When asked his opinion of Wren’s newly completed St Paul’s Cathedral, James II described it as: “Awful, artificial, and amusing.” From this, it’s pretty easy to imagine an historical retrospective with the headline: “James II, architectural philistine.” Or, in a place with differing aesthetic sympathies: “Was James II an early champion of brutalism?”

However, this points to a trap into which comparisons often fall: definitions aren’t set in stone.

What do I mean?
Well, we use definitions to understand and measure reality.

Scientific definitions undergo constant refinement and in ordinary everyday language, meaning mutates through use and misuse. Some contemporary examples of the latter: pretty soon “disinterested” is going to mean “not engaged”, “refute” is going to mean “repudiate”, and “to beg the question” is going to mean “to invite the question”. At the moment, all three of those are, strictly speaking, incorrect. But today’s “strictly speaking” is but a codification of yesterday’s ordinary usage.

Going back to James II’s harsh and damning appraisal of St Paul’s… it wasn’t. At the time, awful meant awe-inspiring, artificial meant well-crafted, and amusing meant inspiring joy.

Pity our poor imaginary headline writers.

Definitions evolve and change and there’s no problem with that just so long as we’re aware of it. Unfortunately, we usually aren’t, and we’re prone to what might be termed “measurement illusion”.

A great illustration of measurement illusion is the Irving Fisher anecdote I mentioned in an earlier article. During the German hyperinflation, Fisher noticed four identical white shirts in a shop window. Three were priced at 32,000 marks, but the fourth was only 11,000.

He asked the shopkeeper if there’d been a mistake.

She replied: “Oh no, there’s no mistake! And do you know, the wholesaler has started charging me 30,000 marks a shirt!?”

A little baffled, Fisher replied: “Then, shouldn’t you be charging me more?” She responded: “Don’t worry, I’m making a tidy profit!” “Are you?” “Oh yes! I bought that one last week and I only paid 8,000 marks for it!”

The woman believed herself to be in profit because she was about to receive 11,000 marks for the shirt, and 11,000 is worth more than 8,000, right?

Not in this case. The 8,000 marks she’d paid for the shirt had had more purchasing power than the 11,000 she was about to receive for it. Money illusion made her loss look like a profit. Money illusion is just one manifestation of the perils of measurement illusion, perils to which none of us is immune.

In mid-October, Nobel prize-winning economist Paul Krugman triumphantly proclaimed: “The war on inflation is over. We won, at very little cost.” He presented a graph in which inflation was currently below 2%. Krugman’s recent track record isn’t good. In 2020, he was confident that inflation wasn’t about to strike…Wrong! In 2021 he was certain that inflation was transitory…Wrong! He also claimed inflation doesn’t hit low-income households particularly hard…abjectly wrong!

Be all that as it may, what’s going on with his 2% claim?

Well, Krugman’s definition of inflation doesn’t include food, rent, mortgage payments or energy costs. Do these things affect the pound in your pocket? They certainly affect it in mine. How easy it is to triumph if you change the definition of winning.

To reiterate, there’s no intrinsic problem with changing definitions, the problems arise either when people fail to realise words don’t mean what they used to, or when malfeasant actors use shifting definitions to pretend that something is the case when it isn’t.
This problem isn’t in any way restricted to economics.

How easy it is to triumph if you change the definition of winning

It doesn’t take too much to be aware of the deliberate stretching of “hurrah” words such as “fairness” or “justice” to include whatever it is that the speaker wants everyone to approve of. The same goes for “boo” words, stretched as they routinely are to include anything the speaker wants to condemn. When definitions are manipulated, so too are people.

To further illustrate measurement illusion, imagine a change in the way deaths in the workplace are recorded. Let’s say there’s a long-established practice of counting only those deaths which happened within 24 hours of an accident. Now, let’s say a new criterion is introduced in which any death within six months of an accident is recorded as a death in the workplace. There may be very good reasons for the change, but politicians, the media, and the public would almost certainly think there had been a sudden upsurge in deaths. No doubt immediate action would be called for. (This of course may have been the motivation for the original change.)

And here’s a real-world example. A few years ago there was a proposal to redefine the word “climate”. Climate is a 30-year moving average of weather. So, right now, the climate of any specific place means the average of its weather variables from 1993 until the present. It was thought, in some quarters, that climate change had rendered this historical standard obsolete. It was proposed that climate be redefined as a five-year moving average. Why? Well, the rationale was that climate change means we should give more weight to recent weather events than to those of two or three decades ago. However, what a very dangerous move that would have been, a definition of climate that would vitiate historical comparisons, a definition with climate change inbuilt: a genuine example of “begging the question”.

Peter Lawlor was the Chief Economist at the German Stock Exchange and continues to advise senior politicians and Wall St institutions. These are strictly his own views