Fax Machines, Midnight Taxis and What Banking Actually Taught Me

By Lasse Mäkelä, Founder, Larzon Capital

This summer I ran into my old Associate at an event in Zurich. We had not seen each other since Merrill Lynch days, and we spent a while laughing about what our days actually looked like back then. At the time it was not quite so funny.

How it started

I received my conditional offer from Merrill Lynch roughly six months before graduating from City University London, where I was reading Banking and International Finance. Conditional on my grades. I remember thinking the stakes had suddenly become very concrete.

I graduated in 1997 and started that summer with a ten-week analyst training programme in Manhattan. Around a hundred recruits from across the world, all sitting in the same rooms, going through the same content, sizing each other up. New York in the summer of 1997 had a particular energy. The markets were moving, dot-com was beginning to take shape, and the sense that something large was happening in finance was everywhere.

Then London.

The first day

My first day in the London office started at nine in the morning. It ended sometime around three the following morning. I do not think this was unusual or designed to send a message (or maybe it was). It was just what the work looked like, and the work started immediately.

I joined the Nordic team and later the TMT coverage team, working with technology, media and telecommunications companies at the peak of the dot-com era. The transactions were significant. I worked on the Sonera privatisation and IPO, the Finnish national telecommunications operator being brought to market in the late 1990s. When Telia, the Swedish incumbent, moved to acquire Sonera, I was involved in that process too. I worked on the O2 IPO when it was spun out of British Telecom, and in various roles alongside major operators including TIM and Vodafone during what was genuinely the most active period for telecoms capital markets in a generation.

As a junior analyst, I was sitting in boardrooms with the Boards, CEOs, CFOs and heads of M&A of large listed and unlisted companies. There was no gradual introduction to that. You were in the room, you were expected to be useful, and the preparation that made you useful happened the night before. I still remember, when my boss said, just before an important meeting: You are not expected to say anything:)

What the physical work actually looked like

The hours were long. Not long by the standards of a busy week, but structurally, consistently, sustainably long in a way that is difficult to explain unless you lived it. Working two or sometimes three times the hours of a normal office job was the baseline, not the exception. At the time, surrounded by other young and equally energised people, it felt like being part of something. My Associate, when I saw him in Zurich this summer, confirmed that his memory of it matched mine, though he had some additional detail I had apparently blocked out.

What I remember most clearly is not the hours themselves but the physical nature of the work. The constraints were not intellectual, they were logistical.

A typical evening might go like this. You spent the earlier part of the night refining a financial model. Once the model was in acceptable shape, you needed to update the presentation. PowerPoint was still relatively new at that time, and Merrill Lynch had a department called Desktop Publishing, staffed by people who actually knew how to use Powerpoint properly. Analysts queued for slots in the DTP. You physically waited your turn while the presentation was reformatted, resized and corrected by someone else, because doing it yourself would take longer and look worse.

Meanwhile, the senior banker on the transaction was in a hotel somewhere, around the world, reviewing an earlier version of the same presentation. His feedback came back by fax. You waited for the fax. If the hotel's machine was broken or out of paper, you waited longer. When the pages arrived, often handwritten across printed slides with arrows and crossings-out, you decoded them, went back to the model, made the relevant changes, and then rejoined the queue for DTP.

And then, if the meeting was early enough in another city, you printed ten copies of a hundred-page presentation, put them in a bag, and moved.

The motorcycle and the flight

There was one occasion involving a Request for Proposal that had to be physically delivered in Helsinki by a specific time in the morning. The printers chose that night to malfunction. There were edits still being made. A motorcycle courier had been booked. I joined the courier, riding through London traffic to Heathrow on a motorcycle rather than waiting in a car. We made the terminal. The gate was already closed. I had a conversation with the gate agent that I suspect I could not have managed if I had not been running on adrenaline, and the gate opened. The proposals made the flight. The timetable was met.

I tell this story not to make it sound impressive but because it illustrates exactly what the work was sometimes made of. The intellectual challenge of the job was real: building complex models, structuring transactions, understanding how to position a company for investors. But a meaningful proportion of the actual time and energy went into physical logistics that had nothing to do with the quality of the thinking. Printing, queuing, waiting, running.

What that environment produced

It is fashionable to look back at those hours and call them unsustainable, which they were, or to suggest the culture was broken, which in some respects it was. But something real also came out of it. When you are in boardrooms as a junior, working on transactions that matter, at a pace that does not allow for passive observation, you compress your learning significantly. Four years of that produced something that would have taken much longer to accumulate anywhere else.

The transactions I worked on introduced me to the full architecture of capital markets at a moment when that architecture was being genuinely tested and reshaped. Sonera, Telia, O2, Vodafone: these were not peripheral situations. The models I built under those conditions taught me to separate what matters analytically from what is noise.

The same work now

I am now reasonably deep into using AI in my daily work, and I find myself thinking about this contrast often.

The physical problem-solving that consumed so much of the analyst experience has largely been replaced. Formatting a presentation no longer requires queuing. Information arrives without fax machines or motorcycle couriers. The model can be refined and the deck updated and the revised version sent across three time zones before anyone has left their desk.

What that means, I think, is that the work has shifted from physical problem-solving to intellectual problem-solving. The logistics no longer eat the time. The question becomes what you do with the time that remains.

I suspect analysts today face a different kind of pressure. The expectation of speed has not decreased because the tools are faster. The marginal value of processing information quickly is lower when everyone has access to the same tools. What differentiates good work now is probably the quality of the judgment applied to the output, rather than the speed of producing it.

When I met my old Associate in Zurich this summer, we agreed on something. The fax machine era was genuinely harder in ways that were mostly pointless. But it also produced a certain tolerance for ambiguity and physical endurance that was not without value. Whether you can learn the same things with better tools is a question I do not think anyone has fully answered yet.

Lasse Mäkelä is the Founder of Larzon Capital, a cross-border M&A advisory firm based in Switzerland. He began his career as an analyst at Merrill Lynch in London in 1997, working on Nordic and TMT transactions through the dot-com era.