This is a thoughtful and measured post Karen. My colleagues and I at the Breakthrough Institute will soon be publishing a thorough literature review covering the current state of the art in research in the field. Hope you can take a look when it's out (in February most likely).
One comment on your post. As you correctly note, there are a variety of different rebound effects operating at various economic scales and scopes. The full global, macroeconomic impact of rebound effects is the scale that matters most to climate and global resource depletion concerns, although micro-scale direct rebounds are more important to individual consumers and businesses, while utilities may concern themselves with locally-bounded macroeconomic impacts only, etc.
After introducing the various direct, indirect and macroeconomic effects, however, you write:
See the paper for the scale of the rebound effect, which is close to 0% for home appliances, 10 - 30% for cars, and 0 - 50% for space cooling.
This is the scale for direct, micro-economic-scale rebound effects only, however, not the full scale of combined direct, indirect, and macroeconomic rebound factors. As you correctly note, the ultimate scale depends quite a bit on what kind of efficiency improvements we're talking about and what kind of economic sector or country that occurs in. Combined economy-wide rebound is going to be quite a bit larger than those figures however. Rebounds are particularly large, and can sometimes lead to backfire -- rebound greater than 100% of projected energy savings -- in developing economies (as you note), in sectors where elasticity of substitution of energy for other factors of production is high (e.g. particular industrial sectors, most notably energy-intensive ones), and when efficiency improvements yield productivity improvements in both energy, as well as other factors (e.g. labor productivity boosts from better and more efficient lighting, for example), which produces an out-sized impact on economic production and output (and associated energy demand).
In the end, while the insight into rebound effects first proffered by Jevons is widely considered a paradoxical notion, the combined impact of rebound effects means we have reason to be skeptical of the ability of below-cost energy efficiency to drive real and lasting reductions in total energy consumption and thus the ability of efficiency measures to significantly contribute to climate and energy security objectives directly. Relying on a linear, direct, and one to one relationship between below-cost energy efficiency improvements and carbon emission reductions, as is almost universally the case in contemporary policymaking, is very likely to lead nations and the world on a dangerous path.
To be prudent then, efforts to reliably reduce greenhouse gas emissions or dependence on depleting fossil fuels that aim to avoid the risk of overreliance on energy efficiency measures should focus primarily on shifting the means of energy production (rather than end use), relying on zero-carbon and renewable energy sources to diversify and decarbonize the global energy supply system.