2017 was the third consecutive year of weakening in the South African residential market, which was always anticipated after a multi-year economic performance stagnation dating back to around 2012, and amid weak business and consumer confidence as concerns surrounding future political and economic policy direction mounted.
2018, being the year of the "run up" to the 2019 general election, could conceivably be another year of policy uncertainty. But some leading economic indicators do at least point to some near-term improvement in the economy, while a change of leadership in South Africa’s ruling party in December appears to have boosted sentiment, if the rand’s subsequent strengthening is anything to go by.
And this is why we would expect a slightly better residential market performance to emerge in 2018.
When one talks of weakening or strengthening in a housing market, it refers not to the strength of housing demand, but rather to the balance between housing demand and housing supply.
One indicator of housing demand that we utilise comes from the FNB Estate Agent Survey, where agents are asked to estimate the average number of “serious” viewers per show house prior to the home being sold.
Their average estimate was 10.61 viewers per show house in the final quarter of 2017, marginally higher than the 10.29 estimate of the prior quarter. Estimates in recent years have moved broadly sideways since 2016 after a prior noticeable decline from a multi-year high of 16.69 viewers reached at a stage of 2013.
Therefore, estate agents did not implicitly point to further weakening in housing demand through 2017.
However, they did point to further weakening in the market, with the prevailing levels of demand remaining insufficient to “mop up” the existing levels of housing supply on the market through 2016 and 2017. The result was a further weakening in the housing market’s balance between supply and demand through 2017.
This was reflected in a further increase in the estimated average time of homes on the market prior to sale to 17 weeks and two days as at the fourth quarter of 2017. This continued a broad rising trend which began early in 2016 from a low of 11 weeks and one day at the beginning of that year.
In addition to a rising average time on the market, an increasing percentage of sellers has ultimately been required to drop their asking price to make the sale. At the end of 2017, this estimate stood at 95% of total sellers being required to drop their asking price, up from a previous 93%, continuing a broad rising trend from 78% as at the second quarter of 2014.
Furthermore, the estimated average drop in asking price has increased in magnitude recently, averaging around -10% for the final two quarters of 2017, after a lesser estimate nearer to -7% in the first two quarters of last year.
In the survey, we also ask respondents for their expectations of market activity levels over the next three months, followed by an open-ended question as to why they expect what they do. The various reasons for their expectations are insightful and, since 2015, a diminished percentage have reported “available stock constraints” as an issue in their lives, from a high of 24% of agents at the start of 2015 to a mere 2% by the final quarter of 2017. This response adds support to the opinion that housing supply did not keep pace with demand.
In terms of housing market “activity”, a term that, for agents, incorporates both buyer interest and activity as well as seller interest and activity, we saw further weakening late in 2017, according to the FNB Estate Agent Survey.
We ask survey respondents for a rating, on a scale of one to 10, of the levels of activity that they experience, 10 being the highest level. The final quarter of 2017 saw a further decline in the activity rating to 5.29, from 5.42 in the previous quarter, and down from a 13-year high of 6.76 reached in early-2014 to the lowest level since the second quarter of 2009.
Despite agents pointing to actual market weakening in 2017, their near term “forward-looking” expectations of activity in the market were on average better than in 2016, albeit not overly strong.
We ask the survey respondents the question as to what they expect activity levels to do in the three months subsequent to the survey, the response being either “increase”, "remain the same”, or “decrease".
We take the responses, assign a +1 to an “increase” response, a zero to an unchanged response, and a -1 to a “decrease” response. We then roll these up into what we term our “Home Buying Confidence Index”, which is on a scale of +1 to -1, a +1 reading implying 100% of respondents expecting an increase in activity and a -1 implying 100% expecting a decrease.
Because seasonal factors play such a large role in expectations from quarter to quarter, we express this index on a four-quarter moving average basis in order to eliminate such seasonal factors and focus on the non-seasonal market issues.
At +0.22 in the final quarter of 2017, the reading was slightly strengthened from the +0.19 reading in the prior quarter. It was also noticeably higher than a +0.13 low reached at the end of 2016, indicating mildly improved expectations through 2017 compared with 2016, despite a weaker actual situation being recorded on average in 2017.
Insightful, though, is to examine the results of an open-ended question in which we ask agents for the reasons as to why they expect what they do for near-term activity levels.
Seasonal factors aside, in the fourth quarter of 2017 survey, 37% of agents cited “economic stress/general pessimism”, while a lowly 6% cited the opposite, i.e. “positive consumer sentiment”.
And examining the fourth-quarter moving averages for percentage of respondents citing the abovementioned two factors, we have seen a steady rise in the “economic stress/general pessimism” percentage to 36.8% for the four quarters to the final quarter of 2017, from only 7.2% back at the start of 2015, while “positive consumer sentiment stood at only 6.7% for the fourth quarter of 2017.
Finally, we examine agents’ average house price growth expectations for the 12 months following the survey date. In the fourth quarter 2017 survey, the average expected price increase over the following 12 months was 3.9%. This represents the second consecutive quarter of slowing in average expectations, from 4.5% as at the second quarter of 2017.
In short, agent expectations were modest through 2017, and similar to what was reflected in national business confidence and consumer confidence indices. The agents also clearly felt the national “negativity” in their day-to-day dealings in the housing market in 2017.
We believe, however, that the national mood has changed significantly since that fourth quarter 2017 agent survey, and that come the first quarter 2018 survey (to take place in February), we may see a slightly better market picture emerge.
We say this because, subsequent to the survey, we had a ruling party conference in December at which there was a change in leadership. While it remains to be seen what such a change will mean policy-wise for South Africa, if the strongly positive reaction of the rand to this news is anything to go by, the national mood appears to have improved on the news.
In addition, since around mid-2017, the SARB’s Leading Business Cycle Indicator, and the OECD’s South Africa version, have both been rising noticeably, and these are often useful indicators of near-term economic performance.
An improved economic growth rate, accompanied by our expectation of stable interest rates, should be supportive of mildly stronger household sector disposable income growth, which in turn should provide some additional support to housing demand.
Our own FNB House Price Index may have also begun to show early signs of market strengthening very late in 2018. When we view it on a month-on-month basis, after adjusting for seasonal factors, we see that after a lull in price growth in mid-2017, there was a late “pick-up” of sorts in December. It is too early to tell as to how sustainable such a month-on-month lift is.
However, it is more the rise in the Leading Business Cycle Indicator, which in part reflects a reasonably solid global economic performance, and a perceived sentiment improvement in South Africa early in 2018 (as reflected in a stronger performance in the rand in recent times), which lead us to expect that, despite agents not yet reporting actual improvement in the market on a national average basis, 2018 should bring about a slightly stronger housing market.
