Global Markets and Trading Hours

The clock is the first market-structure fact

Equity markets are local businesses with local hours, and liquidity is shaped like a wave that travels as the world turns. **Tokyo** trades from 9:00 to 15:00 JST, which is evening into late night in New York. **London** opens at 8:00 London time, and the **London–New York overlap** — roughly 8:00 to 11:30 in New York — is when the two deepest pools of capital are awake at once, which is why the largest currency and cross-border equity volume in the world happens in those hours. When the United States closes, liquidity in U.S. names does not vanish: it moves to **extended hours**, where the book is thinner, spreads are wider, and there is no closing auction to anchor anyone (M11). This has three practical consequences. First, a foreign position has **two closes**, and a fund measured against one of them is trading a different product from a fund measured against the other: an ETF of Japanese stocks listed in New York closes in New York, while its holdings closed hours earlier in Tokyo. Second, the **first print of the day is not a prediction, it is a repricing** — every event that happened while your market slept is compressed into the opening auction, which is why gaps are the normal state of affairs for cross-listed names (M5). Third, a **holiday on one side of a wrapper is a trading day on the other**: your ADR keeps trading in New York on a Tokyo holiday, but the arbitrage that would normally keep it near its home value cannot be executed, because the other leg of the trade is closed. The currency market itself never closes in the way an exchange does — it is an over-the-counter network that runs from the Sydney open on Sunday evening to the New York close on Friday — so the exchange rate that matters for your return is being set even when the shares are not trading. That is why a foreign holding can move while its own market is shut, and why the FX rate and the share price rarely tell you the same story in the same hour. Who is awake, in New York time — Tokyo: 19:00 – 01:00 (previous evening to overnight) · London: 03:00 – 11:30 · London–New York overlap: 08:00 – 11:30 — the deepest window of the day · U.S. regular session: 09:30 – 16:00 · U.S. extended hours: 04:00 – 09:30 and 16:00 – 20:00, thin ←

A foreign stock is two bets

Your return in dollars is the product of two returns: what the shares did in their own currency, and what their currency did against yours. If the local return is `r` and the currency move is `c`, your dollar return is `(1 + r)(1 + c) − 1 = r + c + r·c`. The cross term is small when the moves are small and decisive when they are not: it is the difference between the +2% people expect and the +1.2% they get on a 10% gain with an 8% currency loss. So a foreign equity position is a currency position whether you asked for one or not, and the decision to **hedge** is a separate decision. Hedging — selling the foreign currency forward against your own — removes the currency return and costs the **interest-rate differential** plus a small spread: borrow conceptually at the lower-rate currency and lend at the higher, and the forward price is set so that arbitrage makes the two equivalent. In practice that means hedging a currency with a much lower interest rate than the dollar can cost several percent a year, which will quietly eat the reason you bought the stock. The honest framing is that currency hedging is a decision about volatility and cash flow, not a free option to remove a source of worry: it removes a risk you may or may not be paid to take. The third column of any international return is **what you are comparing against**. A U.S. investor holding a European index cares about the dollar return; a European investor holding the same fund cares about the euro return. The local index can be up 12% in its own currency while the dollar return is negative, and neither number is wrong. Any claim about a foreign market — "Europe outperformed" — is incomplete until it says in whose currency. The same year, three ways to say it — Local index return: +12.0% in euros · Currency return: −6.0% (euro weaker against the dollar) · Unhedged dollar return: +5.28% — 1.12 × 0.94 − 1 · Cost of hedging for the year: ≈ interest-rate differential, say 1.5% ← Adding returns and adding currency moves is the standard error and it always flatters the outcome at small moves and destroys it at large ones. Multiply.

Wrappers: ADRs, ratios and the hours nobody can arbitrage

You usually do not buy shares on a foreign exchange; you buy a **wrapper**. A **depositary receipt** — an ADR in the United States, a GDR elsewhere — is a security issued by a depositary bank that represents a fixed number of ordinary shares, held by the bank in the home market. The **ratio** tells you how many: one ADR for one share, or one for two, or one for ten, chosen so the ADR price lands in a range that suits the local market. A **sponsored** ADR is issued with the company's agreement, which is what allows it to be listed and to carry the company's reporting; an **unsponsored** one is created by a bank without the company's involvement, trades over the counter, and may be less liquid and less transparent. The shares underlying any of them are ordinary shares, and they carry the same economic rights — this is a wrapper, not a different class of stock. Because the wrapper and the shares are two expressions of the same claim, the wrapper's price has a computable fair value: the home price, converted at the spot rate and scaled by the ratio. Anything on top of that is a **premium or discount**, and the arbitrage that closes it has exactly the same shape as an ETF's (M16) with one extra leg: a dealer buys shares in Tokyo, has them converted into ADRs, and sells them in New York — which requires Tokyo to be **open**, the currency to be **convertible**, the shares to be **available for conversion**, and the depositary to be **issuing**. That is why the premium on a broad foreign ADR is small during the overlap and can be wide when New York is trading alone. In the hours when the home market is closed, the New York price is not a measurement of value — it is a **forecast** by whoever is willing to trade, made with no ability to hedge and no ability to arbitrage. This is also why a real premium can persist for other reasons: index inclusion that forces funds to buy the wrapper, a local restriction on foreign ownership that makes the wrapper one of the few legal doors, or the dividend withholding tax that applies differently to the wrapper and the share. As with every premium in this subject, the useful question is not "is it mispriced?" but "which leg of the arbitrage is blocked, and for how long?" The ratio is the exchange rate the wrapper uses before the real one. Two ADRs for ¥6,500 of stock means the ADR moves with half the local price in dollars, which is why the same 10% local rally raises the ADR by 10% — and a ¥ move raises it by however much the currency moved, at any hour New York chooses.

Sources

• Depositary receipts: ratios, levels and creationSEC — investor bulletin on depositary receipts
• Session overlaps and the shape of intraday liquidityBIS — Triennial Central Bank Survey of foreign exchange turnover
• Currency return, whether to hedge, and what hedging costsCFA Institute — currency management for equity portfolios
• Exchange calendars, holidays and half-daysNYSE / Nasdaq — trading calendars
• Following the sun across the world's equity sessionsWorld Federation of Exchanges — market hours and statistics