The only people who do see gains are people who hold amounts in excess of their short term needs -- bitcoin's monetary policy is building the notion that the rich get richer while everyone else is coerced to transact at suboptimal prices right into the heart of it. Bitcoin is fundamentally about exploitation to benefit the wealthy true believers.
It's a deal with Moloch to use bitcoin.
In a sense, this already happens in dollars, in the form of opportunity costs for relatively stable investments like mutual funds. Assuming zero transaction costs, what's the difference between spending Bitcoin that would otherwise appreciate 5% value per year, vs. spending dollars, thereby foregoing the opportunity to invest in a 5%/year mutual fund?
(Obviously, I'm assuming an ecosystem of both fiat and crypto; it might be a different story if we were debating a deflationary-crypto-only marketplace.)
Also: it's worth noting that the steady influx of new coins, deflationary and otherwise, means that if one looks at the entire crypto ecosystem rather than just Bitcoin, the deflationary problem may end up being practically non-existent. For better or worse, the whole thing is effectively a live experiment in the Free Banking model: creating markets for currency itself.
Risk. The decreasing yield of mining ensures that the future value of bitcoin will appreciate (assuming demand stays flat). There is no risk that supply might suddenly increase.
In that scenario, how is spending Bitcoin any different, micro-economically, than foregoing the opportunity to invest in a similarly-appreciating mutual fund? What's absent from my model?
Assuming people are rational actors, they would spend as little as possible and hoard the rest.
So then we need to ask: "how is holding bitcoin different than investing in a similar mutual fund".
Holding bitcoin decreases the money supply driving up value artificially, whereas investing in a mutual fund increases the money supply available for investment and drives economic growth.