Via the excellent Jonathon Mott at UBS:
We estimate the RBA rate cuts and APRA’s removal of its interest rate serviceability floor may improve maximum borrowing capacity by around 14%. However, these changes need to be considered in the context of ongoing tightening, in particular a new HEM methodology, the rollout of Comprehensive Credit Reporting and Open Banking.
A new estimate of the Household Expenditure Measure (HEM, a floor on estimated living expenses in mortgage calculations) is being rolled out by the banks over coming months. Our channel checks indicate the revised HEM is skewed to higher income earners, especially above ~$120k. Although each bank now applies the HEM slightly differently, we estimate that for households with income above $150k (~33% of owner occupied mortgagors by number and 50% by value), the increase in HEM may more than offset the impact of lower interest rates and removal of APRA’s serviceability floor. Given ~50-60% of borrowers are still assessed using HEM assumptions, these changes may have a material impact on the housing market, especially in Sydney and Melbourne given higher median incomes and house prices. Investment property buyers are also generally higher income earners and will also be impacted by these changes. For lower income households, the impact of HEM changes is less material implying the benefit of rate cuts and removal of the serviceability floor may be more effective.
This is the single largest reason why we see housing struggling to recover, needing a lot more stimulus to do so, and only posting modest gains when it does lift.
The last round of housing inflation was all about HEM, previously via Endeavour Equities:
1. Downgrading our outlook for residential property in 2019; expecting peak to trough falls of 25-30% – the worst since 1890.
i) We are downgrading our outlook for residential property in 2019 with peak to trough falls of 25-30% – the worst since 1890. We expect -10 to -15% % in 2019 in addition to falls of -15% in 2018.This means 2014 vintages will see significant losses while many from 2015, 2016 and 2017 will experience negative equity.
ii) Key insights from Endeavour’s 2018 Report “Credit Crunch 2018 as HEM/ NonPrime Bubble Busts” of June 2018 are driving credit and property prices into a major slump. The key revelation that emerged late 2018; all (or almost all) mortgages were based on the HEM as a default from 2012 to 2016, and this has driven a Non-Prime bubble in terms of DSTI ratios.
iii) The key driver of our current property price downgrade is the late 2018 revelation (ASIC vs Westpac) that Westpac and likely other ASX banks did not collect, or certainly did not input, actual borrower expenses into loan serviceability calculators. Instead they applied their own downwardly biased HEM expenses of $32k at all income levels as the default or policy driven input. This upwardly biased Debt Service to Income ratios by 10-15%+. We note ASIC has alleged that this has made all such loans irresponsible lending (all 260,000 mortgages written to 2016 in Westpac’s case).
iv) We expect a continuation of the 2018 Credit Crunch well into 2019 as the HEM/ non-prime bubble busts due to the combined impact of i) real expenses shifting sharply towards a ABS HES Survey reality and ii) amortization of Interest Only loans. Together these impacts are expected to hit loan borrow sizes for aggressively geared borrowers by 46%+, savaging borrowing capacity for the marginal price setter of housing in the boom to 2016 – the highly IO borrower using HEM expenses. The last phrase of this doesn’t make sense in the sentence
2. The size of the Credit Crunch will be directly proportional to the unreasonable of the HEM – very large!
i) The Size of the Credit Crunch is directly proportional to the unreasonableness of the HEM expenses benchmark. Since HEM expense estimates are unreasonably low, the credit crunch will be significant and ongoing as it is increasingly replaced with reasonable expenses that are consistent with Responsible Lending Laws.
ii) The Median Borrower on a HH income of $144k HEM understated expenses by $48k p.a. leading to loan sizes 30%+ or $380k larger than if HES based survey expenses were used. For the median debt which is owned by households on $180k+, the understatement of expenses is considerably larger – up to a total of $80k. This led to loan sizes $640k larger than if HES expenses had been used.
iii) Failure to amortize Interest Only Loans over the non IO periods in serviceability calculators has also inflated loan sizes 20-30%+
Without mortgage fraud, there is no sustained housing recovery coming.