[wdmmg-discuss] Government Income Data

[Lisa Evans]

This is just a quick note to say it is difficult to obtain a rich source 
of income data because of the way HMRC has outsourced much of its data 
management.

For example this request for more information on the databases used to 
store each tax:

http://www.whatdotheyknow.com/request/database_provider_for_each_tax_t

was rejected and given this reply:

"Our IT supplier, Capgemini, under the Aspire contract, holds the 
information on our behalf but not at the level of detail requested. This 
is because Capgemini don't have to collate the information requested at a 
single point in their day to day activities and it would not be easy to 
retrieve as it is dispersed across different databases. Assembling the 
information would exceed the appropriate limit in section 12 FOIA 
consequently HMRC will not be processing your request further. "

Suggestions on how to deal with this are most welcome.

On Wed, 24 Mar 2010, Lisa Evans wrote:

> Our aim is to calculate an individual's contribution to the government income 
> (to personalise the expenditure data we have). However, we do not want to ask 
> more questions than is absolutely necessary (the aim is for at most 10 
> questions).
>
> This requires identifying:
> - the main taxes an individual or family would be subject to
> - the rates and conditions for these taxes
> - the national statistics for these taxes so we can work out the average
>   tax the individual will pay.
>
> At the moment we've collected most of the information in the main CKAN 
> package here:
>
> http://www.ckan.net/package/hmrc-national-statistics-income-tax-and-personal-incomes
>
> Basically, we need to know:
>
>   1. Income -> income tax paid
>    * Here's income and tax by gender and location 
> http://www.hmrc.gov.uk/stats/income_distribution/menu-by-year.htm#311
>      But we are looking for more important factors that will affect tax paid
>   1a: NICs and PAYEs -> how relevant are these?
>   2. Mapping of spending on special areas (smoking, petrol etc) to tax
> income.
>     * e.g. for each £1 spent on cigarettes there is 20p tax revenue
>  3. Mapping of income to VAT consumption (or simply VAT payment) -- crude 
> approach is to divide total VAT by no. of people
>
> Over the next week I'm working on filling out the income section in our data 
> guide:
>
> http://docs.google.com/Doc?docid=0AYn3JiuouxLUZGN0eGIzOWNfNDhjemQ1N2hnbg&hl=en
>
> If you want to help out please dive in.
>
> Lisa
>