An investor who started in 2012 and learned entirely by doing has genuine expertise. They know how markets behaved through a decade of falling rates, low inflation and reliable bond diversification, and their instincts were calibrated by real money in real conditions.
The difficulty is that some of what experience taught was a feature of that period rather than a permanent property of markets. Relationships between asset classes shift, sometimes abruptly, and when they do a self-taught process inherits assumptions it never knew it was making.
Structured material has an advantage here for a reason that has nothing to do with sophistication. It covers periods the learner didn’t live through.
Why Foundational Material Deserves a Second Look
This is where investment courses for beginners become worth revisiting by people well past the beginner stage.
Introductory material covers the long historical record because it has to explain concepts from scratch. That means it describes multiple regimes rather than the current one, including periods where familiar relationships behaved differently. Someone whose knowledge came entirely from recent experience has depth in one regime and no reference points for the others.
The material isn’t teaching anything new to an experienced investor. It’s supplying context that experience alone can’t reach, which is a different thing.
The Relationship That Changed Sign
The clearest example is the correlation between stocks and bonds, which underpins the most widely held portfolio structure in existence.
Research examining this relationship found that the stock-bond correlation stood at +0.35 during the 1970 to 1999 period and at −0.29 from 2000 to 2023, with 60/40 portfolio volatility running near 10.5% in the positive-correlation regime and falling to 8.4% in the negative one.
Read that carefully. The same portfolio construction produced materially different risk depending on which regime it operated in, and the shift between them was abrupt rather than gradual.
Anyone who learned portfolio theory during the second period absorbed a specific, correct-at-the-time idea: bonds cushion equity drawdowns. That was accurate for two decades. It was not accurate for the three decades before.
What Drives the Shift
The correlation isn’t random. The research points to inflation, real rates and government creditworthiness as the main explanatory variables, which means the regime can be reasoned about rather than merely observed.
Macroeconomic forecasters have been explicit about the implications. One assessment concluded that the pre-Covid regime of weakly pro-cyclical inflation is unlikely to return sustainably, with the correlation expected to move back into positive territory, posing challenges for allocation approaches that rely on long-duration fixed income to provide portfolio ballast.
Forecasts are forecasts, and this one may prove wrong. The useful part isn’t the prediction. It’s that the mechanism is identifiable, which makes the assumption testable rather than something to be discovered after the fact.
Assumptions Worth Testing Against Another Regime
A self-taught process usually contains several beliefs formed in one set of conditions:
- That bonds rise when equities fall, which held for two decades and not for the three before
- That drawdowns recover quickly, which reflects the specific recoveries of the past fifteen years
- That cash is a poor holding, which depends heavily on the prevailing rate environment
- That growth outperforms value, which describes a particular period rather than a permanent ordering
- That volatility mean-reverts fast, which has been true recently and not universally
None of these are wrong. They’re conditional, and the condition usually goes unstated because it was invisible while it applied.
Building Regime Awareness Into a Process
Practical steps that don’t require forecasting anything:
- Write down what each holding is relying on, in terms of a relationship rather than a price
- Check those relationships periodically against current data rather than assuming they hold
- Read the historical record for periods before the one you invested through
- Ask what would break each assumption, which usually identifies a macro variable worth watching
- Avoid sizing anything as though a correlation were fixed, since the sign itself can change
The first point is the most useful. A portfolio built on written relationships can be audited. One built on absorbed intuitions can’t, because the assumptions were never made explicit enough to check.
It also changes what a review looks like. Checking whether a holding is up or down takes a minute and says little. Checking whether the relationship it depends on still behaves as expected takes longer and occasionally produces something worth acting on.
What Can’t Be Known in Advance
Nobody can identify a regime change while it’s happening, and the research is clear that these shifts are recognised in hindsight rather than called in advance. Anyone claiming otherwise is describing a capability the evidence doesn’t support.
What’s achievable is narrower and still worth having. An investor who knows which relationships their portfolio depends on, and knows those relationships have historically changed sign, will notice a shift considerably earlier than one who absorbed the current regime as simply how markets work.
