High levels of risk are associated with a higher potential or return, says one of the principles of investment. However, when you have the possibility to compare these two variables and put them among competitors in their historical performance, what comes out is that there are those who take an additional risk to a level of performance and that some people manage to, with a lower risk, be more efficient and get better results.
Among the Afores you can observe various risk profiles and investment styles. On the one hand, there are those who are mandated to be at the top even if it means taking an additional risk; others are interested in being above average; others just want to be in the average to avoid risk and others which are guilty of being conservative in their results.
Investments are dynamic and what worked yesterday will not necessarily work in the future. Therefore, investment areas have to continually reinvent themselves to improve results.
Yields are just part of the big picture. To see it complete, you need to add risk. In this case, it will be expressed as the value at risk (as measured by the standard deviation of daily quotations which is multiplied by a probability value) to obtain the expected loss with a certain percentage of probability that in this case we place 95%.
Graphing risk and performance
To do this exercise we use direct returns (not annualized) of the basic Siefore 2 (SB2) which is the Siefore (Afore fund) which manages the largest amount of assets (36%), equivalent to 50 billion dollars. Additionally, it is focused on federal government and private sector workers whose age is in the range between 46 and 59 years. Therefore, 9 million workers are enrolled in these funds. 17% of assets are invested in equities and the fund has a weighted average maturity of 11 years at May 2016 according to data provided by CONSAR (regulator for Afores).
Three years were chosen because it has been characterized by significant volatility which has demonstrated the ability of the administrator in a changing environment. The Mexican government bond M23 (due in 2023) in the same period fluctuated between a rate of 5.20% and 6.64%, expressed in prices that has meant a change in the order of 10%.
Based on the standard deviation of historical prices of each of the 11 Afores Value at Risk (VaR) parametric 95% confidence interval was calculated. This means that in 95% of the days of operation, the maximum expected loss percentage is being calculated for each Afore. The parametric VaR is obtained by multiplying the standard deviation of prices period by 1.96%.
The balance between risk and return
Plotting risk (horizontal axis or “X”) and performance (vertical axis or “y”) it can be seen that the SB2 Coppel has been the best performing Siefore in the last three years (15.jul-13 to 14 -jul-16) with a VaR of 0.52%. In the graph a vertical line is added from Coppel’s VaR and some Afores are on the left and others on the right side. The left means lower risk and the right is higher risk taking with Coppel as a reference.
The first thing that jumps out at us is that the SB2 Azteca with a degree of risk VaR near Coppel’s (VaR 0.52% of Coppel vs VaR 0.51% of Azteca), paid 4% less than Coppel.
The other point that draws our attention are the two extremes in terms of risk, where on one hand Afore Inbursa appears as the most risk averse (the left side) and Afore Invercap as the highest risk taker (the right side). Inbursa with a VaR of 0.18% paid 10.63%, while Invercap with a VaR of 0.79% had a 15.62% payment. Comparing Invercap with Coppel’s result, it can be said that the risk incurred by Invercap did not pay up, since with two-thirds of their risk, Coppel achieved better results.
In the graph it can be seen that Profuturo and Principal who are in second and third place performance in the period have a small VaR compared to Coppel (about 0.10%) and a yield of between 1.50 and 2% below Coppel. Also, these two Afores did a better risk management than Sura, Banamex and Metlife (in this period) which with a similar VaR had lower returns. XXI-Banorte who also has a VaR similar to this group, had the lowest yield for this level of risk. Their performance is in the second to last position (place 10 of 11) and with a difference of almost 8% vs Coppel. Finally, you have PensionISSSTE who is in position 9 of the 11 Siefores.
In long-term portfolios, exercises like this serve to see if the investment team is doing well or not (in hindsight). Additionally, prospectively, it is important to maintain a constant review of investment scenarios and an analysis of market expectations. In a second plane, it works to determine whether the distribution by asset class (within applicable investment regime) is correct and in a third plane, if the instruments chosen to invest in this asset class are suitable for generating the expected performance. Many times you can have the correct stage and distribution by asset class, but fail in specific assets, for this reason it is important to carry out this exercise routinely and to make the necessary adjustments in time.
Column by Arturo Hanono