Every month, the Bank for International Settlements calculates and publishes the real and nominal effective exchange rates for major currencies. The most recent data, released in mid-February, contained a shock for Japan. They show that the yen’s real effective exchange rate (REER, representing roughly the purchasing power of the currency) is now as low as it was in the early 1970s, when the yen was first floated following the collapse of the Bretton Woods and Smithsonian systems of fixed exchange rates.

As the yen’s purchasing power has weakened, the Japanese have noticed that imported goods are now more expensive, and the media are full of stories about Japanese visitors to New York, London, and Singapore being surprised by the high prices of goods and services there. In Tokyo, for example, a bowl of ramen typically costs less than ¥1,000 ($8.66), but in New York, based on my observations, it costs more than $20. A McDonald’s Big Mac costs the equivalent of $3.38 in Tokyo, as opposed to $5.81 in the United States, according to The Economist.

Similar comparisons can be made for subway fares, Uniqlo Heattech undershirts, five-star hotel rooms, and many other goods and services. Prime property in Tokyo, which was the most expensive in the world 30 years ago, is now much cheaper than in any other major financial center.

So, when and why did Japan become so cheap? The yen’s REER more than doubled from 1973 to its April 1995 peak, but has since lost all of that gain. Some blame the decline on Japan’s 15 years of deflation, while others point to the yen’s sharp nominal depreciation since 2013 under so-called Abenomics, fueled by the Bank of Japan’s quantitative easing and ultra-low interest rates.

By definition, deflation in Japan, combined with the yen’s nominal depreciation against major currencies (while its main trading partners had healthy inflation), contributed to the REER decline. But both deflation and nominal yen depreciation result from the ultimate cause of the yen’s weakness: sluggish Japanese productivity.

That brings us to the Balassa-Samuelson hypothesis, generally considered to be the most appropriate framework for explaining movements in the REER. The theory holds that, under certain assumptions, when productivity increases much faster in a country’s tradable sectors than in its non-tradable sectors, and the difference in the rate of increase is higher than in its trading partners, the REER will appreciate, regardless of whether the exchange rate is fixed or floating. This explains the dramatic yen REER appreciation that began in the 1960s and peaked in April 1995.

In fact, the yen’s REER appreciation under both fixed and floating exchange-rate regimes was regarded as a textbook example of the Balassa-Samuelson effect. So, could the hypothesis also explain the subsequent decline in the yen’s purchasing power to its early-1970s level?

Over the last 30 years, the Japanese economy has suffered from several negative shocks, including the 1997-98 banking crisis and the 2011 Great East Japan Earthquake and Fukushima nuclear power plant accident. But the biggest structural shift in Japan in the last three decades has been the hollowing out of tradable manufacturing sectors.

In the 1990s, Japanese manufacturing firms, including automobile producers, started to build factories in the US and Europe in order to avoid trade conflicts. In parallel, many Japanese companies invested in other Asian countries to benefit from lower-cost labor. More recently, manufacturers have been reluctant to invest in Japan, because the country’s increasing number of retirees and shrinking working-age population have made it less attractive as a market for high-volume products or as a location for labor-intensive factories.

Unsurprisingly, the productivity of factories in Japan has stagnated. Real wages have thus slightly declined in the last 20 years, as the real wage index (2015=100) increased from around 60 in 1970 to 113 in 1997 and then declined to 100 in 2021, resulting in a loss of purchasing power.

True, Japan’s traditionally non-tradable service sector, including retail, hotels, entertainment, and education, is a mixed bag of advanced and backward industries. But the gradual decline in manufacturing productivity has narrowed the previously large productivity gap between the tradable and non-tradable sectors. This reverse Balassa-Samuelson effect explains the decline of the yen’s REER.

The effects of this shift are becoming increasingly clear. Until the mid-1990s, strong productivity growth enabled Japan’s export industries to weather the yen’s continued nominal appreciation. But after the 2008 global financial crisis, and with productivity declining, exporters could no longer cope with the yen’s rise, which accelerated the relocation of operations abroad.

Moreover, Japan’s rapidly increasing social-security expenditures have constrained government budgets for education, science, infrastructure, and defense. This has further eroded the capacity for basic research that can help to increase productivity.

In short, do not blame the BOJ’s monetary policies for the yen’s REER depreciation. It is flatlining productivity in tradable sectors and stagnant real wages that explain the arrival of Cheap Japan.

Ito is a professor at the School of International and Public Affairs at Columbia University and a senior professor at the National Graduate Institute for Policy Studies in Tokyo.

–Project Syndicate