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 Gold production fell during World War I, but gold coins had been withdrawn from circulation and replaced with paper or token coins.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.
Uneasy Money 
Could it be that France was accumulating gold above the 40% reserve ratio because of it’s own debts?
https://en.wikipedia.org/wiki/Mellon%E2%80%93Berenger_Agreement
“In a statement on 28 July 1929, Hoover outlined the terms of the agreement. France would make payments of $35 million in the 1930 fiscal year, with the amount payable rising over the next eleven years to a maximum of $125 million annually. The value of the payments was then $1,680 million.”
Your statement:
“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).”
Okay, so $1.7 billion in gold reserves with a 40% reserve ratio plus another $1.68 billion in debt payments (over eleven years) to the US gets you to about $3.38 billion needed by France over an eleven year period starting in 1929. And that assumes a steady stream of new gold discoveries as well as tight credit conditions in France over the same period.
Was it insane for France to accumulate gold to pay back the US as well as maintain it’s own economy?
What seems to be missing from the “insane bank of France” argument is that world gold discoveries were dwindling by 1931 and France needed a plan of it’s own to pay off it’s war debts above and beyond whatever domestic monetary policy gold needs it had.
LikeLike
David,
You said:
“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.”
As I have demonstrated several times, the French and US gold accumulation did not cause a reduction of gold holdings by the other industrialized countries, in fact these countries increased their gold holdings. The only countries to lose gold in the relevant period were the agricultural countries and Japan which had deteriorating balance of payments situations due to falling commodity prices.
You have never responded to these facts.
A few blogs back I supplied interest rate movement data for the major economies. These showed that interest rates peaked in 1928 and 1929 and declined thereafter until the crisis in sterling. If US and French gold accumulation was causing stress to the rest of the world, interest rates in the rest of the world should have gone up. They did not. They fell.
You said:
“The only reason the Fed raised interest rates in 1928 was a misguided concern with rising stock prices. ”
You say “misguided”.
There was concern that the stock market no longer reflected real values and that there was massive speculation financed on margin and with debt. There was also increasing over-investment in real productive capacity. This situation could not be allowed to continue and would naturally have found a denouement anyway.
You said:
“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.”
That is not correct. Demonstrably, capital flows into France of all types were evident in 1928-1929.
You said:
“Unemployment was a British problem caused by an overvalued exchange rate; it was not a systemic gold-standard problem.”
Given Britain was a significant world economy, its overvaluation of the pound had serious deleterious world wide implications. The overvaluation of the pound was the other side of the coin to the undervaluation of the franc.
A general point. In my reading of your blog over the years it seems to me that you support Cassel’s notion that there was an under supply of gold and that this fostered a tendency to deflation. Now you are arguing the reverse.
LikeLike
Henry,
Thanks for engaging me yet again on this topic. Unfortunately we seem to be repeating ourselves and rehashing the same arguments, but I will keep responding as I have in the past. You say that I have not responded in the past to the facts that you cite. I disagree. I have responded. You may not like my response, which is your right, but I believe that I have responded. In brief, as I showed in my response to Gabriel Mathy, the accumulation of gold by the US, France, Belgium, Netherlands, and Switzerland, was far greater than the total increase in world reserves. Whether it can be accounted for by Canada, Australia, Brazil and Japan I do not know, but I am inclined to doubt that that is the case. Even if it were it would not change the fact that the increased monetary demand for gold caused a rapid increase in the total monetary gold holdings of the world, reducing the quantity of gold available for non-monetary uses and implying a rising relative value of gold in terms of commodities, which under a gold standard implied deflation. That is exactly what was observed starting in the second half of 1929 and deflation only increased for the next several years. That deflation in terms of money convertible into gold was the proximate and most significant factor causing the Great Depression.
I remember well that you cited falling nominal interest rates beginning in 1930. Everyone knows that interest rates fell after the crash. Perhaps you disagree, but most economists believe that the distinction between nominal and real interest rates is important, and given that deflation was raging and expected to continue after the crash, the reduction in nominal interest rates does not demonstrate that monetary policy was not too tight. The stress imposed by the gold-accumulation policies of France and the US on the reserve position of other countries was reflected in the rates in those countries relative to rates in countries accumulating gold, not in the level of rates worldwide.
I am well aware of the concerns about stock-market speculation. I regard those concerns, even if sincerely held, as bogus. You have no basis other than specious claims about over-investment and speculation for the claim that the output expansion in the US and other countries in the 1920s was unsustainable. Hawtrey demolished claims of stock-market overvaluation in The Art of Central Banking (chapter two).
Obviously, gold was flowing into France before 1931. However, it was only after Britain and the sterling bloc left the gold standard in September 1931 that there was a massive speculative movement of gold from the US to France, Belgium, Netherlands, and Switzerland. That was the movement that I was primarily referring to. There was probably also significant movement of gold from Britain earlier in 1931 before Britain wisely decided to end gold convertibility and allow sterling to depreciate, though a more aggressive easing of monetary policy after suspension of convertibility would have hastened an initially sluggish recovery.
Unemployment in Britain in 1929 was about 10%, perhaps slightly less depending on who is counting. That was certainly too high, but that was primarily Britain’s problem not the world’s problem. It was only after gold started appreciating late in 1929 that deflation became a problem for the world. The British unemployment rate nearly doubled by the end of 1930. It doubled because of gold accumulation that caused deflation in all gold-standard countries, not because of British policy. The French used an undervalued franc as a tool of gold accumulation by allowing the quantity of domestic francs to increase only through the sale of gold to the Bank of France.
I (along with Hawtrey) disagree with Cassel that there was a long-run problem of insufficient gold production that would have led to secular deflation. That concern by Cassel was separate and distinct from the concern Hawtrey and Cassel shared that a rapid increase in the monetary demand for gold would cause a sudden and disastrous deflation if steps were not taken to limit the accumulation of gold reserves by the central banks. So there is no inconsistency on my part or that of Hawtrey. Nor was Cassel inconsistent. He had two distinct concerns and it was only the second one that was at issue in 1929-32.
LikeLike
David,
” Whether it can be accounted for by Canada, Australia, Brazil and Japan I do not know, but I am inclined to doubt that that is the case. ”
I have provided the figures in the past. If you seriously considered the point you would study these figures. I think your fixation on France and the US has blinded you to the facts. I have to conclude you are not willing to look because this would undermine the rationale for your fixation on France and the US if it were true.
“The stress imposed by the gold-accumulation policies of France and the US on the reserve position of other countries was reflected in the rates in those countries relative to rates in countries accumulating gold, not in the level of rates worldwide.”
The fact is the nominal rates all fell in unison.
“You have no basis other than specious claims about over-investment and speculation for the claim that the output expansion in the US and other countries in the 1920s was unsustainable.”
The Dow Jones increased almost 4 times in value from 1925 to 1929. I have had professional experience in the markets going back to the mid 1970s. This kind of rise in that time frame is exceptional. I would say Hawtrey was talking through his hat.
“That was certainly too high, but that was primarily Britain’s problem not the world’s problem.”
Britain in the late 1920s was still a significant player in the world economy. Problems in the UK meant problems elsewhere.
LikeLike
Henry,
Thanks so much for providing figures in the past. It may surprise you to read this, but I’m sorry to say that I don’t commit your comments on my blog to memory. And I’m not about to go searching for them. If you want me to consider them, provide them to me again. You keep bringing up the same arguments repeatedly. If you want me repeat commenting on the same arguments over again you can keep providing the figures to me repeatedly.
Nominal rates did fall in unison, but that doesn’t mean that the relationship between rates in different countries did not change. If you want me to address that issue, go ahead and calculate the differences in rates among different countries from 1929 to 1932 yourself and see if they changed. I am not your research assistant.
But that is not the main point. The main point is that, under the gold standard, interest rates were set by central banks with a view to ensuring that there would be no external drain of gold even though those rates were too high, despite falling from 1929 levels, to permit a recovery.
It is certainly possible that stock prices were overvalued in 1929 (I don’t believe in EMH), but that also doesn’t mean that they were unrelated to economic conditions at the time. Even substantial overvaluation of stock prices doesn’t mean that the Great Depression was already baked in. The DJIA from 1929 to the spring of 1930 lost about 40% of its value and then quickly gained back almost half its losses, before resuming its steep downward slide. By 1933 before FDR devalued the dollar and abandoned the gold standard (triggering a 50% increase in the DJIA within 3 months), the DJIA lost well over 80% of its 1929 value. That monumental loss of value wasn’t because the DJIA was overvalued in 1929; it was because of policies followed by the US and France before and after 1929.
Finally, your remark about Hawtrey is not only entirely out of order; it is just foolish.
The British economy was troubled before rejoining the gold standard in 1925 and remained troubled after. The resumption of the gold standard marginally worsened conditions in 1925 but the British economy continued to grow slowly till the Great Depression started, so there is no reason to think that problems in Britain played a central role in the Great Depression
LikeLike
David,
I don’t expect anything from you other than a scholarly approach – it is what is generally evident in your blogging.
I think your blog is one of the most interesting around. It deals with subjects I find interesting. Not only that, you deal with them in an intelligent, thoughtful, considered and balanced way.
I was shocked and stunned to read that: “Whether it can be accounted for by Canada, Australia, Brazil and Japan I do not know, but I am inclined to doubt that that is the case. ”
The figures I have from time to time reported are those assembled by the League of Nations statisticians of the time. If you are not inclined to give them any credence then all I can say is that you are so fixated on the notion that the gold accumulation by the US and France in the late 1920s was the cause of a shortage of gold and the ultimate cause of the Great Depression that you have abandoned your generally scholarly approach to discussing matters to hand.
LikeLike
Thanks, Henry, for your kind words about my blog. They are deeply appreciated.
I have two problems with the data you referred to. First, because there has been a substantial lapse of time since you posted the League of Nations data, and I don’t recall specifics. So I only meant to say that I could not give a specific response to your assertions about the data showed.
Second, and more important, even if the data are exactly as you represented, the data would not disprove my basic argument. Obviously, an increase in the demand for a stock of some good in fixed supply will cause some reallocation of the stock unless the demand of each agent increases by exactly the same amount, in which case there is no reallocation. But even if there were no reallocation at all, the increase in demand would be manifested in an increase in the value of each unit of the stock. Your assertion that not all countries experience a reduction in its gold reserves does not prove that the increase in demand for gold by the Bank of France and the Fed did not cause gold to appreciate. The increase in demand did lead to the appreciation, which led directly to deflation, depression and disaster for humankind.
LikeLike
David,
“Obviously, an increase in the demand for a stock of some good in fixed supply will cause some reallocation of the stock unless the demand of each agent increases by exactly the same amount, in which case there is no reallocation. But even if there were no reallocation at all, the increase in demand would be manifested in an increase in the value of each unit of the stock.”
I would say not necessarily but probably.
“Your assertion that not all countries experience a reduction in its gold reserves does not prove that the increase in demand for gold by the Bank of France and the Fed did not cause gold to appreciate.”
No it does not necessarily but then it does throw doubt on your strident assertion that it did. This doubt is worthy of some consideration. Getting excited about large increases in the US and French gold hoard is getting excited about the obvious, which every commentator I have read has focused on and not noticed what happened elsewhere. Not all the facts have been fully considered.
What also has to be remembered is that c. $400M in gold was coming out of the mines annually adding to supply and that the French accumulation was in part due to German war reparations.
“The increase in demand did lead to the appreciation, which led directly to deflation, depression and disaster for humankind. ”
Which deflation are we talking about precisely? Commodity prices? There are those that argue the secular fall in commodity prices through the 1920s was in part due to the large increase in productivity brought by new technological developments. I think this is an important factor. It would explain the BOP problems experienced by the agricultural countries and their loss of gold.
I am not arguing that the US and French gold accumulation did not put stress on the system. Whether it was catastrophic as you assert is still an open question as far as I am concerned.
The collapse in Wall Street directly affected the solvency of big players and destroyed confidence in the world economy.
I think the latter factor was the most telling feature of the late 1920s. It caused massive redirection of capital to safe havens like France and away from Britain with an overvalued currency.
And France did attempt to ease the pressure on the British by retaining high sterling balances. They were converted to gold in 1931 when it was obvious to all that the pressure on Britain to devalue was overwhelming. Who would hold large tranches of sterling in the face of such circumstances?
LikeLike
Henry,
You say probably, I say necessarily.
An increase in the demand for gold must cause the value of gold to increase. The stock of gold is nearly fixed in the short run. The total production of gold in any year is rarely more than 3-4 percent of the existing stock. So substantial changes in the value of gold in any short period are necessarily caused by changes in the demand for gold not the supply. Almost all the increase in monetary gold reserves from 1928-31 wound up in the vaults of the New York Fed and the Bank of France, so it was clearly the demand of those two institutions that was causing gold to appreciate. Reparation payments did not have to be settled by shipments of gold, they could have been discharged by means of a German export surplus. France, however, undervalued its currency and only allowed the quantity of francs to increase by way of gold deposits at the Bank of France.
There were two sharp periods of deflation 1920-22 and 1929-32. From 1922 to 1928, there was relative overall stability. But even in the 1922-28 period of relative stability, as in every period of relative stability, there were sectors, e.g., agriculture, in which some prices were falling. There’s nothing inherently unusual about that.
The stock market crash was certainly a destabilizing event, but the Great Depression did not follow inexorably from that shock. The destabilizing speculation that started in 1931 was largely the result of a gold standard that had become unmanageable and destructive, although that outcome could have been avoided if the Fed and the Bank of France had avoided the gold accumulation of 1928-30 that caused things to fall apart.
France did provide some forbearance to the Bank of England, but that forbearance was the exception to an otherwise destructive and self-destructive policy.
LikeLike
David,
“The destabilizing speculation that started in 1931 was largely the result of a gold standard that had become unmanageable and destructive,”
At the heart of the 1931 crisis was the overvaluation of sterling.
On the Continent, the collapse of the Danat Bank and Creditanstalt had a lot to do with the politics of German war reparations.
Re the deflation of the late 1920s, a word from the Fed Reserve (Federal Reserve Bulletin, June 1930, p.339):
“There appears to be no evidence in the available information that the price declines in recent years have at any time reflected a general shortage of bank reserves, or of gold, but there are indications that the diversion of funds to this country during the period of high money rates contributed to the difficulties of economic reconstruction in Europe. Furthermore, unfavorable conditions in our bond market in 1929 made it difficult for foreign countries to arrange for long-term financing in this country, and were a further factor tending to delay industrial recovery abroad and to depress the world level of commodity prices.”
This statement is in contradiction to your assertion that there was a shortage of gold in the late 1920s and that this shortage of gold precipitated a general deflation.
And in support of my provision of factual official gold holdings data, more from the Fed (Federal Reserve Bulletin, June 1931, p.303):
“The character of the shift in gold holdings during the year has been commented upon on previous occasions. The more complete figures now available confirm the statements previously made that during 1930 the unusually large addition to the world’s central gold reserves was for the most part acquired by France and the United States and that these countries in addition were the recipients of considerable amounts of gold exported by outlying raw-material producing countries, while the other large commercial countries have maintained their gold reserves at a fairly constant level.”
This statement totally supports the data from the League of Nations which I have reported several times previously, the veracity of which you doubt.
LikeLike
The pound was overvalued from 1925 to 1929 without any crisis, so, no, the overvaluation of sterling was not at the heart of 1931 crisis. What was at the heart of the crisis was the likelihood that Britain would not be stupid enough to tolerate — by remaining on the gold standard — an accelerating deflation with unemployment continuing to increase from an already intolerable rate of close to 20%, nearly double the excessive rate that had been declining, albeit too slowly, from 1925 to 1929.
I agree that the 1931 sterling crisis was precipitated by the failure of the Creditanstalt which was deeply affected by worsening situation in Germany. Things were rapidly falling apart under the pressure of the ferocious deflation unleased by the Fed and the Bank of France. And it was folly for the UK to continue subjecting itself to the punishment associated with staying on the gold standard, and the British government wisely, but belatedly, cut itself loose from the golden chains that were leading to increasing misery and destruction.
Henry, I am shocked that you would expect me to be impressed in the slightest for even a nanosecond by the self-serving assessment of the crisis caused in large measure by the Fed’s own deliberate policy choices. You might as well have quoted to me the self-serving statements of the Bank of France. I regard such statements as entirely worthless and devoid of credibility. I don’t have the figures in front of me, so I can’t comment in detail on the extent to which they are entirely accurate representations of reality. I do note the resort to such weasel words as “other large commercial countries” and “fairly constant level.”
I don’t believe that I have challenged the strict veracity of the League of Nations data. But I have questioned the interpretation of the data that you and apologists for the Fed and the Bank of France have drawn from the data. So I repeat , even if many countries were able to avoid any loss of gold reserves during the 1928-31 period, that in no way detracts for the deflationary effect of the increase in gold demand implicit in the large accumulation of gold reserves by the Fed and the Bank of France. And your repeated citations of that fact are entirely irrelevant to the proposition that French and US gold accumulation had a horrific deflationary effect.
LikeLike
David,
“The pound was overvalued from 1925 to 1929 without any crisis, so, no, the overvaluation of sterling was not at the heart of 1931 crisis.”
So why was it that after Britain left the gold standard in September 1931 the pound, over the short term, collapsed at one point by over 30% and settled for a considerable period over 1932 to a loss of c. 30%. Countries left, right and centre had been cashing in their sterling holdings for British gold. The French held on to some sterling balances.The losses caused by the depreciation cost the Bank of France seven times its capital.
I cannot see how you can argue that the pound was not overvalued.
The factors you mention were the result of the overvalued pound.
” I am shocked that you would expect me to be impressed in the slightest for even a nanosecond by the self-serving assessment of the crisis caused in large measure by the Fed’s own deliberate policy choices. ”
Yes I can understand that thought and I have to say I wonder about it myself. However, on the one hand the Fed says its gold accumulation caused no gold shortage etc.. But the Fed then effectively says its rate hikes caused economic and financial distress abroad. If it was the Bank’s intention to protect and defend itself from any criticism it would not have admitted the latter. So I don’t buy your argument.
“I don’t believe that I have challenged the strict veracity of the League of Nations data. ”
You said above: “Whether it can be accounted for by Canada, Australia, Brazil and Japan I do not know, but I am inclined to doubt that that is the case.”
Your position is fairly clear.
“I don’t have the figures in front of me, so I can’t comment in detail on the extent to which they are entirely accurate representations of reality. I do note the resort to such weasel words as “other large commercial countries” and “fairly constant level.””
I didn’t see the point of presenting the data to you and having it ignored again, so I resorted to generalities.
LikeLike
I never denied that sterling was overvalued. From 1925 through 1928, the overvaluation was stable; after massive deflation was triggered by US and French gold accumulation, the overvaluation of sterling was increased. The increasing overvaluation of sterling from 1929 onward was a symptom, not a cause, of the crisis. The depreciation of sterling after Britain suspended convertibility reflected an expectation that Britain would adopt a reflationary policy to reverse the 1929-31 deflation caused by the US and France. At first, Britain merely halted the deflation but made no attempt to reverse the deflation. Things stopped getting worse, but did not improve until reflation began.
I agree with you that the Fed statement from which you quoted a passage can as easily be construed to be incoherent as it can as it can be construed to be self-serving.
My comment about whether reductions in the gold holdings of Canada, Australia, and Brazil were sufficient to account for the increased gold holdings of other central banks was made with reference to that narrow assertion, not to the veracity of League of Nations statistics in general. I did not ignore your data, but I did not commit the data to memory and I don’t see why you expect me to go searching for it myself when you could easily supply it to me if you have it readily available to you. And I incorporate by reference my earlier comment about the use of weasel words.
LikeLike
David,
“I never denied that sterling was overvalued. ”
I didn’t say you did.
“The increasing overvaluation of sterling from 1929 onward was a symptom, not a cause, of the crisis.”
The overvaluation of sterling was the result of a specific policy implementation of the British Government. It was a fact some years before the gold accumulation by the US and France began. How can it be a symptom?
And I am beginning to see an argument that the accumulation of gold by various countries was the result of and not the cause of the economic dislocation of the late 1920s.
I guess will we have to disagree on what caused what.
“My comment about whether reductions in the gold holdings of Canada, Australia, and Brazil were sufficient to account for the increased gold holdings of other central banks was made with reference to that narrow assertion,…”
The point is that this “narrow assertion” is based on factual League of Nations data.
If you are now genuinely interested in the data I suggest you seek it out for yourself. It can be found in the reports produced by the League of Nations Gold Delegation in the early 1930s. Possibly also in Brown. I am sure that you have available to you the requisite resources.
LikeLike
Henry (12 hours ago):
“I cannot see how you can argue that the pound was not overvalued.”
David (3 hours ago):
“I never denied that sterling was overvalued. ”
Henry (8 minutes ago):
“I didn’t say you did.”
Narrator: He did.
LikeLike
Henry: “Yes I did.” – LOL.
LikeLike