Germany is an outlier

The G4S World Cash Report came out and I was e-leafing through it when it struck me once again just how much Germany is an outlier when it comes to retail payments. The average German wallet contains 103 physical euros, the European Central Bank (ECB) estimated last year, more than three times the figure in France. Bloomberg says that cash is used in 80% of German point-of-sale (POS) transactions, compared with only 45 percent—and falling fast—next door in the Netherlands. I think they must mean 80% by value because the FT says that 48% of retail transactions are in cash (down from 58% a decade ago).

Perhaps it is that Germans are just naturally conservative people. The Roman historian Tacitus (55-117CE) wrote in his history “Germany and its Tribes” that the barbarian inhabitants of that land had traditionally exchanged weapons, slaves, cattle, women and such like to settle up between themselves but that the Romans had introduced them to money. Having changed their medium of exchange once in the last two millennia, perhaps they just don’t want more change for change’s sake. Or perhaps there is another explanation. The use of cash in retail is falling slowly and we all know that Germans prefer to keep some of their money as cash at home rather than in the bank, maybe much of the cash “in circulation” there just isn’t.

Given the suspicion that much of the cash in Germany is stuffed under mattresses rather than circulation in the economy, it was still rather surprising to hear from the Bundesbank that nine in ten of the euro banknotes that they are are never used in transactions. That’s right: nine in ten. Approximately all of the cash printed in Germany is never used. Not rarely, not occasionally, but never. So this led me wonder whether this huge volume of never used banknotes are in “hoards” (that is, legitimate money held outside of the banking system) or in “stashes” (that is, illegitimate money held outside of the banking system). Can it really be that the German predilection for holding some of their money in the form of cash account for these billions of euros in inert paper money?

Well, because of the current unusual circumstances with respect to interest rates and so forth, it’s certainly a plausible hypothesis. The European Central Bank (ECB) interest rate for bank deposits is currently minus 0.4%. Conventional economic theory would predict that at a minus rate, depositors would prefer to hold cash rather than pay the banking system to look after their money for them.

(One of the reasons why economists are interested in getting rid of cash is in order to allow the interest rates to go further into negative territory in order to stimulate economic activity over hoarding.)

Now, it clearly costs something to manage cash over and above the cost of managing an electronic deposit hence it is interesting to speculate what the German “crossover” negative interest rate might be, the modern version of the old “specie point” at which it was cheaper to hold bullion for monetary purposes rather than paper instruments.

The current negative interest rate cost German banks about a quarter of a billion euros per annum. The Bavarian Savings Bank Association sent around a circular to their members some time ago setting out their calculation of the crossover rate, which they calculated as something like -0.2%, or half of the current negative rate. However, as I wrote at the time, this isn’t really a serious calculation because, as it says at the end, it doesn’t take into account the significant costs of cash in transit (CIT) or the additional security expenditure that would be needed to guard cash hoards. But it does make a fun point, at least to me, which is that the existence of the €500 notes has an impact on that crossover rate. Now that the ECB has decided stop printing the 500s, banks will have to store masses of 200s, so the cost of storage and transport will be higher (which, in turn, will put a premium on the 500s in circulation so that they will trade above par). Just as an indication, two billion euros in 200 euro notes weighs about 11 tonnes.

While that calculation may not be complete, it does make the interesting point that although we have passed the crossover point already, no banks have to date decided to store their squillions under the mattress rather than leave them on deposit. It seems to me therefore that Bavarian estimates must be too low and that the costs of transport, security, insurance and so on are actually quite high, so the ECB will be able to push interest rates further negative before it gets close to a genuine crossover point that would see banks investing in larger mattresses.

Show me the money

In 2016, the latest year for which information is available, cash was used for 44% of all consumer transactions in the UK. That was down from 50% the previous year and from 68% a decade earlier. Victoria Cleland, Chief Cashier at the Bank of England says that the value of notes “in circulation” has been increasing year on year for the past decade or so and that “we are still seeing growth in total demand for cash.”

What on Earth are these notes being used for if they aren’t being used to buy things? This isn’t a UK problem. The latest figures from the Bundesbank show that nine out of every ten euro banknotes issued in Germany are never used in payments but hoarded at home and abroad as a store of value, the Bundesbank has found. Not “rarely”. Not “infrequently”. Never. The notes are not in circulation at all but are stuffed under mattresses.

Down under, the Reserve Bank of Australia (RBA) Bulletin for September 2017 notes that the value of notes “in circulation” has gone up 6% per annum for the past decade while the use of has collapsed (from two-thirds of consumers payments down to one-third) over the same period. It goes on to note that higher cash usage may be concentrated in groups not included in the survey of consumers (who may well use cash more often than the average consumer) as well as the shadow economy.

Aha. The shadow economy.

A couple of years ago I was at an event where the Chief Cashier said that only about a quarter of the cash the Bank put’s into circulation is for “transactional purposes” (i.e., used). They have a richer categorisation than the Bundesbank for the rest of it, saying that it is either shipped overseas (i.e., exported), kept outside of the banking system (i.e., hoarded) or used to support the shadow economy (i.e., stashed).

Aha. Stashed.

I wrote a comment piece on this for The Guardian, looking at what the key drivers in each of them might be. The first, cash that is used, is easy. We know that the driver is technology but that the impact is weak. In other words, new technology does reduce the amount of cash in circulation, but very slowly.  Moving on to the next category, I know it’s a rather simplistic analysis, but if the amount of cash that is being hoarded has been growing then that would tend to indicate that people have lost confidence in formal financial services or are happy to have loss, theft and inflation eat away their store of value while forgoing the safety and security of bank deposits irrespective of the value of the interest paid. If, on the other hand the amount of cash that is being stashed has been growing then the Bank of England is facilitating an increasing tax gap that the rest of us are having to pay for. In this context cash is a mechanism for greatly reducing the cost of criminality while it remains a penalty on the poor who have to shoulder an unfair proportion of the cost of cash. In this case, we should expect to see a strategy to change this obviously suboptimal element of policy.

The amount of cash that is being exported is hard to calculate, although the Bank itself does comment that the £50 note (which makes up a fifth of the cash out there by value) is “primarily demanded by foreign exchange wholesalers abroad”. I suppose some of this may be transactional use for tourists and business people coming to the UK, and I suppose some of it may be hoarded, but surely the strong suspicion must be that these notes are going into stashes. Note “primarily”. 

My personal feeling is that stashes have grown at the expense of hoards. In a fascinating paper by Prof. Charles Goodhart (London School of Economics) and Jonathan Ashworth (UK economist at Morgan Stanley), they note that the ratio of currency to GDP in the UK has been rising and argue that the rapid growth in the shadow economy has been a key cause. If you look at the detailed figures, you can see that there was a jump in cash held outside of banks around about the time of the Northern Rock affair, but as public confidence in the banks was restored fairly quickly and the impact of low interest rates on hoarding behaviour seems pretty marginal, there must be some other explanation as to why the amount of cash out there kept rising. Two rather obvious factors that do seem to support the shape of the curve are the increase in VAT to 20% and the continuing rise in self-employment (this came up a couple of times in comments to The Guardian piece), both of which serve to reinforce the contribution of cash to the shadow economy.

There are a awful lot of people not paying tax and simple calculations will show that the tax gap that can be attributed to cash is vastly greater than the seigniorage earned by the Bank on the note issue. Cash makes the government (i.e. us) considerably worse off. In summary, I think think the Bank’s view on hoarding is generous and that it is the shadow economy fuelling the growth in cash “in circulation”. There’s something wrong about this, especially when we know that the cost of cash falls unfairly on the poor. It is time for Bank of England to develop an active strategy to start reducing the amount of cash in circulation, beginning with £50 notes.

The Taylor report is right: we should get cash out of the “gig economy”

The Taylor Report was released today. It’s a report about the “gig economy” and contains a number of proposals for reform in the labour market to modernise the various systems (e.g., tax and benefits) and improve the lot of workers. I don’t propose to comment on any of those proposals, also having recently entered the gig economy myself, I can attest to both benefits and annoyances, but I do want to comment on one point made by the report that was picked up in the media. 

Cash-in-hand payments to builders, window cleaners, plumbers and other trades people should be discouraged through a technology revolution to collect up to £6 billion more in tax, a Government-commissioned review urged today.

From Abolishing cash-in-hand jobs ‘would raise £6bn in tax and benefit workers’ | London Evening Standard

The report notes, entirely correctly, that allowing people to exist in a cash-in-hand economy is not only bad for them (because law-abiding employers get undercut) but that it is bad for the rest of us too. Here’s a short extract from my new book Before Babylon, Beyond Bitcoin on this point:

Professor Charles Goodhart (London School of Economics) and Jonathan Ashworth (UK economist at Morgan Stanley) have studied the subject in some detail. They note that the ratio of currency to GDP in the UK has been rising (as you will recall from Figure 7) and argue that the rapid growth in the shadow economy has been a key cause. In their detailed examination of the statistics, the authors make a clear distinction between the “black economy” (e.g., drug dealing and money laundering) and the “grey economy” of activities that are legal but unreported in order to evade taxation. When your builder offers you a discount for cash and you pay him, you are participating in the grey economy. When your builder offers you crystal meth and you pay him, you are participating in the black economy. They define a total “shadow economy” as the sum of the black and grey economies.

…Two rather obvious factors that do seem to support the shape of the Sterling cash curve are the increase in VAT to 20% and the continuing rise in self-employment, both of which serve to reinforce the contribution of cash to the shadow economy. The Bank say that there is “limited research to confirm the extent of cash held for use in the shadow economy”, but Charles and Jonathan make a reasonable estimate that the shadow economy in the UK could have expanded by around 3% of UK GDP since the beginning of the current financial crisis.

…According to Tax Justice UK, that expansion means that there were £100 billion in sales not declared to UK tax authorities that meant a tax loss of £40 billion in 2011/12 and that will rise to more than £47 billion this year. The IMF have noted that while Her Majesty’s Revenue and Customs (HMRC) is not good at estimating losses outside the declared tax system, which is why their latest estimates for the tax gap are low at £33 billion for 2011/12. And while we all read about Starbucks and Google and other large corporates engaging in (entirely legal) tax avoidance, half of all tax evasion is down to SMEs and a further quarter down to individuals (according to HMRC).  There are an awful lot of people not paying tax and simple calculations will show that the tax gap that can be attributed to cash is vastly greater than the seigniorage earned by the Bank on the note issue. Cash makes the government (i.e. us) considerably worse off.

The suggestion made in the Taylor report should be uncontroversial. However, there are people out there who think that forcing law-abiding persons such as myself to subsidise money launderers, drug dealers and corrupt politicians is a reasonable price to pay because the alternative is unpalatable.

In a world without cash, every payment you make will be traceable.

From Why we should fear a cashless world | Dominic Frisby | Opinion | The Guardian

My old friend Dominic Frisby is of course, completely mistaken about this.  Whether the electronic money in your pocket is completely traceable, completely untraceable, or somewhere in between, is a design decision. As I point out in my new book (did I mention that I had a new book out?) where exactly that dial is set between anarchy and totalitarianism is something that our elected representatives should decide and then ask technologists to deliver. This is subject that I know a rather a lot about and so I can assure you that the technology that we already have is perfectly capable of delivering electronic money anywhere on that spectrum.

My own prediction is based on William Gibson’s prediction in the pages of Count Zero. There, one of the characters in this future fiction notes in passing that “it wasn’t actually illegal to have [cash], it was just that nobody ever did anything legitimate with it”. Therefore I expect to see a variety of different kinds of anonymous electronic value transfer systems that are used to deliver pseudonymous electronic money systems and I expect some of those pseudonymous electronic money systems to be used by banks and others to deliver the special case of wholly traceable payment systems.

That, however, isn’t the point of this post. The point that I want to make is that we need an intelligent and informed debate on what we want to replace cash, since it’s going to happen. It should be society that determines how it wants electronic money to work. Whether cash is going to burn out or fade away, we should be planning its 21st-century replacement now. It’s an interesting question to ask whether that means Bank of England Bitcoins or not!

Csfi jun audience

On which topic I was invited along to take part in the CSFI roundtable on “‘Formal’ digital cash: The currencies of the future?” with Ben Dyson from the Bank of England and Hugh Halford-Thompson of BTL Group last month. The event, held at the London Capital Club, was hugely oversubscribed, which I took to be evidence of renewed City interest in the general topic of digital cash and the specific topic of digital currency.

My good friend Andrew Hilton, long-standing captain of the good ship CSFI, framed the discussion in his invitation ask the basic “what if”. “What if some central bank issued a digital coin that was as widely accepted as a bank note? Or, if not a central bank, what if a group of banks or payments operators issued a similar digital coin?”.

For me, the roundtable was both an opportunity to plug my new book (did I mention that I have a new book out by the way?) and an opportunity to learn in the best possible way: by answering hard questions from smart people. I won’t attempt to summarise the discussion here except to say that there seems to be a lot of confusion about what form a central bank currency might take and it wasn’t limited to the people in the room.

“Such risks could be reduced if central banks offer digital national currencies, which the IMF defines as a ‘widely available DLT-based representation of fiat money’.”

IMF urges central banks to study digital currencies | afr.com

Now, why the IMF would define digital national currencies this way is unclear. A national digital currency, or e-fiat for short, may be implemented in any number of different ways. A “widely-available DLT-based representation” would be only one such option and even then it is not entirely clear what “DLT-based” actually means in this context. For that matter, it is not entirely clear what “DLT” means in this context either.

It’s important to separate the topics to move the conversation along: do we need e-fiat and if we do, then how should it work? To the first point I think the answer is probably yes. To the second point, the answer is “well, it depends”. It depends on what we want the e-fiat to do. Should it deliver anonymity or privacy, for example. Should it work like M-PESA or Bitcoin? That’s a fun discussion. How much would it cost to set up “Bank of England PESA”? It wouldn’t even have 100m accounts and Facebook has a couple of billion. If they were to look at some form of shared ledger solution, where copies of the “national ledger” are maintain by regulated financial institutions (e.g., banks – whereby taking part in the consensus-forming process would be a condition of a banking licence) and the entries in those ledgers related to transfers between pseudonymous accounts (i.e., your bank would know who you are but the central bank, other banks and auditors would not) then it would be a permissioned ledger (without proof of work) that could work pretty efficiently. Either way, my point is that it’s doable, so we ought to do it.