The Kessler effect says that if satellites collide and produce space debris at a rate faster than satellites de-orbit, then we'll get a cascading effect where the number of objects in orbit grows faster and faster, causing more and more collisions, and more and more objects to run into. Since collisions with even a grain of sand at 42,000 miles per hour can take out a satellite, it's more the number of objects out there than the size of them.
Once the Kessler effect is going on in earnest, you can't put anything into space at all without it quickly colliding with things.
If your goal was to cause a Kessler effect, you'd dump bags of sand or pebbles into space in popular orbits, and they'd take out everything. Given that it's a big world, and just about any goal will be adopted by someone sometime once it's easy enough, and this is very easy to achieve, this will probably happen. Even if this weren't ever anyone's goal, I know people already have the goal of dragging asteroids into earth orbit. And asteroids, it turns out, are usually mostly loose collections of sand and pebble sized rocks. Even if nobody did that, the Chinese blew a weather satellite into 150,000 pieces in 2007 (military antisatellite test), and two other satellites accidentally collided at 42,000 miles per hour in 2009, so the existing satellites will decay into this eventually.
So the Kessler effect is starting already, and it will eventually be a fact of life for the world we live in. What do you do once it's there in earnest?
Consider the bags of sand dumped into orbit. They disperse into a cloud, filling an orbit. They'll run into each other, which will make them ever smaller pieces of dust. They'll get static electric charges, so they'll interact with each other and earth's magnetic fields even if they don't directly run into each other. What I'm saying is, there is essentially friction for anything that isn't going the same speed as everything else.
What speed is everything else going? That's the wind. Space doesn't have much of a wind right now, that's why we can have satellites having near misses at 42,000 miles per hour in the first place. But once the Kessler effect is up and running, there will be a wind. If you're going at the same speed and direction as the wind, you're fine. Anything that is seriously different from that is going to get destroyed. You could even gather the particles from the wind and fire them off again, as a form of propulsion.
Low earth orbits tend to be all over the map. Nobody's even trying to set up traffic lanes. Iridium (one of the culprits in the 2009 collision) has polar orbits. Polar orbits are just asking for trouble.
What if you don't have a steady wind, but instead just have lots of marbles flying every which-way? Ignoring turbulence and precession and such, the Kessler wind at any point wants to be the average of the original orbits going through that point. And that wind will be in its own orbit, according to its speed and distance from earth. If the speed is low enough, it won't stay in orbit, it'll fall into the atmosphere and be gone. So: you could launch clouds of dust in retrograde orbit, the average of that and the marbles will be zero, and they'll all fall out of orbit. There's 100% probability of hitting dust clouds, so it would capture everything. And dust tends to be statically charged, so the earth's magnetic field doesn't let it stay in stable orbit for very long anyhow. This may be a way to clean up a mess. We'd have to simulate it to see whether it makes things better or worse.
So, what direction does the Kessler wind blow? Well, obviously, being me, I'm going to want to make it follow a torus. The way to guarantee that is to put all your satellites into torus orbits in the first place. That would have a side effect of reducing the speed and frequency of collisions even without a Kessler wind. All satellites would go out on one side of the plane of the moon's orbit, and all would come back on the other side, and neighbors should all be going nearly the same speed. (But not exactly the same speed otherwise you get clumping.)
Below is a single red satellite in an inclined elliptical high earth orbit, with the earth fixed, from a fixed viewpoint high above earth. Drag the mouse to change the viewpoint, or shift+drag to change the scale. Over time you can see the satellite's orbit precess. The moon precesses as well. Precession of the satellite is mostly due to the moon.
A circular orbit in the plane of the moon's orbit around the earth does not precess. These are very stable orbits. This is nearly the same plane as the earth's orbit around the sun (5 degrees off). The moon's orbital plane does precess itself due to the sun. The plane of the moon's orbit does not match the earth's equator, it is off by 18 to 29 degrees. Precession means a space elevator over earth's equator would need active orbital maintenance, since it will try to drift towards the moon's orbital plane.
If the satellite's elliptical orbit has the major axis in the plane of the moon's orbit around the earth, then the far end of the major axis of the satellite's ellipse precesses in roughly the directly of the satellite's orbit at that point, always. If it sometimes went up and sometimes went down, then I could find some set of elliptical tilted orbits where the major axis always stays in the plane of the moon's orbit without active maintenance, and we could build a torus. Alas, all orbits between the earth and the moon behave this way, regardless of distance, ellipticity, or tilt. Ditto for retrograde orbits between earth and moon. The most stable satellite orbits I think we'll find are saturn-style ring of circular orbits in the plane of the earth's orbit around the sun. Like a torus, it keeps neighbors next to neighbors and naturally avoids high-speed collisions. Unlike a torus, it's a thin band when seen from earth, it does not cover the whole sky.