For the first time since the first quarter of 2014, the average gross yield from residential rental property rose to 8.62 percent in the second quarter of this year, from a revised level of 8.59 percent in the first quarter. This is according to the latest residential yields review by First National Bank (FNB) and TPN, a credit bureau that specialises in residential letting.

The gross yield excludes a buy-to-let investor’s operating costs, which have to be factored into the calculation to arrive at a net yield. Rode & Associates have suggested that about 1.5 percentage points could be subtracted from the gross yield to estimate a net yield, the review says. If this were done, the national average net yield would be about 7.1 percent. This is “not an overly attractive yield”, in light of the recent low capital growth on property and because the prime rate of 10.5 percent translates into an average home loan rate of 11 percent, the review says.

The average national yield also obscures the wide differences in yields by metropolitan area and type of property (see below).

The review says there are signs that the macro-economic environment may be turning in favour of renting, and against homeownership, which bodes well for a further rise in yields for buy-to-let investors.

The combination of higher interest rates and economic uncertainty seems to have put the brakes on first-time home-buying, the review says.

The FNB Estate Agent Survey estimates that first-time home-buying has slowed to 18 percent of total buying, down from 21 percent in the first quarter, and below the multi-year high of 28 percent in the second quarter of 2014.

When aspirant first-time buyers hold off on purchasing, many remain in the rental market for longer, driving demand for rental property. This can contribute to stronger rental growth and slower property-price growth, the review says.

Data from Statistics SA shows a gradual acceleration in year-on-year rental growth since the middle of 2012, from 4.26 percent to 5.3 percent in mid-2016.

“This is not a strong acceleration, but, with average house-price inflation recently slowing, the combination may be just sufficient to start the ball rolling in terms of a national average yield increase and, ultimately, a more attractive buy-to-let opportunity.”

The FNB Estate Agent Survey found that buy-to-let buying, when expressed as a percentage of total home-buying, increased to 10.1 percent in the third quarter, from 7.6 percent in the second quarter. The review cautions that quarter-on-quarter fluctuations could be attributed to “volatility”, not the start of a significant trend.


Yields by region

When it comparing yields by major metropolitan area, the review says there is some truth in the principle that lower risk translates into lower return.

Yields in the second quarter were lowest in Cape Town (7.71 percent), which had the most tenants in good standing (89.52 percent), and Nelson Mandela Bay (8.47 percent), with the third-highest percentage of tenants in good-standing (88.01 percent).

Johannesburg recorded the highest average yield (9.51 percent) in the second quarter. In Gauteng, the percentage of tenants in good standing was 84.98 percent, slightly below the national average of 85.08 percent.

The average yield was 8.94 percent in Ekurhuleni and 8.79 percent in Tshwane.

Ethekwini had the second-highest average yield (9.19 percent). The good-standing percentage in KwaZulu-Natal was a below-average 83.98 percent.


Yields by rental band

The review segments suburbs into five value bands based on the average value of the properties in them. The five value bands are:

* Lower-income areas: the average home value is below R600 000;

* Lower-middle-income areas: between R600 000 and R900 000;

* Middle-income areas: between R900 000 and R1.2 million;

* Upper-middle-income areas: between R1.2 million and R1.5 million; and

* Upper-income areas: the average home value is higher than R1.5 million.

In the second quarter, the median gross yields in each band were:

* Lower income: 9.43 percent (8.83 percent in the first quarter);

* Lower middle income: 9.07 percent (8.73 percent);

* Middle income: 8.49 percent (8.13 percent);

* Upper middle income: 7.84 percent (7.66 percent); and

* Upper income: 6.58 percent (6.13 percent).

The review says gross yields are not the only factor that buy-to-let investors should consider; they should also look at the risk that tenants may not pay their rent.

It is likely that properties in the lower-income band are being rented out for less than R3 000 a month. The percentage of tenants in this rental band who are in good standing (79.75 percent) is below the national average of 85.08 percent. The high yield is compensating investors for this high risk, the review says.

It says that for investors who want a more reasonable trade-off between risk and return, the “sweet spot” remains the lower-income (R3 000 to R7 000 rent) and middle-income (R7 000 to R12 000) bands, where the good standing percentages are 86.44 and 88.56 percent respectively.

In the upper-middle-income band, where rents are likely to be between R12 000 and R25 000 a month, 84.76 percent of tenants are in good standing.

The band where rents are R25 000 a month or more remains the worst performer, with 78.11 percent of tenants in good standing.


Yields by title deed and number of rooms

Yields on sectional title properties continue to out-perform full-title properties, the review says.

Two-bedroom sectional properties performed particularly well, with a gross yield of 9.28 percent (9.11 percent in the first quarter).

Although one might expect that one-bedroom properties, which have the lowest average value, would produce the highest average yield, this segment has a had strong run in terms of demand and price growth, with the result that yields are lower than those for sectional title properties with two or more bedrooms, the review says.

The yield for one-bedroom properties was 8.8 percent in the second quarter (8.86 percent in the first quarter), while the yield for sectional title properties with two or more bedrooms was 9.01 percent (8.86 percent).

With regard to the yields on full-title properties, “smaller is better”. Houses with fewer than three bedrooms had the highest gross average yield of 8.05 percent (unchanged from the first quarter); followed by three-bedroom houses, with 8.38 percent (7.84 percent) and houses with more than three bedrooms, with a yield of 7.55 percent (7.13 percent).



Good standing: TPN classifies tenants as being in good standing if they pay the full monthly rent by the due date, or within the grace period, or late. Tenants are not in good standing if they pay only in part or not at all.


Yield: The annual return on an investment, expressed as a percentage of the capital value. The initial yield is a straightforward calculation in which in the annual rental is divided by the capital value of the property, expressed as a percentage. For example, if the property cost R1 million and the annual rental is R60 000 (or R5 000 a month), the initial yield is six percent. This is the gross yield; it ignores the expenses associated with owning and renting out the property. These expenses must be subtracted from the annual rental to calculate the net yield.



The First National Bank (FNB)-TPN Residential Yield dataset is the combined result of TPN rental data, FNB’s house price data and its automated valuation model (AVM). The methodology takes all of the properties for which TPN rental data exists, uses the AVM to estimate a current value on the property, and then calculates the gross initial yield on all such properties.

Rentals seem to vary far more from the mean than do house prices (possibly because of the absence of professional valuer guidance in the rental market), so the survey uses median yields, not average yields, for rental segments.

The national average yield is a combination of mean and median. The median yields are compiled for the area value bands in the major rental regions (the six major metropolitan areas and “the rest of South Africa”). The median yields of the regional segments are then rolled up into regional and national weighted averages based on weightings determined by the rental volumes in the segments and regions.