Inflation
1.
The rate at which prices are rising.
Usually measured in the UK by the Consumer Prices Index (CPI).
2.
A situation in which prices generally are rising.
This is the usual situation in most developed economies at most times.
Contrasted with the less usual situation of Deflation, when prices generally are falling. However, very low - and negative - inflation has been more common in recent years. For example, measured by the CPI, UK inflation was negative in both September and October 2015.
Points to note
We have taken inflation and deflation above as changes in "prices" or "prices generally" as if they have a clear meaning. But they raise the the questions of "What prices?" and "What does generally mean?"
Since the term was first used "inflations" have proliferated. Retail price inflation, consumer price inflation, services inflation, government sector inflation, health sector inflation, asset price inflation (house prices, vintage cars, industrial assets, etc.), wage inflation, and so on.
And once we know what we are interested in, we need to define it more closely. What exact set of goods or services or assets precisely are we looking at? For each member of the set, how much weight do we give it in the basket - how do we weigh cream cakes against toothpaste? And if we include a device or a machine or a house, which precise type, model or design, and where bought?
And what if our selected type is no longer made, or even the whole product ceases to be relevant? For example in Europe, households no longer launder with wash-boards and mangles (relatively cheap, though requiring much manual labour) but with washing machines (relatively expensive, but with much less manual labour required). And bars of laundry soap have been replaced by "clever" detergents? Or consider a mega-byte memory - that industrial good that fitted on the back of a truck and was used only by the largest companies and governments against the litter of USB memory sticks in drawers in developed-country houses at the start of the 21st century. Adjusting for this effect is called hedonic adjustment - but any two people are likely to value changes differently: it has huge amounts of subjectivity.
And how do we find out prices - read a catalogue, go to the village shop, look online, etc.?
So always have great suspicion of any calculated inflation index and ask yourself if it is the index best suited for what you have in mind - and don't apply it where it may be positively misleading.
And the most general measure of inflation in an economy - the implied GDP deflator - doesn't that include everything and so is most relevant to its society as a whole at the time? Well, it makes an effort, but it has its own problems. It is the (percentage) difference between nominal and real growth - this year's GDP (Gross domestic product) at actual prices and this year's GDP at last year's prices. But estimating GDP is very hard, different methods producing different results.
Even what one should try to include in GDP is disputable. For example, if you employ a care-worker to look after a sick or infirm family member, the service provision is part of GDP. If one of the family does it, it is not. And what about the black economy? If you pay the just discussed care-worker in cash and don't tell the tax man, will that be included? So, for example, internationally in the early 21st century, GDP estimates are being changed to include the "value" of prostitution and illegal drug use as the discussion moves on.
Surveys of the prices in each year are hard too and raise a lot of the same questions as discussed above.
We are dividing an uncertain numerator by an uncertain denominator. So while the implied GDP deflator may be the best "general inflation" measure we have readily available, it should still be treated circumspectly.
Added to which, if we don't quite know what inflation is, we certainly have little idea of what the underlying causes of inflation are, though there are a lot of theories. The choice of underlying cause assumed in any article, speech or policy decision being perhaps as much an exercise in politics and wishful thinking as in economics.
For these reasons it is normally better to undertake forecasting in money terms, rather than in 'real' (inflation-adjusted) terms.
