Topic: What happens if pools try to maximize fees by congesting the network? (Read 491 times)

I was remembering the term for that person, who knows all the doors to bad portals/methods but still chose to go against them, Like a white hat hacker. So that i could call you that. But then i looked at your name an you already saying blackhatcoiner. Well, you raised a great point, TBVH, i had no idea of artificially increasing the pool fee by making a single group just like many shops in the market grouped together and artificially or by protest, want to increase the price of some goods.

Well, i went curious now and i have to do a short research about it and i will try to cover an article on it too. But for now, thanks for the new topic to keep me busy.

Well, you are suggesting, that miners if collude to manipulate fees then why do fee payers can make collude and come together at a single point, so that they only pay a fixed fee, in that way, people can also manipulate the market fee because then miners have to validate those transactions. But AFAIK, still in both scenarios of Miner Collusion and Fee payer Collusion, both need solid connectivity in between them and they are totally decentralized and anonymous to each other then how can 1 contact another, so that they could convince him/her to collude? Well, the possibility of this happening is very low because in this way those pools will lose the percentage that you aforementioned because those who value the BTC and BTC blockchain will leave those pools and as a result, their holding of 69% will decrease.

So everyone who's part of Bitcoin, one way or another. So mining pools included.

You can't kick anyone out of a censorship-resistant, permissionless network.

Completely agree. That's why I made this thread. I just don't think it's entirely against their benefit to pretend there's network congestion.

I'm sorry for not making it clear, but what makes the Bitcoin community includes the developers, the miners, the economic majority, and our fellow users.


I don't think it would be OK for the majority of the community if mining pools themselves turned into dishonest operators in the network. Kicking them out would easily get community consensus.


?

---

The Game Theory in Bitcoin revolves around incentives. It's what makes everything in the network stick together.

Before actually colluding, the pools will be asking themselves, "Would it be worth it to risk long term incentives for a few moments of high fees"?

Yes, but undermining the system their own wealth relies on outweighs the advantage of controlling which transactions are allowed in, apparently. Also, attacking the Bitcoin network can happen for a temporary time, because the miners will switch pool, and it will have serious financial damage to whoever attempts to do it. I don't hold my breath that the same applies if they pretend there's network congestion.

The "community" can't kick mining pools out of the network, only revenue-sharing miners who are mining at their pool can do that. But the problem is, why would they have the incentive to do that, if they too are earning a bump in pay rates from the increased transaction fees?

Ultimately, the game theory design of Proof of Work only includes miners as participants in this game, as how it was envisioned in the whitepaper. So for full node users to wrest control of fees, they would need a second layer to govern. But however it ends up, it is certainly much better than using PoS (where only exchanges and rich entities have the absolute power).

The end-result would still be the same. The community will kick them out of the network, and it would have been better for "those attackers" to be just honest and be paid in Bitcoin. All that money spent for buying ASICs and to pay for electricity = absolutely wasted.

Theoretically, yes, major mining pools could orchestrate such a scheme, artificially inflating transaction fees. But, lets pull back a bit and examine the broader landscape.

This strategy would be tantamount to shooting themselves in the foot. Sure, they may enjoy a short-term windfall from higher fees, but what about the long-term implications? Higher fees could discourage Bitcoin usage, possibly damaging the very ecosystem they rely on. Not to mention, any noticeable shift in transaction patterns could raise alarm bells among observant network participants.

Moreover, such a tactic would require a level of cooperation that's practically unrealistic. Remember, these pools are competitors in a cutthroat industry; the incentives for defection are strong.

maybe for a while pools are happy that there is congestion on the network because the fees are going up, but this is only temporary, it will not be possible for them to maintain this for long considering that it will take quite a lot of resources for them and with increasing transaction fees will also discourage people from using bitcoin and that will disrupt the market.

so even though they want the fees on the network to go up so that the rewards they get are also high, they also have the thought that this will slow down the adoption of bitcoin and annoy people on the network, for example like in the case of yesterday's Ordinal where Bitcoin fees didn't make sense and pools seems to be enjoying this.

I think someone or people who are smart enough to to control that much hash rate won't be too dumb to know that this won't be economically viable long-term unless they aren't motivated by profit but to attack the network.
If that happens, many "customers" will likely go somewhere else where transaction is cheap or stop using the network all together and demand will reduce making the transaction fees to become low again... This could happen in just two weeks. Will they be willing to risk more than 1month of low demand for just 2weeks of illegally hiking the fees?

If 69% of the hashpower is going to successfully collude to control which transactions are valid and which aren't, then there is no need to waste time and effort spamming the network with fake transactions. With more than 50% of the hashpower, they have full control over which transactions are in blocks and which aren't.  They can simply choose the transactions that pay very high fees, overwrite any block from any other pool or miner that doesn't do the same.

I don't believe that there's only a "few people" who want to acknowledge it. Mining pools are centralizing, but it's part of the evolution of the participants in looking for more efficient methods to do their job for the network as well, like the development of ASICs. It's unpreventable, especially ASICs.

But there are developers like Matt Corallo who is developing BetterHash, which would give back the miners control over their hashing power, and pools would be mere coordinators and distributors of mining rewards.


It wouldn't be that simple because everything is about the incentives and being incentivized. Would a group of miners/pools really want to attack the network and risk to be kicked out of it? Or would they prefer to be honest, keep mining Bitcoin, and be rewarded for it?

Your topic has certainly identified a key weakness in Bitcoin's centralisation as a network. Few people want to acknowledge it, but even the idea of "mining pools" as a concept is a form of centralisation that was controversial in the early days and still remains a threat today, with a few pools holding the majority of the hash power as you pointed out. All it would take is collusion either way for malicious activity. The one reason pools became accepted and popular is because they were more profitable, as well as obviously allowed access to the market for individuals with low hash rate that would rarely ever gain any block rewards otherwise.

It all reminds me how Ethereum has recently become centralised, whereas the system was designed in an attempt to further decentralise it. Whereas similarly there are just a few pools that dominate and could easily collude for malicious activity. Maybe people don't like the comparison, but given how much maxi's accuse Ethereum of being centralised (rightly so), they do so without looking in a mirror over Bitcoin's centralisation.

Sure, but with majority of hash rate, you'd expect to mine yours more often than the rest of the miners. But maybe you're right; maybe that risk outweighs the benefit, it's just seems to me that in times when the network is clogged up, raising the high fee by a lot could outweigh the risk.

Parents don't need to be large at all. Think of a parent that is double-spent, whose children are Ordinals.

Don't free market principles apply to frankland-utopia?

i find it strange how blackhat makes a topic about mining pool mangers colluding to attempt to raise the fee's .. via a blackhat stupid scenario method of messing with users mempools..

the  easiest, fastest cheapest option is just have them mining pool managers just select exclusively high tx fee's especially ones they include to themselves thus no real cost to themselves thus they end up setting the min fee the network sees as what gets into the next block majority of the time

all of this chatter about needing to mess with users mempools becomes meaningless and can cause excess cost to pools who stupidly would broadcast tx's with fees to the network unconfirmed as it allows the non colluding pools to pick them up along with users other cheap fee tx.

i think blackhatcoiner was told a stupid scheme by a buddy he knows, but doesnt understand the process himself..  and is asking the community how effective it is and ignoring the basics of how bitcoin actually works to not understand how futile his scenario is in comparison to much easier and cheaper and more effective methods

in short. pools dont need to mess with users nodes database of zero-confirm transactions..
pools just choose to only include high fees in their own blocktemplates.. job done

edit to respond to below
dont you understand that your forum wife pretends to spout "freemarket" like a libertarian loyalist(when he is on a recruitment drive to find people to adore him).. but he is really just playing you.. because its a false excuse to make everyone pay more.. and there is no actual free market..

heck when you ignore his pretend libertarian recruitment buzzwords and listen to what he says once he thinks he has enough recruits for the week. he changes his narrative..
EG when he adores the fee mania and wants people to pay more.. thats capitalism not libertarian freemarket
and no capitalism is not a freemarket stereotype

try not to rely on your forum wifes buzzwords. actually take some time to learn how bitcoin works

If you allow enough time before invalidating it you risk another miner mining them just 5 seconds after so you just lose money and achieve zero.


20 was just a thrown-out number because I didn't know the size of the parent, but still no!
The most important thing is that you're trying this to block the tx from getting confirmed but at the same time rise the median fee of the next block and it doesn't work:
- if you throw like you said 201 tx with the above fees they will simply get confirmed in the next block and the rest of the fees in the mempool will have no time to pick up, but you just wasted 1 btc to the competition
- if you push 1000 highly paid tx at 200sat/b and you try chain them to the mempool forever with parents twice their size but with 1 sats the mempool will not show 200sat/vb since that tx are ~66at/b and they will still get confirmed as they wills till be the first in line.

Either you try to maximize the fees and you risk having them confirmed in minutes, or you either chain them by reducing their overall fees in which case they will obviously not raise the mempool fee average or next block fees.
Again, you can't rise the fees for the next block estimation without putting their tx that will get confirmed in the next block, so gifting bitcoins to other mners for nothing!

It will be far easier to just mine for like 12 hours empty blocks blaming some glitch and then "solve' the issue!

But that would be late. People would have raised their fee rate, and they won't be able to lower it, which is what's this attack all about.

How so? The parent transaction is meaningless in size, comparably to the children. If you have 1000 tx paying 200 sat/vb, and 1 tx paying 1 sat/vb, it's obviously in favor of the miner to include all 201. Why would the explorer show 20 sat/vb?

Invalidating transactions it will mean it will be dropped out of every single node mempool, so leaving the mempool empty again.
If you try to force the chain of child and parents to remain unconfirmed by a low fee from either of them it will bring the other transactions fee down also.So you can have 1000 tx with 200sat/b if you want to get them stuck like this by having either a parent or child with 1 sat/b it will still bring the average fees down also, so you will be having 1001 tx with  20 or whatever sat per byte and every mempool explorer will see that.

You can't get a valid traction that pays over the next block limit stuck in the mempool, there is no way for that other than all miners blacklisting them.

Just for clarity: any post related to "cult groups" and "Bitcoin Core dev team" derails the thread and will be deleted.

I already told you, I'm not debating what they can or cannot do technically. I'm already asking under the assumption that they can do it.


Hey franknbeans, there's a massive difference between the scaling debate and what was at stake during that era, and what's being discussed in the hypothetical situation that OP is asking.

It's a valid question. How sustainable is it when the community sees such dishonesty, and what assurances are there that the miners will continue pointing their hashing power towards dishonest pools?

imagine this scenario

for 18 blocks the total fee is 10sat/byte. meaning an average block is 0.15btc in total fees for 1.5mb block
meaning over 3 hours they get 2.7btc

now imagine for the next 18 blocks the pools collude by creating their own self confirming transaction paying to their own destination where because they do not broadcast pre-confirm to the network. they are the sole includer of their transaction in their block meaning their fees come back to themselves. so no cost
the asics of their pools still get shares of the main 6.25btc so they are happy

and now the colluding pools create a 3 hour situation of only including fees of 40sat/byte of their own self paying transactions)again reminder NO COST)
thus in that 3 hours . USERS start to see delays in users getting confirmations and so users start to do RBF to increase their fee to 40sat+/byte min

now the pools dont have to fill their blocks with only their own transactions when users are now starting to pay that minimum.. so they can include some of there own and users transactions of 40sat/byte plus until its majority users transactions of 40sat/byte tx in a block

now they are starting to earn again from users. but still ensuring they set min confirm fee of 40sat/byte. which continues to make users pay that amount.

and soon you are realising when user pay this amount, the pool is getting 10.8btc per 3 hours instead of 2.7btc and the pool does not have to bloat a block with their own transactions becasue now the users are paying their collusion high fee requirement

---

as for "can the network stop this"
arnt you part of the idiot-speak cult group that dont want to stop high fee's.. you know the guys that want cheap fees rejected unconfirmed and want a "freemarket" of paying premium..
arnt you also part of the same idiot brigade that doesnt want to reject dishonest stuff.. such as ordinals scams that has caused a fee mania via a different technique but your cult wants to continue

yes technically there are many ways to stop it but i am 100% sure your cult and the core devs wont implement it

I'm not debating if they can or cannot do it. I'm merely asking if it's sustainable, and if the pools can actually get away with such dishonesty towards the community. How long until the developers, the users and the economic majority, and the honest miners and pools come together to find a way to kick those pools out of the network? It might start another movement within the community again.

Would the top mining pools take the risk for temporary additional profit? Or will they take the game theoretically smart approach and simply be honest?

Why 250 bytes? Let's take 142 bytes. I made a typo, should have written 100 sat, sorry.
Totally, transaction fee per the smallest transaction would be 0.00011BTC, i.e. 3.52 Bitcoin totally (0.00011 * 32,000). At the moment it will clog mempool because fees are low but 200 sat/byte or 400 sat/byte is no joke for anyone.

It was a typo, sorry. I already wrote above and edited my post.

First of all, 32,000 transactions which pay 0.001 sat/vb isn't a problem. Even if full nodes didn't deny to propagate your transactions, they'd have the least priority. My hypothetical attack is all about pretending there are people paying for priority (when there aren't). Secondly, 0.001 sat/vb is so low, that your transactions will probably never reach the miner. (unless you drop them a message, and then again, I don't think they'd accept as it's really strange)

you dont need to clog the mempool of users to cause fee mania. as proven dozens of times over the years.. the mempool clog of users nodes is the reaction not the cause.
the cause is what pools allow into their blocks

and by the way.. 32,000tx of tx that are ~250bytes. meaning 0.25btc a tx meaning 8000BTC for a 2 hour non event that dissipates after two hors thus wasted 8k btc for a non event..

unlike other fee mania methods involving transactions that do get into blocks. thus can cost a pool nothing if organised right. and can sustain for far far longer

mining pools dont need to first attack users mempools to cause fee mania

however if the topic was "what if core try to maximize fees by congesting the network?!"
then the answer would be doing what they are already doing by messing with the users mempool and relay limits to reject/prune cheap fee's to never reach a pool, causing everyone to pay more just to get seen by a pool

To be honest, it's entirely possible even for one man to clog the whole bitcoin mempool with some tens or hundreds of thousands of dollars, depends how long do you want to ruin the party and then depends if that lit turns into fire.

Just small statistics before I show you an example:

• Roughly 6 blocks are mined each hour.
• Roughly 2600 bitcoin transactions are confirmed in one block, that means 16K transactions are confirmed an hour.

For example, make 32,000 transactions offline with 100,000 sat/vB and broadcast them all at once. You are gonna clog the whole bitcoin mempool for 2 hours, guaranteed!

---

By the way, those big mining companies aren't good guys, they manipulate transaction fees, sometimes. They can cooperate too but if they cooperate, this will destroy the whole bitcoin ecosystem and there will be no winner in this game.

This is not how the free market works. Eventually, there will be pools willing to include my transaction if it only constitutes profit. For example, if they do have a few transactions paying 1 sat/vb, and choose not to include them, and mine empty blocks, then at some point, there will be some miner who will. It will only be loss for them. Simple economics; supply and demand and of block space.

the obvious way. by pools not including your transaction unless your fee meets their desired minimum..
isnt this the whole point of your topic. trying to think of ways to cause fee mania again.. via pools

seems you dont know the basics of most things. because you think its a users mempool that does transaction selections for blocks. maybe you need to realise its actually pools that select the transactions that go into blocks. not users nodes.

but i did find it funny how your scenario in this topic is you trying to mess TEMPORARILY with users mempool of zero-confirm transactions and cancelling them before they get into a block. thus have nothing to do with fees of transaction that pools select of what they want to see in a block.. you know, blocks.. you know the block data where real estimate of priority are found.. you know blocks created by pools.. (or maybe you dont know these things)

but anyway you have much to learn so ill give you a break and give you some time to learn. try not to waste it on social drama. and instead actually learn some stuff

No, I mean the opposite. The child is the big transaction, and the parent is just an unconfirmed one. You can invalidate the big transaction, by double-spending the parent.

How can a pool force you to pay more?

As I said, by invalidating the parent, it's trivial to pretend there's bulk of high-paying child transactions, with minimum cost. When you should do it, is another thing to discuss. I don't know.

If those pools doesn't include TX with high fee rate (which created by themselves) on their block, wouldn't people find it as odd/unusual behavior? For example, mempool.space have block audit[1] and block health[2] feature which can be used to check such behavior easily.


Another question, when or how should they decide to invalidate their transaction?

[1] https://mempool.space/docs/faq#how-do-block-audits-work
[2] https://mempool.space/docs/faq#what-is-block-health

If pool try to maximize fees by congesting the network it can lead to slower transaction times and higher transaction fees for all users. This is because when there are more transactions being processed at once, the network can become congested and the miners may prioritize transactions with higher fees. As a result, users who are willing to pay higher fees may have to wait longer for their transactions to be processed. This can lead to a less efficient network overall, as more resources are required to process a larger number of transactions. This could make using the network more expensive and less accessible for everyday users. While pools may incentivize to maximize fees in the short term it is not a sustainable strategy for the long term health and growth of the network.

because when pools force thousands of users to pay more. the pools can ramp down their own high fee tx's to themselves and and instead let high fee users fill the gap. thus making the users then persist the fee mania

EG
imagine the next 18 blocks are transactions of high fee's of pools sending transactions to themselves in their own block templates thus paying the fees back to themselves.. causing fees to show as 60sat/byte minimum.. without costiong the pools anything..

then users will see a 60sat/byte minimum for last 3 hours and so realise if they want to get a tx to confirm within 3 hours(3x(6blocks/hour)=18) then they need to pay that minimum

then when the pools see normal users paying 60sat/byte minimum they allow those transactions in and not include one of their own. ramping fown their own tx involvement becasue now the users tx start to persist the attack for them.. thus they continue letting in users transactions as long as users pay 60sat/byte min

again by pools not zero-confirm relaying their own tx and only including them in their own block template. it stops competing pools stealing their fee's and thus costs the attacker pools nothing to do this, with just a 3 hour of no external fee income until they can push the fee estimates up to the min amount for users to then pay the pools this higher fee consistantly

I think that even if collusion to engineer a congestion was technically possible, how does that play out in those pool's long term expectation to get incentivized? Such a thing cannot be hidden forever, and the discovery of collusion/dishonesty would make the participants lose confidence in the system, which would probably also make them leave and cause a price crash.

From a game theory point of view, it's better for the pools to do what's best for themselves, but also act honestly and do what's best for the rest of network.

by doing the unconfirmed mess with mempool game.. does not cause fee congestion.
the games of cancelling out transactions before the 31% get to add them just makes blackhatcoiners whole game just a temporary distraction none event of a few minutes if seeing a transaction clump in mempool.. then not.

this playing with mempool and cancelling transactions does not cause any fee mania

the real method to cause fee mania is if the majority pools did include high fee transactions into their own blocks but not pre-relay those transactions to others before their block is solved. that way it costs them nothing but ramps up the fee statistics for other users to need to pay more to tempt a pool to accept them

as for blackhats other scenario vector of using CPFP to swap transactions around at unconfirmed status. again. a temporary none event of messing with mempool that lasts only a few minutes

...
it really would have helped blackhat if he spent the last few years learning how bitcoin works instead of trusting his buddy about how features are explained to him

messing with unconfirmed mempool is just small temporary non event that does not result in high fee confirmations. especially if blackhat is trying to avoid having high fee transactions confirmed in a attacker majority pools scheme. which in of itself would be some basic self enlighting answer to his own question about confirmed fees
yep not including high fees in your own block results in no high fees seen in the blockchain

messing with unconfirmed transactions to cancel high fees before confirmation also makes no high fees appear in the blockchain

and broadcasting unconfirmed high fee's to other pools giving them a chance for the minority pools to add them(in a scenario of not cancelling the high fee/high size transaction). just means the attacker pool(majority) loses value by paying its competitor(minority). which wont be sustainable, so again not going to cause a long term viable fee mania game
so again if blackhat broadcasts unconfirmed transactions and then cancels them to prevent competitor pools adding it. would mean it only sits in a mempool for a few seconds thus causes no fee drama.

Blurry eyes.

I still do not get it; you are saying one can invalidate the parent transaction by mining the smaller (in size) child transaction? But to mine the child transaction you do have to confirm the parent transaction as well, that is not invalidating.
Double spending and CPFP are different things.

- Jay -

What's this? Closed For New Participants? 

Both parent and child transactions are unconfirmed, and the high-paying one is the child, so to invalidate the big one, you only need to double-spend the parent. So, you could pretend the mempool is clogged up, but you'd only need to mine one small transaction for invalidating every big child transaction.

Correct me if I'm wrong, but CPFP does not invalidate the parent transaction, it bumps up the fee for confirmation by a miner who wants the cumulative fees of both transactions.
How can one then use the child transaction to invalidate the parent?

- Jay -

in the short term, anything is possible. In the long term, only what is profitable is possible. if a pool starts to mess with transactions and its return on the provided computing power from miners starts to decrease, individual miners will start to move to the competition. the same will be done by miners who will not want to take part in the attack.

if tx cost pumps, the network will start to produce fewer transactions (because users will start delaying transfers or using altcoins to transfer value) and thus sooner or later most of the transactions mined by the network will be transactions generated and paid for by the attacker. and it doesn't matter if the miner will mine up his spam transactions and the fee will come back to him (cost 0) or someone else's on which he earns (profit) and his spam transactions will be mined by another miner (loss) it's  net zero.

net zero when it comes to revenue and a great loss, when you add the costs of running the ming rig to the calculations  (electricity and equipment depreciation)

That' s a huge resources needed to that, i mean money, yes <10 sat/vb is too low but the sum of it sending in every block sounds a suicide to me. Besides the remaining pools will also gain to that move. Business wise, that is a waste of money without having assurance that their pool will get the mining reward.

firstly..
pools dont need to broadcast unconfirmed tx to a network but then not include them in their own block template.. you have it the wrong way around

a productive method of fee mania is the opposite.. its to not broadcast unconfirmed transactions to the network. so that other pools dont add an attacker pools tx to their block.. but only include high fee tx in their own blocktemplate.. that way the transactions dont end up in a competitors block and only be in the creators block thus whatever fee the pool puts into its own transaction it gets back . thus no loss no cost.

if you did do things your way to broadcast unconfirmed to network but refuse to put your own transactions into your own block template.. then you wont get your fees back and you instead would pay some other pool when they include your transactions in their block. thus cost you alot to do it for "thousands of transactions"

secondly pools dont need to use ordinals to set high fee rates. pools can just put their own transactions into their own blocks (not relayed before block solve.
there is no need for ordinals at all to be used for fee mania..

the only purpose of ordinals is data bloat by putting in stupid 3.96mb meme images to stop normal users transactions from fitting into blockspace because its used up

lastly.. many fee estimators base it on a recent set of confirmed blocks fee amounts. so its more about the fee of the last few blocks. rather than whatever transactions are being relayed around which you hope will never get into a block.
so again if you want to cause a fee mania you have to put transactions with high fees into a block

I guess big pools could team up to do that, but the question is for how long. This is not something that would go unnoticed. Many people (well, people smarter than me) would see that there are thousands of bogus transactions coming from that pool, and they would ask difficult questions to pool owners... Many people would react, and i can't imagine this lasting long...

Besides, there's been congestions before, and within 24 hours, everything was back to normal.

It's not about if it's unlikely, because it is definitely unlikely but we all know fully well that pools and miners could do this coordinated attack if they so please. It's all about what would happen, as OP asks in his main post.

Now to answer OP's question. With Network Congestion comes people who would wait for the network to be less congested so they could carry out their transaction, sure they don't count for the mass majority but in a large-scale maneuver like this one, they'd be a great clincher. I'd say that when they do decide to congest the network so much that you'd have to pay 10x the usual price you pay for transactions, millions of cryptocurrency users in the world will mutiny, will not create any transaction for let's say, the next two weeks or so, or would boycott the usage of what they assume were the reasons for this fee increase and transaction delay. Utter chaos will commence within the industry, which will only lead to the general public getting whatever they want.

They'd have less chance than the 69%. Also, the 69% could invalidate them at any time, and at a very low cost, just by making sure they use CPFP. For instance, a 10kb transaction paying 100 sat/vb, requires only a small transaction spending the parent, and invaliding the 10kb one.

That's open for debate, but I don't expect a decentralized network to rely on mutual decisions made by three for-profit organizations. Eventually, there may be benefits outweighing the principles.

Wouldn't the remaining 31% of the hash rate include those 10 sat/vb transactions in their blocks?  Wouldn't that mean that these pools would be sending their revenue to their competitors?  That doesn't sone like a good way to stay in business.

Furthermore, can Foundry, Antpool, and F2Pool really trust each other?  Isn't there a pretty strong chance that one of the pools will agree to participate with the others in broadcasting these high-value transactions, and then not actually send the transactions, but instead mine a bunch of the transactions that the other 2 send, allowing them to take a whole bunch of revenue from them?

It's very unlikely, and against the game theory. But you don't need everyone. You just need a couple of pools, and an organized approach.

Sounds to me only sustainable to a coordinated majority. Minorities can't survive this, because big pools will quickly earn their good-paying transactions.

Maybe it's just not worth the risk of discrediting the top pools' integrity.

It's unlikely that all pools and solo miners can form a cartel to launch such attack, and if there's a minority who isn't a part of it, they will mine those fake high fee transactions, causing a loss. If there's a conspiracy to actually include them in block and return the fee to themselves, they still will lose those fees to miners who are not part of the attack, plus they will miss out on the fees from real transactions.

I'm not going to completely dismiss the idea of fee manipulation, but if it's possible, it's likely very hard on practice and there's a lot of factors involved.

Honestly, I had never thought about this before, or at least I don't remember thinking about it.

Say that Foundry USA, AntPool and F2Pool (which in total hold about 69% of the hash rate according to btc.com) cooperated to pretend there's network congestion, when there isn't. For example, say the median fee is 1 sat/vb. But, they don't like that, so they broadcast a thousand transactions paying 5-10 sat/vb, to encourage some of the users with 1 sat/vb to raise their fee rate. Pools' transactions don't cost them anything, because they don't include them into their candidate blocks. They just take advantage of the wallet software there exists which tells the user to pay more to have priority.

This isn't limited to when there isn't congestion. Pools could do this right now with Ordinals, and broadcast transactions with 100 sat/vb; that would encourage those paying 50 sat/vb to raise fees. To avoid other pools from including their transactions (and pay 100 sat/vb which is pretty high), they could sometime include a transaction that invalidates their attractive transactions (e.g., spends an UTXO from those).

A problem, from the pool owners' side, is that they can get caught easily for doing it. Every person running a full node will notice unreasonable activity (the miner not including tx that pay much), and blocks mined by pools are known.