How to invest: Rules, not stories

Building a world model

It can be easy to be swept away by the narrative when you’re thinking about investing. Heroes and villains, world-changing technologies and catastrophic politics, bullish cases and bearish views. The problem with this in investing is that you can always find a tale to support your perspective. A buyer and a seller – both with a story to tell.

Which is why we prefer to build evidence-based models. We look for rules, not stories. Collect as much actual economic data as we can, compare it to investment opportunities, and then use statistics to establish likely outcomes.

It’s often termed “macro” investing, and over the long term it works. Economic rules are pretty reliable – following them offers a better than average chance of getting the investment positioning right.

Muted macro, supercharged stories

However, over shorter periods of time, the economic signals are often swamped by the daily noise. You only get twelve inflation readings a year, compared to twelve thousand “cost of living crisis” articles.

And if macro signals are faint at the same time as the stories are strong, it can feel like the rules no longer apply. Recent years have felt a bit like that; whether looking at stock markets or bond markets.

The tale of tech

Let’s start with stocks. In a typical economic-led market, the rules are clear. If growth is strong, companies in cyclical businesses like hospitality, luxury goods and banks do well. If growth is weak or falling, stable companies such as healthcare or utilities shine.

But in the past two years, that rule hasn’t seemed to hold. After the pandemic, the economic forces were confused and contradictory, eating away at the predictive power of traditional models. Economically, nothing went drastically wrong, but at the same time, nothing went amazingly right. Growth was mildly good or sort of bad, and so were the model signals.

Enter tech. The largest sector in the US index didn’t just outperform – it told a blistering tale while doing so. The world-changing potential of AI made boring old economics seem irrelevant. Investors stopped asking “where’s GDP headed?” and started asking “who’s training the best LLM?” or “who has the most data centres?” or “who keeps delivering positive profit surprises?” These questions seem easier to answer, and are often more fun to talk about than the complexity of the global economy.

But history tells us macro matters – eventually. If economic data starts to swing around once more, the best guide will be the macro rules which have worked for decades, rather than waiting to see who’s on the cover of TIME magazine.

Given the lack of signal, our equity positioning this year hovered around our usual amount; neither adding to nor avoiding risk. One could argue that we missed the chance to add to equities in a strong year, but that would be looking back at the story, rather than the rule. We prefer to stick to our rules.

What’s bugging bond markets?

Bonds play an important defensive role in portfolios, protecting against downturns and slumps. Our models look for economic indicators which suggest the likely state of the world, and whether it might be susceptible to change.

If tech was the story which supplanted economics in equities, for bonds the culprit was politics. Our model had a disappointing year. Not disastrous, but disappointing. In a year where rates drifted gradually lower, the model didn’t quite move fast enough.

Our rules look at monthly or quarterly data series. In 2025, monthly data just couldn’t keep pace with events (largely coming from the Oval Office).

Just consider summer 2025. Donald Trump announces tariffs in April, then starts on-off pauses and resumptions which continue to swing sentiment around. The US then bombs Iran, before announcing a number of trade deals. Add in talk of Beautiful Big Bills, Federal Chairs being fired and a government shutdown, and you’ve got a backdrop where fundamentals were drowned out by headlines. The rules, calibrated for macro momentum, simply couldn’t keep pace with political flip-flops.

This isn’t a flaw; it’s a feature. Adjusting the exposure between short or long dated bonds is designed for sustained trends, not tweetdriven turbulence. When businesses get more optimistic, inflation bites, or growth accelerates, the model shines. But in a year where macro was muted and politics noisy, neutrality often beat conviction.

Two stories, one Lesson

Tech and bonds tell different tales, but the moral is the same: context matters. Models thrive and rules work when their assumptions hold. When those assumptions break, performance wobbles. That’s not failure; it’s reality. No model captures every regime.

For equities, the recent regime shift has been tech’s dominance, AI hype, and a market chasing innovation over inflation. For bonds, it was episodic: political shocks hijacking a macro-driven framework.

The essence of rules-based investing is about probabilities, not certainties. It’s about applying rules consistently, even when headlines tempt us to improvise. Because while gut feel can win a battle, process wins the war.

Looking ahead – what brings macro back?

So, what brings the rules back into play? For bonds, the catalyst might well be inflation’s second act, a credit wobble, or an increasingly powerless Donald Trump. For equities, it might be tech fatigue or a growth scare that puts the wider economy back in the spotlight. Either way, macro volatility will return. It always does. And when it does, systematic models will regain their edge.

Until then, balance is key. For equity investors, that means respecting momentum without abandoning diversification. For bond allocators, it means trusting the process, even when signals feel slow.

Maintaining investment discipline is hard. But over time, the benefits compound into something far more rewarding than chasing stories.