The Fed, if it had perfect foresight (about the nominal shock they are experiencing and the amount of cash that will be hoarded by private agents) and had sufficient ability to bypass the banking system to make loans then yes, we could separate the issue of the money supply and a bank failure.

But this isn’t the case. Mass bank failures will lead to a drop in the money supply – and even if they went all out printing money to ensure that the aggregate money supply figure didn’t fall (a printing measure that is mentally inconceivable in the face of mass bank failures) there would be MASSIVE compositional issues, which would come with both output and welfare costs.

And if the argument is regarding whether this is the way Friedman felt, it is useful to not that Friedman himself said:

“We said then and believed then, and I still do, that the Federal Reserve had failed to do what it was originally set up to do. It had permitted a collapse of the monetary system, it had permitted perfectly sound banks to fail by the thousands because of liquidity problems, although it had been set up in 1913 with the objective of preventing that kind of a situation”

http://www.minneapolisfed.org/publications_papers/pub_display.cfm?id=3748

“I am shocked by the confusion in Bruce Barlett’s comment. It seems to be yet another example of what Leland Yeager called “money and credit, still confused.”

The quantity of money includes currency and a variety of bank deposits. Even if deposits fell to zero, the Fed could increase the quantity of currency enough to increase the total quantity of money. For example, suppose that the quantity of currency was $800 billion and the quantity of deposits was $1600 billion. The total money supply would be $2.4 trilion. Suppose the Fed wanted the quantity of money to be $3 trillion. Deposits could fall to zero, and the Fed could create $3 trillion worth of currency. The money supply would then be $3 trillion, with no bank deposits.

Of course, not all the banks failed during the Great Depression. Those that didn’t fail didn’t even increase their reserve deposit ratio to 100%. The money multiplier didn’t fall to 1. And even if they had kept 100% reserves, then the Fed would just need to incease reserves enough to get the quantity of money to the desired level.

Given the actual money multiplier and currency deposit ratio during the Depression, there was always sufficient outstanding government debt for the Fed to use open market operations and raise base money enough to keep the M1 measure of the money supply on its pre-Depression growth path. (Of course, there may have been gold outflows if they had tried, which points to the real problem.)

Still further, increasing the quantity of currency in response to an increase in the demand to hold currency, because people have less trust in banks and want to hold currency rather than deposits is not a “bail out” of the banks.

Failure to meet that demand for currency results in a shortage of currency and broader measures of money, a decrease in nominal expenditure, real output, and prices. The last two things, falling output and prices, hurt bank debtors and can cause banks to fail.

However, if some banks were already going to fail, say small unit banks in midwest communities where drought had devastated the farmers who owed the them money, increasing the quantity of currency to match an increase in the demand for currency would hardly help.

Bernanke’s theory explains how widespread bank failures could cause a decrease in real output even if prices and wages were perfectly flexible so the real volume of expenditures were not impacted by the reduction in nominal expenditures and the nominal quantity of money. Bernanke developed a theory of how bank failures could reduce real output in the context of a new classical model based on continuous market clearing. (That was the style when he was writing.)

The implication of Bernanke’s argument is that even if the Fed expanded the quantity of money enough to maintain nominal expenditure in the economy, the failure of banks could disrupt production. The decrease in the supply of output would result in shortages of goods and inflation. So, bank failures would lead to stagflation.

If Bartlett and Cowen want to appeal to Bernanke to defend Bernanke… well, I guess that is natural. What does that have to do with Milton Friedman?”