Archive for the 'Great Depression' Category

The Rises and Falls of Keynesianism and Monetarism

The following is extracted from a paper on the history of macroeconomics that I’m now writing. I don’t know yet where or when it will be published and there may or may not be further installments, but I would be interested in any comments or suggestions that readers might have. Regular readers, if there are any, will probably recognize some familiar themes that I’ve been writing about in a number of my posts over the past several months. So despite the diminished frequency of my posting, I haven’t been entirely idle.

Recognizing the cognitive dissonance between the vision of the optimal equilibrium of a competitive market economy described by Marshallian economic theory and the massive unemployment of the Great Depression, Keynes offered an alternative, and, in his view, more general, theory, the optimal neoclassical equilibrium being a special case.[1] The explanatory barrier that Keynes struggled, not quite successfully, to overcome in the dire circumstances of the 1930s, was why market-price adjustments do not have the equilibrating tendencies attributed to them by Marshallian theory. The power of Keynes’s analysis, enhanced by his rhetorical gifts, enabled him to persuade much of the economics profession, especially many of the most gifted younger economists at the time, that he was right. But his argument, failing to expose the key weakness in the neoclassical orthodoxy, was incomplete.

The full title of Keynes’s book, The General Theory of Employment, Interest and Money identifies the key elements of his revision of neoclassical theory. First, contrary to a simplistic application of Marshallian theory, the mass unemployment of the Great Depression would not be substantially reduced by cutting wages to “clear” the labor market. The reason, according to Keynes, is that the levels of output and unemployment depend not on money wages, but on planned total spending (aggregate demand). Mass unemployment is the result of too little spending not excessive wages. Reducing wages would simply cause a corresponding decline in total spending, without increasing output or employment.

If wage cuts do not increase output and employment, the ensuing high unemployment, Keynes argued, is involuntary, not the outcome of optimizing choices made by workers and employers. Ever since, the notion that unemployment can be involuntary has remained a contested issue between Keynesians and neoclassicists, a contest requiring resolution in favor of one or the other theory or some reconciliation of the two.

Besides rejecting the neoclassical theory of employment, Keynes also famously disputed the neoclassical theory of interest by arguing that the rate of interest is not, as in the neoclassical theory, a reward for saving, but a reward for sacrificing liquidity. In Keynes’s view, rather than equilibrate savings and investment, interest equilibrates the demand to hold the money issued by the monetary authority with the amount issued by the monetary authority. Under the neoclassical theory, it is the price level that adjusts to equilibrate the demand for money with the quantity issued.

Had Keynes been more attuned to the Walrasian paradigm, he might have recast his argument that cutting wages would not eliminate unemployment by noting the inapplicability of a Marshallian supply-demand analysis of the labor market (accounting for over 50 percent of national income), because wage cuts would shift demand and supply curves in almost every other input and output market, grossly violating the ceteris-paribus assumption underlying Marshallian supply-demand paradigm. When every change in the wage shifts supply and demand curves in all markets for good and services, which in turn causes the labor-demand and labor-supply curves to shift, a supply-demand analysis of aggregate unemployment becomes a futile exercise.

Keynes’s work had two immediate effects on economics and economists. First, it immediately opened up a new field of research – macroeconomics – based on his theory that total output and employment are determined by aggregate demand. Representing only one element of Keynes’s argument, the simplified Keynesian model, on which macroeconomic theory was founded, seemed disconnected from either the Marshallian or Walrasian versions of neoclassical theory.

Second, the apparent disconnect between the simple Keynesian macro-model and neoclassical theory provoked an ongoing debate about the extent to which Keynesian theory could be deduced, or even reconciled, with the premises of neoclassical theory. Initial steps toward a reconciliation were provided when a model incorporating the quantity of money and the interest rate into the Keynesian analysis was introduced, soon becoming the canonical macroeconomic model of undergraduate and graduate textbooks.

Critics of Keynesian theory, usually those opposed to its support for deficit spending as a tool of aggregate demand management, its supposed inflationary bias, and its encouragement or toleration of government intervention in the free-market economy, tried to debunk Keynesianism by pointing out its inconsistencies with the neoclassical doctrine of a self-regulating market economy. But proponents of Keynesian precepts were also trying to reconcile Keynesian analysis with neoclassical theory. Future Nobel Prize winners like J. R. Hicks, J. E. Meade, Paul Samuelson, Franco Modigliani, James Tobin, and Lawrence Klein all derived various Keynesian propositions from neoclassical assumptions, usually by resorting to the un-Keynesian assumption of rigid or sticky prices and wages.

What both Keynesian and neoclassical economists failed to see is that, notwithstanding the optimality of an economy with equilibrium market prices, in either the Walrasian or the Marshallian versions, cannot explain either how that set of equilibrium prices is, or can be, found, or how it results automatically from the routine operation of free markets.

The assumption made implicitly by both Keynesians and neoclassicals was that, in an ideal perfectly competitive free-market economy, prices would adjust, if not instantaneously, at least eventually, to their equilibrium, market-clearing, levels so that the economy would achieve an equilibrium state. Not all Keynesians, of course, agreed that a perfectly competitive economy would reach that outcome, even in the long-run. But, according to neoclassical theory, equilibrium is the state toward which a competitive economy is drawn.

Keynesian policy could therefore be rationalized as an instrument for reversing departures from equilibrium and ensuring that such departures are relatively small and transitory. Notwithstanding Keynes’s explicit argument that wage cuts cannot eliminate involuntary unemployment, the sticky-prices-and-wages story was too convenient not to be adopted as a rationalization of Keynesian policy while also reconciling that policy with the neoclassical orthodoxy associated with the postwar ascendancy of the Walrasian paradigm.

The Walrasian ascendancy in neoclassical theory was the culmination of a silent revolution beginning in the late 1920s when the work of Walras and his successors was taken up by a younger generation of mathematically trained economists. The revolution proceeded along many fronts, of which the most important was proving the existence of a solution of the system of equations describing a general equilibrium for a competitive economy — a proof that Walras himself had not provided. The sophisticated mathematics used to describe the relevant general-equilibrium models and derive mathematically rigorous proofs encouraged the process of rapid development, adoption and application of mathematical techniques by subsequent generations of economists.

Despite the early success of the Walrasian paradigm, Kenneth Arrow, perhaps the most important Walrasian theorist of the second half of the twentieth century, drew attention to the explanatory gap within the paradigm: how the adjustment of disequilibrium prices is possible in a model of perfect competition in which every transactor takes market price as given. The Walrasian theory shows that a competitive equilibrium ensuring the consistency of agents’ plans to buy and sell results from an equilibrium set of prices for all goods and services. But the theory is silent about how those equilibrium prices are found and communicated to the agents of the model, the Walrasian tâtonnement process being an empirically empty heuristic artifact.

In fact, the explanatory gap identified by Arrow was even wider than he had suggested or realized, for another aspect of the Walrasian revolution of the late 1920s and 1930s was the extension of the equilibrium concept from a single-period equilibrium to an intertemporal equilibrium. Although earlier works by Irving Fisher and Frank Knight laid a foundation for this extension, the explicit articulation of intertemporal-equilibrium analysis was the nearly simultaneous contribution of three young economists, two Swedes (Myrdal and Lindahl) and an Austrian (Hayek) whose significance, despite being partially incorporated into the canonical Arrow-Debreu-McKenzie version of the Walrasian model, remains insufficiently recognized.

These three economists transformed the concept of equilibrium from an unchanging static economic system at rest to a dynamic system changing from period to period. While Walras and Marshall had conceived of a single-period equilibrium with no tendency to change barring an exogenous change in underlying conditions, Myrdal, Lindahl and Hayek conceived of an equilibrium unfolding through time, defined by the mutual consistency of the optimal plans of disparate agents to buy and sell in the present and in the future.

In formulating optimal plans that extend through time, agents consider both the current prices at which they can buy and sell, and the prices at which they will (or expect to) be able to buy and sell in the future. Although it may sometimes be possible to buy or sell forward at a currently quoted price for future delivery, agents planning to buy and sell goods or services rely, for the most part, on their expectations of future prices. Those expectations, of course, need not always turn out to have been accurate.

The dynamic equilibrium described by Myrdal, Lindahl and Hayek is a contingent event in which all agents have correctly anticipated the future prices on which they have based their plans. In the event that some, if not all, agents have incorrectly anticipated future prices, those agents whose plans were based on incorrect expectations may have to revise their plans or be unable to execute them. But unless all agents share the same expectations of future prices, their expectations cannot all be correct, and some of those plans may not be realized.

The impossibility of an intertemporal equilibrium of optimal plans if agents do not share the same expectations of future prices implies that the adjustment of perfectly flexible market prices is not sufficient an optimal equilibrium to be achieved. I shall have more to say about this point below, but for now I want to note that the growing interest in the quiet Walrasian revolution in neoclassical theory that occurred almost simultaneously with the Keynesian revolution made it inevitable that Keynesian models would be recast in explicitly Walrasian terms.

What emerged from the Walrasian reformulation of Keynesian analysis was the neoclassical synthesis that became the textbook version of macroeconomics in the 1960s and 1970s. But the seemingly anomalous conjunction of both inflation and unemployment during the 1970s led to a reconsideration and widespread rejection of the Keynesian proposition that output and employment are directly related to aggregate demand.

Indeed, supporters of the Monetarist views of Milton Friedman argued that the high inflation and unemployment of the 1970s amounted to an empirical refutation of the Keynesian system. But Friedman’s political conservatism, free-market ideology, and his acerbic criticism of Keynesian policies obscured the extent to which his largely atheoretical monetary thinking was influenced by Keynesian and Marshallian concepts that rendered his version of Monetarism an unattractive alternative for younger monetary theorists, schooled in the Walrasian version of neoclassicism, who were seeking a clear theoretical contrast with the Keynesian macro model.

The brief Monetarist ascendancy following 1970s inflation conveniently collapsed in the early 1980s, after Friedman’s Monetarist policy advice for controlling the quantity of money proved unworkable, when central banks, foolishly trying to implement the advice, prolonged a needlessly deep recession while central banks consistently overshot their monetary targets, thereby provoking a long series of embarrassing warnings from Friedman about the imminent return of double-digit inflation.


[1] Hayek, both a friend and a foe of Keynes, would chide Keynes decades after Keynes’s death for calling his theory a general theory when, in Hayek’s view, it was a special theory relevant only in periods of substantially less than full employment when increasing aggregate demand could increase total output. But in making this criticism, Hayek, himself, implicitly assumed that which he had himself admitted in his theory of intertemporal equilibrium that there is no automatic equilibration mechanism that ensures that general equilibrium obtains.

The Demise of Bretton Woods Fifty Years On

Today, Sunday, August 15, 2021, marks the 50th anniversary of the closing of the gold window at the US Treasury, at which a small set of privileged entities were at least legally entitled to demand redemption of dollar claims issued by the US government at the official gold price of $35 an ounce. (In 1971, as in 2021, August 15 fell on a Sunday.) When I started blogging in July 2011, I wrote one of my early posts about the 40th anniversary of that inauspicious event. My attention in that post was directed more at the horrific consequences of Nixon’s decision to combine a freeze on wages and price with the closing of the gold window, which was clearly far more damaging than the largely symbolic effect of closing the gold window. I am also re-upping my original post with some further comments, but in this post, my attention is directed solely on the closing of the gold window.

The advent of cryptocurrencies and the continuing agitprop aiming to restore the gold standard apparently suggest to some people that the intrinsically trivial decision to do away with the final vestige of the last remnant of the short-lived international gold standard is somehow laden with cosmic significance. See for example the new book by Jeffrey Garten (Three Days at Camp David) marking the 50th anniversary.

About 10 years before the gold window was closed, Milton Friedman gave a lecture at the Mont Pelerin Society which he called “Real and Pseudo-Gold Standards“, which I previously wrote about here. Many if not most of the older members of the Mont Pelerin Society, notably (L. v. Mises and Jacques Rueff) were die-hard supporters of the gold standard who regarded the Bretton Woods system as a deplorable counterfeit imitation of the real gold standard and longed for restoration of that old-time standard. In his lecture, Friedman bowed in their direction by faintly praising what he called a real gold standard, which he described as a state of affairs in which the quantity of money could be increased only by minting gold or by exchanging gold for banknotes representing an equivalent value of gold. Friedman argued that although a real gold standard was an admirable monetary system, the Bretton Woods system was nothing of the sort, calling it a pseudo-gold standard. Given that the then existing Bretton Woods system was not a real gold standard, but merely a system of artificially controlling the price of a particular commodity, Friedman argued that the next-best alternative would be to impose a quantitative limit on the increase in the quantity of fiat money, by enacting a law that would prohibit the quantity of money from growing by more than some prescribed amount or by some percentage (k-percent per year) of the existing stock percent in any given time period.

While failing to win over the die-hard supporters of the gold standard, Friedman’s gambit was remarkably successful, and for many years, it actually was the rule of choice among most like-minded libertarians and self-styled classical liberals and small-government conservatives. Eventually, the underlying theoretical and practical defects in Friedman’s k-percent rule became sufficiently obvious to cause even Friedman, however reluctantly, to abandon his single-minded quest for a supposedly automatic non-discretionary quantitative monetary rule.

Nevertheless, Friedman ultimately did succeed in undermining support among most right-wing conservative, libertarian and many centrist or left-leaning economists and decision makers for the Bretton Woods system of fixed, but adjustable, exchange rates anchored by a fixed dollar price of gold. And a major reason for his success was his argument that it was only by shifting to flexible exchange rates and abandoning a fixed gold price that the exchange controls and restrictions on capital movements that were in place for a quarter of a century after World Was II could be lifted, a rationale congenial and persuasive to many who might have otherwise been unwilling to experiment with a system of flexible exchange rates among fiat currencies that had never previously been implemented.

Indeed, the neoliberal economic and financial globalization that followed the closing of the gold window and freeing of exchange rates after the demise of the Bretton Woods system, whether one applauds or reviles it, can largely be attributed to Friedman’s influence both as an economic theorist and as a propagandist. As much as Friedman deplored the imposition of wage and price controls on August 15, 1971, he had reason to feel vindicated by the closing of the gold window, the freeing of exchange rates, and, eventually, the lifting of all capital controls and the legalization of gold ownership by private individuals, all of which followed from the Camp David meeting.

But, the objective economic situation confronted by those at Camp David was such that the Bretton Woods System could not be salvaged. As I wrote in my 2011 post, the Bretton Woods system built on the foundation of a fixed gold price of $35 an ounce was not a true gold standard because a free market in gold did not exist and could not be maintained at the official price. Trade in gold was sharply restricted, and only privileged central banks and governments were legally entitled to buy or sell gold at the official price. Even the formal right of the privileged foreign governments and central banks was subject to the informal, but unwelcome and potentially dangerous, disapproval of the United States.

The gold standard is predicated on the idea that gold has an ascertainable value, so that if money is made exchangeable for gold at a fixed rate, money and gold will have an identical value owing to arbitrage transactions. Such arbitrage transactions can occur only if, and so long as, no barriers prevent effective arbitrage. The unquestioned convertibility of a unit of currency into gold ensured that arbitrage would constrain the value of money to equal the value of gold. But under Bretton Woods the opportunities for arbitrage were so drastically limited that the value of the dollar was never clearly equal to the value of gold, which was governed by, pardon the expression, fiat rather than by free-market transactions.

The lack of a tight link between the value of gold and the value of the dollar was not a serious problem as long as the value of the dollar was kept essentially stable and there was a functioning (albeit not freely) gold market. After its closure during World War II, the gold market did not function at all until 1954, so the wartime and postwar inflation and the brief Korean War inflation did not undermine the official gold price of $35 an ounce that had been set in 1934 and was maintained under Bretton Woods. Even after a functioning, but not entirely free, gold market was reopened in 1954, the official price was easily sustained until the late 1960s thanks to central-bank cooperation, whose formalization through the International Monetary Fund (IMF) was one of the positive achievements of Bretton Woods. The London gold price was hardly a free-market price, because of central bank intervention and restrictions imposed on access to the market, but the gold holdings of the central banks were so large that it had always been in their power to control the market price if they were sufficiently determined to do so. But over the course of the 1960s, their cohesion gradually came undone. Why was that?

The first point to note is that the gold standard evolved over the course of the eighteenth and nineteenth centuries first as a British institution, and much later as an international institution, largely by accident from a system of simultaneous gold and silver coinages that were closely but imperfectly linked by a relative price of between 15 to 16 ounces of silver per ounce of gold. Depending on the precise legal price ratio of silver coins to gold coins in any particular country, the legally overvalued undervalued metal would flow out of that country and the undervalued overvalued metal would flow into that country.

When Britain undervalued gold at the turn of the 18th century, gold flowed into Britain, leading to the birth of the British of gold standard. In most other countries, silver and gold coins were circulating simultaneously at a ratio of 15.5 ounces of silver per ounce of gold. It was only when the US, after the Civil War, formally adopted a gold standard and the newly formed German Reich also shifted from a bimetallic to a gold standard that the increased demand for gold caused gold to appreciate relative to silver. To avoid the resulting inflation, countries with bimetallic systems based on a 15.5 to 1 silver/gold suspended the free coinage of silver and shifted to the gold standard further raising the silver/gold price ratio. Thus, the gold standard became an international not just a British system only in the 1870s, and it happened not by design or international consensus but by a series of piecemeal decisions by individual countries.

The important takeaway from this short digression into monetary history is that the relative currency values of the gold standard currencies were largely inherited from the historical definitions of the currency units of each country, not by deliberate policy decisions about what currency value to adopt in establishing the gold standard in any particular country. But when the gold standard collapsed in August 1914 at the start of World War I, the gold standard had to be recreated more or less from scratch after the War. The US, holding 40% of the world’s monetary gold reserves was in a position to determine the value of gold, so it could easily restore convertibility at the prewar gold price of $20.67 an ounce. For other countries, the choice of the value at which to restore gold convertibility was really a decision about the dollar exchange rate at which to peg their currencies.

Before the war, the dollar-pound exchange rate was $4.86 per pound. The postwar dollar-pound exchange rate was just barely close enough to the prewar rate to make restoring the convertibility of the pound at the prewar rate with the dollar seem doable. Many including Keynes argued that Britain would be better with an exchange rate in the neighborhood of $4.40 or less, but Winston Churchill, then Chancellor of the Exchequer, was persuaded to restore convertibility at the prewar parity. That decision may or may not have been a good one, but I believe that its significance for the world economy at the time and subsequently has been overstated. After convertibility was restored at the prewar parity, chronically high postwar British unemployment increased only slightly in 1925-26 before declining modestly until with the onset of the Great Deflation and Great Depression in late 1929. The British economy would have gotten a boost if the prewar dollar-pound parity had not been restored (or if the Fed had accommodated the prewar parity by domestic monetary expansion), but the drag on the British economy after 1925 was a negligible factor compared to the other factors, primarily gold accumulation by the US and France, that triggered the Great Deflation in late 1929.

The cause of that deflation was largely centered in France (with a major assist from the Federal Reserve). Before the war the French franc was worth about 20 cents, but disastrous French postwar economic policies caused the franc to fall to just 2 cents in 1926 when Raymond Poincaré was called upon to lead a national-unity government to stabilize the situation. His success was remarkable, the franc rising to over 4 cents within a few months. However, despite earlier solemn pledges to restore the franc to its prewar value of 20 cents, he was persuaded to stabilize the franc at just 3.92 cents when convertibility into gold was reestablished in June 1928, undervaluing the franc against both the dollar and the pound.

Not only was the franc undervalued, but the Bank of France, which, under previous governments had been persuaded or compelled to supply francs to finance deficit spending, was prohibited by the new Monetary Law that restored convertibility at the fixed rate of 3.92 cents from increasing the quantity of francs except in exchange for gold or foreign-exchange convertible into gold. While protecting the independence of the Bank of France from government fiscal demands, the law also prevented the French money stock from increasing to accommodate increases in the French demand for money except by way of a current account surplus, or a capital inflow.

Meanwhile, the Bank of France began converting foreign-exchange reserves into gold. The resulting increase in French gold holdings led to gold appreciation. Under the gold standard, gold appreciation is manifested in price deflation affecting all gold-standard countries. That deflation was the direct and primary cause of the Great Depression, which led, over a period of five brutal years, to the failure and demise of the newly restored international gold standard.

These painful lessons were not widely or properly understood at the time, or for a long time afterward, but the clear takeaway from that experience was that trying to restore the gold standard again would be a dangerous undertaking. Another lesson that was intuited, if not fully understood, is that if a country pegs its exchange rate to gold or to another currency, it is safer to err on the side of undervaluation than overvaluation. So, when the task of recreating an international monetary system was undertaken at Bretton Woods in July 1944, the architects of the system tried to adapt it to the formal trappings of the gold standard while eliminating the deflationary biases and incentives that had doomed the interwar gold standard. To prevent increasing demand for gold from causing deflation, the obligation to convert cash into gold was limited to the United States and access to the US gold window was restricted to other central banks via the newly formed international monetary fund. Each country could, in consultation with the IMF, determine its exchange rate with the dollar.

Given the earlier experience, countries had an incentive to set exchange rates that undervalued their currencies relative to the dollar. Thus, for most of the 1950s and early 1960s, the US had to contend with a currency that was overvalued relative to the currencies of its principal trading partners, Germany and Italy (the two fastest growing economies in Europe) and Japan (later joined by South Korea and Taiwan) in Asia. In one sense, the overvaluation was beneficial to the US, because access to low-cost and increasingly high-quality imports was a form of repayment to the US of its foreign-aid assistance, and its ongoing defense protection against the threat of Communist expansionism , but the benefit came with the competitive disadvantage to US tradable-goods industries.

When West Germany took control of its economic policy from the US military in 1948, most price-and-wage controls were lifted and the new deutschmark was devalued by a third relative to the official value of the old reichsmark. A further devaluation of almost 25% followed a year later. Great Britain in 1949, perhaps influenced by the success of the German devaluation, devalued the pound by 30% from old parity of $4.03 to $2.80 in 1949. But unlike Germany, Britain, under the postwar Labour government, attempting to avoid postwar inflation, maintained wartime exchange controls and price controls. The underlying assumption at the time was that the Britain’s balance-of-payments deficit reflected an overvalued currency, so that devaluation would avoid repeating the mistake made two decades earlier when the dollar-pound parity had overvalued the pound.

That assumption, as Ralph Hawtrey had argued in lonely opposition to the devaluation, was misguided; the idea that the current account depends only, or even primarily, on the exchange rate abstracts from the monetary forces that affect the balance of payments and the current account. Worse, because British monetary policy was committed to the goal of maximizing short-term employment, the resulting excess supply of cash inevitably increased domestic spending, thereby attracting imports and diverting domestically produced products from export markets and preventing the devaluation from achieving the goal of improving the trade balance and promoting expansion of the tradable-goods sector.

Other countries, like Germany and Italy, combined currency undervaluation with monetary restraint, allowing only monetary expansion that was occasioned by current-account surpluses. This became the classic strategy, later called exchange-rate protection by Max Corden, of combining currency undervaluation with tight monetary policy. British attempts to use monetary policy to promote both (over)full employment subject to the balance-of-payments constraint imposed by an exchange rate pegged to the dollar proved unsustainable, while Germany, Italy, France (after De Gaulle came to power in 1958 and devalued the franc) found the combination of monetary restraint and currency undervaluation a successful economic strategy until the United States increased monetary expansion to counter chronic overvaluation of the dollar.

Because the dollar was the key currency of the world monetary system, and had committed itself to maintain the $35 an ounce price of gold, the US, unlike other countries whose currencies were pegged to the dollar, could not adjust the dollar exchange rate to reduce or alleviate the overvaluation of the dollar relative to the currencies of its trading partners. Mindful of its duties as supplier of the world’s reserve currency, US monetary authorities kept US inflation close to zero after the 1953 Korean War armistice.

However, that restrained monetary policy led to three recessions under the Eisenhower administration (1953-54, 1957-58, and 1960-61). The latter recessions led to disastrous Republican losses in the 1958 midterm elections and to Richard Nixon’s razor-thin loss in 1960 to John Kennedy, who had campaigned on a pledge to get the US economy moving again. The loss to Kennedy was a lesson that Nixon never forgot, and he was determined never to allow himself to lose another election merely because of scruples about US obligations as supplier of the world’s reserve currency.

Upon taking office, the Kennedy administration pressed for an easing of Fed policy to end the recession and to promote accelerated economic expansion. The result was a rapid recovery from the 1960-61 recession and the start of a nearly nine-year period of unbroken economic growth at perhaps the highest average growth rate in US history. While credit for the economic expansion is often given to the across-the-board tax cuts proposed by Kennedy in 1963 and enacted in 1964 under Lyndon Johnson, the expansion was already well under way by mid-1961, three years before the tax cuts became effective.

The international aim of monetary policy was to increase nominal domestic spending and to force US trading partners with undervalued currencies either to accept increased holdings of US liabilities or to revalue their exchange rates relative to the dollar to diminish their undervaluation relative to the dollar. Easier US monetary policy led to increasing complaints from Europeans, especially the Germans, that the US was exporting inflation and to charges that the US was taking advantage of the exorbitant privilege of its position as supplier of the world’s reserve currency.

The aggressive response of the Kennedy administration to undervaluation of most other currencies led to predictable pushback from France under de Gaulle who, like many other conservative and right-wing French politicians, was fixated on the gold standard and deeply resented Anglo-American monetary pre-eminence after World War I and American dominance after World War II. Like France under Poincaré, France under de Gaulle sought to increase its gold holdings as it accumulated dollar-denominated foreign exchange. But under Bretton Woods, French gold accumulation had little immediate economic effect other than to enhance the French and Gaullist pretensions to grandiosity.

Already in 1961 Robert Triffin predicted that the Bretton Woods system could not endure permanently because the growing world demand for liquidity could not be satisfied by the United States in a world with a relatively fixed gold stock and a stable or rising price level. The problem identified by Triffin was not unlike that raised by Gustav Cassel in the 1920s when he predicted that the world gold stock would likely not increase enough to prevent a worldwide deflation. This was a different problem from the one that actually caused the Great Depression, which was a substantial increase in gold demand associated with the restoration of the gold standard that triggered the deflationary collapse of late 1929. The long-term gold shortage feared by Cassel was a long-term problem distinct from the increase in gold demand caused by the restoration of the gold standard in the 1920s.

The problem Triffin identified was also a long-term consequence of the failure of the international gold stock to increase to provide the increased gold reserves that would be needed for the US to be able to credibly commit to maintaining the convertibility of the dollar into gold without relying on deflation to cause the needed increase in the real value of gold reserves.

Had it not been for the Vietnam War, Bretton Woods might have survived for several more years, but the rise of US inflation to over 4% in 1968-69, coupled with the 1969-70 recession in an unsuccessful attempt to reduce inflation, followed by a weak recovery in 1971, made it clear that the US would not undertake a deflationary policy to make the official $35 gold price credible. Although de Gaulle’s unexpected retirement in 1969 removed the fiercest opponent of US monetary domination, confidence that the US could maintain the official gold peg, when the London gold price was already 10% higher than the official price, caused other central banks to fear that they would be stuck with devalued dollar claims once the US raised the official gold price. Not only the French, but other central banks were already demanding redemption in gold of the dollar claims that they were holding.

An eleventh-hour policy reversal by the administration to save the official gold price was not in the cards, and everyone knew it. So all the handwringing about the abandonment of Bretton Woods on August 15, 1971 is either simple foolishness or gaslighting. The system was already broken, and it couldn’t be fixed at any price worth pondering for even half an instant. Nixon and his accomplices tried to sugarcoat their scrapping of the Bretton Woods System by pretending that they were announcing a plan that was the first step toward its reform and rejuvenation. But that pretense led to a so-called agreement with a new gold-price peg of $38 an ounce, which lasted hardly a year before it died not with a bang but a whimper.

What can we learn from this story? For me the real lesson is that the original international gold standard was, to borrow (via Hayek) a phrase from Adam Ferguson: “the [accidental] result of human action, not human design.” The gold standard, as it existed for those 40 years, was not an intuitively obvious or well understood mechanism working according to a clear blueprint; it was an improvised set of practices, partly legislated and partly customary, and partially nothing more than conventional, but not very profound, wisdom.

The original gold standard collapsed with the outbreak of World War I and the attempt to recreate it after World War I, based on imperfect understanding of how it had actually functioned, ended catastrophically with the Great Depression, a second collapse, and another, even more catastrophic, World War. The attempt to recreate a new monetary system –the Bretton Woods system — using a modified feature of the earlier gold standard as a kind of window dressing, was certainly not a real gold standard, and, perhaps, not even a pseudo-gold standard; those who profess to mourn its demise are either fooling themselves or trying to fool the rest of us.

We are now stuck with a fiat system that has evolved and been tinkered with over centuries. We have learned how to manage it, at least so far, to avoid catastrophe. With hard work and good luck, perhaps we will continue to learn how to manage it better than we have so far. But to seek to recreate a system that functioned fairly successfully for at most 40 years under conditions not even remotely likely ever again to be approximated, is hardly likely to lead to an outcome that will enhance human well-being. Even worse, if that system were recreated, the resulting outcome might be far worse than anything we have experienced in the last half century.

Gabriel Mathy and I Discuss the Gold Standard and the Great Depression

Sometimes you get into a Twitter argument when you least expect to. It was after 11pm two Saturday nights ago when I saw this tweet by Gabriel Mathy (@gabriel_mathy)

Friedman says if there had been no Fed, there would have been no Depression. That’s certainly wrong, even if your position is that the Fed did little to nothing to mitigate the Depression (which is reasonable IMO)

Chiming in, I thought to reinforce Mathy’s criticism of Friedman, I tweeted the following:

Friedman totally misunderstood the dynamics of the Great Depression, which was driven by increasing demand for gold after 1928, in particular by the Bank of France and by the Fed. He had no way of knowing what the US demand for gold would have been if there had not been a Fed

I got a response from Mathy that I really wasn’t expecting who tweeted with seeming annoyance

There already isn’t enough gold to back the gold standard by the end of World War I, it’s just a matter of time until a negative shock large enough sent the world into a downward spiral (my emphasis). Just took a few years after resumption of the gold standard in most countries in the mid-20s. (my emphasis)

I didn’t know exactly what to make of Mathy’s assertion that there wasn’t enough gold by the end of World War I. The gold standard was effectively abandoned at the outset of WWI and the US price level was nearly double the prewar US price level after the postwar inflation of 1919. Even after the deflation of 1920-21, US prices were still much higher in 1922 than they were in 1914. Gold production fell during World War I, but gold coins had been withdrawn from circulation and replaced with paper or token coins. The idea that there is a fixed relationship between the amount of gold and the amount of money, especially after gold coinage had been eliminated, has no theoretical basis.

So I tweeted back:

The US holdings of gold after WWI were so great that Keynes in his Tract on Monetary Reform [argued] that the great danger of a postwar gold standard was inflation because the US would certainly convert its useless holding of gold for something more useful

To which Mathy responded

The USA is not the only country though. The UK had to implement tight monetary policies to back the gold standard, and eventually had to leave the gold standard. As did the USA in 1931. The Great Depression is a global crisis.

Mathy’s response, I’m afraid, is completely wrong. Of course, the Great Depression is a global crisis. It was a global crisis, because, under the (newly restored) gold standard, the price level in gold-standard countries was determined internationally. And, holding 40% of the world’s monetary reserves of gold at the end of World War I, the US, the largest and most dynamic economy in the world, was clearly able to control, as Keynes understood, the common international price level for gold-standard countries.

The tight monetary policy imposed on the UK resulted from its decision to rejoin the gold standard at the prewar dollar parity. Had the US followed a modestly inflationary monetary policy, allowing an outflow of gold during the 1920s rather than inducing an inflow, deflation would not have been imposed on the UK.

But instead of that response, I replied as follows:

The US didn’t leave till 1933 when FDR devalued. I agree that individual countries, worried about losing gold, protected their reserves by raising interest rates. Had they all reduced rates together, the conflict between individual incentives and common interest could have been avoided.

Mathy then kept the focus on the chronology of the Great Depression, clarifying that he meant that in 1931 the US, like the UK, tightened monetary policy to remain on the gold standard, not that the US, like the UK, also left the gold standard in 1931:

The USA tightens in 1931 to stay on the gold standard. And this sets off a wave of bank failures.

Fair enough, but once the situation deteriorated after the crash and the onset of deflation, the dynamics of the financial crisis made managing the gold standard increasingly difficult, given the increasingly pessimistic expectations conditioned by deepening economic contraction and deflation. While an easier US monetary policy in the late 1920s might have avoided the catastrophe and preserved the gold standard, an easier monetary policy may, at some point, have become inconsistent with staying on the gold standard.

So my response to Mathy was more categorical than was warranted.

Again, the US did not have to tighten in 1931 to stay on the gold standard. I agree that the authorities might have sincerely thought that they needed to tighten to stay on the gold standard, but they were wrong if that’s what they thought.

Mathy was having none of it, unleashing a serious snark attack

You know better I guess, despite collapsing free gold amidst a massive speculative attack

What I ought to have said is that the gold standard was not worth saving if doing so entailed continuing deflation. If I understand him, Mathy believes that deflation after World War I was inevitable and unavoidable, because there wasn’t enough gold to sustain the gold standard after World War I. I was arguing that if there was a shortage of gold, it was because of the policies followed, often in compliance with legal gold-cover requirements, that central banks, especially the Bank of France, which started accumulating gold rapidly in 1928, and the Fed, which raised interest rates to burst a supposed stock-market bubble, were following. But as I point out below, the gold accumulation by the Bank of France far exceeded what was mandated by legal gold-cover requirements.

My point is that the gold shortage that Mathy believes doomed the gold standard was not preordained; it could have been mitigated by policies to reduce, or reverse, gold accumulation. France could have rejoined the gold standard without accumulating enormous quantities of gold in 1928-29, and the Fed did not have to raise interest rates in 1928-29, attracting additional gold to its own already massive holdings just as France was rapidly accumulating gold.

When France formally rejoined the gold standard in July 1928, the gold reserves of the Bank of France were approximately equal to its foreign-exchange holdings and its gold-reserve ratio was 39.5% slightly above the newly established legal required ratio of 35%. In subsequent years, the gold reserves of the Bank of France steadily increased while foreign exchange reserves declined. At the close of 1929, the gold-reserve ratio of the Bank of France stood at 47.3%, while its holdings of foreign exchange hardly changed. French gold holdings increased in 1930 by slightly more than in 1929, with foreign-exchange holdings almost constant; the French gold-reserve ratio at the end of 1930 was 53.2%. The 1931 increase in French gold reserves, owing to a 20% drop in foreign-exchange holdings, was even larger than in 1930, raising the gold-reserve ratio to 60.5% at the end of 1931.

Once deflation and the Great Depression started late in 1929, deteriorating rapidly in 1930, salvaging the gold standard became increasingly unlikely, with speculators becoming increasingly alert to the possibility of currency devaluation or convertibility suspension. Speculation against a pegged exchange rate is not always a good bet, but it’s rarely a bad one, any change in the pegged rate being almost surely in the direction that speculators are betting on. 

But, it was still at least possible that, if gold-cover requirements for outstanding banknotes and bank reserves were relaxed or suspended, central banks could have caused a gold outflow sufficient to counter the deflationary expectations then feeding speculative demands for gold. Gold does not have many non-monetary uses, so a significant release of gold from idle central-bank reserves might have caused gold to depreciate relative to other real assets, thereby slowing, or even reversing, deflation.

Of course, deflation would not have stopped unless the deflationary expectations fueling speculative demands for gold were reversed. Different expectational responses would have led to different outcomes. More often than not, inflationary and deflationary expectations are self-fulfilling. Because expectations tend to be mutually interdependent – my inflationary expectations reinforce your inflationary expectations and vice versa — the notion of rational expectation in this context borders on the nonsensical, making outcomes inherently unpredictable. Reversing inflationary or deflationary expectations requires policy credibility and a willingness by policy makers to take policy actions – even or especially painful ones — that demonstrate their resolve.

In 1930 Ralph Hawtrey testified to the Macmillan Committee on Finance and Industry, he recommended that the Bank of England reduce interest rates to counter the unemployment and deflation. That testimony elicited the following exchange between Hugh Pattison Macmillan, the chairman of the Committee and Hawtrey:

MACMILLAN: Suppose . . . without restricting credit . . . that gold had gone out to a very considerable extent, would that not have had very serious consequences on the international position of London?

HAWTREY: I do not think the credit of London depends on any particular figure of gold holding. . . . The harm began to be done in March and April of 1925 [when] the fall in American prices started. There was no reason why the Bank of England should have taken any action at that time so far as the question of loss of gold is concerned. . . . I believed at the time and I still think that the right treatment would have been to restore the gold standard de facto before it was restored de jure. That is what all the other countries have done. . . . I would have suggested that we should have adopted the practice of always selling gold to a sufficient extent to prevent the exchange depreciating. There would have been no legal obligation to continue convertibility into gold . . . If that course had been adopted, the Bank of England would never have been anxious about the gold holding, they would have been able to see it ebb away to quite a considerable extent with perfect equanimity, . . and might have continued with a 4 percent Bank Rate.

MACMILLAN: . . . the course you suggest would not have been consistent with what one may call orthodox Central Banking, would it?

HAWTREY: I do not know what orthodox Central Banking is.

MACMILLAN: . . . when gold ebbs away you must restrict credit as a general principle?

HAWTREY: . . . that kind of orthodoxy is like conventions at bridge; you have to break them when the circumstances call for it. I think that a gold reserve exists to be used. . . . Perhaps once in a century the time comes when you can use your gold reserve for the governing purpose, provided you have the courage to use practically all of it.

Hawtrey’s argument lay behind this response of mine to Mathy:

What else is a gold reserve is for? That’s like saying you can’t fight a fire because you’ll drain the water tank. But I agree that by 1931 there was no point in defending the gold standard and the US should have made clear the goal was reflation to the 1926 price level as FDR did in 1933.

Mathy responded:

If the Fed cuts discount rates to 0%, capital outflow will eventually exhaust gold reserves. So do you recommend a massive OMO in 1929? What specifically is the plan?

In 1927, the Fed reduced its discount rate to 3.5%; in February 1928, it was raised the rate to 4%. The rate was raised again in August 1928 and to 6% in September 1929. The only reason the Fed raised interest rates in 1928 was a misguided concern with rising stock prices. A zero interest rate was hardly necessary in 1929, nor were massive open-market operations. Had the Fed kept its interest rate at 4%, and the Bank of France not accumulated gold rapidly in 1928-29, the history of the world might well have followed a course much different from the one actually followed.

In another exchange, Mathy pointed to the 1920s adoption of the gold-exchange standard rather than a (supposedly) orthodox version of the gold standard as evidence that there wasn’t enough gold to support the gold standard after World War I. (See my post on the difference between the gold standard and the gold-exchange standard.)

Mathy: You seem to be implying there was plentiful free gold [i.e., gold held by central banks in excess of the amount required by legal gold-cover requirements] in the world after WW1 so that gold was not a constraint. How much free gold to you reckon there was?

Glasner: All of it was free. Legal reserve requirements soaked up much but nearly all the free gold

Mathy: All of it was not free, and countries suffered speculative attacks before their real or perceived minimum backings of gold were reached

Glasner: All of it would have been free but for the legal reserve requirements. Of course countries were subject to speculative attacks, when the only way for a country to avoid deflation was to leave the gold standard.

Mathy: You keep asserting an abundance of free gold, so let’s see some numbers. The lack of free gold led to the gold exchange standard where countries would back currencies with other currencies (themselves only partially backed by gold) because there wasn’t enough gold.

Glasner: The gold exchange standard was a rational response to the WWI inflation and post WWI deflation and it could have worked well if it had not been undermined by the Bank of France and gold accumulation by the US after 1928.

Mathy: Both you and [Douglas] Irwin assume that the gold inflows into France are the result of French policy. But moving your gold to France, a country committed to the gold standard, is exactly what a speculative attack on another currency at risk of leaving the gold standard looks like.

Mathy: What specific policies did the Bank if France implement in 1928 that caused gold inflows? We can just reason from accounting identities, assuming that international flows to France are about pull factors from France rather than push factors from abroad.

Mathy: So lay out your counterfactual- how much gold should the US and France have let go abroad, and how does this prevent the Depression?

Glasner: The increase in gold monetary holdings corresponds to a higher real value of gold. Under the gold standard that translates into [de]flation. Alternatively, to prevent gold outflows central banks raised rates which slowed economic activity and led to deflation.

Mathy: So give me some numbers. What does the Fed do specifically in 1928 and what does France do specifically in 1928 that avoid the debacle of 1929. You can take your time, pick this up Monday.

Mathy: The UK was suffering from high unemployment before 1928 because there wasn’t enough gold in the system. The Bank of England had been able to draw gold “from the moon” with a higher bank rate. After WW1, this was no longer possible.

Glasner: Unemployment in the UK steadily fell after 1922 and continued falling till ’29. With a fixed exchange rate against the $, and productivity in the US rising faster than in the UK, the UK needed more US inflation than it got to reach full employment. That has nothing to do with what happened after 1929.

Mathy: UK unemployment rises 1925-1926 actually, that’s incorrect and it’s near double digits throughout the 1920s. That’s not good at all and the problems start long before 1928.

There’s a lot to unpack here, and I will try to at least touch on the main points. Mathy questions whether there was enough free gold available in the 1920s, while also acknowledging that the gold-exchange standard was instituted in the 1920s precisely to avoid the demands on monetary gold reserves that would result from restoring gold coinage and imposing legal gold-cover requirements on central-bank liabilities. So, if free-gold reserves were insufficient before the Great Depression, it was because of the countries that restored the gold standard and also imposed legal gold-cover requirements, notably the French Monetary Law enacted in June 1928 that imposed a minimum 35% gold-cover requirement when convertibility of the franc was restored.

It’s true that there were speculative movements of gold into France when there were fears that countries might devalue their currencies or suspend gold convertibility, but those speculative movements did not begin until late 1930 or 1931.

Two aspects of the French restoration of gold convertibility should be mentioned. First, France pegged the dollar/franc exchange rate at $0.0392, with the intention of inducing a current-account surplus and a gold inflow. Normally that inflow would have been transitory as French prices and wages rose to the world level. But the French Monetary Law allowed the creation of new central-bank liabilities only in exchange for gold or foreign exchange convertible into gold. So French demand for additional cash balances could be satisfied only insofar as total spending in France was restricted sufficiently to ensure an inflow of gold or convertible foreign exchange. Hawtrey explained this brilliantly in Chapter two of The Art of Central Banking.

Mathy suggests that the gold-standard was adopted by countries without enough gold to operate a true gold standard, which he thinks proves that there wasn’t enough free gold available. What resort to the gold-exchange standard shows is that countries without enough gold were able to join the gold standard without first incurring the substantial cost of accumulating (either by direct gold purchases or by inducing large amounts of gold inflows by raising domestic interest rates); it does not prove that the gold-exchange standard system was inherently unstable.

Why did some countries restoring the gold standard not have enough gold? First, much of the world’s stock of gold reserves had been shipped to the US during World War I when countries were importing food, supplies and war material from the US paid with gold, or, promising to repay after the war, on credit. Second, wartime and immediate postwar inflation required increased quantities of cash to conduct transactions and satisfy liquidity demands. Third, legislated gold-cover requirements in the US, and later in France and other countries rejoining the gold standard, obligated monetary authorities to accumulate gold.

Those gold-cover requirements, forcing countries to accumulate additional gold to satisfy any increased demand by the public for cash, were an ongoing, and unnecessary, cause of rising demand for gold reserves as countries rejoined the gold standard in the 1920s, imparting an inherent deflationary bias to the gold standard. The 1922 Genoa Accords attempted to cushion this deflationary bias by allowing countries to rejoin the gold standard without making their own currencies directly convertible into gold, but by committing themselves to a fixed exchange rate against those currencies – at first the dollar and subsequently pound sterling – that were directly convertible into gold. But the accords were purely advisory and provided no effective mechanism to prevent the feared increase in the monetary demand for gold. And the French never intended to rejoin the gold standard except by making the franc convertible directly into gold.

Mathy asks how much gold I think that the French and the US should have let go to avoid the Great Depression. This is an impossible question to answer, because French gold accumulation in 1928-29, combined with increased US interest rates in 1928-29, which caused a nearly equivalent gold inflow into the US, triggered deflation in the second half of 1929 that amplified deflationary expectations, causing a stock market crash, a financial crisis and ultimately the Great Depression. Once deflation got underway, the measures needed to calm the crisis and reverse the downturn became much more extreme than those that would have prevented the downturn in the first place.

Had the Fed kept its discount rate at 3.5 to 4 percent, had France not undervalued the franc in setting its gold peg, and had France created a mechanism for domestic credit expansion instead of making an increase in the quantity of francs impossible except through a current account surplus, and had the Bank of France been willing to accumulate foreign exchange instead of requiring its foreign-exchange holdings to be redeemed for gold, the crisis would not have occurred.

Here are some quick and dirty estimates of the effect of French policy on the availability of free gold. In July 1928 when France rejoined the gold standard and enacted the Monetary Law drafted by the Bank of France, the notes and demand deposits against which the Bank was required to gold reserves totaled almost ff76 billion (=$2.98 billion). French gold holdings in July 1928 were then just under ff30 billion (=$1.17 billion), implying a reserve ratio of 39.5%. (See the discussion above.)

By the end of 1931, the total of French banknotes and deposits against which the Bank of France was required to hold gold reserves was almost ff114 billion (=$4.46 billion). French gold holdings at the end of 1931 totaled ff68.9 billion (=$2.7 billion), implying a gold-reserve ratio of 60.5%. If the French had merely maintained the 40% gold-reserve ratio of 1928, their gold holdings in 1931 would have been approximately ff45 billion (=$1.7 billion).

Thus, from July 1929 to December 1931, France absorbed $1 billion of gold reserves that would have otherwise been available to other central banks or made available for use in non-monetary applications. The idea that free gold was a constraint on central bank policy is primarily associated with the period immediately before and after the British suspension of the gold standard in September 1931, which occasioned speculative movements of gold from the US to France to avoid a US suspension of the gold standard or a devaluation. From January 1931 through August 1931, the gold holdings of the Bank of France increased by just over ff3 billion (=$78 million). From August to December of 1931 French gold holdings increased by ff10.3 billion (=$404 million).

So, insofar as a lack of free gold was a constraint on US monetary expansion via open market purchases in 1931, which is the only time period when there is a colorable argument that free gold was a constraint on the Fed, it seems highly unlikely that that constraint would have been binding had the Bank of France not accumulated an additional $1 billion of gold reserves (over and above the increased reserves necessary to maintain the 40% gold-reserve ratio of July 1928) after rejoining the gold standard. Of course, the claim that free gold was a binding constraint on Fed policy in the second half of 1931 is far from universally accepted, and I consider the claim to be pretextual.

Finally, I concede that my assertion that unemployment fell steadily in Britain after the end of the 1920-22 depression was not entirely correct. Unemployment did indeed fall substantially after 1922, but remained around 10 percent in 1924 — there are conflicting estimates based on different assumptions about how to determine whom to count as unemployed — when the pound began appreciating before the restoration of the prewar parity. Unemployment continued rising rise until 1926, but remained below the 1922 level. Unemployment then fell substantially in 1926-27, but rose again in 1928 (as gold accumulation by France and the US led to a rise in Bank rate), without reaching the 1926 level. Unemployment fell slightly in 1929 and was less than the 1924 level before the crash. See Eichengreen “Unemployment in Interwar Britain.”

I agree that unemployment had been a serious problem in Britain before 1928. But that wasn’t because sufficient gold was lacking in the system. Unemployment was a British problem caused by an overvalued exchange rate; it was not a systemic gold-standard problem.

Before World War I, when the gold standard was largely a sterling standard (just as the postwar gold standard became a dollar standard), the Bank of England had been able to “draw gold from the moon” by raising Bank rate. But the gold that had once been in the moon moved to the US during World War I. What Britain required was a US discount rate low enough to raise the world price level, thereby reducing deflationary pressure on Britain caused by overvaluation of sterling. Instead of keeping the discount rate at 3.5 – 4%, and allowing an outflow of gold, the Fed increased its discount rate, inducing a gold inflow and triggering a worldwide deflationary catastrophe. Between 1929 to 1931, British unemployment nearly doubled because of that catastrophe, not because Britain didn’t have enough gold. The US had plenty of gold and suffered equally from the catastrophe.

Monetarism v. Hawtrey and Cassel

The following is an updated and revised version of the penultimate section of my paper with Ron Batchelder “Pre-Keynesian Theories of the Great Depressison: What Ever Happened to Hawtrey and Cassel?” which I am now preparing for publication. The previous version is available on SSRN.

In the 1950s and early 1960s, empirical studies of the effects of money and monetary policy by Milton Friedman, his students and followers, rehabilitated the idea that monetary policy had significant macroeconomic effects. Most importantly, in research with Anna Schwartz Friedman advanced the seemingly remarkable claim that the chief cause of the Great Depression had been a series of policy mistakes by the Federal Reserve. Although Hawtrey and Cassel, had also implicated the Federal Reserve in their explanation of the Great Depression, they were unmentioned by Friedman and Schwartz or by other Monetarists.[1]

The chief difference between the Monetarist and the Hawtrey-Cassel explanations of the Great Depression is that Monetarists posited a monetary shock (bank failures) specific to the U.S. as the primary, if not sole, cause of the Depression, while Hawtrey and Cassel considered the Depression a global phenomenon reflecting a rapidly increasing international demand for gold, bank failures being merely an incidental and aggravating symptom, specific to the U.S., of a more general monetary disorder.

Arguing that the Great Depression originated in the United States following a typical business-cycle downturn, Friedman and Schwartz (1963) attributed the depth of the downturn not to the unexplained initial shock, but to the contraction of the U.S. money stock caused by the bank failures. Dismissing any causal role for the gold standard in the Depression, Friedman and Schwartz (359-60) acknowledged only its role in propagating, via PSFM, an exogenous, policy-driven, contraction of the U.S. money stock to other gold-standard countries. According to Friedman and Schwartz, the monetary contraction was the cause, and deflation the effect.

But the causation posited by Friedman and Schwartz is the exact opposite of the true causation. Under the gold standard, deflation (i.e., gold appreciation) was the cause and the decline in the quantity of money the effect. Deflation in an international gold standard is not a local phenomenon originating in any single country; it occurs simultaneously in all gold standard countries.

To be sure the banking collapse in the U.S. exacerbated the catastrophe. But the collapse was the localized effect of a more general cause: deflation. Without deflation, neither the unexplained 1929 downturn nor the subsequent banking collapse would have occurred. Nor was an investment boom in the most advanced and most productive economy in the world unsustainable as posited, with no evidence of unsustainability other than the subsequent economic collapse, by the Austrian malinvestment hypothesis.

Friedman and Schwartz based their assertion that the monetary disturbance that caused the Great Depression occurred in the U.S. on the observation that, from 1929 to 1931, gold flowed into, not out of, the U.S. Had the disturbance occurred elsewhere, they argued, gold would have flowed out of, not into, the U. S.

Table 1 shows the half-year changes in U.S., French, and world gold reserves starting in June 1928, when the French monetary law re-establishing the gold standard was enacted.

TABLE 1: Gold Reserves in US, France, and the World June 1928-December 1931 (measured in dollars)
Date World ReservesUS ReservesUS Share (percent)French ReservesFrench Share (percent)
June 19289,7493,73238.31,13611.7
Dec. 192810,0573,74637.21,25412.4
2nd half 1928 change31214-1.11180.7
June 192910,1263,95639.11,43614.2
1st half 1929 change692101.91821.8
Dec. 192910,3363,90037.71,63315.8
2nd half 1929 change210-56-1.41971.6
June 193010,6714,17839.21,72716.2
1st half 1930 change3352781.5940.4
Dec. 193010,9444,22538.72,10019.2
2nd half 1930 change 27347-0.53733.0
June 193111,264459340.82,21219.6
1st half 1931 change3203682.11120.4
Dec. 193111,3234,05135.82,69923.8
2nd half 1931 change59-542-5.04874.2
June 1928-Dec. 1931 change1,574319-2.51,56312.1
Source: H. C. Johnson, Gold, France and the Great Depression

In the three-and-a-half years from June 1928 (when gold convertibility of the franc was restored) to December 1931, gold inflows into France exceeded gold inflows into the United States. The total gold inflow into France during the June 1928 to December 1931 period was $1.563 billion compared to only $319 billion into the United States.

However, much of the difference in the totals stems from the gold outflow from the U.S. into France in the second half of 1931, reflecting fears of a possible U.S. devaluation or suspension of convertibility after Great Britain and other countries suspended the gold standard in September 1931 (Hamilton 2012). From June 1928 through June 1931, the total gold inflow into the U.S. was $861 billion and the total gold inflow into France was $1.076 billion, the U.S. share of total reserves increasing from 38.3 percent to 40.6 percent, while the total French share increased from 11.7 percent to 19.6 percent.[2]

In the first half of 1931, when the first two waves of U.S. bank failures occurred, the increase in U.S. gold reserves exceeded the increase in world gold reserves. The shift by the public from holding bank deposits to holding currency increased reserve requirements, an increase reflected in the gold reserves held by the U.S. The increased U.S. demand for gold likely exacerbated the deflationary pressures affecting all gold-standard countries, perhaps contributing to the failure of the Credit-Anstalt in May 1931 that intensified the European crisis that forced Britain off the gold standard in September.

The combined increase in U.S. and French gold reserves was $1.937 billion compared to an increase of only $1.515 billion in total world reserves, indicating that the U.S. and France were drawing reserves either from other central banks or from privately held gold stocks. Clearly, both the U.S. and France were exerting powerful deflationary pressure on the world economy, before and during the downward spiral of the Great Depression.[3]

Deflationary forces were operating directly on prices before the quantity of money adjusted to the decline in prices. In some countries the adjustment of the quantity of money was relatively smooth; in the U.S. it was exceptionally difficult, but, not even in the U.S., was it the source of the disturbance. Hawtrey and Cassel understood that; Friedman did not.

In explaining the sources of his interest in monetary theory and the role of monetary policy, Friedman (1970) pointedly distinguished between the monetary tradition from which his work emerged and the dominant tradition in London circa 1930, citing Robbins’s (1934) Austrian-deflationist book on the Great Depression, while ignoring Hawtrey and Cassel. Friedman linked his work to the Chicago oral tradition, citing a lecture by Jacob Viner (1933) as foreshadowing his own explanation of the Great Depression, attributing the loss of interest in monetary theory and policy by the wider profession to the deflationism of LSE monetary economists. Friedman went on to suggest that the anti-deflationism of the Chicago monetary tradition immunized it against the broader reaction against monetary theory and policy, that the Austro-London pro-deflation bias provoked against monetary theory and policy.

Though perhaps superficially plausible, Friedman’s argument ignores, as he did throughout a half-century of scholarship and research, the contributions of Hawtrey and Cassel and especially their explanation of the Great Depression. Unfortunately, Friedman’s outsized influence on economists trained after the Keynesian Revolution distracted their attention from contributions outside the crude Keynesian-Monetarist dichotomy that shaped his approach to monetary economics.

Eclectics like Hawtrey and Cassel were neither natural sources of authority, nor obvious ideological foils for Friedman to focus upon. Already forgotten, providing neither convenient targets, nor ideological support, Hawtrey and Cassel, could be easily and conveniently ignored.


[1] Meltzer (2001) did mention Hawtrey, but the reference was perfunctory and did not address the substance of his and Cassel’s explanation of the Great Depression.

[2] By far the largest six-month increase in U.S. gold reserves was in the June-December 1931 period coinciding with the two waves of bank failures at the end of 1930 and in March 1931 causing a substantial shift from deposits to currency which required an increase in gold reserves owing to the higher ratio of required gold reserves against currency than against bank deposits.

[3] Fremling (1985) noted that, even during the 1929-31 period, the U.S. share of world gold reserves actually declined. However, her calculation includes the extraordinary outflow of gold from the U.S. in the second half of 1931. The U.S. share of global gold reserves rose from June 1928 to June 1931.

The Hawley-Smoot Tariff and the Great Depression

The role of the Hawley-Smoot Tariff (aka Smoot-Hawley Tariff) in causing the Great Depression has been an ongoing subject of controversy for close to a century. Ron Batchelder and I wrote a paper (“Debt, Deflation and the Great Depression”) published in this volume (Money and Banking: The American Experience) that offered an explanation of the mechanism by which the tariff contributed to the Great Depression. That paper was written before and inspired another paper “Pre-Keynesian Theories of the Great Depression: What Ever Happened to Hawtrey and Cassell“) I am now revising the paper for republication, and here is the new version of the relevant section discussing the Hawley-Smoot Tariff.

Monetary disorder was not the only legacy of World War I. The war also left a huge burden of financial obligations in its wake. The European allies had borrowed vast sums from the United States to finance their war efforts, and the Treaty of Versailles imposed on Germany the obligation to pay heavy reparations to the allies, particularly to France.

We need not discuss the controversial question whether the burden imposed on Germany was too great to have been discharged. The relevant question for our purposes is by what means the reparations and war debts could be paid, or, at least, carried forward without causing a default on the obligations. To simplify the discussion, we concentrate on the relationship between the U.S. and Germany, because many of the other obligations of the allies to the U.S. were offset by those of Germany to the allies.[1]

The debt to the U.S. could be extinguished either by a net payment in goods reflected in a German balance-of-trade surplus and a U.S. balance-of-trade deficit, or by a transfer of gold from Germany to the U.S. Stretching out the debt would have required the U.S., in effect to lend Germany the funds required to service its obligations.

For most of the 1920s, the U.S. did in fact lend heavily to Germany, thereby lending Germany the funds to meet its financial obligations to the U.S. (and its European creditors). U.S. lending was not explicitly for that purpose, but on the consolidated national balance sheets, U.S. lending offset German financial obligations, obviating any real transfer.

Thus, to avoid a transfer, in goods or specie, from Germany to the U.S., continued U.S. lending to Germany was necessary. But the sharp tightening of monetary policy by the Federal Reserve in 1928 raised domestic interest rates to near record levels and curtailed lending abroad, as foreign borrowers were discouraged from seeking funds in U.S. capital markets. Avoiding an immediate transfer from Germany to the U.S. was no longer possible except by default. To effect the necessary transfer in goods, Germany would have been required to shift resources from its non-tradable-goods sector to its tradable-goods sector, which would require reducing spending on, and the relative prices of, non-tradable goods. Thus, Germany began to slide into a recession in 1928.

In 1929 the United States began making the transfer even more difficult when the newly installed Hoover administration reaffirmed the Republican campaign commitment to raising U.S. tariffs, thereby imposing a tax on the goods transfer through which Germany could discharge its obligations. Although the bill to increase tariffs that became the infamous Hawley-Smoot Act was not passed until 1930, the commitment to raise tariffs made it increasingly unlikely that the U.S. would allow the debts owed it to be discharged by a transfer of goods. The only other means by which Germany could discharge its obligations was a transfer of gold. Anticipating that its obligations to the U.S. could be discharged only by transferring gold, Germany took steps to increase its gold holdings to be able to meet its debt obligations. The increased German demand for gold was reflected in a defensive tightening of monetary policy to raise domestic interest rates to reduce spending and to induce an inflow of gold to Germany.

The connection between Germany’s debt obligations and its demand for gold sheds light on the deflationary macroeconomic consequences of the Hawley-Smoot tariff. Given the huge debts owed to the United States, the tariff imposed a deflationary monetary policy on all U.S. debtors as they attempted to accumulate sufficient gold to be able to service their debt obligations to the U.S. But, under the gold standard, the United States could not shield itself from the deflationary effects that its trade policy was imposing on its debtors.[2]

The U.S. could have counteracted these macroeconomic pressures by a sufficiently expansive monetary policy, thereby satisfying the demand of other countries for gold. Monetary expansion would have continued, by different means, the former policy of lending to debtors, enabling them to extend their obligations. But preoccupied with, or distracted by the stock-market boom, U.S. monetary authorities were oblivious to the impossible alternatives that were being forced on U.S. debtors by a combination of tight U.S. monetary policy and a protectionist trade policy.

As the prospects that protectionist legislation would pass steadily improved even as tight U.S. monetary policy was being maintained, deflationary signs became increasingly clear and alarming. The panic of October 1929, in our view, was not, as much Great Depression historiography describes it, the breaking of a speculative bubble, but a correct realization that a toxic confluence of monetary and trade policies was leading the world over a deflationary precipice.

Once the deflation took hold, the nature of the gold standard with a fixed price of gold was such that gold would likely appreciate against weak currencies that were likely to be formally devalued, or allowed to float, relative to gold. A vicious cycle of increasing speculative demand for gold in anticipation of currency devaluation further intensified the deflationary pressures (Hamilton, 1988). Moreover, successive devaluations by one country at a time increased the deflationary pressure in the remaining gold-standard countries. A uniform all-around devaluation might have had some chance of quickly controlling the deflationary process, but piecemeal deflation could only prolong the deflationary pressure on nations that remained on the gold standard.

FOOTNOTES

[1] The United States, as a matter of law, always resisted such a comparison, contending that the war debts were commercial obligations in no way comparable to the politically imposed reparations. However, as a final matter, there was obviously a strict correspondence between the two sets of obligations. The total size of German obligations was never precisely determined. However, those obligations were certainly several times the size of the war debts owed the United States. Focusing simply on the U.S.-German relationship is therefore simply a heuristic device.

[2] Viewed from a different perspective, the tariff aimed at transferring wealth from the foreign debtors to the U.S. government by taxing debt payments on debt already fixed in nominal terms. Moreover, deflation from whatever source increased the real value of the fixed nominal debts owed the U.S.

The Real-Bills Doctrine, the Lender of Last Resort, and the Scope of Banking

Here is another section from my work in progress on the Smithian and Humean traditions in monetary economics. The discussion starts with a comparison of the negative view David Hume took toward banks and the positive view taken by Adam Smith which was also discussed in the previous post on the price-specie-flow mechanism. This section discusses how Smith, despite viewing banks positively, also understood that banks can be a source of disturbances as well as of efficiencies, and how he addressed that problem and how his followers who shared a positive view toward banks addressed the problem. Comments and feedback are welcome and greatly appreciated.

Hume and Smith had very different views about fractional-reserve banking and its capacity to provide the public with the desired quantity of money (banknotes and deposits) and promote international adjustment. The cash created by banks consists of liabilities on themselves that they exchange for liabilities on the public. Liabilities on the public accepted by banks become their assets, generating revenue streams with which banks cover their outlays including obligations to creditors and stockholders.

The previous post focused on the liability side of bank balance sheets, and whether there are economic forces that limit the size of those balance sheets, implying a point of equilibrium bank expansion. Believing that banks have an unlimited incentive to issue liabilities, whose face value exceeds their cost of production, Hume considered banks dangerous and inflationary. Smith disagreed, arguing that although bank money is a less costly alternative to the full-bodied money preferred by Hume, banks don’t create liabilities limitlessly, because, unless those liabilities generate corresponding revenue streams, they will be unable to redeem those liabilities, which their creditors may require of them, at will. To enhance the attractiveness of those liabilities and to increase the demand to hold them, competitive banks promise to convert those liabilities, at a stipulated rate, into an asset whose value they do not control. Under those conditions, banks have neither the incentive nor the capacity to cause inflation.

I turn now to a different topic: whether Smith’s rejection of the idea that banks are systematically biased toward overissuing liabilities implies that banks require no external control or intervention. I begin by briefly referring to Smith’s support of the real-bills doctrine and then extend that discussion to two other issues: the lender of last resort and the scope of banking.

A         Real-Bills Doctrine

I have argued elsewhere that, besides sketching the outlines of Fullarton’s argument for the Law of Reflux, Adam Smith recommended that banks observe a form of the real-bills doctrine, namely that banks issue sight liabilities only in exchange for real commercial bills of short (usually 90-days) duration. Increases in the demand for money cause bank balance sheets to expand; decreases cause them to contract. Unlike Mints (1945), who identified the Law of Reflux with the real-bills doctrine, I suggested that Smith viewed the real-bills doctrine as a pragmatic policy to facilitate contractions in the size of bank balance sheets as required by the reflux of their liabilities. With the discrepancy between the duration of liabilities and assets limited by issuing sight liabilities only in exchange for short-term bills, bank balance sheets would contract automatically thereby obviating, at least in part, the liquidation of longer-term assets at depressed prices.

On this reading, Smith recognized that banking policy ought to take account of the composition of bank balance sheets, in particular, the sort of assets that banks accept as backing for the sight liabilities that they issue. I would also emphasize that on this interpretation, Smith did not believe, as did many later advocates of the doctrine, that lending on the security of real bills is sufficient to prevent the price level from changing. Even if banks have no systematic incentive to overissue their liabilities, unless those liabilities are made convertible into an asset whose value is determined independently of the banks, the value of their liabilities is undetermined. Convertibility is how banks anchor the value of their liabilities, thereby increasing the attractiveness of those liabilities to the public and the willingness of the public to accept and hold them.

But Smith’s support for the real-bills doctrine indicates that, while understanding the equilibrating tendencies of competition on bank operations, he also recognized the inherent instability of banking caused by fluctuations in the value and liquidity of their assets. Smith’s support for the real-bills doctrine addressed one type of instability: the maturity mismatch between banks’ assets and liabilities. But there are other sources of instability, which may require further institutional or policy measures beyond the general laws of property and contract whose application and enforcement, in Smith’s view, generally sufficed for the self-interested conduct of private firms to lead to socially benign outcomes.

In the remainder of this section, I consider two other methods of addressing the vulnerability of bank assets to sudden losses of value: (1) the creation or empowerment of a lender of last resort capable of lending to illiquid, but solvent, banks possessing good security (valuable assets) as collateral against which to borrow, and (2) limits beyond the real-bills doctrine over the permissible activities undertaken by commercial banks.

B         Lender of Last Resort

Although the real-bills doctrine limits the exposure of bank balance sheets to adverse shocks on the value of long-term liabilities, even banks whose liabilities were issued in exchange for short-term real bills of exchange may be unable to meet all demands for redemption in periods of extreme financial distress, when debtors cannot sell their products at the prices they expected and cannot meet their own obligations to their creditors. If banks are called upon to redeem their liabilities, banks may be faced with a choice between depleting their own cash reserves, when they are most needed, or liquidating other assets at substantial, if not catastrophic, losses.

Smith’s version of the real-bills doctrine addressed one aspect of balance-sheet risk, but the underlying problem is deeper and more complicated than the liquidity issue that concerned Smith. The assets accepted by banks in exchange for their liabilities are typically not easily marketable, so if those assets must be shed quickly to satisfy demands for payment, banks’ solvency may be jeopardized by consequent capital losses. Limiting portfolios to short-term assets limits exposure to such losses, but only when the disturbances requiring asset liquidation affect only a relatively small number of banks. As the number of affected banks increases, their ability to counter the disturbance is impaired, as the interbank market for credit starts to freeze up or break down entirely, leaving them unable to offer short-term relief to, or receive it from, other momentarily illiquid banks. It is then that emergency lending by a lender of last resort to illiquid, but possibly still solvent, banks is necessary.

What causes a cluster of expectational errors by banks in exchanging their liabilities for assets supplied by their customers that become less valuable than they were upon acceptance? Are financial crises that result in, or are caused by, asset write downs by banks caused by random clusters of errors by banks, or are there systematic causes of such errors? Does the danger lie in the magnitude of the errors or in the transmission mechanism?

Here, too, the Humean and Smithian traditions seem to be at odds, offering different answers to problems, or, if not answers, at least different approaches to problems. Focusing on the liability side of bank balance sheets, the Humean tradition emphasizes the expansion of bank lending and the consequent creation of banknotes or deposits as the main impulse to macroeconomic fluctuations, a boom-bust or credit cycle triggered by banks’ lending to finance either business investment or consumer spending. Despite their theoretical differences, both Austrian business-cycle theory and Friedmanite Monetarism share a common intellectual ancestry, traceable by way of the Currency School to Hume, identifying the source of business-cycle fluctuations in excessive growth in the quantity of money.

The eclectic Smithian tradition accommodates both monetary and non-monetary business-cycle theories, but balance-sheet effects on banks are more naturally accommodated within the Smithian tradition than the Humean tradition with its focus on the liabilities not the assets of banks. At any rate, more research is necessary before we can decide whether serious financial disturbances result from big expectational errors or from contagion effects.

The Great Depression resulted from a big error. After the steep deflation and depression of 1920-22, followed by a gradual restoration of the gold standard, fears of further deflation were dispelled and steady economic expansion, especially in the United States, resulted. Suddenly in 1929, as France and other countries rejoined the gold standard, the fears voiced by Hawtrey and Cassel that restoring the gold standard could have serious deflationary consequences appeared increasingly more likely to be realized. Real signs of deflation began to appear in the summer of 1929, and in the fall the stock market collapsed. Rather than use monetary policy to counter incipient deflation, policy makers and many economists argued that deflation was part of the solution not the problem. And the Depression came.

It is generally agreed that the 2008 financial crisis that triggered the Little Depression (aka Great Recession) was largely the result of a housing bubble fueled by unsound mortgage lending by banks and questionable underwriting practices in packaging and marketing of mortgage-backed securities. However, although the housing bubble seems to have burst the spring of 2007, the crisis did not start until September 2008.

It is at least possible, as I have argued (Glasner 2018) that, despite the financial fragility caused by the housing bubble and unsound lending practices that fueled the bubble, the crisis could have been avoided but for a reflexive policy tightening by the Federal Reserve starting in 2007 that caused a recession starting in December 2007 and gradually worsening through the summer of 2008. Rather than ease monetary policy as the recession deepened, the Fed, distracted by rising headline inflation owing to rising oil prices that summer, would not reduce its interest-rate target further after March 2008. If my interpretation is correct, the financial crisis of 2008 and the subsequent Little Depression (aka Great Recession) were as much caused by bad monetary policy as by the unsound lending practices and mistaken expectations by lenders.

It is when all agents are cash constrained that a lender of last resort that is able to provide the liquidity that the usual suppliers of liquidity cannot provide, but are instead demanding, is necessary to avoid a systemic breakdown. In 2008, the Fed was unwilling to satisfy demands for liquidity until the crisis had deteriorated to the point of a worldwide collapse. In the nineteenth century, Thornton and Fullarton understood that the Bank of England was uniquely able to provide liquidity in such circumstances, recommending that it lend freely in periods of financial stress.

That policy was not viewed favorably either by Humean supporters of the Currency Principle, opposed to all forms of fractional-reserve banking, or by Smithian supporters of free banking who deplored the privileged central-banking position granted to the Bank of England. Although the Fed in 2008 acknowledged that it was both a national and international lender of last resort, it was tragically slow to take the necessary actions to end the crisis after allowing it to spiral nearly out of control.

While cogent arguments have been made that a free-banking alternative to the lender-of-last-resort services of the Bank of England might have been possible in the nineteenth century,[2] even a free-banking system would require a mechanism for handling periods of financial stress. Free-banking supporters argue that bank clearinghouses have emerged spontaneously in the absence of central banks, and could provide the lender-of-last resort services provided by central banks. But, insofar as bank clearinghouses would take on the lender-of-last-resort function, which involves some intervention and supervision of bank activities by either the clearinghouse or the central bank, the same anticompetitive or cartelistic objections to the provision of lender-of-last-resort services by central banks also would apply to the provision of those services by clearinghouses. So, the tension between libertarian, free-market principles and lender-of-last-resort services would not necessarily be eliminated bank clearinghouses instead of central banks provided those services.

This is an appropriate place to consider Walter Bagehot’s contribution to the lender-of-last-resort doctrine. Building on the work of Thornton and Fullarton, Bagehot formulated the classic principle that, during times of financial distress, the Bank of England should lend freely at a penalty rate to banks on good security. Bagehot, himself, admitted to a certain unease in offering this advice, opining that it was regrettable that the Bank of England achieved a pre-eminent position in the British banking system, so that a decentralized banking system, along the lines of the Scottish free-banking system, could have evolved. But given the historical development of British banking, including the 1844 Bank Charter Act, Bagehot, an eminently practical man, had no desire to recommend radical reform, only to help the existing system operate as smoothly as it could be made to operate.

But the soundness of Bagehot’s advice to lend freely at a penalty rate is dubious. In a financial crisis, the market rate of interest primarily reflects a liquidity premium not an expected real return on capital, the latter typically being depressed in a crisis. Charging a penalty rate to distressed borrowers in a crisis only raises the liquidity premium. Monetary policy ought to aim to reduce, not to increase, that premium. So Bagehot’s advice, derived from a misplaced sense of what is practical and prudent and financially sound, rather than from sound analysis, was far from sound.

Under the gold standard, or under any fixed-exchange-rate regime, a single country has an incentive to raise interest rates above the rates of other countries to prevent a gold outflow or attract an inflow. Under these circumstances, a failure of international cooperation can lead to competitive rate increases as monetary authorities scramble to maintain or increase their gold reserves. In testimony to the Macmillan Commission in 1930, Ralph Hawtrey masterfully described the obligation of a central bank in a crisis. Here is his exchange with the Chairman of the Commission Hugh Macmillan:

MACMILLAN: Suppose . . . without restricting credit . . . that gold had gone out to a very considerable extent, would that not have had very serious consequences on the international position of London?

HAWTREY: I do not think the credit of London depends on any particular figure of gold holding. . . . The harm began to be done in March and April of 1925 [when] the fall in American prices started. There was no reason why the Bank of England should have taken ny action at that time so far as the question of loss of gold is concerned. . . . I believed at the time and I still think that the right treatment would have been to restore the gold standard de facto before it was restored de jure. That is what all the other countries have done. . . . I would have suggested that we should have adopted the practice of always selling gold to a sufficient extent to prevent the exchange depreciating. There would have been no legal obligation to continue convertibility into gold . . . If that course had been adopted, the Bank of England would never have been anxious about the gold holding, they would have been able to see it ebb away to quite a considerable extent with perfect equanimity, . . and might have continued with a 4 percent Bank Rate.

MACMILLAN: . . . the course you suggest would not have been consistent with what one may call orthodox Central Banking, would it?

HAWTREY: I do not know what orthodox Central Banking is.

MACMILLAN: . . . when gold ebbs away you must restrict credit as a general principle?

HAWTREY: . . . that kind of orthodoxy is like conventions at bridge; you have to break them when the circumstances call for it. I think that a gold reserve exists to be used. . . . Perhaps once in a century the time comes when you can use your gold reserve for the governing purpose, provided you have the courage to use practically all of it.

Hawtrey here was echoing Fullarton’s insight that there is no rigid relationship between the gold reserves held by the Bank of England and the total quantity of sight liabilities created by the British banking system. Rather, he argued, the Bank should hold an ample reserve sufficient to satisfy the demand for gold in a crisis when a sudden and temporary demand for gold had to be accommodated. That was Hawtrey’s advice, but not Bagehot’s, whose concern was about banks’ moral hazard and imprudent lending in the expectation of being rescued in a crisis by the Bank of England. Indeed, moral hazard is a problem, but in a crisis it is a secondary problem, when, as Hawtrey explained, alleviating the crisis, not discouraging moral hazard, must be the primary concern of the lender of last resort.

            C         Scope of Banking

Inclined to find remedies for financial distress in structural reforms limiting the types of assets banks accept in exchange for their sight liabilities, Smith did not recommend a lender of last resort.[3] Another method of reducing risk, perhaps more in tune with the Smithian real-bills doctrine than a lender of last resort, is to restrict the activities of banks that issue banknotes and deposits.

In Anglophone countries, commercial banking generally evolved as separate and distinct from investment banking. It was only during the Great Depression and the resulting wave of bank failures that the combination of commercial and investment banking was legally prohibited by the Glass-Steagall Act, eventually repealed in 1999. On the Continent, where commercial banking penetrated less deeply into the fabric of economic and commercial life than in Anglophone countries, commercial banking developed more or less along with investment banking in what are called universal banks.

Whether the earlier, and more widespread, adoption of commercial banking in Anglophone countries than on the Continent advanced the idea that no banking institution should provide both commercial- and investment-banking services is not a question about which I offer a conjecture, but it seems a topic worthy of study. The Glass-Steagall Act, which enforced that separation after being breached early in the twentieth century, a breach thought by some to have contributed to US bank failures in the Great Depression, was based on a presumption against combining and investment-banking in a single institution. But even apart from the concerns that led to enactment of Glass-Steagall, limiting the exposure of commercial banks, which supply most of the cash held by the public, to the balance-sheet risk associated with investment-banking activities seems reasonable. Moreover, the adoption of government deposit insurance after the Great Depression as well as banks’ access to the discount window of the central bank may augment the moral hazard induced by deposit insurance and a lender of last resort, offsetting potential economies of scope associated with combining commercial and investment banking.

Although legal barriers to the combination of commercial and investment banking have long been eliminated, proposals for “narrow banking” that would restrict the activities undertaken by commercial banks continue to be made. Two different interpretations of narrow banking – one Smithian and one Humean – are possible.

The Humean concern about banking was that banks are inherently disposed to overissue their liabilities. The Humean response to the concern has been to propose 100-percent reserve banking, a comprehensive extension of the 100-percent marginal reserve requirement on the issue of banknotes imposed by the Bank Charter Act. Such measures could succeed, as some supporters (Simons 1936) came to realize, only if accompanied by a radical change the financial practices and arrangements on which all debt contracts are based. It is difficult to imagine that the necessary restructuring of economic activity would ever be implemented or tolerated.

The Humean concern was dismissed by the Smithian tradition, recognizing that banks, even if unconstrained by reserve requirements, have no incentive to issue liabilities without limit. The Smithian concern was whether banks could cope with balance-sheet risks after unexpected losses in the value of their assets. Although narrow banking proposals are a legitimate and possibly worthwhile response to that concern, the acceptance by central banks of responsibility to act as a lender of last resort and widespread government deposit insurance to dampen contagion effects have taken the question of narrowing or restricting the functions of money-creating banks off the table. Whether a different strategy for addressing the systemic risks associated with banks’ creation of money by relying solely on deposit insurance and a lender of last resort is a question that still deserves thoughtful attention.

On the Price Specie Flow Mechanism

I have been working on a paper tentatively titled “The Smithian and Humean Traditions in Monetary Theory.” One section of the paper is on the price-specie-flow mechanism, about which I wrote last month in my previous post. This section develops the arguments of the previous post at greater length and draws on a number of earlier posts that I’ve written about PSFM as well (e.g., here and here )provides more detailed criticisms of both PSFM and sterilization and provides some further historical evidence to support some of the theoretical arguments. I will be grateful for any comments and feedback.

The tortured intellectual history of the price-specie-flow mechanism (PSFM) received its still classic exposition in a Hume (1752) essay, which has remained a staple of the theory of international adjustment under the gold standard, or any international system of fixed exchange rates. Regrettably, the two-and-a-half-century life span of PSFM provides no ground for optimism about the prospects for progress in what some are pleased to call without irony economic science.

PSFM describes how, under a gold standard, national price levels tend to be equalized, with deviations between the national price levels in any two countries inducing gold to be shipped from the country with higher prices to the one with lower prices until prices are equalized. Premised on a version of the quantity theory of money in which (1) the price level in each country on the gold standard is determined by the quantity of money in that country, and (2) money consists entirely in gold coin or bullion, Hume elegantly articulated a model of disturbance and equilibration after an exogenous change in the gold stock in one country.

Viewing banks as inflationary engines of financial disorder, Hume disregarded banks and the convertible monetary liabilities of banks in his account of PSFM, leaving to others the task of describing the international adjustment process under a gold standard with fractional-reserve banking. The task of devising an institutional framework, within which PSFM could operate, for a system of fractional-reserve banking proved to be problematic and ultimately unsuccessful.

For three-quarters of a century, PSFM served a purely theoretical function. During the Bullionist debates of the first two decades of the nineteenth century, triggered by the suspension of the convertibility of the pound sterling into gold in 1797, PSFM served as a theoretical benchmark not a guide for policy, it being generally assumed that, when convertibility was resumed, international monetary equilibrium would be restored automatically.

However, the 1821 resumption was followed by severe and recurring monetary disorders, leading some economists, who formed what became known as the Currency School, to view PSFM as a normative criterion for ensuring smooth adjustment to international gold flows. That criterion, the Currency Principle, stated that the total currency in circulation in Britain should increase or decrease by exactly as much as the amount of gold flowing into or out of Britain.[1]

The Currency Principle was codified by the Bank Charter Act of 1844. To mimic the Humean mechanism, it restricted, but did not suppress, the right of note-issuing banks in England and Wales, which were allowed to continue issuing notes, at current, but no higher, levels, without holding equivalent gold reserves. Scottish and Irish note-issuing banks were allowed to continue issuing notes, but could increase their note issue only if matched by increased holdings of gold or government debt. In England and Wales, the note issue could increase only if gold was exchanged for Bank of England notes, so that a 100-percent marginal gold reserve requirement was imposed on additional banknotes.

Opposition to the Bank Charter Act was led by the Banking School, notably John Fullarton and Thomas Tooke. Rejecting the Humean quantity-theoretic underpinnings of the Currency School and the Bank Charter Act, the Banking School rejected the quantitative limits of the Bank Charter Act as both unnecessary and counterproductive, because banks, obligated to redeem their liabilities directly or indirectly in gold, issue liabilities only insofar as they expect those liabilities to be willingly held by the public, or, if not, are capable of redeeming any liabilities no longer willingly held. Rather than the Humean view that banks issue banknotes or create deposits without constraint, the Banking School held Smith’s view that banks issue money in a form more convenient to hold and to transact with than metallic money, so that bank money allows an equivalent amount of gold to be shifted from monetary to real (non-monetary) uses, providing a net social savings. For a small open economy, the diversion (and likely export) of gold bullion from monetary to non-monetary uses has negligible effect on prices (which are internationally, not locally, determined).

The quarter century following enactment of the Bank Charter Act showed that the Act had not eliminated monetary disturbances, the government having been compelled to suspend the Act in 1847, 1857 and 1866 to prevent incipient crises from causing financial collapse. Indeed, it was precisely the fear that liquidity might not be forthcoming that precipitated increased demands for liquidity that the Act made it impossible to accommodate. Suspending the Act was sufficient to end the crises with limited intervention by the Bank. [check articles on the crises of 1847, 1857 and 1866.]

It may seem surprising, but the disappointing results of the Bank Charter Act provided little vindication to the Banking School. It led only to a partial, uneasy, and not entirely coherent, accommodation between PSFM doctrine and the reality of a monetary system in which the money stock consists mostly of banknotes and bank deposits issued by fractional-reserve banks. But despite the failure of the Bank Charter Act, PSFM achieved almost canonical status, continuing, albeit with some notable exceptions, to serve as the textbook model of the gold standard.

The requirement that gold flows induce equal changes in the quantity of money within a country into (or from) which gold is flowing was replaced by an admonition that gold flows lead to “appropriate” changes in the central-bank discount rate or an alternative monetary instrument to cause the quantity of money to change in the same direction as the gold flow. While such vague maxims, sometimes described as “the rules of the game,” gave only directional guidance about how to respond to change in gold reserves, their hortatory character, and avoidance of quantitative guidance, allowed monetary authorities latitude to avoid the self-inflicted crises that had resulted from the quantitative limits of the Bank Charter Act.

Nevertheless, the myth of vague “rules” relating the quantity of money in a country to changes in gold reserves, whose observance ensured the smooth functioning of the international gold standard before its collapse at the start of World War I, enshrined PSFM as the theoretical paradigm for international monetary adjustment under the gold standard.

That paradigm was misconceived in four ways that can be briefly summarized.

  • Contrary to PSFM, changes in the quantity of money in a gold-standard country cannot change local prices proportionately, because prices of tradable goods in that country are constrained by arbitrage to equal the prices of those goods in other countries.
  • Contrary to PSFM, changes in local gold reserves are not necessarily caused either by non-monetary disturbances such as shifts in the terms of trade between countries or by local monetary disturbances (e.g. overissue by local banks) that must be reversed or counteracted by central-bank policy.
  • Contrary to PSFM, changes in the national price levels of gold-standard countries were uncorrelated with gold flows, and changes in national price levels were positively, not negatively, correlated.
  • Local banks and monetary authorities exhibit their own demands for gold reserves, demands exhibited by choice (i.e., independent of legally required gold holdings) or by law (i.e., by legally requirement to hold gold reserves equal to some fraction of banknotes issued by banks or monetary authorities). Such changes in gold reserves may be caused by changes in the local demands for gold by local banks and the monetary authorities in one or more countries.

Many of the misconceptions underlying PSFM were identified by Fullarton’s refutation of the Currency School. In articulating the classical Law of Reflux, he established the logical independence of the quantity convertible money in a country from by the quantity of gold reserves held by the monetary authority. The gold reserves held by individual banks, or their deposits with the Bank of England, are not the raw material from which banks create money, either banknotes or deposits. Rather, it is their creation of banknotes or deposits when extending credit to customers that generates a derived demand to hold liquid assets (i.e., gold) to allow them to accommodate the demands of customers and other banks to redeem banknotes and deposits. Causality runs from creating banknotes and deposits to holding reserves, not vice versa.

The misconceptions inherent in PSFM and the resulting misunderstanding of gold flows under the gold standard led to a further misconception known as sterilization: the idea that central banks, violating the obligations imposed by “the rules of the game,” do not allow, or deliberately prevent, local money stocks from changing as their gold holdings change. The misconception is the presumption that gold inflows ought necessarily cause increases in local money stocks. The mechanisms causing local money stocks to change are entirely different from those causing gold flows. And insofar as those mechanisms are related, causality flows from the local money stock to gold reserves, not vice versa.

Gold flows also result when monetary authorities transform their own asset holdings into gold. Notable examples of such transformations occurred in the 1870s when a number of countries abandoned their de jure bimetallic (and de facto silver) standards to the gold standard. Monetary authorities in those countries transformed silver holdings into gold, driving the value of gold up and silver down. Similarly, but with more catastrophic consequences, the Bank of France, in 1928 after France restored the gold standard, began redeeming holdings of foreign-exchange reserves (financial claims on the United States or Britain, payable in gold) into gold. Following the French example, other countries rejoining the gold standard redeemed foreign exchange for gold, causing gold appreciation and deflation that led to the Great Depression.

Rereading the memoirs of this splendid translation . . . has impressed me with important subtleties that I missed when I read the memoirs in a language not my own and in which I am far from completely fluent. Had I fully appreciated those subtleties when Anna Schwartz and I were writing our A Monetary History of the United States, we would likely have assessed responsibility for the international character of the Great Depression somewhat differently. We attributed responsibility for the initiation of a worldwide contraction to the United States and I would not alter that judgment now. However, we also remarked, “The international effects were severe and the transmission rapid, not only because the gold-exchange standard had rendered the international financial system more vulnerable to disturbances, but also because the United States did not follow gold-standard rules.” Were I writing that sentence today, I would say “because the United States and France did not follow gold-standard rules.”

I pause to note for the record Friedman’s assertion that the United States and France did not follow “gold-standard rules.” Warming up to the idea, he then accused them of sterilization.

Benjamin Strong and Emile Moreau were admirable characters of personal force and integrity. But . . .the common policies they followed were misguided and contributed to the severity and rapidity of transmission of the U.S. shock to the international community. We stressed that the U.S. “did not permit the inflow of gold to expand the U.S. money stock. We not only sterilized it, we went much further. Our money stock moved perversely, going down as the gold stock went up” from 1929 to 1931.

Strong and Moreau tried to reconcile two ultimately incompatible objectives: fixed exchange rates and internal price stability. Thanks to the level at which Britain returned to gold in 1925, the U.S. dollar was undervalued, and thanks to the level at which France returned to gold at the end of 1926, so was the French franc. Both countries as a result experienced substantial gold inflows. Gold-standard rules called for letting the stock of money rise in response to the gold inflows and for price inflation in the U.S. and France, and deflation in Britain, to end the over-and under-valuations. But both Strong and Moreau were determined to prevent inflation and accordingly both sterilized the gold inflows, preventing them from providing the required increase in the quantity of money.

Friedman’s discussion of sterilization is at odds with basic theory. Working with a naïve version of PSFM, he imagines that gold flows passively respond to trade balances independent of monetary forces, and that the monetary authority under a gold standard is supposed to ensure that the domestic money stock varies roughly in proportion to its gold reserves. Ignoring the international deflationary dynamic, he asserts that the US money stock perversely declined from 1929 to 1931, while its gold stock increased. With a faltering banking system, the public shifted from holding demand deposits to currency. Gold reserves were legally required against currency, but not against demand deposits, so the shift from deposits to currency entailed an increase gold reserves. To be sure the increased US demand for gold added to upward pressure on value of gold, and to worldwide deflationary pressure. But US gold holdings rose by only $150 million from December 1929 to December 1931 compared with an increase of $1.06 billion in French gold holdings over the same period. Gold accumulation by the US and its direct contribution to world deflation during the first two years of the Depression was small relative to that of France.

Friedman also erred in stating “the common policies they followed were misguided and contributed to the severity and rapidity of transmission of the U.S. shock to the international community.” The shock to the international community clearly originated not in the US but in France. The Fed could have absorbed and mitigated the shock by allowing a substantial outflow of its huge gold reserves, but instead amplified the shock by raising interest rates to nearly unprecedented levels, causing gold to flow into the US.

After correctly noting the incompatibility between fixed exchange rates and internal price stability, Friedman contradicts himself by asserting that, in seeking to stabilize their internal price levels, Strong and Moreau violated the gold-standard “rules,” as if it were rules, not arbitrage, that constrain national price to converge toward a common level under a gold standard.

Friedman’s assertion that, after 1925, the dollar was undervalued and sterling overvalued was not wrong. But he misunderstood the consequences of currency undervaluation and overvaluation under the gold standard, a confusion stemming from the underlying misconception, derived from PSFM, that foreign exchange rates adjust to balance trade flows, so that, in equilibrium, no country runs a trade deficit or trade surplus.

Thus, in Friedman’s view, dollar undervaluation and sterling overvaluation implied a US trade surplus and British trade deficit, causing gold to flow from Britain to the US. Under gold-standard “rules,” the US money stock and US prices were supposed to rise and the British money stock and British prices were supposed to fall until undervaluation and overvaluation were eliminated. Friedman therefore blamed sterilization of gold inflows by the Fed for preventing the necessary increase in the US money stock and price level to restore equilibrium. But, in fact, from 1925 through 1928, prices in the US were roughly stable and prices in Britain fell slightly. Violating gold-standard “rules” did not prevent the US and British price levels from converging, a convergence driven by market forces, not “rules.”

The stance of monetary policy in a gold-standard country had minimal effect on either the quantity of money or the price level in that country, which were mainly determined by the internationally determined value of gold. What the stance of national monetary policy determines under the gold standard is whether the quantity of money in the country adjusts to the quantity demanded by a process of domestic monetary creation or withdrawal or by the inflow or outflow of gold. Sufficiently tight domestic monetary policy restricting the quantify of domestic money causes a compensatory gold inflow increasing the domestic money stock, while sufficiently easy money causes a compensatory outflow of gold reducing the domestic money stock. Tightness or ease of domestic monetary policy under the gold standard mainly affected gold and foreign-exchange reserves, and, only minimally, the quantity of domestic money and the domestic price level.

However, the combined effects of many countries simultaneously tightening monetary policy in a deliberate, or even inadvertent, attempt to accumulate — or at least prevent the loss — of gold reserves could indeed drive up the international value of gold through a deflationary process affecting prices in all gold-standard countries. Friedman, even while admitting that, in his Monetary History, he had understated the effect of the Bank of France on the Great Depression, referred only the overvaluation of sterling and undervaluation of the dollar and franc as causes of the Great Depression, remaining oblivious to the deflationary effects of gold accumulation and appreciation.

It was thus nonsensical for Friedman to argue that the mistake of the Bank of France during the Great Depression was not to increase the quantity of francs in proportion to the increase of its gold reserves. The problem was not that the quantity of francs was too low; it was that the Bank of France prevented the French public from collectively increasing the quantity of francs that they held except by importing gold.

Unlike Friedman, F. A. Hayek actually defended the policy of the Bank of France, and denied that the Bank of France had violated “the rules of the game” after nearly quadrupling its gold reserves between 1928 and 1932. Under his interpretation of those “rules,” because the Bank of France increased the quantity of banknotes after the 1928 restoration of convertibility by about as much as its gold reserves increased, it had fully complied with the “rules.” Hayek’s defense was incoherent; under its legal obligation to convert gold into francs at the official conversion rate, the Bank of France had no choice but to increase the quantity of francs by as much as its gold reserves increased.

That eminent economists like Hayek and Friedman could defend, or criticize, the conduct of the Bank of France during the Great Depression, because the Bank either did, or did not, follow “the rules of the game” under which the gold standard operated, shows the uselessness and irrelevance of the “rules of the game” as a guide to policy. For that reason alone, the failure of empirical studies to find evidence that “the rules of the game” were followed during the heyday of the gold standard is unsurprising. But the deeper reason for that lack of evidence is that PSFM, whose implementation “the rules of the game” were supposed to guarantee, was based on a misunderstanding of the international-adjustment mechanism under either the gold standard or any fixed-exchange-rates system.

Despite the grip of PSFM over most of the profession, a few economists did show a deeper understanding of the adjustment mechanism. The idea that the price level in terms of gold directly constrained the movements of national price levels across countries was indeed recognized by writers as diverse as Keynes, Mises, and Hawtrey who all pointed out that the prices of internationally traded commodities were constrained by arbitrage and that the free movement of capital across countries would limit discrepancies in interest rates across countries attached to the gold standard, observations that had already been made by Smith, Thornton, Ricardo, Fullarton and Mill in the classical period. But, until the Monetary Approach to the Balance of Payments became popular in the 1970s, only Hawtrey consistently and systematically deduced the implications of those insights in analyzing both the Great Depression and the Bretton Woods system of fixed, but adjustable, exchange rates following World War II.

The inconsistencies and internal contradictions of PSFM were sometimes recognized, but usually overlooked, by business-cycle theorists when focusing on the disturbing influence of central banks, perpetuating mistakes of the Humean Currency School doctrine that attributed cyclical disturbances to the misbehavior of local banking systems that were inherently disposed to overissue their liabilities.

White and Hogan on Hayek and Cassel on the Causes of the Great Depression

Lawrence White and Thomas Hogan have just published a new paper in the Journal of Economic Behavior and Organization (“Hayek, Cassel, and the origins of the great depression”). Since White is a leading Hayek scholar, who has written extensively on Hayek’s economic writings (e.g., his important 2008 article “Did Hayek and Robbins Deepen the Great Depression?”) and edited the new edition of Hayek’s notoriously difficult volume, The Pure Theory of Capital, when it was published as volume 11 of the Collected Works of F. A. Hayek, the conclusion reached by the new paper that Hayek had a better understanding than Cassel of what caused the Great Depression is not, in and of itself, surprising.

However, I admit to being taken aback by the abstract of the paper:

We revisit the origins of the Great Depression by contrasting the accounts of two contemporary economists, Friedrich A. Hayek and Gustav Cassel. Their distinct theories highlight important, but often unacknowledged, differences between the international depression and the Great Depression in the United States. Hayek’s business cycle theory offered a monetary overexpansion account for the 1920s investment boom, the collapse of which initiated the Great Depression in the United States. Cassel’s warnings about a scarcity gold reserves related to the international character of the downturn, but the mechanisms he emphasized contributed little to the deflation or depression in the United States.

I wouldn’t deny that there are differences between the way the Great Depression played out in the United States and in the rest of the world, e.g., Britain and France, which to be sure, suffered less severely than did the US or, say, Germany. It is both possible, and important, to explore and understand the differential effects of the Great Depression in various countries. I am sorry to say that White and Hogan do neither. Instead, taking at face value the dubious authority of Friedman and Schwartz’s treatment of the Great Depression in the Monetary History of the United States, they assert that the cause of the Great Depression in the US was fundamentally different from the cause of the Great Depression in many or all other countries.

Taking that insupportable premise from Friedman and Schwartz, they simply invoke various numerical facts from the Monetary History as if those facts, in and of themselves, demonstrate what requires to be demonstrated: that the causes of the Great Depression in the US were different from those of the Great Depression in the rest of the world. That assumption vitiated the entire treatment of the Great Depression in the Monetary History, and it vitiates the results that White and Hogan reach about the merits of the conflicting explanations of the Great Depression offered by Cassel and Hayek.

I’ve discussed the failings of Friedman’s treatment of the Great Depression and of other episodes he analyzed in the Monetary History in previous posts (e.g., here, here, here, here, and here). The common failing of all the episodes treated by Friedman in the Monetary History and elsewhere is that he misunderstood how the gold standard operated, because his model of the gold standard was a primitive version of the price-specie-flow mechanism in which the monetary authority determines the quantity of money, which then determines the price level, which then determines the balance of payments, the balance of payments being a function of the relative price levels of the different countries on the gold standard. Countries with relatively high price levels experience trade deficits and outflows of gold, and countries with relatively low price levels experience trade surpluses and inflows of gold. Under the mythical “rules of the game” under the gold standard, countries with gold inflows were supposed to expand their money supplies, so that prices would rise and countries with outflows were supposed to reduce their money supplies, so that prices fall. If countries followed the rules, then an international monetary equilibrium would eventually be reached.

That is the model of the gold standard that Friedman used throughout his career. He was not alone; Hayek and Mises and many others also used that model, following Hume’s treatment in his essay on the balance of trade. But it’s the wrong model. The correct model is the one originating with Adam Smith, based on the law of one price, which says that prices of all commodities in terms of gold are equalized by arbitrage in all countries on the gold standard.

As a first approximation, under the Smithean model, there is only one price level adjusted for different currency parities for all countries on the gold standard. So if there is deflation in one country on the gold standard, there is deflation for all countries on the gold standard. If the rest of the world was suffering from deflation under the gold standard, the US was also suffering from a deflation of approximately the same magnitude as every other country on the gold standard was suffering.

The entire premise of the Friedman account of the Great Depression, adopted unquestioningly by White and Hogan, is that there was a different causal mechanism for the Great Depression in the United States from the mechanism operating in the rest of the world. That premise is flatly wrong. The causation assumed by Friedman in the Monetary History was the exact opposite of the actual causation. It wasn’t, as Friedman assumed, that the decline in the quantity of money in the US was causing deflation; it was the common deflation in all gold-standard countries that was causing the quantity of money in the US to decline.

To be sure there was a banking collapse in the US that was exacerbating the catastrophe, but that was an effect of the underlying cause: deflation, not an independent cause. Absent the deflationary collapse, there is no reason to assume that the investment boom in the most advanced and most productive economy in the world after World War I was unsustainable as the Hayekian overinvestment/malinvestment hypothesis posits with no evidence of unsustainability other than the subsequent economic collapse.

So what did cause deflation under the gold standard? It was the rapid increase in the monetary demand for gold resulting from the insane policy of the Bank of France (disgracefully endorsed by Hayek as late as 1932) which Cassel, along with Ralph Hawtrey (whose writings, closely parallel to Cassel’s on the danger of postwar deflation, avoid all of the ancillary mistakes White and Hogan attribute to Cassel), was warning would lead to catastrophe.

It is true that Cassel also believed that over the long run not enough gold was being produced to avoid deflation. White and Hogan spend inordinate space and attention on that issue, because that secular tendency toward deflation is entirely different from the catastrophic effects of the increase in gold demand in the late 1920s triggered by the insane policy of the Bank of France.

The US could have mitigated the effects if it had been willing to accommodate the Bank of France’s demand to increase its gold holdings. Of course, mitigating the effects of the insane policy of the Bank of France would have rewarded the French for their catastrophic policy, but, under the circumstances, some other means of addressing French misconduct would have spared the world incalculable suffering. But misled by an inordinate fear of stock market speculation, the Fed tightened policy in 1928-29 and began accumulating gold rather than accommodate the French demand.

And the Depression came.

An Austrian Tragedy

It was hardly predictable that the New York Review of Books would take notice of Marginal Revolutionaries by Janek Wasserman, marking the susquicentenial of the publication of Carl Menger’s Grundsätze (Principles of Economics) which, along with Jevons’s Principles of Political Economy and Walras’s Elements of Pure Economics ushered in the marginal revolution upon which all of modern economics, for better or for worse, is based. The differences among the three founding fathers of modern economic theory were not insubstantial, and the Jevonian version was largely superseded by the work of his younger contemporary Alfred Marshall, so that modern neoclassical economics is built on the work of only one of the original founders, Leon Walras, Jevons’s work having left little impression on the future course of economics.

Menger’s work, however, though largely, but not totally, eclipsed by that of Marshall and Walras, did leave a more enduring imprint and a more complicated legacy than Jevons’s — not only for economics, but for political theory and philosophy, more generally. Judging from Edward Chancellor’s largely favorable review of Wasserman’s volume, one might even hope that a start might be made in reassessing that legacy, a process that could provide an opportunity for mutually beneficial interaction between long-estranged schools of thought — one dominant and one marginal — that are struggling to overcome various conceptual, analytical and philosophical problems for which no obvious solutions seem available.

In view of the failure of modern economists to anticipate the Great Recession of 2008, the worst financial shock since the 1930s, it was perhaps inevitable that the Austrian School, a once favored branch of economics that had made a specialty of booms and busts, would enjoy a revival of public interest.

The theme of Austrians as outsiders runs through Janek Wasserman’s The Marginal Revolutionaries: How Austrian Economists Fought the War of Ideas, a general history of the Austrian School from its beginnings to the present day. The title refers both to the later marginalization of the Austrian economists and to the original insight of its founding father, Carl Menger, who introduced the notion of marginal utility—namely, that economic value does not derive from the cost of inputs such as raw material or labor, as David Ricardo and later Karl Marx suggested, but from the utility an individual derives from consuming an additional amount of any good or service. Water, for instance, may be indispensable to humans, but when it is abundant, the marginal value of an extra glass of the stuff is close to zero. Diamonds are less useful than water, but a great deal rarer, and hence command a high market price. If diamonds were as common as dewdrops, however, they would be worthless.

Menger was not the first economist to ponder . . . the “paradox of value” (why useless things are worth more than essentials)—the Italian Ferdinando Galiani had gotten there more than a century earlier. His central idea of marginal utility was simultaneously developed in England by W. S. Jevons and on the Continent by Léon Walras. Menger’s originality lay in applying his theory to the entire production process, showing how the value of capital goods like factory equipment derived from the marginal value of the goods they produced. As a result, Austrian economics developed a keen interest in the allocation of capital. Furthermore, Menger and his disciples emphasized that value was inherently subjective, since it depends on what consumers are willing to pay for something; this imbued the Austrian school from the outset with a fiercely individualistic and anti-statist aspect.

Menger’s unique contribution is indeed worthy of special emphasis. He was more explicit than Jevons or Walras, and certainly more than Marshall, in explaining that the value of factors of production is derived entirely from the value of the incremental output that could be attributed (or imputed) to their services. This insight implies that cost is not an independent determinant of value, as Marshall, despite accepting the principle of marginal utility, continued to insist – famously referring to demand and supply as the two blades of the analytical scissors that determine value. The cost of production therefore turns out to be nothing but the value the output foregone when factors are used to produce one output instead of the next most highly valued alternative. Cost therefore does not determine, but is determined by, equilibrium price, which means that, in practice, costs are always subjective and conjectural. (I have made this point in an earlier post in a different context.) I will have more to say below about the importance of Menger’s specific contribution and its lasting imprint on the Austrian school.

Menger’s Principles of Economics, published in 1871, established the study of economics in Vienna—before then, no economic journals were published in Austria, and courses in economics were taught in law schools. . . .

The Austrian School was also bound together through family and social ties: [his two leading disciples, [Eugen von] Böhm-Bawerk and Friedrich von Wieser [were brothers-in-law]. [Wieser was] a close friend of the statistician Franz von Juraschek, Friedrich Hayek’s maternal grandfather. Young Austrian economists bonded on Alpine excursions and met in Böhm-Bawerk’s famous seminars (also attended by the Bolshevik Nikolai Bukharin and the German Marxist Rudolf Hilferding). Ludwig von Mises continued this tradition, holding private seminars in Vienna in the 1920s and later in New York. As Wasserman notes, the Austrian School was “a social network first and last.”

After World War I, the Habsburg Empire was dismantled by the victorious Allies. The Austrian bureaucracy shrank, and university placements became scarce. Menger, the last surviving member of the first generation of Austrian economists, died in 1921. The economic school he founded, with its emphasis on individualism and free markets, might have disappeared under the socialism of “Red Vienna.” Instead, a new generation of brilliant young economists emerged: Schumpeter, Hayek, and Mises—all of whom published best-selling works in English and remain familiar names today—along with a number of less well known but influential economists, including Oskar Morgenstern, Fritz Machlup, Alexander Gerschenkron, and Gottfried Haberler.

Two factual corrections are in order. Menger outlived Böhm-Bawerk, but not his other chief disciple von Wieser, who died in 1926, not long after supervising Hayek’s doctoral dissertation, later published in 1927, and, in 1933, translated into English and published as Monetary Theory and the Trade Cycle. Moreover, a 16-year gap separated Mises and Schumpeter, who were exact contemporaries, from Hayek (born in 1899) who was a few years older than Gerschenkron, Haberler, Machlup and Morgenstern.

All the surviving members or associates of the Austrian school wound up either in the US or Britain after World War II, and Hayek, who had taken a position in London in 1931, moved to the US in 1950, taking a position in the Committee on Social Thought at the University of Chicago after having been refused a position in the economics department. Through the intervention of wealthy sponsors, Mises obtained an academic appointment of sorts at the NYU economics department, where he succeeded in training two noteworthy disciples who wrote dissertations under his tutelage, Murray Rothbard and Israel Kirzner. (Kirzner wrote his dissertation under Mises at NYU, but Rothbard did his graduate work at Colulmbia.) Schumpeter, Haberler and Gerschenkron eventually took positions at Harvard, while Machlup (with some stops along the way) and Morgenstern made their way to Princeton. However, Hayek’s interests shifted from pure economic theory to deep philosophical questions. While Machlup and Haberler continued to work on economic theory, the Austrian influence on their work after World War II was barely recognizable. Morgenstern and Schumpeter made major contributions to economics, but did not hide their alienation from the doctrines of the Austrian School.

So there was little reason to expect that the Austrian School would survive its dispersal when the Nazis marched unopposed into Vienna in 1938. That it did survive is in no small measure due to its ideological usefulness to anti-socialist supporters who provided financial support to Hayek, enabling his appointment to the Committee on Social Thought at the University of Chicago, and Mises’s appointment at NYU, and other forms of research support to Hayek, Mises and other like-minded scholars, as well as funding the Mont Pelerin Society, an early venture in globalist networking, started by Hayek in 1947. Such support does not discredit the research to which it gave rise. That the survival of the Austrian School would probably not have been possible without the support of wealthy benefactors who anticipated that the Austrians would advance their political and economic interests does not invalidate the research thereby enabled. (In the interest of transparency, I acknowledge that I received support from such sources for two books that I wrote.)

Because Austrian School survivors other than Mises and Hayek either adapted themselves to mainstream thinking without renouncing their earlier beliefs (Haberler and Machlup) or took an entirely different direction (Morgenstern), and because the economic mainstream shifted in two directions that were most uncongenial to the Austrians: Walrasian general-equilibrium theory and Keynesian macroeconomics, the Austrian remnant, initially centered on Mises at NYU, adopted a sharply adversarial attitude toward mainstream economic doctrines.

Despite its minute numbers, the lonely remnant became a house divided against itself, Mises’s two outstanding NYU disciples, Murray Rothbard and Israel Kirzner, holding radically different conceptions of how to carry on the Austrian tradition. An extroverted radical activist, Rothbard was not content just to lead a school of economic thought, he aspired to become the leader of a fantastical anarchistic revolutionary movement to replace all established governments under a reign of private-enterprise anarcho-capitalism. Rothbard’s political radicalism, which, despite his Jewish ancestry, even included dabbling in Holocaust denialism, so alienated his mentor, that Mises terminated all contact with Rothbard for many years before his death. Kirzner, self-effacing, personally conservative, with no political or personal agenda other than the advancement of his own and his students’ scholarship, published hundreds of articles and several books filling 10 thick volumes of his collected works published by the Liberty Fund, while establishing a robust Austrian program at NYU, training many excellent scholars who found positions in respected academic and research institutions. Similar Austrian programs, established under the guidance of Kirzner’s students, were started at other institutions, most notably at George Mason University.

One of the founders of the Cato Institute, which for nearly half a century has been the leading avowedly libertarian think tank in the US, Rothbard was eventually ousted by Cato, and proceeded to set up a rival think tank, the Ludwig von Mises Institute, at Auburn University, which has turned into a focal point for extreme libertarians and white nationalists to congregate, get acquainted, and strategize together.

Isolation and marginalization tend to cause a subspecies either to degenerate toward extinction, to somehow blend in with the members of the larger species, thereby losing its distinctive characteristics, or to accentuate its unique traits, enabling it to find some niche within which to survive as a distinct sub-species. Insofar as they have engaged in economic analysis rather than in various forms of political agitation and propaganda, the Rothbardian Austrians have focused on anarcho-capitalist theory and the uniquely perverse evils of fractional-reserve banking.

Rejecting the political extremism of the Rothbardians, Kirznerian Austrians differentiate themselves by analyzing what they call market processes and emphasizing the limitations on the knowledge and information possessed by actual decision-makers. They attribute this misplaced focus on equilibrium to the extravagantly unrealistic and patently false assumptions of mainstream models on the knowledge possessed by economic agents, which effectively make equilibrium the inevitable — and trivial — conclusion entailed by those extreme assumptions. In their view, the focus of mainstream models on equilibrium states with unrealistic assumptions results from a preoccupation with mathematical formalism in which mathematical tractability rather than sound economics dictates the choice of modeling assumptions.

Skepticism of the extreme assumptions about the informational endowments of agents covers a range of now routine assumptions in mainstream models, e.g., the ability of agents to form precise mathematical estimates of the probability distributions of future states of the world, implying that agents never confront decisions about which they are genuinely uncertain. Austrians also object to the routine assumption that all the information needed to determine the solution of a model is the common knowledge of the agents in the model, so that an existing equilibrium cannot be disrupted unless new information randomly and unpredictably arrives. Each agent in the model having been endowed with the capacity of a semi-omniscient central planner, solving the model for its equilibrium state becomes a trivial exercise in which the optimal choices of a single agent are taken as representative of the choices made by all of the model’s other, semi-omnicient, agents.

Although shreds of subjectivism — i.e., agents make choices based own preference orderings — are shared by all neoclassical economists, Austrian criticisms of mainstream neoclassical models are aimed at what Austrians consider to be their insufficient subjectivism. It is this fierce commitment to a robust conception of subjectivism, in which an equilibrium state of shared expectations by economic agents must be explained, not just assumed, that Chancellor properly identifies as a distinguishing feature of the Austrian School.

Menger’s original idea of marginal utility was posited on the subjective preferences of consumers. This subjectivist position was retained by subsequent generations of the school. It inspired a tradition of radical individualism, which in time made the Austrians the favorite economists of American libertarians. Subjectivism was at the heart of the Austrians’ polemical rejection of Marxism. Not only did they dismiss Marx’s labor theory of value, they argued that socialism couldn’t possibly work since it would lack the means to allocate resources efficiently.

The problem with central planning, according to Hayek, is that so much of the knowledge that people act upon is specific knowledge that individuals acquire in the course of their daily activities and life experience, knowledge that is often difficult to articulate – mere intuition and guesswork, yet more reliable than not when acted upon by people whose livelihoods depend on being able to do the right thing at the right time – much less communicate to a central planner.

Chancellor attributes Austrian mistrust of statistical aggregates or indices, like GDP and price levels, to Austrian subjectivism, which regards such magnitudes as abstractions irrelevant to the decisions of private decision-makers, except perhaps in forming expectations about the actions of government policy makers. (Of course, this exception potentially provides full subjectivist license and legitimacy for macroeconomic theorizing despite Austrian misgivings.) Observed statistical correlations between aggregate variables identified by macroeconomists are dismissed as irrelevant unless grounded in, and implied by, the purposeful choices of economic agents.

But such scruples about the use of macroeconomic aggregates and inferring causal relationships from observed correlations are hardly unique to the Austrian school. One of the most important contributions of the 20th century to the methodology of economics was an article by T. C. Koopmans, “Measurement Without Theory,” which argued that measured correlations between macroeconomic variables provide a reliable basis for business-cycle research and policy advice only if the correlations can be explained in terms of deeper theoretical or structural relationships. The Nobel Prize Committee, in awarding the 1975 Prize to Koopmans, specifically mentioned this paper in describing Koopmans’s contributions. Austrians may be more fastidious than their mainstream counterparts in rejecting macroeconomic relationships not based on microeconomic principles, but they aren’t the only ones mistrustful of mere correlations.

Chancellor cites mistrust about the use of statistical aggregates and price indices as a factor in Hayek’s disastrous policy advice warning against anti-deflationary or reflationary measures during the Great Depression.

Their distrust of price indexes brought Austrian economists into conflict with mainstream economic opinion during the 1920s. At the time, there was a general consensus among leading economists, ranging from Irving Fisher at Yale to Keynes at Cambridge, that monetary policy should aim at delivering a stable price level, and in particular seek to prevent any decline in prices (deflation). Hayek, who earlier in the decade had spent time at New York University studying monetary policy and in 1927 became the first director of the Austrian Institute for Business Cycle Research, argued that the policy of price stabilization was misguided. It was only natural, Hayek wrote, that improvements in productivity should lead to lower prices and that any resistance to this movement (sometimes described as “good deflation”) would have damaging economic consequences.

The argument that deflation stemming from economic expansion and increasing productivity is normal and desirable isn’t what led Hayek and the Austrians astray in the Great Depression; it was their failure to realize the deflation that triggered the Great Depression was a monetary phenomenon caused by a malfunctioning international gold standard. Moreover, Hayek’s own business-cycle theory explicitly stated that a neutral (stable) monetary policy ought to aim at keeping the flow of total spending and income constant in nominal terms while his policy advice of welcoming deflation meant a rapidly falling rate of total spending. Hayek’s policy advice was an inexcusable error of judgment, which, to his credit, he did acknowledge after the fact, though many, perhaps most, Austrians have refused to follow him even that far.

Considered from the vantage point of almost a century, the collapse of the Austrian School seems to have been inevitable. Hayek’s long-shot bid to establish his business-cycle theory as the dominant explanation of the Great Depression was doomed from the start by the inadequacies of the very specific version of his basic model and his disregard of the obvious implication of that model: prevent total spending from contracting. The promising young students and colleagues who had briefly gathered round him upon his arrival in England, mostly attached themselves to other mentors, leaving Hayek with only one or two immediate disciples to carry on his research program. The collapse of his research program, which he himself abandoned after completing his final work in economic theory, marked a research hiatus of almost a quarter century, with the notable exception of publications by his student, Ludwig Lachmann who, having decamped in far-away South Africa, labored in relative obscurity for most of his career.

The early clash between Keynes and Hayek, so important in the eyes of Chancellor and others, is actually overrated. Chancellor, quoting Lachmann and Nicholas Wapshott, describes it as a clash of two irreconcilable views of the economic world, and the clash that defined modern economics. In later years, Lachmann actually sought to effect a kind of reconciliation between their views. It was not a conflict of visions that undid Hayek in 1931-32, it was his misapplication of a narrowly constructed model to a problem for which it was irrelevant.

Although the marginalization of the Austrian School, after its misguided policy advice in the Great Depression and its dispersal during and after World War II, is hardly surprising, the unwillingness of mainstream economists to sort out what was useful and relevant in the teachings of the Austrian School from what is not was unfortunate not only for the Austrians. Modern economics was itself impoverished by its disregard for the complexity and interconnectedness of economic phenomena. It’s precisely the Austrian attentiveness to the complexity of economic activity — the necessity for complementary goods and factors of production to be deployed over time to satisfy individual wants – that is missing from standard economic models.

That Austrian attentiveness, pioneered by Menger himself, to the complementarity of inputs applied over the course of time undoubtedly informed Hayek’s seminal contribution to economic thought: his articulation of the idea of intertemporal equilibrium that comprehends the interdependence of the plans of independent agents and the need for them to all fit together over the course of time for equilibrium to obtain. Hayek’s articulation represented a conceptual advance over earlier versions of equilibrium analysis stemming from Walras and Pareto, and even from Irving Fisher who did pay explicit attention to intertemporal equilibrium. But in Fisher’s articulation, intertemporal consistency was described in terms of aggregate production and income, leaving unexplained the mechanisms whereby the individual plans to produce and consume particular goods over time are reconciled. Hayek’s granular exposition enabled him to attend to, and articulate, necessary but previously unspecified relationships between the current prices and expected future prices.

Moreover, neither mainstream nor Austrian economists have ever explained how prices are adjust in non-equilibrium settings. The focus of mainstream analysis has always been the determination of equilibrium prices, with the implicit understanding that “market forces” move the price toward its equilibrium value. The explanatory gap has been filled by the mainstream New Classical School which simply posits the existence of an equilibrium price vector, and, to replace an empirically untenable tâtonnement process for determining prices, posits an equally untenable rational-expectations postulate to assert that market economies typically perform as if they are in, or near the neighborhood of, equilibrium, so that apparent fluctuations in real output are viewed as optimal adjustments to unexplained random productivity shocks.

Alternatively, in New Keynesian mainstream versions, constraints on price changes prevent immediate adjustments to rationally expected equilibrium prices, leading instead to persistent reductions in output and employment following demand or supply shocks. (I note parenthetically that the assumption of rational expectations is not, as often suggested, an assumption distinct from market-clearing, because the rational expectation of all agents of a market-clearing price vector necessarily implies that the markets clear unless one posits a constraint, e.g., a binding price floor or ceiling, that prevents all mutually beneficial trades from being executed.)

Similarly, the Austrian school offers no explanation of how unconstrained price adjustments by market participants is a sufficient basis for a systemic tendency toward equilibrium. Without such an explanation, their belief that market economies have strong self-correcting properties is unfounded, because, as Hayek demonstrated in his 1937 paper, “Economics and Knowledge,” price adjustments in current markets don’t, by themselves, ensure a systemic tendency toward equilibrium values that coordinate the plans of independent economic agents unless agents’ expectations of future prices are sufficiently coincident. To take only one passage of many discussing the difficulty of explaining or accounting for a process that leads individuals toward a state of equilibrium, I offer the following as an example:

All that this condition amounts to, then, is that there must be some discernible regularity in the world which makes it possible to predict events correctly. But, while this is clearly not sufficient to prove that people will learn to foresee events correctly, the same is true to a hardly less degree even about constancy of data in an absolute sense. For any one individual, constancy of the data does in no way mean constancy of all the facts independent of himself, since, of course, only the tastes and not the actions of the other people can in this sense be assumed to be constant. As all those other people will change their decisions as they gain experience about the external facts and about other people’s actions, there is no reason why these processes of successive changes should ever come to an end. These difficulties are well known, and I mention them here only to remind you how little we actually know about the conditions under which an equilibrium will ever be reached.

In this theoretical muddle, Keynesian economics and the neoclassical synthesis were abandoned, because the key proposition of Keynesian economics was supposedly the tendency of a modern economy toward an equilibrium with involuntary unemployment while the neoclassical synthesis rejected that proposition, so that the supposed synthesis was no more than an agreement to disagree. That divided house could not stand. The inability of Keynesian economists such as Hicks, Modigliani, Samuelson and Patinkin to find a satisfactory (at least in terms of a preferred Walrasian general-equilibrium model) rationalization for Keynes’s conclusion that an economy would likely become stuck in an equilibrium with involuntary unemployment led to the breakdown of the neoclassical synthesis and the displacement of Keynesianism as the dominant macroeconomic paradigm.

But perhaps the way out of the muddle is to abandon the idea that a systemic tendency toward equilibrium is a property of an economic system, and, instead, to recognize that equilibrium is, as Hayek suggested, a contingent, not a necessary, property of a complex economy. Ludwig Lachmann, cited by Chancellor for his remark that the early theoretical clash between Hayek and Keynes was a conflict of visions, eventually realized that in an important sense both Hayek and Keynes shared a similar subjectivist conception of the crucial role of individual expectations of the future in explaining the stability or instability of market economies. And despite the efforts of New Classical economists to establish rational expectations as an axiomatic equilibrating property of market economies, that notion rests on nothing more than arbitrary methodological fiat.

Chancellor concludes by suggesting that Wasserman’s characterization of the Austrians as marginalized is not entirely accurate inasmuch as “the Austrians’ view of the economy as a complex, evolving system continues to inspire new research.” Indeed, if economics is ever to find a way out of its current state of confusion, following Lachmann in his quest for a synthesis of sorts between Keynes and Hayek might just be a good place to start from.

My Paper “Hayek, Deflation, Gold, and Nihilism” Is now Available on SSRN

I contributed a chapter entitled “Hayek, Deflation, Gold and Nihilism” to volume 13 of Hayek: A Collaborative Biography edited by Robert Leeson and published in 2018 by Palgrave Macmillan.

I have posted a preliminary draft of that chapter on SSRN. Here is the abstract.

In Hayek’s early writings on business cycle theory and the Great Depression he argued that business cycle downturns including the steep downturn of 1929-31 were caused by unsustainable elongations of capital structure of the economy resulting from bank-financed investment in excess of voluntary saving. Because monetary expansion was the cause of the crisis, Hayek argued that monetary expansion was an inappropriate remedy to cure the deflation and high unemployment caused by the crisis. He therefore recommended allowing the Depression to take its course until the distortions that led to the downturn could be corrected by market forces. However, this view of the Depression was at odds with Hayek’s own neutral money criterion which implied that prices should fall during expansions and rise during contractions so that nominal spending would remain more or less constant over the cycle. Although Hayek strongly favored allowing prices to fall in the expansion, he did not follow the logic of his own theory in favoring generally increasing prices during the contraction. This paper explores the reasons for Hayek’s reluctance to follow the logic of his own theory in his early policy recommendations. The key factors responsible for his early policy recommendations seem to be his attachment to the gold standard and the seeming necessity for countries to accept deflation to maintain convertibility and his hope or expectation that deflation would overwhelm the price rigidities that he believed were obstructing the price mechanism from speeding a recovery. By 1935 Hayek’s attachment to the gold standard was starting to weaken, and in later years he openly acknowledged that he had been mistaken not to favor policy measures, including monetary expansion, designed to stabilize total spending.


About Me

David Glasner
Washington, DC

I am an economist in the Washington DC area. My research and writing has been mostly on monetary economics and policy and the history of economics. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey’s unduly neglected contributions to the attention of a wider audience.

My new book Studies in the History of Monetary Theory: Controversies and Clarifications has been published by Palgrave Macmillan

Follow me on Twitter @david_glasner

Archives

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 3,261 other subscribers
Follow Uneasy Money on WordPress.com