Lesson 3 of 7
Testing a claim against history
Free banking is where the school goes to the archives. A dispute inside the tradition about one Scottish episode shows how hard that test is.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Free banking is where this tradition does economic history: a claim about competitive note issue tested against episodes in Scotland, Canada, Sweden and elsewhere.
- The Scottish case produced a sharp dispute inside the school about whether the banks were free at all and whether the outcome was good.
- The dispute is a lesson in how a historical test fails: the two sides disagree about the facts, the criteria and what would count as success.
Most of this track is theory. This lesson is about the part of the school that spends its time in archives, and about what happened when two members of it read the same episode.
What the two sides read out of the same episode
The thesis under review
- Scottish banking from the early eighteenth century to 1845 was largely free of the central bank.
- Each bank held its own specie reserves.
- The system showed greater cyclical stability than England's.
- The evidence offered for that stability is the lower rate of bank failures.
The reviewer's objection
- The Scottish banks suspended specie payment when England did, in 1797.
- Reserve ratios fell to very low levels in the following decades.
- A standard history records repeated credit expansions and crises there.
- Fewer failures is not evidence of a better system, and may be evidence of the opposite.
A historical test needs agreement on the facts and on the criteria. This one had neither.
Why this is a research programme and not a doctrine
The claim being tested is specific: that competing banks issuing their own notes, with no central bank behind them, are disciplined by the clearing system faster and more reliably than a central authority disciplines them.
That is a claim with a historical record attached. Several periods have been studied by economists in this tradition, and the arguments are conducted with dates, reserve ratios and failure rates rather than with deductions. Whatever else is true of the school, this part of it is empirical in temper.
The Scottish dispute, and what it shows
White's book on British free banking argued the Scottish case in the terms set out above. Rothbard's review, published in the Review of Austrian Economics, disputed both halves.
On freedom, he argues that the Scottish banks suspended payment in specie when the Bank of England did in 1797 and stayed suspended until 1821, which is not what an independent bank on its own specie base can do; and that reserve ratios fell from between ten and twenty percent in the later eighteenth century to between one and three percent in the first half of the nineteenth.
On performance, he argues that a lower rate of bank failures is a weak criterion, because failures are the market's method of restraining credit expansion and their absence may mean the restraint is not working. He cites a standard history of Scottish banking recording expansions and contractions around a long list of crisis years.
A second review in the same volume, by Sechrest, disputes White's thesis on different grounds again.
What a reader should take from it
Not a verdict on Scottish banking. The point is methodological, and it is uncomfortable for everybody.
Three parties to one historical test disagreed about whether the institutions were free, about which outcome measure counts, and about what the measure would show if agreed. That is the normal condition of historical tests in economics, and it is why the evidence lesson in the previous course could not produce a verdict either.
It also shows the tradition arguing with itself in public, in its own journal, about facts. That is worth more, as evidence of intellectual health, than any amount of agreement would be.
Check yourself
What claim does the free banking literature test?
- That central banks cause all recessions
- That competing banks of issue are disciplined by clearing faster and more reliably than by a central authority
- That gold is the only sound money
Show the answer
Answer: B.
The claim is institutional and has a historical record attached, which is what makes it a research programme rather than a doctrine.What is the objection to using bank failures as the measure of success?
- That failure data before 1900 are unavailable
- That failures are the mechanism restraining credit expansion, so few failures may mean the restraint was absent
- That failures are caused by fraud rather than by credit
Show the answer
Answer: B.
Rothbard's review makes exactly this argument and asks for evidence on money, credit and prices instead.What does this dispute illustrate about historical tests?
- That historical evidence is useless in economics
- That the parties can disagree about the facts, the criteria and what the criteria would show
- That the Scottish case was unusually badly documented
Show the answer
Answer: B.
Three participants inside the same tradition disagreed at all three levels, which is the ordinary difficulty rather than a special one.Sources
- Rothbard, 'The Myth of Free Banking in Scotland', Review of Austrian Economics, volume two, 1987, a review of White, Free Banking in Britain: Theory, Experience, and Debate, 1800 to 1845, Cambridge University Press, 1984: the two parts of the thesis under review, the 1797 suspension and its duration, the fall in specie reserve ratios, the argument about bank failures as a criterion, and the citation of Checkland's history of Scottish banking for the list of crisis years.
- Sechrest, 'White's Free-Banking Thesis: A Case of Mistaken Identity', Review of Austrian Economics, volume two, 1987, as a second critical review in the same volume.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- White's book was not obtained for this lesson. Its thesis is reported as his reviewers state it, and a reader who wants the case for free banking should read the book and the theoretical lesson linked above rather than this page.
- No position is taken here on banking regulation of any kind. The lesson is about how a historical claim is tested and why the test was inconclusive.