The consumer price index – or short ‘CPI’, is used to measure changes in the price level of goods and services purchased by households, and is also used to measure inflation.
To measure the CPI, a study is done that investigates what the ‘average’ household (of the particular country for which the CPI is being measured) buys / spends
money on in terms of goods and services. We are in essence working with an imaginary shopping basket/cart which has a whole bunch of stuff in it which is supposed to represent what a typical household spends money on.
When constructing the CPI, a few things are taken into consideration:
- Goods and services which should be in the ‘basket’
- What weight should be assigned to each good/service (to indicate its relative importance in the basket -- eg. Food over perfume)
- What base year is going to be used
- What formula is going to be used
- Prices have to be collected each month to calculate the value of the CPI for that month (as it gets published monthly)
For the first two points, in-depth surveys are undertaken to determine the relative weights, and the goods and services which should be included. Since this type of research is quite time-consuming, it only gets done every five years or so.
The base year reflects the year within which the initial survey/study was done to establish the CPI, and will be the year to which other CPI’s will be compared to.
Let’s look at a simplified version of the process:
In Country X
consumers only purchase water and bread, where the ‘average’ consumer will purchase 10 bottles of water and 5 loaves of bread in a given period of time(“years”). Next we look at the prices for these items within each time period.
Year | Price of water | Price of bread |
2005 | $1 | $3 |
2006 | $2 | $4 |
2007 | $3 | $5 |
Now we calculate each basket’s cost per time period:
2005: (10 x $1) + (5 x $3) = $25
2006: (10 x $2) + (5 x $4) = $40
2007: (10 x $3) + (5 x $5) = $55
Next we pick a base year, and calculate the CPI – we’ll use 2005 as our base year.
The CPI for 2005 is ($25/$25) x 100 = 100 (cost now/cost base year) x 100
The CPI for 2006 is ($40/$25) x 100 = 160
The CPI for 2007 is ($55/$25) x 100 = 220
To see how much prices increased from time period to time period – we simply subtract a 100 (= base year index) from the comparison year. So if we are using 2007 as our comparison year, we can conclude that the prices increased 120% from 2005 to 2007.
Different CPI’s will be measured for different expenditure groups such as: pensioners, urban wage earners, different provincial/metropolitan areas etc.
Since food, energy,
housing and mortgage rates will usually take up quite a bit of ‘weight’ within the ‘basket’ – changes in the prices concerning these will have the biggest impact on changes in CPI.
The CPI gives a good indication in terms of increasing prices, but you can’t use it to measure the cost of living, since the CPI works with a ‘fixed basket’ – while the cost of living will change as people will substitute items for one another as prices move higher / lower (so for instance, if coffee becomes more expensive and tea cheaper, people will start substituting coffee with tea, and so try and maintain the same expenditure level, rather than just going with a ‘fixed basket’ and being non-responsive to price changes). The CPI also does not take into account the introduction of new items. If for instance internet was just introduced in the year 2006 in Country X, it would not yet be included in the CPI for the next couple of years until a new basket survey is done, and so internet prices would not be accounted for until the new survey is done. Another point which is not covered is that of quality. If bread became more nutritious but cost the same, the cost of living would remain the same while living standards would go up.
There's not much to this point, it's just one of those things we've made up to 'measure' and 'evaluate' our economy' while in no way whatsoever measuring anything real, since our economy is all about infinite growth.
Since we're living in a world of major
inequality, the CPI will also in most cased not really be representative -- as with any 'average', you're just drawing the two polarities together (extremely rich and extremely poor) and so you're not really working with anything which really represents the 'cost of living' within this world, since the majority of people are not living an average life -- but a life of misery. This 'misery' is easy to hide away in averages as the abundance of the
rich will quickly "balance" this misery out in numbers. When looking at for instance per capita income, you are also working with averages. So if you have ten people in a group, and 9 out of ten earn $1000 per year and one of the ten makes $100 000 (eg small wealth elite scenario) -- you get a per capita income of $10 900, which is the 'average', but completely not representative. And so with any model in economy where one work with 'averages' you're always working with a
misleading picture which will always portray things to be better for the "average person" than what they really are.
