The Tokyo session

A band of hours, not a building 

Tokyo is one city inside a session that spans at least five financial centres. Wellington quotes first, Sydney follows about an hour later, and Tokyo, Hong Kong, Singapore and Shanghai open within a short window of each other. Naming the whole band after Tokyo is a convention inherited from the period when Japan was the region's dominant financial centre, and it survives because it is short. Singapore and Hong Kong are substantial foreign exchange centres in their own right, and much of what moves in these hours is priced in neither.

The distinction matters because a session is a span of clock time rather than a place. Nothing shuts when Tokyo shuts. What changes at a session boundary is who is at the desk, which currencies they have a commercial reason to deal in, and how many of them there are. The instruments and the quoting carry on unchanged.

Where the band sits on a clock 

Sessions are described in coordinated universal time, because UTC is the one reference that does not move and belongs to no market. Most desks treat the Asian band as running from the point at which Tokyo's morning begins to the point at which London's does. Gulf Standard Time sits four hours ahead of UTC and, unlike most of Europe and North America, does not shift for daylight saving, so the conversion is a single fixed addition.

A session boundary is therefore approximate by construction, and the Asian band is the only one of the three whose relationship to a Gulf clock never moves.

What is active while Asia is at the desk 

Activity concentrates in the currencies whose home markets are open, because the banks, exporters and funds with a commercial reason to deal in them are the participants who are awake. Four groups account for most of what moves.

• The yen complex. USD/JPY is the reference pair of the group, with the crosses against the euro, sterling and the Australian dollar alongside it. Japanese corporates hedge commercial flows here, and a daily benchmark rate is fixed in Tokyo mid morning, which draws a concentration of orders toward one moment in the session.
• The Australian and New Zealand dollars. Both countries publish their economic releases and hold their central bank announcements in local business hours, which fall inside this band.
• The offshore Chinese yuan and the regional currencies quoted against it, which respond to the reference rate published in China each morning.
• Regional index CFDs. Contracts referencing the Nikkei, the Hang Seng and the Australian benchmark are written on exchanges whose hours sit inside this band, so those underlying prices form here and nowhere else in the day.

None of this changes what is available to quote. The 60+ currency pairs YAL lists are quoted continuously from the week's opening to its close, and a pair does not become unavailable because its home market is asleep. What changes is where the interest sits. The one real exception is a share CFD, which can reference a price only while the exchange quoting the underlying share is open, so a contract on a company listed in New York has no price at all during these hours.

Depth, range and the arithmetic of a thin book 

The band is conventionally called the quietest of the three main sessions, and in a narrow sense that is accurate: for the European and North American majors, fewer participants are quoting, so fewer orders rest on either side of the current price. Depth is not a mood. It is a count of the orders sitting at each price level, and it has an arithmetic consequence. An order consumes the resting orders nearest the current price first and then moves to the next level, so an order of a given size travels further through a thin book than a deep one. That one mechanism explains both things practitioners notice here: quoted spreads on non Asian instruments are generally wider than at those instruments' own peak hours, and a single large order moves a price further than its size alone suggests.

Range follows from the same fact. A session range is the distance between the highest and lowest price traded between two points on a clock, and on an ordinary day the Asian range on a European or American instrument is narrower than the range it prints later. For the yen, the Australasian dollars and the regional indices the position reverses, because these are their own peak hours.

The band and a Gulf morning 

For a reader in the Gulf this is the session that arrives at a workable hour, which is a scheduling fact rather than a recommendation, and it runs in both directions. Tokyo's open lands in the early local morning and the band has closed by the early afternoon. London begins around the middle of the Gulf working day and New York in the local evening, and the New York close falls in the middle of the night, so the hours in which the European and American majors do most of their business sit largely outside the region's working day.

That asymmetry is a consequence of longitude rather than a problem with a solution, and it produces a mismatch worth naming. The instruments a reader in the region can watch forming in real time are the Asian ones, while the instruments most written about in English language trading material are the European majors, whose busiest hours arrive after the local working day ends.

Where practitioners disagree 

The first argument is about the word quiet. One tradition treats the Asian session as a low activity stretch to be read rather than dealt in, because the majors move comparatively little while the cost of transacting in them is comparatively higher. Another answers that the description is a function of which instruments are on the screen: the yen complex, the Australasian currencies and the regional indices do most of their business in these hours, so calling the session quiet describes the observer's watchlist rather than the market. Both are internally consistent, which is why the argument does not resolve.

The second concerns whether the Asian range carries information. Several technical traditions treat the high and the low of the band as reference levels for the sessions that follow, reasoning that a range set by fewer participants is more likely to be revisited once the larger ones arrive. The objection is that those levels are widely published and widely watched, which is itself a reason orders cluster near them, so the convention may describe a self referential habit rather than a property of the market. Nothing settles it either way.