Barkley Rosser wrote: "James A., I find your claim that the money supply must be cash to be, well, idiosyncratic. If we evolve to a world without cash where everybody uses debit cards, would you deny that the demand deposits being drawn down by the use of those debit cards is the money supply? " First, I was pointing out how one would employ classical theory (the quantity theory of money and Say's law) to interpret the Great Depression. The classics meant by "money" cash. All other media of exchange they called "money substitutes." I have compiled the classical writers' statements of this (from David Hume to J.S. Mill) in chapter 2 of my _Classical Macroeconomics_ (pp. 30-40; now available in paperback for only $37.50 at www.routledge.com/paperbacksdirect). See also Marshall's _Money, Credit and Commerce_ (1923). Even Keynes (1923) defines money as cash only. Evidently, after reading Fisher, he decided to extend the definition of money beyond where Fisher stopped (the equivalent of modern M1) to include even unspent "overdraft facilities" with banks (Keynes 1930, 1:236), and sometimes to include treasury bills (1936). But one easily gets into trouble (confused) trying to understand the classics when one is not paying close attention to the definitions they used and one reads them with Keynes's new definitions instead. This was the main problem with Keynes himself, and continues to plague Keynesians stuck on his definitions. As Jacob Viner notes in his 1936 review of the _GT_, "The book ... breaks with traditional modes of approach to ... problems. ... no old term for an old concept is used when a new one can be coined, and if old terms are used /new meanings/ are generally assigned to them. The definitions provided, moreover, are sometimes of unbelievable complexity. The old-fashioned economist must, therefore, struggle not only with new ideas and new methods of manipulating them, but also with a /new language/" to make meaning of Keynes's claims (p. 147; my Italics). Now, I wouldn't call demand deposits "the money supply," as Rosser suggests. I include them among people's wealth or financial assets or savings. I also get analytical clarity with that definition. An increase in the demand for such deposits (all else constant) will lower interest rates, not raise them as modern macroeconomists using Keynes's definition of money would like to infer from their "money" supply and "money" demand theory of interest. Would we ever get to a world without cash? I doubt it. There will always be the need for cash. Not all of us would like to leave traces of our transactions (for possible government audit). There also could be electricity failures making the use of debit cards impossible. Imagining giving my grandchild, nephew, or niece a small "cash" gift by exchanging my debit card with him or her, I'd leave to science fiction. Anyhow, I'm glad to share my idiosyncrasy on this issue with the likes of R. Glenn Hubbard (2005, 20). Besides, central banks will never tire of the seigniorage from printing money, would they ever? Rosser also asks: "do you deny that if people get worried about their status or businesspeople lose their optimism and become afraid of investing at any interest rates, that people cannot let their demand deposits pile up, even in a world of no cash?" I don't know what "people [getting] worried about their status" means. But if they don't trust anyone else temporarily to borrow their unspent income (which demand deposit accounts really amount to), they would hoard their incomes in whatever form they are can. True enough, if businesspeople lose their optimism, they would *reduce* their demand for funds to borrow. But I cannot conceive of an economy in which there is a zero demand for credit (savings) or borrowing. Someone always wants to spend more than they have the present means to attain that need. Rosser also writes: "Regarding your argument from Mill that there cannot be a general glut because surpluses of one commodity will be offset by deficits of another one, well, this is simply the claim that not only does supply create income equal to itself, but that aggregate supply always equals aggregate demand. But, as I have noted above, this is simply empirically false. We do see fluctuations of inventories, and a rise in aggregate inventories has long been one of the leading economic indicators that is considered to forecast the possibility of a downturn in economic activity, a recession, indeed one due to a 'general glut.' " Here, I simply urge Rosser to re-read my Mill quotes. If he wants additional quotes, including those from Say, James Mill, and Ricardo, he could read my chapter in Steve Kates 2003 edited book, "Say's Law: Keynes's Success with it Misrepresentation.." I venture to mention it again, in spite of Rosser's earlier sneer at my mention of Kates's edited book, because he seems comfortable dealing with Kates in this exchange. In sum, the classical argument is that there could be unsold goods, but that would be caused by an unsatisfied demand for money (cash), e.g., Mill (3: 574). Thus, "In order to render the argument for the impossibility of an excess of /all commodities/ applicable to the case in which a circulating medium is employed, /money must itself be considered as a commodity/. It must, undoubtedly, be admitted that there cannot be an excess of all other commodities, and an excess of money at the same time" (Mill 1874, 71; emphasis added). This is what some analysts, like Leijonhufvud, term the zero excess demand for goods and money in the aggregate. Believe it or not, I find repeating myself many times tiresome. So I'd leave Rosser to hold on to his Keynesian confusions, if that is his pleasure. James Ahiakpor