Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Origin of Wealth: Evolution, Complexity, and the Radical Remaking of Economics Hardcover – June 1, 2006
"Children of Blood and Bone"
Tomi Adeyemi conjures a stunning world of dark magic and danger in her West African-inspired fantasy debut. Learn more
Customers who bought this item also bought
Customers who viewed this item also viewed
What other items do customers buy after viewing this item?
From Publishers Weekly
Accounting for the creation of wealth has long challenged humanity's best minds. For business readers and academics, Beinhocker is a zealous and able guide to the emerging economic paradigm shift he calls the "Complexity Economics revolution." A fellow of the economic think tank McKinsey Global Institute, he rejects traditional economic theory, based on a physics model of closed systems, in which change is an external disruptive shock. Instead, he outlines an open, adaptive system with interlocking networks that change organically, reflecting the interaction of technological innovation, social development and business practice. Wealth is created to the degree that this interaction decreases entropy in favor of "fit order" that meets human needs, desires and preferences. Beinhocker is sufficiently comfortable with this evolutionary model to advocate a comprehensive redesigning of institutions and society to facilitate it. He argues for corporate policies that favor many small risks over a few big ones and recommends restructuring financial theory to favor growth and endurance rather than short-term gains. Though he asserts that complexity economics can reduce political partisanship and increase social capital, Beinhocker stops short of saying that it cures sexual dysfunction. By the end, the concept emerges as a great idea that the author tries to make a panacea. (June 1)
Copyright © Reed Business Information, a division of Reed Elsevier Inc. All rights reserved.
Eric Beinhocker's The Origin of Wealth ties risk management, incentives, and human psychology together with many other criteria, all under one philosophical framework. --The Motley Fool, December 31, 2007
Author interviews, book reviews, editors picks, and more. Read it now
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
The other 60% is devoted to the author's vision of complexity economics, which regards the economy as a type of "complex adaptive system" (CAS). It's a possible successor to NCE that's based more on modeling, empiricism (a dash, anyway) and metaphors from evolutionary biology and other scientific fields. (More on metaphors below.) As for this part of the book, there's good news and bad news. The bad news: a lot of it is sloppy, garbled, misleading or even wrong. The good (and ironic) news: the author fails to show that the other 40% of the book logically depends on this imaginative but mushy 300 pages.
If you've never read a critique of NCE before, never heard of CAS theory, and/or never worked in an entrepreneurial environment or read books like "Built to Last" or "The Balanced Scorecard" you can learn a lot from reading this book. If you're wondering why I give it only three stars anyway (2.5, actually), you can read the rest of this review. Here is a very abbreviated list of my irritations and disappointments with this book:
1. Unduly half-hearted critique of NCE:
Compared to earlier critiques of NCE by Mandelbrot, Osborne, Ormerod, Keen, and McCauley, among others, EB wimps out. E.g., despite his description of the totally unempirical nature of NCE, he persists in regarding NCE economists as scientists (at 75). He also says "It would be hard to believe" a model of economics that doesn't include the law of supply and demand (at 95). But take a look at M.F. Osborne's "The Stock Market and Finance from a Physicist's Viewpoint" for a demonstration that this "law" is vaporware.
EB doesn't cite Osborne, but he does include Philip Mirowski's "More Heat than Light" (1988) and "Machine Dreams"(2002) in his bibliography. The latter book -- an exhaustively documented, slam-dunk withering indictment of NCE based on the NCE economists' own words -- seems to have made little impression on him; it's only cited once, for a small point. You should read it if you're going to read "Origin of Wealth". In addition to the criticism of NCE, you'll get a thorough historical context for the mathematical modeling that underpins EB's arguments in Parts Two and Three of his book.
2. Ambiguous or careless citation of sources:
A. EB provides an extensive set of footnotes and an impressive bibliography, but checking the citations can be a frustrating task. EB frequently cites to edited volumes of articles or essays -- but almost always to the entire volume, rather than to the specific article that might support his point. Page numbers are rarely included in the cites, so his sources often remain cloaked in ambiguity.
B. More troubling is a discussion of "punctuated equilibrium" (PE) at 173-175. Here EB discusses at length two papers by Jain & Krishna from 2002 -- but neither one mentions PE at all. In fact, no grammatical form of "punctuate" or "equilibrium" appears in either paper, but for one exception that is not relevant to EB's point. This left me concerned that other of his cites relate only tangentially -- or perhaps not at all -- to the points he claims they support.
C. Sometimes the cites even negate his point. In his blog about Adam Smith, Gavin Kennedy has noted a case in which EB misstates the position of Prof. Herb Gintis and his colleagues by 180 degrees (EB at 121, and Kennedy entry for 2006/8/26). (BTW, Prof. Gintis, whose encomium of the book appears a couple of reviews below this one, is affiliated with the Santa Fe Institute (SFI) and is cited approvingly in several places in the book.)
3. Opinions presented as more unified than they are:
A. Why did I just mention SFI? Because it's one of the leading think tanks for complexity theorists, and the homeland of CAS theory. Also because it's financially underwritten by, among others, EB's employer, the consulting group McKinsey & Company. EB is very up-front about disclosing this relationship. Nonetheless, he omits mention of most alternative points of view about complex systems, producing what amounts to a brochure for SFI, and giving the impression that complexity theory is CAS theory. For example, you won't encounter even a paragraph about the philosophy or techniques of the Brussels School (of Nobel laureate I. Prigogine et al.), nor about the attempts of Ralph Stacey and others to integrate these into management theory and practice. (See, e.g., Stacey, Griffin & Shaw's "Complexity and Management" (2000) and related volumes, which focus far more on people and organizations, and far less on computer modeling, than EB.)
B. In other cases, EB uses ambiguous quantifiers to suggest that opinions are more widely-held than they may be in actuality. Consider this statement (at 12): "Modern evolutionary theorists believe that, like gravity, evolution is a universal phenomenon, meaning that no matter whether the algorithm is running in the substrate of biological DNA, a computer program [or] in the economy ... evolution will follow certain general laws in its behavior." Logically, this means that there exist *some* theorists who believe all these things, but rhetorically the statement is made to sound as if their number is legion --despite the fact that the statement concatenates a number of potentially contentious notions. So who are these theorists? Biologists? Cosmologists, even? Checking EB's footnote shows that 4 out of 5 of the cited "modern evolutionary researchers" are computer scientists. So it's not surprising they take this abstract, algorithmic point of view. (The fifth cite is to an edited volume of SFI proceedings, cited as a whole.)
C. A more direct example of misrepresented consensus is the statement "[M]ost researchers believe that the brain is a Boolean network" (at 148). According to one Stanford neurobiology professor with whom I checked, this statement is simply false. While some researchers may believe it's Boolean, more believe the brain is closer to a Bayesian network, and others believe that it's neither Boolean nor Bayesian. (Caveat: Her reply was based on researchers who actually study *the brain* professionally -- researchers who study computers, rocks, etc. may have their own ideas.)
4. Unexplained disunity in EB's own opinions
At times I had the feeling that EB was speaking out of both sides of his mouth. This first came up in connection with hierarchy, which seemed at first to be bad, then good, all within 4 pages. One has to wait about 200 pages for a resolution of this dialectic (compare 152-155 and 356 with 364-367).
Unfortunately, not every contradiction has such a happy ending. EB disses econometrics because it provides only statistical correlations, not causal explanations (at 59). His attitude is quite the opposite when it comes to CAS-based computer models, however. Shortly after noting that it's impossible to predict a model's output without running it (at 232), he goes on: "We may then be able to say things like `For parameter settings of X, there is a 60 percent chance that a high-profit environment will emerge and a 40 percent chance of a collapse into low profits.' Thus, while we may not be able to forecast the model's specific outcomes, by exploring the parameter space and collecting statistics, we can learn a good deal about the model's behavior. In many ways, a deep understanding of how the system works may ultimately be more valuable than being able to make forecasts" (at 233).
Shades of "It's not a bug, it's a feature" -- how did statistical correlations become cool again? And why do they now constitute "deep understanding"? It's a shame for such mysteries to surround one of the key practical points of the whole book.
5. Sins of definitional omission and commission:
A. Try as I might, I couldn't find a definition of "complexity" in this book. EB's reticence to define the term, however, doesn't inhibit him from making quantitative statements incorporating it, such as "[T]he global economy is orders of magnitude more complex than any other physical or social structure ever built by humankind" (at 6), or "The number of SKUs added by an innovation climbs exponentially with the complexity of the artifact" (at 248).
The lack of a definition perhaps gives me no excuse to be surprised to read that "complex designs are inherently modular" (at 197), but I admit this remark made me shout out loud. If you're familiar with definitions of complexity that rely on some monotonically increasing function of the minimal number of binary digits needed to characterize a system (e.g. Kolmogorov complexity, algorithmic complexity, etc.), you will sense that EB got it totally backwards. By such a measure, a messed up pile of papers is more complex than the same papers organized modularly into file folders, chapters etc. EB cites Herb Simon's 1962 paper "The Architecture of Complexity" as support for his comment. Simon's notion of "complexity" was based on his deep interest in hierarchy in organizations; more hierarchy meant more complexity for him. (See Philip Agre's insightful 2003 critique, available on the Web, for the historical context of Simon's essay.) However, most natural scientists, mathematicians and computer scientists haven't followed Simon in their usage of the term. In fact, even SFI tends to ignore Simon (you won't find him mentioned or cited even once in the 700-plus page SFI conference proceedings volume "Complexity: Metaphors, Models and Reality" (1994), nor in SFI guru Brian Arthur's 1996 seminal "Complexity and Economics" paper) -- leaving EB deserted by his homies on this issue. EB's implicit reliance on Simon while attempting (as I discuss below) to ground his own theory directly in thermodynamics and biological evolution thus sets up a kind of culture clash, if not outright confusion, when it comes to understanding this fundamental bit of terminology.
B. EB's use of the phrase "punctuated equilibrium" seems willfully to depart from the usage of biologists and eventually veer out of control. For biologists, PE is a term of art describing sudden appearances of new species in the fossil record; moreover, PE is not evidence against gradual evolutionary change, which inevitably is imperfectly displayed in rocks. (See, e.g. the book by EB's favorite non-scientist philosopher of evolution, "Darwin's Dangerous Idea" by D. Dennett, at 282-299.) As for EB, after dutifully citing the original PE papers by S. Gould and N. Eldredge, he faults Gould for misusing the term (at 173). I mentioned a minute ago how he finds PE in sources that don't use the term at all. As the book progresses, EB begins to apply the term to situations of alternating stasis and *sudden* change (e.g. at 328-329), and eventually uses it to characterize changes in the *environment*, rather than changes in the things the environment is acting on (at 363).
C. EB defines knowledge as "fit" information, i.e. "information that is useful" (at 317). He goes on: "Evolution is a knowledge-creation machine -- a learning algorithm. Think of all the knowledge embodied in the ingenious designs of the biological world. A grasshopper is an engineering marvel, a storehouse of knowledge of physics, chemistry and biomechanics -- knowledge that is beyond the bounds of current human ability to replicate." (Id.)
Question is, *to whom* is such knowledge "useful"? EB doesn't tell us. Any hints? Well, he never suggests that the goal of all biological evolution is to increase humans' knowledge; and although he insists that biological life is "designed" he also insists that there isn't any designer (see 187-188). So that leaves the grasshopper. Is the leg of a grasshopper useful to it qua information, or qua matter? One could argue that it's useful to the grasshopper as matter, not as information, since it seems unlikely that the grasshopper has any awareness of his leg as information. (Similarly, humans didn't have any awareness of DNA until a bit more than 60 years ago.) But suppose we insist that the leg is useful to the grasshopper as information, and we insist the same regarding DNA for humans in, say, year 1800, even though neither subject would have any awareness of this information. I submit that this stretches the term "knowledge" way beyond all usual usage of the term, leading to the question of *what insight* do we win from EB's peculiar definition? (Smarter people than I can also ponder whether it makes sense for the goal of evolution to be creating something useful for an individual, when genes, not individuals, are the unit of selection (at 279).)
6. The good, the bland and the ugly:
Throughout the book, EB insists that when he speaks about economics, evolution, thermodynamics and such stuff, he is *not* speaking metaphorically. E.g., "Economic wealth and biological wealth are thermodynamically the same sort of phenomena, and not just metaphorically" (at 317). However, at times this no-metaphor commitment seems to paint him into a corner from which the only escape routes are triviality or sticky-footed retreat.
A. Throughout the book, EB develops a theme based on thermodynamics. He frequently repeats the mantra of "matter, energy and information," as in: "A business is a person, or an organized group of people, who transforms matter, energy, and information from one state to another with the goal of making a profit"(at 280), or "[I]n economic systems, entropy reduction requires flows of energy, matter and information" (at 410). But this mantra is a bit like listing "concrete, asphalt and streets" (albeit there isn't any E=mc2 for asphalt and concrete) -- it rests on a category confusion. As Rolf Landauer says in the opening sentence of his landmark 1996 paper on information, energy and computation (reprinted in Leff & Rex, "Maxwell's Demon 2" at 341), "Information is inevitably tied to a physical representation, such as a mark on paper, a hole in a punched card, an electron spin pointing up or down, or a charge present or absent on a capacitor." You can't transform information without also transforming matter or energy somehow.
Too bad EB didn't read that sentence. (EB explains the broad gist of Landauer's paper but cites to a secondary source (at 305 and n. 27).) It might have helped him to rethink his no-metaphors commitment when he made remarks like "[I]rreversibility is a necessary ... condition for value creation" (at 306). When two people swap Babe Ruth and Hank Aaron baseball cards, this is "irreversible in a thermodynamic sense" because the traders "would not want to immediately undo their deal" (id.). (How do we know this? Because we could get a market researcher to survey the traders before the trade (id.). How does EB know the outcome of this survey? He doesn't say.) In later chapters, EB gets more deeply into the mess. In the chapter on strategy, he notes that "truly strategic choices are difficult or costly to reverse once made" (at 325) and that "without irreversibility, there is no wealth creation" (id.). A bit later, after narrating the rise and fall of Westinghouse (at 349), he declares "[O]rganizations carry out thermodynamically irreversible transformations on matter, energy and information, converting high-entropy inputs into lower-energy outputs" (at 352).
Where's the mess? If we accept EB's statements about irreversibility and Hank Aaron literally, the result is trivial. The physical process of swapping cards (including friction as they slide across the table or schoolyard playground), and the sloshing around of molecules in each trader's brain may indeed be physically irreversible. But what insight do we get from that? On this basis, all human activity, not just economic activity, is "irreversible in a thermodynamic sense."
For EB to be making a deeper point, he must be trying to address some different level of organization. He seems to want to treat information -- the trader's preferences -- as something different from matter or energy. But Landauer's comment shows that you can't do that. Moreover, if we ignore the physics and chemistry of the deal, then irreversibility becomes much more questionable. What's so impossible about a trader changing his mind? Does it require energy to do this? If this is literal energy, mazel tov -- we're back to the banal interpretation of "irreversible". And if it's some kind of metaphorical energy, there goes our no-metaphors promise.
Similarly, in what sense is a strategic decision irreversible? In the sense that it takes energy and money to reverse it? But in the literal sense, you need the same to reverse any business decision. The difference is only one of degree, not kind. Once you start to make distinctions such as tactical decisions are reversible and and strategic decisions "irreversible", you're in metaphor territory. Moreover, as EB's fable of Westinghouse shows, value created can easily be turned into value destroyed -- nothing irreversible about that.
B. Finally, what is the origin of wealth? The answer is a bit of an anticlimax: It is knowledge -- only treated as an endogenous variable in a CAS model, rather than as an exogenous variable in an equilibrium (NCE) model (at 317). Even EB finds it hard to warm up to this. The real punch line for him seems to come two paragraphs earlier, when he notes "[T]he fitness function of the economy -- our tastes and preferences -- is fundamentally linked to the fitness function of the biological world -- the replication of genes. **The economy is ultimately a genetic replication strategy**"(at 317, emphasis in original).
Here EB makes good on his commitment not to speak metaphorically. But despite his italics, one again has to wonder: so what? Couldn't the same explanation apply to lots of other human behaviors such as art, music, sports and, per Messrs. Dennett and Dawkins, even religion? What new insight should we get from this revelation about economics? Why should we have expected otherwise? The bigger surprise might be if economics were *not* linked to genetic replication. (BTW I ignore for the moment (i) the question of whether evolution really is only about genes, and (ii) EB's failure to mention anything about evolutionary neutrality in his book.)
I'll end here. There are a host of other things that bug me about this book (ambiguities about fitness landscapes for physical technologies; the omission, contrary to my own limited personal experience as patentee, of intuition as a cognitive process in innovation; garbled discussion of Nash 1950 bargaining paper (see also Gavin Kennedy blog); attribution of realization of business plans solely to "management teams", and analogy between such teams and implanted fertilized mammalian eggs (compare pp. 194 & 236); frequent anthropomorphizings, reifications, shifts in who are the agents and patients of evolution, etc. etc.), but I will bite my tongue. As I've noted, the book does have some true virtues. But surely when someone as sharp as Prof. Gintis calls this the most important book on economics in years, he is using a blurbista's hyperbolic license to help out a pal. At least I hope so -- because if intended at face value, it would be a deeply, deeply sad statement about the economics profession.
Some examples of the unevenness of detail: Several times the author described computer simulations, but did not include enough detail for the reader to fully understand how they were set up. For example, in one simulation of evolution, 1's and 0's were referenced as having been strung together to represent strategies in the "prisoner's dilemma" game. Unfortunately, the author did not bother to explain how the strings of digits corresponded to the strategies; therefore there was not enough information to figure out how the simulated evolution actually worked. In another case, over a page is consumed in an unnecessary blow-by-blow description of how a factory manager might increase production capacity, only to see the demand cycle slowing due to a feedback lag.
A distracting aspect of Beinhocker's presentation is the way that he repeatedly refers to mainstream economics as "Traditional Economics" with a capital T, indicating that it is not progressing, is resistant to change, and is obsolete. Of course, like any science, mainstream economics evolves slowly, but I don't think that a neutral observer could say that economists as a whole have not been open to new ideas over time. "Traditional" economics is actually in the process of absorbing some of the results he relates -- for example those from behavioral economics -- but you wouldn't know it from the text.
There are many statements in the book that seem questionable, but are stated as if they are incontrovertible facts. For example: that punctuated equilibrium is a fully accepted part of standard evolution theory (it's still under debate), that success in evolution can be fully defined by a "fitness function" (success in evolution is ultimately defined by success itself), that the second law of thermodynamics applies to the interactions between people (just because something applies to atoms doesn't mean that it applies to people -- consider the strong nuclear force).
The main part of Beinhocker's thesis is that the economic system can be equated to an evolutionary system, with companies consisting of a set of "businesses" that interact with each other in the economy. Businesses that are "fit" replicate, and those that are not disappear. He creates the concept of a hypothetical, written business plan that corresponds to DNA in evolution. This is a neat concept, but the parallel doesn't quite work. DNA "describes" the organism in a condensed, holistic manner. A business plan fully describing a business would have to be as complicated as the business itself.
In the final section, Beinhocker presents implications of his new complexity-based, evolutionary economics for business and policy.
For business, a major implication is that a company should not try to determine up front what strategy will work, but should set up different parallel businesses with different strategies and see how things go before betting on a single strategy. Ignoring the issue of when a company should stop providing capital to a business, -- What if the "fittest" business takes the longest to succeed? -- this is precisely what is suggested by such strategic frameworks as the BCG matrix, which were developed well before complexity economics. Basically the concept comes down to the old saw of not putting all ones eggs in one basket. It's also an implication of CAPM, "Traditional Finance", that diversification is optimal. (By the way, according to the book, "Traditional Finance" is obsolete also!)
(Beinhocker also finds an "implication" that companies should take a stakeholder approach and favor growth over return to shareholders. I cannot see how this follows from the theory, but I'll let it go.)
In terms of policy, The Origin of Wealth gets really crazy. Somehow Beinhocker comes to the conclusion that Behavioral Psychology favors universal health care coverage and a "minimum living wage." (In all fairness, Beinhocker is relating the proposals of a Matt Miller; however, he strongly supports them also.) The text presents these policies as if they are consistent with Strong Reciprocity, but how can these benefits be reciprocal when they apply to everyone regardless of behavior? (He also conveniently ignores moral hazard, and politicians' strong incentives to manipulate programs to benefit themselves.) It's not so much the conclusions -- e.g., no one can argue that the proposed strong elementary education for all is a bad idea -- but the fact that they don't follow logically from the premises.
The bottom line is that describing economics as an evolutionary system is fine, but there is really nothing that you can draw from the resulting model prescribing any particular strategy or policy. In order to see what will happen to the economy as the result of a new policy, one would have to create a simulation as complex as the economy itself -- in essence one would have to create an entire, new economy; or one could just try the policy and wait and see what happens. On the other hand, "Traditional Economics" gives us such theories as the law of supply and demand and interest rate parity. The law of supply and demand may not be perfect, but at least it predicts that a business will reduce sales if it raises prices. Interest rate parity isn't perfect, but at least it predicts that if a central bank raises rates, it should see a strengthening of its currency. Don't give up on mainstream, "Traditional," economics yet!
(If you are considering this book, you should check out "More Than You Know: Finding Financial Wisdom in Unconventional Places" by Michael Mauboussin, which applies some of the same theories of science and psychology to investing; and "Knowledge and the Wealth of Nations: A Story of Economic Discovery" by David Warsh, which presents a history of the development of mainstream growth theory.)
Most recent customer reviews
I must admit that it is just about the best book that I have ever read on the issues/pitfalls of neoclassical...Read more