I say nothing about costs.
But regarding the concept that the
luminous flux (which is measured in
lumens) decreases with the square of distance, this is
plainly wrong.
What decreases with the square of distance is another physical quantity, called
illuminance, and measured in
lux.
One lux is one lumen divided one squared meter.
So, if a given luminous flux, say, 1000 lumen, is propagating inside a cone with a certain angle, say 65°, so that at 1m distance the area of the cross-section of the cone is, say, 0.5 m², you will get an Illuminance of 1000/0.5=2000 lux.
You go further away, at a distance of 2m, where the cross-section area is 4 times larger (2 m²), and the illuminance will be 1000/2=500 lux. The illuminance is now 1/4 of the previous value, but the luminous flux remains the same (1000 lumen)...
What characterises the output of a light source is the luminous flux (lumens). So a claim of generating 35000 lumen from a power source capable of 360W is a reasonable value, as we know that decent LED sources usually are around or exceed an efficiency of 100 lumens/watt.
There is a third quantity characterising how the luminous flux is spread in a narrow or wide beam, and this is the luminous intensity I, measured in candelas. I is the ratio between the luminous flux and the solid angle of the beam (measured in steradians, not in degrees).
Given a certain luminous flux, the luminous intensity will be larger if the beam is narrower.
Also luminous intensity does NOT decrease with distance, apart the case of light propagating in an absorbent medium (such as dirty water).
In this video you can see an explanation on the cause-effect relationships between luminous flux, luminous intensity and illuminance: