Our perception that inflation is the normal condition is no more than a reflection of the experience of people alive today. In 1913, unlike now, a pound or a dollar would have bought the same goods as a century earlier. The longest semi-official price series we have reports a 140 fold rise in prices in the UK since 1750 — but even then all the increase up to 1938 is accounted for by inflation during the Napoleonic and first world wars. Indeed, while the price level roughly doubled during both these episodes, it fell slightly over the rest of the period.
This historical perspective may offer a partial antidote to the fears of investors over the arrival of deflation in Europe, where a euro today buys more than a year ago. But there are other reasons not to panic. For one, there is no qualitative difference between an economy in which prices are rising slightly and one in which prices are falling slightly. Unlike water, liquid at temperatures above 0C and solid below, the consumer price index is a complex statistical construct not a physical fact. You can see and feel the difference between ice and water, and you can skate on one but not the other. Similarly, with a commodity such as petrol, you can tell whether the price at the pump is rising or falling because petrol is a homogeneous product that changes little over time.
But what has been happening with cars, or smartphones, or medical services? In the CPI, the price component for cars comprises the sticker price adjusted for changes in quality. Such quality adjustment is subjective — and, it is generally conceded, too low. So we could have been experiencing deflation for years without realising it.