Hayek on the Creation of Moral Hazard by Central Banks

Mises Institute Archives: https://mises.org/library/why-mises-and-not-hayek
Friedrich Hayek, Ludwig von Mises, free banking,
F.A. Hayek and Ludwig von Mises from the Mises Institute Archives

Some years ago I published a paper on the banking theory and policy views of the important twentieth-century economist Friedrich A. Hayek, entitled “Why Didn't Hayek Favor Laissez Faire in Banking?[1] Very recently, working on a new paper on Hayek’s changing views of the gold standard, I discovered an important but previously overlooked passage on banking policy in a 1925 article by Hayek entitled “Monetary Policy in the United States After the Recovery from the Crisis of 1920.” I missed the passage earlier because the full text of Hayek’s article became available in English translation only in 1999, the same year my article appeared, in volume 5 of his Collected Works. Only an excerpt had appeared in translation in Money, Capital, and Fluctuations, the 1984 volume of Hayek’s early essays.[2]

Hayek wrote the article in December 1924, very early in his career. In May 1924 he had returned from a post-doctoral stay in New York City and had begun participating in the Vienna seminar run by Ludwig von Mises. It is safe to say that the passage I am about to quote reflects Mises’ influence, since the article cites him, and in many ways takes positions opposite to those Hayek had taken in an earlier article that he wrote while still in New York.

The main topic of the 1925 article is the Federal Reserve’s policies in the peculiar postwar situation in which, as Hayek put it, the US “emerged from the war … as the only country of importance to have retained the gold standard intact.” The US had received “immense amounts” of European gold during and since the war (Hayek documents this movement with pertinent statistical tables and charts), and now held a huge share of the world’s gold reserves — more gold reserves than the Fed knew what to do with. European currencies, having left the gold standard to use inflationary finance during the First World War, and not having yet resumed direct redeemability, were for the time being pegged to the gold-redeemable US dollar. This was a new and unsettled “gold exchange standard,” unlike the prewar classical gold standard in which major nations redeemed their liabilities directly for gold and held their own gold reserves. Rather than delve into what Hayek had to say about that topic, I want to convey what he said about banking.

In section 8 of the article (pp. 145-47 in the 1999 translation), Hayek gives a favorable evaluation of free banking as against central banking. Having overlooked this passage, I had previously thought that Hayek first addressed free banking in his 1937 book Monetary Nationalism and International Stability. Hayek does not embrace free banking as an ideal, first-best system, because he thought it prone to over-issue (as I discussed in my 1999 article based on Hayek’s other writings). But he criticizes the Federal Reserve Act for relaxing rather than strengthening the prior system’s constraints against excess credit expansion by American commercial banks.

Hayek begins the passage with a caution that the intended result of creating a central bank, when the intention is to avoid or mitigate financial crises, need not be the actual result:

It cannot be taken for granted that a central banking system is better suited to prevent disturbances in the economy stemming from excessive variations in the volume of available bank credit than a system of independent and self-reliant commercial banks run on purely private enterprise (liquidity, profitability) lines.

By standing ready to help commercial banks out of liquidity trouble, central banks give “added incentive … to commercial banks to extend a large volume of credit.” In modern terminology, a lender of last resort creates moral hazard in commercial banking. A free banking system (my phrase, not his) restrains excessive credit creation by fear of failure:

In the absence of any central bank, the strongest restraint on individual banks against extending excessive credit in the rising phase of economic activity is the need to maintain sufficient liquidity to face the demands of a period of tight money from their own resources.

Hayek’s belief that the pre-Fed US system did not restrain credit creation firmly enough is understandable in light of the five financial panics during the fifty years of the federally regulated “National Banking system” that prevailed between the Civil War and the First World War. He might have noted, however, that the National Banking system was a system legislatively hobbled by branching and note-issue restrictions rather than a free banking system or a system “run on purely private enterprise lines.”[3] The Canadian banking system, lacking those restrictions, did not experience financial panics during this period (or even during the Great Depression) despite having an otherwise similar largely agricultural economy.

Despite the flawed character of the pre-Fed system, Hayek judged that the Federal Reserve Act made the situation worse rather than better by loosening the prevailing constraints against unwarranted credit expansions:

Had banking legislation had the primary goal to prevent cyclical fluctuations, its main efforts should have been directed towards limiting credit expansion, perhaps along the lines proposed — in an extreme, yet ineffective way — by the theorists of the “currency school,” who sought to accomplish this purpose by imposing limitations upon the issuing of uncovered notes. … Largely because of the public conception of their function, central banks are intrinsically inclined to direct their activities primarily towards easing the money market, while their hands are practically tied when it comes to preventing economically unjustified credit extension, even if they should favour such an action. …

This applies especially to a central banking mechanism superimposed on an existing banking system. … The American bank reform of 1913-14 followed the path of least resistance by relaxing the existing rigid restraints of the credit system rather than choosing the alternative path …

Thus the Fed was granted the power to expand money and credit, a power that “was fully exploited during and immediately after the war,” not waiting for a banking liquidity crisis. The annual inflation rate in the United States, as measured by the CPI, exceeded 20 percent in 1917, and remained in double digits for the next three years (17.5, 14.9, and 15.8) before the partial reversal of 1921. Hayek (p. 147) observed ruefully “how large an expansion of credit took place under the new system without exceeding the legal limits and without activating in time automatic countermeasures forcing the banks to restrict credit.” He concluded: “There can be no doubt that the introduction of the central banking system increased the leeway in the fluctuations of the volume of bank credit in use."

Here Hayek reminds us that a less-regulated banking system does not need to be perfect to be better than even well-intentioned heavier regulatory intervention. Good intentions do not equal good results in bank regulation.


[1] Lawrence H. White, "Why Didn't Hayek Favor Laissez Faire in Banking?History of Political Economy 31 (Winter 1999), pp. 753-769. I also published a companion paper on his monetary theory: Lawrence H. White "Hayek's Monetary Theory and Policy: A Critical Reconstruction," Journal of Money, Credit, and Banking 31 (February 1999), pp. 109-20.

[2] F. A. Hayek, “Monetary Policy in the United States after the Recovery from the Crisis of 1920,” in Good Money Part I: The New World, ed. Stephen Kresge, vol. 5 of The Collected Works of F. A. Hayek (Chicago: University of Chicago Press, 1999); F. A. Hayek, Money, Capital, and Fluctuations: Early Essays, ed. Roy McCloughry (Chicago: University of Chicago Press, 1984).

[3] See Vera C. Smith, The Rationale of Central Banking (Indianapolis: Liberty Fund, 1990), chapter 11; and George A. Selgin and Lawrence H. White, "Monetary Reform and the Redemption of National Bank Notes, 1863-1913," The Business History Review 68, no. 2 (1994), pp. 205-43.


  1. Above: "Had banking legislation had the primary gold to prevent cyclical fluctuations…" "goal" not "gold." Freudian slip?

  2. Amen, Brother White. Applying these lessons to today's banking scene, are bankers trading Bitcoin at $10,000+ and financing mortgages (again) with only 3 percent equity or down payments as a policy of judicious restraint when all the signs in the heavens are that existing bubbles soon might pop? A prudent banker might require larger down payments for houses (and cash in full and on the counter for Bitcoin purchases — 100 pct. margin). — Walker Todd, Middle Tennessee State University

  3. He thought that information flowed, was quantized and contained the economic model. He discovers this ideal, algebraic substrate, information. Why does he think it sufficient to model the economy?

    He selected this unit that flows so as to what?? To fill in channels with bell shaped noise distributions, as Shannon concluded. Shannon also called it sphere packing. The concept is that information is the optimal queueing element, in the solution to an algorithm for optimal packing, Huffman proves them both right. The economy is all about queueing.

    The Huffman encoder is the checkout manager. He looks at your basket, and if you have n1 to n2 item, then you goto line k. He keeps all the lines equally busy, sometime changing their boundary values.

    1. This comment may be relevant to another post, but not to this one, unless I'm missing something.

      1. There were other posts on Hayek on the other blogs, so, yes this was an indirect alert about another Hayek topic.

  4. While it's likely that the Fed aggravated the situation somehow, the disastrous pattern of 1914-1917 inflation, followed by the 1929-32 correction, was primarily due to Europe's having gone off gold during WW I and then abruptly trying to get back on in the late 20s – early 30s. The Fed added its own inflation 1917-19 to the ongoing global gold inflation, but that was mostly corrected 1920-22.

    Hayek is probably right that the Fed gave banks and depositors an illusion of safety that made things worse. The National Bank system had many problems and was not a fully free banking system despite free entry, because of the 1865 tax on notes of state banks.

    One major problem with the NBS was the understanding that banks should suspend their deposits if they could not meet their reserve requirements. Reserves should be there to meet withdrawals whether excess or "required". For a reserve requirement to be meaningful, it should be interpreted that loans may only be made out of excess reserves, so that a bank with a reserve shortage would be prohibited from making new loans, not from honoring its existing liabilities.

    The reluctance of central banks to use their abundant gold reserves to meet redemptions, and therefore their great appetite for excess gold reserves, severely aggravated the 1929-33 deflation.

  5. I agree that there is a hazard in being lender of last resort.

    What troubles the situation is the possibility of a run on the bank – something which is not at all necessary to include in the design of the banking system.

    If that possibility is taken away, then any bank failure is well deserved and should not be rescued. The board of directors and the shareholders should be abandoned, and the bank should be rescued to protect the depositors and savers. Save disruptions.

    How can a run on the bank be avoided?

    Not by deposit insurance – which is too expensive.

    Not by a tax-payer bail out


    Every economy goes through cycles of having too little money in spending circulation and then too much spending or spending money in circulation. When people are nervous they save, and when they are over-borrowed they pay down their loans. Then it all goes into reverse.

    The function of monetary policy is to manage this cycle so as to avoid a recession and to optimise the national output by ensuring that there is enough demand to purchase it.

    To do this there is a problem with inflation targeting.

    As prices rise more money in circulation becomes necessary. A higher priced economy needs more money than a lower priced economy.

    If not provided, then a recession looms and the target of optimising the national output is not achieved.

    If we accept that interest rates must rise and fall, we also know that there will be a lot of damage done to the housing sector and other parts of the economy.

    We at the Ingram School of economics say that the first step to take is to re-arrange the way lending and savings contracts are written so that these undulations do not cause needless damage.

    Then we must change the way that currency markets operate so that the price of the trading currency is not mixed up with the price of the international investment currency. Difficult but worth the effort.

    Then we must change the banking system so that banks are no longer allowed to create new money.

    They all become deposit taking institutions.

    The central bank auctions deposits to them for on-lending

    We get a market rate of interest and we get enough credit in circulation to enable the national output to be purchased as far as credit is concerned.

    The overall result is that all prices, costs, and values become free to rise without causing disruption to the economy. This is called creating the KFPP Platform for core price adjustments. All other price adjustments being the result of real economic forces. I explain that below.

    As Keynes said, if all core prices were to double then people would be wholly unaffected. What he actually said was that if money were to halve in value and all prices, costs, asset values and earnings were to double (same thing – money halves in value), then people would be wholly unaffected.
    (‘A tract on Monetary Reform, Macmillan 1923).

    So that is what we need to arrange.

    Then we do not need a strict inflation target. The Keynesian Floating Platform Paradigm = KFPP Platform upon which all prices, costs, and values and earnings are rising as money falls in value, will mop up the surplus money and bring the inflation to an end. People will be largely unaffected. Savings value preserved, mortgage repayment costs rising appropriately and falling every year in actual value to avoid payments fatigue, and so forth. We show how to do this in our university course – read our website http://macro-economic-design.blogspot.com

    So as I said, the economy will enter a new cycle and as most economies grow they will need more new money, even without higher prices.

    When a bank fails, it can be rescued with new money. It costs nothing for the central bank to create it.

    Just let the shareholders go and let the board of directors go.

    No need for a lender of last resort.

    No moral hazard there.

    Just a different way of sharing the new money.

    On sharing the new money of both kinds, the monetary target is to supply all economic output with a similar stimulus – not to embark upon some major road building project of anything like that. And not to ask the government to borrow and spend.

Comments are closed.