There is way to much shit talk in this thread and noones presenting a single shred of fact. Here are some rough numbers to at least shed some light on the matter.

This is an image of a daily study box in December. At the time the best learning ability you could get was around 500%.
The top table is of a person who uses the best possible curio. The total LPH shows the two numbers calculated where in one instance the bolete is replaced by a pearl. The orange highlighted curios indicate possible bot farmed curios as well. The bottom table shows that of a more casual player studying more commonly found curios.
These are some rough calculations of the LPH of a person who have access to some of the high end curios and a person who plays regularly but don't have access to the same high end stuff. The person with the lower tear curios also fills the study less regularly therefor have to think about using curios that last longer to avoid having to constantly fill the study box. The person with the high end curios can also choose to study pearls or boletes. The difference is extreme and very significant as well as require a bot to automatically study.
Clearly it shows how heavy impacts pearls have. With pearls the LPH triples to that of a regular player. Without pearl study the LPH of a very wealthy player is roughly 1.75x faster. Whats not shown on the first study window is how fast some of the very high end curios study as well. If the curios are manually studied twice a day then the efficiency would be reduced significantly and be far less then 1.75x faster. Taking that into account it can be said that the wealthy person who doesn't auto fill study have a 1.5x faster study then that of a regular player. 50% faster LPH can give a specific character more LP but as there is diminishing returns the gaps become smaller over time.
Assuming the 3 types of players would play for a month then this would be there rough estimated LP:
Auto study with pearls: 60 000 LPH * (time in hours) 720 = 43.2 million LP
Wealthy player: 38 400 LPH * (time in hours) 720 * (study time lost when curios are not studying) 0.8 = 22.1 million LP
Regular player: 21 900 LPH * (time in hours) 720 * (study time lost when curios are not studying) 0.9 = 14.1 million LP
As the regular player uses a better strategy to lose less study time its assumed to have a 10% study loss. The wealthy player cant avoid losing study time as some of the curios are to short term and most likely will finish studying when he is offline increasing the study loss to 20%.
Spending all the LP in a single combat starting with level 1 you end up with.
Auto study with pearls: 930 UA
Wealthy player: 664 UA
Regular player: 532 UA
Clearly auto pearl studying is over powered. But the other curios as in bells, floatsams, edels show that is only give marginal advantages. But on the other hand 720 pearls isn't exactly easy to bot either.
Pearls are the problem at the end of the day. Remove bots and wealthy traders can pull in pearls just as effectively or even more so then bots and study them religiously on a specific character. This also proves that the prices have not gone down sense the previous worlds as pearls can be consumed faster then they are traded making there demand stay on the same level as they were before. As of date there is only one single trader that have refused pearls, most likely unaware of how effective pearls really are.
An alternative to the bell and edel farming could also be to make high end curios have a longer study time similar to shewbread making them valued to need less maintenance instead of a short term boost in LPH. Clearly they would need to be made slightly more rare to compensate for the long study times. Farming them would make it pointless as you would just end up with a large stockpile that you couldn't study.
Edit: There is another thing to note around automated bluebell farms. Everyone who talks about bots that collect bells clearly haven't used there brains to farm curios manually. A single intelligent player can prepare 5-7 alts that can spot bluebells. If said person were to place each alt in a swamp in the random vicinity of his main base he could quite literally find a bluebell per hour or more. Log into one character and scout, log out and log into the next one. By the time your finished with the last character on the list the first character would be ready to scout the first swamp on the list again.
This technique could be used to farm a weeks worth of bluebells for a main character in a single day. The whole debate around bot farmed bells is stupid as 7 bluebells takes roughly a week to study and could be gathered by a normal person in just under 2-3hours with enough alts and swamps. When people show pictures of a cupboard full of bluebells they don't really understand that that cub would last for 2 full months of constant study if used on a single character. The only advantage bots give in this particular matter is that you can supply bells to more then one alt.