On efficiency

There is nothing so useless as doing efficiently that which should not be done at all.

Peter Drucker

Surely if there is one thing we can all agree on, it is that efficiency is a good thing. By efficiency I simply mean the ratio between what one puts in and what one gets out. Fuel efficiency in cars is a straightforward example: it is expressed as miles per gallon (or kilometres per litre), stating that for every unit of fuel you put into the car you can travel a certain distance. The further the distance, the more fuel-efficient the car. One comparison site I looked at told me, for example, that I could expect to get 67.5 mpg out of a Ford Fiesta, compared to 13.3 from a Lamborghini Murcielago.

This is not of course the only measure of a car’s performance. I expect if I looked at how long it takes to get from 0-60 mph, the Lamborghini would look a lot better than the Ford. But that is not a measure of efficiency.

Which type of efficiency we choose to measure has a large bearing on whether we consider one thing more efficient than another. For example, industrial agriculture is often considered more efficient than traditional methods, but it depends on what you look at. In terms of calories out per man-hour in, industrial ag looks terrific; but if you look instead at calories out per calorie in – that is, at energy efficiency – the picture is very different.

A subsistence farmer using his own muscle power and that of his draught animals must always obtain at least one calorie of food for each calorie of effort put in – any less than that and he will stop subsisting and start starving. According to one study (Steinhart, J.S. and Steinhart, C.E., 1974, “Energy history of the U.S. food system”, Science, vol. 184), such farmers normally manage to get between 11 and 61 calories out for each calorie they put in. By comparison, another study (Leach, G., 1976, Energy and Food Production) found that UK agriculture in general gets 0.35 calories out per calorie in. I’ve seen estimates for US agriculture of 13 calories in for each calorie out, which works out at around 0.08.

On the face of it, these figures are catastrophically inefficient. The only reason the industrial-farming nations have not all starved to death is that most of the calories going in are coming from fossil fuels. Very little of that energy input is from the person sat in the cab. This is fine until it isn’t: fossil fuels are depleting, and at some point will become prohibitively expensive. That point is not as far in the future as we would like.

Another factor which can distort our perception of efficiency is the tendency to measure it in terms of price. Other things being equal, the cheapest way of doing something should be the most efficient, but other things never are equal. Economists have a delightful euphemism for a cost that a business is not going to pay; they call it an “externality.” The diesel which keeps all those tractors running is produced at considerable environmental cost which is in no way factored into the price. This is still true even if we insist on defining that cost purely in financial terms; it’s going to be an expensive proposition to deal with sea-level rise, for instance.

Because prices are often distorted in this way, the magic of the free market is unable to work as the theory says it should. The price no longer reflects the true cost. It would be interesting to know what the price of a barrel of crude oil really ought to be, once one includes all the externalised costs. As far as I can tell that study hasn’t been done, but I suspect the result would be somewhat higher than the $60-odd the stuff is going for at the moment.

Our love of efficiency has brought us the magic of the Just-in-time supply chain, where the amount of stock sitting in warehouses is minimised as far as possible. Sadly, this can all too easily morph into its evil twin, the Not-quite-in-time supply chain. As I write this, an enormous container ship has got itself wedged across the Suez Canal. Now I am sure that ship looks very efficient in terms of volume of goods moved per dollar, but at the moment it is holding up hundreds of other vessels and goodness know how many gazillion dollars’ worth of stuff. Nobody yet knows how long it will take to unblock the canal.

And this exposes the dark side of efficiency: its antagonism to resilience. Many of the ships queuing up to the south of the canal are oil tankers. Refineries typically keep about a week’s worth of crude on hand in case of interruptions to supply, but if any of those tankers end up having to go the long way around Africa, there will be more than a week’s gap. By the time you read this, no doubt we will have seen how that plays out, but issues like that are not uncommon. Recently, for instance, an earthquake in Japan shut down a semiconductor plant, interrupting production at a number of car factories.

Back in 2000, there was a strike by petrol tanker drivers in the UK which disrupted supplies to supermarkets, amongst other things. It emerged that only a few days’ worth of food was kept in stores. People know this; hence the incidence of panic buying in the early days of the 2020 pandemic. (This is one of those irregular verbs: I am stocking up on essentials, you are panic buying, he is a prepper.) As the saying has it, any society is three square meals away from anarchy.

Resilience implies a certain level of inefficiency. That spare tyre is increasing the weight of your car as well as taking up space, but it also makes the car resilient in the event of a puncture. The most efficient way is not necessarily the best. As with most things in life, a balance needs to be struck. Efficiency, after all, is a one-dimensional measure. It can tell you which car has the highest mpg, but that may not be the only or even the most important factor in your purchasing decision.

Consider the work of Frederick Winslow Taylor, the first person to apply the rigorous empirical methods of science to the workplace. In his case, he was trying to maximise output per man-hour. What he really wanted to achieve, but couldn’t due to the limited technology of his day, was complete automation. Instead he tried to treat the human beings he was studying as if they were robots. He advocated rigid adherence to standards which were to be defined and enforced by management, making Adam Smith’s pin factory a worker’s paradise in comparison.

Now of course Taylor wasn’t optimising for happiness, and indeed it is hard to imagine how one would go about doing such a thing. (The Utilitarians gave it a try, but their definitions beg most of the interesting questions.) If we define efficiency as a ratio between quantities, then only quantifiable things can be considered. That does not however mean that they are the only important things, or even the most important.

Efficiency is a seductive notion. It looks so clear-cut. After all, you can’t argue with arithmetic. Nor is it a bad tool when employed correctly. Fuel economy is certainly something I would take into consideration when buying a car, and it’s useful to have a standard for comparison. But like everything else, dedication to efficiency can be taken too far.

As we have seen with the example of agriculture, something may be efficient by one measure and appallingly inefficient by another. It is important to be aware of which measure we are choosing, and why. Nor is “efficient” a straightforward synonym for “good.” Would the Nazi death camps have been better if they had been more efficient at killing people? There is always a bigger picture than a simple measurement of efficiency will give us.

The next time you see something touted as efficient, look for that bigger picture and ask the awkward questions. You might be surprised by the answers.

Comments are welcome, but I do pre-moderate them to make sure they comply with the house rules.