Horses, trade posts and MAPI losses

  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

yurcick

Colonel
63 Badges
May 22, 2010
1.118
5.732
  • 500k Club
  • Hearts of Iron IV: By Blood Alone
  • Hearts of Iron IV: No Step Back
  • Victoria 3 Sign Up
  • Victoria 2
  • Hearts of Iron IV: Cadet
  • Stellaris
  • Hearts of Iron III
  • Crusader Kings II
  • Crusader Kings III
  • Europa Universalis IV
  • Cities: Skylines Deluxe Edition
  • War of the Roses
This is a compilation of several suggestions I and other people made in different threads here and on Discord. I think it’s worth putting together to form a somewhat complete picture of how to kill several birds with one stone:
  • get rid of MAPI-related money destruction
  • strengthen PB
  • stop cavalrymen from going into battle on foot
  • make transportation demand more significant and less prone to low-supply irregularities

tl;dr In a helicopter view, it consists of the following:
  • adding horse ranch building and a possible horse output for some subsistence buildings (the way they make clothes and such now)
  • adding horse input for "personal transportation" UC PM
  • adding horse input for barracks
  • starting calculating MAPI-related losses for all industries (it's not computationally expensive if you're smart about it)
  • adding an organic “trade post” building, automatically existing in every state. It would employ shopkeepers and use transportation the same way TCs use convoys, and would grow with internal trade growth without capital investment the way TCs do. MAPI-related losses should be profits for the employed shopkeepers and capitalists. Size should scale to provide reasonable income by shopkeeper employed

Here in full:
As most people know, MAPI, introduced in 1.5, means that even with 100% market access, when a supplier and consumer are in different states, consumer pays more money than supplier receives, to introduce some logistic difficulties and incentivize organic industry concentration (e.g. steel mills being put not that far from iron and coal sources). While this is a welcome addition to the game, there is a significant problem in this. Where IRL the difference in prices would translate to profits for the companies that actually move the goods, in game it is simply deleted. Industries in one state produce 1000 units of wood that brings them £20000, industries in another state purchase those 1000 units of wood that costs them £25000, and the £5000 in between simply disappears. This is very undesirable from the economics point of view (especially in the V3 model, where money is supposed to be the only stored resource, not subject to arbitrary imbalances between creation and consumption).

I believe we can get rid of this problem with very minor added player complexity and computational costs, while solving several additional issues along the way.
As I said, IRL these losses are going to transportation companies. Well, let’s create those! The game already has a concept of buildings without capital investment needs, UCs and TCs, and a concept of building existing in (almost) any state, UCs. Let’s make another one that would scale automatically and exist everywhere to absorb the logistics costs as profits. I’ll call them trade posts (TPs), which might not be the best name for urbanized areas, but TC is taken and I couldn’t come up with a better one.
They should turn ugly money destruction into a much more reasonable process: the TPs should take transportation as an input resource and should pay wages (and the consumption of traders on those wages would be a much more natural money sink that would properly benefit the economy in comparison to deletion). Only when the TP is understaffed or when there's an infrastructure deficit, should some money be totally lost. There might be some counterintuitive behaviour when MAPI improves and suddenly TP profits drop, but I think this should mostly be offset with increased possible volumes (as the industries become more profitable), and if not, some added MAPI granularity will help making this more gradual to avoid sudden trader SOL drops (it’s not gradual enough now anyway, from many points of view).
How do we calculate total TP profits, without checking what goods go where, which is computationally impossible? Easily, we just calculate the losses compared to market price, and let the losses and subsequent profits remain local. So in a two-state example with 1000 wood moved, if there is a wood market price of £22, and the wood is actually sold at £20 and bought at £25 (at 100% MA), the traders in the selling state get the £2000, and in the buying state – £3000 (before deducting transportation costs).
Now what we need is transportation supply before railroads are introduced. In fact, 1.5 has already added transportation output for all transport PMs, including the early game, pre-railroad one. Although I do think it would make bigger sense if those consumed horses, and there would definitely need to be more output.
I think this game has long needed horses, and now it’s finally time to add them. For most countries, subsistence farms output of horses should be more or less enough, but for extreme situations we should add a horse ranch building taking an arable land slot.
Apart from UC’s horse carts, horses should be consumed by most barracks in different amounts (possibly in infantry for wagon trains, likely in artillery for artillery transportation, definitely – in cavalry for everything cavalry-related).
There can be different approaches to possible TP PMs. I think they need to be mostly automatic, like TC ones (changed with tech and law progress) and be configured the way that they’d create a reasonably strong early game PB, especially in somewhat industrialized countries. Later, they could employ less shopkeepers and more capitalists.

What do you all think?
 
Last edited:
  • 30Like
  • 4
  • 3
  • 1Love
Reactions:
I've made several mock up screenshots to better illustrate the idea.
While making them, I found out that 1.5.9 actually already has transportation output for all transportation PMs in UC. I hadn't known that, and now I reworded the original post slightly. I still think that horses should be added (although they aren't absolutely necessary for the general idea).

2023_11_26_5.png
2023_11_26_6.png
2023_11_26_1.png
2023_11_26_4.png


Please note that while some numbers are arbitrary and unbalanced (namely transportation production), some are true. For example, Moscow does have almost £18k weekly money destruction at the moment in 1836.
 
  • 2Like
  • 1
Reactions:
What is the meaning of the lost profits made by buildings selling locally at an inferior price of the market price?

Furthermore, the building is selling its wares at the mix of the local price, i.e., losing a lot of money for all its revenues if they come from the market and not locally.

For example, Zulia has military shipyards that produce 71.5 ironclads and only sells 4 in Zulia, selling 52 to the market.

The price in Zulia is 57.1 but the building is accounting all its production as it was sold at the local price.

Is there a trade post adding the difference for this buildings so they can accept to sell all production at local prices? Are they stockpiling the ironclads that have no buyer?

1701014274241.png

1701014444297.png


The best solution is to cut the middle men and buildings calculate revenues by its actual price, internalizing the costs of sells (marketing, transport, etc..)

In my example, 4 ironclads x 57.1 + 52 ironclads * 63.7 = 228,4 + 3.312,4 = 3.540,80. The other 15 ironclads should go to a stockpile.

But this may not be the optimal solution for many other reasons.
 
Last edited:
  • 2
  • 1
Reactions:
What do you make of the lost profits made by buildings selling locally at an inferior price of the market price?

In your suggested system this money is lost as well.
It's not. The difference between selling price and market price goes to local traders in the "exporting" state (see "wood" or "furniture" on my first mock up screenshot above, the local prices one).
The difference between buying price and market price goes to local traders in the "importing" state (see any other non-local good on the screenshot).

Normally (if there's 100% MA and the TP is fully staffed) there's no money lost (although it's still much better to have a higher MAPI to allow the industries to be profitable and to economize on traders' consumption, there's gonna be a lot of them).

For example, Zulia has military shipyards that produce 71.5 ironclads and only sells 4 in Zulia, selling 52 to the market. The other 15 ironclads should go to a stockpile.
I fail to understand what you're referring to when you speak of stockpiles (the only thing that can be truly stockpiled in V3's economic system is money). Do you propose adding them as a concept? This would be a very major change, which doesn't seem reasonable, and anyway is very out of scope of this suggestion.
Anyhow, under the current paradigm where everything is sold and bought instantly, we can always speak of the price at which something is sold and bought in a state. In your example, Zulia's military shipyards would get £4555 if they sold everything at market price of £63.7. However, they sell at the local price of £57.1, receiving only £4083. The difference of £472 should go to local traders.
You could interpret £57.1 as some weighed average between the (higher) market price and some (much lower) hyperlocal one, but this would only confuse you, and isn't necessary for this line of reasoning.
 
It's not. The difference between selling price and market price goes to local traders in the "exporting" state
Not this, the money lost by the building selling at local prices if they are lower than market prices. Is there a policy that mandates buildings to sell at local buyers lowering their profits?
I fail to understand what you're referring to when you speak of stockpiles
Buildings happen to have less buy orders than produced goods.

We understand that exceden goods were sold to the black market. But I believe having stockpiles for military goods is already in the devs mind.
 
Not this, the money lost by the building selling at local prices if they are lower than market prices. Is there a policy that mandates buildings to sell at local buyers lowering their profits?
I think the question is: how would an industry sell at anything other than local prices? The national market prices can be thought of as a benchmark price that local prices are based on. There should be a system for bringing goods to the national market like @yurcick is proposing.

A real world example of what I think the national market price is representing is West Texas Intermediate (WTI), which is one of the main benchmark prices for oil. While the price is a benchmark, the main delivery point for that oil actually being exchanged is in Cushing Oklahoma. So if you want to buy 100k barrels of oil somewhere else, you will also have to pay for transportation for that oil from Cushing (or wherever it actually came from) to the delivery point. That transportation cost is accounted for in the contract to sell the oil, with the buyer paying more that the WTI price unless it is coming from a production point.

This seems a relatively easy problem to solve with the proposed solution and is kind of in place for international trade already, though I would love that trade to also account for the costs by having to buy convoys instead of having them provided by the state for free.
 
  • 4
Reactions:
I think the question is: how would an industry sell at anything other than local prices?
Because it is more profitable.

The whole point of tariffs to exports is to protect local supply because buildings will always sell to the max price if possible.

Market price is not a benchmark price, is the price of the broader market.

Transport availability is accounted by the infrastructure and transport costs are accounted for with transport good.

I agree with you that convoys and infrastructure is provided by the government with no private investment.

Furthermore, as I was saying, transport costs should reflect the actual buyers location to calculate them. It is not the same to sell inside your state that going through four or five states. But transport only affects your costs, not the price the buyer is willing to pay.
 
Last edited:
  • 1Like
Reactions:
Because it is more profitable.
It could be even more profitable to sell at ten times market price. Sadly, no-one's buying.
The prices, displayed and discussed, mean what an industry can receive, per unit, if it sold its output. It can't receive more. Not because there's any "policy" that prevents it, but because the price is a result of market balance.
As it's been correctly explained above with the Texas oil example, it makes perfect sense to speak of local price different from the market price. Not because you're somehow forbidden from selling high on the market instead of selling low locally, but because the buyer who's willing to buy high is far away, and the transportation costs to get goods to the (real or abstract) market, per unit, would reduce your margin and ultimately make it the same as if you sold locally. IRL, of course, there would be more variety of local prices as "costs to get to the market" would depend on a lot of things, namely distance differences (which is now ignored if it's non-zero). But calculating distances for all states for all goods, given that actual consumers might be much closer than the "market" so it's not a 2N calculation, isn't possible for any PC.
 
  • 2
  • 1
Reactions:
I think some sort of logistics centre kind of building popping up makes sense. Not sure about adding horses since I already found the number of goods to be kind of overwhelming but that might just be due to the UI not making it easy to rapidly compare lots of different statistics about the market.
 
  • 4
Reactions:
In fact, there are not many products in this game right now. I am for the horses and for the alteration proposed in this branch. I would like to add more goods (peat - for early heating; animal fur - for the rise of the countries of the north; spices, including salt - for the rise of Indonesia; the division of meat into pork and beef - so that British India does not save on meat; horses - for transport and the army)
 
  • 2
  • 1Like
  • 1
  • 1
Reactions:
In fact, there are not many products in this game right now. I am for the horses and for the alteration proposed in this branch. I would like to add more goods (peat - for early heating; animal fur - for the rise of the countries of the north; spices, including salt - for the rise of Indonesia; the division of meat into pork and beef - so that British India does not save on meat; horses - for transport and the army)
There are now very significant problems with goods substitution (see links in my signature if interested), and unless they are addressed, I wouldn't create more substitutes, as the problems tend to become more severe with variety.
Spices sound great though.
 
  • 5
  • 1
Reactions:
I also like the elegance of this solution in that it eliminates the need for processing every possible pair of states to calculate the transportation costs. It basically comes down to every selling state paying the cost to transport the goods to the market, and every buying state paying the cost to transport the goods from the market to the state. The weirdness of very remote states having similar transportation costs to states next to one another won't be resolved, but that can probably be addressing using state traits or later grouping into submarkets by region or something. This method has the advantage of being far more performant in that for each state and good there would be only 2 calculations, rather than say 25 calculations in a great power with a number of states.

I like the changes to the MAPI so someone is "moving" resources, but I do not understand the desire I see on the forums to add horses as a dedicated building and resource.
I don't think they are necessary either, I would be happy with the horses/draft animals PM just using extra grain/fruit and lots of labor. The key is that there needs to be a way of producing transportation in the early game without relying on urban centers or railroads, even it is wildly inefficient. I would love to see 3 types of logistics/transportation buildings - roads, railroads, and water transportation.
  • Roads could move from: dirt roads (labor intensive) > Paved Roads (trade some tools for labor) > Early Trucks (Automobiles for more labor) > Limited Access Highways (more automobiles)
  • Railroads could exist largely as they do now
  • Water Transportation could be restricted to states with coasts and river traits, and go from: Small Boats (lots of labor, small clippers) > River Barges (labor for more clippers), Steam Barges (steamers), Diesel Barges (steamers and oil). This would also make canals useful as they could raise the limit of water transportation available in a state.
 
  • 8
  • 4Like
Reactions:
I don't think they are necessary either, I would be happy with the horses/draft animals PM just using extra grain/fruit and lots of labor. The key is that there needs to be a way of producing transportation in the early game without relying on urban centers or railroads, even it is wildly inefficient. I would love to see 3 types of logistics/transportation buildings - roads, railroads, and water transportation.
  • Roads could move from: dirt roads (labor intensive) > Paved Roads (trade some tools for labor) > Early Trucks (Automobiles for more labor) > Limited Access Highways (more automobiles)
  • Railroads could exist largely as they do now
  • Water Transportation could be restricted to states with coasts and river traits, and go from: Small Boats (lots of labor, small clippers) > River Barges (labor for more clippers), Steam Barges (steamers), Diesel Barges (steamers and oil). This would also make canals useful as they could raise the limit of water transportation available in a state.
That's something I agree with, I just keep seeing people asking for horses as a trade good.

I think they should rename railroad to "Infrastructure" and give every state a cap of 1 modified by navigable rivers and such. Basic dirt roads just weren't that good for moving large amounts of goods long distance, but they were still a big step up from no road, and it could generate transportation to be used by whoever ends up acting as the middleman. Roads weren't useful for goods movement unless the area was largely undeveloped and you had no other choice. Draft animals eat too much and the roads weren't really useful for long distance vehicle traffic until the interstate system killed the local spurs railroads had built. Pretty much every town in my state, no matter how small, is or was located along a railroad. They're mostly gone now, but you can still see them used as trails and such today.
 
That's something I agree with, I just keep seeing people asking for horses as a trade good.

I think they should rename railroad to "Infrastructure" and give every state a cap of 1 modified by navigable rivers and such. Basic dirt roads just weren't that good for moving large amounts of goods long distance, but they were still a big step up from no road, and it could generate transportation to be used by whoever ends up acting as the middleman. Roads weren't useful for goods movement unless the area was largely undeveloped and you had no other choice. Draft animals eat too much and the roads weren't really useful for long distance vehicle traffic until the interstate system killed the local spurs railroads had built. Pretty much every town in my state, no matter how small, is or was located along a railroad. They're mostly gone now, but you can still see them used as trails and such today.
Are you thinking that they would have all three methods in one building and they would be under different PM categories? I had that thought before, but one issue that I see is that you will want to adjust the mix of methods and levels depending on the situation in your country. For example you may have more levels of water transportation than land transportation if your country has a lot of rivers or coastline, especially if there is not a lot of cheap labor. A country with a ton of labor may just eat the inefficiency of early road transport in order to get some transportation in the market, even if it is relatively high cost compared to later methods. As long as every state can have *some* transportation that will fix a lot of the issues with local only.
 
  • 1
Reactions:
This is a compilation of several suggestions I and other people made in different threads here and on Discord. I think it’s worth putting together to form a somewhat complete picture of how to kill several birds with one stone:
  • get rid of MAPI-related money destruction
  • strengthen PB
  • stop cavalrymen from going into battle on foot
  • make transportation demand more significant and less prone to low-supply irregularities

tl;dr In a helicopter view, it consists of the following:
  • adding horse ranch building and a possible horse output for some subsistence buildings (the way they make clothes and such now)
  • adding horse input for "personal transportation" UC PM
  • adding horse input for barracks
  • starting calculating MAPI-related losses for all industries (it's not computationally expensive if you're smart about it)
  • adding an organic “trade post” building, automatically existing in every state. It would employ shopkeepers and use transportation the same way TCs use convoys, and would grow with internal trade growth without capital investment the way TCs do. MAPI-related losses should be profits for the employed shopkeepers and capitalists. Size should scale to provide reasonable income by shopkeeper employed

Here in full:
As most people know, MAPI, introduced in 1.5, means that even with 100% market access, when a supplier and consumer are in different states, consumer pays more money than supplier receives, to introduce some logistic difficulties and incentivize organic industry concentration (e.g. steel mills being put not that far from iron and coal sources). While this is a welcome addition to the game, there is a significant problem in this. Where IRL the difference in prices would translate to profits for the companies that actually move the goods, in game it is simply deleted. Industries in one state produce 1000 units of wood that brings them £20000, industries in another state purchase those 1000 units of wood that costs them £25000, and the £5000 in between simply disappears. This is very undesirable from the economics point of view (especially in the V3 model, where money is supposed to be the only stored resource, not subject to arbitrary imbalances between creation and consumption).

I believe we can get rid of this problem with very minor added player complexity and computational costs, while solving several additional issues along the way.
As I said, IRL these losses are going to transportation companies. Well, let’s create those! The game already has a concept of buildings without capital investment needs, UCs and TCs, and a concept of building existing in (almost) any state, UCs. Let’s make another one that would scale automatically and exist everywhere to absorb the logistics costs as profits. I’ll call them trade posts (TPs), which might not be the best name for urbanized areas, but TC is taken and I couldn’t come up with a better one.
They should turn ugly money destruction into a much more reasonable process: the TPs should take transportation as an input resource and should pay wages (and the consumption of traders on those wages would be a much more natural money sink that would properly benefit the economy in comparison to deletion). Only when the TP is understaffed or when there's an infrastructure deficit, should some money be totally lost. There might be some counterintuitive behaviour when MAPI improves and suddenly TP profits drop, but I think this should mostly be offset with increased possible volumes (as the industries become more profitable), and if not, some added MAPI granularity will help making this more gradual to avoid sudden trader SOL drops (it’s not gradual enough now anyway, from many points of view).
How do we calculate total TP profits, without checking what goods go where, which is computationally impossible? Easily, we just calculate the losses compared to market price, and let the losses and subsequent profits remain local. So in a two-state example with 1000 wood moved, if there is a wood market price of £22, and the wood is actually sold at £20 and bought at £25 (at 100% MA), the traders in the selling state get the £2000, and in the buying state – £3000 (before deducting transportation costs).
Now what we need is transportation supply before railroads are introduced. In fact, 1.5 has already added transportation output for all transport PMs, including the early game, pre-railroad one. Although I do think it would make bigger sense if those consumed horses, and there would definitely need to be more output.
I think this game has long needed horses, and now it’s finally time to add them. For most countries, subsistence farms output of horses should be more or less enough, but for extreme situations we should add a horse ranch building taking an arable land slot.
Apart from UC’s horse carts, horses should be consumed by most barracks in different amounts (possibly in infantry for wagon trains, likely in artillery for artillery transportation, definitely – in cavalry for everything cavalry-related).
There can be different approaches to possible TP PMs. I think they need to be mostly automatic, like TC ones (changed with tech and law progress) and be configured the way that they’d create a reasonably strong early game PB, especially in somewhat industrialized countries. Later, they could employ less shopkeepers and more capitalists.

What do you all think?
In my opinion, Trade post is just an unnecessary middleman.

This is how I would do it:
1) Calculate how much money has building lost due to MAPI (let's call it MapiLoss)
2) Add MapiLoss to the building's income
3) Add (MapiLoss/50) amount of transportation to inputs
4) MapiLoss from pop consumption goes to an invisible building that hires nobody and buys all transportation it can afford

My solution is better for multiple reasons:
1. It would be easier to understand what MapiLoss is and very it is going.
2. There would be no wave of unemployment when technology increases Mapi.
3. Since it doesn't create a separate building that would shrink/grow, hire and fire people it should require less computation.
4. It would create additional level of strategy where you could use transportation costs to move money between interest groups or player. If you create abundance of transportation then profits for buildings in this state will increase, but profits for railway owners will decrease; If you create deficit of transportation the opposite will be true. This way you could, for example: empower capitalist by providing cheap transport in industrial areas and making transportation in rural areas expensive and owned by capitalists, or you could make transportation expensive everywhere and set railway ownership to government.

Also, you should post this on Reddit as well, if you haven't already, this is a good idea.
 
  • 1Like
Reactions:
Are you thinking that they would have all three methods in one building and they would be under different PM categories? I had that thought before, but one issue that I see is that you will want to adjust the mix of methods and levels depending on the situation in your country. For example you may have more levels of water transportation than land transportation if your country has a lot of rivers or coastline, especially if there is not a lot of cheap labor. A country with a ton of labor may just eat the inefficiency of early road transport in order to get some transportation in the market, even if it is relatively high cost compared to later methods. As long as every state can have *some* transportation that will fix a lot of the issues with local only.
All one building, with "canals and roads" as the basic PM with more levels unlocked similar to ports so you can't truly go all in until you have actual steam engines. You're trying to split it up with "levels" of canal, but I don't think they need to. Some states could start with 1 or 2 extra possible levels of infrastructure from the start, similar to some states having +2 port max level.

So early game, every state can build 1 level of infrastructure but states that were conducive to large canals would have +1 or +2 max level from the start, increase the max level when you unlock experimental railroads, then remove the cap when you unlock steam engines.
 
I made a mod (that i never released though) for the early version that included roads, horse carts, transport costs for about everything, big map update with added road buildings (especially in more developed country's) and province modifiers especially for river and coastal provinces etc.

I can pass the files, most of the work has been done, just need to remake the map in the new version i guess. What i wonder though, is why not try solve this with a mod? It took me about 2 weeks to make it as just 1 person.

Granted, the OP makes it sounds way more simple than it is. Fundamentally, the real issue is that even if you add transport costs, the game can't differentiate between 1) a province next door and one that is far and 2) 2 provinces that that are connected with railroads counting the provinces in between them and those that arn't and 3) provinces that are in the same market but seperated by a body of water and a few more. Whatever you try, these issue tends to bite you.

In the original release, the way to implement a difference between local costs and global costs was to force it so that the whole map started with low infrastructure and that it would have required huge infrastructure building to get it to 100 state infrastructure. In that fashion, there was a price difference aswell between buying local goods and goods from the internal market. So i forced it off that most provinces with lack of infrastructure started around 60 state infrastructure and that a good province would be 90+ infrastructure. This created the same prince differences that MAPi does. I then allowed more infrastructure to be added by buildings like roads and railroads, with an added good called road transport. Factory's and such would then either use road or railroad transport, albeit that railroad transport (as a transport choice for the factory) allowed the province to have more infrastructure than road transport, with other words putting factory's to railroad transport added a tiny bit more of infrastructure than road transport, and both had a cost in terms of the good. Besides that provinces with rivers and coasts had fairly large boni to infrastructure, aswell as ports had, to mimic that water transport was more the prefference. At a later point i was going to add things like canals, bridges and tunnels.

I did not add horses though, rather i abstracted horses to grain (as required to feed them) wood. (as what the cart is made off) and labor besides that there was also non-horse drawn transport, with other words guys pulling carts, which was just a bit more expensive in similar ways and added less infrastructure than horse transport.

I added the mod below, but one would have to do a rollback to near release version to play it. Mind that it might look weird if you lood it up and you see so much provinces lacking infrastructure by design. If people like what they see, i can perhaps make a new mod for current versions that builds on the ideas i had there.

The issue however will always be that the game can't calculate distances, and that unrealistic things happen as a result. What my mod did though in the grander scheme of things was to make certain areas like say new York or Shangai far better for building lots of factory's in because the infrastructure was more effecient due to presence of coast and rivers and flat land and such, besides that it put a fairly big break on development for most country's and favoured western country's that had more road buildings and ports.
 

Attachments

  • transport_mod.zip
    1.020,8 KB · Views: 0
Last edited:
The best solution is to cut the middle men and buildings calculate revenues by its actual price, internalizing the costs of sells (marketing, transport, etc..)
You don't want to do this because it would make the game run extremely slow.

The way that Vic3 works is there is no matching of buyers and sellers, which makes things much faster to compute. All goods are sold at the market price regardless of volume and all goods are bought at market price regardless of volume. The only change from 1.4 to 1.5 is market price is now calculated from the national and local price.

This is why you can have mismatches between buy and sell orders without anything breaking.
 
  • 7
Reactions:
If i were to remake my mod with MAPI in mind, things would be a lot easier. Basicly my old mod mostly did what MAPI is doing now, and so i think MAPI is great. The only caveat is that in my mod the money did go somehwere IF infrastructure was great enough to overcome it, and in that case it went into the pockets of pops trough transport costs.

So, i think the best sollution would be to have an "input transport" option within the factory, to either buy inputs withought addition of transport or with addition of transport. In the case one buys with addition of transport, a transport cost is added but the effects of Mapi are reduced or nullified, it's just that in a province with lack of infratructure (and thus more transport demand than supply) the cost of transport would be large, less so if there would be plenty of infrastructure.

Thus, you either would not pay for a MAPI nullification (in case you either buy cheap localy or you pay the MAPI penalty for buying outside the province) or you would pay for MAPI nullification for the input of that good, in which case the factory would buy at the market price withought MAPI penalty but with an added cost/demand in transport good which would be more costly if transport is expensive in the province. One could additionally add the same for what regards selling the product to the market, aka pay an added transport cost to put it cheaper to the larger market with MAPI nulification where a factory in another provice could buy it similarly with MAPI nulification, but both adding transport costs.

That seems like a very simple and yet very intuitive sollution. I may make a new mod based on that principle.

I wonder atleast if its possible to have a local modifier for MAPI that only affects the factory for which it is chosen. I presume that MAPI is a nationwide modifer though. Otherwise, what you could have is that you could have either "near 0" MAPI (aka, very expensive if impossible to buy/sell the good outside the province) or 100% MAPI (you can sell/buy from outside the province, but you have to pay the reansport cost that comes with that MAPI nullification)
 
Last edited: