Monday 28 March 2011

Are the Greek inflation readings representative of the actual inflation...?

Something that’s been nagging me for a while now is whether the way that we calculate consumer price inflation is representative of actual inflation experienced by consumers presently.

I do not mean that the current basket of goods which is used to calculate the Consumer Price Index (CPI), was not representative of what the typical Greek household consumed at the time that it was compiled.

Nonetheless, presently, the typical Greek household is experiencing a sizeable reduction in its disposable income, which is bound to have some equally important effects on its consumption patterns. Each and every one of us has to fulfill its basic needs first (food, housing, etc.) everything else comes next (consumer discretionary items etc.). When one’s income is reduced the part of it that is being spent in order to fulfill his/hers basic needs, rises as a % of the total, which means that prices fluctuations in these groups of goods affect her/him more. In the current CPI readings the basket compiled at the 2004/2005 household budgets survey is being used.

Typically, the higher the income the lower the part of it that is channeled at food, housing, etc. and vice versa. Here are the most important groups of goods for the typical Greek household, according to their weights in the consumer price index (CPI).


source: statistics.gr

I do not know whether there are guidelines from the EU, or whether the Greek statistical agency (El.Stat) decides when to update the basket of goods used to calculate the CPI on its own, nor am I trying to throw the blame on anyone of the agencies involved here. I’m just wondering whether it’s time to update the basket of goods…

While browsing through the relevant CPI data for Latvia I stumbled upon the weights of individual goods for the typical basket used there. It seems that they are reviewing the representative basket of goods annually. Of course, to be fair it is probably much easier and less time-consuming to carry out such a survey in Latvia than in Greece, but all the same…


source: Latvijas Statistika


As you can see, during the past 2-3 years that Latvia is undergoing a depression-like recession the weights for food and housing are edging upwards. If something like that was done in Greece right now, maybe the CPI readings that we would get would be a tiny bit different. Maybe then they wouldn’t and I am wrong here, but there is only one way to find out…

UPDATE, 6/4/2011: Since I want to be fair, I have to say that as I discovered today, there is a planned update of the representative basket of the CPI based on the 2009 Household Budget Survey. Obviously, I didn't know this at the time that I was writing the post, otherwise I wouldn't write it at all. Here's the link to the relevant press release:
 http://www.statistics.gr/portal/page/portal/ESYE/BUCKET/General/A0515_DKT87_MT_MM_01_2011_12_2099_01_F_EN.pdf


2 comments:

  1. I am not quite aware of the issue analysed here, but are you sure that the variables like fuels and electricity are not included in the housing index for example? Moreover, I would understand your argument if the inflation percentage remained stable the last year but this is not the case. The inflation percentage did grow up after the rise in gas taxation for example as well as after the rise of electricity prices. I do not really unbderstand how you see miscalculations........

    ReplyDelete
  2. I think that you didn't get what I am talking about here, maybe I wasn't clear...CPI is calculated by using a basket of goods, which according to the households finances survey are deemed typical for Greek households...I am talking about the weights of individual goods in computing inflation, whether inflation spiked or fell over that period is irrelevant, what matters is what disposable income did...when disp.income falls, basic goods, like food or housing tend to get a larger part of the remaining income since they are in a large part inelastic...so, when disp.income falls, weights of the goods included in the basket should be adjusted to reflect actual inflation faced by consumers....this is displayed pretty clearly by the chart about Latvia, these are not prices about Housing or Fuel or Food these are their weightings when calculating CPI...I hope I was more clear now...
    If you're not bored you could check the official press release about CPI, to see first-hand what I'm talking about look at martix 1...

    http://www.statistics.gr/portal/page/portal/ver-1/ESYE/BUCKET/A0515/PressReleases/A0515_DKT87_DT_MM_12_2010_01_F_GR.pdf

    ReplyDelete