Tuesday, April 04, 2006

When Statistics Go Bad

I saw these statistics being used recently over at DPF's blog (by a commenter - not David) as a rebuttal that Australia's effective tax rates are lower than ours.

Evolution of the tax burden, 2000-2005
Single persons without children at 100% of average earnings
http://www.oecd.org/dataoecd/50/41/36371703.pdf


Which at first impression the table appears to show:
Australia 28.3%
New Zealand 20.5%

Yay - New Zealanders rejoice; apparently our tax burden is not as bad as we feel it is... Until we read the third line of the heading which actually explains what we are looking at;
Income tax plus employee and employer contributions less cash benefits as a % of labour costs

So in actual fact the taxation burden on an EMPLOYEE shown by these figures is overstated by the rate of both compulsory superannuation (0% NZ, 9% Australia) AND contributions by employer or employee to employer run superannuation schemes.  So even without assessing the final part of the distortion Australia already has a lower taxation burden than us. 


2 comments:

Gardener said...

The moment I heard Helen Clark claim that tax was higher in Australia without any supporting facts I assumed it was higher here.

Rule of thumb - if anything is claimed by the PM without simple supporting statistics - it's spin.

I just wonder how many half wits (including the half wits who can command a front page of the paper any day they want) actually understand anything about the statistics they use to justify their lies.

Unknown said...

Unfortunatly having been an University statistics tutor, I can say it isn't just the half wits that do not understand stats. It is quite scary how many people have passed 2nd/3rd stage statistics papers who do not understand!

Still - I can't complain too much, peoples continuing misunderstanding of simple concepts is what lets me make money in the options markets :-)