This tweet from MacroResilience caught my eye today:
"In a world of interest-bearing money, money = govt bonds & The "liquidity trap" is a permanent condition, not a temporary affliction".And he then expanded his argument in this post.
The first thing to note is that nearly all money is interest-bearing - and not because central banks are paying interest on reserves, although this is a contributory factor. It is because nearly all money now is held in bank accounts and other forms of investment, and nearly all bank accounts and investments carry interest. Actually this has been the case for a long time, but the change in the last decade has been the ease with which people can move money around between accounts, funds and investments. As MacroResilience points out in his post, there is no reason for anyone to leave significant amounts of money in a non-interest bearing account. And I would add that there is no reason for anyone in the Western world to keep large amounts of physical cash, either. So for the first time in history virtually all money bears interest.
The fact that money is interest bearing, and money can be readily exchanged with other safe assets - especially government debt - means that investors have no preference for safe assets over money. Any real difference between yields on safe assets and interest rates on government-insured deposit accounts will be immediately arbitraged away, so it is reasonable to assume that they are now equivalent for any given maturity (deposit accounts are not time-bounded) and liquidity (time deposits don't have immediate access). This is the liquidity trap, and as originally envisaged by Keynes and described more recently by (among others) Paul Krugman, it is a "temporary affliction" in a distressed economy.
Both Keynes and Krugman describe the liquidity trap as coming into play when interest rates are so low that the usual preference for holding interest-bearing investments instead of non-interest bearing cash doesn't apply any more. So for them, the liquidity trap is a phenomenon associated with very low nominal interest rates - an abnormal situation by any standard. And we have had very low nominal rates for five years now, so it would be reasonable to assume that the liquidity trap we now find ourselves in is due to interest rates being near-zero, and that once we have restored the economy to sufficient health to allow interest rates to rise to historic norms, normal service will be resumed.
But that's not actually the current situation. We have interest-bearing money. Yes, interest rates on money are very low at the moment. And therefore - as I explained above - so are yields on the investments which are near-substitutes for money. But if interest rates were to rise, would this change?
I can't see any reason why it should. Because interest-bearing money is freely exchangeable with government debt - and indeed the shadow banking system constantly performs that intermediation - the equivalence between government debt and interest-bearing money would hold at any level of interest rates. We are indeed in a liquidity trap, but it's not because of economic distress and near-zero interest rates. It is because the nature of money has fundamentally changed. Money is no longer just "cash". Money is any financial asset that flows freely and is readily exchangeable for currency.
MacroResilience then goes on to discuss whether targeting real interest rate stability rather than price stability would be a better framework for monetary policy. I don't propose to discuss that here, though I recommend reading his thoughts on this subject. What interests me is the effect of negative interest rates in an environment where government debt and money are near-perfect substitutes.
In this post, Izabella Kaminska noted that yields on US Treasuries are now entirely negative except for the 30-year. But that's where they seem to be stuck - a choke point. And there is also a choke point on interest-bearing commercial deposits, whose rates cannot fall very far below zero because non-interest bearing physical cash is an alternative. Perhaps the problem with the UST yield curve is that it has hit its equivalent of the choke point for negative deposit rates and cannot fall further because investors can switch into cash? In which case, if the Fed wishes to depress yields further the next stage will be to restrict cash issuance and impose negative interest rates on reserves.
But what would happen if the UST yield curve were entirely negative? To understand this we need to break with conventional macroeconomic thinking. Conventional thinking says "so what", more-or-less: it assumes that the shape of the curve would still remain the same even if all points were negative. I don't think so. As I've noted before, the negative interest rate world is a very strange place. It is a distorted mirror image of the normal world. I say distorted, because negative interest rates affect the relationship of the public and private sectors in a way that has serious implications for the ordering of society. I shall come back to this shortly. What is important at this point is the mirror image. When an entire yield curve is negative, I think it will invert. Yields on longer maturities will be lower than yields on shorter ones. This should make sense if you interpret negative yields as indicating that investors are primarily interested in safety rather than yield, and are prepared to pay for it. A 30-year Treasury is the safe asset equivalent of a 30-year fixed rate mortgage - its refinancing date is so far away you don't have to worry about it, so there is effectively no duration risk. It's gold-plated. It would be expensive. Of course, I may be wrong.....but I can't think of any good reason why investors who want yield would invest in government debt at negative yields. They would look for positive returns on higher-risk assets.
A fully negative yield curve would force deposit rates below zero. I've noted that the existence of cash as an alternative would make it difficult for banks actually to charge negative rates, so I would expect them to show up as fees and charges rather than interest rates. However, holding physical cash does have risk, as does holding physical cash substitutes such as gold. Investors may prefer to pay what amount to safe deposit charges. And if holdings of physical assets such as cash and gold were limited or even outlawed, then deposit rates could go deeply into negative territory. This is not as crazy as it sounds: in 1933 the US banned private holdings of gold. Admittedly that was to ensure a plentiful supply of gold to settle its international trade debts, but coercive techniques of this kind could well form part of the armoury of an increasingly desperate government trying to kick-start its economy and fast running out of ideas.
When yields and deposit rates are both negative we are, of course, in the realms of financial repression. But perhaps more importantly, we are also facing fundamental changes to financial institutions and their relationships. When interest rates are fully negative, the normal sequence of maturity transformation reverses. Lenders (depositors) have to pay to lend, so they won't want to lend for very long if at all, while borrowers are paid to borrow so are happy to take out very long-term loans. Borrowing long, lending short.....the normal flows of funds would be reversed. And the effect on banks would be horrible. They would find it difficult to raise funds except from government insured deposits and the central bank, because everyone would be charged for lending to them, and they would be paying people and businesses to borrow from them. This is clearly not a viable model. Private banking as we know it could not possibly continue. But there is something that can borrow long and lend short, and in fact does so at the moment. It is, of course, government.
By "government" in this case I mean both the fiscal authorities and the central bank. I regard the distinction between the two as entirely spurious when money and government debt are near-perfect substitutes and freely exchangeable. Fiscal authorities borrow long - very long, actually. Maturities of 30 years or more are common for government debt, and some debt is perpetual (the UK still has perpetual debt from the Napoleonic Wars). No private institution can borrow at these sorts of maturities, and negative rates pose little threat to the market for safe government debt: investors seeking safety will buy government debt even at deeply negative yields. Conversely, central banks lend short - very short, in fact, typically overnight although they also do term lending and short-term borrowing. So in the negative rate universe, the entity best able to provide banking services would be government.
This is once again consistent with the distorted looking-glass nature of the negative rate world. Things that are private become public: the private sector becomes dependent on the public sector. So banking - currently a private operation - would become a public function, or at least very dependent on public sector support. But we are already a long way down this road, actually. To what extent are banks that rely on central bank funding "private"? If a bank is so large and systemically important that it cannot be allowed to fail, so must be rescued by government if it gets into difficulties, is it really "private" - or is it actually a public utility? And if bank runs can only be prevented by government guaranteeing to reimburse depositors in the event of bank failure, how can insured deposit accounts be regarded as anything other than public liabilities?
Actually, we have been going down this path ever since money started to become interest-bearing and people had alternatives to bank deposit accounts for their savings. So although negative rates would precipitate a private banking crisis and force governments to take over borrowing and lending functions, all that really does is speed up the process. The fact is that the distinction between "private" and "public" has been getting increasingly blurred for a long time now. Every new piece of regulation and every intervention to improve the safety of banks increases their dependence on government, however much we might like to think it does not. The more reliant people and businesses are on banks, the greater the likelihood that government would have to intervene to resolve failures of major banks. Yet banks are becoming more and more fragile, as the nature of money expands to include financial assets that aren't under bank control and people take advantage of non-bank opportunities for saving and borrowing. And as we strip away the layers of mispriced and mis-sold products, rigged rates and crazy lending, what lies underneath - core banking - looks less and less like a viable business proposition.
Negative rates do indeed create weird distortions. But perhaps the end of private banking as we know it is not a distortion. Perhaps it is the inevitable consequence of the fundamental change in the nature of money.
Macroresilience On the folly of inflation targeting in a world of interest bearing money
FTAlphaville The end of Ro-Ro, or is it
Coppola Comment The strange world of negative interest rates
When governments become banks
The nature of money
I am indebted to scepticus on the FT Alphaville post for a most interesting and informative discussion, much of which has fed into the second half of this post. The discussion itself can be found by clicking the link above and scrolling down to the comments, though I do recommend reading Kaminska's post too. In the discussion, I am Cat in a Box.