The sustainability of CO2 emissions is measured in terms of emissions sources divided by its emissions sinks. A ratio above 1.0 means long term increases in atmospheric and oceanic CO2. The further above 1.0 we are, the less sustainable.

To make a very crude approximation, terrestrial emissions sinks scale proportionally with area. (Slightly less crude: sinks scale with forested area.) Generally speaking, countries need to have low emissions per unit area to be sustainable. Having low emissions per-capita but a high population density is no more sustainable than low population density and high emissions per-capita.

China has land area of 9328246 km^2 and emitted 10641789000 tons of CO2 in 2015: that's 1141 tons/km^2/year

USA has 9147590 km^2 land area and emitted 5172338000 tons CO2: 565 tons/km^2/year

Canada has 9093507 km^2 land area and emitted 555401000 tons CO2: 61 tons/km^2/year

EDIT: this was not intended to show China as "the worst." Belgium, for example, emits 3324 tons/km^2/year.

On one hand this leads to philosophical issues of justice like "luck egalitarianism": is it fair that Canadians, by accident of birth, have so much more headroom to emit sustainably than people born to the south or east of them? (Is it fair that any nation's citizens enjoy advantages from their own country's natural resources? I tend to think "yes, it's fair enough" but people may reasonably disagree.)

On the other hand, the pragmatic issue is that most countries are emitting CO2 faster than their geographic area can sink those same emissions. Certainly the largest political blocs are emitting unsustainably. The object of measuring emissions is ultimately to control them, not just to identify the party that deserves the loudest booing while we wreck the climate together. The USA, China, India, EU: none of them is "fine" and all need to make significant progress before their emissions are balanced with their sinks. (Not all of the effort needs to be on abating sources, though that's where the low hanging fruit is right now. Increasing sinking rates with afforestation, promoting soil carbon stores, and/or silicate weathering are also viable options.)

You raise a good point, but I think we have to introduce exports and imports into the calculation. As most countries are not subsistance economies, but rather produce for consumption in other countries (and import other products themselves), those exports have to be attributed to the country whose population consumes them. The products that land e.g. in Wallmarts should be attributed to U.S. CO2 output, whereas meat produced in Argentina but consumed in China should be attributed to China.

Otherwise, the numbers don't make sense: China wouldn't release that much CO2 if nobody would import from China.

I think that Exxon Mobil owns the responsibility for its business's CO2 impacts and can't wiggle away by saying "we're just servants to our customers' buying habits -- lecture them about emissions." Likewise I think that when Chinese factories make money from CO2-intensive exports they also own the responsibility for the corresponding environmental burden.

Of course Chinese factories may decline to voluntarily erode one of their cost advantages by bringing environmental standards up to parity with e.g. the EU, and in that case I think that the EU should use tariffs to ensure that imports don't get a competitive price advantage by way of externalized environmental damage. But that's somewhat different from believing that shoppers buying imports are really the responsible party when it comes to pollution generated by those goods' production.