What did Milton Friedman favor?

Responding to my initial post, David Henderson gets himself into a bit of trouble when he writes:

He's [Tyler] correct that Friedman wanted the Fed to increase the money supply. I don't think I'm pretending when I say that I don't think Friedman advocated bailing out banks during the Depression. As I think Friedman would have, last fall I advocated an increase in the money supply while opposing a bailout. Those two, contra Cowen, are separable.

When it comes to 1929-1931, Friedman favored the Fed a) buying up a lot more bonds, and b) serving as a lender of last resort to failing banks. They are separable but Friedman favored both.

In the Monetary History, Friedman and Schwartz approvingly quote Walter Bagehot about the need to do whatever is required, however bold or desperate, to stop a banking panic.  Part of the passage runs like this:

The way in which the panic of 1825 was stopped by advancing money has been described in so broad and graphic a way that the passage has become classical. “We lent it,” said Mr. Harman [one of the Bank’s more senior directors] on behalf of the Bank of England, “by every possible means and in modes we have never adopted before;…

Here is Charles Goodhart quoting Friedman on why the Fed should have been a lender of last resort to troubled banks.  Or see p.269 of the Monetary History, where Friedman and Schwartz explain how it was too difficult for banks to borrow from the Fed at favorable rates in the early 1930s.  Or read this Friedman interview.

In the comments Bruce Bartlett, channeling Friedman, responded:

There's no way the Fed could have expanded the money supply in the early 1930s without bailing out the banks. How do you think the money supply declined in the first place? It's because banks failed and their deposits disappeared. To keep those deposits from disappearing in an era before deposit insurance would have required keeping bankrupt banks afloat.

Friedman's model was not one of allowing a boost in currency to substitute for the broader monetary aggregatesAn article in The Freeman is clear, if perhaps even a bit exaggerated:

Friedman and Schwartz argued that all this was due to the Fed’s failure to carry out its assigned role as the lender of last resort.

You might try to draw a distinction between "lender of last resort" and "bailout" but Bernanke's emergency lending is usually considered part of the bailout package.  No one is suggesting Friedman would have favored each and every part of the bailouts that we have seen.  The point is that Friedman favored some bailouts in the past and probably would have favored some this time around as well.  You don't have to think he would have voted for the first Paulson proposal.

Oddly, Henderson in his post takes offense because I suggest that libertarians try to run away from the idea that Friedman favored Fed action beyond simple monetary policy.  Henderson then tries to run away from the idea that…Friedman favored Fed action beyond simple monetary policy.

I now recall that a related point was made by Paul Krugman, although I find that piece problematic in some other ways. 

Here are a few sounder history of thought claims:

1. Friedman and Schwartz argued that if the Fed had been more on the ball with monetary policy earlier on, the lender of last resort actions would not have been needed. That is distinct from opposing such actions in the time of necessity, when necessity comes.

2. At times Friedman suggested that the rise of deposit insurance limited the importance of the lender of last resort role.  (How he would have thought about rescuing the shadow banking system is an interesting question.)  A related issue hovers here, namely whether support for deposit insurance constitutes support for bailouts.  It seems to me it does, though my original point does not rely on this judgment.

3. Friedman thought that "simple" monetary policy, combined with "simple" Fed lending would go pretty far in stopping a banking panic, yet this view was not borne out in the very recent crisis.  In any case it's wrong to conclude that Friedman was necessarily opposed to more vigorous action, if such action would turn out to be needed.  If you read Friedman as a whole his focus is not on drawing particular lines to circumscribe Fed action, but rather doing whatever is needed to keep the banking system up and running. 

On the broader issue of the bailouts, read Megan McArdle's latest, excerpt: "To a first approximation, I'd say that the bailouts are the reason that we won't have a single-payer health system or actual national automakers any time soon."


'You might try to draw a distinction between "lender of last resort" and "bailout" but Bernanke's emergency lending is usually considered part of the bailout package.'

You absolutely must do this. For instance, the initial specifics of the AIG deal did not constitute a bailout, and I was fine with it. The revised deal does, and it's terrible. To "lend freely at a penalty rate" is not to bailout.

So, where does this tie in with Friedman's argument that the Fed should basically be a computer? In other words, that later in life he apparently favored taking away much of the Fed's discretionary power. To me that seems to sum up what he told Russ Roberts in the EconTalk interviews. Was he being flip or had he changed his mind?

I don't think one can focus solely on the money supply now or in the 1930s because velocity also fell. Since nominal GDP equals the money supply times velocity, a decline in velocity will have exactly the same effect in terms of reducing GDP as a decline in the money supply. But increasing velocity is a lot harder than increasing the money supply.

Friedman and Schwartz mention that velocity declined as well as the money supply in the early 1930s, but they don't emphasize it. The current downturn, however, is almost entirely a function of a decline in velocity. The decline in wealth resulting from collapse of the housing bubble reduced spending very significantly, thus reducing velocity. Those focusing only on the money supply have been looking at only half the equation.

When velocity falls by a small amount the Fed can compensate by raising the money supply. But when velocity falls a lot, as it has over the past year, the Fed can't increase the money supply fast enough to prevent a decline in nominal GDP, which will result in both falling prices and reduced output. That is why an expansionary fiscal policy is essential to compliment monetary policy. Otherwise the Fed is pushing on a string.

Perhaps a good test of how great an economist is is whether people ask what would s/he do after they are dead.

I think this is a pretty good post, and it highlights Friedman quotes I've read to the effect that von Mises was "unreasonable" when discussing fiscal policy and started screaming that all other economists were socialists.

So I guess the only difference between a New Classicist and a New Keynesian is at what point one decides printing money is "necessary."

As for Greenspan's argument from the 1960s (okay, okay - we all know that it was really Ayn Rand's argument articulated via Greenspan) that capital markets would provide stern enough checks on the financial system to prevent a widespread collapse if we put an end to our seemingly unanimous decision to constantly increase the money supply, I guess no one's really interested in thinking about that anymore.

But isn't it sad that so many people argue so fervently about when to print money to avoid a collapose without ever considering that these crises need never occur in the first place. Truly, capitalism is the only economic system without a philosophical basis. :(

"If you read Friedman as a whole his focus is not on drawing particular lines to circumscribe Fed action, but rather doing whatever is needed to keep the banking system up and running."

I think there is an important point for discussion here. Is "saving banks" synonymous with "saving the banking system"? During the Great Depression, bailing out individual banks may have been crucial for saving the banking system. We know that an entire generation lost its confidence in the banking system and permanently reduced their demand for banking services, disrupting the flow of credit and capital.

I think this key question should be asked: were the recent bailouts key to restoring (confidence and demand for) the banking system, or simply large banks?

The Fed went on and on about confidence, suggesting that the answer was yes, and it was crucial for restoring the banking SYSTEM in 2008-9. Were banks seriously in danger of a permanently lower demand for their services? If not, why isn't that simply resolved by the Fed's lender of last resort function (designed for short term losses in confidence)?

More pointedly, if there were no bailouts, and some major banks failed, do we have reason to believe that new and existing banks wouldn't have expanded to supply banking services (given that demand for those services may have survived the current recession)?

Bernanke has answered many of these questions in this talk:


"Following the Lehman collapse, panic gripped the money market mutual funds and the commercial paper market, as I have discussed. More generally, during the crisis runs of uninsured creditors have created severe funding problems for a number of financial firms. In some cases, runs by creditors were augmented by other types of "runs"--for example, by prime brokerage customers of investment banks concerned about the funds they held in margin accounts. Overall, the role played by panic helps to explain the remarkably sharp and sudden intensification of the financial crisis last fall, its rapid global spread, and the fact that the abrupt deterioration in financial conditions was largely unforecasted by standard market indicators.

The view that the financial crisis had elements of a classic panic, particularly during its most intense phases, has helped to motivate a number of the Federal Reserve's policy actions.19 Bagehot instructed central banks--the only institutions that have the power to increase the aggregate liquidity in the system--to respond to panics by lending freely against sound collateral.20 Following that advice, from the beginning of the crisis the Fed (like other central banks) has provided large amounts of short-term liquidity to financial institutions."

He believes that he was following Bagehot:

"The case of the investment bank Lehman Brothers proved exceptionally difficult, however. Concerted government attempts to find a buyer for the company or to develop an industry solution proved unavailing, and the company's available collateral fell well short of the amount needed to secure a Federal Reserve loan of sufficient size to meet its funding needs."

Time will tell if the govt's actions work out:


For those of us who favored saving Lehman, here's why:


"The 'shadow banking system' at the heart of the current credit crisis is, in fact, a real banking system – and is vulnerable to a banking panic. Indeed, the events starting in August 2007 are a banking panic. A banking panic is a systemic event because the banking system cannot honor its obligations and is insolvent. Unlike the historical banking panics of the 19th and early 20th centuries, the current banking panic is a wholesale panic, not a retail panic. In the earlier episodes, depositors ran to their banks and demanded cash in exchange for their checking accounts. Unable to meet those demands, the banking system became insolvent. The current panic involved financial firms 'running' on other financial firms by not renewing sale and repurchase agreements (repo) or increasing the repo margin ('haircut'), forcing massive deleveraging, and resulting in the banking system being insolvent. The earlier episodes have many features in common with the current crisis, and examination of history can help understand the current situation and guide thoughts about reform of bank regulation. New regulation can facilitate the functioning of the shadow banking system, making it less vulnerable to panic."

In the past, a bank run ( Liquidity Crisis )was stopped by Deposit Insurance. This is MF's view ( Essence of Friedman, p.515 ). The question then was whether or not we had a Collateral Run ( Liquidity Crisis ) in the Shadow-Banking sector after Lehman. If you think so, then , according to MF, only a govt guarantee can stop this panic ( EF p. 515. ). Since there was no Deposit Insurance to stem this panic, the govt needed to step in and provide a kind of Deposit Insurance to stop this run. Bagehot's main point is not the particular rules, but that a panic should be stopped. His entire approach was pragmatic. Also, for Bagehot, expectations are key, as evidenced by his explanation for why the B of E could not be closed down.

This post makes the case:


"Bagehot wrote “in a panic, the holders of the ultimate Bank reserve should lend to all that bring good securities quickly, freely, and readily.† To him, “good securities† included those that, while easily traded in normal times,may have no market value during a panic."

Finally, it is interesting in considering what might cause a financial crisis, and what new measures should be taken if one occurs, MF does not mention Deflation ( EF p.424 ). In that essay, he gives various proposals about what to do should a crisis occur. Some commenters have talked about about a specific Fed regime proposed by MF. I don't see how this would easily work in the present crisis. In order for it to work, it must be credible. This was so important that MF calls for a Constitutional Amendment to make it work ( EF p. 414 ).

Since I'm not an economist and am always willing to be instructed, and MF is very important to my views, it would help me if people could give references to his works where I can look up their points.

But Tyler, you ignored the main point of David's post--reflected in his title, his opening sentence, and his closing paragraph that begins, "But here's the bigger question..."

You accused the libertarians who disagree with you of "pretending" that Milton Friedman said the problem was that the Fed let the money supply collapse. And so David asked (I'm now paraphrasing), "Has anyone ever seen Tyler accuse leftist statists of pretense in the same way?"

I don't remember you doing that. If you have, I apologize for the accusation of a double standard.

The most telling fact indicative of the failings in Friedman's understanding of the money/credit/production nexus is that Friedman as late as 2006 thought that there were no problems in the macro economy and that Greenspan / Bernanke were making no mistakes with money/credit/production policy.

Friedman was the Irving Fisher of the 2000s.

I could picture Friedman saying to himself "I'm one libertarian in a room of 99 Keynesians. I better push hard for the most libertarian option available."

It's harder for me to picture Friedman saying to himself "This is probably a good idea because Milton Friedman promoted it."

Excerpt from John Hussman's commentary is interesting:
"Solving economic problems, to our Fed Chairman, is as easy as throwing money out of helicopters. Not surprisingly, throwing money out of helicopters has been the basic core of his strategy during this crisis. This does not involve complex thought about debt restructuring, moral hazard, incentives, equitable distribution of resources, or other factors. All it requires is the three second tape playing in Bernanke's head - "We let the banks fail in the Great Depression, and look what happened." And then the tape repeats. Never mind that the cause of the upheaval was not the failure of banks per se, but the disorganized Lehman-style failure of banks. The tape isn't long enough to encompass such nuances."

"When velocity falls by a small amount the Fed can compensate by raising the money supply. But when velocity falls a lot, as it has over the past year, the Fed can't increase the money supply fast enough to prevent a decline in nominal GDP, which will result in both falling prices and reduced output. That is why an expansionary fiscal policy is essential to compliment monetary policy. Otherwise the Fed is pushing on a string."

This intellectually obtuse explanation seems to be a crummy excuse to rob trillions of dollars from the average taxpayer and hand it over to the politically connected.

Obviously this "expansionary stimulus" does not need to be channelled through Goldman Sachs, JP Morgan, GE, GM and Lockheed Martin...yet it is. To pretend it isn't and to pretend that this is preferable over letting Goldman Sach fail is absurd. A continuation of these policies will lead to violence and people like Bartlett will be morally responsible in my mind.

The government should have taken the banks into FDIC receivership and sold them off over time.

But what you fail to take into account is the libertarian solution of getting government out of the business of deposit insurance. Instead of FDIC insuring regulated deposits, we had AIG and Lehman and other unregulated banks playing at the game of FDIC.

So, in the fall, we had the libertarian shadow FDIC's failing as the libertarian shadow banks failed causing a run on the shadow banking demand deposits of all those commercial money market funds and the mutual money market funds.

How much money was in the shadow banking sector and how underfunded were the shadow FDICs and how much money did the shadow Federal Reserve have to be the shadow lender of last resort? And what was the shadow Fed - it seems like whoever it it was, they ran for the hills at the first sign of trouble.

In "The Paradox of Money", Friedmann said that in the 60s when both "Capitalism and Freedom" and the" Monetary History" were written he still believed in the Fed because he has not read yet the works by Buchannan and Public Choice school

Holy cow! Tyler just admitted he was a Nazi. I knew it! (j/k)

I'm not an economist, so I'll shut up. But the comments on this thread had been very interested to me. Thanks.

Tyler, there are two different issues here. Friedman would have said:
First, systemic panics are harmful and the TARP stopped them.
Second, the free market way of stabilizing the insolvent banks is partial conversion of bonds to equity, public money is not needed at all.

Friedman would have wanted a way to bankrupt the banks without causing mass hysteria. If that seems like a contradiction in terms, I suggest you read some of Luigi Zingales' work over the past year.

Perhaps there's a distinction to be made between providing liquidity and providing capital, that is, taking ownership.


Friedman advocated an M2 growth rule for decades. He also advocated using open market operations to hit that target. That implies that if the currency deposit ratio changes, the composition of the quantity of money will shift between deposits and currency. If he really was concerned with controlling the quantity of deposits, then he wouldn't have favored targetting M2. He would have favored targetting something else, maybe M2 less currency held by the public. So the total of checkable deposits, savings accounts, and CD's less than 100k would be put on a target growth path.

If Friedman really did think that it was bank lending that was important, then we are left with the puzzle as to why he supported closing the discount window. Shouldn't he have advocated having it wide open so that changes in the currency deposit ratio would be accomodated by advances to banks, so that the banks could continue to fund the same amount of loans?

Maybe you forgot, but there is a concept of "lender of last resort" to the banking system. It has nothing to do with lending money to particular banks, much less insolvent ones. Quote Friedman (and any number of other monetarists) all you want about the Fed failing to act as lender of last resort, but you prove nothing about support for bailouts. The orthodox monetarist position was always lender of last resort to the banking system and not particular banks.

Further, Friedman claimed that the reason banks were insolvent during the Depression was the reduction in output and deflation caused by the decrease in the quantity of money. If the Fed had followed the right policy, the banks would not have been insolvent. That view is going to create all sorts of quotes that will suggest bailouts of the banks. The Fed should have acted in a way that prevented the failure of banks. But that doesn't mean that the Fed should have bailed out banks that would have been insolvent if the quantity of money had remained growing on target.

If all the banks were really insolvent (conditional on nominal income continuing to grow on target,) because they all made bad loans, then closing them all down at once and selling off all their asseta and paying off deposits with the proceeds, and letting people who want to organize new banks do so, is probably not a good policy. I don't think this has anything to do with the Great Depression, and trying to guess what Friedman would favor in the face of general insolvency despite a proper monetary policy by looking at what he said about the Depression is just wrongheaded. One would hope he would have a sensible view.

I certainly agree that monetary policy over the last year could have been worse. And I think there are some libertarians who advocate policies that would have been worse. Freezing base money and liquidating all the insolvent banks (conditional on a big drop in nominal income,) and then letting people start new banks if they wanted, would have been much worse, in my view.

On the other hand, expanding base money even more so that nominal income continued on its previous growth path, and keeping FDIC for the time being so that involvent banks remain open until the FDIC can reorganize them, seems like a better policy to me. Having the conventional banking system expand to replace the failed shadow banking system seems like the better policy to me. I think having the Fed pay interest on reserves balances and target loans in ways that look to be aimed at jump starting securitization markets is an inferior policy.

In my view, the key element of the "bailout" was the Treasury buying stock in banks. I opposed that as much as I opposed proposals to have the Treasury buy bad assets from banks. I have no problem with the Fed lending money to solvent banks. I don't really care about collateral. I don't even care if the Fed lends to insolvent banks, but only as part of an FDIC reorganization process that involves wiping out the stockholders. Generally, I think other bank creditors should take a loss. I think the reality is that debt with longer terms to maturity should be targetted for loss. The key is to make it clear who is safe and who is not. But even if they all are subject to loss, isn't it obvious that it just means that insured deposits would play a bigger role in funding reorganized banks?

It seems like the monetarist/Keynesian nexus is that the money illusion only works on the way down, and like a trained tiger, only works when the feds are holding the whip.

Milton Friedman very clearly supported bailing out the banking system. In fact, he criticized the Depression-era Fed for not doing so. Just watch his PBS documentary "Free to Choose".

Comments for this post are closed