You might like "Less Than Zero: The Case for a Falling Price Level in a Growing Economy". Available for free online at http://www.iea.org.uk/sites/default/files/publications/files...

More or less the author argues that stable nominal spending, ie a constant level of nominal GDP, is pretty much ideal. And would lead to falling prices as you suggest.

If you replace constant level with 'target a level of nominal GDP that rises 4% every year' you have pretty mainstream position.

Inflation measures are indeed somewhat subject. Nominal GDP has less suggement calls.

(It's still useful to try and measure inflation. But perhaps it should not be a policy target.)