[euopendata] Study says charge for public data...

[Peter Krantz]

On Thu, Jan 13, 2011 at 22:40, James McKinney <oxford.tuxedo at gmail.com> wrote:
>
> I agree there is no need. But if a government chooses to have NC
> restrictions in order to generate revenue from commercial use, then,
> assuming it is at all successful in generating revenue, I have yet to
> hear a compelling argument that it should stop doing so.
>

Let's try a pure financial perspecive.

1. Misconception of "revenue"

It depends on how you calculate/define "revenue". I did a case here in
Sweden where I looked at the swedish weather data that only is
accessible under commercial terms. The agency responsible (SMHI) has
some 200 M Sek in annual revenue from selling data. It is of course a
valid question where that money should come from if data was provided
free? [1]

When you look at where the revenue comes from (the first line
customers) I found that 71% is from _other government organizations_.
So the actual revenue from a pan-government perspective is 58 M
instead of 200 M. From this you can also remove any profit made (in
this case 9 M Sek). Of these 49M I have reason to believe (because
there is not enough information) that a large part come from
commercial data companies that in turn resells data (with very little
added value) to other government agencies. So the figure is well below
49M. A similar scenario is for geodata where our agency states in
their annual report that a majority of the revenue comes from other
government agencies.


2. Transaction costs between agencies

It is very likely that transaction costs between agencies could be
considered "waste" from a lean production perspecive. They cost money
for tax payers and provide no tangible benefit.


3. Reduced capability to provide efficient services for other agencies.

Now, if you take an outside-in perspective it is interesting to talk
to those who _didn't buy data_. I have heard of several project
proposals in the public sector that weren't started because _data
would be too expensive to buy to fit in the budget_. Thus, the
second-line effect of agencies not buying data is that there are
projects with a percieved value to society that aren't materialized
because a different agency is financed with fees for data. What is the
inefficiency cost of not running these projects?


4. Unfair competition scenarios

If an agency starts adding value by refining raw data it collects I
would argue that the public sector is in competition with the private
sector (a common scenario I am sure). The advantages of having direct
free access to the raw data and being a government agency with reduced
financial risk makes it difficult to compete for a startup with an
innovative use of technology. If a majority of the customers are from
thepublic sector their are more likely to buy from another agency.


5. Reduced efficiency in the private sector

Weather data is a key component to regulate energy use in buildings
(e.g. planned heating etc). What is the cost of having fewer buildings
install data driven systems? It is likely that a lower price of this
type of equipment due to reduced pricing of weather data could lead to
a large decrease in energy use covering the remainder of the initial
perceived revenue.


The benefit of data appears when it is used. I printed that and put it
on my wall.


Regards,

Peter

[1]: http://www.opengov.se/blogg/2009/vem-betalar-data/ (only in
swedish unfortunately. pls use google translate)