[open-economics] Financial comparisons

Thu Jun 6 19:14:15 UTC 2013

Hi

I did consider this problem while presenting historical budgets and 
accounts for Danish municipalities (for charts see 
http://kommune.politiken.dk/tabel.php?kid=101&&type=ud). Even for a 7 
years period it's not fair to present the numbers in real values but how 
to adjust?

There are multiple official price-indexes all used by government in 
different contexts: consumer-price index, a wage index, netto-prices, an 
index for construction, for road constructions and so on. Parts of the 
budgets are grouped in similar ways. Fx. construction expenses are 
identifiable and these expenses could be adjusted using the 
construction-price-index (same goes for salaries, road-construction 
etc.). This could be a reasonable approach when focusing on a specific 
expenditure post in a historic perspective. However to use different 
price-indexes on different parts of the same budget is obviously 
problematic because the relative size of each type of expenditure 
changes over time.

I choose to use a combined consumer-price and wage index to adjust the 
historical data. This is not perfect but I found it better than using 
multiple indexes for different parts of the budget and better than 
presenting figures in real values.

Best, Niels Erik
---
http://www.buhlrasmussen.eu
(+45) 2680 9492

On 2013-06-06 18:03, Guo Xu wrote:
> Hi Tryggvi,
>
> How far do you want to go back?
>
> Calculating deflators is not trivial, particularly as baskets are
> different and keep changing over time! The standard source for CPI is
> from the Penn World Tables. This should give you data running back to
> 1950.
>
> https://pwt.sas.upenn.edu/php_site/pwt_index.php
>
> The variables of interest to you will be the p* variables that give
> the price levels.
>
> Guo
>
> On 6 June 2013 16:39, Tryggvi Björgvinsson
> <tryggvi.bjorgvinsson at okfn.org> wrote:
>> Hi all and sorry for crossposting,
>>
>> I've been working on "amount normalisation" for OpenSpending to 
>> allow us to
>> do historical comparisons more accurately using real value or look 
>> at
>> amounts at the current nominal value (I hope I'm using the correct 
>> terms).
>>
>> I decided to use consumer price indexes and I've collected the data 
>> from the
>> World Bank. The problem with that data is that many years are 
>> missing for
>> some countries (either the CPI wasn't being collected, isn't being 
>> collected
>> any more or has never been collected).
>>
>> Since I don't know how this is usually done I wanted to ask you what 
>> you
>> think would be the best approach:
>>
>> 1. Use the index that's nearest in time (this won't be accurate 
>> though)
>> 2. Interplolate and guess what the index is (not accurate either 
>> since it's
>> not linear)
>> 3. Just say that we can't do historical comparisons nor compute the
>> real/nominal value
>> 4. Something else
>>
>> --
>>
>> Tryggvi Björgvinsson
>>
>> Technical Lead, OpenSpending
>>
>> The Open Knowledge Foundation
>>
>> Empowering through Open Knowledge
>>
>> http://okfn.org/ | @okfn | OKF on Facebook | Blog | Newsletter
>>
>>
>> _______________________________________________
>> open-economics mailing list
>> open-economics at lists.okfn.org
>> http://lists.okfn.org/mailman/listinfo/open-economics
>> Unsubscribe: http://lists.okfn.org/mailman/options/open-economics
>>
>
> _______________________________________________
> open-economics mailing list
> open-economics at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-economics
> Unsubscribe: http://lists.okfn.org/mailman/options/open-economics