JILPT Research Eye
Thoughts on Working Hour Systems

January 20, 2015
(Originally published on June 25, 2014 in Japanese)


Yutaka ASAO

Project Researcher of JILPT

Recently, working hour systems (or at least some of them) have been a much-debated subject. Aside from the debated policy direction itself, this essay will outline some of the author’s personal thoughts on working hour systems by presenting a few results from a survey conducted previously. It should be stressed that the content of this essay is purely the author’s own personal opinion as a researcher and does not represent the view of JILPT or any other organization.

Although it might be better to show latest data if talking about recent topics[Note 1], the data introduced here are re-aggregated results of the “Comprehensive Survey on Japanese Working Styles” conducted with individuals in 2005[Note 2]. However, the data referred to in this essay seem unlikely to have changed greatly since then.

"Job content" set in the survey

An outline of the survey can be found in the research survey series mentioned in[Note 2]. But, to begin with, the “job content” set in the survey should at least be explained here. Instead of using conventional job classifications, respondents were asked to mark all that apply out of 17 options focusing mainly on the specific nature of their work. The following is the classification categories marked as relevant by at least 200 respondents, and the aggregated results based on them are shown here[Note 3]. People subject to this aggregation were limited to respondents who indicated that they were regular employees.

  • 1. General clerical job (mainly involving desk work)
  • 2. Planning of individual projects (often involving meetings with relevant people)
  • 3. Planning of the overall aims and operation of projects (top management or staff of top management)
  • 4. Management and supervision of others (with subordinates or junior employees)
  • 5. Building or making things at construction sites or manufacturing plants
  • 8. Money/securities handling work
  • 11. IT related work and handling of data-processing equipment, based on specialized knowledge or skill
  • 12. Research and development based on specialized knowledge or skill
  • 13. Research and study based on specialized knowledge or skill
  • 15. Teaching, instructing or advising work based on specialized knowledge or skill

* The numbers correspond to those given to options in the questionnaire.


The way of carrying out a job/ High-discretion job and Low-discretion job

Viewing levels of discretion in carrying out the work by different types of "job content" (Fig. 1), in "Construction or manufacturing work" and "Money/securities handling work" more than half responded that the work was "Specified in manuals, or instructed each time", and the level of discretion was thus relatively low. On the other hand, the level of discretion was generally higher in "Planning" work and work that requires "Expertise"[Note 4].

Fig. 1 How jobs are carried out, by job content

Fig. 1

Click to expand (opens in new window)

Level of discretion on working hours

Meanwhile, when respondents were asked about their level of discretion on working hours (Fig. 2), the proportion responding "Have considerable freedom to decide by myself" is relatively high in "Specialized work involving information" and "Planning" related work. However, across all the categories, the proportion of responses to the effect that working hours were "Decided in advance" was high, reaching 90% even in "Development and research" work. Levels of discretion on how work is carried out are not necessarily in proportion to the level of discretion on working hours.

Fig. 2 Working hours by job content

Fig. 2

Click to expand (opens in new window)

Meanwhile, though due to lack of space the data are not shown here, the proportion responding that they were subject to a flextime or discretionary work system was highest in "Development and research" work (27.4% combined). Thus, it would appear that the actual level of discretion on working hours is not closely related to the working hour system applied

How a decision is made on overtime

Looking at the way to make decisions when working overtime (Fig. 3), it would appear that in Japanese workplaces, the response "Make my own judgment and report to my superior afterwards" was quite common (just under 40% of all regular employees). Seen by job content, in particular, this response accounted for more than 50% in "Planning" related work and "Specialized work involving information". Jobs like this also showed considerably high levels of discretion on whether or not to work overtime[Note 5].

Fig. 3 How a decision is made in working overtime, by job content

Fig. 3

Click to expand (opens in new window)


First, in connection with the recent debate, let me refer to the argument of "regular employee exemption from working hour regulation" the author previously published in a report[Note 6]. There, the author argued that exemptions from working hour regulation should be applied to the type of employees who are highly likely to be subject to personnel management based on the fundamental principle of the regulation, even if the regulation is not applied. For example, these employees should be regarded by the employer as personnel who are indispensable to the running of the business, and this will simply be reflected in their salary amounts, which means more than a certain high level of annual income is a requisite. As a researcher with a background in economics, this view of the author remains unchanged.

The same graph in the aforementioned report is shown below (Fig. 4). Although personnel in managerial position are generally exempt from the application of working hour regulation under the Labour Standards Act, when they themselves were asked, not many replied that “Working time management is non-existent or lenient”[Note 7]. Thus, in terms of the annual incomes of personnel in managerial position (department or section manager class), classified as those subject to a regular working hours system, those subject to a flextime and/or discretionary work system, and those subject to lenient working time management, the proportion of those who were subject to lenient time management exceeded the proportions of those subject to other working hour systems in the income bracket of more than 10 million yen[Note 8]. Although the point has been argued in a somewhat roundabout way, when deciding annual income conditions, incomes around the 10 million yen mark could be used as a criterion for consideration[Note 9].

Fig. 4 Annual incom composition by position and working hour system
(All regular employees)(Japanese Yen)

Fig. 4

Click to expand (opens in new window)

The following is the brief outline of the author's view on working hour systems, based on the above data, etc. In Japanese workplaces, it can be said that a considerable degree of discretion and flexibility is actually given when necessary in connection with ways of carrying out a job. Now along with the growing service economy and the decreasing number of workers engaged in manufacturing and construction accounting only for one-quarter of the total ("Labour Force Survey", 2013 average), there are not many industrial sectors where outcomes increase significantly in proportion to working hours. In such a situation, it is now very common that companies leave the ways of carrying out work to their worker's discretion, though in different forms and degrees. Actually, various efforts are being made to find an appropriate way at each workplace. If an attempt were made to establish this practice as a system, the author would be in favor of the policy direction currently being debated. However, if a mistake is made in designing the system, such as by merely importing a foreign system, there is a concern that, like several other "deregulations", it will only be created but will not function. It could even destroy the "customary practice" that has been built up until now.

“Outcome is not necessarily proportional to working hours” is not the same as “outcome can be measured by something other than working hours”. It is also certain that considerable working hours are needed in order to achieve an outcome, even though not necessarily in proportion to them. The first thing we need to do is discuss on how to measure the outcomes on which remunerations are based so that we (including labor and management) can find an acceptable solution.

Correcting excessively long working hours is also a structural issue that should be addressed now. As can be seen in the data shown above, the problem in terms of discretion lies in the fact that the level of discretion in relation to working hours is low. And moreover, another problem is that discretion in relation to the volume of work has hardly been taken into account at all. This could be considered as the biggest reason why increasing discretion in ways of carrying out jobs tends rather to cause longer working hours. If the regulation that helps to prevent longer working hours is to be done away with, an acceptable system must be proposed in its stead. For example, setting and observing rest hours, enforcing time-off in lieu of holiday work, and so on. In addition, what the author has always proposed is that once one “task” finishes, it should be made customary to take at least a certain length of consecutive leave before making a start on the next “task”.

To summarize the above, if a system is to be created for those exempt from working hour regulation, it is considered essential to introduce mechanism including annual salary system above certain income level, stipulations in labor-management agreements or in rules of employment at least with no opposing opinions (including ways of preventing long working hours), the consent of the persons concerned, and provision of information to labor unions or employee representatives, and their monitoring and  complaint processing. However, it must be added that the author as an economist does not think it necessary to regulate all of these issues through legislation.

Note 1. IIn the case of social surveys, it should also be noted that there are not a few cases where surveys conducted during a period of changing circumstances or too lively debate involve some problems.

Note 2. For an outline of the survey and its results, see JILPT Research Series No.14 “Comprehensive Survey on Japanese Working Styles: Data on diverse ways of working”  (April 2006) (Japanese only).

Note 3. In the following (including the figures), headings will be abbreviated for convenience. Also, No.12 and 13 have been integrated into a single category.

Note 4. As probably clear from the description, jobs in which a relatively large proportion of respondents selected options with a high (low) level of discretion are called jobs with a high (low) level of discretion (the same applies below).

Note 5. Preceding the “Comprehensive Survey on Japanese Working Styles” for about a year, JILPT conducted a survey titled “Questionnaire survey on realities and consciousness of working hours” (July-August 2004).  According to this,  42.0% of respondents replied that they sometimes worked overtime hours without being paid overtime premium (“unpaid overtime work”) (the data excluded personnel of managerial position ; section manager level or higher). As reasons why overtime premium was not paid (up to two answers allowed), the most common response was “I work overtime to produce the outcome that I would find acceptable, and so do not claim overtime pay” at 23.2%, followed by “Due to budgetary constraints, overtime pay would not be paid even if I applied for it” (19.4%) and “Because overtime pay is paid at a fixed rate regardless of hours actually worked” (19.0%).

Note 6. JILPT Labor Policy Report Vol.5 “On Diverse Ways of Working and Related Policy Issues” (March 2006) Chapter 4, 4-3-1 “(3) Regular employees exemption from working hour regulation” (p63-64). (Japanese only)

Note 7. In the “Comprehensive Survey on Japanese Working Styles”, this was only 14.2% among section or department managers; the proportion responding “Normal working hours” was 73.9%.

Note 8. When aggregated by “job content”, more or less the same trend was seen. In jobs using specialist knowledge or skills (information-related, development and research), however, flextime+discretionary working hours accounted for the highest proportion among those earning more than 10 million yen and less than 12.5 million yen, while lenient working hour management accounted for a higher proportion than flextime+discretionary working hours among those earning more than 12.5 million yen  and less than 15.0 million yen.

Note 9. Employees earning upwards of 10 million yen account for 5.7% of all regular employees and 1.8% of regular employees below section manager class (‘no response’ on annual income excluded), according to the “Comprehensive Survey on Japanese Working Styles”.