Afterword
In this following section I explore the math behind some of the numbers defined in the rules as well as the thought that went into them. I hope this section provides some further insight and food for thought.
The most necessary item in such systems, given we are not playing a semi-automated game, is simplicity. This will be the overarching theme and force a lot of the later decisions.
The economic system should reflect these issues:
- Separate Heavy and Consumer Industry Trackers.
- Trade (Overland and Marine)
- Rising and Stagnating Economies
- Rebalanced ROI on heavily industrialized countries.
- War Economy / Deficit Spending
- Industrial Decay
The military/technological system should reflect these issues:
- Maintenance Costs
- Air/Land/Naval force production
- Military Combat System
- Research System
Well... why are these issues so important?
1. Separate Heavy and Consumer Industry Trackers
The Wesworld system doesn't account for anything occurring in an economy besides steel production. Given how importance the part of the industry that isn't involved in war-making potential, this seems quite necessary.
2. Trade (Overland and Marine)
Wesworld doesn't have any rules at all regarding trade. Whilst that simplicity is nice, it also results in very little diplomacy. In Wesworld you get old ships sold and maybe a foreign factory investment once in a blue moon. Not to mention there's no incentive to having such a foreign factory investment. In addition, the lack of trade implies nations are self-sufficient and does not discourage war. Whilst this is not an issue in Wesworld (because of the cumbersome scripting system), it would be if the military system was revamped to be more feasible.
3. Rising and Stagnating Economies
In the world, industrialization of undeveloped nations is an extremely powerful force. It exerts itself on diplomacy and war-planning. A real life example would be China. It's industrialization started to break the status quo and thus forced various re-alignments of the Pacific nations. In a sense, because this is not present in Wesworld, we see the status quo persist very long.
4. Rebalanced ROI on heavily industrialized Countries
Having done some math on the ROcI of various nations in Wesworld, it has come to my attention that not only do undeveloped nations not have any edge in industrialization, they actually have it much more difficult. The Wesworld rules state:
New factories may be built during the simulation. They cost 10 pts to produce. A factory that is half-complete may begin limited production, at the rate of 500 tons per quarter, in the quarter following the investment of 5 points; A factory is fully complete following the investment of 10 points and begins production at a rate of 1,000 tons per quarter following completion. ... In each quarter each factory produces either 1,000 tons of warship material (for warship construction) or 0.1 pts of industrial material (for infrastructure upgrades/construction).
I'll separate the ROCI into 3 cases. I'll first find the time needed to break even and then the amortized ROcI per quarter over a period of 50 years (200 quarters).
1) 1 point per quarter is invested into the factory until half-completion status, then stopped.
It takes 50 quarters to reach half-complete status. It produces 0.05 pts per quarter afterwards so it takes 100 additional quarters to break even. A total of
150 quarters.
The ROCI per quarter (50 years) is (2.5 + 5)/200 = 3.75%
2) The factory is invested in at a rate of 0.1 point per quarter.
It takes 100 quarters to complete but produces 0.05 pt from the 51th quarter to the 100th quarter. At the 100th quarter, 2.5 pts have been produced. It takes an additional 75 quarters to break even. A total of
175 quarters.
The ROCI per quarter (50 years) is (10 + 2.5)/200 = 6.25%
3) The factory is completed immediately.
It takes 0 quarters to complete but only starts producing after the first quarter. It takes 100 quarters to break even. A total of
101 quarters.
The ROCI per quarter (50 years) is (9 + 10)/200 = 9.5%
Practically speaking, if Wesworld USA (30 factories) invested 33% of it's factories towards new factories, it would produce a factory per 10 quarter. The amortized growth per year over 50 quarters is 1% per year. The ROCI for each factory is 8.5%
If we took Romania (6 factories) invested 33% of it's factories, it would produce 1 new factory per 50 quarters. Ignoring the additional factory's production the amortized growth per year over 50 quarters is 1% per year. However, the ROCI for each factory in this case is 2.5%.
This means in naive terms, small and major industrial powers grow at much the same rate given the same investment, but the ROCI is close to 3 times worst for small industrial nations. That is, it is almost 3 times less efficient to build factories in low industrialization nations as compared to high industrialization nations. The former attribute (nominal growth) is true to reality (capital begets more capital), but the latter attribute (ROCI) is undesirable.
Given that growth of companies shrink as their ROE and competition increases (both which are simulated by high industrialization), this is the complete opposite of reality. It is also a death sentence to any nations that wishes to grow. Not only are major nations growing slightly faster in percentage (a lot faster nominally), they are also making a much worst investment.
In the long run (over 200 quarters or 50 years) this gap starts to even out with the US at ~9.5% and Romania at ~8.1% but in the long run we're all dead. This is of course ignoring the absurdly high cost of per factory currently in Wesworld.
5. War Economy / Deficit Spending
In general what we consider "war economy" can be interpreted as deficit spending. Sectors of industry expand only if there is a better profit expanding that sector versus another sector. Whilst industry can never be wholly efficient due to structural costs and other stickiness, it is quite efficient.
War economy doesn't produce "new industry" but rather converts existing industrial sectors to one that the government chooses. As we assume the industry was at a nearly efficient state, such a change means efficiency is lost. There is not enough profit in the sector to sustain the businesses moved. As a result those businesses lose money. To counteract this governments artificially prop the demand (at the target price). Essentially this results in governments paying businesses to remain inefficient. Money doesn't grow on trees and as expected governments do this only be deficit spending.
Wesworld's base system simulated this deficit spending to some extent as the decay of industry after putting the country on war footing. This sort of feature is desirable for various reasons besides it's presence in Wesworld.
6. Industrial Decay
When factories are idle, that indicates excess capacity. Excess capacity is also capacity that is losing money, since they aren't doing anything useful. In reality such factories are closed up and the capacity disappears. It seems to be a lot of hassle to simulate this given our usual opinion is that it's not very important. However, I'd like to share an anecdote to display how fearsome this really is.
I stumbled across this whilst trying to find the pig-iron and steel production of China in 1870 for the joint proposal with Snip. The province of Shanxi in China produced about 160,000 tons of Pig-Iron in 1872. In 1898, the production of pig-iron by Shanxi had dropped to 50,000 tons. These estimated were made by a German and some other foreign nationals estimated the production capacity and confirmed by the provincial governor. In 26 years, the province's production declined by 68.75%. Artisans who had worked on pig-iron had turned to other fields.
The reason was that pig-iron imported from other nations through Guangdong was cheaper. This was not because foreign pig-iron was produced significantly better, but because the poor state of infrastructure in China meant the transportation cost of exporting pig-iron from Shanxi to other provinces was too high relative to imports. One has to remember Shanxi is a mountainous landlocked province. In short, the high cost of transportation turned Shanxi's capacity into excess capacity which decayed as a result.
A more stark example would be that of Hanyang Iron Works in China. In 1910 it produced about 130,000 tonnes of steel. In 1912, it produced 93,336 tonnes of steel. This marks a decline of 28.2% over two years. In this case, civil unrest caused the decline.
------------------------------------------------------------------------------------------------------------------------
1. Maintenance Costs
It was proposed that the lack of maintenance costs in Wesworld lead to the size of Wesworld fleets. Certainly the shear idea that nations in Wesworlds could maintain multiple super-battleships, provided they first built them, seems a bit preposterous. After all, in reality the operating cost of such ships is what lead to the massive retirements in the USN after World War 2.
At the same time, we don't want something like Navalism's maintenance system. As Kasier Kirk put it, "I found swapping between reserve/active/mobilized, and tracking training rates and ammo supplies....burdensome".
Maintenance is a necessary evil to keep forces in check, but not a fun one. We want to simplify this as much as possible.
2. Air/Land/Naval force production
Complaints were made regarding the lack of land/air force simulation. This has been credited as the reason for some absurdity in the land and aeronautical fields. Hence, we want to expand these sections. The naval system is for the most part fine and should remain mostly intact.
Needless to say, these items are important because the military is important.
3. Military Combat System
Wesworld doesn't really face this issue because most wars are scripted, but we had a ton of experience with this in Navalism. Since the scripted war system is seen as undesirable, we clearly need an unscripted combat system.
As experience in Navalism showed, the ad-hoc method is extremely time-consuming and unsatisfactory to the participants. In most of the wars the losses seemed a bit arbitrary, the land/air forces were mostly ignored. In the Chinese Civil War, individual designs were reduced to numbers of X type ships and zones. For a forum devoted to ship design (which Wesworld also is), this meant your ship design essentially didn't matter. Even then combat took so long (about the same as the war on the Korean peninsular here) that essentially I became too busy to continue.
To streamline this, we need some sort of semi-automated military system without resorting to actually playing a game.
4. Research System
Wesworld's lack of a tech system led to many arguments (or so I'm told). I don't remember many arguments over the contents of the tech tree in Navalism, but there definitely was a feeling of unfairness. In Navalism, the number of items you could research at any one time was directly limited by the amount of BP (or tonnes of steel) you produced every half-year.
Hence large nations (industrially) had an advantage over smaller nations. Given that there was no catchup mechanization in Navalism (one was introduced very late in the sim but was still subpar), this forced many nations into research trades. But this wasn't really reflective of reality.
The size of a nation's research program and it's industrial output were not in a causal relationship. They were simply correlated. In reality, the primary force behind research projects and their relative "frontier-ness" is money. Wealthier nations have larger research programs, not necessarily nations have more heavy industry.
Another issue with a concrete tech system is most are tied with dates. The dates themselves are not indicative of how impossible that idea was to realize. For examples most of the things developed in WW2 were not strictly speaking impossible for the era of 1930 to develop (save for certain things that metallurgical and various other things needed to catch up on). It was a matter of the size of investment. If WW2 did not happen, it is likely the technological developments would have remained out of the private sector's investment capabilities out in to the 1950s or 1960s. But it is certainly possible to have developed them earlier given enough interest.
But if we remove years we encounter the problem of hindsight-itis. Even if not done on purpose, it's very likely a technology race will lead there within the sim. One way to prevent this would be to create a highly interconnect system of requirements to reflect realities like poor metal quality. However, that's talk for another section.
------------------------------------------------------------------------------------------------------------------------
Tackling the issues abstractly
All in all we had 10 primary issues that the system should solve. They are, for reminder, the following:
- Separate Heavy and Consumer Industry Trackers.
- Trade (Overland and Marine)
- Rising and Stagnating Economies
- Reduced ROI on heavily industrialized countries.
- War Economy / Deficit Spending
- Industrial Decay
- Maintenance Costs
- Air/Land/Naval force production
- Military Combat System
- Research System
To create a simple and easy to understand system, we first want to find ways to compress this.
I am going to propose an economic system consisting of two currencies and one semi-static value: Heavy, Consumer Industry, and Population.
These will be denoted by the symbols H, C, and P respectively.
- Heavy industry will produce the items needed to construct a vehicle, an aeroplane, and a naval vessel.
- Consumer industry will produce the goods and services not covered by heavy industry. Everything from washing machines to dining establishments.
- Population is what it says on the tin. This is the semi-static value.
Heavy industry and Consumer industry will be related to each side such that neither is more important than the other. This is key because we don't want the other currency to simply be an add-on, otherwise there would be no point to it. An unbalanced relationship between two currencies implies that one of them could serve both. Hence, balancing them ensures that we cover more base (since we noted one currency doesn't simulate enough of reality).
Heavy industry should build both industries, but Consumer industry should decay both industries. Hence in this fashion, both are equally important.
Now why would heavy industry build both? Well, machine parts would fall under heavy industry and such parts are vital to factory construction.
What about consumer industry? Here I inject a little retroactive thinking. We want a way to simulate corruption, civil unrest, different government types, etc. I propose that:
- Consumer industry is responsible for keeping the population happy (food and circus). Hence, if consumer industry is not quite up to snuff for the nation, it should suffer from negative effects. This is industrial decay (this affects both industries). What is "up to snuff" is determined by the amount of Consumer industry versus the size of the population.
Now Heavy and Consumer industry are not exchangeable, except in a special case - war economy. In the previous section, I talked about how war economy is the rearrangement of existing industry and that it is deficit spending because it's inefficient. Hence I propose this:
- When war economy is declared, A of the current Consumer industry is converted into Heavy industry. After war economy ends, the industry is converted back into Consumer industry.
Where A is some value less than 100%. This seems simple, but it actually fulfils a number of functions. First when in war economy, war-making potential increases. Second, the increase is not completely efficient. Third, this inefficient increase comes at the price of deficit spending. Remember how I mentioned consumer industry is responsible for keeping the population happy and industry decays if it doesn't? In war economy, you subtract 50% of your consumer industry! For most if not all nations, this should drop consumer industry below satisfaction point and cause industrial decay! Hence this simple rule accomplishes 5 and 6 of our primary issues.
Another thing to notice is that beyond the point where consumer industry satisfies population demand, consumer industry is worthless! We also notice by definition, consumer industry should be low in undeveloped nations. Hence we establish something we want to trade from developed to undeveloped nations. Now we also need to make a mechanic to prevent a global excess of consumer industry to prevent ridiculous war economies, but we'll address that later.
To recap, the outlined rules cover the primary issues 1, 2, semi-3, 5, and 6. Great! On the economic system of things we only have primary issues 4 left and 3 to wrap-up.
The goal with issue 4 is to have nations with the same investment ratio growing at relatively the same speeds regardless of the level of industrialization, but the ROCI be much much higher on undeveloped nations as compared to ROCI in developed nations. To this we have to keep in mind: What defines undeveloped versus developed? What affects ROCI?
To those two questions I answer:
The "developed-ness" of a nation is proportional to the C/P ratio (Consumer industry per Population).
ROCI is affected by output per year and the cost to build. For simplicity, I won't stray from the 1 Heavy = 1,000 tonnes metric.
As such I propose:
or more generally,
Where k is the additional cost per new industry and C/P serves as a modifier of this cost based on level of industrialization.
It does not require knowledge of the previous sim report save for the amount of H, C, and P in the previous year. This reduces the amount of error that could be introduced relative to some other schemes I had inside my head. It also only requires basic arithmetic, making the bean-counting quite easy.
In addition, this provides an incentive to not build too much consumer industry. It makes Heavy industry very costly!
But what if a player decides to only build Consumer industry, ignoring Heavy industry? Wouldn't it would be unaffected by this rule?
Well, such a player would have nothing in the way heavy production normally. They would be unable to build a military.
In addition, don't forget HEAVY industry is responsible for the building of both industries. Naturally the growth rate of such a player nation would start to slow.
If we also add a rule increasing the cost of consumer industry as you build much (abet at a lesser rate), then it strongly discourages such behaviour.
Where q < k.
At the same time this solves the problem of issue 3. Nations that are heavily industrialized cost a lot more to build in their own nation than to invest in a foreign nation, hence a cause for trade.
As an addendum, population isn't counted as a currency since it's mostly static and cannot be traded/used.
Population should grow at 0.25% per quarter, and this is derived from "Population Growth and Agrarian Change: An Historical Perspective".
------------------------------------------------------------------------------------------------------------------------
All the economic issues are now semi-solved. Let's go to the military issues. As a reminder they are:
- 7. Maintenance Costs
- 8. Air/Land/Naval force production
- 9. Military Combat System
- 10. Research System
First let's tackle issue 8. To reflect the fact that you can't simply decide to produce all tanks one quarter and then all ships another, we'll limit production by infrastructure. I'll explain what I mean by this:
Ships production in Wesworld is limited by number of slips. This requires a bit of reworking, since "slips are dual use, and shouldn't be fancy or hard to assemble." - Kaiser Kirk. Ignoring that for the moment, however, I suggest we do something similar for air and land forces.
Land forces are split into mechanical (artillery, vehicles, rifles) and organic (the person who is a soldier). Keeping track of both, however, would be excess complexity. The maximum production capacity of land forces should be limited by the number of such land factories or Barracks. (Please suggest a better name)
Air forces should be limited by the number and size of air fields. I know air planes are not produced in air fields, but it represents the operable capacity of airplanes. Of course, like slips, it doesn't determine how many airplanes you can keep inoperable. For the simplicity of having only 1 infrastructure to keep track of versus two, I merged the two roles.
But this system still leads to the very large fleet size (and now air/land force). Hence maintenance costs are crucial.
However because maintenance costs imply a necessity to bookkeep all your forces, we don't want to have more than two possible valuations for it. Ideally, we want something that only has 1 valuation, since we can just aggregate the whole force before computing the maintenance cost. However, for certain forces we can't do that. We want there to be differences between the cost of a carrier, submarine, destroyer, and another surface combatant. We want tanks and infantrymen to be counted differently.
Hence I propose this set of rules:
- Naval_maintenance (Quarterly) = Tonnage * Z * 1%
- Naval_cost = Tonnage * W
Where Z = 1.5 for submarines/destroyers, Z = 3 for carriers and W = 2 for submarines. Those not mentioned have default values, Z = 1, W = 1.
If the numbers seem a bit arbitrary, it's because they are. Mostly recycled from familiar numbers in Navalism.
- Air_maintenance_active (Quarterly) = Tonnage * 5%
or
- Air_maintenance_reserve (Quarterly) = Tonnage * 2.5%
- Air_cost = Tonnage * 3
The numbers are derived from some figures of the P51 Mustang. The Mustang cost ~$50,985 to make in WW2, burns ~$14/hr or ~$42/hr including maintenance. The max weight of a Mustang was ~7 tonnes. The expected operational time was ~30 hours. The numbers are all in WW2 dollars. Hence, a squadron (the unit to operate on) of 25 planes, a maximum tonnage of 175t costs $1.27 million WW2 USD. By contrast, the Iowa battleship costs $120 million USD and displaces ~45,000 tonnes Standard. I neglected to count for the cost of the pilot and maintainers, so I'll just simulate it by using the same pay for the pilot (~$420) as a US army grunt in national service. The total maintenance cost for a squadron is 23% of the unit cost and the USD/tonnes ratio is 7. For the Iowa BB, the USD/tonnes ratio is ~2.6. Hence the cost is about 3 times more than that of a BB. Such information is sprinkled around on the internet and isn't very difficult to find so I'll retrain from explicit sources for this.
- Land_maintenance_active (Quarterly) = Tonnage * 4% + No. of Divisions * 0.1
or
- Land_maintenance_reserve (Quarterly) = Tonnage * 2% + No. of Divisions * 0.05
- Land_cost = Tonnage * 50 + No. of Divisions * 0.5
The numbers are derived from figures on the M4A3(76)wet HVSS/T80 Sherman. The cost of one was ~$54,836 and the weight was 33.7t. I can't find a life-estimate for the Sherman, so I'm using the life-estimate of the Panther's engine instead. The engine not being the source of the Panther's low reliability, I'd say it approximates the Sherman. The Panther engine had an average expectancy of 1,000 km. This particular Sherman type has a cruising range of ~161 km on road and a fuel capacity of 168 gal. In 1945, the price of fuel was ~$0.21/gal. Hence the operating cost total, if the engine reached it's limit, would be $219. This is roughly 4% of the base cost. Given the USD/tonnes ratio of 162 for the Sherman and the ~2.6 ratio for the Iowa BB, the cost per tonnage should be ~62 more. Since 62 is not a pretty number, I'll arbitrarily lower it to 50.
[Price of Sherman -
http://web.inter.nl.net/users/spoelstra/g104/cost.htm][Weight, Range, and Fuel Capacity -
http://afvdb.50megs.com/usa/m4sherman.html]
The average pay of a US grunt in foreign service (in civilian terms) was $3,600/yr (~$420 for within US in military terms, ~$540 in foreign). The cost to train the US grunt would be the cost of equipment (~$170) plus the pay per year ($3,600) multiplied by the time they spent in training. Before the rapid expansion of the US army, the training period was ~7 weeks. There are 52 weeks in a year so, roughly 13.5% of the year. Hence the cost of 1 infantryman (with no ammo, etc.) would be ~$650. We can bump this value up a bit to ~$700 to account for ammo, etc. The smallest unit of operation would be the division, which I shall define as 10,000 front-line troops. Hence the cost of 1 division is $7 million USD. The maintenance cost is $5.4 million USD, or 77% the base cost. I shall tie this cost to produce a division to about a fifth the USD/tonne ratio of the Iowa BB, simply because it's the lowest of the ratios. This implies a division costs 0.54 H. I rounded this down for a nicer number.
[US Infantry Training Time -
http://www.ww2gyrene.org/boot_camp.htm][US Infantry Annual Pay -
http://www.usmm.org/barrons.html][Bad Source, US Infantry Equipment Cost -
http://www.armoryblog.com/firearms/cost-equipping-soldier]
I separated the maintenance costs here between active and reserve
This seems to cover issue 8, so let's move to issue
10. Research System Issue 9 is quite complex and not something I want to enter into at the moment.
Ideally, we want research to be something not directly tied to Heavy industry, to go at a higher speed given more experience researching or doing stuff in the field. Hence, research speed is multiplied by theory and practical experience.
However that would require keeping track of two numbers per research fields. Certainly not simple on the bookkeeping so the idea needs to be modified a little.
Remember when I said the currencies needed to balanced? Well currently it's not balanced since Heavy is used to produce things, but Consumer is not. I didn't mind at that point because I had planned Consumer to produce Tech Points, which would make the currencies balanced.
Each point of Consumer industry produces 0.5 Research Point. If someone has already finished researching the technology, the cost to research it decreases 4% of the base cost per quarter until it hits 20% cost. It'll hit this minimum cost 5 years (20 quarters) after it was first researched. Note that it is the amount needed to complete that drops down, and any already invested point is unchanged.
For example, say 1936 artillery costs 500 research points and has already been researched by someone. You start putting in 50 research points per quarter before any decay occurs.
Quarter 1: 50/480
Quarter 2: 100/460
Quarter 3: 150/440
Quarter 4: 200/420
Quarter 5: 250/400
Quarter 6: 300/380
Quarter 7: 350/360 Stop Investing
Quarter 8: 350/340 Completed (You don't get excess points back!)
The person who first researched it would have had to go through 10 quarters to get the same tech at a rate of 50 points per quarter.
Take the same situation but in the year 1941 (5 years or 20 quarters after it was researched). It now costs 100 points and at a rate of 50 points per quarter, you'll complete it in 2 quarters.
To include specialization of research, we'll add bonus research. Every time you use research points in a particular field, you gain 10% of that research point back as fixed research points. These fixed points can only be used in the field they were generated in. They also decay every quarter at a rate of 10%. Once one of these fixed points decays to 0.1, they disappear.