Reserve abundance and inflation: a response to Andrew Lilico
Andrew Lilico identifies some inconsistencies in the Bank of England's theory of money creation. His piece is long and detailed, but in essence his argument is this. The Bank correctly states that competition for reserves (in the form of deposits and other borrowing) is a constraint upon lending for individual banks, but also says that the absence of such competition because reserves are plentiful does not constitute an incentive for banks to lend. Lilico disputes this asymmetry. For him, if reserve scarcity is a constraint on lending, reserve abundance must equally be an incentive to lend.
The Bank states that it is the price of reserves that determines demand for reserves, not their quantity:
Central banks do not typically choose a quantity of reserves to bring about the desired short-term interest rate. Rather, they focus on prices — setting interest rates. The Bank of England controls interest rates by supplying and remunerating reserves at its chosen policy rate. The supply of both reserves and currency (which together make up base money) is determined by banks’ demand for reserves both for the settlement of payments and to meet demand for currency from their customers — demand that the central bank typically accommodates.
Unfortunately the Bank does not distinguish between the pre-2008 world, where bank lending creates demand for reserves, and the post-2008 (QE) world, where the quantity of reserves is entirely disconnected from bank lending and is far more than the quantity required to facilitate deposit movements across the banking system as a whole. We can say that reserves are "abundant".
As with all other abundant goods, reserve abundance should drive down the price of reserves to zero: after all, if banks have all the reserves they need, they aren't going to want to borrow more. If this is correct, then the normal reserve pricing mechanism by which the Bank of England controls inflation will not work. Lilico's argument is that once banks have repaired their balance sheets, they will expand lending massively in response to abundant reserves. If he is correct, then this would cause severe inflationary pressures in the economy which the Bank of England would be unable to control through monetary policy. Only macroprudential regulation (capital and liquidity requirements) would be able to calm banks' incentive to lend excessively, and Lilico fears that those would be watered down.
Lilico implicitly assumes that the presence of abundant reserves in the system as a whole means that individual banks at all times ALSO have abundant reserves. But this is the "fallacy of division". The level of reserves held by individual banks depends on the pattern of deposits and withdrawals (payments). And the level that individual banks desire to hold (since there is no regulated reserve requirement in the UK) depends on management's view of short-term liquidity needs. At any one time, some banks will fall short of the level of reserves that they wish to hold, and others will have more than they wish to hold. Therefore, even though the system as a whole has abundant reserves, banks still need to borrow and lend reserves. As long as there is any sort of interbank market, the central bank will always be able to intervene in that market to control the price of reserves.
When reserves are abundant, the central bank's intervention in the market takes a different form. Although there is still a marginal demand for reserves, that demand can always be met from other banks, so the central bank is unable to influence the price directly. But because the system as a whole holds more reserves than banks actually need, some banks must always hold more reserves than they want. These reserves sit on the central bank's balance sheet, and the central bank pays interest on them at a price that it determines - we can call this a "bid" price. The central bank's intervention in the market therefore determines not the reserve lending price but the reserve holding price. It should be obvious that if one bank is holding more reserves than it needs, another bank that wishes to borrow those reserves must pay more than the interest rate paid by the central bank on reserves. The IOR rate can therefore be used to control the interbank lending rate.
Raising IOR forces the base rate upwards, increasing both the cost of funding for banks and the price of loans to new and existing borrowers for those loans that are priced directly or indirectly off the base rate - which is most of them. There would be both a disincentive to banks to lend (because of rising cost of funding) and a disincentive to customers to borrow (because of rising interest rates). If this isn't inflation control, I would like to know what is.
Nor is raising the IOR rate the only means of controlling the price of money when reserves are abundant. In the pre-2008 world, open market operations (OMOs) were used to adjust the quantity of reserves so as to maintain the desired offer price: broad money was created as a consequence of bank lending, as the Bank of England explains. In the post-2008 world, OMOs (QE) have been used on a massive scale (and in the US and Japan still are being used) to increase the amount of broad money in circulation directly, bypassing bank lending*: the abundance of reserves can be regarded as a side effect of direct broad money creation by the central bank. This makes OMOs useless for controlling the offer price. But similar operations can still be used to control the bid price. The Fed is currently lending out assets purchased under its QE programme in so-called "reverse repos". These can be regarded as secured loans from banks to the central bank at a price that the central bank determines: they can also be regarded as pairs of asset sales and repurchases with a spread between the two. Either way, the reverse repo rate is really another version of the IOR rate. And both the Fed and the Bank of England offer fixed-rate term deposits to banks, which temporarily remove reserves from circulation. Other tools for managing the price of money in a reserve-abundant world will no doubt be developed in due course.
The inability of the central bank to target the lending rate directly when reserves are abundant does not mean that bank lending and associated inflation would be difficult to control with monetary policy, as Lilico implies in his final sentence. It is more accurate to say that there are two different price mechanisms. The pre-2008 version uses the offer price of reserves to control inflation. The post-2008 version uses the bid price.
But I also question Lilico's assertion that abundant reserves will inevitably result in banks going on a lending spree when their balance sheets are repaired. QE does drive down loan/deposit ratios, but lending does not necessarily increase them - it depends on the funding mix, which is to a considerable degree dependent on market conditions, tax policy and microprudential regulation. And the combined burden of monetary policy as I have described it with macro- and micro-prudential regulation and intrusive supervision is likely to keep the lid on bank lending for some considerable time to come.
It's also worth remembering that most credit bubbles grow not in the regulated banking sector, but in unregulated activities. I have no doubt that at some time in the future there will be another credit bubble which will burst spectacularly and cause financial and economic chaos. But it won't be caused by reserve abundance. After all, reserves were not abundant prior to 2008, but that did not impede the excessive growth of credit particularly in the shadow banking system. It will be caused, like all credit bubbles, by irrational exuberance on the part of banks, borrowers and governments.
I also have no doubt that at some time in the future there will be higher inflation than there is at present. Indeed I hope there is: the present disinflationary trend does not bode well for economic growth. But I see no reason whatsoever to believe that reserve abundance will ever cause uncontrollable inflation.
* The fact that central banks have been using QE to create broad money directly makes it appear as if there are two different mechanisms by which broad money is created. But actually the central bank has always been able to influence broad money directly to some extent. As Nick Rowe points out, under normal circumstances OMOs do have some effect on broad money. Since 2008 OMOs (QE) have been used as a partial substitute for bank lending as the primary driver of broad money creation. But QE is nowhere near as efficient as bank lending as a driver of broad money creation - which is why the Bank of England is no longer doing QE and is using quasi-fiscal measures such as Funding for Lending to persuade banks to lend. The Bank of England's paper doesn't mention this.