I've seen some estimates saying that the total percentage of waste coming from households can be as lows as 2%, depending on how you count. For example, this cites an EPA study: http://www.zerowasteamerica.org/Statistics.htm
The same page mentions a higher bound of 20%, but in any case it seems reasonably well established that the vast majority of the waste produced (at least in the US) comes from the manufacturing/industrial process itself, not from households.
A book that goes into quite a bit of detail about this is "Gone tomorrow" by Heather Rogers.
I'd still like to have any real quotation. EPA claims about "Municipal Solid Waste":
"We estimated residential waste (including waste from apartment houses) to be 55 to 65 percent of total MSW generation. Waste from commercial and institutional locations,
such as schools, hospitals, and businesses, amounted to 35 to 45 percent."
See also what screwt writes: the rubble, for example, is not the kind of waste that has an environmental impact of the household waste (e.g. batteries with toxic materials etc). The important thing is environmental impact, not the absolute weight of the waste. I agree however that EPA seems to be too silent about the waste produced by factories.
tl;dr: Building waste is mostly rubble (not really necessary to recycle); these stats are by weight (density of rubble >> density of household waste). Treat these stats & suggestions of "no point doing any household recycling" with caution.
I think these statistics are somewhat misleading.
Waste from building tends to be rubble and dirt. Some of this can be re-used (eg filler for roads), but also the environmental impact of disposing of dirt is not much.
More importantly, the stats on the link above give waste measured by weight. A skip full of rubble (densely-packed stone) weighs far more than a skip full of general trash (loose-packed plastic bottles). A comparison by volume would be more informative[1].
That doesn't go for all building waste, and I don't know what comprises most industrial waste. All in all, I suspect cutting everyone's household waste to zero would make a significant impact on reducing the volume of waste produced nationwide; I don't know though whether it would make as much as a 50% impact.
[1] Obviously there are difficulties comparing by volume since household waste can be compacted to a great degree - my point is that comparison by weight skews heavily to the side of building waste.
Huh. Looking at statistics from Germany, the total waste is seperated into four categories: waste from building demolition and construction, waste from industrial processes, waste from mining, waste from private households and offices. Waste from demolition and construction makes up more than half of the total waste, the remaining three categories are split about evenly.
I'm surprised that the construction waste is that high. Apart from that, it seems that manufacturing and construction waste is less dominant over here than it is in the US. Odd! Maybe I'm reading the numbers wrong or they're counting really differently. FWIW, Wikipedia says that the per capita waste amount is about 500 kg, referring to just the category of private households/offices I assume.
The same page mentions a higher bound of 20%, but in any case it seems reasonably well established that the vast majority of the waste produced (at least in the US) comes from the manufacturing/industrial process itself, not from households.
A book that goes into quite a bit of detail about this is "Gone tomorrow" by Heather Rogers.