FNB Home Loans household and property sector strategist John Loos has just released his latest Property Barometer report regarding Residential Market Macro Stability and Risk, with comments on some of the key themes of the report.
A tougher year for residential mortgage lending ahead, as the residential property market slowly moves into its super-cycle “correction”
More challenging times are at hand for the residential mortgage lending sector. As we look set to experience the fifth consecutive year of stagnating economic growth in South Africa, and possibly the third year of interest rate hiking, we expect the volume and value of bonded residential property transactions to decline for 2016 as a whole.
Everything hangs together, and after some years of decline in the prices of South Africa’s export commodities, on the back of slow global economic growth, gradually rising interest rates, and deteriorating net foreign investment capital flows leading to a multi-year rand slide, it is unrealistic to expect the mortgage lending sector of the economy to defy economic gravity.
We believe that the residential market is entering its super-cycle “correction” phase, caused by a stagnating economy whose policy makers have little left to give in the way of fiscal or monetary stimulus measures. And of course, corrections can hold significant risks for mortgage lenders, depending on how sharp these corrections are and what that does to home values.
After our estimates of two prior years of slowing but positive growth in the value of bonded property transactions by individuals (or “Natural Persons” as the Deeds Office calls them), we expect 2016 to see growth in both the volume and value of this category of property transactions finally turning negative. It is always tough to obtain a data series that reflects household sector mortgage lending trends for the entire markets. As a proxy, we use Deeds data to estimate bonded property registrations by individuals (or “Natural Persons”), and we project a -4% decline in the estimated value of such bonded property registrations to individuals in 2016, after estimated moderate positive growth last year.
This expectation is on the back of our forecast of real economic growth slowing further, from 1.3% in 2015 to 0.5% in 2016, along with a further 50 basis points’ worth of interest rate hiking this year. In the subsequent years of 2017 and 2018, economic growth is only expected to improve marginally to just above 1%, while prime rate is expected to peak at 11% early next year and then move sideways.
Under this weak growth economic scenario, which implies a lack of employment creation and pedestrian household income growth, we project average house price growth to lose further steam, slowing from 6% average in 2015 to 4.8% in 2016, and then lower to near 3% in the outer forecast years of 2017 and 2018. While this low single-digit house price growth will imply a decline in real terms, with CPI inflation forecast significantly higher, such a gradual real price correction would not be too bad a situation for mortgage lenders, with widespread “negative equity” (i.e. where homeowners owe more on their homes than they are worth) avoided.
But there is little room for manoeuvre, and the “downside risks” to economic growth forecasts in a troubled global economic environment make the risks of recession in South Africa high. While interest rate “spikes” have traditionally been the key concern of mortgage lenders, our main concern is more the economic stagnation, and the high risk of recession which may not necessarily be a high interest rate scenario.
Our most obvious risk scenario model simulation is not one where interest rates shoot up sharply. Rather, it is one in which we assume a world that goes into recession, taking South Africa with it. Interest rates in this scenario may even peak lower than our Base Case peak of 11% Prime. However, such a recession would exert additional downward pressure on employment and real disposable income growth of households, and we believe that should such a risk scenario play out, it would likely lead to average house price deflation, a bad situation for mortgage lenders and borrowers alike, as it makes it challenging to “trade out” of properties without losses should a household be under financial pressure.
Whether we face recession in the forecast years, or just very slow positive growth as in our “Base Case”, we believe that the residential super-cycle correction is likely to be more of a “slow puncture” than the short-lived but sharp and painful correction of 2008/9, where a global oil and food price spike took interest rates up more rapidly than the current interest rate hiking cycle.
In addition to slower mortgage lending, we expect some deterioration in residential mortgage credit health in the form of some increase in mortgage loan arrears in the industry. Basing our forecast on NCR (National Credit regulator) data for the industry as a whole, we project residential mortgage arrears of “90 days and longer” to rise slightly to average 3.7% of total value of loans outstanding in 2016, up from around 3.5% in 2015, and gradually to rise, averaging further to 4.2% of total loans in 2017 and 4.8% in 2018. However, this rise would be mild compared to the 6.2% peak of 2010, as it should be under a more moderate interest rate hiking scenario, and given that the mortgage lending sector has had higher lending standards since the end of the pre-2008 boom period.
In short, it appears that tougher times are ahead for the mortgage lending sector, but at this stage it doesn’t look like being an extreme shock. The projection of more moderate deterioration compared to 2008/9 has much to do with the far more gradual way in which the SARB is hiking interest rates this time around. But it also has to do with lower levels of household sector indebtedness, and thus vulnerability, these days, with higher quality lending by the mortgage lending sector since the end of the pre-2008 boom having been a key contributor to lowering the Household Debt-to-Disposable Income Ratio by 11 percentage points since 2008.
But there are never any guarantees that it will remain a “soft landing” scenario for the sector. Sharp outflows of capital, causing big rand depreciations, can force the SARB to speed up rate hiking should it occur. It is thus important that lenders and borrowers alike keep doing the right thing, therefore, and that we continue to lower household sector vulnerability by further lowering the still-high Household Debt-to-Disposable Income Ratio in the next few years.
Modern SARB monetary policy approach crucial in “landing” the residential market and mortgage sector softly
Since the end of the residential market boom in 2008, the positive role of the South African Reserve Bank’s interest rate approach in promoting residential and mortgage market stability can’t be underestimated. Now, as tougher economic times and in our expectation the residential super-cycle correction arrive, its role will be even more crucial, hopefully in engineering a “soft landing”, helping the economy to avoid recession as far as possible.
So far so good. The SARB’s positive contribution goes back a number of years to around 2008. Its aggressive interest rate cutting, from late-2008, to where prime rate entered single-digit territory for the first time in decades, provided relief for households, but always remained above average house price inflation, thus preventing the residential market from becoming another crazy “speculators paradise” where speculators could buy and sell properties in short succession using cheap credit to make quick capital gains.
We believe that the residential market looked like starting to “heat up” late in 2013, which could have attracted a greater level of “over-exuberant” behaviour from various market players, but the SARB nipped this in the bud by “timeously” starting to gradually lift interest rates early in 2014. This hiking couldn’t have come soon enough, in order to sustain a “rational” market. But at the same time, the SARB has hiked at a gradual pace, thus not causing undue harm. In addition, the interest rate approach has been just sufficient to contain household sector credit growth to rates where we have made significant progress in lowering the all-important Household Debt-to-Disposable Income Ratio from an all-time high of 88.8% early in 2008 to 77.8% by late-2015, thereby significantly reducing the household sector’s vulnerability to economic shocks.
With CPI inflation the SARB’s main target, and not the housing market, not all of this may have been planned, but it has worked out well nevertheless, and most important in times of high household indebtedness is the bank’s gradual approach to hiking so as to allow ample time for the household sector and its lenders to adjust.
In fact, this gradualist approach of the modern day SARB is to me a key difference between now and back in the early-to-mid 1980s when the last big residential super-cycle correction happened. The big macroeconomic numbers today are looking about as bad as the early-80s after the Gold Boom, including sliding investor confidence and capital flows, a severe rand depreciation, a huge current account deficit on the balance of payments that needs to be funded but can’t be, and stagnating economic growth. But back then, the SARB hiked interest rates far more rapidly and aggressively to curb the deficit on the current account (i.e. get the country back living within its means), and brought the property “house of cards” tumbling down in the process. Indeed, the central bank needs to address large current account imbalances, so we are in favour of higher interest rates currently, but the speed at which the Bank does this can make or break the economy and the property market.
So the SARB is key to the possibility of the soft landing for the economy and housing market, but admittedly not everything is in the SARB’s hands. The Bank can do nothing about the global economy, while many of the structural constraints in the SA economy also lie outside its sphere of influence. So the soft landing can never be guaranteed.
And even in our main forecast scenario of such a “soft landing”, where house prices are projected to decline in real but not in nominal terms, the expectation remains for a “landing”. There is little to suggest a major change from currently stagnant economic growth rates, so we believe that real house prices have to decline to better reflect the weaker economic fundamentals of recent years, while mortgage lending should slow somewhat too in the near term.
In short, the SARB is playing a crucial part with its interest rate policy. In addition, the mortgage lending sector has played a positive part in terms of better quality lending in recent years since 2008, lowering the Household Mortgage Debt-to-Disposable Income Ratio from a high of 49.2% early in 2008 to 34.8% by the end of 2016, thereby lowering its vulnerability to “shocks”.
But many of the key risks to the residential property and mortgage lending market perhaps emanate more from outside the residential market or the areas of SARB influence. They emanate from structural constraints in our own economy, including a highly unequal skills distribution which sustains a highly unequal employment and income distribution, as well as from the weakness of the global economy.
SARB monetary policy thus has limits to the extent of its influence, but so far so good.
We see financial constraints mounting, but we haven’t yet seen noticeable signs of increasing financial stress
It is important to stress that residential market risk does not always equate to actual market performance at a given time. The two concepts are different things, although rising market risk does often eventually manifest itself in a deterioration in performance. And of course there are natural “leads and lags”.
So, while our Composite Household Sector, Residential Market and Economic Risk rating has been rising since 2012, pointing to mounting risks for the residential market, to date we have not seen clear evidence in the Mortgage Sector of rising financial stress, only mounting financial constraints to date.
A key indicator of mounting financial constraints is that of Real Household Sector Disposable Income growth, which has slowed from a post-recession high of 4.6% in 2011, to a lowly 1.4% in 2015. We expect further slowdown, possibly into negative territory in 2016, as a result of the combination of further slowing in economic growth along with rising consumer price inflation this year.
The result has been recent signs in the Residential Market that the Household Sector has begun to “cut its coat according to its cloth. This, we have witnessed in the form of the start of a decline in transaction volumes at the higher priced end of the Major Metro market during 2015, whereas what we call the Lower Middle Income and Affordable Segments, with average price levels below R1 million, saw volumes still growing positively through last year.
With regard to bonded transactions by individuals (or “Natural Persons”), which relate to the world of mortgage lending, we have seen the growth rate in the average bonded transaction value slow from an estimated high of 12.3% year-on-year for the 32 months to February 2013, to 3.8% for the 3 months to January 2016
When it comes to “financial stress” or “credit health”, however, we have only got as far as seeing no further noticeable improvement (decline in stress levels). However, we have not yet seen clear indications of a rise in financial stress levels.
Through 2015 we saw more or less sideways movement in the high level of “Current” Mortgage Accounts, i.e. those accounts that are in good standing, which remain at high levels in excess of 91% of all mortgage accounts, based on NCR (National Credit Regulator) data. The 3rd quarter 2015 percentage of 91.4% of total accounts being in good standing, on a value basis, was virtually unchanged from the 91.5% as at the end of 2014.
A second indicator of financial stress that we monitor emanates from the FNB Estate Agent Survey. This relates to the estimated percentage of sellers that are selling in order to downscale due to financial pressure. Once again, we see no further sign of improvement (decline). At 14% of total home selling, selling in order to downscale due to financial pressure remains around the lows of the past two years, after a prior multi-year decline. This does not yet represent a meaningful increase, but also points to no further improvement in recent years.
One key contributing factor to seeing little sign of deterioration in the financial stress situation to date, despite two years of interest rate hiking, is believed to be the very slow pace at which the Reserve Bank (SARB) has been hiking rates, i.e. 1.75 percentage points in two years. This does give households a long time to gradually adapt their expenditure and borrowing patterns, which is a key positive regarding the way the SARB operates these days.
However, we don’t believe that the Household Sector can escape interest rate hiking totally unscathed. Often, there can be quite a lengthy lag time between the start of a rise in interest rates, which normally raises the Household Debt-Service Ratio (the cost of servicing the Household Sector debt burden expressed as a ratio of Disposable Income), and the start of a rise in bad mortgage debt.
Therefore, as a result of two years of rise in the Debt-Service Ratio to date, and further interest rate hiking expected this year, we project an increase in industry-wide Mortgage Loan arrears in 2016. From a level of around 3.5% of the value of total Household Sector Mortgage Loans Outstanding, the value of loans longer than 90 days in arrears is projected to rise gradually in the forecast period, to 3.7% for 2016, 4.2% in 2017 and 4.8% for 2018.
However, this rise would not be a severe one, at least not by 2009-10 standards where the percentage peaked at 9.2% in 2010. The forecast for a far more mild rise in arrears would be the positive outcome of a reduction in the level of Household Sector indebtedness relative to income in recent years, as well as a more mild interest rate hiking cycle compared with 2008.