JILPT Research Eye
Wage Increases and the Labor Market

February 26, 2015
(Originally published on July 16, 2014 in Japanese)


Yutaka ASAO

Project Researcher of JILPT

The spring wage negotiations in 2014 resulted in an average rise of 6,298 yen or 2.09% in monthly pay (based on data from 411 companies) according to the final survey results conducted by Nihon Keizai (Nikkei) Shimbun (Japan's leading financial newspaper), carried on May 25th. The article also said that this was higher than the previous year's 4,988 yen or 1.67%, achieving "the first 2% level pay hike in the past 15 years" and 42% of responding companies increased their base pay (known as "base-up").

In 2014, the spring wage negotiations attracted public attention for the first time in years. So it seems opportune to think about wages, and especially wage increases. It should be stressed, however, that the views expressed below are purely the author's own opinions as a researcher and do not represent the views of JILPT or any other organization.

Wage systems or wage level and fluctuations

When considering wages, the main focus of attention can take one of two basic directions: wage systems themselves, or wage level and fluctuations under a given wage system. In recent years, the discussion has mainly revolved around systems. While this partly reflects the trend of so-called "performance-based pay system", it could also be influenced greatly by the stagnant wage increase itself, mainly through spring negotiations.

In 2008, the author was in charge of conducting surveys focusing on wage systems[Note 1]. In one survey, targeting companies with a scale of 30 or more employees,  companies that adopted “age” or “years of service” as elements to decide basic pay decreased in number, and even in companies still adopting these elements, their weight had decreased. On the other hand, there were increases both in the number of companies that adopted “job duties and responsibilities”, “performance” or “ability” as elements, and in the weight of these elements in companies that adopted them. This is what we call a shift from personal attribute elements to work-related elements. Among these elements, however, “performance” is not a particularly prominent element, but “ability” and “job responsibility” are also adopted and prioritized to about the same degree. Thus, rather than a shift from “ability” to “performance”, the survey results suggested a situation in which both elements are used in combination[Note 2].

Wage rise suppression theory

Until recently, wage level and wage fluctuations tended to be discussed in the context of “wage suppression”, against the background of declining or stagnant wage increase rates in spring negotiations [Note 3]. For example, one of the papers published on the website of the Cabinet Office Economic and Social Research Institute (transferred from the former Economic Planning Agency) in 2009 was “Why were wages suppressed?” The paper analyzed the trends in recurring profit margins and real wage increase rates, extracted from data on companies in manufacturing industries with capital of 1 billion yen or more, based on “Financial Statements Statistics of Corporations by Industry” of the Ministry of Finance. It concluded that “The wage increase rate in large manufacturing companies was highly correlated with profit margins until around 2000, but since the turn of the century, wage increases have not kept pace with the vast improvement in profit margins” (p2) [Note 4].

On this occasion, the author attempted to produce a similar graph modeled on the above-mentioned paper, using data from the “Financial Statements Statistics of Corporations by Industry”. Unlike the paper, this one is based on the total of all capital scales in manufacturing industries, and year-on-year changes in wages (salary per employee) have not been converted to real figures (Fig. 1)[Note 5]. Judging from this graph, the correlation between recurring profit margins and wage increase rates appears to have been weak but maintained [Note 6]. However, it might be more accurate to say that in around 1997 and 1998, the wage increase rate experienced a sudden step-down in level (shown by the arrow in Fig. 1). In other words, an “employment anxiety switch” or perhaps a “deflation switch” was turned on during that period. In this kind of psychological mood, wage increases will be suppressed [Note 7]. A phase transition in wage increase occurred [Note 8].

Fig. 1 Trends in recurring profit margin and year-on-year increase in wages per employee (manufacturing industries / all capital scales)


Click to expand

  • Data source: Calculated from the Ministry of Finance, "Financial Statements Statistics of Corporations by Industry".

Some thoughts on the labor market

In a recessionary phase in economic and business cycles, suppression of wage increases based on voluntary agreement between labor and management may be seen positively as an effective functioning of the labor market in the short term. However, given that there is an ongoing situation in which the "deflation switch" cannot be turned off even though the economic phase has (potentially) changed, we should bear in mind that, unless we turn off the switch and increase wage, the economy will fall into a state of so-called "fallacy of composition".

The labor market has multilateral social functions. I have an impression that much debate has been focused on the function of distribution of employment (labor force) alone in recent years. However, equally or the most important is the income distribution function. Arguably the only difficulty in a capitalist economy is that, because resources tend to be distributed not according to people’s actual needs but in terms of the contribution to short-term “profit”, income sometimes tends not to be distributed to the people who need it [Note 9]. In such a situation, the income distribution function of the labor market is a social system with the potential to compensate for this “deficiency” [Note 10].

Only when a certain degree of income has been properly distributed to the people who need it, can a capitalist market economy fully manifest its functions; this will also lead to the prosperity of individual companies.

However, it is not so easy to pull out of a state of “fallacy of composition”. This is because, people just make “rational” decisions, at least within the scope of environmental conditions considered here. What is needed is something to trigger a (reverse) phase transition.

Revival of income distribution based on wage increases?

The author has attempted an estimation of the "wage rise function", which once was the subject of much attention but has not even been used in recent years (Fig. 2). The result suggests a tendency that, though it sounds a bit far-fetched, since around 1998, actual figures for the wage increase rate have generally been below the estimated values obtained by the function expression. Particularly noted is the difference observed in the years 2005-2007, a long period of economic recovery known as "beyond Izanagi" (surpassing the "Izanagi" boom of the late sixties). In the author's judgment, the economy had already fallen into a state of "fallacy of composition" during this period. After that, given setbacks such as the Lehman shock and the Great East Japan Earthquake disaster, there seems to have been no great change until 2013 spring in the attitudes of labor and management toward wage increases. In spring 2014, however, the government announced a campaign of aggressive wage increase as part of measures to break away from deflation, which seems to have become a main impetus and brought a change in the attitudes of labor and management. Signs of a revival in the labor market's income distribution function seem to have appeared.

Even so, we should also bear in mind that this still cannot be called a complete revival. Using this function formula for estimation, the rate of wage increase this spring (2014) is calculated at 3.45%. Given a standard error of 0.82 points, it is expected to reach at least 2.63%. Although the rate of wage increase aggregated by the Ministry of Health, Labour and Welfare has not yet been announced, judging from the Nikkei Shimbun figures mentioned at the outset of this paper, it seems that this figure were not reached. Although one can take the view that the shortfall is supplemented by bonuses and others, the author would rather wait until the wage increase negotiations next spring (2015) before judging whether the phase transition (thought to have occurred in around 1998) has been reversed.

Fig. 2 Trends in actual figures for the spring wage increase rate by major companies and estimates using the regression formula


Click to expand

  • Data source: Estimated from Ministry of Health, Labour and Welfare, "Results of Spring Wage Negotiations by Major Private Companies", id. "Employment Security Service Statistics", Statistics Bureau, Ministry of Internal Affairs and Communications, "Consumer Prices", and Ministry of Finance, "Financial Statements Statistics of Corporations by Industry".
  • (Note 1) The estimation formula is as follows. Spring wage increase rate = 1.995 (3.077)*** + 1.023 (4.626)*** x Consumer price inflation rate (previous year) + 0.376 (0.423) x Ratio of job openings to job applicants (previous year) + 0.799 (2.429)** x Year-on-year difference in recurring profit margin (most recent fiscal year)
    In brackets: t values (*** = significant at less than 1%, ** significant at less than 5%)
    Estimation period: 1986-2013.
  • (Note 2) If the 2013 year-on-year difference in recurring profit margin (most recent previous year) is provisionally calculated as 1.0, the estimated rate of wage increase for major companies in spring 2014 is calculated as 3.45%.
  • (Note 3) Recurring profit margin is calculated from companies with capital of 1 billion yen or more.


Please see the following for the author’s basic thoughts on wages.
"Thoughts on Wages Based on the Pay Gap Between Regular and Non-Regular Employees" (JILPT "Business Labor Trend" July 2010, p38-44) (PDF:861KB) (Japanese only)

Note 1. For details of this survey, see JILPT Survey Series No.52, "Basic Findings for Discussions on Changing Wage and Employment Systems and Male-Female Wage Disparities" (March 2009).(Japanese only)

Note 2. “Performance-based pay system” may be evaluated, through about twenty years of trial and error, not as regulating wages in their entirety but as having settled and become established as a revised version of the “performance assessment” that had been used until then, or as playing a role in the wage system, through the post-adjustment function of wage levels in line with changes in corporate performance.

Note 3. Besides wage increase through labor-management negotiation, wage levels and their fluctuations are influenced by the employment structure, i.e. increase in non-regular employment and the age level of the “baby boomer generation”. In recent years, such changes in the employment structure also have a large effect, but in this paper, the focus is put on wage increase.

Note 4. Ryo Yamada, Akihito Toda and Takaaki Murakami, “Why were wages suppressed?” (Cabinet Office Economic and Social Research Institute, New ESRI Working Paper Series No.12 / July 2009). The graph in question is Fig. 3 on p.12, and the period covered by data is FY1981-2006.open a new window(Japanese only)

Note 5. Nevertheless, since it is not made very clear what specific data were used in the paper in question, there is an element of analogy when making estimates to produce this graph .

Note 6. Even so, since profit and wages are in some ways conflicting with each other, the correlation between them cannot be so strong by its nature. Moreover, even if levels of recurring profit margins were maintained or even improved, the impact on employment and wages would probably be different depending on whether this resulted from an increase in ordinary profit as the numerator, or from relative shrinkage of sales as the denominator.

Note 7. This psychological mood could be called “expectation” in terms of economics. Specifically, the author referred to “expectation” as used in Chapters 5 and 12 of Keynes’ “General Theory of Employment, Interest and Money”. It is not like the “perfect foresight” in the rational expectations hypothesis.

Note 8. “Phase transition” is a physical concept whereby the phase of a substance is switched, as in the case of ice ←→ water ←→ steam under certain temperature conditions, for example. A similar process could also be used to describe socio-economic aspects, albeit reduced in rigor.

Note 9. For example, in Keynes’ “General Theory of Employment, Interest and Money”, the first half may be construed as describing the difficulty caused by separation of saving entities from investing entities.

Note 10. While a detailed discussion will be left to another paper, negotiations held in collective labor relations would make it possible to distribute wages in a way that better reflects people’s needs. To this end, we would need to stop assuming the (marginal) productivity that can be allocated to individual workers.

Downloading Adobe Readeropen a new window To view PDF files, you will need Adobe Reader Software installed on your computer.The Adobe Reader can be downloaded from this banner.