Economic Model II

[Lord God Jinnai]

Well if ur worried about inflation yet, the player should be able to see atleast the history of it, also unemployment.

[Mark_Everson]

Axi:

Here is a diagnostic trace output of the economy so you can see what I have so far...

Everyone else, this stuff is generally Not going to be shown to the player, its just for me to make sure things are working right. Because of all this effort though, your people should not be Boneheads like those in other games...

A few things you should know:

This square is modeled roughly on the province in the econ spreadsheet.

This particular diagnostic output starts after a few turns have run, so all values for parameters should be good and you should ignore warnings that things should be 0.0 or 1.0 because of startup. Also some things are already partly built which explains why a legion is immediately created. (On the first turn most things exaclty match the spreadsheet results.)

The people store food to eat, but don't eat it yet... so enormous stores of food have built up.

"Cost to buy 800 food = 800000.0" is just because a request to buy more than is available just returns a ridiculously high price of 1000/unit for now.


Adding new order to MapSquareEconomy:
"res kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 30.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = 0.0

Adding new order to MapSquareEconomy:
"res kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 30.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = 0.0

Adding new order to MapSquareEconomy:
"Military - Legion" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 40.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99

Adding new order to CivEconomy:
"Military - Legion" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 40.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99

Adding new order to CivEconomy:
"prod kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 30.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99

Estimated cost of Government purchases
using MSEconomy.calculateCostOfGovernmentPurchases(): 39.338245

Estimated cost of ALL Government purchases
using civ.executeGovtEconOrders(testOrReal, orderName): 158.43204

Estimated cost of 'Military - Legion' Government purchases
using civ.executeGovtEconOrders(testOrReal, orderName): 89.79417
Names of all goods:
"Resources" "Services" "Gold" "Food" "Manufactured Goods"

Executing calculateProductionInAllSectors (phase A)

************************************************** ****

Initial sector Stats: (Wages etc = 0 since no final prices yet)

farm:
Amount Produced = 418.2821
Labor = 80.227974
Sites/Resource Input = 90.0
Kapital = 79.470795
Labor marginal prod = 3.6423507
Resources or Sites marginal prod = 0.92560893
Kapital marginal prod = 0.52361256
Labor average prod = 5.213669
Resources or Sites average prod = 4.6475787
Kapital average prod = 5.263344
Wages (Avg. return on labor) = 3.189519
Avg Return on Res or Sites = 0.7225274
Avg Return on Kapital = 0.41839874

res:
Amount Produced = 24.151676
Labor = 4.274698
Sites/Resource Input = 3.0
Kapital = 10.570367
Labor marginal prod = 2.1185687
Resources or Sites marginal prod = 2.558444
Kapital marginal prod = 0.55199915
Labor average prod = 5.6499143
Resources or Sites average prod = 8.050559
Kapital average prod = 2.2848475
Wages (Avg. return on labor) = 4.054452
Avg Return on Res or Sites = 6.1337905
Avg Return on Kapital = 0.33953738

serv:
Amount Produced = 63.75304
Labor = 10.817239
Sites/Resource Input = 1.0
Kapital = 17.518518
Labor marginal prod = 4.070528
Resources or Sites marginal prod = 0.0
Kapital marginal prod = 1.0706208
Labor average prod = 5.893652
Resources or Sites average prod = 0.0
Kapital average prod = 3.63918
Wages (Avg. return on labor) = 3.9556904
Avg Return on Res or Sites = 0.0
Avg Return on Kapital = 1.1905501

Secondary Sector - prod:
Amount Produced = 87.70229
Labor = 4.6800866
Sites/Resource Input = 7.5288453
Kapital = 33.610176
Labor marginal prod = 6.15031
Resources or Sites marginal prod = 5.062621
Kapital marginal prod = 0.51577294
Labor average prod = 18.73946
Resources or Sites average prod = 11.648837
Kapital average prod = 2.609397
Wages (Avg. return on labor) = 4.628359
Avg Return on Res or Sites = 9.621356
Avg Return on Kapital = 0.107452795

gold mining:
Amount Produced = 9.293913
Labor = 2.0
Sites/Resource Input = 1.0
Kapital = 3.638026
Labor marginal prod = 1.6364632
Resources or Sites marginal prod = 2.551742
Kapital marginal prod = 0.5817315
Labor average prod = 4.6469564
Resources or Sites average prod = 9.293913
Kapital average prod = 2.5546582
Wages (Avg. return on labor) = 1.6059709
Avg Return on Res or Sites = 5.0083904
Avg Return on Kapital = 0.30193213

************************************************** ****

Initial Prices:
Cost to buy one food = 0.9177498
Cost to buy 800 food = 800000.0
Cost to buy one res = 1.6118339
Cost to buy one prod = 0.9113822
Cost to buy one serv = 1.0
Cost to buy one gold = 1.0
Cost to buy 5 gold = 4.999999

Executing calculateInitialTaxBase (phase B)
Total tax base = 624.962

Before Start of Taxation Phase

Estimated cost of Government purchases: 39.338245

Executing collectTaxesAndMakeGovtPurchases() (phase C)

Orders Existing:
"food for people" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"consumer stuff" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"Military - Legion" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 40.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"prod kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 30.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = 0.0
"res for prod" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"gold mining kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"farm kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99
"res kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 30.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = 0.0
"serv kapital" amountToSpendLumpSum = 0.0
percentOfLocalTaxesToUse = 0.0, turnNeededBy = None
flexPercent = 20.0, requiredROI = -0.99

*** A shiny new Military - Legion has been built ***

Taxes available at 10% = 62.496197
Infrastructure:
farm kapital: Value = 79.470795, finalPrice = 1.049398
res kapital: Value = 10.570367, finalPrice = 1.171781
prod kapital: Value = 49.61049, finalPrice = 1.171781
serv kapital: Value = 17.518518, finalPrice = 1.1526942
consumer stuff: Value = 202.3009, finalPrice = 1.0800885
food for people: Value = 982.69073, finalPrice = 0.8465466
gold mining kapital: Value = 3.638026, finalPrice = 1.171781
res for prod: Value = 0.0, finalPrice = 1.0
Military - Legion: Value = 22.429804, finalPrice = 1.0


Will Eventually Execute Trade Phase here (phase D)

Before Consumption / Investment...
Market Snapshot:
Resources: Amount = 24.151676, finalPrice = 1.6118339
Services: Amount = 33.722015, finalPrice = 1.0
Gold: Amount = 9.293913, finalPrice = 1.0000229
Food: Amount = 418.2821, finalPrice = 0.8465466
Manufactured Goods: Amount = 53.99213, finalPrice = 1.1908678

(Final Prices = 1.0 since they haven't been set yet.)

Infrastructure:
farm kapital: Value = 79.470795, finalPrice = 1.049398
res kapital: Value = 10.570367, finalPrice = 1.171781
prod kapital: Value = 49.61049, finalPrice = 1.171781
serv kapital: Value = 17.518518, finalPrice = 1.1526942
consumer stuff: Value = 202.3009, finalPrice = 1.0800885
food for people: Value = 982.69073, finalPrice = 0.8465466
gold mining kapital: Value = 3.638026, finalPrice = 1.171781
res for prod: Value = 0.0, finalPrice = 1.0
Military - Legion: Value = 22.429804, finalPrice = 1.0


Executing setFinalPricesAndConvertSpecials (phase E)

Executing calculateConsumptionAndInvestment (phase F)

After Consumption...
Market Snapshot:
Resources: Amount = 0.0, finalPrice = 1.1855707
Services: Amount = 0.0, finalPrice = 1.0
Gold: Amount = 0.0, finalPrice = 0.99983215
Food: Amount = 0.0, finalPrice = 0.85128576
Manufactured Goods: Amount = 0.0, finalPrice = 0.9863353

Infrastructure:
farm kapital: Value = 77.3417, finalPrice = 0.9485534
res kapital: Value = 10.041848, finalPrice = 0.9877018
prod kapital: Value = 47.129963, finalPrice = 0.9877018
serv kapital: Value = 22.51623, finalPrice = 0.9890682
consumer stuff: Value = 328.8777, finalPrice = 0.97829616
food for people: Value = 1464.217, finalPrice = 0.85128576
gold mining kapital: Value = 3.5652657, finalPrice = 0.9877018
res for prod: Value = 24.151676, finalPrice = 1.0
Military - Legion: Value = 21.75691, finalPrice = 0.99180114


************************************************** ****

Final sector Stats:

farm:
Amount Produced = 418.2821
Labor = 80.227974
Sites/Resource Input = 90.0
Kapital = 77.3417
Labor marginal prod = 3.6423507
Resources or Sites marginal prod = 0.92560893
Kapital marginal prod = 0.52361256
Labor average prod = 5.213669
Resources or Sites average prod = 4.6475787
Kapital average prod = 5.263344
Wages (Avg. return on labor) = 3.1750362
Avg Return on Res or Sites = 0.7192465
Avg Return on Kapital = 0.4607817

res:
Amount Produced = 24.151676
Labor = 4.274698
Sites/Resource Input = 3.0
Kapital = 10.041848
Labor marginal prod = 2.1185687
Resources or Sites marginal prod = 2.558444
Kapital marginal prod = 0.55199915
Labor average prod = 5.6499143
Resources or Sites average prod = 8.050559
Kapital average prod = 2.2848475
Wages (Avg. return on labor) = 2.71389
Avg Return on Res or Sites = 4.6699243
Avg Return on Kapital = 0.2859588

serv:
Amount Produced = 63.75304
Labor = 11.21053
Sites/Resource Input = 1.0
Kapital = 22.51623
Labor marginal prod = 4.070528
Resources or Sites marginal prod = 0.0
Kapital marginal prod = 1.0706208
Labor average prod = 5.893652
Resources or Sites average prod = 0.0
Kapital average prod = 3.63918
Wages (Avg. return on labor) = 4.666326
Avg Return on Res or Sites = 0.0
Avg Return on Kapital = 0.75784266

Secondary Sector - prod:
Amount Produced = 87.70229
Labor = 5.2867956
Sites/Resource Input = 5.6144156
Kapital = 47.129963
Labor marginal prod = 6.15031
Resources or Sites marginal prod = 5.062621
Kapital marginal prod = 0.51577294
Labor average prod = 18.73946
Resources or Sites average prod = 11.648837
Kapital average prod = 2.609397
Wages (Avg. return on labor) = 5.3901296
Avg Return on Res or Sites = 4.9594393
Avg Return on Kapital = 0.062942475

gold mining:
Amount Produced = 9.293913
Labor = 2.0
Sites/Resource Input = 1.0
Kapital = 3.5652657
Labor marginal prod = 1.6364632
Resources or Sites marginal prod = 2.551742
Kapital marginal prod = 0.5817315
Labor average prod = 4.6469564
Resources or Sites average prod = 9.293913
Kapital average prod = 2.5546582
Wages (Avg. return on labor) = 1.5940037
Avg Return on Res or Sites = 4.97107
Avg Return on Kapital = 0.3115085

************************************************** ****
payoffRate for farm kapital invest. @10% taxes = 0.15922451

Max units for 0.2 payoff = -2.9146068

payoffRate for res kapital invest. @10% taxes = 0.17363179

Max units for 0.2 payoff = -0.29312423

payoffRate for serv kapital invest. @10% taxes = 1.2731531

Max units for 0.2 payoff = 33.569084

payoffRate for prod kapital invest. @10% taxes = 0.09660947

Max units for 0.2 payoff = -5.0199366

payoffRate for gold mining kapital invest. @10% taxes = 2.0037718

Max units for 0.2 payoff = 8.552042


End of turn

Estimated cost of Government purchases
using MSEconomy.calculateCostOfGovernmentPurchases(): 76.793884

Estimated cost of ALL Government purchases
using civ.executeGovtEconOrders(testOrReal, orderName): 195.88766

Estimated cost of 'Military - Legion' Government purchases
using civ.executeGovtEconOrders(testOrReal, orderName): 118.05056

[This message has been edited by Mark_Everson (edited February 09, 2001).]

[axi]

Wow! I'm in awe! I see alot of things here and all look pretty good.

Why do you have added gold mining among the productive sectors? It is because you've got sites producing money instead of resources, isn't it?

I am currently rearranging my infra spreadsheet. So far it looks really good, with judgment_by_Ideal_Value, direct production orders and subsidies already implemented, but I haven't yet touched the judgement-by-ROI method. My prediction is that it all boils down to a convex, bound-extremum, problem of non-linear programming. It would be very simple if the Percieved ROI was the same for every agent, but it isn't.

You will be hearing from me soon, I hope...

[Mark_Everson]

Hi Axi, glad you like what I've got so far. You're right that Gold produces essentially money. In the system it produces Services which are our version of money 'till we think of a better way to handle it! I put Gold in to have a Special working. We are now basically ready to put in specials like wine, tin, salt, etc. But I want to wait to test out the basic system before loading it down with too many new things.

I'm looking forward to seeing your new approach! If you can give a few fairly detailed examples before you drop a big spreadsheet on us, I'd appreciate it. That way I can see what assumptions etc. are being used without needing to go carefully through a bunch of parameters .

[Mark_Everson]

Hi Starfighter, welcome to the forums!

Actually its more complicated than what you've got, because we're talking about rates of return on investments. Thats the basis we were thinking of using for optimization. I guess you're probably familiar with it... so I won't blab on about it. I *think* his PIBs (and so yours) were supposed to implicitly include a discount rate giving the entire future value of the investment. But he was using a minimum 20% rate of return to justify an investment as an example. That's why his buy recommendations and yours don't match.

A straightforward approach like what you put down won't work, because there will always be many competing worthwhile investments. Sorting them by return on investment is one way to select the best among the alternatives.

[Starfighter08]

It's me again. This time macroeconomics, subject: inflation.

Antique, medieval times:

The material a coin was made of had exactly the worth that was "printed" on. FE: a 1C coin would be made from say an alloy of silver and a metal to make the coin more durable which is worth exactly 1C.

Now if a government was in dire need of money it could produce coins that didn't have the worth "printed" on it. Effect was of course that it "gained" the difference between the printed and the real worth, so the gov't had larger purchasing power. Now the big downside of the deal is that currency as a whole looses worth (more money circulates but there is still the same amount of goods being produced). All the prices climb and the citizens loose purchasing power (you can bet this would disturb people very much ->riots).

Modern time:

The material of the money is no longer worth very much (the paper of a 100$ bill is not even close to 100 bucks). Worth of the currency is ensured by the central bank (for example with its gold reserves) and also by a stable political situation of the country. To pull off the same thing as above the government needs control over the central bank and have it produce more money (for example the gov't runs a deficit and forces the central bank to buy the gov't bonds, this has the effect of increasing the amount of money in the economy: result same as above).

There is also a connection with unemployment. A gov't facing high unemployment can fund gov't programs to reduce it. Also short term effect of having more money is that interest rates drop (or vice versa, if the central bank decides to lower intrest rates more money will be available) which normally also helps the private sector. In bad times it's a trade off between inflation and unemployment in the hope that the economy recovers.

And further if an economy is growing (more goods and services produced) and the central bank doesn't allow the amout of money to grow respectively, economic growth will wentually stop and even turn negative (recession).

PS I you think to add these concepts you must also add different currencies for each civ (why should others bow to your monetary policies?). And I haven't even talked about the effects of inflation on foreign trad (don't worry I wont).

PPS: Although it gives the player a set of new possibilities and challenges I think that this will make the econ model quite complex.

[Richard Bruns]

Thanks for looking over the model. It seems that I didn't make an important point clear enough. I'll repeat it here:

The PIB of sewers would change based on the AMRIB and tech advances like germ theory and ecological awareness. Note that unlike PIOC, PIB is a continuous thing; benefits from infrastructure come in every turn. This monetarized value is an indication of how much people think they would benefit from the sewers. So early on, the PIB would be low, so your people would only invest in the simplest system, a pipe leading into the lake. The AMRIB is unchanged, however, so when tech advances, the PIB would increase as people become more aware of things. Now they would want to invest in more and better sewers, since the percieved benefit is now greater than the cost.

Every unit of water and sewer systems listed is a "good" investment. Even the fifth unit would pay for itself in five turns. But note that the return is not money; it is a representation of happiness and health.

[Starfighter08]

I haven't figured out how that quote works!

Quote (Richard Burns):

"....In the case of multiple needs, the one with the highest PIB to cost ratio would get built first. Suppose the above tile generated 100C in taxes and the models calculated the following PIBs and costs for various improvements: Sewers: 10 C a unit, PIB per unit: 25, 12, 6, 2
Water Supplies: 5 C a unit, PIB per unit: 20, 10, 5, 1

The province would build one sewer unit and 2 water units, even though the model says it is worth it to build all 4 sewer units and all 4 water units, since they would still return 20% of the investmen value each turn. So the province wants to spend 60C to invest in things that would return up to 20%
of the benefit......"

This is a classical cost-benefit decision. At optimum you have following situation: price = marginal benefit. Now let me explain that with your example.

Lets say you have those 100C to spend on Sewers or Water Supplies. Now you have to calculate the net benefit of both investments.

Severs: 1st unit: 15C (25-10), 2nd unit: 2C (12-10), 3rd unit: -4C (6-10)
Water S: 1st unit: 15C (20-5), 2nd unit: 5C (10-5), 3rd unit: 0C (5-5), 4th unit: -4C (1-5).

From the numbers above you can see that only the first two sever units yield positive returns, while the third one costs more than it is worth to you (law of decreasing returns). For the water system units the first three yield positive returns (or exactly 0) and for the forth one its negative.

Result:

2 sewer units and 3 water system units would be built by people who want an optimal coverage. Costs of the whole installation: 35C.

If you happen to have less than the 35C required to build all the units at once, the building queue would start from the highest net benefit ending with the lowest:

1st unit: Sewer or Water happen to have the same net benefit, 15C (choose one)
2nd unit: the one which hasn't been chosen as first, also 15C
3rd unit: Water, 5C
4th unit: Sewer, 2C
5th and last unit: Water, 0C

I hope I haven't lost the audience halfway. If you've ever wondered what a lecture in microeconomics looks like, you just got one. I hope you enjoyed it. (evil grin)

[Starfighter08]

Thanks for your answer. Because such infrastructure doesn't yield monetary gain I treated it under the rational consumer choice model. Now I understand that you see it as an investment.

Does that mean that the local gov't compares investments into the private sector (food, production etc.) and infrastructure (said sewer roads etc.)? If yes, how do you ensure that the city hasn't to wait for a sewer system until the whole economy of the province has been developed (because ROI is FE higher in the industry)? An answer would be appreciated.

BTW did I understand it correctly, that a gov't does compare real monetary gain (higher tax revenue) with monetarized value of say clean environment without further distinction (for example effects on happyness)? What about building a sewer (or other city growth increasing infrastructure) even if it doesn't yet pay back, to lure additional citizens into that city?

BTW2 I really enjoy this discussion and I hope you feel the same way, because I have wished a civ-type game with a market economy for quite some time.

BTW3 I haven't posted about inflation because I like to hear myself speaking, but because I had the impression that somebody asked if inflation is important.

[Mark_Everson]

On inflation... It does keep coming up, but I still think the right decision at least for now is to ignore it. Essentially inflation is a hidden tax. I just don't think we need to set up all the stuff needed to do inflation right when it can be roughly simulated as a straight tax. The only exception I can think of to the inflation-as-hidden-tax argument is like what happened in Spain (and then the rest of Europe) with the huge influx of new world silver. But in our current economic system such huge influxes of money will cause severe dislocation in the economy anyway... So although inflation has been important in history, I really think we don't need to model it explicitly. Y. M. M. V.

BTW, I handle money in a fairly odd way in the current model (money = services). If you are interested in making suggestions on how to do it better, I'd be interested.

Sorry I misunderstood what you are trying to say in your example Richard. I was led to believe your numbers were net present value, because investments in infrastructure never just keep returning forever. There's always a decay in the infrastructure itself that needs to be taken account.

Starfighter, on your point about the government being farsighted enough to understand that a beneficial environment will result in increased immigration... I'm just not sure how sophisticated we can make the AI. Ideally it would be good to have things like that, but I don't want to promise. But in fine-tuning the AI weekend certainly have the case of governments that are more inclined to invest in environmental quality, which will indirectly give the sort of action you are looking for.

[Starfighter08]

I just thought about using a "real" market mechanism (supply and demand) for each of the commodities plus for labor and capital. These markets would be on province level and people would pay and get paid in money. Markets would by definition always be at equilibrium, unless the gov't interferes with administrative prices (FE agriculture in the EU). Also the price differences of the different provincial markets could be used as basis for trade between these provinces. FE high prices for food would cause merchants to buy it in an other province and sell it where the price its higher. The effect is that where food is bought the price rises (local demand plus merchant purchases) and where it is sold the price falls (local supply plus mercahnt sales). This would happen until the price for food is the same throughout the civ (lets assume no foreign trade for the moment). To cross check this you aggregate the demand and supply curves of the provinces, then you get the equilibrium price throughout the civ, which happens to be the same as in all the provinces after trade has "happened".

The supply curve for such a market would be defined by aggregating the production functions of all the suppliers on a certain market.

The demand curve would be the aggregation of the demand curves of the single people (say "heads"), which are affected by preferences (changing with higher PCI as stated in the model) and (after tax) budgets of all the buyers. Being the consumers (or the gov't) for food, production and services and the "industry" for ressources.

The same would apply for the other markets too (services also being produced, labor and capital also being "traded"). Of course the different markets are interdependent. A change on one market affects the otheres too (like in real life).

Once all markets are in equilibrium, they would remain so (no need to do the whole process again) until a serious change happens. FE immigration or death because of plaques, new resources or farm sites discovered etc. After such an event it would take a few iterations to reach equilibrium again.

Although this is a complex mechanism, once you have it working it answers all the questions that can possibly come up. The systems would deal with taxes and tariffs, transport costs (merchants maximizing their profit automatically reduce overall transport costs to minimum economywide), migration, loss of production factors due to war, riots etc. This way it's a lot easier to simulate the interactions of the economy with society, gov't and technological progress thus reducing the need for workarounds and similar things. I dare say that it could even reduced overall need for processing power, making it playable on "normal" computers.

The only questions are:
1. If you can find people who can code all this (I'm sorry I've no experience in coding).
2. If it really doesn't need high end machines to work.

What do you think, is the implementation feasible?

PS If the actual economic model already looks like this,I apologize for bothering you. But if it isn't and you have questions or something isn't clear, please feel free to ask.
[This message has been edited by Starfighter08 (edited March 18, 2001).]
[This message has been edited by Starfighter08 (edited March 18, 2001).]

[Mark_Everson]

Hi Starfighter:

As I think you feared, and also hoped , the model is Already basically what you have outlined. So I think you need to do some more reading in our admittedly poorly-structured discussions... One thing you could do is just follow back the link at the top of this thread to the previous Econ discussion. That has a lot of this stuff in it. There are also a few threads covering the demo 5 Econ model. And also a whole area on merchants. The merchants and main Econ threads are already linked off the respective pages on the web site. For the demo 5 Econ model page it's here: http://apolyton.net/forums/Forum21/HTML/000226.html?69 .

Sorry for the inconvenience, but if I spent my time keeping absolutely everything up to date, I'd never make any progress coding. By the way, about half of what you are looking for is already coded, and you can play with it when demo 5 comes out, perhaps in a few weeks. If you are interested in more details on what demo 5 does, I did post a trace of economic activity higher in this thread. Just look for the really long post with lots of text output and relatively little normal prose.

[Starfighter08]

It seems that I can't add anything of any importance. Sorry for causing you so much trouble but from the economy model description I didn't see how much of the actual micro and macroeconomic theories I learn at university has been applied. I'm more than ever convinced that this game will rock. I'm looking forward to playing Demo 5.

Regards

Starfighter

[Lord God Jinnai]

quote: Originally posted by Starfighter08 on 03-18-2001 10:33 AM I just thought about using a "real" market mechanism (supply and demand) for each of the commodities plus for labor and capital. These markets would be on province level and people would pay and get paid in money. Markets would by definition always be at equilibrium, unless the gov't interferes with administrative prices (FE agriculture in the EU). Also the price differences of the different provincial markets could be used as basis for trade between these provinces. FE high prices for food would cause merchants to buy it in an other province and sell it where the price its higher. The effect is that where food is bought the price rises (local demand plus merchant purchases) and where it is sold the price falls (local supply plus mercahnt sales). This would happen until the price for food is the same throughout the civ (lets assume no foreign trade for the moment). To cross check this you aggregate the demand and supply curves of the provinces, then you get the equilibrium price throughout the civ, which happens to be the same as in all the provinces after trade has "happened". The supply curve for such a market would be defined by aggregating the production functions of all the suppliers on a certain market. The demand curve would be the aggregation of the demand curves of the single people (say "heads"), which are affected by preferences (changing with higher PCI as stated in the model) and (after tax) budgets of all the buyers. Being the consumers (or the gov't) for food, production and services and the "industry" for ressources. The same would apply for the other markets too (services also being produced, labor and capital also being "traded"). Of course the different markets are interdependent. A change on one market affects the otheres too (like in real life). Once all markets are in equilibrium, they would remain so (no need to do the whole process again) until a serious change happens. FE immigration or death because of plaques, new resources or farm sites discovered etc. After such an event it would take a few iterations to reach equilibrium again. Although this is a complex mechanism, once you have it working it answers all the questions that can possibly come up. The systems would deal with taxes and tariffs, transport costs (merchants maximizing their profit automatically reduce overall transport costs to minimum economywide), migration, loss of production factors due to war, riots etc. This way it's a lot easier to simulate the interactions of the economy with society, gov't and technological progress thus reducing the need for workarounds and similar things. I dare say that it could even reduced overall need for processing power, making it playable on "normal" computers.

Your statement is lacking in 1 area:

It ignoes the cost of travel expenses for things. It is always cheaper for a costal area to buy fish than an inland area because the local area need only pay for local travel expenses (boat to shore), but inland areas must pay for both local and inland shipment (by truck FE). This means that no matter how much is shipped, it will always cost more when shipped inland because a local area (with few exceptions) will get dibs on its produced resource (in this case fish) before it exports any. Thus what is exported is dependant upon what is left and is therefor finate.

[Richard Bruns]

quote: Originally posted by Starfighter08 on 03-18-2001 07:32 AM Does that mean that the local gov't compares investments into the private sector (food, production etc.) and infrastructure (said sewer roads etc.)? If yes, how do you ensure that the city hasn't to wait for a sewer system until the whole economy of the province has been developed (because ROI is FE higher in the industry)?

I'm not the expert, but I do believe that the AI compares all investments to see which one is most valuable. But unlike most businesses, it also considers externalities (PIOC and PIB) in that return on investment. So while a sewer provides zero monetary return on investment, it provides a huge return in health and happiness. In the example above, a 10 CC one-time investment in sewers resulted in a 25 CC benefit every turn. That is probably the best return on investment that the AI could find anywhere, so that simple sewer system would probably be built very quickly.

quote: BTW did I understand it correctly, that a gov't does compare real monetary gain (higher tax revenue) with monetarized value of say clean environment without further distinction (for example effects on happyness)? What about building a sewer (or other city growth increasing infrastructure) even if it doesn't yet pay back, to lure additional citizens into that city?

The PIB is assumed to include effects like that. Such immigration effects could be included in the formula that we use to turn happiness values into money benefits. All we need is something simple like "5 CC per unit of happiness" which is determined by playtesting; we play and look at the effects of hapiness, then assign values to it that include all positive effects.