Most PVC in industrialized countries is landfilled or recycled, not burned, primarily because of dioxin but also because it evolves hydrogen cyanide. Nor do the people who do burn tons of PVC optimize their process to produce phosgene (it is primarily derived from the VCM residue in the PVC). One would think they do the opposite, where it's permitted to burn it at all. Burn it at low temperature (a process quantitative decomposition) in an oxygen deficient atmosphere and you can then fraction off the phosgene by gravity (it's two and a half times the density of air, so it sinks).
Note: the process yields quite a bit more hydrogen cyanide than it does phosgene. So there's another hole in avocado's tacit assertion that import controls would preclude production of chemical munitions.
Worse for your fear mongering is that phosgene is a lousy chemical munition. It falls to the ground and then disperses. That worked trench warfare, but would be far less effective under other circumstances.
A typical combustion of PVC yields about 1ppm phosgene which quickly dilutes to far less than that under normal circumstances, hence jveritas' comment that people burning trash don't die isn't terribly meaningful. You don't need a lot to produce a fatal dose, typically 0.1ppm (about 150ug/m3). Given the adversely-optimized yield mentioned above, if you burn a ton of chlorinated plastics, you get about 1gm phosgene, enough to render about 6,000 cubic meters with a fatal dose.
How hard is it to get a ton of chlorinated plastic?
The point is this: if you really want phosgene, combustion of any chlorinated hydrocarbon will do; freon from refrigerators will suffice. It's an important constituent for processing pesticides, herbicides, and plastics which are all part of an industrial economy. It's easy to produce and makes a lousy gas for chemical munitions, but hey, if it fits your fear mongering, go for it.