Main Issue in the Second Round of Wage Hikes: The real issue is to improve the terms of trade

KAWAGUCHI Daiji
Faculty Fellow, RIETI

Since the establishment of the second Shinzo Abe administration, the government has been routinely urging the business community to increase wages. The request can be divided roughly into two features. One is a request for nominal wage hikes in line with a series of monetary easing implemented by the Bank of Japan (BOJ) which is seeking inflation. The other is a request for real wage hikes to counter the effective decline in the labor share, due to an increase in the ratio of non-regular workers and retained corporate profits.

This column argues that should this request be significant, it would be justified from the first perspective, while the second perspective is poorly reasoned.

♦  ♦  ♦

To confirm the second perspective, let us first review the relations between labor productivity and wages over the last two decades to prove that there has been no real change in the way that wages are determined.

According to a standard economic theory that states that marginal productivity (increase in output resulting from adding one unit of a production factor) determines the level of wages, the wage level is in proportion to labor productivity assuming aggregate production technology is the Cobb-Douglas production function and perfect competition in the labor market and goods market.

Labor productivity refers to the value that workers produce per hour, while an hourly wage is the salary that workers receive per hour of labor. Since the value of money changes over time, it should be adjusted by price indices. The gross domestic product (GDP) deflator is generally used for adjusting labor productivity, whereas the consumer price index (CPI) may be used for adjusting wages.

The figure shows the changes in the values of real wages divided by real labor productivity during 1994-2013 to see if the above mentioned theory holds in reality.

Figure: Relations Between Real Wages and Labor Productivity
(ratio of wages to productivity)

Figure: Relations Between Real Wages and Labor Productivity (ratio of wages to productivity)
(Note) Compiled with data from the Ministry of Health, Labour and Welfare, the Cabinet Office, and the Ministry of Internal Affairs and Communications

When hourly wages are adjusted by the CPI, the wage/productivity ratio shows a substantial decline. This means that real wages do not increase despite a rise in labor productivity, contrary to what economic theory predicts. However, when the wages are adjusted by the GDP deflator, the ratio is largely stable, showing that the predictions largely holds true.

The results differ greatly depending on which of the price indices is used for adjusting the wages because the GDP deflator is a measure of the price level of goods that consist of a nation's GDP, while the CPI is a measure of the price level of goods that comprise a representative consumer basket. GDP covers consumption as well as investments, government spending, and net exports (exports less imports). These two indices are based on very different baskets of goods and services and have evolved differently in the past decades. It is known that the changes in the price index for exports less imports explain the gap between them.

According to the BOJ, the export price index (2010 = 100) tumbled from 136.2 in January 1994 to 116.6 in December 2014, while the import price index rose from 76.0 to 129.4 over the same period. With the exception of a sudden fall in yen-based import prices amidst the sharp appreciation of the yen that followed the financial crisis of 2008, Japan saw deterioration of its terms of trade over these two decades with a decline in its export prices and a rise in its import prices.

A decline in export prices and an increase in import prices put a downward pressure on the GDP deflator, whereas a rise in import prices puts an upward pressure on the CPI. This is why Japan in the 2000s experienced a substantial deflation according to the GDP deflator, and a minor deflation according to the CPI. The gap may be attributable to the inclusion of investment goods and government-consumption goods, but it is pointed out that it is due largely to the deterioration of terms of trade.

When examining the theory that real wages are in proportion to real labor productivity, what is appropriate is to adjust both nominal figures with the same price index. Judging from the fact that the relations of the two figures adjusted by the GDP deflator are stable, the standard economic theory more or less has explained successfully the state of wage determination in Japan.

♦  ♦  ♦

The relations between the increase in labor productivity and a rise in wages are observed in the Japanese economy as a whole. This is due to a mechanism in which companies respond to higher productivity with increased outputs as well as increased labor inputs, which triggers labor shortages and a subsequent wage increase. A certain proportional relationship between the improved labor productivity and a wage increase is observed as the technological relationship assumed in the Cobb-Douglas production function generally holds true in the Japanese economy as a whole.

It is clearly appropriate to use the GDP deflator instead of the CPI for adjusting wages because domestic companies' ability to pay would not improve when the prices of imported goods including food and energy increase.

Yet, as the CPI measures the purchasing power of workers, it is true that nominal wages decline at a level greater than the CPI, and thus their standard of living has declined. However, the decline in Japanese workers' purchasing power is caused by a decline in the purchasing power of the Japanese due to the deteriorated terms of trade. While the decline in the purchasing power of Japanese workers is a problem, what is necessary is to improve the terms of trade.

The recent decline in oil prices is good news for Japan in improving its terms of trade, but to improve it further, Japan must take a wide range of policy responses including energy policy, industrial location policy, and international trade policy. In any event, the government should avoid intervening unduly in the labor market by misunderstanding the nature of the issue.

In this article, I am mainly reiterating the claims made by macroeconomists from Hitotsubashi University. Professor Makoto Saito has been pointing out the outflow of Japanese wealth due to the worsening terms of trade. Professor Kyoji Fukao has analyzed quantitatively the impact of the deteriorated terms of trade on real wages.

Overseas, many experts including Harvard University Professor Martin Feldstein and London School of Economics (LSE) Professor John Van Reenen have refuted the argument that the increase in labor productivity in the United States and the United Kingdom would not translate into higher real wages. They also have pointed to the need to pay attention to the difference between the CPI and the GDP deflator. In both countries, the two figures are found to be proportional when additional factors are taken into account.

♦  ♦  ♦

On the premise that real wages rise in proportion to real labor productivity, the government's active intervention into wage setting could only be justified when it is aimed at raising the level of nominal wages in line with the increase in prices to the extent that it would not change real wages. In this case, the government should refer to the GDP deflator rather than the CPI.

For the purpose of overcoming deflation, some may embrace the approach of raising nominal wages to trigger a cost increase and a subsequent rise in output prices. Such an approach must be discussed based on empirical studies using microeconomic data on the extent to which an exogenous wage increase would affect output prices. Hitotsubashi University and the University of Tokyo are conducting microeconomic analysis on price setting, using point-of-sale (POS) data from retailers. I hope the study extends to analysis on the relations between prices and wages.

Some claim that the wage issue should be discussed in combination with the structural issues of a non-regular workforce and working style reform. These claims are understood as the discussion on raising the level of real labor productivity.

Since the relations between real labor productivity and real wages are stable, an increase in real labor productivity leads to higher real wages. However, in the government's efforts to prompt wage increase, higher wages could be achieved more smoothly by focusing on increasing wages in line with the rise in prices, rather than taking up diverse issues of conflicting vested interest among stakeholders, which would only cause the debate to stall.

>> Original text in Japanese

* Translated by RIETI.

March 6, 2015 Nihon Keizai Shimbun

April 24, 2015