Flattening of the Wage Phillips Curve and Downward Nominal Wage Rigidity: The Japanese Experience in the 2010s

In this paper, we examine from both a theoretical and an empirical perspective the validity of the hypothesis that downward nominal wage rigidity (DNWR) induced upward rigidity in wage setting, thereby contributing to the flattening of the wage Phillips curve. We focus in particular on Japanese regular workers, those workers who are characteristically employed on long-term contracts. Our theoretical study, which incorporates long-term employment contracts, indicates that DNWR induces upward wage rigidity through the following two channels: first, due to the lack of sufficient downward wage adjustments during economic downturns, firms may become reluctant to increase wages in economic recovery phases; second, firms contain wage increases even in economic expansion phases as they take into account the risk of pay cuts in the future. The strength of the latter channel largely depends on expected economic growth and its uncertainty. As a result, the wage Phillips curve becomes flatter than would be the case without DNWR. In line with the theoretical result, our empirical study using the panel data of Japanese regular workers reveals that the slower growth of monthly earnings, which excludes bonuses but includes overtime pay, for workers who display a strong degree of DNWR pushed down the growth of monthly earnings at the aggregate level by 0.4 percentage points per year (a range of 0.2 to 0.6 percentage points, given uncertainty regarding the identification of DNWR) between 2010-17. In particular, the channel arising from future pay cut risks became relatively stronger in the late 2010s, when labor market conditions became markedly tighter.