The Kagiso Active Quants Fund is the only South African fund available to the retail investor that bases active stock picking decisions on a statistical model of the cross-section of returns. The model is based on a large body of original academic research.
In this quarter a return of 20.1% was earned by the fund. It outperformed its benchmark (the mean of the other fund managers in the general equity sector) by 5.5%. It outperformed its nearest competitor in the general equity sector by more than 2%. Since inception on 22 April 2004, it has outperformed by 7.73%. It has an annualised total risk (standard deviation) of about 12% which is low compared to the ALSI's 22%.
At the start of the quarter, the fund was structured with a 6.5% tracking error with respect to its benchmark. It started off well and by the end of July had added 350 points of outperformance. However, during August there was an unexpected rate cut. This was a surprise as only one earlier the governor had stated that his policy was the reverse. As a result, the rand weakened and gold shares, in particular, jumped sharply.
While the total value of the portfolio during August was up about 7%, it lagged the benchmark by a full 120 points. In the second half of August an important reassessment of the portfolio's positioning was conducted and the portfolio was repositioned with a 5.5% tracking error (which remains its current level of active risk at present). During September the lost ground was recovered and a further 200 points of outperformance was added beyond that already achieved in prior months.
Due to the deliberately lower beta of the Kagiso Active Quants Fund (which is less than half that of its benchmark) it tends to find it more difficult to outperform when the market does well, as in the past quarter. Its good relative performance over this difficult period augurs well for the future. Primary risks include: remaining underweight in rand hedge and large capitalisation stocks.

The fund commenced trading on 26 April 2004, becoming the first South African fund available to the retail investor that bases active stock-picking decisions on a statistical model of the cross-section of returns. The model is based on a large body of robust academic research.
The fund was initially structured with a 3% tracking error with respect to its benchmark (the mean holdings of the other fund managers in the general equity sector). This was planned as an interim step to gain a 'cushion' of outperformance before ratcheting up the tracking error to more aggressive levels.
Including the significant initial trading costs, the fund had outperformed its benchmark by 1.75% by the end of May. With regard to offsetting transaction costs, we were helped by doing the bulk of our buying on a down day where our cash-levels were higher than that of the benchmark. This degree of outperformance was deemed sufficient to proceed to stage two.
In early June an additional R2 million was added to the fund. This resulted in a large proportion of trade as these funds were invested. In the rebalancing that occurred, the tracking error and forecast alpha of the fund both more than doubled and the fund was structured to have a tracking error of about 6.5%. This is approximately the same amount of tracking error as exhibited by the average active fund in the general equity sector. More than 60 basis points were lost in the rebalancing through transaction costs and (primarily) the influence of bid-ask spreads. By the end of June this cost was made back through performance and the fund's cumulative outperformance was 2.3%.
Looking forward, the main 'birth pains' of the fund have been paid up and this should be relatively advantageous for our future outperformance. The primary risk currently facing the fund is a relative underweight in rand hedge stocks. However, this risk was taken after due consideration so as to load up on high alpha shares as predicted by our quant model. Given the relatively small size of the fund, we are able to purchase certain slightly smaller shares than may have been feasible for a less nimble fund. We have chosen to take advantage of this opportunity.