We should all know that there’s an inherent problem with fractional reserve banking, after all, it was explained back in It’s A Wonderful Life. If we all turn up and ask for our money back at the same time we can’t have it. What we think of as our deposits are all out in the mortgages of our neighbors’ houses. This is what we call a bank run and when one happens the bank goes bust. The solution is to have a central bank that will lend freely but at penal rates of interest – to quote Walter Bagehot who was the first person to work this out.
So, what’s the actual problem? That the underlying assets, those mortgages, and loans are less liquid than the offer to depositors. The customer money can be taken away at will, the bank finds it more difficult to collect that cashback in.
Well, OK. That’s just a problem with banks, that problem’s solved and so what? Except that this problem, what is called “liquidity mismatch”, exists in many other parts of the markets. This means it’s something we’ve got to be aware of and, where we either want or need to, avoid it.
To give an example of something going on right now there’s a commercial property fund in the UK run by M&G which has had to suspend redemptions. This isn’t exactly new as the same fund had to do this a couple of years back too. The fund invests, as we might think, in commercial property. This is not a notably liquid market, selling something takes months at least, perhaps more if a keen price is to be achieved. Which is fine of course, there is no problem with investing in things which are a bit “lumpy”. Except, equally of course, unless there’s a more liquid offer on the other side of this. Which there is here, this is an open-ended fund.
An open-ended fund is akin to our banking system. You go and buy units in the fund and the fund then goes and buys more of the asset, whatever it is. When you sell units you sell them back to the fund which then disposes of assets in order to fund the payout back to you. When that underlying asset is as a liquid, as easy to sell, as the promise being made to you about when you’re going to get your money back then there’s no problem. When you can have your money back from a commercial property fund tomorrow and it takes months to sell a shopping center you can see that there might, possibly, be the occasional problem.
In fact, what happens is that the fund starts to eat itself like Ourobouros. The easy to dispose of assets go first but that causes a run on the remaining liquidity, more rush for the door, what’s left is only the truly illiquid stock. Which has to be disposed of at firesale prices, depressing values further and so on. Or the fund simply closes to redemptions.
It’s easy enough to see how this might apply to something like commercial property. But a reasonable explanation of the problems over the Neil Woodford funds, again in London, is that he was too illiquid. As redemptions flooded in he needed to and did, sell the liquid listed stocks. That left the funds either holding significant positions in thinly traded stocks – difficult to place at anything close to market – or unlisted entire investments. This hasn’t worked out well, investors are locked in and down significant percentages on the previously stated unit values.
A closed-end fund doesn’t have this problem. For we never do sell our stocks back to the fund, instead we find someone else in the market to sell to. That has its own problems in that the stock of the fund might not be liquid, or asset values can wander a long way from the stock price.
So far this should all seem to be pretty obvious, closed-end for illiquid assets, open-ended only for things that are at least around as liquid as the offer to investors on their ability to recoup their cash. Thus, say, a REIT for commercial property seems entirely reasonable. An open fund for holding S&P 500 index stocks is also entirely reasonable. If there’s no liquid market in S&P 500 stocks when we want to cash in our units then we’ve all got very much larger problems than whether we get the check-in 24 hours or not.
The problem with this distinction comes at the margin. As all problems always do come at the margins, as one situation bleeds into another. This is also where all economics takes place, that’s why the great revolution of the 1870s was called the Marginalist one. What, exactly, is on the one side of our dividing line where open is better than closed?
Take the Winkelvii and their idea of an exchange-traded fund – a form of the open-ended fund – containing Bitcoin. What’s the liquidity of the underlying asset here? Sure, it’s possible to trade bitcoin quickly but actually getting cash out of at least some of the exchanges is a little more time-consuming. It’s not possible to offer 24-hour payouts on such an asset. Not, that is, unless a very large portion of the fund is held in cash at all times in which case the fund isn’t going to track the asset very well.
On the other hand, the global copper market is entirely liquid. Perhaps not the physical metal, something I’ve traded myself, but we have an excellent proxy in chitties – warrants – from the London Metal Exchange and the like. So, yes, an ETF holding copper can indeed be immediately liquid for so is underlying.
How about an ETF in tungsten? Nope, that’s not liquid enough. The item that’s widely traded is a tungsten concentrate, something that has to go through a processor and there are too many variations in trace elements and thus qualify for it to be homogenous. Tungsten metal is large all the same but isn’t a liquid enough market (something else I’ve bought and sold over the years). There is also no terminal or futures market for the metal. It’s simply not suitable for an open-ended fund.
Gold? Sure, go for it, registered gold is one of the most liquid markets on the planet. So the too US and many other government bonds, the dollar, pound sterling and so on. Sudanese pounds not so much nor Venezuelan bonds.
None of this means that we shouldn’t invest in illiquid assets. People have been making fortunes out of those Venezuelan bonds, for example, yields have been well over 40% in dollars. It’s that it’s never going to be easy to move a substantial position therefore while it may be profitable it’s not going to be liquid.
What we do have to know is our own investing concerns. Are we insistent that we need to be able to get our money back as and when we wish? If so then we don’t want to be in these illiquid assets. If we’ve money that can be packed off to mature for a number of years then illiquidity in exchange for higher returns is just fine. After all, this is how Venture Capital funds work, you commit the money for 5 or even 7 years and that’s that. Most of the underlying investments fail because that’s just the way that VC goes but they’re not going to allow you to recoup your funds until that fund has matured, no way at all.
Where it all goes awry is when we don’t understand these tradeoffs that we’re implicitly making. The point being that, absent a central bank bailing out banks and their mortgages, there is no magic wand that transforms liquidity. Thus the simple existence of an open-ended fund investing in illiquid assets doesn’t solve the basic problem of the mismatch. It might, indeed it will, gloss it over in normal trading circumstances but not when that pedal hits that metal. In a crisis the underlying liquidity problem will re-emerge.
The useful rule therefore when looking at funds is to check the underlying asset. If this is traded on a wide and liquid market then sure, why not use an open-ended fund? The advantage is that the fund value will never stray far from the asset value. But the greater the illiquid nature of that underlying the better off we are in a closed-end fund. We might find the fund value wandering from that asset value but we do know that we’re always doing to be liquid in our own investment position.