As to your first response:
I’ve always tended to be led by the literature to the conclusion, that the traditional urban elites benefited immeasurably by having an Absolute Monarch, and that their interests were almost always aligned.
For one thing these urban elites tended not to be traders, they tended to be absentee landlords, who centred themselves in urban polities not for any financial reasons (though that might have played a part), but instead to be close to the political pulse of the polity. If anything the Absolute Monarchs were just the largest absentee landlord with taxation and law making power, the interests of the Absolute Monarch were thus directly inline with the urban elites, they had a vested interest in maintaining the status quo with regards to land ownership (there primary income source at least initially). After a while it becomes noticeable that the urban elite began to move away from being independent landlords towards becoming dependants of the state. Families in Egypt, Babylon etc, routinely reoccur throughout the records in government service, in Egypt through tombs, in Babylon through commercial documents, all of which strongly indicate wealth correlating with high office (this did change slowly, glacially, probably tied to increasing power of the bureaucracy as the states grew in size and complexity). That is with regards to the urban elite being absentee landlords (which tends to hold well for the hydraulic societies which by nature favour highly an inverse pyramid for landownership, by virtue of the high costs associated with maintaining the infrastructure).
If the urban elites are traders or craftsmen or medium-small holders, which tended to happen in what is now Syria, Jordon, Israel etc, there is a strong penchant for either limited monarchy or a council of some sort forming to administer the state. It helps that these societies were rain fed which by virtue requires far less capital and organization to manage compared to the irrigation systems (the impetus for an absolute monarch goes out of the window). It also helps strongly that the land in the regions is either marginal, or heavily interrupted (the hills of Israel) which tends to strongly favour smaller landholdings, its much more difficult to build a powerbase out of a large number of small holders, that to build a powerbase out of a small number of oligarchs, let us not also forget that marginal land or heavily interrupted land strongly necessitates crop diversification (which has been linked to reduced dependence on the state, and on divergent interests between landholders) all of which make it even more difficult to centralise power. Besides trader and craftsmen have a vested interest in constraining the power of a monarch or government in general, the Code of Hammurabi was a severe constraint on economic growth (some serious arguments about that, the relative stability it would have bought might have outweighed the dead weight loss of some of its provisions, limiting the ceiling of loans is not a good idea ditto with grain prices, although there is strong evidence that the rates were routinely changed (at least in times of serious need, the price rose by some 400% during a particular siege of some city compared to the normal rate (for the life of me I cant remember where I got it, from one of the dozen books on my desk atm…

), at the very least the government had a vested interest in clamping down on predatory actions from merchants and craftsmen who were for the most part not a favoured group, price gouging, extortion, usury and the like have a long history.
I certainly believe that power struggles between factions led to the founding of new cities, but founding a city with a small number of oligarchs in direct competition to a large centrally controlled city would seem to me to be madness (a King who did not chase his cash cows would be a double fool, for allowing a competitor and for fleecing his own treasuries). Greece has a great history of founding cities because so and so had a fight with so and so (at least in mythology, I’m not sure on the historical accuracy, but it certainly seems possible) but I don’t think it holds at least rationally to me that the majority of cities were founded in a similar way (the literature doesn’t mention this at all, at least what I have my hands on, if I have time I’ll go digging in the journals and try and find something). Certainly in Greece and areas where central governance was not strong then it is highly plausible that cities were founded not more than a few miles from each other in the safety of a neighbouring valley purely as a tax dodge.
Notice: My apologies I think I might have misread, and assumed you were using the above reason as the majority case, you might not have…
Yes I think in the long run, trade and safety are the major reasons (but the first cities to develop would probably have been founded not just for safety reasons (a major consideration) or for the trade reasons (a small market early on can grow to a large market rather quickly). We can also deduce though that Malthusian logic can be applied, to found a city or any population centre requires one simple thing food. If your farmers can produce more than they consume, they can feed specialists, a village can begin to form when the first farmer in the long run produces more food than he and his own can consume, he then trades this surplus or stores it. Once this farmer’s productivity has risen beyond what he can store (time constraints and storage limitations) he is forced to trade or have it waste away. To use Adam Smith, a single person can make but 3 nails a day, a single blacksmith can make 60 nails a day (not the exact numbers but the principle stands), the division of labour. So it pays the farmer to trade his goods to someone else who has a comparative advantage in something, for instance another farmer can make 5 nails on his own a day, 2 above the nails that of which the other farmer can make. Not quite enough yet for the second farmer to branch into a trade, but add another four or five dozen farmers with a surplus and then the blacksmith even in lean years is likely to be able to churn out nails uninterrupted. The rate of farmer to specialist would initially have been high, but with each increase in production of food some of that would have gone to supporting additional specialists, while most of it would have gone to increasing the general population (so yes growth in a sense was possible, before Malthus kicked in, or at least a steadily rising percentage of specialists, although there would have been a point at which the land could not be farmed any more intensely, and adding any further specialists would have caused the system to break down in food shortages…

.
Trade is fairly apparent, during the high times of the Bronze Age, cities were built to maximise trade, not a single Minonian city during the good times was built in a defensive spot (I would need to fact check this) but during the collapse the settlements moved further and further up into the hills as it got increasingly dangerous. The Dorian Greeks likewise built cities in fairly indefensible areas (relatively speaking to what they could have built them in) while the Mycenaean’s built them in more defensible locations. Egypt, well they didn’t build many if any of their cities defensibly (maybe owing to their inability to do so) but also probably owing to the relatively stability of the region for most of its history. The Fertile Crescent Empires oscillated between the two (as much as it could) cities on the periphery seem to me to be on balance built for defensive considerations, while the internal cities appear too built to maximise trade and access.
As to your second point, government at the time had few means of checking demand, they can only basically ask the merchants (and you can guess their response! Yes please we need it, guess who profits best of any capital works). Climate and other issues had a lot to do with it to, but it also boils down to viability, one cannot check if a given region can support 20,000 people very easily, water, food, sanitation and the other essentials might block settlement. It’s more of an issue of not being able to test an area before you judge it to be suitable (test in a meaningful way that is).
As to your third point I think I have covered it in my response to the first post. I do however tend to note the following (from geography I think), typically villages (contain the bare essentials, and many small farms) which feed into towns (which take up excess food from the villages, and provide larger services), which feed into cities (which take up an increasing amount of extra food and provide the largest services) it can make quite an interesting lattice shape, with cities in the middle, with towns emanating outwards, and villages emanating outwards from those. It’s not a set in stone rule, but it tends to hold well.
I’m guessing trade continued on virtually uninterrupted during war, symbolic cuts might be made, but it would not surprise me to find out that merchants just false flagged (pretended to be from city Y not X) and continued on as before just taking a diversion around the combat areas (which are limited by geography and logistics anyways). Payments of bribes would have solved the rest, but yes in larger centrally controlled states with a deeper fiscal pocket, it would have been much harder, but even then false flagging and choosing a different city to trade in might have got one around it… and trade even on a normal basis was dangerous, there is an interesting theory bouncing around that private merchants themselves seldom travelled (not universally but in a few areas this holds relatively well) but actually just paid agents to do the job, the capital was advanced, a return contract was required properly affixed from a reputable or temple source (there have been a few of these found in various places) and of you go, that agent might have been a family member etc. Not to mention most trade was a centralized operation anyway, although it appears that barring a few of the more silly instances, private merchants just purchased licences and there was no other consequence (even then one could conceivably just wriggle around the restrictions with the right mix of bribes, a modified trade route and false flags).