Randers' book tries hard to make an objective forecast of what the world will be like in 2052, taking resource limitations into account. He's very good about saying where his data come from, how he interprets them, and how he extrapolates the trends. If you are one of the many economists, journalists, and other pundits out there who are constantly saying that resource shortages are not a problem and that economic growth will solve all problems, this is a great book to get you started on thinking more realistically about the future. For that reason I give the book five stars.
Randers freely admits that his forecast has very large uncertainties. He admits that there are many wild cards out there. It's always possible that some huge, new oil or gas discovery might be made. Randers makes the point that this would be good for economic growth in the short-term, but would also make climate change worse and delay efforts to improve energy efficiency, so that it's hard to say how much long-term change this would cause. It's possible that some considerably nastier things might happen, from financial meltdown and nuclear war to epidemic disease and ecological collapse. It's even possible that humanity might wake up and decide to put a serious effort into population control and reducing pollution, which would probably mean a much more pleasant future. But Randers is mainly interested in what he sees as the most likely future based on the scientific data available now. If you are a "doomer," Randers' forecast will probably not please you, because he thinks that everyday life for humanity in 2052 will probably not be wildly different from today. He thinks life will be more unpleasant in 2052 in many ways, with climate change at crisis levels and wild nature almost gone. However, Randers thinks that while it's possible that there will be a spectacular human die-off, it will probably hit after 2052.
In my opinion, Randers' forecast has many weaknesses. He's a big fan of carbon capture and sequestration (CCS), solar panels, and wind turbines. He sees the transition to sustainability as involving energy efficiency, yes, but also lots of high tech. I simply cannot agree with him here. I think CCS from power plants cannot work on a scale large enough to make any difference to the climate; I'd say it's more of a coal company boondoggle than anything else.
As for solar panels, Randers is correct that prices for these have been coming down. However, I don't think this trend can continue through 2052. The low energy density of sunlight puts hard limits on what can be done with solar panels. Large-scale electricity generation using solar panels cannot ever provide electricity in amounts sufficient to do things like running toasters and clothes dryers. I think the transition to sustainability would do far better by focusing on low tech solutions, not high tech. Solar energy comes in the form of diffuse heat. It works best when it is used directly to provide diffuse heat--not when used to generate electricity. Think clotheslines for drying clothes, solar hot water heaters on rooftops, and passive solar heating for buildings. Similar hard limits apply for wind turbines. Using a wind turbine to generate electricity, then using the electricity to do something else, necessarily involves losing most of the power in the conversion. We don't often think about the power lost in power generation, lost again when electricity is transmitted through power lines, and lost again when an electrical device is operated. The reason we don't normally think about it is that fossil fuels provide such a dense source of power that such losses don't make a huge difference. This is just not true with solar and wind power. Solar panels and wind turbines are fine for technologies that use only a small amount of electricity--like telephones--but they can't provide enough power to run an industrial society. For more on this, see The Wealth of Nature: Economics as if Survival Mattered. The website lowtechmagazine.com also has great articles on this subject, such as how to build urban areas to maximize the use of direct solar heating in buildings.
Randers also thinks that the current trend toward urbanization will continue, with only a small human population working the land. I have to disagree with him here. I think it is much more likely that in the next few years we will see a big increase in unemployment. At the same time, the decline of fossil fuels will mean a big increase in demand for human labor in the fields. Eventually these trends will meet up, and the percentage of people working at farm labor will go up. Jobs formerly done by chemicals or diesel-powered equipment will be increasingly done by people. Long days using a hoe or a pitchfork will be common. The reason for this is simple: in a world where labor was expensive and fuel to build and run machines was cheap, it made sense to run farms using machines. In a world where unemployed and desperate humans are all over the place and fuel for machines is expensive, running farms with mostly human labor will be what makes sense.
Overall, though, Randers' book is thought-provoking, and useful in some respects. I recommend it.