15. Transcending trade-offs: the Engine of Improvement

For most of human history most peoples’ standard of living remained more or less the same. They hunted, gathered or farmed in the same ways that their ancestors did. Technologies advanced and economies grew very slowly. Then, starting in small ways at first, the rate of innovation accelerated. People discovered recipes like ‘how to make iron’, which enabled them do many other things better and to do new things too.

But every recipe is inadequate in one way or another. In the process of trying to achieve something, we hit this constraint, that boundary, or hurdle, or obstacle or undesirable trade-off that limits what we can do. Which forces us to look for ways to address them.

It sounds so dull, mundane and tedious doesn’t it? Butting your head against brick walls and trying to find ways round them? But don’t knock it! This is where virtually all improvement — all additional wealth creation — comes from.

In this blog I want to show you just how incredibly important this process of head-butting is. And just how big it is, because wherever we look, no matter what time, place, industry or scenario, we find constraints, limits and obstacles that need getting over. They include, for example:

• the time, energy or material cost of inputs or their availability/quality
• the (time, energy or material) cost of the production process itself
• the quality /functionality of what is produced
• the negative externalities created by the production process or the product itself — that is, costs or damage imposed on other parties that need to be addressed.

Let’s take a look at some examples.

Availability of key inputs

• As medieval iron making took off, iron makers used up ever more quantities of charcoal. After a while, deforestation meant good quality charcoal became hard to come by. This prompted the search for alternative sources of fuel such as coal. However, early (shallow) mines were quickly exhausted. This, in turn, spurred the quest for new technologies to help miners dig deeper.
• In its early days, the only way to make cotton thread was to manually spin threads and piece them together using a hand loom. John Kay’s ‘flying shuttle’ (invented in 1733) could do the work of so many hand loom weavers that it created a shortage of cotton thread. Cotton spinners’ inability to keep up with demand constrained the growth of the cotton industry for over 30 years.

Reliability and security of input supply

• In the middle ages, well-placed, well-constructed water mills could do the work of many beasts of burden (including humans). At one point, there were over 70 water mills in just one mile stretch of the River Seine in Paris. But water mills ground to a standstill in dry seasons, prompting the search for alternatives such as windmills (which, of course, can be equally unreliable).
• One of the reasons medieval craftsmen turned from bronze to iron was that the tin needed to make bronze (by mixing it with copper) had to come from distant places such as Cornwall. These supply lines were risky and unreliable.
• One of the reasons why the Middle East is a cauldron of conflict today is great power manoeuvring to ensure continuity of oil supplies.

Input costs

• Early versions of important technologies wasted very high proportions of the inputs they used, creating cost barriers. For example, early steam engines lost more than 95% of their energy inputs to heat. One of the reasons James Watt’s steam engine was so successful was that it used a quarter of the fuel of its predecessor to do the same amount of work.
• LED lighting uses less than 75% the amount of electricity used by previous lights.

Operational constraints

• The early days of electricity production involved workers in a constant battle with their machinery. With the first turbines, typically one third of their blades would break. Bearings constantly overheated, requiring workers to cool them with oil and water … until the resulting smoke drove them out of the premises. They kept relay valves open with corks and tied weights on to levers to keep turbines running when stream pressure dropped below recommended levels. This was more Heath Robinson than a sophisticated production process.
• In the first decades of the industry, motor cars were made by craftsmen. Each component was crafted separately and individually. This meant that to make two cars you needed twice the amount of labour time. There were no economies of scale. In addition, because they were all slightly different, to make two handcrafted components fit together snugly, they each had to be manually re-worked. Some of what these craftsmen produced was exquisite. Some Rolls Royces made 100 years ago are still purring today. But the production process had enormous costs built into it, creating an insurmountable barrier to any growth in the volume of production. It was the use of standardised parts in moving assembly lines that finally transcended these constraints.
• A cotton ball on a cotton bush contains a bunch of short fibres (about an inch long) of irregular thickness. For centuries, the process of combing and winding these fibres to create a continuous, strong thread of consistent thickness required immense manual skill (see above). It took James Hargreaves over a decade to perfect his Spinning Jenny, a machine capable of reproducing this manual skill. Once he did so, he lifted the logjam I talked about above, triggering the exponential expansion of the cotton industry and with that, the industrial revolution itself.
• In the early days of telephony, connecting one caller to another was done via manual telephone exchanges. As volumes grew, the cost of employing thousands upon thousands of human switchboard operators grew so prohibitive that it put a break on the industry’s growth. The development of new automatic switching technologies was a decisive turning point in the growth of modern telecommunications.

Product reliability/maintenance costs

• With the earliest incandescent light bulbs, you were lucky if the filament lasted half an hour before burning out. Modern LED lights are guaranteed to work for thousands of hours.
• The first computers were 1000 times faster than earlier electro-mechanical machines. But the valves they relied upon took up a large amount of space and frequently overheated, creating constant interruptions as they had to be replaced.
• The first aero engines were so unreliable they had to be overhauled once for every 50 hours of flying time. Today’s modern engines are overhauled once every 50,000 hours of flying time: a thousand-fold improvement. We couldn’t have our modern aviation industry without this dramatic improvement in reliability, with its consequential reductions in maintenance costs.

Product performance

• Orville and Wright proved that manned flight was possible in 1909, but performance of early aeroplanes was severely limited. It took decades worth of further innovation to create planes that could take off and land safely, carry heavy weights and carry human beings in comfort.
• Today, modern aero engines are not manufactured in traditional ways by assembling them together from numerous sub-components. Engines built this way simply cannot cope with the rapid and extreme changes in heat and pressure that engines experience. Instead, blades are more akin to ‘grown’ from a single crystal of metallic nickel alloy. This way, they don’t have any in-built fault lines and have far fewer moving parts where things can go wrong. This approach to making engines is one of the reasons for reduced maintenance costs described above.
• You might have thought that, given its deadly potential, the gun would have transformed warfare instantly. But it didn’t. Early cannons were so heavy they were almost impossible to deploy on battlefields. And they were so inaccurate they rarely did much damage to opposing armies. Their main initial use was to frighten the foe with their noise. Likewise, the first handguns were extremely temperamental. It was a struggle to make gunpowder that worked reliably. Every time a shot was fired, a new feed of gunpowder had to be damped down. This took time (while the enemy was advancing on you) and was dangerous: it regularly explodes in users’ faces. Even then, before ‘rifling’ was invented, the chances of firing a shot that actually hit its target were very small.

Physical constraints (such as size and weight)

• The first ‘Newcomen’ steam engines were so heavy and cumbersome that they had to be build in situ. Steam engines could only be used on railways (for example), when powerful new condensers dramatically reduced both their size and weight.
• The modern electronics industry would be impossible without dramatic advances in miniaturisation. The latest generations of microchips are just one fourteen-billionths of a metre long — the size of the smallest viruses. With transistors that are sixty times smaller than the wavelength of light used by the human eye, they are literally invisible.

Quality/functionality

• Everything that is produced has certain positive qualities and certain negative qualities. In Roman times for example, wool and woollen clothing was cheap and plentiful. But it was also hot, heavy and itchy. Silk, on the other hand, was smooth, comfortable, cool and light. But it had to travel from the other side of the world and was extremely expensive. Users had to choose between ‘cheap but unpleasant’ or ‘pleasant but expensive’. Breakthroughs happen when such trade-offs are transcended, as they were with cotton manufacture.

Process/coordination limitations

• Early railways made it possible to move more goods much further, much faster and much cheaper than alternatives such as the canal or horse and cart. However, many of these potential benefits were dissipated by the fact that different railways lines were built to different gauges. Early rail travel took far more time and cost far more than it should because so much time and effort had to spent moving goods and people from rolling stock on one gauge onto another. New infrastructure was required to transcend these limitations.
• In the early 20th century Greater London boasted 65 electrical utilities, using 49 different types of supply systems, 10 different frequencies, 32 different voltage levels for transmission and 24 different voltage levels for distribution. Appliances built to operate one system didn’t work if plugged in another just down the road. London’s ‘let the market free’ approach to electricity generation made its electricity companies the most profitable in the world — at the price of a much slower pace of electrification than the rest of the world. To make the journey from achievability to affordability and availability far-reaching standardisation (e.g. a National Grid) was needed (something the electricity companies fought tooth and nail against).

Negative externalities

Every production process imposes costs on participants and third parties.

• The first electricity generating stations caused so much vibration that legal action over their “intolerable nuisance” was common.
• The growth of agriculture has destroyed natural habitats and sent many species of animals to extinction.
• Many of today’s production processes deplete non-renewable resources and damage the surrounding environment by polluting and poisoning air, water and earth and fostering climate change.
• Historically, many work processes have exploited the weak via institutions such as slavery and child labour.

Hard, detailed, practical work

In a previous blog I said that real wealth creation reaches new heights by standing on the shoulders of giants. This isn’t entirely accurate. Real wealth creation advances on the shoulders of multitudes, via the huge amounts of painstaking, hard, detailed work undertaken by countless un-named people to overcome or transcend each particular specific trade-off, hurdle, obstacle, barrier or side-effect that holds additional wealth creation back. It advances by adapting, refining and tweaking existing recipes or inventing new and different ones.

Is there a limit to this process? There may be. But we’re nowhere near to it. In my next blog I’ll explain why.

• When I talk about ‘generating additional wealth’ I’m not thinking of ‘economic growth’. Wealth creation and economic ‘growth’ are not the same things. In fact, often they are opposites. By additional wealth I mean the material prerequisites of enriched human lives, while the concept of ‘economic growth’ is part of the Money Mirage. ‘Growth’ metrics like ‘gross domestic product’ only value those things that can be measured by money (even if they are not enriching humans’ lives) while ignoring a multitude of activities and processes that do enrich peoples lives but which are not measured by money.

Next in this series: 16. Waste: The ‘Free Lunch’ of Wealth Creation

Previous: 14: The Three As of Wealth Creation

Bibliography

Books and articles I found particularly useful researching this blog include:

• Philip E Auerswald, The Code Economy: A Forty-Thousand Year History, OUP, 2017
• Simon Winchester, Exactly: How Precision Engineers Created the Modern World, William Collins, 2018
• Robert Friedel, A Culture of Improvement: Technology and the Western Millennium, The MIT Press, 2010
• Thomas Hughes Parker, Networks of Power, Electrification in Western Society, 1880–1930, John Hopkins University Press, 1983
• Vaclav Smil, Energy and Civilisation: A History, MIT Press, 2018
• David Edgerton, The Shock Of The Old: Technology and Global History since 1900, Profile Books, 2019