That said, you could argue that data networks are mostly a natural monopoly because it's not feasible or efficient to roll out several redundant fiber/cable networks.
Even if there were no franchise agreements in place, very few companies (excepting ones that have other revenue streams) would roll out a second or third cable/fiber network in a city where there is already one in place. Even if you managed to split the market and get half of the potential customers, you'd need to account for the cost of digging trenches and laying cable (which cable TV companies have long since recouped). Makes it hard to stay solvent at such a disadvantage.
It's why a lot of people think the ideal situation is for a single physical network to be built and then service providers pay for access and compete on service to businesses and customers. With physical networks divorced from service providers, the company or municipality in charge of the actual cable/fiber makes their money from maintaining and improving capacity so they can sell access to more providers. Providers compete by offering the best services and customer support in order to profit and pay for more bandwidth on the physical network.
But yeah, it's more complicated (in terms of both business and networking tech) but it's an ideal that many would like to move toward.
But in the end, I'm not sure what's the best way to handle it. I just know that many would agree that the current methods are not optimum, and possibly detrimental.
This model was used regularly for rail and power networks in Europe, but these have all been privatised in the past few decades citing "cost reductions". The net result is that our infrastructure is deteriorating, consumer prices are rising way faster than inflation, and critical infrastructure is now in the hands of a few international power brokers (e.g. the Dutch national telephone grid is owned by Carlos Slim).