Make population a limiting factor (AKA Dealing with unit spam)

[smartcanuck1988]

I would like to propose the following:

1. Make population growth slightly faster early on.
2. Every building requires 1 Citizen to make it active and get its bonus.
3. Only worked tiles/buildings provide yields, the population itself does not.
4. When you build or purchase a Unit in a city, one Citizen is removed from that city.
5. You cannot reduce a city to less than 1 pop. So a city with 1 pop cannot build or purchase units.
6. Disbanding a unit at a city (or at a tile the city controls) will increase that city’s pop by 1. It will not provide gold or anything else.
7. Settlers remove 2 citizens from a city (and new cities start with 2 pop). This will allow the new city to build at least one unit right away.

Extra suggestion: Settlers can upgrade to Colonists around mid-game. They remove ~5 citizens and start a new city with that many instead of just 2. This can help speed things up.

​

Reasons:

This probably comes from my interest in board games where there is a finite number of physical meeples/cubes to use -which limits the clutter, but makes each unit that much more important. In Civ6, there is a crazy number of all sorts of units because you can always continue to produce them. This suggestion can help with that mess. I am not suggesting that there is a cap on population, but that if you want to mass produce military units, then your cities will suffer.

I think a “Unit” is definitely more important in V and VI compared to stacks of IV, but it seems that the consequences of building one have not increased to match that. I do genuinely believe that when as much of the game elements is accounted for, the game mechanics will fit better together and the gameplay overall will feel more grounded. I think it will make it more meaningful to choose between building a huge war machine or emphasize growing your cities and working science/culture tiles. Similarly, choosing between military vs religious units will also require some thought.


Some negatives to start the debate:

1. If you lose a lot of your units in wars, you may reach a point where you cannot make more units to defend yourself. Balancing the rate of growth would be important.

2. Disbanding units could be a backdoor to shifting population around in a gamey way. Not sure how effective it will be (because you’ll be spending hammers to build the unit), but again it requires balancing this against cost/effect of pro-growth decisions.

3. Population growth formula might get a bit confusing given that the number of citizens will be fluctuating. But a clear UI that always tells the user how things are being calculated should make it clearer.
​

Thanks for reading all that. And I'm happy to hear your feedback on this and ideas in general about how we can make the population more important!

 

[Ryika]

I've played around with those systems, it's just not fun.

You either end up permanently reducing your cities to stay small towns, or you have to increase growth so much that cities that don't construct units become extremely big extremely fast.
It also encourages building units in smaller cities, and keeping cities at low pop in general, because those are the citizens that you get back reasonably fast after spending them.

Overall being to dependent on using up population is also just a terrible system in my opinion, I like the idea, but it doesn't seem to work too well. Or maybe my implementation was bad, but I don't think that was the main issue.
It's also really weird that you propose this as a solution to unit spam, but it does so many other things for really... no reason at all. Unit Spam can easily avoided by other measures:

- Increase Unit Costs by total Units or Units of the same type already on the map
- Increase Unit Base Cost
- Implement a Supply limit
- Increase Upkeep

etc.

 

[smartcanuck1988]

Thanks for the reply! That's great to hear that it is possible to test this out in game. I wish I knew how to mod so I can play around with the ideas.

I agree with you that it's not the simplest fix for unit spam. But I do eventually like to see (or at least try it somehow) the game mechanics be more integrated somehow. Right now you can have large armies from small cities because it's a function of hammers not food. And while you are assembling this large army, the city is growing just fine.

So I think game decisions would become more important if you know you are taking a risk on future growth and prosperity if you focus on a huge military then end up losing many units. It should take time to recover from that. As you say however, the important thing is to balance normal growth of cities. One thing to note is that growth rate shouldn't change after producing a unit because that unit still counts as a "citizen" of that city.

 

[Dicemechanic]

As I said in a lengthy rant elsewhere, I'd prefer to have units cost food - so there is an indirect impact on population, which will only be significant if you have a really large military

 

[smartcanuck1988]

As I said in a lengthy rant elsewhere, I'd prefer to have units cost food - so there is an indirect impact on population, which will only be significant if you have a really large military

I think this sounds good. Essentially anything that makes food/growth the foundation of your civ's "raw workers" capacity.

Whether it's a direct conversion of 1 worker to 1 unit as I suggest, or whether it is indirect by requiring units to consume food as you say. Either way, the point is that linking it to food/growth will make choices you make in one system impact the other.

 

[Boris Gudenuf]

Historically, there were two limits to 'size of army'. In the Ancient to Renaissance/Enlightenment Eras, the major limitation was Cost. This was a combination of two things: first the cost to equip, train and maintain troops that, unless you were pillaging and looting, weren't contributing anything to your economy, which was still in a large part subsistence - not a lot of excess workers to syphon off into the military in the first place, and second, the very inefficient methods of collecting taxes to pay for all that military. The Roman Empire with a population exceeding 50,000,000 went essentially bankrupt trying to maintain an army of less than 500,000 - about 1/10 of what states managed in the middle of the 20th century.
Caveat: some societies were less affected: nomads like the Scythians, Huns, Mongols, Alans, etc. in which every adult male (and some female) was a mounted archer or javelin/spearman in their ordinary lives as herders, could raise impressive armies without taxing their 'economy' as much as labor-intensive, agricultural-based societies.

From the Industrial Era to the Atomic Era the chief limitation is Population: Industrial banking/stock exchanges and institutionalized bureaucratic methods made it relatively easy to raise (or borrow) immense sums of money to pay for the military, but the militaries grew so large that they sucked the majority of workers right out of the economy: in World War Two, everybody had to recruit women and other 'non working' parts of the population into industry, because most of the men between the ages of 18 and 48 or older were in uniform.

In the Information Era, the limitation is a combination of the two: on the one hand, weapons have become ridiculously expensive if they are 'state of the art'. On the other hand, to use sophisticated, expensive new weapons requires troops with a much higher level of education and technical skill than ever before - the number of such people willing to forego Big Bucks in the civilian world for their skills is not large, and Conscription produces costs that add to the cost of the weaponry. You can still field large cheap armies of men on foot armed with automatic weapons and shoulder-fired rocket launchers, but against an army with really modern weapons, they suffice for a guerrilla war only, or they get slaughtered in large numbers (see Iraq in 2001, even when trying to defend their country against a mechanized invasion).

So, the Anti-Spam needs two components: Cost of Units and their Maintenance, and Population Limitations.

For the second, I think Civ V actually had a good mechanic: if your number of units of all types, including workers/Builders and Settlers, exceeded your 'civilian' (city) population, you lost 10% of ALL of your production for each unit exceeding the population limit. This places an immediate and increasingly drastic limitation on Number of Units, especially in the early game when your total Civilization city population may be in single digits for a while.

For the first, ALL military units should have a Maintenance Cost, including the currently free Warriors. This would represent both the reduction in workers to the state and the cost of feeding the troops while they are not economically employed, as well as the cost of their equipment, such as it is.

There are numerous historical Exceptions or Special Cases for this. In the 17th and 18th centuries, for instance, Prussian troops only drilled in uniform for about half the day, for half the year. The rest of the time they were encouraged to take jobs in the towns and villages where they were billeted, so the 'military' population remained (in peacetime) a large component of the Work Force. This could be a Military Card in the Renaissance Era removing the Population Cap during Peacetime for military units. Russia and China both used large numbers of troops as workers on State construction projects, so there might be another Military or Production/Economic Card allowing, say, a military unit stacked with a Builder to add one Charge to the Builder (the Cheap But Unskilled Labor card!)

For the 'barbarian' or Nomad units, Saka Horse Archers and perhaps Scythian (Light) Horsemen, halve the Maintenance Cost relative to similar units in 'regular' armies to reflect the different dynamics of their society/economy. I would balance this by removing the Scythian ability to build two mounted units at once.

These are just 'top of my head' suggestions, but I find that using the historical basis for the problem/solution generally works better than trying to make up a solution for a problem that the game produced in the first place.

 

[Dicemechanic]

Great suggestions. I've just finished reading Caesar's Commentarii de Bello Galli, and problems with supply of food (usually corn) is mentioned on almost every page, hence my suggestion of tying into food. So totally agree about limiting factor being "cost", but cost means more than just Gold, particularly in relatively simple economies where most wealth is measured in terms of land, because land produces food (and other useful resources of course). In early civs, this also represents the cumulative loss of labour to work the fields from fielding large armies (more of a problem for the Gauls than Caesar).

Similarly, if you want to represent loss of production, have a production penalty. Abstracting it all to gold places all the importance on that single element, whereas there's a good argument that the cost should be shared (personally, I'd keep it to Gold and Food, largely because it's simpler and because production maintenance costs isn't what Civ does - we don't have to pay production to maintain districts for example, or upgrade roads).

Your comment about Exceptions and Special Cases is probably the as often as the exception. E.g. Saxon Ffyrd, medieval Knights, etc.etc. Perhaps way to do this is to treat any unit that is Sleeping in home territory with no enemy unit within X hexes has a reduced maintenance cost (with further reductions from Policies as you say).

What you say about early modern armies suggests a key benefit from policies like Universal Suffrage (perhaps not that, but similar). I also agree that the Civ 5 mechanic sounds like an ideal way to take account of the production impact and manage numbers.

 

[Boris Gudenuf]

Great suggestions. I've just finished reading Caesar's Commentarii de Bello Galli, and problems with supply of food (usually corn) is mentioned on almost every page, hence my suggestion of tying into food. So totally agree about limiting factor being "cost", but cost means more than just Gold, particularly in relatively simple economies where most wealth is measured in terms of land, because land produces food (and other useful resources of course). In early civs, this also represents the cumulative loss of labour to work the fields from fielding large armies (more of a problem for the Gauls than Caesar).

The Gallic and Roman armies had two different problems, but representing them doesn't have to be different in a game. The Gauls, like many other civilizations, took their army right out of their economy: they were largely temporary warriors, following a small group of 'Big Men' who were more or less professional warriors. A similar system prevails right up to the European Middle Ages. Classical Greek armies were similar in that the Hoplites were citizen-soldiers, and whenever they were carrying a Hoplon (shield), they weren't farming or managing olive groves, or otherwise usefully producing. The Roman Army by the time of Caesar was a Professional, Army, filled with men who made a career (25 year enlistment) of weapons. They were replaced in the fields, if you will, by having the money to buy food from somewhere else - in Rome's case, Egypt and North Africa, in the other similar army, the Macedonian, from the Crimea using silver from Phillip's mines. So, although food supply is basic to both problems, Gold is the limiting factor. Without it, you can't buy what your soldiers can't produce, with it, you can afford to keep men behind shields instead of behind plows.

Similarly, if you want to represent loss of production, have a production penalty. Abstracting it all to gold places all the importance on that single element, whereas there's a good argument that the cost should be shared (personally, I'd keep it to Gold and Food, largely because it's simpler and because production maintenance costs isn't what Civ does - we don't have to pay production to maintain districts for example, or upgrade roads).

I like the Civ V system of losing Production only after you have reached the point of depletion of your manpower - in game terms, when there are more Units of manpower than there are 'points' of manpower in the cities. Before the 19th century, it rarely happened: Sweden managed to depopulate some of its districts in the early 18th century, but it was light on population to start with and was punching 'way above its weight-class (fighting a war with Russia, Denmark, and Saxony all at the same time and conscripting every able-bodied man out of the farms). It was not until the 'mass armies' of the first half of the 20th century (Industrial-Modern Era) that countries had to 'trade-off' men at work for men at the front. Earlier, as I said, the limiting factors are different. The Roman Empire never ran out of manpower, it outnumbered the barbarians right up to the end, but couldn't afford to mobilize and pay men to act as soldiers because of their wretchedly inefficient tax system and extended borders - they couldn't match needs to resources, because they couldn't get their hands on their own resources.

Your comment about Exceptions and Special Cases is probably the as often as the exception. E.g. Saxon Ffyrd, medieval Knights, etc.etc. Perhaps way to do this is to treat any unit that is Sleeping in home territory with no enemy unit within X hexes has a reduced maintenance cost (with further reductions from Policies as you say).

I would love to see a mechanic in which, say, a Castle (a Fort in the countryside) automatically produces one Maintenance-Free Knight on declaration of war (calling out your Feudal Retainers) BUT each castle reduces all the yields of the tile in which it sits to pay for that knight, neatly encapsulating the problem with the Feudal System - its reduces the resources available to the central government (you, the player)

What you say about early modern armies suggests a key benefit from policies like Universal Suffrage (perhaps not that, but similar). I also agree that the Civ 5 mechanic sounds like an ideal way to take account of the production impact and manage numbers.

The Civilization design team greatly underestimates the effects of the 'soft' factors: literacy and bureaucracy to more efficiently mobilize and collect the resources you have, Nationalism and Patriotism has motivating factors to recruit cheap soldiers, etc. The rise of modern armies and the building of Industrial Infrastructure (canals, railroads, highways) simply could not happen without the prior rise of University Education, near-universal literacy, and the invention of Corporate/Stock Financing through stock markets and central banks. They've made a start in the Civics tree, but so much more could be done to integrate it with the Tech Tree and the units...

 

[Dicemechanic]

They were replaced in the fields, if you will, by having the money to buy food from somewhere else - in Rome's case, Egypt and North Africa, in the other similar army, the Macedonian, from the Crimea using silver from Phillip's mines. So, although food supply is basic to both problems, Gold is the limiting factor. Without it, you can't buy what your soldiers can't produce, with it, you can afford to keep men behind shields instead of behind plows.

Just seems to me that if the issue is Food, and their solution was buying Food, then the game should have Food as the limiting factor AND add a mechanism where you can purchase excess food. I like to over-complicate. Now in game terms, that will never happen, simply because there is no such thing as excess food, certainly in the early game, as it all goes towards growth. So to implement this you'd have to cap the amount of useful excess food or make gold more useful for civs/city states that don't have it, etc. etc. Definitely one for Civ 7, not a mod!

And obviously another factor that comes into play, especially when thinking about Caesar, is that he basically funded his Gallic conquests by selling captive Gauls as slaves: a concept that has been in most Civs but is pretty much absent in Civ 6. An unsavory aspect of human culture for sure, but surely one that was common enough that it really has to be in there.

I would love to see a mechanic in which, say, a Castle (a Fort in the countryside) automatically produces one Maintenance-Free Knight on declaration of war (calling out your Feudal Retainers) BUT each castle reduces all the yields of the tile in which it sits to pay for that knight, neatly encapsulating the problem with the Feudal System - its reduces the resources available to the central government (you, the player)

Amen to that!

 

[Boris Gudenuf]

Just seems to me that if the issue is Food, and their solution was buying Food, then the game should have Food as the limiting factor AND add a mechanism where you can purchase excess food. I like to over-complicate. Now in game terms, that will never happen, simply because there is no such thing as excess food, certainly in the early game, as it all goes towards growth. So to implement this you'd have to cap the amount of useful excess food or make gold more useful for civs/city states that don't have it, etc. etc. Definitely one for Civ 7, not a mod!

When they originally started talking about 'Amenities' in Civ VI, I thought this was a great opportunity to divorce Growth from being purely a Food Function, but they didn't do that at all. You still can, basically, only 'trade' food within your own Empire (see the Trade Route yields in any game) whereas, in game terms, both Rome and Macedon were getting food from City States or other Civilizations, who were making a nice profit on it. As you say, it's obvious we will be getting no such mechanism in Civ VI, if ever...

And obviously another factor that comes into play, especially when thinking about Caesar, is that he basically funded his Gallic conquests by selling captive Gauls as slaves: a concept that has been in most Civs but is pretty much absent in Civ 6. An unsavory aspect of human culture for sure, but surely one that was common enough that it really has to be in there.

IF slavery is to be included, like any other historical factor, the game has to show both the positive and negative side to the gamer. Slavery provides cheap labor, obviously, but it also provides a continuous threat of Slave Revolt requiring investment in security forces. Also, slavery inhibits certain types of technology: labor-saving devices simply aren't profitable when you have cheap slave labor available, as witness Heiro's steam and pneumatic machines designed in Alexandria in the Classical period, but never followed up, because who needs machines when the Roman Legions will bring you a fresh batch of cheap workers with every campaigning season? Ironically, mechanical inventions got a boost in the late Empire, when the Legions were on the defensive and no longer conquering fresh laborers - the water wheel with sophisticated gearing to run everything from trip hammers to flour mills, wind mills and other 'labor-saving' devices...

In other words, slavery gives yo cheap labor, but perhaps removes all the Eurekas from certain Tech Lines while you're using it and increases your chances of generating 'rebels' - perhaps reduces the effects of Amenities from 4 cities each to 3. There are ways to represent the effects, even with Civ V as it is configured now.

 

[smartcanuck1988]

I'd be happy to see the supply limit return and be more prominent on the top info bar, just like the limits on trade routes. My only issue with it is that the production hit is really only for anything you want to build in the future. Production isn't really needed to maintain or operate the buildings you currently have. Especially with the district system now, the production queue (buildings-wise) is somewhat short-lived until you can unlock the next district or wait for the next building to be researched. Hence my point #2 suggestion, that requires citizens to operate the existing buildings to get any yield out of them. It would also make more than one yield matter/suffer in my opinion.

I also agree with the unit maintenance. Although I'm not sure if we should simply give warriors and such 1 gold cost, or should the base cost remain free and then with each era you add +1 cost. Related to gold, I just realized that population doesn't give tax revenue anymore? Seems odd and I hope I'm wrong? I think this should be reintroduced along with a "Tax Office" building for the City Center around Classic/Medieval to permanently increase the tax per citizen.

I love the castles idea!

 

[Boris Gudenuf]

I'd be happy to see the supply limit return and be more prominent on the top info bar, just like the limits on trade routes. My only issue with it is that the production hit is really only for anything you want to build in the future. Production isn't really needed to maintain or operate the buildings you currently have. Especially with the district system now, the production queue (buildings-wise) is somewhat short-lived until you can unlock the next district or wait for the next building to be researched. Hence my point #2 suggestion, that requires citizens to operate the existing buildings to get any yield out of them. It would also make more than one yield matter/suffer in my opinion.

Excess Production has not been a problem in Civ VI, IMHO and from what I glean from other comments: just the opposite, that between military units, civilian units (Builders especially), districts and buildings and wonders you never have enough Production to build everything you want. Right now, in fact, the over-speedy Tech/Civics Progression throws new things to produce at you faster than you can handle them: I find myself regularly building Classical buildings in the Renaissance/Industrial Eras, because I just cannot get around to them earlier!

I also agree with the unit maintenance. Although I'm not sure if we should simply give warriors and such 1 gold cost, or should the base cost remain free and then with each era you add +1 cost. Related to gold, I just realized that population doesn't give tax revenue anymore? Seems odd and I hope I'm wrong? I think this should be reintroduced along with a "Tax Office" building for the City Center around Classic/Medieval to permanently increase the tax per citizen.

There are Civics Cards now that reduce the Maintenance cost of units, and others that 'boost' Gold production, and those should be added to, perhaps with a set of Civics/Techs like Bureaucracy, Tax Farmers, Accounting, Census, and other 'efficiencies' that increase Tax/Gold collection. Another possiblity would be to add Citizen Slots to certain buildings that, when filled by one of your precious Population Points, would increase revenues: a Scribe in the Palace after Writing, a Customs Inspector in the Medieval Era in every Harbor or Market, etc. These would allow the gamer to invest in armies of 'bureaucrats' to increase revenue, but at the cost of possibly lowering overall totals of Food, Gold, Production - or increasing the overall totals and hope that it makes up for the inefficient collection of 'taxes'.

I love the castles idea!

I think it is a near-perfect example of what the game needs to show more of: the interaction among Technologies, Civics, and Units. Many units are actually the product of combinations rather than single 'technological' developments. Other examples would be Greek Hoplites, which were fairly ordinary spearmen with the 'Civics' of Citizen Soldier and Warrior Ethic, or the German 'panzers' which have little to do with technology (as a group, German WWII tanks were mechanically unreliable, ridiculously expensive to produce, and inferior to their opponents' machines is several fundamental respects, like radius of action and operational mobility), but everything to do with 'soft' factors like Mission Tactics, Warrior Ethos, and the Great General Staff (which could be a Wonder or National Wonder requiring an Encampment, Barracks, and Armory)

Instead, Civ VI has 'dumbed down' the Tech Tree and added a Civic Tree to 'compensate' but neglected most of the intricate interaction between soft (Civic) and hard (Technological) factors in history...

 

[smartcanuck1988]

Excess Production has not been a problem in Civ VI, IMHO and from what I glean from other comments: just the opposite, that between military units, civilian units (Builders especially), districts and buildings and wonders you never have enough Production to build everything you want. Right now, in fact, the over-speedy Tech/Civics Progression throws new things to produce at you faster than you can handle them: I find myself regularly building Classical buildings in the Renaissance/Industrial Eras, because I just cannot get around to them earlier!

Hmm that hasn't been my experience (I play on King mainly, maybe that's why). Still, I'm afraid it doesn't address my concern about it really affecting the future buildings and not necessarily what's on the ground right now. For example, if you built up a lot of cultural buildings, and then you are going through a massive war, your culture output should be taking a major hit somehow.

There are Civics Cards now that reduce the Maintenance cost of units, and others that 'boost' Gold production, and those should be added to, perhaps with a set of Civics/Techs like Bureaucracy, Tax Farmers, Accounting, Census, and other 'efficiencies' that increase Tax/Gold collection. Another possiblity would be to add Citizen Slots to certain buildings that, when filled by one of your precious Population Points, would increase revenues: a Scribe in the Palace after Writing, a Customs Inspector in the Medieval Era in every Harbor or Market, etc. These would allow the gamer to invest in armies of 'bureaucrats' to increase revenue, but at the cost of possibly lowering overall totals of Food, Gold, Production - or increasing the overall totals and hope that it makes up for the inefficient collection of 'taxes'.

Yeah the citizens slots you mention are exactly what I had in mind in terms of point #2 in original post. I like the idea of different slots in a building for different functions/bonuses! .. and we can easily add more buildings to represent the production needed to setup the infrastructure and pay salary (since buildings cost maintenance).

I think many of these basic facts of life (like taxation) should be for the most part a permanent thing (so a building is a better representation than a policy) as there are already too many policies to sort through. Techs/Civics themselves can increase the efficiencies, and we can free up the policies for things that make your civ unique. For example, a tax-related policy can convert some of the tax collected into culture/tourism and another one can convert it into some extra food. Both policies would modify the use of your tax, which is generated by worker (generated with Food) at a building (Production/Gold/(Faith)), whose yield is improved by a Tech (Science) or Civic (Culture). I think this would be a great approach to simplify the design but increase the interaction between the different game concepts.

Policies, in combination with the "citizen-to-military" point (#4), can also address what you mentioned about Romans vs Nomads. Romans would be a buildings-heavy empire where citizens are needed to operate them and so there is an indirect limit on building up your military without sacrificing important yields. On the other hand nomads would have unique policies that provide some of the benefits of the buildings. Since policies do not require a citizen to "work" them, this frees up the population to be built up into a bigger army.

I think it is a near-perfect example of what the game needs to show more of: the interaction among Technologies, Civics, and Units. Many units are actually the product of combinations rather than single 'technological' developments. Other examples would be Greek Hoplites, which were fairly ordinary spearmen with the 'Civics' of Citizen Soldier and Warrior Ethic, or the German 'panzers' which have little to do with technology (as a group, German WWII tanks were mechanically unreliable, ridiculously expensive to produce, and inferior to their opponents' machines is several fundamental respects, like radius of action and operational mobility), but everything to do with 'soft' factors like Mission Tactics, Warrior Ethos, and the Great General Staff (which could be a Wonder or National Wonder requiring an Encampment, Barracks, and Armory)

Instead, Civ VI has 'dumbed down' the Tech Tree and added a Civic Tree to 'compensate' but neglected most of the intricate interaction between soft (Civic) and hard (Technological) factors in history...

I fully agree with this. Great example of policies making your civ unique.

 

[Boris Gudenuf]

Hmm that hasn't been my experience (I play on King mainly, maybe that's why). Still, I'm afraid it doesn't address my concern about it really affecting the future buildings and not necessarily what's on the ground right now. For example, if you built up a lot of cultural buildings, and then you are going through a massive war, your culture output should be taking a major hit somehow.

See below: I think there's a way to combine your concerns

Yeah the citizens slots you mention are exactly what I had in mind in terms of point #2 in original post. I like the idea of different slots in a building for different functions/bonuses! .. and we can easily add more buildings to represent the production needed to setup the infrastructure and pay salary (since buildings cost maintenance).

Let's add two things to the game:
1. With any normal set of Buildings and Districts, a city will almost always have more 'slots' in them than it has people to fill them. Therefore, the gamer or the AI will be forced to allocate and re-allocate people points to 'enhance' buildings as required by the situation.
2. The enhancements available from a given Building can vary depending on Civics and/or Technology.

First, I think simply constructing a building should give you something. Even without any dedicated 'Merchants', a market provides a place for the ordinary citizen to trade, buy and sell. It will give a basic Enhancement to commercial activity. Similarly with a Library and other Buldings.
BUT to get the really major Enhancements from the Building, 'specialists' (to re-use a Civ V term) in the form of dedicated population, have to be working on/in them. The Type of Enhancement, though, is Variable.

Good Historical Example: A Temple or Shrine each provide a slot which, when filled, will provide further Enhancement to the Religious output and pressure. This is the Basic Function of both of these buildings. But, with Writing, you also get the opportunity to enact the Civic card: Temple Scribes. This allows you another 'slot' in each religious building which, when filled, provides an Enhancement to Science - you've provided a place to train literate workers in the Temple to work in keeping records throughout your city/civilization. BUT (and this is most important) filling that slot will not get you as much Science enhancement as filling a slot at the Library or University. (BTW, the University can also get an additional slot that enhances Production, as you train Engineers there).
So, you've got a choice: do you emphasize Science, in which case you will build Libraries and fill their slots, or do you emphasize Religion and use the extra Slot to get some Science without investing in all them Science buildings and District?

And if War starts, and you shift all the people to slots in the Barracks, Stables and Armories, you will be losing Enhancements in the buildings they came out of. Guns Or Butter (or Science, or Production, or Religion) - the age old question neatly captured in the game, for once ...

I think many of these basic facts of life (like taxation) should be for the most part a permanent thing (so a building is a better representation than a policy) as there are already too many policies to sort through. Techs/Civics themselves can increase the efficiencies, and we can free up the policies for things that make your civ unique. For example, a tax-related policy can convert some of the tax collected into culture/tourism and another one can convert it into some extra food. Both policies would modify the use of your tax, which is generated by worker (generated with Food) at a building (Production/Gold/(Faith)), whose yield is improved by a Tech (Science) or Civic (Culture). I think this would be a great approach to simplify the design but increase the interaction between the different game concepts.

And, note from above that simply giving you the gamer the opportunity/requirement to allocate population to slots in buildings that vary by Tech/Civic and your immediate Requirements, the game will provide even more ways to differentiate and 'personalize' your Civilization.

Policies, in combination with the "citizen-to-military" point (#4), can also address what you mentioned about Romans vs Nomads. Romans would be a buildings-heavy empire where citizens are needed to operate them and so there is an indirect limit on building up your military without sacrificing important yields. On the other hand nomads would have unique policies that provide some of the benefits of the buildings. Since policies do not require a citizen to "work" them, this frees up the population to be built up into a bigger army.

Good Point, which also, potentially, makes the Nomads more independent of pure Technological or Civic progress to stay competitive - up to a point. At some point (historically, the Renaissance post-Mongol Era) the Nomadic Civ will have to become more like the 'other' Civs and concentrate on infrastructure, technology and civic progress to stay in the game - which, to my mind, makes a separate Nomadic Type Civilization more interesting to play, and far more interesting than cramming such civilizations into a 'normal' agriculture-building emphasis Civ the way all Civ games have attempted so far.