Saturday, October 18, 2014

Korea's Labor Productivity, and How to Interpret Data About Korea

An article titled Seven Reasons Why Korea Has the Worst Productivity in the OECD, from March 2014, has been recently making rounds in TK's Facebook feed again. It was a dumb article at the time of the publication, and it remains dumb today. Regardless, the article continues to receive approving reactions--which merits pointing out exactly what is dumb about this article.

First, the article itself. The author Michael Kocken, writing for Business Korea magazine, begins with this:
Korea was recently named the worst place for worker productivity in the OECD, which was featured in a recent article by this magazine. This news is not surprising for any professional previously or currently working in Korea, as the notorious overtime hours coupled with years of low growth have been a widely-discussed issue over the past few years.
Then the article makes the familiar, banal complaints about Korea's corporate culture:  Korea's corporate structure is too rigid and hierarchical; there is no honest and direct communication; worker distraction from the Internet and smartphones; hungover workers, valuing form over substance, new workers who are poorly equipped, and the need to put in useless "face time."

Typical office scene in Korea. Is this the home of low productivity?

What's dumb about this article?

First, the article's starting premise is flatly untrue. Korea's labor productivity was not the worst in the OECD. Korea's labor productivity per worker in 2012 (which was the most recent data available as of the article's writing) was at 23rd place among the 34 OECD member states. Sure, 23 out of 34 is still in the lower range. But it is a far cry from being at the worst place.

But let's be generous and make an ample allowance between the bottom third and the rock bottom. After all, it would be good for Korea to aspire to be on the above-average side of the OECD. However, even this allowance cannot save this article. The main problem with the article is that the author does not seem to understand what "labor productivity" means. This is apparent from the second sentence of the article's opening paragraph, which refers to Korea's long overtime hours. Even setting aside the factual inaccuracy that TK noted earlier, this is a strange statement.

Why is it strange? Because OECD measures labor productivity by, essentially, dividing "output" by number of hours worked. (The precise methodology is somewhat more complicated, especially on how one defines "output." If you are interested in the actual methodology, you can find it here.) This necessarily means that the longer one works, the lower the labor productivity, because if you increase the denominator while holding the numerator at the same level, the result is always a smaller number. In other words, Korea's labor productivity is low because of long overtime hours, not despite the overtime, as Kocken appears to imply.

This leads us to the most important lesson:  what OECD means by "labor productivity" is not what an ordinary person would think. When OECD states Korea has low labor productivity, the word "productivity" is not being used in the same manner in which regular people talk about being "productive at work." But the latter is exactly how the author Michael Kocken uses the term "productivity." Then the article simply runs with the incorrect understanding of the term, and make the trite, stereotypical complaints about Korea's corporate culture.

(More after the jump.)

Got a question or a comment for the Korean? Email away at

Well then. If Korea's low labor productivity doesn't mean what a regular person may think it means, what does it actually show?

Korea's low labor productivity shows that the value creation in Korea's economy is heavily concentrated. The OECD report that focuses on Korea makes this very clear (at pp. 4-5): Korea's labor productivity is excellent in areas for which Korea is known, and very poor in other areas. Specifically, Korea's labor productivity in manufacturing is world-class, while labor productivity in service industry is below par. (As of 2010, labor productivity in Korea's manufacturing sector ranked second in the OECD.) In fact, labor productivity in Korea's service industry is only half as much as that of Korea's manufacturing industry. This ends up dragging down the whole of Korea's labor productivity per worker, because over 70 percent of Korean workers work in the service industry.

Why is the labor productivity in Korea's service industry so low? There are largely two reasons. First, unlike Korea's manufacturing industry, Korea's service industry does not compete in the global market. For example, Korea's legal industry was a closed market until very recently; foreign law firms will not be able to fully operate in Korea until 2017. Same is true for Korea's medical industry, whose market will not open to international hospitals until much later, if it ever does. Absent world-class competition, it is not a surprise that Korea's service industry does not have world-class labor productivity.

Second reason is that the Korean economy has an inordinately huge number of mom-and-pop shops, which technically belong to service industry. One of TK's favorite Korea trivia is:  Korea has 12.2 restaurants per 1,000 people, whereas Japan only has 5.7 restaurants per 1,000 people and the United States has 1.8. As much as Koreans love their food, gastronomical pleasure alone does not explain why Korea has nearly seven times more restaurants per capita than Americans do.

Because Korea has had a weak social security network, few Koreans could truly afford to "retire"--that is, take their savings, add them with governmental assistance, and stop working. Instead, upon "retiring" from their initial career, most Koreans have to take their savings and open up their own business to generate enough money to get through old age. This leads to a proliferation of restaurants, a business with pretty low initial barrier to entry. In particular, franchise restaurants--which are very easy to open without much experience--makes the barrier to entry even lower. (By the way, this is a large part of the reason why Korea has so many different fried chicken restaurants everywhere.)

And of course, this trend is hardly limited to restaurants: as of 2011, 28.2 percent of all Korea's workers were self-employed, which is a huge proportion compared to the OECD average of 16.1 percent. (In fact, the same number for Korea in 2010 was eye-popping 36.8 percent.) Such mom-and-pop shops naturally have low labor productivity, because the hours tend to be long and the value created tend to be small. And there are far too many of them in the Korean service industry.

The low labor productivity of the mom-and-pop shops weighs down the labor productivity of the entire service industry, which in turn weighs down the overall labor productivity of the Korean economy. This is arguably the most important reason why Korea tends to have low labor productivity.

*                   *                   *

The lesson here is simple: be diligent. If you are going to write about data regarding Korea, actually look at the data. Let the data speak, instead of inserting your own bias.

As shown above, OECD's labor productivity has nothing to do with Korea's corporate culture. If Korea's corporate culture were dragging down Korea's labor productivity, one should expect to see that drag across all sectors. But data says that is not true. Korea's manufacturing sector has world-beating labor productivity--and it is not as if Korea's manufacturing giants (e.g. Samsung, Hyundai, LG) are known for their touchy-feely lack of hierarchy. Having accurate knowledge about what labor productivity indicates, leads to the accurate conclusion:  if Korea is good at something, it has excellent labor productivity in that area; if Korea is not as good, its labor productivity in that area is poor. Simple as that.

Do you think Korean work culture is too hierarchical, to Korean economy's detriment? Fine, make that argument. TK is skeptical of the claim that the corporate culture which created a world-beating manufacturing sector can be a problem somehow, but he is willing to entertain a good argument to the contrary. Find actual data that supports your contention, and we will have a debate. But don't bring in irrelevant data because it kind of sounds like the thing you are going for--because that's just lazy, and laziness is what often leads to dumb articles like this one.

Got a question or a comment for the Korean? Email away at


  1. There is one sense in which the corporate culture is responsible for low productivity - corporations need to increase their mandatory retirement age. 55-56 is far too young, and pensions - as you state - are far too meager.

  2. The poorly-written article does contain some valid points, but unfortunately is not based on a clear development of how these points relate to productivity.

    In the case of labor-intensive manufacturing, organizational execution, rather than individual brilliance, tends to hold more importance, so the top-down hierarchical structure of Korean business may have actually helped productivity more than harming it (one may also argue that the hierarchical business culture in Korea is partially a result of the manufacturing-oriented Korean economy).

    In certain knowledge-intensive industries where the weight of importance shifts more toward individual competence than organization management, a top-down hierarchy becomes less effective. In the case of software development, especially, Korean-style hierarchy can be detrimental, but since the Korean software industry is so insignificant (less than 1.5% of GDP in 2012), it doesn't matter much here.

    1. I work in an American service industry in which the labor productivity is extremely high, i.e. the legal industry. It is just as top-down as any Korean corporation.

    2. I should clarify what I mean by "top-down". It doesn't only refer to the fact that you obey without question. It also refers to the amount of responsibility that is delegated to you, among others. An absolute hierarchy may be less "top-down" simply because the supervisor doesn't give you many orders to begin with (but you would still follow any when she does). In that sense, law firms, both in the U.S. or Korea, can't be really top-down. Else, they wouldn't be trying to hire lawyers from top law schools. You are hiring bright, expensive guys exactly because you want (and need) to delegate a lot of the work to them.

      I'm curious and must ask, though, exactly what Korean company have you worked for, and in what capacity, when you make the comparison that your law firm is just as top-down as any Korean corporation?

    3. When it comes to talk of corporate culture in Asia, people fall into a habit famously enunciated by the Nobel Prize-winning economist Robert Solow: "anytime American economists engage in a discussion of [foreign country's] economy, they end with a blaze of amateur sociology."

      What he means is that people think they're making causal arguments, what they're actually doing is saying "I feel that..."

      If you could point to some empirical evidence from rigorous studies that shows that corporate culture or whatever is the malaise (as opposed to just something that people who go through it dislike) of [some Asian] economy, then I'd like to see it.

    4. "I work in an American service industry in which the labor productivity is extremely high, i.e. the legal industry"

      You said a funny.

      There is the paradox that when measurements are made in isolation from the whole what is deemed good for the part could be a tragedy for the rest.

      In terms of dollars made per man hour, the legal profession may be viewed as a very productive industry. However, for the rest of the economy, the business of legalese are like blood sucking worms, leeches and parasites that take more than they ever put into the economy in services rendered.

      Numerous mom and pop restaurants may not be productive for themselves but what they give up in personal profits are given back manifold to the economy in the form of long hours, hard work, good service, and cheap prices.

  3. Well written, as usual!

    I hate these types of articles because they ALWAYS work backwards from the assumption that Korea is too debilitatingly "Confucian" (whatever that means) and the story is always neatly sculpted to fit that narrative. Then you have the horde of nod-along chuckleheads like non-Asian expats who fancy themselves experts because they've worked in Korea for a bit, and Asian Americans who think that their outsider status in America somehow automatically make them experts on Asian culture. These types of people all give unthinking props to these articles because their entrenched biases are confirmed.

    I agree with the overall argument that there are many aspects of Korean society that need to be examined and reformed. But we're not going to get there when the same lazy argument is made again and again by those who are ignorant and don't care enough to fix their lack of knowledge and understanding. What's worse is that many Koreans, even native ones, have started to internalize this outsiders' viewpoint of their own country that they know best and are now basically parroting foreigners' views in an effort to be more "global" or whatever.

    1. " they ALWAYS work backwards from the assumption that Korea is too debilitatingly "Confucian" "

      The problem with reports like this and other assumptions is that it fundamentally contradicts empirical reality. If Korea's overall productivity is truly as low as it is reported Korea could not be where it is today. Notwithstanding the hours wasted behind the desk perfecting "the art of looking busy" there is productivity somewhere and that productivity raised Korea from indigence to affluence in a very short time.

  4. Thanks for the interesting and insightful post. There's just one point I would contest. You say:

    "In other words, Korea's labor productivity is low because of long overtime hours, not despite the overtime, as Kocken appears to imply."

    He invokes Parkinson's law that "work expands so as to fill the time available for its completion". I think it's clear that he is in fact saying productivity is low because of long hours.

  5. This comment has been removed by the author.

  6. You say:
    "This necessarily means that the longer one works, the lower the labor productivity, because if you increase the denominator while holding the numerator at the same level, the result is always a smaller number. In other words, Korea's labor productivity is low because of long overtime hours, not despite the overtime, as Kocken appears to imply."

    My question is: why should we assume that the 'numerator' must be held constant? You're saying that the output stays the same, even though there are overtime hours being put in. Well, why is the output the same? Isn't that the whole problem? If you put extra hours in you should get more output, and if you don't, then you're not being very productive. So there actually is a productivity problem if people are working extra hours and the output is not proportionate.

    Saying the numerator is held at the same level is admitting that the extra hours are completely wasted (or the actual productivity is done then, and there's little productivity during normal working hours). And what's the reason for the extra hours? It's not 'working backwards' from cultural assumptions - ask any salary man why they don't leave work at closing time and spend more valuable time with their families, and you will get the same answer: that it doesn't look good to leave work before one's superiors do. Why do the superiors work later? I dunno, retribution for having been kept late by their superiors when they were on the lower rungs of the corporate ladder?

    Anyway, cultural explanations aside (because we all know it's a soar topic for you) - the extra hours are not being used well. It lowers productivity. So why add extra, unproductive hours in the first place?

    1. No one said the numerator *must* be held constant. The point is that it makes no sense to say "labor productivity is low DESPITE long hours," because that is a facially inaccurate statement. And remember, Korea's manufacturing sector (which has plenty of overtime and hierarchy) leads the world in labor productivity. If the extra hours are "completely wasted," how do you explain this?

    2. Fine. I was focusing more on what Kocken was focusing on to begin with - productivity in the 'work place', loosely defined as everything other than the manufacturing and hospitality service industries, from which his direct and albeit subjective experience draws. Can I explain why things work better in the manufacturing industries for Korea? Not really. Machine-like efficiency is evident and suitable in machine-related manufacturing, and the martialization of a work force is conducive to such machinization of input-->output?

      I'm not sure why we're having problems with the word 'despite'.
      'Korea's labor productivity is low despite long overtime hours.' This is what you believe Kocken to be implying. It is the assumption that the rate of productivity (output/hours) is at an acceptable rate IF we do not include the overtime hours - IF the denominator were lower (i.e. not adding overtime hours), we would see a more acceptable level of productivity.
      'Korea's labor productivity is low because of long overtime hours.' This is what you claim to be the actual case. And it leads to at least two conclusions: (1) The extra overtime hours added is just throwing off the rate of production overall, because there is little-to-no added output/production in those hours. (2) The extra overtime hours added is an accepted and expected part of the work day, so people do not work as productively during regular hours, and will get a regular working hours day of work done in a longer day of extra hours
      And all of these conclusions, from 'despite', and the two versions of 'because', have a common denominator: extra hours. Extra hours that are bad because they either (1) make the calculation of productivity low, or (2) they make productivity low. Either way, people are working extra, unnecessary hours.

      This was the point of 'labor productivity' in Kocken's article and the previous Business Korea article to which he refers to - HOURS. It's about the overworked status of Korea in terms of HOURS. You cited an article from 문화일보, 2014.03.04, that says Korea is 23rd of the 34 OECD nations in terms of, not labor productivity measured in output/hours, but labor productivity of output/laborer/year - essentially, GDP per laborer per year, with 'value added' (I don't know if 부가가치 기준 refers to Gross Added Value, or...).

      The ranking of Korea in the labor productivity per hour understanding is actually not rock bottom, either, though it is lower - for 2012, it is at 31 of 34 as a percentage related to the U.S.; at 30 of 34 in current USD prices and PPP; and at 26 of 34 in constant USD prices with 2005's PPP. This is all from:

      Anyway, I've learned now that 'labor productivity' can be measured in different ways, and maybe you took issue with Kocken for one understanding of 'labor productivity' that he ignores, and one that you believe most people (and, admittedly, most OECD figures) actually refer to, viz. output/person (sometimes just GDP per capita), or output/laborer (which the 문화일보 article uses, and whose figures I haven't been able to find on that site). But I think an interesting side of measuring labor productivity, and Kocken's focus, is how time is mismanaged and wasted and essentially robbed of office workers who would be happier to spend that time with the family, or a friend, or even alone.

  7. I am not sure whether you would consider this as sound basis to state's one opinion (you demend facts/numbers and I will merely state my observations) but I have worked for almost 11 years for Korean manufacturers in South Korea: first in electronic industry and now for one of the top shipbuilders. I totally agree that as far as manufacturing is concerned the labor productivity is excellent. I have experienced it myself. However, to the contrary of the notion you support in many of your articles, I do think culture plays an important part in it. In fact, its rigidity helps to execute things more efficiently if Koreans are good at something (mind blowing optimization in all areas), or totally slows things down if they are at the level of developing things and not yet excel at them (i.e. when more imagination or flexibility is required). Again it might sound subjective/simplistic but truly it is a careful observation based on direct and long-term experience at Korean chaebols.

    On a more personal note, reading your otherwise open-minded and revealing articles, it always puzzles me why you so vehemently reject every notion that Korean culture might to some degree influence/deepen social phenomenons in Korea...

    1. "it always puzzles me why you so vehemently reject every notion that Korean culture might to some degree influence/deepen social phenomenons in Korea"

      I'm not sure that he does?

      I think it goes without saying that as personality and values influence individual views and behavior, culture shapes nations. But we are also not victims to personality or culture as well as these things are also shaped in turn by learning and experience. If Korea appears rigid today that is not a fait accompli. I think Korea's past as well as contemporary culture suggests she can be quite creative and is becoming more so as continued development demands it.

  8. In short - Koreans need to stop being so goddamn hierarchial, top-down, and uptight. Stop drinking too much soju...loosen up and maybe smoke some weed, and then the labor productivity might improve.


Comments are not available on posts older than 60 days.

Related Posts Plugin for WordPress, Blogger...