# The Bank of England on money creation

Money creation makes people go nuts. Everyone can verify that claim by searching youtube for “money creation” or “money multiplier”, which will turn out a long list of clips telling you the SHOCKING TRUTH about our FRACTIONAL RESERVE BANKING system. Interestingly the money creation craze can be found on both ends of the political spectrum. To socialists, the fact that banks create money out of thin air proves that they are the ultimate force of evil in the world, designed to enslave the working class in a never-ending spiral of debt and compound interest. Libertarians, on the other hand, go “Money out of thin air! Inflation! Theft!”.

Introductory economics textbooks are partly to blame for this confusion. Most of them present an oversimplified model of money creation that goes something like this: there is an initial deposit of \$100 with bank A. Bank A keeps 10% (say) as reserves and lends out the remaining \$90. This new loan ends up as a deposit with bank B. Bank B keeps 10% of it as reserves and lends out \$81, which will end up in bank C, and so on. By the magic of infinite series, the initial deposit of \$100 creates new money in the amount of \$1000; it gets multiplied by a factor equal to the inverse of the reserve ratio.

The reason econ educators like this model is twofold. First, it is an opportunity to employ a neat mathematical result to a “real-world” question. Second, it has the convenient implication for macroeconomic theory that the central bank can directly control the amount of broad money in circulation (the amount of bank deposits), by changing the amount of reserves.

Now the Bank of England is out with an article debunking this simple model. It makes four points that can get lost in many textbook treatments of the subject:

1. The overwhelming share (97%) of the broad money supply is created by private banks and takes the form of bank deposits. That is, (broad) money is an interest bearing asset.
2. An increase in reserves is the result rather than the cause of increased lending. When a bank makes a new loan, it immediately creates a new deposit in the borrower’s bank account. Since the borrower will want to withdraw that money at some point, the bank will want to match up the new loan by new reserves, which it acquires either by attracting new deposits or by borrowing from other banks. So loans create (demand for) reserves, not the other way around.
3. The primary limit to money creation is the profit motif of commercial banks. The profit of a bank from making a new loan depends positively on the interest rate it earns on loans and negatively on the interest rate it pays on deposits. Increased lending tends to decrease the interest rate on loans and leads to an increased demand for reserves which tends to increase the interest rate on deposits. Profit maximizing banks will lend until the marginal loan earns just enough to cover the costs of making it.
4. The central bank influences the broad money supply by setting interest rates, not by setting the amount of reserves. Lowering the interest rate for which banks can get central bank reserves lowers the cost of making new loans, which (as explained above) leads to new loans and thus to an increase in broad money.

Some econ commentators immediately seized upon the article as yet another proof that mainstream economics is all wrong. For David Graeber, the article is a vindication of “the kind of populist, heterodox positions more ordinarily associated with groups such as Occupy Wall Street”, and that it has “effectively thrown the entire theoretical basis for austerity out of the window“.

As Simon Wren-Lewis says, this is all nonsense. First of all, the simplistic money multiplier has been abandoned by “mainstream” economists at least since James Tobin’s 1963 paper. The only real puzzle here is why the simple model is still taught to undergrads – more on that below. Second, there is a difference between “banks create money out of thin air”, which is true, and “banks can create money without limit”, which isn’t, that seems awfully difficult to grasp for the Occupy crowd and economists of the libertarian/Austrian persuasion. Third, if the BoE article has thrown something out of the window then it is the notion, widely held by the money creation nuts, that banking is exempt from the ordinary laws of economics and, worse, that it is some kind of dark secret that academic economists try to hide from the unsuspecting public.

Now as for the question why the money multiplier is still found in intro textbooks, Wren-Lewis has some interesting things to say in this old blog post. But I think he doesn’t quite get to the crucial point, which is that the money multiplier story, although clearly wrong as a literal description of money creation, can still be useful as a quick-and-dirty approach to the money supply. As long as commercial banks’ aggregate demand for reserves is a reasonably stable fraction of broad money, the money multiplier gives the right answers. With a stable demand function for reserves, it doesn’t matter whether we model monetary policy as adjusting the price of reserves (interest rates) or the quantity of reserves. A given change in the interest rate corresponds to a change in reserves and vice versa. Of course, the demand for reserves is not a universal constant, but changes with economic conditions. In particular, the reserve ratio tends to go up in times of financial turmoil such as the last financial crisis. But this does not invalidate the whole concept.

So should we kill the money multiplier as a textbook model of the money supply? I think this is a question of balancing costs and benefits. Clearly, the beauty of the multiplier is its simplicity. But this simplicity comes at the cost of purveying a misleading image of banks mechanically lending out whatever deposits they have on their books. The amount of confusion surrounding money creation evident on youtube channels and in the blogosphere seems to indicate that the costs outweigh the benefits.