Readers of George Selgin's post just below should be aware that there is a an extensive comment section about it, including a few of his own remarks, over at Marginal Revolution.
Now, on to what I promised to discuss in my last post, some time ago. Every monetary system faces a number of major problems to solve or fail at. Here I will discuss how a free banking system, assumed for the sake of concreteness to operate on a fixed exchange rate with gold or on what George Selgin termed a quasi commodity standard (a frozen fiat monetary base), compares with a central banking system, assumed to operate on a floating exchange rate, with inflation as the target. In a future post I will have more to say about targeting the path of nominal GDP, but I will leave it aside now because it is not a system that has yet been put into practice in explicit fashion.
Knowledge. Does the monetary system generate enough knowledge and enable people to use it efficiently enough to coordinate activity well? The major argument in favor of free banking is that the competition it permits generates knowledge of consumer preferences, especially about preferences for saving as opposed to consuming, that a central banking system cannot elicit to the same degree. Unfortunately, it is an argument that mainstream economics still typically ignores.
Technique. Can the monetary system hit its stated targets with a high degree of accuracy? We know from experience that an exchange rate is an easy target to hit in a technical sense, and that it is likewise easy for the public to tell the difference between meeting and not meeting the target. A similar point applies to a quasi commodity money: if banks promise to redeem their liabilities for the quasi commodity money at a specified rate, either they do or they don't. When it comes to a central bank meeting an inflation target, in practice judging whether it is meeting the target has proved harder. The target is generally forward-looking, but judgment about it is backward-looking. As a result, central banks typically have not suffered any punishment for persistently missing targets by what seem like wide margins, given the small tolerance bands within which they claim to operate.
Politics. Can the monetary system be sufficiently insulated from politics that it can coordinate activities in an economically efficient way? This is the question of the Public Choice school of economists. No monetary system provides perfect protection against politically motivated inefficiencies. By design, though, under free banking the government is not both player and referee, whereas under central banking it is a bit of both and where it also owns large commercial banks the combination of roles and the potential for conflicts between them is even stronger. Since the high inflations that plagued many countries in the 1980s and early 1990s, central banks around the world have been remarkably successful in asserting greater independence from political pressure than they previously had. It is still the case, though, that a central bank by design is subject to a higher degree of overt political influence than free banks.
Redeemability. Banks operate with less than 100% reserves, with the level of reserves reflecting their judgments about how big a cushion they need against mistakes in estimating the difference between debits against them and credits in their favor. In a system where the ultimate monetary base, whether gold or a quasi commodity, has a limited supply, if the public decides that it wants to redeem bank deposits for the monetary base quickly and on a large scale, as in a financial panic, banks must contract credit or limit redeemability. Under a floating exchange rate fiat system, the central bank can create further supplies of the monetary base at will, so no system-wide question of redeemability need arise, though the central bank may allow individual commercial banks to fail. The downside of the unlimited power to create reserves is, of course, that it contains the potential for unlimited inflation.
Credibility. How believable are the promises that the monetary authority or banks make? With a fixed exchange rate into gold or into a quasi commodity money (which itself floats), the ability to hit the target and to judge easily whether the target it being hit contribute to credibility. With inflation targeting, credibility is fuzzier. Critics of the gold standard might say that a free banking gold standard goes overboard on credibility, tying banks to a standard that in some cases is best abandoned. That is another topic for a later post, concerning implicit and explicit "option clauses" that some free banking systems have had.
Economic growth. Economists often compare economic growth under different monetary arrangements. To me, growth is secondary in the sense that I do not think it is something that monetary policy can target. A good monetary policy can facilitate growth, by enlarging the sphere of exchanges, but it is only one factor in growth and often a decidedly secondary factor compared to property rights, taxes, regulation, etc. In the cases where monetary policy is decisive for influencing growth, it is usually where it is spectacularly bad rather than spectacularly good.