Bitcoin Fees Are Skyrocketing (arstechnica.com)

An anonymous reader quotes a report from Ars Technica: The cost to complete a Bitcoin transaction has skyrocketed in recent days. A week ago, it cost around $6 on average to get a transaction accepted by the Bitcoin network. The average fee soared to $26 on Friday and was still almost $20 on Sunday. The reason is simple: until recently, the Bitcoin network had a hard-coded 1 megabyte limit on the size of blocks on the blockchain, Bitcoin's shared transaction ledger. With a typical transaction size of around 500 bytes, the average block had fewer than 2,000 transactions. And with a block being generated once every 10 minutes, that works out to around 3.3 transactions per second. A September upgrade called segregated witness allowed the cryptographic signatures associated with each transaction to be stored separately from the rest of the transaction. Under this scheme, the signatures no longer counted against the 1 megabyte blocksize limit, which should have roughly doubled the network's capacity. But only a small minority of transactions have taken advantage of this option so far, so the network's average throughput has stayed below 2,500 transactions per block -- around four transactions per second.

Comments

What will the effects be?

[Bruce Perens]

What happens when Bitcoin crashes? What effects will it have on companies that accept, use, or hold it, market-makers on exchanges and futures, etc. ?

My theory is that it was created by a national actor with the intent of crashing national economies. Not sure it will actually do that, though. But real people will be hurt. Some of them will be people who took the risk themselves and deserve the consequences. But when stocks or currencies crash there are often lots of innocent victims who never made the choice to invest in them.

Re:What will the effects be?

[ShanghaiBill]

What happens when Bitcoin crashes?

If you look at the history of crashes, nearly all have one thing in common: Many people investing with borrowed money. This happened with tulips, the South Sea Bubble, the 1929 crash, and the sub-prime mortgage crash. I am unaware of any bubble that did not involve a lot of people borrowing or buying on margin.

So far that is not happening with bitcoin.

What effects will it have on companies that accept, use, or hold it, market-makers on exchanges and futures, etc. ?

I may be going way out on a limb here, but I think they will lose money.

My theory is that it was created by a national actor with the intent of crashing national economies.

All the bitcoins in the world total to $300B. For comparison, the recent financial crisis wiped out $20 trillion.

Also, the financial crisis hit people at the bottom the hardest, people that were struggling to pay their mortgages, and were then evicted from their homes when they lost their jobs.

Bitcoin is different. The people buying can mostly afford the loss, and much of the value is just paper profits. I bought my stash on a lark when the value crossed $1/BTC, mostly just out of curiosity of how the whole thing worked. They have turned out to be worth way more than I ever expected, but if the crash came tomorrow, it would not make one iota of difference to how I live my life.

But when stocks or currencies crash there are often lots of innocent victims who never made the choice to invest in them.

Stock crashes have way less effect on "normal" people than currency, bond, or real estate crashes, because no one expects equities to be stable. We shrugged off stock crashes in 1987, 1991, 2001. The financial crisis was far worse because no one expected housing prices to crash, and a huge part of our economy relied on their stability. Nobody is relying on bitcoin to be stable, or even as stable as equities.

Re:Well, it's a matter of context.

[omnichad]

Not at all true.

Bitcoin is undergoing massive deflation. Which means you're spending fewer Bitcoin because it has a larger buying power. Transaction fees are actually going up (not just in absolute BTC terms, but relative to BTC spending power), because it's computationally expensive to process those transactions.

Re:Unclear Story

[Jonathan C. Patschke]

That's it.

It doesn't have to be that way. If you're running a node, you could run software that packs transactions into a block by any criteria you choose. However, since the aggregate fees for the block are part of the reward for winning the mining operation, it's in the miners' best interests to pack the largest fees first. Also, since the likelihood of any particular node winning the block is very low, we can talk about the most common strategy as if it were universal.

One of the parts of what makes me really nervous is that fees and trading price feed-back into each other in a damaging sort of way. The way that transactions are expressed is that their components must be spent in totality. So, if you're trying to pay 0.01 BTC, and your smallest previous inbound transaction was 0.1 BTC, you have to refund 0.09 BTC (less fees) back to yourself. Over time, this means your wallet gets fragmented with a bunch of spare change which may make some later transaction expensive because of the number of previous-spends needed to be referenced to get up to that amount. The smaller the pieces of change, the less they can contribute because of the fee overhead in referencing them.

Looking even a little in to the future, this is going to ramp up pressure to rely on off-chain transactions, using the blockchain itself more like inter-bank settling than like personal accounts. That's the exact opposite of the initial sales pitch.