Is BTC really facing security issues?

Author: DaPangDun, independent researcher Source: mirror

On July 9, Justin Bons posted a long article on Twitter, stating that "BTC's security model has been destroyed" and argued according to his logic.

I am very interested in this view, so I tried to analyze this view and arguments.

Justin Bons' logic

First, I sorted out Justin Bons' logic in this tweet:

1. Hash rate is not equal to BTC's network security. The cost of producing hashes (and the cost of miners) is the guarantee of BTC's network security. The current network security is even lower than three years ago.

2. With the reduction of block rewards caused by halving, the extreme situation of high fees on the superimposed chain is unsustainable, and miners' income will and is decreasing.

3. Therefore, to maintain the current level of security, BTC prices need to continue to rise, which is unsustainable.

4. The dilemma can only be solved by canceling the 21 million hard cap or forking.

However, we also need to realize that changes in mining hardware have indeed reduced the cost of obtaining unit computing power. The following figure shows the models and functions of Bitmain's BTC mining machines over the years:

Based on the above picture, I drew a graph of the evolution of machine efficiency, as follows:

You can see that the efficiency of the machine is getting higher and higher. The current equipment is 30-40 times more efficient than 10 years ago;

We also need to note that with the development of energy mining technology and the diversification of energy forms (green energy, etc.), the cost of energy is also decreasing, which will further reduce the mining cost of miners.

Therefore, the security of the overall system cannot be measured entirely according to hashrate. So is the cost of miners a good indicator of system security?

>Miner Cost

Miners are the most important role in "BTC distribution", so the total cost of miners' real and continuous investment can indeed be regarded as the "safety valve" of the network. If you want to attack this network, you need to spend enough cost to achieve the effect, and this attack cost is directly related to the investment cost of miners.

The calculation of miner costs is more difficult because it includes "mining machine fees, maintenance fees, electricity fees, etc.", so we make some assumptions in advance to simplify the process:

1. Mining machine fees and maintenance fees are not considered for the time being, because the former also involves depreciation, discounts, insurance, etc., and the latter is related to the automation level and management level of the mining farm, so only electricity fees are considered (electricity fees are likely to be the largest part)

2. The mining machines in each period are calculated according to Bitmain's best mining machines at that time

3. Electricity costs fluctuate, and here we simply calculate it according to 0.12$/kw

The calculation results are as follows:

It can be seen that the estimated cost consumed by miners is still on the rise overall. Therefore, if the cost of miners is considered to be an important indicator affecting network security, the current network security is still on the rise.

Why does Justin Bons say that security has decreased? Because he uses the income of miners as a measure of security, and the income is only a judgment for the future. The expenditure of miners should be the guarantee of current security.

>Ratio of cost to market value

Of course, there is a problem here. Let's consider a situation:

Use 1M cost to protect 1B assets or use 10M cost to protect 100B assets?

Some people will say that the ratio of 1M/1B is higher than the ratio of 10M/100B, so it is more secure; others say that the total amount of funds of 10M is large, so it is more secure.

Both make sense. My personal opinion is:

When the cost has not reached a sufficient level, the cost will be a better measure, which is a threshold issue; if the cost is already high enough, then the ratio of cost to market value will be a better measure.

Miners' income

If the miner's cost is the most important factor affecting network security, then the miner's income is the most important factor affecting the overall miners' ability to maintain such security.

Miners' income mainly consists of two parts: [Block Reward] [Transaction Fee Subsidy]

>Block Reward

Block reward is relatively simple. We all know that it follows the halving rule. There have been four halvings, corresponding to the following:

1. November 2012, the first halving, the reward from 50BTC→25BTC

2. July 2016, the second halving, the reward from 25BTC→12.5BTC

3. May 2020, the third halving, the reward from 12.5BTC→6.25BTC

4. April 2024 In February, the fourth halving took place, and the reward was reduced from 6.25BTC to 3.125BTC.

Because the cost is calculated in $, we also convert the income into the corresponding $ for calculation. In order to simplify the calculation (reduce the amount of data), I made an assumption:

• Take the average price of BTC for one month (the average of the opening + closing prices) as the BTC price data for each day of this month

• The average block interval is calculated as 10 minutes

The following figure is obtained:

Combining these two pictures and comparing them with the graph of electricity expenditure, we can see that:

• In most of the time, the rewards for miners can cover the costs and make a profit

• The early profits were very high, even if they were sold at the price at that time, which was also the bonus period for the development of the industry

• In the bull market stage, miners also made huge profits, and price was the key factor

• In the last round, the price of BTC started the main upward trend about 4 months after the halving

• At the current stage, judging from the block rewards alone, the profit level has decreased

>Transaction fee subsidy

Another very important fee for miners is the "transaction fee subsidy", which is also considered to be the compensation for the reduction in block rewards caused by halving in the future. This is easy to understand. If the block reward is reduced, then as long as the transactions on the chain are more active, the miners' income can still be guaranteed.

Of course, we also need to consider another factor, which is the gas corresponding to the transaction. If there are too many transactions, the gas will naturally be high and the transaction fee will naturally be pushed up.

Let's use data to look at the actual situation:

From the comparison curve, we can see that:

• The activity on the chain is on the rise in the general trend, but whether it can be sustained is still unknown, because the on-chain activities from 2016 to 2023 are similar;

• If we want to calculate the specific transaction fee subsidy, we need to multiply these two data and process them. Because the amount of data is large, I will use another indicator to observe: [The proportion of transaction fee subsidies in block rewards]

  

  I roughly sampled the data and calculated it, and got the following chart:

  

  Combined with the block rewards, we get the following picture:

  

  Combined with the miners' expenditure, we can know:

  • Because the miners' expenditure level is rising, the miners' income is indeed decreasing

  >BTC price

  I don't think this should be discussed, because the assertion that "BTC prices have been rising" should be easy for rational people to judge that this is impossible. It is impossible for the market value to exceed the world's GDP, right?!

  Although we hope that BTC can continue to perform well in terms of price, it is also restricted by "objective laws". The increase in BTC prices in each round is decreasing, which is an obvious fact.

  Since the price increase is limited and the miners' income will indeed gradually decrease, the expectation of maintaining the current security level may change, which will ultimately affect the security of the network.

  Therefore, the problem mentioned by Justin Bons is real, but whether the solution is to "remove the hard cap" or "fork the big block" as he said, I think this needs another discussion.

  Discussion of solutions

  Miners' income is affected by two aspects: cost and income. We have some solutions in both directions.

  >Cost

  What can be done in terms of cost include:

  • Further improve mining efficiency, develop hardware, optimize algorithms, and further reduce the cost per unit of computing power;

  • Strive for lower-priced energy. Of course, stability also needs to be considered;

  • Look forward to changes in world energy technology. For example, if commercial controlled nuclear fusion is developed, the price of energy will be greatly reduced, which can reduce costs;

  Of course, we also need to realize that miners will inevitably gradually shift from the previous "era of making money by mining" to the "era of mining game". Investors need to fully consider the risks and input-output results before deciding whether to participate.

  This is very similar to our current situation. Everyone will feel that it is difficult to make money because the industry has passed the early bonus period and the competition has become very fierce. It is not the era where you can make money by buying anything. Gambling will be the main theme.

  >Income

  Block rewards

  This part is temporarily unchangeable. According to the halving rule, the income of miners will gradually decrease to 0. This is an indisputable fact. One solution is indeed to "eliminate the hard cap of 21 million" and then set up "long tail rewards". But for the current cognition, it is extremely difficult. BTC's 21 million consensus is extremely powerful. If it changes, it will have a huge impact on the consensus.

  Transaction fee subsidy

  Therefore, the natural idea is to make a fuss about "transaction fee subsidy" and increase this part to make up for the loss of block rewards, and add a certain increase in BTC price to allow this system to operate as long as possible.

  However, the performance of the current BTC main network cannot support large-scale transactions. Gas and block time are both problems. Possible solutions are:

  • BTC block expansion: Make blocks bigger and improve efficiency. This is another solution of Justin Bons; but after the "big and small block dispute" that year, the right to speak was basically controlled by "small block advocates", so it is difficult to start such a plan at present. Of course, it is difficult for me to evaluate other impacts brought about by this plan, such as the impact on the "degree of decentralization", etc., which requires a deeper understanding of the underlying industry.

  • Lightning Network: Make the Lightning Network more popular, so that it can replace part of the world's financial payment, and then add technological innovation to allow the transaction fees on the chain to be shared by many people. The current number of main network transactions per day is about 300,000-500,000, and the number of transactions on Alipay alone on Double Eleven can reach 9.7 billion. The order of magnitude difference here is too large, and there is still a lot of room for development.

  • Expansion of the BTC native network: The current function of the BTC network is still very weak. Based on its expansion innovation, whether it is on the chain or off the chain, as long as the main body increases the number of transactions on BTC, it will contribute to the transaction fees.

  Of course, there are some other ideas, such as: for some joint mining projects, the computing power can be operated in parallel, and additional rewards can be obtained while mining BTC...

  In general, my opinion on Justin Bons's point of view is:

  Hash rate is indeed not the only indicator to measure BTC security. The cost of miners can better reflect the security of the network. The current security has not decreased; with the reduction of block rewards caused by halving, the high fee rate on the chain is unsustainable, and the price of BTC cannot rise indefinitely. The network security problem is real and needs to be discussed. However, it is not only "canceling the hard cap" and "forking" that can solve it. At least for now, there is a certain time and way for us to try to solve this problem through other innovations. Therefore, it is too early to assert that "BTC's security model has been destroyed."