Saturday, 31 August 2013

Laurence Kotlikoff versus Positive Money.

Summary. Kotlikoff and PM agree that as regards money which depositors want to use for transaction purposes and/or which they want to be completely safe, that money should not be invested or loaned on: it should simply consist of or be backed by monetary base.
In contrast, and as regards money which depositors want their bank to lend on or invest, Kotlikoff argues that money should go into a unit trust (“mutual fund” in the US) of the depositor’s choice. As with existing unit trusts, depositor-investors carry the full risk.
As against that, PM says the money should be invested in such a way that the bank and depositor-investors SHARE the risk.
The latter arrangement is essentially a hybrid or compromise between having banks simply store money (without investing it) and straightforward investing (as on the stock exchange). It is argued below that there is no case for that hybrid: it falls between two stools. I.e. Kotlikoff’s system is preferable.


PM’s system is set out in Chapter 6 of “Modernising Money” (2012) – printed by TJ International Ltd. The system is also set out in a work co-authored by Richard Werner and the New Economics Foundation.
Under PM’s system, where depositors want their money to be loaned on or invested, depositors’ money is no longer instant access. They can choose the period of notice required before withdrawal is allowed, plus they can choose the proportion of risk they accept. And the interest they get varies with the amount of risk accepted and with the period of notice.
Under PM’s system, depositor/investors get back £X for every £X they put in, except where the bank goes bust. The authors of Modernising Money (MM) do not actually say what constitutes bank failure (p.201). “Bust” or “insolvent” normally means the value of investments or loans made by the bank has fallen below the amount of cash (£X or whatever) that the bank owes depositor/investors. But the authors don’t say whether the failure of just one type of account triggers bankruptcy proceedings, or whether a larger number need to fail.

Problems with MM’s investment accounts.
1. Assuming bank insolvency is triggered by the failure of just one type of investment account (likely to be a risky type of account), that will involve a lot of inconvenience for those who have chosen safer types of investments. Going through bankruptcy proceedings could take weeks or months.
In contrast under Kotlikoff’s system, if a bunch of people who have made risky investments lose half their money, that is of no concern to, and does not affect those who have chosen safer investments. There is no need for the latter to be messed around by bankruptcy proceedings.
Moreover, if one type of investment account does particularly badly (say assets drop to one third of what’s owed to the account’s depositors) then under MM’s system those depositors will walk away with about a third of what they put in. But that happens AUTOMATICALLY under Kotlikoff’s system. That is, if the assets of a unit trust drop to a third of the initial value of investments, then the value of the “unit trust units” drops by a similar amount. But the under Kotlikoff’s system, the “account” or unit trust battles on: i.e. there is no need for bankruptcy proceedings.

2. On p.183 of MM there is a heading: “The Investment Account will not be money.”
This section deals with possibility that money in an investment account or the investment account itself might still be used as a form of money. The example given is the assignment of money in such an account over to a car dealer in payment for a car.
It’s right to be concerned about this. Indeed Irving Fisher was concerned about this in the 1930s. On p.15 of his book “100% Money” he says “deposits could be used as money so that the lending department also issue money or the quasi money of demand deposits.”
However I doubt this would be much of a problem. My experience of buying cars is that car dealers want one thing and one thing only: cash.
But to the extent that this is a problem, the problem does not arise under Kotlikoff’s system: that is, there is no way unit trust holdings are ever counted as cash (nor are unit trust holdings ever accepted by car dealers). 

3. Under PM’s system, depositor/investors cannot get at their money till after a month or more’s notice. Under Kotlikoff’s system, if depositor/investors they want their money out in a hurry, they can get it. They may make a loss, break even or make a profit in doing so, but at least they can get at it.

4. Page 184 gives us the authors’ reasons for investment accounts of the type they propose rather than a Kotlikoff type system. The authors say that any loss made on investments “will be split between the bank and the holder of the Investment Account. This sharing of risk will ensure that incentives are aligned correctly, as problems would arise if all the risk fell on either the bank of the investor. For example, placing all the risk on the account holder will incentivise the bank to make the investments that have the highest risk and highest return possible..”
Well that problem is dealt with by giving depositor/investors the choice as to what is done with their money!!! (See bottom of p.184). For example, to cater for depositor/investors wanting something ultra-safe banks would doubtless offer accounts that just supplied mortgages to British households with a minimum 20% or so equity stake in their house. That’s as good as 100% safe.
As to what might be called the opposite problem, namely an arrangement where banks have no skin in the game, the authors claim banks would have no incentive to make good investment decisions.
Well that problem is dealt with by existing unit trusts by giving staff a bonus depending on the performance of the trust. Indeed banks are notorious for giving staff an incentive to do things, some of them not in the best interests of customers. In short, giving bank or unit trust staff an incentive to make worthwhile investments, without the bank actually taking a stake itself in the investment is not difficult.

5. At the bottom of p.184 says in respect of investment accounts that “the broad categories of investment will need to be set by the authorities.” That’s debatable.
The stock exchange and the unit trust industry already offer investors a huge range of types of investments, and all without any instructions by politicians or bureaucrats. However, it would probably be an idea to force every bank to make available to customers some sort of very safe type of investment, like the above mentioned mortgages where house owners had a 20% or so minimum equity stake. In fact the relevant accounts / unit trusts could be called “building societies” a term with which UK citizens are familiar and which would a perfectly fair description of the unit trusts concerned.

6. (p.185). This page sets out three types of account that each bank has at the Bank of England (“Operational”, “Investment Pool” and “Customer Fund”).
This is unnecessary bureaucracy. The basic and very simple rule that underlies Kotlikoff’s system is: “transaction/current accounts must be backed by central bank money”. As to investors, they can pretty much do what they want. E.g., and taking a far-fetched example, if someone wants to pay for an investment not by using money but by paying the investee with crates of whisky, that’s of no concern to anyone, apart from the investor and investee.
Indeed, the latter example is not all that far-fetched: there are firms that offer employees shares in the firm related to how long employees have worked for the firm. Such employees are in effect paying for their investment with their labour.

7. Bank runs. Bank runs have taken place throughout history and the crisis in the US was essentially a run on the shadow bank industry. Bank runs occur precisely because of the promise that banks make to return to depositors a specific sum of money. As soon as there is a rumour that a bank won’t be able to return that sum of money, depositors might as well get their money out.  
In contrast, under Kotlikoff’s system, if the general view is that a particular unit trust is doing badly, the value of the relevant trust units falls. Depositors can sell their stakes if they like (probably at a loss), but the bank or unit trust does not face bankruptcy or insolvency for the simple reason that it does not owe any specific sum of money to anyone.
That point was explained in more detail in a Wall Street Journal article by John Cochrane.

8. A possible argument for the MM type investment account is that it enables depositor-investors to get some sort of return on their money, while avoiding the possibility of total loss.
Well the answer to that is that they can do that under  Kotlikoff system by putting some of their money into a safe account (or cash only unit trust) and the rest into a unit trust which lends on or invests money.

Positive Money is heading in the right direction in that the system it advocates involves making depositors choose between on the one hand safe / transaction accounts and on the other, investment accounts. However, PM’s investment accounts are a compromise or hybrid between two extremes: first, lodging money in a bank with the bank not lending on or investing the money, and second, straightforward stock exchange type investments. The hybrid does not achieve anything: it falls between two stools.
In contrast, Kotlikoff’s equivalent to PM’s investment accounts are essentially stock exchange type investments.


Afterthought (3rd Sept. 2013). My dismissal of the above mentioned “operational” and “customer fund” accounts over simplified the issue. I’ll expand on that in a day or two.


No comments:

Post a Comment

Post a comment.