A common response made by economists to my previous post on the overwhelming evidence for endogenous money is: ‘so what?’ Economists rarely disagree about the mechanics, but assert that there are no major implications of endogenous money that differ from their theories. This is a mistake. Reality is so complex that seemingly small errors in a model can have a big impact on its conclusions. As I will show, endorsement of the exogenous money theory causes economists to miss some key features of a capitalist economy.
It is worth noting that the names are perhaps misleading, as both theories contain endogenous and exogenous elements. In the case of exogenous money, the central bank expands the supply of base money, and the reaction of banks to this – how much they ‘lend out’ – is in large part endogenous. In the case of endogenous money, banks create loans as they please, except for an exogenous constraint (the interest rate), set by the monetary authority*. Clearly there is a large crossover between the two theories, but there are also subtle differences, which lead to important conclusions.
A point emphasised by endogenous money proponents is the robust correlation between private debt and other key variables such as growth, stocks and house prices:
Obviously correlation doesn’t equal causation, but it is based a theoretical link Keen often mentions – Minsky’s observation that, in order for aggregate demand to increase, planned spending must be greater than current income, and therefore credit must fill the gap. This is a pithy observation, but it seems to me that, contra Keen/Schumpeter/Minsky, the exogenous money model can account for this: the central bank fills the gap by increasing the money stock. This can cause both an increase in debt and an increase in aggregate demand, but, in contrast to the endogenous story, the increase in debt does not cause the increase in aggregate demand.
In other words, the exogenous story is that central bank expands the money supply, and this increases both debt and AD. The endogenous story is that banks expand credit, which expands aggregate demand and forces the central bank to expand the money stock. Thus both are compatible with the correlation between private debt and growth, although the exogenous story would not necessarily have the correlation so tight, or have private debt moving first every time.
In any case, given the other evidence – that credit money expansion precedes base money expansions, that central banks have failed to control base money in the past, and that anyone who actually works in a bank will tell you they make loans independently of the number of reserves they have – endogenous money appears to have the mechanics correct.
So why does this matter? Well, the exogenous story has the causality backwards: it assumes that banks receive reserves and then ‘lend them out,’ whereas what they actually do is make loans and then balance their reserve requirements afterwards. Obviously this means economics textbooks are wrong about the causal mechanics, but economists will likely plead that it doesn’t really matter. However, it matters for a couple of reasons.
The first is stability: if banks have adequate reserves before they make loans, a bad loan will cause problems for the bank that lent them out. But if the banks depend on each other for reserves, and look after the loans have been made, then bad loans can quickly destabilise the entire system as the availability of reserves dries up, triggering a positive feedback loop rather than a return to normality. Thus the system is highly interlinked, and far more vulnerable to systemic crises.
Even more crucially, endogenous theory means that money is effectively created and destroyed by the banking system. This is because debt-based assets and liabilities expand and contract simultaneously as debts are repaid; in other words, the loan-ee is both the saver and the borrower. When the loan is created it is deposited in the recipient’s bank account; as this is paid down both the asset and liability are discharged. Reserves are a secondary consideration and the availability of them does not affect lending decisions. It is true that, in name, the reserves ultimately come from the central bank. But really the expansion of spending power is an endogenous decision, though the central bank influences it via setting the price of reserves.
Simply put, debt does not cancel out at the macro level; it scales up. Just as it is true for a household that income can be scarce relative to debts, it is true for the economy**, which as a whole can find itself over-indebted.
Economists often realise that disequilibrium and finance are important, but their faulty view of the banking sector causes them to miss the mark. Take neoclassical models such as Krugman’s, which argue debt is merely a redistribution from savers to borrowers, and have to add ‘special case’ considerations to make debt matter. But this is misguided – debt always matters, because money enters the economy primarily as new debt, which the economy must expand to service. Hence, debt must go into productive investments, which create future income streams, rather than bidding up the price of assets. I doubt economists such as Krugman would object to the policy implications of this argument, but their models do not imply it is important.
Thus, the name for endogenous money strongly implies its conclusions: the system is easily destabilised endogenously, and the money stock endogenously expands and contracts to accommodate activity (thus rendering the ‘neutrality of money‘ an absurd proposition in any time frame). The differences in the fundamentals are perhaps more subtle than endogenous money proponents make out, but the conclusions are extremely different, and have strong implications for equilibrium analysis, crises and the relationship between finance and the real economy.
*They can also be constrained by regulation, but that is another story.
**This is the only time I will use a household analogy to communicate a point about the economy as a whole.