CRAB Cardano stake pool: retiring! running a Cardano stake pool likely makes no financial sense

2020-10-13 · computing

I’m sorry to announce that the CRAB Cardano stake pool is retiring. Having followed the development of Cardano from a distance for a couple of years, I was pleased to be a part of the launch of Shelley mainnet 76 days ago. Since then, I have invested weeks of full-time work in the research, setup, testing, debugging, documenting, and promoting the cryptocurrency proof-of-stake pool. On top of that, I’ve had 6 production (and for a while 6 test) servers in my network, along with constant monitoring and maintenance. To date, I have earned 1418.538801 ADA (~ 136 EUR) as a stake pool operator for this significant work (likely 1 EUR/hour or less). Not only that, but I believe the situation regarding operating a Cardano stake pool from the Shelley era was somewhat missold, the situation which has emerged being far from what it seemed like it would be in the run up to the Shelley launch. I have decided to cut my losses, and retire the CRAB stake pool via the proper multi-epoch notice procedure; delegators should thus have time to redelegate elsewhere and also to receive accrued rewards.

reliability vs staking rewards

Cardano is proof-of-stake, which I consider to be of critical long-term importance. I do not see how an energy-hungry proof-of-work system such as Bitcoin improves the world, given that many environmental systems are on the verge of collapse. I was pleased to join the Cardano effort not only because of this and its innovative technologies, but also because the pre-Shelley-launch information stressed the need for experience in operating systems reliably and skilfully. What has transpired, however, is that these skills seem much less important than small differences in staking rewards. These rewards are in turn heavily influenced not only by large delegators (including IOHK/IOG, Emurgo, and Cardano Foundation), but also by small fluctuations in block-minting according to the Law of Small Numbers. It’s become clear even from my experiment since reducing the pool pledge to zero that many delegators either simply do not understand the system, or simply do not care.

Shelley is still far from decentralised

Despite the grand aims for the launch of Shelley, Cardano is still far from decentralised. This is for three main reasons: the network decentralisation parameter is advancing slowly on a sliding scale likely over a period of several months (currently 42%); IOHK/IOG launched with a large number of stake pools of their own (around 20), meaning in fact they dominate both sides of the fence rather than relinquishing any significant control; many of those fortunate enough to have launched well into the Shelley era have created multiple stake pools, meaning there are only a few operators controlling most of the network. I do not blame those stake pool operators for this; after all, one of the ideas of Cardano was that the interests of profit and the network would be aligned. In addition, IOHK/IOG themselves set the tone for the era, by launching so many stake pools themselves, so I do not consider criticism of other operators merited. Nevertheless, Shelley is far from decentralised.

Cardano Foundation gave some SPOs a huge advantage

The Cardano Foundation, which also holds a collosal amount of ADA, has effectively chosen who the approved profitable stake pool operators will be in the near future, by delegating vast amounts of ADA to selected pools. Undoubtedly, this helps to decentralise the network. But for those operators who haven’t yet been selected (the delegations appear to be updated every so often), the large amounts mean that they have an extreme disadvantage, no matter how much of their own work or funds they put into building and promoting the pool. Delegations of over 64M ADA are enough to generate blocks consistently, helping to break even or turn an early profit, and punishing mistakes far less severely than those unsupported by large delegators.

stake pool operator rewards are small

With the race to the bottom in pool margins, even 1% pool margin is seemingly high enough to put many delegators off. After all, there are many large pools with low pool margins, and these of course put pressure on smaller operators to match or exceed their offers. But even 2% or 3% isn’t much by way of return if blocks are not made consistently, and making blocks consistently is impossible without high amounts of stake. Of course, a small stake pool operator such as myself might try to undercut by setting pool margins to 0%, at least until they’re more established. This indeed does seem to help in attracting more delegators. But that also means that only the pool fee will be paid (340 ADA/5 days; ~ 33 EUR/5 days). That on its own might be enough to run a stake pool—but those rewards are only paid in epochs where blocks are made, making the actual rewards even smaller. It’s unclear how it is expected that an experienced SPO’s time will be recompensed, even if the server costs break-even.

staking rewards are also fairly small

Even for those not running a stake pool, Cardano staking rewards are also fairly small. On the surface, it would seem that delegators can earn around 6% on their ADA, which is not bad considering the state of the financial system in many countries. But that doesn’t account for the inflation of the cryptocurrency, accounting for which the rewards drop to around 4%. Of course, this is typically an optimistic long-term scenario; short-term, bad luck and missed blocks can decrease these rewards further. But in the midst of all the scrabbling to maximise even a fraction of a percentage point of interest, it’s useful to remember how much money this actually is. For example, if you have 1000 EUR, the difference between a 4% and a 1% yearly return is only 30 EUR.

And crucially, this is where the current state of cryptocurrency regulation in many countries can cause a far bigger issue: most countries require that such staking income be taxed, some even at the point where it is earned (rather than converted into fiat), and unlike bank interest (minuscule though that is), tax of course cannot be deducted automatically. And so, unless you have a huge amount of cryptocurrency staked, properly accounting for, declaring, and paying tax on those small amounts during the year could easily mean your or your accountant’s additional time exceeds the amount earned. For relatively small or even medium amounts, you might like to consider whether it is even worth the effort compared to buying and selling cryptocurrency like a security.

where should delegators stake instead of CRAB?

This depends on a few things, including how much ADA you have to stake, what your main goal in staking is, and where you’re tax-resident. If your goal is to support the community even without making a profit, choose a small stake pool operator who is unsupported by either big donors or the Cardano Foundation. If you don’t care whether you make any blocks, pretty much any size pool might be acceptable, especially as if a lucky block is awarded, you might make a lot on a particular epoch (but not long-term). This happened to CRAB in Epoch 213, when 3 blocks were made despite having only 381K ADA in active stake.

But if your main goal is to make a small amount of money as ‘interest’, your best bet is probably to go with the largest pool that is not saturated (or close to it), especially those chosen by the Cardano Foundation. Any stake pool with under 5M ADA or so is likely not a good use of your stake, in this case, since there’s a good chance no blocks will be made. Regardless, make sure to stake with a pool which is actually making blocks, even infrequently, as otherwise, there’s no way of knowing whether a pool has just been unlucky or whether in fact it’s set up incorrectly.

If you have a large amount of ADA to stake (millions of ADA), you can afford to support a smaller pool, since your stake will make a difference in profitability. If you’ve seriously got a lot, though, you might like to consider just running a pool yourself—or paying somebody with experience to run one for you. If you’ve not got much stake, and are tax-resident in a country which considers staking rewards as accruing tax even if they’re not sold, or which imposes some non-trivial reporting requirement, you might like to consider not staking at all; the amount you’ll lose in time and fees, whether you file the taxes yourself (which I would advise against) or through an accountant, will likely be far more than you’ll make from staking with Cardano.

what about the future of Cardano?

You might reasonably think, reading this post, that I am now anti-Cardano. That is by no means the case. I think it’s a very promising technology, which I at present continue to hold a significant investment in, and which I watch actively. This is despite the cult-like feel of some areas of the community, and the extreme confidence in ADA becoming the dominant currency of the future (it’s a possibility, but I certainly wouldn’t call it a certainty). ADA appears to be a good cryptocurrency—but there are other good cryptocurrencies, too. And perhaps more importantly, even the best cryptocurrency might never achieve widespread adoption unless financial regulations change rather a lot. But my engagement at this time with running a stake pool is at an end, as will be my support of the open-source independent Cardano Node Docker container images (unless a reason arises for me maintaining them). I won’t test software for (close to) nothing indefinitely, and there’s likely far more money to be gained from other forms of investment (or even traditional employment).