In the midst of the discussion on social security reform, there is an issue that has not been analyzed: the evolution and effectiveness of the regulation in which the multifunds are operating.
The idea of the multi-fund scheme is that they should offer returns compatible with their level of risk. However, an investigation by academics led by Arturo Cifuentes (ClapesUC) shows that it did not happen, in fact, when analyzing periods of five years, most of the time is just the opposite, that is, the most risky has been the fund has less rented. The issue is fundamental to pensions in the country and, until now, had not been touched. Of course the problem is not how the AFPs manage the funds, but the regulatory framework of investment policy that they are required to follow.
Several articles give an account of this situation. The first "Fifteen years of defined contribution: evaluating the experience of the Chilean retirements", by Hans Schlechter (ClapesUC), Bernardo Pagnoncelli and Cifuentes, indicate that the analysis of the evolution of the multifunds shows a negative verdict. While the study reveals that funds are well ranked in relation to their risk (with A being the most risky and E the least risky), "their accumulated returns for long periods of time are not aligned with their risk profiles."
Thus, the article states that "in essence, participants in funds A and B are at greater risk, but do not receive returns that compensate for these risks." And in terms of more than five years, the situation worsens, as all funds converge to similar returns, "which makes us wonder if it would not be better to just keep one type of fund."
This situation particularly affected the younger workers who entered the system around 2007 and probably financed A. The document shows that, taking as an example, one can see that if they invested 100 UF in 2007, they would have only 102.12 UF in November of 2015. Evidence indicates that attempting to control the risk-return profile of funds through asset placement without change in time did not work in the desired direction. The risk-return profiles of the funds are in fact opposed to the desired objectives.
The problem is that the reason for this is precisely that the managers of the funds in the AFPs are bound in this sense, since it is the regulatory framework that forces them to have minimums and maximums of each type of asset. The same authors point out that at the end of the day, these managers are also victims of this regulation.
What is proposed?
In the second study, in which Cifuentes and Pagnoncelli participated with Tomás Gutiérrez and Davi Valladao, a solution is proposed to realign the risks with the returns.
The authors propose that the current system, in which the funds are divided by type of assets, with minimum and maximum values, be replaced by a system of limits measured with VaR (Value-at-Risk) or CVaR (conditional risk assessment) , which limits investments directly to assumed risk, rather than to the asset class. Not only did this align risks with returns, but it would also offer more flexibility to fund managers.
In fact, another paper, which will be presented today by ClapesUC (Pension Funds in Mexico and Chile: a risk-reward comparison, by Schlechter, Pagnoncelli and Cifuentes), makes a comparison of the Chilean multi-fund system with the Mexican. In the latter country, a ceiling is used for asset classes, but without a minimum, and also combined with risk constraints at the portfolio level, which are different depending on each fund.
Under this scheme, the Mexican multi-fund system "showed much more returns in line with the risk-reward profile desired for each fund," the study said. Thus, the returns from these funds are organized as they should theoretically, ie the riskier ones have the highest returns and vice versa.