Fables of the Reconstruction


In just 160 pages of their new book The Fables of the Keiretsu, Tokyo University economics professor Yoshiro Miwa and Harvard Law School professor J. Mark Ramseyer manage to raise doubts about almost all conventional wisdom regarding the structure of the Japanese economy.

A summary of their arguments in Q&A form:

Are Japanese firms organized into informal industrial groups called keiretsu?

Keirestu have never existed, and the concept was invented by Marxists at the Economic Research Institute who wanted to identify a source of “monopoly capital” for the Japanese market.

Do firms in keiretsu arrangements cross-hold other firms stocks, therefore allowing them to pressure each other towards beneficial relationships?

Not really.

Was the pre-war economy controlled by zaibatsu financial cliques?

These firms were only deemed “zaibatsu” because they were successful in the market when other companies were struggling through the Depression.

Do Japanese firms have a “main bank” which will rescue struggling companies even without the explicit contracts to do so?

There is no such thing as a “main bank,” and banks do not automatically bail out their clients.

Is the Japanese economy directed by the soft authoritarian guidance of the central bureaucracy (especially MITI)?

Japanese firms ignore MITI’s direction, and the courts back these firms up when the bureaucracy gets decides to press them.

For the most part, Miwa and Ramseyer supply an abundance of data and objective measures to show that the Japanese economy has never really operated in the “unique” way depicted by fifty years of mainstream scholarship in both Japan and abroad. In the case of keiretsu, they make a very strong case that the characteristic cross-shareholding is actually very rare, the so-called “lunch clubs” are rarely used for group decision-making, and most parts suppliers working under automobile companies make products for a large number of different rivals — not just their supposed keiretsu “parent.” The authors re-analyze the Sumitomo Metals case study, which is supposed to show that MITI punished Sumitomo for refusing to enter into a production cartel. Instead they find that not only did Sumitomo freely challenge MITI’s guidance but got away with it scotch-free.

I find Miwa and Ramseyer’s arguments convincing for the most part, but as someone previously partial to the idea of the Japanese economy being a “developmental state” model rather than a complete “free market,” I would like to see a heavyweight on the other side of the fence argue the opposing view. They are most certainly right about the narrow topics they picked up, but perhaps end up overselling their bigger conclusions. Instead of just correcting the record, they are obsessed with aggressively belittling a certain kind of cultural relativistic economic view where Japanese firms “sacrifice profits for the sake of cultural norms” etc. Chalmers Johnson’s theories on MITI guiding the Japanese economy could maybe benefit from more attention to hard data, but Miwa and Ramseyer seem less interested in dealing with scholars like Johnson and instead channel their wrath towards a more extreme and less existent animal:

“East is east, and west is west, and never the twain shall meet, till earth and sky stand presently at God’s great judgement seat.” As politically incorrect as Sylvester “Rambo” Stallone and Mae “Peal-Me-a-Grape” West, Kipling nonetheless remained a cultural relativist to the end. Dumb down his verse six levels and it captures most of what passes for “theory” among modern cultural relativists and much of what passes for analaysis about Japan. And stripped of their political baggage, the modern cultural relativist and the old-school colonialists like Kipling fascinate for the same reason: they indulge our lust for the exotic and free us from the rules of social science (156).

The authors’ main methodological advice is praiseworthy: Economic theories about certain institutional systems need to incorporate inductive data and match behavior to pre-existing universal understanding before assigning behavioral choices to the black box of “culture.”

That being said, Miwa and Ramseyer show little interest in acknowledging the much bigger question that emerges in light of their refreshed view: Aren’t there many places in the Japanese economic and consumer structures that are drastically different than what is seen in the other mature capitalist economies? Maybe MITI does not completely control industrial policy in Japan, but did the American government ever proclaim that there should be only two car companies and that they would not assist new entrants? Honda, of course, ended up ignoring MITI’s warnings about manufacturing cars and still went on to unbridled success. But I think it still says something about the nature of the Japanese government that it believes it has the power to shape the industrial marketplace, to allow cartels, to essentially not ever prosecute monopolies and oligopolies. This could be a product of historical circumstances (rebuilding after a war, late development, etc.), idiosyncratic decision-making, or ingrained cultural traditions, but Miwa and Ramseyer leave themselves open for attack by failing to mention that there are, at the end of the day, still differences and points of divergence.

The authors maintain that economic actors tend to follow the same logic all over the world, but if we accept this on general terms, it means that differences in output must thus depend on particular market arrangements. On this blog, I have made claims in this vein. For example, Japanese pop music sounds like it does because of the dominance of oligopolistic artist management companies like Johnny’s Jimusho that continually use their market power to crush rivals and keep tastes stable. Johnny’s behavior is no different than what can be expected among firms in the West, but American music companies, for example, never are able to achieve the same kind of economic power over the media to be able to sustain such actions.

What I want to know after reading The Fable of the Keiretsu is, what are the structures of the economy, the government decisions, the particular historical circumstances that are different than what are seen all over the world? Even if keiretsu do not exist, why do you see so much tolerance and/or esteem for monopoly and centralization? This may not be “Confucian” per se, but the sources of these kinds of macro behavior still need to be clarified.

So yes, let us clear the dead weight of myth and the laziness of evoking deep-seeded cultural voodoo. But once we have gone back to the data and put our knees on the ground, we then still have to ask, why and how is a set of universal reactions being shaped and altered to produce a particular result?

W. David MARX (Marxy)
February 12, 2007

Marxy wrote a lot of essays back on his old site Néomarxisme. This is one of them.

16 Responses

  1. Chuckles Says:

    Props for giving this the attention it deserves. It is mostly hot air, as both scholars are adept at perforating strawmen. The fact of the matter is that there are real structures on ground that resemble the Keiretsu, and attacking overgeneralizations is hardly a substitute for scholarly research: One doesnt produce a book simply to argue that, well, not all African governments are *that* corrupt and call it scholarly. The ideological biases of the authors are pretty clear: Like the Old New Revisionists who overplayed the cultural angle, these folks are pulling in another extreme.
    First: Miwa clearly has as his target, Masahiko Aoki. It is pretty much obvious that they are charging him with positions that he has never taken. That a main bank does not exist, Godfather like, is a strawman. Aoki never claimed this. In fact, IIRC, his argument about relational financing is explicitly built on a theory of incentives for banks to enter into such relationships.
    Second: Miwa and Ramseyer simply oversell the case for the nonexistence of Keiretsu. Claiming that Marxists invented Keiretsu is like claiming Marxists invented the Upper Class.
    Third: As Milhaupt points out here:
    there is widespread data contradicting Miwa and Ramseyer. This is simply empirical – and not something to be overlooked for the sake of shattering CW. Furthermore, Milhaupt shows, how relatively easy it is to construe the data cited by Miwa and Ramseyer as being supportive of Aokis thesis.
    Fourth: It is not at all clear that general accounts of the Sumitomo Metals incident evince the dictatorial hand of MITI. To put things in perspective, the Sumitomo incident was hardly the norm: Why cite it as evidence for a thesis that purports to prove a general independence of Japanese firms from MITI? You will notice that in several papers, M&R use trimuphalist language in describing Sumitomos purported victory over MITI, yet describe similar cases: Nisshin Spinning and Idemitsu Kosan: as a case of MITI negotiating with said corporations instead of coercion by fiat. Why is this important? It is important because when discussing Sumitomo they raise a strawman and demolish it; but other key cases illustrate precisely what Aoki has severally conceded: That export based companies hardly needed MITIs guidance anyway and that MITIs strategy is less one of soft authoritarianism and more one of consultation.
    The notion of State directed industrialization is over interpreted by M&R – the examples cited by proponents of this reality illustrate clearly that they are not referring to the Godfatherlike Mafia setup M&R are flailing against.
    Indeed, their own rereading of the Sumitomo incident suggests a culturalist dynamic – the face saving act of MITI – demonstrated in the press releases – what was all that for, if not an identification with the cartel? Is the social psychology of emphasizing centralization now to be suddenly deemed unimportant to the Japanese corporate landscape?

  2. James Says:

    I’m in the process of reading the same book, and while I find it to be very interesting and informative, I also think that it’s a bit too much to simply declare western perceptions of the Japanese style of business as mere fantasy. Sure, there might not be the formal keirestu structures with membership lists or secret luncheons for business planning, but there is certainly something different about the way many businesses function here [compared with the USA]. As you’ve said, monopolies are pretty much the accepted norm for many industries here, and while they might be profitable businesses that don’t require propping up from Keiretsu partner banks and keiretsu stock-buying schemes, there must be some explanation for it [which the authors do not mention].

  3. marxy Says:

    Thanks a lot for posting all of that. I respect the idea of going back into the data to re-examine all of our preconceptions about the Japanese market, but yeah, I felt like there was some relatively incomplete about the conclusions.

  4. marxy Says:

    “I also think that it’s a bit too much to simply declare western perceptions of the Japanese style of business as mere fantasy.”

    They also don’t sufficiently deal with the fact that the keiretsu idea has become widespread in Japanese scholarship as well.

  5. TJJ Says:

    “There is no such thing as a “main bank,” and banks do not automatically bail out their clients.”


    “There is no sexual relationship between Ms. Lewinsky and I”

    Sure, there isn’t NOW, but what about in the past. What if it wasn’t one single main bank but a collection of banks? What if they didn’t ‘automatically’ bail out clients, but would do it 95% of the time?

    Mythbusters they are not.

  6. Alex Says:

    Ramseyer seems to make a living out of mocking Japanenese contructivists, brushing them off as so attached to the uniqueness of Japanese culture and history that they miss cases where well-established theory fits perfectly well to Japan. I haven’t had a chance to read this book yet, but guessing that the conclusions are futher in line with Ramseyer’s thought, I’d imagine he certainly makes some good points, but also works so hard proving the aforementioned constructivists wrong that he takes his reasonings to the extreme, ignoring well documented facts and social trends.

    The fact of the matter is, the developmental state was, at least before the late 1980s, a very real entity, and keiretsu relationships, often highly supported by MITI policy, were huge structural impediments to both trade and investment coming into Japan. So many people argue Japan as different from Europe and America because of well-documented reasons. Nevertheless, Ramseyer very often brings up extremely interesting arguments in his work, constantly putting checks and balances on those that would overemphasize Japan’s distinctive role in the world, yet it’s important to read with a very critical eye.

  7. Brian Says:


    I have some questions i’d like to ask you regarding japanese indie labels, what’s the best way of getting the Q’s privately to you?

  8. marxy Says:

    “I’d imagine he certainly makes some good points, but also works so hard proving the aforementioned constructivists wrong that he takes his reasonings to the extreme, ignoring well documented facts and social trends.”

    You nailed it without even reading it.

    I don’t think Ramseyer is necessarily a bad force in the world, but rather than just broading or redefining the discussion, he seems to just mock everyone else pretty aggressively. The tone is very harsh.

    “I have some questions i’d like to ask you regarding japanese indie labels, what’s the best way of getting the Q’s privately to you?”

    marxy (at) neomarxisme.com

  9. Chris_B Says:

    Well now, this looks worth checking, if nothing else for the fact that even if they may not be right, they may not be wrong either.

  10. Mulboyne Says:

    You can download for no charge their earlier 2001 paper on the myth of the keiretsu from here:


    They ask some fair questions. People are certainly quite free with their usage of the term because they assume every knows what they are talking about. During the internet bubble, when everyone in the US seemed to be financing everyone else, there was talk of the power of cross shareholdings and the emergence of the Microsoft keiretsu, Amazon keiretsu etc. However, if all keiretsu means is “industrial group” then it doesn’t really explain very much.

    By most definitions, Softbank’s group of companies qualifies as a keiretsu but there certainly seems something different about their business arrangements compared with, say, Mitsui. The Taiyo group also looks like a keiretsu and yet founder Kuniyasu Sakai rails against that kind of organization in his books.

    If the authors’ manage to stop people reaching unthinkingly for an ill-defined term to explain some aspect of Japanese corporate behaviour then they will have done some useful work.

    You touch on one of the areas they leave out, though. Whatever, the origins of the term and the truth of the concept, it seems clear that Japanese managers often think they are part of keiretsu and Japanese customers, whether as individuals or organizations, will stop doing business with one company because of a scandal at a keiretsu relative. It’s as if once people hear their behaviour named and described, they say “Yes, that’s it” and then proceed to adjust their actions so they more closely fit the description.

  11. marxy Says:

    I was frustrated by the fact that they proclaim “there are no keiretsu” as in the mythical 100-company strong groups. But are there not smaller units of maybe 20-40 companies that do have the kind of ties that would make them an “industrial group.” They never ask this question. They just debunk the wilder theories of massive economy-wide oligopolies.

    There’s the other question of whether Japan has more vertically integrated huge conglomerates than other economies. They may not, but it’s something worth perhaps bringing up. Sumitomo and Seibu etc. sure own a lot of different things in different industries…

    Instead of limiting their book to a mere 160 pages, they should have added on 50 more to explain all the obvious follow-ups.

  12. michael Says:

    this should have been called ‘deconstruction of the fables’–yet,
    it does seem a little ridiculous: no keiretsu? merely an invention
    of marxist economists? now that’s some revisionist history for you!

  13. Chris_B Says:

    “It’s as if once people hear their behaviour named and described, they say “Yes, that’s it” and then proceed to adjust their actions so they more closely fit the description.”

    The network effect (?) has amazing unintended consiquences. Perhaps this goes along way towards explaining what seem to be very conservative business decisions which turn out to be penny wise and pound foolish. Must chew on this a bit more.

  14. Aceface Says:

    I’m NOW reading my decade old 積ん読 book “The Weight of the Yen” by R.Taggart Murphy.(japanese trans.)and I’m in no position of telling what’s right or wrong about the latest work on the Japanese economy of which I have’nt even read.
    But two things.

    East is East:
    “As politically incorrect as Sylvester “Rambo” Stallone and Mae “Peal-Me-a-Grape” West, Kipling nonetheless remained a cultural relativist to the end. ”
    Why do people still keep misquote Kipling since the rest of the verse goes:
    “But there is neither East nor West,Border,nor Breed,nor Birth,When two strong men stand face to face,tho’they come from the end of the earth!”
    He could have been a sexist though.

    No keiretsu:
    Any of you have ever attended the funeral of a Japanese salaryman?
    I have.
    The latest was that of my father held last October who died at the age of 64.

    As soon as my father died we phoned a lot of numbers including that of my father’s office of which he retired a year before.The guy who took the phone told me “We’ll call Nikkei for the obituary right away”.And next day there was a little piece showed up in the downside of the paper.
    That attracted many flowers sent by ex-business associates with name tags of the company and the guy who sent it along with his title in the office.
    Didn’t know what to do with them,until three of my father’s former colleague showed up and helped my mother and I with funeral desk works.
    Since the departed had worked in three companies at different times(two of them are keiretsu of the first one),I couldn’t tell who was who and which was which.But the people from my father’s office knew exactly what they have to do.They started to set all the flowers surrounding the coffin exactly in order by the company rank of the keiretsu.

    A life of the salaryman,in the midst of the keiretsu hierarchy even at his deathbed.

    No keiretsu?I don’t believe it.

  15. Mulboyne Says:

    The authors say “at root, the keiretsu do not exist”. This is an extreme claim and I’d be surprised if they really sought to deny that business groups and conglomerates existed in Japan.

    As Chuckles noted, their real target is those who make claims about how keiretsu relations influence capital allocation, industrial policy, competition etc. They believe a number of characteristics of the keiretsu – which are virtually received opinion – don’t actually exist. The also believe that you can explain a lot of corporate behaviour without needing the concept of keiretsu to do any heavy lifting.

    My own observations of seeing a “parent” with less than a 5% stake regularly send people into a group company and effectively determine its destiny makes me sceptical of anyone who denies these kinds of ties exist. I do think that people make a grave mistake in thinking that when they see a keiretsu, there will naturally follow a fixed pattern of behaviour.

  16. Ken Says:

    Given that MITI ceased to exist in 2001, I wonder when the book was published. Also, along with what Mulboyne said above, it’s hard to deny that such business groups exist in Japan when not only are cartels legal under certain circumstances, they are and have been approved by the relevant government ministries.