On Thursday, October 8, around five in the afternoon, hundreds of families in San Carlos de Apoquindo were where one is when it rains hard: in their homes. The home is a refuge, the place where one protects the intimate vulnerability and fragile sweetness of the family. Within minutes, the mud that came down through Quebrada Honda rushed through their doors and shutters, swept away trucks and beds, buried armchairs and dreams, entombed pets and memories. What came afterward is well known: the classic Chilean "blame game." When the refuge becomes a trap, blaming nature is not enough, and the little game of "who said what, when" doesn't matter: what matters is that something that should have been protected was not protected.
I remembered this while reading about Fund E. Fund E is the affiliate's home when it rains outside: it is where those about to retire and those who don't want surprises take shelter. Between January and September, it posted one of its worst results in history. Simply put: the AFPs bet that long-term U.S. rates would fall... and they rose. To be fair, they were not the only ones making that bet. But they made it.
Let's put things in perspective. In 2024, facing another resurgence of pension withdrawal fever, I wrote that the thesis of the AFPs' historically poor performance did not withstand basic arithmetic, and I stand by that. Our low pensions are explained mainly by evasion, insufficient contributions, gaps in employment, and a retirement age that has remained frozen even as we live longer. That is precisely why I say it clearly: what happened with Fund E is indefensible. And I am not alone: a storm of criticism has fallen on the fund managers, even from economists who rarely agree.
The industry argues that there are no realized losses, only valuation losses. But the drop in the share value is a real loss for anyone who needs their savings today. Let them explain that to someone about to retire who does not have years to wait. The final impact will depend on when and how one retires, but the risk of retiring at the worst possible moment is real. Most people cannot postpone retirement or finance the wait with other assets.
The saying goes that it's not the pig's fault, but the one who feeds it slops. But sometimes the pig is partly to blame. That something is not prohibited does not exempt the risk managers of the country's most sophisticated investors from evaluating the consequences of their actions. And, I'm sorry, it also does not exempt the boards from a strategic analysis broader than mere financial mathematics. That is what they are there for: to demand a perspective that integrates the technical side, questions its assumptions, and accounts for the risks their models underestimate. Not to confuse complying with the rules with exhausting their responsibility, nor to resignedly endorse the myopic mechanics of the technicians.
One has to ask what was discussed at those tables, who asked the uncomfortable questions, and what got approved. Those boards are full of people I respect and admire, but in this... there is no other way to say it.
So the pig bears some blame, but there was also slops. What failed in the design? At least four things.
First: cheating at solitaire. We confused the conservative label with real risk. Fund E had little equity exposure, but it could accumulate, via bonds and derivatives, extremely high exposure to interest rates. Regulatory limits did not prevent losses improper for a conservative fund. Two contracts with the same nominal value can have very different sensitivities. That is why it is not enough to limit instruments or amounts: you have to control the portfolio's aggregate risk, its effective duration, its response to extreme scenarios, and the liquidity those require.
Second: where does Vicente go? Where the people go. The minimum return is measured against the industry average. Deviating from the predominant strategy risks regulatory costs if you end up with worse returns than the rest, even if you are trying to reduce a systemic risk. It does not force you to follow the herd, but it incentivizes not straying from it.
Third: there is no one deafer than he who does not want to hear. The warnings came back in 2024. In 2025 it was addressed by the coordination instances among regulators, the Superintendency proposed limits to the Technical Investment Council in November, and the Central Bank spelled it out in black and white in its Financial Stability Report. The regulation arrived in April 2026, with a 210-day deadline. Action was taken, but one has to ask whether it was in time and with sufficient scope. When the losses surfaced, the transition was not yet complete. That does not prove that acting faster would have avoided them, but it obliges us to examine whether the regulatory timelines were appropriate.
Fourth: with someone else's gunpowder, anyone can shoot. The reserve requirement makes the AFP share in the losses and partially aligns its fate with that of the affiliate. But it is not insurance against systemic losses, nor does it oblige it to compensate them if everyone loses at once. Its capital cost may be insufficient relative to the risk it transfers to affiliates. It would be worthwhile to discuss capital requirements and responsibilities proportional to the actual risk of the portfolios, especially in extreme scenarios.
There is urgency. In April 2027, the generational funds debut, with reference portfolios and new investment rules. These are advances, but questions remain: Will they prevent an apparently conservative portfolio from accumulating excessive duration, liquidity, and derivatives risks? Will the calculation of the reference portfolios be safe from reproducing, de facto, the herd effect? If we don't learn, we will build new neighborhoods on the same ravine. And the mud may not have finished coming down.
The State faces high financing needs and maturities while local fixed income adjusts. The Central Bank had warned that derivative margin calls could require liquidity and, under stress, force sales of local assets. If the AFPs liquidated significant positions in Treasury bonds, or persistently reduced their demand, they would push their prices down and the State's rates up. There is no evidence that this has already happened, but the channel exists and deserves attention. Thus, the flood that hit pension savings could also splash onto fiscal accounts; and in budgetary matters we do not need any more "muddy messes."
Beware of hiding behind the system's "usual" performance. It's like saying there were no floods in 99.9% of the neighborhoods, or that 99.9% of the buildings in Concón have no sinkholes. It makes no difference. That catastrophic 0.1% is enough to delegitimize authorities and administrators, regulators and representatives, builders and councilmembers; and, above all, to embolden the merchants of discontent, those who always arrive after the flood, never before, with their creative ideas and their TikTok videos. We already see them, enthusiastic, splashing around in the mud.
NEWSLETTER Opinion Saturday, AM Ideas in tension, opposing views, and clear analysis: elements to reveal the issues that divide opinions and will set the agenda. By signing up you are accepting the Terms and Conditions and Privacy Policies of La Tercera.