mitigant by lenders is also likely to be high, which perhaps moderates the mortgage credit risk of these households for lenders. 3.

Renters with investment properties: Dependence on negative gearing?

Amongst renting households 13% own other investment property and 10% have an investment property with housing debt on it; prima facie 3.5% of households are in this category (being 35.1% x 10%). This segment might be potentially vulnerable to the extent that the viability of the other residential property investment is dependent on negative gearing tax benefits (according to the HILDA Survey, in 2014 47.1% of households owning investment housing were negative geared, sample ATO personal tax return data suggests that almost two-thirds of investment property is negatively geared).

Baillieu Holst ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 4

BAILLIEU HOLST RESEARCH 09 September 2016

Figure 8: …the high income and the wealthy

Figure 7: Investment property ownership skewed to… Owners of residential investment property – by income quintile

Owners of residential investment property – by wealth quintile

100.0%

100%

90.0%

90%

80.0%

80%

70.0%

70%

60.0%

60%

50.0%

50%

40.0%

40%

30.0%

30%

20.0%

20%

10.0%

10% 0%

0.0% 2006

2010 Bottom

2nd

2006

2014

Middle

3rd

2010 Bottom

Top

2nd

Middle

2014 3rd

Top

Source: HILDA Release 14.0

4.

Middle income owners: Highest level of housing gearing

Middle income households holding property debt have the highest level of housing debt % housing assets amongst all the income quintiles at 51% (vs. 47% for all households), notably in other property gearing (64% vs. 60% for all households). Amongst middle income households 34% hold owner occupier housing debt and 9% hold investor housing debt. Figure 9: Middle income households relatively geared…

Figure 10: …particularly in relation to other property

Median housing debt % housing assets by income quintile for households with housing debt

Median housing debt % housing assets by income quintile for households with housing debt

60%

70%

50%

60%

50%

40%

40% 30% 30% 20%

All Income Quintiles

1st

2nd

3rd

4th

5th

All Income Quintiles

Property Gearing

1st

2nd

Own Home Gearing

3rd

4th

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

2006

2002

2014

2010

0%

2006

0%

2002

10%

2002

20%

10%

5th

Other Property Gearing

Source for both charts: RBA Statistical Tables

5.

Older households: Rising incidence of owner occupier housing debt

Older households have over time (2002 to 2014) seen a rising incidence of owner occupier housing debt being held, namely from 22% to 36% amongst 55-64 year olds and from 5% to 14% amongst 65-74 year olds. However, amongst such households the level of gearing – owner occupier housing debt % owner occupier housing assets – is relatively low (30% for 55-64 year olds; 17% for 65-74 year olds). The existence of housing gearing amongst retirement and pre-retirement aged households creates a risk that such debt may not be easily serviced if / when employment income ceases, but the credit risk associated with this is mitigated by: 1) The premium levels of median wealth levels enjoyed by both these household segments (circa $750,000 each); 2) The ability to access superannuation assets to service such debt (such households have reached or are soon approaching eligibility age); and 3) As illustrated in the following charts, median home equity of older households is relatively high. Bringing the data of both charts together, a position of negative equity would only be achieved for households aged 65+ following (an unlikely) 95% decline in house prices, compared to a 28% decline for 25-34 year old households:

Baillieu Holst ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 5

BAILLIEU HOLST RESEARCH 09 September 2016

Figure 11: Median home values similar…

Figure 12: …but home equity rises sharply with age

Median home value – by age cohort

Median home equity of home owners – by age cohort

$600,000

$600,000

$500,000

$500,000

$400,000

$400,000

$300,000

$300,000

$200,000

$200,000

$100,000

$100,000

$0

$0 25-34

35-44

45-54 2002

2006

2010

55-64

65+

25-34

35-44

2014

45-54 2002

2006

2010

55-64

65+

2014

Source for both charts: HILDA Release 14.0

6.

Property gearing generally has risen over time

Where property gearing is held, median housing debt % housing assets has increased from 36% (2006) to 47% (2014); bottom quantile income households are least geared (27%) and middle income households the most geared (51%). Gearing has declined over time amongst the lowest income quintile households but risen for other households (notably the 3rd and 4th middle income quantile households). Owner occupier housing gearing has increased from 40% (2002) to 50% (2014) over time; investment property gearing has also increased from 53% (2002) to 60% (2014), with investor property continuously more geared through time than owner occupier gearing. As the HILDA Survey states: "Despite the fall in home ownership and the decline in mean home wealth between 2010 and 2014, mean home debt among all households rose in a sustained fashion. In 2014, mean home debt was, in real terms, nearly double its 2002 level."

Baillieu Holst ABN 74 006 519 393 www.baillieuholst.com.au Please read the disclaimer at the end of this report.

Page 6

BAILLIEU HOLST RESEARCH

This document has been prepared and issued by: Baillieu Holst Ltd ABN 74 006 519 393 Australian Financial Service Licence No. 245421 Participant of ASX Group Participant of NSX Ltd

Baillieu Holst Ltd ABN 74 006 519 393 Australian Financial Service Licence No. 245421 Participant of ASX Group Participant of NSX Ltd

www.baillieuholst.com.au

Analysts’ stock ratings are defined as follows: Buy: The stock’s total return is expected to increase by at least 10-15 percent from the current share price over the next 12 months. Hold: The stock’s total return is expected to trade within a range of ±10-15 percent from the current share price over the next 12 months. Sell: The stock’s total return is expected to decrease by at least 10-15 percent from the current share price over the next 12 months.

Disclosure of potential interest and disclaimer: Baillieu Holst Ltd (Baillieu Holst) and/or its associates may receive commissions, calculated at normal client rates, from transactions involving securities of the companies mentioned herein and may hold interests in securities of the companies mentioned herein from time to time. Your adviser will earn a commission of up to 55% of any brokerage resulting from any transactions you may undertake as a result of this advice. When we provide advice to you, it is based on the information you have provided to us about your personal circumstances, financial objectives and needs. If you wish to rely on our advice, it is important that you inform us of any changes to your personal investment needs, objectives and financial circumstances. If you do not provide us with the relevant information (including updated information) regarding your investment needs, objectives and financial circumstances, our advice may be based on inaccurate information, and you will need to consider whether the advice is suitable to you given your personal investment needs, objectives and financial circumstances. Please do not hesitate to contact our offices if you need to update your information held with us. Please be assured that we keep your information strictly confidential. No representation, warranty or undertaking is given or made in relation to the accuracy of information contained in this advice, such advice being based solely on public information which has not been verified by Baillieu Holst Ltd. Save for any statutory liability that cannot be excluded, Baillieu Holst Ltd and its employees and agents shall not be liable (whether in negligence or otherwise) for any error or inaccuracy in, or omission from, this advice or any resulting loss suffered by the recipient or any other person. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Baillieu Holst Ltd assumes no obligation to update this advice or correct any inaccuracy which may become apparent after it is given.

Melbourne (Head Office) Address Level 26, 360 Collins Street Melbourne, VIC 3000 Australia Postal PO Box 48, Collins Street West Melbourne, VIC 8007 Australia Phone +61 3 9602 9222 Facsimile +61 3 9602 2350 Email [email protected] Adelaide Office Address 1, 341 Payneham Road Marden, SA 5070 Australia Phone +61 8 7074 8400 Facsimile +61 8 8362 3942 Email [email protected] Bendigo Office Address Cnr Bridge & Baxter Streets Bendigo, VIC 3550 Australia Postal PO Box 40 North Bendigo VIC 3550 Australia Phone +61 3 4433 3400 Facsimile +61 3 4433 3430 Email [email protected] Brisbane Office Address Level 18, 333 Ann Street Brisbane, QLD 4000 Australia Phone +61 7 3232 1110 Email [email protected] Geelong Office Address: 16 Aberdeen Street Geelong West Vic 3218 Postal PO Box 364 Geelong Vic 3220 Australia Phone +61 3 5229 4637 Facsimile +61 3 4229 4142 Email [email protected] Newcastle Office Address Level 1, 120 Darby Street Cooks Hill, NSW 2300 Australia Postal PO Box 111 The Junction, NSW 2291 Australia Phone +61 2 4037 3500 Facsimile +61 2 4037 3511 Email [email protected] Perth Office Address Level 10, 191 St Georges Terrace Perth WA 6000 Australia Postal PO Box 7662, Cloisters Square Perth, WA 6850 Australia Phone +61 8 6141 9450 Facsimile +61 8 6141 9499 Email [email protected] Sydney Office Address Level 18, 1 Alfred Street Sydney, NSW 2000 Australia Postal PO Box R1797 Royal Exchange, NSW 1225 Australia Phone +61 2 9250 8900 Facsimile +61 2 9247 4092 Email [email protected]

Baillieu Holst Ltd ABN 74 006 519 393 www.baillieuholst.com.au

Page 7