Yen: Now and Then Again
What a 30% drop in the exchange rate could mean for Japanese stocks
By: Verdad Research & Nikita Romanov
The Japanese yen has plunged 30% relative to the dollar this year. How will this major currency move impact Japanese companies and thus investors in Japanese equities?
We compiled a data set of all publicly listed Japanese firms with financial data from every quarter over the last 20 years. We combined that with firm-level data on the percentage of each company’s revenue that was derived from foreign or domestic markets and the percentage of their assets that were reported as domestic or overseas assets. We then tested how changes in the yen predicted future changes in these companies’ financials—and the extent to which this varied depending on the proportion of international versus domestic revenues and assets.
We found, in general, that a cheaper yen had a positive impact on Japanese companies’ financials. Our analysis of the cross section of all publicly traded Japanese firms during this time showed that revenue going forward grew faster for Japan’s exporters (30% of companies with highest % of revenue from foreign markets) versus domestic sellers (30% of companies with lowest % of revenue from foreign markets) when the yen was cheaper in the future, aligning with the basic microeconomic theory that Japanese exports would have had more globally competitive prices before converting back to yen, leading to higher demand and, hence, higher immediate revenue and eventual volume demand. The finding here is in line with those of a 2001 study on the topic.
We illustrate this relationship below, plotting the rolling two-year forward growth rate for exporters and domestic companies alongside the average yen rate across those forward two years (up is a cheaper yen for the yen/dollar rate). This chart shows that revenue grew faster when the yen was cheap, especially for exporting companies. However, the economic benefits from a cheaper yen take about nine months to flow through into financials, with benefits lasting for about two years.
Figure 1: 2Y FWD Revenue Growth vs 2Y FWD Average Yen/Dollar Rate Change
Source: Compustat. All publicly listed Japanese stocks above $25M in market cap, excluding REITS and financials.
But Japan is an island nation that imports most of its raw materials. So while a cheaper yen benefits exporters on revenue, more expensive imports often have a negative impact on margins, meaning that the relationship between earnings and the yen has been less powerful than the relationship between revenue and the yen.
The first chart below shows changes in input prices (think material costs to produce) and output prices (think sales prices) for Japanese firms since 1973. In 2012, as the yen began to devalue, both price indicators shot up. The second chart shows if Japanese firms were experiencing excess demand for their products (too many orders) or excess supply (too much production capacity) over the same period. Only in the late 1980s, 2017, and today have Japanese firms experienced excess demand.
Figure 2: Output and Input Price and Excess Supply and Demand Survey Results
Source: BoJ Tankan Survey of Japanese companies, 1 July 2022. In the first chart, up is increased prices. In the second chart, up is excess demand, down is excess supply/capacity.
As long as we keep in mind the time lag (i.e., the nine-month delay in full revenue growth and the stabilization of input costs mentioned above), a cheaper yen was good for the financials of Japanese firms, especially for exporters.
But before you assume the yen will stay cheap and rush out to buy all the exporting firms for the next few years, we have to check how efficiently the market priced the expected growth.
We analyzed returns for the same period as shown in Figure 2 above. We formed 3x3 portfolios based on the export intensity and offshoring of assets of individual Japanese firms each quarter. For simplicity, we just cut the last decade in half, with the start of the Abenomics devaluation from 2012 to 2017 being the “cheaper yen era.” During this era, all stocks, and especially exporter stocks, did well as the yen got and stayed cheaper for the next few years. Exporters compounded a small premium relative to domestic firms. Over the full decade, though, there wasn’t any meaningful return differential between exporters and domestic firms. In both the short and long haul, asset location (offshore intensity) didn’t seem to be particularly predictive.
Figure 3: Cheaper Yen Era and Decade of Returns on Japanese Equities
Source: Compustat. All publicly listed Japanese stocks above $25M in market cap, excluding REITS and financials.
Note the premiums above for exporters during the "Cheaper Yen Era" were significantly lower than the relative growth rates we showed you in figure 2, where exporters grew revenue at roughly 2x the rate of domestic firms for most of the period. This suggests that much of the expected financial statement impact of the yen devaluation was known by the market and was very quickly priced into exporter stocks. And here we have the benefit of hindsight that the yen stayed cheap for this whole era, ignoring the risk of a sizable reversal in the yen that did not occur.
Indeed, if you knew the yen’s forward one-year level relative to today in any given quarter, the long-exporter, short-domestic firm portfolio was a fantastic trade. Below we plot the one-year forward exporter portfolio premium over time if you knew the future yen/dollar strength.
Figure 4: Exporter Premium vs 1Y FWD Yen/Dollar Strength
Source: Compustat. All publicly listed Japanese stocks above $25M in market cap, excluding REITS and financials.
Unfortunately for us, we don’t know this information today. There is generally no persistent serial autocorrelation in the movement of FX pairs. Simple FX forecasts (like “no change”) generally outperform more complex econometric models despite having far less storytelling power.
There’s certainly no shortage of storytelling in today’s headlines on the yen. However, the yen’s future rate is largely a political prediction predicated on the future action of central banks in response to inflation expectations. Predicting inflation is the profession of central bankers, and it seems fair to say that they aren’t (and weren’t) particularly successful at it, despite their best efforts and the “forecasting” advantage of being able to influence it directly.
But even if you could nail the analytical predictions, the above out-of-sample check generally supports the 2001 paper’s finding that “investors are able to anticipate the impact of exchange rate fluctuations [on stock prices] with no significant delay.” One’s correct predictions may already be largely priced in. Or, as Oaktree’s Howard Marks reminded us recently, most correct consensus macroeconomic forecasts are not a source of abnormal excess returns.
Our study found, in line with prior research, that a cheaper yen appeared to benefit the financials of Japanese companies as a whole and Japan’s global exporters especially. Markets generally adjusted to changes in FX rates very rapidly down to the individual firm level.
Acknowledgement: This piece was co-authored by Nikita Romanov, a senior at Johns Hopkins University studying applied mathematics and philosophy. Speaking both Russian and Japanese at a native level, Nikita is looking for opportunities in finance, preferably with exposure to international markets.