Because encapsulation prevents people from seeing that code (sort of).Another example of encapsulation is it can enforce devs to adopt better practice. It leads to less undefined behaviour.

As a dev, I should only be modifying/impacting code that I need to be interacting with. I shouldn't be able to accidentally modify X code. If other devs adopt encapsulation, I'm no longer able to accidentally interact with code I shouldn't be able to. This reduces undefined behaviour which is a huge part in producing a stable app.

If my system is allowing for tons of (irrelevant) places to interact with otherwise irrelevant code, I could end up introducing bugs/issues every time I PR code. But if my team adopts encapsulation, it reduces that risk.

In other words, it reduces dependencies. Every dependency you add to a module increases the chance of bugs and increases the complexity of that module. So it's harder to work with a system if there are 10000000 potential dependencies vs only a few. It's cleaner and more clear.