Type of activity	Corresponding occupation
Cooking	Student chef
Cleaning	Building cleaning
Laundry	Laundryman
Sewing and knitting	Sewing machine operator
Miscellaneous family affairs	Janitor
Shopping	Janitor
Child care	Kindergarten teacher
Elderly/nursing care	Nurse assistant
Volunteer work	Weighted average for service industries

( billion yen, % )

(Note) Data on the number of working hours represent the weekly averages per capita per day, that have been calculated using total average number of hours (whole week) worked by workers age 15 or over by activity type, which are shown in the "Survey on Time Use and Leisure Activities." People not engaged in a relevant activity are also included in the calculation of averages.

(3) Assessed value of unpaid work by gender

3.Annual assessed value of unpaid work per capita by gender

2.Assessed value of child care for married couples

1. Cooking and food service industry

(Note 1) Service output value of laundry = Assessed value of unpaid work *2.856 (output value of laundry industryyenwages and salaries of the same industry). The base data have been quoted from the input-output tables for 1990.

(Note 1) Day-care and senior citizens welfare institutions, etc. consist of the child welfare business, elderly welfare business, mentally handicapped and physically disabled welfare business, rehabilitation care business, and other social insurance and social welfare mentioned in the "Private Nonprofit Institutions Income and Expenditure Survey 1996," Economic Planning Agency.

(Note) Compensation of employees consists of "salaries and wages," "employers" contributions to social security schemes, etc. in respect of their employees." Figures for fiscal 1995 ("Prefectural Accounts 1995," Economic Planning Agency) were used, since figures for 1996 were not available.

However, "unpaid work valued as paid by gender" takes the ratio of paid work to unpaid work separately for men and women and applies the gender specific average pay rates to each. "Valued pay rate for domestic work" and "valued pay rate for industry" apply pay rates more disaggregated by industry and maintain the disaggregation by gender.

Moreover, the scope of unpaid work in the UK is significantly larger than that in Japan. So it is difficult to directly compare the figures of time use in the two nations (see the following reference table).

Therefore, applying the UK's method to Japan after making the scope of unpaid work in Japan as much as close as possible to that in the UK, the total assessed value of unpaid work in Japan is estimated below:

Japan (1996)
time for unpaid work 295 (minutes) / time for paid work 250 (minutes) = 1.18
Total wages and salaries = 239,377.8 billion yen
The total assessed value of unpaid work = 282,465.8 billion yen
62% of GDP (at current factor cost)

There are two main reasons for the significant difference of the percentage of GDP between the two nations. One is that the scope of unpaid work in the UK is larger than that in Japan, even after adjustments. The other is that the ratios of unpaid/paid hours are quite different because the paid working hours are longer in Japan than in the UK. Moreover, the GDP ratio of the wages and salaries is 55% in the UK and 52% in Japan, a factor that has influenced the results slightly.

Acknowledgements
We would like to acknowledge the help of Mr. Uzair Rizki (Office for National Statistics of the UK) who gave us detailed information about the methods used for monetary valuation of unpaid work in the UK and offered useful commentary on an earlier draft.

5. Issues to be Noted

Attention needs to be paid to the following issues when making use of our estimates.

(1) Significance of non-economic value

Unpaid work could frequently have special value or effects for the provider and the beneficiary, because such activities are unpaid and devoted. For example, child care is a form of expression of love by parents for their children, and contributes to the formation of bonds between parents and children. In that sense, unpaid work has, in some cases, non-economic value rather than economic value, in the form of provision of a type of service.

Our estimate does not attempt to assess such non-economic value. Monetary valuation under the OC method can be regarded as an estimate that considers subjective valuation by the provider, because the provider chose to spend valuable time on unpaid work. The method, however, does not attempt to value or estimate respect or appreciation by the beneficiary for the unpaid work or effects of housework on the formation of bonds among family members.

(2) Problems with the scope of unpaid work

The scope of unpaid work is fixed basically in accordance with third-party criteria. Some people, however, argue that the scope should be slightly widened. For example, according to some, "schoolwork" and "studies and researches" should be included in unpaid work, because such activities have some labor elements in that they are conducted in order to improve the potential ability to gain income in the future.

On the other hand, some overseas estimates include "Travel" for unpaid work in unpaid work, regarding it as ancillary to unpaid work. Our estimate, however, excluded travel as a whole from unpaid work, as the statistics used did not specify the reasons for the travel, with the exception of commuting.

(3) Limitation of time use survey data

Our estimate is based on the results of the distribution of time use by people, as given in the "Survey on Time Use and Leisure Activities" conducted every five years by the Management and Coordination Agency. With regard to the monetary valuation of unpaid work, the survey has several limitations, in addition to problems in the handling of "travel," since the purpose of the survey is not to clarify the conditions of unpaid work.

One such limitation is that because the "Survey on Time Use and Leisure Activities" does not have categories for some of the activities that are theoretically unpaid work, it is impossible to assess the monetary value of such activities under an independent classification. For example, house maintenance and gardening are regarded as unpaid work and are included in equivalent surveys in other nations. In the "Survey on Time Use and Leisure Activities," however, such activities are included in "housework" or "hobbies and amusements," resulting in possible underestimation of unpaid work compared with other nations. Our estimate gives a breakdown of housework. The "Survey on Time Use and Leisure Activities," however, does not have such a breakdown; therefore, our estimate referred to the "National Time Use Survey" conducted by NHK in this regard. As a result, the assessed value for each sub-section of "housework" is not necessarily accurate.

Furthermore, people often engage in two types of unpaid work at the same time. Mothers, for example, do housework while taking care of their children. Therefore, it is desirable to assess such activities in a precise manner, and some of the surveys conducted abroad double-count such activities. The "Survey on Time Use and Leisure Activities," however, covers only main activities, and accordingly, our estimate evaluates only the main activities.

(4) Problems with applied wage

The OC method uses the average wages for respective genders and age groups. As a result, a disparity in wages between men and women is directly reflected in the assessed monetary value of unpaid work by sex. Because of this, although time spent by women on unpaid work is longer than that spent by men, the difference in the assessed value becomes smaller, or the assessed value for men could turn out to be larger than that for women, because the applied wage for men is higher than that for women. The burden of unpaid work by sex can be more simply revealed by time use data.

On the other hand, there are some arguments on corresponding occupations used under the Replacement Cost Method Specialist Approach (RC-S method). Since it is difficult to agree on a single correspondence, we chose occupations on which statistics were available. In addition, wages of occupations or housework employees used under the RC method are lower than those of other types of jobs. Consequently, assessed value under the RC method tends to be lower than that obtained under the OC method.

(5) Problems with output valuation

In our estimate, the monetary value of unpaid work is obtained by multiplying hours by wage. Because of this, monetary value is high if working hours are long or the applied wage is high, regardless of the quantity and quality of the service actually provided by the relevant unpaid work.

The value of service in a market economy should essentially be evaluated by the market price based upon the quantity and quality of the service. In view of this, in the field of monetary valuation of unpaid work, efforts are being made to develop a method to assess the value of the service itself (or output) produced by the relevant unpaid work.

With regard to output valuation, some theoretical achievements have been reported abroad, but no method has been established yet. Besides, a considerable amount of base data is necessary for such valuation, a problem that must be addressed in the future.

6. Base Data

The outlines of time and wage data that constituted the base for monetary valuation of unpaid work in 1996 are shown below.

(1) Time data

The base data are the results of a survey on time use published in the "Survey on Time Use and Leisure Activities 1996." Data for sub-sections of "housework" have been calculated by dividing the total hours used for "housework" in the proportion (see Note) that is given in the "National Time Use Survey 1990" (NHK).

When the provider is engaged in two or more different types of work at the same time, the hours spent are recorded as those spent on the main activity.

(Note) The proportion of time spent on each type of housework has been adjusted in accordance with the survey by NHK in 1995 (aftercode system), to estimate such a proportion in 1996.

(2) Wage data

1.OC method

The base data used were wages per hour by gender and age group, which were obtained by dividing the prescribed wages in respective genders and age groups by the prescribed actual working hours in the relevant category shown in "Basic Survey on Wage Structure 1996."

The average wage for all workers calculated by the same method is 1,749 yen, while the average for men is 1,976 yen and that for women is 1,255 yen.

(yen)

Age	Men	Women
15-19	957	894
20-24	1,174	1,083
25-29	1,464	1,270
30-34	1,784	1,376
35-39	2,057	1,418
40-44	2,272	1,369
45-49	2,450	1,342
50-54	2,543	1,327
55-59	2,311	1,253
60-64	1,742	1,167
65-	1,568	1,169

2.RC-S method

The base data used were wages per hour by occupation type, which were obtained by dividing the prescribed wages in respective occupation types by the prescribed actual working hours in the relevant category shown in "Basic Survey on Wage Structure 1996."

On the other hand, under the simplified RC-S method, the average wage for the corresponding occupation (1,030 yen) was applied to housework (cooking, cleaning, laundry, sewing and knitting, and miscellaneous family affairs). As for other activities, wages that were applied under the RC-S method were also used under the simplified RC-S method.

3.RC-G method

The national average wage (880 yen) was calculated from wages paid for housework support service by region, which are shown in the "Survey of Wages of Home Workers FY1996," and was used as base data.