Essentially, if you think that a currency will be worth more tomorrow than it is today there is little reason for investment (or even spending!). It quickly becomes a spiral where no one wants to spend. Inflation provides a nice kick in the pants for people to put their money to good use.
This isn't a concern yet for Bitcoin because in the scheme of world economies BTC simply doesn't matter (it could go away tomorrow without any noticeable effects), but if in the future it becomes a critical thing it would be a major concern.
For one thing it assumes a massive collective irrationality where people's expectations are all rendered berserk and plunged into a negative time preference. It's a very tough gambit to make that people can withdraw their propensity to consume to such a high extent. It's tough to presume that the heterogeneous stock of capital and the time structure of production will just stand still to a deflationary pressure and not readjust to add more stages or adjust the price spreads in between. [1] Of course, BTC being a global currency means it exists in competition and per Gresham's law can always be driven out. Not a catastrophe.
One of the more idiotic ideas to come out of neoclassical economics is the assumption that a negative time preference is irrational or impossible, when most of us have one (pension, passing wealth on to one's kids, saving up to buy big ticket items, etc.).
Indeed, what would really be irrational is spending all of your money immediately.
Economists agree: deflation is either good, or bad, or irrelevant: http://ftalphaville.ft.com/2015/03/23/2122452/economists-agr...
https://mises.org/library/hayek-paradox-saving
The "deflation makes people not spend" is a hand-wavy argument. There is zero empirical evidence for it.
Actually with negative interest rates proving possible (almost nobody thought they were a few years ago) if you had a deflationary currency you could always use negative interest rates to control demand. I am sure some smart economist has looked into this.
