David Andolfatto recently constructed a model intended to ‘reconcile’ (or I think more accurately to distinguish between) ‘mainstream’ and ‘heterodox’ views of the macroeconomic importance of money and banking. More specifically, he wants to answer the question: does the ability of banks to ‘create money’ when they issue loans give bank lending a greater impact on aggregate demand than other forms of leverage? His conclusion is that in general it does not. I believe his model is wrongly specified to determine this question – it essentially pre-decides by its structure. To demonstrate this, I have identified the features that he omits from his model and attempted to incorporate them within the structure of the model he himself uses. My full response is in the form of a downloadable pdf.
In various aspects the model fails to describe an actually-existing monetary economy. First, David fails to include relevant empirical regularities by, for example, centralising the banking process, allowing taxes to be paid out of the total income of ‘savers’ (rather than only that part of their income that is received in the form of money), and allowing important transactions essentially to be carried out by barter. He thus fails to characterise the most important features that make money ‘special’. Secondly, the model leaves out capitalism. In the model the ‘investors’ do not, as capitalists do, acquire labour to make a monetary profit from their production, but instead simply consume the residual of their own output. Thirdly, he does not account for true uncertainty and how this is countered by liquidity preference.
I attempt to add the following to the structure of David’s model:
- All money, whether state or bank originated, is uniform in its relation to the real economy.
- Money movement follows ‘loanable funds’ – an aggregate increase in funding leads to an aggregate increase in money, or at least in its flow.
- Expectations and their realisation or otherwise are more clearly modelled.
- Workers and Investors do not always exhaust their purchasing power – with the consequence that bank loans may not be repaid, reducing banks’ capacity to issue new loans.
Under my adjustments to David’s model, banks and bank leverage are special, particularly because their role in the economy means that a failure of expectations can have an impact beyond the immediate period. This impact only becomes manifest in the model because I allow for liquidity preference, in that potential consumers opt to hold money rather than spend it. This requires a recognition that the future is truly uncertain and so there is an individual advantage in maintaining liquidity. The adjusted model also emphasises that bank money issue, as well as price expectations and output expectation failure, may all increase the price level, although these depend on monetary velocity assumptions.
The pdf also has two appendices with greater detail on The Nature of Financial Liabilities and Why Money is Special and some comments on the role of ‘Shadow Banking’.