Banking Stats – Page 5 – Digital Finance Analytics (DFA) Blog

APRA Releases 2017 Stress Test Summary

APRA has released some information relating to their 2017 stress tests in their “APRA Insight Issue Three 2018“. We discussed the results in a recent Adams/North video. We were not impressed!

This is what APRA has now released. Compared with the bank by bank data the FED releases, its VERY high level!

Stress testing plays an important role for both banks and APRA in testing financial resilience and assessing prudential risks. This article provides an overview of APRA’s approach to stress testing, with insights into how APRA developed the 2017 Authorised Deposit-taking Institution (ADI) Industry Stress Test. This builds on the recent speech by APRA Chairman Wayne Byres, “Preparing for a rainy day” (July 2018), which outlined the results from the 2017 exercise.

Why stress test

The aim of stress testing is, broadly, to test the resilience of financial institutions to adverse conditions, including severe but plausible scenarios that may threaten their viability. The estimation of the impact of adverse scenarios on an entity’s balance sheet provides insights into possible vulnerabilities and supports an assessment of financial resilience. It can be a key input into capital planning and the setting of targets, and in developing potential actions that could be taken to respond and rebuild resilience if needed.

Stress tests are particularly important for the Australian banking system, given the lack of significant prolonged economic stress since the early 1990s. As a forward-looking analytical tool, stress testing can help to improve understanding of the impact of current and emerging risks if a downturn were to occur. Stress tests provide an indication, however, rather than a definitive answer on the impact of adverse conditions, and the results will always reflect the inherent uncertainty that exists in any scenario.

Scenarios – the starting point

The starting point in developing a stress test is to design the scenarios. This is the foundation of the exercise, and it is important that scenarios are calibrated effectively and are well targeted. For industry stress tests, APRA collaborates with both the Reserve Bank of Australia (RBA) and Reserve Bank of New Zealand (RBNZ) on the design of scenarios and the economic parameters that define them. In 2017, APRA also engaged with the Australian Securities and Investments Commission (ASIC) in the design of the operational risk scenario.

Scenarios should be aligned to the purpose of the exercise. At a high level, these are guided within APRA by several objectives – namely to:

assess system-wide and entity-specific resilience to severe but plausible scenarios;

improve stress testing capabilities across the industry; and

support the identification and assessment of core and emerging risks.

In line with these objectives, APRA developed two scenarios for the 2017 industry stress test. The first was a macroeconomic scenario with a China-led recession in Australia and New Zealand. This was considered the most severe but plausible adverse economic scenario for the banking system at the time.[1] In this scenario, there was assumed to be a fall in Australian GDP of 4 per cent, while house prices declined 35 per cent nationally over three years. The chart below provides an indication of the severity of the scenario: it shows that the assumed peak unemployment rate in the scenario was similar to the 1990s recession in Australia, the United States’ experience in the global financial crisis (GFC), as well as recent overseas regulatory stress tests.

Chart 1: International comparison: Peak unemployment rate

The second scenario involved the same macroeconomic parameters, with an additional operational risk event added. This was designed to test bank resilience beyond traditional economic risks and to consider other vulnerabilities. In this operational risk scenario, APRA asked the participating banks to identify a material operational risk event involving conduct risk and/or mis-selling in the origination of residential mortgages.[2]

The banks chosen for the exercise were selected based primarily on scale, enabling a clear system-wide view to be generated. The 2017 stress test covered 13 of the largest banks with aggregate assets of nearly 90 per cent of the system, as well as the leading lenders mortgage insurers. It included both the major banks and smaller entities that have a significant presence in particular regions or asset classes.

The testing process

The 2017 stress test was run in two phases. In the first phase, banks generated results using their own stress testing methodologies and models.[3] The results of this phase varied by differences in the risk characteristics of each bank, as well as by how they interpreted and modelled the scenarios.

In the second phase, APRA provided more prescription with specified risk estimates for loan portfolios and other assumptions. This enabled greater comparability across the banks and more consistency in the results. The APRA estimates were developed based on the banks’ results from the first phase, APRA’s own internal modelling, historic and peer stress testing benchmarks, internal and external research and expert judgement.

The estimates were also differentiated based on inherent risk characteristics. For example, riskier interest-only loans were assumed to be more likely to default and losses were estimated to be greater for loans with higher Loan to Valuation Ratios (“LVRs”). For the operational risk scenario, APRA defined key elements and estimates to be applied by the banks, based on research of overseas experience as benchmarks for potential impacts.

Interpreting the results

The table below summarises some key results from the macroeconomic scenario. As important as the quantified outcomes are the lessons learned and implications of the exercise. Stress testing should not be an academic exercise, but should be used to inform assessments of resilience, risks and capacity to respond to stress.

In analysing the results, APRA assessed not only the impact on capital, but also the effects on profitability and loan portfolios. The differences between phase 1 (based on bank’s own modelling) and phase 2 (based on APRA risk estimates) can also shed light on bank stress testing modelling capabilities.

Macroeconomic scenario – aggregate results Phase 1 Phase 2

Peak to trough decline in Common Equity Tier 1 capital -2.87% -3.21%

Peak to trough decline in Return on Equity -9.86% -12.05%

Peak credit loss rate[4] 0.81% 0.90%

Macroeconomic scenario

Given the design of the scenario, the impact on the participating banks was material. The results show that in this scenario the decline in profitability was severe and occurred quickly. Return on equity (ROE) in aggregate fell materially in the first two years before recovering after year three. The impact on profitability led to a significant fall in capital in both phases, as shown in Chart 3. There was, however, a wide range in results across banks in both phases.[5]

Macroeconomic scenario – aggregate results	Phase 1	Phase 2
Peak to trough decline in Common Equity Tier 1 capital	-2.87%	-3.21%
Peak to trough decline in Return on Equity	-9.86%	-12.05%
Peak credit loss rate[4]	0.81%	0.90%

Chart 2: Aggregate return on equity

Chart 3: Cumulative CET1 impact – Macroeconomic scenario

The losses were driven by bad debts in residential mortgage lending, corporate lending and other credit portfolios, as well as lower net interest income and losses on large single counterparties. Across the mortgages portfolios of the banks, aggregate losses were similar in both phases and although the mortgage portfolio contributed the largest aggregate loss, the loss rate was lower than business lending and other consumer lending portfolios.[6] The banks also modelled the impact of the scenario on liquidity and funding positions; most banks were able to maintain their liquidity with a liquidity coverage ratio (LCRs) above 100 per cent (or initiated strategies to restore their LCR within a reasonable timeframe).

Chart 4: Aggregate cumulative credit losses

Operational risk scenario

In the second scenario, the impact of the operational risk event led to a more severe capital outcome, as shown in Chart 5. The operational risk events modelled by the banks in Phase 1 represented a wide range of potential risks, including broker fraud, inappropriate product design and sales practices, inappropriate verification and documentation, overstatement of valuation errors at origination, and serviceability errors. The impact of the operational risk event in Phase 2 based on APRA-defined assumptions was, however, more severe.

Chart 5: Cumulative CET1 impact – Operational risk scenario

Conclusions and lessons learned

Following industry stress test exercises, APRA provides participating entities with formal feedback. In the 2017 stress test, APRA noted the importance of ongoing improvements in modelling capabilities and developing better internal model governance and discussions on results. In practice, a stress event is unlikely to play out exactly as designed and simulated in hypothetical scenarios, reinforcing the need to continue to stress test and enhance stress testing capabilities.

In APRA’s view, the results of the 2017 exercise provide a degree of reassurance: ADIs remained above regulatory minimum levels in what was a very severe stress scenario. In addition, the results were presented before taking into consideration management actions that would likely be taken to rebuild capital and respond to risks in the scenarios.[7] The impact on profitability, loan portfolios and capital would, however, be substantial: this underlines the importance of maintaining strong oversight and prudent risk settings, unquestionably strong capital and ongoing crisis readiness.

Footnotes

The RBA’s Financial Stability Review in October 2016 highlighted the risks posed by high and rising levels of debt in China, at a time of slowing growth and signs of excess capacity.
Overseas experience in the GFC has highlighted that operational risk events can impact the financial system alongside an economic downturn. For example, in the US, there was a significant impact from sub-prime mortgage lending, while in the UK, there was additional stress related to the mis-selling of payment protection insurance.
In order to ensure consistency for the exercise, APRA provided guidance and templates for results.
This represents the credit loss rate in the most severe year of the stress. Aggregate losses over the duration of the stress period were higher.
The range presented in the charts is the interquartile range (middle 50 per cent of entities).
Banks determine credit losses by calculating the likelihood of the loans defaulting (“probability of default”) and then the actual loss on the defaulted loans (“loss given default”) which is then applied to the stressed value of the asset. For mortgages, insurance on higher LVR loans helps to mitigates losses.
These actions include raising equity, loan repricing, cost cutting, tightening lending standards and balance sheet measures. Within this set, the cornerstone action was typically equity raisings, as a relatively quick step. In the operational risk scenario, the participating banks raised around $40bn in equity, almost twice the level as during the GFC. The assumption that banks were last to market challenged thinking around the potential capacity and pricing implications.

Discussing Home Price Dynamics – Live Event Recording

I recently ran our monthly YouTube live stream event, in which we discussed the trajectory of home prices, and why we expect more falls in the months ahead.

The event is available to watch on YouTube, complete with the chat room questions and answers in real-time.

You can learn more about the 60 Minutes segment I participated in, as well as the latest trends in lending, home prices and sales transactions.

The “killer slide” is this one:

The number of property transfers are way, way down, and not just in Sydney. This is based on ABS data released yesterday.

Upping The Ante 10 Years Later – The Property Imperative Weekly – 15 Sept 2018

Welcome to the Property Imperative weekly to 15th September 2018, our digest of the latest finance and property news with a distinctively Australian flavour.

Watch the video, listen to the podcast or read the transcript.

scrollings=”no”

On the 10^th Anniversary of the failure of Lehman Brothers, the consensus seems to be that the financial system is still stressed, under the impact of sky high global debt, artificially low interest rates and asset bubbles. The shadow is long, and the risks high. I discussed this on ABC Radio Sydney, and also in a Video Post with Robbie Barwick from the CEC. Perhaps of most concern is the lack of acceptance that we have a problem, with the RBA this week recognising that household debt is high, but declaring it manageable and the Housing Industry Association calling for a relaxation of lending standards to support housing construction. That is in my view the last thing we need. The truth is, pressures on households, and tighter lending standards mean more price falls will follow. Those who follow my analysis will know I run four scenarios, including the one, the worst case, where prices could drop 40-45% from their highs over the next few years. This is the angle which the upcoming 60 Minutes programme, to be aired tomorrow, Sunday is driving at.

Just remember this is one of four scenarios! But its rated a 20% probability now.

There was more evidence this week as to the issues under the hood. For example, Domain says that whilst housing affordability has improved in all capitals where property prices have started to decline, the median multiple is still well above affordable housing thresholds in several capital city markets. They said that drawing on Domain price data and adjusted census income data, the change in price and the median multiple across capital city markets, since the respective peaks, was analysed.

While the house price to income ratio is a simple, standard indicator for understanding affordability — particularly across countries — it is far from comprehensive. Other affordability metrics still spell out tough times ahead for homeowners. Rental affordability, mortgage serviceability and the deposit hurdle are also vital considerations. But Domain says that as of June 2018, data shows the median income household in Sydney would require 59.8 per cent of weekly income to service an owner-occupied mortgage (assuming a 5.2 per cent variable rate on a loan-to-value ratio of 80 per cent). This is down from 64.4 per cent at the peak of the latest cycle

Another angle is credit scoring, as Banking Day called out, as the remaining three Big Four banks are reportedly getting ready to join NAB as participants in the new Comprehensive Credit Reporting regime. This means a massive database will share their customers’ full credit history with each other for the first time from the end of this month, at which point comprehensive credit reporting will be a foregone conclusion with the remaining major banks. The new data-sharing regime will allow lenders to better verify loan applications and assess credit risk by accessing the full repayment history of a potential customer, including their total debts. The major lenders have pushed ahead with the changes following pressure from the prudential regulator, The Australian reported, noting that ANZ said it had been testing positive data reporting since the end of June, although the data was not shared with the public at this stage. The big banks’ embrace of the new regime would put pressure on others to sign up, since only lenders who supplied comprehensive reporting to the credit bureaus would have access to the data, Australian Retail Credit Association chairman Mike Laing told The Australian. “If they don’t join then the people who intend to borrow money but not pay it back will quickly find out which ones are not in the system and they’ll go to the lenders who don’t have access to verifiable data. So it’s risky for a lender not to take part once most of the data is in there”.

And yet another angle. Between 2008 and 2012, the number of self-managed super funds grew by 27 per cent to nearly half a million. That was more than 40 per cent of the growth of the whole superannuation system. The global financial crisis coincided with the Howard government lifting the ban on superannuation funds borrowing money. As a result, self-managed super funds have rushed to take advantage and racked up $32 billion in debt in little more than a decade. The Financial System Inquiry in 2014 recommended that borrowing by superannuation funds be banned. It’s a view shared by Saul Eslake, the former ANZ Bank chief economist, who describes the decision to allow super funds to borrow as “the dumbest tax policy of the last two decades.” “The last thing Australians really needed in the last 20 years is yet another vehicle or incentive for Australians to borrow more money in order to speculate on property prices continuing to rise,” Mr Eslake said.

Overlaying that is the perennial problem of property spruikers trying to persuade people to borrow big to buy, and tip their newly acquired, heavily leveraged, property into a self-managed super fund. Super fund borrowing is known as “limited recourse” — which means if the fund can’t pay off the loan, the bank can’t go after any other assets — just the property in question. Remember this was at the heart of the sub-prime mortgage fiasco 10 years ago, which morphed into the global financial crisis. Whilst not wanting to be alarmist, Saul Eslake is concerned with what he’s seeing now in self-managed super funds with their limited recourse borrowing. “You might have thought that someone would have heard the term ‘limited-recourse borrowing’ and recognised that there were some significant risks associated with it that we could have done without in the Australian context.”

And CoreLogic Reported that the combined capital cities returned a final auction clearance rate of 55.3 per cent last week, a slight improvement on the 55 per cent over the week prior when volumes were lower. There were 1,916 homes taken to auction last week, up on the 1,748 held the previous week. While one year ago, a higher 2,258 auctions were held with a 66.9 per cent success rate.

Melbourne returned a final auction clearance rate of 60 per cent this week; an improvement not only over the week but the highest seen since May, with clearance rates for the city remaining within the mid-high 50 per cent range up until this week. The improved clearance rate was across a higher volume of auctions week-on-week, with 891 auctions held, increasing on the 805 held the week prior when 57 per cent sold.

Sydney’s final auction clearance rate came in at 50.6 per cent last week across 656 auctions, falling on the week prior when a 53.8 per cent clearance rate was returned and auction volumes were a similar 664.

As usual the performance across the smaller auction markets was mixed last week, with clearance rates improving in Adelaide, Brisbane and Canberra, while Perth’s final clearance rate fell.

The Gold Coast region was the busiest non-capital city region last week with 56 homes taken to auction, although only 26.4 per cent sold. Geelong was the best performing in terms of clearance rate with 88 per cent of the 34 auctions successful.

And this week, CoreLogic is tracking 1,882 capital city auctions this week. If we compare activity to the same week last year volumes are down 25 per cent on the 2,510 auctions held one year ago.

And finally, APRA released their quarterly property exposure data to June this past week. APRA release their quarterly property exposure lending stats for ADI’s today. There are some interesting data points, and some concerning trends and loosening of standards recently. I will focus on the new loan flows here. First the rise in loans outside serviceability continues to rise, now 6% of major banks are in this category a record, reflecting first tightening of lending standards, but second also their willingness to break their own rules! This should be ringing alarms bells. APRA?

Foreign Banks are writing the greater share (relative percentage) of 80-90% LVR loans. Other lenders tracking lower.

Foreign Banks are lending more 90+ LVR loans in relative percentage terms.

New investor loans are moving a little higher for Credit Unions and Major Banks, suggesting a growth in volumes.

The share of interest only loans dropped below 20% but is now rising a little, as lenders seek to grow their books.

All warning signs, especially when as APRA reports ADIs’ residential term loans to households were $1.62 trillion as at 30 June 2018. This is an increase of $86.6 billion (5.6 per cent) on 30 June 2017. Of these: owner-occupied loans were $1,076.4 billion (66.4 per cent), an increase of $76.7 billion (7.7 per cent) from 30 June 2017; and investor loans were $544.0 billion (33.6 per cent), an increase of $9.9 billion (1.9 per cent) from 30 June 2017. Debt is sky high, the grow rate must be slowed substantially – there are rumours of more tightening to come, we will see.

Looking at the local markets, the ASX 100 was down at the end of the week, ended up at 5,065.90, up 29.8 on the day, and it continues to underperform compared with the US markets. In the banking sector, NAB ended the week at 27.35, after they announced they would not follow the lead of Westpac, CBA and ANZ for now by not lifting their variable mortgage rates, for now. NAB closed up 0.18% on the day. ANZ, who it was announced with be subject to civil proceedings from ASIC for an alleged continuous disclosure breach in relation to a $2.5 billion institutional share placement undertaken by the ANZ in 2015. Their shares rose 0.32% on Friday to 28.15. CBA who took some further knocks this week thanks to further evidence of poor practice in CommInsure in the Banking Royal Commission, among others in the industry. They ended the week at 71.50, and up 0.45% today. And Westpac ended the week at 27.76 up 0.69% on Friday. Despite the relatively benign employment figures out this week, still at 5.3%, the Aussie ended the week at 71.54 and down 0.57% on Friday. The downward trajectory is clearly in play. This risks importing inflation into the local economy.

Looking across to the USA, many investors may be inclined to dismiss yet another headline on global trade and focus on the more granular aspect of the markets. But make no mistake, the markets were gyrating with the twists in the saga between the U.S. and its trading partners. The latest salvo came Friday, when Bloomberg reported that Trump instructed aides the day before to proceed with tariffs on about $200 billion more in Chinese products, but that the announcement has been delayed as the administration considers revisions based on concerns raised in public comments.

Earlier in the week, China had welcomed an invitation by the United States to hold a new round of trade talks. The Trump administration had invited Chinese officials to restart trade talks, the White House’s top economic adviser said on Wednesday. In addition to those tariffs, Trump has said he’s ready to add an additional $267 billion in tariffs “on short notice if I want.”

Earlier in the week, Beijing indicated it will ask the World Trade Organization for permission to impose sanctions on the U.S. as part of a dispute over U.S. dumping duties that China started in 2013.

And there’s still the revamp of NAFTA to consider. The U.S. and Canada have been in talks to bring Canada into a new agreement between the U.S. and Mexico, but there have been on announcements to far. Talks are expected to continue through Monday.

Beyond the US manufacturing sector – for example Boeing is still pretty strong, at 359.80, while Caterpillar ended down 0.44% to 144.90; the potential spill over into the consumer sector impacted a range of stocks, with Whirlpool down 1.68% to 123.21, Walmart down 0.56% to 94.59 and Mattel was up 1.49% to 16.35. Among the financials, Morgan Stanley was at 48.19, a little higher on the day, but still well down on March highs. The S&P 500 ended up 0.03% to 2904.98, as did the Dow Jones Industrial Average to 26,154, while the NASDAQ was down just a little to 8,010.

Apple got the type of promotional attention some companies can only dream of when the eyes of tech lovers and investors alike were glued to its keynote event for details on its new products, especially phones. Apple announced Wednesday its new iPhone product line. Shares of Apple rose the day before the event in anticipation of the kind of surprise announcement for which former CEO Steve Jobs was famous. The stock sold off as details about the new iPhones arrived and shares ended the day lower. But shares bounced back on Thursday, leading the overall tech sector higher, despite negative analyst commentary about the price of the iPhone XR. Apple ended the week down 1.14% to 223.84.

Bucking the recent trend that’s made investors nervous about price pressure, the latest data showed inflation cooling. First, figures showed wholesale prices fell unexpectedly. Producer price index decreased 0.1% last month. In the 12 months through August, the PPI rose 2.8%. Economists had forecast the PPI rising 0.2% last month and increasing 3.2% from a year ago. The core PPI decreased by 0.1% from a month earlier and rose 2.3% in the 12 months through August. Analysts had predicted core PPI to increase 0.2% month on month and 2.7% on an annualized basis.

Next, retail inflation rose less than anticipated. The consumer price index advanced 0.2%, missing expectations for a gain of 0.3%. In the 12 months through July, the CPI increased 2.7%, below forecasts for a reading of 2.8% and down from 2.9% in July. The core CPI increased by 0.1% from a month earlier, below forecasts for a gain of 0.2%. The annual increase in the so-called core CPI was 2.2%. Economists were looking for it to hold steady at July’s 2.4% advance. But despite these softer inflation numbers, traders ended the week still predicting a more-than-80% chance of the Federal Reserve hiking rates at its December meeting on top of the expected boost this month.

Bond yields rose sharply this week, owing to confidence that the Federal Reserve will lift rates for a total four times this year. The rise was particularly strong Friday, when the United States 10-Year yield topped 3% briefly. A big reason for that was Friday’s retail sales numbers.

The August retail sales numbers were disappointing at first blush, rising 0.1%, compared with expectations for 0.4%. But July’s gain was revised up to 0.7% from 0.5%. That revision gave market watchers some more confidence that the U.S. could see GDP growth of 4% in the third quarter, which would all but guarantee another rise in rates in December.

Gold ended the week lower at 1,198, down 0.82%, with preference for the US Dollar as a safe haven. And Copper fell 2.61%, well down on the start of the year, with demand slowing. Oil prices were higher to 69.00, up 0.60% on Friday, reflecting concerns about supply thanks to Hurricane Florence, and trade concerns. Of course, with the lower Aussie, this means fuel prices will rise further ahead.

Finally, Bitcoin is still making lower highs, even though the cryptocurrency has seen slightly higher lows. The key is going to be when bitcoin trades back above $7,000. There is a trend line connecting all the recent highs going back to early 2018. If BTC can bust above that level, it will likely take out the high at $7,350 and make a higher high. Once that happens, institutions may start buying heavily and upside could be back above $10,000 within months. That said, it ended the week down 1.15% to 6,488.

According to Bloomberg, Morgan Stanley plans to offer trading in complex derivatives tied to the largest cryptocurrency, according to a person familiar with the matter, joining other Wall Street firms in creating ways for clients to play the digital currency market. The U.S. bank will deal in contracts that give investors synthetic exposure to the performance of Bitcoin, said the person, who asked not to be identified because the information is private. Investors will be able to go long or short using the so-called price return swaps, and Morgan Stanley will charge a spread for each transaction, the person said. Citigroup is developing a new mechanism for trading cryptocurrencies known as digital asset receipts, a person with knowledge of the plans said earlier this month. Goldman Sachs is exploring derivatives on Bitcoin called non-deliverable forwards, and is considering a plan to offer custody for crypto funds.

Finally, today a couple of quick reminders, first the 60 Minutes programme tomorrow evening and our live stream event on Tuesday at 20:00 Sydney, where you can discuss with me the latest on the outlook for home prices, as well as all our other analysis. You can bookmark the event by using this link. I look forward to your questions in the live chat.

If you value the content we produce please do consider joining our Patreon programme, where you can support our ability to continue to make great content.

Rates Up, Lending Down

Today’s video post discusses the implications of the latest official lending data and announced rate rises.

Investment Lending Slides, But Overall Credit Higher

The RBA released their credit aggregates to June 2018 today. Overall credit grew 0.3% in the month to $2.84 trillion, up $9.7 billion. to a new record.

Within that, owner occupied housing lending rose 0.6% or $6.6 billion to $1.18 trillion, while investment lending fell $800 million, down 0.1% in seasonally adjusted terms, or rose $1 billion, up 0.2% in original terms. (I have no idea what adjustments the RBA makes, its not disclosed!).

Investment lending fell to 33.5% of the portfolio. Total lending for housing is a new record $1.77 trillion, and remember this is at a time when housing debt to income is knocking on the 200 door, and we are one of the most in debt nations on the planet. Least we forget, loans need to be repaid, eventually!

Business lending in seasonal terms rose 0.4%, up $4.1 billion to $921 billion, and fell to 32.2% of all lending – we see a continued fall in the proportion of lending to business, as opposed for housing, which is not good.

Personal credit rose $600 million, up 0.4% in original terms or fell $300 million in seasonally adjusted terms down 0.2%.

The monthly seasonally adjusted numbers highlight the slide in investor lending, and the stronger owner occupied lending.

Finally, we also estimate the growth in on-bank lending, by taking the original RBA data, and comparing this with the ADI data from APRA also out today.

In essence, the relative share of home lending going to the non-banks is rising, to around 7% of all loans, and the bulk of the loans being written are for owner occupied borrowers.

A caveat here, as the non-bank segment of the data will always be a bit off, because there is less timely data captured from this small, but growing part of the market. Something which APRA needs to address.

So more of the old same old, same old, housing lending still growing way above inflation and wages, forcing housing debt higher, at the expense of business investment.

We have not fundamentally addressed the credit elephant in the room. Despite all the noise.

Perhaps the regulators would like to tell us, how much debt is too much? We clearly have not hit their pain threshold yet, despite the rising financial stress in many households.

Mortgage Credit Growth Accelerates In May

APRA has released their monthly banking statistics to end May 2018. After last months drop, we were waiting to see whether the loosening announced by APRA would show up, and yes, this month there was a rise in both the growth of owner occupied and investment lending!

scrollings=”no”

Total portfolio balances rose by 0.38% to $1.63 trillion, which would translate to be a 4.6% annualised growth rate, well above inflation and wages growth if this rate continued. Thus household debt still grows ever larger (a ratio of 188.6 household debt to income according to the RBA, last December), despite being at record and risky levels.

Within that, owner occupied loans rose 0.52% in the month to $1.08 trillion, up $5.5 billion while investment lending rose just 0.13% to $555 billion, up $712 million. Or in annualised terms, owner occupied loans are growing at 6.2% while investment loans are growing at 1.5%. Investment loans now make up 34.06% of all loans, which is still very high but falling.

Turning to the individual lenders, there is little to be seen at the total portfolio level, with CBA leading the owner occupied lending, and Westpac the investment side of the ledger.

However, the individual portfolios within the lenders are more interesting, with Westpac still leading the way in investment lending portfolio growth, alongside Macquarie and NAB. However CBA and ANZ both saw their investor portfolio balances fall, while still expanding their owner occupied portfolios. Bank of Queensland dropped their balances in both owner occupied and investment lending this month. Clearly different strategies are in play.

Later we will get the RBA numbers, and we will see what the total market trends look like. We suspect non-banks will be growing faster than ADIs.

But overall, this appears to show a willingness to continue to let debt run higher to support home prices, so we are still on the same debt exposed path, should interest rates rise further, as is likely, as we discussed recently. Sound of can being kicked down the road once again!

Mortgage Growth Slows Again In April

APRA has released their monthly banking statistics for ADI’s to April 2018. And now we are really seeing the tighter lending standards biting. In fact, Westpac apart, all the majors have reduced their investor property lending.

APRA reports for each lender the net total balance outstanding at the end of each month and by looking at the trends we see the net of loan roll offs and new loans. This is important, as we will see.

At the aggregate level, total lending for mortgages from ADI’s rose 0.2% in the month, up $3.0 billion to $1.62 trillion. Within that owner occupied loans grew by 0.29% or $3.1 billion to $1.07 trillion while investment loans fell slightly, down 0.01% of $42 million. As a result the relative share of investment loans fell to 34.14%.

The trend movement highlights the significance of the fall (the August 2017 point is an outlier thanks to reclassification), this is the weakest result for years. At an aggregate level, over 12 months this would translate to a rise of just 2.3%, significantly down from the 5-6% range of recent months.

Turning to the individual market shares, at the aggregate level there was little change, other than Westpac’s share of investor loans grew to 27.6%, up from 26.1% in January 2016

Here is the trends from the big four. Westpac has continued to increase share of investor loans. CBA has fallen away the most significantly.

There has been almost no shift in owner occupied loan shares.

The monthly changes in value tell the story, with Westpac and Macquarie growing the value of their portfolios, across both owner occupied and investment lending. We see falls in net investor lending elsewhere.

We conclude that Westpac executed a different strategy in terms of mortgage origination compared with its competitors. They may be offering selective discounts to attract particular types of business, or they may have different lending standards, or both.

It is quite possible we will see other lenders trying to compete relative to Westpac in the investment loan domain ahead, as lending growth is needed to sustain profitability. But demand is also falling, so we expect lending momentum to continue to weaken, with the consequential impact on home prices and bank profitability. We are entering credit crunch territory!

Weakening UK Household Finances Pose Risks

The UK household sector’s worsening financial health reduces consumer resilience to income or interest rate shocks and presents risks for UK consumer loan portfolios, Fitch Ratings says in a new report. Consumer credit has been a key driver of rising household debt.

UK households’ swing into an aggregate net financial deficit position over the last year is almost unprecedented, having only previously occurred briefly during the late 1980s. Greater residential investment is a factor, but the household saving ratio has declined steadily since the mid-1990s to an historical low of 4.9% last year. This has been partly due to lower private pensions saving relative to income.

More recently, the UK household debt to income ratio has risen, retracing more than a third of its post-global financial crisis decline (in contrast with the US), and is likely to keep rising given the sector’s financial deficit. This has been largely driven by growing consumer credit, notably car loans.

Low interest rates mean the rise has yet to increase the debt service burden. We calculate that the effective interest rate on all UK household debt fell to 3.3% last year. The fall in the effective interest rate has saved household borrowers GBP20 billion-25 billion in interest payments since 2009.

But weaker household finances reduce the resilience of consumer spending – by far the largest demand component of UK GDP – to shocks. A major interest rate shock appears unlikely (we forecast the UK base rate to rise gradually, to 1.25% by end-2019), but a more immediate shock could come from tightening credit supply. The impact of the Brexit referendum on real wages may be fading, but Brexit uncertainty creates risks of a bigger shock to growth and employment.

Any performance deterioration in UK consumer loans would be from a very strong level. The charge-off rate on bank-financed consumer credit hit a record low of 1.7% in 4Q17. We forecast a modest rise in UK unemployment this year, to 4.7%, implying that charge-offs, which are closely correlated with changes in unemployment, will head back up to 2%-3%.

Fitch-rated consumer ABS deals have substantial headroom to absorb any performance deterioration. For example, the lowest long-term charge-off assumption we currently apply to any prime UK credit card trust is 5%, following the sharp increase in competition in credit card lending since 2013.

UK auto ABS deals have also performed strongly and benefit from structural credit protection, although if unemployment rose and consumer demand fell, both credit risk and residual value risk, which has become more prominent with the rise of the personal contract purchase (PCP) product, could increase.

UK banks are highly exposed to UK households, but mostly through mortgages, with consumer credit accounting for just 10% of banks’ lending to the sector. Recent stress testing by the Bank of England highlighted strong capital buffers against severe consumer credit losses. Nevertheless, high household debt is a constraint in our assessment of UK banks’ operating environment, and currently caps UK domestic banks’ Viability Ratings in the ‘a’ range, all else being equal.

Non-bank financial institutions are more exposed, although their specialist nature means any impact from deteriorating performance would vary across the industry.

Home Lending Slides In March

The ABS released their housing finance data today to March 2018. The trends are pretty clear, lending is slowing, and bearing in mind our thesis that lending and home prices are inextricably linked, this signals further home price falls ahead, which will be exacerbated by even tighter lending standards we think are coming. Debt is still rising faster than inflation or wages.

Starting with the original first time buyer data we see a rise in volumes, reflecting the incentives in NSW and VIC. On a rolling average basis, volumes are strongest in VIC.

Overall volumes were 17.4% of new loans written compared with 17.9% in February, which is a little higher than 2016, but is appears to be drifting lower now, suggesting that ahead volumes of new loans may fall a little.

Looking at the FTB month on month movements, we see a 5.9% uplift in flows which equates to an extra 515 loans last month. The average loan size rose by 2.3% to$335,400.

Looking at our overall first time buyer tracker, we see a fall in overall volumes, as the number of first time buyer investors falls away, as captured in our household surveys.

Turning to the 12 month rolling trend, we see that both owner occupied and investor loan flows are slowing, with investor lending shrinking faster.

The proportion of investor loan flows slid again (excluding refinance) to 43.6%.

Looking in more details at the moving parts, in trend terms, lending for owner occupied construction fell 1.1%, owner occupied purchase of new dwellings fell 0.2%, and purchase of other established dwellings rose 0.2%. So overall, owner occupied lending flows, excluding refinance, were flat at $14.8 billion.

Refinance of existing owner occupied loans rose by 0.5% to $6.4 billion, and on the investor side, construction of new property for investment purposes fell 3.8%, purchase of existing property for investment purposes fell 0.7% and purchase of existing property for investment by other entities (e.g. super funds) fell 0.2%. So overall investment flows fell 0.9% to $11.5 billion.

Finally, the total of all finance trend fell overall 0.2% to 32.7 billion.

And in original terms, the stock of all housing loans rose $6.7 billion, or 0.4% in the month, which equates to 4.9% annualised, still well above inflation or wages growth which is at 2%.

So, the overall growth of lending for housing is still sufficient to lift household debt even higher. So far from losing lending controls, regulators should be seeking to tighten further. There is no justification for growth above inflation or wages, because stronger momentum will only lift the burden on households even further, and inflate the banks’ balance sheets, which flatters their performance.