This is the second of two posts about Rolls-Royce and its bankruptcy in February 1971.

In the last one, we left Rolls on the morning of 4 February as the engineering group called in the receivers, having run out of money trying to build a large jet engine - the RB211. (If you haven’t read it, or listened to the A Long Time in Finance podcast we did on this, you can find the post here and the pod here.)

We heard how the news shocked the nation and dismayed the workforce, large numbers of whom faced redundancy as creditors sought to salvage something from the wreckage. In this second part we’ll look at what happened afterwards; how the RB211 project was resurrected against the odds, and how insolvency proved not just a turning point in Rolls’s story but the essential foundation of its future success. Without this bankruptcy, I’ll argue, Rolls simply wouldn’t be the world beating £120bn company it is today. Our podcast on this second part can be found here.

Now read on..

The sparing of the RB211

One thing was clear in the wake of the bankruptcy. Much of Rolls would have to be resurrected in some form; particularly the military aero engine business (around a quarter of group sales) as well as the nuclear submarine propulsion side. These performed essential national tasks, and couldn’t simply be closed or sold to the highest bidder.

The big question was over the RB211; the engine on which Rolls’s bosses had literally bet the company. Its future was much less certain. Pressing on required not just financial aid from the British government, but the backing of Lockheed, the large US aircraft manufacturer, which was the sole customer for the engine.

Having ordered the RB211 to power its new wide body Tristar aircraft, Lockheed held a contract that was highly advantageous, and could (in theory) force the UK company to pay hundreds of millions in compensation. But Lockheed was also in a mess. The Tristar had taken too long, cost too much to develop, and sales were not strong. There was a real risk that Lockheed itself might go bust if the Tristar were further delayed and launch customers started backing out, leaving it with huge un-recoupable costs. In short, the two companies were locked in a form of death embrace.

Up till the bankruptcy, both the British and American governments had stayed on the sidelines, wary of weighing into a commercial dispute while there remained a chance the two could find a way through unaided. But after Rolls went bust, they started to play a much more central role.

At first, the British government had to decide whether the RB211 should be saved. As I said in the last post, there were doubts. The Minister for Aviation Supply, a Gloucestershire pig farmer called Fred Corfield, felt there might not be a big enough market to justify completion, given that Pratt & Whitney and GE had both already produced similar large turbofans and snagged the largest share of the wide body market (supplying the Boeing 747 and Douglas DC10 respectively). There were concerns about how much the RB211 would cost to complete, how long it would take, and whether the Tristar might itself be axed.

Another critical question was over the revolutionary three spool design that had proved so problematic: could it be made to work technically? And that brings us back to the telex the Rolls board received on the morning of 4 February from its technical head Ernest Eltis, which reported the engine had finally produced sufficient thrust on the test bed to meet Lockheed’s performance criteria.

Fortified by this, the Minister of Defence, Lord Carrington, commissioned a study from a team of veteran Rolls engineers led by Stanley Hooker (developer of the wartime Merlin and the Welland, Rolls’s first jet engine) into how long and how much the RB211 would take to finish. The prognosis was quite (if not overwhelmingly) positive: they felt the engine could be made to work with a 6 month delay and a further cash injection of £120m (roughly double the £60m Rolls had sought the previous autumn, none of which had yet been spent).

Hooker was promptly appointed to the board of Rolls-Royce (1971) Ltd, the new company containing the aero engine and nuclear business the government was buying out of receivership, and told to get on with it. (In the interim ministers indemnified the receiver, Rupert Nicholson, against any claims from creditors for running up further costs and agreed to drip finance into the project.) Hooker, meanwhile, assembled a sort of “Dirty Dozen” cadre of wartime Rolls engineers - many in their 60s and 70s - who came out of retirement to help out.

On arrival he found many of the existing team “completely demoralised” and “looking for someone to blame”. “I called a meeting of the entire engineering staff and explained the exact situation,” he said. “I then appealed to their loyalty to the good name of Rolls-Royce to get the RB211 quickly certificated and delivered to Lockheed. I promised 100% support to their efforts and asked any doubters to leave at once.” Within days morale had improved and Hooker was “able to put my finger on several crucial faults and have them rectified very quickly indeed..”

As the engineers put their hands to the pump, the government had to decide how to handle the onerous Lockheed contract. Did it still even have a customer? Lockheed was looking over its shoulder at the airlines who had ordered the Tristar, worried that they might pull out.

The US company even re-ran its engine competition to consider replacing the RB211 with GE’s CF6. Yet, in a funny way, the high tech British engine was the key selling point for the Tristar over Lockheed’s main airframe rival, the DC10, because of its fuel efficiency and relatively low maintenance costs - always assuming it worked. Switching engines also meant even more delays. Lockheed sucked its teeth and decided to stick with Rolls.

It was increasingly clear the two projects were inextricably tied together. So when the British government discreetly explored the possibility of Lockheed ditching the contract, Lockheed’s chairman Dan Haughton was adamant: he wanted “engines not damages”. To ram home the point, he went round his own customers, confirming they would stick by Tristar even if the cost went up and deliveries were delayed. (TWA said it was relaxed because its own finances were so stretched it didn’t mind waiting a bit longer.)

Lastly, Haughton persuaded Richard Nixon’s administration to kick in $250m of soft loans which unlocked a further $400m in private finance - sufficient to tide Lockheed over till cash from sales started to roll in. (This was a close call - a sceptical Congress approved the loan by a single vote - and rejection could have killed the project: the extra funding for Lockheed was a key condition of the UK’s own financial rescue of the RB211.)

On 25 March 1971 therefore, less than two months after the bankruptcy, a new deal was struck, under which Rolls would sell 555 engines with revised prices, delivery dates, and an allowed increase in engine weight. To sweeten it all, the UK not only put more money into Rolls but ran around cajoling British carriers to buy the Tristar. In addition to state-owned BOAC and BEA, ministers put the squeeze on overseas private but dependent carriers such as Cathay Pacific in the British colony of Hong Kong. Overall, about a fifth of Tristars were sold in the end to British airlines.

The reckoning

The path was now open to restructure Rolls and, throughout 1971 and 1972, the company was retrieved from bankruptcy. State-owned Rolls-Royce (1971) took over the aero engine and nuclear business, while the famous motor car division was hived off and finally sold to the stock market in 1973, raising £37m. The creditors didn’t do that badly. Debenture holders were repaid in full and unsecured creditors eventually recovered around 62.5p in the pound.

Under its new chairman, Kenneth Keith, a former City banker, Rolls introduced more financial discipline. The bosses implicated in the RB211 debacle were gone; Pearson when the company filed for bankruptcy, while Huddie, the boss of the aero engine division, never returned from his nervous collapse. Both were heavily criticised in the official report into Rolls’s failure for their financial irresponsibility. One manager noted the change of direction under Keith’s leadership. “The first thing we had to learn was that the company was not a playground for engineers to amuse themselves.”

Hooker and his team got the RB211 working ahead of time and it was certified in April 1972. The first Eastern Airlines Tristar flew on 26 April and Hooker became the third Rolls executive to be knighted for the RB211 - the only one associated with success not disaster.

Getting the Tristar airborne was, however, far from the end of the story. The RB211 was destined to have a much bigger role in Rolls’s future. (Comparatively speaking, the L-1011 Tristar was a failure, selling just 250 units before being discontinued in 1982 - many fewer than the DC10 or Airbus A300, the latter finally taking to the skies in 1974.)

Central to this was its 3-spool technology which had proved so tricky to perfect. Once mastered, it proved surprisingly forgiving to scale up without a major redesign. The RB211 went quickly from delivering 40,000lb of thrust at launch to 50,000 by 1973, a step-up that persuaded Boeing to offer the engine to customers of the 747. Nearly 10% of the 1,574 Jumbos built ended up being powered by the Rolls engine - an unexpected vindication of the “American strategy”.

Meanwhile its fuel efficiency became a key selling point following the 1973 oil crisis. The Australian flag carrier Qantas switched to Rolls powered Jumbo Jets because the RB211 burned 7% less fuel than its P&W powered fleet - a saving of around $1m a year. The engine’s relative quietness was also a positive with the tightening of environmental standards from the late 1970s on.

Dealing with nationalisation

The rescue did come with one significant political uncertainty. Rolls-Royce (1971) Limited ended up as a nationalised company during that less-than-golden era when ministers routinely meddled in state-owned businesses to promote social objectives such as employment-based policies and regional development. Few nationalised industries - think cars, railways, steel or shipbuilding - emerged from the 1970s with their reputations or performance enhanced.

Fortunately, the company’s new chairman, Kenneth Keith, proved strong enough to keep the meddlers at arm’s length - even the arch meddler, Tony Benn, who served as Labour’s industry secretary from 1974-5, when he concocted a mad plan to direct even private industry to make investments in what were essentially government priorities through a new body, the National Enterprise Board.

To get a flavour of these years, take this entry from Benn’s own voluminous diaries about a meeting with Keith in February 1974 to discuss the new government’s priorities and what it expected of Rolls. After Keith recited his prior vow only to take the chairman’s job “so long as I am not buggered about by junior ministers and civil servants and officials”, Benn recorded what he described as a “bitter exchange”. At its conclusion, he wrote: “[Keith] said: ‘I hope we can get on well together with you as my Minister.’ I added, ‘And the owner of your company.” So he said, ‘But not my employer’. I said, ‘I am inclined to say to you what my sergeant said to me in the war: ‘You play ball with me and I will play ball with you’.”

Whether Keith played ball or not, Benn was gone within 18 months and he got on much better with the successor Eric Varley, who took a less hands-on approach. When Mrs Thatcher was elected in 1979, Keith persuaded the new industry secretary Sir Keith Joseph to remove Rolls from oversight of the National Enterprise Board entirely and take a backseat approach - essentially putting money in when required and not intervening too much.

This was timely. Although the RB211 had proved itself the early 1980s were tough. Sales languished as the weak economy made airlines wary of buying new aircraft, while in 1982 Lockheed discontinued the Tristar, still the RB211’s mainstay. Rolls experienced several years of losses, and there was a moment in 1984 when, mindful both of the need to start planning for an RB211 successor, and the Thatcher government’s growing interest in privatising Rolls, the company entered into a partnership with GE which could have led it to give up being a major player in large aero engines.

This saw GE take a 15% stake in the medium sized RB211-E4 in return for Rolls taking a similar stake in a new 60,000lb thrust GE engine. Essentially, had the deal rolled out as planned, Rolls would have given up on the big end of the market and focused on mid sized turbofans. The move was greeted with grim satisfaction by the Financial Times in its report: “Rolls-Royce faces up to reality”.

“By swapping a share in one of its new engines for a stake in one of General Electric’s, Rolls has finally moved away from a course which it has followed in the large civil engine market for the last 20 years - a course which has taken this proud engineering company into bankruptcy, and which more recently has left it with an increasingly weak position in the market for high thrust commercial engines,” wrote Richard Lambert and John Makinson. The article also cited Ralph Robins, later Rolls’s CEO, saying that “to develop the RB211 up to 60,000 pounds of thrust would have effectively required the designers to start with a clean sheet of paper. On this basis the project could have cost $1-1.5bn or more.” This was money that state-owned RR neither had nor was the Thatcher government likely to give it.

But.... once again, the doomsters spoke too soon. The RB211’s revolutionary design rode to the rescue. By scaling up the fan on the front, it turned out Rolls could raise the thrust to the level required without an expensive redesign - establishing what from 1990 became known as the Trent family of turbofans (uniquely the RB211 had kept its original development designation, and was never given a river name, like all the company’s other jets). By 1986, Rolls was winning Boeing contracts in competition to the GE engine it had a stake in, and by the end of that year, the alliance was dead.

Enter the Trent

Few products have fulfilled the hopes of their creators more completely than the RB211. Together with the Trent, essentially an evolved variant, the engine conceived by Denning Pearson, David Huddie and Adrian Lombard has built Rolls into the £120bn giant it is today. Thanks to its three spool technology, Rolls could make the leap from 60,000 pounds of thrust to 90,000 and more without developing a new engine. The Trent cost just £400m to develop and even today, the range is based on technology first developed nearly 70 years ago.

Rolls has built an enviable position at the apex of the large aero engine business. Take market shares as a measure: in 1987, at privatisation, Rolls had around 10% of the market for sales of big turbofans, up from zero in 1971 when it was nationalised. Three years later, that had doubled to around 20%, and by 2000 its share was up to around 30%.. Remember also the market itself expanded dramatically after the Cold War wound up in 1991 ( the year before Rolls sold its first civil engines to the Russian manufacturer Tupolev). So while Rolls had delivered just 128 wide body jet engines in 1982, it delivered 525 by 1995 and 1,541 by 2001. By 2022, Rolls held a 50% market share of new engine deliveries, which took its installed global base to a new high of 36%.

Conclusion

It’s impossible to look at Rolls today without acknowledging the centrality of the RB211 - the very engine that brought it down. This failure ended the careers of the men who decided to build it, and can be seen (as it was in the 1973 official report) as an irresponsible gamble.

But could Rolls have avoided bankruptcy in 1971 and still be the same company it is today - essentially the number one large jet engine maker? There is (I think) no world in which that outcome would not have required the RB211 - or something very like it - to be built.

On the podcast, Neil and I went through various permutations to see if a counterfactual version existed, but we couldn’t see it. Here are four scenarios.

• Not build the RB211 and stick with smaller jets - or even retreat to defence like much of the rest of the British aerospace industry. Rolls might not have gone bust but it would be a much smaller company, and its surviving civil business (if it had one) would be a bit like its European competitors (Safran, Fiat Avio, BMW) in producing narrow body engines in joint ventures. It would also be much more dependent on military sales, which would have been a worse place to be between 1991 and about now.

• Build the RB211 but finance it more sensibly, i.e. with much more equity. It’s hard to see how Rolls could have funded the £200m odd the RB211 ultimately cost to develop, even with a substantial dollop of launch aid. The company did raise around £20m in a share and debenture issue around the time it took on the project, which was not wildly well received. Doubling the shares in issue to bet on a revolutionary engine without a top tier customer? Not a chance.

• Build the RB211 but in a joint venture with a US manufacturer. To be fair, Rolls did investigate this: they tried to cut a partnership with Allison, the aero engine subsidiary of General Motors for whom it would have been a way into jets, but couldn’t strike a deal. Otherwise, it’s hard to see who they might have gone to. P&W and GE would never have helped Rolls get into the large turbofan market. It’s also questionable whether Rolls would have ended up where it is in large jets had it gone down the unstable JV route.

• Perhaps the most intriguing: junk the American strategy, don’t go with Lockheed but do Airbus instead, which would have allowed Rolls to get a better deal and a less rushed development timetable. This has some genuine plausibility, in that Airbus would certainly have been game, and the A300, when it came to market in 1974, proved a big success, selling more than twice as many units as Tristar. Rolls could have become Airbus’s main engine supplier from the off, not a bad outcome given the European manufacturer’s subsequent sales, while by going with Lockheed, it didn’t get a look in until 1995. So maybe. But while this all looks great, remember it’s all based on hindsight. The situation Pearson and Rolls actually confronted in 1968 was one where the Americans were building actual wide body designs while the A300 was a paper plane in search of a consortium to build it. Twin jet planes still couldn’t fly the Atlantic. It was just a step too far.

So a case can be made that bankruptcy really was the turning point for Rolls. That said it doesn’t explain everything. For instance, why did nationalisation not derail things, as it did at so many other state owned firms? Well there was obviously Sir Kenneth Keith and his tough minded determination to keep politicians at arm’s length. But maybe its also because Britain is quite good at aerospace, that aircraft engines were and are a growth business, and the defence side meant that (unlike so many potentially world beating British tech companies today) a foreign competitor couldn’t simply snap Rolls up. Lots of other things came right for the RB211 too, such as the oil crisis and noise regulation which helped to telegraph its relative virtues.

We ended by nominating our heroes from the story. I picked Lord Carrington, whose intervention kept the RB211 alive at a time when Treasury voices were seeking to kill it, and Stanley Hooker, the wartime engineer who assembled his splendid “Dirty Dozen” of old engineering hands and helped to rescue the engine from death row. Neil nominated the receiver Rupert Nicholson, who held Rolls together through bankruptcy, and Keith for his expert handling of Tony Benn. Lastly we both acknowledged the extraordinary cohesive culture of the company, which endured through the darkest hours and allowed it to retain the vital skills that could otherwise have been scattered to the winds. Sue Allen, the employee whose loyalty to Rolls we noted at the start, was far from alone.

So, in summary, this isn’t a very British story in some ways: a company makes a long term risky bet on the future of its industry (in this case big turbofans, fuel efficiency, noise regulations) and instead of falling to bits it all pays off… Hurrah!

We recently recorded another episode of our series on big turning points in the lives of companies and industries. Up till now, I’ve been turning the research notes into little essays. But that proved a bit time consuming and suddenly I thought: why not just publish the notes, basically as is but slightly edited. That would speed the whole thing up!

The story of how Rolls-Royce collapsed in 1971 and what it meant is probably my favourite episode so far. If you want to listen to the podcast it’s here. It’s not only an epic story, but one that goes against the grain of Britain’s post war industrial story. For once a company with brilliant ideas and technology actually had the guts to go all out to join the big league. Oh, and it’s one of our few nationalisations that proved a resounding success.

So here is the first part of the story: how Rolls-Royce changed from being a car company, first into a car producer that also made aircraft engines, and then the most innovative jet engine maker in the world. And how it came the most almighty cropper in the spring of 1971.

The Fall

“In February 1971 I was working in Turbine Engineering in B Wing of Moor Lane. On the previous day we’d heard rumours that ‘something’ was happening because managers were being summoned to secret meetings. Soon after arrival on Thursday 4th we found the switchboard wouldn’t allow any external calls.. I can’t remember what time we heard the news of the bankruptcy that morning but I do know that not much work was done for the rest of the day… or the day after. Many of us were in denial, refusing to believe that our company would cease to exist.”

That was Sue Allen, a young engineer based in the Rolls aero engine division in Derby. And her story is typical of those employees who experienced the company’s spectacular collapse in 1971.

A few days earlier she’d joined a group of Rolls workers who’d travelled to London to appeal for a government rescue as the company slid towards bankruptcy. They had marched on parliament and attended a rally at the Friends Meeting House on Euston Road, where, among the speakers demanding action, was none other than Bristol MP and nationalisation enthusiast, Anthony Wedgwood Benn, aka Tony Benn.

Now, as the receivers arrived in Derby, a particularly sore point was the large number (around 4,000 out of 42,000 workers) of redundancies Rolls immediately announced. Selected on a “first in, last out” basis, there was also another feature: all married women were asked to leave. Sue had wed seven months earlier, so she was out on her ear, leaving the company at the end of March.

You might have thought that was that. But it wasn’t the end of Sue’s story. She went on to return to Rolls in 1974, retired in 2004 and worked as a contractor for another 15 years (so until 2019), and is still on the Rolls-Royce Heritage Trust. She later wrote (rather touchingly) of her employment: “I hope the company viewed its support and investment in my education as a profitable one.”

But first let’s rewind and look a bit more closely at what actually happened that day, 4 February 1971.

At 8.45am, the Rolls board met and resolved that, subject to talks with Lockheed and Her Majesty’s Government, the RB211 engine programme “would be discontinued, should be regarded as being suspended for the time being.” The RB211 was the large turbofan engine Rolls was developing for the American aircraft maker to power their new Tristar aircraft. The collapse was the product of massive delays and cost overruns on this project.

A letter was issued asking that a receiver be appointed for Rolls-Royce Limited. The board also requested that dealings in the company’s shares be suspended on the stock exchange.

At around the same time, the Board received a telex from Ernest Eltis, Rolls’s technical director. After an overnight performance test run, an RB211 engine 10011 of batch 3 had achieved thrust of 38,500lb - essentially meeting the power commitment to Lockheed. It was the first clear indication that, despite all the problems, the RB211 programme could be a technical success..

The immediate reaction

The first reaction to the collapse was one of profound shock. Although people in the City had long known Rolls was in trouble, the public couldn’t quite believe that the company had actually failed.

It wasn’t seen as one of the “lame duck” industries of the 1960s - car making, steel and shipbuilding - which were perceived to be in decline. “The whole country was stunned,” one government official told Fortune magazine. “It was like hearing Westminster Abbey had become a brothel.” A New York Times headline best summed it up: “Death Sentence for Symbol of National Pride”.

In Britain the newspapers warned that 20,000 jobs could be lost, and 4,000 Rolls employees were immediately made redundant (including Sue Allen), with thousands more being laid off at plants in the company’s supply chain. “We are fighting the Battle of Britain all over again,” a shop steward said at one of Rolls’s Scottish plants.

In Derby there was even a run on the local building society over fears it had invested heavily in now worthless Rolls shares. (A groundless fear; building societies couldn’t own equities, and even the proportion of Rolls employees with loans from the society (a more reasonable concern given the amount of skilled employment the company accounted for in the area) was found to be a small percentage of its mortgage book.)

Of course, there was no question Rolls could be allowed simply to vanish entirely. It did far too much critical national work, including keeping the RAF flying and the Royal Navy’s nuclear submarines at sea. Edward Heath’s government was keeping close tabs on the situation. On the day the company collapsed the PM held an emergency Cabinet meeting.

But the idea the troubled RB211 engine programme would survive was far from certain. In the House of Commons, the Minister for Aviation Supply Frederick Corfield (a Gloucestershire pig farmer who seems to have taken his cue from sceptical Treasury ministers) argued the RB211 was unnecessary for Britain to stay in the “big aero engine league”.

He questioned whether there was even a big market for the engine which, he said, was “to a large extent behind engines of equivalent power and performance” (by which he meant Pratt & Whitney’s JT9-D, which powered the Boeing 747, and GE’s CF-6).

Now as I’ll discuss, the premise of this whole story is that Rolls’s collapse was the TURNING POINT for the company. In this article I’ll chart its failure and, in the second, the amazing recovery to what Rolls is today - a £100 billion company, world number one at making large jet engines - and how the triumph in the second is linked directly to the failure in the first (indeed a necessary condition).

But first things first: how did Rolls get into this terrible situation? And how did the troubled RB211 programme come to bring it down?

How Rolls got into aviation

First, a bit of history.

When people talk about Rolls they often think first of the cars - Silver Clouds, Spirits and so on. But by 1971, motor cars were only about 10% of the company’s sales. Most came from gas turbine technology - and particularly making engines for planes.

Rolls had started making aircraft engines in the First World War - for the fairly obvious reason the government badly needed them for its growing wartime air force. Britain’s military air fleet expanded prodigiously during the conflict: from just 30 operational aircraft in 1914 to a staggering 22,000 by the end. Also if you built motor car engines, as Rolls did, turning your hand to the aircraft kind didn’t involve acquiring many new skills.

By this point, Rolls was already a company in transition. It had lost one of its two founders. Charles Rolls, an enthusiastic pioneering aviator (and holder of the splendidly named Gordon Bennett trophy for the longest balloon ascent) who had perished in an air crash in 1910 - the first Briton to do so. Meanwhile his partner, Henry Royce, never exactly hale and hearty, had started his decline into chronic ill-health, that necessitated him giving up management duties and retreating to a villa on the south coast from where he continued to send engineering drawings to HQ in Derby.

The result was that as early as 1914 Rolls was more of a professionally managed business than a family or founder-led concern; one that was increasingly meritocratic. There was however always a proviso: to run the show, you had to be an engineer. And a key test of whether you had the right stuff in the 1960s was whether you had been at Rolls between 1939 and 1945, and participated in the heroic development of the first jet engines.

In the inter war years Rolls remained a pretty balanced business - cars and aero engines (aviation had shrunk after the First World War because demand for military aircraft declined dramatically.) Even after 1939, when it was again a vital supplier to the RAF, producing the Merlin piston engine for Spitfires and Lancasters, it was possible for Rolls to ride both horses. What changed everything was the jet.

The jet race: Britain v Germany

Rolls wasn’t initially involved when jet engines started to be developed in earnest in the mid 1930s as engineers operating independently in two different (and increasingly antagonistic) countries - Britain and Germany - began unravelling the mysteries of how to utilise hot gases rather than propellers to achieve powered flight.

Jets liberated air travel from the physical limitations imposed by propellers. So in a prop engined plane above around 400mph, the propeller tips begin to exceed the speed of sound, creating shockwaves and increasing drag. This makes further gains in speed ever harder to achieve. Another limitation involves altitude: prop aircraft can only go so high because the thin air above a certain ceiling makes propellers less effective.

This then was the allure. If (and it was a big if) jet propulsion could be made to work, it offered a way to build fighter planes that could go higher and faster than anything then available - potentially a war winning combination.

Britain’s effort was led by Frank Whittle, a highly-strung RAF pilot and engineering genius whose hand-to-mouth efforts had been largely ignored by the government until the eve of the war. By 1942 it looked like Britain was losing: the Germans had successfully tested their first prototype in 1939 while it took Britain two further years to get its first Gloster airframe to fly.

The government had appointed the Rover Car Company to be Whittle’s industrial partner in prototyping the first commercial engines. But Rover had little expertise in aviation and had also fallen out big time with Whittle who, never the easiest of men, was going a bit doolally under the pressure. (Apparently he was not only smoking three packets of cigarettes a day, but was suffering from various stress related ailments including severe headaches and insomnia, eczema and heart palpitations. To cap it all, he was sniffing benzedrine to sustain his 16 hour workdays and then taking tranquillisers at night to help him sleep.) When Rover’s parts didn’t work properly he accused them of sabotage, being in German pay and so on. Rover was getting very fed up.

Enter Ernest Hives, then the General Manager of Rolls. He’d been told about the potential of these wonder engines by his chief designer, Stanley Hooker, who’d been involved in trying to help Rover construct a workable compressor. Hives came to think that unless Rolls stepped in and helped, the Germans were going to win the jet race and - if they did - the Allies would be defeated in the skies.

In late 1942, he went to see Sir Stafford Cripps, Minister of Aircraft Production, to demand that Rolls be drafted into the jet project to speed it up. New to his ministerial post, Cripps didn’t appreciate this aggressive lobbying and told Hives to stick to building Merlin engines. At which the Rolls boss basically had a meltdown.

‘But people are dying needlessly!’ he erupted. ‘It’s wrong, I tell you. It’s stupid and it’s wrong, and I won’t let it go on like this. There should be questions in Parliament about it, headlines in the papers. The world should know that this country is making a mistake, and I shall see to it that the world does know, I shall see to it myself! We’ve got to have jets or more people will be killed. And they’re young.’

Whether Cripps was moved by this peroration, history does not relate. But he gave way on Rolls’s participation and the upshot was the famous “five shilling dinner” involving Hives and Rover’s chairman Spencer Wilks in December 1942 at the Swan and Royal Hotel in Clitheroe, Lancashire. In exchange for a Rolls tank engine plant in Nottingham - Rover handed over the entire jet project, including the development facility at Barnoldswick in Yorkshire. (The “RB” in the RB211 designation stands for Rolls-Royce Barnoldswick.)

The rest is history. By 1944, Rolls had produced its first jet engine, the Welland, which went into the Gloster Meteor aircraft. And while the Germans did introduce the jet first - if only by three months - by then they they were running out of pilots and petrol. Their engines were also pants - largely because they didn’t have the right materials to cope with all those hot gases. And when the war ended, they lost their lead entirely when the whole German aviation industry shut down for ten years.

More bigger jets and into civil aviation

The jet changed everything. For one thing, Rolls now needed a whole new approach to manufacturing as jets couldn’t be built on similar production lines to piston engines. Pretty quickly, the aero engine business mushroomed and, increasingly, Rolls’s fortunes were tied to sales of these machines.

In the 1940s and 1950s, that wasn’t a bad place to be. A time of hectic growth - first in military then in civil aviation - it saw manufacturers racing to make aircraft engines more powerful. So while the tiny Welland produced 1,600lb of thrust in 1944, just six years later Pratt & Whitney (which entered the jet business using technology developed by Rolls) had produced the J57 to power the first jet bombers. This generated 10,000lb of thrust - more than six times as much.

But developing jets was a risky and expensive process. When P&W launched the J57, an executive noted the engine took 1.3 million man hours of design time; roughly twice that taken for a piston engine. It also cost a massive $150m to develop - roughly $2bn in 2025 money.

American manufacturers were advantaged in they could call upon the vast resources of the Pentagon, and (later) the massive US market for civil aviation to absorb the costs involved. Rolls had a much smaller national customer base, and while initially it could sell its products to other European countries, many were recovering from the war and, to the extent they had ambitions in aviation, harboured hopes of building jets themselves.

But all this was in the future. Early jets had only military customers. Once they’d seen the performance, air force commanders (at least in the major military powers) were keen to yank piston planes from front line squadrons and replace them with fast jets as quickly as possible.

Airlines however proved much warier. In his history of commercial aviation, T.A. Heppenheimer explained their misgivings: “Airlines as commercial enterprises in no way could accept the cost and performance limitations of jets. The jet engines that the airline executives knew about were still fuel guzzlers, and the aircraft they powered would still be limited in range. And jet engines of the day still demanded much maintenance, which would drive costs even higher by taking aircraft out of service, reducing their ability to generate revenue.” Ralph Damon, president of TWA, put it succinctly: “The only thing wrong with the jet planes of the day is they won’t make money.”

To turn these machines into viable power plants for airlines, manufacturers needed to do three things - none of them cheap or easy:

• Make them much more reliable (in the Korean War, almost a quarter of the 1,100 planes the US air force lost were due to engine failure and a government report said “Pilots were no longer keen to join the Air Force if they had to learn to fly fast jets”.)

• Increase the power output so you could carry sufficient fare paying passengers to make the exercise profitable for the carrier.

• Make them much more fuel efficient so the aircraft had greater range.

It was Britain that made the daring first leap into civil jet aviation, launching the De Havilland Comet, the world’s first passenger jet, in 1952. While too small, underpowered, and ultimately unsuccessful, the Comet proved that jets could be extremely popular with customers. Until they started crashing due to structural failures, Comets were often full.

The Americans really picked up the baton when Juan Trippe’s Pan Am made the first big move, becoming the launch customer for the 150-seat Boeing 707 jet, which entered service in 1958. Douglas followed with its first big jet plane, the DC8, the following year. These aircraft were powered almost exclusively by the emerging US duopoly of Pratt & Whitney and GE.

The “American strategy”

As civil aviation grew, airlines wanted bigger aircraft to exploit the economies of scale. And by the early 1960s, Rolls realised the next generation of aircraft was going to be roughly twice as large as the existing one. If they wanted to stay at the apex of jet aviation they needed to produce more powerful engines. At the time the biggest Rolls engine in service was the Conway, which produced thrust of around 17,500lb. The next generation would need twice that - around 30,000 to 40,000lb.

These new large aircraft were certain to be American made. Both the money and the market were overwhelmingly on the other side of the Atlantic. In the late 1950s, around 85% of the world’s civilian aircraft were built in America, and around 65-70% for US airlines. (The Airbus, which would become Europe’s competitor, was still only a distant dream.)

What Rolls needed was an American aircraft maker to snuggle up to. This was known within the company as “the American strategy” and it became an increasing obsession as the 1960s dragged on.

Rolls thought it had the technology to succeed: the turbofan, a leap forward in engine design and one of Whittle’s many brilliant ideas it had pioneered commercially. So, early jet engines (or turbojets) were like a tube, sucking all the air in the front and pushing it through a compressor, before mixing it with fuel and igniting it, and then squelching it out of the back as a stream of hot gas.

Turbofans introduced a key difference: they mounted a large fan on the spool at the front that had a larger diameter than the rest of the engine. As air was sucked in through the fan, some of it flowed round the side of the engine rather than through it. The proportion going round the side is called the “bypass ratio”.

The bypass effect offered advantages that would help to make big planes economic :

• More power and more range. While turbojets eject gasses at high speed, jet engines are at their most efficient when the exhaust stream is as slow as possible. The (slower) air going round the sides of a turbofan slows the whole exhaust stream, which has two beneficial effects: it increases thrust and makes the engine more fuel efficient.

• Slow air also makes less noise. This would become an important criterion both for consumer choice and regulation.

Rolls actually had a lead in turbofans having developed the world’s first commercial model - the Conway - in the late 1950s. But it was relatively small and had a by-pass ratio of just 0.3:1. Most large turbofans now have a bypass ratio of 10:1. By the mid 1960s, the Americans had not only caught up - helped by bulging US military budgets - but had produced larger versions with higher bypass ratios and much more thrust.

Rolls knew that to succeed with the American strategy they would not only need a so-called high bypass ratio turbofan, but one that was so advanced - and had such advantages over their US rivals - that a big plane manufacturer would go for it despite them not being American. This is what became the RB211.

Selling the dream

In the mid 1960s, the big jet revolution took off. Boeing started to develop the first “widebody” passenger plane - the 747 - which would enter service in 1970. Rolls duly made a pitch to supply the engines but Boeing went with P&W, as it always did.

Then, all of a sudden, several interesting opportunities opened up:

• McDonnell Douglas announced it was building the DC10 (a three engined plane as opposed to the 747’s four)

• Lockheed, which had dropped out of civil aviation in the late 1950s and had never built a jet passenger airliner, proposed a return with the Tristar (also, as the name suggests, a tri-jet)

• There was an intriguing new European option - the Airbus A300 - a large twin jet that would consequently require more powerful engines. But this remained at a conceptual phase, and would be built by a consortium of European aircraft manufacturers that was still at the stage of composing itself.

Of the three, the nascent Airbus consortium was by far the keenest. The Europeans actively wanted to buy a large (50,000lb of thrust) version of what Rolls was calling its “Advanced Technology Engine” (the larger version was dubbed the RB207 and the smaller, producing around 30,000lb, the RB211).

But here the American fixation kicked in. Rolls had to have its American partner and as selling to Airbus would have required it to build two designs -the RB207 and the RB211 - it decided to say no on grounds of financial risk.

Airbus were furious. The A300 programme manager Roger Beteille said that at a stroke Rolls had made the project unviable; leaving them the choice of stopping or producing “a superb glider”. The decision would also have long term commercial consequences. It not only led Rolls to miss out on sales of the A300, which proved in the long term a much bigger seller than either the DC10 or the Tristar; the company would not sell a single engine to Airbus until 1995.

But Rolls’s top brass were totally convinced the American strategy was the right one. In early 1967, the company’s chairman Denning Pearson had presented market analysis to the board showing that sales of its five engines in production would fall dramatically from £58.9m in 1969 to just £3.5m in 1975.

This was due to poor sales prospects for many of its existing engines. Only the Spey would continue selling well, and even its prospects were limited because of weak demand for Rolls powered British aircraft such as the Hawker Siddeley Trident, the BAC1-11 and Vickers VC10) in a market increasingly dominated by Boeing. Convinced that retreating mainly to short haul and executive jets would condemn the aero engine business to decline, he urged his colleagues to embrace competition with the big two US makers - P&W and GE.

A distinguished engineer who had run a Merlin production plant during the war, Pearson had taken the company into civil aviation in its aftermath. He explained his thinking: “It appeared to us that if we were to stay in the civil business which we had built up since the war, we needed to get into a United States-built airframe as first engine choice.” He added: “Building a new engine would not guarantee that we stayed in business. Not building one would guarantee that we went out of business.” Years later an official report into the company’s failure would flag up this fixation, noting that “a contract with a US manufacturer had become for them not merely a consummation devoutly to be wished, but a task in which failure would be unthinkable.”

It was with this mindset that Rolls went in to bat with McDonnell Douglas and Lockheed.

The “Advanced Technology Engine”

The engine Pearson was touting was the “Advanced Technology Engine”; a pioneering design, and much more advanced than its US rivals - being potentially quieter, more fuel efficient and more reliable.

This was achieved by adding an extra gizmo to the turbofan - a third engine spool, which not only allowed the turbine and compressor to run at different speeds, but optimised the compressor by running different parts at varying rotational speeds. It made for a shorter engine which was more rigid and hence more reliable. One downside was that it added weight but Rolls planned to get round this by using composites to make the fan blades.

What was clear was that there was a lot of technical risk involved in building the RB211, with the potential that it might take longer and cost more than anticipated. Which brings us to the other big constraint Rolls was under: its finances. A few years before, in 1966 it had bought out a rival, Bristol Siddeley Engines, and taken on a lot of debt to do so. (The acquisition was defensive: to stop Bristol partnering with P&W to offer a “European” version of its large turbofans.) So on top of all the other challenges, Rolls was strapped for cash.

The deal

The DC10 beauty parade came first, and Rolls didn’t win it. That left only the Tristar, and it is hard to overstate how hard they fought: Pearson and the aero engine division boss David Huddie almost lived in the US for the year during the interminable pitches and negotiations. Basically they won on price - offering a figure that GE (who were trying to get both the Tristar and the DC10) couldn’t match. In March 1968, the deal was signed.

The general mood was they had secured the company’s future. The press were ecstatic. The RB211 was supposedly Britain’s biggest ever export order. The Observer estimated that - if you bundled in future parts and servicing sales - it would bring in £1,000m and boost the balance of payments. Both Pearson and Huddie were knighted.

In reality however they were taking a massive risk. Lockheed was the weakest partner they could have gone with; more like a start up coming back into commercial aviation after a decade out. It also had the thinnest order book - selling to the weakest two (Eastern and TWA) of America’s big four airlines.

Because the American Strategy was such an article of faith, and the Tristar was its last chance, the contract was extremely favourable to Lockheed. It offered the company fixed price terms for deliveries out to 1971, and thereafter pre-agreed escalations to 1974. Rolls could only raise its prices outside this framework if Lockheed hiked the price of the airframe. There were also severe penalties if deliveries were late, and binding guarantees surrounding performance and weight.

Funding the RB211

Rolls weren’t stupid. They knew that launching the RB211 would take more capital than they could sensibly spare, costing an estimated £65m to develop - equivalent to £1bn in 2025 money. So they applied for “launch aid” - essentially a soft loan from the government repayable through a levy on engine sales once the manufacturer had recovered its development costs. Normally the top limit for launch aid was 50%, but they managed to wring 70% out of the government (partly because the minister they negotiated with was aviation enthusiast Tony Benn).

At this stage launch aid, which had been introduced in 1960, didn’t take account of the risk that a single launch-aided project might sink the company doing it. That would later change!

So, to summarise, Rolls embarked on the RB211 project with:

• A technologically advanced and unproven new design

• A weak balance sheet

• A partner that was coming back into civil aviation and whose Tristar had been ordered by Eastern and TWA - the weakest of the “Big Four” US airlines

• A very demanding contract driven by management’s central objective of hooking a US aircraft maker..

What could possibly go wrong?

The crisis strikes

The contract was signed and the Launch Aid raised. Rolls assembled a team of 700 engineers to develop the RB211, working from an initial design completed in October 1967.

Almost immediately things unravelled. The company’s top engineer, Adrian Lombard (one of the wartime designers who’d developed the Welland) died of a brain haemorrhage, leaving the team without its top trouble shooter. The technology threw up heaps of problems - the engine didn’t produce enough thrust, the combustion chamber became too hot and was prone to melting. One of the key innovations to keep weight down - using composite material for the fan blades - couldn’t resist bird strikes.

By 1970 the RB211 was a year behind schedule and, given additional development costs and the stringent terms in the contract, it became clear Rolls would struggle to make any profit on each engine sold. Essentially the company was now in a kind of doom loop. With each technological bump in the road the cost of the project rose ever higher, but Rolls couldn’t raise the price of the engine to compensate. It was also clear the RB211 would be late and might also fail to meet the performance requirements laid down by Lockheed. Either exposed Rolls to oceans of damages on top.

One solution might have been to sit down with Lockheed and renegotiate. But here lay a further wrinkle: the costs of the Tristar itself (which Lockheed had also sold to carriers on price) were threatening to bankrupt the American company. So it was unwilling to make any concessions on the Rolls deal that might add to its own difficulties.

By the summer of 1970 Rolls was in deep crisis. The straw/camel’s back moment came with the realisation in March that the composite fan blades wouldn’t work. Rolls had to go back to heavier titanium, involving further costly re-engineering. Huddie went on sick leave in June after a nervous breakdown, and in September, Rolls finally fessed up that development costs had more than doubled to £170m. It didn’t have the money to build the RB211, which still didn’t work properly.

Rolls needed more launch aid from the government (breaching a supposedly iron rule that the UK never chipped in for overruns) which the government reluctantly granted in the autumn subject to a bunch of conditions. Among them was a change at te top and Pearson was demoted to deputy chairman. Lockheed was aware of the crisis - but remained sanguine. Its top brass concluded that Rolls was so important to the UK that the government would surely bail it out.

Yet while that might have been a reasonable assumption before Britain’s general election in June 1970, it was now questionable. Edward Heath’s new government had been elected pledging to let the market do its work and not to protect so-called “lame ducks”.

On 18 January 1971, the Rolls board learned the costs had risen again to £203m. The need to redesign some components would impose a further 6-12 month delay, putting the company in line for massive penalties. Meanwhile the banks were threatening to pull funding because the numbers looked so bad. Both Rolls and Lockheed looked to the other for concessions, but neither wanted to make the first move.

At the start of February, the government finally lost patience and pulled the plug. No sensible concessions having been offered by Lockheed, they would not support Rolls further and the company filed for bankruptcy. (The concessions the government wanted were: abandonment of the existing contract, a 12 month delay and a 33% increase in the engine price).

Conclusion

We end up on 4 February with all the classic hallmarks of a British industrial disaster; one where a famous old company overextends itself, goes bust, and either disappears entirely or emerges a shadow of its former self. That was certainly the expectation at the time.

An official report from 1973 cast all the blame on the management. They had underestimated the size of the engineering task and the resources required to deliver. Also of entering into a ranch-betting contract on unduly tight terms. I’ll come back to that judgment later on.

But what we know is that 1971 wasn’t the end of the story. Rolls recovered and is now the world leader in the same big jet industry (and especially the bit of it that the RB211 was designed to take it into). So in part two I’ll look at how Rolls came back.

We recently released the second in our “Turning Point” series, which looks at key events that changed the course of a company or industry’s history, transforming it for good or ill.

This time we examined the story of Eastman Kodak, a company totally steeped in a culture of relentless innovation, which stumbled when faced with the challenge of the digital camera - a device it invented itself. You can listen to the episode here.

Founded in 1881 by George Eastman, a bank clerk turned camera enthusiast from Rochester, New York, Kodak pretty much single-handedly created the consumer market for photography. Eastman started out selling a machine that coated dry plates - the somewhat cumbersome process by which photographs were made at that time.

Pre-prepared plates was certainly an improvement on the previous method, which required the photographer to do his own chemistry. But Eastman wasn’t satisfied with the idea of customers having to lug around bulky plates. He started experimenting with rolls of flexible film and in 1885 received a patent for this newfangled medium.

The first Kodak camera in 1888 was sold pre-loaded with enough film for 100 exposures. Once the pictures were taken the owner would send the camera back to Eastman with $10 and the company would develop the film, insert a new roll, and send it back.

By making camera equipment more portable and separating the process of photo taking from film development, Eastman admittedly took an axe to his dry plate business. But he made photography more accessible and hence vastly more popular. Meanwhile his innovative new apparatus made sure that Kodak was the principal beneficiary of the resulting craze.

Mass adoption drove down the costs of Eastman’s camera. Having sold for $25 at the time of its invention, the price fell to $5 by the mid 1890s, and just $1 by 1901 when the Brownie was introduced. (Even better, as rivals copied Eastman’s new system, their customers had to buy his film rolls - increasingly the real engine of Kodak’s profitability.

What about the name Kodak? Where did that come from? Well, in addition to being a superb innovator, Eastman was also an enthusiastic marketeer. He devised the name because he wanted one that would work in any language, was entirely invented so would be easier to trademark, and started and ended with a “k” - a letter he liked because he thought it strong and incisive. Oh and he also invented the company’s slogan which lasted for decades: “You press the button; we do the rest.”

In his five decades at the helm of Kodak, Eastman stamped his mark firmly on the company. It remained at the forefront of research and development, opening one of the world’s first corporate research laboratories just before the first world war. Kodak continued to push at the frontiers of photography, developing colour film between the wars, and once again sacrificing its highly profitable and market-leading black and white franchise, while successfully supplanting it with Kodachrome, the colour films it continued to sell until 2009.

And it was at Kodak’s Rochester research labs in the mid 1970s, that a young electrical engineer, Steve Sasson, made an invention that would ultimately prove as earth shaking as that first Box Brownie.

When we spoke to Steve about his Kodak career, which he started fresh out of college in 1973, he explained he wasn’t a typical recruit: “Traditionally, Kodak hadn’t hired many electrical engineers because fundamentally the company was based on mechanical engineering.”

Sasson was put in the apparatus division laboratories, which focused on equipment design rather than the core film business. And it was here one day, at the conclusion of a project, that his supervisor Gareth Lloyd asked him to fill in time “until something useful [came] along”. One option “was to do some modelling on exposure control systems for Excel movie cameras,” Sasson recalled. “The other was to look at a new type of imaging device that had just been commercially made available, called the Charge Coupled Device.” He chose the latter, and thus opened the door to the asteroid that would ultimately crash into Kodak: the handheld digital camera.

It’s worth noting upfront that Sasson didn’t develop the first digital camera. This already existed, albeit not in consumer form. For instance there was the Multi Spectral Scanner, a sophisticated imaging device mounted on the Landsat 1 satellite, launched in 1972, which was used to take pictures of earth from space. But these were multi million dollar bits of equipment, far removed from consumer products.

What made such cameras possible was a type of integrated circuit called a Charge Coupled Device, or CCD, that had been invented in Bell Labs in 1969. By 1974 these had been commercialised by Fairchild Semiconductor and Sasson’s first task when given his project was to buy one - but only one mind “as they were kind of expensive, they were several hundred dollars,” Sasson recalled.

A CCD is basically a light-sensitive integrated circuit that acts as an image sensor, converting photons (light) into electrical charges to create digital images. It works by using tiny pixels (capacitors) on a silicon chip to trap electrons generated by light, creating a proportional charge for each part of the image. The CCD transfers these trapped charges across the chip to a readout register, which counts the amount of charge from each pixel and converts it into a digital value that a computer uses to build the final image.

The first CCDs were pretty primitive. The one Sasson used for his camera had just 100 x 100 pixels (or 0.01 megapixels). Compare that to the latest iPhone which has 8,000 x 6,000 pixels (48 megapixels). Yet while the picture quality was far too poor for any consumer product, the potential was clear. “I remember reading about this revolutionary device in the New Scientist,” Sasson recalled. “It just seemed to me that given the fact a year or two earlier you couldn’t have got anything like 100 times 100 pixels on a chip, that in a few years time it’ll be 1,000 times 1,000. And then there is no real limit to how good the definition can get.”

Initially, Sasson played around with the device. But then he started to think, wouldn’t it be cool to build a portable camera: “Initially I thought I have to capture an image with this because that’s what it’s supposed to do; a two dimensional charge pattern.” Next he figured out how to store the image by turning it into numbers. “Then I thought, well, what I’ve got is an electronic version of an optical pattern or a digital image. And I thought: wouldn’t it be great to put it into a portable package?” Finally he built a playback device that allowed you to store the image on a cassette and then press a button to show it on a screen.

The machine Sasson finally unveiled to bemused colleagues in December 1975 looked a bit like a toaster and weighed eight and a half pounds. But rudimentary as it was, it was a demonstration of a complete photographic system. “Anybody in that room could have picked up that camera, taken a picture, put the tape in the machine, pushed the button, and could have seen that picture,” Sasson said.

The invention caused a stir and Sasson was subsequently invited to show his machine to a host of managers and engineers. “There was a lot of technical curiosity, but there was also caution about - don’t extend this idea too much into things,” Sasson said. “Because what I was basically proposing was a replacement for consumer photography.”

Now here we come to the story that’s conventionally told - that Kodak essentially told him to put a sock in it: film photography was too valuable to risk with some newfangled digital method. It’s even said that one Kodak boss told him: “That’s cute; just don’t tell anyone about it.”

But that’s not exactly how Sasson remembers things. The question he remembers most being asked was not “how does it work?” but “why?” Why make the switch from one technology to the other? What problem was digital photography solving? In what way would it make things better for consumers, and what might drive them from chemical film to digital? How long before the mechanism produced a similar-to-film quality image?

“They didn’t know how far this would go, but they were certainly aware that any discussion publicly would create some concern and uncertainty around the photographic system that we had today,” Sasson said. Even so, Kodak certainly didn’t put its head in the sand. “They thought about it immensely,” Sasson said. “They spent a whole lot of money and R&D on this stuff. What really made an impression on them was the idea that a 25 year old kid with a soldering iron and a voltmeter could build this thing out of parts he found around a laboratory.”

In 1978, Kodak patented Sasson’s application, and in 1981 Sony introduced the first electric camera - the Mavica. It wasn’t a digital camera; it recorded analogue images on a floppy disk which were displayed on a TV screen. But it showed the way the market was potentially going, and Kodak commissioned a study from its head of market intelligence, Vince Barabba, about the future of photography.

This concluded unhelpfully that there was good news and bad news if you relied on film for your business. The bad news, Barabba thought, was that digital would eat film in the long run. But the good news was that Kodak had at least a decade to prepare. In some ways the Mavica actually revealed the problems that lay ahead: image quality was poor, digital photography equipment was still expensive and there were unresolved problems with the interoperability of various components, such as cameras, displays and printers. Until these could be resolved, the physical album ruled supreme.

Despite the apparent let off, Kodak continued wrestling with the digital problem. The first CCDs were black and white, but in 1976, Bryce Bayer, a scientist working at Kodak Research Labs invented a filter that allowed digital shots in colour. (His filter remains a key component in all smartphones today.) There was a sustained R&D effort. “We weren’t public about it but we were spending a lot of money developing megapixels, encoding algorithms, compression algorithms, and leveraging the technology that was being developed for other fields like microprocessors and DRAMs and the like,” he said.

Kodak knew that to get to 35mm film quality, they would require multiple megapixels. So they started to build megapixel sensors, producing their first one in the mid 1980s. They even demonstrated it at a show because “people were losing confidence that Kodak was working on this.” What’s certainly true is that much of the intellectual property that went into digital photography originally came from Kodak. “We were absolutely not asleep,” Sasson said.

In the late 1980s, Kodak developed the first digital SLR (single lens reflex) camera that used one of the company’s megapixel sensors with image compression similar to the JPEG technology in use today. It was way ahead of its time and the marketing people were cautious. “They said: ‘We don’t want to do this. Why should we sell this? Where’s the business model that replaces the one we have?” recalled Sasson.

The problem was that digital required Kodak’s elephantine business - still dominated by chemical film - to pirouette like a ballerina. Kodak was still huge in the 1980s - with 140,000 employees, a $20bn-plus market cap, and a huge global market share in film sales. At their peak, gross margins on film were running at close to 70 per cent. Lacking George Eastman’s daring, the digital opportunity didn’t seem big enough for the board to bet the farm.

So instead Kodak hedged its bets. There was a strange $500m diversification into a film system called Advantix which imported some digital features but dodged the fundamental switch. You could preview shots on the camera and indicate how many prints you wanted but it still used Kodak film. Advantix was a flop and the $500m had to be written off. In 1989, playing on its chemicals expertise, Kodak tried to diversify into pharmaceuticals, spending billions on a company called Sterling Drug. But that didn’t work either and Sterling had to be broken up and the bits sold off. Afterwards, a new CEO, Kay Whitmore, promised that in future Kodak would stick to its core film and photographic equipment business.

In the early 1990s, Kodak finally introduced its first digital camera - the DCS. But it was very expensive and had a limited market. And while the company messed around with Advantix, Sony and others introduced the first real mass consumer digital cameras. For a long time however, it didn’t seem to matter too much, chemical sales held up, only peaking in 2001. The following year, digital camera sales overtook film ones for the first time.

The end, when it came, was swift and crushing. Between 2002 and 2006, film sales fell by 60%, crushing Kodak’s profitability. (And that despite the fact that the digital screens weren’t great, there were only websites like Flickr to share snaps on electronically, and printers remained inadequate.) To survive Kodak had to cannibalise itself, first selling its imaging division (basically X-ray machines) in 2007; and then flogging its patents - the product of decades of Eastman inspired R&D.

What really killed the business though was the smartphone. “Yeah, the smartphone encompassed not only the image capture but the printing aspect,” said Sasson. “When they all get encompassed in a single device there’s a massive disruption to your whole business model.” More significantly, iPhones also changed the way people thought about photography as a pastime. It was no longer about preserving memories (or “Kodak moments”, as they had once been known) but rather about sharing immediate experiences - a development for which smartphones were perfectly designed but for which Kodak had no answer. It turned out that Eastman’s success hadn’t been built on the super-profitable film business after all. In reality he’d created a sticky social pastime which allowed him to make money from selling high-priced film. The smartphone transformed that pastime. In 2012, the company filed for bankruptcy.

Looking back now, Sasson thinks it’s hard to see how his old employer, with so much profitability riding on consumer film, could have foreseen the radical shift and pivoted in time.

He recalled coming to London a few years ago to win an award for his innovation. There he met a fellow winner, Martin Cooper, the Motorola employee responsible for placing the first mobile phone call around the time Sasson was building the first digital camera. “We were trying to figure out which was the worse company to work for because they were both resistant to accepting new ideas,” Sasson said. “But neither of us ever thought about combining cameras and telephones. It’s an evolution I don’t think anyone anticipated but it turned out to be exactly what the public eventually would adopt.”

Kodak is still going, and recently reintroduced Kodachrome colour film, sensing a traditional photography revival. But many see its fate as a cautionary tale - of how an entrepreneurial dynamo became a dinosaur, too complacent and wealthy to see the writing on the wall. Sasson sees it differently. “I think it’s actually a really bad characterisation to call Kodak a dinosaur,” he said. “They were inventing and developing many of the things we take for granted today in terms of the digital architecture we all use. They just couldn’t see any way to make money out of it.”

Here’s the link to the first episode if you want to take a listen.

I thought I’d write up these stories on Business Adventures as they are really precisely that - adventure stories. We start with ASML, the biggest European company you have never heard of, and pretty much the Continent’s only tech giant.

So how did this anonymous Dutch business come to be the world’s biggest producer of machinery used in printing integrated circuits on silicon chips, and a company worth around $500bn? Well, it’s a complicated story and one that involves American scientists, all conquering Japanese exporters, and rednecks beating the crap out of Toyotas with sledgehammers.

Perhaps the best place to start is with the development of the transistor in the 1950s, an event that kickstarted what we know as Silicon Valley. Its inventor, William Shockley, set up a company to exploit his brilliant idea. But he was apparently a man so tyrannical and impossible to work with that his top employees (known by him as the “traitorous eight”) all scarpered, setting up a competitor, Fairchild Semiconductor.

From there the eight and the people they employed went on to found an astonishing number of key US tech firms including Intel and AMD - a constellation referred to as ‘the Fairchildren”..

The first semiconductors were primitive: technologists only had the ability to fit a few transistors onto an integrated circuit board. But it was clear the more you could squeeze in, the more tasks the chip could perform - and perform them more quickly. So people set to work ingeniously shrinking the circuits. And a few years later, in 1965, one of the early pioneers of semiconductors and a traitorous Shockley refugee, Gordon Moore, made the empirical observation that the number of transistors on a microchip doubled roughly every two years, while costs halved.

“Moore’s Law” has proved a remarkably durable predictor of the pace of development. And key to it was how small you could make these circuits on a silicon wafer. Their size has shrunk from 10,000 nanometers in the 1970s to 20-60 today, allowing Moore’s Law largely to hold good. It was in this constant search for shrinkage that ASML came to play a key role.

So, let’s head back to Holland and the company. The man behind ASML was a Dutch engineer and entrepreneur called Arthur del Prado. Given he is pretty much the alpha and omega of the European tech industry, it’s amazing he’s not better known. (I couldn’t find an obit in either the FT or the WSJ from when he died, aged 85, in 2016). Yet to give a sense of his importance, he created three tech firms, all involved in chipmaking equipment. Today they have a collective value of almost a trillion dollars.

Born in 1931 in the Dutch East Indies, del Prado went to America in the 1950s to study at Harvard. For some reason he never completed his degree but instead headed west to California, where he joined one of the semiconductor companies founded by a former employee of Shockley’s (but not one of the celebrated eight) - Knapic Electro-Physics.

Returning to his homeland in the early 1960s, he initially marketed US technology in Europe, but in 1968 established ASM (Advanced Semiconductor Materials) to make machinery to produce silicon chips.

Fast forward two decades and, having built ASM into a successful (if still quite small) business, del Prado was on the lookout for new opportunities. This was a time when European governments and the EU were trying to encourage companies to invest in technologies of the future; one of which was semiconductors.

There was only one Dutch company making silicon chips then: the giant electronics firm Philips. And because of its involvement, and the need for faster, more capable chips to stay competitive, researchers from the company had been looking into lithography, and had developed a novel process which both allowed circuits to be imprinted more quickly, and also spoiled far fewer of the silicon wafers than previous techniques, thus dramatically increasing yield.

There was one big obstacle: perfecting it and winning customers would take a lot of capital and marketing nous, which Philips, already a sprawling conglomerate, was reluctant to commit. It needed either to find a partner or shut the project down. Del Prado’s company ASM was just down the road from Philips’ base in Eindhoven and he made it his mission to persuade his giant neighbour to share its technology with him. At first Philips was dubious: it didn’t think ASM was big enough to be a credible partner. But Del Prado persisted and on 1 April 1984, he was triumphant. ASM and Philips created a joint venture company called ASM Lithography. ASM contributed $2.1m in capital, while Philips’ contribution was its intellectual property. Four years later, ASML was spun off by the partners as an independent listed company, which it remains.

The timing was propitious. From the 1960s scientists had been perplexed by what they saw as the fast approaching limit to the further shrinkage of integrated circuits. Lithography worked by projecting light onto a silicon wafer: the beam shone through a mask and essentially imprinted the circuit into the silicon.

The problem with so-called optical lithography (you’re using normal light, albeit at low end of visible range spectrum) was that at a certain point, you run into a problem called diffraction. Miniaturise too much and diffraction of light as it passes through the mask makes the patterns become blurry and indistinct. So the race was on to find a better form of lithography..

Scientists investigated several alternatives. One involved the use of electron beams, which IBM developed from the 1960s to the 1990s. The basic problem with this was that it was very slow (speed of production is another requirement of Moore’s Law) so while developers used it to make prototypes, they didn’t use it for production.

A second involved using X-rays - really short wavelength light. IBM also looked into this over roughly the same period it was pursuing electron beam lithography, along with a host of other US researchers: Bell Labs, Hughes Aircraft, and Westinghouse. X-rays were a better option. The main drawback was the expense involved in producing the light source; you needed synchrotrons which cost $25m a pop.

Fortunately given the complexity involved in replacing optical lithography, the scientists got cracking long before the optical way hit the buffers. Moore’s Law just about kept on going, partly because the industry proved more resourceful than expected at continuing to advance with natural light. It found ways to work round the diffraction problem, or at least to push out the limits. Indeed there is now a second law called Sturtevant’s law, which says that “the end of optical lithography is 6-7 years away and always will be”.

In our podcast, which we made with Brian Potter, an academic from the Institute for Progress, a Washington based think tank, Brian goes through the various twists and turns in the process by which researchers groped their way towards cracking the problem of X-ray lithography. I won’t go into all the details, it’s too complicated and involves incredibly sophisticated mirrors, but you can read his brilliant account here

The long and the short of it is that what emerged in the 1990s was a whole new way of doing lithography - Extreme Ultraviolet (EUV) Lithography. And the centre of this research was not Holland: it was the United States. (Although both Japan and the EU had their own publicly funded programmes.)

That it was in America was slightly unexpected, as by then the US had almost entirely vacated the market for lithographic machines. As in so many other areas, it had been supplanted by Japanese firms, and by the 1990s Nikon had around 40%, and Canon another 30%, of the global market. The only other significant player was ASML with about 20%.

The impetus for the research came from the customers - US chipmakers which saw EUV as the way to gain an edge. They briefly roped in the US government to fund it, until that was scrapped by Congress in 1996. Whereupon Intel took the financial lead, channeling huge sums into a research consortium called EUV-LLC, with the support of Motorola, AMD, IBM and Micron. And finally, by the turn of the Millennium, they were able to show that EUV was commercially viable.

At which point EUV-LLC was faced with a dilemma: the IP they had developed was supposed to be reserved for US industry, but there wasn’t actually an American-owned business that could take it on. One lithography company, Silicon Valley Group, was up for sale, and Washington was trying to stop it being bought by the Japanese. (In the end ASML bought it.) Another, called Ultratech, had already decided not to pursue EUV.

The obvious companies to commercialise the project were the Japanese. But at this point American xenophobia came to the fore. Remember those images from the 1980s, of American workers, and sometimes auto industry bosses, bashing up Toyotas and Nissans with sledgehammers? A mood akin to this gripped the US chipmakers in EUV-LLC and officials in Washington. They were damned if they were going to give the Japanese the means to sweep them aside again.

Canon and Nikon were firmly cold-shouldered. Nikon decided against pursuing a participation in EUV-LLC because there was so much controversy. And when Canon did apply its application was promptly vetoed by the US government. Which left only one credible corporate participant standing - the unthreatening and herbivorous Dutch corporation ASML (which helpfully acquired Silicon Valley Group in 2000, keeping it out of the hands of the Japanese). Not being Japanese, ASML wasn’t seen as a strategic threat (basically, it wouldn’t advantage Japanese chip makers against the Americans). Meanwhile Intel’s main focus was on ensuring it had access to the next generation of lithography tools, and it was Intel that argued that ASML should - even though it was foreign - get a crack at a license. So in the end ASML, in partnership with Carl Zeiss, became the only company to cross the finishing line.

Fast forward two decades, and it’s clear that was a brilliant move for the Dutch firm. From a market cap in 2000 of just $7.5bn, its value has grown 66-fold to its current level (much of that coming after 2010, which is when the first EUV machines started to be introduced commercially). Meanwhile, its Japanese rivals have lost their edge. Nikon’s market cap has fallen from $5.3bn in 2000 to just $3.9bn today (partly, of course, that’s due to the smartphone). And while Canon has done better, its market cap at $29bn is a fraction of ASML’s.

It’s clear that despite its somewhat anonymous presence in the market place - there is no Dutch Jeff Bezos or Elon Musk hogging the headlines - ASML is a brilliantly run company. This article gives you some sense of how the management system works.

In Britain we often curse ourselves for not being able to capitalise commercially on our innovations. Were the Americans happy about giving the big prize to the Dutch, we asked our guest Brian Potter, or did they have misgivings about how it had all worked out? He said the disappointment was more around what happened to Intel (another company whose value is today dwarfed by ASML).

“Intel invested all this time, money and effort to get this technology up and running and commercially viable,” he said. “But then in the 2010s they decided not to take the next step with it, and stick with optical lithography. They figured at the time EUV was not quite ready. TSMC, the Taiwanese chipmaker, ended up making the big bet on EUV as the next generation technology. And that was part of what let TSMC leap ahead of Intel.”

I recently wrote an article for the Telegraph about a series of unexplained electrical substation fires across the country. There had been eight in the space of a few weeks between March and mid May, when normally you might expect one or two every few years.

The James Bond baddie Auric Goldfinger (played in the film by the great Gert Frobe) had a simple rule of thumb when it came to how many times seemingly innocent mishaps could recur before he perceived a more sinister pattern. “Once is happenstance; twice coincidence; three times is enemy action,” he said.

So was it enemy action in this case, or perhaps something even more sinister; that the people charged with running the network weren’t keeping it in sufficiently good repair?

If you’re interested, you can read the whole article here. It describes eight fires starting in Lancing, West Sussex, in early March and concluding on 12 May with a second fire in the space of a few weeks at the Aberdeen Place substation in Maida Vale, West London.

The most significant of these erupted on 20 March, when a transformer exploded at the large North Hyde substation near Heathrow Airport, severing power to 71,655 domestic and commercial customers in the West London area and also knocking out flight operations at Heathrow.

Europe’s biggest aviation hub closed for more than a day, disrupting over 1,300 flights and almost 300,000 passengers, as well as costing the airlines an estimated £50-100 million in lost revenue. “It was an eloquent demonstration of how a single point of failure could bring chaos to a crucial piece of the British economy,” I wrote.

But was it enemy action? Well, it was hard to find any hard evidence of foreign sabotage being responsible for the damage, despite some unconnected but hard evidence of Russia having been involved in at least one arson attack on a Ukrainian owned warehouse in East London.

Certainly none emerged in any of the eight cases, even though a question mark hangs over a fire at a substation in Glasgow close to the BAE warship yard in Govan (whose electricity was actually knocked out by the fire). Two unnamed individuals have been arrested in connection with a “wilful fire”, but the Scottish Crown Office does not appear yet to have decided whether to prosecute. Watch this space.

Therefore I had some sympathy for the opinion offered by a former member of British military intelligence, Col Philip Ingram, who concluded that, while not completely beyond the bounds of possibility, sabotage seemed less likely.

“Experts I have spoken to with access to very sensitive areas say they have no evidence of anything untoward going down,” he said. Instead, he’d come to believe the fires were more a case of cock-up than foreign conspiracy. “My primary view is that old infrastructure and poor maintenance - or the lack of maintenance - may be responsible for a lot of what we are seeing,” he observed.

This week we received a bit more information to support the colonel’s revised suspicions. On 30 June, the UK grid overseer, NESO. issued its final report on the events of 20 March. You can read the report here.

As I’d pointed out in my article, perhaps the most striking fact buried in NESO’s interim findings (published in early May) was the sheer age of the equipment involved: “The 275 kilovolt (kV) transformer blamed for the blaze was one of two installed when the substation was itself built in 1968, and manufactured by Hackbridge & Hewittic, a British brand name that disappeared in the early 1970s.”

The final report gave as the “most likely cause” for the failure of this venerable item that “moisture [had entered] the bushing causing a short circuit. The electricity likely then ‘arced’ (causing sparks) which combined with air and heat to ignite the oil, resulting in a fire.” So-called bushings are the connections through which high voltage power enters the transformer to be stepped down to a lower voltage. The oil is designed to be an isolator.

NESO also pointed out this failure had not been unanticipated. In 2018, seven years before the fire, “an elevated moisture reading in one of SGT3’s bushings had been detected in oil samples taken in July 2018”, it said. This is more significant than it first sounds. Moisture ingress is a particular concern with transformers, a former CEGB engineer told me, as it can reduce the insulation properties and increase the risk of electricity arcing and fires. Such contamination inevitably becomes more likely with age.

Indeed, according to the guidance issued by National Grid Electricity Transmission - the owner of the North Hyde transformer - such readings indicate ‘an imminent fault and that the bushing should be replaced’. But despite this rule, and the age of the equipment in question, not only was the bushing not replaced but the transformer received no basic maintenance overhaul between 2018 and the date of the fire.

NESO also noted that the fire had spread because the substation did not have proper firewalls between the three transformers on site, which allowed the fire to spread from one to the next and ultimately took all three off line. Although such firewalls have been mandatory in new substations since the 1970s, there is no requirement to retrofit. The fire suppression system at North Hyde was also known to be faulty.

Subscribe now

Two points emerge from all this that are concerning to my mind. First is that owners like National Grid seem so laid back about maintenance. In this case, despite a host of red flags about the state of the equipment, the last basic maintenance on the North Hyde transformer had been done in 2018, and the last major overhaul in 2012 - more than a decade before.

Second is the tendency for such companies to rely on ever older equipment. Britain now has an extremely aged network, with around 40% of the assets on it dating from before 1975 - the last time the country invested heavily in the grid as it completed its electrification.

Of course, just because gear is old doesn’t mean it’s not capable of working adequately. And engineers say the old transformers on the system are fantastically robust and were always capable of lasting far more than the 40 years their manufacturers designed them for. But there’s a suggestion that the companies have a less than innocent incentive for deferring their replacement.

It all goes back to privatisation in 1989, when the new private transmission and distribution companies popped out of the Central Electricity Generating Board, its Scottish cousins, and the 12 local area boards. In the new world, regulators not only set prices; they determined how much companies could spend on (and charge the customer for) replacing worn out electrical gear. Margaret Thatcher’s government wanted the system to reward efficiency, so if the companies could do it for less than the watchdog’s capex “allowance”, they got to keep the difference.

Initially, most of these companies were listed on the stock market, but over the years, like much of British infrastructure, they were snapped up by specialist investors. For instance, UK Power Networks is owned by funds connected to Li Ka-shing, the Chinese billionaire, while Northern Powergrid belongs to Berkshire Hathaway, the investment vehicle of Warren Buffett.

“As one might expect, these financially-driven owners responded to the regulatory incentives. Over the past decade, they have generally underspent their allowances for replacing equipment, banking the extra income and helping to make electricity transmission and distribution one of the highest margin sectors in the country,” I wrote. Between 2014 and 2021, transmission companies undershot their replacement capex allowance by 28%, according to Ofgem data, while for distribution companies between 2016 and 2023, the figure was 12%.

Recent analysis by CommonWealth, a left of centre think tank, estimates they collectively spent just £3.1 billion annually last year on replacing equipment against an approved budget of £3.6 billion.

The risk with this approach - which splits the “gains” between the companies and their customers - is that it perversely encourages the companies to skimp on replacing assets - even when that capex might be needed. By setting broad rules which require companies to deliver outcomes such as network performance targets while letting them decide how (and if) to spend the allowance, Ofgem may think it is empowering companies to be more efficient. But insiders point to a culture of “asset-sweating” that has left most substations with primitive analogue control systems, ancient concrete gantries, and a lack of modern monitoring gear designed to forestall outages.

“This old gear can’t last forever,” warns Mathew Lawrence of CommonWealth, adding that when companies are challenged about skimping on capex, they say: “‘Ah, but we will catch up when it’s really needed’. It’s like a football manager who loses his first five games saying he’ll win the next seven.”

Many electricity insiders dismiss the idea that elderly gear poses any risk to the network; or indeed that it has degraded the service in any way. According Simon Gallagher of UK Network Services, a consultant: “The networks are performing better than they ever have.”

He provided me with some data about the number of transformer failures for two distribution companies, UK Power Networks (Li Ka-shing’s company) and the central and southern England part of SSE Networks (SSEN) over the past three years. The UK Power Networks graph is above and here below are the SSEN figures.

Gallagher’s interpretation was that “transformer fires , while rare and highly visible, have always occurred. There is no evidence of a recent spike in fires or failure rates”. My view was that it was very hard to know what the data is actually saying - not least because it only covers three years and just three of the 12 regional Distribution Network Operators or DNOs.

But it is worth pointing out that in the case of both groups it did show an increase - of 28% over 3 years for UK Power Networks and 23% for SSEN (S. England).

My point however is not that the system is necessarily collapsing now: we just can’t know. And as Gallagher says, “the lights stay on 99% of the time”, while in NESO’s latest annual transmission review, so-called loss of supply incidents actually fell in number over the past five years, although the aggregate amount of electricity lost was greater.

The real worry is the backlog of capex that is inevitably building up. Some time that stuff will have to be replaced: but when? “Old bicycles tend to break more often,” was the comment one former National Grid engineer made to me. What happens when a bunch of elderly transformers does finally start to give up the ghost in a sudden rush?

It’s not an unimaginable outcome. After all, the strain on the UK’s electricity system is rising. Until now, the network’s assets have aged against a backdrop of declining usage, easing the pressure on hard-pressed substations. Britons consumed just 318 terawatt hours of electricity last year; 23% less than the peak year of 2005.

Now, as Keir Starmer’s government presses on towards Net Zero, usage should start shooting up as more sectors (like motor transport and domestic heating) flip from fossil fuels to electricity. Other power hungry applications such as AI are also coming over the horizon. By 2030, NESO anticipates demand will have grown by almost a third to 411 terawatt hours.

The increment to supply that will be needed will come from renewables, such as wind and solar. These create other concerns, such as “dirty power” - a phenomenon linked to intermittent renewable generation sources where, instead of operating at a stable 50 hertz, the power supply’s frequency jumps around and becomes irregular.

“There is no escaping the fact that dirty power imposes a far greater burden on network equipment like transformers,” says one electricity expert. There are also concerns about demand spikes from electric vehicle charging putting old transformers, switchgear and cables under heavy strain.

My worry is that an avalanche of equipment failures could put the whole network in a terrible mess. Years of thin orders have killed off firms like Hackbridge and Hewittic which built the gear during the great construction wave in the 1960s and early 1970s. GEC has gone, and companies like ABB have closed their British factories. The UK now has just one transformer plant - GE Vernova - based in Stafford. The average delivery time for a step up transformer is currently running at around 200 weeks in the US.

In a Linked in post, Simon Gallagher accused me of relying on “unnamed sources”, and my article of lacking supporting data and adopting an unduly alarmist tone. I wish it was alarmist. But I am not sure it is shouting “fire” to warn that our regulatory system may have left us worryingly dependent on ageing gear, with insufficient engineers to build or install replacements if needed in a hurry. It hardly looks like a good place.

What we must hope is that our old equipment keeps working while we rebuild the supply chain and train up all the skilled engineers needed. “It’s hard to tell when equipment like this is going to start failing,” says Tim Stone, an infrastructure and energy expert. “ What I fear is that it could resemble Ernest Hemingway’s description of going bankrupt in the way it happens - first slowly then suddenly.”

“I want to have the security of rare earth. We’re putting in hundreds of billions of dollars. They have great rare earth, and I want security of the rare earth, and they’re willing to do it.”

Donald Trump is hardly the first political leader to want to claw back the cost of military services rendered by seizing large mineral concessions in the territory of an ally-cum-client. Wind back the clock to 1918 and much the same thought stirred in the minds of victorious Allied statesmen. Having spent much blood and treasure liberating Arabia from the Ottomans in the First World War, they saw an opportunity to lay claim to giant concessions in the wilderness of Mesopotamia - lands thought to be pregnant not with rare earths but with that no less bewitching commodity: oil.

Not only could this solve one of Britain and France’s biggest headaches: their chronic shortage of hydrocarbons in a world where the war had demonstrated beyond all doubt the vital strategic importance of possessing them. (After all, had not the Allies, as Lord Curzon observed, “floated to victory on a wave of oil”?). It would also free them from their strategic dependence on America, to whom they had looked during the war almost exclusively for petroleum supplies, and whose growing isolationism was a source of grave concern.

The story of how the Allies revived a strange pre-war relic, the Turkish Petroleum Company, and repurposed it as perhaps the greatest concession hunting vehicle the world has ever seen, tells us a lot about how these deals work - and also offers a warning: that trying to solve with a deal one strategic headache - the need for minerals - can over time lead to its replacement by many more.

A Long Time in Finance, the podcast I do with Neil Collins, recently made an episode about this story with James Barr, author of A Line in the Sand, a history of the Ottoman carve-up at the end of the First World War, which you can listen to here. As James rightly observed: “There was nothing Turkish about the Turkish Petroleum Company”. It was stitched together just before 1914, initially with a group of European investors looking to recoup some of their country’s historic losses on dishonoured Ottoman bonds. Oil was the game, and their impresario and agent was an Armenian entrepreneur, Calouste Gulbenkian, a man who, far more than Trump ever will, truly exhibited the “art of the deal”.

When the Sultan’s Grand Vizier finally submitted to Gulbenkian’s silky blandishments and awarded the concession on the very same day Archduke Franz Ferdinand was shot in Sarajevo, in a letter that left most of the small print subject to future negotiations, it was to an unlikely coalition of future enemies, with Deutsche Bank owning 25%, Royal Dutch/Shell 22.5%, the UK government-owned Anglo-Persian Oil Company (later BP) 47.5% and Gulbenkian clutching tight his famous 5% share (initially a life interest, but later cunningly transmuted into a perpetual equity stake). By rights, of course, the whole deal should have been stillborn, given the almost immediate outbreak of hostilities. But the Turkish Petroleum Company had secured itself one important piece of future-proofing: it wasn’t a Turkish company. Registered in London, it operated under English Law.

That would prove helpful when the Allies came to pick up the pieces in 1918. Defeated in war, the Ottomans had lost much of their territory, not least the Mesopotamian provinces where the putative concession lay. As these were under British military occupation, London might have seemed to have the whip hand. But Asquith’s government had got itself into a strategic muddle during the conflict, promising territories to the Arabs if they revolted against their Turkish masters, and in 1916 carving up the Middle East in a way that inadvertently handed most of the concession lands to its key wartime ally, the French. (Sir Mark Sykes, the bumptious MP who negotiated the so-called Sykes-Picot pact for Britain, gave no thought to hydrocarbons as he grandly drew a line across Syria “from the e of Acre to the last k of Kirkuk”.)

The need to navigate all these conflicting deals and promises explains the complex horse-trade that gave birth to the new TPC - later renamed the Iraq Petroleum Company. At this point, it should be remembered, this was still a paper concession, no oil having been found. Under the emerging deal, Britain kept most of northern Mesopotamia, which it annexed to a new Mandate territory, Iraq, under the rule of its old Hashemite ally, Prince Feisal, while the French were given the Deutsche Bank stake to keep them quiet. (Lloyd George was helped by the fact that Clemenceau, the French leader, was far too grand to engage in grubby hydrocarbon diplomacy. “When I want some oil, I’ll find it at my grocers,” he once reportedly sniffed.)

Nor was this the end of the post-war wheeler-dealing. The Americans too had to be bought off when they threatened to back Mustafa Kemal’s new nationalist government over where the Turkish border with Iraq actually lay. Figuring that a smaller stake in something of value was better than a bigger share of nothing at all, the British opted for compromise.

The TPC’s final structure bore the stamp of all these backstairs deals. Anglo-Persian was still top dog, with a 23.75% stake and a prior right to a 10% royalty off the top. The other main partners, each with 23.75%, were France’s new national oil champion Compagnie Francaise des Petroles (CFP), Royal Dutch/Shell, and the Near East Development Corporation (a consortium of five US oil companies). Naturally Gulbenkian still had his 5%.

The deal with the new Iraqi kingdom granted these concessionaires the sole right to develop oil fields within the (ill-defined) old boundaries of the Ottoman empire (a deal called the “Red Line Agreement”) on terms that were pretty sweet for them but hardly generous to their host government. The TPC essentially delivered oil to the partners through a not-for-profit entity, meaning most of the value and tax were captured downstream in the partners’ home countries. Iraq received a royalty of four shillings per ton (equivalent to £0.20 for every 7.5 barrels) of oil produced with promises of future adjustments to peg this amount to world prices.

The new royal Iraqi government swallowed all this because … well, it had pretty much no cards to play. Setting aside such practicalities as Iraq’s chronic lack of capital or petroleum engineers, it was in a weak place politically: the Hashemite house had just been booted out of Syria by the French while Feisal’s father, Sharif Husein, would soon be turfed out of Mecca by the Saudis. With the French irreconcilably opposed to Feisal ruling anywhere on their turf, and the British and Americans aligned over their determination to obtain and exploit Iraqi oil come what may, it seemed as if the Hashemites might possibly end up with zilch to show for their wartime revolt against the Ottomans.

The IPC concession - extracted with courteous menaces - would come back to haunt everyone involved - and not just the Iraqis. The first unexpected factor was the sheer size of the find when oil was finally struck in 1927. The Baba Gurgur field near Kirkuk turned out to be the largest in the world, and the dazzling scale of the economic opportunity sparked immediate tensions between Iraq and the company. The former’s flat rate royalty made it very focused on maximising production, while the British and the Americans, who had deposits elsewhere, saw Iraq as a sort of strategic reserve to be developed at leisure - an attitude intensified by the collapse in world oil prices after 1929.

There were also fierce rows between IPC’s own partners, especially the French who, lacking oil reserves elsewhere, didn’t appreciate the laid back British and American approach to development and also wanted (for reasons of energy security) that IPC’s pipeline should run to the Mediterranean to run through their own Syrian territory; something the British were most reluctant to accept. (In the end IPC built a pipeline from Kirkuk which bifurcated with branches ending at Haifa in Palestine and Tripoli in Lebanon, an asset that deepened Britain’s strategic commitment to Palestine, and determination to retain its foothold there).

Security, meanwhile, proved elusive. While, from Iraq’s perspective, the IPC deal protected the country from its predatory neighbours, Turkey and Iraq, it actually contributed to the internal opposition the monarchy faced. The country’s commercial elite grumbled at the lack of employment and contracts from IPC, and blamed these humiliations on the King’s supine reluctance to stand up for their interests. Meanwhile, stifling Anglo-French control of the Middle East, plus the growing problem of Jewish immigration into Palestine, stirred the embers of Arab nationalism. In 1941, beguiled by the possibility of an anti-British alliance with the Axis, the Iraqis revolted under a nationalist premier, Rashid Ali al-Gailani, and Britain had to divert troops from the fight against the Germans and Italians in the Western desert to quell the unrest.

It was a harbinger of the future. The great one-sided concessions were creaking, and would not long survive the war. More generous financial deals - such as the 50/50 profit split Aramco agreed with the Saudis in 1950 - made a mockery of the tight-fisted old royalty arrangements and stirred pressure for better terms across the Middle East. The British were slow to scent the wind changing and Anglo-Iranian (the old Anglo-Persian) - still wedded to the proconsular style of their Edwardian heyday - fell out with Iran’s prime minister Mohamed Mossadegh so badly in 1951 that its prized Iranian assets were nationalised. For a moment war loomed, and while the nationalisation was eventually reversed by a CIA-MI6 sponsored coup, the crisis revealed the extent to which cash-strapped Britain’s military policy was stretched by its need to defend private oil interests abroad.

In 1958, the Hashemites paid the price for their perceived poodle-like relationship with foreigners. King Feisal II and his family were slaughtered in a nationalist coup and the king’s body dragged through the streets of Baghdad by a jeep, ushering in first military rule and then decades of the murderous Sunni fascism of the Baath Party, ultimately under Saddam Hussein. Yet not even the loss of the concessions (IPC’s Iraqi fields were finally nationalised in 1972), would change the fundamental premise of Western policy, which was to treat all borders in the Middle East as sacrosanct, to be defended by force of arms if necessary. As late as the 1990s, America, Britain and France were sending hundreds of thousands of troops to Kuwait to preserve the lines so airily drawn by Mark Sykes in 1916.

Today, with Washington, Beijing and Moscow vigorously rekindling the 19th century concept of spheres of interest, it’s hardly surprising that mineral imperialism too is back in fashion. Perhaps it’s even inevitable, given each Great Power’s suspicion that the others are already employing this Victorian play book to secure their own supplies. But no one should harbour any illusions about the security of such deals, their souring impact on the politics of the put-upon host country, and the difficulty of navigating the long-term strategic commitments that follow. Brits and Americans were fighting again in Iraq less than two decades ago.

The Turkish Petroleum Company long outlived its founders, with the final fragment of the old concession vehicle, the Abu Dhabi Petroleum Company (formerly Petroleum Development (Trucial Coast) Limited) only expiring as recently as January 2014 when its 75-year concession ended. (The heirs of the arch dealmaker Gulbenkian were there at the death of course.) Great fortunes were made - mainly by the private owners. But that was a function less of their acumen than Iraq’s vast oil riches. Ukraine offers no such certainties. As to what lies beneath its soil and its value, no one is yet sure.

By the banks of the River Lagan, not far from the centre of Belfast, stands a modernist structure that looks from some angles like the hull of a massive ship. Opened in 2012, the Titanic Experience is one of Northern Ireland’s biggest tourist attractions: a museum dedicated to the story of the world’s most infamous passenger liner. Inside, visitors mill around the cavernous galleries, peering at mock-ups of the doomed ship’s lavish interior, CGI films of the sinking and maquettes of the shipyard that built the mighty vessel. On a patch of quayside just beyond are the remains of the slipway where the construction work was done.

It might seem odd to start an article about Harland & Wolff, as I did recently, by describing the museum dedicated to its most notorious product, RMS Titanic. But last summer, when the famous Belfast shipyard sank into a massive financial mire, it looked briefly as if it might become considerably less relevant to Northern Ireland’s economy than the museum which commemorates just one of the 1,700 vessels it built.

The company’s financial difficulties weren’t new. It had been struggling for decades, first with the decline of the passenger liner, once its mainstay, and then the unstoppable rise of the ultra-productive Asian commercial shipyards; first Japan, then Korea and later China. The Royal Navy stopped building warships there in the late 1960s. As a consequence, the yard had shrunk relentlessly, shedding staff and failing to replace worn out equipment as it popped in and out of state ownership. The last time Harlands actually launched a ship (and then it was a German kit design) was in 2003.

A near-death experience in 2019, when its long running owners, the Norwegian Fred Olsen shipping group, decided to close it was followed by an unlikely reprieve: a Scottish entrepreneur called John Wood bought it for a few million quid and reversed in Infrastrata, a quoted company whose only asset, a gas storage venture called Islandmagee Energy, was little more than a few signboards on some fields outside the Antrim port of Larne.

There was the usual wild rejoicing that the yard had been rescued and lots of talk about how a renaissance was on the way. I wrote about it at the time, expressing doubt that much palatable could be conjured from these unpromising ingredients:

“I am delighted by the news that Infrastrata have purchased the Harland & Wolff shipyard and retained the skills of the existing workforce,” said Northern Ireland secretary Julian Smith in response to Tuesday’s £6m rescue of the ailing Titanic constructor. But he might want to keep the champagne on ice. Infrastrata describes itself as a global energy business. It has one significant project — a gas storage scheme in Ulster — which still awaits its final investment decision. The company generates zero revenues from continuing operations, and the 2018 accounts contain a long passage entitled “material uncertainty relating to going concern”. This details the auditor’s worries about its need for funding and the carrying value of the gas project.

Last summer it became clear that, sadly, Harlands hadn’t turned the corner. The end was brutally sudden. In July, the new trade secretary Jonathan Reynolds rejected its application for taxpayer guarantees worth £200 million that were crucial to securing a Ministry of Defence contract vital to its future. The government had identified “a very substantial risk that taxpayer money would be lost”, Reynolds said. Within a few weeks Harlands was dumped into administration. In January, it was retrieved by Navantia, a Spanish state-owned shipbuilder, which was its partner in the critical MoD contract that had brought Harlands down.

I have been thinking a lot about Harlands recently in light of various newspaper opinion columns urging the government to use rearmament to spark a massive industrial reshoring in Britain. Harlands is in many ways the prime example of an attempt to do that - and also shows how hard it is to pull off. There was actually a genuine case for reviving Harlands as a shipbuilder and (perhaps equally importantly) a thriving repair and maintenance hub. Britain might have given up on commercial shipbuilding decades ago, due to its inability to recognise that small riverine yards and outdated manufacturing practices weren’t going to cut it in the modern world. But it still has a navy, and hence a need to preserve the capacity to build warships both for domestic use and (if the designs are good) for export.

This need to preserve warship building capacity had led to a more managed decline after the Second World War than in the commercial sector, where output plummeted almost to zero by the 1980s. True, the number of warships built per decade also fell - but steadily, and orders were concentrated in a smaller number of yards to keep them viable. There was just about enough predictable work to preserve a critical mass with the necessary skills.

That at least was the model while the Cold War lasted. Post 1991, however, it was completely torn up. Huge cuts were made to the navy, with the MoD both reducing warship numbers and lengthening the ordering cycle. The result was a catastrophic order famine: so while British yards delivered more than 100 naval vessels in the 1980s, by the 2000s that had fallen by two thirds to about 30. Roll forward to the 2010s, and there were just two proper shipbuilders left - BAE Systems and Babcock - with three just-about-viable yards; two warship facilities and one, in Barrow in Furness, devoted to submarines. Firms like Harlands survived (or didn’t) on scraps of fabrication work, or repair contracts.

At this point, it started to occur to the MoD that it might have shot itself in the foot, especially when the ordering tempo inevitably picked up - mainly because the bulk of the fleet’s frigates, support ships and whatnot were all either at the end of their useful life or well beyond it. Having only a handful of yards to choose from allowed BAE and Babcock to hold them to ransom - both on price and the delivery schedules to which yards would commit. (BAE, for instance, has taken more than a quarter of a century to build just seven submarines.) Not only would all this extend the time needed to renew the fleet, it would also add to the cost. Yet having been burned once, the big two were in no hurry to take on risk and add new capacity to build complex ships for which there was (generally) only one somewhat unreliable customer.

It was a situation that presented an opportunity for yards like Harlands - which had extensive, if run-down, facilities. For instance, it had one of the only two dry docks in the UK large enough to take the Royal Navy’s new aircraft carriers. But the big prize was getting back into the shipbuilding business as a “third force” to which the MoD would apportion business to keep the big two on their toes. Soon after the acquisition, Wood started talking about regenerating the old Belfast yard. The amount of text in Infrastrata’s accounts devoted to the gas storage business dwindled, and in 2020, the company quietly dropped its name and rechristened itself after its fabled subsidiary: Harland & Wolff.

You might wonder why, if it was that easy, no one else came forward to recapitalise Harlands. Well, there was a big obstacle: money. Regenerating the knackered yard in Belfast would cost an absolute ocean of cash, and no-one credible was in a hurry to advance it if they had to rely on the government handing on enough profitable military contracts to make the whole exercise worthwhile. Investors had learned a bitter lesson from the great ordering famine. Naturally, Wood himself didn’t have many resources. He’d charmed a few millions out of the shareholders on the promise of untold riches to come, but if he was serious he needed access to debt finance, and lots of it.

The banks however weren’t keen, and not just because the opportunity was intrinsically risky. Buried in his CV, Wood had glossed over an unfortunate episode in the 2000s when he had previously owned a shipyard, Southampton based MSC Europe. He and a longtime business partner Chris Rogers claimed they’d been forced out of business when a customer failed to make payments on a large contract. The pair then enraged those left out of pocket by buying the company out of administration themselves and (briefly) continuing trading. Rogers was jailed in 2015 for an unrelated fraud. The best Wood could do was to raise £2 million from unconventional lenders at sky-high interest rates and by giving them security over all the assets - a drop in the ocean. The only way this Gordian knot could be cut was through the intervention of the state.

Enter at this point a deus ex machina; namely the clown car that was Boris Johnson’s government. On taking office in 2019, Johnson inherited a worthy plan from his predecessor Theresa May, the National Shipbuilding Strategy, which was a sort of attempt to think a way out of the mess the MoD had got itself into by running down the shipyards. The idea was for ministers to commit to long-term programmes, built in British yards, and give the industry much more visibility on orders to come. This in turn would spark productivity boosting investment. Containing the magical word “renaissance”, it appealed to Johnson’s boosterish temperament, as well as offering the intoxicating possibility of much Winston Churchill style cosplay posing in shipyards beside massive warships. He and his colleagues started taking Wood surprisingly seriously. In 2020, after Harlands had acquired a second bankrupt shipyard (at Appledore in Devon), Johnson came down for the reopening. Photographed alongside Wood, both in bright red H&W jumpsuits, the prime minister declared the yard had “a great future” and promised “a good enough stream of contracts coming through to drive jobs and growth here in Devon”.

The contracts he had in mind turned out to be a class of vessels known as Fleet Solid Support (FSS) ships. These were 39,000 ton monsters operated by the Royal Fleet Auxiliary whose purpose was to supply the giant aircraft carriers the RN had just acquired with such things as food and ammunition. They were sorely needed. Only one of the RFA’s existing store ships was compatible with the new flat-tops, and that was well beyond the end of its useful life and also lacked crew (it has since been laid up permanently, leaving the carriers with no UK store ships, and embarrassingly dependent on foreign navies helping out).

The way the MoD works is that warship orders can only be built in UK yards, while non-warship orders can be tendered overseas. (The RFA’s Tide Class tankers were, for instance, built entirely in South Korea). While the status of destroyers and carriers is clear, the definition has a fuzzy boundary. Politicians are sometimes tempted to favour classifying naval-adjacent ships as warships (even at the expense of higher costs and longer build times) if that allows them to link orders to British jobs - especially in marginal constituencies. Admirals who just want the ships are keener on tendering them to more efficient yards overseas.

Conceived in 2015, the FSS order had been round the houses on this one and had, despite a chorus of complaint from unions and industry leaders (including the author of the National Shipbuilding Strategy, Sir John Parker) been plunked in the non-warship category, opening it up to foreign bids. A host of international yards had duly tendered. But in September 2020, brandishing the National Shipbuilding Strategy, Ben Wallace, the labrador-like defence secretary changed tack, tore up the process, and declared the FSS ships to be warships that had to be at least partially constructed in a UK yard, or yards (these days ships are generally built in blocks, often by different companies, and then brought together for assembly). In theory, this was quite a win for British shipbuilding and a chance to put the new strategy into practice. But it also raised a problem: where exactly would these three giant ships now be built?

As ever with Johnson, no one had got round to thinking this through. The truth was that BAE and Babcock had more than enough work to keep their existing facilities busy, and were wary of investing to fulfil the FSS order to the MoD’s timetable. Anyway, the whole point was to divvy the contract up to someone new. There were few other available options: Cammell Laird in Birkenhead was really a repair yard and still in the throes of completing its first ship for decades: the research vessel Sir David Attenborough (a ship that had become notorious after a government-sanctioned public competition picked for it the name “Boaty McBoatface”). So the MoD looked more widely and lighted on the Spanish state-owned shipbuilder Navantia, which was keen to enter what it saw as the lucrative UK defence market.

Britain might not have the Royal Navy of yesteryear, but it remains one of Europe’s largest defence customers. Navantia was willing to bid competitively, and dole out a large chunk of the work to less efficient British yards such as Harlands. It also fitted well with Johnson’s half-baked levelling up agenda and his desire to safeguard the union by doing something for Northern Ireland’s Unionist community, still bruised by his Brexit deal. It would right a historic wrong, which had left the Province with defence spending of £97 a head against the UK average of £370. Team Resolute was born.

Cue ministers draping themselves in the Union Flag in the usual manner. But all this dicking around had, of course, pushed back the FSS service entry date by three years to 2031, making it certain there would be a long and costly capacity gap between the last old store ship, Fort Victoria, leaving service and the new ones coming in. There were other nagging questions about the deal too. Why had MoD had put one the delivery of its biggest projects in the less-than-safe hands of Wood, and his run down shipyards which - as of 2019 - had just 61 employees (60 in Belfast and one in Devon)?

What happened next is perhaps an object lesson in how not to run a critical defence project. While the Spanish had overall control, and hence were always likely to keep as much of the fabrication work as they could for themselves, there was still an obligation to do about 60% of the work in two UK shipyards, including the final assembly. The contract therefore provided around £80 million to help with “recapitalisation” (i.e. hiring and training a skilled workforce, and renewing perished or obsolete plant and equipment, of which there was lots). But that money was staged and much of the investment had to come up front. So the project was under-resourced from the beginning, and also depended on the executive capacity of Wood, who had never managed anything remotely on this scale.

These worries did not however trouble Wallace. Having decided that Wood was his man, the defence secretary doubled down. Vice Admiral Paul Marshall told sceptical MPs that Harlands had been thoroughly vetted, adding that it was “in no one’s interest for us to be placing contracts in unviable yards”. The MoD even pushed some naval business itsway - a £55 million contract to recondition an old mine countermeasures vessel for the Ukrainian navy. (Harland’s total revenue for the previous year was just £18 million.) This, it was hoped, would unblock some finance as well as help to train up some of the skilled workers that would be needed when the FSS ships came along. And in one sense, it succeeded: Wood was able to charm some US hedge fund managers at a firm called Riverstone into lending him $70 million, later expanded to $115m - albeit at an eye-watering 14% interest rate. The idea was that this could be refinanced at lower rates when UK Export Finance, a government agency underwritten by the Treasury, guaranteed £200 million of H&W’s debts, something the MoD was encouraging them to do.

But quickly it became clear that Wood was pretty much out of his depth, running an organisation an order of magnitude larger than anything he’d ever run before. “It wasn’t very easy working for John,” said Seena Shah, who once headed up PR at the firm and recalled his angry tirades at subordinates. These became more frequent, and the company ran through seven general managers at its Belfast yard in 18 months. To get people in quickly, pay went through the roof, making the likelihood of profitability even more elusive.

A quoted company, Harlands’ shares tracked steadily downwards, reflecting investors’ lack of confidence in its ability to deliver. And as the pressure mounted, Wood diverted his attention to new projects he thought might keep the stock price up. He bought two more shipyards, this time in Scotland, with the promise to bring in big contracts to fabricate wind turbine masts. But a deal with Saipem, an Italian company, ended up with the contract being terminated early. There was a bizarre diversion into building ferries for the Isles of Scilly Steamship Group, which, when Harlands’ tender was rejected, morphed into a mad plan to actually own and operate a competing ferry line on the route.

The collapse of Harlands last summer pushed back the FSS delivery schedule yet again, and added more cost to the deal (some estimate the number at around £100m) because of the need to restructure the contract. The Treasury’s suspicion that it would lose money on any guarantee it issued was borne out when new directors put in by the lenders appointed the auditors PWC to investigate the possible “misapplication” by management of £25 million of company money. Harlands was eventually sold to Navantia for £93 million - considerably less than the £175 million the US hedge fund ended up lending. Unsecured creditors may get just 2p in the pound.

What has the MoD got for all this chopping and changing? Well, despite siphoning extra cash from its already stretched budgets and deferring delivery, it is not clear much has been done to expand Britain’s shipyard capacity. In November, Lord Coaker, a junior defence minister, confirmed the FSS ships were no longer warships - thus reversing the change made by Wallace. In theory the plan remains unchanged to build blocks in Britain as well as Spain before doing the final assembly in Belfast. But sources inside Harlands say that all the steelwork for the first of the three ships has been switched to Navantia’s Spanish yards with only final assembly likely to take place in the UK. Work on recapitalising the Belfast shipyard - paused during last summer’s crisis - has slowed. One manager described the whole experience as “a bit of a kick in the teeth” for those who had come in to turn the yard round.

Last week’s announcement of a 0.2 percentage points of GDP increase in defence spending prompted much media excitement, with analysts rushing to draw up “fantasy fleets” of all the new frigates and destroyers that could flow from these expanded resources. But the hard fact remains that, even were an expansionary course now set, and it certainly hasn’t been, little capacity exists to build extra new warships. Despite the additional facilities being opened by BAE on the Clyde and the new “frigate factory” Babcock has erected at Rosyth, there are no plans to materially accelerate the construction of the existing order book - something that rather undermines the Starmer government’s claims to have moved to a “war footing”.

Turning things around is no easy undertaking. For all the brave talk of “reshoring”, restoring even a fraction of the capacity lost will require more than a few press releases and three-day-wonder initiatives. Think of all the barriers: shipbuilding (who’d have thought) is an energy intensive activity and Britain has the developed world’s highest energy costs right now. The skills base and supply chain has all but atrophied. It’s very hard to see the UK breaking back into the commercial shipbuilding market in anything other than a pretty niche and specialised capacity, and even that would need some massive investments, which would take heroic levels of conviction for any investor to make.

If military shipbuilding is to be the focus, the state must take the lead, working patiently with industry to foster the confidence that can lead to new investments. If needs be, it should make them itself (indeed, it might have been a better use of everyone’s time if the UK, rather than Infrastrata, had rescued Harland & Wolff in the first place). Step one would be to preserve those existing facilities that could provide yard capacity in future. “There are not a lot of places you can build large ships in the UK,” says Pete Sandeman of the website Navy Lookout, who points out that the navy has a lot of big ship orders, such as the Multi Role Ocean Surveillance Ship (MROSS) project and strategic lift cargo vessels in the pipeline. “If we get rid of Belfast, these could end up being built in foreign yards.”

Most of all, the MoD needs to be more consistent and more, well, serious. Until a consistent path is followed, backed by credible plan with budgets to match (and even direct investment where needed) the idea of serious rearmament will remain just that - talk - while places like Harlands’ Belfast yard will come to look ever more like the shipbuilding museum next door.

Gordon Tait still remembers the visit that Cecil Parkinson, then Margaret Thatcher’s energy secretary, paid to the atomic plant at Dounreay on Scotland’s northern tip in the spring of 1988 - and the sense of betrayal that followed it. “He came up here and told us: ‘You are doing a great job here chaps, this is really good’, the rangy Scottish engineer said. “Then he went back to parliament in London and told them: ‘We are going to shut down the fast reactor programme’.”

Tait’s face crinkles ruefully at the memory. He is 60 now, and semi-retired, but back then he was a fresh-faced assistant shift manager just out of university, who expected to spend his entire career working on the Dounreay reactor. The project was one of those incredibly bold bets that Britain placed on technology immediately after the Second World War. Despite a towering national debt and many other calls on the public purse, Winston Churchill’s government established a giant physics experiment in an attempt to give the UK an enduring lead at the cutting edge of civil nuclear power.

The task was to crack a tantalising problem: how to build an economically viable reactor that not only burnt nuclear fuel but produced as a by-product more fissionable material than it was consuming. This seemed to hold out the alchemist’s promise of inexhaustible, bounteous power.

I’ve been twice to Dounreay over the past two years to look at what remains of a project that once drew thousands of engineers and scientists from across the country to an almost deserted part of Scotland. Tait’s family were part of the influx, coming originally from Glasgow, and he was a second generation employee. His father had arrived in the 1950s, drawn by the prospect of cutting-edge research into what seemed like a new miracle energy source. “Nuclear reactors and nuclear physics were his life’s work and what he believed in,” Tait recalled.

When Parkinson’s announcement came in July 1988, it fell on Dounreay and the neighbouring town of Thurso like a thunderbolt. The government, which was preparing the ground to privatise Britain’s electricity industry, no longer saw the virtue of pumping £50 million a year into researching advanced nuclear technology, the energy secretary said. The expectation was “that commercial deployment of fast reactors in the United Kingdom will not now be required for 30 to 40 years.” The plant wouldn’t close all at once. The reactor would stay open till 1994 and the associated fuel reprocessing plant three years after that. But then it would be over, with just a skeleton staff left to help clear up the site.

Dounreay’s story is worth considering as the UK starts trying to revive its nuclear industry after more than three decades of neglect. One consequence of the hiatus is that the country has little to show for its past investment. Pretty much all the intellectual property from Dounreay ended up not in Britain but in Japan, at its Monju fast reactor project. Dounreay’s staff are largely retired, their skills lost. “When Thatcher shut the plant, she didn’t say we won’t need fast reactors ever, she said we won’t need them for 30 years and blow me she was right,” said Tait. “Where she was wrong was when she said we’ll not retain the knowledge and buy them from overseas. I guess she thought her pal Ronnie Reagan would be the purveyor. But look at who supplies all the reactors now: Russia and China. In the interim we have lost all the knowledge.”

Before we get on to the story of Dounreay and its project however, it’s worth asking a question; why fast reactors? Why go to all the trouble of cracking such a moonshot technology in the early 1950s - just a few years after the UK had struggled to perfect its first primitive atomic “piles” at Windscale in Cumbria? It sprang partly from a desire to leapfrog rivals - then the USA and Soviet Union - in the evolving technology of civil nuclear energy. Sir Christopher Hinton, the poker-playing head of the UKAEA’s nuclear energy activities, was a particularly fervent believer. But there was also a more practical reason: energy security. One of the big barriers to adopting nuclear power was the perceived scarcity of uranium, whose main sources were then the Belgian Congo in Central Africa and Soviet-controlled Czechoslovakia. Short supply not only made the element expensive, messing up the economics which required nuclear energy to be competitive with coal; it also encouraged military authorities to hog supplies of fissile material for the purpose of making bombs.

The attraction of fast reactors was their promise to get round this problem by squeezing almost unlimited power out of nuclear fuel. A fast reactor can produce 60 times more usable energy from a given quantity of uranium than any of the so-called thermal reactors. And by irradiating a “blanket” of more uranium that is wrapped around the core, it becomes possible to breed more plutonium, itself a powerful energy source. For a country like the war-battered UK, without oil or indigenous uranium, and whose coal supplies were dwindling, this held out the dizzying vision of long-term energy security.

But if that was its promise, the technology also posed plenty of challenges. Such as the need to use liquid sodium, a highly volatile element, as a coolant, and highly enriched uranium as a fuel. Both were entirely new areas. What would happen if the plant malfunctioned? Hadn’t the government’s own chief boffin, Lord Cherwell, called the fast reactor design “a controlled bomb”? Caithness in northern Scotland was chosen for its remoteness, being, as the civil servants put it, “appropriate to a first experiment in an unknown field of development.. of a potentially hazardous nature.” The desire for an economic boost in the depressed region overcame any local reservations. The local MP, Sir David Robertson, fervently championed the plant, seeing it both as a chance to revive the district and to bolster his own slender majorities.

It took the UK Atomic Energy Authority the best part of a decade to get to grips with the intricacies of the design. Over that time, nearby Thurso mushroomed to accommodate the influx first of construction workers, and later of scientists and engineers. The town’s population more than tripled from 3,000 to nearly 10,000 and for a time it had the second largest secondary school in Scotland. Whole housing estates were built - the first in Caithness with garages and underfloor heating - while social and commercial opportunities expanded. “A lot of the top scientists in Britain were coming to Thurso,” said Heather McLean, then a pupil but later a teacher and deputy head at Thurso High School. “All the children’s organisations in the town - scouts, guides and so on - went from having one company to having three or four. Along with the town amateur dramatics, the Thurso Players, you had the Dounreay Players. Aspects of the town life grew. Overall, people were aware that Dounreay made more things possible.”

Yet always the spectre at the feast was the question of whether fast reactor technology would be economically viable, and, if it was, whether the UKAEA would build a commercial plant at Dounreay. In 1966, that question appeared to be answered when the government sanctioned the construction of a 250 megawatt prototype commercial fast reactor on the site. When Frank Cousins, the Minister for Technology, announced the decision to the House of Commons, he said the government had “taken into account the economic and social circumstances of Caithness including the considerable investment in housing and schools in Thurso”. In the town, the sigh of relief was audible. “People felt it was a reward for all their hard work and the risk they’d taken,” said Innes Miller, then a local farmer.

By the seventies, however, the tides were turning against fast reactor technology. Uranium turned out to be far more abundant than expected. A series of finds round the world meant that production rose from 3,713 tonnes in 1950 to 68,052 tonnes in 1980, by which time there was a glut and the price had plunged. There was little need to go through the expense of reprocessing nuclear fuel to eke it out. Meanwhile, fears about nuclear proliferation raised questions about the merits of reprocessing itself. Was it really wise to recover plutonium from spent nuclear fuel rods, especially if you then shipped it to customers abroad? In 1977, President Jimmy Carter closed the Barnwell reprocessing plant in South Carolina on proliferation grounds, and announced the US would take no further part in this activity.

The biggest problem though lay with the prototype reactor itself. It proved extremely difficult to build and construction overran by three years, only completing finally in 1974. Even once up and running, reliable operation remained elusive. The plant’s “load factor” - the proportion of time it was producing electricity and hence earning revenue - didn’t get above 15% until the mid 1980s. Even in the cloudy UK, a solar farm achieves around 11%. Given the high underlying costs of fast reactor technology - the need for highly enriched uranium and elaborate (and persistently fallible) cooling systems - these numbers were disastrous. By the late 1980s, with nuclear energy heavily out of fashion after the Chernobyl accident, the plant’s fate was sealed.

Three decades have passed since Dounreay’s fast reactor closed. The plant’s after life has proved much more labour intensive than the UKAEA expected. There are still around 2,000 people working at the site, and, while a good number are contractors bussed in to do a specific decommissioning task, the clean-up still contributes some £200 million a year to the community. Dounreay experienced no deep post-closure slump like the coal mining communities after the pits shut down. North Sea oil - whose production was peaking in the 1990s - offered alternative avenues for well-paid, skilled engineering employment. But slowly the realities of Caithness life reasserted themselves: local opportunities and amenities dwindled, the UKAEA estates were sold off, and Thurso is again depopulating.

Like many nuclear communities, the inhabitants mainly mourn the loss of the secure high paid jobs that the plant brought, as well as the extraordinary social and scientific influx that - for a time - made Thurso one of the “brainiest towns in Britain”. Dounreay remains a licensed nuclear site and, despite the Scottish government’s ban on new nuclear development which remains in place, many townspeople hope they will build a new reactor before all the skills are finally lost. “They say they’re thinking of new atomic and I do hope they do something,” said Elizabeth MacDonald, whose parents’ farm was absorbed by one of the so-called atomic estates. “There wasn’t much for us in the county before Dounreay came.”

What lessons though can be gleaned from Dounreay’s experience? We are, after all, at a point where those Thursoan dreams are not wholly unrealistic. Western countries like the UK are again looking to nuclear power as a replacement for fossil fuels - this time for climate reasons. One lesson might be that it’s worth thinking hard about pursuing nuclear innovation instead of going with existing, tried-and-tested technologies. These may have drawbacks - but they are at least known to work. While the Dounreay experiment didn’t stop the UK pursuing other designs in the 1960s and 1970s - such as the Magnox gas cooled reactors, or looking at heavy and light water types - there is a sense the UKAEA became too focused on cutting-edge technology rather than surveying the commercial landscape and going with what actually did the job (exhibit A: the innovative but problem-plagued AGR).

Sticking with proven designs seems doubly wise given the need to start building quickly if nuclear is to pay any meaningful role in reducing carbon emissions. The industry does know how to construct - albeit too slowly and expensively - light water reactors that can reliably achieve load factors of 90-95%, and last for maybe up to 100 years. Does it make sense - in the manner of Christopher Hinton in the 1950s - to dart off in search of novel and untested methods to do the same job?

The recent atomic revival has witnessed a dizzying burst of atomic innovation - the first since the 1960s. There are ambitious plans to develop advanced designs using high temperature gas, and so-called pebble bed reactors (first conceived in 1944). These are - in theory - “walk away” safe, meaning, at least in theory, you could put them in your back garden rather than on a special secluded site. Bill Gates’ Terra Power is even building a fast reactor design - albeit one which doesn’t require the complex chemical reprocessing plant to capture the plutonium bred in the reaction. Yet many of these designs exist only on paper, and none has proved itself commercially.

Of course if these advanced reactors work, they could be incredibly helpful in solving some of the problems still thrown up by nuclear power, such as issues around waste and safety. But who is to say that - like Dounreay’s fast reactor - they won’t prove incredibly hard to build, or un-viably unreliable when finally switched on? The UKAEA spent 30 years chasing the fast reactor and never got there, which was not for the want of trying, or the application of brilliant scientists and engineers. We simply don’t have that much time again.

All photographs copyright Michael Collins

Start a Substack

This letter was conceived as a homage to the doyen of all business journalists: John Brooks (1920-93). A longtime staff writer at the New Yorker, where he specialised in financial topics, Brooks wrote about many of the great business stories of the twentieth century, from the 1929 crash to the merger mania of the 1980s.

Business Adventures was the title of a 1969 collection of his essays, which Bill Gates once hailed as “the best business book I have ever read”. Gates was unstinting in his praise for the way Brooks turned “eye-glazing topics (eg, price-fixing scandals in the electronics market) into rollicking narratives,” and how “he invited the readers to draw their own conclusions about the lessons” and, what’s more, how those lessons still applied.

One of my favourite of Brooks’s stories concerns an astonishingly far sighted American entrepreneur called Clarence Saunders, who at the end of the First World War essentially invented the modern supermarket. Saunders also knew a thing or two about marketing, and called his chain Piggly Wiggly Stores. (Asked by someone why he’d picked the name, he reputedly replied: “So people will ask that very question.”)

Saunders’ story, however, had an unhappy ending. Having floated his company on the stock market, he became infuriated when some Wall Street speculators shorted the shares. Determined to strike a blow for the “honest promoter”, he tried to corner the shares and put the speculators out of business. Instead, he ended up ruining himself and losing the lot. I made a podcast about this story. If you are interested you can listen to it here.

This letter doesn’t hope to hit heights of narrative style and insight that Brooks brought to his craft. What it does aspire to do is to shine a light on business and financial history; often a subject that analysts and journalists - in their quest to anticipate the next twist and turn in markets - tend to ignore.

Posts will often be long and informal in style, looking at stories that might shine a light on things we are presently concerned with; from the search for fuels to replace fossil-based energy to the merits and demerits of state ownership. It will consider how innovations, discoveries, and energy transitions happen, and the decline and fall of once-great industries, as well as how hard capacities, once lost, are to re-acquire. It will look at the social impact of companies on the communities that hosted them, and how scandals and mistakes repeat themselves. Sometimes, they will just be (hopefully) entertaining yarns.

What can we learn from financial and business stories? Perhaps not much more than the fact that human nature never changes; that it’s very rarely different this time, that business promoters and their financial sponsors are incorrigible optimists, and that financial and intellectual capital takes a very long time to build.

If there is one thing we can know from old narratives, it is how the story ends. And from that, as Gates said, we can draw our own lessons. It is surprising how often these lessons still apply.

Subscribe now