The ECB, the Bank of Japan and the People's Bank of China

The three things that actually differ 

Every institution here sets the price of short term money in its own currency, and through that influences the rates at which everything else in that currency is borrowed and lent. Three things separate them, and nearly every misreading of a foreign institution comes from taking an announcement to be about a different one of the three than it is. The mandate: what the law requires the institution to achieve, and which objective ranks higher where they conflict. The instrument: whether the lever is a price, a quantity of money, or the exchange rate itself. And transparency: how much reasoning is published, how quickly, and whether the vote is visible.

The European Central Bank 

The founding treaty gives the ECB a primary objective of price stability and, without prejudice to it, a secondary duty to support the Union's general economic policies. The ranking is the load bearing part: where one objective is explicitly subordinate, an argument for loosening policy to support activity has to survive a prior question about whether price stability permits it. The Governing Council, the Executive Board together with the governors of the euro area national central banks, has defined price stability as a numerical inflation target over the medium term, symmetric in the sense that undershooting it is treated as no more acceptable than overshooting it.

That composition points at a structural fact with no equivalent elsewhere in this lesson. One monetary policy is set for economies that are not synchronised, and several sovereign governments borrow in a currency none of them individually controls, so the yields on their debt can move apart and an easing decided in one place can arrive weakened in another. Hence a working vocabulary about transmission and fragmentation, and instruments aimed at transmission itself sitting beside the three official rates: the rate paid on funds deposited overnight, the rate on the main refinancing operations, and the rate on the marginal lending facility. Where reserves are abundant, no bank lends below what it can earn leaving money at the central bank overnight, so the deposit rate is the effective floor and the number that matters.

Communication runs on a fixed shape. The decision is published at a scheduled minute, then a prepared statement is read and questions taken a short while later, so the decision and its explanation reach the market as two separate events with a measurable gap between them. Accounts of the discussion follow weeks after. No member by member voting record is published, and no individual's own view of where rates should go; quarterly staff projections are the closest thing to a published path. The press conference therefore carries more of the week's information than the decision does.

The Bank of Japan 

The Bank of Japan Act gives the Bank price stability, described as contributing thereby to the sound development of the national economy, and responsibility for the stability of the financial system. A Policy Board of nine decides, the Governor, two Deputy Governors and six other members, and the vote is published. What makes the institution different is not that mandate but the problem it has worked on for a generation, the inverse of the one this module has described: inflation persistently below the target rather than above it, with the policy rate already at or beneath zero.

An institution that cannot move its main price far moves quantities instead, which is why the toolkit here is the widest of the three: asset purchases at announced scales, lending programmes aimed at bank credit, and, in one framework the Bank has run, a target for the yield on a longer dated government bond, held near a stated level by buying whatever quantity that required. A yield target changes what an announcement is about. The news becomes the target level and the width of the band around it, and a widening of that band appears on no calendar as a rate rise yet reprices every instrument anchored to the old band. The standing objection is that an administered yield stops carrying the information a discovered one carried, because the institution holding it there ends up owning much of the market whose price it quotes.

One institutional detail matters more to anyone watching the yen than any of that, and it is the one most often got wrong. Authority over foreign exchange intervention in Japan rests with the Ministry of Finance, not with the Bank, which executes in the market as the Ministry's agent. So a currency reaching levels the authorities are unwilling to accept produces comment, and at times action, from a finance ministry, while the rate decision is taken by a different committee on a different calendar. A release that moves the yen is not always a monetary policy release. On transparency the Bank sits between the other two: decision, vote, quarterly outlook report, press conference, minutes, and full transcripts after a decade.

The People's Bank of China 

The first thing to get right is the name. This is the People's Bank of China, and it is not the Bank of China, which is a separate organisation entirely: a large commercial bank with branches, retail customers and a stock market listing. The People's Bank operates under the State Council rather than as an authority independent of government, and its monetary policy committee is advisory rather than deciding. Its statutory objective is to maintain the stability of the value of the currency and thereby to promote economic growth, read domestically as inflation and externally as the exchange rate, which places an exchange rate objective inside the mandate itself.

The toolkit is a set of levers rather than one price. Reserve requirement ratios set the proportion of deposits banks must hold rather than lend, so changing them alters how much credit the system can create with no rate announced at all, and repurchase operations and a medium term lending facility set the rates at which the central bank supplies funds to banks. The mechanic with no counterpart in the other two institutions is the daily reference price: each morning, before onshore trading opens, a central parity rate for the currency against the dollar is published, and onshore trading that session is permitted only within a stated band either side of it. The band is a hard limit for the session, not a preference, and how much discretion goes into the reference price is not published, which is why practitioners read it as a communication rather than a quotation.

The band travels with the reference price, so a session's permitted range is settled before the session opens, and a currency pinned against one edge of it describes pressure the price is not permitted to express. The same currency has a second price offshore, traded outside the mainland and outside the band, and the gap between the two is followed because it measures what the onshore price is not being allowed to do. Transparency is otherwise where the distance from the ECB is widest: no press conference of the same kind, no published vote, and no statement setting out a stance in the terms the previous lesson on hawkish and dovish language described. Changes commonly become visible in the terms of an operation or in the drift of the reference price, so reading this institution is inference from what has been done rather than interpretation of what has been said.

What a managed exchange rate does to transmission 

An economy cannot simultaneously hold its exchange rate fixed, allow capital to move freely across its borders, and run a monetary policy set for domestic conditions. The three are jointly impossible for arithmetic rather than ideological reasons: if capital moves freely while the exchange rate cannot, any gap between the domestic rate and the anchor economy's rate is a difference capital will move to collect, and the flows required to hold the exchange rate still drag the domestic rate back toward the anchor's. An authority wanting a fixed rate and a policy of its own has to restrict the movement of capital. A managed float sits between the corners, holding the rate within limits rather than at a point, which buys some of each and settles none of it.

That constraint changes the chain this module has been building. The chain runs from inflation to rates to the currency, and every link assumes a currency free to move and capital free to chase a difference between two rates. Where the rate is managed, the last link is administered rather than discovered: pressure that would have appeared in the currency appears instead in the reserves the authority buys or sells and in the reference price it publishes next, and where capital movement is restricted the interest difference channel has less to work with. A release then reprices what the currency is not permitted to express: domestic rates, local equity, the offshore price of the same currency, and the currencies and commodities of the economies that sell into it. Watching a freely floating instrument as a proxy is a persistent habit here, and a habit rather than a rule, because the relationships drift and the proxy is a different instrument with drivers of its own. The cases where a rate is held against an anchor rather than merely managed are taken up in the lessons that follow.

Which instruments each most directly affects 

The instruments below are the ones conventionally described as most directly exposed to each institution. Directly is doing the work, because a decision anywhere in this lesson can reach almost any instrument through some chain or other. None of it states what any instrument will do.

• The ECB. Euro denominated currency pairs, European government bond markets and European equity index contracts. Because the euro is among the most heavily traded currencies, its meetings also register in pairs that do not contain it, through the dollar leg those pairs share.
• The Bank of Japan, and separately the Ministry of Finance. Yen pairs first, then Japanese equity index contracts, conventionally described as sensitive to the yen because much of the earnings behind the index is earned outside Japan. The yen's long standing conventional role as a funding currency is also why sharp yen moves are so often observed alongside moves in instruments with no obvious connection to Japan.
• The People's Bank of China. The onshore and offshore prices of the yuan, and mainland and Hong Kong equity indices. At one remove, industrial commodities and the currencies of economies that export heavily into China, which are the instruments practitioners most often reach for as proxies, with the limits of proxies noted above.

Where practitioners disagree 

The first argument is whether a quantity is comparable to a price. When an institution announces that it will buy a stated quantity of bonds, or hold a yield near a stated level, one tradition treats the announcement as equivalent to a rate change of some size and spends its effort estimating that equivalence. Another argues that most of the effect is a signal about the future path of the overnight rate, in which case the same announcement made without the signal would do very little. The two readings imply different things about what happens when a programme ends, and the occasions on which one has ended have not settled it.

The second is how to read an institution that does not explain itself. One tradition treats every operation as a message and reads its size, tenor and timing as deliberate. Critics answer that most operations are routine liquidity management with no content whatever, and that a discipline finding a signal in every roll will be wrong about most of them. The wider version of the argument is whether transparency helps at all: publishing votes, projections and a path either anchors expectations and spreads repricing across many small moments, or commits a committee in public and concentrates repricing into the moment the path is revised. The three institutions here sit at three different points on that spectrum.