On May 11, a major event took place in the bitcoin world, yet it had no negative effect on the price of the coin and much less effect on mining than it would seem it should. This event is known as the "halving," and it means the reward for mining bitcoins was suddenly cut in half.

This happens about every 4 years, and is the reason that the total number of bitcoins mined in the future, and thus the lifetime supply, is constrained. It was designed in from the start to attain this goal.

To understand what this is all about, you do need to understand some of the complexities of bitcoin mining. Bitcoin transactions are recorded into the immutable ledger that is known as the "blockchain" by machines (and their owners) called miners. These miners need to be super-high-performance computing hardware, and they are expensive to buy and they need a lot of electricity to run.

If you run one, spending that money, then you (or rather the pool of fellow miners you join) will win at mining a block with some predictable regularity. Each time your group mines a block, it gets the "reward" of 12.5 bitcoins, today valued around $120,000. Mining is a profitable business if the hardware and electricity needed to reliably gain that reward costs less than that $120K. If it costs much less, you have a nice profit margin.

If that profit margin is fat, this encourages people to want to buy more mining equipment of their own, so they can get in on it. It's meant to be naturally regulating, so that people will invest in running mining gear, and get an acceptable profit margin, like any business. It also means the margins should never get too fat.

On May, 11 the reward suddenly dropped to 6 1/4 coins. From $120K to $60K, roughly. Any mining equipment that was not extremely profitable before the halving should have become unprofitable immediately.

It's more complex because there are two costs, and possibly more. There are capital costs, like buying a mining device and a server room, and there are operating costs, namely electricity. Once the capital costs are spent, the question of whether it is profitable to run the gear depends primarily on whether the reward exceeds the operating cost. Even if your gear is old and slow by today's standards, if it can pay for its electricity, you will keep it on. If not, you should turn it off.

For people who pay for their electricity, though, it is the major part of the cost. All the equipment running in real electricity markets should have taken a fairly serious hit at the halving, and on a purely rational basis, all but the most efficient and newest gear should have shut down instantly at the halving.

That didn't happen on July 9, 2017 at the last halving. It didn't happen at all. It didn't happen so much that it's a puzzle. What kept all that equipment on. And this time, it happened, but only to a small extent. The Bitcoin "Hashrate" -- which sums up all the mining power out there dropped from its all time peak of 136 exahash to 92 3 days later. But the longer term moving average today sits at the same level it was at back at the start of April -- there was a bigger drop back in March, for example, with no halving at all.

How can that be?

• Since electricity is such a big cost, many miners have sought places where they either get super cheap electricity, or even "free" (flat monthly rate) electricity. In the latter case they are effectively cheating people who wrote unwise contracts, like housing or offices that come with electricity included in the rent. If you don't pay extra to use more electricity, then there is no reason to shut off your gear.
• The people with the absolute latest and most efficient mining gear can still make money at half the reward. They may be a much larger proportion of the mining community than would be expected
• Some people are willing to mine even at what would be a loss. This can include people who hold large amounts of bitcoin and are willing to lose money to prevent any hiccup in the network which might erode confidence in the coin, and the price. If you hold hundreds of millions of dollars in bitcoin, it is worth it to lose a few million to smooth out the blip.
• Another group that might mine at a loss are people who are in China, where bitcoin use is restricted, and foreign currency transactions are regulated. A bitcoin miner can pay for gear and electricity with RMB, and then sell them overseas for USD -- thus turning RMB into USD at a market rate without interference by the currency controls. A halving is painful, but still might make things worth it.
• Within the bitcoin system, it was expected that as the reward decreased, miners could start demanding higher transaction fees to make up for it. Fees have not increased at a level that would compensate for the loss of mining reward, though. Miners can't "demand" higher fees; instead they can only refuse to record transactions unless they offer high enough fees, and people start realizing this and offering them. There was no announcement of "hey, everybody, after the halving, please bump your fees a lot" and it did not happen.
• Some suggest the miners might be speculating -- holding on to the coins they mine in the hope they go up, like any other speculator. And the price at the last halving was around $650. It's been above that ever since. But most importantly, if it is costing your more electricity to mine than you make in reward, you can simply turn off the electricity and buy the coins to speculate with using dollars. There is no reason to mine at a loss.

In theory, at the moment of the halving, some -- and one would incorrectly predict a lot -- of the miners would shut down. This would instantly reduce the network "hashrate" -- the measure of how much mining power is in operation. Bitcoin is not designed to handle this. If the hashrate were cut in half, it could be a catastrophe -- blocks would take twice as long to be issued, the maximum transaction volume would be cut in half. For the upcoming halving, the difficulty does not recalculate for 1008 blocks after the halving, which would normally be one week, but would be two long weeks at a 50% hashrate. In those two weeks, a lot could happen, including panic.

But again, this did not happen in 2016, not even a little. There was a serious drop in hashrate after the Dec 2012 halving, but that was in a day when bitcoin was fairly immature, and people were in it just for the excitement. In 2016, the hashrate dropped around 10%, but it did not happen instantly (as it would if based on purely rational electricity cost) and it recovered in reasonable time. 10% is not a calamity.

On top of that, in early March of 2020, Bitcoin's price dropped almost in half during the market crash. This was effectively a halving, done a bit more slowly -- and the hashrate did decline but recovered. Indeed, there are those in the Bitcoin world who believe that halvings cause the price of the coin to increase -- and this is a self-fulfilling prophecy of course, as people buy up coins in anticipation of that. The idea behind this belief is that since new coins are minting at half the rate, there has been a sudden drop in the supply of available coins because most coin owners seem to be "HODLers" rather than sellers and spenders. (HODL is a famous typo of hold, and represents a philosophy of solidarity in holding on to coins to help them go up.)

The recent increase in the price of Bitcoins is, like most of them, hard to explain. Some feel it was a flight from the stock market. Others assert it was due to anticipation of the halving, which of course comes with zero surprise and thus will be priced into the market one way or another.