By these means they can get along and do their business with an amount of notes having an aggregate real value substantially less than before. For example, the notes in circulation become worth altogether $20,000,000 instead of $36,000,000, with the result that the next inflationary levy by the Government, falling on a smaller amount, must be at a greater rate in order to yield a given sum. When the public take alarm faster than they can change their habits, and, in their efforts to avoid loss, run down the amount of real resources, which they hold in the form of money, _below_ the working minimum, seeking to supply their daily needs for cash by borrowing, they get penalised, as in Germany in 1923, by prodigious rates of money-interest. The rates rise, as we have seen in the previous chapter, until the rate of interest on money equals or exceeds the anticipated rate of the depreciation of money. Indeed it is always likely, when money is rapidly depreciating, that there will be recurrent periods of scarcity of currency, because the public, in their anxiety not to hold too much money, will fail to provide themselves even with the minimum which they will require in practice. Whilst economists have sometimes described these phenomena in terms of an increase in the velocity of circulation due to loss of confidence in the currency; nevertheless there are not, I think, many passages in economic literature where the matter is clearly analysed. Professor Cannan’s article on “The Application of the Apparatus of Supply and Demand to Units of Currency” (_Economic Journal_, December 1921) is one of the most noteworthy. He points out that the common assumption that “the elasticity of demand for money is unity” is equivalent to the assertion that a mere variation in the quantity of money does not affect the willingness and habits of the public as holders of purchasing power in that form. But in extreme cases this assumption does not hold; for if it did, there would be no limit to the sums which the Government could extract from the public by means of inflation. It is, therefore, unsafe to assume that the elasticity of demand is necessarily unity. Professor Lehfeldt followed this up in a subsequent issue of the _Economic Journal_ (December 1922) by a calculation of the actual elasticity of demand for money in some recent instances. He found that between July 1920 and April 1922, the elasticity of demand for money fell to an average of about ·73 in Austria, ·67 in Poland, and ·5 in Germany. Thus in the last stages of inflation the prodigious increase in the velocity of circulation may have as much, or more, effect in raising prices and depreciating the exchanges than the increase in the volume of notes. The note-issuing authorities often cry out against what they regard as the unfair and anomalous fact of the notes falling in value _more_ than in proportion to their increased volume. Yet it is nothing of the kind; it is merely the result of the