You say hot potato, he says endogenous

David Glasner finally chimed in on the subject of endogenous money. I pretty much agreed with everything he said on the topic of bank money endogeneity. Basically, the financial system adjusts to a reduced demand  for bank money by destroying that money rather than keeping it in circulation and forcing all prices to adjust. The process by which this occurs is an arbitrage process. This is the classical theory of money that Glasner describes in his book Free Banking and Monetary Reform, and it applies equally to the modern banking system since banks make their deposits convertible into central bank money.

David said something interesting:

So while I think that bank money is endogenous, I don’t believe that the quantity of base money or currency is endogenous in the sense that the central bank is powerless to control the price level.

I was curious about his claims that modern central bank (CB) money is not endogeneous and left some comments on his post trying to drill down on this issue. It seems to me that, much like old-fashioned gold standard CBs and modern private banks, many modern inflation-targeting CBs also have effective convertibility regimes, ie. they are governed by the redemption principle. This regime allows for arbitrage. In other words, David's classical theory of money applies just as well to modern CB money as it does to competitively-supplied banking money and central bank liabilities convertible into gold.

As Nick Rowe points out here (and my post here), modern inflation-targeting central banks including the Bank of Canada, Australian Reserve Bank, and others, are not so different from old-fashioned gold standard CBs, the only difference being that rather than offering convertibility at a fixed rate into gold, modern CBs offer convertibility at  a floating rate into bonds (Nick calls this a CPI standard, I see it as floating rate bond-convertibility, same thing in the end). In between changes to that floating rate, those with access to the CBs "conversion window" - effectively those who can conduct open market operations with the CB - can engage in arbitrage between the external, or secondary market for bonds, and the private price set at the conversion window. This arbitrage mechanism withdraws money from the system or adds to it. Its existence renders modern CB money endogenous (or at least more endogenous than before). Using less exact terminology, once issued, modern (inflation-targeting) CB money never becomes a hot-potato. Increases in the public's demand for CB money draws it out of the system via open market operations while decreases in this demand push that money back into the CB via the same.

That being said, CB money can easily become hot-potato money. Should an inflation-targeting CB foresake its inflation targeting regime and cease offering daily withdrawal/deposit mechanisms (ie open market purchases and sales) priced using some sort of consistent rule, its liabilities have effectively become hot-potato, or exogenous. The modern Federal Reserve, which no longer targets the Federal funds rate by withdrawing/adding to reserves using open market purchases/sales of bonds, is a good example of this. There is no formal process which by the mountain of excess reserve balances might be withdrawn should the demand to hold them collapse, we only have indications from Ben Bernanke that some sort of draining process will occur. Over the last four years, the Fed has surely become a more "hot potato" central bank than the BoC, for instance. Thus David's classical theory surely applies more to the BoC than the Fed, although I'm not sure where he stands on this.

This hot potato vs not issue also came up at Lars Christensen's blog. I pointed out to him that I don't think that the distinction is a core one - its just a matter of how liabilities are structured.

I think we can agree that corporate stock is a hot-potato asset. Once issued, it circulates endlessly. Stocks can also be highly liquid -companies can issue new stock to buy services and many sorts of assets rather than using cash.

But if the corporation agrees to repurchase all its stock at $100 and sell unlimited stock at $105, then that stock is no longer hot-potato. Should the firm issue excess stock to buy services, that stock will quickly be returned for $100.

So much like stock, I’d say that central bank money is not necessarily either exogenous (hot potato) or endogenous. Like the stock example above, it depends on how the asset itself is structured and what options it provides its holders. 

So we're not talking about foundational differences here.

Lastly, while Nick Rowe's post From Gold Standard to CPI Standard is becoming one of my all time favorite Rowe posts, we like it for very different reasons, I think. Nick wanted to use the progression to show that if the old system was reserve-constrained, so is the current one. I'm using it to show that if the old system didn't emit hot potato money, neither does the new one.

Addendum: David Glasner follows up Nick Rowe’s Gold Standard, and Mine