Reserve management and the carry trade

More than a quarter of century ago the Bank of Japan (BOJ) started to really amass

significant amounts of US government debt. At the time Japan was running a huge trade

surplus, the vast majority of which was coming from trade with the United States. So, the

Bank of Japan and the Ministry of Finance took the decision to invest vast swathes of that

surplus, back with the US government, buying US treasury bonds.

This seemed like a great deal for Japan because at the time the US treasury was paying them

close 7% to hold their debt. At the same time the Japanese government was issuing its own

debt at close to just 1%. So was born the ‘Japanese carry trade’.

Now, of course on the face of it, this seemed like a no brainer- making 7% for holding AAA

US government debt and paying only 1% to finance your own. Consequently, the Japanese

just couldn’t but help themselves and kept piling it on. Within a few years their total

holdings of US bonds passed one trillion dollars.

Now, that was all well and good, but eventually it did mean the BOJ became highly sensitive

to the level of the USDJPY. So by 2011/2012 when Japan had amassed over $1 trillion of US

debt that was even more the case. Over the preceding decade the USDJPY had fallen from

above 125 to as low as 75, so they found themselves sitting on a huge currency loss. A loss

that pretty much wiped out anything they had made in that previous decade from the ‘carry


So, what did the BOJ do then? They intervened to buy even more US dollars. A move that

subsequently worked out very well for them indeed because by the middle of 2015 the

USDJPY price went back to 125 again.

Another plus side was their US treasury holdings did make them a huge amount of money

too, offsetting that earlier currency loss. That’s because when US yields collapsed after the

financial crisis the Japanese were sitting on a huge paper profit with the price of the US

bonds they owned through the roof.

So, it’s worth remembering that the BOJ is always sensitive to the exchange rate-always.

When we start talking numbers on this scale it does get really quite scary. Each 1 Yen move

in the USDJPY exchange rate; say from 100 to 99 results in an immediate loss of $10 billion-

big numbers right? That’s the scale of the P+L swings when one has reserves of that


Perhaps, also worth noting, is the fact that more recently, whilst US 10year yields have been

rising above 3%, the Japanese equivalent is still camped pretty much near zero, currently at

just 0.08%.

So, why am I telling you all this? Simply, because I trust you can now appreciate the

importance of reserve managers in determining longer term currency direction. What you

may not know; is that Japan is actually no longer the biggest holder of US government debt

because they were overtaken by China a couple of years back. Currently, between them

they now hold around $2.2 trillion of US government IOUs.

In mentioning this, forgetting any rhetoric to the contrary, let’s also be clear; the PBOC

(Peoples Bank of China) are equally sensitive when it comes to the level of the USDCNY,

naturally for exactly the same reasons as their Asian counterparts.

Now, here’s a few more facts for you. As of September 2018, surprisingly the third biggest

holder of US debt is Brazil ($317bln), even more amazing, Ireland is 4th with around $290bln

and creeping up in 5th place is the UK with $276bln. In fact, UK investors have steadily been

increasing their holdings of US treasuries since October 2017 when we collectively held

around $225bln.

OK, so of course the UK has got some way to go before catching the Japanese and the

Chinese, but the point is; this in effect is a short play on the GBPUSD. Indeed, a GBP

shortage that has increased by around $50bln over the past year-following the Japanese and

Chinese central bank examples perhaps?

So, there’s clear evidence if ever you needed it, of just one aspect of how the markets have

been effectively shorting the pound since Brexit? Leaving aside the exchange rate for one

minute; its worth noting the current yield differential between 10year US bonds at 3.10%

and 10year gilts at 1.4%. Not quite the 6% that drove the BOJ originally into that kind of

trade, but a ‘positive carry’ nonetheless.

Naturally, should the pound collapse in the weeks and months ahead, then it will work out

very nicely for those that have put their money to work with Uncle Sam. Conversely, if it

doesn’t, then there’s certainly going to be an entirely different price to pay.

Key Data releases due later today

23/11- 1.30pm Canadian September retail sales and October CPI inflation report

23/11- 2.45pm US November Manufacturing and services PMI data