In the next few days Ill be posting a series on the Keynesian multiplier in Monetary Circuit Theory but first it is necessary to clear up a misunderstanding about the investment=savings identity. This confusion famously led Basil Moore to claim that there was no Keynsian multiplier as it violated the investment=savings identity. Similar arguments have been made by other authors such as Wyne Godley. The treatment of the ‘identity’ is debated by authors such as Grasner and Sumners but to my mind the foundational problem has not been tackled.
The basis of this identity is the national income accounts where assuming a simple closed economy with no government sector and no imports /exports
Y=national income, C=consumption and I=investment.
From this if one deducts consumption from national income Y-C one is left with a residual. Using the Keynesian definition of savings as unconsumed income then one is left with a savings=investment ‘identity. Note also that the accounting convention is that unspent inventory is treated as investment.
The problem with this approach is that it is stock flow incoherent, and assuming it to be coherent can make some treatments of the Keynesian multiplier stock-flow incoherent as well.
National flow of funds accounts are flows, but unspent income ‘hoarded’ money is a stock.
There is also the hidden assumption that investment is funded from a fund of savings, a ‘loanable fund’, when we know that most is funded through endogenous credit creation.
Even in its own terms the ‘identity’ assumes a static economy. Imagine an economy where all goods depreciate at a constant rate over their lifetime and replacement goods are just as efficient as their predecessors’. In such economy if savings were made at a steady rate to replace capital goods then there would be no growth as at each point in time with depreciation and debt amortization investment=savings. A steady state non-growth world.
A more coherent approach needs different starting assumptions.
1) Rather than two assumed states of money, consumption or investment, there are four, expand money balance, consumption, investment, extend credit. From this we can distinguish between desire to hold money (after Wray) from liquidity reference (after Keynes). The desire to consume and invest drives the demand for money, if this exceeds current income credit is taken out. The desire to hold cash being a residual after the desire to hold money and expenditure.
2) Investment is primarily through endogenous money creation by financial institutions. This money creation through a change in debt expands aggregate demand and aggregate supply.
3) Over the term of the loan the financial institution will need to fund it – that is find ex-poste savings for the investment. Over the term of the loan there will be equal monetary contraction through debt amortization and savings as there was monetary expansion at the beginning of the term. (This is essentially Hawtrys theory of the Credit Cycle/Impulse)
4) This of course assumes a broad equilibrium position of banks remaining profitable and being able to attract funds over their discount window. If they are not able to do so then they may change their interest rates to either attract more savers or borrowers or lever/de-lever to maintain their deserved leverage/reserve ratio. As well as deposit funding financial institutions may also attract savings in the form of equity.
Over a particular period of time T0-T1 there will be a flow of funds in successful bank loans so that investment=savings. But what of bad loans, excessive unsold inventory, malinvestments etc? In double entry terms these items only ‘balance’ because of write offs to ensure they balance over the term.
In dynamic terms their is absolutely no investment=savings identity. Rather for an investment over term X their is a commitment from the institution making the investment to rebuild its balance sheet by attracting savings and receiving profits so that a profit is made on the loan. In other words the return on the investment must be greater than the savings attracted for the loan to be made. It is a committment to a future flow to rebalance the stock that has been created.
5) All of this is just another way of expressing the M-C-M’ approach of Marx in refuting Says law. Capitalist growth is profit orientated and profits require the return on investments to be greater that the money saved or created to invest. This happens because investment for capital creation results in more efficient processes that generate a greater return than their investment.
Armed with these concepts we will attempt a more coherent approach to the Keynesian multiplier and examine Moore’s critique in the next article.
Note I missed out a key argument here, productive investment returns (discounted) =savings over the term of a loan and the law of large numbers indicates that with no change in net debt in the economy investment (present value)=savings, but this is true only so long as there is no net change in debt, it does not apply generally.