Formation of Expectations and Social Reform: Implications for countermeasures for the low birthrate, gender equity, and employment system reform

Visiting Fellow, RIETI

At an interview by on May 7, 2013, Katsuhito Iwai reportedly commented on "Abenomics" as follows:

Expectations can be a driving force for society

"Capitalism is a system where people with money but no ideas lend money to those with ideas but no money, and ideas get realized thereby. During a deflation, people gain only by holding money and speculating on money itself, which causes a credit crunch. An expectation for inflation will make people shift their speculation on money to investment in ideas and, furthermore, investment in things.

In that sense, expectations not only make money work as money but also affect the economy itself significantly. The present economic policies are often described as having 'only expectations but no substance,' but, in a monetized economy, expectations can be the substance."

Professor Iwai's comments are uniquely thought-provoking. The capitalism that he describes is a daily reality in Silicon Valley's information technology (IT) industry. In Chicago, on the other hand, there is frequent talk on macroeconomic and social issues based on microeconomic ideas. Expectations are discussed, based on Robert Lucas's rational expectations theory, in the context that the formation of expectations neutralizes policy interventions. However, Professor Iwai's argument emphasizes that expectations for monetary value in an uncertain macroeconomic situation increase the incentive for a specific speculative action and cause social change. When the formation of expectations goes beyond being a result of objective information received by individuals and becomes socially diffusive, it can be a driving force for social reform.

The effect of the formation of such expectations goes further than the effect of expectations for deflation or inflation. I was recently invited to report at a government advisory panel on countermeasures for low birthrate. I stressed the importance of a society where work-life balance (WLB) can be attained based on the results of empirical data analysis. To my surprise, one of the panel members—whose name shall remain undisclosed as this was a closed-door panel—stated that "WLB is not important as a measure to address low birthrate." The member argued that "close to 80%" of the causes of low birthrate in Japan are "due to late marriage and non-marriage" and that the remaining percentage is due to post-marriage birthrate decline. Therefore, he stated that if half of the latter is due to lack of WLB, then "its influence is approximately 10% of all causes." When I heard this, I immediately sensed what was erroneous with the argument, which was confirmed later. The member had implicitly assumed that the effect of WLB state on individual behavior occurs only after marriage. Generally speaking, however, low rate of first marriage or "late/non-marriage" is considered an endogenous variable of low birthrate (a phenomenon likely to be caused by factors that affect birthrate) and thus is hardly an independent cause. In particular, late/non-marriage cannot be considered causally unrelated to WLB state in society. In rebuttal, another member of the panel pointed out an important aspect that "some women are hesitant to get married because, once they do so, it will be difficult to do both work and childrearing." On my part, I was not prepared with empirical evidence for a counterargument, and, in accordance with academic standard, withheld immediate criticism. This column intends to provide a supplementary counterargument on the topic.

Late marriage and non-marriage related to the low compatibility of childrearing and post-marriage employment

What is important here is the effect of formation of expectations, as Professor Iwai argues. I analyzed the relation between the chronological change in total first marriage rate (TFMR) and WLB indices (a comprehensive index with two major components, "compatibility rate of work and childrearing" and "compatibility of workplace flexibility"), which I had used in analyzing the relation between the change in women's labor participation rate and the change in birthrates for Organization for Economic Co-operation and Development (OECD) countries. As indicated in the table below, the rate of first marriage is decreasing on average (late/non-marriage is increasing) as shown by the negative intercept. However, countries with high WLB, in particular the "compatibility of work and childrearing" reflecting the societal state of childcare leave, income compensation for such leave, and the availability of nurseries and childcare centers, have a statistically significant lower rate of change on late/non-marriage. While this analysis takes into account the effect of unobserved heterogeneity in marriage rate among countries, a more rigorous analysis that considers the effects of other variables that could potentially generate spurious relations will be necessary. Yet, among the three OECD countries with the highest "degree of compatibility of work and childrearing," that is, Denmark, Sweden, Finland in this order, Denmark increased the rate of first marriage between 1980 and 2002, while many countries experienced a considerable decrease during this time. Sweden and Finland, too, had a relatively slow trend towards late/non-marriage. Hence, the relation shown in the table below is unlikely to be spurious. If the "compatibility between childrearing and work" prevents the decrease of first-marriage rate, then we cannot deny causality based on the formation of expectations by never-married women in regards to their post-marriage situation. In Japan, more than 60% of women leave their jobs to raise children, and unmarried women consider how their lives will become after marriage by observing the choices of married women. The assumption that late/non-marriage and WLB are unrelated thus leads to mistaken policy measures.

Table: An Analysis of Change in the Total First Marriage Rate in 1980-2002Table: An Analysis of Change in the Total First Marriage Rate in 1980-2002

A policy vision that supports the formation of new employee expectations is needed

I have repeatedly emphasized that employers' presumption on "women leaving the company sooner or later" causes statistical discrimination against women and thus generates a higher rate of women leaving their job: a self-fulfilling prophecy. This expectation is different from the above-mentioned relation between WLB state and late/non-marriage because the expectation is imposed by the employers, instead of the women themselves.

Today in Japan, relaxation of requirements for dismissal on firm's economic grounds is being discussed as part of employment system reform. Generally speaking, I agree with the promotion of employment liquidity. However, I also hear such discussions as the division of regular employees with indefinite term contract into distinct types: the conventional "generalist-type" whose roles are not limited to specific job categories, and "job-type" employees whose roles are confined to specific job types and can be terminated when they become redundant. I feel that a desirable society should provide equal opportunities to a diverse set of people regardless of employment status, increasing the utilization of individuals' potential capabilities. This is not regarding the also important but frequently discussed issue of providing "a more comprehensive safety net assuming increase in unemployment rate." What is currently left out is a vision regarding what kind of expectations will be formed for diverse employees through system reform, and how society will change as a result. For employees towards whom companies' human resource investment is not primarily applied, labor productivity is largely dependent on the incentive of employees to invest in themselves. This incentive depends on whether those employees can form an expectation for future returns from self-investment, such as gaining new knowledge or skills, or making efforts to improving the quality of their present work for purpose of career development. If Japanese companies treat only "permanent employees" with long years of service and hours of work as the core of human resource and continue to marginalize other types of employees, reform will become merely a reorganization of the status system and will deprive diverse employees of hope for the future. The employment system reform should be reviewed from the perspective of changing Japanese society to allow formation of expectation that "effort and self-investment will pay off" among employees with diverse working styles, and not just among permanent employees.

May 21, 2013

July 2, 2013