Economic model

[gardas714]

Hi there,

I'm not writing to criticize the current economic model (that's been done a few times . Rather, if there are people here who would like to develop an alternative one, please let me know and we can coordinate. I'm an economist (university prof), and have a very clear view about how to introduce market forces into OpenTTD. Would love some company to introduce a different dynamic to the game.

Thanks!

[speeder]

I am interesting in seeing what will come from this.

I am yet to find a game with an actually good economic model.

[jfs]

This very forum is as good as any place

If you want to talk to those of us who develop the game, the IRC channel is the best place to come. (Most people are active during European evening.)

[gardas714]

So glad to see some responses!

The irc plan sounds great. I'm in Europe, so it should be easy to do.

Will see you there soon!

[gardas714]

Just had a first conversation about this on IRC.

Users were FLHerne, gardas714, and Yexo.

Main ideas:

0. While it could be interesting to introduce a new economic model, including supply and demand forces, this represents a paradigm shift, in relation to the current game goals.

1. A current fun factor of OpenTTD right now is building long, intricate routes. However, in a pure economic model, if a coal plant opens it shouldn't give priority to coal that comes from far away. For example, if a competitor starts sourcing coal from close by, it should be able to sell it at a lower cost and steal part of the business of the first, farther one.

2. Introducing competition could be interesting, but wouldn't work with the typical 1024^2 maps used in many multiplayer sessions. There, players seem to shy away from competition, and focus on building their own thing instead. So, smaller maps would be needed.

3. One of the features of the current economic model is that money is never really a constraint. A goal for an alternative economic models would be to prolong the phase in which a) there isn't enough money to do everything; b) the potential winner takes time to determine.

3.1. One way to potentially accomplish this would be to make simultaneous construction very expensive, ie, some convexity in the cost of producing "too much at once".

3.2. Another idea to solve point 3 is to make it especially profitable to connect many modes of transportation together, ie, build a multi-modal transportation network. For example, one may be able to transport lots of food by train, but smaller cities may be too small to demand a lot of it. So, it could make more sense to transport the food to a trucking station, that then moves it to smaller towns in a more economical way. Airports could also be very expensive to maintain, thus increasing economics of having multi modes. So, the game could start out with small, direct transport operations, and end in large transportation networks competing with each other.

--------------
That was fun! If you have more discussions on the chat, feel free to add the conclusions to this thread, so we can keep a collection of ideas.

[odisseus]

The means to encourage short-distance transportation is already present in the game — it's called subsidies. By tweaking the subsidy multiplier and cargo payment rates it should be possible to make unsubsidised transportation unprofitable. Think of it as industries paying you for delivering the cargo they want instead of just any cargo.

The Busy Bee game script also has a somewhat related purpose. It assigns each player a set of tasks, which require delivering certain amounts of cargo to certain industries in a limited time span. However, the tasks are completely optional, and there is no reward for completing them. There are also no requirements as to the sources of the cargo.

Competition between players definitely fell out of fashion after the advent of large maps. However, limiting the choice of profitable routes should force players to compete for those. If the reward for transportation were made independent of the travel distance and journey time, the player with the smallest operating costs (which doesn't always mean the shortest route) would win.

The notion that money is not a constraint holds only for the default game settings. There are multiple servers that make the players take care of their finances even in the advanced stages of the game. Usually this is achieved through increased costs of construction, track maintenance, or vehicle operation. For example, one of the servers provides narrow-gauge tracks, which are very cheap but also very slow, several types of conventional track with different speed limits, and vacuum track which costs a million per tile.

In fact, there is an option called "Infrastructure maintenance" which punishes overbuilt and inefficient networks. It also makes the airports very expensive, because under the default settings they are a lazy way to get mountains of cash.

A multi-modal transportation route with reloads between legs is actually less efficient in real life than a single leg, and I don't see why it should be otherwise in the game.
However, your idea about forcing the player to distribute the cargo between small consumers is valid, and there already are ways to achieve that. For passengers and mail, it is as easy as enabling Cargodist with symmetric distribution. With this setting, each station will attract roughly the same amount of cargo as it generates.

For other types of cargo, there are two different approaches. In the FIRS economy, some industries need limited amounts of supplies to boost production. Any amount past the limit is still accepted but essentially wasted, as there is no further increase in production. In the ECS economy, each processing industry has a stockpile that increases and shrinks depending on the industry's production rate, which in turn depends on how much of its output is transported. When the stockpile gets full, the industry stops accepting cargo until some of the stored materials are used up.

Not all of these features would work together smoothly, or at all. Subsidies are disabled when Cargodist is on; FIRS, ECS and the default industry set are mutually exclusive; increased operation costs might make short routes unprofitable. But most of the ideas you mentioned already exist in some form; tying them together and providing the missing parts would still require a substantial amount of work, but you won't have to implement them from scratch.

[gardas714]

Thank you. This is really useful in terms of thinking about the different pieces of the puzzle, and how they can be addressed by the current resources.

A follow-up question: On top of these resources, is there an implemented system that makes it less profitable for me to supply coal to a plant, once a rival also starts supplying to it from a different source? (Thinking about direct competition forces.)

[jfs]

To some extent, if you use an industry set where industries can only accept a limited amount of cargo. If the power plant can at most burn 1000 tons a month, it should perhaps reject cargo in excess of that. Now consider the situation where two transport companies try to service the power plant, company A delivering up to 750 tons of coal monthly from a nearby coal mine, and company B delivering up to 750 tons month from a distant coal mine. Assuming the same extraction costs and purchase price of the material between the power plant and the mines, the power plant should prefer to buy 750 tons transported by company A, and 250 tons transported by company B, since the transportation costs via company A would likely be cheaper. So the distant coal mine should only be shipping out 250 tons monthly since the power plant wouldn't be willing to buy more from them. However, you would also need to keep track of some kind of quality of service ratings, if company A is unreliable about delivering the coal on time then perhaps the power plant should still prefer to buy more transported by company B.

Really, you'd have to have the transport companies bidding on cargo delivery contracts.

[gardas714]

Thank you; very useful. Indeed, making it completely "accurate" would imply incorporating bidding, etc. A mess.

A possible workaround for this is the Cournot model, which simply penalizes the price based on the total overall quantity produced. For example, if two companies are willing to supply 750 tons each, the buyer can always absorb all of the production (i.e., 1500 tons) - assuming a constant elastic demand function, for example (econspeak). However, price is penalized by total production. Incorporating service quality is also possible: in this case, actual prices paid to each company are adjusted by their service qualities.

Since shipments don't arrive all at the same time, implementing this would probably require keeping track of shipments in the last "t periods" per buyer, in order to calculate the current price. (Finally, changes in demand by the buyer over time can also be incorporated, by moving the price up and down.)

[LaChupacabra]

Hey gardas
I certainly support the introducing of changes that would improve the current economic model. I write my proposals myself, unfortunately they are not completed due to lack of time and I cannot publish them fully yet (I will put some of them below). However, reading your suggestions I get the impression that you don't know very well this game ... and its hundreds of additions.
Most of the problems you write about can be solved with newGRF addons or scripts, existing or ones that can be created. It is not always necessary to modify the core of the game itself.

As for additions

3.1. One way to potentially accomplish this would be to make simultaneous construction very expensive, ie, some convexity in the cost of producing "too much at once".
3.2. Another idea to solve point 3 is to make it especially profitable to connect many modes of transportation together, ie, build a multi-modal transportation network. For example, one may be able to transport lots of food by train, but smaller cities may be too small to demand a lot of it. So, it could make more sense to transport the food to a trucking station, that then moves it to smaller towns in a more economical way. Airports could also be very expensive to maintain, thus increasing economics of having multi modes. So, the game could start out with small, direct transport operations, and end in large transportation networks competing with each other.

In terms of costs, there is a Base Cost Mod addon that allows you to change almost everything. In addition, many other newGRF have the ability to quite accurately choose the cost or rates for transport. The fact is, however, that many addons have these costs chosen almost randomly and it is often difficult to balance the game well - it is sometimes very laborious and requires many tests. This is a problem of refining the addons themselves and perhaps the lack of cost patterns that the creators of the addons could use, not the problem of the game - I see no major need to interfere in the game itself on this theme. But I think that some changes would be useful when it comes to searching for such addons, because it is not really easy to find it in this whole stack currently. Even if someone has known the game for years, they may not notice new interesting additions at all, because they are invisible.
Here I would have a proposal, not closely related to economics, but useful in finding the necessary additions.

[+] Spoiler

New windows for downloading and selecting addons, enabling them to be sorted and easier to find



I will try to do the description another time

1. A current fun factor of OpenTTD right now is building long, intricate routes. However, in a pure economic model, if a coal plant opens it shouldn't give priority to coal that comes from far away. For example, if a competitor starts sourcing coal from close by, it should be able to sell it at a lower cost and steal part of the business of the first, farther one.

Basically, OTTD is not a simulator of a trading company, but a transport company. Here the player doesn't sell anything but merely transports and receives payment for the transport.
I think, however, that it would be possible to create such an newGRF addon, where the game would be about trade and not only transport. Even a few years ago, a certain addon was created that went in this direction: Buy, Process, Sell for ECS industries, where you pay for raw materials and earns only on products; transport distance does not matter. A good development of such an addon would be the addition of towns enterprises that would accept such products but in limited quantities so that the player doesn't pack everything in one, closest this time place. The only change in the game that would be very useful here would be to add the possibility of purchasing exclusive transport rights for the enterprise production, as it can be done with the city. Or enabling the purchase of an enterprises. Some time ago, on the occasion of another proposal, I posted such graphics...

[+] Spoiler

Enterprise management.png (27.25 KiB) Viewed 4825 times

When it comes to scripts

Thanks to them, it is possible, among others, to introduce taxes (Villeges is Villeges, Highway Toll). In part, they can also be used to generate subsidies (Think Globaly Act Localy). They can also affect the amount of transport rates and set goals. Unfortunately, the scripts have the disadvantage that they cannot be easily combined at the moment, as is the case with newGRF. The disadvantage of scripts is also that they cannot conveniently display a lot of important information. I don't know if it would be possible to combine them, so that you can add their functions as in the case of newGRF, but if so, maybe it would be worth going in this direction. However, there are also elements that would be worth adding to the game itself as a permanent or optional element.

Important note

A possible workaround for this is the Cournot model, which simply penalizes the price based on the total overall quantity produced...

In general you are right. The problem is that such solutions in this game will not work completely. I see that you like competition in the game, me too, but this solution would work only in very specific conditions of small maps. In the normal game, players avoid interfering in the interests of other companies and direct competition with them. On many servers this is simply prohibited. Imagine someone transporting coal from a very long distance, making great money, and then suddenly some guy connected to the power plant and supplies coal from a nearby mine. According to your proposal, this would cause a huge drop in the income of that company. Do you know what the effect of such changes would be? Why create something that would cause conflicts? Did you think, that some players would want to do it angrily to tease someone? Another disadvantage would be that players would avoid cooperation. They would not help in the supply of raw materials, because they would know that this will reduce the profits of the company they want to help. However, someone who already has a network built wouldn't connect new, closer mines, because it would be economically pointless.

[LaChupacabra]

As for the changes in the game

In my opinion, if we are to talk about significant changes in the economic model, I see here two ways that could very well be one. These are changes that would need to be added to the game to be fully functional:

1a. Variable transport rates
Dependent on the date of the game and the percentage of a given good transported globally - the greater part of the global production of a given goods is transported, the lower the rates.
1b. Cargodist improvement
You can now carry loads wherever and how you want and the cargodist has not changed that in any way, but it could. Cargodist could set directions of transport already at the enterprise or city level - goods at the station would be generated with a designated goal, which would be another enterprise or city and only than stations.
1c. Improved station rating
An alternative to changes in Cargodist. The station (transport) rating for selected loads would depend on the number of directions in which transport takes place and, for some loads, on the size of cities / population within the range of the station relative to the global population.
2. Extension of subsidies
Odisseus and jfs wrote about this, and I am also of the opinion that the extension of subsidies could change the most in the game. Thanks to this element, a very different economic model can be created. Subsidies could take the form of contracts. They could be part of or replace the standard economy. In the latter case, the contracts would be the core business. They would last for a much longer time and sometimes would be indefinite. Contracts could specify different rates for transport as well as other requirements such as time or type of transport. The selection of connections would not be random. The conclusion of the contract could take the form of a 'first come first served' competition or commitments, then appropriate service would have to be provided at a given time, otherwise the company would pay fines. Both forms could be used. The contract could relate to the construction and operation of a specific A-B connection as well as to supply the city or enterprise with specific goods from any source.

In addition, there are several other things about economy that can be improve:

3. Company shares
The current model is so distorted that it can be called a joke at most. The player should decide whether he wants to sell his shares and not anyone else. Shares are sold to raise funds for the development of the company. If someone has shares, e.g. 25%, then they should also receive such a percentage of that company's profit. The new share model could largely be modeled on, I think simple and understandable for every rules of the well-known Dragons' Den television program. By selling 100% of shares it would also be possible to transfer the company to someone.
4. Loan
Extension of the amount (0-100 million) and interest rate (1-20%) range. The maximum amount of loan that varies during the game depends on the company value and the interest rate depends on the risk assessment (company value, ratings by local authorities, stations ratings, global situation: e.g. a crisis). A larger range of initial loans would allow to create more interesting scenarios.
5. Own funds
Adding the possibility of owning small funds and starting the game without taking a loan (0-100k).
6. Taxes
An optional element aimed at equalizing opportunities between new and developed companies. They would be charged every quarter or every year, on income and on possession. The latter tax would be an effective element to reduce excessive accumulation of wealth. In both cases, a tax-free amount would be included, so that small and new companies could grow freely.
7. Extended recession
Greater diversity in the impact of the crisis on the economy (change in rates for transport, including one product, interest rate change, hyperinflation) and different duration (depending on the difficulty setting from 1-6 to 6-36 months)
8. Company value rating
Dependent to a greater or total extent on projected income and not on account balance. The current rating causes that profitable companies fail, whose balance due to investing was accidentally a few times negative. In contrast, abandoned companies that permanently bring losses will last several dozen years, because earlier in a few years they have accumulated much more than all operating costs for several hundred years ahead.
9. Removing of conditional payments (transfer)
Elimination of unjustified negative income. Payment would be unconditional, in an amount adequate to the distance traveled and time. The player would receive a penalty for disappearing cargo from the station. Revenue from London - Singapore - Paris transport would be full for both courses, but interest on this route would be marginal. This approach would be probably better in case of shared infrastructure.
10. Measuring the distance between stations
The distance between the nearest elements of the station should be measured, not between labels - it's sometimes distorts the sense of competition very much.
11. Simple improvement of subsidies
Allowing to work with Cargodist, extending the multiplier range from 1.5-4 to 1.5-10 or more, adding settings changing the duration of subsidies, their frequency and preferred distances. If possible, allowing newGRF additions to influence subsidies would also be a good option.
12. New finance window
In addition to partially changing the layout to improve readability, the window would contain additional panels: shares (visible if the company owns or divides them), taxes (if included) and detailed financial information. In the case of an improved loan, information on the current interest rate would be shown. Some or even all of the elements could have additional windows that would display detailed information. I have several versions prepared, this is one of them:

[+] Spoiler

New finances window v4.png (64.87 KiB) Viewed 4823 times

13. New graphs window
Changes would mainly concern readability. First of all, the window would have a dark background on which all colors (even black) are much more visible. The designation of periods would also change - instead of completely illegible and unnecessary abbreviations of month names, Roman numerals would be used, which are very often used for quarter designations. A new window would allow convenient switching between different graphs. New graphs could also be added: profitability, investments (it will be easy to see who is active) and global production and transport of goods - the latter two could be linked to the element of variable rates for transport. There would also be a few other changes that I will describe other times.

[+] Spoiler

New graph window - companies v1.0.png (54.54 KiB) Viewed 4823 times

For some of the proposals I have prepared a more detailed description with details of the solutions. But it's not for today...

[gardas714]

Although I had been thinking of an alternate economic model, basically a different game mode, allow me to comment on a few ideas:

As for the changes in the game

1a. Variable transport rates
Dependent on the date of the game and the percentage of a given good transported globally - the greater part of the global production of a given goods is transported, the lower the rates.

I think this is a *great* idea. It does not encourage direct competition per se, but allows for a "soft competition" component. Makes a lot of sense to me.

2. Extension of subsidies
Odisseus and jfs wrote about this, and I am also of the opinion that the extension of subsidies could change the most in the game. Thanks to this element, a very different economic model can be created. Subsidies could take the form of contracts. They could be part of or replace the standard economy. In the latter case, the contracts would be the core business. They would last for a much longer time and sometimes would be indefinite. Contracts could specify different rates for transport as well as other requirements such as time or type of transport. The selection of connections would not be random. The conclusion of the contract could take the form of a 'first come first served' competition or commitments, then appropriate service would have to be provided at a given time, otherwise the company would pay fines. Both forms could be used. The contract could relate to the construction and operation of a specific A-B connection as well as to supply the city or enterprise with specific goods from any source.

I always felt subsidies were "low powered incentives" in the game. My main beef is the deadlines, but this is a matter of taste... In essence, I agree that subsidies are broken, but I'm less clear on how fun these contracts would be.

3. Company shares
The current model is so distorted that it can be called a joke at most. The player should decide whether he wants to sell his shares and not anyone else. Shares are sold to raise funds for the development of the company. If someone has shares, e.g. 25%, then they should also receive such a percentage of that company's profit. The new share model could largely be modeled on, I think simple and understandable for every rules of the well-known Dragons' Den television program. By selling 100% of shares it would also be possible to transfer the company to someone.

I especially like the last point. My take: usually companies don't own shares of competitors... there's a clear conflict of interest. There could be a stock market that is willing to buy shares, but I'm less clear about buying shares from competitors. If that happens, I should be able to veto decisions (if I have 50% of a company or more). For example, I should be able to forbid my competitor from transporting a resource, serve a city, etc.

A different idea is for me to be fired with higher likelihood if I don't have 50% of a company. The board can vote me out, for example, if I'm underperforming in relation to my competitors.

An optional element aimed at equalizing opportunities between new and developed companies. They would be charged every quarter or every year, on income and on possession. The latter tax would be an effective element to reduce excessive accumulation of wealth. In both cases, a tax-free amount would be included, so that small and new companies could grow freely.

This dynamic happens in most games. Non-linear taxes could help, but if they are too powerful, they could take some of the fun away. In some games, the followers are the ones who can benefit from technical innovations first. When a new technology comes in, it could be more expensive for the leader to convert its technology, for example, because it has a larger network. Their conversion costs could be higher. An alternative would be to introduce enough randomness into demand over time, so that some initial investments wouldn't pay off forever. There are probably other ways to think about that.

I'll write about the other points later!

[jfs]

Idea for potentially improving subsidies:
When a subsidy is offered, you have to accept it. Accepting the subsidy gives you an instant up-front payment of cash, and a deadline to begin transporting the cargo on the route requested.
If you have not completed and made functional the route before the deadline, the up-front payment is converted into a loan (ideally with high interest rate, assuming the game is modified to allow for multiple distinct loans with different interest rates/terms active at the same time), but there may be a secondary deadline where completing the route will still yield some bonuses later.
When you do complete the line and begin transporting cargo on the line, there could be multiple modes of subsidy. One could be the "classic" of getting a bonus payout as a percentage of the usual income. Another could be a flat monthly bonus for keeping the route active. The duration the subsidy is active for could vary by the subsidy, and could perhaps be dynamic (like "until one of the towns reaches 5k pop"), or even eternal.

Accepting a subsidy involving an industry could perhaps also protect that industry from closure, until the deadline.