24-07-2000 04:08									#1
	Economy Model Version 0.2 Put an end to popups!

I am posting this in it's own thread just to make sure that everybody sees it.

It is a pretty rough start, but it starts from the basics and works it's way up. Unfortunately it is so complex that it is almost impossible to sit down and calculate what will happend the next turn via some know parameters. Therefor we will propably need some economy program (only text based) to balance it off, after the model has been somewhat settled, but before it is included in the game.

Ok, here goes:

ECONOMY MODEL VERSION 0.3


INTRODUCTION

Economy is one of the major factors in the game. Economy is the foundation that man y of the game's aspects. Trade, production, politics and much more. Therefor a good economy model is crucial for a good civgame. Civ2's economy system was ridiculously bad. Not only had it very little to do with reality, it also required far too much micromanagement, and it neglegted a lot of interesting options that could be possible with a good economy model.

Making a good economy model is, however, a very complex matter. Guildmaster showed us what economy should do. Now I will try to show how we could make it do that.

My economy model is based on real economic theory. It has only been slightly changed to meet the needs of a computergame. It is based on a microperspective, as it focuses on the individual consumer and producer. As far as I know this is the best way to calculate prices. Of cause I am no expert in economics. It should, however, generate some macroeconomic effects, which is good. The model is really complex and requires a LOT of calculations by the computer every turn (unless we find a more simple intercity trade system this number might escalate into millions of needed calculations). If this is a problem then we will have to figure out something simpler. If not then let's go!


DEMAND AND PRICE

The people of your civ are consumers. They demand goods. What they demand is determined by their income level. The more wealthy they get the more goods they demand. The basic price for 1 item of a good is 10 credits. This is the same for all goods. This price will, however, change due to the supply/demand situation. My basic population unit is 10,000 people. 10,000 people will eat 10 food per turn. Food is the basic good. People will buy food before they buy anything else. This will mean that the first 100 credits people get will be used to buy food, at the basic price of 10 credits per food item. When people's income rises above 100 they will demand other things. Maybe they will use the next 50 credits they earn for small consumer goods, to a total of 5 items of that. The next 30 could be used for clothes, the next 20 for more small consumer goods and so forth. This will mean that at a basic price of 10 credits per item a PU with an income of 190 credits demand 10 food, 6 small consumer goods and 3 clothes. How people get payed is described later. In the late game people will have incomes that are way larger than these. Today people will have incomes that propably exceeds 5000 credits. This will mean that they will demand loads of goods of all different sorts. When people get very high incomes they will also demand services, which will be delt with a little different than goods.

The demand changes when the price changes. The amount the demand drops when the price rises with 1 is called the elasticity of that good. If the elasticity is 100 then the demand will drop by 100 when the price rises with 1. This means that if the price is 15 then the demand will be 500 lower than the basic one. If the elasticity of a good is 20 and the price is 8 then the demand will be 40 higher higher than the basic one. The elasticity is individual for each good and will not change. I am not sure if this is a good system, though, as the demand is very different in a large city compared to a small one (a demand change of 100 with a total demand of 100000 is a lot different than if the total demand is 400), so perhabs using percentages is better, so an elasticity of 10 means that demand will drop 10% when the price rises 1. This system will mean that the demand graph will not be a straight line, but would get closer and closer to 0 as the price rises, but would never reach it. Or we could use a system where it drops 10% of the basic demand when the price rises with 1. This has the disadvantage that with elasticity 20 the price can never exceed 15. I am not sure what system we should use.

Supply of a good is the amount of it available in the city. It is determined before the turn and before the demand is determined and will not change untill the next turn. This will mean that no matter what the price is the supply is the same. I know that in economic theory supply usually rises when demand rises. But in the game I think that this should not be in the short run that a turn is. In the longer run supply will rise when the price is high. But this includes more factors, which I will explain later.

The city is the basic unit in the model. Demand, supply and price is calculated for each good in each city. I will therefor look at the individual city when explaining this: Via the income/demand algorithm, which is the same for all classes (there is different classes with different incomes and therefor different demands) the total basic demand for a good is calculated. The basic demand is when the price is 10 per item. The way the price is calculated is best shown via a simple supply/demand graph. The price is the x-axis, the amount is the y-axis. On this graph the supply is a horizontal line, meaning that it is a certain amount no matter the price. The demand is a line with the total basic demand as the amount at the price 10. The elasticity of the good is the inclination of the line. It is always negative. This all means that the supply and demand graphs will meet at a certain place. The price at this point is the end price. The new demand is the consumption. It is shared among the consumers, so each consumer gets the same percentage as it got from the basic demand. This is then done for each good.

If the elasticity of all goods were 10 (using the third elasticity system, although I am still not sure what we should choose) then people will use the same amount of credits on all goods no matter what the price, as a price rise would result in an equally sized demand drop. Of cause this is not always so. If the price on food goes up then people will not just demand less of it. In stead they would reduce their expenses on other goods and use more money to buy food. On the other hand if the price on champagne (not that it should be a good in the game, it's just an examble) suddently doubled then people might not want to spend as many money on champagne as they did before. In stead they might spend even less money on it, and in stead buy other things. All this has the effect that the end expence people have on different goods might not be the same as the basic one. This has the effect that people, when the demand for all goods have been calculated, might end up spending more or less money than they have. If this is the case then the demand will of cause have to change. If people spend more money than they have then they will reduce their demand on the "last" good that they buy (the one that they use their last money on after having bought food and the other basic goods). This will be done for all income classes and a new demand is calculated for some goods. This is put into the supply/demand algorithm, and a new price is calculated, so income and consumption will match. If people have spent less money than they have then they will spend more money on the basic goods (starting with food and moving upwards) if a high price has made them demand less of these goods than their basic demand is. Otherwise they will buy some more of the goods that they would have bought if they had a higher income. Again demand and price is calculated to make sure that it matches.

When a good has a low elasticity it means that it is an essential good that people really want. They will therefor go really far to get it, meaning that consumers will overbid each other, in an attempt to get more of the god. Of cause the supply wont change no matter how high the price gets, so people only makes it worse for themselves when they push the price upwards. The next turn, however, a high price will make the supply rise.

When there is a lack of food (less food than people eat) the price on food will be pushed upwards. The price system here needs to make sure that the upper class will push the price up to a point where the lower classes can no longer afford it, making sure that the upper class gets enough food and the lower classes get less food than if it was shared equally among the people. The same thing could happend with other basic goods. This will mean that it might be best for you in some situations to rationalize a good (divide it equally among your people) as it, in the food examble is better to have everybody eat 80% of the food they need (at which point they will still live) in stead of a small upper class get all the food they need when the rest of the people die and riot as they only get 50% of the food they need. Of cause there will also be disadvantages with rationalizing goods, which will be explained later.


SUPPLY AND INCOME

Before I described the supply as being a horizontal line. This is realistic when looking at one moment in time, as there is only a certain amount of goods available. In economic theory, however, supply rises when the price rises, as when the price on a good is high then a lot of people will want to produce it. I think that in our game it would be better to not do this, as it creates a much more vibrant and living economy when the supply changes over time. And it will change over time. Between each turn supply will change according to the price. If the price is high the production (supply) will rise, when it is low it will drop. Therefor in the long run supply will follow price. The lag in supply is what will make the economy dynamic. Without it it would be static. Excactly why supply follows price will be described later.

Production is started and run by the private sector (or sometimes by the government).

First I will look at production. As Cylon correctly mentioned you need some things to produce. These are:

Startup capital: The amount needed varies from good to good. To start producing cars or space shuttles you need a very high startup cost, where other sectors require only little. Some sectors would not require any startup costs. The startup costs is what makes trade between cities profitable. In stead of wasting lots of money to start production of all goods in all cities the cities can specialize in a few goods and then trade with each other, which makes sure that they can produce more and thus get wealthier. The startup capital will not depend on the amount produced, and would be required one time when starting production of a good in a city, unless the city stops producing the good. In that case startup costs are needed again for production of that good to start again in that city. Maybe the startuo costs should actually go to buying actual machines.

Raw materials and half fabrikata: These will also have a price, but here the producers would be consumers (normal people don't demand iron ore or raw oil). The supplyers would be either be farmers (in the case of cotton and such) or other producers, who work in the raw materials sector. Price will be determinated for this as it is for goods. The amount needed of this is determined only by the size of the production. The beauty of this is, that we can have an economic crisis developing realistically: a war in a major oil exporting country emerges, and blocks this country's oil export. The world supply of oil drops, the price rises and everybody will have to pay more to get oil. As oil is needed for a lot of different goods the production costs of these rises. This means higher prices and/or higher prices on goods, without the wages goes up. This means fewer goods for all.

Money: To produce 1 item of a good you also need some money. This covers all the small expenses that can not be put into any of the other categories.

Land: Although it is usually included in economic models I think we should rather include it in the startup capital.

Labour: This comes from the workers. Labour is treated as a good like any other. The supply is the total amount of people in the working ages in the city. There is, however, differences between the labour market and the other markets. First the supply wont change very easy. The only way this can happend is with more or less people. As it will take ~20 years for kids to be old enough to work, this takes long. Migration is also needed. We must develop a migration model, with living standards (amount of goods the workers can afford and the access to other things like libraries), distances and cultural differences as primary factors. The demand for labour will not be decided by the wages themselves, but rather on the profitrate of the production and therefor indirectly on the prices on the goods. The basic wage for 10,000 people is 100 credits in the beginning of the game. This will allow them only to buy the food they eat. If the wages for several turns in a row doesn't give people enough money to buy the food they eat they will begin moving to the rural areas, where they will become independant farmers.

Labour is propably the most interesting of the productive factors. Each 10,000 people unit would, as previously mentioned, get 100 credits per turn in the beginning. They would create 100 labour per turn. The wage would be changed by unemployment or lack or workers. This will either make the wage go up or down. The productivity (labour per people unit) would be decided by technology level (new advances would rise it), and would propably rise a little every turn (like 2%). This percentage would depend on how fast technological development is being made. There should be some way to replace labour with capital (by buying machines), and I am not yet sure how city improvements like factories (which should most oftenly be built automatically by the private sector) should be handled, but it should be possible to fit it into this system. The wages would propably rise as productivity rose. But it shouldn't be automatic. It should be possible to have wages that were low although productivity was high. We will need to find some way to calculate wages, including supply/demand, productivity and more. Labour unions (which you could allow or ban) would put the workers stronger, and would make sure that they would not decrease in wage. Unemployment welfare (a public expence - you would set the welfare at a certain level per 10,000 people group. This would mean that people would never work for less than 10% more than this level, which in itself could create unemployment). Perhabs we could include a few ecucation levels, and independant supply and demand for each of these. This could, however, turn out to be too complex.

A good would require certain amounts of each of the productive factors. Cars might have startup costs of 10,000,000 credits (this number might be way off), and 1 car item could require 1 iron, 1 oil and possibly more raw materials and halffabrikata, and 20 labour. This would be different for each good. Each city has different avalibility of each productive factor (some cities have loads of cheap labour, others a lot of free capital) and this determines what goods will be produced in what cities.

The private sector is divided into two types (apart from the nobility and the independant farmers who only produce food, cotton, silk and such. Food production would propably be handled in a slightly different way than the others, but would use the same rules. For now I am focusing on the production of manufactured goods), the bourgeoisie and the little bourgeoisie. The first are large producers with employees, the latter is small producers who work in their own company and have no employees. They would not have much capital. To portray this I have chosen to only let them work in sectors where no startup costs is needed. They would not recieve wage, but would get the surpluss they make. This will make them vulnerable in a crisis. The relationship between the bourbeoisie and the little bourgeoisie would be shown by the economic concentration level (EC). This would show the percentage of workers being employees working for the bourgeoisie. If it is 0 then all would be independant little bourgeoisie's. The EC would be determined by the economic factors, but generally It would rise unless you did something about it (it would be beautiful if the economy model could do this by itself). There would be an EC for both farmers and producers. The bourgeoisie would be a very little group, who didn't worked. Their number would be decided by the EC, but they would not be a part of the workers.

Trade is a special sector. Traders do not make any goods themselves. In stead they trade goods. For trade (everything but complete autarky) to excist traders are needed. 10,000 traders can trade x number of items per turn. The number would of cause need to be balanced, and would rise with technological advancement. As income traders recieve a certain percentage of the price the consumers pay for the good. The rest of the purchase price goes to the producer. Trade would not need any startup costs, so the little bourgeoisie could also work in the trade sector. Employed workers in the trade sector would get the same wage as all other workers.


SURPLUS AND INVESTMENT

Each class has a savings level (SL). It shows the percentage of the income after tax, that is used for investment. For the bourgeoisie and the little bourgeoisie this is very high, and for the workers it would be pretty low (0 untill they became really wealthy). The investment is used to start new production, to buy the neccesary raw materials and hire the necessary workers to make goods that can be sold the next turn or, if the potential sale is large enough, start production of a new good in the city. Again the little bourgeoisie would only operate in sectors with no startup costs. Every turn it is checked what the profit (sale minus expenses) for all goods in the city is. The most investment will be made in sectors where the profit is high. A simple calculation method is used to calculate the investment in each sector. If a sector looses money then the next turn there will be less investment made there. If in the previous turn 1000 credits were invested in a sector, but it only brings back sales of 800, then the next turn the investment in that sector will only be 800 credits. In a sector where sales equals expenses or where there is actual profit then there will be invested the same amount of money here as the last turn. The sectors with profit will also face an increase in investment. It will work so that the investment capital beyond the amount that was available the previous turn is pooled together, and shared between the sectors with profit. It will work so that the profit for each good is calculated and compared with the profits for other goods. Examble: In a simple world there is 1000 availible credits still to be invested, and 3 sectors with profit. The first gives a profit on 3 credits per 10 invested, the second 2 and the last 1. This means that the first recieves 3/(3+2+1)=3/6=50% of the availible investment, the second recieves 2/6=33% and the last one 1/6=17%. Note that in a sector with demand but no supply there will be no price and no comsumption. In that case the potential price (the price at which demand is 0) is used to determine a potential profit, and so investment is likely to be made in that sector.

Most investment would create profit. Every turn the total investment of each class would be calculated, the profit of this amount is calculated and sent back to the investing class as income shared equally between the class members. The total income is calculated, with the savings ratio a new investment amount is calculated and so forth.


INTERCITY TRADE

The city is the basic unit in the economy model. Perhabs we will later have to make it the province, to reduce the amount of calculations needed, but for now it will be the city. Each city would have it's own price on the different goods. Consumers would buy goods from the cheapest availible source, whether this is in the city or not. To move goods between cities some money would be needed to take care of the transportation. This amount depends on the movement points needed to travel between the two cities. This makes sure that roads and railroads reduces this cost, and that traveling 3 hexes across grassland isn't like traveling 3 hexes in thick jungle. Technological advances would reduce this cost. In the beginning of the game the movements costs would be high. This would make out-of-city trade expensive, and limit intercity trade to cities right next to each other. In the endgame movement costs would be reduced to virtually nothing, creating a world market price on most goods. Intercity trade might turn out to be highly complex, and involve millions of calculations every turn. I have therefor not yet figured out excactly how this would work. But of cause local suppliers would have to compete with out-of-city ones. If they can not they will loose market shares, production would decrease and unemployment would be the result. This would especcially be catastrophic for you if the supplier was foreign. This would mean that production would decrease in your civ and that your people's living standards would drop. But of cause, in the long run and in the big picture division of labour is good and raises the overall living standards.

When shipping costs makes intercity traded goods as expensive as those produced inside the city then these cities becomes one market for that/these goods. Of cause local goods would still have the advantage of not needing to pay shipping costs. But supply and demand would be calculated for these cities as a whole. Of cause this will prove to be very complex when several markets overlap each other. I am not sure if it will turn out to be too complex to work. This is why this part especcially needs some work. It is, however, a very important part of the government model.


INTERCITY INVESTMENT

Capital will flow to where the profit is highest. This also includes other cities than the home one. Capital would have some kind of max limit of where it is willing to invest. Untill the 18th century this limit would not exceed the city borders - people will only invest in their own city. The max limit would grow rapidly, untill it covered the entire world in the sixties or so. Within the max limit investment would flow freely. This would mean that the profit in those cities slowly begins to move towards each other. The cities become more alike. When capital owners invest in other cities than their own they would create jobs there, but take the profit home toe their own city and demand goods there. You as the government could forbid your capital owners to invest in foreign civs (which would upset them) or you could forbid foreign capital owners to invest in your civ (which would make them happy, but upset other civs) or you could set a certain tax on foreign people investing in your civ.

The profitrate drops the more investment there has been made. This is because the price drops the more supply there is. This will also mean that in a free international economy capital will flow from places with a lot of available capital (developed civs) to places with only a little available capital (less developed civs). Just like in the real world.


GOVERNMENT ACTIONS

The economy runs on it's own. You don't HAVE to do anything. But to make the economic development go the way you want it to go, and to benefit from a thriving economy you have some tools to use. First of all you are able to buy stuff from your people. These are all the things you need to maintain your civ, like city improvements and units. I am not sure how to handle the city improvements in our game, but they would require labour, goods and raw materials. With units hiring the men would be seperated from buying the equipment. Here I will just handle the equipment. Each unit would have it's own equipment, which would cover everything they need. There would be one warrior equipment, one tank equipment etc. A unit would be 1 item of the equipment and the men. Equipment can possibly be stored, not quite sure. Equipment can also be sold to civs or bought from them. Of cause selling equipment to a civ would upset the civs in war with it. Equipment can also be transported between cities, like all other goods. I have not yet figured it out, but either the private sector produces equipment with a market price calculation, or the government simply hires some people for the production untill it's done. When making equipment you could choose to buy from the cheapest source, which would save you money, but might be foreign, or to buy it from a certain city, which would boost that city's economy. When making things with your own people as labour there would be two opportunities: Either you hire the people for the wage they would otherwise get. This would upset the bourgeoisie and the nobility a little as it would push the wages upwards, or you could force people to work for you at a wage that you decide, which would upset the workers a lot. In the last case you would need military units to keep the workers from rioting. The less money you paid them the more unhappy they would be. You could pay these workers nothing if you wanted, but this would mean that they couldn't buy any food, so they would all die the next turn. A slavery model is also needed for the game. Basically you decide whether to allow it or not, and then it would work on it's own. You could also use slaves. Slaves would come from your civ or from conquored cities, would only need food, but would need units to keep them from rioting.

One of the best things about this economy model is, that it allows virtually unlimited amount of economic management from the player. At the beginning of the game the player would decide how much involvement he wants. This goes from simply setting a global income/expences level to an extreme level of involvement, where you can set individual sales tax on each good in each city, or to make a progressional taxation system. All with realistic effects in the economy. You can also make tariffs between cities inside your civ. This might be needed in the early game, if the nobility was strong enough to forbid you to tax them. So if you wanted money to go to war and other stuff, you would need to get income from everywhere you can, even if it will halt the economic development in your civ. Your income is used to buy units, improvements, roads etc. It would also be used for more permanent expenses like paying administrators, scientist, public welfare for unemployed people and more. You could also use your money to start actual production of goods yourself, if you wanted to become a socialist/communist state. Here you can decide whether to simply use the market forces or to share the goods you produce evenly among your people. No matter what (but more if you used the latter one) you would need remarkably more administrators the larger your public sector gets. When starting production you might earn money on it. There should propably also be some other penalties for using public production, like slow supply changes and more.


Ok. This is it for now. I was hoping to have a more developed model to present to you, but as this will need a lot of more work I thought that I might as well get some responce from you guys, and use this for the next version.

	24-07-2000 04:14									#2
	Browse Apolyton AD-FREE

Oh yeah.

I have been looking for some economy models on the web. I have found a few, but although I have only skimmed through them I doubt that I can actually understand them...

I would wish that we had a real economist to help us.

	25-07-2000 11:57									#3
	Enter the AD-FREE zone

Okay, anyone who doesn't understand basic algebra, look away

Demand:
I believe the forumla
D = E [-(P-10)]
were D = demand and E = elactisy and P = Price
would be good enough for this.
Setting up this, with a price rise of 1 and an elactisity of 100 gives
D = 100 [-(1)]
D = -100
The Demand has changed, so that it is 100 less.
Another example:
Elactisty: 20, Price 8
D = 20[-(-2)]
D = 20(2)
D = 40
So the product would be in demand +40.

quote: The elasticity of the good is the inclination of the line. It is always negative.

How does this work? Are you sure you didn't mean positive?
Example:
If inclanation of the graph is negative, it will slant L-R Downwards. This means that the new price would be negative. Does this make sense. If I understand you system this is how it works. I have it all drawn on paper in front of me


All over this is an unebliveably complex, yer brilliant system, allowing a wide range of options. I don't know if the computer will be able to handle it though. We'll have to ask Amajyee or Dan on this. Apart from that let's get the 'bugs' sorted out, and get a great system. I might post a link on the Civ3 forum, to attact more attention to this (unless someone else does that first)

	25-07-2000 19:12									#4
	Support Apolyton buy from Amazon

I think you have done great job, Joker. This is excactly what we need; it is simple enough to make, yet provides all necessary complexity for the scale of our game. And to answer heardie's question, computer is especially suitable for the number-crunching required by these kinds of systems; as Joker pointed out, his economy model is almost pure mathematics.

Implementing your model will propably one of the easier tasks, since we simply need to do some functions that process numerical data. The complexity comes from many variables, and balancing off the model, but that can be done by anyone, when the model is running. From computer's point of view, complexity comes from large number of calculations, but we have to remember average pc's of today make hundreds of millions of calculations per second. It will take only a second to make all calculations, perhaps all in all several seconds, since we need to go through all cities and gather the info required by calculations. And we can probably make the amount of calculations smaller with some clever code design, as I told before.

So, let's give some thought to complete the model, then we program a "test bench" for it, and then you guys can play-test and balance it. I'm sure everything will work out; if it takes too much time to make the calculations, which I doubt, we can try to reduce the complexity somehow.

All in all everything in your model looked good, when I skimmed it through. I will later make a more thorough commentary of it, but so far it looks great! Perhaps you could integrate the companies to it soon?

	26-07-2000 19:51									#5
	Support Apolyton, pre-order Civilization IV

I am currently reading some economy stuff I have found on the web, including some models. It would be great to have a real economical model that we could use, and then use this for a huge remake of my model.

It is pretty complex stuff, though.

	26-07-2000 20:06									#6
	Support Apolyton buy from Amazon

Oh. About the corporations:

I think that they should simply be represented by the private sector (the bourgeoisie), so they will make stuff and invest in foreign countries etc, and with the government model, in which they could make suggestions, demands and more, towards you and towards foreign civs.

	29-07-2000 01:27									#7
	Put an end to popups!

Excuse me for interfering, but as we Clashers have a little bit more experience on such things, let me give you some free advice (but spare me the tomatoes ):

1) Don't let one guy develop a game model all by himself. He is bound to take more time and he is in danger of overlooking some critical points. Not to mention that he might dissapear from the boards, leaving you empty-handed. The cooperation of at least two developpers is required (one had better be a programmer)and this work had better be here, on public grounds.

2) Do not let your models stay on pen and paper. Since you have alot of programmers among you, it's better to follow the "Bazaar" approach, as my fellow F_Smith has repeatedly suggested. If you lack the ability to begin some test-code on the model, try modelling on a spreadsheet (as a matter of fact try developping on a spreadsheet). This will give faster results.

3) Do not develop "models in a jar"; always have in mind that the various models have profound interconnections, especially those on economy, government, society and demography. Copatibility and à peu près equality in depth and development should be primary concerns.

I wish Clash had followed these principles from the beginning; as a matter of fact I wish it was following them even now...

------------------
"In a time of universal deceit, telling the truth is a revolutionary act."
George Orwell

	30-07-2000 00:48									#8
	Support Apolyton buy from Amazon

Thanks for the advice axi.

I agree with all of them. Unfortunately we are not very many OC3'ers at the moment, and several of the previous members have left us pretty fast. This makes it hard to get several people to do one model. Apart from this it seems as if I am the only one interested in this part of the game at the moment .

	01-08-2000 00:53									#9
	Full PM-box? Change here!

Heardie:

Sorry for the delayed responce. But here goes:

The demand formula:

Well. I am not really sure whether to use the elasticity as a percentage of the demand, or to use it as a set amount.

The formula would look like this:

Demand = D + E (10 - P)

Where D is the original demand, E is elasticity and P is the price.

As elasticity is a positive number it means, that when the price is higher than 10 the demand will be lower than the original one, and when it is lower than 10 it will be higher than originally. The problem is, whether to make the elasticity a set number, or a percentage of the original demand. Both has it's advantages and disadvantages.

I am currently working on a way to replace this demand calculation system with a simpler and better one. This is, sadly, a very dificult job. I will of cause notify you when I come up with something.

Besides this I am working on a large new version of my model, which should include more on capital and investment.

The inclination of the line:
If you look at the demand function as a graph, with price out the x-axis and amount up the y-axis, the demand will be high when the price is low and low when the price is high. This will mean that the line will have a high y-value when the x-value is low, and vice versa. This way the graph will have a negative inclination.


PS: I hope you will sone go through this system really critically, and point out all it's flaws. this way we can begin working out a new and better version of it.

	01-08-2000 16:57									#10
	Full PM-box? Change here!

About elasticity, I think you should go for the percentage of the original value system. That makes more sense. About commenting the model, I'm working on it, as many other things, too... hopefully soon.

	01-08-2000 20:22									#11
	Support Apolyton or Terrorists Win

The problem with using a percentage is, that the maximum price for a good (the price at which point the demand will be 0) will always be the same, no matter what the basic demand is.

Another problem with the price and deman deal is, that I would like to make it possible for the upper class, in a time where there is a lack of food or any other basic good, to push the price upwards untill only they could afford to buy the food they wanted. This would mean that in stead of everybody getting 70% of the food they need (which would keep most of them alive) the upper class would get 100% of the food they need and the lower class might only get 50% of the food they needed - which would mean that many people in the lower class would die, or riot. This would give you the possibility to rationalize food, which would reduce such problems.

None of the demand functions that I have thought of untill now can do such a thing.

	02-08-2000 03:28									#12
	Support Apolyton, buy Call to Power 2

Correct me if I'm wrong, but shouldn't there be some situations where the government (in this case the player) can decide what the price of a good is? Particularly in communist countries, but also during the depression some of our own American politicians used that technique to try and stem the crisis.
SO how are we going to work that in?

Also each trade good needs to have a certain quantity per unit, to determine how much of the good is mined vs. how much of it is sold and taxed and etc. Par example, we over here tax gasoline by the gallon, buy it sometimes by the barrel, etc. All kinds of things. I propose that each trade good for simplicity's sake be mined by the x, sold by the x, taxed by the x, and etc.
Here's another thought... shouldn't we also concern ourselves with trade good quality? Perhaps every good can be assigned a "grade" to determine how good it is for use with whatever it is used for. Kinda like how some crude oil requires more refinement, thus you get less gasoline per barrel of oil and thus claims a lower price than high-grade oil. The same grading scale can be used with all trade goods I think, like perhaps a "grade A" car might represent a Bently or a Ferrari, while a "Grade B" car could be a Mercedes or BMW and so on down the line. With this grading scale, you could take the GNP or average income level of a civ and determine the grade of a trade good with the highest demand. ie. we could say that for a country with a low income level you couldn't sell grade A or B cars because no one can afford them, but in a wealthier country nobody wants to buy grade F cars because they suck so bad. Same could be said for just about any trade good.
Perhaps the origional manufacturer or supplier of a good (the one that invents it) would get a +1 grade bonus no matter how terrible it may be in the present. ie. French sparkling wines are grade A wines but are not as good as California sparkling wines which are also A wines. They are A wines because of the inventor of sparkling wine is french. What think you of this idea?

	02-08-2000 16:15									#13
	Remove this text

Oh, sorry, I was meaning the system where the new demand is elasticity times the last demand level. Elasticity could be positive or negative, depending on situation. I think you said with this system the demand level could never reach zero in this system.

Guildmaster:

I agree that the quantity of goods should always be x.

I really like the grading system. It would be great to have it in the game, but the complexity is frightening me. Having 7 grades for each good would make the amount of goods seven times larger. We need to think about it - if it is possible, we should try to make the grade system. Just let's make the simple system first, and find out how complex and time-consuming it is. Perhaps we can implement this idea in some simple way, that doesn't increase the complexity very much. To be found out...

	02-08-2000 19:17									#14
	Inflate your Upload Space

Just a quick thought:

The 'price elasticity of demand' for a product is actually dependent on how necessary that product is. Whether it's a need, or a want, and how much it is needed or wanted.

People will still buy a certain quantity of food, no matter how much it costs (within reason), and they won't consume much more of it no matter how cheap it gets. Low elasticity.

Luxury items are exactly the opposite.

	03-08-2000 04:45									#15
	Support Apolyton or Terrorists Win

The grading system could work like this:
The game would track the standard price of each good, and each grade of the good would be a set percentage of that good.

Say C grade is the standard, a Volkswagon for $20k-$25k. An A grade would be 1000% the std price for a $250k Bently or Lamboroghini, while a BMW would command a 200% price increase over standard. Grade D might be 75% and so on, that would make pricing the different grade goods real easy.

As for demand, the overall demand for the good remains the same. People are going to need cars regardless, and generally they are going to demand the same ammount regardless of the grade. However, the demand for each grade is strictly a factor of income level. If the income level isn't high enough for the populous to meet the demand (ie. they all need cars but nobody can afford them) they would simply buy a lower grade vehicle... which would in turn mean they buy from someone who manufactures the lower grade cars.

Also note that the profit return wouldn't be as high for lower grade goods, but the mass would be there to make up the difference. Mind you under a capitalist government, all this is done for the player so they don't have to touch it.

Note also, that raw minerals when mined from the earth might also have a grade, to signify some deposits are simply of better quality than others.

	03-08-2000 17:32									#16
	Support Apolyton, buy Civilization 2

F_Smith: I think that's what elasticity is used.

Guildmaster: Yes, that sounds as a plan. I think that would work quite well. The people should perhaps have a demand for every good, and a grade; that figure might be a floating point number, for example the grade they want for their cars might be 3.3; mostly grade 3 cars, some grade 4, and little everything else. This would decide the amount of money they would spend on cars. Grade system sounds great now! I think we should try can we make it work.

	05-08-2000 17:09									#17
	Support Apolyton, buy Alpha Centauri

F_Smith:

Yeah, that was what I had in mind for elasticity.


Guildmaster:

I think that the graded goods could very well work, and would also be a good addition to the economy model. But I think that we should start out by making a more simple model without graded goods, and then introduce them later on.

One thing: I am not sure whether I think that raw materials should be graded. How are we to decide the difference between a car made by grade A iron and one made by grade C iron? It might add too much complexity. I think having the finished goods being graded is propably enough.

	06-08-2000 08:15									#18
	Support Apolyton, buy Civilization III: Complete

The only difference between one grade ore another would be purity, or content. Say you would kneed 10 tons of grade c gold ore to extract one ton of gold, while grade a gold ore is basically pure gold. It only comes to mind when I read an article in the paper about crude oil found in Venezuela that was more like paste, and was much more expensive to refine than say oil found in the middle east. But i see what you're saying, grading raw materials wouldn't really add too much, other than to say some deposits are more profitable than others.

	06-08-2000 20:51									#19
	Support Apolyton, buy Alpha Centauri

Guildmaster:

Well, I think that it would propably be much simpler and easier to have raw material deposits that gives different amounts of raw materials per turn, in stead of different grades of raw materials.

But I am starting to like the graded consumer goods. How about making it so that when people can afford to buy a certain amount of a good, then they will upgrade to a lower amount of a higher grade of that good?

So when 1000 people can afford to buy 20 car grade F, then they will upgrade to 15 car grade E (which would automatically cost the same). This would give us the opportunity to have graded goods, with all the trade and specialization that comes with it, but without needing complex price calculations for every single grade of every single good.

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

	07-08-2000 08:54									#20
	Browse Apolyton AD-FREE

So wait, if 20 people who can afford grade F cars automatically buy 15 grade E cars instead, what are the other 5 people goin to drive?
Perhaps I didn't fully understand what you meant?

The way I meant it, say you have a demand... some 1000 cars needed in France. You look at the wealth demographics and find 2% extremely wealthy and 80% extremely poor with an %18 middle class. So, 20 Bently Arnage will be needed (2% of the population, 2%of the demand) also 180 Toyota Camrys and also 800 Yugos. To get Bently you have to buy from England so you buy 20 Bentlys from England, 180 Toyotas from Japan, and 800 Yugos from Yugoslavia.
This is a basic outline of the graded consumer goods vs. wealth demographics. Either way, 1000 cars are demanded and 1000 cars supplied. I'm assuming we're going into more detail than just rich/poor/middle class so there's also more than three grades.

	11-08-2000 19:28									#21
	Tired of ads?

Well, I don't think a car unit would actually mean one car. It would more mean the amount of cars that 1000 people would drive.

My concept was, that when 10000 people get wealthy enough to buy 20 grade F cars (which would propably mean 2 cars per family) they would in stead upgrate to 15 grade E cars (which would automatically cost the same). This would mean that the wealthiest half of these people would have 2 E cars per family, and the less wealthy part has 1 E car per family.

I think your system might create some problems, as the luxury cars are much more expensive than the cheap cars. This will mean that the wealthy people will spend more money on cars the the poor ones, but it also means that in a market economy it would create an independant supply and demand for each grade of the car. And thereby we might end up having 100s of different goods.

I think that although graded goods would be a great addition to the economic model, we should start out by making a more simple model, and then gradually build it up to be more advanced.

	12-08-2000 09:00									#22
	Get a bigger avatar today!

Firstly, we already have a good basic model and this is adding a little complexity to it, just like you suggest.

Second, to have more that one demand for any given supply is exactly the thing I was looking for.

(still using the car example) If people automatically bought a higher grade car thet would mean lower grade manufacturers would be automatically be out of business, even though the demand is the same. If there are 2 cars per household needed why would they settle for only one simply because it's more luxurious even though they needed two?
Or perhaps one car in every three households, why would people then go to one in every 4 or five just because they can afford a better model?
I think that if youe example were accurate, auto industries in America would be dead compared industries in England and Italy who for the most part stick to grade A Rolls, Bently, Ferrari, etc. because instead of meeting the demand for volume say ten execs who all need their own car would get together and buy 1 Bently Arnage. It just isn't logical.
Now on the other hand, if the opposite were true, only American and Japanese cars would be sold because of the sheer volume needed to meet the demand at a low cost (ie. a lower grade) But that would mean no one is buying British or Italian cars.
As for the complexities involved with a multi-grade demand structure, they really are quite simple. While the demand for said good (in this case cars) is a factor of this and that and everything else, demand for each grade of that good is solely a factor of wealth demographics which makes it quite simple to calculate.

The main strength of this system is that it allows smaller countries without the mass resources of world leading empires to enjoy a portion of an industry simply by producing small ammounts of an extremely high grade good.

	13-08-2000 23:03									#23
	Increase Your PM Length

Well, I really don't think my model is ready for implementation. I am thinking about ways to improve it, even at the most basic level, but it is pretty dificult.

I would rather wait a little while untill adding further complexity to the model.

The demand model for graded goods:
I think it could make sence. First there is no such thing as needing 2 cars. I think that it will make sence to use a system as the one I proposed. First it is really simple to do, and it still offers the advantages of the graded goods. All it needs is a simple function for the costs of the different grades of the goods. Imagine that whenever 10,000 people can afford 3 cars of one grade, they automatically upgrade to 2 cars of the next grade, and so forth. This will make sure that no family buys 19 grade F cars.

I don't think that you understood my system. Of cause people would not upgrade to higher grades if there is less than 1 car per family. It is easy to make sure that such a thing will not happen.

Examble: In a society there are 3 classes. The poor, who demand 20 car units per 10,000 people, the middleclass, who demand 40 car units per 10,000 people, and the wealthy, who demand 120 cars per 10,000 people. This means that the poor have in average 2 grade D cars per family (10 cars per 10,000 people would mean 1 per family). As they have less than 3 cars per family they will not upgrade. The middleclass can afford 4 grade E cars per family. This means that they will upgrade to 4*2/3=2,7 grade C cars per family. The wealthy have 12 grade D cars per family, which means that they will upgrade to 3,6 grade A cars per family.

This system makes sure that the demand for each grade of a good is put into the intitial demand for the good. It offers a lot of flexibility, as both the amount of cars bought and the grades of the cars are included into the same function, and at the same time it is simple enough to make it workable with a minimum of extra calculations. Simply have a function, in which a grade B costs 2/3 of a grade A, a grade C 2/3 of a grade B and so forth, and everything can be calculated.

I also have to say, that I don't think we will need grades for all goods. In luxury items such as cars or wine it is really good, as it offers a civ the ability to make a niche production, as you said. But I see no need to grade food.

	14-08-2000 04:56									#24
	Inflate your Upload Space

I actually would see a use (not so important though) for graded food items, especially with the advent of artificial additives and colorings, pesticides, preservatives, hormone-injected livestock, "Organic" produce, etc.
Perhaps that's taking it a bit too far though, although I see perhaps the green movement going on in Cuba right now where all agriculture is done by organic means (and producing more per acre too!) as opposed to industrial chemical fertilizers. To top it all off the food tastes better... or so I've heard as it's difficult to get goods from Cuba here in the states.
Either way, different agricultural technologies can affect the food crop output and health of a contry so perhaps foodstuffs don't need to be graded. Or if they do, Grade A sandwich might be a prosciutto & gorgonzola on rye while Grade F sandwich would be bologna & PP cheese on generic bread...

But for luxury items and MFG goods

Please explain better, but I don't see how your system wouldn't change the ammt of a good being bought? If a demand is for 20 car units and only 15 are supplied where are the other 5? Or are you saying that each grade represents a point scale where 20 car units isn't an exact figure? Best I gather from your example is that 1 grade A car would represent 12 grade F cars? If this is the case then what's stopping everyone from buying all grade A of a good, since you can meet the entire demand with a higher quality good just by buying fewer? Then who would buy, who would make, and what's the point of having lower grade goods?
Also, if there is a demand for 2 or even 3 or 4 cars in one family why would the family opt to buy only one high quality car, the high demand would indicate either a different kind of vehicle (car for driving, truck for working) in which case 2 still are needed and 1 higher grade car wouldn'd suffice... or more likely there is 2, 3, or 4 drivers in the one family, and while the husband needs the car to go to work, the wife also needs the car to go to work, and the kids one has college to go to and the other is still in high school but they all need to drive because there is a lack of public transit. Why would such a family opt for only 2 or even 3 cars when 4 people need their own? In addition, I was under the assumption that the ammt of a good in each unit would remain constant? If 15 D goods were the same as 20 F goods wouldn't that change the ammt of a good?

I dunno maybe I'm just confoosed

	14-08-2000 05:18									#25
	Support Apolyton, buy Civilization III: Complete

OoK here goes...

The total demand for a good is outlined by your trade minister. Say a province/city in your country has a demand for 100 cars (or units of cars). This represents all the people in the province/city who want to buy new cars just for the sake of it, old cars that break down, new drivers that come of age, perhaps a new model meets a need that hadn't been met before. Or for whatever reason, a total of 100 cars are needed as a "trade demand"

The wealth demographics for this province/city are as follows,
1% are "Very Rich"
5% are "Rich"
29% are "Middle class"
45% are "Lower Middle Class"
20% are "Poor"

So, of ALL the goods sold to these people:
1% grade A
5% grade B
29% grade C
45% grade D
20% grade F

This means of the 100 cars demanded,
1 grade A car
5 grade B cars
29 grade C cars
45 grade D cars
20 grade F cars

SO then you look at the supply, if the supply meets these at market value nothing happens. The cars bet bought the people ar cool, everyone is happy. Now suppose the available supply were as follows:
A cars: 0
B cars: 0
C cars: 15
D cars: 10
F cars: 129
In this case, the price index for C and D cars would go up and all those would be bought, and also 75 F cars would need to be bought to meet the total demand. BUT since the people demanding them got the lower quality goods they will not be happy and this will add to their discontent. If 75 F cars weren't available to meet the total demand then a lot of people in you empire would go "car hungry" since a trade demand is not met. Either way, demands are not met. (A smart player would take advantage of this and subsidize the development of a luxury car)
Now suppose the supply were as follows:
A cars: 15
B cars: 12
C cars: 4
D cars: 1
F cars: 2
Since the supply of A and B cars is so much higher than the demand they will sell for a much lower price to consumers demanding lower grade goods, meaning a large portion of the people will be quite happy they get better than they asked for. BUT still, the total demand is not met so a lot more people will be less happy about it.

	14-08-2000 11:32									#26
	Support Apolyton or Terrorists Win

Guilmaster's text seemed reasonable, somewhat like that it would work. In the final example there was a flaw though; it was over-simplified. If there are only high grade cars available, prices might go down, but still the poor couldn't afford them; that would create demand for lower-grade cars, which would encourage import of lower-quality cars. If those are not available, in a long run they are started to be produced, to meet the demand.

Also, if your poor people don't have enough money, they will not buy those cars; but I think that is taken into account when calculating the demand.

I also thought that this system could be used succesfully also with military equipment; there could be grade A combat fighters or tanks, for example. Perhaps also the buildings and facilities could be graded? You could build higher-grade temples, to make your people more happy, and your neighbors envious.

	14-08-2000 16:01									#27
	Support Apolyton, buy Galactic Civilizations

I still see some flaws in your system, Guildmaster.

How do you know when people are rich or poor? What if the wealth drops, and the rich gets poorer, when will they buy grade B cars in stead of the grade A ones? And what if the price on cars change? Another thing is, that you can not seperate the income of the people and the demand for a good. Maybe the poor people simply can't afford any cars! I think that the system is inflexible to changes in society.

But the main objection I have to the system must be, that it calculates the price on each grade individually. And with 5 grades and 50 goods this will simply create WAY too many calculations for the player, as trade, supply and demand will have to be calculated for every single grade of all goods.

With my system the "wealth" of the people is automatically calculated via their initial demand for cars. If the price on cars change (or if you change the tax on cars), the "wealth" of the people will change accordingly. And it does not mean that people will just pool together to buy fewer cars of a higher grade. It will only mean that when the amount of cars per family (or per person) reaches a certain level, people would rather have fewer cars of a higher grade than buy more cheap cars (wealthy people would rather have 2 Ferraris than 100 Fiat 500's).

I don't think that the demand for cars can be seen as a demand for a certain number of cars. It would rather be a wish to spend a certain amount of money on cars (this amount would be changeable if the price on cars change or if the income of people change), and then the size of this amount will determine how many cars of each grade will be bought.

Graded food:
I really don't think we need this complexity. Of cause in modern times we would eat food of a higher quality than in ancient times, but it might as well be portrayed by having us eat much more food than they did, as we eat a lot of meat, which is less costefficient than grain.


Amjayee:
I generally don't like grading military equipment. Like the training levels it would create a lot of uncertainty, as noone really know what grade is most costefficient.

I don't really like grading improvements either. I think that you should propably just buy an improvement, and then you would have to spend some money every turn to actually use the improvement.


Anyway, I have finally figured out a better way to handle the supply/demand calculations. It is not much different from the earlier method, but it is generally better. I don't have time to post it now, but I will propably do so later today!

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

	14-08-2000 19:25									#28
	Support Apolyton, buy Alpha Centauri

I thought that wealth we already know, and we were already using wealth to calculate initial demand/supply.

I also don't see why the player is going to have to calculate all this? It seems to me the computer ought to be able to do that.
Say yo have a pop-up menu for buying goods...
You say you have a demand for 3 A cars, so you go to your trade advisor who says to buy 3 grade A cars from England will cost 3600 ea. but 3 A cars from Italy will cost 3300 ea. which one will you buy?
Then you also need 12 B cars... Germany has 8 B cars available for 200 ea. while japan has plenty for 333 ea. and Canada is making 14 for 311 ea. So you buy 8 german cars and then 4 canadian cars
and so on and so forth.

	15-08-2000 00:54									#29
	Support Apolyton buy from Amazon

The initial demand will be a certain amount of the good at a certain price. This means that the amount of money people spend on a good is included in the demand. But as the more expensive cars will cost much more than the cheap ones the wealthy people will spend much more money on cars than the poor ones. But this amount is in your system not included in the initial demand. In stead it becomes a function of the wealth of the people, without looking at what people spend on other goods. And therefor we would not really know how much people spend on each good.

My system pretty much does the same thing as your system. It just offers a way to calculate the demand for each grade in a way that is integrated into the demand function. Your system doesn't do this. Therefor I think that my system is not only further developed compared to yours, it is also more flexible.

Calculations:
Of cause the computer will do that. All of that. Including buying and selling goods. You shouldn't have to buy goods from other civs. Your people will do all this automatically. You will, as the government, set the tax level, set tariffs, make embargoes, build units and all other sort of fun things. This will mean that the amount of micromanagement will be decreased heavily.

What I meant about calculations was, that the computer might not be able to handle them. If we have 500 provinces and 50 goods all with 5 grades, this means that the computer will have to calculate 5*50*500=125,000 different prices. Calculating every single price might easily require 10 calculations, and this means that there will be over 1,000,000 calculations to be made. On top of this the calculations for the trade between provinces will escalate this amount even further, as you will have to check the price plus the shipping costs for every province in all provinces. This could mean that in the end there could be over 100,000,000 calculations having to be made every single turn. And this is simply more than an average home computer can do within a reasonable timeframe.


Anyway, to move away from the graded goods I would like to present the new and improved demand and price model:

Demand is handled pretty much like before. When looking at a single good the initial demand is like before when the price is 10 per item. the elasticity describes how much the demand will decrease when the price rises with 1. The elasticity is a percentage of the total demand. For basic goods, like food, the elasticity is really low (less than 5%), and for luxury goods, like wine, it is high (above 10%).

The supply function is what has been changed the most. Before it was a horizontal line. Now it has a positive inclination, meaning that when the price rises the supply rises too. At the beginning of the game the inclination of the supply line for food will be at a level where supply equals demand when the price is 10. Of cause when people get wealthier and there is more of them the demand function will be moved upwards. This will mean that the price of the good will rise.

The supply graph will have to change to make sure that the price wont continue to rise. The beginning of the supply graph will not change - when the price is 0 the supply will be 0. It will always be so. What will change is the inclination of the supply graph.

This will happend via a simply algorithm. Each good has an "optimal" income per sold item. Income is of cause sales - expenses. This optimal income per item is a certain number. For food it might be 3 credits. If the price on food is 10, and the total expenses in the food production didived by the amount of food produced is 6, then the income per item is 4.

When the income per item is different from the optimal income per item the supply graph will change (it will propably not change untill the next turn). When the income per item is higher than the optimal income the inclination of the supply graph will rise with a certain, small percentage (perhabs 1 or 2%). When the income per item is higher than the optimal income the inclination of the supply graph will drop with the same percentage.

This will make sure that the supply graph will follow the demand graph in the long run, but that there is still room for changes in the supply graph without changes in the demand graph.

Imagine that the demand suddently rise with a considerable amount. This will mean that the price on goods will rise, as a higher production requires a higher price (due to the positive inclination of the supply graph). But as the producers have a great time with high prices the amount of producers eventually rise (the inclination of the supply graph rises), which forces the price downwards. The price will continue to drop (over a considerable amount of time) untill the equilibrium is reestablished.

Another examble will be where a war in a major oil exporter occur. This means that the amount of oil produced drops (the inclination of the supply graph for oil drops). This means that the producers of consumer goods will have to pay more for the oil. This means that their income per item drops, which causes the inclination of the supply graphs to drop. This again means that the price on goods rise, without the income of the people does. This means that the real income of people (the amount of goods they can buy) drops. A recession has been made without any form of artificial algorithms!

The total demand the people has is calculated via their income like before. The income is used to calculate how much they demand of all goods (when the price of the goods is 10 per item). The actual price setting is done by first taking the most basic demand - food. The total demand is calculated, the price is determined, and it is calculated how many credits people use on food, and how many they have left. If they still have money left the next thing in the demand function is taken, a price is determined and again the people's spend on the good is calculated, and the amount they have left is calculated. Then the next good is taken, then the next etc. If the people run out of money they will simply consume what they have bought so far. If they end up spending more money than they have then the amount they spend on the last good they bought will be changed to make sure that they spend the same amount as they earn. The other goods they buy will not have to be changed by this.

I think I am done now. I now actually think that no futher work is needed on the demand, supply and price. If anything I have written is not clear please tell me. I am pretty sure that this part of the economy model is ready to go. Of cause the actual production of the goods, where the income comes from etc is still not quite settled. But we'll get there!

At least now we are one step closer to perfection.

I have some ideas on how to do the production of the goods. But I will not post them right now due to lack of time (I also have to post in the population thread).

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

	15-08-2000 09:47									#30
	Got spare money?

Odd but I don't think you understand my system?

The ammt of money people are willing to spend on any good depends on a)the price of the good and b)how much money people have. Beyond that it's not too complicated. So if the income level is lower than the price of the lowest grade of good obviously the demand will be low. The demand increases when the income level rises so that people can afford it... so that the demand of each good is already calculated in the base price.

As for calculations, even 1o million calculations take very little time when the calculations are as simple as
x(a)=10c
x(b)=3c
x(c)=c
x(d)=.7c
x(f)=.4c

but the big thing that still concerns me is this: IF there is a demand for X ammt of a good assuming the people can afford that ammount, why would they buy less of a good just because it's abetter good? If I'm making a cabinet and I need 15 nails, I'm not going to use 10 nails just becasue the're better nails. if 15 people need 15 cars why would they buy 10 cars just because they are better cars what are the other 5 people going to drive? You still haven't explained this?