Thoughts on the future of Bitcoin

One of the advantages of being in a virtual adventure is that I can be in several paths at the same time.
Tonight, I decided to do a small incursion in Blockchain territory.
The blockchain landscape is, at the moment, exciting, frightening, uncertain and promising. On the one hand, we see a move of
several countries threatening to regulate the cryptocurrencies space; we’ve seen China banning crypto miners after having received them with open arms… after having gone through exactly the same cycle in 2014; we’ve seen Korea threaten to ban crypto trading, after being the biggest champion of the technology in recent years.
On the other hand, we see Venezuela launching its own crypto-currency, possibly with Russian help; and the Marshall Islands issue a cryptocurrency that is legal tender in the country.
Meanwhile, the cryptoworld is reeling back from an exceptional end of 2017, and if that was the year of ICOs let loose, 2018 may well be the year of the chickens coming home to roost, with SEC and other regulators establishing what is a security or not, and going through past and would be coin-offerings to bring to justice those who sometimes unwittingly defied the law.

All of this reached popular news outlets relatively easily, at least in sites like Coindesk, Investopedia and the Financial Times. But this is, after all, just a part of the blockchain territory. The technology is far wider than this, and I’m more interested in guessing where it can lead and what will be its uses a few years from now.

As still an amateur in this field, I am slowly building my opinions and vision, and recognize that they are, at this stage, still tempered by naivety. That said, and given that caveat, I’d like to share with you my random thoughts about the crypto field. I’ll limit my thoughts today to Bitcoin, and leave other topics for another day.

Bitcoin

Bitcoin is the originator of all cryptocurrencies and the blockchain technology. As such, it has all the advantages of the first mover and managed to carve a household name for itself. After a tumultuous 2017,it has definitely reached pop-culture status and is no longer an obscure technology only in the hands of techies and shadowy figures.

It has gone through a speculative bubble in 2017, and is now in uncertain territory, trying to stabilize around 10k USD after having hit almost 20k in late December 2017. After the year that broke all records, and catapulted Bitcoin suddenly into the public conscience, the question now is whether it will be able to hold to the gains of 2017. And for the optimists, whether it will reach the outlandish values predicted for 10 years from now.

It is clearly the major cryptocurrency in the world. But while it has today around 40% (all my figures in this paper are liberally rounded, because of how fast these values move. At the time of writing, this percentage was more around 44%, but 40% is a rounder number) of the total crypto market capitalization, and is above values of around 35% at the beginning of 2018, but remember it was comfortably above 70% until the end of 2017. This is not a sign that Bitcoin has gone down in price, but only that new money has come in and diversified into other altcoins looking for quick gains. So, I think Bitcoin is still healthy. But there are things in Bitcoin that I don’t like and I think will have to be solved if it is to become the universal cryptocurrency.

High Fees and Delays

In late 2017, the fee for a Bitcoin transaction hit over $30.00 USD and it could take over a day for your purchase to appear in your wallet. This was grossly contrary to the promise in Satoshi Nakamoto’s white paper that Bitcoin provided a much faster and cheaper way to transact money than traditional banks. This is especially true if you consider that fees do not depend on the value of the transaction, but rather on its size in bytes. Thus, paying 30 dollars in fees for a micro transaction of a few dollars would be absurd. It changed Bitcoin to be a valid means of exchange only for very large transactions, and not for the day-to-day trades each of us might want to do.

Since then, fees and settlement time have gone down and are back at healthy levels. At the time of writing, the estimated fee for a transaction is $0.22 USD. What caused this immense fluctuation?

The direct cause for this spike was speculative frenzy in the last two months of 2017 (I will repeat myself many times in this post referring to this period, so from now on I may prefer on occasion to refer to it only obliquely). But an underlying force that has also led to it was the conservative stance in the debate about solving Bitcoin’s scalability.

This is the first hurdle that Bitcoin will have to solve if it wants to last. If it doesn’t, I believe it will be overtaken by other altcoins which position themselves also as cryptocurrencies and are looking to become the dominant coin: for example, Monero, Litecoin, Bitcoin Cash and Dash. Bitcoin cannot rely on having been the first currency and being the current behemoth: Bitcoin was not perfect at birth and still isn’t, and many competitors have appeared and will continue to do so to improve on Bitcoin’s faults. Bitcoin still has the size and volume to prevail, but it has to improve.

Network Effect

That said, it does not have to be perfect. It only has to be more convenient than the alternatives. I’m convinced that there will not be room for many cryptocurrencies simultaneously being globally used. That would suggest that consumers would be able to choose different currencies according to their use cases, and keep a stash of several coins to use in each circumstance. On the contrary, once we pass the speculative stage of cryptocurrencies, I think people will have as few of these as possible, because that requires less management. And if they are going to use only one, they are going to pick the one that is most convenient: the fastest, cheapest and with the widest network. There is room for a few of these simultaneous only if none manages to dominate all the others in all fields.

But is that possible?

The widest network comes as well with the widest support and the widest reach. It will have more miners and so a higher capacity to deal with transactions, although that is counterbalanced as well by more transactions overall. But because of its size, it is conceivable that developers will have better support and more resources and be able to improve more consistently, therefore becoming cheaper and faster than the competition. This is a big advantage of Bitcoin, because it already has the largest network.

But it is not a guarantee. If Bitcoin’s fees become permanently in the tens of dollars, then it will be pointless to transact values under 1000 dollars. There will be other alternatives to cover that niche, and these will then have the ability to grow and oust Bitcoin from its territory, just because they will be more convenient for everyday use. And it may well be the case that no single crypto-currency will be able to handle a truly global number of users, in which case several cryptocurrencies may be able to divide the market, more or less in the same way that mobile communications companies do.

Let’s backtrack now a bit: how did the fees become so high in 2017?

Bitcoin had impressive growth in 2017, but a particularly crazy November and December. The value has corrected since 2018 started (I wonder how much of this was due to a December effect as the one there used to be in stocks, and whether it will repeat at the end of 2018) and we’ve been in a consistent decline since then. November and December were truly manic, with an euphoria in the air that drew large quantities of new money to Bitcoin, and a large number of transactions flooded the network. This explains the two effects above: with many more transactions competing for miners in the network, transaction delays obviously increased (mining capacity did not increase especially, or at least, certainly not at the same rate). Secondly, because fees are not fixed, and users can choose how much to pay to have a transaction included in a block, a simple supply-demand effect explains how the fees increased so much. Now that this local bubble seems to have burst (we may still be in a more systemic bubble), traffic is lower and consequently so are delays and fees.

It is also useful to notice that all the spikes followed by crashes in the history of Bitcoin (and there were a few over the years) were accompanied and actually supported by a similar exceptional increase in transaction volume.

What does this mean for the future? I’m guessing that Bitcoin will stabilize in 2018 eventually at a plateau, and that it will still be higher than the beginning of 2017. If that is the case, this will have been another spike like the one in December 2013 or November 2015 and that as before Bitcoin will be able to grow again but from a higher base.

I am thinking whether fees can be correlated to stability. When the price is booming and going consistently up, euphoria brings many new buyers and transactions increase. When prices go down consistently and the market becomes scared, fear will bring create many sellers and transactions increase. In both cases, fees tend to go up. If fees are low for a significant amount of time, I’d hypothesize this corresponds to a stable plateau where the market is essentially in equilibrium and believes the current price is fair. This is a good point to strengthen the currency, and I think is where we are at the moment.

The key for the network effect to trigger off is widespread acceptance.There is a tipping point of adoption that, once passed, creates a self-reinforcement cycle, where more people use Bitcoin because there is enough adoption to make it convenient. This leads to more businesses using Bitcoin, more products being available and more reasons for people to use it, leading to yet more users.

At this stage, I still don’t see the necessary support in the number of merchants. Bitcoin is at the moment too volatile to be used as a currency, because a merchant who sells a service today cannot really tell whether the value of its sales will have tumbled tomorrow. For merchants to arrive, we need price stability, or at least a narrow band of price variation. Because of 2017 and the recognition of blockchain as a valid technology, I think that even a period of lowering prices will not scare merchants away. This is the time to experiment with calm, without having to weather storms of high volatility, to put solid prototypes in place and pilot businesses. This may be just what Bitcoin needs to become solid.

Once such merchants become comfortable in their market, others will join, and Bitcoin’s value as a means of exchange will also grow, likely causing network growth. But I’m convinced Bitcoin can only do so it if it keeps itself viable for low-value transactions, with low fees and short settlement times.

Scalability

This brings me to the major technical problem of Bitcoin: scalability. From its early days (at least 2010), Bitcoin blocks have had a limit size of 1Mb. Bitcoin Classic forked with a limit of 2Mb. Then, the introduction of SegWit has allowed this limit to increase up to around 4Mb, and SegWitx2 would allow to go near 8Mb, but I’m not sure of its status at the moment.

If the network effects set in, the subsequent growth may be exponential, or at least polynomial, and even these current improvements wouldn’t prevent a saturated network. You can look at them as an exponential improvement (1Mb, 2Mb, 4Mb, 8Mb) if you are an optimist, but that is only true if the doubling of the size ca be sustained. If it isn’t, and we get stuck here, then it’s no more than a linear improvement.

So, why can’t we increase the block size or make it uncapped? Why is there debate in the community?

Different people are affected differently by this option. Users would prefer to have large blocks in order for transactions to be processed at a higher bandwidth and lower price; but miners may have an incentive to keep blocks small. It may not be obvious at the start, so here is why: miners get their reward from two streams: the Bitcoins they mine every time they publish a block to the chain; the transaction fees they collect from the mining process. The Bitcoin reward halves every 4 years. The last time it halved was in 2016, and the next one will be in 2020. Notice that this is an inverse exponential: even much before than 2140 (about when the last coin will be mined because fees will go under the minimum unit of 1 Satoshi), the Bitcoin reward will be so minimal that it won’t count for much. Miners, therefore, will have to be compensated by the mining fees. Fees are paid per transaction byte, and so a larger block would warrant more fees. But this would be negated by also taking longer to mine. On the other hand, lower-cap blocks will promote a backlog with many transactions waiting to be included in a block. This leads to a fee market dictated by supply and demand which will raise the fees paid to the miners. This is exactly what happened in the 2017 spike.

There are still other reasons against an expansion. Larger blocks lead to larger times for creation and propagation of a block, which ultimately means longer settlement times. Many miners are still not capable of handling blocks of that size.
They also increase the storage costs for full nodes, and this can lead to loss of nodes and concentration of too much power in a few players. If these decide to collude, or are all corrupted by an adversary, the security of the network fails. But this may be also a problem with current block limits. Currently, the blockchain is around 160Gb. If you want to run a full node, you have to hold this volume. Sure, that can be stored locally in a home computer, but the chain is poised to grow, and if the Bitcoin network grows exponentially, so will the space needed. This will discourage people from running full-nodes, which will decrease mining capacity and make the problem worse. Will we end up in a situation where the network will be dominated by relatively few miners, opening the possibility of a realistic 51% attack?

I’m more excited by the Lightning Network, which effectively reduces the number of transactions that need to go on the blockchain. Nevertheless, I don’t think I understand the concept fully enough. From what I get, the LN works by opening a channel between two parties, and allowing them to make many successive transactions until one or both parties decide to close the channel and commit the final balances to the blockchain. This is great for repeat customers with frequent transactions but does not seem to solve anything for occasional purchases. What would be needed (and forgive me if the LN can already do that) is for a merchant to open a channel that can be accessed by multiple buyers at the same time and close it, say, at the end of the day, or every hour. On the other hand, keeping a channel open for a month because that one customer only makes one purchase a day means that you are mostly not using the blockchain, but something else.

In the end, I think there will be no solution to the scalability problem without drastically reducing what is kept in the blockchain. We may even have to come to the point where we don’t keep the full history any more, but only witnesses of past events (blocks), or intermediate checking points that serve as virtual genesis blocks, avoiding the need to check and download the whole history. Perhaps this is the purpose of side-chains (which I have not investigated yet). Perhaps the idea of a fully decentralized blockchain will be contested and will evolve into a federation of chains, with sub-networks validating each constituent. The problem will then be at the intersecting points, and on how trust will be achieved here.

Externalities

There is another point of concern to me: the environmental impact. Bitcoin requires a very high energy consumption that miners, and ultimately users, have to pay for. According to the Digiconomist, Bitcoin mining now spends about the same energy per year as several countries, like Algeria, Greece or Israel. Not micro-countries, either. There is an ethical debate here: is it right to consume this amount of energy only for calculation to support a money transaction network? How much more will this expenditure grow as the network grows, and what impact will it have on natural resources?

This may be a breaking factor, and one that is pushing countries to adopt measures. China has invoked this worry to curtail the actions of miners in the country (although other reasons underneath it may have to do with regulation) and Canada / Quebec, that initially placed itself to absorb the miners fleeing from China, has also now stepped back on its support. Proof-of-work is computationally very expensive, and by design will gradually become harder as we go. But it is only deployed to enforce trustless decentralization. If some breakthrough (probably in cryptography) came about, Bitcoin would be cheaper, faster and would suddenly be much more acceptable for end users. It would also be much less ethically controversial. Such one-off innovations are not to be discarded. Once, Public-Key cryptography (at the basis of all wallets) was also unimaginable.

Finally, what about regulation? In what way will this regulation manifest?

Coming from the tech side, this is a subject I don’t understand very well, so take what I say with a grain of salt. The argument goes that “Governments fear Bitcoin, and therefore seek to regulate it”.This probably takes the guise of taxing cryptocurrency gains, requiring Know-Your-Customer and Anti-Money-Laundering measures from the exchanges and establishing rules, similar to those in the financial markets, to what the issuers of tokens and currencies can do.

I think the one crucial point here is the anonymity of Bitcoin. The government does not know its users, it can’t see their transactions, and so can’t control them. This was, at least, the libertarian promise that sustained Bitcoin at the start of its life.

But I don’t think this is all that true anymore. With wide acceptance, users are now directed to exchanges in order to obtain their Bitcoin. It’s less shadowy and users are protected, but their privacy is compromised.

It has been said that Bitcoin is used by criminals to evade taxes, escape embargoes and trade stealthily. But that is not because of Bitcoin’s architecture, it’s in spite of it. Bitcoin is a fully public ledger that everyone, including governments, can inspect and trawl through. Spending patterns can be detected and transaction networks can surely be identified. What is not easy to do is the link between the blockchain and the real world, because all identities in the blockchain are public keys of some sort (disguised as wallet names). The link betweek key holder and person name is broken. However, there are many ways in which users can restore that link, for example by announcing to the world their wallet in order to receive payments or trading through an exchange that must know their identity.

Actually, if governments really want or need to trace someone, the blockchain is the ideal tool for that: every coin can be traced from its origin through its many holders to its current wallet. This would greatly help against tax evasion, corruption and money-laundering, so I can see governments investing a good deal on de-anonymizing the network. KYC and AML regulations for exchanges is an obvious step, but I’m sure there will be other covert techniques that will not be publicized.

What is the impact of this in the future of Bitcoin? Let’s take for granted this regulation will take place, and that also similar to SEC regulations. Economically, I don’t see why this should be a bad thing for regular consumers. But those on the fringes and concerned with their privacy may well break away and seek other means of exchange. Such means have always existed: cash, gold, favours, service. But Bitcoin has the conveniency of being easy to acquire and to exchange at a distance.

I don’t think Bitcoin will be made illegal as a means of exchange. If it did, it would probably not cease to exist, but go underground back to the Dark Web. And some other obscure cryptocurrency (now that the technology is understood) would emerge somewhere, stay under the radar and keep servicing all the same illegal needs.
Regulation, then, will likely allow Bitcoin to be legal but enforce limits and responsibilities in its public trading. Although it may well be bad for speculation, in the long term it will add credibility to the coin and possibly facilitate its wider adoption.

That’s it for tonight. But let me know your thoughts, don’t be shy to comment.