Vollgeld: Our response to Martin Wolf

This week, Financial Times commentator Martin Wolf argued that the Swiss should vote in favour of Vollgeld- a plan to fundamentally change the way money is created. We believe this is a risky experiment.

Open to ideas

We have discussed some pros and cons of Vollgeld elsewhere and Wolf raises some very valid questions. In discussing some of them, we should first emphasise that we are not against alternative setups of the monetary system per se. There are many developments, tech-related and otherwise, which prompt financials and banks, in particular, to fundamentally rethink their business models and balance sheets. Why not add monetary reform to the mix? In fact, when designing the system from scratch with today’s technology and knowledge, one would probably exclude physical cash (way too easy to counterfeit and abuse) and include a form of central bank digital cash. After all, the idea that the public should have access to money at its source, and not only via an intermediary like a bank, is fundamentally appealing.

But money creation is not the source of the problem

Yet Vollgeld goes a few steps beyond introducing central bank digital cash. It strips banks of their money creation ability. This is often justified by pointing to repeated banking crises. Yet most, if not all of those crises, started with credit. That means they affected the asset side of banks’ balance sheets. Crises tend to be called ‘credit crises’ and not ‘money (creation) crises’ for a reason. Money, on the other hand, is a bank liability and is at best a by-product, not the source, of the problem. It should also be noted that credit necessarily involves two parties: a lender and a borrower. Credit cycles are not only supply-driven but also reflect mood swings among borrowers.

Shadow banking would get a boost under Vollgeld

Wolf is well aware of all of this, of course, but notes that in a crisis (whatever its source), a bank will be rescued to protect its liabilities, in particular, deposit holders. By disconnecting deposits from bank assets, banks no longer need to be rescued. This sounds like a logical solution. However, depositor protection is just one reason for the elaborate system of bank oversight that has been erected. Credit crises show that a sudden stop in credit supply is detrimental to an economy. As depositors, we may be glad our money is safe, but if we then lose our jobs because of a deep recession, we are hardly better off.
Moreover, the most recent and severe crisis began with lending by non-bank entities in the US. ‘Parallel banking’ or ‘shadow banking’ is a channel starting with originate-to-distribute lenders and ending with money market funds. This channel performs functions that are normally done by banks: transforming non-liquid, long-duration and risky assets into highly liquid, short-term and (perceived) low-risk liabilities. We all know where US shadow banking ended in 2008. The point here is, Vollgeld, or any other monetary reform aimed at money and money creation, does not address the issue of the parallel banking channel mimicking banks. In fact, this parallel banking channel receives a boost under Vollgeld as traditional bank lending is curtailed and made less competitive by the inability to create money.

Strict regulation still needed

Wolf rightly states that investors will bear the risk. But is that sustainable? In a Vollgeld system, lenders will compete for funds by offering attractive terms. Yield swings may contribute to booms and busts, just as they do in bank-based lending. In addition, parallel banks will try to shape their liabilities into something that resembles money and traditional deposits as closely as possible, in terms of liquidity and perceived risk – as US money market funds started to do back in the 1970s. Investors will be lulled into believing these “near monies” are like money, also in terms of safety. A belief that will hold until the next crisis – when investors will call for bailouts (which were duly established e.g. in 2008).

So in order to avoid the proliferation of such ‘near monies’ and the inflation of the next credit bubble outside the regulated banking system, lenders, both bank and non-bank, will need to be tightly regulated from the start, which annuls one of the touted advantages of Vollgeld-like systems, namely less regulation.

But if we need to maintain strict regulation anyway, regardless of where money is held, why make the switch to an untested alternative setup with uncertain benefits? A system too, where the central bank has the difficult task to make crucial judgements about the pace of growth in both money and credit?

Lack of detail

As Wolf acknowledges, regulation and supervision have changed a lot over the past ten years, giving supervisors a host of new tools to prevent crises and if needed, wind down banks while imposing losses on holders of equity and debt. At the same time, Vollgeld and sister proposals at this time are lacking in detail. Do the Swiss really want to subject their large financial sector to such an unprecedented experiment? Better to first work it out in more detail, and start experiments on a more limited scale.

This post first appeared on THINK.ING.com

Swiss Vollgeld referendum: Time to prepare for another Brexit shock?

The Swiss vote on monetary reform on 10 June. A “Yes” to Vollgeld is a low-probability, but a high-impact event. Ring any Brexit bells? 

Vollgeld: a Brexit-sized referendum in the Alps

On 10 June, the Swiss vote on “Vollgeld”. This referendum asks whether the Swiss want to reform their monetary system in a fundamental and unprecedented way. The referendum has largely remained under the radar outside of Switzerland. That may, however, be a mistake. Recent polls indicate a rejection is not a done deal. A “Yes” would first herald a period of great uncertainty, in which the Swiss Franc could suffer. In the longer term, more expensive and scarce credit hurt the economy, but the franc may benefit from deflationary policies.

A rejection is not a done deal

Moreover, the Brexit referendum of 2016 should have taught the forecast community some modesty. And the louder “the experts” recommend to stay away from something, the more inclined people appear to gravitate towards it. After all, to paraphrase a British minister during the Brexit campaign, “people have had enough of experts”. In Switzerland too, the “elite”, including the government, the Swiss central bank (SNB), the banks and a number of scholars, have come out against Vollgeld. If only for that reason, a “Yes” vote should not be excluded – although to be clear, a rejection remains our base case.

What is this Vollgeld and why should I care?

Which brings us to the impact. Here too, an important parallel with Brexit is that a “Yes” would herald a period of uncertainty. In short, Vollgeld would convert commercial bank deposits into SNB liabilities, temporarily replacing the gap on commercial bank balance sheets with SNB loans. Commercial banks would then no longer be allowed to create money when lending.

This basically reverses the order for banks. Currently, they can lend and worry about funding the loan later. Under Vollgeld, they first have to collect the funds, and can then go on and lend these afterwards. That sounds like a simple tweak, but it has big consequences. Vollgeld and various alternatives have been debated among economists for many years. But it has never really been tried – let alone in a developed economy with a big banking sector which finances 33% of its balance sheet with deposits.

A fundamental reform with wide-ranging consequences

Vollgeld is a complicated monetary reform about which books can be (and have been) written. We shortlist some consequences here. We discuss them at greater length in our main report.

  • Current accounts are banks’ most stable and cheap sources of funding. As these are lost (they are converted into SNB liabilities), bank lending will become more expensive and supply may be reduced. Bank profitability will suffer.
  • Current accounts will become more expensive for depositors, as the costs of e.g. cybersecurity and maintaining the payment system will have to be paid for directly.
  • The central bank will, whether it wants to or not, get a more politicised role. Vollgeld is created and doled out to the government or citizens “for free”, so it is basically helicopter money. The SNB has to decide on the amount of helicopter money to be created. Also, it will have to “ensure the supply of credit to the economy”, providing additional funding to banks if necessary.
  • Contrary to what proponents argue, Vollgeld will not materially reduce the incidence of crises. After all, financial crises tend to stem from an overextension of credit, not an oversupply of money. In a Vollgeld system, lenders will try and find ways to circumvent the limitations Vollgeld imposes on funding. In addition, reliance on non-bank lending may increase. Non-bank lending was where the 2008 financial crises started in the United States.

“The [Vollgeld] initiative raises unrealistic expectations… [it] is an unnecessary and dangerous experiment”
Thomas Jordan, SNB President. The SNB on Vollgeld (in German)

So what happens after a “Yes”?

Should the Swiss adopt Vollgeld, then (like with Brexit) initially uncertainty will dominate. This would be a negative for the Swiss Franc. In the longer term, Swiss banks and the economy

If adopted, uncertainty will dominate

will suffer from reduced credit. The SNB will find it much more difficult to reverse helicopter money creation than to reverse more traditional monetary policy (raising rates, or selling assets to reverse quantitative easing). Therefore, the central bank will want to avoid creating too much money in the first place and adopt a conservative, deflationary monetary policy. This may be a positive for the franc. The positive effect may be reinforced by the safe haven status provided by Francs held directly at the SNB.

Is a fudge still possible after a “Yes”?

The Vollgeld initiative spells out the changes it wants to see implemented in the law. Yet the follow-up to the Swiss “mass immigration” referendum (which the EU judged to be a violation of free movement principles) shows that the Swiss Parliament — which has come out against Vollgeld — may try and water it down. This comes with political risks, though. If Vollgeld is adopted with a great majority, its proponents are likely to call for another referendum if they feel Parliament is not implementing Vollgeld properly. A thin majority, on the other hand, increases political leeway for Parliament to come up with a fudge.

Read our full report “The Swiss Vollgeld referendum: an unprecedented experiment”.
This article originally appeared on THINK.

Cryptovaluta decentraal? Dat had je gedacht!

Vandaag zat ik in de Tweede Kamer om mijn visie op cryptovaluta te geven. Hier het paper dat ik vooraf aan de Kamerleden heb toegezonden. In het paper betoog ik dat cryptovaluta in theorie decentraal van opzet zijn, maar dat in de praktijk zaken er een stuk traditioneler, namelijk meer gecentraliseerd, aan toe gaan.

  • Dat begint met het handelen in cryptovaluta. De meeste mensen die trots melden in cryptovaluta te investeren, doen dat helemaal niet. Ze beleggen in feite in cryptovaluta-derivaten. Grofweg 70% van de handel vindt namelijk plaats via crypto-beurzen. Deze beurzen hebben zelf wel een wallet op de blockchain(s). Maar de handelaren op de beurs hebben zelf geen cryptovaluta op de blockchain. Ze hebben slechts een claim op de beurs, toevallig uitgedrukt in cryptovaluta. Klinkt dat als een bankrekening? Daar lijkt het wel een beetje op. Met dit verschil dat er geen enkele regulering is voor deze beurzen. Het gaat dan ook wel eens mis.
  • Daarnaast is het beurslandschap sterk geconcentreerd: zo’n driekwart van de handel loopt via vier beurzen.
  • Dan zijn er de miners. Miners clusteren samen in pools. In theorie kunnen miners met 51% van (de computerkracht in) het netwerk in handen, hun wil opleggen aan het netwerk. De grootste drie mining pools beheersten bij Bitcoin al enige tijd meer dan de helft van het netwerk.
  • En tenslotte de programmeurs. Hoewel de broncode van de meeste cryptovaluta’s vrij beschikbaar is, en iedereen eraan kan bijdragen, is er in de praktijk een klein groepje programmeurs dat verreweg het meeste werk doet. In het geval van Bitcoin zijn er uiteindelijk drie mensen die iedere verandering in de code moeten autoriseren. Ook hier is de macht dus sterker gecentraliseerd dan je misschien voor een “decentraal” project zou verwachten.

Conclusie wat mij betreft: niet alles is wat het lijkt. Cryptovaluta en blockchain zijn prachtige innovaties (en banken zijn druk bezig met het toepassen van blockchain). Maar als de revolutionaire decentrale theorie van cryptovaluta in aanraking komt met de realiteit van luie gebruikers, snelle jongens en criminelen, blijkt het toch anders te lopen. En ontstaan machtsconcentraties die er heel traditioneel uit zien. Met als belangrijk verschil dat er geen enkele vorm van regulering of toezicht is. Caveat emptor!

Download het volledige paper hier.

Zie ook:

Cryptocurrency en blockchain: voorbij de hype

Op vrijdag 19 januari 2018 heeft het symposium Cryptogeld, georganiseerd door IvO Center for Financial law & Governance (ICFG), plaatsgevonden bij De Nederlandsche Bank te Amsterdam. Cryptogeld zoals de Bitcoin is een hot topic. Er spelen tal van belangrijke vragen. Gaat cryptogeld het geld zoals wij dat nu kennen vervangen? Wat zou dit kunnen betekenen voor het klassieke betalingssysteem en de rol van centrale banken? Is cryptogeld veilig? En hoe zit het met investeringen in cryptogeld? Is er sprake van een bubble? Is cryptogeld kwetsbaar voor witwassen? Deze en andere vragen kwamen aan bod tijdens dit symposium met gespecialiseerde sprekers uit de sector, toezichthouders en wetenschap.

Waarom Bitcoin een niche-valuta zal worden

Lees op ING.com/THINK ons nieuwe stuk, Why bitcoin is destined to become a niche asset. Samenvatting in één alinea:
Op lange termijn heeft Bitcoin weinig te bieden aan het brede publiek. Het zal waarschijnlijk weer een niche-valuta worden voor een kleine groep adepten. Wat zij als grote voordelen zien van Bitcoin, zijn tegelijkertijd hindernissen die een breder gebruik in de weg staan. De huidige hoge prijs van Bitcoin heeft een twijfelachtige basis, omdat het Bitcoin-platform open source is en gemakkelijk “geforked” of gecopy-paste kan worden. Alleen netwerkeffecten en overstapkosten rechtvaardigen enige waarde, maar die effecten konden wel eens veel kleiner zijn dan gedacht.

Uitgebreidere samenvatting:

  • De “echte” waarde van bitcoin hangt af van het toekomstige gebruik. Als men zich er massaal van afkeert, kan de waarde naar nul dalen. Aan de andere kant, in het onwaarschijnlijke scenario dat Bitcoin alle wereldwijde betalingen overneemt, zou de waarde boven de $1mln kunnen stijgen.
  • Maar juist als betalingssysteem is Bitcoin aan het falen. Het is nu vooral een valuta om te kopen en te houden. De enige overgebleven rechtvaardiging om in Bitcoin te investeren is dan de aanname dat anderen er in de toekomst nóg meer voor willen betalen.
  • Waarom zal Bitcoin het brede publiek uiteindelijk niet aan weten te spreken?
    • Regulering: de decentrale natuur maakt Bitcoin moeilijk te reguleren. Overheden en toezichthouders zullen decentrale financiële netwerken waarschijnlijk nooit echt gaan waarderen. Een negatieve gebeurtenis, zoals een prijscrash gevolgd door publieke verontwaardiging, kan een repressieve toezichtsreactie uitlokken.
    • Werken zonder tussenpersoon: een kerndoel voor de oorspronkelijke Bitcoin-adepten. Maar het brede publiek houdt er niet van geen rechten, garanties, juridische dekking of wat dan ook te hebben. Zij willen gewoon gemakkelijk en betrouwbaar toegang tot hun geld, en een helpdesk kunnen bellen als ze hun wachtwoord kwijtraken.
    • Schaalbaarheid: het Bitcoin-netwerk raakt verstopt en transactiekosten (gemiddeld $8 in november) maken het onaantrekkelijk voor kleine betalingen.
    • Volatiliteit: de waarde van “gewoon” geld wordt gestuurd door de centrale bank. Het aanbod bitcoins ligt vast en de waarde fluctueert met de vraag. Dit maakt Bitcoin inherent volatiel.
    • Energieverbruik: de prijs van het omzeilen van tussenpersonen is een zeer hoog electriciteitsverbruik.
    • Governance: blockchain is goed in handhaving van regels, maar over het opstellen van die regels is niets bepaald. Dit maakt het implementeren van innovaties tot een langzaam en pijnlijk proces. Bovendien kan de macht geconcentreerd raken bij enkele partijen (miners, in het geval van Bitcoin).
  • De huidige Bitcoindominantie lijkt gebaseerd op de gedachte dat Bitcoin altijd de belangrijkste cryptocurrency zal blijven. Inderdaad is cryptocurrency, zoals bijvoorbeeld ook sociale media, onderhevig aan netwerkeffecten en overstapkosten. Dit creëert een “winner takes all”-dynamiek. Maar met cryptocurrency zouden de netwerkeffecten en overstapkosten wel eens een veel kleiner obstakel kunnen zijn dan gedacht. Doordat Bitcoin open source is, kan het gemakkelijk geforkt of gekopieerd worden. Het is gemakkelijk om een kloon of erg gelijkend alternatief te creëren. Bitcoin is misschien schaars op zijn eigen blockchain, maar die blockchain zelf kan oneindig gekopieerd worden. Als bitcoin “digitaal goud” is, dan zijn forken en copy-pasten succesvolle vormen van “digitale alchemie”.

Why bitcoin is destined to become a niche asset

ING.com/THINK has our new report, Why bitcoin is destined to become a niche asset. One-paragraph summary:
In the long term, Bitcoin has little to offer to a wider audience, and will likely return to being a niche product for a select group of enthusiasts. What they regard as key benefits, may actually be impediments to wider adoption. Moreover, bitcoin’s high value today is based on shaky foundations, given that the platform is open source and can therefore be forked and copy-pasted easily. Scope for such cryptocurrency debasement is limited only by network effects and switching costs, but those may be smaller than expected.

Slightly longer summary:

  • The “true” value of bitcoin depends on its future use case. If users would en masse lose interest, then it could end at zero. On the other hand, in the unlikely scenario that Bitcoin takes over all worldwide payments, its value could rise beyond $1m.
  • Yet as Bitcoin is failing as payment system, and is now primarily used as asset to hold, the only remaining justification for investing in bitcoin is the assumption that others are willing to buy bitcoin at higher prices in the future.
  • Why won’t Bitcoin appeal to mass audience?
    • Regulation: Its decentralised nature makes it difficult to regulate. Governments and regulators may not ever come to like decentralised financial networks at all. A negative event, such as a price crash followed by public outcry, could trigger a regulatory crackdown.
    • Intermediaries: working without intermediaries is cherished by a core group of Bitcoin enthusiasts. The mass audience however dislikes having no rights, no recourse, no guarantees, no legal coverage, nothing. They just want secure, reliable and hassle-free access to their money, and a help desk to call when they lose their password.
    • Scalability: The Bitcoin network is currently clogged and the current level of transaction fees (average $8 in November) makes it very unattractive for small payments.
    • Volatility: while the value of “ordinary” money is managed by the central bank, Bitcoin’s supply is fixed and its value depends greatly on demand. This makes it inherently volatile.
    • Energy use: in the case of bitcoin, the price of taking out intermediaries is very high electricity consumption.
    • Governance: blockchain is great at rule enforcement, but does not provide at all for rule setting. This lack of governance makes implementing innovations slow and painful. Moreover, power may get concentrated in the hands of a few (miners, in the case of Bitcoin).
  • The current Bitcoin dominance appears built on the idea that bitcoin will remain the cryptocurrency of choice forever. Indeed, cryptocurrency, like other internet services, are subject to “network effects” and “switching costs”, creating a “winner takes all”-dynamic. However, with cryptocurrencies, these network effects and switching costs may be lower than thought. This is especially the case given Bitcoin’s open-source, forkable, clonable nature. It is easy to create a clone or close substitute. This means that Bitcoin may be scarce on its own blockchain, but its blockchain is in infinite supply. If bitcoin is “digital gold”, then forking and copy-pasting are successful forms of “digital alchemy”.

Eurozone bank lending is no impediment to ECB tapering

Eurozone bank lending growth is clearly alive, but kicking only very gently. The perfect data release for the ECB, a few hours before it is expected to announce its tapering plans.

As markets are preparing for what could be the most exciting ECB press conference of the year, later today, the ECB released one last important set of data this morning, on monetary and credit developments in the Eurozone. On this front, things are moving along steadily. Bank lending to households was stable at 2.7%YoY in September, with lending to non-financial businesses ticking up slightly to 2.5% from 2.4%.

This basically means that bank credit will not be an issue for Draghi this afternoon. The ECB-president can convincingly argue that he has jolted bank lending back to life with QE, TLTRO and negative rates. Back in early 2014, households and businesses were, in net terms, paying off bank loans. But those days of deleveraging appear long off today – at least, for the Eurozone as a whole. Bank lending has been recovering slowly but very steadily over the past two years. To be fair, bank lending growth is still negative in Spain, Portugal, Ireland, Greece and the Netherlands. But in most of these countries, the growth rate is slowly turning less negative. So it may not yet be “mission accomplished” here, but the ECB can certainly claim that its policies have helped to mend bank lending markets, and that things are moving in the right direction.

At the same time, the ECB’s unconventional policies have not led to an outright lending boom. Sure, bank lending growth rates looks a bit exuberant in Belgium (6.8%YoY for households and 8.7% for businesses), and to a lesser extent in France (6.2% for households, 5.5% for businesses). But these growth rates remain single-digit and do not appear to forebode the imminent popping of a bubble. So in sum, monetary and credit developments do not provide any reason to slam the brakes.

In sum, today’s monetary and credit data enables a gentle “lower for longer” tapering message. We expect the ECB to lower monthly purchases to €25bn while extending them to year-end 2018. At the same time, lack of a credit boom creates room for Draghi to signal that rates will remain low beyond 2018.

Why Bitcoin transactions are more expensive than you think

Bitcoins are sometimes marketed as a low-cost alternative to traditional payments but they’re not as cheap as you’d think. What’s going on?

Confronting the Mad Max Problem

A core element of cryptocurrency is the lack of a central authority. Nodes on the network verify transactions which are rewarded with transaction fees and in the case of bitcoins, newly minted bitcoins go with each verified block of transaction. From the verifying nodes’ perspective, these new bitcoins are mined. Hence they are referred to as “miners”.

As explained in our report, one of the central issues of cryptocurrencies is trust. How can the rest of the cryptocurrency network trust the verification work done by miners? I’d like to call this the Mad Max problem. In a Mad Max world, with no law enforcement, your base assumption has to be that nobody can be trusted. How to do transactions take place in such a world without anyone getting robbed?

Replacing trust with raw computing power

For example, malevolent miners could verify blocks of fraudulent transactions in which bitcoin is taken from victims and sent to their own wallets, or where the same bitcoin is spent several times. How do network nodes know that the blocks presented by miners are truly valid?

The innovative concept applied by bitcoin is proof-of-work (POW) system. By making sure that verifying transactions is a costly business, the integrity of the network can be preserved as long as benevolent nodes control a majority of computing power. Together, they will dominate the verification (mining) process. Read Satoshi Nakamoto’s original white paper for a more detailed explanation here.

To make the verification (mining) costly, the verification algorithm requires a lot of processing power and thus electricity. In fact, the website Digiconomist has constructed a Bitcoin Energy Consumption Index, estimating bitcoin energy consumption. And the results are sobering. At the time of writing, verifying one transaction on the bitcoin blockchain consumes about 200kWh.

Privatising gains, socialising losses?

This number needs some context. 200kWh is enough to run over 200 washing cycles. In fact, it’s enough to run my entire home over four weeks, which consumes about 45 kWh per week costing €39 of electricity (at current Dutch consumer prices).

Let’s put this another way. To process your bitcoin transaction, which might not cost you anything, 200kWh of electricity is used. Powering the entire Bitcoin blockchain currently, costs over 2200MW which is more than what the biggest Dutch energy plant, the Eemshavencentrale requires.

This might make you wonder why you’re not charged €39 for the electricity used? The answer is simply the block reward. The miner whose block is selected to be added to the chain currently receives BTC12. At current BTC prices, the block reward clearly and vastly outweighs electricity costs. Mining is a no-brainer for individual miners, but the benefit to society at large is much less obvious.

Looking for more sustainable alternatives

Bitcoin’s energy costs stand in stark contrast to payment systems that have the luxury of working with trusted counterparties. E.g. Visa takes about 0.01kWh (10Wh) per transaction which is 20000 times less energy.

But blockchain technology could be used in a setting with trusted nodes as well, for example between banks. And this would abolish the need for expensive proof of work.

But operating in a setting without trusted authorities was one of the core goals of the original bitcoin project. At the same time, the cryptocurrency community is aware of the sheer energy consumption issue. Therefore, it is looking for alternatives solutions to the Mad Max problem.

One alternative may be Proof of Stake. Miners are not asked to show they put in work (computing power) in validating but to commit valuable resources beforehand, indicating they have a stake in the proper outcome. For example, miners may have to put an amount of cryptocurrency in escrow which is only released if no fraud is detected, otherwise forfeited.

That sounds like a smart idea. However, it implies that only those wealthy enough to be able to put resources in escrow can join the mining process. This creates a plutocracy, which sits uncomfortably with cryptocurrency’s anarchistic and libertarian roots.

My conclusion is that finding a sustainable and fair solution to the Mad Max Problem is one of the biggest challenges for the cryptocurrency community today.

This article appeared on THINK.ING.com.