Abstract :
- A potential financial crisis originating in Japan was reported by the media in early August 2024, linked to the end of the yen carry trade investment strategy.
- This strategy, which consists of taking advantage of low interest rates in Japan to borrow in yen in order to invest in higher-yielding assets in other currencies, came to an end when the Bank of Japan raised its key interest rates, causing some volatility in international capital markets.
- Despite the lack of reliable data to estimate the scale of yen carry trade, this article attempts to logically trace these flows and identify the countries that have benefited from them since 2022. These include the United States, France, Hungary, and Singapore.
- The impact of unwinding yen carry trade strategies appears to be very limited. However, even without a significant impact, a second wave of effects could occur between now and the end of 2024 if the yen continues to rise against other currencies.
Read the article: yen-carry-trade-the-crisis-that-will-not-happen-note
« A new financial crisis is looming »: this was the headline in the media in early August 2024, in connection with the fall in stock market indices in Japan[1]. This stock market incident came after the Bank of Japan’s surprise rise of nearly +20 basis points in key interest rates. This event prompted a reaction from certain international market operators, who had to put an end to their yencarry tradeinvestment strategy and unwind their positions.
The purpose of this note is to briefly review the mechanisms involved in this chain of events and provide some initial analysis to understand the potential repercussions and contagion effects on other countries.
What actually happened?
Faced with very low inflation and sluggish economic activity for several years, even decades, the Bank of Japan (BoJ) embarked on a highly accommodative monetary policy. This consisted of promoting an environment of extremely low (even negative) interest rates and injecting large amounts of liquidity into the economy.
This low interest rate environment in Japan has attracted keen interest from certain categories of investors, who have implemented strategies known as « yen carry trades » strategies: borrowing in yen at very low interest rates to purchase financial assets denominated in high-yield currencies (mainly bond debt securities in emerging countries and a wider range of investments in developed countries[2]), thereby generating short-term gains. The exchange rate also plays a role in the transaction. Once the profit has been made, the financial returns on the foreign currency investment are converted into yen, part of which is used to repay the initial yen loan[3], making the transaction even more profitable when the yen depreciates over the same period.
On the other hand, once the BoJ raises interest rates and the yen tends to appreciate, the yen carry trade becomes less advantageous: the interest rate differential with other regions of the world becomes less favorable and an appreciation of the yen exposes investors to losses. With the BoJ’s unexpected rise in its key interest rate on July 31, 2024, investors were caught off guard and « unwound » their yen carry tradestrategies: they sold their foreign currency-denominated assets and bought yen to repay their initial loans in that currency. This event caused a wave of panic on international financial markets and was behind the fall in the Japanese stock market.
While it is very difficult to calculate the scale of yen carry trade strategies, the chart above provides a clearer picture of the phenomenon:
- Between the end of 2021 and 2023, Japanese banks tended to increase their short-term yen-denominated assets (blue curve) to USD 162 billion, reaching a total amount of close to USD 500 billion. These assets, mainly loans and deposits, can be used as a proxy to measure the scale of the funds used for carry trade[5].
- This increase in Japanese banks’ external assets coincided with a widening interest rate differential with the United States (red dotted line), a period during which the BoJ kept its rates close to 0% while the Federal Reserve (Fed) entered a cycle of tightening financing conditions.
- During this period, net speculative positions in yen on the foreign exchange market, as calculated by the Commodity Futures Trading Commission (CFTC), were short (yellow curve).
- Ultimately, short-term yen loans from Japanese banks were probably intended to fuel investors’ quest for higher yields, a particularly attractive proposition given that the market was on average short on the yen, betting on its depreciation and reinforcing the appeal of yen carry trade strategies.
When the BoJ unexpectedly changed its monetary policy stance and the yen regained ground against other currencies, net speculative positions on the yen also changed (see sharp decline in the yellow curve in Q3 2024, with a net long position now). The slight reduction in the interest rate differential with the United States (but also with other countries) and the change in the yen’s trajectory forced investors exposed to yen carry trades to unwind their positions and repatriate their funds, to the detriment of beneficiary countries exposed to greater capital flow volatility.
Which countries are potentially exposed to contagion effects?
The unwinding of yen carry trade strategies is not without consequences, particularly for countries that benefited from capital inflows in 2022-2023, a period during which these strategies were booming. These countries could face volatility in capital flows and stock market movements.
As it is extremely difficult, if not impossible, to track the path and calculate the actual amounts of carry trade flows, it seems necessary to provide a practical reminder to better understand this phenomenon:
- The starting point, in the current configuration, is to observe an increase in Japan’s short-term external assets in yen (see previous section).
- The counterpart to Point 1 is that the rest of the world’s short-term external liabilities in yen would have increased since 2022. It is therefore necessary to identify the countries that have benefited from a surplus of external loans in yen from Japan, in order to deduce that it is investors from these countries who have borrowed in yen in order to carry out carry trade operations. Unfortunately, this data is not available. Using several proxies, however, some countries stand out with increases in external liabilities in yen ranging from tens of billions of USD (Germany, China and Hong Kong, the United States, France, and the United Kingdom) to more modest amounts (Australia, Brazil, Canada, South Korea, Italy, Luxembourg, the Netherlands, and Switzerland) over this period.
- Using the list of countries revealed in Point 2, the ideal would be to observe the cross-flows of securities acquisitions, all currencies combined except yen, between 2022 and 2023 (data for the first half of 2024 not yet available). The aim would therefore be to identify the countries that benefited from capital inflows from the countries listed in Point 2 during this period, ideally short-term flows corresponding to a carry trade strategy. Although the IMF provides data of this type for certain countries, unfortunately it does not cover the desired period. In this analysis, a proxy is therefore used with incoming portfolio investment flows (net of repayments), the origin of which is global in the absence of cross-flows. If these flows tended to increase more significantly during the period 2022-2023 than during the period 2010-2021, then the countries concerned can be considered to have potentially benefited from yen carry trade strategies. A Z-score calculation[9] is used to better identify these countries, although the results should be interpreted with caution in the case of developed countries[10].
The results are presented in the chart below:
How should the graph be interpreted? The further to the right of the graph a country is (a Z Score strictly greater than 0), the more it tended to attract portfolio investment flows in 2022-2023 in proportions higher than its average for the years 2010-2021. Logic would suggest that this surplus of inflows could be linked to yen carry trade strategies. The higher a country is on the graph, the greater the share of its 2023 GDP represented by cumulative portfolio investment inflows in 2022-2023, meaning greater exposure to capital flow volatility during potential unwinding of yen carry trades.
What are the results? Emerging economies are generally more sensitive to capital flow volatility[11], but most emerging countries do not appear to be particularly exposed, with the majority located in the lower left part of the chart[12]. This is probably because central banks in developed countries raised their key rates at around the same time as emerging countries, and despite higher interest rate differentials between Japan and emerging countries, investors favored developed countries with a more favorable risk/return ratio. Several countries appear to have low exposure and are unlikely to experience significant repercussions (Saudi Arabia, Morocco, Nigeria, Poland), unlike Romania and Hungary. Among developed economies, certain countries stand out with higher levels of exposure: France, the United States, Austria, Singapore, and Finland.
Risks limited at this stage
Three main risk channels can be analyzed to estimate exposure to the unwinding of yen carry trade strategies, with changes in: sovereign yields, exchange rates against the yen, and direct and indirect effects on the performance of major stock indices.
Changes in stock market index performance: a significant impact
The shock was quite severe in early August, but ultimately proved short-lived. There are two groups of countries that can be used to understand the impact on stock markets. First, there are the countries directly affected by carry trade investments in yen, where asset classes other than sovereign bonds could be affected and thus have an impact on the stock market, which is also highly interconnected with other financial centers. Secondly, there are countries that have strong trade relations with Japan and whose stock markets may have fallen in response to the decline in the Nikkei 225.
Most countries in the first category saw their benchmark stock market indices fall significantly between July 31 and mid-August, before returning to their pre-July 31 levels in the second half of August (United States, Austria, Bulgaria, France). While some have not yet returned to this level, they are close and tend to see their index recover at the same rate as the Japanese index (Singapore, South Korea, Hungary, and Romania, for example). For the second category, countries that export heavily to Japan, the impact on financial markets followed a trend fairly similar to that of the countries in the first category, where a recovery quickly took place after a brief decline in indices, with the exception of three countries (China, the United Arab Emirates, and Italy), although this does not appear to be significant.
Changes in sovereign yields: low impact at this stage
In principle, countries that had benefited from yen carry trade since 2022 should have seen a wave of capital outflows when yen carry trade strategies were unwound in early August. For countries where these amounts were high (relative to their long-term average or as a percentage of GDP, see previous section), these capital outflows could be significant, meaning that sovereign bond yields in these countries should have risen[13].
Of the 17 countries that could have potentially benefited from yen carry trades, almost all have not seen their bond yields increase since July 31, with the exception of Bulgaria (for maturities of less than 5 years, nearly +20 basis points, which is not a significant change at this stage) and Nigeria.
Exchange rate movements against the yen: impact to monitor over time
Carry trade is a short-term investment strategy, i.e., with a time horizon of less than one year. The bulk of carry trade volumes in 2022 and 2023 have probably already matured. However, carry trades that are refinanced at maturity or undertaken later (between the second half of 2023 and the first half of 2024) may not have matured by August 2024.
Furthermore, the Fed now seems more inclined to cut interest rates, following the example of other central banks (in the eurozone and even the UK), which would inevitably lead to a reduction in the interest rate differential with Japan, even if the BoJ maintains the status quo. In such a scenario, the yen would tend to strengthen, at least in the short term. These two factors (yen appreciation and narrowing yield spreads) would in principle increase the likelihood of further unwinding of yen carry trades. As a result, a second wave of capital flow volatility could be seen in H2 2024. As mentioned in the first part, yen carry trade strategies are exposed to losses when the yen appreciates, as has been the case since the end of July 2024. The magnitude of the movements will depend on the volume of yen carry trades still in play.
The chart above shows that, at this stage, the yen is rapidly gaining ground (light blue bar) compared to the period of depreciation that began a year ago at most (yellow bar). As the size of the light blue bars approaches or exceeds that of the yellow bars, new unwinding would take place and therefore pose a threat to the countries concerned (especially the United States, South Korea, Hungary, and the eurozone, Singapore, Switzerland, and Romania to a lesser extent, as shown in the chart).
However, unlike at the end of July, market operators will probably be more alert to signals that could lead to a potential new rise in interest rates in Japan and will not be caught off guard. This « surprise effect » played a major role in the unwinding of yen positions in August 2024. This vigilance could potentially lead to increased use of currency risk hedging tools (such as swaps) by the most exposed investors. Thanks to these hedging tools, investors engaged in a yen carry trade strategy would thus reduce their exposure to substantial losses in a scenario of continued yen appreciation and would not necessarily be forced to abruptly unwind their positions, which would mitigate the effects of a second wave.
While it is very difficult to measure and track carry trade investment flows, the crisis mentioned at the beginning of the month is unlikely to materialize. Given the direction of monetary policy (which is relatively more accommodative elsewhere in the world than in Japan, with a few exceptions), yen carry trade strategies are likely to lose momentum over the coming months and quarters.
On the other hand, in this context of narrowing interest rate differentials between Japan and the rest of the world, the current trend of yen appreciation should in principle continue, as several indicators seem to suggest, such as JPY/USD speculative positions (see yellow curve in the first chart). This aspect should be monitored closely, or at least followed up on in the short term.
V.L Article completed on September 5, 2024
[1] The Nikkei 225 stock index lost 19.5% of its value between July 31 and August 5, 2024, before recovering but still posting a decline for the month of August (-4.4% from July 31 to August 21).
[2] Notably those issued by emerging countries with limited inflation, as they offer high yields. These also include securities denominated in USD or EUR, whose yields have been more attractive since 2022 with the rise in key interest rates in both regions. Several articles suggest that the US technology market has been one of the beneficiaries of these investments.
[3] It should be noted here that this type of investment does not necessarily involve the use of currency hedging, for two main reasons: i) the desire to save on the cost of purchasing a hedging tool, or ii) the high cost of hedging tools for yen positions in recent years. Thus, these yen positions were not systematically hedged in the context of carry trade. Not hedging these positions maximized the gain on the transaction as long as the yen depreciated, but also exposed investors to substantial losses in the event of depreciation, thus encouraging them to quickly unwind their yen positions.
[4] The Nikkei 225 index is composed mainly of large Japanese exporting companies, which generate significant revenues in USD. Consequently, the appreciation of the yen was interpreted as a future weakening of their revenues.
[5] While this proxy is effective, it is not perfect: i) it tends to overestimate carry trade amounts, given that some loans, in particular, may be used to meet other yen needs (e.g., to settle international trade transactions in yen) ii) it also underestimates carry trade amounts because it does not take into account the external assets of other Japanese institutions (insurers, pension funds, etc.), or Japanese investors of all kinds who themselves borrow in yen to engage in carry trade.
[6] These effects are documented in academic literature (Hutchinson, Sushko (2012) and Cheung, Cheung, He (2012)).
[7] Academic literature proposes methods for identifying these countries, as in this article by Heath, Galati, McGuire (2007): the United States, the eurozone, the United Kingdom, Caribbean financial centers (Panama, Cayman Islands, Bahamas, etc.), and to a lesser extent Switzerland, Hong Kong-Macau-Singapore.
[8] By cross-referencing the International Monetary Fund’s balance of payments database on external debt securities with maturities of less than one year denominated in yen (mainly loans and deposits) with the Bank for International Settlements’ banking database on total liabilities in yen (all maturities, all liability instruments combined) for all sectors in all countries.
[9] The Z Score was calculated by subtracting the average of portfolio inflows (in billions of USD) over the period 2010-2023 from the cumulative portfolio inflows (in billions of USD) over the period 2022-2023, and dividing the result by the standard deviation of the series in question. This technique makes it possible to compare the results of different countries and identify those that deviate significantly from their long-term average.
[10] This type of proxy is particularly suitable for emerging countries, where portfolio investments appear to be more affected by carry trade strategies than equities. On the other hand, equity investments are also widespread in developed economies as part of carry trade strategies benefiting from high liquidity. However, including them in the analysis could overestimate carry trade volumes in developed countries, where transactions involving this type of asset do not necessarily correspond to short-term strategies, which would therefore skew the results. The choice of proxy may therefore seem questionable here, or at least less relevant for developed economies.
[11] Emerging countries are often more vulnerable and tend to experience greater repercussions on their balance of payments depending on changes in their exchange rates and imported inflation, variations in their foreign exchange reserves, or even to cover short-term external financing needs, etc.
[12] Countries that tended to attract less portfolio investment flows in 2022-2023 or made more repayments for this category of investment.
[13] With a sudden increase in the supply of these securities, bond prices would in principle tend to fall and, due to the inverse relationship between bond prices and yields, yields would tend to « tighten » and therefore increase.
[14] A country experiencing economic difficulties, particularly in terms of inflation and currency risk, which fuel tensions on bond yields regardless of what happens in Japan.
