It is important to understand that, due to significant resistance by home sellers to house price declines, in times of economic and residential demand slowdown the residential market can often move away from market equilibrium price.
Average time of homes on the market rises
Therefore, the average market house price level, as depicted by a house price index, is not necessarily the “market equilibrium” price level. Often, in times of market weakness, part of the weakness is reflected in the average transaction price, or its inflation rate, and part of it will be reflected in a longer average time that it takes to sell a home.
The key question around the estimated average time of properties on the market is what would be the average time on the market that reflects market equilibrium? The answer to this is subjective, but FNB’s view is that the level is around three months average time on the market. From 2014 to early-2016, the estimated average time had been moving broadly sideways at levels around twelve weeks, i.e. slightly less than three months, and this was a time with very mild average house price growth in real terms (zero average house price growth in real terms theoretically reflecting a well balanced market)
Through 2016 and into 2017, the market appears to have been moving broadly away from that equilibrium. In the FNB Estate Agent Survey for the first quarter of 2017, FNB did see a decline in the average time of homes on the market, after three prior quarters of increase. This, FNB thought, may be the start of some very mild recovery in the residential market, with two consecutive quarters of rise in the FNB Residential Activity Rating also being witnessed up to and including the first quarter of this year.
However, it was not yet to be, and the second quarter saw both the Residential Activity Rating recede, as well as the average time of homes on the market resume its rise from thirteen weeks and four days in the first quarter to fifteen weeks and four days in the second quarter.
A greater percentage of sellers is required to drop their asking price
A second question related to price realism is where FNB asks the agents to estimate the percentage of sellers ultimately being required to drop their asking price to make the sale.
While the overwhelming majority of sellers do tend to start high and allow themselves to be bargained down as a strategy, FNB nevertheless finds this estimated percentage of sellers having to drop their asking price to have crept up. From a multi-year low of 78% in the second quarter of 2014, the market weakening since then has brought about a mild upward trend to 92% of all sellers dropping their asking price in the second quarter of 2017, according to the respondents’ estimates.
But interestingly, the estimated magnitude of the average price drop continues to diminish
Interestingly, though, the estimated magnitude of decline, for those being required to drop their asking price, has broadly declined, from -8% for much of 2015 to -6.8% by the second quarter of 2017. So more are dropping their price but by a lesser magnitude.
The shift away from equilibrium/deteriorating price realism is a delayed response to weaker demand.
The noticeable increase in the national average time on the market since early last year was arguably the lagged response to a slowing in housing demand through 2014 and 2015. During those two years FNB saw interest rates rise, economic growth continue to slow, and housing demand slow as a result.
A key residential “demand-side” question that is asked to the survey respondents, in the FNB Estate Agent Survey, is to give an estimate of how many serious viewers per show house they get before making the sale.
From a multi-year high average of 14.42 estimated serious viewers per show house for the four-quarters of 2013, FNB saw a noticeable decline to 10.66 average for the four quarters of 2015. Thereafter, the broad movement has been more-or-less sideways up to the present, averaging 10.87 viewers for the four quarters up to and including the second quarter of 2017.
While according to this indicator, demand no longer appears to be declining, the average number of serious viewers since early 2016 has moved at a lower level than in prior years, and this is seemingly at a level where demand is not strong enough to mop up available residential supply.
Stock constraints diminish
It is difficult to gauge the strength of supply of residential stock through asking survey respondents for their opinion. But when asking agents about their market expectations in the near term, FNB allows them to provide a list of factors that influence their expectations, both in a positive and a negative way
After the percentage of agents citing “stock constraints” as a key factor had intensified noticeably from 2012 to early-2014, assisted by relatively low levels of residential building activity since the end of the building boom in 2008, they began to diminish through 2015, as one would expect in most slowing demand environments. In the second quarter of 2017, FNB saw a decline from the previous quarter’s 12% of agents to 8%, after two prior quarters of increase, and the percentage citing stock constraints is now far below the 24% high of early 2015.
The percentage of agents citing “Ample Stock” equaled those citing “Stock Constraints”, i.e. also 8%, after having generally been below the “Stock Constraint” estimate in recent years.
In short, agents have perceived stock constraints to have eased significantly since 2015.
Namibia, the weak link in the Rand Area. Gauteng, the strong point in terms of realism and balance.
When breaking down the key indicators of price realism into the major survey regions, a key feature is the weakness in Namibia relative to South Africa. The once booming Namibian market is no longer. Whereas South Africa’s estimated average time on the market was fifteen weeks and four days in the second quarter of 2017, Namibia’s average time had risen to as high as twenty-three weeks and six days.
To boost survey sample size when breaking down the survey into regions (to reduce volatility), we resort to a second quarter moving average.
For the first two quarters of 2017, the average estimated time of homes on the market in South Africa was 14.57 weeks. By comparison, Namibia averaged a lengthy 22.36 weeks. The income levels of Namibia can no longer sustain the high price levels following that country’s meteoric rise in house prices over a decade-and-a-half.
Gauteng, the strong point in terms of realism and balance.
Within South Africa, the shift away from market equilibrium, or towards less realistic pricing, has taken place largely in the country’s Coastal Metros, where FNB has seen the average time on the market rise to twenty weeks and three days in the second quarter of 2017.
Notably, the recently very strong City of Cape Town has also seen its average time on the market rise noticeably, from thirteen weeks and six days in the previous quarter to eighteen weeks and four days in the second quarter of 2017.
The Gauteng region, on the other hand, has become the “solid” region, and appears to have improved in health, averaging twelve weeks time on the market in the second quarter. This is right on our subjective “market equilibrium” .
Using second quarter moving averages to boost sample size at city level, the tjree coastal cities (Cape Town, Ethekwini and Nelson Mandela Bay) are noticeably weaker, all three averaging above sixteen weeks time on the market for the first half of 2017.
By comparison, Joburg has averaged a heathier 12.86 weeks and Tshwane Metro an impressive 9.57 weeks for the same two quarters
In the City of Cape Town, FNB believes that the recent rise in average time on the market has much to do with mounting home affordability challenges after a run of very strong house price inflation in that region in recent years.
However, just because Joburg and Tshwane have a far lower average time on the market does not mean that they are markets with strong demand and high levels of seller pricing power. To the contrary, these two cities still have relatively high estimates for the percentage of sellers being required to drop their asking price to make the sale. Joburg’s two-quarter average was 89% of total sellers being required to drop the price, while Tshwane’s average was 95%, in the first two quarters of 2017.
Far from having strong housing demand, therefore, it would seem that after a few recent years of weakness in demand in these markets their sellers are more ready to accept a price drop.
Interestingly, though, is that while Cape Town’s average time on the market has risen, suggesting a weakening in the level of demand relative to supply at the high prevailing price levels, still 29% of that city’s agents perceived “stock constraints” as an issue in their lives in the first two quarters of 2017. This is far higher than any other city, whereas only 4% of agents in Joburg and zero in Tshwane reported stock constraints.
However, FNB does expect that this percentage will decline noticeably in the near term in Cape Town, given that homes appear to be staying on the market for longer of late.
The lower end is the “hot spot” in both South Africa and Namibia
Examining the average time of homes on the market by Income Area segment, both South Africa and Namibia showed very wide differentials between the Lower Income Area end and the High Net Worth end, with the Lower Income end having far shorter average times on the market.
While high end homes do generally take longer to sell, we believe that the wide differential, i.e. twenty-four weeks in the case of the High Net Worth segment and eight weeks in the Lower Income segment in the case of South Africa, also in part reflects a stronger lower end of the market.
On a national average basis, a resumption of the rise in average time of homes on the market prior to sales points to a housing market moving away from equilibrium, and into a space where supply exceeds demand.
However, a look at the major regional breakdown points to two very different housing markets. Gauteng appears very much in equilibrium, averaging twelve weeks time on the market, an average which has declined slightly.
By comparison, it is the three Coastal metros whose average times on the market have risen markedly, and who take the national average higher.
Since the FNB Estate Agent Survey has been started, it has been an interesting feature of the housing market that the Gauteng market seems to be a more realistic one in general, more swiftly moving back to “equilibrium” than the major coastal city markets.
This was shown in 2011/12. The market was trying to recover from the 2008/9 recession (and widespread financial stress) at the time. Gauteng started its broad shortening in the average time on the market from mid-2011 onward, whereas the Coastal Metros only saw their average time peak as late as the second quarter of 2012 at twenty-one weeks and three days, and only starting the declining trend there-after.
This time around, it appears to be the same, Gauteng having seemingly started its recovery with a decline in average time on the market in the past 2 quarters, while the Coastal regions’ average time is still on a rising path.
The explanation for this greater realism and adaptability in Gauteng is not yet understood.