Economy Model Version 0.2

[korn469]

yeah i've been busy with school, work, and my gf for the past few day, and i wasn't able to connect with apolyton over the weekend

korn469

p.s. i'll try to post a little after i get off from work tonight

[Guildmaster]

It's not just this thread, but all of them. I presume that everyone else is like me: Busy with school at this time...

------------------
Peace and trust can win the day despite all your losing
-Led Zepplin

[amjayee]

Yes, I'm a little busy too, like also VetLegion. I don't intend to push people around. I'm doing as much as I can, everyone else can do the same. We are in no serious hurry, since we don't have deadlines...

[The Joker]

Finally I am able to post here again.

I have been spending my time off Apolyton (as much as I could - like the rest of you I have been buissy) to create a really simple version of my economy model. It is almost done, so I will propably post it tomorrow. It should be ready to program, so if Amjayee (or somebody else, that is, if we have other programmers at the moment...?) could program it as a textbased program it would be great. It should teach a lot of things about how economy works, and it should hopefully make it easier for us to expand on it. By adding more and more features to the model, while testing it and program it all on the way, it should make it possible for us to create a great economy model.

------------------
"It is only when we have lost everything
that we are free to do anything."
- Fight Club

[korn469]

i can't wait to see the model that you have came up with maybe it will kickstart discussion again

korn469

[amjayee]

Sounds great Joker, I can't wait to see your model. I'm trying to do the same with the population model. It would be really good to test and analyze the models without the graphics. That way we would make them a lot better. About programmers, VetLegion is a little busy right now, but he's with us. Dan is even more busy, but he said he can help us out if needed; otherwise, he doesn't have much time left for us. But that doesn't matter in this phase, when we are designing. With good design docs, it's easier to get programmers - usually those people don't care much about the design stuff, they just want to code.

[The Joker]

This is the first demo of the textbased economy model.. It should include only one province, and no map. It would simply include a set number of workers who could work in any sector, and a certain productivity.

I will try to describe how it should work. There should be enough information here to create the demo, but if there is something that I have not explained enough about just let me know. The demo should be in a province with 100,000 people in it. All of these people will work - so there will be no age groups for now. The population will also remain constant, unless we change it. This is because the economy model will not yet be linked with the population model. In this early demo all people would be self employed. They would own their own business and take home the profits. Later we can always include employees and employers. In this demo there will be 3 goods - food, housing and clothing. We should wait a bit untill we include graded goods, demand groups and other fancy features. For now let’s just see if we can make something work. Other features not included here are capital and raw materials. For now labour will be the only thing required to make each type of good.

I will make the basic population entity be 10,000 people. This will be called a PU (pop unit). A PU doesn’t really have any meaning in the game, it will just be a short way for me to make assumptions conserning the population. In the game 1 PU would eat 10 food per turn to be healthy. In this demo the amount of food required to remain healthy doesn’t really matter, as the amount of people will be constant no matter how little people eat. In the game as well as in the demo the basic price of 1 unit of a good is 10 credits. In this demo there will be 1000 credits flowing around in this virtual society. These will be divided evenly among the people, so each PU has 100 credits to spend (note: I think that since all the people in this demo belong to the same class, they should always share the money they have as a class, so they will always be equal even though the people working in one sector might take home more profit than ones working in another – all the money earned by each member of a class will simply be collected in a pool and split equally among the members of the class). Again all of these will be used for consumption, as no money investment is needed. All that is required to create goods in labour. In this demo 1 PU will create 100 labour. This labour is what creates the goods. I will assume that 5 labour will create 1 item of any of the three goods in the game. This will mean that 1 PU can create 20 items of any good. Note that productivity, amount of people or credits in the demo and all other variables should be easily changeable via a txt file or something - so we can change e.g. productivity or amount of money and see what happens.

The first turn everybody will be working in the food sector. But since this will create a huge surplus in food workers will start moving to the other sectors, untill an equilibrium is obtained. The main purpose (at least to me) of this economy model demo is to see where this equilibrium will be, how long it will take it to get there and how it is best made possible.

The primary challenge in this economy model is to make sure that workers move to the other sectors of the game. This happens via the demand function, which has shown to not only be one of the most important and basic aspects of the econ model, but also one of the most dificult to create.

I suggest using a basic predefined demand function for this demo. Via the price on the goods and the income of the people it is possible to determine what people will buy.

1 PU will use it’s income on certain goods. It will work in a way that I have described before. When the price for all goods is 10 credits per item people will buy first one thing, then another, then a third and so on. The list of goods bought could be the following:

Good 1-5 bought is food
Good 6 bought is housing
Good 7-8 is food
Good 9 is clothing
Good 10 is food
Good 11 is housing
Good 12 is food
Good 13 is clothing
Good 14 is food
Good 15-16 is housing
Good 17 is food
Good 18-19 is clothing
Good 20-22 is housing
Good 23 is food
Good 24 is clothing
Good 25-28 is housing
Good 29 is food
Good 30-32 is clothing
Good 33-38 is housing
Good 39-40 is food
Good 40-43 is clothing
Good 44-50 is housing

This also means that the first 50 credits earned are spent on food, the next 10 on housing etc.

This should be enough. I have not really spent a lot of time thinking about this list. But it does make sure that when people are poor they will spent most of their money on food and when they become wealthier they will spent a lower percentage of their income on food.

Examble from the list: The price on 1 item of each good is 10. Some people have an income of 240 credits per PU. This means that they can buy 24 goods. Via the list we will know that they buy 12 food, 7 housing and 5 clothing.

The key element of the economy model demo is to figure out a way to handle the price setting.

I have to ways to handle the price setting. A simple one, that is easy to make, and a more advanced one that I am not yet sure how should work. The advanced one will have goods that have an elasticity other than 10%.

Simple version:
The amount of goods bought always equals the supply (the supply only changes between turns, so when looking at a single turn it is constant, no matter what the price is). People always spend the same amount of money on each good when their income is constant, no matter what the prices of the goods are. A high price will simply mean that people will buy less of the good.

The price per item of a good=(money spent on good)/(supply of good)

So if the supply of food is 300 and people spend 5000 credits on food the price per food unit will be: 5000/300=17

The price of a good will be used to determine whether workers will shift sector (from one with a low price to one with a high one). This version is this simple. The advantage of it is, that it is really easy to implement. The disadvantage is, that I think it will make certain economic crisises impossible to portray in the game.

Advanced version:
It starts out like the simple version. It will just include a way to make it so that a change in the price on a good will change the amount of money people will spend on it. So if the price on food rises people are likely to spend more money on food. The problem with this version is, that it means that when the price on certain goods is high people will spend more money on them and less on some other goods. But what other goods? It requires a somewhat advanced way to calculate what a change in price means to the amount of money spent on each good. I have a feeling that it is only because I have been thinking so much about this problem that I am unable to solve it. There may be a really simple solution to it that I just can’t see (there is an old Danish saying that goes something like this: ”I can’t see the forest for all the trees” – that is a bit like how I feel)

These are just suggestions, and in the demo we should propably have both (start with the simple one and include the advanced one when it is done).

The demand function for each good is being calculated, and a price on each good is determi-ned. The supply seen in amount will never change no matter what the price is. If the supply is 0 then the price of the good would be infinite. In stead of using infinity we should just be using some very high price which would be constant for each good (for food it might be 500 or something, for luxury goods it would be significantly lower). The price will be used not only to define the price at which goods are bought/sold, it will also be used to determine the amount of workers that change to work in another sector. The higher the price on a good, the more people will begin to work in that sector. I am not completely sure how to determine this.

But I have some ideas. Let’s look at the initial situation. There are 100,000 people in this society, and they’re all farmers. This means that they will create 200 food per turn. Each PU has an income of 100 credits, so their initial demand (when the price on each good is 10) is 8 food, 1 housing and 1 clothing. This will mean that the total initial demand for food is 80. The supply is 200 items. The demand for housing is 10, but since the supply is 0 the price of housing will be where the demand is 0 – again a really high number. The same thing for clothing. But since there is no other things to buy other than food all the money the people have will be spent on food. This means that with 200 food and 1000 credits the price will be 5 credits per food.

This means that the price for food is 5 and for housing and clothing it would be really high. Of cause this will mean that it is likely that people will move from producing food to producing these other items. But how many people will do that? It must be a number large enough to make the economy flexible, and yet small enough to make sure that there will still be people working in the food sector. And at the same time the equilibrium should not be obtained in just 1 turn.

Since I have not yet figured out excactly how to do this, and I am simply too tired to think about it now. But making this demo ready to be programmed should be an easy task, if we put some work into the two areas that I have problems with: The advanced demand version and the way to handle a change in the amount of workers working in each sector.

I should have some time tomorrow to sit down and think about this, but I hope that some of you will also post some comments or ideas to this model.

And, as always when I'm posting models, if there is anything you don't understand or find weird, please let me know, and ask me. I think I have it all figured out in my head, but when I write it down I end up doing it in odd internal monologues, which may make it impossible to actually understand what I have written.

But I really think this is the right way to go! In stead of developing complex models that we can not at all overview and really don’t know how functions we should start very simple and gradually balance the model out as we include more complexity and features to it.

Well, I'm going to bed, so goodnight!

------------------
"It is only when we have lost everything
that we are free to do anything."
- Fight Club

[korn469]

quote: But I have some ideas. Let’s look at the initial situation. There are 100,000 people in this society, and they’re all farmers. This means that they will create 200 food per turn. Each PU has an income of 100 credits, so their initial demand (when the price on each good is 10) is 8 food, 1 housing and 1 clothing. This will mean that the total initial demand for food is 80. The supply is 200 items. The demand for housing is 10, but since the supply is 0 the price of housing will be where the demand is 0 – again a really high number. The same thing for clothing. But since there is no other things to buy other than food all the money the people have will be spent on food. This means that with 200 food and 1000 credits the price will be 5 credits per food.

The Joker

i think your system is close but that it should work slightly differently...

ok we have 200 food units and 10 population units

each population needs to eat 10 food units, and they might be willing to pay fairly high prices for those food units...how ever if food prices went down that people would eat more food, maybe 12 units or 15 or even 18 units of food if price didn't matter but they wouldn't eat more than 18 no matter how low prices were...i mean even if a loaf of bread was only a penny i would only eat so much bread a day, however if bread was ten dollars a loaf i would eat much much less but the important thing is that when a product is virtually free there is only so much of it you want, i mean if bread cost 2 pennies then they lowered the price from 2 cents to 1 cent i wouldn't buy any more bread...and on the same had if a loaf of bread costed 5 million dollars and they lowered the price to one million dollars i wouldn't buy five times as much (i wouldn't buy any)

but lets say that in your example that people would pay the following prices for good

10 units of food 1.00 pfu (per food unit)
12 units of food .90 pfu
15 units of food .75 pfu
18 units of food .60 pfu
19 units of food .01 pfu

what i'm trying to say is that people assign a certian priority to a good and they will only pay a cerain amount for it, right now a gallon of gasoline costs $1.41 where i live and it hasn't effected my driving habits too much...but if gas cost $14.10 a gallon or $141.00 then i certainly would drive much much less, however if gas only costed $.14 cents a gallon i wouldn't drive that much more than i already do

supply is low prices go up then either supply will increase or demand will decrease...if supply cannot meet demand then prices will increase until supply and demand reach equilibrium

or

supply is high prices go down then supply will go down or businesses will go bankrupt because prices are so low...if supply is too high then eventually prices will drop below what it actually costs to produce the good (and if prices don't drop demand won't increase so good will go unsold)

does that help any or do i need to explain what i'm trying to say in a better fashion?

korn469

[amjayee]

Good, Joker! I'm sure I can come up with something with that model. I will need to try start programming the basic things in the model to see how it goes, and what problems it might have. It is difficult to say how that would work. You said that you can't see the forest for the trees; actually, there is a saying that game designers can't see the trees for the forest, and the programmers can't see the forest for the leaves of the trees! Anyway, it is possible that the system is almost too simple to work properly; I just have to try to find that out.

I think Korn made good points. I will once again collect all things from this model in one place, distilling them to simple sentences, describing what I think the system should work like:

-people have minimum demand they need to survive. They will do anything to get that amount.

-people have a maximum demand; they will not buy anymore that product no matter how much they have money.

-housing: people try to have at least some kind of shelter. If they have money, but there's no supply of houses, they will build their own houses.

-When demand for example for houses rises, some people will start producing houses; that might depend on the initiative of the people or something, and how much smaller the demand is than the supply.

-combining the two former sentences: if there is not enough supply for houses, and people have money, some people will build their own houses; some people from those will find out it's fun and possibly profitable to build houses, and start building houses for other people as well.

-when supply for products is higher than the demand, the prices will go lower AND supply will go smaller, until surplus production is sufficiently small.

These were some things to consider. Tonight I will try to program the framework for the demo, and find out how to get the people buy and produce things. It will take some days before the actual supply/demand/prices things are needed. Until that, continue to discuss the model...

[korn469]

also keep in mind that everything has a production cost associated with it and that the goal of a business is to make a profit, also we need to make our model complex enough so that price is not the only factor which determines what goods people buy

i mean if there was only a dollar price difference between 2000 model BMW luxery car and a 1982 Yugo in the real world everyone would buy the BMW...however if there was a 30,000 dollar price difference then the choice would be more in favor of the Yugo

but if prices drop below what it cost to produce a product then a business has a very real chance of going bankrupt

also we must not forget capital and resources in our model but so far i think that it is a good start

korn469

[F_Smith]

Just wanted to throw a term out:

Joker, in his post above about consumption changing due to price, is called 'Price Elasticity of Demand'. Each type of good is thought to have one.

The more of a 'need' a good is, the less 'elastic' it's demand curve.

Sorry, just a random thought. Carry on, good work.

[S. Kroeze]

Dear Joker,

I think you are moving in the right direction and that the model you presented might prove to be a good starting point. The fact that it is still relatively simple is in my opinion only an advantage. And I agree that food, clothing and housing are the most essential human needs. One could possibly add water and fuel.

Yet I think you should add some basic resources (grain, wood, mud(for bricks), stone, animals, perhaps metal) as soon as possible. The production of food is in a way different because it can actually be consumed in the form it has- besides some cooking. Production of clothing and housing is more complicated: first wood, bricks and/or stone should be produced and collected before the actual house can be built.

Another element soon to be taken into consideration is the durability of an article: in general food has to be consumed immediately, clothing will last about twenty years, while a house can stand straight for some 300 years. This will influence the behaviour of consumers. When you live in an agreeable house you will not buy a second one; only the rich and famous can afford such extravagance.

I agree with the remarks made by Korn but would again emphasize the exceptional importance of food (and water). When their existence is at risk people are willing to spend all their money and sell all property to obtain food and water. Their only alternative is death or migration. So during shortage of food prices will soar sharply because in this case demand is very inelastic.

An important point is when and how quickly people will change their profession. In general people are rather conservative so my guess is that only 2%(just an intelligent guess) of the population would choose another profession the next year to make more profit, unless they are forced into it. This conservative attitude will automatically cause economic problems. Many people will remain working in a once prosperous sector of the economy when they could earn much more by learning a new profession. Especially older people are no longer able to learn another way of life. They just die out.

When you add some difference in purchasing power and also luxury goods, which you have to produce first, the model might prove to be quite interesting. Production of luxuries should be highly profitable but dangerous for the economy as a whole, because it attracts workers that could sometimes better be employed elsewhere.

Finally a citation:

"Capital is made possible by saving. If resources are consumed they are obviously not available for capital cumulation. If you eat your cow today you cannot hope to have your milk tomorrow. Only by forgoing present consumption can a society cumulate capital. It is generally admitted that in any agricultural society, given the low per capita income, per capita saving is -in absolute terms- rather low. This circumstance is badly aggravated by the way saved resources are used. Temples, pyramids, mansions, jewellry, warfare, and so forth generally absorb a large quota of resources squeezed out of current income. Furthermore, pre-industrial societies are typically characterized by inadequate transport facilities. Mass transportation is generally non-existent and communications are costly and insecure. Consequently any pre-industrial society must keep inventories in much larger proportion to current production than any industrial society does. This is true for any type of commodity, but particularly so for basic necessities. Such inventories are a form of investment, i.e., of capital accumulation, but with a 'stabilizing' character. Generally investment of a 'developmental' character is very small in any agricultural society.

In all agricultural societies of our past we find that, mainly because of limitations of energy sources known and exploited, the great mass of people could hardly afford to satisfy anything but the more elementary needs, food, clothing, and housing, and even these at rather unsatisfactory levels. Correspondingly, most of the available resources are employed in agriculture, textile manufacture, and building.

Of these three sectors, agriculture is always by far the predominant one. It absorbs the greater quota of available capital and labour. Further, it somehow represents the pivotal point around which all other activities tend to revolve. Building makes a large use of timber. And textile manufacture uses materials -wool or linen, cotton or silk- that are also produced 'in the fields'."

(source: C.M.Cipolla:'The Economic History of World Population',1962)

Sincere regards and good luck!

S. Kroeze

[korn469]

hmmm i was just thinking and i had an idea...maybe the relative amount of venture (free?) capital available in a nation could determine how quickly new businesses appear in that nation...of course this would be one of the later ideas to be implemented but a nation where funding is easy to obtain would encourge economic growth

think about which country do u think you'd be more likely to get funding to start a new business...the USA or Uganda (i'm not picking on Uganda i'm just making a point)

korn469

[The Joker]

Korn:

I agree that the current, simple demand function is simply too inaccurate. But I don't think there should just be a maximum on the amount of goods people want to buy. In stead we will need a better function to describe demand.

I never thought I would say this, but I actually wish I had my old highschool mathbook now. It would help when finding the function we should use.

quote: supply is low prices go up then either supply will increase or demand will decrease...if supply cannot meet demand then prices will increase until supply and demand reach equilibrium

I would actually prefer, at least for now, that supply didn't change within one turn. In stead it should change between turns. This will mean that when finding the price and demand for a good the supply will be constant no matter what the price is. I know that this is not completely realistic, but if the supply is to change together with the demand we would require a whole new level of complexity in our functions. Cause when the supply changes then the demand for labour also changes. This changes the wages, which again changes the demand. This will mean that we would need differential calculations, something I would prefer to avoid.

Therefore, what I think should happen when the price is high is, that the demand will drop to match the supply. But before the next turn the high price would mean that a lot of people and capital would move over to this particular sector, which would increase supply the next turn. If this system is done right this should work good. This would just reverse when the price is low - the next turn supply would have decreased. This way the economy would always move towards an equilibrium.

quote: also keep in mind that everything has a production cost associated with it and that the goal of a business is to make a profit, also we need to make our model complex enough so that price is not the only factor which determines what goods people buy

Yeah, I'm working on that. But first I would like to have some supply and demand functions that works great, so we don't have to worry about it anymore.


Amjayee:

quote: You said that you can't see the forest for the trees; actually, there is a saying that game designers can't see the trees for the forest, and the programmers can't see the forest for the leaves of the trees!

I can imagine!

quote: -people have minimum demand they need to survive. They will do anything to get that amount. -people have a maximum demand; they will not buy anymore that product no matter how much they have money. -housing: people try to have at least some kind of shelter. If they have money, but there's no supply of houses, they will build their own houses. -When demand for example for houses rises, some people will start producing houses; that might depend on the initiative of the people or something, and how much smaller the demand is than the supply. -combining the two former sentences: if there is not enough supply for houses, and people have money, some people will build their own houses; some people from those will find out it's fun and possibly profitable to build houses, and start building houses for other people as well. -when supply for products is higher than the demand, the prices will go lower AND supply will go smaller, until surplus production is sufficiently small.

Hmm... I would actually prefer if we have an all mathematical model. I will try to think of actual functions to model supply and demand.

Besides this I'm just looking forward to that demo!

BTW: In the Civ3 Forum I made a thread about our name change to GGS about 3 weeks ago. But since I was locked out of Apolyton I couldn't really tell you all about it. Sir Shiva, a civer from India wants to know what kind of programmers we need. I have told him that we needed C++ programmers, but as a programmer yourself you could propably go into more details.


F Smith:

Yeah, I know. Is there any way to find the actual elasticities of different goods? Also, do you know what type of function would best describe demand?


S. Kroeze:

It is really great that you make these posts once in a while. They really help us out. Thanks!

For this basic model I would mostly just like to find a way to handle supply and demand. I have been thinking about how to implement raw materials and capital/investment (I actually think I have a pretty good investment idea), but I don't want to spend time implementing these things before the very basics of the model is set and working.

------------------
"It is only when we have lost everything
that we are free to do anything."
- Fight Club

[korn469]

hmmm i've been thinking and this is what i have came up with

ok supply is a function of demand...supply will increase if there is a great demand and supply will decrease if there is little demand

S(d)

as prices go demand goes down

hmmm so tear this function apart

S(d)=d/((P)^2 + 1)

where d is demand p is price and s is supply...

we'd have to keep track of d...hmmm more thought is needed on this but i must go now

korn469