The point is that you won't ever gain the 40 lbs if you only eat the 20 cal/day extra.
Imagine you start eating the 20 cal/day extra. Initially, your weight climbs at 1 lb/year. However, by the time you gain that one pound, you have to spend an additional 11 cal/day to maintain that weight, so now you're gaining at a bit less than 1/2 lb/year. By the time you (asymptotically) approach having gained 20/11 ~ 1.9 pounds (which really only happens at infinity, but let's imagine you splurged to put yourself over), then you're at maintenance for your new weight. In order to gain more, you have to increase your uptake again.
As you gain weight, you accumulate metabolic load, so it takes a serious surplus of calories to continue to gain in the long term.
The steady-state condition of obesity requires a large ongoing surplus of calories to maintain.
Plausible fixes are incorporating more fiber in the diet to increase the feeling of satiety, reduce portions generally, cut out high-calorie drinks, and start exercising, even if only moderately but consistently.
Succinctly, the amount you "should" eat is not the amount to maintain your current weight, but the amount you need to maintain a healthy weight.
His actual point is at the end: 'Maybe when we get fat it’s because those physiological, metabolic and genetic factors you mentioned are dysregulating our fat tissue, driving it to accumulate too much fat'. He might be wrong about this, but his point is that the difference between an obese and non-obese person cannot be fully explained by an intake difference of 20 calories per day over a lifetime. There must be something else in play.
If that's his point, it's wrong. The intake difference between the obese person and the non-obese person is just shy of 1 big mac/day after 40 years (440 cals maintenance + 20 cals excess).