The Real Bills Doctrine
Summary of the Real Bills Doctrine
Inflation is caused by the rate of spending exceeding the capacity of production at current prices. In order to have a consistent theory of why inflation happens, you must have a model of how changes in the value, types and quantity of assets affect the rate of spending. Under a gold standard, the nominal price of gold (or any other arbitrary commodity) is fixed and banks must convert their paper money to gold at a specified exchange rate. Central banking versions of the gold standard limit the central bank to producing only so much money as it has gold to back the currency. By fixing the amount of money, the gold standard is designed to keep the rate of spending and hence inflation under control. The underlying theory is that the amount of money determines the amount of spending. The Real Bills Doctrine is similar to the gold standard, except that it states that increases in the money supply will not be inflationary as long the central bank backs them with holding of any good assets, not just gold or some other commodity. So long as the central bank increases its holdings of good assets at the same rate as it increases the amount of money, the increase in the money supply will not be inflationary.
Conventional monetary policy has no impact on the economy under the Real Bills Doctrine. The central bank increases the money supply but simultaneously increases the value of the assets they hold to back the money. Conventional fiscal policy cannot affect the rate of spending either under this model. Government spending financed by taxation simply moves spending from the private sector to the public and government spending financed by bond issue affects neither the money supply nor the assets held by the central bank. If the model treats government bonds as equivalent to paper money, then perhaps fiscal policy could stimulate the economy, but bonds are not net wealth since they finance themselves through taxation and not increases in capital. Only fiscal policy financed by printing money could have an impact. If a government desired inflation, they would either have to print money without increasing the central bank’s assets or order the central bank to destroy assets they had. If a central bank held a government bond to maturity and did not ask the Treasury to repay them with taxation, that would be equivalent to a helicopter drop and would cause inflation.
Professor Sproul implies throughout his writings that a change in the money supply is inflationary iff the nominal price of the backing of the currency changes. However, inflation is not caused by a change in the price of a paper dollar relative to gold or silver, but rather a change in value relative to goods purchased. For example, during the early years of the Great Depression, the nominal value of gold remained at $20.67 per troy ounce and yet prices fell in the neighborhood of 30%. When FDR devalued the dollar relative to gold and increased the central bank’s assets, the price level rose.
As Irving Fisher pointed out in his classic article “The Debt-Deflation Theory of Great Depressions“, when one bank sells its assets, that drives down the price of those assets, pushing other banks toward insolvency. Even if a central bank was fully solvent based on current market prices of the capital it holds, it could be rendered insolvent by downward movements of the price of its’ capital. Furthermore, such a movement would be deflationary, not inflationary as the Real Bills Doctrine would suggest. As prices of capital fell, the assets of household’s balance sheet would decline causing them to reduce spending to rebuild them. Only when prices adjusted downwards to restore real money balances would equilibrium be restored. Under the Real Bills Doctrine, even small fluctuations of the central bank’s balance sheet could cause swings in the price level (see the section under Inflation). In Sproul’s example, a fall of 20% in the value of the central bank’s capital value will cause a drop in the value of money by 50%. Real world central banks can be even more highly leveraged than the example, so this implies that relative price shifts in capital can be highly disruptive.
Like the Post-Keynesian school of thought, the underlying theory is that the net assets held by the public determines the rate of spending; the composition of assets held does not. Unless people will sit on money just as they would sit on the capital they sell to the central bank, the rate of spending will increase and thus inflation will occur. Illiquid assets produce lower spending than liquid assets. Depreciating assets produce high spending levels than assets which are dropping in value. When a central bank performs an open market operation, it is exchanging an asset which is illiquid and appreciating for one that is liquid and depreciating. Therefore, I disagree with the Real Bills Doctrine in this regard; I believe that the real money balance savings rate is highly dependent on the expected return from holding money compared to alternative investments.
Do Fiat monies Exist?
I agree with Professor Sproul’s assessment that temporary suspension of convertibility does not affect the source of value of an eventually-convertible currency. However, in order for a suspension of convertibility not to affect the value of the currency, the central bank must credibly promise to eventually convert at such a rate as to compensate the holders of currency for their forgone interest income. In other words, they must promise to pay implicit interest when they return the currency to convertibility. In America, at least, no such promise exists. Few expect the Fed to convert dollars to gold, silver or even Twinkies at any point during our life spans. Even if they did, no exchange rate has been specified and investors have no guarantee that they will earn any interest at all on their dollar holdings.
Professor Sproul does not deny that central banks which do not offer to convert paper money to a physical asset or claim on capital exist. Throughout the majority of the developed world, the only assets which central banks hold are government bonds, which are themselves promises to pay fiat currency. Sproul states that in that case, paper money is instead backed by taxation. I think that most economists would agree that one source of value for government issued paper money is its ability to be used to pay taxes. I do not think, however, that this means fiat money does not exist. It’s a definitional argument: is paper money backed only by taxation “fiat”? Some would say yes, some no, but in the end, there is no way to resolve the issue other than coming up with a definition of what “fiat” means.
On page 11 of his article “There’s No Such Thing as Fiat Money“, Sproul states that if fiat monies existed, central banks could compete with each other and drive down the profit on issuing currency to 0. So long as issuers of currency are able to establish trademarks on their currencies, others will have no ability to undercut them unless they offer some other sort of advantage. The Mexican central bank cannot issue dollars, they can only issue pesos. Those pesos are not able to settle U.S. taxation debt, U.S. firms are not allowed to pay their workers in pesos, etc. Each government uses its monopoly on force to keep competitors at bay and so in the most extreme circumstances are outside fiat monies able to displace the resident government’s fiat currency, such as during a hyperinflation. Producing a good that people value is not free lunch. Central banks do produce seigniorage revenue for their governments, in fact, that’s one of the main reasons why they exist. The 0 profit condition only exists in perfectly competitive markets and currency is one of the least competitive markets there is.
It’s important through all of this that a model is only a map of reality. Any model both illustrates some aspect of reality while simplifying reality. It is impossible for a model to be true or false, it can only be useful or not useful, which in turn depends on the particular situation. Institutions are social constructs, and money especially so. Money has value because people believe it to have value. The coordination point is maintained by the monopoly on force exercised by governments and by mutually compatible incentives. So long as people believe fiat currency does not require a backing, nor do they look at the central bank’s balance sheet to determine how much their money is worth, backing is unnecessary and irrelevant.