The Japan News by The Yomiuri Shimbun

Home > Opinion > The Great East Japan Earthquake


The Great East Japan Earthquake

Service Lives of Building
-Actual Life-Spans Determined by Factual Investigation

Yukio Komatsu
Professor, Faculty of Sciences and Engineering, Waseda University

A Regulation Unparalleled in the West

Buildings have service lives. This seems obvious to Japanese, but might be seen as strange by Westerners.

The most typical service life is the life-span defined by Ministry of Finance directive - for example, 50 years for reinforced concrete offices, and 22 years for wooden houses. This was defined in order to eliminate arbitrariness when calculating depreciation for income tax or business tax purposes, but it is often referred to when calculating asset values. Perhaps, because the term service life was used instead of the more accurate depreciation period, it is also generally understood to indicate the number of years a building can be used for. I have carried out extensive research into building life-spans, and have keenly felt the huge impact that this service life has on the way Japanese look at buildings.

Once, thinking it would help my understanding of differences in building life-spans, I asked several specialists about the service lives of buildings in other countries. I came to understand that, apparently, developed Western countries don't have defined life-spans like Japan does. While I have not made a formal investigation into this, I found this surprising, but at the same time, it made sense. In the West, land and the buildings on it are handled as a single entity, so just as there is no service life for the land, there is no service life for buildings. This is a fundamentally different understanding of buildings than we have in our country, where people think of buildings as sooner or later becoming unusable.

Earlier, I mentioned life-spans, but here I would like to differentiate between life-spans and service lives. They are often used interchangeably, but I define service lives to be the number of years that it is decided a building can be used for, while life-span by itself refers to the actual number of years a building exists, and is not a decided-on figure. Service lives are nothing more than figures defined for use in tax calculations, and should have been decided by those that actually use the buildings.

Factual Investigation Using Housing Registries

Since around 1982, I have been using statistical methods to factually investigate actual building life-spans. First, let me discuss how to determine building life-spans. The first method that may come to mind is to find a building being torn down, and check how old it is. This method would tell one how long the life-span of a specific building is, but it would not necessarily be fair to say that this is the life-span of a typical building. One might then think that it would be sufficient to do the same with a greater number of buildings, but in reality, this is not so simple. First, it is difficult to obtain information about buildings being torn down, and then there is no guarantee that the history of any building in question can be obtained. If there were some sort of ledger which contained information for all buildings, this kind of investigation approach would be possible, but that is not the case.

However, the average life-span of the Japanese people is announced every year by the government. The Ministry of Health, Labor and Welfare estimates this based on the vital statistics, so put more accurately, this should be called the average remaining life-span of a newborn infant. I thought about using this same approach to buildings, and realized that I could use the property tax housing registry. Average human life-spans are estimates of the average length a group of newborn infants will live if they follow the mortality order (the percent chance of death for each age). As can be seen from this method of determining life-span, average life-span is hypothetical, and changes depending on specific circumstances. For example, if an epidemic occurs and the death rate rises, the average life-span will shorten. However, the life-spans of those who did not contract the disease will not have their own life-spans shortened by this change in the average. It is important to understand that the figures derived are hypothetical figures which reflect conditions at the time.

Is There any Scientific Basis for Service Lives?

Housing registries are created by individual towns, cities, and other municipal bodies, and contain the information they need to levy property taxes on buildings (houses). They do not cover all buildings, but because they are used for taxation purposes, there are notifications when buildings have been destroyed, which is useful for investigating life-spans. With assistance from various persons, I was able to obtain data, on a year-by-year basis, on the number of existing buildings, and the number of buildings that were torn down. I used these to estimate the average life-spans of buildings, and found that actual average life-spans were very different from legally defined service lives. I also found that, in terms of consequences due to structural materials, while wooden houses are specified as having shorter life-spans than reinforced concrete buildings, this was not always true in reality. From an architectural engineering standpoint, building life-spans are considered as being decided by the durability of the building materials used, but how well they are used also appears to have a significant impact.

Perhaps due to improvements in building materials in recent years, and the economic environment, average life-spans are gradually rising. The most recent analysis results show an average life-span of 50 to 60 years or longer for every type of building. Unlike average human life-spans, this refers to the point at which 50% of the buildings are still standing.

Looking at the establishment of legally-defined service lives, it is unclear if they possess any scientific foundation. It would be truly regrettable if people found themselves misled by these figures, which are merely defined for legal purposes, and it resulted in their tearing down of buildings which are still usable.

Survival Rate Curve and Average Life-Span

Yukio Komatsu
Professor, Faculty of Sciences and Engineering, Waseda University

1949 Born in Tokyo
1978 Graduated with Doctorate in Engineering from School of Engineering at the University of Tokyo (Majored in Architecture)
1978 Assistant at Faculty of Engineering, The University of Tokyo
1982 Assistant professor at Faculty of Engineering, Niigata University
1990 Assistant professor at School of Engineering, Yokohama National University
1998 Professor at Faculty of Science and Engineering, Waseda University
Currently professor at Department of Architecture, School of Creative Science and Engineering, Waseda University

2008 Winner of Architectural Institute of Japan Prize (Thesis) "Study on Life Time Estimation of Buildings [Tatemono no Jumyo Suikei ni kansuru Kenkyu]"