The Halma Discipline

Do less of what underperforms. Do more of what outperforms. Repeat for 20 years.

The unglamorous compounder

In 1997, David Barber gave a speech explaining how Halma — then a £0.5bn engineering conglomerate — had delivered 22.5% compound annual EPS growth for twenty years, ranking among the top performers in the UK.

The strategy was almost aggressively boring.

No dramatic restructurings, no transformational acquisitions, no pivots to hot sectors. Just a clear financial model (20-30% EPS growth, 40% ROCE, self-financing), applied with relentless discipline across a portfolio of small industrial businesses.

Twenty-seven years later, Halma is a £12.8bn FTSE 100 company. It has raised its dividend by 5% or more for 46 consecutive years — a record unmatched in the index. The same playbook, still working.

Centipede, not caterpillar

Barber called it the centipede approach. Rather than betting on big strategic moves, you make incremental improvements across the portfolio. Do less of what gives below-target returns. Do more of what exceeds targets. Each small positive step compounds over time.

This is virtually risk-free compared to the alternative.

Big strategic pivots look decisive — they generate board excitement, consultant fees, and press coverage — but they also carry execution risk, integration risk, and the risk of being wrong about where the world is heading. Most fail to deliver.

The centipede just keeps walking. Trim the underperformers, reinvest in the outperformers, repeat. No single decision is bet-the-company. The aggregate effect, over decades, is transformation.

Halma went from tea trading to life-saving technology through this process — not a visionary leap, but thousands of small steps in roughly the right direction.

Quality isn’t fungible

Financial advisors love P/E arbitrage. Buy a company at 8x earnings, fold it into your 20x multiple, watch the market value the combined earnings at the higher multiple. The maths works on paper.

Barber’s insight: the maths destroys value.

A company earning 40% ROCE doesn’t want acquisitions that dilute this, regardless of how the P/E arithmetic looks. The market adjusts. Investors aren’t stupid — they’ll re-rate your combined entity lower once they see the blended quality.

His formulation: “Never pay enough for a good acquisition; never pay too little for a bad one.” The quality of earnings matters more than the quantity. Halma rejected deals that would have boosted short-term EPS but diluted long-term returns on capital.

This requires saying no to opportunities that look good on spreadsheets. Most companies can’t do it. The pressure for growth, any growth, is too strong.

Grow the market, not your share

Halma targets niches — businesses with fewer than 10 home competitors and fewer than 80 globally. Market shares of 60-70% are common.

Short-term operators see this as a problem. You’ve saturated the market. Where’s the growth?

Long-term operators see it differently. High margins from market leadership fund investment in growing the total market — new applications, new geographies, new adjacent problems your technology can solve — rather than fighting competitors for slices of a fixed pie.

When you dominate a niche, market expansion is your growth strategy. The competitive moat funds the R&D that widens the moat further.

This only works if you resist the temptation to diversify into unfamiliar markets where your advantages don’t translate. Halma’s risk hierarchy put bolt-on acquisitions (in familiar fields) at the bottom and diversifications at the top. Seventy percent of their deals were bolt-ons.

Holding the frame

What made this work wasn’t the individual insights — plenty of investors understand compounding, quality, and focus. It was the systematic application over decades.

Have a clear, simple mathematical model. Halma’s targets were set in the mid-1970s and barely changed. Everyone knew what good looked like.

Keep main objectives always in mind for quick decision reference. When an opportunity arose, the framework was already there. No strategy retreats needed.

Systematically reshape the group to conform to the model. Not occasionally, not when convenient — systematically. Every capital allocation decision measured against the same criteria.

Don’t be blown off course by opportunistic arguments. Short-term advisors will always have clever reasons to deviate. The model says no.

Question institutions focused on short-term arithmetic. Investment banks, consultants, even your own executives will optimise for quarterly metrics. The discipline is holding the longer frame.

What survives

The speech was given in 1997. Since then: the dotcom crash, the financial crisis, a pandemic, multiple recessions. Halma kept compounding.

The approach isn’t complicated. It’s just difficult to sustain. Boards get impatient. New management wants to make their mark. Activists demand action. The temptation to do something dramatic is constant.

The companies that compound over decades are rarely the ones making headlines. They’re the ones running the same playbook, year after year, resisting the pressure to be interesting.

Halma’s centipede is still walking.

Connects to Library: Marginal Gains · Process Power · Counter-Positioning

See also: Reading Guide for the complete collection of Field Notes.