“After ten years they let you cook the eggs…”

That line was from Jiro Dreams of Sushi.

Think of this film as a learning by doing model.  Bosses would like to invest in training workers, but they fear the workers will leave them high and dry, unable to recoup their investments.  Bosses therefore train workers excessively slowly, keeping them as apprentices in the meantime.  Only in the end stages of training do the workers learn how to handle the high-margin items, namely the sushi itself.  Furthermore Japanese customers demand high quality, which make it difficult for an incompletely trained worker to open his own sushi bar.  As long as there are many very good sushi bars, this equilibrium with well-informed customers can persist and sustain long-term worker training.  Quality is inefficiently high, and productivity in the service sector is inefficiently low, while personal service quality is inefficiently high (let him greet and bow to customers before he learns how to shape the rice), but training occurs and the elderly retain lots of social and economic bargaining power.

Young workers earn not so much, but can cash in on equity (i.e., open their own sushi bar) later in their lives.  They are not promising marriage prospects for young women.

Imagine a shock which limits the future profitability of sushi bars, such as fish depletion or greater competition from foreign foods or from cheaper sushi produced by lower-skilled workers.  This will shift the composition of apprentices toward somewhat older individuals, and indeed the movie suggests this has happened under Jiro.

Jiro: “I have been able to keep at the same line of work for seventy-five years.”  The viewer does not expect anyone else in the movie to be making the same claim, years from now.  In the meantime, such an economy is not good at reallocating labor in response to sectoral shifts.

At age 85 Jiro holds three Michelin stars, although his restaurant has only ten seats and the bathroom is outside and down the hall.

They serve slightly smaller portions to the female customers, so that everyone in a party finishes their portion at more or less the same time.

Addendum: The new “SushiBot” makes 3,600 pieces of sushi an hour, albeit at lower quality.


" Quality is inefficiently high, and productivity in the service sector is inefficiently low"

It is amazing to me how you "market efficiency" guys just feel that you can spout off about anything and everything through your ideological prism.

Are there government controls requiring this type of trianing? Are there "unions" or trade groups that limit market forces? Is there some actual study that has been done that supports your views that you are not citing? What is your basis for claiming that this is somehow "inefficient" and not otherwise based upon the very peculiar tastes of Japanese diners that demand this suprising level of quality?

You conclusion smacks of market distortion but there is no sign of it in your reasoning.

It may astonish you to learn that mainstream economics does, in fact, have a concept that private nonstate incentives can lead to inefficient outcomes.

Can't people just express opinions anymore?

"Hey Tyler, are you shorting the Japan Sushi Index? No? Then you really don't believe that!!!"

You may wish to google things like "externalities".

I did some work for the owner of two Manhattan Japanese restaurants some years ago. In the course of this work I got a look at his tax returns which managed to show essentially zero profits while nevertheless not really paying anybody very much. This despite a fairly elegant house in a fairly elegant suburb.

A year or so later I spoke to his lawyer and asked how the restaurants were doing. He said well, but the former owner had sold them. "How could he, I asked, when they showed no profits?" Oh, he didn't really sell them... he "gave" them to his chef. Ah, the joys of hidden equity and hidden capital payments and thinly disguised indentured servitude.

The whole society is shifting to a somewhat older composition, too.

Let a thousand sushi bots bloom.

Are you able to watch this movie anywhere?

It's opened in Philadelphia-- try tracking down the website.

Kew Gardens Cinema's in kew gardens NY, NY 11415 (I watched it there yesterday)
Most fascinating thing about the movie is that 95% of what makes sushi good is the prep work. Actually putting the sushi on the rice and serving is really just showmanship. Preparing the rice, preparing the fish, obtaining the right quality fish, is what makes the sushi good and requires the most skill and the apprentices do that. Jiro, does not. Jiro checks quality and is the "face" of the restaurant when he serves the sushi. He teaches all of techniques to his apprentices, and most of the technique is simply experience: how long to cook something, when to turn something, how long to marinate, what temp. to hold the sushi at, it's all done by experience and "feel".

Frankly, this seems like a triumph of marketing, in the way that sushi chefs have trained Japanese audiences to perceive such subtle quality differences in the slicing of raw fish.

“SushiBot” makes 3,600 pieces of sushi an hour, albeit at lower quality.

Is there an evidence that they are of lower quality? Chances are high that in a double blind test it would be discovered that it does either equal or better job. It's probably the snobbish bullshit along the same lines as more expensive wine = better wine or screw cap = low quality wine.

Not sure that "chances are high." Try "chances are completely unknown that...insert assertion."

I'm feeling you on the wine, but fund a blinded study of sushi and we'll talk. Crazy thing about the scientific method: hypotheses are hypotheses until...

Actually, no. The priors in this case are very obviously "if it looks like a duck, and quacks like a duck....". It's just a roll, you know. So my bet would be on very high chances.

...and especially on a price-normalized basis. According to Jiro's website their Sush meal costs around $300.

If it woiks, don't fix it: Let the inefficionistas open their own sushi bars... .

Its already happened. At least in the US, there are no shortage of sushi places ran by (often non-Japanese) people who were never trained by real sushi chefs. While anyone is free to mock the sushi these places produce, the sheer number of them suggest that they probably dominate the industry in both volume and revenue.

Hell, looking at the prices this guy charges and his volume, I wouldn't be surprised if the average American supermarket makes more profit from sushi than this guy.

Re: ” Quality is inefficiently high, and productivity in the service sector is inefficiently low”

One of the things I noticed when my wife became pregnant is that in the U.S., you are strictly advised against consuming any raw fish when pregnant. In Japan, there is no such prohibition. I thought this might be a cultural thing, but it turns out listeriosis is a major concern when consuming sushi in the U.S., while it is essentially unheard of in Japan.

I suppose one could argue that the occasional food poisoning is a reasonable tradeoff for lower cost, but what's not clear is that the U.S. lies at the point of greatest efficiency between Japan or, say, India.

Maybe if it wasn't a little bit dangerous Americans would be less interested.

Listeriosis rates have been falling in the US (apart from the recent deadly cantaloupes), and Japan apparently doesn't track them. There have been some studies showing that listeriosis infection in Japanese retail fish is common (like 10%), so maybe there's some genetic resistance protecting the Japanese

Quality is inefficiently high
This implies a very narrow view of efficiency — perhaps to be expected from an economist. There is more to life than mere economic efficiency. In a society that prizes quality the texture of life is different in a way that is hard to convey with words. Can't you kindly let the Japanese keep their quality? But no, you must strong-arm and wheedle them into TPP (taken over by the US as a handy instrument of power from a handful of tiny countries like Peru), destroying half the Japanese agriculture and accepting the American way in health care, insurance and finance, and the sovereignty of American IP, trade and corporate law.
productivity in the service sector is inefficiently low
People can consume only so much services, there's just twenty-four hours in a day and people have to sleep too. Suppose productivity in the service sector improved by 100%. Half the workers there are fired and what do they do? Solder together iphones? The Chinese do that cheaper. Get a degree in puppetry? We see how that works here. Those that can get a useful degree are already doing it. So where do they go? Right, on the dole. What for? Beats me, but apparently it is efficient and increases productivity.

Are you following the lump of labor fallacy?

No, I am not. Subsistence can employ any number of people (this trivially refutes the fallacy). Ultimately all people will be employed, even if their work is being an object of care for the various social services. I am saying that increasing productivity by means of degrading people into the underclass may be efficient for some measures of efficiency, but in reality it destroys pieces of society and converts them into economic output. The process looks efficient just because we have no good way of valuing the destroyed pieces, or the desire to do so.

"Ultimately all people will be employed, even if their work is being an object of care for the various social services."

This sounds like the concept of "tax expenditures."

Yeah, but I feel it's less enlightening in this discussion to express the phenomenon as 'tax expenditures' seeing as we're talking about employment and productivity.

I think he means that the quality is higher than the customers need because it takes you 10 years before you are allowed to touch the eggs. I was cooking my own eggs at age 5.

the quality is higher than the customers need
Aren't the customers the best judge of that? I heard there is this thing called 'market'. You know, where people can buy what they want and not what somebody thinks they need.
I was cooking my own eggs at age 5
And running a sushi shop at ten?

I think that's begging the question, right? Are the customers demanding that quality or is the nature of the apprenticeship pushing it on them.

It's the same thing I see in academic research and engineering education. The market is being dictated to because they are inefficient from the perspective of the market itself, not because of my opinion.

No, I wasn't running a sushi shop by 10. I'd have to be at least 15 before I could have run a sushi shop by Japanese custom.

Do you hear many complaints from the Japanese that the quality of sushi, or anything else for that matter, is too high? For recent news about quality of academic research see this article (originally in Nature). As for engineers, I hear (ok read) many complaints that the quality of engineers is too low, at least in IT, not that it is too high. The quantity of high-quality engineers may be too low, but that is a slightly different problem.

Training in IT isn't anything like training a sushi chef. It is not uncommon to get a entry level job (internship) while in college, and it is not uncommon for interns to have root passwords to production machines. If the sushi industry were ran in this fashion, you would expect apprentices with 2 years of training to be doing everything important. Things like greeting the customer would be left to minimum wage workers.

I wasn't comparing them, I was replying to Andrew's remarks about engineering education. OTOH your comparison is apples to oranges. The interns that are given root passwords to production machines have probably been hacking since primary school, that's easily 10 years' worth of experience (even if it is of an irregular and informal quality). The ones that have only their bachelor's degree in their backpack normally won't get within a mile of production machines, or production code for that matter.

BTW things like greeting the customer are often left to minimum wage workers — part-timing high-schoolers, students and freeters.

I suspect that people who this sushi chef takes on as apprentices have cooked since primary school too. You have to have a serious love of cooking to be willing to sign up for this kind of thing, especially considering that the pay is probably quite low as well. (10 seats times 300 dollars each means 3000 dollars per day, which really isn't very much to pay their staff with.

I don't get your argument? Are you arguing against productivity increases in general?

I don't think Tyler's remarks can be read as arguing against productivity increases in general. As for my argument above, I am not arguing against productivity increases in general either. That target is too big for me to argue against. I can't say I am arguing it, but I do feel that a) there appear to be diminishing societal returns to productivity, b) that we are at a point where these diminishing returns have started to kick in, and c) that part of the mechanism by which our society is offsetting these diminishing returns is the consumption of social capital by means of converting it into economic output. We are cutting off the parts that are no longer productive enough and leave them to rot, sometimes literally.

The remark that still confuses me is what you mean by the "consumption of social capital"? How do we use it to offset diminishing returns?

I think high productivity is a mis-guided target of your ire. What might be lacking are the social nets / transfer payments to cushion the sub-productive cohorts.

The phenomenon I label “consumption of social capital” is described e.g. in Charles Murray's recent book. Certainly there was more social capital in Fishtowns 50 years ago than there is now, so where and why did it go? I think his idea is that social capital was supported and buttressed against natural decay by gainful employment in a way that welfare cannot exactly duplicate (empirical observation). This meant that resources were being transferred, through a very long chain which I can only dimly imagine, from production into social capital, and it certainly was economically inefficient. Now we have cut this chain and increased efficiency, but social capital no longer has the external support from production, so it decays. But it decays in places and strata of society that are easy to ignore, for various reasons, so we do.

I think high productivity is a mis-guided target of your ire.
Maybe it is. After all, I am myself a part of high productivity. But I want to invite you to think where the quest for high productivity ends. Suppose productivity grew so much that material goods can be distributed free. What do the people do with their lives? I am reminded of "The Fence" by Simak on the one hand and "The Final Circle of Paradise" by Strugackis on the other.

What might be lacking are the social nets / transfer payments to cushion the sub-productive cohorts.
We have been at this for quite a bit of time, haven't we? Transfer payments cover only the economic aspect. Workfare might be better if people didn't know it was workfare, but the sub-productive cohorts are not intelligent enough to disbelieve the evidence of their own eyes. Segregation and isolation would be called for.

So, one might say the Japanese put the fish in inefficient?

the Japanese put the fish in inefficient

Does not parse.

Its a pun, read it outloud.

The idea of serving smaller portions to the lady customers, so that everyone finishes at the same time, is a good one.

I'm not sure. You rarely see a woman who eats as fast as a man.

I want to point out that the sushi bot can make top quality sushi. As the movie points out, 95% of what makes the sushi top quality is the prep work. the last 5% is presentation. If you give the sushibot top quality rice, prepared perfectly (as they show in the movie) and top quality fish as shown in the movie, then the sushibot can make top quality sushi. The bottleneck in the sushi production is not the actual making of the rolls and pieces, the bottleneck for top quality sushi production is the limited supply of top quality ingredients, and how long it takes to prep everything. From the movie, what makes sushi good is how the chef is able to soften firm meat, and firm up soft meat, how much to marinate, etc etc, it's all about taste.

Interestingly in New York City, Bloomberg has made it almost impossible to have top quality sushi ala a Jiro type establishment due to the strict rules the NYC health department has on the allowable temperatures to hold food at (140degF). A top quality sushi restaurant in NYC should not have a sanitary score higher than a B. Notice, in the movie Jiro's restaurant does not have those mini refrigerators at the bar that hold the slabs of fish.

I love you guys!!!

You do some poorly thought out rant about sushi chefs and immediately hide behind "externalities" when your logic and relavancy is challenged. However you blindly defend corporate America against any attempts by government to deal with the externalities created by our clearly inefficient system of corporate subsidy through the universal grant of limtied liability for any business enterprise no matter how dangerous or damaging to the environment or anything else.

California has both wall-to-wall sushi schools and sushi restaurants. I'm not sure that has meant "better average sushi" or even "better best sushi" for all the economic efficiency. Which are we after here ;-), good food or many restaurants?

We are after economic efficiency ahd high productivity.

Best addendum ever.

Comments for this post are closed