In a credit money system, the vast majority of money in circulation is created not by the central bank but by commercial banks. Furthermore, government deficit spending increases the total amount of money circulating in the economy (unless this money expansion is actively neutralised). Therefore the combination of fiscal authorities and commercial banks can create all the money required by the economy. Indeed it can create far too much, potentially triggering inflation. The job of the central bank is not to create money, but firstly to facilitate payments and secondly to limit the creation of money.
To those of you who are firm believers in the supremacy of central banks, this will seem like heresy. So let me explain how this works with the help of a simple example.
Imagine a sovereign country which issues its own fiat currency. The currency floats freely against other currencies. There is an extensive commercial banking system and a central bank which acts as lender of last resort for the private banks. The reserve requirement is zero and reserves are provided on demand to settle payments made by commercial banks. Most people in the country have bank accounts and the majority of transactions take place through banks. At the start of this example, reserves are zero and there is little physical cash, and consequently the central bank's balance sheet is very small. And (incredibly) the fiscal budget is exactly balanced in every period so that legacy debt doesn’t complicate the example. At the start of this example, therefore, the government's fiscal position is zero. It has no money. What little money is in circulation comes mainly from commercial bank lending to the private sector.
So our government puts together its budget for the year. It takes a while for tax receipts to arrive, and in the meantime it needs to pay wages to employees, benefits to recipients, payments to suppliers. We are used to governments funding their deficits by issuing bonds in advance of spending (more on this shortly), but let us assume for a moment that this government chooses not to pre-fund its deficit spending. How would it make these payments?
It would make them through a commercial bank, of course. Whether or not its transaction account at that bank actually contains any money is irrelevant. The government is the most creditworthy borrower in the country and the sole issuer of banking licences (which confer the right to create sovereign money). In the absence of pre-funding, therefore, the commercial bank would allow the government to make these payments by running an overdraft.
As the Bank of England explains, money is created when banks lend:
Whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s bank account, thereby creating new money.An overdraft is simply a bank loan of indeterminate duration. Overdraft financing of government spending therefore results in M3 expanding by the amount of the government's deficit.
This is, of course, monetary financing of the government deficit. It could be inflationary. So having forced a commercial bank to create money to fund its deficit, the government then drains the newly created money, leaving M3 at its original level.
The principal mechanism used to drain money created by government spending is bond issuance. Payments to government by purchasers of new bond issues are intermediated through the government’s commercial bank account, funding its deposit account and eliminating any overdraft. When a loan or overdraft is paid down, money is destroyed (eliminating the money created when the loan was made). Issuing new bonds ex post therefore reduces M3, while ex ante issuance prevents M3 expanding. Term deposits directly with the government (for example through NS&I) have the same effect.
And so do tax revenues. Bonds are redeemed as tax revenues are received. When the government’s spending (including bond interest payments) is entirely “financed” by bonds which are subsequently redeemed from tax revenues, the increase in M3 from government spending is wholly offset by the reduction in M3 from private sector purchases of new bonds, and their redemption from tax revenues is a wash. Regardless of how much the government spends, if it is wholly offset by bond issuance subsequently redeemed by taxation there is no net new money in the economy.
Of course, government spending may be expansionary for other reasons. Government expenditure mostly goes to people on low to middle incomes, who are likely to spend much of it, so increased government spending may raise the velocity of money even if M3 is unaffected. And investment expenditure can generate returns well in excess of the amount invested.
Government borrowing from central banks is inflationary when the absence of an external borrowing constraint enables the government to spend extravagantly and increased velocity arising from this pushes up prices and wages. This effect can be significant, but it would equally apply if government was borrowing from a captive commercial bank backed by unlimited central bank liquidity support. Prohibiting central bank financing but not commercial bank financing of government deficits is therefore absurd. Admittedly, if commercial banks are unwilling to fund the government deficit, the government can force the central bank to print physical currency and deliver it by a variety of mechanisms: helicopter drops, jars in the ground, brown envelopes, wheelbarrows…..But frankly, with a modern banking system, any government that had to resort to physical cash would already be in very deep trouble. After all, if it is so untrustworthy that domestic banks won’t lend to it, its days are numbered anyway.
Commercial bank lending to government is not really subject to central bank control. Central banks limit commercial bank lending to the private sector by controlling both the price of lending and its availability. But as long as central banks are willing to supply reserves, and governments are willing to pay interest on their overdrafts, commercial banks will always create money for their governments. Central banks can discourage or prevent commercial banks lending to the private sector - but they can't stop commercial banks supporting their own government.After all, the central bank itself is beholden to its own government. The independence of central banks is entirely fictional.
But what about QE? If government bond issuance drains money, then surely a central bank purchasing bonds must create it.
Indeed it does. If a central bank - or a commercial bank, for that matter - buys all the bonds issued by a government in a budgetary cycle, M3 increases by the amount of the government's deficit exactly as if bonds had not been issued. So QE should be inflationary - and not because of magical multiplier effects arising from increased bank reserves, as
But there is little evidence that QE has much effect on consumer prices, although it does have a significant positive effect on asset prices. This seems to be because the distribution of the M3 increase arising from QE is very different from that arising from the original government spending.
QE returns money to those who bought bonds, increasing bank reserves (M0) in the process. But increasing bank reserves does not force banks to lend. Returning money to the rich does not encourage them to distribute it more widely in the economy. Swapping one safe asset for another does not encourage the fearful to stop hoarding and spend money. And although QE might reduce yields on government bonds, this wouldn’t make any difference to monetary conditions: making it cheaper for governments to neutralise money growth doesn’t enhance money growth. Weakening the currency might help exporters but it wouldn’t improve domestic demand - indeed it might depress it still further by raising the price of imports.
It seems likely that because of its unfortunate distributional effects, QE simply cannot adequately replicate the velocity effects of direct government spending. Where the money goes matters even more than how much is created.The opportunity cost of relying on QE instead of fiscal stimulus after a severe negative demand shock, when fiscal multipliers are very large, is therefore considerable. And so is the harm done by premature fiscal consolidation undertaken in the belief that monetary policy will offset the negative effects. As I have shown here, it is painfully evident that QE does not fully offset the contractionary effects of fiscal consolidation.
So far, we have been talking about a currency-issuing sovereign with its own central bank. The Eurozone is more complex. But essentially, the same mechanism works there too. Spending in Euros by member state governments via Eurozone banks increases M3. Bond issuance by member state governments drains that increase, as does taxation. Central bank funding of government deficits is outlawed by Article 123 of the Lisbon Treaty, but commercial bank funding is not, so in theory there is nothing to prevent a Eurozone government forcing its banks to lend to it.
In fact they already do. But it's not a happy arrangement - and that is because rather than bank loans or overdrafts, governments have forced banks to buy bonds, exposing those banks to the risk of heavy losses if the market price of those bonds collapses. Since M3 rises when commercial banks buy bonds, these purchases have increased Eurozone M3. I hate to think what it would be without them. But this unfortunate arrangement is known as the sovereign-bank "doom loop", and so far no solution to it has been found.
So why not replace those bonds with bank loans? Indeed why not end bond issuance completely, and fund government borrowing requirements entirely from bank lending, as Richard Werner of Southampton University has suggested? Hallelujah, we have a solution to the Eurozone crisis. Let all governments borrow directly from their banks to reflate their economies.
If only it were that simple. Eurozone member states have tied themselves into a treaty which requires them to balance their budgets over the business cycle and limit variation to 3% of GDP. To enforce compliance with these rules, the ECB has several times threatened to withdraw liquidity support from banks. Banks denied liquidity support cannot facilitate government spending. So even if Eurozone governments ended bond issuance tomorrow and funded their borrowing requirements entirely from bank lending, there would still be significant restriction of money growth due to ECB-enforced fiscal consolidation. And because the ECB, like all central banks, lacks the power to force commercial banks to lend, it cannot adequately offset this fiscal consolidation.
The only institutions that can create the money used for economic transactions in the real economy are commercial banks, and the only authority that can compel commercial banks to lend is the fiscal authority. Faux monetary expansion by central banks is no substitute for the real expansion caused by government deficit spending via commercial banks. So when the economy is on the floor, monetary conditions are tight and banks not lending to the private sector, unchain the fiscal authority - and stop worrying about deficits.
Printing clever weird stuff - Principles & Interest