One of the major arguments against Bitcoin as a potential replacement for fiat currencies is that, since the supply of Bitcoin will eventually become fixed, it must necessarily cause deflation since the demand for currency will continue to rise. This argument, as it turns out, is based either on a conflation of the different kinds of money supply or on a belief that the ratio of “broad” to “high-powered” money cannot adjust sufficiently.
"High-powered money" is the part of the money supply that Bitcoin could constitute: cash and bank reserves. "Broad money," on the other hand, is the total liquidity available in the economy: demand deposits, short-term treasuries and commercial paper, money market accounts, et cetera. These can be used directly for transactions (via check, wire transfer, direct asset transfer, etc.), or they can be converted quickly to a form that can be used for transactions. Either way, they satisfy people’s demand to hold currency.
It turns out that the multiplier between high-powered and broad money increases steadily over time, even when regulation and monetary policy otherwise remain static. Milton Friedman mentions this phenomenon in this podcast with Russ Roberts, but you can see it for yourself just by looking at the Fed’s and BoE’s own statistics since their inception.
Along with credit cards and innovative mobile payment systems like Pay with Square and SMS payments, this takes care of most situations, but the demand for good ol’ cash also increases over time. One only need look back to the days of gold-backed free banking in the US, Canada, and Scotland to see that we’ve already solved this problem: bank notes. We’re already seeing various entities issuing physical representations of Bitcoin. This would be similar, but backed by the total credit and assets of the bank, even though they’d still be denominated in Bitcoin.
Of course, anyone with even a passing familiarity with US banking history would point out that banks can fail, causing people to lose their life’s savings, and that with bank notes even the money in your wallet would be gone. Well, it turns out that bank failures in the United States have almost exclusively been the product of extremely bad overregulation. In particular, banks were not allowed to branch out of state and sometimes even out of the local area, meaning they could not diversify. For example, a bunch of banks in Texas failed just because the oil price went down. As if that weren’t enough, many states required banks to hold a large fraction of their reserves in that state’s debt, and then (predictably) kept defaulting on their debt since they had captive lenders. If you don’t believe me, Canada had nationwide branching and not a single bank failure during the Great Depression.
Bitcoin is actually pretty darned inconvenient in cases where you would want to use cash; it takes a random amount of time averaging to ten minutes to complete a transaction. In fact, it’s even inconvenient for most online purchases, since it’s non-repudiable and doesn’t offer the protections of credit cards. Along with bank notes and various innovations, a fixed quantity of actual Bitcoin should prove sufficient to address the demand without the need for deflation. And since in any real world implementation banks would likely also hold precious metals, Bitcoin would only be one of several reserve currencies anyway.