The FT is calling out for feedback on how people are getting on with AI in planning their retirement. I know Mary and her advanced bump so am sharing a link to her request which you can access and leave your comment using this sharing link.
Sharing link to leave your comment
If the link has run out, please drop me a line at henry@agewage.com. Here is what Mary is asking for and you can copy across the FT email money@ft.com
There are some great comments on Mary’s request.
I just wonder if over the longer term passive index tracking will not always outperform AI generated portfolio selection.
With passive index tracking you are effectively following the money in a widely diverse portfolio with much lower costs (transaction and management costs combined). The impact of both market stars and black holes is diminished and are adjusted for just as fast as with AI generated trading.
The urge to out-perform the market is probably entirely psychological and is rarely achieved. Will AI try to do the same?
It’s just not possible for everyone to out-perform the market.
That’s not the point of active investment.
If the (pension) system deliberately supports disinvestment and punishes the risk takers (indeed by the word of the law may criminalise risk (ie investment)), then the outcome is that everyone does worse than the counter factual of an invested symbiotic pension system.
If we (actively) all invest in our system ( economic and pension) then yes there will be some who win more than others, but the greater good is achieved for more people by active investment.
This goes back to the 2004 act (?) which legally created the conflict between employers and Trustees and led to flight from Trustee boards of ‘business people’ ( ie those with more of a propensity to invest and grow).
The Treasury ( via a weak DWP) stepped into the governance gap via TPR, consequently directed schemes to ‘de-risk’ supporting the (convenient ) flight to gilts.
The folly of this is now well understood (but unspoken, least ‘the industry’ looks silly) and the Treasury like angry soviets are demanding “growth, growth, growth”, but of course the mandarins are not by design capable of grasping at all what is needed to let this happen – ie less direction from Govt / Treasury and less regulation.
The Pension Bill can help if it encourages and requires Trustees and sponsors to carefully look at the longer term impact of their investment and funding decisions, importantly framed within the context of growth. That needs to be supported by a contemporary TPR remit aligned to growth, and not as currently the case by protecting the overfunded PPF, which currently the TPR protects by eliminating risk ( ie growth) in the private sector schemes.
We really could not have created a system better suited to stifling growth if he had tried.
No growth, no pensions.
Asked about passive investing, my preferred AI (Perplexity) answered as follows:
In passive investing, the “tyranny of the benchmark” is the way index benchmarks end up dictating what investors own, how “risk” is defined, and how success is judged, often in ways that are misaligned with real-world goals like capital preservation and/or long-term wealth building.
What the “tyranny” means
• A benchmark (like the S&P or MSCI World or FTSE or Bloomberg ) becomes the anchor: portfolios are built to hug it, and performance is measured relative to it, not in absolute terms.
• This creates “benchmarkism”: the priority shifts from “Is this sensible, risk‑aware, long‑term investing?” to “Am I close enough to / slightly ahead of the index this month or quarter or year?”.
How it shows up in passive investing
* Owning more because it’s big, not because it’s attractively priced. Market‑cap‑weighted indices automatically give the highest weights to the largest, most expensive stocks or the most indebted issuers in bond indices. When sectors or names become very crowded (eg big US tech), index investors must keep buying them as they grow, reinforcing concentration and potentially inflating bubbles.
* Risk defined as “deviation from the index”. Tracking error becomes the main risk metric: the bigger the deviation from the benchmark, the more “risky” a portfolio is considered, even if that deviation is arguably safer in absolute terms (eg holding more cash or higher‑quality bonds).
* This can discourage defensive moves, because being too different from the index can get a manager fired by some clients or a fund’s reputation penalised, even when it protects capital.
Forced herding and misallocation of capital
• Because capital flows chase benchmark membership and weights, money tends to go where the index goes, rather than where fundamentals are strongest.
• In sovereign and credit markets, being labelled “emerging market” or being in a popular index bucket can matter more for flows than the country’s or issuer’s actual fundamentals.
Benchmarks as de facto policy constraints
• Regulators, consultants, and consultant-led boards often evaluate funds almost entirely on benchmark-relative performance, embedding benchmark hugging into mandates and risk rules.
• This can lock large pools of capital (pensions, insurance, etc.) into benchmark‑like portfolios even when that may not best serve long‑term beneficiaries.
Why it’s a “tyranny”
• It narrows the feasible choice set: anything too far from the index is treated as unacceptable, regardless of its absolute risk/return merits.
• It can increase systemic fragility: if a large share of assets is managed to the same indices, flows and rebalancing rules can amplify price swings and bubbles/ bursts/busts.
• It distorts incentives: managers are rewarded for not falling behind the benchmark more than for preserving capital, which can push them into the most crowded, expensive areas of the market.
How thoughtful investors may respond
For an individual or institution using passive investing, acknowledge the benchmark as a useful reference but refuse to be ruled by it:
• Use indices as tools for reviewing broad exposure and cost control, not as a definition of what you “must” own.
• Judge success mainly on absolute goals (eg real return over inflation, drawdown limits) rather than narrow benchmark outperformance.
• Be willing to accept some tracking error if it meaningfully reduces overdiversification or extreme concentration or improves resilience (eg capping single‑stock weights or tilting to quality).