This book was lacking in several respects:
1. The numbers behind the graphs are not provided and are not available so you cannot do any further analysis yourself. The graphs themselves are also drawn in such a way that it is hard to extract the numbers using a ruler.
2. The problem of survivorship bias. They claim that while the 16 countries analysed are an incomplete list (only 70% of world GDP in 1900), this is not a big problem, they feel. Their message that stocks do well in the long run supposedly remains intact, however they do not provide any solid evidence of this. The countries left out of course suffered terrible performance, with total confiscation of assets in most cases and major losses in others.
The countries left out include: Russia, China, Eastern Europe, Latin America. As an example, Argentina was the wealthiest country 100 years ago but was left out. They claim that their criterion for inclusion was the availability of data, but Switzerland was included even though the data is incomplete.
In my opinion, some attempt should have been made to adjust for this problem.
3. No assessment is made of the issue of capital controls etc as an impedement to implementing the world indexing strategy. It is simply assumed that equal dollar indexing could be implemented without any costs, and with no taxes.
All in all, this book fails to provide a realistic and convincing assessment of global investment returns in the real world.
Victor Niederhoffer uses this book to justify his bullishness on stocks, Sorry Vic, no cigar.
First, although much is made of the quality of the data much of the data is from other authors and sources and is used blindly. In other words much of what the authors have done is collation.
Second, there are several countries that ostensibly did better than the U.S and Britain during the twentieth century but little is said about this fact and indeed the authors spend some pages justifying the emphasis on the U.S. and Great Britain during the twentieth century. But the reasons they give are hollow. There may have been justification but greater depth was in order.
Third, although there is a great deal of information in the book several graphs are presented in a confusing manner i.e. the same color line is used for several curves on the same graph making interpretation almost impossible.
Fourth, reading the book is in many places a lot like taking the SAT or GRE graph interpretation section. There is much here but it would have been much more impressive if the authors had drawn some deeper conclusions than was the case. Probably many will attempt to justify this deficiency by stating that the purpose of the book was the assembly of the data in one place and that others will use the data for developing profound interpretations and conclusions. But I don't believe that will happen: the authors understand that the main users of the book will be institutional and personal investors. That is sad for there is much to be interpreted here in a deep and satisfying manner.
Fifth, the authors clearly suscribe to the idea that the data in this book will help readers determine the most likely future returns of the stock markets. But I suspect strongly that William Bernstein is correct and that past returns are best at determining future risks of investing and rather less usefull for determining future returns. The case the authors make for using the data for determining future results is laughable: in essence everybody else does it. The hypothesis may be true but that is hardly a case for it.
Sixth, it would have been fascinating for the authors to have analysed how open some of these societies were during various periods in the twentieth century and related that to investment performance.
Those that believe a CD or soft copy would have been better because of the more detailed access to the data have missed the point entirely.
In the final analysis there is enough data here to keep political economists, sophisticated investors, and many others occupied for many years-if nothing else they can attempt to fill in some of the gaping holes. For instance, many in the U.S. believe that our inflation numbers for several years now are "cooked" to some extent. Certainly this is true for some other countries as well. Yet every bit of the inflation data is presented as the gospel with no challenge or analysis. If the authors did do some analysis here (or even if the primary sources did) and found the "cooking" insignificant or significant then it seems they should have presented their findings in a more complete fashion. Is the data reliable or not? That is the question. This is just one example out of the entire book. Hardly a page goes by without similar questions arising. Too bad the authors could not be bothered to answer any of them
Finally, it is particulary fun to read Triumph of the Optimists in conjunction with Adventure Capitalist.