Since Friedmann and Schwartz's Monetary History (1963) has had such a major impact on the debate about the causes of the Great Depression, it seems desireable to explain at greater length than possible in the body of my text my reasons for finding their explanations unsatisfactory and implausible.
Interpretations of the theoretical aspects of the book is rendered difficult by its being primarily a historical and not a theoretical work. It provides a dense and scholarly account of monetary development and the evolution of policy in the ninety three years after 1867. The four depression years take 120 pages of the 700 pages of the book (a fifth of the space being footnotes). Within that are around 40 passages varying in length from a paragraph to a sentence which state or imply a view about the causality of the depression. In addition there are about 20 more such passages in the final chapter. These passages are scattered and embedded in not difficult, but dense and demanding, discussions of statistical developments and conflicting views and personalities, themselves requiring attentive reading: they are therefore difficult to do justice to on a first reading, or to extract and collate afterwards.
The massive nature of F&S's history is dictated by the structure of each of its main chapters. Chapter 7 on the Great Depression starts with a general review, covering almost everything discussed later. Section 2 appears to be constructed on the plan of discussing chronologically different monetary aggregates and sub aggregates and the relations between them (ratios and identities); but much else is included within that framework. Section 3 is about bank failures, already discussed once; and section 4 about financial events abroad. Then come two sections discussing, first, monetary policy and contemporary discussions of it, and, second, Friedman's comments thereon. Repetition in analytic narrative is impossible to avoid, but F&S have allowed many layers of it to remain.
The steady thoroughness of their prose acts to induce acceptance of preannounced propositions long before evidence in support of them comes into sight. Chapter 8, for example, is entitled "The Great Contraction", and since that phrase is used in a dual sense, it functions as a suggestion of causal connection. The Great Contraction can apply, on the one hand, to the great decline in real output, but it can apply also (and more aptly) to the contraction of the money stock. The impression given is thus that the two were inextricably intertwined, and that the monetary contraction was the essence of it. That is indeed what F&S believe, and say on the third page; but to which they never later give precision, or, even by the end, much evidential support.
The title of the final section of Chapter 7 is equally indicative. On emight have expected here a connected discussion of whether monetary contraction might have been not cause but the effect of the real contraction, a question of which the authors are aware: had that been the case, it could have been difficult for the authorities to stop money contracting. Instead, the question is begged. The title of the final section is "Why Was Monetary Policy So Inept", - as if by this time the authors had earned their licence to drop the appearance of scholarly caution and objectivity - and we are given an explanation (of something that has never been shown to require explanation) in terms of a "shift in power" within the Federal Reserve System (p 411).
In the following paragraphs I try to analyse F&S's argument by collating statements made at different points. I will discuss it under four headings:
1) why the real depression occured;
2) why the monetary contraction occured;
3) the effects of bank failures, capital adequacy and the flight to cash;
4) the authors identification of three occassions on which in their view the Federal Reserve acted in a sharply restricted fashion
Underlying everything that F&S say on the occassion of The Depression is the assumption that (in the short term as well as the long) the "velocity" of money reflects the "money holding propensities of the community" (p679). Under the normal ceteris paribus procedure, tastes can be taken to be unchanged. By this logic, the authors are enabled to write as though a movement of the monetary aggregate itself implied a simple and direct sympathetic influence on real output.
To one of another way of thinking, this line of argument appears tendentious. In my view, clarity of thought requires recognition of the fact that money supplied by the banking system is not necessarily in equilibrium with the stock the public want to hold, and that in the short term therefore velocity is no more than a ratio between two magnitudes determined in partial independence.F&S's view is connected with the fact, discussed further below, that they give little or no weight to the role of banks as lending institutions.
1/. Though the general tone of what F&S say about the great depression clearly implies that it was caused by the monetary contraction, their statements are often hedged and unclear, and appear at times contradictory. (for instance in the opening to the Chapter on "The Great Contraction" after discussing the changes in attitude to the question whether "money matters" they remark that the "contraction is in fact a tragic testimonial to the importance of money forces" (p. 300). It is tragic presumeably becuase of the poverty and unemployment it caused, and the implication clearly is that that was due or largely due to "monetary forces".
One statement is:
prevention or moderation of the decline in the stock of money, let alone the substitution of monetary expansion, would have reduced the contraction's severity, and almost as certainly its duration. The contraction might still have been relatively severe. But it is hardly conceivable that money income could have declined [as much as it did]. (p301).
This statement however comes just after another one that seems to run in the other sense:
True, as events unfolded, the decline in the stock of money, and the near collapse of the banking system can be regarded as a consequence of non-monetary forces in the United States, and monetary and non-monetary forces in the rest of the world.(p300)
But by the end of the book they appear quite clear.
there is..... only one sense in which a case can be made for the proposition that the monetary decline was a consequence of the economic decline. (p691, also p694).
This is a view which they say is possible only if the Federal Reserve System is held to have been a prisoner of contemporary opinion. That, as they say, is not a matter of economic inter-relations, and is not what we are concerned with.
A Monetary History does not provide a connected view of why, in the authors opinion, the depression happened.
2/. F&S for the most part take a simple and extreme view of the cause of monetary contraction:
the monetary authorities could have prevented the decline in the stock of money - indeed could have produced almost any desired increase in the stock of money. (P.301: note the almost]. The monetary decline from 1929 to 1933 was not an inevitable consequence of what had gone before. It was the result of policies followed during those years (p.699).
On the question of how the authorities could have controlled the money stock, they answer, by "extensive open market purchases" as a means to increase bank reserves: What they say of the first year of the depression seems to express their view:
It has been contended with respect to later years (particularly during the years after 1934....) that increases in high-powered money, through expansion of Federal Reserve credit or other means, would simply have added to bank reserves and would not have been used to increase the money stock.... we shall argue later that the contention is invalid, even for the later period. It is clearly not relevant to the period from August 1929 to October 1930. During that period, additional reserves would almost certainly have been put to use promptly. Hence the decline in the stock of money is not only arithmetically to the decline in federal Reserve credit outstanding: it is economically a direct result of that decline. (pp341-2)
F&S, however, never seriously consider the possibility that they were held back by factors other than their reserve position. That is becuase they take practically no note of the asset side of banks' balance sheets. It is striking that, though the book contains many charts and table of bank liabilities it has only one table of bank assets (and then for selected years only). Their conceptual picture allows little place for the role of banks as financial intermediaries who live by striking a balance between the needs of two types of client: depositors (the public as asset owners) and borrowers (the public as investors and debtors). Nor does it allow for the fact that banks as intermediaries operate in a world of uncertainty, that banks' assessment of their client's creditworthiness varies, and that in the short term their lending determines the size of the money stock.
If these effects are accepted as important, the stock of money has to be seen as determined at least in large part by the business situation. On my view, the scale of bank lending during the Great Depression must have been greatly restrained by banks' bad debts, by banks' increased caution in lending to customers whose profit expectations had deteriorated drastically, and by capital inadequacy. It must have been these considerations, not the size of bank reserves, that imposed a limit on the size of the money stock.
(F&S admit that capital adequacy was a serious problem (p330) but argue that the authorities could and should have remedied this by buying securities to raise their price (hence increasing bank capital) and lending to banks after 1932 through the Reconstruction Finance Corporation. But the former would have been only partial indirect help, and the latter would have augmented banks' capital only if the authorities had taken bad assets in exchange for good.
On this view, the fact that the banks' reserves even by the end of the depression had fallen little, though banks' total assets/liabilities fell by about a third is presumptive evidence that reserves remained on the whole ample. (I admit that there may have been particular episodes when banks were constrained by reserve shortage: see further below).
F&S comment exclusively from their own point of view, and can see no force in the opinions of the authorities at the time based on considerations which they dismiss. They quote, for instance, the exchange of views inside the Federal Reserve System in May and June 1930, when most governors felt that there was already "an abundance of funds in the market" (p. 371): as one observed "we have been putting out credit in a period of depression where it was not wanted and could not be used" (p.373). F&S remark "these views seem to us...confused and misguided". To me they seem common sense. Again in it's report for 1930 the Reserve Board described its policy as one of "monetary ease". This draws from F&S the comment that "this is a striking illustration of the ambiguity of the terms "monetary ease" and tightness..... it seems paradoxical to describe as "monetary ease" a policy which permitted the stock of money to decline in fourteen months [as much as it did], (p. 375). But if one admits that the powers of the authorities are limited, it is not paradoxical at all.
3/. Though F&S rightly devote a good deal of space to the three successive waves of bank failures, these do not take a central place in their argument. At times they speak of their having had a contractional effect on the stock of money. But it becomes clear that what is meant is that the crises drained banks of reserves which depressed the money stock only because (in their view) the monetary authorities did not offset this effect so as to maintain bank reserves.
(The second banking crisis had far more severe effects on the stock of money than the first. P.314, see also P.648).
Hence they can say:
the bank failures were important not primarily in their own right but because of their indirect effect. If they had occured to precisely the same extent without producing a drastic effect on the stock of money, they would have been notable but not crucial. If they had not occurred, but a correspondingly sharp decline had been produced in the stock of money by some other means, the contraction would have been at least equally severe and probably more so. (p. 352, see also p. 357).
I have argued in the main text that the bank failures did indeed have a major effect in accuating the depression, but quite largely by reducing the amount of bank credit available to business - a question that F&S do not consider.
F&S have a curious argument which runs like this. The bank failures provoked a flight from deposits to currency: the former deposits-currency ratio had originally been preferred; hence the flight to cash must have made money less attractive and thus must have reduced the demand for money (that is, deposits and currency). Paradoxically, therefore, the bank failures, by their effect on the demand for money, offset some of the harm they did by their effect on the supply of money. That is why we say that if the same reduction in the stock of money had been produced in some other way, it would probably have involved an even larger fall in income than the catastrophic fall that did occur (p. 353). To be consistent they should also argue that if bank failures had been prevented from causing the money stock to fall (which, as they have already argued, could, and should have been done), then the effect would have been to raise income. The paradox in this argument springs I think from Freidman's practice of treating velocity as if it represented tastes and propensities, and was an independent force. In my view, velocity changes because the banks adjust lending, and thus the stock of money, to changes in income with a lag of several years: only in the very long term does velocity inter alia depend on the demand to hold money.
4/. At the very end of their book there is a passage in which they identify three crucial episodes in US history when the authorities took what is held to be incontrovertibly contractive action. The passage is worth discussing in detail because it is difficult to reconcile with others of their statements, and hence reveals more of their lines of thought: and also becuase it leads me to qualify what I have said already.
After first commenting that "it is often impossible and always difficult to identify accurately the effects of the actions of the monetary authorities, they then proceed:
on three occasions the system deliberately took steps of major magnitude which cannot be regarded as necessary or inevitable economic consequences of necessary changes in money income and prices. The dates are January to June 1920, October 1931, and July 1936 to January 1937. These are the three occasions - and the only three - when the Reserve system engaged in acts of commission that were sharply restrictive: in January 1920 by raising the discount rate from 4.75 per cent to six per cent, and then in June 1920 to 7%, at a time when member banks were borrowing from Reserve banks more than the total of their reserve balances: in October 1931 by raising the rediscount rate from 1.5 per cent to 3.5 per cent within a two week period, at a time when a wave of failures was engulfing commercial banks.... and indebtedness was growing: in July 1936 and January 1937, by announcing the doubling of reserve requirements in three stages, the last effective on May 1 1937, at a time when the Treasury was engaged in gold sterilisation, which was the equivalent of a large-scale restrictive open market operation. There is no other occasion in Federal Reserve history when it has taken explicit restrictive measures of comparable magnitude - we cannot even suggest possible parallels (p 688-9).
It seems that the restrictive action by the authorities in 1931 which F&S have in mind is not high interest rates as such (for the discount rate was low in 1931), but the authorities acting in a way that put further pressure on the banks at a time when, being already short of reserves, they were particularly vulnerable to pressure. This is perhaps the same point as that made in several passages that discuss bank failures in 1929-33, which blame the authorities for allowing bank reserves to fall (see pp 318 and 356).
In looking at any particular critical juncture it must be impossible for a commentator or historian to tell whether a) reserves where adequate because bank lending had fallen first and reduced the need for reserves; or, (b) reserves were at some stage indeed inadequate, and that had forced banks to curtail their lending, and thus to exert a restrictive effect on output. In looking at the whole span of the depression, I have already expressed doubt about whether the banks could have been short of reserves (since, over the whole period, the ratio of reserves to deposits rose). F&S are, however, here speaking of a critical period, in which banks were being drained of reserves by the flight to cash. Banks were therefore particularly dependent on action by the authorities to replenish reserves, and it is quite conceivable that the authorities were insensitive to their needs at this juncture. If that is the criticism, I could see some force in it - though it hardly seems to qualify as what F&S describe as "explicit restrictive measures" of a major sort. But (as I argue in the text) it was in any case perhaps chiefly the nature of existing financial institutional arrangements that was to blame.
Having dealt with vaious particular steps in F&S's particular argument, I will in conclusion give two broad general reasons for discounting a monetarist explanation of the Great Depression.
First, F&S at times admit that there were non-monetary influences at work; but they never explain how these can be integrated into a monetarist explanation. The point is that one must believe that the fall in income had a major effect on personal consumption, and the great fall in output a major influence on business investment, and the interaction of the two (once the process started) a dominating influence. At most, then, monetary influences could have initiated the depression, or worsened it when it started, but they could not have been the sole dominating influence through its course. The parallel fall in the money stock cannot then be taken as evidence of the latter's causal significance.
Second, there is an international dimension which F&S do not face up to which creates a similar difficulty for their argument. Depression in the United States was so deep that it must, be held to have largely determined the coincident depressions in other countries; and in the transmission process non-monetary channels (international trade) must have been important. Now in other countries also there was coincidentally monetary contraction. Does this mean that in other countries the causation ran from non-money to money (the opposite of what is said to have happened in the USA)? Or is it argued that monetary authorities in all countries coincidentally made identical mistakes?
This comes from this link.