Saturday, 18 July 2020

Positive Money’s flawed “Money We Trust” report.

“Money We Trust” is a work or report produced by Positive Money about so called “Central Bank Digital Currency” (CBDC) which is a name for the type of money that would arise if central banks made accounts at central banks available to everyone, something that several central banks have been actively considering recently. This work is long: a bit over 20,000 words, thus I’ve only had time for one “read-thru”.

Unfortunately the work contains a number of obvious errors, as follows. 

First (p.8) it claims one of the advantages of CBDC as being: “Increasing financial stability by providing a risk-free alternative to bank deposits.”

That idea is flawed because governments already guarantee deposits at private / commercial banks via deposit insurance and bail outs for banks in trouble. Of course in the event of a complete ban on private banks creating money (along the lines  of full reserve banking / Sovereign Money) then there’d be a definite need for CBDC or similar. But this Positive Money report (unlike much of PM literature produced since PM was founded) does not advocate full reserve banking. That is, it deals with the possibility of having CBDC run alongside traditional accounts at private / commercial / high street banks (henceforth “commercial banks”).  (See also para at the bottom of p.18 which starts “A carefully designed….”.)

Tool kits and helicopter money.

Next (p.8) the report goes along with an alleged merit of CBDC put by Ben Dyson, namely “Increasing the toolkit available for monetary policy by making policies such as helicopter money or ‘QE for people’ easier to implement and potentially overcoming the ‘zero lower bound’ of conventional monetary policy.” See also p.27.

It isn't clear exactly why CBDC would help “QE for the people”, which presumably means a cash handout to everyone. The problem is that not everyone would have a CBDC account, just as not everyone currently has an account at a commercial bank. handouts for everyone is very difficult to implement. A better way (and it’s no coincidence that this is how QE has actually been implemented in recent years) is simply to use the extra cash to spend more on whatever the government of the day thinks merits priority: health, education, more generous state pensions and unemployment benefit, defence, tax cuts, you name it.

Commercial banks can’t be “undermined”?

Setting up a CBDC system would obviously attract deposits away from existing commercial banks and the report devotes much effort to considering how to avoid treading on the toes of existing commercial banks too much. For example at the bottom of p.18 it says “A carefully designed and implemented CBDC would likely become a new anchor of trust, allowing complex monetary systems to continue operating effectively in the face of increased demand for digital payments. However, in order to achieve that, CBDC must not inadvertently undermine the structures that allow for private monetary instruments - most importantly bank deposits - in order to maintain trust in all parts of the monetary system.”

And again on p.46, the report refers to the “critical risk” allegedly involved when the share of deposits held by commercial banks falls too far.

Unfortunately there is a glaring clash between that idea and one of the basic ideas advocated by Positive Money since its foundation, namely full reserve / Sovereign Money, which involves a total abolition of traditional bank deposits. The report completely fails to deal with that clash: indeed neither the phrase “Sovereign Money” nor “full reserve” appear in the report.

Moreover the above idea that CBDC increases “trust” is rather contradicted by p.24 which claims that too large a movement of depositors to CBDC would impair “trust in the overall payments system”.

There’d be a flight to CBDC?

Next, the report claims there might be a flight of depositors from commercial banks to CBDC. E.g. the report says (p.19) “The implementation of a CDBC will pose some risks to financial stability, which need to be effectively mitigated. In the previous section, it was concluded that CBDC would compete for market share with cash and bank deposits in the provision of both the store-of-value and medium-of-exchange functions. When it comes to the facilitation-of-trust function, a CBDC - just like cash - would have the upper hand due to its issuance by the central bank, and therefore if it were to compete rather than complement private monetary instruments, it could rapidly erode trust in bank deposits or significantly decrease their market share, generating new risks for the monetary system.” (See also p.20).

Well the problem with that idea is that already pointed out, governments already stand behind commercial bank deposits, thus those depositors would have no particular reason to flee to CBDC.

Moreover, an advantage of commercial bank deposits (from depositors’ point of view) is that interest is earned on them. Or if they don’t pay interest, then certainly interest earned by private banks from lending out money helps defray the cost of running those bank accounts.

Of course that inherent advantage of commercial bank accounts can be countered by having government (i.e. taxpayers) pay interest on CBDC accounts. But that raises the question as to whether taxpayer should pay interest to the relatively well off simply for hoarding money. Milton Friedman and most advocates of MMT argue/d that normally no interest should be paid on government liabilities. (That point is considered in more detail below).

Our existing CBDC system: NSI.

Another reason for doubting there’d be a major flight to CBDC accounts is that in the UK and doubtless some other countries, a form of CBDC already exists: in the form of accounts at the government run savings bank, National Savings and Investments in the UK. Of course NSI does not offer the same flexibility as commercial bank accounts, e.g. NSI does not offer debit cards. Nevertheless, depositors at NSI can withdraw money at 24 hours notice. Thus if depositors at commercial banks were remotely concerned about the safety of their accounts, they’d hold the bulk of their money at NSI, while doing transfers between NSI and their commercial bank accounts when the balance at the latter was larger or smaller than they wanted.

The report does not mention NSI.


The report devotes much attention to the allegedly harmful effect of disintermediation. For example (p.19) the report says  “Our analysis concludes that if central banks can control the overall market share of CBDC, they can effectively mitigate the risk of excessive bank disintermediation.” See also p.27.

The report does not define the word disintermediation, but online definitions are very much in line with common sense, i.e.  the word is defined there as cutting the amount of intermediation: “cutting out the middle-man” as Wikipedia puts it.

So in the case of banks, disintermediatio is taken here to mean lenders lending direct to borrowers rather than lending via a bank and/or a net reduction in the amount of lending and borrowing via banks without any corresponding increase in non-intermediated lending.

Now the problem with the idea that a significant movement of depositors from commercial banks to CBDC would result in harmful levels of disintermediation is that that idea looks odd when considered alongside the policy which Positive Money has advocated since its foundation, namely a very large scale movement of depositors to CBDC: to be exact, a total ban on deposits at commercial banks, with all deposits moving to CBDC which is what happens under full reserve / Sovereign Money.

The reason that not much disintermediation occurs under full reserve is that in the absence of being able to fund themselves to a large extent via deposits, banks can perfectly well fund themselves via equity and bonds, just like every other corporation.

Of course the latter could turn out to be a more expensive method of funding, which would lead to higher rates of interest for mortgagors and other borrowers, but any idea that would be a big problem is very questionable and for several reasons as follows. (The report actually refers to a possible rise in interest rates on p.44, passage starting “Another problem…”).

First, interest rates are currently at a record low. To illustrate, mortgagors in the 1990s were paying almost three times the rate of interest they do nowadays. The sky did not fall in in the 1990s.

Second, low interest rates are not an unmixed blessing: it is mainly low interest rates that have caused the huge rise in the real price of houses over the last twenty years or so.

Third, the report claims excessive disintermediation might harm “financial stability” (p.22, passage starting “If the introduction of CBDC…”)

Well clearly a sudden and disorganisted introduction of CBDC could lead to instability. But if that change is done in an organised manner, there is no reason for instability: in particular if banks have time to cut their liabilities at the same rate as their assets, there wouldn’t be a problem.

Fourth, another reason for thinking a significant amount of disintermediation would do no harm is that as pointed out by Mervyn King in his “Bagehot to Basle” speech, the assets and liabilities of UK banks expanded a massive ten fold relative to GDP between around 1970 and 2000. Quite what we’ve gained from that, apart from catastrophic bank failures of the sort that happened in 2007/8 is a mystery. Thus a big reduction in bank activity (in the form of “disintermediation”) would not necessarily do any harm.

Moreover, there’s an endless list of worthies who keep going on about excessive debts. But who creates most of those debts? It’s banks!

And even more absurd is the fact that a significant number of those worthies who oppose any reduction in the activity of banks also complain about the amount of debt (most of which, as mentioned above, is gratis banks). Lib Dem politician Vince Cable was famous for indulging in that self contradiction.

Fifth, as the more clued up readers of this article will by now have gathered, this all leads to the crucial question as to what the optimum or GDP maximising amount of intermediation is. (Alternatively, if you don’t like the idea of maximising GDP for environmental reasons, then let’s say “output per hour maximising” – in the hope that people use their increased output to work fewer hours.)

Well it is widely accepted that banks under the existing fractional reserve system are recipients of various subsidies and privileges, which means at least on the face of it that the banking industry is currently larger than its optimum size. (I actually spell out in more detail the subsidies / privileges which I think are the crucial ones in this article of mine. But the important point here, to repeat, is that banks under the existing system are subsidised and hence are too large, which in turn means a bit of disintermediation would actually be BENEFICIAL).

Interest on CBDC accounts.

The report deals with this on p.31.  Clearly the number of people opting for CBDC accounts can be manipulated by adjusting the interest paid to CBDC depositors. But as already mentioned, there is a problem there, namely that Milton Friedman and most MMTers claim interest on government liabilities is not justified, except in emergencies. Certainly prior to the 2007/8 crisis it was not normal for central banks to pay interest on the only form of CBDC that then existed, that is reserves held by commercial banks at central banks (if we ignore the NSI).

It could of course be argued that paying interest on CBDC deposits encourages people to save with a view to holding  more of those deposits, which in turn cuts demand, which in turn enables government to spend more: specifically create new money and spend it into the private sector, which enables the above depositors to acquire their desired stock of CBDC money.

But the problem with that argument is that most public spending is of a current rather than capital nature. And it makes no sense to borrow to fund current spending, rather than fund that spending out of tax. There is doubtless not much difference between those two options during the actual year in which the spending takes place. But suppose government borrows $X in year 1 so as to fund $X of current spending. Government will then be faced with having to pay interest on that sum till the end of time, which means that after twenty or thirty years, funding that particular tranche of spending will end up costing taxpayers vastly more than had the expenditure been funded via tax.


The report claims in its concluding section that CBDC would help ensure the “benefits of future monetary stimulus are more equitably shared than those of the past”. Unfortunately no reasons are given for that idea. The reality is that a highly unequal society would be perfectly compatible with CBDC as would an egalitarian society. Though against that, if there is a particular need for as many people as possible to have some sort of bank account, that would probably be easier to arrange under CBDC (and in particular under what might be called “extreme CBDC, i.e. full reserve banking) in that CBDC accounts could be made free of bank charges. That would mean taxpayers in general were subsidising those who in the absence of free CBDC accounts would not be able to afford CBDC accounts or normal bank accounts.

1 comment:

  1. The author of "Money We Trust" put a response to the above article of mine here:


Post a comment.